Food Microbiology, 4th, Doyle

Food Microbiology: Fundamentals and Frontiers 4TH EDITION

Food Microbiology: Fundamentals and Frontiers 4TH EDITION

EDITED BY

Michael P. Doyle

Center for Food Safety University of Georgia Griffin, GA 30223-1797

Robert L. Buchanan

Center for Food Safety and Security Systems University of Maryland College Park, MD 20742

ASM Press Washington, DC 20036

Copyright © 2013 by ASM Press. ASM Press is a registered trademark of the American Society for Microbiology. All rights reserved. No part of this publication may be reproduced or transmitted in whole or in part or reutilized in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage and retrieval system, without permission in writing from the publisher. Disclaimer: To the best of the publisher’s knowledge, this publication provides information concerning the subject matter covered that is accurate as of the date of publication. The publisher is not providing legal, medical, or other professional services. Any reference herein to any specific commercial products, procedures, or services by trade name, trademark, manufacturer, or otherwise does not constitute or imply endorsement, recommendation, or favored status by the American Society for Microbiology (ASM). The views and opinions of the author(s) expressed in this publication do not necessarily state or reflect those of ASM, and they shall not be used to advertise or endorse any product.

Library of Congress Cataloging-in-Publication Data

Food microbiology : fundamentals and frontiers / edited by Michael P.   Doyle, Robert L. Buchanan.—4th ed.     p. ; cm.   Includes bibliographical references and index.   ISBN 978-1-55581-626-1—ISBN 978-1-55581-846-3 (e-book)   I. Doyle, Michael P. II. Buchanan, Robert, 1946  [DNLM: 1. Food Microbiology. QW 85] 664.001¢579—dc23 

2012038536

doi:10.1128/9781555818463 Printed in the United States of America 10  9  8  7  6  5  4  3  2  1 Address editorial correspondence to: ASM Press, 1752 N St., N.W., Washington, DC 20036-2904, USA Send orders to: ASM Press, P.O. Box 605, Herndon, VA 20172, USA Phone: 800-546-2416; 703-661-1593 Fax: 703-661-1501 E-mail: [email protected] Online: http://estore.asm.org

Contents

Contributors   ix Preface   xvii

I. Factors of Special Significance to Food Microbiology   1.  Physiology, Growth, and Inhibition of Microbes in Foods   3



Thomas J. Montville and Karl R. Matthews

  2.  Antimicrobial Resistance   19 Lu Zhang, Jennifer Cleveland McEntire, Rosetta Newsome, and Hua Wang   3.  Spores and Their Significance   45 Peter Setlow and Eric A. Johnson   4.  Microbiological Criteria and Indicator Microorganisms   81 Jean-Louis Cordier and the International Commission on Microbiological Specifications for Foods   5.  Biosecurity: Food Protection and Defense   91 Shaun P. Kennedy and Frank F. Busta

v

Contents

vi

II. Microbial Spoilage and Public Health Concerns   6.  Meat, Poultry, and Seafood   111 John N. Sofos, George Flick, George-John Nychas, Corliss A. O’Bryan, Steven C. Ricke, and Philip G. Crandall   7.  Milk and Dairy Products   169 Obianuju N. Nsofor and Joseph F. Frank   8.  Fruits and Vegetables   187 Frédéric Carlin   9.  Nuts, Seeds, and Cereals   203 Linda J. Harris, Joseph R. Shebuski, Michelle D. Danyluk, Mary S. Palumbo, and Larry R. Beuchat

III.  Foodborne Pathogenic Bacteria 10.  Salmonella Species   225 Haiping Li, Hua Wang, Jean-Yves D'Aoust, and John Maurer 11.  Campylobacter Species   263 Ihab Habib, Lieven De Zutter, and Mieke Uyttendaele 12.  Enterohemorrhagic Escherichia coli   287 Jianghong Meng, Jeffrey T. LeJeune, Tong Zhao, and Michael P. Doyle 13.  Cronobacter Species   311 Franco J. Pagotto and Kahina Abdesselam 14.  Yersinia enterocolitica   339 Roy M. Robins-Browne 15.  Shigella Species   377 Rachel Binet and Keith A. Lampel 16.  Vibrio Species   401 James D. Oliver, Carla Pruzzo, Luigi Vezzulli, and James B. Kaper 17.  Clostridium botulinum   441 Eric A. Johnson 18.  Clostridium perfringens   465 Bruce A. McClane, Susan L. Robertson, and Jihong Li 19.  Bacillus cereus   491 Per Einar Granum and Toril Lindbäck 20.  Listeria monocytogenes   503 Elliot T. Ryser and Robert L. Buchanan

Contents 21.  Staphylococcus aureus   547 Keun Seuk Seo and Gregory A. Bohach 22.  Epidemiology of Foodborne Diseases   575 Craig W. Hedberg

IV. Nonbacterial Pathogens 23.  Mycotoxins   597 Marta H. Taniwaki and John I. Pitt 24.  Foodborne Viral Pathogens   619 Lee-Ann Jaykus, Doris H. D’Souza, and Christine L. Moe 25.  Bovine Spongiform Encephalopathy   651 Paul Brown and Linda A. Detwiler 26.  Helminths in Meat   673 H. Ray Gamble and Dante S. Zarlenga 27. Helminths Acquired from Finfish, Shellfish, and Other Food Sources   697 Eugene G. Hayunga 28.  Protozoan Parasites   713 Ynes R. Ortega

V.  Preservatives and Preservation Methods 29.  Physical Methods of Food Preservation   737 Ahmed E. Yousef and V. M. Balasubramaniam 30. Chemical Preservatives and Natural Antimicrobial Compounds   765 P. Michael Davidson, T. Matthew Taylor, and Shannon E. Schmidt 31.  Biological Control of Foodborne Bacteria   803 Thomas J. Montville and Michael L. Chikindas

VI. Fermentations and Beneficial Microorganisms 32.  Fermented Dairy Products   825 Mark E. Johnson and James L. Steele 33.  Fermented Vegetables   841 Fred Breidt, Roger F. McFeeters, Ilenys Perez-Diaz, and Cherl-Ho Lee 34.  Fermented Meat, Poultry, and Fish Products   857 Steven C. Ricke, Ok Kyung Koo, and Jimmy T. Keeton

vii

Contents

viii

35.  Cocoa and Coffee   881 Sterling S. Thompson, Kenneth B. Miller, Alex S. Lopez, and Nicholas Camu 36.  Beer   901 Iain Campbell 37.  Wine   915 Mickey E. Parish and Graham H. Fleet 38.  Probiotics and Prebiotics   949 Erika A. Pfeiler and Todd R. Klaenhammer

VII. Advanced Techniques in Food Microbiology 39.  Genomics and Proteomics of Foodborne Microorganisms   975 Grace L. Douglas, Erika A. Pfeiler, Tri Duong, and Todd R. Klaenhammer 40.  Predictive Microbiology   997 E. Van Derlinden, L. Mertens, and J. F. Van Impe 41.  Microbial Risk Assessment   1023 Juliana M. Ruzante, Richard C. Whiting, Sherri B. Dennis, and Robert L. Buchanan 42. Hazard Analysis and Critical Control Point System: Use in Managing Microbiological Food Safety Risks   1039 Robert L. Buchanan and E. Noelia Williams 43.  Molecular Source Tracking and Molecular Subtyping   1059 Peter Gerner-Smidt, Eija Hyytia-Trees, and Timothy J. Barrett Index   1079

Contributors

Kahina Abdesselam Bureau of Microbial Hazards, Health Products and Food Branch, Food Directorate, Health Canada, Sir F.G. Banting Research Centre, P/L2204E, Ottawa, Ontario, K1A 0K9, Canada V. M. Balasubramaniam Department of Food Science and Technology, The Ohio State University, 2015 Fyffe Court, Columbus, OH 43210 Timothy J. Barrett Office of the Associate Director of Science, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333 Rachel Binet Center for Food Safety and Applied Nutrition, Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740 Gregory A. Bohach Department of Basic Sciences, Mississippi State University, Mississippi State, MS 39762 Fred Breidt U.S. Department of Agriculture, Agricultural Research Service, and Department of Food, Bioprocessing & Nutrition Sciences, NC State University, Raleigh, NC 27603 Paul Brown Commissariat à l’Énergie Atomique (CEA), Service d’Étude des Prions et des Infections Atypiques (SEPIA), 92265 Fontenay-aux-Roses, France

ix

x

Contributors Robert L. Buchanan Center for Food Safety and Security Systems, College of Agriculture and Natural Resources, 0119 Symons Hall, University of Maryland, College Park, MD 20742 Frank F. Busta National Center for Food Protection and Defense, University of Minnesota, 1954 Buford Ave., 120 LES, St. Paul, MN 55108 Iain Campbell International Centre for Brewing and Distilling, Heriot-Watt University, Edinburgh EH14 4AS, United Kingdom Nicholas Camu Cocoa Fermentation Department, Barry Callebaut, Aalstersestraat 122, 9280 Lebbeke, Belgium Frédéric Carlin INRA, UMR408, Sécurité et Qualité des Produits d’Origine Végétale, F-84000 Avignon, and Université d’Avignon et des Pays de Vaucluse, F-84000 Avignon, France Michael L. Chikindas Department of Food Science, Rutgers - The State University of New Jersey, New Brunswick, NJ 08901-8520 Jean-Louis Cordier Nestec Ltd., Nestlé Quality Assurance Center, CH-1800 Vevey, Switzerland Michelle D. Danyluk Department of Food Science and Human Nutrition, Citrus Research and Education Center, University of Florida, 700 Experiment Station Road, Lake Alfred, FL 33850-2290 Jean-Yves D’Aoust Food Directorate, Health Products & Food Branch, Health Canada, Ottawa, Ontario, Canada P. Michael Davidson Department of Food Science and Technology, University of Tennessee, Knoxville, TN 37996-4591 Lieven De Zutter Department of Veterinary Public Health and Food Safety, Ghent University, Faculty of Veterinary Medicine, Salisburylaan 133, B-9820 Merelbeke, Belgium S. B. Dennis Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740 Linda A. Detwiler Department of Pathobiology and Population Medicine, College of Veterinary Medicine, Mississippi State University, Mississippi State, MS 39762

Contributors Grace L. Douglas Department of Food, Bioprocessing & Nutrition Sciences, North Carolina State University, Raleigh, NC 27695-7624

Michael P. Doyle Center for Food Safety, University of Georgia, Griffin, GA 30223 Doris H. D’Souza Department of Food Science and Technology, University of Tennessee, Knoxville, TN 37996 Tri Duong Department of Poultry Science, Texas A&M University, 2472 TAMU, College Station, TX 77843 Graham H. Fleet Department of Food Science and Technology, The University of New South Wales, Sydney, New South Wales 2052, Australia

George Flick Department of Food Science & Technology, Virginia Tech, Blacksburg, VA 24061 Joseph F. Frank Department of Food Science and Technology, University of Georgia, Athens, GA 30602-7610 H. Ray Gamble National Research Council, 500 Fifth Street NW, Washington, DC 20001 Peter Gerner-Smidt Enteric Diseases Laboratory Branch, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333 Per Einar Granum Department of Food Safety and Infection Biology, Norwegian School of Veterinary Science, Oslo, Norway Ihab Habib Department of Food Safety and Food Quality, Ghent University, Faculty of Bioscience Engineering, Ghent, Belgium, and Division of Food Hygiene and Control, Alexandria University, High Institute of Public Health, Alexandria, Egypt Linda J. Harris Department of Food Science and Technology, University of California, One Shields Avenue, Davis, CA 95616-8598 Eugene G. Hayunga Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892 Craig W. Hedberg Division of Environmental Health Sciences, School of Public Health, University of Minnesota, Minneapolis, MN 55455

xi

xii

Contributors Eija Hyytia-Trees Enteric Diseases Laboratory Branch, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333 The International Commission On Microbiological Specifications For Foods Lee-Ann Jaykus Department of Food, Bioprocessing and Nutrition Sciences, North Carolina State University, Raleigh, NC 27695-7624 Eric A. Johnson Department of Bacteriology, Botulinum Toxins Laboratory, Food Research Institute, University of Wisconsin, 1550 Linden Drive, Madison, WI 53706 Mark E. Johnson Center for Dairy Research, Department of Food Science, University of Wisconsin – Madison, Madison, WI 53706-1565 James B. Kaper Department of Microbiology and Immunology, Center for Vaccine Development, University of Maryland School of Medicine, 685 West Baltimore St., Baltimore, MD 21201 Jimmy T. Keeton Department of Nutrition and Food Science, 122 Kleberg Center, 2253 Texas A&M University, College Station, TX 77843-2253 Shaun P. Kennedy National Center for Food Protection and Defense and College of Veterinary Medicine, University of Minnesota, 1954 Buford Ave., 120 LES, St. Paul, MN 55108 Todd R. Klaenhammer Department of Food, Bioprocessing & Nutrition Sciences, Schaub Hall, Box 7624, North Carolina State University, Raleigh, NC 27695-7624 Ok Kyung Koo Center for Food Safety, Department of Food Science, University of Arkansas, 2435 North Hatch, Fayetteville, AR 72704 Keith A. Lampel Center for Food Safety and Applied Nutrition, Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740 Cherl-Ho Lee Graduate School of Biotechnology, Korea University, Seoul, 136-701 Korea Jeffrey T. LeJeune Food Animal Health Research Program, Ohio Agricultural Research and Development Center, Wooster, OH 44691 Haiping Li Division of Food Processing Science and Technology, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 6502 S. Archer Rd., Bedford Park, IL 60501

Contributors Jihong Li Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, 420 Bridgeside Point II Building, 450 Technology Drive, Pittsburgh, PA 15219 Toril Lindbäck Department of Food Safety and Infection Biology, Norwegian School of Veterinary Science, Oslo, Norway

Alex S. Lopez 6621 Creeping Thyme Street, Las Vegas, NV 89148 Karl R. Matthews Department of Food Science, School of Environmental and Biological Sciences, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-8520 John Maurer Department of Population Health, University of Georgia, Athens, GA 30602 Bruce A. McClane Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, 420 Bridgeside Point II Building, 450 Technology Drive, Pittsburgh, PA 15219 Jennifer Cleveland McEntire Leavitt Partners, 299 South Main Street, Suite 2400, Salt Lake City, UT 84111 Roger F. McFeeters U.S. Department of Agriculture, Agricultural Research Service, and Department of Food, Bioprocessing & Nutrition Sciences, NC State University, Raleigh, NC 27603 (retired) Jianghong Meng Department of Nutrition and Food Science, University of Maryland, College Park, MD 20742 L. Mertens BioTeC – Chemical and Biochemical Process Technology and Control, Department of Chemical Engineering, KU Leuven, B-3001 Leuven, Belgium Kenneth B. Miller Microbiology Research Technical Center, Hershey Foods Corporation, 1025 Reese Avenue, Hershey, PA 17033-0805 Christine L. Moe Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, GA 30322 Thomas J. Montville Department of Food Science, School of Environmental and Biological Sciences, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-8520 Rosetta Newsome Institute of Food Technologists, Chicago, IL 60607

xiii

xiv

Contributors Obianuju N. Nsofor U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740 George-John Nychas Agricultural University Athens, Laboratory of Microbiology & Biotechnology of Foods, 75 Iera Odos, Athens 11855, Greece Corliss A. O’Bryan Department of Food Science, University of Arkansas, Fayetteville, AR 72701 James D. Oliver Department of Biology, University of North Carolina at Charlotte, 9201 University City Blvd., Charlotte, NC 28223 Ynes R. Ortega Center for Food Safety, University of Georgia, 1109 Experiment Street, Griffin, GA 30223-1797 Franco J. Pagotto Bureau of Microbial Hazards, Health Products and Food Branch, Food Directorate, Health Canada, Sir F.G. Banting Research Centre, P/L2204E, Ottawa, Ontario, K1A 0K9, Canada Mary S. Palumbo Western Center for Food Safety, University of California, One Shields Avenue, Davis, CA 95616-8598 Mickey E. Parish U.S. Food and Drug Administration, 5100 Paint Branch Pkwy., College Park, MD 20740 Ilenys Perez-Diaz U.S. Department of Agriculture, Agricultural Research Service, and Department of Food, Bioprocessing & Nutrition Sciences, NC State University, Raleigh, NC 27603 Erika A. Pfeiler Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740 John I. Pitt CSIRO Animal, Food and Health Sciences, P.O. Box 52, North Ryde, New South Wales 1670, Australia Carla Pruzzo Department of Biology, University of Genova, Corso Europa 26, 16132 Genova, Italy Steven C. Ricke Center for Food Safety, Department of Food Science, University of Arkansas, 2650 North Young Avenue, Fayetteville, AR 72704-5690 Susan L. Robertson Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, 420 Bridgeside Point II Building, 450 Technology Drive, Pittsburgh, PA 15219

Contributors Roy M. Robins-Browne Department of Microbiology and Immunology, The University of Melbourne, Victoria 3010, and Murdoch Childrens Research Institute, Royal Children’s Hospital, Parkville, Victoria 3052, Australia J. M. Ruzante Pew Health Group, The Pew Charitable Trusts, 901 E St. NW, Washington, DC 20004 Elliot T. Ryser Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824-1225 Shannon E. Schmidt Pecan Deluxe Candy Company, Dallas, TX 75212-6308 Keun Seuk Seo Department of Basic Sciences, Mississippi State University, Mississippi State, MS 39762 Peter Setlow Department of Molecular, Microbial and Structural Biology, University of Connecticut Health Center, Farmington, CT 06030-3305 Joseph R. Shebuski Corporate Food Safety and Regulatory Affairs, Cargill, Incorporated, 15407 McGinty Road West, Wayzata, MN 55391 John N. Sofos Center for Meat Safety & Quality, Department of Animal Sciences, Colorado State University, Fort Collins, CO 80523 James L. Steele Department of Food Science, University of Wisconsin - Madison, 1605 Linden Drive, Madison, WI 53706-1565 Marta H. Taniwaki Instituto de Tecnologia de Alimentos, Av. Brasil, 2880, Campinas, São Paolo CEP 13070-178, Brazil T. Matthew Taylor Department of Animal Science, Texas A&M University, College Station, TX 77843-2471 Sterling S. Thompson Microbiology Research Technical Center, Hershey Foods Corporation, 1025 Reese Avenue, Hershey, PA 17033-0805 Mieke Uyttendaele Health Products & Food Branch, Health Canada, Ottawa, Ontario, Canada E. Van Derlinden BioTeC – Chemical and Biochemical Process Technology and Control, Department of Chemical Engineering, KU Leuven, B-3001 Leuven, Belgium

xv

xvi

Contributors J. F. Van Impe BioTeC – Chemical and Biochemical Process Technology and Control, Department of Chemical Engineering, KU Leuven, B-3001 Leuven, Belgium Luigi Vezzulli Department of Biology, University of Genova, Corso Europa 26, 16132 Genova, Italy Hua Wang (chapter 2) Department of Food Science and Technology, The Ohio State University, 110 Parker Food Science and Technology Building, 2015 Fyffe Court, Columbus, OH 43210 Hua Wang (chapter 11) Division of Microbiology, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740 R. C. Whiting Exponent, Inc., Center for Chemical Regulation and Food Safety, 17000 Science Dr., Suite 200, Bowie, MD 20715 E. Noelia Williams Department of Nutrition and Food Science, College of Agriculture and Natural Resources, University of Maryland, Skinner Hall, College Park, MD 20742 Ahmed E. Yousef Department of Food Science and Technology, The Ohio State University, 2015 Fyffe Court, Columbus, OH 43210 Dante S. Zarlenga U.S. Department of Agriculture, Agricultural Research Service, 10300 Baltimore Avenue, Beltsville, MD 20705 Lu Zhang Department of Food Science and Technology, The Ohio State University, 110 Parker Food Science and Technology Building, 2015 Fyffe Court, Columbus, OH 43210 Tong Zhao Center for Food Safety, University of Georgia, Griffin, GA 30223

Preface

The history of civilization is intimately linked with the ability of humankind to acquire sufficient food so that we could devote time and resources to pursue the arts and sciences. While historians tend to focus on the contributions of agriculture, those gains would have been for naught if means for preserving foods had not been developed, first by techniques developed over the centuries by observation and trial-and-error, and more recently by the ever increasing application of science and engineering. At the core of these advances is our knowledge of food microbiology. Well before Antonie van Leeuwenhoek described his “living animalcules,” many of the conditions that controlled microbiological spoilage had been identified empirically. However, it was the emergence of the science of microbiology that moved food preservation from an art to a science, allowing foods to be processed, distributed, and marketed with a high degree of confidence in terms of both the product’s quality and safety. Thus, food microbiology has been a major part of the discipline since its very earliest days. The scope of food microbiology is highly inclusive, interfacing with virtually all microbiology subdisciplines (e.g., public health microbiology, microbial genetics, fermentation technologies, microbial physiology). Furthermore, food microbiologists have been in the forefront of many microbiological concepts and advances. For example, the development of biofilms and the ability to detect low numbers of metabolically stressed microbes from highly complex matrices are two areas where food microbiologists are providing critical insights into the behavior of microbiological systems. Furthermore, new research topics have arisen as a result of the unique challenges facing food microbiologists, such as predictive microbiology, probiotics, microbial risk assessments, and naturally occurring antimicrobials.

xvii

xviii

Preface Since its initial publication in 1997, the goal of Food Microbiology: Funda­ mentals and Frontiers (FMFF) has been to provide an advanced reference that explores both the breadth and depth of food microbiology. As such, it provides only a limited review of the basic principles, concepts, and techniques of food microbiology. FMFF has been designed to be used in combination with one of the many good, general texts used to teach food microbiology at the undergraduate level. This allows FMFF to build on that knowledge base to address advanced concepts and techniques and to provide the reader with insights into where food microbiology is going in the next 5 years, based on the advances in the past 5 years. Thus, new editions have been introduced every 4 to 6 years to capture current advances in food microbiology and emerging food microbiology concerns. For example, a key focus of the 3rd edition was the expanding insights gained through the application of molecular biology to food microbiology. That tradition continues in the 4th edition. The intervening period between the 3rd and 4th editions has witnessed a rapid advancement of our basic knowledge of foodborne microorganisms (e.g., the differentiation between virulent and avirulent strains), the expansion of the types of foods associated with foodborne disease events (e.g., the increased recognition of dried foods and fresh produce as vehicles of pathogenic microorganisms), the continued globalization of the food industry, and the search for new prevention and intervention technologies to enhance the quality and safety of foods. As before, the 4th edition has been organized into seven sections based on the various roles that microorganisms play in the production, processing, and protection of foods. These begin with a section on “Factors of Special Significance to Food Microbiology,” which includes chapters on the impact of the food environment on the growth and survival of foodborne microorganisms, antimicrobial resistance, endospores, proper use of microbiological criteria and indicator organisms, and the expanding role of food defense in the overall protection of the food supply. The second section, “Microbial Spoilage and Public Health Concerns,” considers four broad classes of foods: (i) meat, poultry, and seafood, (ii) milk and dairy products, (iii) fruits and vegetables, and (iv) nuts, seeds, and cereals. These commodity-oriented chapters introduce the types of microbiological spoilage associated with foods and explain how the food environment selects for these forms of spoilage. Further, the chapters identify the types of microbial hazards that have been associated with these commodities. The third and fourth sections of the 4th edition are devoted to “Foodborne Pathogenic Bacteria” and “Nonbacterial Pathogens.” These sections cover all of the major foodborne microbial agents associated with foods. Each chapter explores diseases associated with the pathogen, the epidemiology and etiology of the pathogen, the microorganism’s virulence determinants, and control measures used to prevent foodborne disease. There has been a concerted effort in this edition to expand these chapters to include advances that have been made in the control of the pathogen along the food chain. The fifth section, “Preservatives and Preservation Methods,” provides detailed consideration of the three general approaches to food preservation, namely, physical methods, chemical methods, and biological methods. Each chapter covers the key underpinning concepts as well as recent advances in finding alternatives to traditional preservation techniques. The sixth section, “Fermentations and Beneficial Microorganisms,” covers the wide range of foods that are produced by fermentation processes, including some that may not be commonly known to undergo a fermentation step. There is a strong emphasis in these chapters on explaining how recent advances in food microbiology are dramatically changing our

Preface

xix

approaches to traditional methods of using microorganisms to produce unique food and beverage products. Finally, the seventh section, “Advanced Techniques in Food Microbiology,” includes five chapters on different advanced systems to investigate, evaluate, and manage microbiological concerns related to foods. These include (i) application of advanced genomics and proteomics, (ii) predictive microbiology and related mathematical modeling, (iii) quantitative microbial risk assessment, (iv) food safety risk management systems, and (v) molecular source typing and subtyping. The 4th edition of FMFF represents a substantial update from the previous editions, and we are confident that you will find it a valuable resource both now and for years to come. As you use this edition, we would appreciate your taking the time to provide the editors feedback regarding the subject matter covered and additional materials that should be considered for future editions. Thank you in advance for your assistance. Michael P. Doyle Robert L. Buchanan

Factors of Special Significance to Food Microbiology

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch1

Thomas J. Montville Karl R. Matthews

Physiology, Growth, and Inhibition of Microbes in Foods

Food microbiologists must understand microbiology, molecular biology, and food systems and be able to inte­ grate them to solve problems in complex food ecosys­ tems. This chapter addresses three issues: (i) the ability of bacteria to use different biochemical pathways to gener­ ate the energy required to grow under adverse condi­ tions in foods; (ii) the interaction of bacteria and foods in ecosystems in which the cells may exist in a variety of physical and physiological states and in which the roles of intrinsic and extrinsic factors are discussed; and (iii) the kinetics of microbial growth.

Microbial Physiology and metabolism All things progress to the state of maximum random­ ness in the absence of energy input. Since life is an or­ dered process, all living things must generate energy. Foodborne bacteria do this by oxidizing reduced com­ pounds. Oxidation occurs only in a chemical couple in which the oxidation of one compound is linked to the reduction of another. In the case of aerobic organisms, be they bacteria, yeasts, or fungi, the carbon source is oxidized to carbon dioxide, oxygen is reduced to water,

1

and 38 ATP molecules are generated. Most of the ATP is generated through oxidative phosphorylation in the elec­ tron transport chain. In oxidative phosphorylation, the energy of the electrochemical gradient generated when oxygen is used as the terminal electron acceptor drives the formation of a high-energy bond between inorganic phosphate and an adenine nucleotide, typically ADP for the formation of ATP. Anaerobic bacteria, which lack functional electron transport chains, must reduce an in­ ternal compound through the process of fermentation. They generate only one or two moles of ATP per mole of hexose catabolized. In this case, ATP is formed by sub­ strate level phosphorylation and the phosphate group is transferred from an organic compound to the adenine nucleotide.

Glycolytic Pathways: Carbon Flow and Substrate Level Phosphorylation Embden-Meyerhof-Parnas Pathway

The most commonly used pathway for glucose catabo­ lism (glycolysis) is the Embden-Meyerhof-Parnas (EMP) pathway (Fig. 1.1). In many organisms, the pathway is bidirectional (i.e., amphibolic) and can synthesize

Thomas J. Montville and Karl R. Matthews, Department of Food Science, School of Environmental and Biological Sciences, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-8520.



1.  Growth, Survival, and Death of Microbes in Foods

9 Factors of Special Significance

Figure 1.1  Major catabolic pathways used by foodborne bacteria. Figure 1.2 Major catabolic pathways used by foodborne bacteria. doi:10.1128/9781555818463.ch1f1

EMP pathway is used to produce pyruvate, which is glucose, glycogen, or starch. The overall rate of glycoly­ then reduced by lactate dehydrogenase, forming lactic sis isand regulated by theNAD. activity of phosphofructokinase. acid regenerating Some Lactobacillus species, This as enzyme convertsplantarum fructose-6-phosphate to fruc­ such Lactobacillus (137), are charactertose-1,6-bisphosphate. Phosphofructokinase activityare is ized as “facultatively heterofermentative.” Hexoses subject to allosteric the binding of their preferred carbonregulation, source andwhereby are metabolized by the AMP or ATP at one site inhibits or pentoses stimulates, homofermentative pathway. If only arerespec­ availtively,the the of fructose-6-phosphate at able, cellphosphorylation shifts to a heterofermentative mode. When the enzyme’s site. Fructose-1,6-bisphosphate grown at lowactive hexose concentrations, these bacteriaacti­ do

not make enough fructose-1,6-bisphosphate to activate vates lactate dehydrogenase below) so that them the flow their dehydrogenase.(see This also causes to of carbon to pyruvate is tightly linked to the regenera­ shift to heterofermentative catabolism. tion of NAD when pyruvate is reduced to lactic acid. Another key enzymeAcid of theCycle EMP pathway is aldolase. The Tricarboxylic The tricarboxylic ultimate fermentation endcycle products the The acid (TCA) links generated glycolytic by pathcatabolism of pentoses and hexoses are2 partially deter­ ways to respiration. It generates NADH and FADH 2 as mined by for which enzyme converts the sugars smaller substrates oxidative phosphorylation while to providing units. Aldolase cleaves substrate one molecule of fructose-1,6additional ATP through level phosphorylation.

1.  Physiology, Growth, and Inhibition of Microbes in Foods bisphosphate to two three-carbon units: dihydroxyace­ tone phosphate and glyceraldehyde-3-phosphate. Other glycolytic pathways use keto-deoxy-phosphogluconate (KDPG) aldolase to make two three-carbon units or phosphoketolase to produce one two-carbon compound and one three-carbon unit. Substrate level phosphoryla­ tion generates a net gain of two ATP molecules when 1,3-diphosphoglycerate and phosphoenolpyruvate do­ nate phosphoryl groups to ADP.

Entner-Doudoroff Pathway

The Entner-Doudoroff pathway is an alternate glyco­ lytic pathway that yields one ATP molecule per molecule of glucose and diverts one three-carbon unit to biosyn­ thetic pathways. In aerobes that use this pathway, such as Pseudomonas species, the difference between form­ ing one ATP molecule by this pathway and forming two ATP molecules by the EMP pathway is inconsequential compared to the net 34 ATP molecules formed from oxidative phosphorylation. In the Entner-Doudoroff pathway, glucose is converted to 2-keto-3-deoxy-6phosphogluconate. The enzyme KDPG aldolase cleaves this to one molecule of pyruvate (directly, without the generation of an ATP molecule) and one molecule of 3phosphoglyceraldehyde. The 3-phosphoglyceraldehyde is then catabolized by the same enzymes used in the EMP pathway with the generation of one ATP molecule by substrate level phosphorylation using phosphoenol pyruvate as the phosphoryl group donor.

Heterofermentative Catabolism

Heterofermentative bacteria, such as Leuconostoc spp. and some Lactobacillus spp., have neither aldolases nor KDPG-aldolase. The heterofermentative pathway is based on pentose catabolism and yields one ATP molecule. The pentose can be transported into the cell from the environment or generated intracellularly by decarboxyl­ ation of hexoses. In either case, the pentose is converted to xylulose-5-phosphate with ribulose-5-phosphate as an intermediate. The xylulose-5-phosphate is cleaved by phosphoketolase to a glyceraldehyde-3-phosphate and a two-carbon unit that can be converted to acetaldehyde, acetate, or ethanol. Although this pathway yields only one ATP molecule, it offers cells a competitive advan­ tage by allowing them to utilize pentoses, which homo­ lactic organisms cannot catabolize.

Homofermentative Catabolism

Homofermentative bacteria in the genera Lactococcus, Pediococcus, and some Lactobacillus species produce lactic acid as the sole fermentation product. The EMP pathway is used to produce pyruvate, which is then



reduced by lactate dehydrogenase, forming lactic acid and regenerating NAD. Some Lactobacillus spp., such as L. plantarum, are facultatively heterofermentative. Hexoses are their preferred carbon source and are me­ tabolized by the homofermentative pathway. If only pen­ toses are available, the cell shifts to heterofermentative fermentation. When grown at low hexose concentrations, these bacteria do not make enough fructose-1,6-bisphos­ phate to activate their lactate dehydrogenase. This causes them to shift to heterofermentative catabolism.

Tricarboxylic Acid Cycle

The tricarboxylic acid (TCA) cycle links gylcolytic path­ ways to respiration. It generates NADH2 and FADH2 as substrates for oxidative phosphorylation while providing additional ATP through substrate level phosphorylation. For each glucose molecule, the TCA cycle catabolizes 2 pyruvate + 2ADP + 2FAD + 8NAD ® 6CO2 + 2ATP + 2FADH2 + 8NADH. Succinic acid, oxaloacetate, and α-ketoglutarate link the TCA cycle to amino acid bio­ synthesis. The TCA cycle is used by all aerobes, but some anaerobes lack all of the enzymes required to have a functional TCA cycle. The TCA cycle is also the basis for two indus­ trial fermentations important to the food industry. The microbial production of citric acid by Aspergillus niger and Aspergillus wentii and of glutamic acid by Corynebacterium glutamicum depends on mutations that affect α-ketoglutarate dehydrogenase and cause TCA intermediates to accumulate.

Aerobes, Anaerobes, the Regeneration of NAD, and Respiration

The flow of carbon to pyruvate always consumes NAD, which must be regenerated for continued catabolism. When NADH2 is oxidized to NAD, another compound must be reduced, i.e., serve as an electron acceptor. Aerobes use molecular oxygen as the terminal electron acceptor during oxidative phosphorylation. As electrons travel down the electron transport chain, protons are pumped out, forming a proton gradient across the mem­ brane. This proton gradient can be converted to ATP by the action of the F0F1 ATPase. Oxidation of NAD(P)H2 yields three ATP. Oxidation of FADH2 yields two ATP. ATP and NADH are thus, in a sense, interconvertible. Nitrate and sulfur can also serve as terminal electron ac­ ceptors in “anaerobic respiration.” Anaerobes, in contrast, have a fermentative metabo­ lism. Fermentations oxidize carbohydrates in the ab­ sence of an external electron acceptor. The final electron acceptor is an organic compound produced during

Factors of Special Significance

 carbohydrate catabolism. In the most obvious case, py­ ruvic acid is the terminal electron acceptor when it ac­ cepts an electron from NADH and is reduced to lactic acid. Some anaerobes are aerotolerant and can generate more energy in the presence of low levels of oxygen than in its absence. For example, some lactic acid bacteria have inducible NADH oxidases that regenerate NAD by reducing molecular oxygen to H2O2. This spares the use of pyruvate as an electron acceptor and allows it to be converted to acetic acid with the generation of an ad­ ditional ATP. These lactic acid bacteria have an NADH peroxidase which detoxifies the H2O2. Obligate anaer­ obes cannot detoxify H2O2 and die when exposed to air.

Bioenergetics

All catabolic pathways generate energy with which the bacteria can perform useful work. The preceding sec­ tion on microbial biochemistry stressed the role of ATP in the cell’s energy economy, but transmembrane gra­ dients of other compounds play an equally important role. Transmembrane gradients release energy when one compound moves from high concentration to low con­

centration (i.e., “with the gradient”). This energy can be coupled to the transport of a second compound from a low concentration to a high concentration (i.e., “against the gradient”). The most important of these gradients is the proton gradient. The proton motive force (PMF) has two components. An electrical component, the membrane potential (∆Ψ), represents the charge potential across the membrane. The transmembrane pH gradient (∆pH) is the second component. Together, these constitute the PMF as stated by the equation PMF = ∆Ψ − z∆pH. In this equation, z = 2.3 RT/F, where R is the gas constant, T is the absolute temperature (in degrees Kelvin), and F is the Faraday constant. The factor z converts the pH gradient into milli­volts and has a value of 59 mV at 25°C. The PMF is defined as being interior negative and alkaline, resulting in a negative value. (In the equation above, z∆pH is not being subtracted from ∆Ψ but makes this negative term more negative.) PMF values can be as high as −200 mV for aerobes, or in the range of −100 mV to −150 mV for anaerobes. Protein phosphorylation, flagellar synthesis and rotation, reversed electron transfer, and protein transport use PMF as an energy source.

Figure 1.2  Proton motive force can be generated by respirations, ATP hydrolysis, end prod­ uct reflux, or anion exchange mechanisms. doi:10.1128/9781555818463.ch1f2

1.  Physiology, Growth, and Inhibition of Microbes in Foods PMF is generated by several mechanisms (Fig. 1.2). The translocation of protons down the electrochemical gradient during respiration generates a proton gradient. The oxidation of NADH is accompanied by the export of protons to generate three ATP. The proton gradient is converted to ATP by the F0F1 ATPase when it is driven in the direction of ATP synthesis. The bacterial F0F1 ATPase is nearly identical to chloroplast and mitochon­ drial F0F1 ATPases. The F0F1 ATPase is reversible. Aerobes use it to con­ vert PMF to ATP. Anaerobes convert ATP to PMF. Maintaining internal pH homeostasis may be the princi­ pal role of the F0F1 ATPase in anaerobes. In this case, it hydrolyzes ATP to pump protons out of the cytoplasm and maintain a constant internal pH. Internal pH not only influences the activity of cytoplasmic enzymes but also regulates the expression of genes responsible for functions ranging from amino acid degradation to viru­ lence. Most bacteria maintain their internal pH (pHi) near neutrality, but lactic acid bacteria can tolerate lower pHi values and expend less ATP on pH homeosta­ sis. Acid-induced death is the direct result of an exces­ sively low pHi. Because of their limited capacity for ATP generation, some lactic acid bacteria also generate ∆pH by ATPindependent mechanisms. The electropositive excretion of protons with acidic end products has been demon­ strated for lactate and acetate. For example, under some conditions, Lactobacillus plantarum excretes three pro­ tons per molecule of acetate, thus sparing one ATP. The antiport (see below) exchange of precursor and prod­ uct in anion-degrading systems, such as the malate2−: lactate1− exchange of the malolactic fermentation, might contribute to the generation of ∆Ψ. Bacteria have evolved several mechanisms to achieve similar ends. The accumulation of compounds against a gradient (i.e., transport) is work and requires energy. In the case of primary transport systems and group trans­ location, this work is done by phosphoryl group transfer (Fig. 1.2). Secondary transport systems are fueled by the energy stored in the gradients that constitute the PMF.

MICROBIAL GROWTH KINETICS The growth of bacteria, yeasts, and molds is character­ ized by growth curves having four parts. These are the lag, exponential (logarithmic or log), stationary, and death phases. Food microbiology is concerned with all four phases of microbial growth. These phases and the growth kinetics described below are based on binary replication of DNA (i.e., one copy makes two copies, which make four copies, which make eight copies, etc.).



In bacteria, which replicate by binary fission, increases in cell number closely correspond to the replication of DNA. Thus, the growth curves can be plotted as the number of cells (CFU/ml) on a logarithmic scale or log10 CFU/ml versus time. This method may not be appropri­ ate for yeasts, which replicate by budding, because each cell may have many buds at a given period. In this case, the log of the optical density will give accurate estimates of growth. Fungi are even more problematic since they grow both by branching, which is exponential, and hy­ phal elongation, which is linear. In this case, it is the log of the cell dry weight that is generally plotted against time. During the lag phase, cells adjust to their new envi­ ronment by inducing or repressing enzyme synthesis and activity, initiating replication of DNA, and in the case of spores, differentiating into vegetative cells. The length of the lag phase depends on temperature, the inoculum size (larger inocula usually have shorter lag phases), and the physiological history of the organism. If actively growing cells are inoculated into an identical fresh me­ dium at the same temperature, the lag phase may van­ ish. Conversely, these factors can be manipulated to extend the lag phase beyond the time at which some other food quality attribute (such as proteolysis or browning) becomes unacceptable. Foods are generally considered microbially safe if obvious spoilage precedes microbial growth. However, “spoiled” is a subjective and culturally biased concept. It is more prudent to cre­ ate conditions that prevent growth altogether. During the exponential or log phase of growth, bac­ teria reproduce by binary fission. Thus, during expo­ nential growth, classical reaction kinetics can be used to describe the change in cell numbers. Food microbi­ ologists often use doubling times as the kinetic constant to describe the rate of logarithmic growth. Doubling times (td), which are also referred to as “generation” times (tgen), are related to a variety of kinetic constants as shown in Table 1.1. The influence of different parameters on a food’s final microbial load can be illustrated by manipulating the equations in Table 1.1. Equation 1a states that the num­ ber of bacteria (N) at any time is directly proportional to the initial number of organisms (N0). Thus, decreas­ ing the initial microbial load 10-fold will reduce the cell number at any time 10-fold, although at extended times the population from the lower inoculum may reach the same final number. Because the instantaneous specific growth rate (µ) and time are in the power function of the equation, they have more marked effects on N. Consider a food for which N0 = 1 × 104 CFU/g and µ = 0.2 h−1 at 37°C. After 24 h, the cell number would be 1.2 ×

Factors of Special Significance



Table 1.1  First-order kinetics can be used to describe exponential growth and inactivation Growtha

Thermal inactivationb

1a. N = N0eµt

1b. N = N0e−kt

2a. 2.3log(N/N0) = µ∆t

2b. 2.3log(N/N0) = −(k∆t)

3a. ∆t = [2.3log(N/N0)]/µ

3b. ∆t = −[2.3log(N/N0)]/k

4a. td = 0.693/µ

4b. D = 2.3/k 5b. Ea =

Irradiationc 1c. N = N0e−D/D0

2.3RT1T2 9 ´ z 5

a N, cell number (CFU/g); N0, initial cell number (CFU/g); t, time (h); µ, specific growth rate (h−1); td, doubling time (h). b k, rate constant (h−1); D, decimal reduction time (h); Ea, activation energy (kcal/mol); T1, T2, reference temperature and test temperature (°K), respectively. c D0, rate constant (h−1); D, dose (Gy).

106 CFU/g. Reducing the initial number 10-fold will re­ duce the number after 24 h 10-fold, to 1.2 × 105 CFU/g. However, reducing the temperature from 37 to 7°C has a much more profound effect. If one makes the simplifying assumption that the growth rate decreases twofold with every 10°C decrease in temperature, then µ will be de­ creased eightfold, to 0.025 h−1 at 7°C. When equation 1a is solved using these values (i.e., N = 104 e0.025 × 24), then N at 24 h is 1.8 × 104 CFU/g. Both time and temperature have much greater influence over the final cell number than does the initial microbial load. Equation 3a can be used to determine how long it will take a microbial population to reach a certain level. Consider the case of ground meat manufactured with an N0 of 1 × 104 CFU/g. How long can it be held at 7°C before reaching a level of 108 CFU/g? According to equation 3, t = [2.3(log 108/104)]/0.025, or 368 h. Food microbiologists frequently use doubling times (td) to describe growth rates of foodborne microbes. The relationship between td and µ is more obvious if equa­ tion 2a is written using natural logs [i.e., ln(N/N0) = µ∆t] and solved for the condition where t = td and N = 2N0. Since the natural log of 2 is 0.693, the solution for equation 2a is 0.693/µ = td (equation 4a). The aver­ age rate constant k, defined as the number of genera­ tions per unit of time (i.e., 1/tgen), is also used by applied Table 1.2  Representative specific growth rates and doubling

times of microorganisms

Organisms and conditions

m (h−1)

td (h)

  Optimal conditions

2.3

0.3

  Limited nutrients

0.20

3.46

Psychrotrophs, 5°C

0.023

Molds, optimal conditions

0.1–0.3

microbiologists. The instantaneous growth rate con­ stant, µ, is related to k by the equation µ = 0.693k. Both rate constants characterize populations in the exponen­ tial phase of growth. Some typical specific growth rates and doubling times are given in Table 1.2.

Food ecosystems

Foods as Ecosystems

Ecosystems are composed of the environment and the organisms that inhabit it. Ecosystems are also impacted by the generation and utilization of energy. The food ecosystem is composed of intrinsic factors, which are inherent to the food (i.e., pH, water activity [aw], and nutrients) and extrinsic factors, which are external to it (i.e., temperature, gaseous environment, and the pres­ ence of other bacteria). This distinction is somewhat artificial, since gasses can diffuse into the food, the pres­ ence of other bacteria can physically change the com­ position of the food, and even the distinction between aerobic and anaerobic can change over time. Foods can be heterogeneous on a micrometer scale and may con­ tain several distinct microenvironments. Heterogeneity and its associated gradients of pH, oxygen, nutrients, aw, etc., are key ecological factors in foods. The organisms in the ecosystems can exist in a va­ riety of physiological and physical states. They can be injured, unculturable, or communicating in structured communities. All of these factors influence the bacteria in food.

Physiological States

Bacteria

30 6.9–20

Cells exist in many different physiologies and forms. Cells “injured” by sublethal stressors are unable to grow on selective media but can still be cultured on nonselective media (Fig. 1.3, top). When injured cells repair, they regain their ability to grow on selective

1.  Physiology, Growth, and Inhibition of Microbes in Foods

Injury

Figure 1.3  Data illustrating injury and repair of bacteria (top) compared to VNC bacteria (bottom). Cells that undergo injury when sublethally stressed show smaller populations when plated on a selective medium (dashed line) than do those plated on a nonselective medium (solid line). As the cells re­ pair, resistance to selective agents is regained and the popula­ tion levels obtained on the selective medium approach that of the nonselective medium. In the case of VNC cells (bottom), a temperature downshift (¯) results in the loss of the ability to be enumerated on any medium. Culturability is regained upon temperature upshifts (�). doi:10.1128/9781555818463.ch1f3

media, even though the total number of viable cells does not increase, as evidenced by the constant viability ob­ tained on nonselective media. Cells that are “viable but nonculturable” (VNC) are fundamentally different from those that are injured (Fig. 1.3, bottom). VNC cells can­ not be cultured on any medium, even though other mea­ surements of physiological activity indicate that they are still viable. Bacteria are also more social than previously thought. They can respond to environmental conditions via signal transduction or even to their own population level by quorum sensing. Bacteria may even (and per­ haps predominantly) exist as structured communities in biofilms.



Injury is defined as the inability of cells exposed to sub­ lethal stress to grow on selective media, while retain­ ing culturability on nonselective media (14). Injured cells are more sensitive to selective agents and may have increased nutritional requirements. Microorganisms may be injured by sublethal levels of stressors such as heat, radiation, acid, or sanitizers. Injury due to freezing at −20°C for 24 hours can cause a 2-log10 apparent reduction of an Escherichia coli O157:H7 popu­ lation. When injured cells of Listeria monocytogenes are enumerated on selective media, their D values can be much lower than those obtained with the enumeration of the same injured cells on nonselective media. Sanitizer tests using selective media may indicate that listeriae are killed. However, viable cells can be recovered by using listeria repair broth. Cells considered killed by the first method are shown to be merely injured by the second method. The type of food influences both injury and subsequent recovery. Molecular events associated with injury are complex and are still being defined. Injury is influenced by time, temperature, concentration of injuri­ ous agent, strain of target pathogen, and experimental methodology. Injury can encompass physiological and genetic changes in a microorganism. For example, bacterial cells exposed nearly instantaneously to high heat often suffer greater loss of viability than cells exposed incrementally to the same final temperature. Incremental exposure permits the regulation of heat shock genes that enhance cell survival under more-extreme conditions. This phe­ nomenon has been demonstrated in both gram-positive and gram-negative foodborne pathogens. Similarly, Salmonella or E. coli O157:H7 organisms exposed to extremely high concentrations (50 ppm) of sodium or calcium hypochlorite fail to recover. However, at lower concentrations both pathogens exhibit injury based on differential plating. The fermentation ability and vi­ ability of Saccharomyces cerevisiae are significantly decreased during freeze-thaw injury. The use of frozen dough baking involves freeze-thaw treatment, resulting in freeze-thaw injury to the yeast cells. An insufficiency of copper ion homeostasis may be one of the causes of freeze-thaw injury in yeast. Microbial injury is important to food safety for sev­ eral reasons. (i) If injured cells appear dead due to the use of selective media in thermal resistance studies, the thermal resistance will be underestimated and the D values will be errantly low. (ii) Injured cells that escape detection at the time of postprocessing sampling may repair before the food is consumed and present a safety or spoilage problem. (iii) The “selective agent” may be

10 a common food ingredient such as salt, organic acids, humectants, or even suboptimal temperature. Data illustrating injury are shown in Fig. 1.3. Cells subjected to a mild stress are plated on a rich nonse­ lective medium and a selective medium containing 6% NaCl. The difference between the numbers of colonies on the two media represents injured cells. (If 107 CFU/ ml of a population are enumerated on the nonselective medium and 104 CFU/ml can grow on the selective me­ dium, then 9,990,000 CFU/ml are injured.) Specialized enumeration media are often required because growth and gene expression of an organism cultured on nonse­ lective media may be different from those expressed by organisms on selective media. Cells injured by heating, freezing, and detergents usually leak intracellular constituents from damaged membranes. Membrane integrity is reestablished during repair. Osmoprotectants can prevent or minimize freeze injury in L. monocytogenes. Oxygen toxicity also causes injury. Recovery of injured cells is often enhanced by adding peroxide-detoxifying agents such as catalase or pyruvate to the recovery medium or by excluding oxy­ gen through the use of anaerobic incubation conditions or adding Oxyrase (which enzymatically reduces oxy­ gen) to the recovery medium. Repair is the process by which cells recover from in­ jury (i.e., regain the ability to grow on selective media). Repair requires de novo synthesis of RNA and protein and often manifests as an extended lag phase. The ex­ tent and rate of repair are influenced by environmental factors. The repair of cells injured under similar condi­ tions may require dramatically different periods for full recovery based on the temperature at which the injured cells are held following injury (e.g., longer recovery at 4°C versus 37°C). For example, injured L. monocytogenes can start to repair immediately at 37°C, but repair at 4°C is delayed for a week or more.

Viable but Nonculturable

“Viable but nonculturable” is defined as a state in which cells cannot be cultured on any medium, even though their viability is demonstrated by nonculturable methods, as explained below (3, 8). Salmonella, Campylobacter, Escherichia, Shigella, Vibrio, and other genera can exist in a state wherein they are viable but cannot be cultured. The differentiation of vegetative cells into a dormant VNC state is a survival strategy for many nonsporulat­ ing species. The VNC state is morphologically different from that of the “normal” vegetative cell. During the transition to the VNC state, rod-shaped cells shrink and become small spherical bodies (which are not spores). It takes from 2 days to several weeks for an entire popu­

Factors of Special Significance lation of vegetative cells to become VNC. Changes in membrane fatty acid composition occur in Vibrio dur­ ing entry into the VNC state. The viability of VNC cells is demonstrated through cytological methods. The structural integrity of the bac­ terial cytoplasmic membrane can be determined by the permeability of cells to fluorescent nucleic acid stains. Bacteria with intact cell membranes stain fluorescent green, whereas bacteria with damaged membranes stain fluorescent red. Iodonitrotetrazolium violet can also identify VNC cells. Respiring cells reduce iodonitrotet­ razolium violet to form an insoluble compound detect­ able by microscopic observation. Cell populations as determined by culture methods may actually be orders of magnitude greater when quantified using nonculture methods. Experimental data in Fig. 1.3 illustrate a Vibrio vulnificus population that appears to have died off (i.e., gone through a 6-log reduction in CFU/ml), although >105 cells per ml are quantified as viable by nonculture methods (7). VNC cells can also be identified by their substrateresponsive metabolism. When VNC cells are incubated with yeast extract (as a nutrient) and nalidixic acid or ciprofloxacin (an inhibitor of cell division), their elonga­ tion can be quantified microscopically. The method(s) used varies, depending on the bacteria for which the state of VNC is to be demonstrated. This method de­ tects VNC cells of L. monocytogenes. Other meth­ ods have been used to demonstrate the VNC state for Streptococcus faecalis, Micrococcus flavus, and Bacillus subtilis. Additional methods for detecting VNC cells are being developed as understanding of bacteria at the molecular level and techniques for genetic manipulation advance. Reverse transcriptase PCR, reporter genes (e.g., green fluorescent protein tagged and lux tagged), and other methods are used to provide evidence of protein synthe­ sis. The detection of specific RNA by reverse transcrip­ tase PCR is one such method. Alternatively, reporter genes such as those used in green fluorescent proteintagged and lux-tagged methods can identify cells that are synthesizing proteins, even though they cannot be cultured. Because the VNC state is most often induced by nutrient limitation in aquatic environments, it might appear irrelevant to the nutrient-rich milieu of food. However, the VNC state can be induced by changes in salt concentration, exposure to hypochlorite, and shifts in temperature. Foodborne pathogens in nutri­ tionally rich media can become VNC when shifted to refrigerated temperature. When Vibrio vulnificus populations are shifted to refrigeration temperatures,

1.  Physiology, Growth, and Inhibition of Microbes in Foods the entire population of 105 viable cells become non­ culturable (<0.04 CFU/ml) but maintain lethality in mice, where they can be resuscitated. VNC campylo­ bacters have been resuscitated when injected into fertil­ ized chicken eggs and held at 37°C for 48 h (2). E. coli and Salmonella enterica serovar Typhimurium can enter a VNC state following chlorination of wastewa­ ter (10). Although unculturable, they may still present a public health hazard. Temperature changes can induce the VNC state; some organisms enter this state at low and others at high temperatures. When starved at 4 or 30°C for more than a month, Vibrio harveyi became VNC at 4°C but remained culturable at 30°C. In con­ trast, E. coli entered the VNC state at 30°C but died at 4°C. There is also evidence that yeasts can also enter a VNC state. Yeasts in the VNC state were observed in Botrytisaffected wine following addition of high concentrations of sulfur dioxide to stop the alcoholic fermentation (4). The high levels of alcohol and sulfur dioxide create an extremely hostile environment for yeast. In order to sur­ vive such a harsh environment, the yeast enters a VNC state. A “refermentation” can occur when the yeast exits the VNC state. Resuscitation of VNC cells is demonstrated by an increase in culturability that is not accompanied by an increase in the total cell numbers. The return to cul­ turability can be induced by temperature shifts or the gradual return of nutrients. The same population of bacteria can go through multiple cycles of the VNC and culturable states in the absence of growth. Inhibitors of transcription and translation inhibit resuscitation. The addition of catalase or sodium pyruvate to media re­ stores culturability of VNC E. coli O157:H7 and Vibrio parahaemolyticus (9). This suggests that the transfer of cells to nutrient-rich media initiates rapid production of superoxide and free radicals. Catalase hydrolyzes H2O2. Sodium pyruvate degrades H2O2 through decar­ boxylation of the α-keto acid to form acetic acid and CO2. VNC cells are found in marine, soil, and gastroin­ testinal environments. Indeed, as many as 99% of bac­ teria in the biosphere may be unculturable. Increased awareness of the VNC state may lead to a reexamina­ tion of our reliance on cultural methods to monitor the viability of microbes in foods. The mechanisms of VNC formation, the mechanisms by which VNC cells resus­ citate, and the mechanisms that regulate resuscitation are largely unknown. The relationship between viability and culturability needs further discussion. Does VNC reflect a true state of bacteria or, as the name implies, simply our inability to culture them?

11

Quorum Sensing and Signal Transduction

The explosion of papers on cell-to-cell communication gives new perspectives to many issues in food micro­ biology (5). Quorum sensing and signal transduction both regulate genes that would be superfluous to iso­ lated cells but are advantageous to large populations. Cellular communication occurs by two fundamen­ tally different mechanisms: signal transduction and quorum sensing. Two-component signal transduction systems are comprised of a membrane-spanning histi­ dine kinase sensor and a response regulator protein. Typical two-component systems are found in Bacillus and E. coli. Similar three-component systems are used extensively by lactic acid bacteria, which can excrete small, often antimicrobial, peptides as the autoinducer. Quorum-sensing systems are based on the diffusion of autoinducers across cell membranes to initiate a spe­ cific physiological response when a threshold concen­ tration is reached. When the threshold concentration is reached, the autoinducers interact with intracellular regulator proteins to modulate gene transcription.

Quorum Sensing

Microorganisms use quorum sensing to determine if they have a sufficient population size to express a given phenotype (1). The term “quorum sensing” is derived from human legislative bodies in which a quo­ rum (i.e., a certain number of participants) is required before action can be taken. In microbial quorum sens­ ing, cells produce a signal compound that diffuses into the environment. When the bacterial population is low, the extracellular concentration of the signal mol­ ecule remains low and the signal molecule continues to diffuse away from the cell. However, when there is a large population of cells that produce the signal mol­ ecule, its extracellular concentration increases and it diffuses back into the cell. This elicits the phenotypic response. This mechanism is fundamentally differ­ ent from signal transduction. Rather than acting on a transmembrane protein, the signal compound diffuses across the membrane into the cell. It binds to a regu­ lator protein that affects transcription of regulon(s) to elicit a cellular response. The signal compound is made by a gene product of the same regulon and hence is autoinduced. In gram-negative bacteria, N-acyl homoserine lac­ tones (abbreviated in the literature as both AHLs and HSLs) generally act as signaling molecules. These are referred to as AI-1 (autoinducer 1) and are synthesized by AHL synthase, encoded by the luxI gene. AHLs ob­ tain their species specificity from their differing acyl side chains. AI-2, originally thought to be unique to Vibrio,

Factors of Special Significance

12 is a furanosyl-bromide diester product of LuxS. LuxS is encoded by luxS, which is also involved in the synthe­ sis of the newly discovered AI-3. At high concentration, these molecules bind to and activate a transcriptional activator, which in turn induces target gene expression.

Examples of Cell Signaling in Foodborne Microbes

Given the current excitement about quorum sensing, it is tempting to speculate that it has a role in spoilage. However, there are few actual data to support this. Many studies have used a bioluminescence response in V. harveyi as evidence of quorum sensing in the foodborne or­ ganism under investigation. However, such studies have little meaning unless the autoinducing compound has been isolated and the regulated phenotype has been identified in that foodborne microbe. Campylobacter, Salmonella, E. coli O157:H7, Enterobacteriaceae, Pseudomonas, Aeromonas, Shewanella, and Photobacterium produce a bioluminescence response in V. harveyi. Such signals have been detected in broth, chicken soup, milk, bean sprouts, vacuum-packaged beef, fish fillet, and turkey. A more rigorous study casts doubt on the linkage of quorum sensing and spoilage. Thin-layer chroma­ tography and mass spectrometry were used to demon­ strate that the bioluminescence-inducing compound in samples of commercial vacuum-packaged meat was

N-3-oxo-hexanoyl homoserine lactone. However, meat inoculated with wild-type strains and meat inoculated with AHL synthase knockout mutants spoiled at the same rate. Furthermore, addition of halogenated furones (quorum-sensing inhibitors) did not influence spoilage, leading to the conclusion that quorum sensing does not regulate spoilage in vacuum-packaged meat. Molecules in foods may mimic or alter quorumsensing signaling systems of spoilage and pathogenic bacteria in food. Probiotic bacteria are thought to have health-promoting effects on the host. The precise mech­ anism by which probiotic bacteria such as Lactobacillus acidophilus achieve this is still not entirely understood. Probiotic bacteria have been shown to prevent infection with E. coli O157:H7 in mouse models. Research sug­ gests that probiotics produce small biologically active molecules that interfere with quorum-sensing systems of E. coli O157:H7, limiting the ability of the pathogen to adhere to tissues and cause lesions (6). The yeast S. cerevisiae, commonly used in bread making, exhibits quo­ rum-sensing behavior. The communication molecules are the aromatic alcohols tryptophol and phenyletha­ nol. The transition between the solitary yeast form and the filamentous form is regulated by quorum sensing. There are many phenotypes for which a role for cell signaling has been established at a genetic level (Table 1.3). Many of these effects are pleiotropic. The discov­

Table 1.3  Examples of quorum sensing in food microbiology Organism L. monocytogenes

Signal system Signal transduction 2 component

Phenotype

Genetic involvement

Growth at low temperature and high salt concentration

kdpE, orfX (rsbQ homolog)

Virulence

pclA, hly, actA, inlA, host srcFR (encodes kinase that acts on actin)

S. aureus

Signal transduction 2 component

Pleiotropic effects on cytotox­ ins, enterotoxins, proteases

arg (accessory gene regula­ tor locus) activated by RAP signaling peptide

Lactic acid bacteria

Signal transduction 3 component

Bacteriocin production

cln locus in Carnobacterium piscicola, pln locus in L. plantarum, nis locus in L. lactis

S. enterica serovar Typhimurium

Quorum sensing, AI-2

“Fitness” in chickens

Pleiotropic effect of luxS

Enteropathogenic E. coli

Quorum sensing, AI-2

Flagella, formation of attach­ ment and effacement lesions

Pleiotropic (?) effect of luxS, qse (quorum sensing regula­ tor), ee (enterocyte efface­ ment) locus

V. cholerae

Quorum sensing, AI-1, AI-2, other

Virulence

Activates virulence regulon by repressing hapR

1.  Physiology, Growth, and Inhibition of Microbes in Foods ery of autoinduction in bacteriocin-producing lactic acid bacteria explains the hitherto puzzling loss of bac­ teriocin production by cells that still have the requisite genes (and the ability to coax it back by adding super­ natants from normally producing cultures) and the fact that some strains produce bacteriocins on agar but not liquid media. Four criteria should be met before quorum sensing is attributed to a given organism: (i) the production of signal compound is specific to an event, (ii) the signal ac­ cumulates extracellularly, (iii) a unique response is gen­ erated when the signal compound reaches a threshold concentration and is recognized by a specific receptor, and (iv) the response goes beyond metabolism or detoxi­ fication of the signal compound. Rarely are these criteria met. The widespread produc­ tion of AI-2 by many microbes may have little to do with quorum sensing but rather be related to its role as a waste product of the activated methyl cycle.

Transduction

Two-component signal transduction systems consist of a histidine kinase receptor and a response regulator. An extracellular “trigger” molecule binds at the N terminus on the “out” side of membrane-spanning protein ki­ nase. Unlike quorum sensing, the trigger molecule does not diffuse into the cell. The protein kinase transduces (i.e., transmits) the signal across the membrane through a conformational change that increases the kinase activ­ ity at its cytoplasmic side. The increased kinase activity phosphorylates a response regulator protein. The phos­ phorylated response regulator protein can modulate gene expression, enzymatic activity, flagellar rotation, or other phenotypes. The signal molecules of grampositive bacteria are usually small posttranslationally processed peptide signals. Lactic acid bacteria use a three-component signal transduction system. The signal is a small peptide coded by structural genes on the operon. It is excreted, some­ times after posttranslational modification. When the peptide reaches a certain extracellular concentration, it binds to a specific receptor, transduces a signal to phos­ phorylate a response regulator, and upregulates its own synthesis. This has been studied extensively for the lan­ tibiotic nisin.

Biofilms

A biofilm is an aggregation of cells, often of multiple species, into heterogeneous complex structures that are attached to a solid surface (11). Biofilm formation is a multistep process. First, the solid surface undergoes a conditioning process that allows cells to be absorbed by

13

weak reversible electrostatic forces. Biopolymer forma­ tion follows rapidly and anchors these cells. The syn­ thesis of the matrix polymer may be upregulated by quorum sensing when the local concentration of cells increases by adsorption. This is followed by the forma­ tion of microcolonies having defined boundaries that allow fluid channels to run through the biomatrix. This may be considered a primitive circulatory system. Such a circulatory system requires higher-level differentiation, quorum sensing, or some kind of cell-to-cell communi­ cation to prevent undifferentiated growth from filling in these channels, which bring nutrients and remove wastes. Finally, cells can be sloughed off the biofilm to initiate new biofilms or take up residence in other envi­ ronments (such as food). The role of quorum sensing in biofilm formation has been established for some bacte­ ria, but conflicting evidence exists for the role of quo­ rum sensing in biofilm formation for many foodborne pathogens. Cells in biofilms can be more resistant to heat, chemi­ cals, and sanitizers than are planktonic (i.e., free, single) cells. The lethality to L. monocytogenes of a combina­ tion of sodium hypochlorite and heat is approximately 100 times lower in biofilms than for free cells. Increased chemical resistance is attributed to the very slow growth rates of cells in biofilms and not to a diffusional barrier created by the biomatrix. Indeed, cells in the nutrientdepleted interior of the microcolony may be in the VNC state. Biofilms pose special challenges to the food indus­ try. Foodborne pathogens E. coli O157:H7, L. monocytogenes, Yersinia enterocolitica, and Campylobacter jejuni form biofilms on food surfaces and food contact equipment, leading to serious safety issues. Only proper cleaning can ensure that the cells in the nascent biofilm can be reached by sanitizers before they become recal­ citrant in fully developed biofilms. Trisodium phos­ phate is effective against E. coli O157:H7, C. jejuni, and Salmonella serovar Typhimurium cells in biofilm. Inactivation of E. coli O157:H7 biofilms occurs through the synergistic effect of an alkaline cleaner and a bac­ teriophage. A more novel approach uses bacteriophage engineered to express a biofilm-degrading enzyme. This treatment reduces the biofilm cell count by approxi­ mately 3 log. Other methods for control of biofilms in­ clude superhigh magnetic fields, ultrasound treatment, proteolytic and glycolytic enzymes, and high pulsed electric fields. The design of equipment with smooth highly polished surfaces also impedes biofilm formation by making the initial adsorption step more difficult (see the review by Simoes et al. [12]). Protein expression can vary depending on whether cells are in a biofilm or planktonic. In C. jejuni, a foodborne

14 pathogen commonly associated with raw poultry, an array of proteins is upregulated in biofilm-associated cells compared to planktonic cells. Expression of proteins in­ volved in general (GroEl and GroES) and oxidative (Tpx and Ahp) stress responses, adhesions (Peb1 and FlaC), and biosynthesis and energy generation are altered. There is a desperate need for research that will yield a better understanding of biofilms. Planktonic cells are easy to study, and pure culture is the foundation of mi­ crobiology as we know it. However, in most natural en­ vironments bacteria reproduce on surfaces rather than in liquids. To a large degree, food microbiologists study microbes in their domesticated (i.e., planktonic) setting rather than their natural, attached state.

Factors That Influence Microbial Growth

Intrinsic and extrinsic factors are both important to mi­ crobial growth. Extrinsic factors such as temperature are external to the food. Those factors inherent to the food are intrinsic factors. These include natural food compounds, preservatives, the oxidation-reduction po­ tential, aw, and pH. Most of these factors are covered separately in the chapters on physical and chemical methods of food preservation. The influences of tem­ perature, pH, and aw are particularly important and covered in some depth below.

pH Intracellular pH (pHi) must be maintained above some critical level at which intracellular proteins become irre­ versibly denatured. Three progressively more stringent mechanisms, the homeostatic response, the acid toler­ ance response (ATR), and the synthesis of acid shock protein, maintain a pHi consistent with cell viability. These have been studied most extensively in S. enterica serovar Typhimurium. The homeostatic response helps cells maintain viabil­ ity at mildly acidic external pH (pHo) values, typically greater than 6.0. The homeostatic response maintains pHi by modulating the activity of proton pumps, anti­ ports, and symports to increase the rate at which pro­ tons are expelled from the cytoplasm. The homeostatic mechanism is constitutive and functions in the presence of protein synthesis inhibitors. The ATR is triggered by a pHo of 5.5 to 6.0 and main­ tains pHi of >5.0 at pHo values as low as 4.0. In L. monocytogenes, ATR appears to involve the membrane-bound F0F1 ATPase proton pump. In Enterobacteriaceae, at least four regulatory systems, an alternative sigma fac­ tor, a 2-component signal transduction system (PhoPQ), the major iron regulatory protein Fur, and Ada (involved in adaptive response to alkylating agents), are involved

Factors of Special Significance with acid survival. Loss of the gene encoding the gen­ eral transcription factor σB in L. monocytogenes dimin­ ishes acid tolerance but has no effect on virulence in a mouse model. Induced ATR in E. coli O157:H7 alters the expression of 86 genes, of which 6 are important for survival at low pH. The ATR differs for log-phase and stationary-phase cells. In Salmonella Typhimurium, OmpR is critical to stationary-phase ATR but not to the log-phase ATR. Acid-adapted Salmonella cells have increased resistance to a low-pH gastric environment, which may increase virulence (see the review by Wesche et al. [13]). The ATR may confer cross-protection to other envi­ ronmental stressors. Acid adaptation increases heat and freeze-thaw resistance of E. coli O157:H7. The expo­ sure of Salmonella serovar Typhimurium cells to pH 5.8 for a few cell doublings renders the cells less sensitive to sodium chloride and heat. Survival of acid-adapted L. monocytogenes exposed to nisin is approximately 10-fold greater than that of nonadapted cells. Acidadapted L. monocytogenes cells also have increased re­ sistance against heat shock, osmotic stress, and alcohol stress. Acid adaptation of E. coli O157:H7 enhances thermotolerance. The third way that cells regulate pHi, the synthe­ sis of acid shock proteins, is triggered by a pHo from 3.0 to 5.0. Acid shock proteins are trans-acting regu­ latory proteins. The majority of acid-induced proteins in L. monocytogenes are common for the responses to acid adaptation and acid stress, but some are unique. Three stationary-phase-dependent acid resistance sys­ tems protect E. coli O157:H7 under extremely acid (pH 2.5 or less) conditions. These include the oxida­ tive or glucose-repressed system, the glutamate de­ carboxylase system, and the arginine decarboxylase system. DNA-binding proteins (Dps) interact with DNA to form stable complexes that protect the DNA from acid-mediated damage. An organism deficient in Dps is less likely to survive under highly acidic conditions. Survival of an E. coli O157:H7 dps mutant is signifi­ cantly less (4-log CFU/ml reduction) than that of the parent strain (1-log CFU/ml reduction) after acid (pH 1.8) exposure. External pH (pHo) can regulate the expression of genes governing proton transport, amino acid degra­ dation, adaptation to acidic or basic conditions, and even virulence. The expression of the Y. enterocolitica inv gene in laboratory media at 23°C but not at 37°C seems paradoxical, since its expression is required for infection of warm-blooded animals. However, at the pH of the small intestine (5.5), the inv gene is expressed at 37°C. The yst gene, which codes for a heat-stable

1.  Physiology, Growth, and Inhibition of Microbes in Foods enterotoxin in Y. enterocolitica, is regulated similarly. The toxR gene, which controls expression of cholera toxin in Vibrio cholerae, is regulated in part by pH. In Salmonella, exposure to low pH enhances survival in macrophages. S. enterica serovar Dublin virulence genes are induced by low pH. Exposure of Salmonella serovar Enteritidis to pH 10 or 1.5% trisodium phosphate sig­ nificantly increases thermotolerance.

Osmoregulation Controlling aw has been used for centuries as a means to preserve food. aw is a measure of the available water in a food, reflecting the moisture content that under nor­ mal conditions can be exchanged between the product and the environment. The controlled flow of water by a bacterial cell is important to ensure survival. Bacteria must be capable of rapidly responding to the loss or gain in water associated with changes in osmolarity of the environment. This section will focus on the physi­ ological and genetic responses that bacteria utilize in osmoregulation. Foods are complex environments that can present major barriers to osmoregulation. Bacteria experience hyperosmotic shock when they are transferred to envi­ ronments that have higher solute concentrations than are found in the cytoplasm. This results in the efflux of water from the cell until the osmotic activity of the cytoplasm and that of the environment are balanced. Conversely, transfer of bacteria to an environment low in solutes causes rapid water entry into the cell, which must be offset by rapid release of solutes to the environ­ ment. Osmoregulation is the ability to control the influx and efflux of solutes from the cell. The movement of water is essentially passive. Research indicates that a range of genes are associ­ ated with osmoregulation. Bacteria must accumulate compatible solutes in high-osmolarity environments. Compatible solutes include quaternary amines, amino acids, sugars, and a range of peptides. E. coli accu­ mulates trehalose in order to sustain growth at high osmolarity. The genes encoding for trehalose synthetic enzymes are under the control of the rpoS sigma factor. The pool size of compatible solutes proline and beta­ ine are controlled by the ProP and ProU transport sys­ tems, respectively. Osmoregulation in Staphylococcus aureus is similar to that of E. coli, regulating the trans­ port of proline and betaine. There is a specific system for proline (PutP) and betaine (BPI) and a less specific system for accumulation of both proline and betaine (BPII). Similar to other organisms, L. monocytogenes uses proline and betaine as compatible solutes to grow at high osmolarity. However, this organism also uses

15

peptides as a source of compatible solutes. S. cerevisiae accumulates glycerol as a compatible solute, which is partially controlled by the high-osmolarity glycerolsignaling systems. Bacteria are often presented with many challenges to survival in a food matrix. Their response requires a coordinated response. In E. coli, Salmonella, and L. monocytogenes, specific stress-induced sigma factors are central to such global responses. Therefore, expo­ sure of a bacterium to a given food preservation regime may actually increase its ability to survive exposure to other stresses linked with food preservation. In E. coli, the RpoS sigma factor (ss) regulon includes over 50 dif­ ferent genes that ultimately confer resistance to a wide range of stress conditions (osmotic stress, starvation, and low pH). Food microbiologists must recognize that exposure of a foodborne pathogen to a sublethal food preservation regime may actually enhance the ability of the pathogen to survive future stresses associated with food preservation. Various microbes have different aw requirements. Decreasing the aw increases the lag phase of growth, decreases the growth rate, and decreases the number of cells at stationary phase. Gram-negative species usually require the highest aw. Gram-negative bacteria such as Pseudomonas spp. and most Enterobacteriaceae usu­ ally grow only above aw of 0.96 and 0.93, respectively. Gram-positive non-spore-forming bacteria are less sen­ sitive to reduced aw. Many Lactobacillaceae have mini­ mum aw near 0.94. Some Micrococcaceae grow below aw of 0.90. Staphylococci are unique among foodborne pathogens because some strains can grow at a mini­ mum aw of about 0.86 under aerobic conditions or 0.90 under anaerobic conditions. Toxin production may re­ quire more favorable conditions. Most spore-forming bacteria do not grow below aw of 0.93. Spore germina­ tion and outgrowth of Bacillus cereus is prevented at aw of 0.97 to 0.93. The minimum aw for Clostridium perfringens spore germination and growth is between 0.97 and 0.95. Several yeast species grow at aw levels lower than those of bacteria. Salt-tolerant species such as Debaryomyces hansenii, Hansenula anomala, and Candida pseudotropicalis grow well on cured meats and pickles at NaCl concentrations of up to 11% (aw = 0.93). Some xero­ tolerant species (such as Zygosaccharomyces species Z. rouxii, Z. baillii, and Z. bisporus) grow on and spoil foods such as jams, honey, and syrups having high sugar content (and correspondingly low aw). Molds generally grow at lower aw than foodborne bacteria. The most common xerotolerant molds belong to the genus Eurotium. Their minimal aw for growth is

Factors of Special Significance

16 0.71 to 0.77, while the optimal aw is 0.96. True xero­ philic molds such as Monascus (Xeromyces) bisporus do not grow at aw of >0.97 to 0.99. The relationship of aw to mold growth and toxin formation is complex. Under marginal environmental conditions of low pH, low aw, or low temperature, molds may be able to grow but not make toxins.

Temperature Temperature and gas composition are the primary ex­ trinsic factors influencing microbial growth. The influ­ ence of temperature on microbial growth and physiology cannot be overemphasized. While the influence of tem­ perature on growth kinetics is obvious and covered here in some detail, the influence of temperature on gene expression is equally important. Cells grown at re­ frigerated temperature express different genes and are physiologically different from those grown at ambient temperature. Later chapters provide organism-specific details about how temperature regulates phenotypes ranging from motility to virulence. A rule of thumb in chemistry suggests that reac­ tion rates double with every 10°C increase in tempera­ ture. This simplifying assumption is valid for bacterial growth rates only over a limited range of organismdependent temperatures (Fig. 1.4). Bacteria are classi­ fied as psychrophiles, psychrotrophs, mesophiles, and thermophiles according to how temperature influences their growth. Both psychrophiles and psychrotrophs grow, albeit slowly, at 0°C. True psychrophiles have optimum growth rates at 15°C and cannot grow above 25°C. Psychrotrophs, such as Clostridium botulinum type E, have optima of ~35°C and cannot grow above 40°C. Because these food­

borne pathogens, and even some mesophilic S. aureus strains, can grow at <10°C, conventional refrigeration cannot ensure the safety of a food. Additional barriers to microbial growth should be incorporated into refriger­ ated foods containing no other inhibitors. Several metabolic capabilities are important for growth in the cold. Homeoviscous adaptation enables cells to maintain membrane fluidity at low temperatures. As temperature decreases, cells synthesize increasing amounts of mono- and diunsaturated fatty acids. The kinks caused by the double bonds prevent tight pack­ ing of the fatty acids into a more crystalline array. The membrane’s physical state can regulate the expression of genes, particularly those that respond to tempera­ ture. The accumulation of compatible solutes at low temperatures is analogous to their accumulation under conditions of low aw. Foods serve as a rich source of compatible solutes for foodborne bacteria. Compatible solutes in foods that can be used directly by bacteria include peptides, amino acids, ectoine, betaine, sugars, taurine, and carnitine. Complex proteins and phospho­ lipids can serve as precursors of solutes. The production of cold shock proteins (CSPs) con­ tributes to an organism’s ability to grow at low tem­ peratures. CSPs appear to function as RNA chaperones, minimizing the folding of mRNA, thereby facilitat­ ing the translation process. Streptococcus thermophilus CSPs are maximally expressed at 20°C. There is a 9-fold induction of csp mRNA. Pretreatment at 20°C increases survival approximately 1,000-fold compared to nonadapted cells. E. coli CSPs are categorized into two groups. Class I proteins are expressed at low levels at 37°C and increase dramatically after a shift to low temperature. Class II CSPs increase only fewfold after

Figure 1.4  Relative growth rates of bacteria at different temperatures. doi10.1128/9781555818463.ch1f4

1.  Physiology, Growth, and Inhibition of Microbes in Foods a downshift in temperature. Cold shocking L. monocytogenes from 37 to 5°C induces an array of 12 CSPs with molecular weights ranging from 48,000 to 14,000. Expression of the fri gene, encoding ferritin, protects L. monocytogenes against multiple stresses including cold and heat shock. Exposure of E. coli O157:H7 to cold stress decreases its acid tolerance. Temperature regulates the expression of virulence genes in several pathogens. The expression of 16 pro­ teins on seven operons on the Y. enterocolitica virulence plasmid is high at 37°C, weak at 22°C, and undetect­ able at 4°C. Similarly, the gene(s) required for virulence of Shigella spp. is expressed at 37°C but not at 30°C. The expression of genes required for L. monocytogenes virulence is also temperature regulated. Cells grown at 4, 25, and 37°C all synthesize internalin, a protein re­ quired for penetration of the host cell. Cells grown at 37°C, but not at 4 or 25°C, are hemolytic. However, the hemolytic activity is restored during the infection pro­ cess. Temperature influences expression of V. cholerae toxT and toxR genes, which are essential for cholera toxin production. Maximal expression occurs at 30°C, whereas at 37°C expression is decreased or abolished. In enterohemorrhagic E. coli, temperature modulates transcription of the esp genes; synthesis of Esp proteins is enhanced when bacteria are grown at 37°C. Esp pro­ teins are required for signal transduction events leading to the formation of the attaching and effacing lesions linked to virulence. The growth temperature can influence a cell’s thermal sensitivity. L. monocytogenes cells preheated at 48°C have increased thermal resistance. Holding listeria cells at 48°C for 2 hours in sausages increases their D value at 64°C 2.4-fold. This thermotolerance is maintained for 24 h at 4°C. Subjecting E. coli O157:H7 cells to sub­ lethal heating at 46°C increases their D value at 60°C 1.5-fold. Two proteins, putatively GroEL and DnaK, increase following heat shock. In short, the heat shock response and regulated synthesis of heat shock proteins (HSP) in gram-negative bacteria can differ markedly from those of gram-positive bacteria. Many HSP are molecular chaperones (e.g., DnaK and GroEL) or ATPdependent proteases (e.g., Lon and ClpAP) and function in protein folding, assembly, transport, and repair under stress and nonstress conditions. Shock proteins synthe­ sized in response to one stressor may provide crossprotection against other stressors. Exposing B. subtilis to mild heat stress enables the organism to survive not only otherwise lethal temperatures but also exposure to toxic concentrations of NaCl. Heat-adapted (50°C for 45 min) Listeria cells are more resistant to acid shock. Similarly, sublethal heat treatment of E. coli O157: H7 cells increases their tolerance to acidic conditions. Yeasts

17

require heat shock transcription factor (HSF) in order to grow above 35°C. An increase in protection against oxidative stress associated with HSF occurs following heat shock.

Hurdle Technology

Instead of setting one parameter to the extreme limit for growth, hurdle technology deoptimizes a variety of fac­ tors such that lower concentrations of each are needed to inhibit growth. Hurdle technology is most effective when it combines two stressors that act by different mechanisms. For example, a limiting aw of 0.85 or a limiting pH of 4.6 prevents the growth of foodborne pathogens. Hurdle technology might obtain similar in­ hibition at pH 5.2 and an aw of 0.92. Hurdle technology assaults multiple homeostatic processes. In acidic condi­ tions, cells use energy to pump out protons. In low-aw environments, cells use energy to accumulate compat­ ible solutes. Maintenance of membrane fluidity also re­ quires energy. When the energy needed for biosynthesis is diverted into maintenance of homeostasis, cell growth is inhibited. When homeostatic energy demands ex­ ceed the cell’s energy-producing capacity, the cell dies. Hurdle technology can encompass the use of antimicro­ bial agents and technology including the use of ozone and the application of irradiation in conjunction with shifts in pH and aw to inhibit microbial growth. Hurdle effects can be additive or synergistic. Claims of synergy should be supported by a quantitative anal­ ysis, such as the use of an isobologram (Fig. 1.5). To construct an isobologram, the MIC of compound A is

Figure 1.5  Determination of synergy through the use of isobo­lograms. A synergistic effect is demonstrated when the MICs of mixtures (s) of compound A and B fall below the line connecting the MIC of B and the MIC of A. doi:10.1128/9781555818463.ch1f5

18 plotted on the x axis. The MIC of compound B is plot­ ted on the y axis. A line is drawn to connect the points. Experiments determined the MICs of A and B in various concentrations, and these are plotted on the isobolo­ gram. If the points fall on the line, the effects are addi­ tive. If they fall below the line, the effects are synergistic. If they fall above the line, there are antagonistic effects. Novel approaches to hurdle technology include the use of bacteriocin with high pressure, pulsed electric fields, and other antimicrobials such as lysozyme and lactoferrin. Control of S. aureus in pasteurized milk can be achieved using phage-encoding endolysin and nisin. Combining the antimicrobials causes a 64-fold and 16-fold reduc­ tion in MICs of nisin and endolysin, respectively. The mechanism(s) by which the synergistic effect is achieved is not yet known but may be linked to better access to their respective cleavage and binding sites or associated with an enhanced peptidoglycan hydrolase activity.

CONCLUSION Microbial growth in foods is a complex process gov­ erned by genetic, biochemical, and environmental fac­ tors. Much of what we “know” about foodborne microbes must be held with the detached objectivity required of an unproven hypothesis. Many of the de­ velopments in molecular biology and microbial ecology are newly detailed in this edition of Food Microbiology: Fundamentals and Frontiers.

References 1. Annous, B., P. M. Fratamico, and J. L. Smith. 2009. Quorum sensing in bacteria: why bacteria behave the way they do. J. Food Sci. 74:R24–R37.

Factors of Special Significance 2. Chaveerach, P., A. A. H. M. ter Huurne, L. J. A. Lipman, and F. van Knapen. 2003. Survival and resuscitation of ten strains of Campylobacter jejuni and Campylobacter coli under acid conditions. Appl. Environ. Microbiol. 69:711–714. 3. Colwell, R. R., and D. J. Grimes (ed.). 2000. Nonculturable Microorganisms in the Environment. ASM Press, Washington, DC. 4. Divol, B., and A. Lonvaud-Funel. 2005. Evidence for vi­ able but nonculturable yeast in botrytis-affected wine. J. Appl. Microbiol. 99:85–93. 5. Dunny, G. M., and S. C. Winans (ed.). 1999. Cell-Cell Signaling in Bacteria. ASM Press, Washington, DC. 6. Medellin-Pena, M. J., and M. W. Griffiths. 2009. Effect of molecules secreted by Lactobacillus acidophilus strain La-5 on Escherichia coli O157:H7 colonization. Appl. Environ. Microbiol. 75:1165–1172. 7. Nilsson, L., J. D. Oliver, and S. Kjelleberg. 1991. Resuscitation of Vibrio vulnificus from viable but noncul­ turable state. J. Bacteriol. 173:5054–5059. 8. Oliver, J. D. 2005. The viable but nonculturable state in bacteria. J. Microbiol. 43(Spec. no.):93–100. 9. Oliver, J. D. 2010. Recent findings on the viable but non­ culturable state in pathogenic bacteria. FEMS Microbiol. Rev. 34:415–425. 10. Oliver, J. D., M. Dagher, and K. Linden. 2005. Induction of Escherichia coli and Salmonella Typhimurium into the viable but nonculturable state following chlorination of wastewater. J. Water Health 3:249–257. 11. Romeo, T. (ed.). 2008. Bacterial Biofilms. SpringerVerlag, New York, NY. 12. Simoes, M., L. C. Simoes, and M. J. Viera. 2010. A review of current and emergent biofilm control strategies. LWTFood Sci. Technol. 43:573–583. 13. Wesche, A. M., J. B. Gurtler, B. P. Marks, and E. T. Ryser. 2009. Stress, sublethal injury, resuscitation, and virulence of bacterial foodborne pathogens. J. Food Prot. 72:1121–1138. 14. Wu, V. C. H. 2008. A review of microbial injury and recovery methods in food. Food Microbiol. 25:735–744.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch2

Lu Zhang Jennifer Cleveland McEntire Rosetta Newsome Hua Wang

2

Antimicrobial Resistance

Antimicrobials are widely used in human medicine, agricultural production, and food processing and have been essential for ensuring human and animal health as well as the safety of our food supply. Unfortunately, the resistance of microbes (both commensals and pathogens) to commonly used antimicrobials is rising on a global scale. Particularly, the specter of antibiotic-resistant (ART) microbes is a worldwide concern due to the public perception of an oncoming postantibiotic era. Consumers are concerned about the increased difficulty in treating infections caused by ART bacteria, but few understand the real risk factors involved in the rapid emergence and dissemination of these ART bacteria. Difficult-to-treat hospital-acquired infections caused by ART pathogens and opportunistic pathogens, such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycinresistant enterococci, have captured public attention. In 2003, over one-quarter of nosocomial enterococcal infections were resistant to vancomycin, and almost 60% of nosocomially acquired S. aureus infections were MRSA. This represents 12% and 11% increases, respectively, in percent resistance compared to the previous 5-year averages (19). Antibiotic resistance (AR) also has economic implications: in the mid-1990s, an estimated $4 billion of the >$7 billion in human antibiotic sales targeted nosoco-

mial infections caused by ART bacteria (61). The Centers for Disease Control and Prevention and the medical community are promoting changes in antibiotic-prescribing practices and encouraging prudent use of antibiotics to reduce the selective pressure that increases the prevalence of ART organisms. These recommendations include renewed emphasis on preventing the spread of infection and targeting specific antibiotics for specific pathogens to limit the use of broad-spectrum antibiotics (18). However, emerging evidence suggests that the development, dissemination, and persistence of ART bacteria and their resistance-encoding genes are much more complicated than previously thought. The true impact of the food chain, particularly commensal bacteria and agriculture practices beyond antibiotic usage, on AR dissemination is just becoming recognized. The potential correlation of resistance to various antimicrobials has yet to be revealed. Accordingly, control strategies will likely need to be updated for targeted and effective mitigation (135).

ANTIMICROBIALS AND RESISTANCE The term “antimicrobial resistance,” in a broad sense, describes the decreased susceptibility of a multitude of microbes to a broad spectrum of single or multiple

Lu Zhang and Hua Wang, Department of Food Science and Technology, The Ohio State University, 110 Parker Food Science and Technology Building, 2015 Fyffe Court, Columbus, OH 43210. Jennifer Cleveland McEntire, Leavitt Partners, 299 South Main Street, Suite 2400, Salt Lake City, UT 84111. Rosetta Newsome, Institute of Food Technologists, Chicago, IL 60607.

19

SMP_Food Microbiology_CH02.indd

19

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

20 agents, with the temporary or permanent ability of an organism and its progeny to remain viable and/or multiply under conditions that would destroy or inhibit other members of the strain (24). The resistance to distinct agents such as antibiotics, food antimicrobials (e.g., acids and bacteriocin), and disinfectants will be discussed in this chapter. Various AR mechanisms, whether due to permanent genetic and structural alterations or to temporary adjustment of metabolic machinery such as stress responses, formation of biofilms and spores, etc., directly contribute to the persistence of the microbes in the environment or, for pathogens, to infection in their host. Therefore, such resistance determinants can also be considered broadly as virulence factors in pathogens (31). The parallels between the mechanisms of AR and mechanisms of organic acid and bacteriocin resistance are shown in Table 2.1.

Antibiotics and AR Types of Antibiotics

Depending on the mode of action, antibiotics can be categorized into bactericidal and bacteriostatic agents. Bactericidal antibiotics normally target key cell structures and activities, such as the biosynthesis of cell wall (b-lactam antibiotics) and DNA (fluoroquinolones), whereas bacteriostatic antibiotics commonly interfere with protein synthesis, inhibiting the proliferation of bacterial cells (67). The three types of commonly used bactericidal antibiotics include the members of the blactam antibiotics family, aminoglycosides, and quinolones, whereas macrolides, telithromycin, sulfonamides, and tetracycline are bacteriostatic antibiotics (103).

Bactericidal antibiotics b-Lactam antibiotics.  b-Lactam antibiotics are most commonly used in clinical therapy for infections. b-Lactam antibiotics refer to antibiotic agents containing a b-lactam in the molecular structure and sometimes to those that are b-lactamase inhibitors (although they are not true antibiotics). b-Lactam antibiotics are bactericidal, interfering with the biosynthesis of the peptidoglycan layer in bacterial cell walls by irreversibly blocking penicillin-binding proteins, including carboxypeptidases, endopeptidases, and transpeptidases. Penicillin-binding proteins are a group of proteins that facilitate crosslinking of newly synthesized peptidoglycan to the existing cell wall structure (41). Once treated with b-lactam antibiotics, susceptible bacterial cells develop a weak cell wall and are eventually subject to cell lysis. Four major classes of antibiotics and their derivatives belong to the b-lactam antibiotics category, including penicillin,

SMP_Food Microbiology_CH02.indd

20

cephalosporins, carbapenems, and monobactams. Based on their antimicrobial activities and antibacterial spectra, they can be further divided into different groups.

b-Lactamase inhibitors.  Apart from the above four families, b-lactamase inhibitors are also recognized as blactam antibiotics. One of the most common (b-lactam antibiotics) mechanisms of resistance in bacteria is carried out by genes encoding b-lactamase, which hydrolyzes intracellular b-lactam molecules. To be most effective, b-lactamase inhibitors are administered together with b-lactam antibiotics to minimize the effects of bacterial resistance. b-Lactamase inhibitors are not true antibiotics, but rather antibiotic enhancers (105). Aminoglycosides.  Aminoglycosides are another type of bactericidal antibiotic. Their key structure includes an aminocyclitol ring, with different glycosidic linkages and side chains differentiating the members of this family (68). Aminoglycosides affect bacterial cells by displacing cations such as Mg2+ and Ca2+ on the outer bacterial membrane, thereby disrupting membrane permeability. Moreover, aminoglycosides can also be bacteriostatic, as they impair growth of bacterial cells by binding to the 30S subunit of the bacterial ribosome, thereby inhibiting protein synthesis. Susceptible bacteria are primarily aerobic gram-negative bacteria, such as Klebsiella spp. and Pseudomonas aeruginosa, whereas most grampositive bacteria and anaerobes are not susceptible to this antibiotic class (98). Representative members of the aminoglycosides group include gentamicin, kanamycin, neomycin, and streptomycin. This group of antibiotics is primarily used for treating infections on the skin’s surface and in the respiratory system and is normally combined with other types of antibiotics for clinical therapy. Quinolones.  Quinolones are also bactericidal, as they inhibit DNA synthesis in bacterial cells by binding to DNA gyrase and DNA topoisomerase, both of which are involved in DNA replication, transcription, and repair systems (34, 138). Quinolones are effective against many types of gram-negative bacteria, including Klebsiella spp., Escherichia coli, Salmonella spp., and Shigella spp. (12). Ciprofloxacin, gemifloxacin, moxifloxacin, and trovafloxacin are examples of quinolone antibiotics. Bacteriostatic antibiotics Bacteriostatic antibiotics are different from bactericidal antibiotics in that they inhibit the growth of bacteria rather than killing them. Intracellular nucleotide or protein synthesis is most commonly targeted by this type of antibiotics, and a broad spectrum of bacterial species are

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

21

Table 2.1  Major classes of antibiotics and their applicationsa Antimicrobial classes (selected examples)

Mode of action/spectrum

Speciesb

Disease Disease Growth treatment prevention promotion

Aminoglycosides (gentamicin, neomycin, streptomycin)

Inhibits protein synthesis/broad spectrum

Beef cattle, goats, p­oultry, sheep, swine

X

X

Beta-lactams   Penicillins (amoxicillin, ampicillin)

Inhibits cell wall synthesis

Beef cattle, dairy cows, fowl, poultry, sheep, swine

X

X

Plants

Humans

X (certain plants)

X

X

  Narrow-spectrum c­ephalosporins (cefadroxil)

X

X

  Expanded-spectrum c­ephalosporins (cefuroxime)

X

  Broad-spectrum c­ephalosporins (ceftiofur)

Beef cattle, dairy cows, poultry, sheep, swine

X

X

Chloramphenicol

Inhibits protein synthesis/broad spectrum

X

Florfenicol

Inhibits protein synthesis/broad

Cycloserines (cycloserine)

Inhibits cell wall synthesis/narrow

X

Glycopeptides (vancomycin)

Inhibits cell wall synthesis/narrow

X

Beef cattle

X

Ionophores (monensin, Disrupts osmotic salinomycin, semduramicin, balance/narrow lasalocid)

Beef cattle, fowl, goats, poultry, r­abbits, sheep

X

X

Lincosamides (lincomycin)

Inhibits protein synthesis/narrow spectrum

Poultry, swine

X

X

Macrolides (tylosin, t­ilmicosin, erythromycin)

Inhibits protein synthesis/narrow

Beef cattle, poultry, swine

X

X

Monobactrams (aztreonam)

Inhibits cell wall synthesis/broad

Polypeptides (bacitracin)

Inhibits cell wall synthesis/narrow

X

X

X X

Fowl, poultry, swine

X

Fluoroquinolones Inhibits DNA (e­nrofloxacin, danofloxacin) synthesis/broad

Beef cattle

X

Streptogramins (virginiamycin)

Beef cattle, poultry, swine

X

Sulfonamides (sulfadimethoxine, sulfamethazine, sulfisoxazole)

Inhibits folic acid synthesis/broad spectrum

Beef cattle, dairy cows, fowl, poultry, swine, catfish, trout, salmon

X

Tetracyclines (chlortetracycline, oxytetracycline, tetracycline)

Inhibits protein synthesis/broad spectrum

Beef cattle, dairy cows, fowl, honey bees, poultry, sheep, swine, catfish, trout, salmon, lobster

X

X

X

X X

X

X

X

X

X

X

X

X (certain plants)

X

(Continued)

SMP_Food Microbiology_CH02.indd

21

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

22 Table 2.1  Major classes of antibiotics and their applicationsa (Continued) Antimicrobial classes (selected examples)

Mode of action/spectrum

Disease Disease Growth treatment prevention promotion

Speciesb

Plants

Humans

Bambermycin

Inhibits cell wall synthesis/narrow

Beef cattle, poultry, swine

X

X

X

Carbadox

Inhibits DNA synthesis/narrow

Swine

X

X

X

Novobiocin

Inhibits DNA gyrase/narrow

Fowl, poultry

Spectinomycin

Inhibits protein synthesis/narrow

Poultry, swine

X

X

X

X

X

Adapted and modified from reference 133a. Poultry includes at least one of the following birds: broiler chickens, laying hens, and turkeys. Fowl includes at least one of the following birds: ducks, pheasants, and quail. a

b

affected. For example, sulfonamides are competitive inhibitors of dihydropteroate synthase (82), an enzyme that plays a key role in folic acid synthesis. The lack of folic acid will eventually lead to the breakdown of nucleotide synthesis and interruption of cell proliferation. Four groups of sulfonamides have been defined based on their modes of absorption/excretion. These include short- to mediumacting sulfonamides, long-acting sulfonamides, gastrointestinal (GI) tract sulfonamides, and topical sulfonamides. Commonly used sulfonamides include sulfamethoxazole, sulfadoxine, sulfasalazine, and silver sulfadiazine. Tetracyclines are different from sulfonamides in that they inhibit bacterial growth by impairing protein synthesis. This type of antibiotic blocks aminoacyl tRNA by binding to the 30S subunit of bacterial ribosomes (1). Tetracyclines can inhibit a broad spectrum of bacteria, including most gram-positive bacteria and gramnegative bacteria, aerobes, and anaerobes. Tetracyclines are divided into three groups depending on their pharmacokinetic characteristics. These include short-acting tetracyclines (oxytetracycline and tetracycline), intermediate-acting tetracyclines (demeclocycline), and longacting tetracyclines (doxycycline and minocycline). The macrolides/ketolides are another type of bacteriostatic antibiotics and are capable of binding to subunits of the 50S ribosome, thereby inhibiting transpeptidation and translocation during protein synthesis. Macrolides have a relatively broad antibacterial spectrum, especially against gram-positive bacteria. Members of the macrolide family include erythromycin, clarithromycin, dirithromycin, and telithromycin, and they are often used to treat infections of the respiratory tract.

Mechanisms of AR in Bacteria

Bacteria can become resistant to specific types/groups of antibiotics in many different ways, which can be

SMP_Food Microbiology_CH02.indd

22

summarized as (i) innate resistance; (ii) reducing the intracellular concentration of corresponding antibiotics; (iii) inactivatin­g corresponding antibiotics; (iv) mutations at binding sites of corresponding antibiotics; and (v) formation of additional metabolic pathways that compensate for the loss of key enzymes or substrates (caused by corresponding antibiotics) (Table 2.2) (64). Bacterial cells with innate resistance to a particular antibiotic naturally lack the molecule or pathway targeted by the drug; therefore, the drug has no effect on the survival of the microbes. Bacteria can also become resistant to antibiotics by reducing the intracellular concentration of the antibiotic, which can be achieved with a reduction of membrane permeability (decreased import) and adoption of efflux pumps (increased export). Gram-negative bacteria generally have better control over importing molecules, because their outer membranes have a smaller size exclusion limit than the peptidoglycan layer surrounding gram-positive bacteria. Hydrophilic solutes diffuse into gram-negative bacteria primarily through porins on the outer membrane. Bacterial mutants that lack porins may present a broad spectrum of resistance. Decreased permeability to chloramphenicol is regarded as a cause of resistance in Haemophilus influenzae and Pseudomonas cepacia (14, 15). In gram-positive Staphylococcus epidermidis, glycopeptide resistance may result from overproduction of glycopeptide binding sites within the cell wall peptidoglycan (118). Bacillus subtilis mutants resistant to the protonophore carbonyl cyanide m‑chlorophenylhydrazone (CCCP) have altered membranes with reduced amounts of C16 fatty acids and increased ratios of iso:anteiso branches. The CCCPresistant mutants are cross-resistant to other inhibitors, including 2,4‑dinitrophenol, tributyltin, and neomycin (49).

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

23

Table 2.2  Examples of antimicrobial resistance mechanisms in bacteria Mechanism

Action

Export

Specific Nonspecific

Destruction

Antibiotics

Organic acids

Tetracycline Phenicols Organic solvent tolerance

Bacteriocins

F0F1ATPase pumps out p­rotons, anions a­ccumulate intracellulary

Not applicable, bacteriocins not in cytoplasm

Specific or general b-Lactamases and cephalosporinases

Not applicable

Protease, specific “bacteriocinase”

Modification

Specific

Acetylation, adenylation, methylation, or phosphorylation of aminoglycosides Acetylation of phenicols

Not applicable

Dehydroreductases can inactivate lantibiotics like nisin

Altered receptors

Specific

Penicillin binding proteins Ribosome DNA gyrase RNA polymerase

No receptor required

Probable, but not reported to date

Membrane composition

General

Altered membranes in resistant E. coli and bacilli

May affect permeability

Demonstrated for nisin resistance

Energy-dependent efflux pumps are another mechanism of reducing intracellular antibiotics in ART bacteria. The efflux pumps can be specific to single antibiotics, or they can export multiple antibiotics. For example, both multidrug-resistant pumps and tetracycline-specific pumps were discovered in tetracycline-resistant bacteria. The coordination of reduced membrane permeability and the presence of efflux pumps likely can contribute to high-level resistance to antibiotics. For example, it was determined that active efflux pumps, as well as a reduction in the amount of major porins, contributed to increased resistance to chloramphenicol in some Enterobacter strains (45). Antibiotic inactivation can be achieved by degradation or modification. An example of antibiotic inactivation is by b-lactamase, which is carried by many bacterial species. b-Lactamase is capable of hydrolyzing the b-lactam ring in b-lactam antibiotics, which inactivates their antimicrobial activity. Currently four types (classes A, B, C, and D) of b-lactamases are recognized, and each type possesses its own characteristic nucleotide/amino acid sequence. Some lactamases confer resistance to various b-lactam antibiotics, whereas others are responsible for resistance to a single type of b-lactam antibiotic. It was determined that many types of b-lactamases are also resistant to a variety of b-lactamase inhibitors. In some situations, antibiotics are not degraded but modified by resistant bacteria. For example, aminoglycosides can be modified by bacterial phosphotransferase and acetyltransferase, thereby losing their capability to bind the ribosome subunit (120). Chloramphenicol is inactivated by acetylation, which is catalyzed by the chloramphenicol acetyltransferase of resistant bacteria (115).

SMP_Food Microbiology_CH02.indd

23

Resistant pathogens can modify aminoglycosides by a variety of enzymes, including methylases, acetyltransferases, nucleotidyltransferases, and phosphotransferases (120). Another strategy for gaining resistance is to modify (mutate) antibiotic recognition sites on the intracellular structure/enzyme of bacterial cells, so that the corresponding antibiotics cannot recognize them. This mechanism is involved in the development of resistance to many types of antibiotics, such as quinolones (mutations on target sites in DNA gyrase), tetracycline, and macrolides (mutations on corresponding ribosome subunits) (115). The formation of alternative metabolic pathways is responsible for resistance to some antibiotics that target critical metabolic activities of bacterial cells. For example, sulfonamides are effective against bacteria by competing with para-aminobenzoic acid, a substrate involved in folic acid synthesis, which is critical for nucleotide production. Bacteria can inhibit sulfonamide activity by producing additional para-aminobenzoic acid, which has been observed in many sulfonamideresistant bacteria. In summary, various mechanisms have been adopted by bacteria to provide resistance to different types of antibiotics. Depending on the mechanism, resistance to certain antibiotics can be of low, medium, or high level. Bacteria can be resistant to single antibiotics or become multidrug resistant via one or multiple AR mechanisms. Recently, Liu and Pop (81) created the Antibiotic Resistance Genes Database by compiling publicly available information. The database includes annotation of each AR gene and resistance type with information on

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

24 AR profile, mechanism of action, ontology, and external links to sequence and protein databases. The database also supports sequence similarity searches and has a tool for characterizing common mutations conferring antibiotic resistance. The site provides useful information for AR prevalence studies and for identification and characterization of resistance mechanisms.

Development of AR in Bacteria

As mentioned, resistance to antimicrobials can be due to genetic and structural changes or temporary adjustment of metabolic machinery. AR due to genetic changes can be transmitted to progeny by vertical transmission and to other susceptible bacteria by horizontal gene transmission (HGT) mechanisms. Resistance due to a particular mode of living or stress response is transient and is not transferable.

Vertical gene transmission Both mutated genes (resulting from mutations in genes encoding the antibiotic recognition site, alternative metabolic pathways, membrane porin/efflux pump, or antibiotic degradation systems, etc.) and acquired AR genes can be transferred by vertical transmission to offspring during bacterial growth. Mutations may be caused by errors in DNA synthesis, chemical changes induced by mutagens, or incorrect repair of damage-induced singlestrand breaks. Spontaneous mutations occur at varied rates in different organisms regardless of the presence of antibiotics. The reported frequency of mutation ranges from 10–6 to 10–10 per bacterial generation (30). Normally, without the selective pressure from antibiotic usage, the mutants would account for only a very small portion of the total bacteria population and generally would not be detected. Not only does antibiotic usage select for ART population, but emerging data also suggest that constant exposure to inhibitory levels of antibiotics may significantly increase the mutation rate in bacteria and lead to multidrug-resistant offspring (66). It was determined that sublethal levels of antibiotic treatment can lead to an increase of reactive oxygen species, resulting in heterogeneous increases in the MIC of a variety of antibiotics, regardless of the type of bactericidal antibiotics used. This free-radical-based mechanism illustrates an additional effect that antibiotic usage can have on the evolution of multidrug-resistant bacteria (66). Horizontal gene transmission HGT mediates the spread of resistance determinants (AR genes) to susceptible bacterial cells, most likely by mobile gene elements, such as plasmids, transposons, and integrons (131). The transfer can occur via conjugation,

SMP_Food Microbiology_CH02.indd

24

transformation (uptake of free DNA), and transduction. Unlike the time-consuming mutation accumulation process, resistant units can be disseminated rapidly among microorganisms, within or across species and genera, via HGT events; therefore, HGT mechanisms play a key role in the rapid emergence of AR in microbial ecosystems. It is worth noting that the immunity gene(s) in antibioticproducing microbes that protects the organisms from the antimicrobial compound (if applicable) can potentially serve as AR determinant(s) if acquired by others.

Mobile genetic elements in HGT.  Three important mobile gene elements (plasmid, transposon, and integron) are involved in HGT more commonly than others. A plasmid is a type of circular DNA that self-replicates independently of chromosomal DNA. Plasmids may carry beneficial genetic traits, such as additional genes for metabolic activities, and can be readily transferred to another host bacterial cell via conjugation (80). Various AR gene-encoding, evolutionarily related plasmids have been isolated from clinical, environmental, food, and human microfloras, indicating that plasmids are an important shuttle for disseminating AR in nature. A transposon is a type of mobile DNA element that can “cut/copy and paste” itself using a transposase enzyme. Once separated from its original position, a transposon can integrate into a specific or random site on genomic or plasmid DNA (91). Transposons often contain an AR gene, thus contributing to AR dissemination. An integron is a mobile DNA element with a site-specific DNA recombination system that is capable of recognizing and capturing mobile gene cassettes (51). Depending on the content of the gene cassettes, integrons can be divided into AR integrons, which carry mainly AR genes and can reside on plasmid and genome DNA, and superintegrons (large genomic DNA elements containing an array of genes that encode a variety of adaptive functions, such as virulence and metabolic activities). Resistance integrons are an important contributor in the dissemination of AR throughout the food chain. More than 70 AR genes have been detected in AR integrons, and a single integron may contain multiple AR genes. Compared with integrons, insertion sequences are small and compact DNA elements that primarily code for mobility. Once an insertion sequence contains an AR gene, it becomes a mobile vessel capable of disseminating the resistance genes within and between bacterial populations. HGT mechanisms.  Common HGT mechanisms include conjugation, transformation, and transduction. Conjugation is the transfer of genetic material from donors to recipients through direct contact between the cells.

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

25

Susceptible recipient cells may become resistant to corresponding antibiotics by acquiring an AR-encoding genetic element (plasmid or transposon) from the donor cell. The frequency of conjugal transfer of functional genetic elements is affected by both the genetic compatibility of the donor-and-recipient pair and the cell concentration. Apart from conjugation, it is also possible to transform susceptible bacterial cells with cell-free DNA segments containing AR genes, leading to acquired resistance in these cells to the corresponding antibiotics. A bacterial cell able to absorb naked DNA and integrate into the genome is called a competent cell, and the process is called transformation. The frequency of transformation is commonly low. However, given the high concentration of free DNA in certain natural and host environments, the intake of DNA by bacteria via transformation may be a critical route in modifying the microbial ecosystem in such cases. Transduction is a form of HGT mediated by bacterial phage, during which DNA from the donor cell can be packaged into the phage by mistake and then transmitted to the recipient cells in the subsequent round of the infection and eventually integrated into the host (recipient) genome or plasmid. Besides facilitating AR dissemination, transduction also plays an important role in bacterial evolution. For example, DNA sequence data suggest that a bacterial phage likely was involved in the evolution of Escherichia coli O157:H7, during which Shiga toxin-encoding genes were initially transmitted from another toxin-producing bacterium to an E. coli strain via bacterial phage infection (39). Due to the extensive involvement of transduction in microbial evolution, including pathogens, it is important to reevaluate the long-term safety impacts of phage-based intervention strategies intended to control foodborne pathogens.

Impact of the Food Chain on AR Antibiotic Usage in Food Animal and Agriculture Production

The usage of antibiotics extends beyond human medicinal purposes. Antibiotics are used in food and aquaculture production to treat infected animals, prevent infection of animals potentially exposed to pathogens, and promote growth. To a lesser extent, antibiotics are also used in plant agriculture production to prevent diseases and improve productivity. Estimates of the combined total use of antibiotics, including human and animal use, vary widely. According to a recent report from the FDA, the total domestic sales and distribution amount of antimicrobial drugs approved for use in food animal production in 2009 was close to 29 million

SMP_Food Microbiology_CH02.indd

25

pounds (43). Thus, on a mass basis, agricultural practices, including therapeutic treatment, contribute to a significant portion of the antibiotics introduced into the biosphere. The growth-promoting function of antibiotics is only partly understood and is the subject of heated debate (44, 122). It is generally believed that the increasing prevalence of AR is directly related to increased antibiotic use, both prudent and frivolous, in clinical and agricultural settings. Many argue that the widespread use of antibiotics at subtherapeutic levels exacerbates the selective pressure that enriches microbial populations for ART bacteria and suggest that growth promotion and disease prevention could be achieved through nonantibiotic alternatives and novel management practices (113). As a result of the debate, many European countries have discontinued the administration of antibiotics for growth promotion. Denmark essentially banned the use of antibiotics for growth promotion in food animal production in an effort to decrease the prevalence of a­ntibiotic-resistant human pathogens. While the total use of antibiotics in animals in Denmark decreased by 30% from 1997 (before the ban) to 2004, there has been a 68% increase in the use of antibiotics for therapeutic purposes during the same period (140). This makes it difficult to assess the impact of this mitigation strategy. A 2002 meeting of the World Health Organization assessing the impact of the Danish ban on growthpromoting antibiotics revealed that ART enterococci decreased. For example, resistance of Enterococcus faecium to virginiamycin declined from approximately 60% to 10% on broiler meat from 1997 (the peak of virginiamycin use) to 1999 (when there was no virginiamycin use). However, the incidence of Salmonella, Campylobacter, and Yersinia infections in humans was not affected (140). In the United States, organic farming is increasing in response to consumer demand. A small 3-state study compared the rates of pathogen isolation in conventional swine production (indoor rearing) versus antimicrobialfree farms (outdoor rearing) (46). The prevalence of both Salmonella and Toxoplasma was statistically higher in the antimicrobial-free production systems, although it is not clear how long the antimicrobial-free farms were in operation or if they had always been antimicrobial-free. In 2007, Cox et al. proposed a model that estimated the increased load of Campylobacter on chickens when animals were reared without antibiotics (26). Hence, the risks and benefits of antibiotic use in animal agriculture need to be carefully weighed. Agricultural antibiotic usage extends beyond animal use, presumably to a much lesser extent. In the United States, fruit plants may be sprayed with oxytetracyline

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

26 and streptomycin. Since there are no surveillance systems for ART bacteria on crops, it is difficult to estimate the extent of ART bacteria on plant products. Although evidence is lacking for the direct transfer of AR determinants between plant pathogens and human pathogens under natural conditions (94), the impact of plant antibiotic usage on the environmental commensal flora and, consequently, pathogens is anticipated. There is very limited use of antibiotics in U.S. aquaculture. Only sulfadimethoxine, ormetoprim, sulfamerizine, and oxytetracycline are approved for use, although sulfamerizine is no longer commercially available. The extent of antibiotic use in other countries that export seafood and produce is unknown and presumed to be much higher. The presence of antibiotic residues is a common reason for seizure of imported seafood. As indicated by Cabello (16), the use of prophylactic antibiotics in aquaculture compensates for unsanitary or otherwise poor fish production practices. Antibiotics administered through fish feed can be deposited in the aquaculture animals and the environment and may select for AR in environmental flora. These microbes, in turn, can be further involved in AR gene dissemination. Cabello’s review suggested that the AR determinants in Salmonella DT104 originated from fish pathogens and resulted from HGT. Heuer et al. (55) also described the human health consequences of antimicrobial use in aquaculture, noting the likelihood of transmission of drug-resistant pathogenic bacteria to humans from the aquaculture reservoir and the evidence for exchange of AR genes between fish pathogens and human bacteria. Briefly addressing risk management options, the authors mentioned that the use of effective vaccine strategies by the Norwegian salmon industry reduced by 99% antimicrobial use in aquaculture between 1987 and 2007, despite an increase in production.

Foodborne ART Bacteria

In the past decade, a variety of ART foodborne pathogens have been isolated and characterized, indicating that the food chain can serve as an avenue for transmitting ART bacteria to humans. Although it is important to realize the significance of the contribution of foodborne pathogens to foodborne illness, pathogens account for only a small percentage of the microbes associated with foods, with the number of foodborne ART pathogens being even smaller. Hence, one could argue that foodborne ART pathogens may not represent a major source for transmitting ART bacteria to humans (135, 137). However, considering the prevalence of ART commensal bacteria and the large size of the AR gene pool, it is likely that human beings are constantly exposed to ART

SMP_Food Microbiology_CH02.indd

26

commensal bacteria through the food chain (137). The impact of ART commensal bacteria on food safety and public health is not well recognized or characterized, and more research is needed to fill this knowledge gap so that strategic approaches can be developed for mitigating the transmission of ART bacteria through the food chain.

ART foodborne pathogens Extensive research during the past few decades has led to in-depth understanding of microbe-specific mechanisms in AR emergence, dissemination, and persistence for a small number of zoonotic pathogens. However, an enhanced understanding of differences in AR prevalence among pathogen species and serovars within the same genus is still needed. Such information may enable a better assessment of the relative contribution of different animal hosts and environmental factors, including selective pressure, on AR development and persistence for a given pathogen population. Furthermore, a comprehensive understanding of the relative contribution of pathogens in AR development in complex microbial ecosystems along the food chain, from the natural environment to acquisition by animal hosts, is needed for conducting a proper risk assessment and developing a well-targeted mitigation strategy. The following sections briefly discuss several important foodborne pathogens as well as nonpathogenic E. coli strains and their resistance to antibiotics. Salmonella.  Considering the large number of Sal­ monella serotypes (more than 2,500) and the differences in AR among different Salmonella serotypes, it is difficult to make generalizations regarding AR in Salmonella. This is compounded by differences that occur among Salmonella serotypes in the incidence of the human illnesses they cause. For example, the incidence of foodborne illness caused by Salmonella enterica serovar Typhimurium decreased by 40% between 1996–1998 and 2004, whereas Salmonella serovar Newport-related foodborne illnesses increased by 40% during the same period (20). Changes in serotype prevalence should be considered when examining trends in AR, so that the potential impact of AR on human health can be evaluated in the context of pathogen prevalence. Despite the variation that occurs in AR, pathogenicity, and incidence of human illness associated with different serotypes of Salmonella, there are some general relationships that occur among Salmonella species and AR. AR rates vary depending on the product (e.g., chicken versus turkey or layers versus broilers), country, year, and the type of antimicrobial. Gyles (50) proposed that

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

27

differences in AR profiles observed among Salmonella serotypes in different countries or regions could be the result of dissemination of dominant Salmonella clones specific to these locations. However, in general, resistance to tetracycline and streptomycin in Salmonella is high, whereas ciprofloxacin resistance is usually low, if it exists at all. Extended-spectrum cephalosporin resistance in Salmonella is of particular concern because this class of antibiotics is commonly used to treat salmonellosis. Extended-spectrum cephalosporin resistance is mediated by genes encoding b-lactamases. One gene, blaCMY-2, has been found in plasmids carried by Salmonella isolates worldwide. b-Lactamase-mediated resistance in Salmonella is encoded mainly by transposons and transmissible plasmids (17). Over the past several years, Salmonella Newport MDR-AmpC, which is resistant to at least four additional antibiotics, has received much attention due to its epidemic spread. The blaCMY-2 gene was identified in resistant isolates examined and was successfully transferred from a representative Salmonella isolate to E. coli by conjugation (110, 142).

Campylobacter.  Approximately one million cases of campylobacteriosis are estimated to occur annually in the United States (95, 119). While sporadic cases are typically associated with poultry, outbreaks are most often associated with consumption of raw milk, unchlorinated water, and produce (4). Occasionally, a Campylobacter infection must be treated with antibiotics, of which erythromycin or a fluoroquinolone is most commonly prescribed, followed to a lesser extent by tetracycline and gentamicin. Fluoroquinolones are also used in animal agriculture in different countries. In the United States, fluoroquinolones had been used to control disease in poultry. However, the FDA withdrew in 2005 the approval of the fluoroquinolone enrofloxacin, marketed as Baytril, for use in poultry because the agency deemed that fluoroquinolone use in poultry increased the prevalence of fluoroquinolone-resistant Campylobacter, which potentially compromised the treatment of human campylobacteriosis (42). Resistance to fluoroquinolones in Campylobacter is due to a single chromosomal mutation in gyrA (83), unlike most other pathogens, which require multiple mutations for clinically relevant resistance. There appears to be no fitness cost associated with fluoroquinolone resistance in Campylobacter, and in fact, the mutation in GyrA confers a fitness advantage, increasing colonization of the pathogen in chickens (85). Resistance to fluoroquinolones can be detected shortly after exposure to the antibiotic. This is in contrast to resistance developed to macrolides,

SMP_Food Microbiology_CH02.indd

27

which requires repeated exposure to the antibiotic for resistance to occur (83). The macrolide erythromycin is also used to treat human campylobacteriosis. In a study using chickens colonized with macrolide-sensitive Campylobacter species C. coli and C. jejuni, the effect of erythromycin dose on developing resistance was examined (70). One group of chickens was given a subtherapeutic dose of tylosin (a macrolide), and the other was given a therapeutic dose. Approximately 63% of the Campylobacter isolates were resistant to erythromycin after the subtherapeutic dose was administered, whereas only 11.4% were resistant among the isolates from the group treated with the therapeutic dose (70). This finding was consistent with the results of a laboratory study in which longterm use of subtherapeutic doses of tylosin, but not therapeutic treatments, selected for macrolide-resistant Campylobacter (78). It appears that C. coli has higher frequencies of macrolide resistance than C. jejuni; however, the reasons for this difference are unknown (83). Macrolide resistance in Campylobacter most commonly results from point mutations in 23S rRNA, which result in modifications to the antibiotic. Active efflux mediated by CmeABC is another mechanism that works synergistically with antibiotic modification in conferring macrolide resistance (83). Compared to other antibiotics, tetracycline resistance in Campylobacter is relatively high and is mediated by the tet(O) gene. The tet(O) gene in Campylobacter is most often carried on a plasmid but occasionally is chromosomally encoded. The protein product of this gene binds to the ribosome, protecting it from the inhibitory action of tetracycline (83).

Shigella.  When considering antibiotic-resistant Shi­ gella, it is important to understand that humans, not food animals, are the host for this pathogen. Hence, human cases of shigellosis are not due to carriage of the pathogen in food-producing animals or to contamination of crops with manure from food animals. Human fecal contamination of food, either on the farm or during food preparation, is the root source of shigellosis. The infection is classified as a foodborne illness because food is often the vehicle (not the source) for transmitting Shigella. The National Antimicrobial Resistance Monitoring System (NARMS) has maintained a database of antimicrobial-resistant enteric bacterial isolates from humans and food, animals, and fowl, including human Shigella isolates, since 1996. In 2010, most (81.8% of 407 isolates) of the Shigella species reported were Shigella ­sonnei (21), and most (91.2%) of the NARMS Shigella isolates were resistant to streptomycin; 40.8% were

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

28 resistant to ampicillin; 48.2% were resistant to tri­ methoprim-sulfamethoxazole; 30.2% were resistant to sulfamethoxazole/sulfisoxazole; 31.7% were resistant to tetracycline; 10.1% were resistant to chloramphenicol; 4.4% were resistant to nalidixic acid; 1.7% were resistant to ciprofloxacin; less than 1% were resistant to gentamicin, ceftiofur, ceftriaxone, and ceftriaxone; and 0% were resistant to amoxicillin-clavulanic acid and kanamycin. Emergence of cefotaximase-producing Shigella isolates in the United States has become a concern because cefotaximase-type b-lactamases are among the extended-spectrum b-lactamases that confer resistance to extended-spectrum cephalosporins and significantly compromise shigellosis treatment options (21). A study by Wong et al. (139) on antimicrobial resistance trends among Shigella isolates from patients with shigellosis in New York City between June 2006 and February 2009 revealed that 70% of 477 patients received an antibiotic for shigellosis, and 11% of these were treated with an antibiotic to which their Shigella infection was resistant. High levels of resistance among the isolates were found against amoxicillin-clavulanate, ampicillin, and trimethoprim-sulfamethoxazole, and resistance at a lower level to ciprofloxacin was also reported.

Listeria monocytogenes.  Infection with Listeria monocytogenes is typically treated with a combination of ampicillin and aminoglycoside. Rates of AR in this pathogen have not been studied to the same extent as in other pathogens such as Salmonella and Campylobacter. Because L. monocytogenes is considered widely distri­ buted in the environment, it is not strictly a zoonotic organism and has not been considered problematic in terms of AR. Hence, it is not included in any of the NARMS surveys. Recently, Prazak et al. (109) reported that 95% of the Listeria isolates from Texas cabbage and environmental or water samples were resistant to two or more antibiotics, with 85% of those resistant to penicillin. More recent studies (22, 40, 97, 106, 141) have revealed that L. monocytogenes isolates from raw meat and a variety of retail foods in several countries outside the United States are ART. A study of food products in China reported multidrug resistance (i.e., resistance to ³2 antibiotics) in 63.9% of 72 Listeria isolates and resistance to only one antibiotic in 19% of isolates (22). A study by Pesavento et al. (106) revealed that eight Listeria isolates were resistant to ampicillin and nine were resistant to methicillin. This finding is of note because ampicillin is used for treatment of listeriosis and Listeria can potentially transfer the methicillin resistance gene to Enterococcus spp., thereby compromising the treatment for Enterococcus infections. Yücel et al. (141)

SMP_Food Microbiology_CH02.indd

28

also reported Listeria resistance to ampicillin (five isolates), nalidixic acid (nine isolates), cephalothin (nine isolates), trimethoprim-sulfamethoxazole (six isolates), and kanamycin (one isolate) among the isolates from meat products in Turkey. Lyon et al. (86) examined the antimicrobial resistance profiles of 14 L. monocytogenes actA types (n = 157) isolated from a poultry processing plant and found that the prevalence of antimicrobial resistance among all actA types and the diversity within resistotypes were both low. All isolates were sensitive to the antibiotics used to treat listeriosis; and except for tetracycline, resistance was not related to type or lineage. In the same study (86), it was also observed that persistence of L. monocytogenes in a poultry further processing plant was not related to antimicrobial resistance and that there was little difference in antimicrobial resistance variability between clinical and environmental isolates.

Escherichia coli.  Pathogenic strains of E. coli may be among the more publicly recognizable foodborne pathogens, although only certain pathogens such as enterohemorrhagic E. coli (e.g., E. coli O157:H7) cause illness. Human infections caused by enterohemorrhagic E. coli are not typically treated with antibiotics. This makes AR in these strains practically less relevant from a public health standpoint. However, when animal and human clinical isolates of E. coli O157:H7 have been tested for antibiotic susceptibility, resistance against tetracycline, streptomycin, and sulfamethoxazole is most commonly found. While innate stress resistance is outside the scope of this chapter, this is a noteworthy characteristic of E. coli O157:H7. A major outbreak of E. coli O157:H7 was associated with unpasteurized apple juice, an acidic product for which conventional knowledge regarding E. coli deemed that this pathogen would not survive. However, the enhanced ability of E. coli O157:H7 to survive acidic conditions by acid stress response mechanisms is an innate quality and has been confirmed in multiple studies (58). The vast majority of E. coli strains are nonpathogenic and are part of the natural microbiota of many animals, including humans. E. coli organisms make up about 1% of the colonic flora (33, 89). A European Food Safety Authority (EFSA) Community Summary Report (2010) on antimicrobial resistance in zoonotic and indicator bacteria from animals and food in the European Union from 2004 to 2007 indicated that resistance to antimicrobials was common among indicator (commensal) E. coli organisms, with resistance being greater among isolates from fowl and pigs than cattle, and varying among

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

29

fowl and pig isolates from the 20 member states providing information. In North America, Alexander et al. (2) investigated the effect of subtherapeutic use of chlortetracycline plus sulfamethazine, chlortetracycline, virginiamycin, monensin, and tylosin on the development of AR in E. coli in fecal samples from feedlot steers (n = 300) in Canada. They found that cattle not receiving antimicrobials prior to arrival at feedlots were carrying E. coli strains that were resistant to tetracycline and ampicillin and that clonal dissemination of ampicillin-­resistant E. coli appeared to readily occur between animals within the same pen. They also found that administering subtherapeutic levels of chlortetracycline in combination with sulfamethazine increased the prevalence and cell numbers of tetracycline- and ampicillin-resistant E. coli, and tetracycline-resistant E. coli populations were higher in animals fed a grain-based diet than in those fed a silage-based diet. The investigators concluded that the GI and environmental factors influencing the development and dissemination of antimicrobialresistant E. coli in feedlot cattle are complex and that other factors in addition to the use of antimicrobials need to be considered when assessing the acquisition of AR by bacteria, such as relationships between environmental stressors and AR genes. Morley et al. (100) compared differences in AR between non-type-specific E. coli (NTSEC) and Salmonella enterica recovered from pen floors of feedlot cattle in Colorado raised without exposure to antimicrobial drugs (40 pens and 4,557 animals) and those reared using conventional practices (44 pens and 4,913 animals). They determined that overall the prevalence of resistance among all NTSEC isolates was very low (<1% for 9 of 15 drugs, 1 to 5% for more than 2 drugs, 6 to 15% for 2 drugs, and >15% for only 1 drug, including those antibiotics of interest in zoonotic transmission to humans, e.g., potentiated penicillins, cephalosporins, quinolones, and aminoglycosides) and that conventional feedlot production methods did predictably or uniformly increase the prevalence of anti­ microbial resistance among fecal NTSEC compared with rearing methods that restrict exposure to antimicrobial drugs. The investigators suggested that it is possible that the impacts of different antimicrobial drug exposures were overwhelmed by environmental exposures. Marshall et al. (89) concluded that although there is strong evidence for direct transfer of antibiotic resistance traits from animal- to human-associated microorganisms, there is still little direct evidence of transfer of resistance genes from animal-associated E. coli, Bacteroides, or Enterococcus spp. into comparable human microflora members, which subsequently resulted in antibioticresistan­t infections.

SMP_Food Microbiology_CH02.indd

29

The EFSA Community Summary Report indicated that E. coli is thought to be a reservoir of resistance genes that may be transferred to pathogens and that resistance in E. coli over time may serve as an “early warning system” for resistance in potentially pathogenic bacteria (37). However, recent studies have revealed that although nonpathogenic E. coli strains are commensals of the GI tract of animals and humans, they are not the major and earliest carriers of AR genes in various ecosystems from foods to the human gut microflora (137, 141a). Hence, proper selection of an AR indicator should be based on a comprehensive understanding of the AR ecology, which is described below.

Staphylococcus aureus.  Toxin production by large cell numbers of Staphylococcus aureus causes foodborne illness. Ingestion of the bacterium itself does not. There has been much publicity in the medical community and among the general public regarding MRSA. MRSA usually causes nosocomial infection rather than foodborne illness, although recent reports suggest potential zoonotic (or foodborne) transmission due to the finding of common genotypes in both animal and human MRSA isolates (79). However, since there are no reports of increased toxin production by MRSA, while this pathogen is of great clinical significance, its AR has no effect on staphylococcal food poisoning. Commensal bacteria and AR Although the idea that commensal flora members might play an important role in the development of resistance in bacterial pathogens was raised at least 40 years ago, this concept did not receive serious attention and there was lack of strong scientific evidence supporting this hypothesis. From a regulatory agency’s perspective, the immediate threat of AR to human health is due to pathogens, not commensals; therefore, studies on ART commensal bacteria have received minimal attention compared to AR of pathogens. In addition, AR was not considered to be a major health problem when new antibiotics were developed on a regular basis. Moreover, a technical challenge for microbiologists is the lack of methods to culture all the bacterial species in complex ecosystems and determine gene transmission by such members of the microbiota to enable the measurement of the overall impact of unidentified commensal bacteria on AR dissemination. It has been much more direct to study mechanisms of AR in designated bacterial species such as particular pathogens or opportunistic pathogens, including Salmonella, Campylobacter, E. coli, enterococci, and enterobacteria, than in commensal bacteria. Therefore, for a long time the influence of commensal bacterial populations on AR

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

30 dissemination, and particularly the size and breadth of the resistant population and the spectrum of bacteria involved, has remained unknown.

Roles of commensal bacteria in AR dissemination.  ­ Andremont (6) reported that horizontal transmission of AR genes between commensal and pathogenic microorganisms in ecosystems is a much more likely event than direct gene dissemination from one pathogen to another. AR gene reservoirs in commensal microbes in various environmental and host ecosystems have been identified (48, 71, 101, 102, 111, 117). Furthermore, Nandi et al. (101) determined that gram-positive bac­ teria, not gram-negative Enterobacteriaceae such as E. coli, served as a major reservoir of class 1 AR integrons in poultry litter. Luo et al. (84) determined that an inherited, clumpingassociate­d high-frequency conjugation system, previously considered beneficial for disseminating fermentationrelated traits in Lactococcus lactis, a beneficial bacterium known for its application in dairy fermentation, can enhance in laboratory settings more than 10,000 times the transmission of a broad-host-range drug-resistant plasmid, pAMb1. Since similar clumping-associated highfrequency conjugation systems have been previously reported in Enterococcus, Bacillus, and Lactobacillus, this finding likely has broad biological implications. This study revealed for the first time that commensal bacteria could serve as not only a reservoir but also a facilitator in AR gene dissemination. Furthermore, it illustrates that beneficial bacteria, including fermentation starter cultures and probiotic microbes, are not exempt from AR gene dissemination.

Magnitude and spectrum of the commensal ART bacteria.  Recent studies have revealed that ART bacteria are prevalent in commercial cheese and other retail food products, including seafood, meat, poultry, produce, and delicatessen and restaurant foods (76, 137). A survey conducted during 2004–2005 with samples from grocery chain stores in Columbus, OH, involving multiple types and brands of commercial cheese products, revealed that most of the cheeses contained large numbers of ART bacterial cells, with tet and erm genes detected in about 10% of the isolates assayed (137). The level of AR in the gene pool was as high as 109 copies of AR gene/gram of food in some products, and AR gene carriers involved a broad spectrum of commensal bacteria, including Streptococcus thermophilus, L. lactis, Enterococcus spp., and Pseudomonas spp., with some being used as fermentation starter cultures (88, 137). Due to

SMP_Food Microbiology_CH02.indd

30

limitations of the cultivation conditions that were used, the reported data were likely an underestimate of the actual AR status of the commensal bacteria in the ecosystems. Kastner et al. (63) also reported that starter culture and probiotic strains used in fermented dairy and meat products contained antibiotic resistance-encoding genes. A study by Duran and Marshall (35) revealed the existence of a broad spectrum of ART commensals and pathogenic bacteria in ready-to-eat shrimp products. Because many of these products were directly consumed without further processing, the data suggest that the public consuming these products was often exposed to high level of ART bacteria, including AR genes, even without being exposed to antibiotic treatments or a clinical environment. The use of new methodologies in lieu of the standard selective-enrichment-based approach played a key role in more accurately assessing the extent of the AR situation in the commensal bacterial population. For instance, instead of testing for AR susceptibility of individual isolates, Duran and Marshall (35) applied an antibiotic disk diffusion method directly to food samples and successfully isolated a large number of ART bacteria associated with shrimp products. The resistant bacteria were further identified by gas chromatographyfatty acid methyl ester and API tests, revealing a broad spectrum of ART commensal and pathogenic bacteria. Using a direct-plating method supplemented with DNA profiling techniques, such as 16S rRNA gene sequence analysis and denaturing gel gradient gel electrophoresis coupled with 16S rRNA gene fragment sequence analysis, studies by the Wang research group isolated a wide variety of ART commensal bacteria from dairy, seafood, meat, produce, and delicatessen and restaurant foods (76, 137). These new methods are also applicable to detecting ART bacteria in a variety of hosts and environmental ecosystems.

AR amplification, dissemination, evolution, and persistence in microbial ecosystems.  Animal and human fecal wastes can be major sources of ART bacteria (101, 117, 126). However, contrary to common belief, evidence indicates that the initial and rapid development of the ART bacterial populations in the host GI tract, at least in the case of Tetr bacteria, likely is independent from exposure to antibiotic treatment as well as food intake (124). Kinkelaar and Wang (65) observed that Tetr bacteria and a tetM gene pool developed rapidly (within a few days following regular birth), in the GI tracts of infants, without being exposed to antibiotic treatment. ART bacteria against additional antibiotics have also been found to be prevalent in

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

31

infant fecal samples (141a). In a study involving homebred hamsters, not only were Tetr bacteria isolated from both parents and their offspring, none of which had been treated with antibiotics, but the Tetr gene patterns from Tetr isolates from the baby hamsters were similar to those of the mother but different from those of the father (74). These data suggest that the ART bacteria that initially developed in the infant GI tract were acquired from the mother, either through the birth canal or other contact that eventually resulted in oral exposure of the infants, independent of therapeutic antibiotic exposure and regular food intake. This conclusion is in agreement with recent data on the development of the microbiota in animal and human GI tracts, suggesting that transmission of both specific pathogens such as Salmonella spp. and the general microbial population is largely from mother to infant (9, 77). This is also in agreement with findings from a study of a remote Bolivian community in which ART bacteria were isolated from 73% of the population’s fecal samples (8), although antibiotics were not used in agriculture and rarely used by humans in the community. ART bacteria colonizing the host digestive system can further serve as an “immobilized” source of AR genes for potential HGT events, regardless of the presence of antibiotics. A constant supply of large populations of AR gene-containing bacteria through daily food consumption, coupled with occasional colonization and HGT events, will inevitably have an impact on the AR status of the host gut microflora (137). Jacobsen et al. (60) observed that, after inoculating a tetracycline­resistant Lactobacillus plantarum strain and a nonresistant Enterococcus faecalis strain into the digestive tract of germ-free rats, HGT occurred even in the absence of selective pressure and that the administration of tetracycline caused only a slight increase in transmission of AR. Evidence also suggested AR gene transmission among Bacteroides spp. and bacteria of other genera in the human colon, indicating that the colon likely provides a suitable environment for HGT events to occur (121). Based on quantitative analysis, the cell populations of ART bacteria in fecal samples were much greater than those in foods consumed during daily food intake, suggesting the ART bacteria were enriched in the host GI tract, even in the absence of antibiotic exposure (135). Once released into the environment, these ART bacteria can then spread to other human and animal hosts through contact with water, food, and the environment (23, 104, 126, 130). Hence, proper waste treatment is a critical control point for targeted mitigation of ART bacteria (135).

SMP_Food Microbiology_CH02.indd

31

Many human pathogens and their ART derivatives do not cause disease in animal hosts, such as L. monocytogenes, Salmonella, and E. coli O157:H7, but can cause disease in humans when consumed via raw, smoked, fermented, or undercooked foods. Marshall et al. (89) report there is strong evidence for direct transmission of AR traits from animal- to human-associated microorganisms, such as Salmonella, Vibrio, Campylobacter, Yersinia, and Listeria. Kieke et al. (65) observed that e­xposure to poultry was associated with acquiring a quinupristin-dalfopristi­n resistance gene and inducible quinupristin-dalfopristin resistance in human fecal E. faecium. Johnson et al. (62) determined that many human-source drug-resistant fecal E. coli isolates in the U.S. population were more likely to have originated from poultry than humans, whereas drug-resistant poultry isolates likely were derived from drug-susceptible poultry isolates. Stanton and Humphrey (128) isolated Tetr Mega­ sphaera elsdenii strains having a novel mosaic gene containing hybrids of tetO and tetW from swine fed with or without chlortetracycline. Results revealed a likely role of commensal bacteria not only in the preservation and dissemination of AR in the intestinal tract but also in the evolution of resistance. A significant portion of AR genes from resistant bacteria are quite stable in the absence of antibiotic selective pressure (76). Some AR genes have been integrated into the host genome, and others may reside on plasmids with special stabilization mechanisms (135). For example, pRE25-like plasmid with the toxin-antitoxin (TA) stabilization system has been found to be widely distributed among Enterococcus spp. and other bacteria (87, 114). The TA system was also found responsible for persistence of the vanA-encoding plasmid in vancomycin-resistant enterococci isolated from the clinical and farm environments (99, 127). In addition, a TA-independent plasmid stabilization mechanism is likely responsible for the persistence of plasmid pM7M2, although the plasmid is closely related to pRE25 (75).

Antimicrobial Uses in the Food Chain and Resistance

Besides antibiotics, microorganisms often encounter many antimicrobial hurdles during the journey from farm to fork. Sanitizers and disinfectants such as quaternary ammonia compounds (QACs), chlorine-containing agents, and peroxyacids are often used in the food s­ystem, from production to processing, and in household environments. During food processing, raw meats may be decontaminated with a variety of physical processes or acidic rinses. Fruits and vegetables may be washed in chlorine rinses and/or packed in modified atmosphere

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

32 environments. While antibiotics are not permitted in food products, whether as residues from production or as a means to inhibit the growth of pathogens and spoilage organisms in the final products, various food antimicrobials are added to food to control the growth of both pathogens and spoilage microbes. Many natural food antimicrobials, such as nisin, lysozyme, lactoferrin, essential oils, and organic acids, not classified as antibiotics, have been used in food products (29). Triclosans are used in food processing plants and are now being introduced into many consumer products. Some of these antimicrobial compounds may be removed by washing or inactivated by consumers before consumption; others may be consumed along with the foods.

Microbial Resistance to Food Antimicrobials, Sanitizers, and Disinfectants

Overall, microbial resistance to food antimicrobials, sanitizers, and disinfectants is less well understood and more difficult to characterize and quantify than resistance to antibiotics, and standardized methods have not been developed, nor are surveillance systems in place. Active export mechanisms contribute in part to bacterial resistance to many agents, including antibiotics, but their potential involvement in bacteriocin resistance has not been reported. However, strains of nisin-producing Lactococcus lactis contain immunity mechanisms encoded by genes nisI and nisFEG, located on a transposon (13, 69). Acquisition of the immunity gene(s) can lead to acquired nisin resistance in recipient cells (129). As detailed in chapter 30 (Chemical Preservatives and Natural Antimicrobial Compounds), the intracellular accumulation of protons from organic acids acidifies the cytoplasm, inhibiting many metabolic processes. The cell tries to maintain pH homeostasis by pumping these protons out of the cytoplasm using ATP as an energy source. The resultant energy depletion is a major factor in the inhibition caused by organic acids. Cells have numerous mechanisms, some of which are inducible, to survive in acidic environments, including proton pumping, changes in the cell envelope, and the production of acid shock proteins and chaperones, as well as generation of alkali in the immediate external environment (25). In addition, enzymatic degradation of preservatives, either specialized or general, but different from the enzymes that inactivate antibiotics, may also contribute to resistance to antimicrobials. For example, some bacteria metabolize citric acid, rendering it ineffective in their presence. Many proteases inactivate bacteriocins in a nonspecific fashion. Heir et al. (54) investigated the molecular epidemiology and disinfectant susceptibility of L. monocytogenes isolates

SMP_Food Microbiology_CH02.indd

32

from food processing plants and human infections. The prevalence of QAC-resistant L. monocytogenes is much higher in isolates from food processing facilities than in clinical isolates. While QAC resistance could increase the prevalence of L. monocytogenes in the processing plant, these strains or serotypes may not be the dominant causes of human listeriosis. No correlation between resistance and pulsed-field gel electrophoresis profile or between resistance and persistence was observed. Isolates of L. monocytogenes resistant to the bacteriocin nisin have been reported and characterized. While the genetic determinant of the resistance is currently unknown, these strains typically have a more rigid cell membrane, which may physically impair attachment or insertion of nisin molecules (90, 96). Nisin is approved for use in some foods to prevent clostridial outgrowth, but not to control L. monocytogenes. It is not known if nisin resistance results in increased survival of L. monocytogenes in foods treated with nisin and potentially increased foodborne illness. Some strains of S. aureus are resistant to QACs, either due to plasmid acquisition or chromosomal alteration that resulted in efflux of the sanitizer (11, 47, 125). While this may help the pathogen persist in the food processing environment, most cases of foodborne illness related to S. aureus are related to postprocessing contamination, making the industrial significance of QAC resistance questionable. While some L. monocytogenes and S. aureus strains were described as “resistant” to QACs, the levels of QACs at which “resistance” was observed are several­ fold lower than what is used industrially (59). While these organisms may be classified as resistant in a laboratory, they are likely still susceptible to QACs at the levels normally used in food processing plants.

Adaptation to Stressors at Sublethal Levels

Microbes may be innately resistant to certain food antimicrobials, but sensitive organisms do not typically mutate or acquire resistance to food antimicrobials. However, exposure to subinhibitory levels of antimicrobials may cause temporary adaptation, so that subsequent exposure to levels that are normally lethal is less effective. Bacterial adaptation is the term used to describe temporary phenotypic changes in response to stress. Adaptation is more common for food antimicrobial agents than antibiotics. New genetic material is not required for bacteria to adapt; instead, the stressor quickly activates certain existing pathways and mechanisms to produce a physiological response that helps the organism withstand the stressor. In contrast to the antibiotic selecting for resistance, in the case of adaptation, the antimicrobial causes the observed resistance.

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

33

Some mechanisms for adaptation are known. The synthesis of stress response proteins, including heat shock proteins, is triggered by low levels of the stressor, and alternative sigma (s) factors are involved in the process. These stress response proteins protect the cells from subsequent related or unrelated stressors. In many cases, these proteins serve a general protective function. In addition, the alternative s factors are involved in regulating additional metabolic pathways, likely with universal impact. For example, exposure to low levels of acid may activate the stress response system, providing those cells protection from subsequent heat exposure as well as acid exposure (7). Acid adaptation and the acid tolerance response are briefly discussed in chapter 30 of this volume. There are conflicting data on the persistence of adapted phenotypes. The duration of the phenotype may depend on the underlying mechanism. For example, in the acid tolerance response (ATR) (chapter 30), stress response proteins are quickly produced, but few studies have demonstrated their persistence over time and with subsequent generations. Quorum sensing allows cells to communicate with each other and exchange information in a nongenetic fashion. Signal compounds secreted by stressed cells may also be sensed by surrounding cells, including the nonstressed progenies, and impact the corresponding gene expression, until the signal becomes sufficiently diluted.

Cross-Resistance and Coselection

The ability of a bacterial strain to display resistance to more than one antimicrobial can also be due to crossresistance or coselection. In cross-resistance, the multidrug resistance may be due to the same resistance m­echanism that applies to multiple antibiotics. For example, an alteration in a cellular target may make a cell resistant to all antibiotics that act at that target site; acquiring a general drug efflux may increase the resistance to several antibiotics. In cases of coselection, the microbe may be resistant to a number of different antibiotics, each with distinct mechanisms of action. The different resistance determinants are generally genetically linked. This also applies to resistance against antibiotics, heavy metals, and other biocides (32, 116). Further, as mentioned previously, the exposure to subtherapeutic drug levels may increase the mutation rate and lead to multidrug resistance (66). In a farm study, Alexander et al. (2) evaluated the effect of subtherapeutic administration of antibiotics on the prevalence of antibiotic-resistant E. coli in feedlot cattle, finding that subtherapeutic administration of tetracycline in combination with sulfamethazine increased the prevalence of tetracycline- and ampicillin-

SMP_Food Microbiology_CH02.indd

33

resistant E. coli. However, because this resistance may also be related to additional factors such as diet and addi­ tional oral exposure through environmental contacts, it is important to characterize the corresponding molecular mechanism(s) and confirm the correlation with wellcontrolled laboratory studies. The multiple-antibiotic resistance (mar) operon, which controls the expression of chromosomal genes involved in intrinsic multidrug susceptibility and resist­ ance in Enterobacteriaceae, is of special interest in food microbiology. This global regulator (marRAB) can be expressed by a number of pathogens, including E. coli O157:H7 and Salmonella serovar Enteritidis. Activation of the operon in Salmonella serovar Enteritidis by chlorine, sodium nitrite, sodium benzoate, or acetic acid, all of which are used by the food industry, cause increased resistance to antibiotics such as tetracycline and ciprofloxacin (108). Some sanitation agents and disinfectants may also cause coresistance to other antimicrobial compounds, such as antibiotics. A review by Hegstad et al. (53) cites several studies that examine the relationship between bacterial resistance to QACs and clinically important antibiotics. The cross-resistance can be attributed to nonspecific resistance mechanisms such as altered membrane composition and efflux pumps. An EFSA panel on biological hazards considered the possible effects of chlorine dioxide, acidified sodium chlorite, trisodium phosphate, and peroxyacids as used for poultry carcass decontamination. The EFSA Scientific Opinion is that there are currently no published data to conclude that application of these antimicrobials under the proposed conditions of use will lead to acquired reduced susceptibility to these substances or lead to resistance to therapeutic antimicrobials (36). A few studies have examined the response of ART bacteria to food processing treatments such as heat. Compared to susceptible cells, ART Salmonella and L. monocytogenes are as sensitive to heat as their susceptible counterparts (7, 134). However, there is a relationship between AR and heat resistance; activation of the AR confers increased protection against heat (7). This can be explained by the AR mechanism. The induction of stress response proteins (due to exposure to acid) results in increased heat resistance, for which stress response proteins also play a role. This exemplifies the need to examine situations of cross-resistance on a case-by-case basis; broad assumptions about crossresistance cannot be made without an understanding of the mechanism(s). One study has linked nisin resistance with AR (56). S. aureus nisin-resistant isolates were resistant to antibiotics. The MICs for the antibiotics increased as much

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

34 as 30-fold in the nisin-resistant strains. However, similar studies with Listeria monocytogenes revealed no significant increase in resistance to antibiotics (27, 92). Nisinresistant L. monocytogenes and Clostridium botulinum are more sensitive to food preservatives such as low pH, salt, sodium nitrite, and potassium sorbate (90). Their increased acid sensitivity is caused by increased ATPase activity, which decreases intracellular energy stores (93). Studies (5) have also explored cross-resistance between decontaminants and antibiotics in strains of L. monocytogenes and Salmonella serovars Enteritidis and Typhimurium. Although standard deviations were not provided, the data appear to reveal little to no increase in the MICs of the pathogens to peroxyacetic acid or trisodium phosphate. However, for acidified sodium chlorite (ASC), the MIC increased between 1.88- and 2.71-fold. Exposure to sublethal concentrations of ASC was related to cephalothin resistance in one strain of L. monocytogenes and one strain of Salmonella. For both strains of Listeria but neither of the Salmonella strains, ASC exposure was associated with increased resistance to chloramphenicol. For both Salmonella serovars, ASC exposure was associated with increased streptomycin resistance. The authors cite previously identified mechanistic relationships but do not speculate on mechanisms that may cause the cross-resistanc­e observed in this study.

CRITICAL ISSUES, GAPS, AND FUTURE DIRECTIONS

Proper Data Interpretation and Risk Communication

Despite the large AR gene pool in commensal bacteria, resistant isolates differ in their ability to disseminate the resistance-encoding traits within microbial ecosystems. For instance, the transmission of AR genes from foodborne Enterococcus spp., Staphylococcus spp., Lactococcus spp., and Carnobacterium spp. to human pathogens and residential bacteria, such as Streptococcus mutans and E. faecalis, was illustrated by natural gene transformation and electroporation (76, 137) and from Lactobacillus plantarum to Lactobacillus rhamnosus, Lactococcus lactis, Listeria innocua, Enterococcus faecalis, and L. monocytogenes by conjugation (38); however, attempts to transmit AR genes from S. thermophilus and Pseudomonas sp. to S. mutans and E. coli by the same methods were not successful in the lab setting (data not shown). These data suggest that while it is generally true that commensal bacteria can play an important role in

SMP_Food Microbiology_CH02.indd

34

AR dissemination, the risks may still vary among the organisms. AR in S. thermophilus is probably less of a concern than in other lactic acid bacteria, particularly enterococci, which are prone to HGT and known as second-class (opportunistic) pathogens but first-class problem makers. Enterococci were among the most common ART bacteria with mobile AR genes isolated from fermented sausage, cheese, and many other readyto-eat, restaurant, and delicatessen food items and able to disseminate AR genes by HGT in laboratory settings (76, 137). The impact of the commensal microbe AR issue could be a major one for the food industry. However, despite the large scale of the issue, prevalence of ART bacteria in fermented dairy products on the U.S. market decreased sharply in recent years due to rapid responses from the industry. Besides screening cultures and removing potentially problematic strains from the market, scientists further conducted critical control point analysis throughout the dairy fermentation process in manufacturing plants. It is worth noting that even though the main culture supplies from the starter culture companies are clean, some plant-maintained adjunct cultures still contain ART lactic acid bacteria, which may contribute to the low-level, sporadic incidences found in commercial products (75a). Rapidly addressing the AR issue associated with fermented dairy products resulting in reduced prevalence of AR bacterial contamination revealed how effective such a targeted mitigation strategy can be. Meanwhile, ART commensal bacteria and probiotic strains, including lactic acid bacteria, were still being isolated from specialty cheese products and other fermented products in the European Union and developing countries in which effective AR bacteria reduction strategies have not been put into place (52, 112). Furthermore, probiotics are broadly used in both food animal and aquaculture production as feed supplements, with the intention of replacing antibiotics as growth promoters to reduce the dissemination of AR; however, the safety screening of such probiotic microorganisms is much less than that of probiotics used for human consumption. Lactic acid bacteria as well as enterococci were found among the list of ART bacteria frequently isolated from seafood products, and their AR genes can be transmitted to human-pathogenic and resident bacteria in laboratory settings (76).Therefore, without a thorough understanding of the scientific underpinnings, some mitigation strategies might be problematic themselves. The data suggest that not only is proper safety screening critical for starter cultures and probiotic strains intended for beneficial applications in both living hosts and the environment, but there is also a

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

35

need for effective communication to stakeholders of the scientific information regarding safety use applications. Regardless of whether transfer studies are conducted in vivo or in vitro, the experimental conditions under which they are conducted should always be specified; otherwise, the conclusions derived from such studies may not be directly applicable in the real world. For example, a number of studies have failed to demonstrate the transferability of AR genes from certain beneficial bacterial strains to selected recipients; however, the experimental conditions may have limitations (such as the compatibility of the donor and recipient pairs that were selected). Alternately, when HGT events are successfully demonstrated under experimental conditions, this indicates the likelihood that such events can occur, but not necessarily in the real world. Because of the complexity of microbial ecosystems, as well as the many possibilities of influences that environmental and host conditions can have on the acquisition and dissemination of AR, results from both in vitro and in vivo studies should be appropriately qualified, and generalization of results should not be provided unless well substantiated.

Selective Pressure from Farm to Fork

Gyles (50) has attempted to analyze AR trends globally. He observed that countries with more restricted use of antibiotics generally have lower incidences of AR in animal populations. In countries where AR is high even though the use of specific antibiotics in animals has been eliminated (e.g., ceftiofur resistance in E. coli in Belgium, where cephalosporin use has not been permitted since 2000), it is postulated that the AR determinants are retained because of coselection for other r­esistance agents. Singer et al. (123) studied a normally functioning dairy to examine changes in resistance profiles and genetic diversity of E. coli in ceftiofur-treated and -untreated cattle, whose bacterial flora was susceptible to ceftiofur prior to treatment. They determined that ceftiofur treatment provided a window for detection of the ceftiofur-resistant E. coli genotype and phenotype but did not cause an emergence or amplification of lasting AR. They concluded that finding resistant isolates following antibiotic treatment is not sufficient to estimate the strength of selection pressure or to demonstrate a causal link between antibiotic use and the emergence or amplification of resistance. They also observed that the genotype of the resistant strain did not resemble any of the susceptible E. coli strains in the population and reported that it did not appear that the plasmid carrying the blaCMY-2 gene was transferred to other E. coli cells as a result of selection pressure. However, it is possible

SMP_Food Microbiology_CH02.indd

35

that the ceftiofur-resistant E. coli strain could serve as a source for transmitting the gene to other bacterial genera not addressed in this study. Consideration of a variety of data such as colony counts, community DNA PCR, and genotyping of isolates was important for their interpretation of the findings. Understanding the characteristics and prevalence of background populations of ART bacteria and AR genes is important to making informed conclusions regarding the relationship between antibiotic use and AR. The Centers for Disease Control and Prevention (CDC) monitors AR of foodborne pathogens isolated from human infections and reports this information through the NARMS. While the data are reported for different foodborne pathogens, there is no attribution of the infection to specific sources, such as foods. NARMS data collected by the U.S. Department of Agriculture track AR rates in foodborne pathogens isolated from food animals but do not address the dynamics of AR in food animals during production or from food animals to foods (Fig. 2.1). Slaughter techniques and contamination rates of the raw product, consumer handling, especially storage and cooking of foods, and human dose response are all variables that can influence the transmission of ART foodborne pathogens and complicate the correlations of on-farm AR rates with human illness. In 2002, NARMS was expanded to include surveillance by the Food and Drug Administration Center for Veterinary Medicine of ART Salmonella, Campylobacter, E. coli, and Enterococcus in raw retail meats. These data will help to provide insights into the transfer of ART pathogens from food animals to meat. However, since meat is generally not consumed raw, NARMS will still be unable to provide a direct relationship between ART bacteria from food animals and human foodborne illness acquired from food animals. Importantly, foodborne pathogens account for only a small percentage of the microbes present on meat; hence, ART foodborne pathogens do not represent a major source for transmitting AR genes to microbes in humans. AR in foodborne commensal bacteria could be a useful indicator for early warning and risk assessments of the association of ART pathogens from food animals with human illnesses; however, NARMS does not address commensal bacteria currently. There are several points along the farm-to-table continuum where interventions can influence the selection, dissemination, and control of ART bacteria such as in meat (Fig. 2.1). Decreasing the use of antibiotics, particularly in humans and agriculture, will decrease the selection of ART bacteria; however, it is not possible to eliminate a low level of resistant mutants. Good

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

36

Probiotics

ART microbes in feed

Onsite environmental ART microbes

Probiotics

Safety screening essential

Food Animals Cleaning (CCP)

ART microbes Meat products

Safety screening essential

Humans

Contaminated or decontaminated Amplification, HGT in processing and cooking by With antibiotic processor and consumers (CCP) Without (growth promoter Without Antibiotic antibiotic or therapy) antibiotic treatment

Amplification, HGT

ART microbes in feces

ART microbes in feces

Compost for possible reduction (CCP)

Amplification or reduction, HGT

ART microbes in manure Amplification or reduction, HGT

Waste treatment for reduction (CCP)

Wastes impact offsite environment ART bacteria pool (mostly commensals in water and soil) Produce, etc.

Proper processing for reduction (CCP)

Figure 2.1  Flow of ART microorganisms from farm to fork. CCP, critical control points for mitigation. doi:10.1128/9781555818463.ch2f1

sanitation practices pre- and postharvest, along with prudent farm management practices including utilization of vaccines and proper probiotic microbes, may limit the transfer and colonization of both ART and antibioticsusceptible pathogens among animals. Generally, ART bacteria and their antibiotic-susceptible counterparts are equally sensitive to heat treatments; hence, proper cooking of meat and proper food handling practices in food preparation to prevent cross-contamination are effective control measures to prevent dissemination of foodborne pathogens, whether ART or not. If all antibiotic use was suddenly discontinued, would AR in bacteria disappear or be maintained? Until recently, it was thought that, in the absence of the selective pressure caused by an antibiotic, an AR determinant would be lost and the bacteria would revert to antibiotic sensitivity. Whether AR results from chromosomal mutations or gene acquisition, there is often an associated “fitness cost.” In other words, the

SMP_Food Microbiology_CH02.indd

36

ability of a cell to resist an antibiotic often comes at the expense of another cellular function. For example, ribosomal mutations that prevent antibiotics from binding to their ribosomal target may reduce the rate of protein synthesis (10). If AR is no longer advantageous, the cell would benefit from reverting to the wild type by allowing that protein synthesis to proceed unhindered. Generally speaking, the accumulation of mutations is neutral and sometimes harmful or even lethal to bacterial cells, and bacteria often adopt elaborative mechanism(s) to minimize the negative biological impact by the mutation. In the absence of selective pressure, in any natural environment with a mixed bacterial population, mutations may get reversed, or acquired AR may gradually be lost during population proliferation if they do not provide a survival advantage to the host bacteria. However, reversion does not always happen. In these cases, compensatory mutations alleviate the fitness cost and the resistance phenotype is maintained

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

37

(73). In the case of Denmark, where the discontinuation of antibiotic growth promoters in livestock production decreased the overall antibiotic use, the prevalence of ART enterococci did decrease (140). However, a certain level of glycopeptide-resistant enterococci was found to be persistent in Norwegian poultry and poultry farms several years after the removal of the antibiotic selection pressure, due to the presence of the plasmid stabilization mechanism (127). Hence, a comprehensive understanding of the underlying mechanism(s) is the key to proper data interpretation. In addition, in the absence of antibiotic selective pressure, certain resistant mutants may also survive and dominate due to various survival advantages (fitness) and persistence mechanisms (such as the TA system). For example, fluoroquinolone-resistant Campylobacter (harboring a gyrA mutation) possesses a fitness advantage over the wild type, resulting in dominance of resistant Campylobacter in the chicken host even in the absence of antibiotic selective pressure (85). Dantas et al. (28) recently reported the finding of phylogenetically diverse soil bacteria, many of which are closely related to human pathogens, subsisting on natural and synthetic antibiotics widely distributed in the environment, which indicated an unrecognized reservoir of multiple-AR machinery and the possibility that pathogens could obtain AR genes from environmentally distributed superresistant microbes subsisting on anti­ biotics. More than one-half of the bacterial isolates they identified belonged to the orders of Burkholderiales and Pseudomonadales. Hence, high levels of AR prevalence were found in various hosts and environmental eco­ systems without direct exposure to antibiotics. Various pathways and maintenance mechanisms could have contributed to the phenomena. A review by Allen et al. (3) explored the presence and spread of AR in non­ agricultural, nonclinical environments, addressing selection pressures in the environment, movement of resistance genes (e.g., via physical forces such as wind and watershed, animals, and humans), and antibiotics and resistance in natural communities, and concluded that more detailed studies of environmental reservoirs of resistance are crucial to mitigating infection in the future. These studies delivered the consistent message that a comprehensive understanding of the impact of commensal bacteria and the mechanisms involved in AR emergence, dissemination, and maintenance is essential for effective AR mitigation (135). A comprehensive understanding of microbial AR and adaptation mechanisms and contributing factors can further enable food microbiologists and veterinarians to develop effective intervention strategies to improve

SMP_Food Microbiology_CH02.indd

37

overall food safety. While exposure to sublethal stresses may result in stress adaptation, concurrent exposure to multiple stresses may cause cell death due to metabolic exhaustion as bacteria try to maintain their cellular integrity and homeostatic balance. This is commonly referred to as “hurdle technology” (72). This approach should decrease AR prevalence, since a bacterium would have to be resistant to multiple stressors to survive. Again, understanding the mechanisms of resistance will aid in the selection of the specific “hurdles” to minimize potential resistance. In other cases, instead of sublethal exposure resulting in cross-resistance to other agents, such as the activation of the stress response proteins, the cells’ susceptibility to other stresses may be enhanced. For example, after exposure to alkali cleaning solutions, four of five strains of L. monocytogenes were as sensitive to heat as, or more sensitive to heat than, unexposed cells, and all were more sensitive to free chlorine, benzalkonium chloride, and cetylpyridinium chloride, components of sanitizers, than controls (133). A mechanistic understanding may also help to identify opportunities of collateral sensitivity, whereby the cellular changes resulting in resistance leave the cell more vulnerable to other types of antimicrobial agents. For example, several studies suggest that nisin resistance results in physiological changes that decrease resistance to other food safety interventions such as heat (30), acid, salt, sorbates, and nitrite (90). This was also true for some cephalosporins, which are normally ineffective against L. monocytogenes. The nisin-resistant strain was highly sensitive to expanded-spectrum and broadspectrum cephalosporins, at concentrations where the wild type was virtually unaffected (94). The converse may also be true: resistance to some antibiotics increases sensitivity to nisin. A penicillin-resistant S. aureus mutant was 50-fold more sensitive to nisin (132). Bacillus licheniformis, resistant to the bacitracin it produces, is highly sensitive to detergents, likely due to a specific membrane change (107).

Recommendations

Several recommendation documents have become available since 2009, and an important trend is the recognition that the food chain and HGT are a part of the AR continuum (5a, 24a). A recent expert report (136) concluded that: systematic studies for a comprehensive understanding, both at the macroscopic and microscopic levels, of AR connected to the food chain, is central to design targeted and integrated intervention strategies for effective mitigation.

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

38 Recommended strategies include conducting fundamental studies at the ecosystem level to (i) develop novel experimental and systematic approaches and systematically investigate the AR ecology such as main resistance gene reservoirs and key microbial players within the ecosystems, at and between various links in the food chain from farm to table, including pre- and postharvest environment (agriculture and aquaculture farms, surrounding soil, air, and water, processing, transportation, storage, retail chain, etc.), raw and processed foods, as well as animal and human hosts; (ii) reveal the impact of natural and implemented factors (such as the application of antibiotics, compost, processing treatment, etc.) including dosage effect on the evolution and mitigation of AR in the corresponding ecosystems . . . [136]

And Fundamental studies at the individual microorganism (both pathogen and commensal) level are needed to (iii) reveal molecular mechanisms involved in AR origination, dissemination, persistence and environmental fitness as related to the food system, and (iv) investigate the impact of natural and implemented factors on the evolution and persistence of AR in such organisms as well as ecologically relevant bacterial species; . . . [136]

Also recommended is developing effective mitigation strategies and outcome measurements by pursuing the following directions: (v) identify critical control points for AR based on sophisticated and ecological measures and risk assessment outcomes, and develop and implement agriculture, aquaculture and industrial practices to minimize and contain the spread and persistence of AR in the pre- and post-harvest food environment, products and host ecosystems; (vi) conduct studies with a focus on disease prevention and biosecurity, such as developing vaccines or alternatives for subtherapeutic uses of antibiotics in animal production; (vii) develop and implement integrated research, education, and outreach programs engaging academic, government agencies, industry and consumers including the lay public for effective mitigation; (viii) design and implement studies to measure the impact or effect of potential interventions on existing AR at the macro or micro level. [136]

CONCLUDING REMARKS Antibiotics play a vital role in improving human, animal, and plant health. However, their effectiveness is endangered when their targets become resistant. The human medical community has begun reducing frivolous usage of antibiotics to decrease the selective pressure that generates ART microorganisms. Similar efforts are ongoing in animal agriculture through prudent-use practices.

SMP_Food Microbiology_CH02.indd

38

Because sick animals must be treated, the complete elimination of the use of antibiotics in animals is not practical and may be counterproductive. Efforts to minimize the agricultural use of antibiotics must be balanced to decrease the emergence of antibiotic-resistant microbes while not increasing the level of pathogens associated with food products, as this could result in an overall net increase in foodborne diseases. It is important to recognize that prudent use is not simply about whether to use or not to use antibiotics, but more importantly which, how, and when to use them (137a). Not all manifestations of AR result in increased foodborne illness or affect the ability to address it. However, from a broader ecological perspective, AR may still raise questions. For example, how might AR in commensal microbes and soil microflora affect foodborne illness? Is AR of minimum concern in foodborne pathogens whose illnesses are not treated with antibiotics, such as E. coli O157:H7? What role can food microbiologists and veterinarians play in combating the emergence of ART pathogens? Food microbiologists have no influence over the use of antibiotics in animal agriculture, veterinary medicine, or human medicine, which generates the selective pressure that can increase the prevalence of antibiotic-resistant microbes. Food microbiologists do use bacteriostatic and bactericidal agents other than antibiotics, such as food antimicrobials, sanitizers, and disinfectants, and should be aware of the prevalence and impact of microbial resistance to these agents. Standardized methods to assess resistance to food antimicrobial agents and sanitizers are needed. There is a continuing need to determine what combinations of microbe(s) and antimicrobial agent(s) represent the greatest resistance challenges for food safety. This information is difficult to ascertain, since FoodNet data do not distinguish foodborne illness caused by ART microbes from those caused by susceptible microbes, and they lack robust food attribution information. With a better understanding of the AR profile of the microbes causing illness and better information on the farm-to-table path of these microbes, we may be able to alter practices along that continuum and design intelligent approaches to counter resistance. This should be a continuing effort because microorganisms are constantly evolving and finding new ways to respond to antimicrobials. An important area that has been neglected in the past is the role of commensal bacteria, including beneficial bacteria, in AR dissemination and maintenance. Because these bacteria are present in large cell numbers throughout the food chain from production and processing to consumption, it is the responsibility of

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

39

food microbiologists to fully understand their impact on food safety and public health and minimize AR gene carriers and microbes that are prone to HGT. Finally, ART bacteria appear to respond to food preservation interventions in the same way as their susceptible counterparts. Therefore, interventions against pathogens in foods should be equally effective for antibiotic-resistant and antibiotic-susceptible cells.

References   1. Alekshun, M. N., and S. B. Levy. 2007. Molecular mechanisms of antibacterial multidrug resistance. Cell 128:1037–1050.   2. Alexander, T. W., L. J. Yanke, E. Topp, M. E. Olson, R. R. Read, D. W. Morck, and T. A. McAllister. 2008. Effect of subtherapeutic administration of antibiotics on the prevalence of antibiotic-resistant Escherichia coli bacteria in feedlot cattle. Appl. Environ. Microbiol. 74:4405–4416.   3. Allen, H. K., J. Donato, H. H. Wang, K. A. CloudHansen, J. Davies, and J. Handelsman. 2010. Call of the wild: antibiotic resistance genes in natural environments. Nat. Rev. Microbiol. 8:251–259.   4. Allos, B. M. 2001. Campylobacter jejuni infections: update on emerging issues and trends. Clin. Infect. Dis. 32:1201–1206.   5. Alonso-Hernando, A., R. Capita, M. Prieto, and C. Alonso-Calleja. 2009. Comparison of antibiotic resistance patterns in Listeria monocytogenes and Salmonella enterica strains pre-exposed and exposed to poultry decontaminants. Food Control 20:1108–1111.   5a. American Academy of Microbiology. 2009. Antibiotic resistance: an ecological perspective on an old problem. http://academy.asm.org/images/stories/documents/antibioticresistance.pdf   6. Andremont, A. 2003. Commensal flora may play key role in spreading antibiotic resistance. ASM News 69:601–607.   7. Bacon, R. T., J. R. Ransom, J. N. Sofos, P. A. Kendall, K. E. Belk, and G. C. Smith. 2003. Thermal inactivation of susceptible and multiantimicrobial-resistant Salmonella strains grown in the absence or presence of glucose. Appl. Environ. Microbiol. 69:4123–4128.   8. Bartoloni, A., F. Bartalesi, A. Mantella, E. Dell’Amico, M. Roselli, M. Strohmeyer, H. G. Barahona, V. P. Barron, F. Paradisi, and G. M. Rossolini. 2004. High prevalence of acquired antimicrobial resistance unrelated to heavy antimicrobial consumption. J. Infect. Dis. 189:1291–1294.   9. Bettelheim, K. A., A. Breadon, M. C. Faiers, S. M. O’Farrell, and R. A. Shooter. 1974. The origin of O serotypes of Escherichia coli in babies after normal delivery. J. Hyg. (London) 72:67–70.   10. Bilgin, N., F. Claesens, H. Pahverk, and M. Ehrenberg. 1992. Kinetic properties of Escherichia coli ribosomes with altered forms of S12. J. Mol. Biol. 224:1011–1027.

SMP_Food Microbiology_CH02.indd

39

  11. Bjorland, J., M. Sunde, and S. Waage. 2001. Plasmidborne smr gene causes resistance to quaternary ammonium compounds in bovine Staphylococcus aureus. J. Clin. Microbiol. 39:3999–4004.   12. Blondeau, J. M. 1999. A review of the comparative invitro activities of 12 antimicrobial agents, with a focus on five new respiratory quinolones. J. Antimicrob. Chemother. 43(Suppl. B):1–11.   13. Broadbent, J. R., W. E. Sandine, and J. K. Kondo. 1995. Characteristics of Tn5307 exchange and intergeneric transfer of genes associated with nisin production. Appl. Microbiol. Biotechnol. 44:139–146.   14. Burns, J. L., L. A. Hedin, and D. M. Lien. 1989. Chloramphenicol resistance in Pseudomonas cepacia because of decreased permeability. Antimicrob. Agents Chemother. 33:136–141.   15. Burns, J. L., P. M. Mendelman, J. Levy, T. L. Stull, and A. L. Smith. 1985. A permeability barrier as a mechanism of chloramphenicol resistance in Haemophilus influenzae. Antimicrob. Agents Chemother. 27:46–54.   16. Cabello, F. C. 2006. Heavy use of prophylactic antibiotics in aquaculture: a growing problem for human and animal health and for the environment. Environ. Microbiol. 8:1137–1144.   17. Carattoli, A. 2003. Plasmid-mediated antimicrobial resistance in Salmonella enterica. Curr. Issues Mol. Biol. 5:113–122.   18. Centers for Disease Control and Prevention. 2004. 12 steps to prevent antimicrobial resistance among long-term care residents. Centers for Disease Control and Preven­ tion, Atlanta, GA. http://www.cdc.gov/drugresistance/ healthcare/ltc/12steps_ltc.htm.   19. Centers for Disease Control and Prevention. 2004. National Nosocomial Infection Surveillance (NNIS) System Report, data summary from January 1992–June 2004, issued October 2004. Am. J. Infect. Control 32:470–485.   20. Centers for Disease Control and Prevention. 2005. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food— 10 sites, United States, 2004. MMWR Morb. Mortal. Wkly. Rep. 54:352–356.   21. Centers for Disease Control and Prevention. 2012. National Antimicrobial Resistance Monitoring System for Enteric Bacteria (NARMS): Human Isolates Final Report, 2010. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, GA.   22. Chao, G., X. Zhou, X. Jiao, X. Qian, and L. Xu. 2007. Prevalence and antimicrobial resistance of foodborne pathogens isolated from food products in China. Foodborne Pathog. Dis. 4:277–284.   23. Chee-Sanford, J. C., R. I. Aminov, I. J. Krapac, N. Garrigues-Jeanjean, and R. I. Mackie. 2001. Occurrence and diversity of tetracycline resistance genes in lagoons and groundwater underlying two swine production f­acilities. Appl. Environ. Microbiol. 67:1494–1502.   24. Cloete, T. E. 2003. Resistance mechanisms of bacteria to antimicrobial compounds. Int. Biodeterior. Biodegrad. 51:277–282.

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

40   24a. Codex Alimentarius Commission. 2009. Report of the Third Session of the Codex Ad Hoc Intergovernmental Task Force on Antimicrobial Resistance. Joint FAO/ WHO Food Standards Programme, Codex Alimentarius Commission. http://www.codexalimentarius.net/download/report/730/al33_42e.pdf   25. Cotter, P. D., and C. Hill. 2003. Surviving the acid test: responses of gram-positive bacteria to low pH. Microbiol. Mol. Biol. Rev. 67:429–453.   26. Cox, L. A., D. A. Popken, and R. Carnevale. 2007. Quantifying human health risks from animal antimicrobials. Interfaces 37:22–38.   27. Crandall, A. D., and T. J. Montville. 1998. Nisin resistance in Listeria monocytogenes ATCC 700302 is a complex phenotype. Appl. Environ. Microbiol. 64:231–237.   28. Dantas, G., M. O. A. Sommer, R. D. Oluwasegun, and G. M. Church. 2008. Bacteria subsisting on antibiotics. Science 320:100–103.   29. Davidson, P. M., and M. A. Harrison. 2002. Resistance and adaptation to food antimicrobials, sanitizers, and other process controls. Food Technol. 56:69–78.   30. Davies, J. 1994. Inactivation of antibiotics and the dissemination of resistance genes. Science 264:375–382.   31. Davies, J., and D. Davies. 2010. Origins and evolution of antibiotic resistance. Microbiol. Mol. Biol. Rev. 74:417–433.   32. Davis, I. J., A. P. Roberts, D. Ready, H. Richards, M. Wilson, and P. Mullany. 2005. Linkage of a novel mercury resistance operon with streptomycin resistance on a conjugative plasmid in Enterococcus faecium. Plasmid 54:26–38.   33. Diez-Gonzalez, F., T. R. Callaway, M. G. Kizoulis, and J. B. Russell. 1998. Grain feeding and the dissemination of acid-resistant Escherichia coli from cattle. Science 281:1666–1668.   34. Drlica, K., and X. Zhao. 1997. DNA gyrase, topoisomerase IV, and the 4-quinolones. Microbiol. Mol. Biol. Rev. 61:377–392.   35. Duran, G. M., and D. L. Marshall. 2005. Ready-to-eat shrimp as an international vehicle of antibiotic-resistant bacteria. J. Food Prot. 68:2395–2401.   36. European Food Safety Authority. 2008. Assessment of the possible effect of the four antimicrobial treatment substances on the emergence of antimicrobial resistance—scientific opinion of the Panel on Biological Hazards. http://www.efsa.europa.eu/en/scdocs/scdoc/ 659.htm.   37. European Food Safety Authority. 2010. The Community Summary Report on antimicrobial resistance in zoonotic and indicator bacteria from animals and food in the European Union in 2004-2007. EFSA J. 8:1309. doi:10.2903/j.efsa.2010.1309.   38. Feld, L., E. Bielak, K. Hammer, and A. Wilcks. 2009. Characterization of a small erythromycin resistance plasmid pLFE1 from the food-isolate Lactobacillus plantarum M345. Plasmid 61:159–170.   39. Feng, P., K. A. Lampel, H. Karch, and T. S. Whittam. 1998. Genotypic and phenotypic changes in the emer-

SMP_Food Microbiology_CH02.indd

40

  40.

  41.   42.

  43.

  44.   45.

  46.

  47.   48.   49.

  50.   51.   52.

  53.

gence of Escherichia coli O157:H7. J. Infect. Dis. 177: 1750–1753. Filiousis, G., A. Johansson, J. Frey, and V. Perreten. 2009. Prevalence, genetic diversity and antimicrobial susceptibility of Listeria monocytogenes isolated from open-air food markets in Greece. Food Control 20:314–317. Fontana, R., P. Canepari, M. M. Lleo, and G. Satta. 1990. Mechanisms of resistance of enterococci to beta-lactam a­ntibiotics. Eur. J. Clin. Microbiol. Infect. Dis. 9:103–105. Food and Drug Administration. 2005. Docket No. 2000N-1571, OC 2005194. Animal Drugs, Feeds, and Related Products; Enrofloxacin for Poultry; Withdrawal of Approval of New Animal Drug Application, p. 44048–44049. FR Doc. 05-15223, 1 August 2005. U.S. Food and Drug Administration, Rockville, MD. Food and Drug Administration. 2009. 2009 Summary report on antimicrobials sold or distributed for use in foodproducing animals. http://www.fda.gov/downloads/ ForIndustry/UserFees/AnimalDrugUserFeeActADUFA/ UCM231851.pdf. Gaskins, H. R., C. T. Collier, and D. B. Anderson. 2002. Antibiotics as growth promotants: mode of action. Anim. Biotechnol. 13:29–42. Gayet, S., R. Chollet, G. Molle, J. Pages, and J. Chevalier. 2003. Modification of outer membrane protein profile and evidence suggesting an active drug pump in Enterobacter aerogenes clinical strains. Antimicrob. Agents Chemother. 47:1555–1559. Gebreyes, W. A., P. B. Bahnson, J. A. Funk, J. McKean, and P. Patchanee. 2008. Seroprevalence of Trichinella, Toxoplasma, and Salmonella in antimicrobial-free and conventional swine production systems. Foodborne Pathog. Dis. 5:199–203. Gillespie, M. T., B. R. Lyon, and R. A. Skurray. 1989. Gentamicin and antiseptic resistance in epidemic methicillin-resistant Staphylococcus aureus. Lancet i:503. Gilliver, M. A., M. Bennett, M. Begon, S. M. Hazel, and C. A. Hart. 1999. Antibiotic resistance found in wild rodents. Nature 401:233–234. Guffanti, A. A., S. Clejan, L. H. Falk, D. B. Hicks, and T. A. Krulwich. 1987. Isolation and characterization of uncoupler-resistant mutants of Bacillus subtilis. J. Bacteriol. 169:4469–4478. Gyles, C. L. 2008. Antimicrobial resistance in selected bacteria from poultry. Anim. Health Res. Rev. 9:149–158. Hall, R. M., and C. M. Collis. 1995. Mobile gene cassettes and integrons: capture and spread of genes by sitespecific recombination. Mol. Microbiol. 15:593–600. Hammad, A. M., Y. Ishida, and T. Shimamoto. 2009. Prevalence and molecular characterization of ampicillin-resistant Enterobacteriaceae isolated from traditional Egyptian Domiati cheese. J. Food Prot. 72: 624–630. Hegstad, K., S. Langsrud, B. T. Lunestad, A. A. Scheie, M. Sunde, and S. P. Yazdankhah. 2010. Does the wide use of quaternary ammonium compounds enhance the selection and spread of antimicrobial resistance and thus threaten our health? Microb. Drug Resist. 16:91–104.

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

41

  54. Heir, E., B. Lindstedt, O. Rotterud, T. Vardund, G. Kapperud, and T. Nesbakken. 2004. Molecular epidemiology and disinfectant susceptibility of Listeria monocytogenes from meat processing plants and human infections. Int. J. Food Microbiol. 96:85–96.   55. Heuer, O. E., H. Kruse, K. Grave, P. Collignon, I. Karunasagar, and F. J. Angulo. 2009. Human health consequences of use of antimicrobial agents in aquaculture. Clin. Infect. Dis. 49:1248–1253.   56. Hossack, D. J. N., M. C. Bird, and A. A. Fowler. 1983. The effects of nisin on the sensitivity of microorganisms to antibiotics and other chemotherapeutic agents, p. 425–433. In M. Woodbine (ed.), Antimicrobials and Agriculture. Butterworth, London, United Kingdom.   57. Reference deleted.   58. International Commission for the Microbiological Spec­ ifications of Foods. 2005. Anonymous Microorganisms in Foods 6, 2nd ed., p. 490. Springer, New York, NY.   59. Institute of Food Technologists. 2006. Antimicrobial resistance: implications for the food system. Institute of Food Technologists, Chicago, IL. http://www.ift.org/ knowledge-center/read-ift-publications/science-reports/ expert-reports/antimicrobial-resistance.aspx.   60. Jacobsen, L., A. Wilcks, K. Hammer, G. Huys, D. Gevers, and S. R. Andersen. 2007. Horizontal transfer of tet(M) and erm(B) resistance plasmids from food strains of Lactobacillus plantarum to Enterococcus faecalis JH2-2 in the gastrointestinal tract of gnotobiotic rats. FEMS Microbiol. Ecol. 59:158–166.   61. John, J. F., Jr., and N. O. Fishman. 1997. Programmatic role of the infectious diseases physician in controlling antimicrobial costs in the hospital. Clin. Infect. Dis. 24:471–485.   62. Johnson, J. R., M. R. Sannes, C. Croy, B. Johnston, C. Clabots, M. A. Kuskowski, J. Bender, K. E. Smith, P. L. Winokur, and E. A. Belongia. 2007. Antimicrobial drug-resistant Escherichia coli from humans and poultry products, Minnesota and Wisconsin, 2002-2004. Emerg. Infect. Dis. 13:838–846.   63. Kastner, S., V. Perreten, H. Bleuler, G. Hugenschmidt, C. Lacroix, and L. Meile. 2006. Antibiotic susceptibility patterns and resistance genes of starter cultures and probiotic bacteria used in food. Syst. Appl. Microbiol. 29:145–155.   64. Kaye, K. S., H. S. Fraimow, and E. Abrutyn. 2000. Pathogens resistant to antimicrobial agents. Epidemiology, molecular mechanisms, and clinical management. Infect. Dis. Clin. N. Am. 14:293–319.   65. Kieke, A. L., M. A. Borchardt, B. A. Kieke, S. K. Spencer, M. F. Vandermause, K. E. Smith, S. L. Jawahir, E. A. Belongia, and the Marshfield Enterococcal Study Group. 2006. Use of streptogramin growth promoters in poultry and isolation of streptogramin-resistant Enterococcus faecium from humans. J. Infect. Dis. 194:1200–1208.   66. Kohanski, M. A., M. A. DePristo, and J. J. Collins. 2010. Sublethal antibiotic treatment leads to multidrug resistance via radical-induced mutagenesis. Mol. Cell 37:311–320.

SMP_Food Microbiology_CH02.indd

41

  67. Kohanski, M. A., D. J. Dwyer, and J. J. Collins. 2010. How antibiotics kill bacteria: from targets to networks. Nat. Rev. Microbiol. 8:423–435.   68. Kotra, L. P., J. Haddad, and S. Mobashery. 2000. Aminoglycosides: perspectives on mechanisms of action and resistance and strategies to counter resistance. Antimicrob. Agents Chemother. 44:3249–3256.   69. Kuipers, O. P., M. M. Beerthuyzen, P. G. de Ruyter, E. J. Luesink, and W. M. de Vos. 1995. Autoregulation of nisin biosynthesis in Lactococcus lactis by signal transduction. J. Biol. Chem. 270:27299–27304.   70. Ladely, S. R., M. A. Harrison, P. J. Fedorka-Cray, M. E. Berrang, M. D. Englen, and R. J. Meinersmann. 2007. Development of macrolide-resistant Campylobacter in broilers administered subtherapeutic or therapeutic concentrations of tylosin. J. Food Prot. 70:1945–1951.   71. Lancaster, H., D. Ready, P. Mullany, D. Spratt, R. Bedi, and M. Wilson. 2003. Prevalence and identification of tetracycline-resistant oral bacteria in children not receiving antibiotic therapy. FEMS Microbiol. Lett. 228:99–104.   72. Lestner, L., and G. W. Gould. 2002. Multiple Hurdle Technologies. Kluwer Academic, New York, NY.   73. Levin, B. R., V. Perrot, and N. Walker. 2000. Compensatory mutations, antibiotic resistance and the population genetics of adaptive evolution in bacteria. Genetics 154:985–997.   74. Li, X., K. Sun, L. Zhang, Y. L. Li, and H. H. Wang. 2010. The involvement of animal host in the enrichment of antibiotic resistance, abstr. no. 037-45. Abstr. Inst. Food Technol. Annu. Meet. Chicago, IL.   75. Li, X., V. Alvarez, W. J. Harper, and H. H. Wang. 2011. Persistent, toxin-antitoxin system-independent, tetracycline resistance-encoding plasmid from a dairy Enterococcus faecium isolate. Appl. Environ. Microbiol. 77:7096–7103.   75a. Li, X., Y. Li, V. Alvarez, W. J. Harper, and H. H. Wang. 2011. Antibiotic resistance mitigation in dairy fermentation. Appl. Environ. Microbiol. 77:7171–7175.   76. Li, X., and H. H. Wang. 2010. Tetracycline resistance associated with commensal bacteria from representative ready-to-consume deli and restaurant foods. J. Food Prot. 73:1841–1848.   77. Liljebjelke, K. A., C. L. Hofacre, T. Liu, D. G. White, S. Ayers, S. Young, and J. J. Maurer. 2005. Vertical and horizontal transmission of Salmonella within integrated broiler production system. Foodborne Pathog. Dis. 2:90–102.   78. Lin, J., M. Yan, O. Sahin, S. Pereira, Y. J. Chang, and Q. Zhang. 2007. Effect of macrolide usage on emergence of erythromycin-resistant Campylobacter isolates in chickens. Antimicrob. Agents Chemother. 51:1678–1686.   79. Lin, Y., E. Barker, J. Kislow, P. Kaldhone, M. E. Stemper, M. Pantrangi, F. M. Moore, M. Hall, T. R. Fritsche, T. Novicki, S. L. Foley, and S. K. Shukla. 2010. Evidence of multiple virulence subtypes in nosocomial and c­ommunity-associated MRSA genotypes in companion animals from the upper midwestern and northeastern United States. Clin. Med. Res. 9:7–16.

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

42   80. Lipps, G. 2008. Plasmids: Current Research and Future Trends. Caister Academic Press, Norwich, United Kingdom.   81. Liu, B., and M. Pop. 2009. ARDB—antibiotic resistance genes database. Nucleic Acids Res. 37:D443–D447.   82. Lopez, P., M. Espinosa, B. Greenberg, and S. A. Lacks. 1987. Sulfonamide resistance in Streptococcus pneumoniae: DNA sequence of the gene encoding dihydropteroate synthase and characterization of the enzyme. J. Bacteriol. 169:4320–4326.   83. Luangtongkum, T., B. Jeon, J. Han, P. Plummer, C. M. Logue, and Q. Zhang. 2009. Antibiotic resistance in Campylobacter: emergence, transmission and persistence. Future Microbiol. 4:189–200.   84. Luo, H., K. Wan, and H. H. Wang. 2005. High-frequenc­y conjugation system facilitates biofilm formation and pAMbeta1 transmission by Lactococcus lactis. Appl. Environ. Microbiol. 71:2970–2978.   85. Luo, N., S. Pereira, O. Sahin, J. Lin, S. Huang, L. Michel, and Q. Zhang. 2005. Enhanced in vivo fitness of fluoroquinolone-resistant Campylobacter jejuni in the absence of antibiotic selection pressure. Proc. Natl. Acad. Sci. USA 102:541–546.   86. Lyon, S. A., M. E. Berrang, P. J. Fedorka-Cray, D. L. Fletcher, and R. J. Meinersmann. 2008. Antimicrobial resistance of Listeria monocytogenes isolated from a poultry further processing plant. Foodborne Pathog. Dis. 5:253–259.   87. Maki, T., M. D. Santos, H. Kondo, I. Hirono, and T. Aoki. 2009. A transferable 20-kilobase multiple drug resistance-conferring R plasmid (pKL0018) from a fish pathogen (Lactococcus garvieae) is highly homologous to a conjugative multiple drug resistance-conferrin­g enterococcal plasmid. Appl. Environ. Microbiol. 75: 3370–3372.   88. Manuzon, M. Y., S. E. Hanna, H. Luo, Z. Yu, W. J. Harper, and H. H. Wang. 2007. Quantitative assessment of the tetracycline resistance gene pool in cheese samples by real-time TaqMan PCR. Appl. Environ. Microbiol. 73:1676–1677.   89. Marshall, B. M., D. J. Ochieng, and S. B. Levy. 2009. Commensals: underappreciated reservoir of antibiotic resistance. Microbe 4:231–238.   90. Mazzotta, A. S., K. Modi, M. L. Chikindas, and T. J. Montville. 2000. Inhibition of nisin-resistant (Nisr) Listeria monocytogenes and Nisr Clostridium botulinum by common food preservatives. J. Food Sci. 65:888–890.   91. McClintock, B. 1950. The origin and behavior of mutable loci in maize. Proc. Natl. Acad. Sci. USA 36:344–355.   92. McEntire, J. C. 2003. Relationship between nisin resistance and acid sensitivity of Listeria monocytogenes. Ph.D. thesis. Rutgers University, New Brunswick, NJ.   93. McEntire, J. C., G. M. Carman, and T. J. Montville. 2004. Increased ATPase activity is responsible for acid sensitivity of nisin-resistant Listeria monocytogenes ATCC 700302. Appl. Environ. Microbiol. 70:2717–2721.   94. McManus, P. S., V. O. Stockwell, G. W. Sundin, and A. L. Jones. 2002. Antibiotic use in plant agriculture. Annu. Rev. Phytopathol. 40:443–465.

SMP_Food Microbiology_CH02.indd

42

  95. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5:607–625.   96. Ming, X. T., and M. A. Daeschel. 1993. Nisin resistance of foodborne bacteria and the specific resistance responses of Listeria monocytogenes Scott A. J. Food Prot. 56:944–948.   97. Miranda, J. M., B. I. Vazquez, C. A. Fente, P. Calo-Mata, A. Cepeda, and C. M. Franco. 2008. Comparison of antimicrobial resistance in Escherichia coli, Staphylococcus aureus, and Listeria monocytogenes strains isolated from organic and conventional poultry meat. J. Food Prot. 71:2537–2542.   98. Moellering, R. C., Jr. 1983. In vitro antibacterial activity of the aminoglycoside antibiotics. Rev. Infect. Dis. 5(Suppl.):S212–S232.   99. Moritz, E. M., and P. J. Hergenrother. 2007. Toxinantitoxi­n systems are ubiquitous and plasmid-encoded in vancomycin-resistant enterococci. Proc. Natl. Acad. Sci. USA 104:311–316. 100. Morley, P. S., D. A. Dargatz, D. R. Hyatt, G. A. Dewell, J. G. Patterson, B. A. Burgess, and T. E. Wittum. 2011. Effects of restricted antimicrobial exposure on antimicrobial resistance in fecal Escherichia coli from feedlot cattle. Foodborne Pathog. Dis. 8:87–98. 101. Nandi, S., J. J. Maurer, C. Hofacre, and A. O. Summers. 2004. Gram-positive bacteria are a major reservoir of Class 1 antibiotic resistance integrons in poultry litter. Proc. Natl. Acad. Sci. USA 101:7118–7122. 102. Osterblad, M., K. Norrdahl, E. Korpimaki, and P. Huovinen. 2001. Antibiotic resistance. How wild are wild mammals? Nature 409:37–38. 103. Pankey, G. A., and L. D. Sabath. 2004. Clinical relevance of bacteriostatic versus bactericidal mechanisms of action in the treatment of Gram-positive bacterial infections. Clin. Infect. Dis. 38:864–870. 104. Peak, N., C. W. Knapp, R. K. Yang, M. M. Hanfelt, M. S. Smith, D. S. Aga, and D. W. Graham. 2007. Abundance of six tetracycline resistance genes in wastewater lagoons at cattle feedlots with different antibiotic use strategies. Environ. Microbiol. 9:143–151. 105. Perez-Llarena, F. J., and G. Bou. 2009. Beta-lactamas­e inhibitors: the story so far. Curr. Med. Chem. 16: 3740–3765. 106. Pesavento, G., B. Ducci, D. Nieri, N. Comodo, and A. Lo Nostro. 2010. Prevalence and antibiotic susceptibility of Listeria spp. isolated from raw meat and retail foods. Food Control 21:708–713. 107. Podlesek, Z., A. Comino, B. Herzog-Velikonja, and M. Grabnar. 2000. The role of the bacitracin ABC transporter in bacitracin resistance and collateral detergent sensitivity. FEMS Microbiol. Lett. 188:103–106. 108. Potenski, C. J., M. Gandhi, and K. R. Matthews. 2003. Exposure of Salmonella Enteritidis to chlorine or food preservatives decreases [corrected] susceptibility to antibiotics. FEMS Microbiol. Lett. 220:181–186. 109. Prazak, M. A., E. A. Murano, I. Mercado, and G. R. Acuff. 2002. Antimicrobial resistance of Listeria mono-

Manila Typesetting Company

11/08/2012 06:46AM

2.  Antimicrobial Resistance

110.

111.

112.

113. 114.

115. 116. 117. 118.

119.

120.

121.

122.

123.

43

cytogenes isolated from various cabbage farms and packing sheds in Texas. J. Food Prot. 65:1796–1799. Rankin, S. C., H. Aceto, J. Cassidy, J. Holt, S. Young, B. Love, D. Tewari, D. S. Munro, and C. E. Benson. 2002. Molecular characterization of cephalosporin-resistant Salmonella enterica serotype Newport isolates from animals in Pennsylvania. J. Clin. Microbiol. 40:4679–4684. Ready, D., R. Bedi, D. A. Spratt, P. Mullany, and M. Wilson. 2003. Prevalence, proportions, and identities of antibiotic-resistant bacteria in the oral microflora of healthy children. Microb. Drug Resist. 9:367–372. doi: 10.1089/107662903322762806. Rizzotti, L., F. La Gioia, F. Dellaglio, and S. Torriani. 2009. Characterization of tetracycline-resistant Streptococcus thermophilus isolates from Italian soft cheeses. Appl. Environ. Microbiol. 75:4224–4229. Rosen, G. D. 1996. Pronutrient antibiotic replacement standards discussed. Feedstuffs 75:11–13. Rosvoll, T. C. S., T. Pedersen, H. Sletvold, P. J. Johnsen, J. E. Sollid, G. S. Simonsen, L. B. Jensen, K. M. Nielsen, and A. Sundsfjord. 2010. PCR-based plasmid typing in Enterococcus faecium strains reveals widely distributed pRE25-, pRUM-, pIP501- and pHTbeta-related replicons associated with glycopeptide resistance and stabilizing toxin-antitoxin systems. FEMS Immunol. Med. Microbiol. 58:254–268. Ruiz, J. 2003. Mechanisms of resistance to quinolones: target alterations, decreased accumulation and DNA gyrase protection. J. Antimicrob. Chemother. 51:1109–1117. Russell, A. D. 1997. Plasmids and bacterial resistance to biocides. J. Appl. Microbiol. 83:155–165. Salyers, A. A., A. Gupta, and Y. Wang. 2004. Human intestinal bacteria as reservoirs for antibiotic resistance genes. Trends Microbiol. 12:412–416. Sanyal, D., and D. Greenwood. 1993. An electron microscope study of glycopeptide antibiotic-resistant strains of Staphylococcus epidermidis. J. Med. Microbiol. 39:204–210. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. Shaw, K. J., P. N. Rather, R. S. Hare, and G. H. Miller. 1993. Molecular genetics of aminoglycoside resistance genes and familial relationships of the aminoglycosidemodifying enzymes. Microbiol. Rev. 57:138–163. Shoemaker, N. B., H. Vlamakis, K. Hayes, and A. A. Salyers. 2001. Evidence for extensive resistance gene transfer among Bacteroides spp. and among Bacteroides and other genera in the human colon. Appl. Environ. Microbiol. 67:561–568. Shryock, T. J. 2000. Growth promotion and feed antibiotics, p. 735–743. In J. F. Prescott, J. D. Baggot, and R. D. Walker (ed.), Antimicrobial Therapy in Veterinary Medicine. Iowa State Press, Ames. Singer, R. S., S. K. Patterson, and R. L. Wallace. 2008. Effects of therapeutic ceftiofur administration to dairy cattle on Escherichia coli dynamics in the intestinal tract. Appl. Environ. Microbiol. 74:6956–6962.

SMP_Food Microbiology_CH02.indd

43

124. Sliwa, J. 2007. Resistance genes in our food supply. EurekAlert http://www.eurekalert.org/pub_releases/200705/asfm-rgi051707.php. 125. Smith, K., C. G. Gemmell, and I. S. Hunter. 2008. The association between biocide tolerance and the presence or absence of qac genes among hospital-acquired and community-acquired MRSA isolates. J. Antimicrob. Chemother. 61:78–84. 126. Smith, M. S., R. K. Yang, C. W. Knapp, Y. Niu, N. Peak, M. M. Hanfelt, J. C. Galland, and D. W. Graham. 2004. Quantification of tetracycline resistance genes in feedlot lagoons by real-time PCR. Appl. Environ. Microbiol. 70:7372–7377. 127. Sorum, M., P. J. Johnsen, B. Aasnes, T. Rosvoll, H. Kruse, A. Sundsfjord, and G. S. Simonsen. 2006. Prevalence, persistence, and molecular characterization of glycopeptide-resistant enterococci in Norwegian poultry and poultry farmers 3 to 8 years after the ban on avoparcin. Appl. Environ. Microbiol. 72:516–521. 128. Stanton, T. B., and S. B. Humphrey. 2003. Isolation of tetracycline-resistant Megasphaera elsdenii strains with novel mosaic gene combinations of tet(O) and tet(W) from swine. Appl. Environ. Microbiol. 69:3874–3882. 129. Stein, T., S. Heinzmann, I. Solovieva, and K. Entian. 2003. Function of Lactococcus lactis nisin immunity genes nisI and nisFEG after coordinated expression in the surrogate host Bacillus subtilis. J. Biol. Chem. 278:89–94. 130. Stine, O. C., J. A. Johnson, A. Keefer-Norris, K. L. Perry, J. Tigno, S. Qaiyumi, M. S. Stine, and J. G. Morris. 2007. Widespread distribution of tetracycline resistance genes in a confined animal feeding facility. Int. J. Antimicrob. Agents 29:348–352. 131. Syvanen, M. 1985. Cross-species gene transfer; implications for a new theory of evolution. J. Theor. Biol. 112:333–343. 132. Szybalski, W. 1953. Genetic studies on microbial cross resistance to toxic agents. II. Cross resistance of Micrococcus pyogenes var. aureus to thirty-four antimicrobial agents. Antibiotics Chemother. 3:1095–1103. 133. Taormina, P. J., and L. R. Beuchat. 2002. Survival of Listeria monocytogenes in commercial food-processin­g equipment cleaning solutions and subsequent sensitivity to sanitizers and heat. J. Appl. Microbiol. 92: 71–80. 133a. U.S. General Accounting Office. 1999. Food Safety: the agricultural use of antibiotics and its implications for human health. RCED-99-74. http://www.gao.gov/archive/1999/rc99074.pdf 134. Walsh, D., J. J. Sheridan, G. Duffy, I. S. Blair, D. A. McDowell, and D. Harrington. 2001. Thermal resistance of wild-type and antibiotic-resistant Listeria monocytogenes in meat and potato substrates. J. Appl. Microbiol. 90:555–560. 135. Wang, H. H. 2009. Commensal bacteria, microbial ecosystems and horizontal gene transmission: adjusting our focus for strategic breakthroughs against antibiotic resistance, p. 267–281. In L. Jaykus, H. H. Wang, and L. S. Schlesinger (ed.), Foodborne Microbes: Shaping the Host Ecosystems. ASM Press, Washington, DC.

Manila Typesetting Company

11/08/2012 06:46AM

Factors of Special Significance

44 136. Wang, H. H. 2010. Antibiotic resistance mitigation: a complicated issue begging for targeted investigation. Microbe 5:504–505. 137. Wang, H. H., M. Manuzon, M. Lehman, K. Wan, H. Luo, T. E. Wittum, A. Yousef, and L. O. Bakaletz. 2006. Food commensal microbes as a potentially important avenue in transmitting antibiotic resistance genes. FEMS Microbiol. Lett. 254:226–231. 137a. Wang, H. H., and D. W. Schaffner. 2011. Antibiotic resistance and mitigation: how much do we know and where should we go from here? Appl. Environ. Microbiol. 77:7053–7059. 138. Willmott, C. J., and A. Maxwell. 1993. A single point mutation in the DNA gyrase A protein greatly reduces binding of fluoroquinolones to the gyrase-DNA complex. Antimicrob. Agents Chemother. 37:126–127. 139. Wong, M. R., V. Reddy, H. Hanson, K. M. Johnson, B. Tsoi, C. Cokes, L. Gallagher, L. Lee, A. Plentsova, T. Dang, A. Krueger, K. Joyce, and S. Balter. 2010. Antimicrobial

SMP_Food Microbiology_CH02.indd

44

resistance trends of Shigella serotypes in New York City, 2006-2009. Microb. Drug Resist. 16:155–161. 140. World Health Organization. 2003. Impacts of antimicrobial growth promoter termination in Denmark. WHO/ CDS/CPE/ZFK/2003.1. World Health Organization, Geneva, Switzerland. http://www.who.int/salmsurv/en/ Expertsreportgrowthpromoterdenmark.pdf. 141. Yücel, N., S. Citak, and M. Önder. 2005. Prevalence and antibiotic resistance of Listeria species in meat products in Ankara, Turkey. Food Microbiol. 22:241–245. 141a. Zhang, L., D. Kinkelaar, Y. Huang, Y. Li, X. Li, and H. Wang. 2011. Acquired antibiotic resistance: are we born with it? Appl. Environ. Microbiol. 77:7134–7141. 142. Zhao, S., S. Qaiyumi, S. Friedman, R. Singh, S. L. Foley, D. G. White, P. F. McDermott, T. Donkar, C. Bolin, S. Munro, E. J. Baron, and R. D. Walker. 2003. Characterization of Salmonella enterica serotype Newport isolated from humans and food animals. J. Clin. Microbiol. 41:5366–5371.

Manila Typesetting Company

11/08/2012 06:46AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch3

Peter Setlow Eric A. Johnson

Spores and Their Significance

Members of the gram-positive Bacillus and Clostridium spp. and some closely related genera respond to slowed growth or starvation by initiating the process of sporulation, and the resultant spores can cause practical problems in food microbiology. The molecular biology of sporulation and spore resistance in Bacillus subtilis has been extensively studied (6, 47, 60, 83, 129, 130, 149, 152, 155). With the recent availability of genome sequences of clostridia of industrial and medical importance, a molecular understanding of sporulation in clostridia is also being revealed (70, 118, 120). This chapter describes the fundamental basis of sporulation and the problems that spores present to the food industry.

PHYLOGENY OF SPOREFORMERS The primary sporeformers of significance in foods are Bacillus, Clostridium, Anoxybacillus, Desulfotomaculum, Geobacillus, Paenibacillus, and Sporolactobacillus. These species have received extensive recent phylogenetic analysis (35), which provides insight into their evolution and physiological characteristics. Sporeformers are classified in phylum XIII of the Firmicutes based primarily

3

on sequencing of genes encoding small-subunit rRNA. Overall, the sporeformers are of low DNA mol% G+C and possess a gram-positive cell wall structure. Other systematic, s­pecies, and physiological characteristics have been outlined in detail (35). Interestingly, food-relate­d bacteria classified in the Firmicutes include Listeria, Staphylococcus, Streptococcus, and Lactobacillus, and it has been hypothesized that these genera have lost the ability to sporulate during evolution (33). Despite the ability of sporeformers to remain in a dormant state for long periods of time, their rate of evolution has been i­ndicated to be the same as that of non-spore-forming bacteria (98). Sporulation has been most extensively studied in B. subtilis and forms the basis of the description in this chapter. The first obvious morphological event in sporulation is an unequal cell division. This creates the smaller prespore or forespore compartment and the larger mother cell compartment. As sporulation proceeds, the mother cell engulfs the forespore, resulting in a forespore within a mother cell, and eventually the mother cell lyses. Since the spore is formed within the mother cell, it is termed an endospore.

Peter Setlow, Department of Molecular, Microbial and Structural Biology, University of Connecticut Health Center, Farmington, CT 060303305. Eric A. Johnson, Department of Bacteriology and Food Research Institute, University of Wisconsin–Madison, Madison, WI 53706.

45

46 Throughout sporulation, gene expression is ordered not only temporally but also spatially, as some genes are expressed only in the mother cell or the forespore. The pattern of gene expression during sporulation is controlled by the ordered synthesis and/or activation of five new sigma (s; specificity) factors for RNA polymerase and many DNA-binding proteins. As sporulation proceeds, there are striking morphological and biochemical changes in the developing spore. It becomes encased in two and sometimes three novel layers: a large layer of peptidoglycan (PG) termed the spore cortex, whose structure differs from that of growing-cell PG, a number of spore coat layers, and in some species, an exosporium. Both the coats and exosporium contain proteins unique to spores (59). The spore’s central region or core accumulates a huge depot (³10% of spore, dry weight) of pyridine-2,6-dicarboxylic acid (dipicolinic acid [DPA]) (Fig. 3.1), as well as a large amount of divalent cations, and the core loses much water (47). In addition, the developing forespore synthesizes a large amount of novel small acid-soluble proteins (SASP), some of which coat the spore chromosome and protect the DNA from damage (149, 152, 153, 155). As a result of these and other changes, the spore becomes metabolically dormant and resistant to harsh conditions including heat, radiation, and chemicals. Despite the spore’s extreme dormancy, if given the appropriate stimulus (often a sugar or amino acid), the spore can return to life via spore germination (117, 119, 151). Within minutes of exposure to a germinant, spores lose their unique characteristics, including loss of DPA by excretion, loss of the cortex and SASP by degradation, and loss of spore resistance. Completion of germination allows progression into outgrowth, when metabolism of endogenous and exogenous compounds begins and macromolecular synthesis is initiated. Eventually the outgrowing spore is converted back into a growing cell. Detailed study of sporulation, spores, and spore germination and outgrowth has been motivated by a number of factors, one of which is the attraction of this developmental system. Another motivating factor is the major role played by sporeformers in food spoilage and food-

Figure 3.1  Structure of DPA. Note that at physiological pH both carboxyl groups will be ionized. doi:10.1128/9781555818463.ch3f1

Factors of Special Significance borne diseases; related to this factor is the recognition of the potential for the use of spores of Bacillus anthracis as an agent of bioterrorism. While much knowledge has been gained in studies motivated by these disparate factors, the amount of cross talk between individuals working on either basic or applied aspects of spore research has never been optimal. Hence, one purpose of this chapter is to highlight the state of knowledge of molecular mechanisms of sporulation, spore dormancy, germination, and outgrowth. This review will focus on molecular mechanisms, most of which have been examined in B. subtilis. This bacterium is neither an important pathogen nor an important agent of food spoilage. However, its natural transformability, as well as an abundance of molecular biological and genetic information, including the early completion of its genomic sequence, has made B. subtilis the model sporeformer for mechanistic studies on sporulation, spore germination, and spore resistance. The completed genome sequences of >70 gram-positive spore-forming species are now available, and comparison of these sequences indicates that there is a tremendous degree of conservation among genes whose products are involved in sporulation, spore resistance, and spore germination. In addition, this conservation extends to many similar mechanisms, including regulatory mechanisms, involved in sporulation, spore resistance, and spore germination in these various species (33, 57, 90, 118, 119, 152, 153, 187).

SPORULATION

Distribution of Sporeformers

The sporulating bacteria discussed in this chapter form heat-resistant endospores that contain DPA and are refractile or phase bright under phase-contrast micro­ scopy. Most studies on sporulation, spores, and spore germination have been carried out with species of either the aerobic bacilli or the anaerobic clostridia. However, members of many closely related genera form similar spores (33, 155).

Induction of Sporulation

Induction of sporulation in the laboratory is generally by nutrient limitation by either exhaustion of nutrients during growth or shifting cells from a rich to a poor medium. Addition of an inhibitor (decoyinine) of guanine nucleotide biosynthesis also leads to sporulation. Although these methods cause most cells in a culture to sporulate, this may not be how sporulation is induced in nature. Even cells in a growing culture produce a small but finite number of spores, with the percentage of

3.  Spores and Their Significance spores increasing as the culture’s growth rate decreases (32). Massive sporulation of a cell population generally takes place only when cells enter stationary phase. However, sporulation is not an obligatory outcome of entry into stationary phase, and many stationary-phase events are attempts to access new sources of nutrients such that sporulation either will not take place or is at least delayed. These stationary-phase events include the following (4, 22, 26, 87, 104, 115): (i) synthesis and secretion of degradative enzymes such as amylases and proteases; (ii) synthesis and secretion of antibiotics such as gramicidin or bacitracin; (iii) in some species, synthesis and release of protein toxins; (iv) development of motility; (v) killing and cannibalism of sister cells in the population; and (vi) in a few species (e.g., B. subtilis), development of genetic competence (Fig. 3.2). Although these phenomena are not necessary for sporulation, they

47 are often regulated by mechanisms that modulate gene expression during sporulation. Like sporulation, competence is also regulated in a cell density-dependent manner via the secretion of small peptides termed competence pheromones. There are also drastic changes in stationary-phase cell metabolism, with these changes extending into sporulation. Some of these changes include catabolism of polymers like poly-b-hydroxybutyrate formed in vegetative growth and initiation of oxidative metabolism due to synthesis of tricarboxylic acid (TCA) cycle enzymes. Many of these enzymes are not present in the developing forespore. Consequently, the mother cell and forespore have different metabolic capabilities. Progression into sporulation generally requires regulatory signals that lead to the derepression of genes expressed only in stationary phase, although usually not the products of genes encoding degradative enzymes, antibiotics, toxins, and proteins involved in motility or

Figure 3.2  Morphological, biochemical, and physiological changes during sporulation of a rod-shaped Bacillus cell. In stage 0, a cell with two nucleoids (N) is shown; in stage IIi, the mother cell and forespore are designated MC and FS, respectively. Note that the forespore nucleoid is more condensed than that in the mother cell. Stage IIii is not shown in this scheme, and the forespore nucleoid is not shown after stage III for clarity. The time of some biochemical and physiological events, such as forespore dehydration and acquisition of types of resistance to different chemicals (all lumped together as “chemical resistance”), stretches over a number of stages. The data for this figure are taken from reference 155. doi:10.1128/9781555818463.ch3f2

48 competence (83, 128). However, induction of sporulation does require completion of chromosome replication, repair of DNA damage, and induction of synthesis of TCA cycle enzymes. Entry into sporulation is also increased in cells growing at high cell density by small molecules secreted into the growth medium (87, 93). In general, extracellular signals are extremely important in directing decisions about the immediate and ultimate fates of stationary-phase cells of sporeformers, as these cells have a variety of developmental pathways available, including competence, motility, biofilm (including extracellular matrix) formation, and sporulation, and most of these pathways are not mutually exclusive (93).

Morphological, Biochemical, and Physiological Changes during Sporulation

Sporulation can take as little as 8 hours and is divided into eight stages based largely on morphological characteristics (Fig. 3.2). Growing cells are in stage 0, and in stage I the sporulating cell’s two nucleoids align in an axial filament that can be observed in electron micrographs. The first morphological feature of sporulation easily seen in the light microscope is the formation of an asymmetric septum dividing the sporulating cell into the mother cell and forespore compartments (129). However, the mechanism guiding the asymmetric placement of the sporulation septum is not completely clear. Immediately after asymmetric septation, only ca. 30% of the chromosome destined for the spore is present in the forespore; the remainder is subsequently transferred into the forespore, and then both mother cell and forespore compartments contain complete and apparently identical single chromosomes (148). However, the forespore chromosome is initially more condensed than the mother cell chromosome. A biochemical marker for late stage II is the synthesis of alkaline phosphatase. During stage II, genes in the two compartments of the sporulating cell may be expressed differentially. Following septum formation, the mother cell plasma membrane engulfs the forespore, surrounding the forespore with two complete membranes, the inner and outer forespore (or spore) membranes (see below). There are a number of changes that occur in the transition from stage II to stage III, leading to the subdivision of stage II into three substages, two of which are shown in Fig. 3.2. In the transition from stage III to stage IV, a large PG structure termed the cortex is laid down between the inner and outer forespore membranes. The cortex PG has a structure similar to that of cell wall PG, but with a number of differences (described below). The spore’s

Factors of Special Significance germ cell wall is made at the same time as the cortex but appears to have the same structure as growing cell wall PG (132). During the stage III-to-stage IV transition, the forespore synthesizes glucose dehydrogenase and SASP. The developing forespore acquires full UV resistance and some chemical resistance at this time, and the forespore chromosome adopts a ring-like shape due to the binding of specific SASP (129). Late in stage III, the forespore pH decreases by 1 to 1.5 units and forespore dehydration begins. In the stage IV-to-stage V transition, the proteinaceous coat layers are laid down outside the outer forespore membrane (59). The coats of spores of Bacillus species can contain ³70 proteins, almost all of which are unique to spores. The reason for the plethora of coat proteins is not clear, as many B. subtilis coat proteins can be lost with no apparent phenotypic effect. Forespore g-radiation resistance begins to be acquired during this period, as is further chemical resistance, and forespore dehydration continues (149, 152). During the stage Vto-stage VI transition, the spore core’s depot of DPA is accumulated following DPA synthesis in the mother cell. DPA uptake is paralleled by uptake of divalent cations, predominantly Ca2+, but many Mg2+ and Mn2+ cations as well (47). The great majority of these cations are in the spore core, presumably in a 1:1 complex with DPA. The precise state of these compounds in the spore core is not known, although the amount of DPA accumulated exceeds its solubility. During this period, the spore’s central region or core undergoes the final process of dehydration. Because of the high ratio of solids to water in the spore core at this stage, the spore appears bright in phase-contrast microscopy. It has been suggested that the spore core is in a glass-like state (1), but this suggestion now appears incorrect (88, 174). Because of permeability changes in the spore membranes, most likely the inner membrane, the spore at this stage stains poorly, if at all, with common bacteriological stains, in particular nucleic acid stains. The spore also becomes metabolically dormant during this period and acquires further g-radiation and chemical resistance (149). Finally, in the transition to stage VII, autolysins are produced in the mother cell, resulting in its lysis and release of the spore. The eight stages are particularly useful in characterizing various asporogenous or spo mutants that have no defect in growth but are blocked in a particular stage in sporulation. These spo mutants are given an added designation denoting the stage in which they are blocked. Hence, spo0 and spoII mutants are blocked in stages 0 and II, respectively. The various stages have also allowed correlation of biochemical changes with mor-

3.  Spores and Their Significance phological changes (Fig. 3.2). However, the sporulation scheme outlined above is an oversimplification for several reasons. First, since the various stages are intermediates in a continuous developmental process, it is probably somewhat misleading to think of the stages as discrete entities. Second, the scheme is only for rodshaped bacteria that sporulate without terminal swelling of the forespore. There are many sporeformers in which the forespore compartment swells considerably and the mother cell elongates such as the toxigenic clostridia Clostridium botulinum and C. tetani; and some sporeformers (e.g., Sporosarcina species) grow as cocci, and their sporulation septum is placed symmetrically (35, 155, 194).

Regulation of Gene Expression during Sporulation

Much of our knowledge of regulation of gene expression during sporulation has been derived from analysis of spo mutants in B. subtilis, and asporogeny can be caused by mutations in >75 genes (130). The identification of biochemical or physiological markers for the sporulation stages provides another aid to understanding gene regulation during sporulation (Fig. 3.2). Analysis of these markers in spo mutants indicates that sporulation is primarily a linear sequence of events, although development in the two cell types must be coordinated. In general, spo0 mutants exhibit no s­porulation-specific markers, spoII mutants exhibit only stage 0-specific events, etc. While analysis of spo mutants has furnished a broad outline of regulation of gene expression during sporulation, molecular genetics has provided a detailed understanding of this process (60, 129, 130). Sporulation requires transcription of many new genes, as has been elegantly demonstrated by microarray technology (38, 163, 184). These changes in transcription during sporulation are directed in large part by changes in the specificity of the cell’s RNA (E) polymerase due to synthesis of five new s factors that have promoter specificities distinct from that of the main housekeeping sigma factor, sA. The sporulation-specific sigma factors direct transcription in the predivisional cell (sH), the mother cell compartment (sE followed by sK), and the forespore compartment (sF followed by sG). In addition, changes in sporulation-specific gene expression require modulation of transcription by (i) activation or inactivation of s factors by covalent and noncovalent modification; (ii) regulatory communication to coordinate mother cell and forespore development; (iii) synthesis of DNA binding proteins to activate or repress transcription; and (iv) degradation of DNA binding proteins and s factors (33,

49 60, 83, 129, 130). This information has provided a detailed picture of the control of gene expression during sporulation that is striking not only in its complexity but also in the redundancy of its control mechanisms. The following discussion of these control mechanisms has been simplified and concentrates on major regulatory gene products. Initiation of B. subtilis sporulation requires expression of spo0 gene products in vegetative cells (129). The most important of these spo0 gene products is Spo0A, the response regulator half of a two-component regulatory system. These signal transduction systems transmit signals, often by binding to DNA and affecting transcription, when an aspartyl residue in the response regulator is phosphorylated by a sensor kinase (129). In growing cells, Spo0A is primarily in the dephosphorylated state. However, under conditions that initiate sporulation, the level of Spo0A rises, as does its degree of phosphorylation, the latter through action of multiple sensor kinases. In B. subtilis the majority of phosphate on Spo0A is derived via a phosphorelay initiated by phosphorylation of Spo0F (Figs. 3.3 and 3.4). The phosphate is then transferred from phosphorylated Spo0F (Spo0F~P) to Spo0A by Spo0B. There are two major kinases, KinA and KinB, as well other minor kinases that initiate the phosphorelay (124). Mutations in either kinA or kinB have little or no effect on sporulation, but a kinA kinB double mutant is asporogenous. In addition to the k­inases, there are many phosphatases that can dephosphorylate Spo0A~P, either directly or through dephosphorylation of Spo0F~P (Fig. 3.3). This multiplicity of kinases and phosphatases allows a variety of environmental signals to be integrated to determine intracellular levels of Spo0A~P, and a threshold concentration of Spo0A~P is needed to initiate sporulation by modulating the expression of ³100 genes (45, 46). A number of regulators of the kinases and phosphatases that modulate Spo0A~P levels have been identified (45, 46, 129), and overexpression of kinases that generate

Figure 3.3  Some gene products and reactions that affect levels of Spo0A~P. Spo0E is a phosphatase that acts on Spo0A~P; RapA and RapB are phosphatases that act on Spo0F~P (155). doi:10.1128/9781555818463.ch3f3

Factors of Special Significance

50

Figure 3.4  Regulation of gene expression during sporulation. The effect of Spo0A~P on repressors is negative; other effects of regulatory molecules on reactions are generally positive, although the effect of signals may be positive or negative. The enclosure of the pro-s factors and s factors denotes that at this time these factors are inactive. This figure is adapted from that in reference 155. doi:10.1128/9781555818463.ch3f4

Spo0A~P triggers sporulation in growing cells (42, 46). Not all cells in a population choose to sporulate, and the major factor influencing this decision is the Spo0A~P level, which is quite heterogeneous between individual cells in sporulating populations, and stochasticity in the overall phosphorelay appears to play the major role in the heterogeneity in Spo0A~P levels (22, 34, 183). The phosphorylation of Spo0A increases its affinity for binding to sites upstream of several key genes, although different sites have different affinities for Spo0A~P (46). The abrB gene has a high affinity for Spo0A~P, and binding of Spo0A~P decreases abrB transcription. Since AbrB is labile, a decrease in abrB transcription rapidly decreases intracellular AbrB concentration. AbrB is a repressor of a number of genes normally expressed in the stage 0-to-stage II transition, including spo0H, which encodes sH. AbrB is also an activator for synthesis of a second repressor, termed Hpr, that represses additional stage 0-expressed genes, especially genes for several proteases. In addition to AbrB and Hpr, there is a third repressor, termed SinR, that represses other genes expressed in the stage 0-to-stage II transition, including those of the spoIIG and spoIIA operons. SinR action is blocked by synthesis of a protein termed SinI that binds to

SinR and blocks its action. Synthesis of SinI is stimulated by Spo0A~P and probably repressed by AbrB and Hpr. There is also a fourth repressor, CodY, that represses many genes whose expression is needed early in stationary phase/sporulation, including genes whose products are involved in competence, motility, and the TCA cycle (83, 105, 159). CodY also represses genes required for sporulation, including spo0A. The repressor function of CodY requires branched-chain amino acids and GTP, and the latter effect likely explains why decoynine, an inhibitor of guanine nucleotide biosynthesis, can induce sporulation in growing cells (159). The consequences of this regulation are that early in the stage 0-to-stage II transition, many genes normally repressed during vegetative growth are derepressed through a decrease in GTP levels and an increase in the level of Spo0A~P (Fig. 3.4). Among these stage 0-to-stage II genes is spo0H, encoding sH, and early in sporulation there is an increase in EsH transcription (129). As Spo0A~P levels increase, concentrations become sufficient to bind the promoters of the spoIIG and spoIIA operons from which SinR has been removed, triggering the transcription of these two operons. The spoIIA operon is transcribed by EsH and encodes three proteins, with the third cistron (spoIIAC)

3.  Spores and Their Significance encoding another sigma factor, termed sF. The spoIIAB gene encodes an inhibitor of sF function, whereas spoIIAA encodes a SpoIIAB antagonist; SpoIIAB is also a protein kinase that can phosphorylate SpoIIAA (60, 129). Prior to septation, sF is inactive in the sporulating cell due to its interaction with SpoIIAB, and interaction of SpoIIAB with SpoIIAA is blocked by phosphorylation of SpoIIAA by SpoIIAB. The spoIIG operon is transcribed by EsA and encodes two proteins, with the second cistron, spoIIGB, encoding another s factor for RNA polymerase, sE (60, 129, 130). Unlike sF, sE is synthesized as an inactive precursor termed pro-sE. The first gene of the spoIIG operon (spoIIGA) is the protease responsible for processing pro-sE to sE. Action of SpoIIGA on pro-sE requires gene expression under sF control in the forespore (Fig. 3.4). Although spoIIG is transcribed before septum formation, sE is not generated until after septation and then only in the mother cell. sF is also produced before septum formation and is maintained in an inactive state by interaction with SpoIIAB as noted above. However, following septation, sF becomes active in the forespore, as SpoIIAB leaves sF and interacts with SpoIIAA. This process is promoted largely by dephosphorylation of SpoIIAA~P by the SpoIIE phosphatase that resides in the sporulation septum (8, 25, 190), but it is not completely clear how sF activity is confined to the forespore (36). EsF transcribed genes include rsfA, spoIIR, gpr, and spoIIIG (60, 129, 130, 184) (Fig. 3.4). RsfA is a DNA-binding protein that modifies the specificity of EsF. SpoIIR promotes pro-sE processing by SpoIIGA in the mother cell shortly after septation, and this requirement for spoIIR transcription in the forespore for pro-sE processing in the mother cell is the first example of cross talk between the two compartments of the sporulating cell that coordinate mother cell and forespore development. The gpr gene encodes a protease that acts on SASP in the first minutes of spore outgrowth, and spoIIIG encodes another s factor, termed sG, that is responsible for the bulk of forespore-specific transcription (184). However, sG is not active following its synthesis in stage III, at least in part because of the product of the sF-controlled csfB gene (21, 74). Ultimately, mother cell-specific event(s), including transcription of the spoIIIA operon by EsE, are required for generation of active EsG (Fig. 3.3 and 3.4). At least one SpoIIIA protein is an essential component of a channel between the mother cell and forespore compartments, and although the molecules that pass through this channel have not been definitively identified, they may be nutrients (16, 102). This is the second example of cross talk between mother cell and forespore, although the precise mechanism for regulation of sG activity in the

51 forespore is not clear (145). The generation of sE only in the mother cell and the synthesis of sG only in the forespore now establish compartment-specific transcription during sporulation. As noted above, EsG transcribes a number of genes expressed only in the forespore. As the third example of regulatory cross talk, one such gene (spoIVB) is responsible for communicating with the mother cell and coordinating gene expression in this compartment (Fig. 3.4; see below). The ssp genes are a large set of genes transcribed by EsG. These genes encode the SASP that are major protein components of the spore core (see below) (149, 150). Transcription of genes by EsG is modulated by the DNA binding protein SpoVT, and the spoVT gene is transcribed by EsG (184). In contrast to EsG, EsE transcribes genes only in the mother cell, including genes needed for cortex biosynthesis and some genes (cot genes) needed for coat formation and assembly (38, 130). EsE also transcribes gerR and spoIIID, which encode DNA binding proteins that modulate EsE action, resulting in transcription of different classes of EsE-dependent genes (17, 33, 83). One such Es-dependent gene that requires SpoIIID for transcription is sigK, which encodes a final new s factor, termed sK. In B. subtilis, the sigK gene has an intervening sequence that is removed only from the mother cell genome by a recombinase. Expression of the recombinase is regulated such that generation of an intact sigK gene just precedes sigK transcription. However, this intervening sequence is absent in the sigK genes of other Bacillus species (60, 129, 130). As is the case with sE, sK is synthesized as an inactive pro-sK and is processed proteolytically about 1 hour after its synthesis. Conversion of pro-sK to sK in the mother cell requires expression of a gene (spoIVB) in the forespore (the third example of cross talk) and participation of the sE-controlled spoIVF operon in the mother cell (Fig. 3.4). SpoIVFB is the protease that processes pro-sK, and SpoIVFA is an inhibitor of SpoIVFB function. EsK also transcribes sigK (in conjunction with SpoIIID), the genes for DPA synthase, other cot genes, and gerE, which encodes a DNA-binding protein that modulates EsK specificity (Fig. 3.4). Most knowledge of gene expression during sporulation ends at this point, including information about genes that may be required for mother cell lysis. However, the gene for one autolytic enzyme (CwlC) is transcribed by EsK (130). The preceding picture of regulation of gene expression during sporulation is derived predominantly from detailed studies with B. subtilis. However, as noted above, genomic sequence data indicate that many key sporulation regulatory genes such as spo0A, spoIIGA,

Factors of Special Significance

52 spoIIGB, spoIIIG, and sigK are present in other sporeformers, including Clostridium spp. (118, 120, 187). The striking conservation of both the sequences of the shared proteins and the organization of these genes strongly suggests that regulation of sporulation across Bacillus and Clostridium species is quite similar, and this has generally been determined to be the case (33, 57, 90, 187).

THE SPORE

Spore Structure

The structure of the dormant spore is very different from that of a growing cell (Fig. 3.5), as some spore structures, including the exosporium and coats, have no counterparts in growing cells. The outermost spore layer, the exosporium, varies significantly in size between species, being prominent in spores of Bacillus cereus, B. anthracis, and Bacillus thuringiensis and some Clostridium species, and may not be present in spores of other species such as B. subtilis (59, 165). This structure can exclude large molecules such as antibodies and may play some role in spore pathogenesis. The exosporium is composed of lipid, carbohydrate, and protein, including glycoprotein, and the proteins are unique to spores. Under the exosporium are the spore coats. In B. subtilis spores, a number of distinct coat layers are seen in

electron micrographs. The coats are composed primarily of protein, with ³70 different proteins in B. subtilis spore coats, all of which are unique to spores. Some coat proteins are essential for assembly of the coat structure, but specific roles for many coat proteins are not known. Posttranslational modifications of coat proteins, including dityrosine and g-glutamyl-lysine cross-links, may also be important in spore coat structure. The coats protect spore PG from attack by lytic enzymes and the spore’s inner layers against many chemicals. However, the coats play no major role in maintenance of spore resistance to heat or radiation (47, 152). Underlying the coats is the outer forespore membrane, a functional membrane in the developing forespore but likely not functional in the dormant spore (152). The protein composition of this membrane is different from that of the inner forespore membrane. Underlying the outer forespore membrane is the cortex. Cortical PG is structurally similar to cell wall PG, but with several differences (132): cortical PG contains diaminopimelic acid even if vegetative cell PG contains lysine, and ca. 65% of the muramic acid residues in cortical PG lack peptide residues. Although some muramic acid residues contain a single d-alanine, much is present as muramic acid-d-lactam (MAL), a compound not present in growing-cell PG. A number of genes involved in cortex synthesis have been identified and char-

Figure 3.5  Structure of a dormant spore. The various structures are not drawn precisely to scale, especially the exosporium, whose size varies tremendously between spores of different species. The relative size of the germ cell wall is also generally smaller than shown. The positions of the inner and outer forespore membranes, between the core and the germ cell wall and between the cortex and coats, respectively, are also noted. doi:10.1128/9781555818463.ch3f5

3.  Spores and Their Significance acterized, and the pathway for MAL synthesis during sporulation is fairly well understood (50, 120). While the timing of cortex biosynthesis and cross-linking during sporulation has been determined, the distribution of cross-links throughout the cortex is not known. This is unfortunate, as the spore cortex is likely responsible for the dehydration of the spore core and hence for spore dormancy and much resistance (see below). Between the cortex and the inner forespore membrane is the germ cell wall, and germ cell wall PG appears to be identical to vegetative cell PG (132). The next structure, the inner forespore membrane, is a functional membrane and an extremely strong permeability barrier slowing the entry of almost all molecules, including perhaps even water, into the spore core (152, 174, 185). A lipid probe in this membrane is largely immobile, suggesting that this membrane has a relatively “frozen” structure (31). However, the inner membrane’s phospholipid content is similar to that of growing cells (55). The volume surrounded by the inner forespore membrane appears smaller than expected based on this membrane’s phospholipid content. However, the core’s volume expands significantly upon completion of spore germination and in the absence of membrane synthesis (31). A lipid probe in the inner membrane also becomes fully mobile at this time. Finally, the central region or core contains the spore’s DNA, ribosomes, and most enzymes, as well as DPA and most divalent cations. There are also unique gene products in the dormant spore, including the large SASP pool (ca. 10% of spore protein), much of which is bound to spore DNA (149, 150, 155; see also below). A notable feature of the spore core is its low water content (47). While vegetative cells have ca. 4 g of water per gram of dry weight, the spore core has only 0.4 to 1 gram of water per g of dry weight. The core’s low water content likely plays a major role in spore dormancy and in spore resistance to a variety of agents. The core’s low water content is also likely the reason for the immobility of ions and protein in the spore core (30). In contrast to the low water content in the spore core, the water content in other regions of the spore is similar to that in growing cells.

Spore Macromolecules

Some proteins in the spore core and outer layers are not present in growing cells. The SASP are particularly noteworthy, and some of these proteins play a major role in spore resistance (152, 153). SASP are synthesized in the forespore during stage III of sporulation, when their coding genes are transcribed primarily by EsG, and are located in the spore core. There are two kinds of major

53 SASP in Bacillus spp., the g-type and the a/b-type. The 75- to 105-residue g-type SASP comprise ca. 5% of spore protein and do not bind other spore macromolecules. Their only known function is to be degraded during outgrowth, providing amino acids for metabolism and protein synthesis. g-Type SASP degradation is initiated by the SASP-specific germination protease (GPR). The g-type SASP are encoded by a single gene in Bacillus spp. but are not present in spores of Clostridium species. The sequences of g-type SASP are not homologous to those of other proteins, and these proteins’ sequences have diverged significantly during evolution. The a/b-type SASP make up 3 to 5% of total spore protein and are also in spores of Clostridium species. In contrast to g-type SASP, a/b-type SASP are coded for by up to seven genes, all of which are expressed in parallel, although in most species two proteins are expressed at high levels and the remainder at lower levels. However, all a/b-type SASP have similar properties in vitro and in vivo (see below). The amino acid sequences of these small (60- to 75-residue) proteins are highly conserved both within and across species, although they too have no sequence homology to other proteins (150, 153). The a/btype SASP are also degraded to amino acids in the first minutes of spore outgrowth in a GPR-initiated process. The a/b-type SASP are DNA-binding proteins, both in vivo and in vitro (86, 152, 153). In vivo these proteins saturate the spore chromosome and provide much of the spore DNA’s resistance to various treatments (see below). Binding of a/b-type SASP to DNA is not sequence specific but exhibits a preference for GC-rich regions, although AT-rich regions are bound. The proteins bind to the outside of the DNA helix, interacting primarily with the sugar phosphate backbone. This binding alters the DNA structure (86), provides resistance to chemical and enzymatic cleavage of the DNA backbone, alters the DNA’s UV photochemistry, and greatly slows DNA depurination (see below). The spore nucleoid also contains HBsu, the major protein found on the vegetative cell nucleoid; this protein covers ca. 5% of spore DNA and modulates the effect of a/b-type SASP on spore DNA properties (153). A number of proteins present in vegetative cells are absent from spores. These include amino acid and nucleotide biosynthetic enzymes that are degraded during sporulation and resynthesized during spore outgrowth (117). Other proteins, including enzymes of amino acid and carbohydrate catabolism, are present at similar levels in spores and cells, and spores also contain enzymes for RNA and protein synthesis and DNA repair. However, at least one protein needed for initiation of DNA replication may be absent. Enzyme activities

Factors of Special Significance

54 common to both cell and spore proteins are almost always due to the same gene product. In general, spore and cell rRNAs and tRNAs are similar if not identical, although much tRNA in spores lacks the 3¢-terminal A residue and little spore tRNA is aminoacylated. Spores do contain some mRNA that has been identified by microarray techniques (10, 75), although the function of this spore mRNA is not clear. Spore DNA appears identical to growing cell DNA but has a different structure due to different associated proteins as noted above.

Spore Small Molecules

Spores differ from growing cells in their small molecules (Table 3.1), which are located in the core. The small amount of core water and the huge depot of DPA and

Table 3.1  Small molecules in cells and spores of Bacillus

species

Content (mmol/g dry wt) Molecule(s)

Cellsa

Sporesb

ATP

3.6

£0.005

ADP

1

AMP

1

Deoxynucleotides

0.59

<0.025d

NADH

0.35

<0.002e

NAD

1.95

0.11e

NADPH

0.52

<0.001e

NADP

0.44

0.018e

Acyl-CoA

0.6

<0.01e

CoASHf

0.7

0.26e

CoASSX

<0.1

0.54e

3PGA

<0.2

5–18

Glutamic acid

38

24–30

DPA

<0.1

410–470

Ca

 

380–916

Mg2+

 

86–120

Mn2+

 

27–56

7.6–8.1h

6.3–6.9h

g

2+

H+

0.2 1.2–1.3 c

a Values for B. megaterium in mid-log phase are from references 149 and 155. b Values are the ranges from spores of B. cereus, B. subtilis, and B. megaterium and are from references 149 and 155. c Value is the total of all four deoxynucleoside triphosphates. d Value is the sum of all four deoxynucleotides. e Values are for B. megaterium only. f CoASH, free CoA. g CoASSX, CoA in disulfide linkage to CoA or a protein. h Values are expressed as pH and are the ranges from B. cereus, B. megaterium, and B. subtilis.

divalent cations have already been noted (Table 3.1). The ions in the spore core are immobile, as is at least one normally mobile protein (30), consistent with a dearth of free water in spores. The pH in the spore core is 1 to 1.5 units lower than in a growing cell or the mother cell compartment (Table 3.1). Spores of most species accumulate a large depot of 3-phosphoglyceric acid (3PGA) (Table 3.1) shortly after and in response to the forespore pH decrease (117). However, spores have small amounts of nucleoside triphosphates and reduced numbers of pyridine nucleotides or acyl-coenzyme A’s (CoA), although the “low-energy” forms of these compounds are present (Table 3.1). The high-energy forms are lost from the forespore late in sporulation (Fig. 3.2). Much of the CoA in spores is in disulfide linkage, some as a CoA disulfide and some linked to protein (Table 3.1). The function of these disulfides is unknown, but they are reduced early in spore outgrowth. Spores also have low levels of most free amino acids but often have high levels of glutamate (Table 3.1).

Spore Dormancy

Spores are metabolically dormant, catalyzing no metabolism of endogenous or exogenous compounds. The major cause of this dormancy is undoubtedly the low water content of the spore core, which precludes protein mobility and enzyme action (30, 47, 152). A reflection of this dormancy is that the spore core contains at least two enzyme-substrate pairs that are stable for years but interact in the early minutes of spore outgrowth, resulting in substrate degradation. These two enzyme-substrat­e pairs are 3PGA-phosphoglycerate mutase (PGM) and SASP-GPR (117). Although regulatory mechanisms other than dehydration initially stabilize 3PGA and SASP in the developing forespore despite the presence of PGM and GPR, the lack of PGM and GPR action on their substrates in dormant spores over very long periods may be largely due to spore core dehydration.

SPORE RESISTANCE The spore’s metabolic dormancy is undoubtedly one factor in its survival in the absence of nutrients. A second factor is the spore’s extreme resistance to potentially lethal treatments, including heat, radiation, chemicals, desiccation, and various types of plasma (47, 152, 153, 189). Representative data of resistance or susceptibility of growing cells and spores of B. subtilis to different potentially lethal treatments are shown in Table 3.2. There are, however, some species of Bacillus that form much more resistant spores than does B. subtilis. Spore resistance is due to a variety of factors, with

3.  Spores and Their Significance

55

Table 3.2  Killing and mutagenesis of spores and cells of B. subtilis by various treatmentsa A. Freeze-drying Survival (%) No. of freeze-drying cycles 1

Cells

b

Wild-type spores

a−b− sporesc

100 (<0.5)

7 (14)

2

3 B. 10% Hydrogen peroxidea

Survival (%) Time of treatment (min) 2.5

Cells

Wild-type spores

0.3

92

b

5

88

10

a−b− sporesc 50 10 (14)

20

50

60

6 (£0.5)

0.1

C. UVd Dose to kill 90% of the population (J/m2) Cellsb

Wild-type spores

a−b− sporesc

40

315

25

D. Moist heat D value Treatment temp (°C)

Cells

b

Wild-type spores

a−b− sporesc

95

14 min (£0.5)

85

360 min (£0.5)

15 min (13)

105 h

10 h

2.5 yr (£0.5)

2.8 mo (18)

65

<15 s

22 E. Dry heat

D value Treatment temp (°C)

Cellsb

120 90

Wild-type spores

a−b− sporesc

33 min (12) 5 min

5.5 min (12)

Data from references 43, 149, and 155. Values in parentheses are the percentages of survivors with asporogenous or auxotrophic mutations when spores undergo 30 to 99% killing. b Cells in log phase growth. Similar results have been obtained with wild-type and a−b− cells. c These spores lack the two major a/b-type SASP and thus ca. 80% of the a/b-type SASP pool. d UV irradiation with light predominantly at 254 nm. a

spore core dehydration and a/b-type SASP involved in many types of resistance, whereas the impermeability of the spore’s inner membrane may be involved in only one. Since different and sometimes multiple factors contribute to different types of spore resistance, it is not surprising that spore resistance to different treatments is acquired at different times in sporulation (Fig. 3.2). In recent years, some mechanisms of spore

resistance to heat, UV, and chemicals have been elucidated. The following discussion of spore resistance concentrates on B. subtilis spores because of the detailed mechanistic data available for these spores. However, studies with spores of other Bacillus species, in particular studies of heat resistance, have revealed that factors involved in resistance of B. subtilis spores are also involved in resistance of spores of other bacterial genera

56

Factors of Special Significance

as well as Bacillus species, but the relative importance of particular factors in spore resistance may vary between species.

g-radiatio­n resistance is the lack of knowledge of the precise lethal damage, presumably to spore DNA, caused by g-radiation.

Spore Freezing and Desiccation Resistance

Spore UV Radiation Resistance

Growing bacteria are killed during freezing and desiccation, unless special precautions are taken. The mechanism of this killing is not clear but may involve DNA damage. In contrast to the sensitivity of growing cells to freeze-drying, spores are resistant to multiple exposures to freeze-drying (Table 3.2). The a/b-type SASP are one factor providing spore resistance to freeze-drying by preventing DNA damage (Table 3.2). The spore’s DPA is another factor providing spore desiccation resistance, as spores deficient in DPA are more sensitive to desiccation than their DPA-replete counterparts, and killing of DPA-less spores by desiccation is also due to DNA damage (146). However, bacterial spores do not achieve desiccation resistance by accumulating sugars such as trehalose (152).

Spore Pressure Resistance

Spores are more resistant to high pressures (>50 megapascals [MPa]) than are growing vegetative cells (154). In general, spores are killed more rapidly at lower pressures (50 to 300 MPa) than at higher pressures (400 to 600 MPa). The most effective way to kill pressuregerminated spores is by heat, and hence pressure treatments are often carried out at elevated temperatures. However, spore germination by lower pressures proceeds by the activation of the spore’s nutrient germinant receptors (nGRs), and the action of nGRs decreases at elevated temperatures. In contrast, germination promoted by higher pressures does not require the nGRs but may involve a change in the permeability of the spore’s inner membrane allowing DPA release, and the rate of this process increases as the temperature rises. Even if the high-pressure-germinated spore goes no further in germination, this DPA-less spore is more heat sensitive than is the dormant spore.

Spores are generally 7 to 50 times more resistant than growing cells to UV radiation at 254 nm, the wavelength giving maximal killing (111, 152) (Table 3.2). Spores are also more resistant than growing cells at longer and shorter UV wavelengths. UV resistance is acquired by the developing forespore ca. 2 hours before acquisition of heat resistance (Fig. 3.2), in parallel with synthesis of a/b-type SASP. The a/b-type SASP are essential for spore UV resistance, but spore coats, cortex, and core dehydration are not. However, for some species, pigments in spores’ outer layers may provide some UV protection (152, 153). Spore UV resistance is largely due to a different UV photochemistry of DNA in spores and in growing cells. The major photoproducts formed by UV irradiation of growing cells or purified DNA are cyclobutane-type dimers between adjacent pyrimidines. The most abundant of these are between adjacent thymine residues (TT) (Fig. 3.6A), with smaller amounts formed between adjacent cytosine and thymine residues (CT) or adjacent cytosine residues (CC). In addition, UV irradiation of cells or purified DNA generates various 6,4-photoproducts (64PP) also formed between adjacent pyrimidines.

Spore g-Radiation Resistance

Spores are more resistant to g-radiation than are growing cells, and g-radiation resistance is acquired 1 to 2 h before acquisition of heat resistance (110). The a/btype SASP are only one factor contributing to spore gradiation resistance (103), but the other factors have not been elucidated. The low water content in the spore core would be expected to provide protection against g-radiation, but variations in spore g-radiation resistance with levels of spore core hydration have not been studied. One impediment to understanding spore

Figure 3.6  Structures of (A) cyclobutane-type TT dimer and (B) 5-thyminyl-5,6-dihydrothymine adduct (spore photoproduct). The positions of the hydrogens noted by the asterisks are the locations of the glycosylic bond in DNA. doi:10.1128/9781555818463.ch3f6

3.  Spores and Their Significance All these photoproducts can be lethal as well as mutagenic. In contrast, UV irradiation of spores generates few cyclobutane-type dimers and 64PPs but rather large amounts of a thyminyl-thymine adduct initially termed “spore photoproduct” (SP) (Fig. 3.6B). The yield of SP as a function of UV fluence in spores is similar to the yield of TT as a function of UV fluence in growing cells, and SP is a potentially lethal photoproduct. Hence, the difference in UV photochemistry alone between DNA in growing cells and spores is insufficient to explain spore UV resistance, as there must be a difference in the capacity of cells and spores to repair TT and SP. Indeed, spores have two major mechanisms for SP repair, both of which operate in the first minutes of spore outgrowth. One mechanism is an excision repair system that repairs TT and other lesions in growing cells. Spores lacking this repair system are two- to threefold more UV sensitive than are wild-type spores. The second repair system, unique to both spores and SP, monomerizes SP to two thymines without excision of the lesion and is very error-free. Spores lacking this SP-specific repair system are 5- to 10-fold more UV sensitive than are wild-type spores, and those lacking both repair systems are 20- to 40-fold more UV sensitive. SP-specific repair is catalyzed by the spore photoproduct lyase (Spl), an iron-sulfur protein that uses S-adenosylmethionine (SAM) as a cofactor. Spl is a member of the “radical SAM” family of enzymes that use an [Fe-S] center plus SAM to generate a catalytic adenosyl radical. In contrast to enzymes of excision repair that are present in growing cells and spores, Spl is present only in spores. The major factor causing the altered UV photochemistry of spore DNA is the saturation of DNA with a/b-type SASP; all a/b-type SASP appear relatively interchangeable in this regard. Spores lacking ca. 80% of these proteins (a−b− spores) are more UV sensitive than are growing cells (Table 3.2), and UV irradiation of a−b− spores generates significant amounts of TT and 64PP and reduced amounts of SP. UV irradiation of DNA saturated with any of a number of purified a/b-type SASP generates SP and minimal TT, CT, CC, or 64PP. However, yields of SP as a function of UV fluence in these complexes are ca. 10-fold lower than yields in spores. This difference is due to the spore’s DPA, which acts as a photosensitizer. The change in the UV photochemistry of DNA upon binding of a/b-type SASP indicates that the DNA in this complex has an altered structure, and this is a structure in between that of A- and B-DNA (86). During spore outgrowth, a/b-type SASP are degraded to amino acids, a process initiated by GPR (151). Because of the photosensitizing action of DPA, spores early in outgrowth are more UV resistant than are dormant

57 spores due to DPA release prior to SASP degradation. However, as SASP degradation proceeds, this elevated UV resistance falls to that of vegetative cells.

Spore Chemical Resistance

Spores are more resistant than growing cells to chemicals, including aldehydes, oxidizing agents (Table 3.2), phenols, chloroform, octanol, alkylating agents such as ethylene oxide, iodine, and detergents, as well as to pH extremes and lytic enzymes such as lysozyme (99, 110, 152). Resistance of spores to these agents is acquired at different times in sporulation (Fig. 3.2). For enzymes such as lysozyme, spores that lack coat protein due to chemical treatment or mutation are sensitive to cortex degradation by lytic enzymes, and coats protect spores against digestion by bacteriovores (59, 84). Coat-defective spores are more sensitive to many chemicals, including oxidizing agents, perhaps because coat proteins serve as a “reactive armor” that inactivates toxic chemicals before they reach more-sensitive targets further within the spore. The slow passage of hydrophilic molecules across the spore’s inner membrane is also important in spore resistance to chemicals (152). Some chemicals, including formaldehyde, alkylating agents, and nitrite, kill spores at least in part by DNA damage, and the a/b-type SASP are essential in protection against inactivation of both Bacillus and Clostridium spores by these agents (89, 110, 152). In contrast, many oxidizing agents do not kill spores by DNA damage, but rather by damaging the spore’s inner membrane such that even if the treated spores are germinated artificially, the inner membrane ruptures, leading to spore death (152). Some detailed information is available about the points raised above for oxidizing agents such as peroxides that kill growing cells by several mechanisms, a major one being generation of hydroxyl radicals that cause DNA damage. However, spore killing by hydrogen peroxide or organic hydroperoxides such as t-butylhydroperoxide (tBHP) is not accompanied by significant DNA damage or mutagenesis (Table 3.2) (152). Consequently, DNA in spores must be well protected by these agents. While the spore coat, low core water content, and lowpermeability inner membrane may play some role in resistance to H2O2 and tBHP, these factors do not directly protect spore DNA, and this protection is due to saturation of spore DNA with a/b-type SASP, with a−b− spores being much more sensitive to these peroxides (Table 3.2). In addition, wild-type spores are not appreciably mutagenized by H2O2 or tBHP, whereas a−b− spores are heavily mutagenized (Table 3.2). These results suggest that saturation of DNA with a/b-type SASP provides such good protection against DNA damage that killing

Factors of Special Significance

58 of wild-type spores by such peroxides is by other mechanisms. However, in a−b− spores, the rates of DNA damage caused by peroxides are greatly increased such that DNA damage is a significant cause of spore death. Supporting this simple model is that spores of wild-type, but not a−b− strains, acquire one component of H2O2 resistance during sporulation in parallel with accumulation of a/b-type SASP. However, a/b-type SASP binding to DNA is not the only factor in spore hydrogen peroxide resistance. Indeed, an additional component of this resistance is acquired during sporulation at the time of final spore core dehydration. This may reflect roles in hydrogen peroxide resistance for spore core dehydration or the low permeability of the spore’s inner membrane or the fact that the loss of reduced pyridine nucleotides from the developing spore (which also occurs at this time) decreases production of hydroxyl radicals through the Fenton reaction. Unlike the situation in growing cells, enzymes such as catalases and alkylhydroperoxide reductases and the DNA-binding protein MrgA play no role in spore resistance to peroxides.

times at lower temperatures, as spore D values increase 4- to 10-fold for each 10°C decrease in temperature (47). Consequently, a spore with a D value at 90°C (D90°C) of 30 min may have a D20°C value of many years. Knowledge of the identity of the target(s) that is damaged in spore killing by moist heat is crucial to an understanding of spore moist heat resistance. This target is not DNA, as spore moist heat killing is associated with neither DNA damage nor mutagenesis (152) (Table 3.2). However, recent work has indicated that spore killing by moist heat is through damage to one or more proteins, although the identity of specific target proteins is not known (28, 29). Sublethal heat treatment can also damage spores in some way, with this damage being repairable during spore outgrowth (62). Although the specific damage is not known, it may well be damage to one or more proteins (27, 28). Perhaps not surprisingly, moist heat resistance can vary substantially between individual spores in populations (49, 196), although the reason for this has not been elucidated. In contrast to our lack of knowledge about the mechanism(s) of killing spores by heat, there is more information on factors that modulate spore heat resistance, as discussed below.

Spore Heat Resistance

The extreme moist heat resistance of spores, the spore resistance most familiar to food microbiologists, has great implications for the food industry and is probably the most studied spore resistance property (47, 152). Spore moist heat resistance is remarkable, as spores of many species can withstand 100°C for many minutes. Heat resistance is often quantified as a Dt value, which is the time in minutes at temperature t needed to kill 90% of a cell or spore population. Generally, D values for spores at a temperature of t + 40°C are approximately equal to those for their vegetative cell counterparts at temperature t. The extended survival of spores at elevated temperatures is paralleled by even longer survival

Sporulation Temperature

Elevated sporulation temperatures increase spore moist heat resistance (9, 47, 110, 152). Indeed, spores of thermophiles generally have much higher moist heat resistance than do spores of mesophiles. Since spore proteins are generally identical to cell proteins, spore proteins are not intrinsically heat resistant. Presumably, proteins from thermophiles are more heat stable than those from mesophiles, accounting for the higher moist heat resistance of spores of thermophiles. However, spores of the same strain are also more heat resistant when prepared at higher temperatures (Table 3.3). This is probably due

Table 3.3  Heat resistance of B. subtilis spores prepared at different temperatures with

different ions and with or without a/b-type SASPa Spore

Prepn temp (°C)

Mineralization

H2O (g/g of spore core wet wt)

D100°C (min)

B. subtilis

50

Native

0.335

45

 

37

Ca

0.425

37

 

37

Native

0.50

8.9

 

37

H

0.571

2.7

 

20

Native

0.55

4.9

B. subtilis 168 wild type

37

Native

0.37

360b

B. subtilis 168 a−b−

37

Native

0.37

15b

a b

Data from references 9 and 43. D85°C.

2+

+

3.  Spores and Their Significance to the reduced core water content in spores prepared at higher temperatures (9) (Table 3.3), although how higher sporulation temperatures cause reduced spore core water content is not known. Elevated sporulation temperatures do not alter spore resistance to dry heat or UV radiation but do increase spore resistance to a number of chemicals (152). The latter effect may be due to changes in levels of spore coat proteins as a function of the sporulation temperature. In contrast to the situation in growing bacteria, heat shock proteins play no role in spore heat resistance (110, 152).

a/b-Type SASP

The killing of spores by moist heat is not by DNA damage, as neither general mutagenesis (Table 3.2) nor DNA damage accompanies spore killing, even though high temperatures could cause DNA depurination. Therefore, spore DNA is well protected against moist heat. Protection of DNA in both Bacillus and Clostridium spores against moist heat damage appears to be due to the saturation of spore DNA by a/b-type SASP (89, 152, 153), and a−b− spores of B. subtilis have D values 5- to 10-fold lower than those of wild-type spores (Tables 3.2 and 3.3). In addition, while moist heat killing of wild-type spores generates <1% mutants in survivors, such killing of a−b− spores produces up to 18% mutants among survivors (Table 3.2). Killing of a−b− spores by moist heat is also accompanied by a large amount of DNA damage, including abasic sites, and a/ b-type SASP markedly slow rates of DNA depurination in solution (152, 153). Wild-type spores are much more resistant to dry heat than to moist heat and exhibit a high level of DNA damage and mutagenesis accompanying dry heat killing, whereas a−b− spores are more sensitive to dry heat than are wild-type spores and exhibit dry heat resistance similar to that of dry vegetative cells (Table 3.2). These findings suggest that (i) a/b-type SASP are a major factor in increasing the dry heat resistance of wild-type spores, (ii) a/b-type SASP provide significant DNA protection against dry heat damage, but (iii) dry spores are so well protected against mechanisms of heat killing other than DNA damage that at elevated temperatures DNA damage does eventually kill dry spores. In support of these findings, a/b-type SASP slow the depurination caused by dry heat treatment of purified DNA, although not as much as for DNA in solution (152).

Spore Mineralization

Spores accumulate large amounts of divalent cations late in sporulation, approximately in parallel with DPA. Analyses of mutations in B. subtilis that eliminate DPA

59 synthetase have revealed that DPA-less spores are much less moist heat resistant than are their DPA-replete counterparts (152). Much of the latter effect is likely due to the increased core water content of DPA-less spores (see below). However, DPA also provides significant protection to spore DNA against damage by either dry or moist heat when a/b-type SASP are absent (146). DPA is also required for dormant spore stability, as DPA-less spores germinate spontaneously, although they can be stabilized by other mutations (146). The amount and type of mineral ions accumulated also affect spore heat resistance (9, 47). Analyses of spores of several species from which mineral ions have been removed by titration with acid and the spores then back-titrated with a mineral hydroxide give the order of spore heat resistance with different cations as follows: H+ < Na+ < K+ < Mg2+ < Mn2+ < Ca2+ < untreated. Despite the clear role for mineral ions in spore heat resistance, the mechanism of this effect remains unclear. Alteration of spore mineralization can alter spore core water content (Table 3.3), which presumably causes a significant effect on heat resistance. However, mineralization may also affect spore heat resistance independently of its effects on core water content (Table 3.3).

Spore Core Water Content

Low core water content is the major factor contributing to spore heat resistance. Dehydration of the spore begins in the stage III-to-stage IV transition and continues through stages IV and V, with final dehydration taking place approximately in parallel with acquisition of full spore heat resistance (47, 152). Synthesis of the spore cortex is essential both for effecting this dehydration and for maintaining the dehydrated state of the spore core. This is likely due to the ability of PG to change its volume upon changes in ionic strength and/or pH. If an expansion in cortex volume is restricted to one direction, i.e., towards the spore core, core water would be extruded via mechanical action. Although the precise mechanism of this process is unclear, the important role of the cortex in heat resistance is shown by the inverse correlation between spore heat resistance and the volume occupied by the spore cortex relative to that of the core (47). This correlates not only across species but also in a single species in which spore cortex biosynthesis has been altered by mutation (132). Presumably, the volume of the spore cortex influences the degree of core dehydration, and the amount of cortex and presumably its mechanical strength are crucial to the spore’s ability to maintain core dehydration during heat treatment. Another factor determining spore core water content is DPA, as DPAless spores have higher core water levels than otherwise

Factors of Special Significance

60 identical DPA-replete spores (152). Presumably, the core volume normally occupied by DPA is occupied by water in DPA-less spores. As noted above, DPA-less spores are more sensitive to moist heat than are DPA-replete spores. Studies of spores from many species have revealed a good correlation between core water content and moist heat resistance over a 20-fold range of D values (9, 47) (Fig. 3.7). However, at the extremes of core water contents, D values vary widely, presumably reflecting the importance of other factors such as sporulation temperature, cortex structure, etc., in modulating spore moist heat resistance (47). One value that is unfortunately missing from analyses of core water content is the amount of core water that is free water. Studies of protein and ion movement in spores have indicated that these molecules are relatively immobile (30, 110), consistent with a dearth of free water in the core. Presumably, the low water content in the spore core causes heat resistance as well as long-term spore survival by slowing water-driven chemical reactions such as DNA depuri-

nation and protein deamidation. A low water content also stabilizes proteins against denaturation by restricting their molecular motion. It would be informative to know the precise amount of water associated with spore core macromolecules in order to calculate the degree of their stabilization by the low core water content.

DNA Repair

DNA repair is essential for spore resistance to UV and chemicals that kill spores by DNA damage. However, spore resistance to moist heat is not altered by loss of DNA repair functions, including those mediated by RecA (110). This is not surprising because moist heat does not kill wild-type spores by DNA damage. However, spore resistance to dry heat, a treatment that does kill spores by DNA damage, is markedly decreased by loss of various DNA repair activities, including RecA and other proteins (110, 152, 184).

SPORE ACTIVATION, GERMINATION, AND OUTGROWTH

Activation

Although spores are metabolically dormant and can remain so for many years, if given the proper stimulus they return to active metabolism within minutes through spore germination (Fig. 3.8). Spore populations often germinate more rapidly and completely if activated prior to germinant addition (76, 117, 151). However, the requirement for activation varies widely among spores of different species and between individual spores in populations (49). A number of agents cause spore activation, including low pH and some chemicals, although the most widely used agent is sublethal heat. The precise changes induced by spore activation are not clear but may involve reversible changes in protein structure (76, 197).

Germination

Figure 3.7  Correlation of spore heat resistance and protoplast (core) water content of lysozyme-sensitive spore types from seven Bacillus species that vary in thermal adaptation and mineralization. The figure is from the work of Gerhardt and Marquis (47) with permission. The numbers refer to spores of various species: 1, G. stearothermophilus; 2, “Bacillus caldolyticus”; 3, Bacillus coagulans; 4, B. subtilis; 5, B. thuringiensis; 6, B. cereus; and 7, Bacillus macquariensis. The letters denote the sporulation temperature or the mineralization of the spores of various species as described in the original publication. doi:10.1128/9781555818463.ch3f7

The precise period encompassed by spore germination, as distinguished from subsequent outgrowth, has been given a number of definitions. For the purposes of this review, we consider germination as those events taking place without the need for metabolic energy. During germination, a dormant spore with a cortex and a large pool of DPA and mineral ions is transformed into a germinated spore in which the cortex has been degraded, DPA and most mineral ions have been excreted, and the core water content has become that of a growing cell (117, 151). However, conversion of the germinated spore into a growing cell requires exogenous nutrients. The initiation of spore germination can be triggered by many compounds, including nucleosides, amino acids,

3.  Spores and Their Significance

61

Figure 3.8  Spore activation, germination, and outgrowth. The events in activation are not known, hence the question mark. The loss of the spore cortex and the hydration and swelling of the core are shown in the germinated spore. The figure is adapted from Fig. 3 in reference 151. doi:10.1128/9781555818463.ch3f8

sugars, salts, Ca2+ with DPA (CaDPA), appropriate PG fragments, and long-chain alkylamines (117, 119, 151, 156). CaDPA and alkylamines are relatively universal germinants. In contrast, the nutrients that act as germinants vary from species to species, and metabolism of such germinants is not required. l-Alanine is a common nutrient germinant, and its action is often inhibited strongly by d-alanine. Interestingly, l-alanine germination may be modulated by a spore-specific alanine racemase that can generate d-alanine (23, 101). The stereospecificity for nutrient germinants is due to these germinants’ interaction with specific nGRs that are generally the products of tricistronic ger operons. Sporeformers usually have between 2 and 7 of these operons that are expressed in the developing forespore during sporulation (117, 119, 139, 151). Different nGRs have different, although sometimes overlapping, specificities for nutrient germinants and can cooperate and even exhibit synergy in responding to germinant mixtures and may all colocalize in the spores’ inner membrane (3, 7, 36a; K. K. Griffiths and P. Setlow, unpublished results). Loss of all nGRs almost completely eliminates the spore’s germination with nutrients. However, in spores of a few species, in particular Clostridium difficile, nGRs have not been definitively identified, yet these spores respond to specific germinants (119, 139) and have been suggested to contain some type of specific germinant receptors (136). The nGRs are probably most often a complex of the three proteins encoded by individual tricistronic ger operons, although variations in this arrangement may exist (119, 139). Two nGR subunits appear likely to be integral membrane proteins, and nGRs are present exclusively in spores’ inner membrane. Unfortunately, how nGRs function is not known, and analysis of the primary sequences of their subunits has not been informative in this regard.

In addition to the genes encoding nGRs, there are other ger genes in B. subtilis in which mutations alter spore germination (117, 119, 151). For some of these additional ger genes their function is known, whereas for others it is not. In general, ger genes identified in B. subtilis have counterparts in other Bacillus species, and genes encoding some of these genes, in particular those encoding nGRs, are also in Clostridium species (119, 139). However, the gerD gene essential for B. subtilis spore germination via nGRs is not present in Clostridium species (119, 151). The earliest biochemical events in spore germination after nutrient germinant addition are releases of protons and other monovalent cations (117, 119, 151). Release of DPA and its divalent cations is next, as is uptake of some water by the spore core, and hence there must be changes in the inner spore membrane’s permeability. However, the mechanism of such changes is not known, although proteins encoded by the spoVA operon may be involved in DPA release and roles for specific antiporters in cation release have been suggested but by no means proven (119). DPA release can be completed in as little as 30 seconds for an individual spore, although individual spores initiate DPA release after very different lag times following germinant addition, lag times that can be minutes, hours, or even longer (82, 123, 195). Spores with extremely long lag times are termed superdormant and are of special concern to the food industry. Such superdormant spores have been isolated from populations of spores of a few Bacillus species, and factors contributing to spores’ superdormancy have been identified (48, 49). In B. subtilis, the replacement of released DPA and ions by water increases the core water content by ~30% (117, 151). This reduces spore moist heat resistance but has little effect on resistance to other agents and is not sufficient for resumption of enzyme

62 action or protein mobility (30). The release of DPA and other ions has been termed stage I of germination, and germination can be halted after stage I by mutations that block degradation of the spore cortex. Initiation of cortex degradation is the next event in germination and allows a rapid approximately twofold increase in spore core volume through water uptake once the cortex that restricts core expansion is removed. This is termed stage II of germination and leads to spore outgrowth. The importance of cortex lysis in germination has focused attention on cortex-lytic (lysozyme-like) enzymes (CLEs). A number of CLEs have been identified from spores of different species including CwlJ, SleB, SleC, SleL, and SleM (117, 119, 151). These enzymes are located in the spore’s outer layers and are active only on PG containing MAL. This is consistent with the lack of cortex degradation during germination of spores of a B. subtilis cwlD mutant, as CwlD is required for MAL formation. CLEs’ specificity prevents these enzymes from hydrolyzing the germ cell wall during spore germination. The function of the CLEs in vivo has been established in spores of several Bacillus species, in which SleB and one or two CwlJ homologs play redundant roles in cortex degradation (119, 147, 151). However, cwlJ sleB spores of Bacillus species complete stage I of spore germination. CwlJ and SleB in spores are in a mature potentially active form and thus must be regulated. While regulation of SleB is not understood, CwlJ activation requires either exposure to CaDPA released in stage I of germination or high concentrations of exogenous CaDPA. In contrast to Bacillus spores, spores of at least two Clostridium species, C. difficile and Clostridium perfringens, contain a single essential CLE, SleC, present in spores as a zymogen that is activated by proteolysis early in germination (119). However, the mechanism of regulation of SleC activation is unknown.

Outgrowth

Following completion of stage II of spore germination, core hydration is equivalent to that of a growing cell protoplast, and protein mobility is restored and enzymatic reactions begin in the core (30, 117, 151). These reactions include utilization of the spore’s depot of 3PGA to generate ATP and NADH, degradation of SASP initiated by GPR, catabolism of some amino acids produced by SASP degradation, and catabolism of exogenous compounds. Carbohydrate metabolism early in outgrowth uses glycolysis and/or the hexose monophosphate shunt but may stop at acetate, as spores often lack enzymes of the TCA cycle. In Bacillus megaterium, endogenous energy reserves can support ATP production during the first 10 minutes of outgrowth.

Factors of Special Significance RNA synthesis begins in the first minutes of outgrowth, using nucleotides stored in the spore or generated by breakdown of spore RNA. The RNAs made at this time have been studied in B. subtilis and are composed of many mRNAs (75), and protein synthesis begins shortly after RNA synthesis. All components of the protein-synthesizing machinery stored in the dormant spore appear functional, with the exception of some tRNA lacking the 3¢-terminal adenosine residue that is repaired by tRNA nucleotidyltransferase early in outgrowth. Endogenous amino acids derived largely from SASP breakdown can support most protein synthesis in the first 25 minutes or so of outgrowth, but completion of outgrowth requires exogenous nutrients (117, 151). The spore regains the ability to synthesize amino acids, nucleotides, and other small molecules due to synthesis of biosynthetic enzymes at defined times in outgrowth, but regulation of gene expression during outgrowth has not been well studied. DNA replication is not generally initiated until late in outgrowth, although DNA repair can occur well before DNA replication, as spores contain deoxynucleoside triphosphates by the first minutes of outgrowth (117, 151). However, the importance of DNA repair in the first minutes of outgrowth, other than repair of UV damage, has not been well studied. During spore outgrowth, the volume of the outgrowing spore continues to increase, requiring the synthesis of membrane and cell wall components. One question that remains intriguing is whether there are genes that are needed for outgrowth but no other stage of growth. A number of mutations affecting outgrowth (termed out mutants) have been isolated in B. subtilis, and most have been identified (117). While the functions of some of these genes have been established, to date none has been shown to function solely in outgrowth.

PRACTICAL PROBLEMS OF SPORES IN THE FOOD INDUSTRY Spore-forming bacteria and heat-resistant fungi pose specific problems for the food industry. Three species of sporeformers, C. botulinum, C. perfringens, and B. cereus, are well known to produce toxins that can cause illness in humans and animals (51, 53, 58, 186), and many species of sporeformers cause spoilage of foods (65, 66, 104, 167). Some strains of Clostridium butyricum and Clostridium baratii can also produce botulinal neurotoxin, but these bacteria are rare, and their spores appear to have a lower heat resistance than their nontoxigenic counterparts (58, 68, 69). Sporeformers causing foodborne illness and spoilage

3.  Spores and Their Significance are particularly important in low-acid foods (equlibrium pH, ³4.6) packaged in cans, bottles, pouches, or other hermetically sealed containers (“canned” foods), which are processed by heat (167). Certain sporeformers also cause various types of spoilage of high-acid foods (equlibrium pH, <4.6) (13, 104). Psychrotrophic sporeformers have been increasingly recognized to cause spoilage of refrigerated foods (2, 12, 20, 73, 85, 106, 108, 188). Fungi that produce heat-resistant ascospores are an important cause of spoilage of acidic foods and beverages such as fruit and vegetable products (131, 177). Fungal spores generally are more sensitive to heat and chemical compounds than bacterial endospores (131, 177). Food spoilage by spore-forming bacteria was discovered by Pasteur during investigation of butyric acid fermentation in wines (14). Pasteur was able to isolate a bacterium he termed Vibrion butyrique, which is probably the same microorganism now classified as C. butyricum, the type species of the genus Clostridium (19). Endospores were discovered independently by Ferdinand Cohn and Robert Koch in 1876, soon after Pasteur had made microbiology famous (77). Investigations by Pasteur and Koch led to the association of microbial activity with the safety and quality of foods. During investigation of the anthrax bacillus, B. anthracis, the famous Koch’s postulates were born; these stated that a disease is caused by a specific microorganism and also led to the development of pure culture techniques. Diseases and spoilage problems caused by sporeformers have historically been associated with foods that are thermally processed, because heat selects for survival and subsequent growth of spore-forming microorganisms. The process of appertizing, or preserving food sterilized by heat in a hermetically sealed container, was invented in the late 1700s by Appert (5), who believed that the elimination of air was responsible for the long shelf lives of thermally processed foods. The empirical use of thermal processing gradually developed into modern-day thermal processing industries. In the late 1800s and early 1900s, several North American and European scientists were instrumental in developing scientific principles to ensure the safety and prevent spoilage of thermally processed foods (52). As a result of these studies, thermal processing of foods in hermetically sealed containers became an important industry (133, 141). In early studies of canned food spoilage, Prescott and Underwood (133) and Russell (141) revealed that endospore-forming bacilli caused the spoilage of thermally processed clams, lobsters, and corn. The classic studies of Esty and Meyer (41) provided definitive values of the heat resistance of C. botulinum type A and B spores and

63 helped rescue the U.S. canning industry from its near demise due to botulism caused by commercially canned olives and other foods. Quantitative thermal processes, understanding of spore heat resistance, aseptic processing, and implementation of HACCP (hazard analysis and critical control point) concepts also resulted from developments in the canned food industry.

Low-Acid Canned Foods

The U.S. Food and Drug Administration (FDA) and the U.S. Department of Agriculture Food Safety and Inspection Service (USDA-FSIS) define a low-acid canned food as one with a finished equilibrium pH of ³4.6 and a water activity (aw) of ³0.85. The regulations regarding thermal processing of canned foods are described in the U.S. Code of Federal Regulations (21 CFR, parts 108 to 114). The USDA has regulatory oversight of products that contain at least 3% raw red meat or 2% cooked poultry, and the FDA regulates other products. A description of the thermal process, including the environment within the facility, equipment specifications, and formulations, must be filed with the proper governmental agency prior to commercial processing. The processor must report process deviations or instances of spoilage to the FDA or USDA. Low-acid canned foods are packaged in hermetically sealed containers, often cans or glass jars but also plastic pouches and other types of containers. These “cans” (as collectively defined in this chapter) containing low-acid food products must be processed by heat to achieve commercial sterility (126, 142), which is a condition achieved by application of heat that inactivates microorganisms of public health significance, as well as any microorganisms of non-health significance capable of reproducing in the food under normal nonrefrigerated conditions of storage and distribution. Commercial sterility is an empirical term to indicate a low level of microbial survival and provision of shelf stability (126, 142). Preservation procedures such as acidification or lowering water activity by brining or other means can be used to attain commercial sterility. Chemical preservation procedures are often combined with a reduced heat treatment (51). Newer preservation procedures that are increasingly being adapted by industry include high-pressure processing, pulsed-field electric fields, pulsed X ray, pulsed light, UV radiation, ionizing radia­tion, plasma, and ultrasonics (107, 134, 173, 189, 191–193). Generally, these methods are effective in inactivating vegetative cells but are relatively ineffective compared to heat in killing spores. Newer developments in food preservation (134) have also renewed interest in the mechanisms of spore inactivation.

64 In low-acid canned foods, the primary goal is to inactivate spores of C. botulinum because this bacterium has the highest heat resistance of microbial foodborne pathogens. The degree of heat treatment applied varies considerably according to the class of food, the spore numbers, pH, storage conditions, and other factors. For example, canned low-acid vegetables and uncured meats usually receive a 12D process or “botulinum cook” (see below). Lesser heat treatments are applied to shelf-stable canned cured meats as well as to foods with reduced water activity or other antimicrobial factors that inhibit growth of spore-forming bacteria. In practice, the spore content of ingredients and the cleanliness of the cannery environment are of major importance for successful heat treatment of low-acid canned foods (104). Certain foods and food ingredients such as mushrooms, potatoes, spices, sugars, and starches may contain large numbers of C. botulinum and spores of other sporeformers, which may be monitored for their spore input to a process. Honey and certain other foods can harbor C. botulinum spores and should not be fed to infants less than one year of age because only a few spores may be sufficient to cause infant botulism (170). Researchers in the early and mid-1900s described thermal processes to prevent botulism and spoilage in canned foods (reviewed in references 52 and 161). Pflug (125–127) refined the semilogarithmic microbial destruction model and designed a strategy to achieve the required heat process FT value. He also introduced the concept of the probability of a nonsterile unit on a one-container basis. This reasoning logically explains the traditional 12D term, which designates the time required in a thermal process for a 12-log reduction of C. botulinum spores. In thermal processing, two values have historically been used to describe the thermal inactivation of an organism: the D value is the time required for a 1-log reduction of a microorganism, and the z value is the temperature increment required to change the D value by a factor of 10 or 1 log10 (167). While the D value represents the resistance of a microorganism to a specific temperature, the z value represents the relative resistance of a microorganism to inactivation at different temperatures (95, 167). The thermal process for the 10−11 to 10−13 level of probability of a botulism incident occurring will depend on the initial number of C. botulinum spores present in a container. This value can be quite high, for example, 104 for a container of mushrooms, or very low, such as 10−1 or less for a container of a meat product (54). Pflug (125–127) emphasized that the thermal processing industry should prioritize the process design to protect against (i) public health hazard (botu-

Factors of Special Significance lism) from C. botulinum spores, (ii) spoilage from mesophilic spore-forming microorganisms, and (iii) spoilage from thermophilic microorganisms in containers stored in warm climates or environments. Generally, low-acid foods in hermetically sealed containers are heated to achieve a temperature of 121°C (250°F) or equivalent at the center or the most heat-impermeable region of a food for 3 to 6 minutes. This ensures inactivation of the most heat-resistant C. botulinum spores with a D121°C(250°F) of 0.21 min and a z of 10°C (18°F). Economic spoilage is also avoided by achieving ca. 5D killing of mesophilic spores that typically have a D121°C of ca. 1 min (Table 3.4). Foods that are distributed in warm climates of tropical or desert areas require a particularly severe thermal treatment of ca. 20 min at 121°C to achieve a 5D killing of Thermoanaerobacterium (Clostridium) saccharolyticum, Geobacillus stearothermophilus, and Desulfotomaculum nigrificans because these organisms have a D­121°C of ca. 3 to 4 min (Table 3.4). Such severe treatment can have a detrimental impact on nutrient content and organoleptic qualities, but it ensures a shelf-stable food. Predictive models have been outlined to simulate growth of proteolytic C. botulinum during cooling of heated foods (24, 71). The concept of semilogarithmic or first-order kinetics of heat inactivation of spores and the use of D and z values have been carefully reevaluated (121, 122). It was concluded that populations of spores often differ in heat sensitivities and that non-log-linear death kinetics more accurately explain spore inactivation. Analysis using non-log kinetics can explain the tailing and curving that are commonly observed in heat inactivation of spore populations. The kinetics of heat inactivation of spore populations closely followed a cumulative Weibull distrubition, i.e., log S = b(T)tn(T), where S is the survival ratio and and b(T) and n(T) are temperature-dependent coefficients (122). Kinetic models for inactivation and recovery of bacterial spores have been described and have been valuable in evaluating and developing improved procedures within the food industry (24, 61, 72). Heat treatments are commonly applied to aseptically processed low-acid canned foods, whereby commercially sterilized cooled product is filled into presterilized containers followed by aseptic hermetic sealing with a closure in a sterile environment. This technology was initially used for commercial sterilization of milk and creams in the 1950s and then encompassed other food products such as soups, eggnog, cheese spreads, sour cream dips, puddings, and high-acid products such as fruit and vegetable drinks. Aseptic processing and packaging systems have the potential to reduce energy, packaging, and distribution costs (95).

3.  Spores and Their Significance

65

Table 3.4  Heat resistance of sporeformers of importance in foodsa Approx D valueb (min) at: Bacterium

80°C

85°C

90°C

95°C

100°C

110°C

120°C

Sporeformers of public health significance

 

 

 

 

 

  Group I C. botulinum types A and B

50

 

7–30

1–3

0.1–0.2

1–30

0.1–3

0.03–2

 

 

 

0.01

 

 

3–200

0.03–2.4

 

  Group II C. botulinum type B   C. botulinum type E

0.3–3

  B. cereus

 

3–145

0.3–18

2.3–5.2

 

 

 

 

 

 

  B. subtilis

7–70

6.9

0.5

  B. licheniformis

13.5

0.5

 

  B. megaterium

1

 

 

0.1–0.5

 

 

  C. perfringens Mesophilic aerobes

 

 

  Bacillus polymyxa

4–5

  Bacillus thermoacidurans

11–30

2–3

 

 

  Alicyclobacillus acidoterrestris

16

2.6

 

 

 

 

 

 

 

 

Thermophilic aerobes

 

 

  G. stearothermophilus

100–1,600

1–6

  B. coagulans

20–300

2–3

Mesophilic anaerobes

 

 

 

 

 

 

 

  C. butyricum

4–5

0.4–0.8

 

 

 

 

 

80–100

21

0.1–1.5

  Clostridium sporogenes   Clostridium tyrobutyricum

13

 

 

 

 

 

 

Thermophilic anaerobes

 

 

 

 

 

 

 

<480

  D. nigrificans   T. thermosaccharolyticum

400

2–3 3–4

 

Sources: see the text for references. b pH, ca. 7; aw, >0.95. a

Spores of C. botulinum can survive for long durations in high-acid foods with pH values of £4.6 (51). Growth of C. botulinum and outbreaks of botulism in high-acid foods have been reported. Some outbreaks from consumption of high-acid foods were associated with inadvertent increases in pH, such as by growth of molds with catabolism of organic acids (51, 68).

Bacteriology of Sporeformers of Public Health Significance

Three species of sporeformers, C. botulinum, C. perfringens, and B. cereus, are well known to cause foodborne illness (Table 3.5). Certain other species of Bacillus such as Bacillus licheniformis, B. subtilis, and Bacillus pumilus have also been reported to sporadically cause foodborne disease through production of toxins (53),

and rare strains of C. butyricum and C. baratii produce type E and F botulinal toxins, respectively (58, 69). Devastating incidences of intestinal anthrax caused by ingestion of contaminated raw or poorly cooked meat have been reported (56). Spores of B. anthracis have recently been of concern as agents of bioterrorism (40, 64). The principal microbial hazard in heat-processed foods and in minimally processed refrigerated foods is C. botulinum. The genus Clostridium consists of grampositive, anaerobic, spore-forming bacilli that obtain energy by fermentation (19, 35, 58, 69). The species C. botulinum is a heterogenous collection of strains that differ widely in genetic relatedness and phenotypic properties but all have the property of producing a characteristic neurotoxin of extraordinary potency (69,

Factors of Special Significance

66 Table 3.5  Growth requirements of sporeformers of public health significancea Inhibitory condition Bacterium Group I C. botulinum Group II C. botulinum B. cereus C. perfringens a

Minimum pH

NaCl concn (%)

Minimum aw

Temp range for growth (°C)

4.6

10

0.94

10–50 3.3–45

5.0

5

0.97

4.35–4.9

ca.10

0.91–0.95

5–50

5.0

ca. 7

0.95–0.97

15–50

Sources: see the text for references.

171). C. botulinum and other pathogenic bacteria produce spores that swell the mother sporangium, giving a spindle or club-shaped appearance (Fig. 3.9). Spores of C. botulinum types B and E frequently possess an exosporium (114, 165), and type E characteristically produces appendages (Fig. 3.10). C. botulinum is commonly divided into four physiological groups (I through IV) on the basis of phenotypic properties (44, 69, 120). Group I (strongly proteolytic strains producing neurotoxin types A, B, and F) and group II (nonproteolytic strains producing neurotoxin serotypes B, E, and F) are the two groups of concern in food safety (44, 51, 68, 120, 158). Strains in groups I and II have certain properties that affect their ability to grow in foods. The spores of group II have considerably less heat resistance than group I spores (44, 68, 72), but they can grow and produce toxin at refrigeration temperature. The ability of C. botulinum to grow at low temperatures has raised concern that refrigerated food products could lead to botulism outbreaks (44, 68, 94, 97, 120). It has been estimated that when conditions are optimal, a 1,000-fold increase in cell numbers from a single or low-number spore inoculum can be achieved at 25°C in 10.4 hours (120). The minimum temperature for growth and neurotoxin formation has been a matter of debate and has been reported as 3.0 to 3.3°C in 5 to 7 weeks (120). Most strains of C. botulinum grow very slowly below 5°C. It has been recommended for refrigerated products with a shelf life of greater than 5 days and that receive a

Figure 3.9  Transmission electron micrograph (×50,000) of a longitudinal section through a spore and sporangium of C. botulinum type A, showing the characteristic club-shaped morphology. doi:10.1128/9781555818463.ch3f9

heat treatment killing <6 log10 of psychrotrophic spores of C. botulinum that additional intrinsic preservation factors should be included to ensure their safety from botulinum toxin formation (51, 188). The genome sequences of ca. 20 strains of C. botulinum are available in GenBank. The properties garnered from the genome sequences related to food safety and botulism are described in the chapter on C. botulinum (chapter 17). With respect to the sporulation process, genome analyses have revealed that two components of the Bacillus phosphorelay signal transduction pathway, Spo0F and Spo0B, are absent from essentially all Clostridium species, indicating that the initiation of sporulation occurs through direct phosphorylation of the Spo0A transcription factor by sensor histidine kinases. Moreover, there is no conservation of amino acid sequence or structure for sensor domains of these kinases, indicating that each species may respond to different initiating signals (164, 181, 187). Bioinformatic analyses of the genome sequence should also reveal characteristics underlying spore activation, germination, spore resistance, and properties governing growth and toxin formation in foods (120). Genomic analyses potentially could be used to identify novel targets for growth inhibition (120). C. perfringens is widespread in soils and is a normal resident of the intestinal tracts of humans and certain animals (58, 160). C. perfringens can grow extremely rapidly in high-protein foods such as meats that have been cooked to eliminate competitors and are inadequately cooled, allowing it to produce in the intestinal tract an enterotoxin that causes diarrheal disease (chapter 18). It also produces a variety of other extracellular toxins and degradative enzymes, but these are mainly of significance in gas gangrene and diseases in animals (157, 160). C. perfringens differs from many other clostridia in being nonmotile, reducing nitrate, and carrying out a stormy fermentation of lactose in milk. Contributing to the ability of C. perfringens to cause foodborne illness is its ubiquitous distribution in foods and food environments, the formation of resistant endospores that sur-

3.  Spores and Their Significance

Figure 3.10  Electron micrographs of C. botulinum type B (A) and E (B) showing characteristic exosporium in types B and E and appendages in type E. Micrographs courtesy of Philipp Gerhardt from spores produced in E.A.J.’s laboratory. doi:10.1128/9781555818463.ch3f10

vive cooking of foods, and an extremely rapid growth rate in warm foods (6 to 9 minutes at 43 to 45°C) (chapter 18). Sporulation of C. perfringens is often difficult to obtain in many laboratory media and in many foods, and Duncan-Strong or related complex media are generally used. When spores do occur, they are large, oval, and centrally or subterminally located and swell the cells. The optimum temperature for growth of vegetative cells is ca. 43 to 45°C (109 to 113°F), and in rich media or in certain foods at the optimum temperature for growth, doubling times as short as 6 to 9 minutes have been observed. Due to the rapid growth of C. perfringens in many foods, cell numbers sufficient to cause foodborne illness can be produced rapidly (chapter 18). Germination of spores and growth of cells can occur at up to 50°C (ca. 122°F). C. perfringens does not generally grow below 15°C (68°F), and true psychrotrophic strains of the bacterium have not been isolated. C. perfringens can grow over a pH range of 5.0 to 9.0 or 8.5, and the optimum is ca. 6.5 (chapter 18). At temperatures below 45°C (113°F), some strains will grow at pH 5. As with most bacteria, organic acids such as acetate, lactate, and citrate are much more effective than are mineral acids for inhibiting growth of C. perfringens. Growth of most strains is also inhibited by 5 to 6% sodium chloride (Table 3.5). Sodium nitrite in cured meat products also inhibits growth. Conditions for growth of C. perfringens in many foods have been reviewed (chapter 18). The heat resistance of C. perfringens spores is strain dependent and varies considerably. In general, two

67 classes of heat sensitivity are common. Heat-resistant spores have D90°C (D194°F) values of 15 to 145 min and z values of 9 to 16°C (16 to 29°F), compared to heatsensitive spores which have D90°C of 3 to 5 min and z of 6 to 8°C (11 to 14°F). The spores of the heat-­resistant class generally require a heat shock of 75 to 100°C (167 to 212°F) for 5 to 20 min in order to germinate. The basis of the wide variation in heat resistance is not currently understood. The spores of both classes may survive cooking of foods and may be stimulated by heat shock for germination during the heating procedures. Both classes can cause diarrheal foodborne illness, although it would be expected that the heat-resistant forms would be a more frequent cause of illness. Foodborne illness caused by C. perfringens nearly always involves temperature abuse of a cooked food, and most foodborne illnesses caused by C. perfringens could be avoided if cooked foods were eaten immediately after cooking or rapidly chilled and reheated before consumption to inactivate vegetative cells. The objective in prevention of foodborne illnesses is to limit the multiplication of vegetative cells in the food. Since the spores are widespread and are resistant to heat, they will often survive the cooking procedure, germinate, and rapidly outgrow to large vegetative cell populations if the rate of cooling is inadequate. This property has led to USDA performance standards for cooling in the production of certain ready-to-eat meat and poultry products. The USDA-FSIS draft guidelines state that cooked meat products should be cooled at a rate adequate to prevent a 1-log10 increase of C. perfringens (176). If foods are not rapidly cooled, they should be held at 60°C (140°F) or higher. For cooled foods, the maximum internal temperature should not remain between 54.4°C (130°F) and 26.7°C (80°F) for more than 1.5 hours or between 26.7°C (80°F) and 4.4°C (40°F) for more than 5 hours. For products that receive a pasteurization step, the entire process must not allow greater than 1 log10 growth in the product during processing and cooling. Predictive models have been proposed for evaluation of C. perfringens growth and inactivation (176). The genus Bacillus contains only two species, B. anthracis and B. cereus, that are recognized as definitive human pathogens. B. cereus can produce a heat-labile enterotoxin causing diarrheal illness and a heat-stable toxin giving an emetic response in humans (chapter 19). Generally, the bacterium must grow to very high numbers (>106 cells/g of food) to cause human illness. B. cereus is closely related to B. megaterium, B. thuringiensis, and B. anthracis, but B. cereus can be distinguished from these species by biochemical tests and the absence

68 of toxin crystals. Other bacilli, including B. licheniformis, B. subtilis, and B. pumilus, have been reported to cause foodborne disease outbreaks, primarily in the United Kingdom (53). B. cereus spores occur widely in foods and are commonly found in milk, cereals, starches, herbs, spices, and other dried foodstuffs (100). They are also frequently present on the surfaces of meats and poultry, probably because of soil or dust contamination. Investigators in Sweden reported isolation of B. cereus from 47.8% of 3,888 different food samples. In the United Kingdom, B. cereus was isolated from 98/108 (91%) of rice samples. The bacterium causes spoilage of raw and unpasteurized milk (20), and foods containing dried milk, such as infant formulas, may possess fairly high levels of spores or cells. B. cereus grows over a temperature range of approximately 10° to 48°C (50° to 118.4°F), with an optimum of 28° to 35°C (82.4° to 95°F) (Table 3.5). Psychrotrophic strains that produce enterotoxin in milk have been isolated (182). The doubling time at the optimum temperature in a nutritious medium is 18 to 27 minutes. Several strains can grow slowly in sodium chloride concentrations of 7.5%. The minimum water activity for growth is 0.95. The bacterium grows over a pH range of approximately 4.9 to 9.3, but these environmental limits for growth are dependent on water activity, temperature, and other interrelated parameters. Spores of B. cereus are ellipsoidal and central to subterminal and do not distend the sporangium. Spore germination can occur over the temperature range of 8 to 30°C (46.4 to 86°F). Spores from strains associated with foodborne illness had a heat resistance of D95°C (203°F) of ca. 24 minutes. Other strains were determined to have a wider range of heat resistance. It has been suggested that strains involved in foodborne illness have higher heat resistances and therefore are more apt to survive cooking. B. cereus spores are hydrophobic and attach to food contact surfaces (63). Since B. cereus is widespread in nature and survives extended storage in dried food products, it is not practical to eliminate low numbers of spores from foods. Control against foodborne illness should be directed at preventing germination of spores and preventing multiplication of large populations of the bacterium. Cooked foods should rapidly and efficiently be cooled to less than 7°C (45°F) or maintained above 60°C (140°F) and should be thoroughly reheated before serving. Other factors affecting food safety and spoilage by B. cereus are described in chapter 19. B. anthracis has rarely been associated with gastrointestinally mediated anthrax (64), generally resulting

Factors of Special Significance from occupational exposure such as in tanning of hides, but gastrointestinal anthrax has also been reported from consumption of spore-contaminated meat (40, 64). In late 2001, the deliberate release of B. anthracis spores revealed that the pathogen could also cause disease and death as a bioterrorist agent, including by dissemination in foods (40, 64). The resistance properties of spores of B. anthracis have been reviewed and should provide a framework for control (40, 138, 162), although little is known regarding these spores’ resistance in various foods.

Heat Resistance of C. botulinum Spores

Group I C. botulinum type A and B strains can produce spores of remarkable heat resistance and are the most important sporeformers in public health safety of thermally processed foods. The classic investigation on the heat resistance of C. botulinum spores was carried out by Esty and Meyer (41) as a result of commercial outbreaks of botulism caused by consumption of canned olives and certain other canned vegetables. They examined 109 C. botulinum type A and B strains at five heating temperatures over the range 100° to 120°C (212° to 248°F). They determined that the inactivation rate was logarithmic between 100° and 120°C and depended on the spore concentration, the pH, and the heating menstruum. They also determined that 0.15 M phosphate buffer (Sorensen’s buffer), pH 7.0, gave the most consistent heat inactivation results, and their use of a standardized system enables comparisons of heat resistance by researchers today. The use of a reproducible system is valuable in periodically determining the heat resistance of new spore crops. Extrapolating the data of Esty and Meyer gives a maximum value for D121.1°C of 0.21 min for C. botulinum type A and B spores in phosphate buffer (178). The thermal processing industries have used D121°C as a standard in calculating process requirements. Proteolytic C. botulinum type F spores have a heat resistance D98.9°C of 12.2 to 23.2 min and D110°C of 1.45 to 1.82 min (44, 69, 97, 120), which is much lower than that of type A spores. Spores of nonproteolytic C. botulinum types B and E have much lower heat resistance than proteolytic type A and B strains. C. botulinum type E spores have a D70°C(158°F) value ranging from 29 to 33 min and a D80°C(176°F) value from 0.3 to 3 min, depending on the strains. The z value ranged from 13 to 15°F. These values are comparable to those reported by Ohye and Scott (114), who obtained D80°C values of 3.3 and 0.4 min for spores of two type E strains. Spores of nonproteolytic C. botulinum type B have heat resistance considerably higher than that of type E spores. Scott and Bernard (143) determined that the D82.2°C value of spores

3.  Spores and Their Significance of nonproteolytic type B strains ranged from 1.5 to 32.3 min compared with a D82.2°C value of 0.33 min for a type E strain. Media containing lysozyme can significantly enhance recovery of group II C. botulinum spores because lysozyme substitutes for spore-lytic enzymes that are inactivated by heat (94). D values at 85° and 95°C were 100 and 4.4 min, respectively, for spores of strain 17B and 45.6 and 2.8 min, respectively, for spores of strain Beluga E on medium containing lysozyme. The thermal resistance of C. botulinum spores is strongly dependent on environmental and recovery conditions. Heat resistance is markedly affected by acidity (72). Esty and Meyer determined that spores had maximum resistance at pH 6.3 and 6.9, and resistance decreased markedly at pH values below 5 or above 9. Increased levels of sodium chloride or sucrose and decreased aw increase the heat resistance of C. botulinum spores (168). Sugiyama (169) determined that spores grown in media containing fatty acids increased their heat resistance. C. botulinum spores coated in oil were more resistant to heat (169). In common with Bacillus spp., sporulation of C. botulinum at higher temperatures results in spore crops with greater heat resistance (179). Little is known regarding the compositional factors contributing to heat resistance of C. botulinum spores. The metal composition of purified group I C. botulinum spores is probably markedly different from that of spores of Bacillus (78–80). The minerals required for sporulation and mechanisms of heat resistance of C. botulinum and Bacillus spp. may be different (78, 79). Unlike Bacillus spp., which require manganese for sporulation, C. botulinum type B sporulation was enhanced by zinc and inhibited by copper (78). During sporulation, C. botulinum accumulated relatively high concentrations of transition metals, particularly zinc (ca. 1% of cell dry weight) and iron and copper (0.05 to 1%), and iron accumulation can enhance DNA’s sensitivity to mutagenesis (91). Spores containing increased contents of iron or copper were more rapidly inactivated by heat than were native spores or spores containing increased manganese or zinc (79). Spores of C. botulinum groups I and II are highly resistant to g-irradiation compared with vegetative cells of most microorganisms, and it is probably not practical to apply irradiation to foods to inactivate these spores. C. botulinum spores have a D value of 0.1 to 0.45 megarads (2.0 to 4.5 kGy). C. botulinum spore resistance to g-­radiation varies depending on the C. botulinum type; proteolytic types A, B, and F are most resistant (81). C. botulinum spores are also highly resistant to ethylene oxide but are inactivated by halogen sanitizers and by hydrogen peroxide (81). Hydrogen peroxide is commonly used

69 for sanitizing surfaces in aseptic packaging, and halogen sanitizers are used in cannery cooling waters. Alternatives to hydrogen peroxide such as peracetic acid are receiving renewed interest due to the deleterious effects of hydrogen peroxide on packaging equipment and materials (11, 81). Assessments of biocides and food preservatives in sporicidal efficacy have been reviewed (81, 140).

Incidence of Foodborne Illness Caused by C. botulinum

The epidemiology of botulism has been thoroughly reviewed (44, 51, 69). Fortunately, the incidence of botulism caused by commercial foods is very low. It has been estimated that about 30 billion cans, bottles, and pouches of low-acid foods are consumed in the United States each year. From 1940, when heat­processing principles were firmly established, through 1975, fewer than 10 botulism outbreaks and fewer than four deaths were caused by inadequately commercially canned foods in the United States (96, 109). From 1971 through 1982, however, botulinum toxin was detected in several commercial canned foods such as mushrooms, salmon, soups, peppers, tuna fish, beef stew, and tomatoes. Survival of spores and toxin production during this period was caused mainly by underprocessing or by container leakage following processing. The detection of botulinal toxin in canned mushrooms and in canned salmon in the 1970s and 1980s prompted U.S. regulatory agencies to recommend that chlorine or sanitizers be used in cooling water. Botulism has been transmitted via several commercial low-acid “canned” foods, including chopped garlic in oil, cheese sauce, bean dip, and clam chowder (51, 69). These incidences resulted from temperature abuse of products labeled “keep refrigerated” and the absence of inhibitory conditions other than temperature. C. botulinum is a primary safety concern in process cheese products (51, 175). These products are pasteurized and have a low level of competitive microbiota, and their safety is controlled primarily by aw, pH, and the presence of phosphate salts (175). With the trend for shelf-stable process cheese products and nonstandard sauces, C. botulinum has continued to be a major concern in this industry.

Ready-To-Eat Meat and Poultry Products

Although botulism from commercial products has been quite rare due to excellent control during thermal processing, botulism from tainted home-canned and improperly fermented products has occurred sporadically throughout the world (44, 67, 68). The heat resistance of C. botulinum spores is often not understood by home canners, and 20 to 30 botulism cases occur each year

70 in the United States with a current case-fatality rate of about 10%. Botulism is more common in some countries such as Poland and China, where improper home canning of meats and poor fermentation of soybean curd occur relatively frequently (68). During the 1980s and early 1990s, the United States and certain other countries observed a resurgence of botulism in restaurantprepared foods, most often caused by poor temperature control of the prepared foods (51, 68). Inadvertent temperature abuse of foods has resulted in botulism outbreaks. Botulism has occurred from potatoes that were wrapped in foil, baked, and then held at room temperature until they were used for preparing salads (51, 172). During cooking, vegetative organisms are killed, but the spores of C. botulinum survive and grow in the anaerobic environment created by wrapping the potatoes in foil (51). In April 1994, an outbreak of botulism in Texas affected 23 individuals, 17 of whom were hospitalized (44, 63). This was the largest botulism outbreak in the United States since 1983. The food vehicle was a potato-based dip (skordalia), which was prepared using foil-wrapped potatoes that were left at room temperature after baking. These examples illustrate ways in which changes in food processing, elimination of antimicrobials, and relying solely on refrigeration can result in incidences of botulism. Changes in formulation, processing, or packaging of food products can lead to botulism. To assure the botulinogenic safety of a food with potential of supporting C. botulinum growth, it is recommended that laboratory challenge tests be performed. Guidelines have been recommended for such challenge studies (37, 51, 112, 172).

HACCP, FSO, SIAFE, and Prevention of Foodborne Disease by Sporeformers

The safety of thermally processed low-acid foods is enhanced by application of risk management programs including HACCP, Food Safety Objectives (FSO), and Stepwise and Interactive Evaluation of Food Safety by an Expert System (SIAFE), among others (24, 124, 135, 166). These systems entail a systematic and quantitative risk assessment program to ensure the safety of foods. They were designed to have strict control over all aspects of the safety of food production including raw materials, processing methods, the food plant environment, personnel, storage, and distribution. In practice, the identification of potential hazards for a given process and meticulous control of critical control points are required. Methods for HACCP and FSO quality assurance programs for thermally processed foods have been described (24, 124, 135, 166).

Factors of Special Significance SPOILAGE OF FOODS BY SPOREFORMERS Bacteria and fungi cause enormous economic losses of foods by spoilage. It has been estimated that 20 to 30% of commercial foods undergo microbial spoilage (161). With the global increase in population and food consumption, technologies to prevent spoilage would greatly help to alleviate food shortages and spoilage and contribute to food security. Variation and physiological states of sporeformers contribute to their ability to grow and spoil foods (39). Spore-forming bacteria and fungi are responsible for spoilage of several classes of foods.

Spoilage of Acid and Low-Acid Canned and Vacuum-Packaged Foods by Sporeformers

Thermally processed low-acid foods receive a heat treatment adequate to kill spores of C. botulinum but not sufficient to kill more-heat-resistant spores of mesophiles and thermophiles. The sporeformers of major importance are species of Bacillus, Geobacillus, Clostridium, Thermanaerobacterium, and Alicyclobacillus (65, 161) (Table 3.6). The resistance properties of the spores enable them to survive pasteurization and certain other processes that inactivate vegetative cells and subsequently to grow in certain foods and cause spoilage. The principal spoilage microorganisms and spoilage manifestations are presented in Table 3.6. The principal classes of sporeformers causing spoilage are thermophilic flat-sour organisms, thermophilic anaerobes not producing hydrogen sulfide, thermophilic anaerobes forming hydrogen sulfide, putrefactive anaerobes, facultative Bacillus mesophiles, butyric clostridia, lactobacilli, and heat-resistant molds and yeasts (13, 104). Spoilage sporeformers have certain physiological properties that influence their propensity to spoil foods. Bacillus species are aerobic and generally not osmotolerant (161). Some species can be psychrotrophic, including B. cereus, which can spoil refrigerated dairy products; others such as B. subtilis are mesophilic and can spoil bakery products, whereas others such as G. stearothermophilus are thermophilic and spoil foods that are canned or in hermetically sealed packages. Clostridia range from being strict anaerobes to aerotolerant, and most species are not osmotolerant, although Clostridium sporogenes can grow at an aw of ³0.93. Clostridia generally spoil foods of low oxygen/reduction potential such as canned or vacuumpackaged foods. However, the microbiota of a food can lower the oxygen/reduction potential, allowing clostridia to proliferate. The main species causing spoilage are C. sporogenes, C. butyricum, and T. saccharolyticum. Practical control of these bacteria includes monitoring

3.  Spores and Their Significance of raw foods entering the cannery, particularly sugars, starches, spices, onions, mushrooms, and dried foods, to limit the initial spore load in a food product, adequate thermal processing depending on subsequent storage and distribution conditions, rapid cooling of products, chlorination of cooling water, and implementing and maintaining good manufacturing practices within the food plant. Psychrotrophic strains of Bacillus have been isolated from spoiled dairy products (20). Acid and acidified foods with an equilibrium pH of £4.6 are not processed sufficiently to inactivate all spores, because most species of sporeformers do not grow under acidic conditions and inactivation of all spores would be detrimental to food quality and nutritional composition. Certain foods such as cured meats and hams do not receive a thermal process sufficient to inactivate sporeformers and hence must be kept under refrigerated conditions for microbial stability. These classes of foods present opportunities for the growth of sporeformers that do not present a public health hazard but can cause economic spoilage (13, 104). Most of the spoilage bacteria and their characteristic spoilage patterns have been recognized for many years (104, 167). Alicyclobacillus species are highly acid tolerant and grow at pH values of 2.0 to 6.0. They are also moder-

71 ately thermophilic but not osmotolerant, growing above an aw of 0.98. They have been recognized to spoil acidic products such as pasteurized fruit or vegetable juices improperly cooled or stored at high temperatures (104). Some heat-resistant fungi can spoil acidic foods, particularly fruit products (131, 177). While most filamentous fungi and yeasts are killed by heating for a few minutes at 60 to 75°C, heat-resistant fungi produce thick-walled ascospores that survive heating at ca. 85°C for 5 minutes. The most common genera of heat-resistant fungi causing spoilage are Byssochlamys, Neosartorya, Talaromyces, and Eupenicillium (131, 177). Certain heat-resistant fungi also produce toxic secondary metabolites collectively known as mycotoxins (131, 177). To prevent spoilage of heat-treated foods, raw materials should be screened for heat-resistant fungi and strict good manufacturing practices and sanitation programs should be followed during processing. Additionally, manipulation of aw and oxygen tension and the application of antimycotic agents can be used to prevent fungal growth. In practice, the inherent spore contamination of foods and food ingredients and of the cannery environment contributes to spoilage problems. Dry ingredients such as sugar, starches, flours, and spices often contain

Table 3.6  Spoilage of canned foods by sporeformersa Type of spoilage

pH

Major sporeformers responsible

Spoilage defects

Flat-sour

³5.3

B. coagulans, G. stearothermophilus

No gas, pH lowered. May have abnormal odor and cloudy liquor.

Thermophilic anaerobe

³4.8

T. thermosaccharolyticus

Can swells, may burst. Anaerobic anaerobe and products give sour, fermented, or butyric odor. Typical foods are spinach, corn.

Sulfide spoilage

³5.3

D. nigrificans, Clostridium bifermentans

Hydrogen sulfide produced, giving rotten egg odor. Iron sulfide precipitate gives blackened appearance. Typical foods are corn, peas.

Putrefactive anaerobe

³4.8

C. sporogenes

Plentiful gas. Disgusting putrid odor. pH often increased. Typical foods are corn, asparagus.

Psychrotrophic clostridia

>4.6

Aerobic sporeformers

³4.8

Bacillus spp.

Gas usually absent except for cured meats, milk is coagulated. Typical foods are milk, meats, beets.

Butyric spoilage

³4.0

C. butyricum, Clostridium tertium

Gas, acetic and butyric odor. Typical foods are tomatoes, peas, olives, cucumbers.

Acid spoilage

³4.2

Bacillus thermoacidurans

Flat (Bacillus) or gas (butyric anaerobes). Off odors depend on organism. Common foods are tomatoes, tomato products, other fruits.

 

<4

Alicyclobacillus acidoterrestris

Flat spoilage with off flavors. Most common in fruit juices, acid vegetables, and also reported to spoil iced tea.

a

Sources: see the text for references.

Spoilage of vacuum-packaged chilled meats. Production of gas, off flavors and odors, discoloration.

72 high levels of sporeformers (100, 104). Spore populations can also accumulate in a food plant, such as thermophilic spores on heated equipment and saccharolytic clostridia in plants processing sugar-rich foods such as fruits. The development of processing procedures to prevent or minimize spoilage depends on physical and chemical methods to inactivate the target organisms and on the formulation of foods and use of microbial inhibitors to prevent growth.

CLASSES OF FOODS SPOILED BY SPOREFORMERS

Spoilage of Dairy Ingredients and Finished Products

Thermophilic sporeformers can be a major cause of spoilage in the manufacture of dairy ingredients. Of particular importance in dairy ingredient manufacture are the sporeformers G. stearothermophilus, Geobacillus thermoleovorans, and Anoxybacillus flavithermus (15, 20). The presence of these spoilage thermophiles can lead to off flavors and textures through their formation of degradative enzymes, and these bacteria are indicators of poor hygiene, particularly in heated areas of the plants such as evaporators. Many of the thermophilic bacilli form biofilms that are difficult to control by clean-in-place and other hygienic procedures. Clostridia are not generally a major concern in dairy spoilage, although Clostridium tyrobutyricum and related bacteria can cause gas formation “blowing” and butyric acid odors in hard cheeses. The presence of such pathogens as C. botulinum, C. perfringens, B. cereus, and potentially C. difficile is a concern in finished dairy products, particularly infant formulas and medical foods. Heat-resistant clostridia, including C. sporogenes and Clostridium thermosaccharolyticum, can cause swelling of cans or packages through gas production and off flavors through production of degradative enzymes in ultrahigh-temperature-canned milk products.

Refrigerated Meat Products

Psychrotrophic clostridia can spoil refrigerated meat products in several ways, including the production of noxious gases such as hydrogen sulfides, formation of putrefactive amines, and protein and carbohydrate degradation. They can also cause swelling of cans and packages. The primary species of Clostridium involved in ready-to-eat meat and poultry spoilage are C. laramie, C. estertheticum, C. gasigenes, C. butyricum, and C. pasteurianum (2, 20, 73, 85, 104, 188). In particu-

Factors of Special Significance lar, C. estertheticum and C. gasigenes can cause “blown­ pack” spoilage, which occurs in refrigerated batches of vacuum-packaged meats and involves the production of large quantities of gas, putrid odors, and a metallic sheen on the meat (2, 20, 85). Mesophilic bacilli such as Bacillus coagulans can also cause off odors typified as medicinal and phenolic.

Fruit and Vegetable Juices

Sporeformers are a serious cause of spoilage in several types of fruit and vegetable juices (104, 116, 180). Alicyclobacillus spp. have been recognized as a particular problem, leading to medicinal off flavors. Souring and gassing also occur in dairy-based products, caused by thermophilic bacilli and clostridia. Shelf-stable, highacid products can be spoiled by various clostridia and bacilli, including C. pasteurianum. These sporeformers can also lead to gas formation and acid production in fruit-containing products such as pastries.

Bakery Products

Various bacilli, including B. subtilis, B. pumilus, and B. cereus, cause a defect in bakery products known as ropiness, as well as fruity odors reminiscent of overripe melons. This is followed by enzymatic activities leading to discoloration, changes in texture, and stickiness due to the production of extracellular polysaccharides (104, 161). Spores are present in flours and other raw ingredients and subsequently introduced into the dough, in which they survive the baking process and then proliferate to cause the ropiness. This type of spoilage can mostly be prevented by careful monitoring of raw ingredients for the presence of aerobic sporeformers.

Canned Foods

The heat resistance properties of bacterial endospores make them a likely culprit in the spoilage of canned goods (104, 167). Well-known sporeformers, including G. thermophilus, C. sporogenes, B. coagulans, T. saccharolyticum, and D. nigrificans, cause various types of spoilage including gassing, container swelling, and flatsour spoilage (Table 3.6). Control of these heat-resistant sporeformers requires careful monitoring of ingredients, design of the heat process for the intended market and shelf life, and excellent sanitization.

DETECTION OF FOOD SPOILAGE SPOREFORMING ORGANISMS AND THEIR PREVENTION Physical, chemical, and microbiological methods for determination of food spoilage have been published

3.  Spores and Their Significance elsewhere (104, 161). The need for safety in working with potential biowarfare agents such as C. botulinum and its neurotoxins must be carefully considered and applied (113).

SPORES AS PROBIOTIC AGENTS Species of bacterial sporeformers are being evaluated as probiotic agents to prevent disease and improve the health of animals and humans (137). The safety of aerobic endospore-forming bacteria has been evaluated. Other than bacteria in the B. anthracis group, including B. cereus, which cause well-known diseases, and sporadic toxin production in foods by strains of B. licheniformis, B. pumilus, and B. subtilis, most other species of Bacillus spp. are believed to be nonpathogenic to healthy humans and animals (92). Safety assessments have been outlined (92). Antibiotic resistance, genetic modification, and gene transfer in the gastrointestinal tract are among the major concerns. Bacillus spores have been proposed as probiotic agents for human use (144). Their primary benefit is inhibition of pathogens. The mechanisms of persistence and inhibitory action towards pathogens have not been elucidated but may include competition for adherence and nutrients and production of inhibitory metabolites. For animal use, including poultry and monogastric animals, spore probiotics can enhance food conversion and possibly reduce pathogen infection (18). Several companies are investigating Bacillus spp. as probiotics for humans and animals.

MODELING GROWTH OF SPOREFORMERS IN FOODS There is abundant information regarding the behavior of microorganisms in foods, and it is often useful to generate statistical models to quantify safety risks in foods. Two general types of models have mainly been used in food microbiology: (i) those analyzing experimental growth and survival data using simple and higher-order polynominals; and (ii) theoretical models derived from basic scientific principles and computer analysis, used to predict microbial survival. Statistical models can be particularly useful to define important variables and predict microbial behavior in advance of practical testing (112). The determination of the kinetics of germination and lag phase of populations of C. botulinum and the development of predictive models have had an important impact on food safety (120). Single-cell-based approaches have also been used in modeling of spore germination and the lag phase of C. botulinum to bypass the limita-

73 tions of population models such as spore concentrations and interactions and bioavailability of germinants and nutrients (120). Some food processors and academics are beginning to use models involving neural nets. These are valuable because they are adaptive and improve in precision and predictive capability over time. An example of a model used extensively in the dairy industry is that of Tanaka et al. (175) for preventing growth of C. botulinum in process cheese. Although models can provide guidelines for food safety, ensuring microbial control generally needs to be ascertained by laboratory challenge studies. Work in the laboratory of P.S. has been supported by the Army Research Office, the NIH (GM19698), the USDA, and a Multi-University Research Initiative (MURI) award from the U.S. Department of Defense. Research in the laboratory of E.J. has been supported by the USDA, NIAID, and the industrial sponsors of the Food Research Institute.

References   1. Ablett, A. H., P. J. Lillford, and D. R. Martin. 1999. Glass formation and dormancy in bacterial spores. Int. J. Food Sci. Technol. 34:59–69.   2. Adam, K. H., S. H. Flint, and G. Brightwell. 2010. Psychrophilic and psychrotrophic clostridia: sporulation and germination processes and their role in the spoilage of chilled, vacuum-packaged beef, lamb, and venison. Int. J. Food Sci. Technol. 45:1539–1544.   3. Reference deleted.   4. Reference deleted.   5. Appert, N. 1810. L’Art de conserver pendant plusieurs annees toutes les substances animales et vegetales. (Translated by K. G. Bitting, Chicago, IL, 1920.) In S. A. Goldblith, M. A. Joslyn, and J. T. R. Nickerson (ed.), Introduction to the Thermal Processing of Foods. 1961. AVI Publishing Co., Westport, CT.   6. Aronson, A. I. 1993. Insecticidal toxins, p. 953–964. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and Other Gram-Positive Bacteria: Biochemistry, Physiology, and Molecular Genetics. American Society for Microbiology, Washington, DC.   7. Atluri, S., K. Ragkousi, D. E. Cortezzo, and P. Setlow. 2006. Co-operativity between different nutrient receptors in germination of spores of Bacillus subtilis and reduction of this co-operativity by alterations in the GerB receptor. J. Bacteriol. 188:28–36.   8. Barak, I., and A. J. Wilkinson. 2005. Where asymmetry in gene expression originates. Mol. Microbiol. 57:611–620.   9. Beaman, T. C., and P. Gerhardt. 1986. Heat resistance of bacterial spores correlated with protoplast dehydration, mineralization, and thermal adaptation. Appl. Environ. Microbiol. 52:1242–1246.   10. Bettegowda, C., X. Huang, J. Lin, I. Cheong, M. Kohli, S. A. Szabo, X. Zhang, L. A. Diaz, Jr., V. E. Velculescu,

Factors of Special Significance

74

  11.

  12.

  13.   14.   15.   16.

  17.

  18.

  19.

  20.

  21.

  22.

  23.

  24.

G. Parmigiani, K. W. Kinzler, B. Vogelstein, and S. Zhou. 2006. The genome and transcriptomes of the antitumor agent Clostridium novyi-NT. Nat. Biotechnol. 24:1573–1580. Blakistone, B., R. Chuyate, D. Kautter, Jr., J. Charbonneau, and K. Suit. 1999. Efficacy of oxonia active against selective spore formers. J. Food Prot. 62:262–267. Broda, D. M., J. A. Boerma, and T. Brightwell. 2009. Sources of psychrophilic and psychrotolerant clostridia causing spoilage of vacuum-packaged chilled meats, as determined by PCR amplification procedure. J. Appl. Microbiol. 107:178–186. Brown, K. L. 2000. Control of bacterial spores. Br. Med. Bull. 56:158–171. Bulloch, W. 1938. The History of Bacteriology. Oxford University Press, Oxford, England. Burgess, S. A., D. Lindsay, and S. H. Flint. 2010. Thermophilic bacilli and their importance in dairy processing. Int. J. Food Microbiol. 144:215–225. Camp, A. H., and R. Losick. 2009. A feeding tube model for activation of a cell-type specific transcription factor during sporulation in Bacillus subtilis. Genes Dev. 23:1014–1024. Cangiano, C. G., A. Mazzone, L. Baccigalupi, R. Isticato, P. Eichenberger, M. De Felice, and E. Ricca. 2010. Direct and indirect control of late sporulation genes by GerR of Bacillus subtilis. J. Bacteriol. 192:3406–3413. Cartman, S. T., and R. M. La Ragione. 2004. Spore probiotics as animal feed supplements, p. 155–161. In E. Ricca, A. O. Henriques, and S. M. Cutting (ed.), Bacterial Spore Formers. Probiotics and Emerging Applications. Horizon Bioscience, Norfolk, United Kingdom. Cato, E. P., W. L. George, and S. M. Finegold. 1986. The genus Clostridium, p. 1141–1200. In H. A. Sneath, N. S. Mair, and M. E. Sharpe (ed.), Bergey’s Manual of Systematic Bacteriology, vol. 2. Williams & Wilkins, Baltimore, MD. Champagne, C. P., R. R. Laing, D. Roy, A. A. Mafu, and M. W. Griffiths. 1994. Psychrotrophs in dairy products: their effects and their control. Crit. Rev. Food Sci. Nutr. 34:1–30. Chary, V. K., P. Xenopoulos, and P. J. Piggot. 2007. Expression of the sF-directed csfB locus prevents premature appearance of sG activity during sporulation of Bacillus subtilis. J. Bacteriol. 189:8754–8757. Chastanet, A., D. Vitkup, G. C. Norman, J. S. Liu, and R. M. Losick. 2010. Broadly heterogeneous activation of the master regulator for sporulation in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 107:8486–8491. Chenokova, O. N., S. A. McPherson, C. T. Steichen, and C. L. Turnbough, Jr. 2009. The spore-specific alanine racemase of Bacillus anthracis and its role in suppressing germination during spore development. J. Bacteriol. 191:1303–1310. Chotyakul, N., G. Velazquez, and J. A. Torres. 2011. Assessment of the uncertainty in thermal food processing decisions based on microbial safety objectives. J. Food Eng. 102:247–256.

  25. Clarkson, J., I. D. Campbell, and M. D. Yudkin. 2004. Efficient regulation of sF, the first sporulation-specific sigma factor in B. subtilis. J. Mol. Biol. 342:1187–1195.   26. Claverys, J. P., and L. S. Hålvarstein. 2007. Cannibalism and fratricide: mechanisms and raison d’être. Nat. Rev. Microbiol. 5:219–229.   27. Coleman, W. H., and P. Setlow. 2009. Analysis of damage due to moist heat treatment of spores of Bacillus subtilis. J. Appl. Microbiol. 106:1600–1607.   28. Coleman, W. H., D. Chen, Y.-q. Li, A. E. Cowan, and P. Setlow. 2007. How moist heat kills spores of Bacillus subtilis. J. Bacteriol. 189:8458–8466.   29. Coleman, W. H., P. Zhang, Y.-q. Li, and P. Setlow. 2010. Mechanism of killing of spores of Bacillus cereus and Bacillus megaterium by wet heat. Lett. Appl. Microbiol. 50:507–514.   30. Cowan, A. E., D. E. Koppel, B. Setlow, and P. Setlow. 2003. A soluble protein is immobile in dormant spores of Bacillus subtilis but is mobile in germinated spores: implications for spore dormancy. Proc. Natl. Acad. Sci. USA 100:4209–4214.   31. Cowan, A. E., E. M. Olivastro, D. E. Koppel, C. A. Loshon, B. Setlow, and P. Setlow. 2004. Lipids in the inner membrane of dormant spores of Bacillus species are immobile. Proc. Natl. Acad. Sci. USA 101:7733–7738.   32. Dawes, I. W., and J. Mandelstam. 1970. Sporulation of Bacillus subtilis in continuous culture. J. Bacteriol. 103:529–535.   33. de Hoon, M. J. L., P. Eichenberger, and D. Vitkup. 2010. Hierarchical evolution of the bacterial sporulation network. Curr. Biol. 20:R735–R745.   34. de Jong, I. G., J. W. Veening, and O. P. Kuipers. 2010. Heterochronic phosphorelay gene expression as a source of heterogeneity in Bacillus subtilis spore formation. J. Bacteriol. 192:2053–2067.   35. De Vos, P., et al. (ed.). 2009. Bergey’s Manual of Systematic Bacteriology, 2nd ed., vol. 3. The Firmicutes. Springer, Dordrecht, The Netherlands.   36. Do, J. H., M. Nagasaki, and S. Miyano. 2010. The systems approach to the prespore-specific activation of sigma factor SigF in Bacillus subtilis. Biosystems 100:178S–184S.   36a. Dodatko, T., M. Akoachere, S. M. Muehlbauer, F. Helfrich, A. Howerton, C. Ross, V. Wysocki, J. Brojatsch, and E. Abel-Santos. 2009. Bacillus cereus spores release alanine that synergizes with inosine to promote germination. PLoS One 4:e6398.   37. Doyle, M. P. 1991. Evaluating the potential risk from extended shelf-life refrigerated foods by Clostridium botulinum inoculation studies. Food Technol. 45:154–156.   38. Eichenberger, P., M. Fujita, S. T. Jensen, E. M. Conlon, D. Z. Rudner, S. T. Wang, C. Ferguson, T. Sato, J. S. Liu, and R. Losick. 2005. The program of gene transcription for a single differentiating cell type during sporulation in Bacillus subtilis. PLoS Biol. 2:e328.   39. Eijlander, R. T., T. Abee, and O. P. Kuipers. 2011. Bacterial spores in food: how phenotypic variability complicates prediction of spore properties and bacterial behavior. Curr. Opin. Biotechnol. 22:180–186.

3.  Spores and Their Significance   40. Erickson, M. C., and J. L. Kornacki. 2003. Bacillus anthracis: current knowledge in relation to contamination of food. J. Food Prot. 66:691–699.   41. Esty, J. R., and K. F. Meyer. 1922. The heat resistance of the spores of Bacillus botulinus and allied anaerobes. XI. J. Infect. Dis. 31:650–663.   42. Eswaramoorthy, P., D. Duan, J. Dinh, A. Dravis, S. N. Devi, and M. Fujita. 2010. The threshold level of the sensor histidine kinase KinA governs entry into sporulation in Bacillus subtilis. J. Bacteriol. 192:3870–3882.   43. Fairhead, H., B. Setlow, and P. Setlow. 1993. Prevention of DNA damage in spores and in vitro by small, acidsoluble proteins from Bacillus species. J. Bacteriol. 175:1367–1374.   44. Franciosa, G., P. Aureli, and R. Schechter. 2003. Clostridium botulinum, p. 61–89. In M. D. Bier and J. W. Miliotis (ed.), International Handbook of Foodborne Pathogens. Marcel Dekker, New York, NY.   45. Fujita, M., J. E. Gonzalez-Pastor, and R. Losick. 2005. High- and low-threshold genes in the Spo0A regulon of Bacillus subtilis. J. Bacteriol. 187:1357–1368.   46. Fujita, M., and R. Losick. 2005. Evidence that entry into sporulation in Bacillus subtilis is governed by a gradual increase in the level and activity of the master regulator Spo0A. Genes Dev. 19:2236–2244.   47. Gerhardt, P., and R. E. Marquis. 1989. Spore thermore­ sistance mechanisms, p. 17–63. In I. Smith, R. Slepecky, and P. Setlow (ed.), Regulation of Procaryotic Development. American Society for Microbiology, Washington, DC.   48. Ghosh, S., and P. Setlow. 2009. Isolation and characterization of superdormant spores of Bacillus species. J. Bacteriol. 191:1787–1797.   49. Ghosh, S., P. Zhang, Y.-q. Li, and P. Setlow. 2009. Superdormant spores of Bacillus species have elevated wet heat resistance and temperature requirements for heat activation. J. Bacteriol. 191:5584–5591.   50. Gilmore, M. E., D. Bandyopadhyay, A. M. Dean, S. D. Linnstaedt, and D. L. Popham. 2004. Production of muramic d-lactam in Bacillus subtilis spore peptidoglycan. J. Bacteriol. 186:80–89.   51. Glass, K. G., and E. A. Johnson. 2001. Formulating low-acid foods for botulinal safety, p. 323–350. In V. K. Juneja and J. N. Sofos (ed.), Control of Foodborne Organisms. Marcel Dekker, New York, NY.   52. Goldblith, S. A., M. A. Joslyn, and J. T. R. Nickerson. 1961. An Anthology of Food Science, vol. 1. Introduction to the Thermal Processing of Foods. AVI Publishing, Westport, CT.   53. Granum, P. E., and T. C. Baird-Parker. 2000. Bacillus species, p. 1029–1039. In B. M. Lund, T. C. BairdParker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Food, vol. II. Aspen Publishers, Gaithersburg, MD.   54. Greenberg, R. A., R. B. Tompkin, B. O. Blade, R. S. Kittaka, and A. Anelis. 1966. Incidence of mesophilic spores in raw pork, beef, and chicken in processing plants in the United States and Canada. Appl. Microbiol. 14:789–793.   55. Griffiths, K., and P. Setlow. 2009. Effects of modification of membrane lipid composition on Bacillus subtilis

75

  56.   57.

  58.

  59.   60.   61.

  62.   63.   64.

  65.   66.

  67.

  68.

  69.

  70.

sporulation and spore properties. J. Appl. Microbiol. 106:2064–2078. Guillemin, J. 1999. Anthrax. The Investigation of a Deadly Outbreak. University of California Press, Berkeley, CA. Harry, K. H., R. Zhou,, L. Kroos, and S. B. Melville. 2009. Sporulation and enterotoxin synthesis are controlled by the sporulation-specific sigma factors SigE and SigK in Clostridium perfringens. J. Bacteriol. 191:2728–2742. Hatheway, C. L., and E. A. Johnson. 1998. Clostridium: the spore-bearing anaerobes, p. 732–782. In W. J. Hausler and M. Sussman (ed.), Topley and Wilson’s Microbiology and Microbial Infections, 9th ed., vol. 3. Edward Arnold, London, England. Henriques, A. O., and C. P. Moran, Jr. 2007. Structure, assembly, and function of the spore surface layers. Annu. Rev. Microbiol. 61:555–588. Hilbert, D. W., and P. J. Piggot. 2004. Compartmental­ ization of gene expression during Bacillus subtilis spore formation. Microbiol. Mol. Biol. Rev. 68:234–262. Hornstra, L. M., A. Ter Beek, J. P. Smelt, W. W. Kallemeijn, and S. Brul. 2009. On the origin in (preservation) resistance of Bacillus spores: input for a “systems” analysis approach of bacterial spore outgrowth. Int. J. Food Microbiol. 134:9–15. Hurst, A. 1983. Injury, p. 255–274. In A. Hurst and G. W. Gould (ed.), The Bacterial Spore, vol. 2. Academic Press, London, England. Husmark, U., and U. Ronner. 1992. The influence of hydrophobic, electrostatic and morphologic properties on the adhesion of Bacillus spores. Biofouling 5:335–344. Inglesby, T. V., D. A. Henderson, J. G. Bartlett, M. S. Ascher, E. Eitzen, A. M. Friedlander, J. Hauer, J. McDade, M. T. Osterholm, T. O’Toole, G. Parker, T. M. Perl, P. K. Russell, and K. Tonat for the Working Group on Civilian Biodefense. 1999. Anthrax as a biological weapon. Medical and public health management. JAMA 281:1735–1745. Ingram, M. 1969. Sporeformers as food spoilage organisms, p. 549–610. In G. W. Gould and A. Hurst (ed.), The Bacterial Spore. Academic Press, London, England. International Commission on Microbiological Specifications for Foods. 1996. Microorganisms in Foods 5. Characteristics of Microbial Pathogens. Blackie Academic & Professional, London, England. Johnson, E. A. 1991. Microbiological safety of fermented foods, p. 135–169. In J. G. Zeikus and E. A. Johnson (ed.), Mixed Cultures in Biotechnology. McGraw Hill, New York, NY. Johnson, E. A., and M. C. Goodnough. 1998. Botulism, p. 724–741. In W. J. Hausler and M. Sussman (ed.), Topley and Wilson’s Microbiology and Microbial Infections, 9th ed., vol. 3. Edward Arnold, London, England. Johnson, E. A. 2006. Clostridium botulinum and Clostridium tetani, p. 1035–1088. In S. P. Borriello, P. R. Murray, and G. Funke (ed.), Topley and Wilson’s Microbiology and Microbial Infections, 8th ed. Hodder Arnold, London, England. Jones, S. W., C. J. Paredes, B. Tracy, N. Cheng, R. Sillers, R. S. Senger, and E. T. Papoutsakis. 2008. The

Factors of Special Significance

76

  71.

  72.

  73.   74.

  75.

  76.   77.

  78.   79.

  80.   81.

  82.

  83.   84.

  85.

transcriptional program underlying the physiology of clostridial sporulation. Genome Biol. 9:R14. Juneja, V. K., and H. M. Marks. 1999. Proteolytic Clostridium botulinum growth at 12–48°C simulating the cooling of cooked meat: development of a predictive model. Food Microbiol. 16:583–592. Juneja, V. K., B. S. Marmer, J. G. Phillips, and A. J. Miller. 1995. Influence of the intrinsic properties of food on thermal inactivation of spores of nonproteolytic Clostridium botulinum: development of a predictive model. J. Food Saf. 15:349–364. Kalinowski, R. M., and R. B. Tompkin. 1999. Psychrotrophic clostridia causing spoilage in cooked meat and poultry products. J. Food Prot. 62:766–772. Karmazyn-Campelli, C., L. Rhayat, R. CarballidoLópez, S. Duperrier, N. Frandsen, and P. Stragier. 2008. How the early sporulation sigma factor sigmaF delays the switch to late development in Bacillus subtilis. Mol. Microbiol. 67:1169–1180. Keijser, B. J., A. Ter Beek, H. Rauwerda, F. Schuren, R. Montijn, H. van der Spek, and S. Brul. 2007. Analysis of temporal gene expression during Bacillus subtilis spore germination and outgrowth. J. Bacteriol. 189:3624–3634. Keynan, A., and Z. Evenchik. 1969. Activation, p. 359– 396. In G. W. Gould and A. Hurst (ed.), The Bacterial Spore. Academic Press, New York, NY. Keynan, A., and N. Sandler. 1984. Spore research in historical perspective, p. 1–48. In A. Hurst and G. W. Gould (ed.), The Bacterial Spore, vol. 2. Academic Press, London, England. Kihm, D. J., M. T. Hutton, J. H. Hanlin, and E. A. Johnson. 1988. Zinc stimulates sporulation in Clostridium botulinum 113B. Curr. Microbiol. 17:193–198. Kihm, D. J., M. T. Hutton, J. H. Hanlin, and E. A. Johnson. 1990. Influence of transition metals added during sporulation on heat resistance of Clostridium botulinum 113B spores. Appl. Environ. Microbiol. 56:681–685. Kihm, D. J., and E. A. Johnson. 1990. Hydrogen gas accelerates thermal inactivation of Clostridium botulinum spores. Appl. Microbiol. Biotechnol. 33:705–708. Kim, J., and P. M. Foegeding. 1993. Principles of control, p. 121–176. In A. H. W. Hauschild and K. L. Dodds (ed.), Clostridium botulinum. Ecology and Control in Foods. Marcel Dekker, New York, NY. Kong, L., P. Zhang, P. Setlow, and Y.-Q. Li. 2010. Characterization of bacterial spore germination using integrated phase contrast microscopy, Raman spectroscopy and optical tweezers. Anal. Chem. 82:3840–3847. Kroos, L. 2007. The Bacillus and Myxococcus developmental networks and their transcriptional regulators. Annu. Rev. Genet. 41:13–39. Laaberki, M. H., and J. Dworkin. 2008. Role of spore coat proteins in the resistance of Bacillus subtilis spores to Caenorhabditis elegans predation. J. Bacteriol. 190:6197–6203. Lawson, P., R. H. Dainty, N. Kristiansen, J. Berg, and M. D. Collins. 1994. Characterization of a psychrotrophic Clostridium causing spoilage in vacuum-packed

  86.

  87.   88.   89.

  90.

  91.

  92.

  93.   94.

  95.

  96.   97.

  98.   99. 100. 101.

cooked pork: description of Clostridium algidicarnis sp. nov. Lett. Appl. Microbiol. 19:153–157. Lee, K. S., D. Bumbaca, J. Kosman, P. Setlow, and M. J. Jedrzejas. 2008. Structure of a protein-DNA complex essential for DNA protection in spores of Bacillus species. Proc. Natl. Acad. Sci. USA 105:2806–2811. Leisner, M., K. Stingl, E. Frey, and B. Maier. 2008. Stochastic switching to genetic competence. Curr. Opinion Microbiol. 11:553–559. Leuschner, R. G. K., and P. J. Lillford. 2003. Thermal properties of bacterial spores and biopolymers. Int. J. Food Microbiol. 87:8–14. Li, J., D. Paredes-Sabja, M. R. Sarker, and B. A. McClane. 2009. Further characterization of Clostridiuum perfringens small, acid-soluble protein-4 (Ssp4) properties and expression. PLoS One 7:e6249. Li, J., and B. A. McClane. 2010. Evaluating the involvement of alternative sigma factors SigF and SigG in Clostridium perfringens sporulation and enterotoxin synthesis. Infect. Immun. 78:4286–4293. Loeb, L. A., E. A. James, A. M. Waltersdorph, and S. J. Klebanoff. 1988. Mutagenesis by the autoxidation of iron with isolated DNA. Proc. Natl. Acad. Sci. USA 85:3918–3922. Logan, N. A. 2004. Safety of aerobic endospore-forming bacteria, p. 93–105. In E. Ricca, A. O. Henriques, and S. M. Cutting (ed.), Bacterial Spore Formers. Probiotics and Emerging Applications. Horizon Bioscience, Norfolk, United Kingdom. López, D., and R. Kolter. 2009. Extracellular signals that define distinct and coexisting cell fates in Bacillus subtilis. FEMS Microbiol. Rev. 34:134–149. Lund, B. M., and M. W. Peck. 1994. Heat resistance and recovery of spores of nonproteolytic Clostridium botulinum in relation to refrigerated, processed foods with extended shelf-life. J. Appl. Bacteriol. Symp. 76:115S–128S. Lund, D. 1975. Thermal processing, p. 31–92. In M. Karel, O. R. Fennema, and D. B. Lund (ed.), Principles of Food Science. Part II. Physical Principles of Food Preservation. Marcel Dekker, New York, NY. Lynt, R. K., D. A. Kautter, and R. B. Read, Jr. 1975. Botulism in commercially canned foods. J. Milk Food Technol. 38:546–550. Lynt, R. K., D. A. Kautter, and H. M. Solomon. 1982. Differences and similarities among proteolytic strains of Clostridium botulinum types A, B, E and F: a review. J. Food Prot. 45:466–474. Maughan, H. 2007. Rates of molecular evolution in bacteria are relatively constant despite spore dormancy. Evolution 61:280-288. McDonnell, G., and A. D. Russell. 1999. Antiseptics and disinfectants: activity, action and resistance. Clin. Microbiol. Rev. 12:147–179. McKee, L. H. 1995. Microbial contamination of spices and herbs: a review. Lebensm. Wiss. Technol. 28:1–11. McKevitt, M. T., K. M. Bryant, S. M. Shakir, J. L. Larabee, S. R. Blanke, J. Lovchik, C. R. Lyons, and J. D. Ballard. 2007. Effects of endogenous D-alanine synthesis and autoinhibition of Bacillus anthracis germina-

3.  Spores and Their Significance

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113. 114.

tion on in vitro and in vivo infections. Infect. Immun. 75:5726–5734. Meisner, J., X. Wang, M. Serrano, A. O. Henriques, and C. P. Moran, Jr. 2008. A channel connecting the mother cell and forespore during bacterial endospore formation. Proc. Natl. Acad. Sci. USA 105:15100–15105. Moeller, R., P. Setlow, G. Horneck, T. Berger, G. Reitz, P. Rutberg, A. J. Doherty, R. Okayasu, and W. L. Nicholson. 2008. Role of major small, acid-soluble spore proteins, spore specific and universal DNA repair mechanisms in the resistance of Bacillus subtilis spores to ionizing radiation from X-rays and high energy charged (HZE) particle bombardment. J. Bacteriol. 190:1134–1140. Moir, C. J. (ed.). 2001. Spoilage of Processed Foods: Cause and Diagnosis. Australian Institute of Food Science and Technology (NSW Branch), Waterloo DC, New South Wales, Australia. Molle, V., Y. Nakaura, R. P. Shivers, H. Yamaguchi, R. Losick, Y. Fujita, and A. L. Sonenshein. 2003. Additional targets of the Bacillus subtilis global regulator CodY identified by chromatin immunoprecipitation and genome-wide transcript analysis. J. Bacteriol. 185:1911–1922. Moorhead, S. M., and R. G. Bell. 1999. Psychrotrophic clostridia mediated gas and botulinal toxin production in vacuum-packed chilled meat. Lett. Appl. Microbiol. 28:108–112. Morris, C., A. L. Brody, and L. Wicker. 2007. Nonthermal food processing/preservation technologies: a review with packaging implications. Packaging Technol. Sci. 20:275–286. Moschonas, G., D. J. Bolton, J. J. Sheridan, and D. A. McDowell. 2009. Isolation and sources of ‘blown pack’ spoilage clostridia in beef abattoirs. J. Appl. Microbiol. 107:616–624. NFPA/CMI Container Integrity Task Force, Micro­ biological Assessment Group Report. 1984. Botulism risk from post-processing contamination of commercially canned foods in metal containers. J. Food Prot. 47:801–816. Nicholson, W. L., N. Munakata, G. Horneck, H. J. Melosh, and P. Setlow. 2000. Resistance of Bacillus endospores to extreme terrestrial and extraterrestrial environments. Microbiol. Mol. Biol. Rev. 64:548–572. Nicholson, W. L., A. C. Schuerger, and P. Setlow. 2005. The solar UV environment and bacterial spore UV resistance: considerations for Earth-to Mars transport by natural processes and human spaceflight. Mutat. Res. 571:248–264. Notermans, S., P. in’t Veld, T. Wijtzes, and G. C. Mead. 1993. A user’s guide to microbial challenge testing for ensuring the safety and stability of food products. Food Microbiol. 10:145–157. Nulens, E., and A. Voss. 2002. Laboratory diagnosis and biosafety issues of biological warfare agents. Clin. Microbiol. Infect. 8:455–466. Ohye, D. F., and W. J. Scott. 1957. Studies in the physiology of Clostridium botulinum type E. Aust. J. Biol. Sci. 10:85–94.

77 115. Ordal, G. W., L. Marquez-Magana, and M. J. Chamberlin. 1993. Motility and chemotaxis, p. 765– 784. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and Other Gram-Positive Bacteria: Biochemistry, Physiology and Molecular Genetics. American Society for Microbiology, Washington, DC. 116. Orr, R. V., and L. R. Beuchat. 2000. Efficacy of disinfectants in killing of spores of Alicyclobacillus acidoterrestris and performance of media supporting colony development by survivors. J. Food Prot. 63:1117–1122. 117. Paidhungat, M., and P. Setlow. 2002. Spore germination and outgrowth, p. 537–548. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and Its Closest Relatives: from Genes to Cells. ASM Press, Washington, DC. 118. Paredes, C. J., K. V. Alasker, and E. T. Papoutskis. 2005. A comparative genomic view of clostridial sporulation and physiology. Nat. Rev. Microbiol. 3:969–978. 119. Paredes-Sabja, D., P. Setlow, and M. R. Sarker. 2011. Germination of spores of Bacillales and Clostridiales species: mechanisms and proteins involved. Trends Microbiol. 19:35–94. 120. Peck, M. W. 2009. Biology and genomic analysis of Clostridium botulinum. Adv. Microb. Physiol. 55:183–265. 121. Peleg, M., and M. B. Cole. 1998. Reinterpretation of microbial survival curves. Crit. Rev. Food Sci. 38:353–380. 122. Peleg, M., and M. B. Cole. 2000. Estimating the survival of Clostridium botulinum spores during heat treatments. J. Food Prot. 63:190–195. 123. Peng, L., D. Chen, P. Setlow, and Y.-Q. Li. 2009. Elastic and inelastic light scattering from single bacterial spores in an optical trap allows monitoring of spore germination dynamics. Anal. Chem. 81:4035–4042. 124. Perni, S., R. R. Beumer, and M. H. Zwietering. 2009. Multi-Tools for food safety risk management of steam meals. J. Food Prot. 72:2638–2645. 125. Pflug, I. J. 1987. Endpoint of a preservation process. J. Food Prot. 50:347–351. 126. Pflug, I. J. 1987. Factors important in determining the heat process value, FT, for low acid canned foods. J. Food Prot. 50:528–533. 127. Pflug, I. J. 1987. Calculating FT-values for heat preservation of shelf-stable, low acid canned foods using the straightline semilogarithmic model. J. Food Prot. 50:608–615. 128. Phillips, Z. E., and M. A. Strauch. 2002. Bacillus subtilis sporulation and stationary phase gene expression. Cell. Mol. Life Sci. 59:392–402. 129. Piggot, P. J., and D. W. Hilbert. 2004. Sporulation of Bacillus subtilis. Curr. Opin. Microbiol. 7:579–586. 130. Piggot, P. J., and R. Losick. 2002. Sporulation genes and intercompartmental regulation, p. 483–518. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and Its Closest Relatives: from Genes to Cells. ASM Press, Washington, DC. 131. Pitt, J. I., and A. D. Hocking (ed.). 1997. Fungi and Food Spoilage, 2nd ed. Blackie Academic & Professional, London, England.

Factors of Special Significance

78 132. Popham, D. L. 2002. Specialized peptidoglycan of the bacterial endospore: the inner wall of the lockbox. Cell. Mol. Life Sci. 59:426–433. 133. Prescott, S. C., and W. L. Underwood. 1897. Micro­organisms and sterilizing processes in the canning industries. Technol. Q. 10:183–199. 134. Rahman, M. S. (ed.). 2007. Handbook of Food Preservation. Marcel-Dekker, New York, NY. 135. Rajkovic, A., N. Smigic, and F. Devlieghere. 2010. Contemporary strategies in combating microbial contamination in the food chain. Int. J. Food Microbiol. 141:S29–S42. 136. Ramirez, N., M. Liggins, and E. Abel-Santos. 2010. Kinetic evidence for the presence of putative germination receptors in C. difficile spores. J. Bacteriol. 192:4215–4222. 137. Ricca, E., A. O. Henriques, and S. M. Cutting (ed.). 2004. Bacterial Spore Formers. Probiotics and Emerging Applications. Horizon Bioscience, Norfolk, U.K. 138. Rice, E. W., N. J. Adcock, M. Sivaganesan, and L. J. Rose. 2005. Inactivation of spores of Bacillus anthracis Sterne, Bacillus cereus, and Bacillus thuringiensis by chlorination. Appl. Environ. Microbiol. 71:5587–5589. 139. Ross, C., and E. Abel-Santos. 2010. The Ger receptor family from sporulating bacteria. Curr. Issues Mol. Biol. 12:147–158. 140. Russell, A. D. 1998. Assessment of sporicidal efficacy. Int. Biodeterior. Biodegr. 41:281–287. 141. Russell, H. L. 1896. Gaseous fermentations in the canning industry, p. 227–231. In Twelth Annual Report of the Agricultural Experiment Station of the University of Wisconsin. University of Wisconsin, Madison, WI. 142. Schmitt, H. P. 1966. Commercial sterility in canned foods, its meaning and determination. Assoc. Food Drug Off. U. S. Q. Bull. 30:141–151. 143. Scott, V. N., and D. T. Bernard. 1982. Heat resistance of spores of non-proteolytic type B Clostridium botulinum. J. Food Prot. 45:909–912. 144. Senesi, S. 2004. Bacillus spores as probiotic products for human use, p. 131–141. In E. Ricca, A. O. Henriques, and S. M. Cutting (ed.), Bacterial Spore Formers. Probiotics and Emerging Applications. Horizon Bioscience, Norfolk, United Kingdom. 145. Serrano, M., A. Neves, C. M. Soares, C. P. Moran, Jr., and A. O. Henriques. 2004. Role of the anti-sigma factor SpoIIAB in regulation of sG during Bacillus subtilis sporulation. J. Bacteriol. 186:4000–4013. 146. Setlow, B., S. Atluri, R. Kitchel, K. Koziol-Dube, and P. Setlow. 2006. Role of dipicolinic acid in resistance and stability of spores of Bacillus subtilis with or without DNA-protective a/b-type small acid-soluble proteins. J. Bacteriol. 188:3740–3747. 147. Setlow, B., L. Peng, C. A. Loshon, Y. Q. Li, G. Christie, and P. Setlow. 2009. Characterization of the germination of Bacillus megaterium spores lacking enzymes that degrade the spore cortex. J. Appl. Microbiol. 107:318–328. 148. Setlow, P. 1993. DNA structure, spore formation and spore properties, p. 181–194. In P. J. Piggot, P.

149. 150. 151. 152. 153. 154.

155.

156.

157.

158. 159. 160. 161. 162.

163.

164. 165.

Youngman, and C. P. Moran, Jr. (ed.), Regulation of Bacterial Differentiation. American Society for Microbiology, Washington, DC. Setlow, P. 1994. Mechanisms which contribute to the long-term survival of spores of Bacillus species. J. Appl. Bacteriol. 176:49S–60S. Setlow, P. 1995. Mechanisms for the prevention of damage to the DNA in spores of Bacillus species. Annu. Rev. Microbiol. 49:29–54. Setlow, P. 2003. Spore germination. Curr. Opin. Microbiol. 6:550–556. Setlow, P. 2006. Spores of Bacillus subtilis: their resistance to radiation, heat and chemicals. J. Appl. Microbiol. 101:514–525. Setlow, P. 2007. I will survive: DNA protection in bacterial spores. Trends Microbiol. 15:172–180. Setlow, P. 2008. Effects of high pressure on spores, p. 35– 52. In C. Michiels, D. H. Bartlett, and A. Aertsen (ed.), High-Pressure Microbiology. ASM Press, Washington, DC. Setlow, P., and E. A. Johnson. 2007. Spores and their significance, p. 35–67. In M. P. Doyle and L. R. Beuchat (ed.), Food Microbiology: Fundamentals and Frontiers, 3rd ed. ASM Press, Washington, DC. Shah, I. M., M. H. Laaberki, D. L. Popham, and J. Dworkin. 2008. A eukaryotic-like Ser/Thr kinase signals bacteria to exit dormancy in response to peptidoglycan fragments. Cell 135:486–496. Shimizu, T., K. Ohtani, H. Hirakawa, K. Ohshima, A. Yamashita, T. Shiba, N. Ogasawara, M. Hattori, S. Kuhara, and H. Hayashi. 2002. Complete genome sequence of Clostridium perfringens, an anaerobic flesheater. Proc. Natl. Acad. Sci. USA 99:996–2001. Smith, L. D. S., and H. Sugiyama. 1988. Botulism. The Organism, Its Toxins, the Disease, 2nd ed. Charles C. Thomas, Springfield, IL. Sonenshein, A. L. 2007. Control of key metabolic intersections in Bacillus subtilis. Nat. Rev. Microbiol. 5:917–927. Songer, J. G. 1996. Clostridial enteric diseases of domestic animals. Clin. Microbiol. Rev. 9:216–234. Sperber, W. H., and M. P. Doyle (ed.). 2009. Compendium of the Microbiological Spoilage of Foods and Beverages. Springer, New York, NY. Spotts Whitney, E. A., M. E. Beatty, T. H. Taylor, Jr., R. Weyant, J. Sobel, M. J. Arduino, and D. A. Ashford. 2003. Inactivation of Bacillus anthracis spores. Emerg. Infect. Dis. 9:623–627. Steil, L., M. Serrano, A. O. Henriques, and U. Volker. 2005. Genome-wide analysis of temporally regulated and compartment-specific gene expression in sporulating cells of Bacillus subtilis. Microbiology 151:399–420. Stephenson, K., and J. A. Hoch. 2009. Evolution of signalling in the sporulation phosphorelay. J. Bacteriol. 191:7296–7305. Stevenson, K. E., and R. H. Vaughn. 1972. Exosporium formation in sporulating cells of Clostridium botulinum 78A. J. Bacteriol. 112:618–621.

3.  Spores and Their Significance 166. Stringer, M. 2005. Summary report. Food safety objectives—role in microbiological food safety management. Food Control 16:775–794. 167. Stumbo, C. R. 1973. Thermobacteriology in Food Processing, 2nd ed. Academic Press, New York, NY. 168. Sugiyama, H. 1951. Studies on factors affecting the heat resistance of spores of Clostridium botulinum. J. Bacteriol. 62:81–96. 169. Sugiyama, H. 1952. Effect of fatty acids on the heat resistance of Clostridium botulinum spores. Bacteriol. Rev. 16:125–126. 170. Sugiyama, H. 1986. Mouse models for infant botulism, p. 73–91. In O. Zak and M. A. Sande (ed.), Experimental Models in Antimicrobial Chemotherapy, vol. 2. Academic Press, New York, NY. 171. Sugiyama, H. 1980. Clostridium botulinum neurotoxin. Microbiol. Rev. 44:419–448. 172. Sugiyama, H., M. Woodburn, K. H. Yang, and C. Movroydis. 1981. Production of botulinum toxin in inoculated pack studies of foil-wrapped potatoes. J. Food Prot. 44:896. 173. Sun, D.-W. 2005. Emerging Technologies for Food Processing. Academic Press, New York, NY. 174. Sunde, E. P., P. Setlow, L. Hederstedt, and B. Halle. 2009 The physical state of water in bacterial spores. Proc. Natl. Acad. Sci. USA 106:19334–19339. 175. Tanaka, N., E. Traisman, P. Plantinga, L. Finn, W. Flom, L. Meske, and J. Guffisberg. 1986. Evaluation of factors involved in antibotulinal properties of pasteurized process cheese spreads. J. Food Prot. 49:526–531. 176. Tewari, G., and V. J. Juneja. 2007. Advances in Thermal and Non-Thermal Food Preservation. Blackwell, Ames, IA. 177. Tournas, V. 1994. Heat-resistant fungi of importance to the food and beverage industry. Crit. Rev. Microbiol. 20:243–263. 178. Townsend, C. T., J. R. Esty, and F. C. Baselt. 1938. Heat-resistance studies on spores of putrefactive anaerobes in relation to the determination of safe processes for canned foods. Food Res. 3:323–346. 179. Trent, J. D., M. Gabrielson, B. Jensen, J. Neuhard, and J. Olsen. 1994. Acquired thermotolerance and heat shock proteins in thermophiles from the three phylogenetic domains. J. Bacteriol. 176:6148–6152. 180. Tribst, A. A. L., A. de Souza Sant’ Ana, and P. Rodriguez de Massauger. 2009. Microbiological quality and safety of fruit juices—past, present, and future perspectives. Crit. Rev. Microbiol. 35:310–339. 181. Underwood, S., S. Guan, V. Vijayasubhash, S. D. Baines, L. Graham, R. J. Lewis, M. H. Wilcox, and K. Stephenson. 2006. Characterization of the sporulation initiation pathway of Clostridium difficile and its role in toxin production. Mol. Microbiol. 59:1000–1012. 182. Van Netton, P., A. Van de Moosdijk, P. Van de Hoensel, D. A. A. Mossel, and I. Perales. 1990. Psychrotrophic

79

183.

184.

185.

186. 187.

188.

189.

190.

191. 192. 193. 194.

195.

196.

197.

strains of Bacillus cereus producing enterotoxin. J. Appl. Bacteriol. 69:73–79. Veening, J. W., L. W. Hamoen, and O. P. Kuipers. 2005. Phosphatases modulate bistable sporulation gene expression pattern in Bacillus subtilis. Mol. Microbiol. 56:1481–1494. Wang, S., B. Setlow, E. M. Conlon, J. L. Lyon, D. Imamura, T. Sato, P. Setlow, R. Losick, and P. Eichenberger. 2006. The forespore line of gene expression in Bacillus subtilis. J. Mol. Biol. 358:16–37. Westphal, A. J., P. B. Price, T. J. Leighton, and K. E Wheeler. 2003. Kinetics of size changes of individual Bacillus thuringiensis spores in response to changes in relative humidity. Proc. Natl. Acad. Sci. USA 100:3461–3466. Willis, A. T. 1969. Clostridia of Wound Infections. Butterworths, London, England. Wörner, K., H. Szurmant, C. Chiang, and J. A. Hoch. 2006. Phosphorylation and functional analysis of the sporulation initiation factor Spo0A from Clostridium botulinum. Mol. Microbiol. 59:1000–1012. Yang, X., S. Balamurugan, and C. O. Gill. 2011. Effects on the development of blown pack spoilage of initial numbers of Clostridium estertheticum spores and Leuconostoc mesenteroides on vacuum packaged beef. Meat Sci. 88:361–367. Yardimci, O., and P. Setlow. 2010. Plasma sterilization: opportunities and microbial assessment strategies in medical device manufacturing. IEEE Trans. Plasma Sci. 38:973–981. Yudkin, M. D., and J. Clarkson. 2005. Differential gene expression in genetically identical sister cells: the initiation of sporulation in Bacillus subtilis. Mol. Microbiol. 56:578–589. Zeki, B. 2009. Food Process Engineering and Technology. Academic Press, Boston, MA. Zeuthen, P., and L. Bogh-Sorensen (ed.). 2003. Food Preservation Techniques. CRC Press, Boca Raton, FL. Zhang, H., and G. S. Mitttal. 2008. Effects of highpressure processing (HPP) on bacterial spores: an overview. Food Rev. Int. 24:330–351. Zhang, L., M. L. Higgins, and P. J. Piggot. 1997. The division during bacterial sporulation is symmetrically located in Sporosarcina ureae. Mol. Microbiol. 25:1091–1098. Zhang, P., W. Garner, X. Yi, J. Yu, Y-q. Li, and P. Setlow. 2010. Factors affecting the variability in the time between addition of nutrient germinants and rapid DPA release during germination of spores of Bacillus species. J. Bacteriol. 392:3608–3619. Zhang, P., L. Kong, P. Setlow, and Y. Q. Li. 2010. Characterization of wet heat inactivation of single spores of Bacillus species by dual-trap Raman spectroscopy and elastic light scattering. Appl. Environ. Microbiol. 76:1796–1805. Zhang, P., P. Setlow, and Y.-q. Li. 2009. Characterization of single heat-activated Bacillus spores using laser tweezers Raman spectroscopy. Optics Express 17:16480–16491.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch4

Jean-Louis Cordier International Commission on Microbiological Specifications for Foods

Microbiological Criteria and Indicator Microorganisms

The concepts and principles for the establishment of microbiological criteria were elaborated in the mid-1980s by the International Commission on Microbiological Specifications for Foods (ICMSF) (9). These concepts have been used to develop recommendations for criteria for foods in international trade or for specific criteria for pathogens such as Listeria monocytogenes (10, 11). They have also been the basis of the Codex Alimentarius Commission (CAC) document “Principles for the Establishment and Application of Microbiological Criteria for Foods” (2). While CAC recognizes only the general category “microbiological criterion,” other national, transnational, and international organizations, trade associations, and other stakeholders in the food chain often describe microbiological standards, guidelines, or specifications that can be differentiated as follows: Microbiological standard is a mandatory criterion included into a law or a regulatory ordinance. Microbiological guideline is an advisory criterion issued by either authorities, industry associations, or food manufacturers. Such guidelines are indicative of what can be expected for certain mi-

4

crobiological parameters when a food is manufactured according to best hygienic or manufacturing practices. Microbiological specification is an element of purchasing agreements between a buyer and a supplier of a raw material or a food product. Their use may be mandatory or advisory depending on the agreements between the two parties. The term “microbiological criterion” will be used throughout this chapter, since the underlying principles for their establishment and application are valid for all three subcategories. Microbiological criteria have traditionally been developed around significant pathogens, relevant commensals, and hygiene indicators as reflected in the ICMSF cases (13). They are widely used today to discriminate between acceptable and unacceptable lots of food products. In order to address the need for science-based risk assessment, the ICMSF has explored the potential applications of risk assessment techniques to microbiological issues (12). This initial work has triggered an evolution from hazard-based to risk-based food safety management

Jean-Louis Cordier, Nestlé Ltd., Nestlé Quality Assurance Center, CH-1800 Vevey, Switzerland.

81

82 as observed during the last decade. Numerous discussions have since then taken place to develop the concepts further. These activities have culminated in the publication of the document “Principles and Guidelines for the Conduct of Microbiological Risk Management (MRM)” by the CAC (3). The evolution of the traditional metrics, including microbiological criteria, to include additional risk-based metrics has taken place over recent years. This includes a better understanding of the performance and limitations of microbiological criteria. These elements will be discussed in subsequent sections, as well as the role of microbiological indicators.

MICROBIOLOGICAL RISK MANAGEMENT METRICS Details on the establishment and implementation of MRM metrics have been outlined in Annex II of the “Principles and Guidelines To Perform Microbiological Risk Management (MRM)” adopted by CAC in 2007 (3). The scope of this document was to provide a framework for the MRM process to CAC as well as to CAC members and member organizations. The purpose was also to provide guidance to the food industry and other stakeholders who design, validate, and implement control measures ensuring the manufacture of safe food that consistently achieves the targets defined in the MRM metrics. The principles for their establishment and implementation are outlined as follows in the document (3) and are applied in addition to those described in the guidelines themselves: 1. The establishment and implementation of MRM metrics should follow a structured approach, with both the risk assessment phase and the subsequent risk management decisions being fully transparent and documented. 2. MRM metrics should be applied only to the extent necessary to protect human life or health and set at a level that is not more trade restrictive than required to achieve an importing member’s appropriate level of protection. 3. MRM metrics should be feasible, appropriate for the intended purpose, and applied within a specific food chain context at the appropriate step in that food chain. 4. MRM metrics should be developed and appropriately implemented so they are consistent with the requirements of the regulatory/legal system in which they will be used.

Factors of Special Significance Metrics used by competent authorities to express the expected level of control have been based traditionally on the following criteria: The Product Criterion (PdC) specifies chemical or physical characteristics of a food (e.g., pH, water activity), which, if met, contributes to the safety of a food product. Different elements are related to (and impact) a PdC such as the frequency and level of contamination, the effectiveness of control measures, the conditions under which the product will be used, and other parameters ensuring that it will not have the pathogen at an unacceptable level when the product is consumed. Each of the factors that may impact the effectiveness of a PdC has to be considered in a transparent manner. The Process Criterion (PcC) specifies the conditions of treatments to be applied at a specific step of the manufacturing process to achieve the targeted level of control of a microbiological hazard. Again, there are a number of parameters, such as the initial levels of the pathogen, the type and resistance of the target pathogen, and the impact of the food matrices, that influence the expected effect. The Microbiological Criterion (MC) is used for the examination of foods at a specified point of the food chain to determine the compliance of a food to a preestablished limit. Such a criterion can be used as a direct control measure on lots manufactured or, in the frame of a food safety control system, to periodically verify that the system is functioning as intended. Elements pertaining to a microbiological criterion are defined in “Introduction to Sampling Plans” below and also require the definition of necessary actions to be taken should a tested (lot) product exceed the established MC. These metrics have existed for a long time but have been, and are still, evolving over time. This is linked to the evolution of the management of food safety issues from a hazard-based to a risk-based approach. Thus, the evolving focus to manage such issues has led to the establishment by CAC of more-quantitative risk-based metrics outlined below.

Food Safety Objectives

The Food Safety Objectives (FSO) metric expresses the maximum frequency and/or concentration of a pathogen in a food at the time of consumption that provides or contributes to the appropriate level of protection defined by a government. This risk-based limit, established by competent authorities, should be achieved operationally within the food chain, while flexibility is given to in-

4.  Microbiological Criteria and Indicator Microorganisms dividual elements such as primary production or manufacturing. While the FSO represents the final outcome, it does not specify how individual steps will contribute.

Performance Objectives

The Performance Objectives (PO) metric represents an operational risk-based limit at a specific point in the food chain, i.e., the maximum concentration of a microbiological hazard in a food at the point in the food chain that should not be exceeded in order to achieve the established FSO. Since it contributes to achieving an FSO, a PO should be considered and established in the light of previous and subsequent steps and their corresponding POs. Industry may find it beneficial to establish its own POs, taking into account its position within the food chain, the impact of subsequent steps including the intended use of the end product, and the degree of risk it is willing to tolerate.

Performance Criterion

The Performance Criterion (PC) expresses the outcome achieved by a single or a combination of control measures, usually microbiocidal and/or microbiostatic control measures. This metric articulates the desired reduction and/or maximum acceptable increase of the hazard. The PC usually describes the required outcome to achieve the defined PO. The PC needs, therefore, to take into account variability of pathogens, food composition, and processing parameters along the segment of the food chain for which the PO has been established.

INTRODUCTION TO SAMPLING PLANS Traditionally and historically, microbiological criteria have been established to determine whether a lot of product was suitable for commercial distribution and consumption. Acceptability of such a product was defined as the compliance to requirements for certain microorganisms (including parasites) and/or their toxins or metabolites. Requirements have been expressed as absence or maximal numbers or concentrations of these parameters per unit(s) of mass, volume, area, or lot. Microbiological criteria have been and are still widely used to make a decision upon analysis of a food for defined parameters. The analytical results obtained are compared to the established requirements and serve as a basis to decide whether a lot is acceptable or needs to be rejected. The design of meaningful microbiological criteria used for lot acceptance is therefore a key step in the decision process. It must be considered that the development of such criteria is a complex process requiring the

83

appropriate knowledge and information, resources, and efforts. They should therefore be developed only when there are clear benefits and justified need and when they are effective and practical in serving their purpose (2). Microbiological criteria, if deemed necessary and justified, are established to allow assessment of one or more of the objectives listed below: • •

•

•

the safety of a (specific) food the hygienic quality of a food, in particular the adherence to good hygiene and manufacturing practices the suitability of a food or raw material for a particular usage such as its consumption or further processing the acceptability of a food or ingredient manufactured under unknown conditions

The groundwork for the establishment of microbiological criteria was initially developed during joint consultations of the World Health Organization (WHO) and Food and Agriculture Organization (FAO). Since then, the outcome of these consultations has been continuously evolving with the input of the ICMSF, national governments, and academic researchers. These efforts resulted in the preparation of a document on the principles for the establishment of microbiological criteria issued by CAC (2). The revision of this document was initiated in 2010 to take into account new developments in food safety management systems and in particular the MRM concepts. To fulfill the purpose of the objectives, a microbiological criterion needs to be established for a specific food or a defined category of food showing similar characteristics, taking into account a number of elements including the following: • •

•

•

•

• •

•

•

evidence of recognized or potential hazards to health microbiological status of the raw materials used to manufacture the food effect of processing on the microbiological status of the food likelihood and consequences of microbial contamination during processing likelihood and consequences of growth during subsequent handling, storage, and use of the food intended use of the food category(s) of consumer likely to be consuming the food cost-benefit associated with the application of the criterion point of the food chain at which the criterion will apply

A microbiological criterion is then established by translating the general principles above into specific

84 elements of microbiological criteria that should consist of the following: •

•

•

•

•

•

•

a statement of the microorganism of concern and/or its toxins/metabolites where applicable, as well as the reason for the concern microbiological limits considered as appropriate to fulfill the objective the specific point(s) of the food chain at which the criterion would apply the number of analytical units that need to comply with the established limits a sampling plan defining the number of field samples to be drawn and the procedures to sample and handle them the size of the analytical unit as well as the appropriate analytical methods (qualitative or quantitative) capable of providing the adequate response with respect to the limits actions to be taken when the criterion or individual elements thereof are not met

Microorganisms need to be relevant for the food in question and its process and, as stated above, may encompass bacteria, viruses, yeasts and molds, algae, parasitic protozoa, and helminths as well as their toxins or metabolites. Microbiological criteria include normally relevant pathogens, hygiene indicators, and/or spoilage organisms. The current approach in the formatting of recently issued microbiological criteria is to make a clear distinction between food safety-related parameters and process hygiene indicators (5, 7). The different elements constituting a microbiological criterion define the requirements and their stringency. They are usually called sampling plans and are characterized by the parameters n, c, m, and M, which are defined as follows: n is the number of samples that must be analyzed; c is the maximum allowable number of defective sample units in a two-class plan or of marginally acceptable sample units in a three-class plan; m is a microbiological limit that in a two-class plan separates good from defective quality or in a three-class plan separates good quality from marginally acceptable quality; and M is a microbiological limit that in a three-class plan separates marginally acceptable quality from defective quality. Sampling plans can be subdivided into attributes or variables plans. The first are the most appropriate when no or only little information is available on the performance of the process and hence its impact on the microbial flora. This is typically a situation encountered by regulatory authorities at a port of entry or by manufacturers purchasing raw materials from a new and thus

Factors of Special Significance poorly known supplier. Variables plans are suitable only when there is sufficient knowledge on the nature and frequency distribution of microorganisms within lots of a particular food. This type of plan would be suitable for manufacturers having a deep understanding of their process and thorough monitoring of their production, such as would be expected from a well-established hazard analysis and critical control point (HACCP) program. Attributes plans can be further subdivided into twoclass and three-class plans. Two-class plans are used to accept or reject individual lots based on the analysis of a specified number of sample units (n) using qualitative or quantitative criteria to discriminate lots and to make a decision regarding, for example, populations of a microorganism(s) above or below a defined concentration or presence/absence of a microorganism(s). While two-class plans are usually used for pathogens, this is not exclusive and this type of plan exists as well for indicators. An example is sampling plans for Enterobacteriaceae in infant and follow-up formulae with n = 10, c = 2, and m = 0 (in analytical units of 10 g) (5). The statistical analyses for evaluating two-class attribute-based sampling plans are different for presence/ absence data and for stratified quantitative data (18). Three-class plans are used in situations where the quality of the product can be divided into three different classes according to the concentrations of microorganisms within the sample units. While analytical results exceeding the limit m are undesirable, in a three-class plan a limited number of samples as defined by c can nevertheless be accepted. However, results exceeding the upper limit M make the product unacceptable. This is typically used when establishing criteria where quantitative data are stratified into three groupings. The statistical analyses for three-class plans are different from those of two-class plans (13).

PERFORMANCE OF SAMPLING PLANS Sampling plans for finished products are applied by authorities or by manufacturers to detect noncompliant lots of products, i.e., lots presenting a defect. The defect rate of manufactured food products relates to the actual fraction of servings that are contaminated (13). This defect rate is an important element to understand in the context of microbiological sampling and testing activities, whether they are applied to within-lot or to lot-by-lot testing. Given attributes plans have the ability to detect a given level of contamination and thus to reject a nonconforming lot of product with a certain probability.

4.  Microbiological Criteria and Indicator Microorganisms This ability is defined as the performance of the sampling plan and can be calculated using calculation sheets such as the ones published by the ICMSF (www.icmsf.org), and sampling plan performances are illustrated for the ICMSF cases in Table 4.1. As illustrated in Ta­ble 4.1, the performances of sampling plans show they have limitations in their ability to detect very low levels of contamination (i.e., defect rates). The more effective the implemented hygiene control measures are, the lower the defect rate will be, and the greater the number of samples that would need to be tested in order to increase the probability of detection. However, due to the destructive nature of microbiological testing, the required laboratory infrastructure and capacity, and the associated high costs, testing cannot be expanded indefinitely. An additional limitation of within-lot testing as performed by authorities, for example, at the port of entry or during distribution, is frequently the absence of knowledge and information on the conditions under which a particular food has been manufactured. In addition, such testing is performed only sporadically, and data are not necessarily used and managed in the most effective way. Examples of recently issued microbiological criteria with attendant probabilities have been published by CAC for ready-to-eat foods, powdered infant formulae, and bottled water (4–6). Details as well as additional examples are provided by Legan et al. (16), van Schothorst et al. (18), and ICMSF (14).

85

Microbiological criteria formulated as “absence” of one or more pathogens have been established by some governments and companies and represent a special case. Occasionally, such requirements do not even specify the target pathogen or, even more frequently, do not associate them to any specific sampling plan. As a consequence, such requirements do not express any tangible target or numerical tolerance. These types of criteria are not compatible with current MRM metrics and therefore do not provide any quantifiable target against which the effectiveness of implemented food safety measures could be compared and measured. The simple statement “absence” does not take into account that even sampling plans with c values of 0 and an extremely high number (n) of samples do not ensure the complete absence of a pathogen. A “c = 0” also does not take into account the discriminatory power of a sampling plan but emphasizes rather the wish of not detecting a specific pathogen in a given set of samples. For example, in the case of a limit of 5% lot defect rate, a sampling plan of n = 95, c = 1, m = 0 (in analytical units of 25 g) would be more stringent than a traditional sampling plan of n = 60, c = 0, m = 0 (in analytical units of 25 g), rejecting a greater number of defective lots. However, in addition to increasing the analytical burden, it would give the impression that the presence of the pathogen is being “tolerated,” despite the fact that the 95-sample plan is actually more protective.

Table 4.1  Example of the performance of sampling plans for the 15 ICMSF cases Cases, example of sampling plan and calculation of their performance at 95% probability of rejection (standard deviation = 0.4; geometric mean rounded to 2 significant figures). Calculations performed with spreadsheet NEWsampleplans2_05 (www.icmsf.org). Type of hazard

Conditions reduce hazard

Conditions cause no change in hazard

Conditions may increase hazard

Utility

Case 1 (n = 5, c = 3, m = 100, M = 1,000) 335 CFU/g

Case 2 (n = 5, c = 2, m = 100, M = 1,000) 220 CFU/g

Case 3 (n = 5, c = 1, m = 100, M = 1,000) 145 CFU/g

Indicator

Case 4 (n = 5, c = 3, m = 100, M = 1,000) 335 CFU/g

Case 5 (n = 5, c = 2, m = 100, M = 1,000) 220 CFU/g

Case 6 (n = 5, c = 1, m = 100, M = 1,000) 145 CFU/g

Moderate hazard

Case 7 (n = 5, c = 2, m = 100, M = 1,000) 220 CFU/g

Case 8 (n = 5, c = 1, m = 100, M = 1,000) 145 CFU/g

Case 9 (n = 10, c = 1, m = 100, M = 1,000) 80 CFU/g

Serious hazard

Case 10 (n = 5, c = 0, m = 0/25 g) 1 cell in 50 g

Case 11 (n = 10, c = 0, m = 0/25 g) 1 cell in 110 g

Case 12 (n = 20, c = 0, m = 0/25 g) 1 cell in 235 g

Severe hazard

Case 13 (n = 15, c = 0, m = 0/25 g) 1 cell in 170 g

Case 14 (n = 30, c = 0, m = 0/25 g) 1 cell in 360 g

Case 15 (n = 60, c = 0, m = 0/25 g) 1 cell in 740 g

Factors of Special Significance

86 Additional factors such as the nonhomogeneous distribution of contaminants and lack of technologies or procedures applied during processing to eliminate pathogens do not allow yet the manufacture of products such as raw meat or poultry totally free of pathogens. Considering these limitations, there is a need for a compromise between the desire to achieve “absence” of pathogens and the feasibility of reaching this goal as well as the ability to detect pathogens through the application of sampling plans. The limitations of sampling plans for finished products to ensure their safety as illustrated in this section emphasize the need to place microbiological testing in the broader framework of the overall food safety management system. To do so, it is certainly important to have a more complete picture of elements and factors contributing to the contamination of food products as discussed in the following section.

MICROBIOLOGICAL PROFILE OF A FOOD PRODUCT The microbiota of a food product, including commensals, hygiene indicators, and pathogens, is a result of the history of the product along the whole food chain. It is a dynamic system that evolves through the different steps of the food chain, from primary production through finished product to final preparation and consumption by the final user. The composition of the microbiota of the finished product as well as the levels of individual microorganisms or groups thereof is a function of the initial flora of individual raw materials, which is influenced and evolving at each step of the food chain due to

the conditions and factors contributing to their decrease or their increase. Elements contributing to changes of the microbiota are illustrated in Fig. 4.1 for industrially processed foods. Similar schemes and considerations can also be established for unprocessed or minimally processed foods as well as to other segments of the food chain such as primary production, distribution, marketing, or the preparation and handling by the final consumer in catering or institutional facilities or even at home. The fate of individual pathogens along a processing line as illustrated in Fig. 4.1, or any other type of food preparation scheme in the food chain, can be described using the conceptual formula defined by ICMSF (13): H0 − SR + SI £ FSO or PO In this conceptual formula, FSO is the Food Safety Objective and PO the Performance Objective; H0 is the initial level of the pathogen under consideration, SR is the total (cumulative) reduction that is achieved through one or several consecutive bactericidal steps, and SI is the total (cumulative) increase that can occur due to growth and/or recontamination. All elements of the formula are expressed in log10 units. While this conceptual formula is usually used to describe the fate of pathogens, it can very well be used to describe the fate of other microorganisms such as relevant hygiene indicators or spoilage organisms and hence reflect the hygienic performance of the process in question and the resulting microbiological quality of the product. With respect to the illustrated example, the impact of the raw materials submitted to a kill step, such as a

Figure 4.1  Schematic flow diagram of a food process, including raw materials (RM), a heat treatment (Heat), and food contact surfaces in the processing line (PL) as well as the processing environment located after the heat treatment (PE) up to the finished product (FP). doi:10.1128/9781555818463.ch4f1

4.  Microbiological Criteria and Indicator Microorganisms heat treatment, on the microbiota of the finished product will be determined by the type and levels of the initial microbiota and the magnitude of the kill step. In the case of raw materials added after the heating process, no further reduction will occur other than a possible quantitative modification related to the addition ratio of the different raw materials. Microorganisms such as pathogens or hygiene indicators present in the processing environment can lead to contamination of the intermediate product. This will occur either indirectly, from the production premises and external surfaces of equipment, or directly, through contaminated food contact surfaces. While direct contamination by food handlers is unlikely to occur in industrially manufactured products, particular events such as human interventions in the processing line can nevertheless contribute occasionally. Such direct contamination may become significant in productions requiring important human handling such as food service or catering operations. Production processes being dynamic, the occurrence of such contaminations may be continuous in certain cases and sporadic in others, and their importance will depend on the levels in the environment or in the processing lines. These elements will determine the occurrence and the distribution in the finished products. The consequences of such cross-contaminations will be the presence or an increased concentration of these microorganisms in intermediate and finished products. Microorganisms surviving a kill step or being introduced after this step may, given favorable intrinsic or extrinsic conditions, further increase due to growth at specific steps of the process, during distribution or during preparation by the final consumer.

PERFORMANCE OF CONTROL MEASURES Microbiological criteria have played and still play an important role in defining whether a food is acceptable or not and whether an individual lot complies with established limits. Sampling and testing are, however, rarely considered as an effective tool to control microbial hazards. The safety of food products, and hence the protection of consumers, can only be ensured through the implementation of effective control measures. These control measures encompass actions and activities used to prevent, eliminate, or reduce food safety hazards to an acceptable level. Control measures falling under good hygiene practices (GHP) applied by food manufacturers are usually generic by nature and therefore applicable to the manufacture of different product categories. They are

87

qualified as prerequisite programs (PRPs) and designed to minimize contamination from the processing environment (PE and PL in Fig. 4.1). Examples of control measures that are typically considered GHP measures are given below: •

• • •

• • •

Layout of processing lines and zoning to control flows of air, pressure conditions within premises, movement of personnel, and mobile equipment as well as raw and packaging materials or intermediate products. Equipment design and installation for cleanability. General cleaning and sanitization procedures. Preventive maintenance to minimize breakdowns and risks associated with unscheduled interventions. Appropriate waste management. Appropriate pest management. Training and behavior of personnel on PRPs, in particular, those related to microbiological considerations.

The application of PRPs is valid as well for practices applied in different parts of the food chain such as good agricultural practices upstream and good distribution practices downstream. These programs represent a mandatory basis for the implementation of specific control measures as defined in hazard analysis and critical control point programs such as Operational (OPP) and Critical Control Points (CCP). Specific control measures require a thorough validation in order to demonstrate their effectiveness in controlling relevant pathogens to the desired levels. Their performance during processing is determined through monitoring of relevant process parameters such as temperature, time, flow rates, pressure, etc., against critical limits established during validation. In this context, testing of finished products against microbiological limits is normally performed during the validation procedure as well as during the ongoing verification of the overall performance of the food safety management system. General control measures such as PRPs, as well as certain operational PRPs such as specific cleaning and sanitization measures, can be verified through microbiological testing. Such measures are usually designed to prevent postprocess contaminations occurring after specific control measures such as a heat treatment (CCP). PRPs are combinations of measures aiming, in the first place, at preventing the ingress of pathogens into areas close to the processing line. In case ingress has nevertheless occurred, the second line of defense is avoiding their establishment in processing areas and their subsequent multiplication and finally their dissemination into different parts of the plant and into the processing line, thus posing a direct threat to the finished product.

88 The establishment of testing plans for an appropriate verification of the effectiveness of control measures is more complex than the ones limited to finished products. They need to take into consideration all elements contributing to the presence of microorganisms and, in particular, of pathogens in the finished products. They will therefore have to include sampling plans for raw materials added after the control measure(s) and the processing environment and processing lines, i.e., product contact surfaces or intermediate product as appropriate. Environmental testing programs and the way to develop and apply them are discussed in detail by ICMSF (14). Specific examples have been provided by ICMSF (13) for different types of products, and the principles outlined details for ready-to-eat products and powdered infant formulae by CAC (4, 5). The first step is usually to determine a baseline of the microbiota, focusing on the relevant pathogen(s) and hygiene indicator microorganism(s) present in the processing environment and processing lines, either food contact surfaces and/or intermediate product. If no or only few historical data are available, it is important to collect this information through investigative sampling. Data for indicators can be obtained rapidly and at reduced costs and will provide sufficient initial information on the general hygiene status of the premises to allow developing a complementary sampling program for pathogens. This baseline represents the situation and levels under which the plant and process are considered to be under control and therefore the risk of a product contamination is minimal or negligible. This baseline is used as a benchmark to evaluate results and data generated through the sampling and testing activities defined in the sampling program against the established internal limits. This is done to verify and confirm that control is consistently maintained over time. In case findings demonstrate deviations such as the presence of a pathogen in one or more sampling points in areas critical to the process and exposure to contamination or levels of the hygiene indicator exceeding internal limits, the actions need to be taken to restore control. Under such conditions, the testing frequency of the finished product is usually tightened, and it becomes essential to perform a thorough investigative sampling to determine the source and origin of the contamination in order to take the appropriate measures.

INDEX AND INDICATOR MICROORGANISMS Attempts to use certain microorganisms or groups of microorganisms requiring only simple, more-rapid, and

Factors of Special Significance inexpensive methods to predict the presence of pathogens and to avoid expensive and lengthy testing have been made for more than 30 years. Initial work to define microorganisms able to “represent” pathogens was focused on water, and the concepts of model organisms or surrogate organisms (17) have been put forward by different authors such as Havelaar and Pot-Hagebom (8) and Payment and Franco (17). Over time, two different terms have been used to describe and define such microorganisms, namely, index organisms and indicator organisms, and they are well established today. The distinction between index and indicator organisms was discussed initially by M. Ingram in 1977 (1, 15). Index microorganisms have been defined as microorganisms, groups of microorganisms, or microbial metabolites whose presence in numbers exceeding a specified limit would indicate the possible presence of pathogens showing a similar behavior and ecology. The intended role and purpose of index organisms are therefore to be used as a direct predictor of the presence of a specific pathogen. Such a prediction, however, is possible only if a statistically valid correlation has been established between the organism used as index and the “associated” specific pathogen. This correlation is, however, difficult to establish, in particular for pathogens that are found only sporadically and at low levels of index organisms. Indicator microorganisms, on the other hand, have been defined as microorganisms, groups of microorganisms, or microbial metabolites whose presence in numbers exceeding specific limits would indicate a failure in the adherence to GHPs. The intended role and purpose of indicator organisms are thus to serve as an indirect predictor of the presence of a pathogen. It is therefore an indication of an increased risk related to a deviation of the implemented hygiene control measures. The direct testing for the specific pathogen cannot, however, be replaced by the sole testing for indicators, and combined programs including both of them are the rule. It should be noted that several publications discuss, along with hygiene indicators, microorganisms, groups of microorganisms, or metabolites qualified as quality indicators. However, such quality indicators are usually related to physicochemical or organoleptical parameters and hence quality attributes of a food product. While their presence and growth can be associated with defective control measures leading to spoilage of food products, they are, to our knowledge, not used to predict the presence of pathogens and hence not within the scope of this chapter. The use of indicators has been discussed by many authors, in particular in trying to define appropriate or ideal properties to fulfill their role both in raw materials, processing environment, and processing lines and in

4.  Microbiological Criteria and Indicator Microorganisms the end product. Several of the key properties are listed below (12) and should be evaluated for any organism or metabolite evaluated for its use as indicator: •

•

•

•

•

• • •

•

• • •

history of concomitant presence of indicator(s) and associated pathogen or its toxin presence usually at higher levels than associated pathogen presence indicative of an increased risk for faulty practices or faulty processes survival or stability similar to or greater than that of the target pathogen growth behavior similar to or faster than that of the hazard easily detectable and/or quantifiable Identifiable characteristics need to be stable. Methods for indicator organisms need to fulfill the same requirements as the one for pathogens, i.e., they need to be reliable and validated; in addition, they should be more rapid and less expensive. Quantitative results should show a correlation between indicator concentration and level of the pathogen. Results need to be applicable to process control. Analyst health is not at risk. Analytical method is suitable for in-plant use.

Taking into consideration that the behavior of a hygiene indicator should be similar to that of the associated pathogen, both will follow similar if not parallel patterns in the same processing facility. Although the outcome will be different, the conceptual formula of the ICMSF illustrated in “Microbiological Profile of a Food Product” above can be used to describe their fate and provide useful information and insight on whether the targeted FSO and PO are likely to be achieved or not. Numerous publications have been devoted to trying to determine the most appropriate hygiene indicator(s) for relevant pathogens, and it is not the purpose of this chapter to provide an exhaustive list of “ready-to-use” indicators. Enterobacteriaceae and Listeria spp. are certainly the most frequently used hygiene indicators in food processing and are associated with Salmonella and L. monocytogenes, respectively. This is also reflected in their inclusion in numerous microbiological criteria. They are, however, not the only ones, and others such as coliforms, fecal coliforms or Escherichia coli, fecal enterococci, sulfite-reducing clostridia (e.g., Clostridium perfringens), aerobic mesophilic counts, or certain metabolites such as ATP, thermonuclease, or alkaline phosphatases have been described as being suitable under certain circumstances.

89

The choice of an appropriate hygiene indicator will depend on the knowledge of a number of factors related to the products manufactured and, in particular, its microbiota, the microbiota of the raw materials, the processing conditions applied, and their effect on the microbiota, the food’s composition and characteristics and their effect on the behavior of microorganisms, the microbial ecology of the processing environment and lines, the aim of implemented control measures such as PRPs, and the effects of different types of cleaning and sanitization procedures, among others. An important but frequently underestimated or poorly managed aspect of environmental verification programs is the use and interpretation of generated data. Since hygiene indicators are associated to specific pathogens, their use is also associated with “predicting” and “anticipating” deviations and defects. It is therefore of utmost importance that the data generated, often at a higher frequency and in higher numbers than the ones for pathogens, are used in the most appropriate and beneficial way to fulfill their task. Failures in the interpretation and management of these data will jeopardize the purpose of verification testing and, in particular, of correcting deviations identified at an early stage.

References 1. Buchanan, R. L. 2000. Acquisition of microbiological data to enhance food safety. J. Food Prot. 63:832–838. 2. Codex Alimentarius Commission. 1997. Principles for the Establishment and Application of Microbiological Criteria for Foods. CAC/GL 21-1997. Food and Agriculture Organization, World Health Organization, Rome, Italy. 3. Codex Alimentarius Commission. 2007. Principles and Guidelines for the Conduct of Microbiological Risk Management (MRM). CAC/GL 63-2007. Food and Agriculture Organization, World Health Organization, Rome, Italy. 4. Codex Alimentarius Commission. 2007. Guidelines on the Application of General Principles of Food Hygiene to the Control of Listeria monocytogenes in Foods. CAC/ GL 61-2007. Food and Agriculture Organization, World Health Organization, Rome, Italy. 5. Codex Alimentarius Commission. 2008. Code of Hygienic Practice for Powdered Formulae for Infants and Young Children. CAC/RCP 66-2008. Food and Agriculture Organization, World Health Organization, Rome, Italy. 6. Codex Alimentarius Commission. 2011. Code of Hygienic Practice for Collecting, Processing and Marketing of Natural Mineral Waters. CAC/RCP-1985, rev. 2011. Food and Agriculture Organization, World Health Organization, Rome, Italy. 7. European Commission. 2007. Commission Regulation (EC) No 1441/2007 of 5 December 2007 amending Regulation (EC) No 2073/2005 on microbiological criteria for foodstuffs. Off. J. Eur. Union L 322:12–29.

90 8. Havelaar, A. H., and W. M. Pot-Hagebom. 1988. F-specific RNA bacteriophages as model viruses in water hygiene: ecological aspects. Water Sci. Technol. 20:399–407. 9. International Commission on Microbiological Spec­if­ ications for Foods. 1986. Microorganisms in Foods 2. Sampling for Microbiological Analysis: Principles and Specific Applications, 2nd ed. University of Toronto Press, Toronto, Canada. 10. International Commission on Microbiological Specifi­ cations for Foods. 1994. Choice of sampling plan and criteria for Listeria monocytogenes. Int. J. Food Microbiol. 22:89–96. 11. International Commission on Microbiological Specifi­ cations for Foods. 1997. Establishment of microbiological safety criteria for foods in international trade. World Health Stat. Q. 50:119–123. 12. International Commission on Microbiological Specifi­ cations for Foods. 1998. Potential application of risk assessment techniques to microbiological issues related to international trade in food and food products. J. Food Prot. 61:1075–1086. 13. International Commission on Microbiological Specifi­ cations for Foods. 2002. Microorganisms in Foods 7.

Factors of Special Significance Microbiological Testing in Food Safety Management. Kluwer Academic Publishers, New York, NY. 14. International Commission on Microbiological Specifi­ cations for Foods. 2011. Microorganisms in Foods 8. Use of Data for Assessing Process Control and Product Acceptance. Springer, New York, NY. 15. Jaykus, L. A., and P. McClure. 2010. Introduction of Microbiological Indicators in the Food Industry. Biomérieux University Notebook 01 (http://www.biomerieux-industry. com/upload/NoteBook-1.pdf Accessed 13 November 2011. 16. Legan, D. J., M. H. Vendeve, S. Dahms, and M. B. Cole. 2000. Determining the concentration of microorganisms controlled by attributes sampling plans. Food Control 12:137–147. 17. Payment, P., and E. Franco. 1993. Clostridium perfringens and somatic coliphages as indicators of the efficiency of drinking water treatment for viruses and protozoan cysts. Appl. Environ. Microbiol. 59:2418–2424. 18. Van Schothorst, M., M. H. Zwietering, T. Ross, R. L. Buchanan, M. B. Cole, and the International Commission on Microbiological Specifications for Foods. 2009. Relating microbiological criteria to Food Safety Objectives and Performance Objectives. Food Control 20:967–979.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch5

Shaun P. Kennedy Frank F. Busta

Biosecurity: Food Protection and Defense

Many food-related habits from prior generations and practices of ancient cultures, such as certain religious dietary guidelines or preferences for brewed or boiled fluids, have some basis in preventing foodborne illness. Yet the problem of food safety, as we currently understand it, is a relatively modern concern. Even today, food security, i.e., access to sufficient calories, is a dominant enough issue that food security and price volatility of foods were the focus of the 2011 World Food Day (43). Over 30 countries were assessed in 2011 as having at least some significant region with food insecurity (42). Global food system pressures have also had an impact on social unrest, best exemplified by the food riots of 2008 and the backdrop of food issues as part of the Arab Spring of 2011. At the same time, developed societies have moved from having concern over food security to enjoying a relative abundance of food. With this relative abundance of food in the more developed world, however, the possibility of becoming ill from the food itself has become a more significant concern. Sometimes an overall shift in relative health risks and the abundance of food resulting from improvements in the food supply chain drive new foodborne illness risks. This is especially true where the desire for quick-to-prepare, ready-to-eat

5

foods such as fresh fruits and vegetables has resulted in consumers being exposed to a broader array of potentially naturally contaminated foods from around the globe with limited food safety treatment options. The intentional contamination of food for economic gain or to cause harm and consequently the concern for food biosecurity also have a very long history. Historical examples of the use of contaminated food and water as weapons go back at least as far as the Athenians’ contamination of drinking water for the city of Kirrha of the Amphictyonic League with the plant root of Helleborus (Christmas rose) in 590 to 600 B.C., causing severe gastrointestinal illness that rendered the city defenseless against the ensuing attack. Contamination of wine with mandragora by the Carthaginian General Maharbal and various historical instances of plague-infested animal or human bodies dumped into water supplies continued as examples of intentional food contamination from the Roman times forward (72). During World War II, the Japanese Army experimented with the use of food for the delivery of pathogens such as Bacillus anthracis, Shigella spp., Vibrio cholerae, Salmonella enterica serovar Paratyphi, and Yersinia pestis (28, 50). It is only very recently, however, that protecting food

Shaun P. Kennedy and Frank F. Busta, National Center for Food Protection and Defense and College of Veterinary Medicine, University of Minnesota, 1954 Buford Ave., 120 LES, St. Paul, MN 55108.

91

92 from intentional contamination has become a significant concern on regional, national, and international levels. While food systems had been identified as vulnerabilities well beforehand (60, 90), the terrorist attacks in the United States on 11 September 2001 significantly elevated concerns about food system biosecurity (39). The intersection of incredibly efficient, highly integrated food supply systems, new and untraditional pathogens, and individuals or organizations potentially using those same very efficient systems to inflict public health or economic harm on a massive scale is the new face of food system biosecurity. This chapter examines the risk of intentional food contamination, primarily with microbiological agents, and provides an overview of current methods for evaluating risk and defining appropriate interventions.

HISTORICAL PERSPECTIVE There have been many intentional food contamination incidents, including those cited above, that involved the use of diseased animals, toxic plants and chemicals, or microorganisms as a malicious act. These have been generally localized and often not based on any detailed understanding of the contaminant introduced. The rising number of contamination events over the last 30 years (32, 74) suggests that there are increasingly more individuals and organizations that see the food system as a target of opportunity for attack and is parallel with the overall increase in terrorism events in general that can be tracked on the Global Terrorism Database (77). Some events have been politically motivated, such as the use of Salmonella by the Rashnishee cult in Oregon in 1982 (108). Others include those that have been economically motivated, such as the numerous instances of using pesticides/poisons in China. An example of a politically motivated and targeted biological event was the intentional contamination of an Australian-run commissary for Iraqi police with Salmonella that resulted in 350 casualties in 2006. This very low technology attack utilized a traditional foodborne illness vehicle, raw chicken, spoiled in this case intentionally, to contaminate the food being served to the police (6). The simplicity of the attack, however, is illustrative of the ability to utilize bacterial agents for intentional food contamination. This is not surprising, as we have been doing this accidentally on both family and large-population scales for all of history. In fact, to date, unintentional food contamination events provide the most striking examples of how certain foods could be used to affect a significant number of consumers, including geographically dispersed events. A 1994 outbreak of Salmonella

Factors of Special Significance serovar Enteritidis infections resulting from accidental contamination of ice cream in Minnesota involved an estimated 224,000 individuals (54), a 1985 outbreak of Salmonella serovar Typhimurium infections associated with milk in Illinois involved 60,000 consumers (92), and the Salmonella serovar Saintpaul outbreak associated with peppers sickened tens of thousands. The Peanut Corporation of America peanut paste contamination with Salmonella (19) and the Escherichia coli O104:H4 outbreak associated with fenugreek sprouts in Germany (40, 89) are examples of the scale of harm that could be caused, based on both the number of illnesses and the total amount of product contaminated. Such illustrations of the potential scope of a food system contamination event, combined with the knowledge that far more aggressive microorganisms, microbially derived toxins, or nonmicrobial agents could be used, drive the current emphasis on food system protection and defense. While biosecurity is generally associated with only normally occurring biological contaminants, effectively mitigating intentional contamination threats starts with a firm food safety foundation but requires significant additional considerations.

INTENTIONAL VERSUS UNINTENTIONAL CONTAMINATION Both intentional and unintentional food contaminations are of concern when the contamination can result in illness due to the lack of any further processing steps, including home preparation, to eliminate the deleterious impact of the contamination. This generally means that the contaminant is stable in the food, survives final preparation, and is not organoleptically obvious, i.e., provides no flavor or other sensory signals to suggest the presence of the contaminant. The subsequent foodborne illness event can result in either morbidity or mortality, along with a concomitant economic impact. For either intentional or unintentional contamination, the risk management control strategies include identifying foodcontaminant combinations of potential risk/vulnerability and then the insertion of controls to reduce the risk/vulnerability. The controls could be inserted at any point, from preharvest or preslaughter inputs through the point of consumption. A case in Taiwan in which high dioxin levels in pasture resulted in high dioxin content in milk illustrates how even environmental inputs can be of concern (67). For control of the more traditional biological contaminants of concern for food safety, the hazard analysis critical control point (HACCP) system (see chapter 42, this volume) is perhaps the most important example of the approaches that are in use. Intentional

5.  Biosecurity: Food Protection and Defense contamination, while it begins with HACCP principles of selection of potential contaminants and where and how they could be introduced to what foods and at what levels, necessitates other considerations. In traditional food safety, foodborne illness generally results from an overall system risk management failure that enables the introduction, survival, or growth of the contaminant to reach levels high enough to cause harm. In most cases, food safety issues arise primarily from equipment, process, or operator failure. This could be because not all reasonably foreseeable risks were identified (an important consideration in the United States, given that the Food Safety Modernization Act explicitly extends food safety requirements to “all reasonably foreseeable risks”), resulting in insufficient efforts to provide interventions to mitigate the risks. Extensive efforts undertaken across the food industry, academia, and government to identify all reasonably foreseeable risks and then develop control strategies have dramatically improved food safety, reducing the probability of a wide range of anticipatable contamination events. However, failures can and do occur across the food system, e.g., a refrigeration unit not holding temperature and thus allowing microbial growth, poultry process water cross-contamination, undercooked meat or poultry at retail or in the consumer’s home, or consumers eating known high-risk foods such as unpasteurized milk or raw oysters. Foodborne illness from intentional contamination results not from system failure, but instead from intentional attacks on a system that defeat the in-place controls. This could be because the controls and detection strategies in place for natural contamination could have identified and contained the intentional contamination but were actively overridden or bypassed. More worrisome, however, are intentional contamination attacks that succeed because the contamination “could not happen,” e.g., the agent is not normally present or of any realistic concern. Such scenarios provide limited incentive to firms for incurring the incremental cost and complexity to prevent them, increasing the potential for such an intentional contamination to cause harm. Metal detectors, magnets, and screens are designed to catch lowfrequency, accidental contamination with metal, glass, or other particulate agents and keep them out of the food supply. These same tools are also useful for catching intentional contamination events using similar materials (102). Intervention strategies for events that do not normally occur, however, not only are often economically unviable but also, in many cases, may not exist. If the microorganism or contaminant is not normally present at any stage in the preharvest through consumption sys-

93 tem, then there is usually very little published work on inactivation or detection strategies. This is understandable, as there was also little justification for such work, but concerns about food system terrorism have elevated in recent years. A World Health Organization report, Terrorist Threats to Food: Guidance for Establishing and Strengthening Prevention and Response Systems (122), and a U.S. Government General Accounting Office report, Food Processing Security (38), provide international and national reviews of the reasons for concern. Homeland Security Presidential Directive 9, Defense of United States Agriculture and Food (17), enacted on 30 January 2004, elevated the food and agriculture sector to the status of a critical U.S. infrastructure in need of significant efforts to protect it from intentional harm. These are among the drivers behind investigations into how to counter intentional contamination of food and water with pathogens and other agents that could cause significant and catastrophic morbidity, mortality, or economic harm. A sampling of microbial pathogens and toxins of concern are listed in Table 5.1, along with some of the potential chemical agents of concern provided for comparison; several references that go into significant detail on the agents of concern and illness progression (3, 36, 46, 63, 99) and food protection more broadly (87) are available. Some very virulent biological agents, e.g., smallpox virus or Ebola virus, are not listed in Table 5.1, given their limited stability outside their host as naturally expressed (49) and thus the anticipated difficulty of using them to contaminate food, although there are indications that some might be more feasibly stabilized than anticipated. In addition, should one stabilize such viruses, their efficiency as a weapon via delivery vehicles other than food may make their deployment as a weapon via food contamination less likely. In addition, while many food pathogens are readily accessible, the quality and safety control systems in place in foods in which they are a traditional safety concern would often be sufficient to minimize the risk of illness, but not if they are genetically modified. Recent events suggest that both conventional and modified foodborne illness agents still may be of concern, as a Jihadist post following the 2011 German sprouts outbreak illustrates: I say, and may Allah help us to success, the qualities of the E. coli, as well as the ability to develop it into biological weapon, bio-engineered in a laboratory, make the E. coli a most attractive candidate and a significant element in biological warfare, spreading violently, and killing silently, irritating the enemies and tearing their guts apart. (119).

Factors of Special Significance

94 Table 5.1  Potential food contamination agentsa Agent

LD50

Heat sensitivity

Solubility of toxin

Organoleptic

Microbiological agents and their toxins Alpha amanitin (109)

1.0 µg/kg of body wt in mice, oral (109)

MP, 255°C; thermally stable at normal cooking temp (57)

Water, alcohols (109)

NA

Bacillus anthracis

>108 spores in rabbits, oral (68)

Heat stable up to 159°C for 2 h (20)

Water, alcohols

No odor/taste (25)

Botulinum neurotoxin

0.001 µg/kg in mice, oral (34)

Anaerobic spores destroyed by boiling 10 min; moist heat at 120°C for 10 min destroys (34)

Water (51)

Current food contaminant

Clostridium perfringens epsilon toxin

0.1–5.0 µg/kg in mice, oral (34)

Freezing causes loss of viability

Water

Current food contaminant

Diacetoxyscirpenol

120 µg/kg in mice, oral (109)

Heat stable; MP, 161–162°C (109)

Moderately polar solvents (109)

NA

Francisella tularensis

108 CFU in humans, oral (34)

Heat sensitive, killed by heating at 55°C for 10 min (34)

Water

NA

Shiga toxin-producing E. coli

0.002 µg/kg in mice, oral (34)

Optimum growth, 37°C; viable growth range, 7–8 to 46°C

Water (20)

Current food contaminant

Staphylococcal enterotoxin

1–25 µg/kg in humans, oral Heat stable for minutes at (16) 100°C and resistant to freezing (34)

Water (20)

T-2 toxin

1,210 µg/kg in mice, oral (34)

MP, 151–152°C, thermally stable at normal cooking temp and times–decomposes at 816°C in 30 min (34)

Soluble in lower alcohols and polar solvents (109)

No odor/taste (69, 101)

Yersinia pestis

100 CFU, oral (18)

Inactivated at 55°C for 15 min (34)

Water (20)

Prior contamination

Abrin

0.04 µg/kg in mice, oral (34)

Heat stable to 60°C for 30 min (109)

Slightly soluble in water (25)

NA

Aconitine

100 µg/kg in mice, oral (34) MP, 204°C (109)

Water at 310 µg/ml (109)

Conotoxins

5 µg/kg in mice, oral (34)

NA

Water (20)

No odor or taste (109)

Cyanide

30 µg/kg in rats, oral (27)

Heating can release irri­ tating or toxic gases. MP, −13.4°C (hydrogen cyanide); MP, 563.7°C (sodium) (27)

Water at 48 g/100 ml at 10°C (27)

Faint bitter almond odor, odorless when dry, slight odor of HCN when moist (27)

Digoxin

177.8 µg/kg in humans, oral 76.7 µg/kg in mice, oral (109)

Heat stable, decomposes at 235°C (109)

Insoluble in water, soluble in dilute alcohols (109)

Odorless, bitter taste (57)

Fluoroacetic acid

5 µg/kg in rats, oral; 7 µg/ kg in mice, oral (109)

MP, 35.2°C; boiling point 165°C (109)

Soluble in water at 50 µg/ml at 25°C

Odorless powder (109)

Other agents for comparison

5.  Biosecurity: Food Protection and Defense

95

Table 5.1  Potential food contamination agentsa (Continued) Agent

LD50

Heat sensitivity

Solubility of toxin

Organoleptic

Nicotine sulfate

85.5 µg/kg in mice, oral; 83.0 µg/kg in rats, oral (109)

Decomposes on heating, producing toxic fumes (57)

Water, alcohol

None for crystals, aqueous solution has tobacco/fishy odor (109)

Organophosphates (as a class)

150 µg/kg in rats, oral; 600 µg/kg in mice, oral (109)

Varies by specific toxin but general MP is 35–36°C; generally thermally stable to cooking (109)

Water, 55 µg/ml at 20°C (109)

Range of no odor to pungent garlic like (109)

Picrotoxin

150 µg/kg in rats, oral (109)

MP, 203°C; thermally stable to cooking (109)

Water at 3,030 µg/ml (ambient temp), 20% boiling (109)

Odorless, bitter taste (27)

Ricin

3–5 µg/kg in mice, oral (34) Stable but detoxifies at 80°C (10 min) and 50°C (50 min) (109)

Water (109)

No odor/taste (109)

Saxitoxin

263 µg/kg in mice, oral (109)

Heat stable, cooking does not remove (109)

Water and methanol (109)

No odor/taste (109)

Strychnine

2,350 µg/kg in rats, oral 20 µg/kg in mice, oral (109)

MP, 275–285°C (109)

Water at 20% (109)

Odorless, pure material has a bitter metallic taste and bitter taste in solution containing 1.4 µg/ml (27)

Tetrodotoxin

8.0 µg/kg in mice, oral 200 µg/kg in cats, oral (34)

Darkens above 220°C without decomposing (109)

Soluble in diluted acetic acid, slightly soluble in water (109)

No odor (109)

a

LD50, 50% lethal dose; MP, melting point; NA, not applicable. References are indicated in parentheses.

When introduced into foods in which they are not normally associated or introduced in the production system at a point after standard processing steps that would inactivate them, conventional food pathogens pose a significant risk. If the food-pathogen combination is not typical, the quality and safety systems for that food may not adequately mitigate the risk. For example, the detection system may not work for the food-pathogen combination, or the postcontamination processing may not be sufficient to inactivate the pathogen. In addition, uncommon associations of illness, pathogen, outbreak profile, and food could slow the public health system response as it tries to find the source of the contamination (110), a possibility demonstrated by, as only one well-documented example, the misdiagnosis of the cause of a botulism outbreak in 1963 (64). In addition to slow epidemiologic attribution, the potential for delayed diagnosis of the disease itself due to unfamiliarity of many physicians with the disease presentation has been reported in other contexts, such as zoonotic diseases seen more commonly in one region than another, as is the case with hantavirus

in the United States (31, 70). The most likely result of conventional food pathogens used for intentional contamination will be high morbidity and relatively lower mortality. Conventional pathogens are therefore not the only focus of efforts to prevent catastrophic public health consequences through intentional contamination of food. It should also be noted that conventional foodborne pathogens could be used to great effect in causing significant economic harm if, through an intentional contamination resulting in significant morbidity, even if low mortality, the event still results in significant loss of public confidence in a segment of the food system or in the government’s ability to protect the public. Microorganisms and their toxins not normally associated with food, especially those listed as select agents by the Centers for Disease Control and Prevention (26), are of much greater concern. Many of these agents are more difficult to produce at the level of purity and in populations or volumes necessary to result in a significant contamination of the food system than some of the conventional foodborne pathogens. In addition, these

96 agents are now more tightly controlled and regulated in most developed countries, so that obtaining significant quantities can be difficult. However, several characteristics make both select agents and genetically engineered microorganisms potentially more attractive pathogens for intentional contamination of the food system. Such microorganisms and their toxins are significantly more infective or toxic than typical foodborne pathogens, requiring much less material for a successful attack. There are also typically no inactivation processes or detection systems developed and validated for detecting or quantitating them in foods. Consequently, the pathogen could be consumed at infective/toxic doses without detection. At first disease presentation, the initial health care providers’ diagnosis of potential select agents could be more limited than expected (35), some studies suggesting a successful identification rate as low as 10%, due to lack of familiarity with the disease presentation (31). Outbreak investigations in which bioterrorism was considered have had lag times in Centers for Disease Control and Prevention notification from 2 to 27 days of problem identification (5). In the early stages of an outbreak investigation, the fact that many of these agents are not generally associated with food could mean that public health officials would have to evaluate a wide range of possible sources of the causative vehicle beyond food, and this would dramatically slow any response. The combination of high infectivity and toxicity, unproven inactivation and detection systems, and uncertain public health system response, along with the shock value of the agents themselves, could, for some groups, make select agents and genetically engineered microorganisms more attractive as terrorism agents, even given the significantly higher technical expertise potentially required to utilize them. There may be cases, however, where the anticipated higher degree of technical difficulty in utilizing a select agent may be overestimated if the original assessment did not include enteral exposure. Intranasal administration options for therapeutics are more limited than oral options, while occasionally more effective, because of effective delivery challenges that translate to threat agents as well (106). In addition, the purification of agents necessary for aerosol delivery of threat agents may actually be counterproductive for food system delivery effectiveness if the nonpurified form provides some level of protection for the agent over the purified form, as is the case for certain spores. The potential for a spore to survive or germinate from the point of introduction through consumption, including potential terminal treatments should it germinate, is not broadly understood in foods that were not the initial focus of things like retort processing (88, 105, 123, 124).

Factors of Special Significance Because the list of potential agents is extensive and varied, evaluation of intentional food contamination vulnerabilities presents a different set of concerns than preventing accidental contamination. First, there is the overall compatibility of the microorganism with the food, i.e., its technical attractiveness. If the pathogen is stable in the food matrix and it is not inactivated by conventional processing, it could be of concern. This includes food and microorganism combinations in which the food supports growth of the microorganism. In addition, the ease with which the microorganism can be mixed into the food product and the readiness with which it can be dispersed and remain visually unidentifiable are important considerations. If the population of the microorganism necessary to provide an infective dose turns a clear fluid product turbid, for example, that food-microorganism combination is not a likely threat scenario. Just as important are other organoleptic impacts of the microorganism or contaminant on the food item. Negative public health consequences would be avoided if consumers would not eat sufficient product due to noticeable differences in expected organoleptic quality, in much the same way as conventional spoilage by microorganisms renders food inedible, thereby possibly preventing illness that could be caused by pathogens that had also grown in the food. In the case of an intentional contamination of ground beef with nicotine sulfate in Michigan, the burning sensation of the contaminant was one of the clues during the investigation (13). Early detection would be expected for foods with off flavors or any other significant organoleptic shift. The potential risk based on characteristics of the specific food, i.e., the relative attractiveness it might have as a vehicle for delivery of the microbial agent based on compatibility factors, is another important consideration. Large batch sizes, thorough mixing after contamination, and the absence of any postcontamination terminal treatment are factors that increase vulnerability due to the amount of product that could be contaminated with a harmful dose of the pathogen. The speed of distribution of the product and the rate of consumer consumption also have a direct impact on the severity of the public health consequences from the intentional contamination. The fear-and-rage impact of the event is also affected by any preferential consumption by more vulnerable populations, especially children (100). Beyond such specific technical considerations, another aspect of the food that could increase its vulnerability for intentional contamination is the lingering potential impact on consumer confidence and the economy. If the food or commodity is targeted in a way that maximizes consumer concerns about the near- or long-term abil-

5.  Biosecurity: Food Protection and Defense ity to protect that food item from future attacks, then recovery from the attack would be slower and therefore result in a more vulnerable food or commodity. Since it is unlikely that any attack would be significant enough to negatively impact the total availability of food in developed countries, any economic impacts due to consumption changes are at the firm, manufacturer, item, commodity, or category level, reflecting a shift of purchasing patterns. The $200 million loss to the Chilean grape industry resulting from suspected intentional contamination of grapes with cyanide in 1989 (44) and the deepening of the bankruptcy challenges of a restaurant chain following an outbreak of hepatitis A associated with eating food items containing green onions at the restaurant (81) are examples of the economic impacts of food incidents at the sector or firm level. More recently, the Salmonella serovar Saintpaul outbreak associated with produce cost Florida tomato growers over $100 million (112). The Peanut Corporation of America Salmonella contamination resulted in the largest recall in U.S. history and an estimated $1 billion in industry losses (19). Similar to any natural disaster, there are also real economic losses resulting from the event itself, affecting lives and future productivity, and these could be substantial. There also could be significant trade implications if the attack resulted in a restriction of trade due to trading partners’ concerns about the safety of the food type, something more commonly considered in animal and meat trade and the official animal disease status, e.g., the foot-and-mouth disease status of a given country. Intentional contamination of a particular food becomes more attractive as a target as the magnitude of various economic consequences increases. It is important to note that an intentional contamination event could result in no direct or significant public health consequences and yet result in significant or catastrophic economic harm, as was the apparent case in Chile.

INTENTIONAL FOOD CONTAMINATION RISK MANAGEMENT Intervention strategies to prevent attacks on the food system face different challenges than traditional food safety efforts; decisions have to be made on investments and efficacy without the regular natural occurrences of an event. In such cases, stakeholders are understandably reluctant to invest to protect against such events (10). Addressing the economic models that could be useful is beyond the scope of this chapter, but there are technical considerations on how to evaluate intervention strategies that are useful to review. Although potentially difficult to validate, measuring the success of an intervention

97 to reduce terrorism risk can be achieved in the absence of any ability to demonstrate failed attacks. The general approach is to evaluate the degree to which the intervention reduces the vulnerability of each subsystem within the overall infrastructure of concern (121). In the critical infrastructure of the food system, there are multiple approaches currently being used to assess vulnerability or risk at all levels, from the single-unit level to national infrastructure and global supply systems. Independent of the risk assessment approach, it is relatively straightforward to evaluate the success of a particular intervention. First, one analyzes the vulnerability of the particular section or operation within the food system prior to applying the intervention. Then, either after deploying the intervention or in a theoretical construct, the section or operation in the food system is analyzed again. If the vulnerability of the food system section or operation is reduced, then the intervention can be considered successful. This does not, however, answer whether or not the intervention is economically justified, which represents an entirely new set of analyses. Given their widespread use by federal and state agencies, operational risk management (ORM) and “criticality, accessibility, recuperability, vulnerability, effect, and recognizability” (CARVER [see below]) variations are worth reviewing as risk assessment tools for food system operations. In addition, there are many new tools being developed by academia and the private sector either for risk assessment or for development of food system protection plans that could be of use (1, 78, 79).

Operational Risk Management

ORM is a tool that originated from efforts in the U.S. National Air and Space Agency (NASA) and U.S. Department of Defense (DoD) to reduce the risk of failure of aircraft, space missions, and weapons. It is still actively used today by units such as the U.S. Naval Safety Center (118) and was adopted by the U.S. Food and Drug Administration (FDA) Center for Food Safety and Nutrition for early food system risk assessments (38). In total, ORM is a five-step process for identifying and managing risks: (i) identify the hazards; (ii) assess the potential consequence of the hazards; (iii) determine which risks to manage with which interventions; (iv) implement the interventions; and (v) assess the success of the interventions and modify as necessary. Traditionally in ORM, risk is defined as a function of the severity of the failure and the probability of the failure. Since intentional contamination is not a stochastic event, there is not a predictive function for estimating the probability. For food defense, probability can best be considered as the probability of success if an

Factors of Special Significance

98 appropriately skilled person or group tried to contaminate the food system. The severity and probability are evaluated on separate scales. For any food item, commodity, facility, unit operation, or other definable subset of the food system, one can conduct this analysis and come up with a location on an assessment grid, which compares the severity against the probability to allow focusing on those events with the highest combination of both. The basic scales used in ORM as they could be applied to the food system are noted in Table 5.2. The general approach is to select the food system(s) of concern, identify the points of vulnerability (hazards), and conduct a severity versus probability assessment for the vulnerability. Given the breadth of information necessary to complete an ORM assessment, it generally requires a team of experts on the food system or facility, with additional facilitators with detailed knowledge of the ORM approach and any threat information that might not be available to the users. This may include specific details on microbial or chemical contaminants of concern. Importantly, successful utilization of ORM does not require sensitive details on potential agents. Only general characteristics need to be introduced, such as determination of how much of a contaminant could reasonably be added at a particular stage of the food supply chain and potential mitigation steps downstream of that stage such as thermal treatments or filtration. Depending on the result, the facilitator or expert can narrow down the scope of the potential vulnerability. Through a simple charting or scoring exercise, one can then evaluate each of the vulnerabilities and rank them as to those of greatest risk of allowing a catastrophic event. Intervention efforts can then be prioritized for implementation to reduce the system vulnerability. The ORM process then comes full circle with the implementation of the chosen intervention(s), evaluating the success of the intervention, and repeating the risk assessTable 5.2  Operational risk management semantic scales Severity scale

Probability scale

Very high (catastrophic public health impact, high morbidity and mortality)

Very high (system continuously vulnerable)

High (significant public health impact, primarily morbidity, some mortality)

High (system regularly vulnerable)

Medium (some morbidity, no mortality)

Medium (system sporadically vulnerable)

Low (no real impact)

Low (system seldom vulnerable) Very low (system rarely vulnerable)

ment process. This cycle is repeated until the resulting risk reaches an acceptable level. A representation of the ranking grid that could be used for this purpose is presented in Table 5.3.

CARVER+Shock

CARVER, like its further refinement, CARVER+Shock, is a strategy for completing risk assessments in use by both FDA and the U.S. Department of Agriculture (USDA) in analyzing points of vulnerability (25, 30). CARVER+Shock is adapted from DoD evaluation systems that were developed to identify targets of greatest effect on an enemy as well as those of greatest vulnerability (58, 62). The Strategic Partnership Program Agroterrorism, a joint effort by the FDA, USDA, Department of Homeland Security (DHS), and the Federal Bureau of Investigation, in partnership with industry, utilized CARVER+Shock to assess food and agriculture vulnerabilities (117). CARVER+Shock is comprised of seven elements that are utilized to evaluate the vulnerability of a food operation by analyzing each node within a system. While the detailed Strategic Partnership Program Agroterrorism results are classified, in each of the nearly 40 assessments there were vulner­abilities identified for the use of biological agents as threat agents. Modified to fit the concerns presented by intentional food contamination, the CARVER+Shock elements have been defined as follows: Criticality: The degree to which the public health or economic consequences are nationally significant. High scores equate to catastrophic morbidity, mortality, or economic harm. Accessibility: Physical access to target, the ability of the perpetrator to gain access to the point of contamination and escape undetected. Recuperability: Overall system resiliency as measured by the time required to bring the system back into operation, with low scores for only days to recover and high scores for recovery going a year or longer. Vulnerability: Attack feasibility as viewed by the potential for a successful attack. This includes both the ability to introduce enough of the material of concern to cause harm and the potential for subsequent processing to reduce that risk. Effect: Direct loss from the attack as defined by the fraction of the food system that has been impacted by the attack. Recognizability: Ease of target identification is a measure of the degree of specialized knowledge needed in order to identify the point for the intentional contamination.

5.  Biosecurity: Food Protection and Defense

99

Table 5.3  Operational risk management: grid for categorizing Probability Severity

Very high

High

Medium

Low

Very high

++++

++++

++++

+++

+

High

++++

+++

+++

++

+

Medium

+++

++

++

+

+

Low

++

+

+

+

+

Shock: Combined health, economic, and psycho­ logical impacts of the attack, which is a measure of the overall impact. Importantly, the economic and psychological impacts of an attack may not require any morbidity or mortality if they result in substantial lack of public confidence in the food system or the government. To apply CARVER+Shock, the first step is to define the scales to use for each of the rating elements. In selection of the scales, it is very important to avoid potential overweighting of any individual component. The scales developed by FDA and USDA implementation have been modified from the original DoD scales to render them more applicable to the food system. To allow for sufficient discrimination, 10-point (1 through 10) semantic scales are used for each of the rating categories. For the FDA/USDA evaluations, the 10-point scale is aggregated into five semantic rating groups, giving the evaluator the option of a “high” and a “low” score for each semantic rating. Using the chosen scales, the next step is to evaluate the food facility or operation in order to identify unique nodes. These nodes, when taken in total, should represent the entirety of the production system, while they individually represent known or likely differences in vulnerability. For the model system depicted in Fig. 5.1,

Very low

there would be 32 initial nodes for evaluation. Each of the unit operations shown, such as blending or air injection, and each transfer between unit operations represent unique nodes for risk assessment. Each node for a system is then evaluated against each of the CARVER+Shock semantic scales, and a score for each element is assigned. As noted for the utilization of ORM, a team of experts is generally required to complete this evaluation. A composite score is compiled for the food system or for the section of the food system that is under consideration. These ratings allow for a rational comparison of vulnerabilities across portions of a food system, as the nodes with the highest scores represent a greater vulnerability and thus those of greatest potential value for implementing interventions or counter­ measures. While generally utilized to assess a specific food system or facility, the same strategy could be utilized to make comparisons across multiple food systems to identify those with the greatest need for interventions or countermeasures, although the results are not quantitatively comparable. The selection of interventions, implementation, and assessment could then be conducted in an ORM-like process. For ease of use and for further details, downloadable versions of CARVER+Shock are available from FDA (114). Assessing the overall risk through ORM is simpler than with CARVER+Shock in that it has only two rating

Figure 5.1  General ice cream production process flow diagram. doi:10.1128/9781555818463.ch5f1

100 elements for ranking risk, severity, and probability and requires less training. Conversely, however, it may also require greater expertise in the specific food system under consideration. In some cases, a desirable strategy may be to use CARVER+Shock to identify the food system, facility/operation, or node of greatest concern and then allow those who work within that system or facility/operation to use ORM to minimize specific vulnerabilities. One resource in examining options for minimizing vulnerabilities is the FDA Mitigation Database (115).

National Risk and Criticality Assessments

DHS leads biennial risk assessments across agencies for biological, chemical, radiological, and nuclear threats, including the Biological Terrorism Risk Assessment (BTRA). The first BTRA in 2006 included a limited number of food systems in the assessment but still resulted in significant consequence considerations for food system threats. DHS reacted to critique of the methodology of the BTRA (29) through a series of revisions for the 2008 and 2010 updates of the BTRA. The BTRA is a probabilistic assessment approach that attempts to incorporate aggressor intent and capabilities, ease of agent acquisition, ease of agent introduction, probability of agent survival through to target populations and probable public health (primarily) consequences. For each of these and other steps in the overall event tree, probabilities are assigned. As is true for assessments with ORM, CARVER+Shock, or other tools, one of the challenges for the BTRA is its use of probabilistic techniques for deterministic events and the fact that no one approach yet fully addresses the terrorism risk assessment needs (15, 41, 84). The BTRA and its companion assessments utilize the intelligence community to deal with part of this by getting their input into probable motives, goals, and capabilities of terrorist individuals and organizations. Similar to other expert-opinion-based assessments, the assessments are only as good as the subject matter experts, the parameterization of their inputs, and the available information of potential aggressor capabilities and objectives. While the BTRA has significantly improved in both the range of foods considered and the experts accessed in each biennial assessment, even with the Chemical Terrorism Risk Assessment (CTRA) expanding to 10 foods for 2012, it is challenging to represent the full range of risk represented by potential intentional contamination of the food system. What is clear is that, even without access to the actual assessments but only considering the range of potential consequences that could be derived from available supply chain information, in each of the foods considered in the assessments microbial agents could be used to devastating effect.

Factors of Special Significance Given the challenges of performing data- and resourceintensive risk assessments across all critical infrastructures, DHS and its sister agencies have tried to focus their limited available resources by working with the states to identify which critical infrastructure systems, elements, subsystems, and nodes are most critical. Through a reporting system each year in the United States, states submit a list of nominated elements to DHS for consideration as critical infrastructure of national significance (in 2010 termed “Level 1/Level 2”). Prior to 2010, no food and agriculture elements or nodes had been accepted as Level 2, but through successful deployment of the Food and Agriculture Criticality Assessment Tool (FASCAT) (76), in 2010 nearly 10% of the nation’s designated critical infrastructure was food and agriculture. While the actual detailed assessments are not publicly available, the tool used to understand the scale of consequences and infrastructure impact, including guidance on its use, is available online (www. foodshield.org). In nominating a food and agriculture system as critical, the states are now required to provide a scenario for how that system could potentially be compromised. This has significantly included the use of microbiological threat agents as intentional contaminants into food products.

Food System Interventions

Independent of the system used to identify which food system, facility, or operation is vulnerable to intentional contamination, the first levels of interventions are basic security considerations. Standard perimeter and access control, often referred to as “guns, gates, and guards” (14, 93, 111), represents the most basic level, but more extensive efforts including facility and operation­specific physical and human security considerations are required. Important for their successful implementation, any intervention needs to have a clear technical justification (73). Many of these are already normal practices for most food operations, such as restricting access to locations within a facility to those with a need for access in order to complete their task. Human resource functions become much more important, because dealing with disgruntled workers and screening out intentional terrorists attempting to become employees with access to points of high vulnerability are important and necessary interventions. A major challenge in implementing appropriate human resources controls is ensuring their compliance with personal privacy and civil rights regulations. There are many other facility- or operation-specific interventions that may be considered, which are independent of a specific microbiological or chemical agent. Many of these are detailed in the various guidance documents

5.  Biosecurity: Food Protection and Defense developed by the FDA and the USDA to assist food system firms in developing a food security system as well as the FDA mitigation database (23, 45). When considering defense against contamination with microbial agents that may be of unique concern to a specific food item, the characterization of the food itself, the process and safety/quality systems already in place, and final retail/food service/consumer handling practices must also be considered. If the agent has a negative impact on the basic performance of the food, then it is not a combination at risk. In some cases, foods that would seem similar have very different impacts on the viability of the agent in the process, with B. anthra­ cis having very different survival rates in sugared egg yolks versus salted egg yolks (61). Efforts to see if micro­ biological toxins might interfere with basic product attributes have established some that do, such as sodium arsenite, sodium cyanide, and sodium nitrite causing color changes in fruit juices. There are many others that do not, such as ricin, a-amanitin, botulinum neuro­ toxin, picrotoxin, and staphylococcal enterotoxin B, which have no negative impact on yogurt fermentation (86, 116). If the microbial agent has a specific sensitivity to pH, the product either may not be at risk normally from that agent or it may be possible to shift the pH enough to reduce the long-term survivability of the agent. The same could be true for other physiochemical characteristics of the food. More likely, however, is that a food processing modification, such as microfiltration to remove spores (107), or an inhibitor, such as nisin (75), could be introduced to reduce or eliminate the activity of the microbial pathogen so that even if it is introduced into the product, it will not result in morbidity or mortality. Pasteurization is routinely used to control normally prevalent or anticipated accidental or natural microbial contamination in many foods and is effective against a number of the threat agents of concern, including Francisella tularensis and Y. pestis (assessed using Yersinia pseudotuberculosis) (115). Either the current pasteurization processes or more aggressive versions could be applied to certain highly vulnerable foods to mitigate potential harm from their being used to deliver specific agents. This obviously requires that the organoleptic quality of the food not be compromised through the process. In the United States, a majority of the dairy industry has done this already for some of their products to reduce the vulnerability of those foods by in­creasing the pasteurization schedule (37), but this approach is not effective across all similar threat agents, with B. an­ thracis spores requiring higher temperatures or longer times than botulinum neurotoxin (80, 124). Similar interventions could be applied to other foods to mitigate

101 their vulnerability, including processing inter­ventions, e.g., high-pressure processing and ionizing radiation (65). High-pressure conditions utilized for control of Listeria monocytogenes would also be effective for a 5log reduction of Y. pseudotuberculosis strain 197 and F. tularensis (96, 116). In all cases, however, it is important to design the food system such that intentional contamination after the insertion of additional interventions is extremely unlikely or is readily detectable/traceable (59), so that the food does not become a more attractive target than it was originally. Detection and diagnostics, both new technology platforms and integration of new combinations of platforms and systems, are fields of intense interest and activity in the broadly defined area of homeland security. As applied to the food and agriculture critical infrastructure, detection and diagnostic tools have the potential to prevent or contain intentional contamination events. Just as in the case of food safety, however, detection and diagnostic systems have limitations. The traditional challenges of sampling strategies, preanalytical processing, and time to detection are amplified, while concerns about false positives and false negatives are also elevated when applied to detecting contamination events that should not happen. This requires new approaches on how to evaluate the utility of any detection intervention, with regard to both its feasibility to contain the unlikely event and the cost of implementation. An overview of technologies currently available for use in analyzing food items for potential contamination can be found in a range of articles (2, 4, 11, 12, 21, 22, 47, 95, 97, 98). Novel concepts being explored include magnetic and fluorescent multiplexed beads (85); laser desorption/ ionization (8); sandwich immunoassays (125); resequencing arrays (104); mammalian cell biosensors (9); aptamers (66); Raman spectroscopy (52); and a range of techniques based on nanoparticles (82, 83, 91, 94) and PCR (53, 71). The following presents an approach for evaluating detection and diagnostic system intervention options that takes the principles of “detect to warn,” “detect to treat,” and “detect to recover” described by former Rear Admiral Richard Danzig (33) and translates them into a mental construct for use in the food system of detect to prevent, detect to protect, and detect to respond or recover. “Detect to prevent” means an overall strategy that enables positive confirmation of contamination before the finished food item leaves the facility in order to eliminate any chance of public health consequences. This can be at the farm level, such as field-packed produce, all the way through to food processing facilities or large-

102 scale commissaries, and at any point in between. The firm incurs the economic cost of the detection strategy, any further economic impact from actual contamination events being limited to disposal of contaminated product and facility decontamination. “Detect to protect” still strives to prevent any public health consequences from intentional food contamination, but after the food has left the facility of concern. While the time frame varies greatly based on the food item, there is a lag time from when food leaves any harvest, processing, or handling facility to when it is purchased for consumption by consumers. If the detection system can enable the supply chain to maintain control and prevent the sale of contaminated products, then public health and large-scale economic consequences are averted. The firm still has an economic loss, both direct, from moving product through the supply chain, lost sales, and disposal costs, and indirect, through customer and consumer loss of confidence for not preventing or containing the contamination before distribution. “Detect to recover” is the response strategy to enable rapid identification of intentional contamination that has moved into consumer consumption in order to quickly contain the event and minimize its impact, which could also be called “detect to regret.” Here the economic consequences are far more significant, and there are important to catastrophic public health consequences. In many ways, this is similar to how current foodborne illness response systems work. When food safety issues are not identified prior to distribution and consumption, resulting in a foodborne illness outbreak, our public health and food systems work to identify and contain the outbreak while learning from it to prevent future outbreaks. If the agent is broadly present at an infective or lethal level, however, the consequences from an intentional event would be far more dire than those from historical natural foodborne illness outbreaks. In anticipation of this, there have been efforts on using vaccines as a mitigation strategy, but this would require a broad range of vaccines, as the threat agent will not be known in advance (7). In this case, the economic losses go beyond the firm or impacted commodity to include losses due to illness and overall macro­ economic impacts, including reduction in long-term economic growth (103). In all cases, often overlooked are the decontamination and disposal requirements for the contaminated food. While for some agents there are approved sanitizers that can be used, when tested in the context of the food matrix the use concentration allowed is insufficient, with actual levels required being as much as 1 order of magnitude higher than the labeled allowed-use level (55, 56).

Factors of Special Significance To illustrate how these three approaches can be applied, Fig. 5.2 shows a generic process model for egg processing incorporated into the USDA–Food Safety and Inspection Service Model Food Security Plan for Egg Processing Facilities guidance document (113). While detection technologies could be deployed during or after each unit operation shown, such as storage or grading, the arrows indicate potentially preferred points for inserting a detection intervention. The decision of where to insert a detection intervention is based on factors such as the ease of obtaining a representative sample (balance tanks better than grading), the potential risk of contamination at that point (washing versus breaking), and proximity to the terminus of a set of sequenced operations. To use the diagram in Fig. 5.2 to design a detection strategy for any food item, one needs additional information beyond the identified points of potential detection. The time to detection, specificity, and sensitivity of the

Figure 5.2  Generic egg processing model. doi:10.1128/9781555818463.ch5f2

5.  Biosecurity: Food Protection and Defense technique need to be combined with an understanding of the velocity of the food item from that point forward, so that the detection intervention can be categorized as a detect-to-prevent, detect-to-protect, or detect-to-recover strategy. In some cases, the most practical strategy for use of detection technologies could be a rapid technique for identifying suspect food items soon enough to allow diversion prior to final loss of supply chain owner control, followed by a highly specific technique to ensure that the suspect contaminant is present. These technical challenges are in addition to the difficulties in establishing sampling strategies, acquiring samples, and preanalytical sample preparation. While the likelihood of higher levels of the agent from intentional contamination than from unintentional contamination with a foodborne pathogen reduces sampling uncertainty challenges, this likelihood does not eliminate such challenges (120). Whatever detection strategy is selected, however, the economic considerations will be paramount. The cost of the detection system implementation itself could be a significant barrier to industry adoption and could require regulatory or customer actions to drive use of the technologies. Beyond those costs are the costs of false positives. These costs drive the need for total detection system false-positive rates to be significantly lower than for most of the individual detection technologies now available. The scope of food production is such that if the scale of detection desired is the fluid bulk truck, for example, then a false-positive rate of 1/100,000 would result in nearly 40 false positives a year in something like fluid milk. If that false positive could not be rapidly confirmed, it would result in the investigation of 40 potential terrorism events every year. Reflecting the high demands on food defense detection were the original performance targets for the Food Biological Agent Detection System, a DHS program, which included the following: multiplexed benchtop unit; total sample testing cycle time of £20 minutes; limit of detection (toxin) of 0.04 nanograms per milliliter or (microbial) 1 cell per gram; false-positive rate of £1 in 1,000,000; false-negative rate of £1 in 1,000; cost per instrument, <$50,000 per unit; and cost of consumables, £$5.00 per multiplexed sample. While all of these targets are challenging individually, the combination is not yet achievable, even though they are now out of date. Initially targeted for botulinum neurotoxin detection in fluid milk, the time to result was set to be in line with the antibiotic residue tests that are done on every tanker of milk received at processing plants, the preferred location for vulnerability reduction. Since then, the leading residue test (Charm Sciences, Inc.) has reduced its time to result to

103 2 minutes, setting the bar higher still as industry manages operations to the new, more rapid standard. In summary, ensuring that the United States and the world benefit from the overarching goal of an abundant, nutritious, and safe food system is more complex now that intentional contamination is a very real and credible threat. The overall food safety risk management systems that have developed over the last few decades serve as a sound and necessary foundation for approaches that will protect the food system from intentional abuse, but they are not sufficient. HACCP and other food safety systems designed to control naturally occurring microbial pathogens cannot anticipate and protect against individuals committed to using the food system as a means of public health or economic terrorism without significant modification. Approaches to harden the food system against intentional contamination and enhance its inherent biosecurity are being adapted from ORM, CARVER, CARVER+Shock, and other systems from military and law enforcement experience and new concepts in intelligent adversary modeling, among others. Countering the potential risks posed by intentional use of microbiological pathogens or other contaminants requires a continual reassessment and optimization of the food system. While similar to the continual improvement approach dictated by food safety systems to prevent unintentional microbiological contamination of food products, food system defense represents an additional set of challenges. These include recognizing that a successful attack on the food system would not necessarily result in any direct morbidity or mortality. Relevant concerns cover the full range from basic microbiology to food system supply chain management to legal protection of potentially sensitive private sector information. Ensuring that the food system is not intentionally compromised will require that all aspects be addressed, with food microbiology being a crucial, but not the only, element.

References 1. AIB International. https://www.aibonline.org/auditservices/­ index.html. Accessed 21 October 2011. 2. Akbulut, D., K. A. Grant, and J. McLauchlin. 2004. Development and application of Real-Time PCR assays to detect fragments of the Clostridium botulinum types A, B, and E neurotoxin genes for investigation of human foodborne and infant botulism. Foodborne Pathog. Dis. 1:247–257. 3. Arnon, S., R. Schechter, T. V. Inglesby, D. A. Henderson, J. G. Bartlett, M. S. Ascher, E. Eitzen, A. D. Fine, J. Hauer, M. Layton, S. Lillibridge, M. T. Osterholm, T. O’Toole, G. Parker, T. M. Perl, P. K. Russell, D. L.

Factors of Special Significance

104

4.

5.

6.

7.

8.

9. 10.

11.

12. 13.

14.

15. 16. 17.

Swerdlow, and K. Tonat. 2001. Botulinum toxin as a biological weapon. Medical and public health management. JAMA 285:1059–1070. Arranz, I., W. R. G. Baeyens, G. Van Der Weken, S. De Saeger, and C. Van Peteghem. 2004. HPLC determination of fumonisin mycotoxins. Crit. Rev. Food Sci. Nutr. 44:195–203. Ashford, D. A., R. M. Kaiser, M. E. Bales, K. Shutt, A. Patrawalla, A. McShan, J. W. Tappero, B. A. Perkins, and A. L. Dannenberg. 2003. Planning against biological terrorism: lessons from outbreak investigations. Emerg. Infect. Dis. 9:515–519. Associated Press. 2006. Insurgent suspected of mass poisoning of Iraq police, The Australian, 11 October 2006. www.theaustralian.news.com.au/story/0,20867, 20560653-2703,00.html. Accessed 19 October 2011. Atkins, H. S., M. Morton, K. F. Griffin, M. G. M. Stokes, J. P. Nataro, and R. W. Titball. 2006. Recombinant Salmonella vaccines for biodefence. Vaccine 24:2710–2717. Ayyadurai, S., C. Flaudrops, D. Raoult, and M. Drancourt. 2010. Rapid identification and typing of Yersinia pestis and other Yersinia species by matrixassisted laser desorption/ionization time-of-flight (MALDITOF) mass spectrometry. BMC Microbiol. 10:285. doi:10.1186/1471-2180-10-285. Banerjee, P., and A. K. Bhunia. 2010. Cell-based biosensor for rapid screening of pathogens and toxins. Biosens. Bioelectron. 26:99–106. Bazerman, M. H., and M. D. Watkins. 2004. Predictable Surprises: The Disasters You Should Have Seen Coming and How To Prevent Them. Harvard Business School Press, Cambridge, MA. Bennett, R. W. 2005. Staphylococcal enterotoxin and its rapid identification in foods by enzyme-linked immunosorbent assay-based methodology. J. Food Prot. 68:1264–1270. Blyn, L. B. 2006. Biosensors and food protection. Food Technol. 60:36–41. Boulton, M., M. Stanbury, D. Wade, J. Tilden, D. Bryan, J. Payne, and B. Eisenga. 2003. Nicotine poisoning after ingestion of contaminated ground beef. MMWR Morb. Mortal. Wkly. Rep. 52:413–416. Brooks, L. F. 2004. Testimony, Committee on Govern­ ment Reform, Subcommittee on National Security, Emerg­ ing Threats, and International Relations. Washington, DC. http://www.gpo.gov/fdsys/pkg/CHRG-108hhrg/html/ CHRG-108hhrg96313.htm. Accessed 8 August 2012. Brown, G. G., and L. A. Cox, Jr. 2011. How probabilistic risk assessment can mislead risk analysts. Risk Anal. 31:196–204. Burows, W. D., and S. E. Renner. 1999. Biological warfare agents as threats to potable water. Environ. Health Perspect. 107:975–984. Bush, G. W. 2004. Homeland Security Presidential Direc-­ tive/HSPD-9: Defense of United States Agriculture and Food. http://www.dhs.gov/xabout/laws/gc_1217449547 663.shtm. Department of Homeland Security, Washington, DC. Accessed 21 October 2011.

18. Carus, W. S. 2001. Bioterrorism and Biocrimes: The Illicit Use of Biological Agents Since 1900. Center for Counterproliferation Research, National Defense University. Washington, DC. 19. Cavallaro, E., K. Date, C. Medus, S. Meyer, B. Miller, C. Kim, S. Nowicki, S. Cosgrove, D. Sweat, Q. Phan, J. Flint, E. R. Daly, J. Adams, E. Hyytia-Trees, P. GernerSmidt, R. M. Hoekstra, C. Schwensohn, A. Langer, S. V. Sodha, M. C. Rogers, F. J. Angulo, R. V. Tauxe, I. T. Williams, and C. Barton Behravesh. 2011. Salmonella Typhimurium infections associated with peanut products. N. Engl. J. Med. 364:601–610. 20. CBWInfo. 2006. Factsheets on chemical and biological warfare agents. http://www.cbwinfo.com/intro.html. Accessed October 2011. 21. Cebula, T., E. W. Brown, S. A. Jackson, M. K. Mammel, A. Mukherjee, and J. E. LeClerc. 2005. Molecular applications for identifying microbial pathogens in the post9/11 era. Expert Rev. Mol. Diagn. 5:432–445. 22. Cebula, T. A., S. A. Jackson, E. W. Brown, B. Goswami, and J. E. LeClerc. Chips and SNPs, bugs and thugs: a molecular sleuthing perspective. J. Food Prot. 68:1271–1284. 23. Center for Food Safety and Applied Nutrition. 2005. USDA, FDA and DHS sign agreement with NASDA to make nation’s agriculture and food supply more secure. http://www.fda.gov/NewsEvents/Newsroom/Press ­Announcements/2004/ucm108357.htm. Accessed 21 October 2011. 24. Center for Food Safety and Applied Nutrition. Guidance documents. http://www.fda.gov/Food/GuidanceComplia nceRegulatoryInformation/GuidanceDocuments/default. htm. Accessed 21 October 2011. 25. Centers for Disease Control and Prevention. Bioterrorism agents/diseases. http://www.bt.cdc.gov/agent/agentlist. asp. Accessed 21 October 2011. 26. Centers for Disease Control and Prevention. 2006. HHS and USDA Select Agents and Toxins: 7 CFR Part 331, 9 CFR Part 121, and 42 CFR Part 73. http://www.select­ agents.gov/Select%20Agents%20and%20Toxins%20L ist.html. Accessed 21 October 2011. 27. Centers for Disease Control and Prevention, Agency for Toxic Substances and Disease Registry. 2006. http:// www.atsdr.cdc.gov/. Accessed 21 October 2011. 28. Christopher, G. W., T. J. Cieslak, J. A. Pavlin, and E. M. Eitzen. 1997. Biological warfare: a historical perspective. JAMA 278:412–417. 29. Committee on Methodological Improvements to the Department of Homeland Security’s Biological Agent Risk Analysis, National Research Council. 2008. Department of Homeland Security Bioterrorism Risk Assessment: A Call for Change. The National Academies Press, Washington, DC. 30. Conner, C. F. 2006. Testimony of the Honorable Charles F. Conner, Deputy Secretary, United States Department of Agriculture Before the U.S. Senate Committee on Agriculture, Nutrition, and Forestry Subcommittee on Research, Nutrition and General Legislation. http:// www.usda.gov/documents/DepSecTest_jan09.pdf. Accessed October 21, 2011.

5.  Biosecurity: Food Protection and Defense 31. Cosgrove, S. E., T. M. Perl, X. Song, and S. D. Sisson. 2005. Ability of physicians to diagnose and manage illness due to category A bioterrorism agents. Arch. Intern. Med. 165:1–8. 32. Dalziel, G. R. 2009. Food defence incidents 19502008: a chronology and analysis of incidents involving the malicious contamination of the food supply chain. http://www.rsis.edu.sg/cens/publications/reports/ RSIS_Food%20Defence_170209.pdf. Accessed 21 October 2011. 33. Danzig, R. 2003. Catastrophic Bioterrorism—What Is To Be Done? Center for Technology and National Security Policy, National Defense University, Washington, DC. 34. Darling, R. G., and J. B. Woods. 2004. Medical Management of Biological Casualties Handbook, 5th ed. USAMRID, Fort Detrick, MD. 35. Dembek, Z. F., R. L Buckman, S. K. Fowler, and J. L. Hadler. 2003. Missed sentinel case of naturally occurring pneumonic tularemia outbreak: lessons for detection of bioterrorism. J. Am. Board Fam. Pract. 16:339–342. 36. Dennis, D. T., T. V. Inglesby, D. A. Henderson, J. G. Bartlett, M. S. Ascher, E. Eitzen, A. D. Fine, A. M. Friedlander, J. Hauer, M. Layton, S. R. Lillibridge, J. E. McDade, M. T. Osterholm, T. O’Toole, G. Parker, T. M. Perl, P. K. Russell, and K. Tonat. 2001. Tularemia as a biological weapon: medical and public health management. JAMA 285:2763–2773. 37. Detlefsen, C. 2005. Dairy industry vigilant in addressing food security. Cheese Market News 25:1. 38. Dyckman, L. J. 2003. Food-Processing Security: Voluntary Efforts Are Under Way, but Federal Agencies Cannot Fully Assess Their Implementation. GAO (03342). U.S. General Accounting Office, Washington, DC. 39. Dyckman, L. J. 2003. Bioterrorism: A Threat to Agriculture and the Food Supply. GAO (04-259T). U.S. General Accounting Office, Washington, DC. 40. European Food Safety Authority. 2011. Shiga toxinproducing E. coli (STEC) O104:H4 2011 outbreaks in Europe: taking stock. EFSA J. 9(10):2390. 41. Ezell, B. A., S. P. Bennett, D. von Winterfeldt, J. Sokolowski, and A. J. Collins. 2010. Probabilistic risk analysis and terrorism risk. Risk Anal. 30:575–589. 42. Food and Agriculture Organization of the United Nations. 2011. Food security statistics. http://www. fao.org/economic/ess/ess-fs/en/. Accessed 21 October 2011. 43. Food and Agriculture Organization of the United Nations. 2011. World Health Day. http://www.fao.org/ getinvolved/worldfoodday/en/. Accessed 21 October 2011. 44. Reference deleted. 45. Food Safety and Inspection Services. 2005. Food Security and Emergency Preparedness Security Guidelines. http://www.fsis.usda.gov/food_defense_&_emergency_ ­response/index.asp. Accessed 21 October 2011. 46. Franz, D. R., P. B. Jahrling, A. M. Friedlander, D. J. McClain, D. L. Hoover, R. W. Bryne, J. A. Pavlin, G. W. Christopher, and E. M. Eitzen. 1997. Clinical recognition and management of patients exposed to biological warfare agents. JAMA 278:399–411.

105 47. Garber, E. A., R. M. Eppley, M. E. Stack, M. A. McLaughlin, and D. L. Park. 2005. Feasibility of immunodiagnostic devices for the detection of ricin, amanitin, and T-2 toxin in food. J. Food Prot. 68:1294–1301. 48. Reference deleted. 49. Harper, G. J. 1961. Airborne microorganisms: survival test with four viruses. J. Hyg. 59:479–486. 50. Harris, S. H. 2003. Japanese biomedical experimentation during the World-War-II era, p. 463–506. In D. E. Lounsbury and R. F. Bellamy (ed.), Military Medical Ethics, vol. II. Office of the Surgeon General, Department of the Army, United States of America, Washington, DC. 51. Hauschild, A. H. W., and K. L. Dodds. 1993. Clostridium botulinum. Ecology and Control in Foods. Marcel Dekker, Inc., New York, NY. 52. He, L. B. Deen, T. Rodda, I. Ronningen, T. Blasius, C. Haynes, F. Diez-Gonzalez, and T. P. Labuza. 2011. Rapid detection of ricin in milk using immunomagnetic separation combined with surface-enhanced Raman spectroscopy. J. Food Sci. 76(5):N49–N53. 53. He, X., S. McMahon, T. A. McKeon, and D. L. Brandon. 2010. Development of a novel immuno-PCR assay for detection of ricin in ground beef, liquid chicken egg, and milk. J. Food Prot. 73:695–700. 54. Hennessy, T. W., C. W. Hedberg, L. Slutsker, K. E. White, J. M. Besser-Wiek, M. E. Moen, J. Feldman, W. W. Coleman, L. M. Edmonson, K. L. MacDonald, and M. T. Osterholm. 1996. A national outbreak of Salmonella Enteritidis infections from ice cream. N. Engl. J. Med. 334:1281–1286. 55. Hilgren, J., K. M. J. Swanson, F. Diez-Gonzalez, and B. Cords. 2009. Inactivation of Yersinia pseudotuber­ culosis, as a surrogate for Yersinia pestis, by liquid biocides in the presence of food residue. J. Food Prot. 72:392–398. 56. Hilgren, J., K. M. J. Swanson, F. Diez-Gonzalez, and B. Cords. 2009. Susceptibilities of Bacillus subtilis, Bacillus cereus, and avirulent Bacillus anthracis spores to liquid biocides. J. Food Prot. 72:360–364. 57. IPCS Inchem. Chemical Safety Information from Intergovernmental Organizations. http://www.inchem­ .org. Accessed 21 October 2011. 58. Jones, T. S. 2005. NAVMC Directive 3500.86. http:// ebookbrowse.com/navmc-dir-3500-86-pdf-d102 394294. Accessed 8 August 2012. 59. Jotcham, R. 2005. Authentication, antitamper, and track-and-trace technology options to protect foods. J. Food Prot. 68:1314–1317. 60. Kahn, A. S., D. L. Swerdlow, and D. D. Juranek. 2001. Precautions against biological and chemical terrorism directed at food and water supplies. Public Health Rep. 116(1):3–14. 61. Kahn, S. A., K. Sung, M. S. Nawaz, C. E. Cerniglia, M. L. Tamplin, R. W. Phillips, and L. C. Kelley. 2009. The survivability of Bacillus anthracis (Sterne strain) in processed liquid eggs. Food Microbiol. 26:123–127. 62. Kavanaugh, J. J. 1996. Air Force Handbook 31-302. U.S. Air Force. U.S. Department of Commerce, National Technical Information Service, Springfield, VA.

Factors of Special Significance

106 63. Klietmann, W. F., and K. L. Ruoff. 2001. Bioterrorism: implications for the clinical microbiologist. Clin. Microbiol. Rev. 14:364–381. 64. Koenig, M. G., A. Spickard, M. A. Cardella, and D. E. Rogers. 1964. Clinical and laboratory observations on Type E botulism in man. Medicine 43:517–545. 65. Lado, B. H., and A. E. Yousef. 2002. Alternative foodpreservation technologies: efficacy and mechanisms. Microbes Infect. 4:433–440. 66. Lamont, E. A., L. He, K. Warriner, T. P. Labuza, and S. Sreevatsan. 2011. A single DNA aptamer functions as a biosensor for ricin. Analyst 136:3884–3895. 67. Lee, T. 2005. Dioxin in local cows’ milk meets with EU standard, EQPF says. The China Post http://www. chinapost.com.tw/news/2005/09/28/69302/Dioxin-in.htm 68. Lincoln, R. E., J. S. Walker, F. Klein, A. J. Rosenwald, and W. I. Jong. 1967. Value of field data for the extrapolation in anthrax. Food. Proc. 26:1558–1562. 69. Locasto, D., M. Allswede, and T. M. Stein. 2005. CBRNE - T-2 Mycotoxins. http://www.emedicine.com. Accessed 21 October 2011. 70. MacNeil, A., T. G. Ksiazek, and P. E. Rollin. 2011. Hantavirus pulmonary syndrome, United States, 19932009. Emerg. Infect. Dis. 17:1198. 71. Matero, P., T. Pasanen, R. Laukkanen, P. Tissari, E. Tarkka, M. Vaara, and M. Skurnik. 2009. Real-time multiplex PCR assay for detection of Yersinia pestis and Yersinia pseudotuberculosis. APMIS 117:34–44. 72. Mayor, A. 2004. Greek Fire, Poison Arrows & Scorpion Bombs: Biological and Chemical Warfare in the Ancient World, p. 99–118, 145–169. The Overlook Press, Peter Mayer Publishers, Inc., Woodstock, NY. 73. Miller, A. J., C. L. Hielman, S. Droby, and N. Paster. 2005. Science and technology based countermeasures to foodborne terrorism: introduction. J. Food Prot. 68:1253–1255. 74. Mohtadi, H., and A. Murshid. 2006. A global chronol­ ogy of incidents of chemical, biological, radioactive and nuclear attacks: 1950-2005. http://www.ncfpd.umn. edu/Ncfpd/assets/File/pdf/GlobalChron.pdf. Accessed 21 October 2011. 75. Montville, T. J., T. De Siano, A. Nock, S. Padhi, and D. Wade. 2006. Inhibition of Bacillus anthracis and potential surrogate bacilli growth from spore inocula by nisin and other antimicrobial peptides. J. Food Prot. 69:2529–2533. 76. National Center for Food Protection and Defense. 2011. Food Ag Sector Criticality Assessment Tool. http://www. ncfpd.umn.edu/index.cfm/research/fascat/.Accessed 21 October 2011. 77. National Consortium for the Study of Terrorism and Responses to Terrorism. 2011. Global Terrorism Database. http://www.start.umd.edu/gtd/. Accessed 21 October 2011. 78. National Infrastructure Institute: Center for Infrastructure Expertise. http://www.ni2cie.org/CARVER2.asp. Accessed 21 October 2011. 79. Newkirk, R. 2010. A risk and vulnerability assessment methodology for food systems. DHS Science Conference— Fifth Annual University Network Summit, 30 March to

80.

81.

82.

83.

84. 85.

86.

87. 88.

89.

90.

91.

92.

1 April 2011. http://www.orau.gov/dhssummit/2010/­ presentations/March10/Panel06/newkirk_ryan.pdf. Accessed 21 October 2011. Novak, J. S., J. Call, P. Tomasula, and J. B. Luchansky. 2005. An assessment of pasteurization treatment of water, media, and milk with respect to Bacillus spores. J. Food Prot. 68:751–757. Oltmanns, R. 2004. Puts & calls: A Tale of Two Vegetables (OK, a Veggie and a Fruit). Corporate communications key in times of crisis. Crisis Manager, 1 September 2004. http://www.bernsteincrisismanagement. com/nl/crisismgr040901.html. Pal, S., and E.C. Alocilja. 2009. Electrically active polyaniline coated magnetic (EAPM) nanoparticle as novel transducer in biosensor for detection of Bacillus an­ thracis spores in food samples. Biosens. Bioelectron. 24:1437–1444. Pal, S., E. Setterington, and E. C. Alocilja. 2008. Electrically-active magnetic nanoparticles for concentrating and detecting Bacillus anthracis spores in a direct-charge transfer biosensor. IEEE Sensors J. 8:647–654. Parnell, G. S., C. M. Smith, and F. I. Moxley. 2010. Intelligent adversary risk analysis: a bioterrorism risk management model. Risk Anal. 30:32–48. Pauly, D., S. Kirchner, B. Stoermann, T. Schreiber, S. Kaulfuss, R. Schade, R. Zbinden, M.-A. Avondet, M. B. Dorner, and B. G. Dorner. 2009. Simultaneous quantification of five bacterial and plant toxins from complex matrices using a multiplexed fluorescent magnetic suspension assay. Analyst 134:2028–2039. Principato, M., T. Boyle, J. Njoroge, R. L. Jones, Jr., and M. O’Donnell. 2009. Effect of thermal processing during yogurt production upon the detection of staphylococcal enterotoxin B. J. Food Prot. 72:2212–2216. Rasco, B. A., and G. E. Bledsoe. 2005. Bioterrorism and Food Safety. CRC Press, Boca Raton, FL. Rasooly, R., and P. M. Do. 2010. Clostridium botuli­ num neurotoxin type b is heat-stable in milk and not inactivated by pasteurization. J. Agric. Food Chem. 58:12557–12561. Robert Koch Institute. 2011. Report: Final Presentation and Evaluation of Epidemiological Findings in the EHEC O104:H4 Outbreak, Germany 2011. Robert Koch Institute, Berlin, Germany. Robertson, R. E. 1999. Food Safety: Agencies Should Further Test Plans for Responding to Deliberate Contamination. GAO(RCED-00-3). U.S. General Accounting Office, Washington, DC. Rong-Hwa, S., T. Shiao-Shek, C. Der-Jiang, and H. Yao-Wen. 2010. Gold nanoparticle-based lateral flow assay for detection of staphylococcal enterotoxin B. Food Chem. 118:462–466. Ryan, C. A., M. K. Nickels, N. T. Hargrett-Bean, M. E. Potter, T. Endo, L. Mayer, C. W. Langkop, C. Gibson, R. C. McDonald, R. T. Kenney, N. D. Puhr, P. J. McDonnell, R. J. Martin, M. L. Cohen, and P. A. Blake. 1987. Massive outbreak of antimicrobial­resistant salmonellosis traced to pasteurized milk. JAMA 258:3269–3274.

5.  Biosecurity: Food Protection and Defense   93. Salerno, R. M., and J. G. Koelm. 2002. Biological Laboratory and Transportation Security and the Biological Weapons Convention. SAN No:20021067P. Sandia National Laboratories, Albuquerque, NM.   94. Sanvicens, N., C. Pastells, N. Pascual, and M.-P. Marco. 2009. Nanoparticle-based biosensors for detection of pathogenic bacteria. Trends Anal. Chem. 28:1243–1252.   95. Sapsford, K. E., C. R. Taitt, N. Loo, and F. S. Ligler. 2005. Biosensor detection of botulinum toxoid A and staphylococcal enterotoxin B in food. Appl. Environ. Microbiol. 71:5590–5592.   96. Schlesser, J. E. 2009. Inactivation of Yersinia pseu­ dotuberculosis 197 and Francisella tularensis LVS in beverages by high pressure processing. J. Food Prot. 72:165–168.   97. Sharma, S. K., B. S. Eblen, R. L. Bull, D. H. Burr, and R. C. Whiting. 2005. Evaluation of lateral-flow Clostridium botulinum neurotoxin detection kits for food analysis. Appl. Environ. Microbiol. 71:3935–3941.   98. Sharma, S. K., and R. C. Whiting. 2005. Methods for detection of Clostridium botulinum toxin in foods. J. Food Prot. 68:1256–1263.   99. Sidell, F. R., E. T. Takafuji, and D. R. Franz. 1997. Medical Aspects of Chemical and Biological Warfare. Office of the Surgeon General, Department of the Army, United States of America, Washington, DC. 100. Slovic, P. 1987. Perception of risk. Science 236:280–285. 101. Stark, A.-A. 2005. Threat assessment of mycotoxins as weapons: molecular mechanisms of acute toxicity. J. Food Prot. 68:1285–1293. 102. Stier, R. 2003. The dirty dozen: ways to reduce the 12 biggest foreign materials problems. Food Safety Magazine 03:6–7. 103. Stinson, T. 2007. The national economic impacts of a food terrorism event: initial estimates of indirect costs, p. 145–158. In H. W. Richardson, P. Gordon, and J. E. Moore II (ed.), The Economic Costs and Consequences of Terrorism. Edward Elgar, Cheltenham, United Kingdom. 104. Taitt, C. R., A. P. Malanoski, B. Lin, D. A. Stenger, F. S. Ligler, A. W. Kusterbeck, G. P. Anderson, S. E. Harmon, L. C. Shriver-Lake, S. K. Pollack, D. M. Lennon, F. Lobo-Menendez, Z. Wang, and J. M. Schnur. 2008. Discrimination between biothreat agents and ‘near neighbor’ species using a resequencing array. FEMS Immunol. Med. Microbiol. 54:356–364. 105. Tamplin, M. L., R. Phillips, T. A. Stewart, J. B. Luchansky, and L. C. Kelley. 2008. Behavior of Bacillus anthracis strains Sterne and Ames K0610 in sterile raw ground beef. Appl. Environ. Microbiol. 74:1111–1116. 106. Therapeutic Intranasal Drug Delivery. http://intranasal .net/overview/default.htm. Accessed 19 October 2011. 107. Tomasula, P. M., S. Mukhopadhyay, N. Datta, A. Porto-Fett, J. E. Call, J. B. Luchansky, J. Renye, and M. Tunick. 2011. Pilot-scale crossflow-microfiltration and

107 pasteurization to remove spores of Bacillus anthracis (Sterne) from milk. J. Dairy Sci. 94:4277–4291. 108. Torok, T. J., R. V. Tauxe, R. P. Wise, J. R. Livengood, R. Sokolow, S. Mauvais, K. A. Birkness, M. Skeels, J. Horan, and L. R. Foster. 1997. A large community outbreak of salmonellosis caused by intentional contamination of restaurant salad bars. JAMA 278:389–395. 109. TOXNET. 2005. Toxicology Data Network. http:// www.toxnet.nlm.nih.gov. National Library of Medicine, National Institutes of Health, Bethesda, MD. Accessed 21 October 2011. 110. Treadwell, T. A., D. Koo, K. Kuker, and A. S. Kahn. 2003. Epidemiologic clues to bioterrorism. MMWR Morb. Mortal. Wkly. Rep. 118:92–98. 111. Tucker, J. B. 2003. Biosecurity: Limiting Terrorist Access to Deadly Pathogens. Peaceworks 52. U.S. Institute of Peace, Washington, DC. 112. Tumin, T. 2009. Visualizing food safety: seeing the linkages in a networked world. A Fire Under Embers. http://blogs.law.harvard.edu/fireunderembers/. Accessed 21 October 2011. 113. U.S. Department of Agriculture—Food Safety and Inspection Service. 2005. Model Food Security Plan for Egg Processing Facilities. http://www.fsis.usda.gov/PDF/ Model_FoodSec_Plan_Eggs.pdf. Accessed 21 October 2011. 113a. U.S. Food and Drug Administration. 2005. Food security awareness training course. http://www.fda.gov/ Training/ForStateLocalTribalRegulators/ucm120929. htm. Accessed 21 October 2011. 114. U.S. Food and Drug Administration. 2011. CARVER+­ Shock http://www.fda.gov/Food/FoodDefense/CARVER/ default.htm. Accessed 19 October 2011. 115. U.S. Food and Drug Administration. 2011. Food Defense Mitigation Strategies Database. http://www. fda.gov/Food/FoodDefense/ucm245544.htm. Accessed 21 October 2011. 116. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. 2009. Summaries of Competitive Food Defense Research Reports, 2007. http://www.fda.gov/Food/FoodDefense/FoodDefense­ Programs/FoodDefenseResearchReports/ucm175833. htm. Accessed 21 October 2011. 117. U.S. Food and Drug Administration, Department of Homeland Security, U.S. Department of Agriculture, and Federal Bureau of Investigation. 2008. Strategic Partnership Program Agroterrorism (SPPA) Initiative Final Summary Report September 2005-September 2008. http://www.fda.gov/Food/FoodDefense/FoodDefense Programs/ucm170509.htm. Accessed 21 October 2011. 118. U. S. Navy. Naval Safety Center. http://www.safetycenter.navy.mil/. Accessed 21 October 2011. 119. U.S. Senate Subcommittee on Oversight of Government Management, the Federal Workforce, and the District of Columbia. 2011. Agro-Defense: Responding to Threats Against America’s Agriculture and Food System. Hearing, 13 September 2011. (Testimony provided by Colonel John T. Hoffman [Ret.]) Government Printing Office, Washington, DC.

108 120. Whitaker, T. B., and A. S. Johansson. 2005. Sampling uncertainties for the detection of chemical agents in complex food matrices. J. Food Prot. 68:1306–1313. 121. Woodbury, G. 2005. Measuring prevention. Homeland Security Affairs 1:1–9. 122. World Health Organization. 2002. Terrorist Threats to Food: Guidance for Establishing and Strengthening Prevention and Response Systems. World Health Organization, Geneva, Switzerland. http://www.who. int/foodsafety/publications/general/en/terrorist.pdf. Accessed 21 October 2011.

Factors of Special Significance 123. Xu, S., T. P. Labuza, and F. Diez-Gonzalez. 2008. Inactivation kinetics of avirulent Bacillus anthracis spores in milk with a combination of heat and hydrogen peroxide. J. Food Prot. 71:333–338. 124. Xu, S., T. P. Labuza, and F. Diez-Gonzalez. 2006. Thermal inactivation of Bacillus anthracis spores in cow’s milk. Appl. Environ. Microbiol. 72:4479–4483. 125. Zhang, X., F. Liu, R. Yan, P. Xue, Y. Li, L. Chen, C. Song, C. Liu, B. Jin, Z. Zhang, and K. Yang. 2011. An ultrasensitive immunosensor array for determination of staphylococcal enterotoxin B. Talanta 85:1070–1074.

Microbial Spoilage and Public Health Concerns

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch6

John N. Sofos, George Flick, George-John Nychas, Corliss A. O’Bryan, Steven C. Ricke, and Philip G. Crandall

Meat, Poultry, and Seafood

Recent issues and developments regarding the microbial spoilage and safety of meat, poultry, and seafood products, which constitute the group of muscle foods, are ­addressed in this chapter. These products have many similarities, but also distinct characteristics. For example, they all are rich substrates providing nutrients that allow extensive microbial growth causing spoilage and foodborne illness, as well as products of fermentation. They differ, however, in the types of microbial contaminants they contain, as well as in handling procedures, which leads to different products within these food groups being sources of specific pathogens or undergoing characteristic types of microbial spoilage (216). Hence, this chapter provides an overall presentation of microbiological issues associated with all muscle foods, which is then followed by individual sections addressing, in sequence, specific spoilage and safety issues and their control for meat, poultry, and seafood.

microbial Ecology of Muscle Foods

Types and Origins of Initial Microfloras

It is generally accepted that bacteria are absent, undetectable, or present in very low numbers within muscle

6

tissues of healthy live food animals (i.e., meat, poultry, and seafood). The processes of animal slaughtering and carcass dressing, or catching of seafood, allow microbial contaminants to be deposited on the exposed cut surfaces of muscle and adipose tissue. Contamination originates from the external animal surface, including the gastrointestinal tract, as well as from the environment, including air, soil, water, equipment surfaces, and humans. As shown in Table 6.1, exposed muscle tissues may become contaminated with a vast array of gramnegative and gram-positive bacteria and fungi (47, 117, 169, 171, 216, 273, 275, 276). The initial prevalence and extent of microbial contamination on muscle foods may vary depending on the origin of the animal and the sanitation procedures and hygienic practices employed during production, transport, handling, and processing of the product. Specific parameters that affect the composition and numbers of the microflora include environmental conditions, such as wet or muddy hides, which may contain higher populations of bacteria indigenous to soil, whereas micro­organisms of fecal origin are generally more common when the hide is soiled with fecal material.

John N. Sofos, Center for Meat Safety & Quality, Department of Animal Sciences, Colorado State University, Fort Collins, CO 80523. George Flick, Department of Food Science & Technology, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061. George-John Nychas, Agricultural University of Athens, Laboratory of Microbiology & Biotechnology of Foods, 75 Iera Odos, Athens 11855, Greece. Corliss A. O’Bryan, Steven C. Ricke, and Philip G. Crandall, Department of Food Science, University of Arkansas, Fayetteville, AR 72701.

111

Microbial Spoilage and Public Health Concerns

112

Table 6.1  Genera of microorganisms commonly found on meats, poultry, and seafooda Type of muscle food Meat and poultry Microorganisms

Gram reaction

Fresh

–

X

Processed

Vacuum packaged

Fish

Bacteria Achromobacter Acinetobacter

–

XX

X

X

X

Aeromonas

–

XX

X

X

X

Alcaligenes

–

X

Alteromonas

–

X

X

Arthrobacter

–/+

X

X

X X

Bacillus

+

X

X

Brochothrix

+

X

X

XX

Buttiauxella

–

X

Campylobacter

–

X

Carnobacterium

+

X

Chromobacterium

–

X

Citrobacter

–

X

Clostridium

+

X X

X

X

X

XX X

X

Corynebacterium

+

Cytophaga

–

Enterobacter

–

X

X

X

X

Enterococcus

+

XX

X

XX

X

Escherichia

–

X

X

Flavobacterium

–

X

Hafnia

–

X

Halobacterium

–

Janthinobacterium

–

Klebsiella

–

X

Kluyvera

–

X

Kocuria

+

X

Kurthia

+

X

X X X X

Lactobacillus

+

X

Lactococcus

+

X

Leuconostoc

+

Listeria Microbacterium Micrococcus Moraxella Morganella

–

Paenibacillus

+

X

Pantoea

–

X

X

X X

XX

XX

X

X

X

+

X

X

+

X

X

X

+

X

X

X

–

XX

X

X X X

X X

Photobacterium

–

Proteus

–

X

Providencia

–

X

X

X (Continued)

6.  Meat, Poultry, and Seafood

113

Table 6.1  Genera of microorganisms commonly found on meats, poultry, and seafooda (Continued) Type of muscle food Meat and poultry Microorganisms

Gram reaction

Fresh

Processed

Vacuum packaged

Pseudomonas

–

XX

X

Rahnella

–

X

Serratia

–

X

Shewanella

–

X

X

X

Staphylococcus

+

X

X

X

Streptococcus

+

X

X

Vibrio

–

X

Weissella

+

X

Yersinia

–

X

X

Candida

XX

Cryptococcus

X

Debaryomyces

X

Hansenula

X

Pichia

X

Rhodotorula

X

Torulopsis

X XX

X

Saccharomyces XX X

X

Alternaria

X

X

Acremonium

X

Aspergillus

X

Aureobasidium

X

Trichosporon Molds

Cladosporium

XX

Chrysosporium

X

X

Fusarium

X

X

Geotrichum

XX

X

Monascus

X

Monilia

X

X

Mucor

XX

X

Neurospora

X

Penicillium

X

XX

Rhizopus

XX

X X

Scopulariopsis

 

XX X

Botrytis

Sporotrichum

XX

Thamnidium

XX

a

XX

X

X, known to occur; XX, most frequently isolated; +, positive; –, negative. Updated from Nychas et al., 2007 (216).

X X

X X

Yeasts

Fish

114 Decontamination treatments, applied to certain muscle foods during harvest or slaughter, which are discussed in subsequent sections, also determine the type and extent of initial microbial contamination that occurs on carcasses and muscle tissues (108, 216, 277).

Microbial Cell Attachment and Biofilms

The first event in muscle food contamination is the attachment of microbial cells to a surface, which is followed by colonization. The composition of the initial microbiota on meat surfaces may be affected by differences in the rate of attachment of bacterial strains. Pseudomonas is considered to attach more readily onto muscle tissue surfaces than other spoilage bacteria (216). Preevisceration spray-washing or decontamination of carcasses is applied immediately after hide removal in the United States with the goal of removing contamination before strong attachment of cells occurs on the carcass surface (277, 284). In general, bacterial cell attachment on biotic (e.g., muscle tissue) or abiotic (e.g., equipment) surfaces may be influenced by factors such as surface characteristics, properties of the substrate, and physiological stage, surface characteristics, and motility of the cells (107, 124, 216, 265). Biofilms consist of bacterial cells encapsulated in an exopolysaccharide matrix that allows them to adhere to surfaces and each other and protects them from adverse conditions (281). Cells form microcolonies or clusters enclosed within the hydrated matrix, and pores or channels throughout the structure allow transport of oxygen, nutrients, and waste. Cell matrices form a network that facilitates formation and maintenance of the biofilm structure and increases cell resistance to sanitizers. As discussed by Sofos (281), biofilms may form in all areas of food processing environments, including floors, walls, pipes, and drains, as well as on all materials commonly used in meat processing, including stainless steel, aluminum, nylon, Teflon, rubber, plastic, Buna-N, glass, etc. Hard-to-clean and -sanitize crevices in conveyor belts, pasteurizers, gaskets, and dead spaces become sites of biofilm establishment (281, 283). Pathogens such as Listeria monocytogenes may persist in food plants for months and up to several years (303). Creation of microbial harborage sites, niches, and biofilms in food processing environments is undesirable because they serve as sources of contamination (285). Mono- or multispecies biofilms are formed by spoilage as well as pathogenic bacteria, including Pseudomonas, Listeria, Salmonella, Campylobacter, Escherichia coli, and lactic acid bacteria (LAB) (281, 283). Recent research findings relative to E. coli O157: H7 biofilms were summarized by Sofos and Geornaras

Microbial Spoilage and Public Health Concerns (285) and Sofos (281, 283) as follows: the pathogen may attach to food contact surfaces at 4°C as well as 15°C, demonstrating the need for proper cleaning and sanitation programs in all plant environments; allowing beef residues to dry led to cell attachment as well as entrapment on surfaces; the strength of cell attachment was greater on dry than on moist surfaces, demonstrating the importance of proper cleaning and sanitizing of equipment surfaces after each use; cell attachment to beef-boning equipment surfaces varied among strains; the extent of biofilm formation was similar for quorumsensing-positive and -negative strains; spoilage bacteria may outgrow E. coli O157:H7 in biofilms; biofilm formation may be more extensive on plastic than on stainless steel beef contact surfaces; the types of equipment surface material did not affect biofilm cell sensitivity to sanitizers; if the presence of biofilms is suspected or known, sanitizers should be applied at the highest allowable concentrations for extended dwell times; however, adequate cleaning is always essential before sanitation. Regarding L. monocytogenes biofilms, Sofos and Geornaras (285) and Sofos (281, 283) concluded the following: cells have the ability to adhere to various food contact surfaces used in food processing/service and at home, including polyethylene, polypropylene, and laminate, and if surfaces are not properly cleaned, formed biofilms are resistant to sanitizers; multispecies biofilms, containing high levels of L. monocytogenes, survived for up to 14 days on high-density polyethylene­ and polypropylene surfaces at ambient temperature; survival of bacteria in biofilms was greater on rough than on smooth high-density polyethylene surfaces; sanitizers (acetic or lactic acid, sodium hypochlorite, quaternary ammonium, or hydrogen peroxide-based) were especially effective against younger biofilms, more than against older biofilms; a lactic acid-based sanitizer (pH 3.03) was the most effective, whereas quaternary ammonium-based sanitizers of higher pH (10.5 to 11.5) were more effective than those of lower pH (6.2 to 8.7); among products commonly found in households, effectiveness against pathogens (Salmonella, E. coli O157: H7, and L. monocytogenes) increased in the following order: household bleach (0.0314%) > hydrogen peroxide (3%) > undiluted vinegar > baking soda (50% sodium bicarbonate); pathogen sensitivity followed the order of Salmonella > E. coli O157:H7 > L. monocytogenes; sanitizer activity increased at warm temperatures (55°C); and sanitation of cutting boards should be performed after each use, or at least daily, in order to achieve maximum efficacy. The meat processing and related industries should be aware of biofilm issues and implement validated clean-

6.  Meat, Poultry, and Seafood ing and sanitation programs effective in biofilm control. Biofilm formation may be prevented by avoiding conditions that lead to cell attachment and selection of conditions that do not allow microbial growth, which is oftentimes not possible. Proper cleaning and sanitation are necessary for biofilm prevention or for removal and inactivation when prevention of formation fails. It is important to realize that strong attachment of cells and biofilm formation on food and equipment surfaces affect the efficacy of antimicrobial interventions applied to carcasses, meat, or equipment (281). Bacterial cells in biofilms may be as much as 500 times more resistant to sanitizers than free-flowing planktonic cells of the same species. This is also demonstrated by evidence that the concentration of sanitizers and exposure time may have to be increased 10- to 100-fold in order to be effective against cells in biofilms compared to planktonic cells. Biofilm removal and inactivation are achieved by combining proper cleaning and sanitizing agents, adequate exposure time, proper temperature, and mechanical ­ action. Generally, extensive scrubbing with proper chemicals is important for biofilm removal. Incomplete biofilm removal may promote cell growth (279, 281, 283, 285).

MICROBIAL SPOILAGE OF MUSCLE FOODS

Development of the Spoilage Association

It is important to recognize that one-fourth of the world’s food supply and 30% of landed fish are estimated to be lost due to microbial activity causing spoilage (110). Muscle food spoilage, being the result of chemical, enzymatic, and microbial activities, was described in detail by Nychas et al. (216). A spoiled food is not necessarily unsafe, if pathogens are absent; therefore, spoilage is considered an economic loss and usually does not receive the same attention as pathogens. It is important to note, however, that economic losses due to spoilage, as well as associated food waste and loss of consumer confidence, are also of importance. Hence, it is necessary to understand the causes and mechanisms of microbial spoilage in order to minimize losses and provide a food supply of high quality and adequate shelf life (229). As the inherent protective barriers (i.e., skin, hide, scale, and shells) are destroyed at slaughter, or during catching of fish and shellfish, the resulting tissues ­become exposed to environmental microbial contamination. In addition to reduced contamination levels (discussed in the following paragraphs), decontamination of meat with acid solutions may lead to changes in the dominating microbes associated with the tissue and, if combined with

115 long-term storage, may allow ­ development of yeasts, whereas hot water or steam decontamination may select for other types of spoilage microorganisms (171, 216, 251, 277). The levels and types of initial microbiotas also vary due to the level of hygiene, temperature, air flow, gaseous atmosphere, and other influences encountered during product handling, processing, and storage, as they select which of the initial or new contaminants will prevail and form the dominant microbiota as the product ages. In general, relatively few microbial types dominate through selection during storage based mostly on storage temperature and the gaseous composition of the surroundings (66). Aerobic or facultative anaerobic gram-negative bacteria such as pseudomonads dominate under cold aerobic conditions, whereas the dominant microflora consists largely of LAB in vacuum-packaged products (49, 216). The selection of the microbial flora and the subsequent chemical changes that occur also depend on microbial interactions (217, 306). Competition for nutrients (e.g., glucose and protein nitrogen) or chemical elements (e.g., iron), metabiosis or antibiosis (i.e., production of a favorable or unfavorable environment), and cell-to-cell communication (i.e., quorum sensing) can contribute to the selection of the dominant flora (126, 218). For example, Pseudomonas spp. may inhibit the growth of Shewanella putrefaciens or enhance the growth of L. monocytogenes. This is due to the ability of Pseudomonas to utilize glucose and/or produce siderophores at higher rates than S. putrefaciens or to provide useful substrates (e.g., protein hydrolysates) in the case of L. monocytogenes (216). According to Pennacchia et al. (231), Pseudomonas spp., Carnobacterium divergens, Brochothrix thermosphacta, Rahnella spp., and Serratia grimesii, or closely related species, were detected in beef at time zero, whereas Photobacterium spp. occurred in most meat samples stored in air or in vacuum packages. Interestingly, analysis by culturing meat specimens enabled detection of microbial species (e.g., several species of Serratia or Rahnella and Leuconostoc) that were not detected by analysis of DNA extracted directly from meat (231). Ercolini et al. (79) used random amplified polymorphic DNA-PCR and 16S rRNA gene sequencing and identified 50 mesophilic and 29 psychrotrophic isolates from beef. The most frequent species in both mesophilic and psychrotrophic populations were Carnobacterium maltaromaticum and C. divergens, whereas Acinetobacter baumannii, Buttiauxella spp., and Serratia spp. were identified among the mesophiles and Pseudomonas spp. were common among psychrotrophs. Doulgeraki et al. (66) isolated a total of 266 LAB from minced beef

116 stored at 0, 5, 10, and 15°C aerobically or under modified atmosphere packaging (MAP) consisting of 40% CO2–30% O2–30% N2 in the presence or absence of oregano essential oil. Analysis by 16S rRNA sequencing demonstrated dominance of Leuconostoc spp. during aerobic storage at 5, 10, and 15°C, as well as during MAP storage at 10 and 15°C, whereas Lactobacillus sakei prevailed during aerobic storage at 0°C, as well as during MAP storage at 0 and 5°C. Sporadic species included Leuconostoc mesenteroides, Weissella viridescens, Lactobacillus casei, and Lactobacillus curvatus. Species of Leuconostoc and L. sakei have been asso­ ciated with spoilage of cold stored vacuum or MAP meat (78). Hence, different microbial species dominate in different environments and contribute to muscle food spoilage as they release different volatile compounds (216).

Types of Muscle Food Spoilage

Destruction of the inherent protective barriers (i.e., skin, hide, scale, and shells) and natural defense mechanisms (e.g., lysozyme and antimicrobial peptides) of live animals at slaughter allows tissues to undergo rapid microbial decomposition, depending on the dominating extrinsic conditions (e.g., temperature, packaging, and processing method). A food is considered spoiled when consumers reject it based on undesirable sensory characteristics (216). Changes catalyzed by microbial or native meat enzymes as well as other chemical reactions in muscle foods lead to formation of off-flavors, offodors, discoloration, slime, or other changes in physical appearance or chemical characteristics and texture, which make the food unacceptable for consumption. However, characterization of a food as spoiled is based on the subjective judgment of each consumer, which may be influenced by cultural and economic considerations and personal background, as well as the sensory acuity of the individual and the intensity of the change. However, as spoilage progresses, most consumers usually agree that gross discoloration, strong off-odors, and the development of slime constitute spoilage that is ­undesirable (216). Spoilage changes associated with muscle foods vary depending on the types of microorganisms that dominate, muscle type (e.g., high or low pH, poultry thigh or breast, and enzyme activity), product composition (e.g., sugar and lipid content), storage environment (e.g., temperature and gas composition of packs), and duration of storage (216). According to Pennacchia et al. (231), the microbes most commonly involved in meat spoilage are Pseudomonas spp., Enterobacteriaceae, B. thermosphacta, and LAB; their contribution to spoil-

Microbial Spoilage and Public Health Concerns age varies largely depending on oxygen availability (171). In muscle foods stored aerobically at cold temperatures, spoilage is caused through oxidative metabolism by aerobic or facultative anaerobic gram-negative bacteria such as Pseudomonas (Pseudomonas fragi, P. fluorescens, P. putida, and P. lundensis), S. putrefaciens, and Photobacterium phosphoreum (125, 289). Other bacteria present in low numbers (217) in chilled muscle foods stored aerobically include B. thermosphacta and cold-tolerant Enterobacteriaceae (e.g., Hafnia alvei, Serratia liquefaciens, and Enterobacter agglomerans). LAB, although detected in the aerobic spoilage flora of chilled meat, are considered important only in the aerobic spoilage of lamb (136). Muscle souring is caused by growth of LAB and B. thermosphacta (216). LAB are mostly responsible for spoilage of meat in restrictedoxygen environments. Spoilage differs among cooked, cured, heat-­processed, fermented, or dried products of varying water activity and pH (147, 148). Spoilage during storage of perishable, processed or cooked, uncured meats is due to surviving or postcooking microbial contaminants. Dominant spoilage microorganisms may include micrococci, streptococci, lactobacilli, and B. thermosphacta (145). Recontamination with nonproteolytic bacteria results in the development of sour odors, whereas recontamination with proteolytics results in putrid odors due to the breakdown of amino acids. Spoilage of canned muscle food products is usually due to spoiled raw materials, inadequate thermal processing allowing survival of heat-resistant mesophilic sporeformers, slow cooling or storage at high temperatures that allow proliferation of thermophilic sporeformers, or reintroduction of microorganisms through postprocessing leakage (216).

Substrate Utilization, Chemical Changes, and Spoilage Compounds

Spoilage processes of red meat, poultry, and seafood have been reviewed previously (114, 117, 119, 147, 216, 273). The first source of energy for microorganisms growing on muscle foods is glucose, which is metabolized more rapidly by the obligate aerobic pseudomonads than by facultative anaerobes such as B. thermosphacta and oxidative strains of S. putrefaciens (306). Pseudomonads predominate due to their greater affinity for oxygen and faster growth (216, 217). Lactate is the energy source metabolized after glucose, under both aerobic and anaerobic conditions (69, 70, 215, 306). Utilization of glucose and lactate is at slower rates in MAP than in aerobically stored products. The third main energy source used by bacteria in muscle foods is amino acids. Details regarding compound metabolism and spoilage

6.  Meat, Poultry, and Seafood development are summarized by Nychas et al. (216). Overall, the development of organoleptic spoilage is related to the microbial metabolism of nutrients, such as sugars and free amino acids, and the release of undesired volatile metabolites. In cold storage, these activities are performed by psychrotrophic bacteria, hence compromising the value of low temperature as a preservation intervention for muscle foods (79). Metabolic end products associated with spoilage of muscle tissues are the result of substrate modifications by both microbial activity and indigenous muscle enzymatic as well as nonenzymatic chemical reactions and physical changes. However, although present, the contribution of indigenous muscle enzymes to spoilage is minor compared to the enzymatic action of microbes, which is the most important contributor to proteolytic muscle food spoilage (217, 305). Overall, chemical changes may be the result of competition among aerobic gram-negative species or facultatively anaerobic grampositive bacteria (216). Crustaceans are an exception to this rule because endogenous tissue enzymes in their hepatopancreas cause rapid postmortem muscle breakdown that is independent of microbial proteases (45). This is the primary reason that lobsters, crabs, and crayfish are kept alive until cooking; their death after harvesting results in rapid decomposition of edible tissues. The need for microbial population densities of greater than 7 log CFU in order for proteolysis to become sensorially evident (e.g., slime formation and sulfur/ammonia odors) in muscle foods has been disputed. Studies have revealed that proteolysis occurs even during the early stages of storage, regardless of microbial populations and the presence of low-molecular-weight (e.g., glucose) metabolites (216, 293). While many spoilage microorganisms can produce lipases, oxidative rancidity of muscle fat occurs when oxygen reacts with unsaturated fatty acids and compounds such as aldehydes, ketones, and short-chain fatty acids are produced (114). Autoxidation is of particular importance in the deterioration of highly unsaturated lipids in fatty fish and pork. Phospholipids present in muscle tissue membranes are also rich in oxidation-susceptible unsaturated fatty acids (216). Various types of muscle food spoilage are characterized by the total microbial activity rather than the activity of specific enzymes. Hence, overall spoilage of muscle foods should be considered in the context of aerobic versus anaerobic conditions (216). Pseudomonads metabolize sequentially d-glucose and l- and d-lactic acid, whereas use of d-glucose is preferred to that of dl-lactate. The extracellular oxidation of glucose and glucose-6-phosphate causes a transient accumulation of

117 d-gluconate and pyruvate and an increase in the concentration of 6-phosphogluconate (216). Under aerobic conditions, odors of metabolic products of amino acids such as sulfides, methyl esters, and ammonia are usually the first manifestation of spoilage of chilled meat, poultry, and seafood (50, 52, 72). Besides pseudomonads, particularly P. fragi, other bacteria that form such volatile compounds are S. putrefaciens, Proteus, Citrobacter, Hafnia, and Serratia (51, 188). Concentrations of acetone, methyl ethyl ketone, dimethyl sulfide, and dimethyl disulfide increase during aerobic storage of ground beef at 5 to 20°C. Hydrogen sulfide, another indicator of spoilage, is not produced by pseudomonads, whereas dimethyl sulfide is not produced by Enterobacteriaceae (51). Hydrogen sulfide combines with myoglobin to form a green discoloration. Amines such as putrescine, cadaverine, histamine, tyramine, spermine, and spermidine may be formed in ground pork, beef, poultry, and fish stored at cold temperatures (72, 76, 170). Under conditions favoring their growth (e.g., limited oxygen and low temperature), Enterobacteriaceae may also be important in muscle food spoilage, utilizing mainly glucose and glucose-6-phosphate; exhaustion of these substrates is followed by degradation of amino acids (116). Enterobacteriaceae produce ammonia and volatile sulfides such as hydrogen sulfide and malodorous amines from amino acids. Details on other compounds found in spoiled muscle foods are presented by Nychas et al. (216). Gram-positive bacteria, especially the LAB, are, in general, considered unimportant microbial contaminants of aerobically stored muscle foods, whereas B. thermosphacta may be of importance in the spoilage, particularly on fatty surfaces, of pork, lamb, and fish (68). Aerobically, B. thermosphacta utilizes glucose and glutamate but no other amino acid. Under both aerobic and anaerobic conditions, the organism has a greater spoilage potential than lactobacilli, producing several end products (216). Under facultatively anaerobic conditions, products of fermentation (e.g., lactic and acetic acids) accumulate and, in addition to causing ­spoilage, may also have antimicrobial activity (216). Acetic acid, as a product of further oxidation of lactic acid, has been used in models to evaluate the quality of beef, poultry, and fish (50, 126, 161). In such situations, the putrid odors associated with storage in air are replaced by relatively inoffensive sour odors (115). However, since the amounts of accumulated acetic, isobutanoic, l-isopentanol,­ and d-lactic acids are relatively small compared to the endogenous l-lactic acid of normal pH muscle, it is difficult to detect these off-odors. Diacetyl/ acetoin and alcohols causing dairy/cheesy odor in beef

118 stored in gas mixtures containing CO2 are produced by B. thermosphacta and heterofermentative LAB (216). Since gram-negative bacteria are inhibited in lowoxygen environments, spoilage of muscle foods stored under oxygen-limited/carbon dioxide-enriched atmospheres is due to undefined activities of LAB, B. thermosphacta, S. putrefaciens, and P. phosphoreum—the last two in fish (53, 305). However, production of sulfur compounds, such as propyl esters, 3-methylbutanol compounds, and formic and acetic acids, leads to the question as to whether this inhibition is due to carbon dioxide enrichment or oxygen limitation; pseudomonads are sensitive to CO2 at low temperatures, where its solubility increases (216).

Muscle Food Spoilage and Quality Evaluation

In addition to food safety, the meat, poultry, and seafood processing industries need methods to measure freshness and quality and to predict the shelf life of their products. Regulatory authorities also need reliable methods for inspection purposes. In searching for such methods, it is important to understand that differences between species, variation in geographical origin, and most importantly, changes in preservation technologies (e.g., introduction of new processing and packaging methods) may alter microbial spoilage patterns of muscle foods. Therefore, it is important to continuously develop reliable methods (216). Sensory and microbiological analyses are most often used to evaluate freshness or spoilage of meat, poultry, and seafood. A disadvantage of sensory analysis, which is probably the most acceptable and appropriate method to assess food freshness or spoilage, is its reliance on trained panelists, which makes it costly and unattractive for routine use (50). Traditional microbiological analyses (total viable counts) are too general, can be lengthy and destructive to tested products, and can provide misleading results. Therefore, efforts have been made (216) to develop quantitative and objective measures of biochemical changes in muscle foods (e.g., the K-value based on nucleotide catabolism), production of amines, ammonia, trimethylamine, and sulfur compounds, or physical changes (e.g., the Torrymeter, Freshtester, or RT Freshmeter). Studies have examined the correlation between microbial growth and chemical changes during spoilage in order to develop indicators of muscle tissue quality as well as spoilage (50, 126, 217). Metabolites for such use may be those that are specific (e.g., gluconate) to certain microorganisms (e.g., pseudomonads) or are common (e.g., acetic or lactic acid) to different members of the microbial population; in both cases, however, the spoilage information provided may

Microbial Spoilage and Public Health Concerns still not correctly identify spoiled or unspoiled product. Factors that complicate objective evaluation of spoilage include application of antimicrobial interventions, variable types of packaging, and temperature of storage. The ideal spoilage indicator according to Nychas et al. (216) should (i) be absent or present at low levels in fresh and unspoiled tissue; (ii) be produced by the microflora dominating during spoilage; (iii) increase with storage time; and (iv) correlate well with sensory analysis. Identification of such an ideal indicator is a difficult task. In recent years, the search for novel and rapid analytical approaches for quantitative monitoring of meat spoilage has focused on the development of biosensor, electronic nose, electronic tongue, Fourier transform infrared (FTIR) analysis, and Raman spectroscopy procedures. Complementary advances include the application of advanced statistical methods (discriminant function analysis, clustering algorithms, and chemometrics) and intelligent methodologies (neural networks [artificial or not], fuzzy logic, evolutionary algorithms, and genetic programming). Through the use of various multivariate analysis techniques, such as principal-components analysis partial least squares models and two types of artificial neural networks (i.e., multilayer perceptron [MLP] and fuzzy [ARTMAP]), it may be possible to determine the time elapsed in relation to the degradation of muscle foods by using simple potentiometric measurements. Additional information on spoilage/quality detection can be found in articles by Ammor et al. (5), Argyri et al. (7), Bazemore et al. (21), Gil et al. (111), Herrero (133), Nieminen et al. (206), Panagou et al. (227), Van Boekel (310), and Wilson and Baietto (324).

FERMENTED MEATS, BACTERIOCINS, AND PROBIOTIC FOODS Fermented foods are products of metabolic processes of various microbial types, including bacteria, yeasts, molds, or combinations thereof, that convert food commodity substrates into various types of desirable foods or ingredients such as vitamins and enzymes. There is a vast array of fermented muscle food products available throughout the world (228, 238). Traditional fermented foods predate written history and were developed to upgrade plant and animal materials by producing more acceptable foods and to prevent the growth of spoilage and pathogenic microorganisms without the need for cold storage. Common fermented foods are sausages, which were invented by the Sumerians about 5,000 years ago and were popular among ancient Greeks and Romans (41).

6.  Meat, Poultry, and Seafood Major microbial groups used in production of fermented sausages and similar products are LAB and coagulase-negative cocci. Depending on the product, groups such as yeasts and enterococci may also play a role (238). The fermentation process involves complex biochemical and physical reactions resulting in significant changes in the initial product characteristics (239). The sensorial profile of the final product is affected by the dominating microbial strains. Talon et al. (295) ­reviewed the diversity of the microbiotas both in the environment and in traditional fermented European sausages. Rantsiou and Cocolin (238, 239) have reviewed the application of molecular methods for faster and more precise identification of microbial strains isolated from fermented sausages. In addition to their crucial role in meat fermentations, LAB are the most recognized and investigated producers of microbial antagonists, such as bacteriocins (298). Bacteriocins, however, are known to exert only a transitory bactericidal effect against pathogens such as L. monocytogenes; often, there is regrowth of the pathogen in bacteriocin-containing foods. The regrowth may be due to factors that severely limit growth of bacteriocinproducing cells (e.g., restricted nutrient availability); decreased bacteriocin action as a result of adsorption onto food particles, fat, and protein; the presence of curing agents; the emergence of bacteriocin-resistant cells; and/ or bacteriocin degradation by proteases of food and/or microbial origin (168, 298). Although abundant in the environment and extensively researched in the past 30 years, the bacteriocins described above, with the exception of nisin, have not received regulatory approval in the United States and so are not commercially used in foods (287). Recently, in the United States, probiotic-containing foods have gained consumer acceptance as health­promoting, functional foods (62, 221). Hence, there is interest in developing fermented sausages based on probiotic microorganisms. Results of related research are still preliminary; hence, evaluations of the human health effects of fermented meats produced by probiotic bacteria are pending (62). There is also interest for use of LAB as probiotics in fish products (176).

PATHOGENS IN MUSCLE FOODS In general, the hazards that compromise the safety of meat, poultry, and seafood are of physical, chemical, and biological nature. Physical hazards include bone chips and foreign materials such as metal, glass, wood, plastic, and stones. Chemical hazards include natural and synthetic environmental contaminants and toxins, such as

119 residues of animal drugs and pesticides or of chemicals present in the animal’s environment, and excessive use of food additives in processed products. Safety-­related concerns may also include fecal and manure contamination; animal identification and traceability issues; animal welfare and type of animal production system; complete and routine implementation of hazard analysis critical control point (HACCP) at the production and processing level, after proper food handler training and consumer education; and transmissible spongiform encephalopathies (TSEs). These issues have been presented in some detail by Doyle and Erickson (67), Sofos (279, 280), and Sofos and Geornaras (285). Biological hazards associated with meat, poultry, and seafood include pathogenic bacteria, viruses, parasites, toxigenic molds, and TSEs. Major challenges related to microbial pathogens include foodborne illness outbreaks and deaths, associated product recalls from the marketplace, and regulatory compliance problems. Hence, the most important muscle food safety issues of current worldwide concern are the need to control traditional as well as new, emerging, or evolving pathogens, including those of increased virulence at low infectious doses or resistant to antibiotics or to food processingassociated stresses caused by physical factors (e.g., heat, cold, drying, and radiation) and chemical agents (e.g., acids, salts, and sanitizers) (251). Other issues are crosscontamination of water and foods of plant origin with enteric pathogens, associated animal manure disposal issues, and potential for implementation of pathogen control interventions at the farm level. This chapter highlights only selected concerns (279). The most important foodborne bacterial pathogens in land muscle foods are enterohemorrhagic E. coli (including E. coli O157:H7), Salmonella, Campylobacter spp., and L. monocytogenes; bacterial pathogens of concern in seafood are Vibrio spp., L. monocytogenes, and Salmonella. Additional pathogenic bacteria that may be transmitted through consumption of contaminated muscle foods include Staphylococcus aureus, Clostridium perfringens, Yersinia enterocolitica, Bacillus cereus, other Bacillus spp., Clostridium botulinum, Aeromonas, Arcobacter, Brucella, Shigella, Enterobacter, Plesiomonas shigelloides, Helicobacter, and potentially Mycobacterium and Clostridium difficile (12, 147, 148). Relative to parasitic and viral disease agents, swine can transmit trichinosis (Trichinella spiralis), sarcocystosis (Sarcocystis spp.), and toxoplasmosis (Toxoplasma gondii); poultry transmit toxoplasmosis; and beef cattle transmit tapeworms (Taenia spp.), Sarcocystis spp., and, especially through fecally contaminated water, Cryptosporidium parvum

120 (cryptosporidiosis). Viral agents, such as norovirus, hepatitis A virus, and enteroviruses cause most foodborne disease cases in the United States, but their transmission is mostly associated with poor sanitation, cross-­contamination from infected food handlers during preparation and serving, or inadequate cooking (169, 171, 255, 256, 279). Mycotoxins in meat products are reviewed by Bailey and Guerre (13). Bovine spongiform encephalopathy (BSE) has emerged as a major animal health issue in recent years, especially because of its potential involvement in human TSE such as new variant Creutzfeldt-Jakob disease (vCJD). The issue of BSE will continue being of interest mostly as a target for eradication, whereas viral agents affecting food animals, such as avian influenza, will likely always need attention for prevention or containment (279, 280, 285, 321). According to the U.S. Centers for Disease Control and Prevention (CDC), pathogens (and some associated foods) responsible for most foodborne illness are Campylobacter (poultry), E. coli O157:H7 (ground beef, leafy greens, and raw milk), L. monocytogenes (deli meats, unpasteurized soft cheeses, and produce), Salmonella (eggs, poultry, meat, and produce), Vibrio (raw oysters), norovirus (many foods; e.g., sandwiches and salads), and Toxoplasma (meats). Viral pathogens are of major concern in food service, whereas bacterial pathogens such as E. coli O157:H7 and other Shiga toxin-producing E. coli (STEC), Salmonella, and Campylobacter continue to be of importance in the safety of raw meat and poultry, as does L. monocytogenes in ready-to-eat processed products (www.cdc. gov/foodsafety). Muscle food safety challenges may be associated with changes in food animal production, product processing and distribution practices, including aquaculture; increased international food trade; changing consumer needs and preference for minimally processed products; increased worldwide muscle food consumption; large numbers of consumers at risk for infection; emerging pathogens and pathogen changes that may be associated with increased virulence and resistance to control or clinical treatment; advances in microbial detection methodologies; inadequate food handler and consumer education and training in proper food handling; and increased interest, awareness, and scrutiny by consumers, news media, and consumer activist groups (279, 280, 285, 321). Risks to consumers from consumption of stored muscle food products may also be due to the presence of biogenic amines such as histamine, putrescine, spermidine, and spermine. Amines found in fresh meat and fish products (primarily scombroid species such as

Microbial Spoilage and Public Health Concerns tuna, mahimahi, and mackerel) under aerobic or vacuum/MAP storage may lead to scombroid poisoning, a severe, and sometimes fatal, allergic reaction. Biogenic amine formation in some products has been attributed to Enterobacteriaceae; however, tyramine can also be formed by some strains of Lactobacillus (216). Proper sanitation and storage temperature and storage time limitation should minimize human health problems associated with biogenic amines in muscle foods (162).

ANTIMICROBIAL RESISTANCE Bacterial resistance to antimicrobials, especially antibiotics used in human medicine, is of major concern (146, 279). Important antibiotic resistance issues in the United States include the development of Campylobacter resistance to ciprofloxacin; Salmonella enterica serovar Typhimurium DT104 R-type ACSSuT pentaresistant (ampicillin, chloramphenicol, streptomycin, sulfamethoxazole, and tetracycline) strains; Salmonella serovar Newport R-type MDR-Amp C strains resistant to multiple antimicrobials (e.g., ampicillin, chloramphenicol, streptomycin, sulfamethoxazole, tetracycline, amoxicillin-clavulanic acid, cephalothin, cefoxitin, ceftiofur, and ceftriaxone); and vancomycin-resistant Enterococcus faecium and Enterococcus faecalis, which are important in hospital infections (58, 67, 104, 146, 279, 322, 330). In recent years, monophasic Salmonella serovar Typhimurium-like strains (e.g., serovar 4,[5],12:i:-) have emerged in pigs, cattle, and humans and may be a new health issue for animals and humans (82). The use of antimicrobials in food animal production and their role in promoting resistance in foodborne pathogens are subjects of an ongoing debate (146). Several studies have revealed that human fluoroquinoloneresistant Campylobacter infections have increased worldwide following approval of the drugs for use in animal production (330). Although a ban on antibiotic use in animal agriculture is often proposed, it is unknown how such an action might affect the extent of contamination of animal food products with resistant or nonresistant pathogens. As for humans, treatment of animal diseases, including those of pets, with antibiotics should continue for humane reasons. Regarding microbial food safety, the practical issue is whether ­antibiotic-resistant pathogen strains are of similar or higher resistance to common food-processing treatments compared to sensitive counterparts. Higher ­resistance of antibiotic-resistant strains should be of concern, whereas similar overall resistance would indicate that approaches used for pathogen control should be of ­ similar effectiveness against both antibiotic-sensitive and -resistant

6.  Meat, Poultry, and Seafood strains. The limited studies available have not proven the need for such a concern (146). Nevertheless, issues related to clinical use and effectiveness of antibiotics are real and need attention. A commonsense approach for control is to not overuse or abuse antibiotics in animals and humans; prudent use is recommended. Decisions should be based on risk analysis and examination of all issues associated with each specific type of antibiotic application and concern (146, 279). A logical approach for control of antibiotic use consists of the following: prevention of disease before treatment; preferential use of antibiotics of lesser importance to human medicine; and treatment with alternative methods such as vaccines, probiotic or competitive exclusion microbes, antimicrobial peptides, and bacteriophages. Research is needed to address the following: mechanisms of antimicrobial resistance, genetic transfer of resistance determinants, resistance gene transfer from environmental to human gastrointestinal flora, ecology of resistance reservoirs, potential of antibiotics to enter animal production environments through waste streams, fate of antimicrobials within environmental ecosystems, association between resistance and virulence, mechanisms for animal growth promotion by antimicrobials, new (or alternatives to) antibiotics, and improvement of prudent-use guidelines (146, 279).

ENVIRONMENTAL CONTAMINATION CONCERNS Sofos (279) has indicated that the contribution of meat and wild animals, through their feces and manure as sources of environmental, water, and food contamination, as well as the direct animal-to-human ­transmission of pathogens, to human health issues needs serious consideration. In recent years, E. coli O157: H7 and Salmonella have been linked with outbreaks of illness caused through consumption of a variety of nonbeef products, including fruit juices, salad sprouts, ­mayonnaise, spinach, lettuce, onions, cantaloupes, water­melon, tomatoes, peppers, almonds, chocolate, and dry breakfast cereal (www.cdc.gov). Most of these events have been attributed to environmental crosscontamination and transfer of enteric pathogens from animal hosts to products of non-animal origin. If not properly composted, processed, and handled, pathogencontaminated animal manure may lead to environmental pollution and water contamination. Direct use of manure as a fertilizer or use of contaminated water for irrigation or washing may lead to contamination of nonmeat or nonpoultry foods, including seafood. Sofos (279) concluded that the food animal industry

121 needs to address such issues irrespective of whether wild animals and birds or human negligence contributes to such concerns. Climate change could influence the spreading of pathogens among livestock by its effect on the biology of pathogens and potential vectors, land use, farming practices, animal and environmental factors, and establishment of new microenvironments or microclimates. According to Gale et al. (105), the recent emergence of bluetongue virus in livestock in Great Britain demonstrates the potential effects of climate change on the occurrence, distribution, and prevalence of livestock ­diseases. In addition to bluetongue virus, an orbivirus transmitted by Culicoides midges (flies), other exotic arboviruses infecting livestock such as epizootic hemorrhagic disease virus, Akabane virus, and bovine ephemeral fever virus are probably transmitted by the same midge species; climate changes could enable their spread in other parts of Europe. It is also believed that liver fluke is expanding its range in Great Britain because of warmer, wetter conditions that favor its mud snail intermediate host. In ­general, it is believed that flooding contributes to the prevalence or spreading of pathogens, including C. botulinum and Bacillus anthracis, that are transmitted by the fecal-oral route. There is a need for risk assessments of factors that may be directly affected by climate change or that may be indirectly affected through changes in human activity, such as land use (e.g., deforestation), transportation and movement of animals, intensity of livestock farming, and habitat change (105).

MUSCLE FOOD TRACEABILITY Traceability is the ability to maintain credible custody of the identification of animals and their products from production to retail and is considered an essential tool to protect animal and public health (55, 262). European Union (EU) legislation requires animal food traceability systems, based on product labeling. Generally, traceability is becoming an important issue for the meat industry, as it can play a major role in food safety-related risk management practices and in product authentication (262). The use of traceability systems for food products derived from individual animals (e.g., steaks and chops) is increasing. However, traceability of compound products, such as ground beef, is more complicated and difficult to apply in commerce and hence may be limited to the place and date of production. However, demands for traceability are increasing with global competition and food safety concerns. Animal identification and traceability gained importance with the emergence and need for containment of

122 BSE in various countries. Traceability implementation is expanding as food product recalls are becoming more common and global, and as associated technologies such as electronic ear tags, retinal scanning bar codes, radio frequency identification devices, global positioning systems, and smart chips for biological markers are developed and their use becomes economical and practical. Implementation of effective systems can be very useful in tracking, containment, and recalls of animals or their products when necessitated for public health or other concerns (262, 263, 279). The overall success of animal identification and traceability efforts, however, depends on complete and mandatory implementation of effective systems. Issues to be considered in the process of establishing animal identification and tracking systems include selection of the proper technology, maintenance of confidentiality, selection of precision requirements, and costs. The system should include identification numbers for premises and animals, and it should cover feed, animal, and meat (269, 270, 279). A related recent development is biotracing, which has a similar approach. Under biotracing, biological agents, such as microorganisms or their toxins, are traced (backward) and/or tracked (forward) in the food/feed chain. Biotracing, however, is more difficult to implement because currently biological agents cannot be labeled with unique markers (18, 137). Full food chain tracing and tracking of biological contamination (biotracing) will be feasible with progress in and integration of detection technologies, improvements in molecular marker identification, better knowledge of pathogenicity markers, and improved modeling approaches. This will lead to improvements in timely implementation of effective interventions when problems develop (137). In contrast to HACCP, which focuses on critical control points, biotracing addresses the whole production chain, from the farm through transportation, distribution, and storage (222).

MEAT PROCESSING AND MICROBIAL CONTROL

Overall Approach for Microbial Control

Muscle foods are rich in nutrients and provide a suitable environment for rapid microbial proliferation; hence, they need to be adequately preserved in order to maintain quality, delay spoilage, and ensure safety. Preservation is achieved through application of various processing procedures (e.g., heating, freezing, refrigeration, drying, additives, fermentation, acidification, packaging, and various combinations thereof), which

Microbial Spoilage and Public Health Concerns lead to the production of a variety of muscle food product types, some of which are ready-to-eat, in addition to fresh raw products that need to be cooked before consumption (300, 301). The logical strategy for microbial control with the goal of extending shelf life and improving muscle food safety is to apply adequate cleaning, good sanitation, proper hygiene, and effective antimicrobial intervention technologies in order to (i) harvest and ship for slaughter and processing food animals with low contamination levels; (ii) reduce potential for transfer of microorganisms to carcasses, meat, and seafood from live animals, water, and the environment; (iii) apply safe and effective decontamination interventions, when approved, for reduction of microbial levels on carcasses or meat; (iv) apply processes (e.g., heat, high pressure, irradiation, when approved and useful) in order to reduce or eliminate, by killing microorganisms, microbial contamination on processed or cooked products; (v) avoid or minimize cross-contamination at all stages of the chain, from production, slaughter, processing, and preparation to consumption; and, (vi) store products under lowtemperature and packaging conditions that inhibit growth of surviving microorganisms. Proper design, validation, implementation, verification, and documentation of this approach, through a process management system, such as HACCP, is the best available strategy for ensuring food safety and quality (159, 160, 169, 171, 273, 274, 277–279, 283, 291). In general, antimicrobial hurdles used or evaluated in foods include physical interventions such as low and high temperature; nonthermal novel processing technologies, including ionizing radiation (gamma and electron beam), high hydrostatic pressure, pulsed electric fields, sonication, ultrasonic waves, UV light, pulsed UV light, and microwaves; packaging methods such as MAP, including vacuum packaging; high-oxygen, lowoxygen, or oxygen-free controlled atmosphere packaging; active packaging; smart packaging; coatings; and antimicrobial edible films. Physicochemical hurdles include acidity or low pH, low water activity, modified ­ oxidation-reduction or redox potential (Eh), and application of chemical antimicrobial agents as food formulation ingredients or externally applied solutions or preparations. Biological interventions include introduction of starter cultures as microbial competitors (LAB) or as antimicrobial preparations (such as nisin) (171). Control of pathogenic bacteria in foods is commonly achieved with application of more than one, individually sublethal, antimicrobial treatment (hurdle technology concept) in the form of multiple sequential or ­simultaneous hurdles (175). This approach is consid-

6.  Meat, Poultry, and Seafood ered more frequently as consumers prefer an adequate food supply that is safe, of good quality, wholesome, nutritional, and economically affordable, as well as free of additives, convenient, and exposed to only minimal processing. However, under such conditions, foodborne pathogenic bacteria may become resistant to stresses such as acid, cold, heat, drying, anaerobiosis, decontamination interventions, and sanitizers, if applied at individually sublethal levels. This could lead to cell adaptation or selection of resistant populations that may develop multiple resistances or cross-protection to other stresses (251). Presently, implementation of the ­multiple-hurdle approach is mostly empirical and experience based, because knowledge of mechanisms of microbial sensitivities and antimicrobial activity is limited and does not allow intelligent identification of hurdle combinations for synergistic antimicrobial effects against desirable cell targets. However, the goal should be to maximize antimicrobial effects and product safety in complex food systems without creating stress-adapted pathogens that may be more difficult to control, leading to failure of preservation systems. Therefore, knowledge is needed on mechanisms of action as well as microbial cell functions, in order to optimize multiple-hurdle systems. Proper implementation should cause multiple cell injuries or metabolic exhaustion and lead to pathogen death (251, 277, 279, 280).

Muscle Food Decontamination

As concerns associated with foodborne illness linked to consumption of contaminated animal products increased in the 1990s, the practice of decontaminating meat and poultry carcasses has become common in countries such as the United States, Canada, and Australia. The goal of decontamination is to reduce the prevalence and numbers of pathogenic bacteria, such as STEC, verocytotoxin-producing E. coli, Salmonella, and Campylobacter, through application of physical or chemical agents and their combinations in the form of multiple sequential decontaminating interventions at various steps of the slaughtering, dressing, and boning processes (182, 183, 277–279, 284, 291, 292). Efforts to reduce contamination introduced into slaughterhouses with live animals include total or partial external animal washing or hair removal and are followed by decontamination technologies applied to carcasses immediately after hide removal but before evisceration, as well as at the end of the dressing process or before and during carcass chilling. Recently, decontamination is also applied on chilled carcasses before boning and on meat cuts and trimmings at packaging or before grinding or processing into nonintact or brine-injected

123 products (277, 286). Interventions for reduction of contamination on carcasses, meat, and poultry are based on treatments with water or steam (e.g., knife-trimming, steam-vacuum, cold or hot water, and steam pasteurization) at various temperatures and pressures and with chemical solutions, mostly organic acids (e.g., lactic, acetic, and citric), as well as chlorine or chlorine dioxide, trisodium phosphate, acidified sodium chlorite, peroxyacetic acid, and cetylpyridinium chloride; compounds such as hydrogen peroxide, ozone, protein compounds such as lactoferrin, have also been considered (25, 139, 171, 274, 277, 284, 291). The effectiveness of these treatments in reducing microbial contamination is affected by factors such as water pressure, temperature, the chemicals used and their concentration, duration of exposure, method of application, and time or stage of application during processing (277). Decontamination interventions, such as hot water, steam, acetic acid, or lactic acid, cause bacterial reductions of 1 to 3 log CFU on carcasses. It is important to emphasize that decontamination must be implemented as part of an integrated food hygiene system and should be based on use of multiple sequential interventions at different points during slaughter (30, 31, 182, 183, 268, 277, 284, 291). Decontamination may also be applied to fish (143). According to the U.S. Department of Agriculture (USDA)–Food Safety and Inspection Service (FSIS), decontamination agents may be approved for use if they (i) are “generally recognized as safe”; (ii) do not result in adulterated product; (iii) do not create a need for labeling (i.e., added ingredients); and (iv) can be documented with scientific studies as being effective (274, 277). EU regulations, in principle, allow application of water, including hot water, and chemical solutions for decontamination of carcasses. However, each proposed application needs to be individually approved and chemical decontamination should be followed by water treatment to remove residues in order to avoid the need for labeling (25, 82, 140, 171). The following are issues that should be addressed when selecting or approving and using decontaminating agents: (i) the safety of application relative to workers, the environment, and consumers; (ii) potential undesirable effects on product quality; (iii) the potential for spreading and redistribution of bacterial cells over the carcass surface or their penetration into the tissue; and (iv) the potential development of stress-resistant pathogenic bacteria (216, 251, 277).

Nonthermal Processing Treatments

In order to meet consumer preferences for changes associated with less food processing, the industry is

124 c­ onsidering or using alternative nonthermal preservation technologies such as high hydrostatic pressure, highpressure processing (HPP), ionizing radiation, pulsed X rays, ultrasound, pulsed light, pulsed electric fields, high-voltage arc discharge, magnetic fields, dense-phase carbon dioxide light pulses, natural biopreservatives, and active packaging systems (11, 171). Such ­processes have been developed and investigated, and some of them are used in certain commercial products, including muscle foods. Nonthermal processes may be more energy efficient and better able to preserve food quality than conventional processes such as thermal treatments (196). These processes are able to inactivate vegetative microorganisms, including pathogens, but not bacterial spores. Effectiveness may be increased if nonthermal and thermal technologies are used in milder combinations according to the multiple hurdle approach. The benefits and limitations associated with commercial use of nonthermal technologies are discussed by Aymerich et al. (11).

High-Pressure Processing

HPP is a nonthermal technology applying pressures of up to 1,000 MPa for variable lengths of time in order to extend food product shelf life and safety without major modifications in nutrient content and sensory properties (245). By inactivating microorganisms as well as endogenous enzymes, HPP preserves the sensorial characteristics and extends the safe shelf life of food products. It should be noted, however, that different types of microorganisms are of different sensitivities to HPP; vegetative cells are the most sensitive, followed by fungi, whereas viruses and bacterial spores are the most resistant (63, 106, 172, 211, 244, 331). HPP inactivates microorganisms through cell membrane damage, solute loss, protein denaturation, and enzyme inactivation (211, 241, 244). Its ability to act on multiple cellular targets allows its use in combination with other treatments for better success through synergism. Used in proper combinations with mild heat, it may even achieve some bacterial spore inactivation. Food is not heated and may not be deformed because HPP is isostatic (i.e., uniform and instant pressure transmission) and adiabatic (i.e., little temperature variation within the product), with temperature increases of only approximately 3°C per 100 MPa (241, 325). HPP is applied to a range of foods, including meat products such as cooked ham, fish, and precooked meals (211). Its use in ready-to-eat meat products, such as ham, for control of postprocessing L. monocytogenes contamination has gained popularity in recent years. It is marketed as a natural alternative to chemical anti-

Microbial Spoilage and Public Health Concerns microbials in order to meet regulatory requirements for antimicrobial alternatives applied to products allowing growth of the pathogen before consumption (101). HPP has also been considered as an alternative to heat in inactivation of STEC in fermented sausages (223). Since fermented sausages were implicated in STEC outbreaks, some countries established regulations requiring 3- to 5log unit reductions of E. coli O157:H7 in such products. Safe dry-fermented sausages produced without a heating step to eliminate STEC have been difficult to obtain. HPP offers the potential of improving dry-fermented sausage safety without affecting unheated product quality (223). There is evidence, however, that under certain conditions, HPP may actually modify the texture, flavor, appearance, and color of meats (186, 187, 244). According to Rivas-Cañedo et al. (245), HPP causes changes in muscle mechanical properties and the color of fresh beef varies after HPP due to myoglobin denaturation, heme displacement or release, and ferrous oxidation. In contrast to beef and pork, poultry muscles are not drastically discolored because of their lower myoglobin content. HPP of meat was also associated with decreases in some alcohols and aldehydes and higher levels of 2,3butanedione and 2-butanone. In addition, branchedchain alkanes and benzene compounds can migrate from the plastic packaging material into the meat (63, 245). McArdle et al. (187) concluded that HPP may alter meat structure, color, and lipid oxidation levels, as it causes lipids to become susceptible to attack by molecular oxygen. Pressure-induced changes could be used to enhance the value of lower-quality muscles.

Irradiation

Although treatment of foods with ionizing radiation has about 100 years of history and it is scientifically established as safe, its application in muscle foods is limited (86). Treatment of meat products with ionizing radiation such as gamma rays or high-energy electrons and X rays can kill microbial pathogens and indigenous microflora and hence extend shelf life and enhance food safety (174). In the United States, irradiation was approved for fresh and frozen poultry meat in 1992 and for fresh red meat in 1997 (97). Irradiation is effective in reducing L. monocytogenes and Salmonella serovar Typhimurium inoculated on the surface of restructured pork loins in a dose-­dependent manner. The maximum irradiation dose permitted for meat and poultry depends on the type of product and whether it is treated in the refrigerated or frozen state (237). Maximum doses of 4.5, 7.0, and 3.0 kGy are permitted for pathogen reduction in uncooked/chilled meat,

6.  Meat, Poultry, and Seafood uncooked/frozen meat, and fresh or frozen poultry, respectively (271). The U.S. meat industry has petitioned for the application of electron beam radiation for treatment of beef carcass sides after slaughter in order to inactivate E. coli O157:H7 and for inactivation of L. monocytogenes on ready-to-eat meat and poultry products. Besides consumer resistance in purchasing irradiated meat and poultry, conflicting reports on the potential for negative effects of irradiation on meat color and odor exist (26, 237). Interventions proposed to preserve color during irradiation include feeding livestock dietary vitamin E or conjugated linoleic acid and treatment of fresh products with antioxidants (233, 237). Nam et al. (201) reported that irradiated restructured pork loins treated with rosemary-tocopherol and double packaging had lower values for thiobarbituric reactive substances than vacuum-packaged controls after 10 days of refrigerated storage. In contrast, however, the rosemary-tocopherol combination did not affect production of sulfur volatiles in irradiated pork; double-packaging reduced sulfur volatiles significantly, and rosemary-tocopherol effectively reduced hexanal.

Muscle Food Packaging Systems

Fresh muscle foods are packaged to avoid additional microbial contamination and cross-contamination; reduce moisture and weight loss; delay chemical and microbial spoilage; and where applicable, to ensure an oxymyoglobin or cherry red color in red meats at retail or customer level. Existing meat packaging systems range from overwrapping with paper or air-permeable films, for short-term chilled storage and/or retail display, to a large variety of MAP systems including vacuum packaging, bulk-gas flushing, or MAP systems with up to 100% carbon dioxide or other gas combinations for long-term chilled storage (167, 171). Recently, innovative measures for improving the quality and extending the shelf life of packaged muscle foods have been developed, based on technologies such as barrier film, active packaging, nanotechnology, microperforation, irradiation, plasma, and far-infrared ray treatments. Each technology, however, has drawbacks that need to be considered and resolved before application (112, 169, 171). Although storage at cold temperatures is the most common strategy to delay bacterial growth and enhance the shelf life and safety of muscle foods, shelf life can be further enhanced by storage under MAP conditions, including vacuum packaging (257). Bacterial growth, usually that of pseudomonads, is faster in polyvinyl chloride (PVC)-wrapped than in vacuum-packaged products; vacuum packaging reduces total plate counts and favors growth of Lactobacillus. Meat browning in PVC pack-

125 ages occurs due to oxygen-stimulated metmyoglobin formation even when bacterial numbers are low, whereas dipping or injecting fresh meats with antioxidant solutions, such as sodium tripolyphosphate and ascorbate, maintains meat color (44). Vacuum packaging is an acceptable method for light-colored pork and chicken and minimizes brown discoloration in red meats; however, the dark purplish deoxymyoglobin color is not acceptable by U.S. consumers of beef (44). According to McMillin (189), MAP involves the removal and/or replacement of the atmosphere surrounding the product before sealing in vapor barrier packaging materials. In general, MAP storage involves replacing the air surrounding the food with gas mixtures (e.g., CO2–O2–N2 and other nontoxic gases) (266). An additional, but technically different form of MAP is retail case-ready packaging, in which meat is cut and packaged at a centralized location and transported for display at retail. While storage under controlled atmospheres maintains gas composition, in MAP the concentration of CO2 is reduced during storage due to absorption by the food and the permeability of the packaging material (266). MAP extends shelf life characteristics of meat, but anoxic forms of MAP without carbon monoxide (CO) do not generate the preferred bloomed red meat color. Further, MAP with oxygen (O2) may promote lipid and pigment oxidation (189). Cell growth rate decreases with increased CO2 levels and temperature decreases because the gas dissolves and decreases the pH of food (266). There is a concern that limiting the growth of spoilage microorganisms under reduced-oxygen conditions may enhance pathogen growth and toxin production before there is evidence of spoilage. Hence, in reduced-oxygenpackaged products for which refrigeration is the sole barrier to outgrowth of nonproteolytic C. botulinum, if cells and spores have not been destroyed (e.g., ­vacuumpackaged raw fish, unpasteurized crayfish, or crab meat), the temperature must be maintained at 3.3°C or below through consumption in order to prevent growth and toxin formation. While temperature control by processors is usually possible, its maintenance during transportation, retail, and home storage may be inadequate. The use of antimicrobial agents, time-temperature integrators, or frozen storage are potential alternatives that may be used to prevent pathogen growth (216).

MEAT

Microbial Contamination

As with other muscle foods, the type and extent of microbial contamination on red meat carcasses and

126 ­ roducts are determined by the conditions of cleanliness p and sanitation associated with animals before slaughter, the processing environment and equipment, hygienic practices, animal and product handling during harvesting and processing, and conditions during product storage and distribution. Initial contamination levels may be in the range of 102 to 107 aerobic mesophiles per cm2 depending on processing operation and carcass site (169, 171). Common types of microorganisms found on fresh meat carcasses include gram-negative rods and micrococci such as Pseudomonas spp., Enterobacteriaceae, Acinetobacter spp., Alcaligenes spp., Moraxella spp., Flavobacterium spp., Aeromonas spp., Staphylococcus spp., Micrococcus spp., coryneforms, and fecal streptococci. Present in lower cell populations are LAB, B. thermosphacta, and Bacillus and Clostridium spores, as well as parasitic agents and enteric viruses (171, 216). The most important pathogens associated with red meat include Salmonella, STEC, C. perfringens, Campylobacter jejuni, Campylobacter coli, L. monocytogenes, S. aureus, and Y. enterocolitica. Levels of pathogens on meat carcasses can vary from 1 to >30 most probable number (MPN)/cm2. The prevalences of pathogens may differ within an animal species, as prevalence of Salmonella was higher on cow and bull compared to steer and heifer carcasses (68). Pathogen prevalence on carcasses may also be affected by the season of the year. For example, the prevalence of E. coli O157:H7 on cattle in North America is higher in the summer and early fall months.

Spoilage Spoilage of Fresh Meat

As indicated, Pseudomonas spp. are the dominating spoilage microbes on aerobically stored meat at cold temperatures, whereas gram-positive bacteria cause spoilage under vacuum packaging and other MAP conditions (216). While microbial activity, rate of growth, and spoilage processes are similar for lean and adipose tissues, spoilage odors may be associated with lower cell numbers of bacteria on adipose tissue, where most of the available glucose is depleted when populations reach 6 log CFU. Since lipolytic enzymes are not produced by microorganisms until carbohydrates are exhausted, fatty tissue spoilage does not depend on microbial lipase production (212, 213). In addition, amounts of soluble components (e.g., glucose, glycolytic intermediates, and amino acids) are lower in adipose than lean tissue, and soluble nutrients from the underlying tissue are not readily available (114, 212, 213). As a consequence, adipose tissue has less lactic acid than muscle tissue, and therefore, its pH is higher (i.e., approaching 7.0). Hence, spoilage of

Microbial Spoilage and Public Health Concerns adipose tissue may be comparable to that of dark, firm, dry (DFD) meat and may involve growth of S. putrefaciens (213). Growth rates of psychrotrophic bacteria, such as H. alvei, S. liquefaciens, and Lactobacillus plantarum, were more rapid on fat than on lean beef and pork tissues, but there was little difference in growth rates of other bacteria. Generally, spoilage of fat before lean tissue is unlikely, considering the restricted growth on dry fatty carcass surfaces or the presence of muscle tissue purge in packaged meats (216). Muscle tissues with pH of >6.0 appear darker than normal red meat due to increased muscle respiration rates, which decrease the depth of oxygen penetration and therefore reduce the oxymyoglobin formation, resulting in DFD meat (128). Excessive animal stress or exercise before slaughter induces animal metabolism levels that deplete muscle glycogen levels and lead to reduced lactic acid production and a higher ultimate muscle pH. In addition to beef, this condition may also appear in muscles of pigs and other meat animals. Absence of glucose and higher pH values allow faster degradation of amino acids by pseudomonads and detection of spoilage at lower bacterial cell densities (6 log CFU) than in normal meat (205). Under vacuum packaging or other MAP conditions, DFD meat also spoils rapidly by developing green discolorations due to growth of S. liquefaciens and S. putrefaciens, which outgrow LAB at the higher pH and in the absence of glucose and glucose-6-phosphate (118). The green discoloration is associated with formation of hydrogen sulfide from cysteine or glutathione by S. putrefaciens; hydrogen sulfide reacts with myoglobin to form green sulfmyoglobin. S. liquefaciens, even at low populations, produces small amounts of hydrogen sulfide that are associated with DFD meat spoilage (216). Pale, soft, exudative (PSE) muscle tissue or meat is a condition that affects 5 to 20% of pig carcasses; it may also occur in poultry and to a lesser extent in beef (216). The meat appears pale in color and soft in texture, and fluids are exudated from the muscle (114). PSE meat is the result of postmortem glycolysis decreasing the muscle pH to its ultimate value while the muscle temperature is still high, and this condition is associated with porcine stress syndrome, a sensitivity of the animals to mild stress that may lead to malignant hyperthermia and death. It is unclear whether the lower pH causes PSE meat to spoil slower than meat of normal pH. However, amounts of soluble low-molecular-weight compounds should be similar in PSE and normal meat; hence, spoilage development should be at similar rates (216). Comminuted products, including ground meat, spoil faster than externally contaminated intact meat cuts because of higher levels of initial contamination present in

6.  Meat, Poultry, and Seafood

127

trimmings of larger surface area exposed to contamination, cross-contamination among trimmings, microbial spreading during grinding, and the release of fluids and nutrients from cells ruptured due to grinding (169, 171, 216). The types of dominating microorganisms and spoilage defects are similar in intact and comminuted products of the same origin. Dominant microflora members in aerobically stored comminuted products include Pseudomonas, Acinetobacter, and Moraxella, whereas LAB may dominate in the interior due to oxygen limitation. Enterobacteriaceae are more frequently found in comminuted than in intact products. Spoilage microbes dominating in spoiled refrigerated fresh sausage are similar to those present in other fresh meats but may also include B. thermosphacta (216).

spores is variable and that there is a need for specific conditions that allow spore germination and growth (1, 328). Postpackaging heat shrink treatments of vacuum packs have the potential to accelerate the onset of clostridial blown-pack spoilage (1). Prediction or prevention of this type of spoilage is still difficult. Good management practices during meat processing are not always sufficient for its prevention. For control, Moschonas et al. (199) suggested avoiding higher temperature (90°C for 3 s or 70°C for 10 s) during heat shrinking of packages, which could stimulate spore germination, as well as storing vacuum-packaged meats at lower temperatures (e.g., −1.5°C). Adam et al. (1) discussed research needs related to meat spoilage caused by psychrotolerant clostridia.

Meat Spoilage by Psychrotrophic Clostridia

Spoilage of Processed Meats

While certain species of clostridia are well recognized as major food safety concerns due to toxin production, others, known in general as psychrotolerant clostridia, cause spoilage in vacuum-packaged meats such as “deep tissue” or “bone taint” spoilage. Some clostridia are also known as causing surface spoilage and “blownpack” spoilage (1). These unusual types of spoilage of vacuum-packaged refrigerated fresh and cooked meat and poultry products are caused by psychrophilic, psychrotrophic, or psychrotolerant clostridia, such as the generally closely related species C. laramie, C. beijerinckii, C. lituseburense, C. algidicarnis, C. algidixylanolyticum, C. estertheticum, C. frigidicarnis, and C. gasigenes, which can germinate, grow, and sporulate at 2°C or lower (0°C) in meat of normal pH (27, 163). As the meat undergoes proteolysis, the spoilage is characterized by muscle softening, large amounts (blown pack) of hydrogen sulfide (especially by C. estertheticum and C. gasigenes) and associated offensive odors, and release of purge or exudate, while the initial pinkish red becomes green (1, 197, 198, 199). Such spoilage has occurred in the United States, Europe, South Africa, and New Zealand. The microbes involved are not considered pathogenic, but the meat is organoleptically unacceptable. Blown-pack spoilage has been found in beef primal cuts, lamb, venison, cooked dog rolls, precooked turkey, roast beef, and sous-vide products, resulting in major financial losses as it makes the meat unacceptable for consumption. Initially, contaminated product spoils on the surface, with little gas accumulation (surface spoilage), and the bacteria are found in the purge or by swabbing the meat surface. Blown-pack spoilage occurs sporadically and only in a fraction of the packs in a consignment. This indicates that contamination with

Processed products including frankfurters, bologna, cooked or fermented sausage, and luncheon meats develop spoilage defects such as slime, souring, and greening. Slime is usually confined to the surface of the product and is associated with growth of yeasts, Lactobacillus, Enterococcus, and B. thermosphacta in the presence of moisture (216). These microbes metabolize sugars to produce organic acids that result in souring on product surfaces under casings or packaging films. While greening of fresh meats may be due to hydrogen sulfide production, greening of nitrite-cured meat products may develop in the presence of hydrogen peroxide. The compound, formed during growth of catalase-negative bacteria when surfaces of vacuumpackaged or MAP products are exposed to air, reacts with nitrosohemochrome to produce green choleglobin. This is possible because muscle catalase, which breaks down hydrogen peroxide, is inactivated during cooking. The most common causative agents of such greening are Lactobacillus viridescens and species of Streptococcus and Leuconostoc, which may be introduced in the product before or following processing (216). Putrid spoilage of cured meats by gram-negative psychrotrophic bacteria under refrigeration is inhibited by reduced water activity (145). These products, however, are spoiled by lower-water-activity-tolerating lactobacilli or micrococci. While lactobacilli predominate in spoiled vacuum-packaged or MAP products, aerobic micrococci predominate on cured meats stored aerobically. Inclusion of glucose in cured meats allows the formation of a slimy dextran layer by Leuconostoc species or other bacteria, including B. thermosphacta. Drycured products spoil mostly by yeasts and molds because the low water activity, presence of nitrite, and exposure to smoke suppress bacterial growth (302). Growth of

128 spoilage fungi in these products may be inhibited with addition of antifungal agents. Dry-cured meats may also be spoiled because microbial growth occurred before proper and adequate salt penetration, curing, and drying, whereas when properly prepared and stored, dried meat products remain stable. Xerotolerant molds are inhibited at water levels of about 15% and below (216).

Food Safety Pathogens in Meat

Of all the pathogens associated with muscle foods listed in previous sections, the most important of current concern in fresh meat are Salmonella, Campylobacter, and E. coli serotype O157:H7, whereas L. monocytogenes is of concern in ready-to-eat products. Additional pathogens receiving attention are STEC other than E. coli O157:H7 and TSE prions, and C. difficile and Mycobacterium avium subsp. paratuberculosis have recently been suggested as potential concerns (279). In general, the prevalence rates of pathogens in cattle and beef vary considerably among surveys. Rhoades et al. (242) estimated the relative prevalence of pathogens at different stages of production and processing by calculating average prevalences observed in multiple surveys, weighted by sample number. Based on the data, the mean prevalence rates (and ranges of means from individual surveys) of E. coli O157:H7 were 6.2% (0.0 to 57%), 44% (7.3 to 76%), 0.3% (0.0 to 0.5%), and 1.2% (0.0 to 17%) for feces, hides, chilled carcasses, and raw beef products, respectively. The corresponding rates for Salmonella were 2.9% (0.0 to 5.5%), 60% (15 to 71%), 1.3% (0.2 to 6.0%), and 3.8% (0.0 to 7.5%), whereas for L. monocytogenes the mean prevalence rates were 19% (4.8 to 29%), 12% (10 to 13%), and 10% (1.6 to 24%) for feces, hides, and raw beef products, respectively. Seasonal variation was evident in many surveys, with fecal prevalence rates for E. coli O157:H7 and Salmonella generally higher in the warmer months. Prevalence is usually affected by pathogen type, animal type and age, feed, housing, and general meat production practices (208, 242).

Shiga Toxin-Producing or Verotoxigenic E. coli

E. coli O157:H7 is considered the most important human pathogen present in ruminants, which are its natural reservoir. The pathogen has been implicated in large outbreaks as well as in sporadic cases of hemorrhagic colitis and the sometimes fatal hemolytic uremic syndrome (HUS) (12). Hence, it has received consider-

Microbial Spoilage and Public Health Concerns able attention by regulatory authorities, public health agencies, industry, and researchers (113). E. coli O157: H7 has been recognized as a foodborne pathogen in beef since 1982−1983 but did not dominate beef safety­associated concerns until 1993, following a major outbreak associated with a fast food restaurant chain. Recently, however, evidence that some additional Shiga toxin-producing E. coli serotypes are also involved in human illness has been accumulating. This issue is gaining attention and may lead to new regulatory requirements and industry action (71, 81, 113, 127, 165). Enterohemorrhagic E. coli strains, including STEC strains, reside in the gastrointestinal tract of cattle, in part colonizing the terminal colon, and are then spread with feces in the environment, subsequently contaminating animal exteriors and muscle foods of land origin, water, soil, and consequently foods of plant origin (12, 165). Some animals shed large numbers (>105 CFU/g) of cells of E. coli O157 and are considered “supershedders,” constituting the major source of such contamination in the environment and foods (9). Certain animals are considered “persistent shedders” because they shed E. coli O157:H7 cells for long periods of time (33). It is important to identify factors that result in persistentor supershedding cattle in order to develop better interventions to reduce E. coli O157:H7 contamination of feces. STEC serotypes other than O157:H7 that are of concern for transmission on meat and cause human illness include, in order of association with human illness in 2009, O26, O103, O111, O121, O45, O145, O124, O118, O69, and O128 (www.cdc.gov). They can cause mild diarrhea, severe bloody diarrhea (hemorrhagic colitis), or in some cases, HUS, characterized by micro­ angiopathic hemolytic anemia, thrombocytopenia, and acute renal failure (144, 165). STEC produce either Shiga toxin 1 (Stx1), which is essentially identical to the Shigella dysenteriae type 1 toxin, Shiga toxin 2 (Stx2), or a combination of the two toxins. The Shiga-like toxins, or verocytotoxins, are the major virulence factor of STEC (see chapter 12). Many reviews of STEC and their association with ruminants and meat have recently been published (71, 81, 113, 127, 165). According to a review by Hussein (144), prevalence rates of E. coli O157:H7 in beef cattle were 0.3 to 19.7% in feedlots and 0.7 to 27.3% on pasture, and corresponding rates of non-O157 STEC were 4.6 to 55.9% and 4.7 to 44.8%, respectively. In meat products, prevalence rates of E. coli O157:H7 were 0.01 to 43.4% on whole carcasses, 0.1 to 54.2% in ground beef, 0.1 to 4.4% in sausage, and 1.1 to 36.0% in various retail cuts, whereas corresponding prevalence rates

6.  Meat, Poultry, and Seafood of non-O157 STEC were 1.7 to 58.0%, 2.4 to 30.0%, 17.0 to 49.2%, and 11.4 to 49.6%, respectively (144). Of 162 STEC serotypes isolated from beef products, 43 were also detected in HUS patients and 36 are known to cause other human illnesses. Of 373 STEC serotypes isolated from cattle feces or hides, 65 were detected in HUS patients and 62 are known to cause other human illnesses (144). In addition to beef, Fratamico et al. (104) isolated 58 STEC serotypes from swine feces and ­determined that 13%, 6%, 80%, 21%, 6.4%, 4.6%, 42.9%, 11.4%, and 0.46% of the isolates carried the stx1, stx2, stx2e, estIa, estIb, fedA, astA, hly933, and cdt-III genes, respectively. None of the strains possessed the elt, bfp, faeG, fanA, fasA, fimF41a, cnf-1, cnf-2, eae, cdt-I, or cdt-IV genes. Osés et al. (224) reported that although lamb meat consumption has not been associated with STEC outbreaks, sheep may be an important reservoir, as the results of a study revealed that E. coli carrying stx1, stx2, and eae virulence genes were most prevalent in slaughterhouses (69%), whereas E. coli strains with the eae gene alone were found more frequently in the processing plant (32%), and stx1and stx2-positive E. coli isolates were predominant in butcheries (9 to 10%). However, Vettorato et al. (312) determined that virulent E. coli were not common in butcheries. Xia et al. (326) screened for Shiga toxin genes among 7,258 E. coli isolates collected by the U.S. National Antimicrobial Resistance Monitoring System retail meat program from 2002 to 2007. Seventeen isolates (16 from ground beef and 1 from a pork chop) were positive for stx genes, including 5 isolates ­positive for both stx1 and stx2, 2 positive for stx1, and 10 positive for stx2. The 17 STEC strains belonged to 10 serotypes: O83:H8, O8:H16, O15:H16, O15:H17, O88:H38, ONT:H51, ONT:H2, ONT:H10, ONT:H7, and ONT: H46. None of the isolates contained eae, whereas seven carried enterohemorrhagic E. coli hlyA. All except one isolate exhibited toxic effects on Vero cells. Subtyping of the 17 STEC isolates by pulsed-field gel electrophoresis yielded 14 distinct restriction patterns. Hence, according to Xia et al. (326), retail meats, mainly ground beef, were contaminated with diverse STEC strains. Such data indicate the need to control these pathogens in order to enhance meat safety (144). Based on limited existing data, it is expected that interventions used to control E. coli O157:H7 in beef should also be effective against other STEC serotypes.

Other Bacterial Pathogens of Potential Concern

Additional pathogenic bacteria that are currently described as potentially transmitted through meat prod-

129 ucts and are responsible for human illness include Mycobacterium avium subsp. paratuberculosis (M. paratuberculosis) and C. difficile. Mycobacterium paratuberculosis causes Johne’s disease, a chronic enteritis in cattle and other ruminants (195, 200). Humans develop Crohn’s disease, also a chronic inflammatory condition of the intestinal wall, with an appearance similar to Johne’s disease in animals. Based on this similarity, it has been suggested that M. paratuberculosis may also be involved in Crohn’s disease; however, it should be noted that intestinal tissues from Crohn’s disease patients do not contain large numbers of acid-fast M. paratuberculosis; hence, the relationship of M. paratuberculosis with Crohn’s disease remains to be confirmed (75, 87, 200, 259). Samples of blood, liver, kidney, lymph nodes, and muscle tissue from the carcasses of five cows with advanced Johne’s disease as well as samples of cooked muscle tissues and cooked hamburger patties that contained chopped mesenteric lymph nodes were tested for the pathogen. M. paratuberculosis cells were detected at >103 CFU/g from 7 of 15 liver and mesenteric and ileocecal lymph node samples, and from 5 of 15 kidney and superficial inguinal and prescapular lymph node samples (200). M. paratuberculosis was also detected in 1 and 6 of 50 samples of raw, chilled, or frozen meat, respectively; in 1 of 15 samples of meat cooked to 61°C; and in 1 of 40 samples of meat cooked to £ 70°C. In addition, the microbe was detected in 2 of 4 samples of mesenteric lymph nodes cooked to 61°C, but not in samples cooked to £ 70°C. Based on this study, M. paratuberculosis may be present in meat from infected animals at low numbers, but it is likely inactivated when meat is cooked to a well-done condition (200). Nevertheless, minimizing exposure of humans to M. paratuberculosis is an appropriate precautionary measure (123). C. difficile has also received attention in recent years as a potential meatborne pathogen. This is because the organism is considered an emerging pathogen of increased incidence and severity in nosocomial infections. In addition, studies have revealed an overlap between calf and human isolates, including two of the predominant outbreak types (i.e., 027 and 017); further, C. difficile has been isolated from retail meat samples. However, C. difficile types from human and animal populations have not been compared in detail (248). C. difficile is a gram-positive, sporogenic, anaerobic bacillus that can cause, particularly following antibiotic treatment, intestinal disease (57, 157, 193). C. difficileassociated disease has now become a significant clinical problem in North America and Europe; most cases have been documented in hospitals. Primary risk factors for

130 acquiring C. difficile disease are long-term hospitalization, age of >65 years, and antibiotic therapy. Overall, the incidence and severity of the disease seem to be increasing, and reports of severe cases with a community onset are also increasing. In addition, some of the community-acquired cases have been associated with a low-risk population (i.e., young individuals, without previous antibiotic therapy or hospitalization). Potential sources of the organism are farm (horses, pigs, cows) and domestic (dogs, cats) animals (193, 248). A survey in The Netherlands found C. difficile in 8 of 500 meat samples tested (i.e., 1 from lamb [6.3%] and 7 from chicken meat [2.7%]). However, the strains isolated belonged to ribotypes different from those most frequently found in human patients, except for ribotype 001, which was found in one chicken meat sample. According to De Boer et al. (57), this suggests that matrices other than meat may serve as sources of C. difficile. Songer et al. (288) found that 37 (42.0%) of 88 retail meat samples yielded C. difficile, including 42.4% of beef, 41.3% of pork, and 44.4% of turkey products. Ready-to-eat products were more commonly, but not significantly, culture positive (11/23; 47.8%) than uncooked meats (26/65). Most C. difficile isolates were recovered from pork braunschweiger sausage (62.5%) and ground beef (50.0%). C. difficile was also isolated from 3 of 100 retail ground meat samples in Austria (157). The authors (157) concluded that the presence of one isolate of human origin could be due to contamination by human fecal shedders during food processing rather than confirming zoonotic potential. Current evidence is inadequate to indicate that C. difficile contamination of meat can result in human infection (57, 155, 319, 320).

Transmissible Spongiform Encephalopathies

TSEs, also known as prion diseases, comprise an extensive group of neurodegenerative diseases in animals and humans (318). One such disease is BSE, a progressive neurological disease affecting the central nervous system of cattle, which was first diagnosed in the United Kingdom in 1986. Scrapie is another prion disease affecting sheep and goats and has been known for over 200 years. Another TSE is Creutzfeldt-Jakob disease (CJD) in humans and the variant type of CJD (vCJD), which was first reported in the United Kingdom following the discovery of BSE. The causative agent of TSE is considered to be an abnormal form of prion protein, but the details of its pathogenic mechanism are unclear (307). Proper controls for BSE should be implemented based on science and risk assessment (184, 307). BSE preventive controls, such as feed bans and control of specified risk materials during slaughter, have led to its

Microbial Spoilage and Public Health Concerns containment and mitigation. However, it is important that established programs be implemented worldwide and that international collaboration continue in order to prevent similar or related disasters in the future (184, 279). See chapter 25 for additional information regarding TSE.

Processing and Preservation Regulatory Requirements for Pathogen Control

Concerns about E. coli O157:H7 STEC in the early 1990s led to regulatory changes in the United States (96), including the following: reinforcement of a policy requiring removal by knife trimming of visible soil on carcasses (“zero tolerance”) before washing or decontamination; declaration of E. coli O157:H7 as an “adulterant” in ground beef (96) and in all other nonintact beef products (100); and revision of inspection regulations to require formal sanitation standard operating procedures, implementation of HACCP, and compliance with performance criteria for E. coli biotype I to verify process control (done by the processor) and Salmonella as a verification of HACCP and for tracking pathogen reduction (done by USDA-FSIS). Additional FSIS directives to industry in the past 5 to 15 years include reevaluation of HACCP plans, and testing of ground beef and ground beef raw materials recently included testing of trimmings derived from steaks/roasts (“bench trim”) for E. coli O157:H7 (www.fsis.usda.gov). Another U.S. regulation, enacted in 2003, addressed control of L. monocytogenes in ready-to-eat meat and poultry products (101) that may be contaminated after processing and allow growth of the pathogen during distribution and storage, even at refrigeration temperatures, before consumption. An associated development that may be considered is the increasing list of chemical antimicrobials permitted for use to reduce pathogen contamination in various types of meat and poultry products in the United States, as described in FSIS Directive 7120.1 (http://www.fsis.usda.gov/About_FSIS/labeling_&_consumer_protection/index.asp). Based on these developments, the meat industry has responded with changes in their meat slaughter, processing, and preservation procedures in order to comply with new regulatory r­equirements and provide the public with safe meat products (277, 279). In addition to regulatory microbiological performance criteria, the meat processing industry, especially the ground beef sector, has imposed microbial specifications and other requirements on their raw beef suppliers. An effective pathogen control program should include activities employed preharvest

6.  Meat, Poultry, and Seafood or in the field, postharvest or during processing in the plant, at retail and food service, and at home (171, 251, 277, 291).

Preharvest Pathogen Control

Reasons for preharvest pathogen control include reduction of pathogen sources and levels in order to reduce direct animal-to-human transmission, water contamination, and produce or plant origin food contamination through animal feces, manure, and contaminated water. Antimicrobial interventions considered or explored for control of animal contamination in the field include animal diet modifications, use of feed additives or supplements, antibiotic treatments, bacteriophage therapy, vaccination, competitive exclusion, prebiotics or probiotics; and animal production management practices include animal pen management, use of clean feed and chlorinated water, clean and unstressful transportation of animals to slaughter, clean lairage, and animal cleaning before slaughter. With the exception of good production and management practices, and to some extent feeding of probiotics, the remaining approaches are still in the experimental stage or of limited use (169, 171, 274, 277, 278, 291).

Carcass Decontamination

Carcass or meat decontamination interventions are instantaneous or of short duration, of mild intensity, and inadequate for complete microbial inactivation; and if not properly applied, they may cause pathogen spreading, cross-contamination, or penetration into the tissue. They should be examined as to whether they cause cell inactivation, injury, or antimicrobial resistance through selection/adaptation/cross-protection or alteration of the metabolic activity of survivors. Other issues to be examined include potential changes in microbial ecology, such as inhibition of normal gram-negative bacterial flora members and selection of yeasts or LAB by acid treatments and gram-negative bacteria by water/ steam treatments (251, 277, 279). Available data indicate that numbers of beef recalls for E. coli O157:H7 contamination were low from 1994 to 1997 (2 to 6 recalls per year), when the ­pathogen was becoming an increasing concern and procedures to deal with it and methodologies for its detection were being developed. As methods improved and scrutiny increased, for the years 1998, 1999, 2000, 2001, and 2002, recalls increased respectively to 13, 10, 28, 27, and 34. Then, total recalls for 2003, 2004, 2005, 2006, 2007, 2008, 2009, and 2010 were 10, 6, 4, 8, 22, 13, 15, and 7, respectively (FSIS data [www. fsis.usda.gov]). For the USDA-FSIS ground beef testing

131 program for the years 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, and 2009, E. coli O157:H7-positive rates were 0.86, 0.78, 0.30, 0.18, 0.17, 0.17, 0.23, 0.47, and 0.30%, respectively. Salmonella prevalence in ground beef for all plants combined, as tested by USDA-FSIS under the HACCP pathogen reduction regulation of 1996, were reduced from a baseline of 7.5% in 1994 to 6.4, 4.3, 3.3, 2.8, 2.6, 1.7, 1.6, 1.1, 2.0, 2.7, 2.4, and 1.9% in 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, and 2009, respectively. U.S. FoodNet (active surveillance) data obtained by the CDC for cases of E. coli O157: H7 per 100,000 population were 1.73, 1.1, 0.9, 1.06, 1.31, 1.2, 1.12, and 0.99 for the years 2002, 2003, 2004, 2005, 2006, 2007, 2008, and 2009, respectively, compared to 2.7 in 1996; corresponding data for non-O157 STEC were 0.33, 0.46, 0.57, 0.45, and 0.57 for 2005, 2006, 2007, 2008, and 2009, whereas HUS cases for 2005, 2006, 2007, 2008, and 2009 were 0.94, 1.63, 2.01, 1.75, and 1.40, respectively. Overall, decontamination interventions are useful because they reduce total contamination (1 to 3 log CFU) and pathogen prevalence, assist plants in meeting regulatory criteria and industry specifications, reduce crosscontamination, improve product quality, and should increase product safety by reducing the probability of illness if the product is undercooked (277, 279).

Nonintact/Marinated/Tenderized Meat Products

As discussed by Sofos et al. (279) and Sofos (282), a major proportion of steaks and roasts derived from muscles of lower tenderness, which in the United States constitute 74% of the beef carcass, may be subjected to mechanical tenderization, moisture enhancement, marination, or restructuring to yield nonintact products of increased tenderness, juiciness, and flavor for use in hotel, restaurant, and institutional settings. The total annual servings are estimated at 36 billion in the United States (22). Nonintact meat products include intact meat cuts such as chucks, ribs, tenderloins, striploins, and top sirloin butts and rounds that are injected with marination, flavoring, moisture-enhancing, or tenderizing solutions or mechanically tenderized by treatment with solid- or hollow-needle injectors or blades, or with cubing, frenching, or pounding devices (282, 285, 286). Also included are any comminuted products processed by chopping, grinding, flaking, or mincing, as well as manufacturing beef trimmings destined to be processed into formed items such as gyros (286). An important public health concern associated with these products is the potential translocation or

132 e­ntrapment of pathogen cells, such as E. coli O157: H7, into their interior from the surface of intact beef cuts and through cross-contamination (e.g., by needle injection and/or recycling of brines). There is concern that consumers may perceive nonintact products as intact and hence undercook them, without considering the potential presence of contamination in the interior of the product (99, 100, 282, 286). Subsequently, E. coli O157:H7 in the interior may survive cooking and cause illness, especially if the injected ingredients interfere with inactivation or increase heat resistance (286). The public health risk associated with nonintact meat products in the United States is evidenced by outbreaks of E. coli O157:H7 illness linked to consumption of such products (99, 100, 102). Since 1999, the FSIS has included nonintact products in the category of adulterated if samples are found contaminated with E. coli O157:H7; raw ground beef contaminated with the same pathogen was declared adulterated in 1994 (98). The safety of nonintact meat products has been examined by risk assessments (99, 100) by the U.S. National Advisory Committee on Microbiological Criteria for Foods (202) and in various reviews (24, 164, 279, 282, 285, 308). Results summarized by Sofos and Geornaras (286) reveal that the risks associated with nonintact meat products may be controlled through implementation of effective decontamination interventions, application of approved and effective antimicrobial treatments to subprimals before tenderization, proper chilling and rotation of injection solutions, potential use of antimicrobials in injection brines, effective sanitation and temperature controls, and cooking procedures that are selected based on product characteristics. It is also important that manufacturers follow industry-recommende­d best practices (4, 22, 203).

Control of L. monocytogenes in Ready-To-Eat Products

As summarized by Sofos and Geornaras (285), numerous studies have demonstrated, in ready-to-eat products, the antilisterial effects of generally recognized as safe chemicals, including potassium and sodium lactate, acetate, diacetate, and acetic and lactic acids. Postlethality physical treatments, such as radiant heating, flash steam heating, steam pasteurization, or hot water immersion may also be applied to control L. monocytogenes in such products. Postprocessing immersion of frankfurters, bologna, ham, smoked sausage, and turkey breast, formulated with or without antimicrobials, in solutions of acetic or lactic acid, nisin, benzoate, sorbate, and their combinations resulted in reductions in initial populations of the pathogen and inhibitory or bacteriocidal effects against survivors during

Microbial Spoilage and Public Health Concerns product storage, varying with type of product, concentration of antimicrobial, length of exposure time, type or combination of antimicrobials, and sequence of exposure. Other chemicals with the potential to control L. monocytogenes in ready-to-eat products include acidic calcium sulfate, lauric arginate, pediocin, and cetylpyridinium chloride. In summary, postprocessing chemical antimicrobial treatments could potentially be considered as options by meat processors for alternatives 1 and 2 of the USDA-FSIS Listeria control regulation (101, 103) to reduce or eliminate the pathogen on products contaminated at the postprocessing stage. Processors, however, need to develop and validate their own formulations to fit their product specifications and expectations (171, 279, 285).

Meat Packaging Systems General

The bright red color occurring in fresh aerobically packaged meat is preferred by consumers when purchasing fresh meat because they consider it an indicator of freshness and wholesomeness (153). Various packaging technologies have been developed to promote and maintain oxymyoglobin (the red color) on the surface of red meat. It is important, however, that any packaging technology that prevents metmyoglobin formation not “mask” microbial spoilage and off-odors or growth of pathogenic bacteria; otherwise, the consumers could have a false sense of wholesomeness and safety at the time of purchase (142). Technological advances in packaging systems, equipment, and plastic materials have provided improvements in MAP of meat. However, additional technological and logistical considerations are necessary in the proper use of such systems for raw fresh meat. Popular MAP approaches have involved enclosure in master packs of airpermeable overwrapped product in trays, low-O2 formats of shrunk-film vacuum packaging, MAP with nitrogen (N2) and carbon dioxide (CO2), high-O2 MAP, and no O2 MAP with low CO (0.4%) plus CO2 and N2 (CO-MAP). All these have advantages and disadvantages (189).

Aerobic Packaging

When stored aerobically, in order to appear bright red, fresh red meat is placed on Styrofoam trays and overwrapped with PVC film (44). The relatively high oxygen permeability of PVC film allows meat surfaces in contact with oxygen to develop the attractive bright red color due to the reaction with myoglobin and hemoglobin to form oxymyoglobin and oxyhemoglobin, respectively. This economical and easy-to-use technology is widespread in retail stores. A major disadvantage of PVC meat overwraps is the short shelf life (e.g., 5 to 7 days for steaks

6.  Meat, Poultry, and Seafood or roasts and less for ground meat) due to browning caused by meat pigment oxidation to form metmyoglobin. Another disadvantage is the frequent occurrence of leaky packages due to the susceptibility of PVC films to punctures and tearing (44).

High-Oxygen MAP

High-oxygen (70 to 80% O2, plus 20 to 30% CO2) MAP systems have been used as an alternative to wrapping with PVC film to produce fresh meat that is bright red and may last as long as 10 to 14 days (compared to 3 to 7 days in air). However, exposure to oxygen allows aerobic bacterial growth and development of rancid flavors (44). In commerce, this system is used in conjunction with natural antioxidants such as rosemary, to provide protection from rancid off-flavors. Overall, the shelf life in this system is greater than in air but much less than in vacuum (44, 189, 220, 257, 258). High-oxygen packaging involves the use of coextruded polyamide (nylon)-polyethylene films in which the nylon provides strength and the polyethylene provides gas and water vapor barrier properties and heat sealability. The system includes shipment of PVC retail packages to stores in a case-ready format for display, which allows supermarkets to offer retail fresh meat products at lower cost as they avoid the expense of in-store packaging. The elevated oxygen levels used in high-oxygen MAP saturate the meat pigments with oxygen, which slows down surface metmyoglobin formation. CO2 exceeds levels of 20% in these MAP systems and acts as an antimicrobial. In addition, the MAP film is more puncture resistant than PVC film (44). High-oxygen MAP may accelerate lipid oxidation and the generation of off-flavors as well as darkening of cut bone surfaces and premature browning during cooking. Rancid flavor development may be inhibited through the injection of muscle cuts with antioxidants, such as sodium tripolyphosphate or rosemary extracts (254, 260). However, no acceptable antioxidant treatment is available for ground beef in high-oxygen MAP because of labeling limitations (44). An additional issue in ground beef stored in high-oxygen MAP is associated with premature browning during cooking. Ground beef exposed to high oxygen appears fully cooked at temperatures as low as 57°C, a temperature lower than the recommended internal 71°C for destruction of E. coli O157:H7 (261). This makes cooked color unacceptable as the sole indicator of doneness (44).

Carbon Monoxide Packaging

A second option for MAP is a “no-oxygen” environment, similar to vacuum packaging, which uses 0.4% CO in

133 combination with CO2 and N2. The advantage of this approach is that it offers all the advantages of vacuum packaging (i.e., extended shelf life and suppression of oxidized, off-flavors) together with an attractive red color similar to that of oxygen-exposed fresh meat (258). Since 2002, CO has been permitted as a MAP gas for use during meat distribution in the United States (247), and in 2004, low levels of CO were allowed for retail fresh meat packaging (44). Use of low CO levels is the most recent meat MAP technology for meat (COMAP). It is an essentially oxygen-free MAP with low levels (0.4%) of CO, 20 to 30% CO2, and the remainder N2 (44, 154). As listed by Cornforth and Hunt (44), the advantages of CO, compared to aerobic or high-oxygen MAP for meat, include the following: desirable red color and low microbial counts for up to 28 and 35 days in ground beef and steaks or roasts, respectively; better flavor acceptability without oxidized flavors; no bone darkening; no premature browning during cooking; potentially improved tenderness because of reduced protein oxidation; and additional action of endogenous tenderizing enzymes during the extended storage. The disadvantages of CO-MAP include the negative image of CO by consumers who consider CO as potentially hazardous and concerns that the spoiled fresh meat appearance may be masked (44, 258). The market share of case-ready packaged meat increased from <50% of all fresh meat packages in 2002 to 64% in 2007 (236). According to one marketing system, retail fresh meat cuts from all livestock species are individually wrapped in PVC film and distributed in a “master bag” under an atmosphere of 0.4% CO, 20 to 100% CO2, and 0 to 80% nitrogen. At the retail store, the master bag is opened, and individually wrapped cuts are placed in retail display in air and without further contact with CO. In another system, the retail package, not just the bulk pack, contains 0.4% CO (44). The red color of CO-treated meat changes gradually to a brownish discoloration when cuts are stored under aerobic conditions, while the continuous presence of CO maintains the red color (44, 142). Ramamoorthi et al. (237) suggested that CO-MAP could be used to preserve the color of beef irradiated at doses sufficient to reduce microbial loads to safe levels during storage. Potential health issues to consider with use of CO in packaging of foods is that it binds more strongly than oxygen to hemoglobin, which would impair oxygen transport to muscle tissues. In human physiology, CO regulates blood fluidity and flow by inhibiting vasomotor tone, smooth muscle cell proliferation, and platelet aggregation. It is noted, however, that the breakdown of hemoproteins by heme oxidase results in the formation of small amounts

134 of CO. Normal CO production in the body leads to a carboxyhemoglobin concentration of about 0.5%. The average carboxyhemoglobin level (from both endogenous and environmental CO) is 1.2 to 1.5% and 3 to 4% in nonsmokers and smokers, respectively. No adverse effects are observed by CO exposure at levels below 35 ppm for 1 h (44). Meat products have a long history of exposure to CO as a component of wood smoke. From a regulatory standpoint, MAP gases are regarded as processing aids, not food additives (247), and as such, they are not listed on the product label. Therefore, consumers are not informed that CO or other gases are used in the packaging of the product. One option considered for consumer awareness is labeling of the presence of gases (44, 258).

Smart Packaging Systems

The term “smart packaging” covers a range of new packaging concepts with the objective to maintain or extend product quality and shelf life. Most such systems are classified in the categories of active packaging or intelligent packaging, and in recent years there has been increased interest in their use in meat products (80, 167, 171, 220). Active packaging is the term used to describe the incorporation of additives into packaging systems: applied as loose within the package, attached to the inside of the packaging material, or incorporated into the packaging material. Active packaging should contribute to food preservation more than just by serving as a barrier to the environment; it should extend shelf life by modifying conditions within the package. Additives for active packaging systems include oxygen scavengers, carbon dioxide scavengers and emitters, moisture control agents, and antimicrobials (220). In recent years, there has been a growing interest in the use of natural antimicrobials, especially nisin, in food packaging applications. An extruded composite food packaging film containing pectin, polylactic acids, and nisin was developed to inhibit L. monocytogenes (156). Studies have also addressed development of more natural, disposable, potentially biodegradable, and recyclable food packaging materials; incorporation of natural antimicrobials into biobased films; and development of edible films and coatings (32, 42, 77, 80, 185). Intelligent packaging systems are described as those that monitor conditions within packaged foods during transportation and storage and provide information relative to product quality. This is accomplished with sensor technologies, indicators (e.g., integrity, freshness, and time-temperature indicators), and radio frequency identification. Additional development is necessary for such packaging systems to become used in the meat industry (167, 171, 220).

Microbial Spoilage and Public Health Concerns POULTRY

General

Approximately 31% of meat consumed worldwide is poultry; the only meat eaten more frequently than poultry is pork (95). The two principal types of poultry meat consumed are chicken and turkey, with turkey being only 14% of worldwide poultry meat production (149).

Poultry Meat Contamination

The microbial population associated with poultry carcasses at the end of processing is a combination of the bacteria associated with the skin and feathers on the live birds and the population that is acquired during processing (38). Carcasses at the end of processing have a population of mesophilic bacteria ranging from 101 to 104 CFU/cm2 (304). Bacterial populations associated with carcasses at the end of processing are predominantly gram negative and include Acinetobacter/Moraxella spp., Enterobacteriaceae, Flavobacterium spp., and Pseudomonas spp. (14, 129, 179, 180). During refrigerated storage, gram-negative psychrotrophic bacteria of the Pseudomonas/Moraxella/Acinetobacter group become predominant (304). Each step in processing raw poultry influences the level of spoilage bacteria on the product, with some steps leading to an increase in the microbial load and others leading to a decrease.

Processing General Procedures

Poultry in industrialized countries is brought to a central processing plant from different farms. Birds arrive in crates or containers that hold large numbers of birds that can be easily unloaded. Workers manually transfer the birds individually to moving shackles that convey the birds to an area for stunning. Electrical stunning is currently the most common method and involves the birds moving on the shackle line until their heads come into contact with a brine solution. In the United States, settings are usually low voltage and high frequency, whereas in the EU higher voltages and lower frequencies are mainly used (234). Another method of stunning is controlled atmosphere stunning and is found with increasing frequency in the EU. In this stunning procedure, birds are placed into an atmosphere low in or lacking oxygen and containing argon, or nitrogen, or carbon dioxide, or a combination thereof (234). A third method of stunning that is just coming into limited use in the United States and the EU is known as a low atmospheric pressure system, whereby a vacuum pump reduces oxygen tension

6.  Meat, Poultry, and Seafood in the atmosphere (316). The stunning process renders the birds unconscious as they proceed to the killing and bleeding stage and from there to scalding and picking.

Scalding

Scalding is a process in which birds are immersed in a hot medium, usually water, to ensure easy removal of the feathers and cuticle. Temperatures in the scalding tanks are regulated depending on the final product desired. The “soft scald” method is popular in Mexico and the EU and uses scald temperatures between 50 and 52°C. The “hard scald” method is popular in the United States and uses temperatures of 56 to 63°C. Soft scalding does not cause significant damage to the outer skin layers, the stratum corneum, or cuticle and maintains the yellow-pigmented layer of the skin intact. The product of soft scalding is fresh poultry with yellow skin, which is highly desirable in some parts of the world as an indicator of “corn-fed” or healthy birds (252). Conventional scalders consist of a single-tank, two-pass system, or there are countercurrent or counterflow scalders with three tanks and a two-pass system in which water mixes across the two lines of carcasses within each tank. Cason et al. (36) sampled water and carcasses from both types of scalders installed side by side in the same processing plant and found that although numbers of aerobic bacteria were significantly reduced in the third tank of the counterflow scalder compared to the second tank or compared to the single tank of the conventional scalder, the numbers of aerobic bacteria in carcass rinses were not affected by scalder design. Steam scalding is another option, but steam causes damage to the outer cuticle of the skin, which makes it unsuitable for the chilled-product market (16).

Picking

Picking is the process of removing feathers from the carcass after the scalding process. Mechanical pickers are large pieces of equipment with rubber fingers that grab the feathers and pull them from the follicles. Picker fingers contribute to cross-contamination in poultry processing plants by damaging the intact skin, allowing bacteria to become lodged underneath the surface and subsequently proliferate (297). In addition, water sprays are used during defeathering, which may create an aerosol of bacteria that can contaminate other carcasses and equipment (8). A system of dry slaughter, which excludes water from all processes, has been tested in the EU. In the absence of scalding tanks, the birds are defeathered by mechanical means and then waxed by immersion in a paraffin bath maintained at 60°C, peeled, and hand finished (40). Valnegri et al. (309) determined carcass

135 c­ontamination levels in this system and found low levels of contamination from total aerobic mesophilic bacteria, coliforms, and E. coli throughout the dryslaughter­ process but observed an increase in S. aureus in the postwaxing phase.

Evisceration

Evisceration is the removal of the intestinal tract by a series of interconnected machines. Potential problems at this step include muscle damage to the thigh or back of the carcass and the possibility of intestinal rupture. Leakage of the contents of the intestinal tract onto the carcass increases bacterial load on carcasses. Fecal contamination of carcasses, assumed to be a primary pathway for carcasses to be contaminated with pathogens, is often used as an indicator of process control during slaughter in the United States (191).

Chilling

Carcasses are chilled after evisceration to reduce microbial growth in order to maintain food safety and increase shelf life. The two main approaches include water and air chilling, and some combinations of the two may be also used. Rigor mortis processes are still taking place during the chilling phase, so chilling time and temperature have the ability to affect the quality of the finished product. A very rapid chilling rate will result in coldshortening, whereas a chilling rate that is too slow will result in high microbial counts (2, 16). There are two systems for water chilling; one is counterflow water chilling, whereby the flow of the chilling water runs opposite to the flow of spinning poultry carcasses, and the other is parallel water chilling, with water and birds flowing in the same direction. This provides an obvious means for cross-contamination of carcasses, which is dependent on bacterial contamination of carcasses before chilling, along with the amount of water overflowed and replaced per carcass and the ratio of carcasses to water in the chiller (232). Spoilage microbes may actually be reduced in the commercial immersion chilling system. Allen et al. (2) determined that water immersion chilling reduced coliforms and pseudomonads from the cell numbers prechill, although the opportunity for cross-contamination with pathogens does exist. Air chilling is the most common method used in Europe and has the advantages of lack of moisture pickup by the carcass and a dry final product that does not have exudates when packed in trays (252). In order to facilitate cooling and avoid overdrying of skin, the carcasses may be sprayed with water (252). Barbut et al. (17) compared carcasses processed in the same plant on

136 the same day by water and by air chilling. They reported that the numbers of aerobic bacteria were similar for the two groups, but the levels of psychrotrophic bacteria, Enterobacteriaceae, pseudomonads, LAB, and B. thermosphacta recovered from water-chilled carcasses were slightly less than the numbers recovered from air-chilled carcasses, indicating that these bacteria may be injured or inactivated by chilling in water but not by chilling in air (17). Huezo et al. (138) determined that air-chilled carcasses lost from 2.2 to 3.5% of prechill weight, compared to a moisture uptake range of 3.4 to 14.7% during immersion chilling. Breast skin of immersion-chilled carcasses was significantly lighter than the breast skin color for air-chilled carcasses, although storage lightened the skin color of air-chilled carcasses. Carroll and Alvarado (34) determined that air chilling improved color, marination yield, and tenderness and increased the shelf life of packaged breast fillets. However, Zhuang et al. (332) determined that if the same deboning time is used for both types of carcasses, the flavor and texture are not influenced by chilling method.

Carcass Decontamination General

Carcass decontamination in the United States and Canada frequently follows a multiple-hurdle approach (292). In the EU, decontamination with anything but potable water is discouraged, although current legislation does not ban chemical decontamination, but approval must be given by the European Commission (83, 140). Decontamination methods include physical, chemical, and biological processes.

Water-Based Interventions

Washing with water is a common practice in processing plants and is effective in removing visible contamination such as feathers and soil. The effectiveness of hot water for carcass decontamination has been studied by several researchers. Reductions of aerobic bacteria of approximately 1 log CFU were achieved by immersion in water at 75°C for 30 s but resulted in tearing of the skin when carcasses were manipulated (235). The same researchers determined that a treatment of 70°C for 40 s did not negatively affect the skin but still gave a 1-log CFU reduction of aerobic bacteria (235). Avens et al. (10) used boiling-water treatment of carcasses and determined that immersion for 2.5 min produced the maximum results and time periods of up to 4 min did not produce greater reductions in aerobic bacteria. Effects on the skin were not addressed because these researchers were studying the effects on ground chicken produced

Microbial Spoilage and Public Health Concerns from carcasses immersed in the boiling water, and the skin was removed prior to grinding of the meat (235). A comparison of spraying and immersion is made difficult by the different experimental conditions used in the various studies. However, Sinhamahapatra et al. (267) directly compared immersion and spraying and found only slight differences in aerobic counts, although reduction of coliforms and generic E. coli was approximately 0.5 log CFU greater for immersion than for spraying. Steam has been investigated as an alternative to the hot water decontamination process (194). Avens et al. (10) were able to reduce naturally occurring aerobic bacteria by more than 3 log CFU/cm2 using steam at 98°C for 3 min. A lower temperature (90°C) and shorter exposure time (0.2 or 0.4 min) resulted in a <1-log CFU/g reduction in naturally occurring aerobic bacteria, and significant damage to the skin was observed (323). Steam at atmospheric pressure applied to naturally contaminated chicken breast portions resulted in a 1.65-log CFU/cm2 reduction in total viable bacteria, although this treatment did not extend shelf life compared to untreated controls (151). Application of steam at atmospheric pressure for times up to 20 s reduced the cell numbers of inoculated C. jejuni and E. coli on whole chicken carcasses by as much as 3 log CFU, but the treatment caused shrinkage and color change of the skin (152). Electrolyzed water is being used in many countries as a sanitizer in processing areas. Electrolyzed water is produced by electrolysis and results in acidic electrolyzed water with a pH of 2 to 3 and an active chlorine content of 10 to 90 ppm and basic electrolyzed water with a pH of 10 to 13. Immersion chilling in acidic electrolyzed water was more effective than chlorine in reducing total aerobic counts, coliforms, and Salmonella serovar Typhimurium on poultry carcasses immediately after immersion and after 7 days of refrigerated storage (85). When applied by spray, none of the treatments, including acidic electrolyzed water, was significantly different from no treatment either immediately after spray or during refrigerated storage (85). The effect of acidic electrolyzed water was compared to that of chlorinated water or tap water by spraying on broiler carcasses. At time zero, significantly fewer psychrotrophs were recovered from carcasses sprayed with chlorinated or acidic electrolyzed water than from those sprayed with water (135). Psychrotrophic bacteria increased on all carcasses during refrigerated storage, but on day 14 significantly fewer psychrotrophic bacteria were recovered from carcasses treated with electrolyzed water than from those treated with either tap or chlorinated water (135). The USDA-FSIS has approved the use of ozone in recycling water in poultry chill tanks. Sheldon and Brown

6.  Meat, Poultry, and Seafood (264) evaluated the effect of ozone in chiller water on broiler carcasses and determined that aerobic counts were lower during refrigerated storage for ozone-treated carcasses than for controls. Another study comparing spraying ozonated water with immersion revealed that aerobic counts and coliforms were reduced only approximately 0.5 log CFU by spraying and approximately 0.8 log CFU by immersion (85).

Chlorine-Based Interventions

Chlorine has traditionally been used in the food industry as a sanitizer for food contact surfaces and as an antimicrobial treatment for foods. In the United States, poultry processors may use up to 50 mg of chlorine/ liter in the water used in bird washers and immersion chillers. Northcutt et al. (210) determined that adding chlorine or increasing water temperature did not cause further reductions of aerobic bacteria, coliforms, or Campylobacter over those obtained with a plain water spray. In another study, addition of chlorine to chiller water decreased recovery of E. coli, coliforms, total aerobic bacteria, and Campylobacter by less than 0.5 log CFU/100 ml of carcass rinsate but had no effect on the recovery of Salmonella (28). The efficiency of chlorine is limited in the presence of high organic loads and at pH levels above 7.0 (178). In addition, some researchers have reported that the low levels of chlorine permitted in chiller water will not inactivate Salmonella attached to skin (296). Chlorine dioxide has also been approved for use in poultry processing, is more effective than chlorine in reducing aerobic bacteria in water and on carcasses (19, 179), and is more effective against attached Salmonella (314). Hypochlorite (50 ppm) used as a spray on artificially contaminated poultry carcasses reduced aerobic bacteria, E. coli, C. jejuni, and Salmonella cell numbers, with the pathogens having the greatest decrease of approximately 2 log CFU (209). Sodium hypochlorite, however, used at 2% as a spray did not reduce coliforms or E. coli but reduced aerobic bacteria and Salmonella serovar Typhimurium by less than 1 log CFU (85). The same amount of sodium hypochlorite used in an immersion chiller did reduce aerobic bacteria, coliforms, and E. coli cell numbers but not Salmonella (85). Sodium hypochlorite (50 ppm) in a pilot-scale immersion chiller did not reduce Salmonella more than tap water only (249). A study of acidified sodium chlorite (ASC) for use as a prechill treatment in poultry processing revealed that 500 ppm ASC was effective in a dip solution for reducing total aerobic bacteria, E. coli, and total coliforms (166). Increasing the level of ASC to 1,200 ppm reduced coliforms by an additional 1 log CFU/ml of rinsate

137 but had no greater effect on aerobic bacteria or E. coli (166). A comparison of immersion and spray treatment with ASC revealed slightly larger reductions of aerobic bacteria and coliforms with immersion (267). ASC at a level of 1,200 ppm reduced mesophilic bacteria on chicken legs by approximately 2 log CFU/g of skin, psychrotrophs by nearly 1 log CFU/g, pseudomonads by 1.7 log CFU/g, L. monocytogenes by 1 log CFU/g, and Salmonella by 3 log CFU/g (60, 61). Cetylpyridinium chloride (CPC) (0.5%) used in an inside/outside bird washer on carcasses artificially contaminated with Salmonella serovar Typhimurium produced a 2-log CFU reduction in Salmonella, as well as a 2-log CFU reduction of total aerobic bacteria (329). An increase to 0.5% CPC increased reductions by 0.4 log CFU (327). Chicken skins artificially contaminated with C. jejuni and subsequently dipped in a 0.5% solution of CPC had a >4-log reduction of Campylobacter (243).

Organic acids

Addition of 0.5% acetic acid to scalder water did not reduce the levels of coliforms or aerobic bacteria; however, spray treatments after every step of processing reduced bacterial levels more than water sprays alone (250). A study involving spraying or immersion of carcasses in 2% acetic acid produced large reductions of mesophilic bacteria, Salmonella, E. coli, and coliforms when carcasses were sprayed, but immersion chilling produced low or no reductions in the same microbes (95). Immersing chicken wings in 2% lactic acid reduced aerobic bacteria by more than 2.5 log CFU compared to controls (150). Chicken breast treated with 2% lactic acid by dipping exhibited decreased levels of total aerobic bacteria, pseudomonads, and Enterobacteriaceae (6). A comparison of immersion and spraying produced inconsistent results. Reductions reported for aerobic bacteria for both methods have been minor (250, 267), but Sakhare et al. (250) reported greater reductions of coliforms by immersion, whereas Sinhamahapatra et al. (267) determined no treatment differences. Citric acid at 10% reduced inoculated C. jejuni on poultry carcasses slightly more when sprayed rather than by immersing carcasses, and slightly greater reductions were observed when citric acid was increased to 15% (73). Immersing naturally contaminated chicken legs in 2% citric acid for 15 min reduced aerobic bacteria, coliforms, and Enterobacteriaceae by more than 1 log CFU each (60). Dipping chicken legs into a solution of 3% citric acid resulted in a decrease in mesophilic bacteria throughout 8 days of storage, although there was no significant difference in psychrotrophic bacteria

138 except on day 1 (122). Counts of inoculated L. monocytogenes were lower on citric acid-treated legs than on controls by more than 1 log CFU on day zero and during refrigerated storage (122). Legs treated with 3% citric acid remained acceptable in color and odor throughout 8 days of refrigerated storage compared to controls, which were rejected after 6 days (122). Citric acid infused into boneless skinless chicken breast meat was effective in reducing counts of L. monocytogenes and Salmonella serovar Typhimurium (225). Treatment with 10% citric acid (wt/vol) reduced mesophilic bacteria, coliforms, and psychrotrophic bacteria on chicken legs more than with a water rinse at time zero and during 72 h of refrigerated storage, but sensory analysis revealed that the meat was unacceptable in taste and odor (64). Since organic acids often affect the quality of poultry products, an alternate approach is to combine acids with other antimicrobials. Peracetic acid and hydrogen peroxide have been combined in a ratio of 15% peracetic acid and 10% hydrogen peroxide (PAHP). Bauermeister et al. (20) evaluated the effectiveness of 85 ppm PAHP in poultry chill tanks. The PAHP reduced the prevalence of Salmonella-positive carcasses exiting the chiller by 92% compared to a 57% reduction with 30 ppm chlorine (20). They also determined that PAHP reduced Campylobacter on postchill carcasses by 43% compared with 13% with chlorine. Milillo and Ricke (192) studied the effects of the salts of organic acids alone and with heat on Salmonella in a model chicken meat medium. Treatment with acidified solutions of sodium acetate, sodium lactate, sodium pyruvate, or sodium butyrate did not significantly reduce Salmonella serovar Typhimurium compared to the control. However, when combined with heat (55°C), reductions compared to controls of 1.1, 2.3, 3.6, and 3.9 log CFU were observed for sodium salts of acetate, lactate, pyruvate, or butyrate, respectively.

Trisodium Phosphate

Trisodium phosphate (TSP) used at 12% on chicken legs inoculated with S. enterica serovar Enteritidis, L. monocytogenes, P. fluorescens, and B. thermosphacta reduced the counts of all these bacteria by approximately 1 log CFU compared to controls at day zero and at 5 days of refrigerated storage, with P. fluorescens being more susceptible to TSP treatment than L. monocytogenes (59). A 0.1% solution of TSP used as a dip for chicken breast skin inoculated with C. jejuni reduced counts by 2 log CFU immediately after treatment, and counts were below detection after 5 days of refrigerated storage compared to 5 log CFU in controls (226).

Microbial Spoilage and Public Health Concerns

Spoilage General

As spoilage bacteria multiply at refrigeration temperature, metabolic by-products accumulate on the meat, leading to off-odors and slime production (134). The bacteria initially metabolize the free available glucose, which is usually enough to produce a population of approximately 108 CFU/cm2 (116). After the glucose is consumed, the bacteria begin to metabolize amino acids, which leads to the development of off-odors. The wide range of microorganisms that can be present on poultry carcasses are listed in Table 6.1. The predominant spoilage microbes for poultry stored aerobically under refrigeration are members of the genus Pseudomonas (191). Yeasts occur less frequently, but Candida and Debaryomyces have been detected on both fresh and spoiled carcasses (313). The shelf life of poultry depends on a combination of factors, including the number and types of microorganisms initially present, storage temperature, and atmosphere and packaging conditions (148). The chilling process before packaging, whether by air or water, influences the type and number of contaminants on the carcass in the package.

Effects of Chilling Methods

The purpose of chilling, regardless of the method, is to maintain the breast temperature between −0.5 and +4°C (232). The method of chilling is believed to affect the growth of microorganisms, meat quality, and shelf life (88). Allen et al. (2) determined that water immersion chilling reduced microbial contamination of carcasses, although initially the numbers of pseudomonads tended to increase. They also determined that air chilling had little overall effect on microbial contamination of the skin, but air chilling with a water spray caused an increase in the numbers of pseudomonads on the carcass (2). Sanchez et al. (253) determined that psychrotrophic bacteria counts were significantly higher on immersionchilled broilers than on those that were air chilled. In contrast, other researchers have revealed that air chilling and immersion chilling produced comparable reductions of bacteria and reported that chilling reduced E. coli or coliform cell numbers by 90% (138). Mead (191) has suggested that the microbiological content of water- and air-chilled carcasses may differ because of differences in scalding treatments. Air-chilled carcasses are frequently scalded at lower temperatures in order to have an acceptable appearance, enabling more bacteria to survive postchilling (190). Barbut et al. (17) determined microbial counts of air-chilled and i­mmersionchilled carcasses produced at the same processing plant

6.  Meat, Poultry, and Seafood and revealed that the numbers of aerobic bacteria on the two types of carcasses were similar, but the numbers of psychrotrophic bacteria, Enterobacteriaceae, pseudomonads, LAB, and B. thermosphacta on water-chilled carcasses were all 0.4 log CFU less than those enumerated on air-chilled carcasses (17). They speculated that some of the bacteria were injured or inactivated by chilling in water, but not by chilling in air.

Effects of Packaging and Atmosphere

Packaging materials and methods and the composition of the atmosphere are other factors that can influence the spoilage of poultry meat. Possibly the most commonly used packaging system in retail markets is the conventional Styrofoam tray, with or without an absorbent pad liner, overwrapped with a PVC film. During refrigerated storage of poultry, Pseudomonas and psychrotrophic bacterial counts increase as storage time increases, with counts reaching 8 log CFU/g by 8 days of storage (39). Under these conditions, storage temperature is the main factor that determines shelf life because of its effect on growth of spoilage bacteria (65). Other packaging methods include vacuum packaging and MAP (15). Modified atmospheres generally suppress the growth of the usual spoilage microbes, but other species, including LAB and B. thermosphacta, begin to predominate (56). Vacuum packaging of whole and cut-up poultry extends shelf life, provided the chill temperature is maintained at approximately 1°C (158). When shelf life was extended to 20 to 25 days, offflavor­s developed before off-odors became noticeable in the vacuum-packaged meat (158). Shewanella putrifaciens, a sulfide-producing bacterium, is a prominent spoilage microbe in vacuum-packaged poultry (120). The use of increased CO2 within poultry packages has also been studied. Ogilvy and Ayres (219) determined that 25% CO2 was the maximum amount to be used before the meat became discolored. Gill et al. (120) determined that broilers packaged with increased CO2 in an oxygen-impermeable film had a shelf life of 7 weeks at 3°C and 14 weeks at −1.5°C.

Biogenic Amines as Quality Indicators

Biogenic amines are low-molecular-weight organic bases that may be produced by bacterial enzyme action on proteins or free amino acids in stored meats (315). Histamine is a biogenic amine that causes scombroid poisoning from fish products, and the maximum concentration allowed in fish is 100 mg/kg in the EU and 50 mg/kg in the United States (177). Recently, biogenic amines have been investigated as spoilage indicators in meats, including poultry (15). The bacteria most commonly present on stored poultry are biogenic amines

139 producers (109). Bunkova et al. (29) studied in broth cultures production of the biogenic amines histamine, tyramine, putrescine, cadaverine, agmatine, spermine, and spermidine by gram-negative bacteria isolated from poultry skin. They reported that Enterobacteriaceae produced tyramine, agmatine, putrescine, and cadaverine; Aeromonas spp. produced putrescine and cadaverine; but none of the pseudomonads tested produced any biogenic amines. Ntzimani et al. (214) determined that the biogenic amine concentrations correlated well with microbial counts and sensory changes in smoked turkey fillets during chilled storage.

Newer Methods for Shelf Life Determination

There are new methods for directly characterizing microorganisms in a habitat that do not require enrichment culture or isolation (132). These methods typically examine total microbial DNA or RNA in mixed microbial populations to identify individual organisms (141). Takahashi et al. (294) applied denaturing gradient gel electrophoresis (DGGE) to quickly identify spoilage bacteria in meat processing plants and meat products. Ercolini et al. (78), using DGGE, determined that the genera of spoilage bacteria differed during storage time as well as under different packaging conditions of beef products. This method was used to monitor microbial shifts on poultry carcasses that were held at ambient temperature (130). They determined that as aerobic plate counts (APC) remained static during the first 26 h, fewer bands were seen on the DGGE. A large shift in banding patterns was observed after approximately 44 h, indicative of spoilage microbes increasing both in numbers and in variety.

Food Safety Salmonella

Poultry is a major vector for illnesses caused by Salmonella (181). Proper cooking of poultry products produces a Salmonella-free product, although in-home microwave cooking is not recommended for raw poultry products because of the inconsistent heating that can occur (272). Salmonella contamination of poultry meat can occur throughout the entire poultry production and processing continuum. Poultry of any species has the potential to acquire Salmonella from the parent flock (vertical transmission), the hatchery (horizontal transmission), the growing environment, or contaminated feed (230). Feeding 106 Salmonella cells to a breeder hen resulted in 10% of the eggs laid being contaminated with Salmonella, although egg production did not decrease and Salmonella was not recovered from all fecal samples from the hens (299). Salmonella from nests and

140 the environment can penetrate the shell of the egg (48). Regardless of whether the Salmonella enters the egg via the hen or the environment of the laying house, there is a great potential for cross-contamination to other eggs and chicks at the hatchery. Even though only a few eggs of contaminated flocks may contain salmonellae, the problem is amplified in the hatchery, where 71% of shell fragments and 80% of conveyor belt samples have been found to be contaminated (46). This increase in contamination can occur because even eggs that contain high cell populations of Salmonella do not kill the chick and the eggs can still hatch, and the Salmonella is disseminated by the fanforced air (35). The grow-out environment also provides an opportunity for contamination of chicks that come from the hatchery free of Salmonella. One study of several poultry farms in different states revealed that 8% of fresh feed samples, 21% of samples from feed in troughs, 12% of soil samples, and 10% of litter samples were positive for Salmonella (246). The presence of Salmonella in the grow-out house, especially in litter, before placement of a new flock leads to higher frequencies of Salmonella at later stages of production (317). The potential for cross-contamination at the processing plant also exists. In one study of flocks slaughtered in succession at one processing plant, birds of 13% of the flocks were Salmonella positive before slaughter, but carcasses of 55% of the flocks were Salmonella positive after slaughter (240). Even when only Salmonellanegativ­e flocks were slaughtered, carcasses after slaughter were at times Salmonella positive, indicating crosscontaminatio­n either during live haul or from processing plant equipment (240).

Campylobacter

Campylobacter organisms are common enteropathogens of humans in developed countries, with C. jejuni being the most commonly isolated species. Handling and eating poultry is recognized as a major risk factor for human infection (131). Current knowledge indicates that most poultry flocks become infected on the farm (204). Acquisition of Campylobacter is age dependent, with newly hatched chicks being Campylobacter-free until at least 10 days of age (84). Campylobacter occurrence in flocks also appears to have a seasonal component, with the highest contamination rate being from June through August (74). Bird-catching teams and vehicles used for live transport are also a means of crosscontamination from farm to farm (3). A flock positive for Campylobacter entering the processing plant contaminates equipment, with nearly 100% of samples being positive at all stages of processing (23, 74). In one study,

Microbial Spoilage and Public Health Concerns only 12% of chickens on a farm prior to transport were positive for Campylobacter, but after transport 56% of the chicken exteriors were positive for Campylobacter, indicating that transport and holding prior to processing contributes to the high levels of Campylobacter present postprocessing (290).

SEAFOOD

General

Bacteria, viruses, parasites, and naturally occurring toxins in fresh and processed fish and shellfish present the potential for foodborne illness in both healthy and immunocompromised individuals. Preharvest contamination of fish and shellfish is due either to naturally occurring microbes or to those introduced into the environment through animal and human pollution or agricultural runoff. The threat of illness arising from handling or consumption of fish is dependent upon the species, harvest location, prevailing environmental conditions, postharvest handling and processing, marketing, and ultimate storage conditions. The major illnesses resulting from cooked fish consumption are ciguatera and scombroid toxicity. Cold-smoked fish also present a potential health hazard because there is no postprocessing treatment to eliminate pathogens that may be present or to stop their proliferation. The number of illnesses arising from cold-smoked fish consumption is relatively minor, because it is not a major product of commerce. The risk of acquiring illness from shellfish consumption may increase due to environmental conditions in the growing waters, harvesting methods, processing operations, and handling during marketing. Some shellfish present a much greater risk than others. Most shrimp, crabs, crawfish, lobsters, and other crustaceans, for example, are consumed fully cooked, and hence most bacteria, viruses, and parasites are either inactivated or reduced to a level at which infection may not occur. For many individuals, the greatest risk from shrimp consumption is an allergic reaction. However, many molluscan shellfish are customarily consumed raw or with minimal heating. The consumption of raw molluscan shellfish (primarily oysters, clams, and mussels) has been a major concern to national and international health agencies, because harmful bacteria in these products have caused foodborne illness outbreaks and mortality. Some of the microbes of concern include bacterial pathogens Vibrio spp., Salmonella, E. coli, L. monocytogenes, C. botulinum, Campylobacter, and Aeromonas hydrophila; hepatitis A virus; and the parasite Cryptosporidium.

6.  Meat, Poultry, and Seafood It is not surprising that two emerging pathogens, Y. enterocolitica and E. coli O157:H7, have not been identified in either growing waters or products. While surveys for Y. enterocolitica have revealed that this bacterium is widely occurring (having been found in beef, lamb, pork, oysters, shrimp, crabs, and water), its prevalence in fish and shellfish is rare overall. Additionally, most of the isolates obtained are avirulent, and seldom does a virulent strain occur in these foods. A principal source or reservoir of virulent Y. enterocolitica is pigs. Similarly, few foodborne outbreaks have been linked to E. coli O157:H7, but in rare cases, salmon has been identified as a vehicle of transmission.

Fish General

For many years, wild fisheries were the major source of the fish supply. However, overfishing, environmental change, habitat destruction, limited entry, and harvest caps have placed strictures on the quantity and quality of fish available for market. Within recent years, aquaculture has become a major source of fish, although available species are limited. The aquaculture industry in the United States has an excellent record regarding the production of safe, high-quality products. The microbial composition of aquacultured products is important to many individuals and agencies. Of particular concern are those microorganisms considered to be human pathogens. Some of the reasons for this interest are that (i) producers and processors need information to establish adequate HACCP plans, to comply with government regulations and buyer specifications; (ii) the information is useful to aquaculture producers seeking to identify how effluent from their operations may impact compliance with environmental regulations; and (iii) the type of microflora present in an aquaculture facility could have an impact on worker safety. Worker safety concerns are especially relevant to indoor recirculating aquaculture systems. In the 1960s, it was suggested that the bacterial flora of fish was a reflection of its environment. At that time, research also revealed that the gill flora was similar to that of the skin, which reflected that of the environment. Since then, numerous studies have supported the relationship between the growing water environment and the aquatic microbes raised in that environment. Aquaculture or aquafarming in ponds enriched with human and animal excreta has been a long-standing tradition in some countries. Recent studies in parts of Africa have revealed the high potential usefulness of human and animal wastes in the optimal utilization of limited water resources and

141 associated food production. However, the use of these wastes presents distinct health problems through the passive transmission of animal pathogens to the aquacultured animal residing in contaminated or polluted water. While fish may carry pathogens passively in their intestines, gills, and body surfaces and may subsequently infect people who handle, prepare, or eat the fish raw or partially cooked, there is minimum risk to those who consume the fully cooked fish, although cross-contaminatio­n during food preparation is a concern. Many microbial surveys on the presence and significance of human pathogens in various aquaculture systems, such as flowthrough, pond, and recirculating systems, have been conducted (90). Information on the bacterial composition of product reared in net pens is scarce. The reasons for this lack of information lie in net pen culture’s recent emergence as a major form of aquaculture and the relative difficulty of sampling in that environment. The types of pathogenic microorganisms found in the final product and in the growing waters of several species of warm- and coldwater fish are listed in Table 6.2. The microbial flora most frequently identified in products were Salmonella and C. botulinum. The presence of Salmonella is understandable because birds, snakes, turtles, and other wildlife all harbor the microorganism and because nutrients in the water, feed, and feces provide an excellent food source for maintenance and proliferation of the bacterium. Salmonella was present in eel aquaculture systems in both outdoor and indoor ponds. Clostridium botulinum is a ubiquitous organism found in soil. Ponds having earthen structures or significant sediment provide an ideal reservoir for many clostridial species. It is also not surprising that Aeromonas spp. were consistently isolated from the fish-growing water. Aeromonas bacteria are common aquatic microbes and can cause disease in fish and other aquatic animals. Aeromonas hydrophila and Plesiomonas shigelloides are Vibrio-like bacteria recently recognized as causes of foodborne gastroenteritis. Aeromonas hydrophila has been identified as an opportunistic pathogen in immunocompromised hosts, but more recently it has been determined to cause enteritis in healthy hosts as well. Plesiomonas shigelloides has been identified as a potential agent of diarrhea in humans, but it has not received the attention given to the vibrios and other enteric bacterial pathogens. The major pathogens obtained from fish raised in recirculating aquaculture systems include Aeromonas, S. aureus, L. monocytogenes, Vibrio cholerae, Vibrio parahaemolyticus, and Vibrio vulnificus. Most of these are waterborne bacteria, so their presence in a

Microbial Spoilage and Public Health Concerns

142

Table 6.2  Finding of various studies assessing human pathogens associated with pond-raised finfisha Associated human pathogen Fish

A

B

C

D

E

ND

ND

ND

¨b

¨b

F

G

H

I

J

K

Finished product   Channel catfish   Striped bass

ND

ND

ND

¨

ND

  Rainbow trout

ND

ND

ND

¨

ND

  Tilapia

ND

ND

ND

¨

ND

ND

¨

ND

¨

  Brown trout ¨

  Pike

¨

  Carp ¨

  Channel catfish

¨ ¨

In growing water   Rainbow trout

¨

  Brown trout

¨ ¨

  Pike   Brown trout

ND

  Pike

ND

  Salmon

¨

  Tilapia

¨ ¨

  Carp   Channel catfish

¨

¨

¨

  Pathogens: A, Campylobacter jejuni/coli; B, Escherichia coli (O157:H7); C, Klebsiella pneumoniae; D, Plesiomonas shigelloides; E, Vibrio cholerae; F, Listeria monocytogenes; G, Salmonella sp.; H, Clostridium botulinum; I, Yersinia enterocolitica; J, Aeromonas sp.; K, Staphylococcus aureus. ND, not detected.   b Detected in summer only. a

r­ecirculating aquaculture system is not unexpected. The high stocking densities used in recirculation systems commonly subject fish to increased physiological stress. This crowding and stress could enhance the spread of bacterial pathogens among the fish, resulting in overall bacterial loads that may be higher than in fish from less intensive aquaculture systems or from wild stocks. Also, because recirculating systems are designed to conserve water, elimination of the bacteria may be difficult or i­mpossible once colonization of a system occurs. Studies have revealed that human and fish pathogens will form biofilms on biofilters, inside pipes, and in growing tanks. The dynamics between the biofilm, the animals in culture, and the growing water have not been characterized. Other pathogens are often absent because many recirculating systems are located indoors and are thus protected from external sources of contamination such as soil, animals, and the environment.

Biogenic Amines

Cadaverine, putrescine, and histamine are diamines that may be produced postmortem from the decarboxylation of specific free amino acids in fish or shellfish tissue. The decarboxylation process can proceed through two biochemical pathways: endogenous decarboxylase enzymes naturally occurring in fish or shellfish tissue or exogenous enzymes released by the various microorganisms associated with the seafood product. Endogenous production of diamines is insignificant compared to the exogenous pathway. The nature of the microflora and the composition of the product affect the amount of decarboxylase a bacterial cell may release. In general, histamine, putrescine, cadaverine, tyramine, tryptamine, b-phenylethylamine, spermine, and spermidine are considered the most important biogenic amines in foods. However, bphenylethylamine, spermine, and spermidine are not end products of bacterial decomposition in fishery products.

6.  Meat, Poultry, and Seafood Fish muscle is naturally rich in free amino acids, which can increase even more postmortem. The high content of proteolytic enzymes in the intestinal tract is responsible for rapid autolysis and the high free amino acid content in fishery products. Amino acid formation depends on the harvesting season and feeding activity before capture. For example, fish harvested in summer or during the feeding season quickly liberate large quantities of lysine and arginine. The activity of amino acid decarboxylase depends on a range of factors, including the presence of fermentable sugars, pH, and redox potential. The variable influence of environmental temperature, the nature of the microflora, decarboxylase activity, and intestinal tract contents on the rate of biogenic amine formation may be major reasons for the discrepancies that have been reported in the literature concerning levels of biogenic amines in fresh and processed fish. Another reason for discrepancies may be poor experimental design. Regardless of the discrepancies, it is clear that a high amino acid content and bacterial activity could rapidly result in an elevated concentration of biogenic amines if the proper controls are not in place. Biogenic amines, particularly histamine, have been implicated as the causative agent in a number of scombroid food poisonings. There is a wide variation in individual susceptibility to biogenic amines. Clinical symptoms are more severe in people taking medications that inhibit enzymes that normally detoxify histamine in the intestine. Histamine exerts its effects by binding to receptors on cellular membranes in the respiratory, cardiovascular, gastrointestinal, and hematological/immunological systems and the skin. The symptoms of histamine poisoning generally resemble the symptoms encountered with immunoglobulin E-mediated food allergies and usually appear shortly after the food is ingested, with a duration of up to 24 h. Symptoms may be gastrointestinal (nausea, vomiting, and diarrhea), circulatory (hypotension), cutaneous (rash, urticaria, edema, and localized inflammation), or neurological (headache, palpitations, tingling, flushing or burning, and itching). Antihistamines can be used effectively to treat the symptoms. Taking into account the uncertainties reported into the literature, there is a consensus that histamine levels above 500 to 1,000 ppm are considered potentially dangerous to human health based on the concentrations found in food products involved in incidents of histamine poisoning. Even less is known about the toxic doses of other amines. Threshold values of 100 to 800 ppm for tyramine and 30 ppm for phenylethylamine have been reported. In estimating the toxic levels of biogenic amines, one should consider the amount of food

143 consumed in the presence of other amines in the food or other dietary components and the use of alcohol and medicine. An additional concern, especially if nitrite is used in cold-smoked products, is that secondary amines such as putrescine and cadaverine can react with nitrite to form carcinogens. Fish often associated with histamine poisoning are the scombroid fish belonging to the families Scomberesocidae and Scombridae. Fish included in these families are the tunas, bonito, mackerels, bluefish, and saury. Tuna and mackerel are the most common fish associated with the poisoning, but other fish are also associated with outbreaks of scombroid poisoning. Examples include mahimahi, sardines, anchovies, herrings, and marlin. The relationship between the type of fish and the number of cases of biogenic amine poisoning associated with it may reflect consumption patterns for that specific fish. Recently, the U.S. Food and Drug Administration (FDA) (21CFR123) established a guidance level for histamine of 50 ppm to ensure the safe consumption of scombroid or scombroid-like fish and recommended the use of other data to judge fish freshness, such as the presence of other biogenic amines associated with fish decomposition. A maximum average histamine content of 100 ppm has been established in the EU for acceptance of tuna and other fish belonging to the Scombridae and Scomberesocidae families. The EU has suggested that in the future a maximum of 300 ppm for total biogenic amines in fish and fish products may be an appropriate legal limit. It is important to note, however, that there may be a type of food poisoning that does not arise from high levels of histamine. Hence, a low histamine level may not be an absolute assurance of a safe product. It may be more appropriate to say that the absence of decomposition in the fish renders it a safe product. As such, a safe product would have no evidence of spoilage, including odors of decomposition, high histamine levels, and other amines such as cadaverine. It was reported that very large amounts of histamine could be given orally without causing adverse effects. This effect was due to the conversion of histamine to inactive Nacetylhistamine by the intestinal microflora. Human subjects given up to 67.5 mg histamine orally did not produce any subjective or objective symptoms of histamine poisoning. However, in another study, subjects receiving 36 mg or more of histamine subsequently developed symptoms associated with histamine toxicity. Symptoms also appeared upon consumption of tuna sandwiches containing 100, 150, and 180 mg doses of histamine. Generally, high histamine levels cause a toxic response, but subsequent research has revealed that other factors may also be responsible. It has been concluded that there is no relationship between the concentrations of six amines (including

144 histamine, cadaverine, and putrescine) and the onset of scombrotoxic symptoms. The deleterious effects in relation to the amount of histamine ingested at one meal have been reported as follows: mild poisoning at 8 to 40 mg; disorders of moderate intensity at 70 to 1,000 mg; and severe responses at 1,500 to 4,000 mg. Histamine appears not to be the sole factor in causing toxicity, because cases have also resulted from the ingestion of low concentrations of histamine. Strong evidence indicates that biogenic amines such as putrescine, c­adaverine, spermine, and spermidine in fish tissue can potentiate the toxic effect of histamine by inhibiting intestinal histamine-metabolizing enzymes such as diamine oxidase, potentiating histamine uptake, and liberating endogenous histamine in intestinal fluids. It has been reported that fish implicated in a scombroid poisoning incident also had high levels of inhibitors that interfere with histamine metabolism. Monoamine oxidase inhibitor drugs used for the treatment of depression, hypertension, and tuberculosis have also been ­observed to potentiate the toxic effect of histamine. Studies have revealed that the levels of cadaverine in toxic or decomposed fish are generally several times greater than the levels of putrescine. When cadaverine was administered through stomach catheters simultaneously with histamine, peroral toxicity was observed in guinea pigs. The high cadaverine contents of mackerel in comparison with herring, both stored at 10°C, could be responsible for mackerel often being implicated in scombroid poisoning and not herring; histamine levels in the two fish were similar. Cadaverine and putrescine, as well as other diamines, have been suggested to facilitate the transport of histamine through the intestinal wall and increase its toxicity. There is the possibility that bacterial endotoxins, which are widespread, could induce hypersensitivity to histamine. These compounds are complex, heatstabl­e, lipopolysaccharide materials produced by gram-negativ­e bacteria. Endotoxin is capable of inducing histamine release in animals (sometimes called endotoxin shock) similar to that seen in anaphylaxis. However, extremely low levels of endotoxin have been detected in both good tuna and tuna known to have caused illness in humans. From these discussions, it is clear that concentrations of biogenic amines producing observable toxicity may differ significantly, depending on a variety of circumstances. Also, although a variety of histamine potentiators are known, there is not a clear understanding of the level and the manner by which this synergism occurs. Biogenic amines from fish and shellfish are among the foremost causes of seafood-related morbidity worldwide. Seafood

Microbial Spoilage and Public Health Concerns harvesters and processors should be keenly aware of the conditions that exacerbate the formation of these compounds in their products, the relative tendency of different areas of a fish carcass to support rapid formation of these amines, and the effects of different bacterial floras on the formation of these toxins. The prevalence of biogenic amines in fish depends on several factors. In general, concentrations in newly caught fish are low. For example, cadaverine values ranged from 1.16 to 10.36 ppm in high-quality rockfish, salmon steaks, and shrimp, and putrescine levels ranged from 1.36 to 6.30 ppm in high-quality lobster tails, salmon steaks, and shrimp. Another study revealed that high-quality tuna had cadaverine and putrescine values of 0.24 to 5.32 and 0 to 1.84 ppm, respectively. One particularly broad study examined the amounts of histidine and histamine formed in 21 aquatic species during spoilage. The conclusions were consistent with those from other studies, i.e., that more histamine is produced in the red-muscle fishes, such as tuna and mackerel, than in white-muscle species, such as rockfish. Biogenic amine prevalence also varies from year to year. For example, the results of a 3-year study focused on the biogenic amine content in 102 samples of albacore tuna (Thunnus alalunga) harvested off the U.S. northwest coast from 1994 to 1996 revealed that, depending on the year, there were significant differences in amine levels in these fish. Total levels of the six amines detected (spermine, spermidine, putrescine, cadaverine, histamine, and tyramine) varied from 5.9 to 56.5 ppm. These levels were probably lower because the samples were frozen or chilled aboard the boat and then immediately frozen after reaching the dock and kept at −40°C until analysis. Spermine was present at higher levels, followed by spermidine, histamine, putrescine, cadaverine, and tyramine. Researchers have observed no difference in amine levels in upper versus lower loin light muscles; however, dark muscles contained higher concentrations of spermidine (Table 6.3). Samples from the intestinal wall contained high amine levels. A variety of microorganisms can produce biogenic amines. The production of cadaverine and putrescine by microorganisms is not surprising, since studies have revealed that the covalent linking of cadaverine and putrescine to peptidoglycan is necessary for normal microbial growth. As such, production of these amines supports the continued growth of microbial colonies on the surface of the fish. Several inoculation studies on both culture media and fish have revealed that the bacteria Morganella spp., Proteus morganii, Proteus spp., H. alvei, and Klebsiella spp. can produce histamines and other biogenic amines. Similarly, when the relationship between microfloras

6.  Meat, Poultry, and Seafood

145

Table 6.3  Levels of biogenic amines in light and dark muscles and the intestine wall of albacore tunaa Mean level of biogenic amines (ppm) (SD) Sample

Spermine

Spermidine

Histamine

Putrescine

Cadaverine

Serotonin

Total

Light muscle, upper loin

0.68 b (0.12)

0.26 c (0.07)

0.00 b

0.22 ab (0.07)

0.13 b (0.02)

0.00 b

1.29 c (0.17)

Light muscle, lower loin

1.21 b (0.26)

0.25 c (0.05)

0.00 b

0.14 b (0.05)

0.11 b (0.06)

0.00 b

1.77 c (0.37)

Dark muscle

2.50 ab (0.97)

0.79 b (0.18)

0.00 b

0.06 b (0.03)

0.07 b (0.05)

0.00 b

3.42 b (0.72)

Intestine wall

5.35 a (2.46)

3.63 a (1.18)

0.52 a (0.25)

0.43 a (0.16)

1.96 a (0.59)

4.38 a (1.33)

16.3 a (4.59)

  a Values were calculated by using 0 for not detected levels (spermine, spermidine, histamine, putrescine, and cadaverine were not detectable at levels of <0.08 ppm; serotonin was not detectable at <0.18 mg/100 g). Mean values followed by the same letter in the same column do not differ significantly (P £ 0.05 by Tukey’s honestly significant difference test). Data from reference 121.

on horse mackerel (Trachurus japonicus) and dominant spoilage bacteria was investigated, results revealed that Pseudomonas I/II, Pseudomonas III/IV-NH, Vibrio, and Photobacterium were dominant when high levels of putrescine, cadaverine, and histamine were detected. Most studies also revealed that the potential for these microorganisms to produce toxic levels of biogenic amines is increased at abusive temperatures. The activity of decarboxylase enzymes produced by bacteria can be an indirect measurement of the potential for biogenic amine formation. A study revealed that 14 bacterial isolates (Acinetobacter lwoffii, A. hydrophila, C. perfringens, Enterobacter aerogenes, Enterobacter spp., H. alvei, Morganella morganii, Proteus mirabilis, Proteus vulgaris, Proteus spp., P. fluorescens/putida, P. putrefaciens, Pseudomonas spp., and Vibrio alginolyticus) from mackerel tissue were capable of producing ­decarboxylase activity (production of histamine, cadaverine, and putrescine) when incubated in Spanish mackerel at 0°C, 15°C, and 30°C. Other bacteria also have strong histidine decarboxylase activities, including Klebsiella pneumoniae, Klebsiella planticola, Alteromonas putrefaciens, Photobacterium phosphoreum, Staphylococcus xylosus, Cedecea lapagei, Cedecea neteri, P. shigelloides, Providencia spp., Lactobacillus curvatus LTH 975, Lactobacillus buchneri LTH 1388, Serratia spp., and Escherichia spp. Various studies have revealed that putrescine and cadaverine levels solely could be used as an index of freshness. However, the role and significance of putrescine and cadaverine in food safety and biogenic amine poisoning is yet to be established. Histamine is a more likely causative agent in cases of scombrotoxin poisoning than putrescine and cadaverine; however, levels of putrescine and cadaverine can serve as a useful indicator of whether fish products may present a hazard to consumers.

Results of research on production of biogenic amines by different microorganisms on different culture media and on fish likely to be cold-smoked, are summarized in Tables 6.4 and 6.5, respectively. Results of studies of histamine production by bacterial isolates from fish and grown on culture media at different temperatures and incubation times are summarized in Table 6.6. Freezing fish before it is subjected to temperature abuse can reduce the formation of certain biogenic amines, thereby minimizing the risk of scombrotoxin poisoning. Mahimahi that was frozen for 24 weeks and then incubated at 32°C exhibited greatly reduced levels of histamine. When the fish were frozen for 40 weeks before incubation at 32°C, almost no histamine formation occurred during the incubation period. These results suggest that microorganisms responsible for ­ histamine formation were inactivated during the freezing process. This hypothesis was later confirmed when the specific activity of histidine-decarboxylase in the halophilic histamine-forming bacteria Photobacterium phosphoreum and Photobacterium histaminum remained at 27 to 53% of the initial value after 7 days of storage at −20°C. During this time, the viable cells decreased by more than 6 log10 CFU. High-quality mackerel stored for 1.5 years at −14, −21, and −29°C and then temperature abused had no measurable histamine formation. However, if fish are temperature abused prior to freezing, histamine and other biogenic amines may still be present in toxic amounts. Therefore, it is important to know the temperature history of frozen fish because outbreaks of scombrotoxin poisoning can be caused through the ingestion of freeze-thawed fish and its products if they were previously temperature abused. When salt levels of 8% (wt/wt) were applied to sardines, the lag phase for aerobic bacteria growth was increased. The generation time and lag phase of histamine

Microbial Spoilage and Public Health Concerns

146

Table 6.4  Production of biogenic amines by bacteria growing on culture media Histamine-producing bacteria

Histamine concn

Morganella spp.

Temp (time)

1,000 ppm 1,000 ppm

25°C (24 h) 25°C (19 h) followed by 5°C (100 h) 5°C (100 h)

0 ppm Proteus spp.

Large

Proteus morganii

>200 nM/ml Large

Enterobacter aerogenes

>200 nM/ml

Klebsiella pneumoniae

Large

15, 30, 37°C (<24 h)

Hafnia alvei

Large

30, 37°C (>48 h)

Citrobacter freundii

Large

30, 37°C (>48 h)

Escherichia coli

Large

30, 37°C (>48 h)

Lactobacillus (3 strains)

2.2 mg/ml

producers increased at room temperature and with ice storage, respectively. Salt appears to reduce the activity of histamine producers at both room and refrigerated storage temperatures. A recent study revealed such bacteria were inhibited in cold-smoked salmon stored for 5 weeks at 5°C. In this study, salt levels were at 5% (wt/wt) and smoke was linearly proportional to the salt

15, 30, 37°C (<24 h)

and smoke content (the greater the concentration, the greater the inhibition). No synergistic inhibition between the two factors was observed. Although the temperatures used for a hot-­smoking ­process may inhibit histamine-producing bacteria, cold­smoking does not expose the fish to temperatures ­sufficiently high to inhibit their activity. A study on cold-

Table 6.5  Production of biogenic amines by bacteria incubated on fish Concn (ppm) of: Bacterium(a)

Fish

Histamine

Proteus morganii

Tuna

>50 <50

Acinetobacter Aeromonas hydrophila Clostridium perfringens Enterobacter aerogenes Enterobacter spp. Hafnei alvei Morganella morganii Proteus spp.   P. vulgaris   P. mirabilis Pseudomonas spp. Vibrio alginolyticus

Spanish mackerel Spanish mackerel Spanish mackerel Spanish mackerel Mackerel Spanish mackerel Spanish mackerel Spanish mackerel Spanish mackerel

Detectable

Other biogenic amines

Temp (°C)

>1

24, 30 15 0

>1

0

>1

0

>1

0

Detectable >1

0

>1

0

>1

0

>1

0

>1

0

6.  Meat, Poultry, and Seafood

147

Table 6.6  Production of histamine on culture media by microorganisms isolated from fish Microorganism

Fish

Histamine concn (ppm)

Temp (time)

Proteus morganii

Skipjack, Jack mackerel, sardine

>1,000

35°C (24 h)

Hafnia alvei

Skipjack, Jack mackerel

Proteus spp.

Skipjack, Jack mackerel, sardine

>1,000

35°C (24 h)

Klebsiella

Skipjack, Jack mackerel

Morganella morganii

Tuna

>1,000 >1,000 (?) >1,000 (?)

37°C (18 h) 7, 19, 30°C (24 h) 15, 25°C

Klebsiella spp.

Tuna

>1,000

37°C (18 h)

Enterobacter aerogenes and E. cloacae

Tuna

500–1,000

37°C (18 h)

Citrobacter freundii

Tuna

<250

37°C (18 h)

Proteus mirabilis

Tuna

<250

37°C (18 h)

Proteus vulgaris

Tuna

37°C (18 h) 7, 19, 30°C (24 h) 35°C (24 h)

Sardine

<250 >1,000 100–2,000

E. agglomerans

Tuna

<250

37°C (18 h)

Serratia liquefaciens

Tuna

<250

37°C (18 h)

Providencia stuartii

Sardine

150–1,000

35°C (24 h)

Vibrio spp.

Sardine

100

35°C (24 h)

Tuna

25.8 >1,000 of other biogenic amines

4°C (6 days) 37°C (24 h)

Stenotrophomonas maltophilia

smoked mackerel (Scomber scombrus) revealed that histamine did increase during the smoking process but the rate of histamine formation could be controlled by limiting the temperature and time of the smoking process. The production of biogenic amines during chilled storage (5°C) of cold-smoked salmon (Salmo salar) from three smokehouses was studied over a 2-year period. Results revealed the production of biogenic amines was unlikely to result in histamine poisoning in humans as indicated by epidemiologic data. Some samples exceed the defect action level of 50 ppm established by the FDA for Scombridae and 100 to 200 ppm by EU regulations for Scombridae and Clupeidae, but no samples reached toxic levels of 500 ppm (81a, 307a). However, the production of biogenic amines is highly variable and difficult to predict. A study of cold-smoked fish may explain why scombrotoxin formation at toxic levels does not occur in cold-smoked fish. During the first 2 weeks after cold smoking, gram-negative bacteria (those primarily responsible for biogenic amine formation) were dominant. Gram-negative bacteria then progressively decreased, while gram-positive bacteria, dominated by several species of LAB, increased.

Three bacterial suspensions (with final cell populations of 106 for Klebsiella oxytoca, M. morganii, and H. alvei) were inoculated onto yellowfin tuna (Thunnus albacares). Vacuum- and nonvacuum-packaged samples were stored at various refrigerated temperatures for up to 15 days and examined for growth and histamine formation. Samples stored at 2°C contained 120 ppm histamine, whereas samples stored at 10°C contained 2,000 ppm histamine. Vacuum packaging did not provide any beneficial effect in controlling histamine production and bacterial growth; hence, low-temperature storage was more effective than vacuum packaging in minimizing histamine formation. The growth of common aerobic spoilage bacteria be­ longing to genera such as Pseudomonas, Flavobacterium, Micrococcus, and Moraxella is inhibited by carbon dioxide in MAP fish during refrigerated storage. Inhibition of these common psychrotrophic spoilage bacteria increases the shelf life, permitting growth of a different type of spoilage flora (i.e., the slower-growing gram-positive bacteria, including Lactobacillus spp.). Inhibition of the gram-negative bacteria by MAP may result in an initial reduction in the rate of histamine formation, thereby providing an increased margin of product safety.

148 Research using additives to control scombrotoxin formation yielded minimally encouraging results. Potassium sorbate at 0.5% inhibited the growth of biogenic amine-forming bacteria and histamine production. High salt concentrations (3.5 to 5.5% NaCl) inhibits histamine production by histamine-forming bacteria. Fish treated with a 1% potassium sorbate solution contained less histamine than a control during 2 days of storage at 4°C. MAP combined with 1% potassium sorbate retarded the growth of M. morganii during 3 days of storage at 4°C. However, after longer storage times, few treatments are effective in controlling formation of biogenic amines. It is also apparent that the shelf life extension provided by MAP of fish is only effective if sanitary conditions combined with proper temperature control are maintained from the time of harvest. The histamine content of irradiated mackerel samples increased gradually during storage (at 4°C) of samples inoculated with M. morganii. After 8 days, however, histamine levels reached 2,023 ppm and 2,064 ppm for samples irradiated with doses of 0.5 and 2.0 kGy, respectively, from an initial concentration of 412 ppm. Morganella morganii grew by approximately 2.0 and 0.7 log CFU in samples irradiated with 0.5 kGy and 2.0 kGy, respectively, during the 8 storage days. While many processing operations may mitigate the production of histamine and other biogenic amines, fish must receive appropriate handling from harvest through consumption if scombrotoxin poisoning is to be avoided. Some processing methods initially retard the growth of histamine bacteria and the production of biogenic amines, but only for a brief time interval. Failure to develop and implement a food safety plan based on HACCP principles could result in the marketing of toxic fish and fish products.

Shrimp General

Shrimp are the most important exportable aquatic product in the global seafood market. Worldwide, about 75% of shrimp production, whether cultured or wild caught, originates from developing countries. In contrast, 70 to 75% of global shrimp consumption occurs in developed countries. Shrimp is marketed cooked or raw, peeled or unpeeled, and with or without breading or other coatings. Examination of fresh and frozen shrimp reveals that spoilage of this product is largely due to biochemical changes induced by microbial populations and to a lesser degree by endogenous enzymes and chemical compounds in the shrimp.

Microbial Spoilage and Public Health Concerns

Microbial Hazards

Most studies on human pathogens in shrimp have been conducted with the product instead of in the growing water of the shrimp. The presence of V. cholerae, V. parahae­molyticus, and V. vulnificus is not unexpected, because shrimp are grown in marine environments (89). The similarity of compositions between wild and cultured shrimp, despite being obtained from different habitats, suggests an influence of the host on the establishment of the gut flora. Vibrios attach to and colonize the exoskeletons of crustaceans, which provide vibrios a means of survival and growth in the aquatic environment. The numbers of vibrios that often occur on a single shrimp carapace are considered to be a public health concern. Coliforms can be found in harvested shrimp as well as in the water and sediment in which they are reared. Coliform counts in water samples may be affected by a variety of factors, including timing of sampling with respect to water exchanges on the farm, ambient temperature at time of sampling, and stocking density. Levels tend to be lower if sampling coincides with water exchange, cold weather or colder water temperatures, or low stocking densities. In a pond, it is presumed that feed and manure are the major sources of E. coli. Salmonellae are associated with pond water, sediment, and shrimp throughout the culture cycle, including the prestocking period, farming phase, and harvest. The survival rate of the microorganism is enhanced by nutrients, manure, and feed present in the pond system and the favorable interaction of various biological and physical factors. Untreated chicken manure used to fertilize the ponds and droppings from aquatic birds represent significant potential sources of Salmonella. Bacteria of the genus Vibrio are ubiquitous in the marine and estuarine ecosystems in which shrimp are typically farmed. Among more than 20 Vibrio species known to be associated with human disease, V. cholerae, V. parahaemolyticus, and V. vulnificus are of greatest concern. Depending on the species involved, the clinical manifestations vary, ranging from gastroenteritis to septicemia and wound infection. Vibrio cholerae is ubiquitous in marine and brackishwater ecosystems and has been isolated from pond mud, water, and shrimp samples in Thailand and India. The major sources of this Vibrio species in shrimp samples may be sediment, water, and feed. However, natural shrimp food such as algae, plankton, and invertebrate animals such as copepods and zooplankton could also be contributors. Dalsgaard et al. (54) studied the presence of V. cholerae in a total of 107 samples that included water, sediment, shrimp, feed, shrimp gut, and

6.  Meat, Poultry, and Seafood

149

chicken manure. Thirty-three percent of the samples contained V. cholerae non-O1 serotype, which is less severe than V. cholerae O1, the bacterium that causes epidemic cholera. The occurrence of this bacterium was not significantly influenced by water, salinity, temperature, dissolved oxygen, or pH. The results indicate that V. cholerae non-O1 is ubiquitous in aquatic environments where shrimp culture is practiced under a variety of conditions. Vibrio spp. are present in shrimp during all phases of harvest or farming. These microorganisms are most commonly found in shrimp and other crustaceans. Crustaceans have an exoskeleton made of chitin, and some bacteria, including Vibrio spp., have the ability to use chitin as a nutrient source. Various chitin sources can stimulate the growth of V. cholerae O1 and the production of cholera toxin. The Vibrio strains are very low toxin producers in filtered water alone, hence indicating the need for nutrient sources to fuel this activity. The microorganism V. parahaemolyticus has been found in shrimp from the Philippines and in shrimp pond sediment and water from Thailand. Sediment, water, and feed have all been identified as sources of the bacterium. However, there is a natural association between pathogenic V. parahaemolyticus and shrimp, especially with shrimp shell chitin as a nutrient source for this ­pathogen. Vibrio vulnificus has also been detected in shrimp. Vibrio spp. have been detected in shrimp and ponds even when counts of fecal microorganisms were low. This implies that there may not be a correlation between the levels of fecal coliforms and the presence of Vibrio spp., even though fecal coliforms have long been used as indicators of problematic bacterial contamination.

Raw and Processed Shrimp

Results of a study of the microbiological content of 1,264 samples of individually quick-frozen, peeled, and deveined raw shrimp (pond-raised Penaeus monodon) and 914 samples of cooked ready-to-eat shrimp are provided in Table 6.7. Among the raw shrimp samples, 96% had an APC of less than 105 CFU/g, and 74% had an APC of less than 104 CFU/g. APC of cooked, ready-to-eat shrimp were less than 104 CFU/g in 99% of the samples (93). Table 6.7  APC of individually quick-frozen raw and

cooked ready-to-eat shrimp analyzed from January 1994 to December 1995

Sample Raw shrimp Cooked shrimp

No. of samples analyzed

% of samples with APC (CFU/g) of:

102

103

1,264

35

39

914

79

20

Coliforms were present in 15% of frozen raw shrimp but in only 3% of frozen cooked shrimp samples (94). Studies have revealed that the total APC can increase even in adequately refrigerated shrimp, whereas the levels of ­coliforms and thermotolerant coliforms increase only in product that has been temperature abused. Raw shrimp contain coagulase-positive staphylococci, whereas cooked shrimp generally are negative for this microorganism. Since the presence of staphylococci in raw shrimp is typically attributed to workers, the absence of Staphylococcus in cooked product indicates that no postprocessing contamination has occurred. Substantial numbers of aeromonads capable of growth at low temperatures are present in freshwater prawns, which influences the shelf life of iced freshwater prawns, i.e., between 12 and 16 days. Pseudomonas, Aeromonas species A. hydrophila, A. veronii, A. boivar, and A. sobria, and S. putrefaciens have been identified as spoilage bacteria for fresh prawn (92). A study by Lalitha and Surendran (173) has revealed that after 26 days of iced storage the APC of freshwater prawns exceeded 107 CFU/g; an APC of 106 CFU/g is considered unsafe. An international study of aquacultured shrimp in which 1,234 samples from 103 shrimp aquaculture farms in six countries were analyzed for the presence of fecal coliforms, E. coli, and Salmonella revealed a significant relationship (P = 0.0342) between the log number of fecal bacteria and the probability that any given sample would contain Salmonella. The likelihood of any given sample containing Salmonella increased by 1.2 times with each 10-fold increase in either fecal coliform or E. coli concentration. The statistical relationship between Salmonella cell numbers and that of both fecal coliforms and E. coli was highest in grow-out pond water. The likelihood of identifying Salmonella in grow-out pond water increased 2.7 times with each log increase in fecal coliform concentration and 3.0 times with each log increase in E. coli concentration. Salmonella is not part of the natural flora of the shrimp culture environment, nor is it inherently present in shrimp grow-out ponds. The occurrence of Salmonella in shrimp from aquaculture operations is largely related to the concentration of fecal bacteria in the source of the grow-out water (Table 6.8).

Shellfish Microbial Hazards

The microfloras in estuarine and marine environments include various members of the Vibrionaceae family, some of which are pathogenic or potentially pathogenic

Microbial Spoilage and Public Health Concerns

150 Table 6.8  Sample types and Salmonella prevalence on

shrimp aquaculture farms

No. of samples tested

No. (%) of Salmonella-positive samples

Feces

65

6 (9.2)

Holding pond water

40

1 (2.5)

Other

5

0 (0)

Sample type

a

Pond sediment

225

2 (1)

Pond grow-out water

261

9 (3.5)

Ice

16

0 (0)

Processing water

22

3 (13.6)

Drinking water

117

5 (4.3)

Probiotics, fertilizer

25

1 (4)

Shrimp

247

4 (1.6)

Shrimp feed

63

0 (0)

Source sediment

25

6 (24)

Source water

120

6 (5)

Wastewater

3

1 (33)

1,234

44 (3.6)

Overall  

a

Includes crabs and frogs.

to humans and constitute a potential health threat for consumers of raw or partially cooked bivalves (89). More than 35 species of Vibrio have been identified, and more species are being added as new scientific information becomes available. In addition to V. cholerae, the most widely known pathogenic Vibrio, a number of Vibrio species may cause disease. Vibrio parahaemolyticus, V. mimicus, and V. vulnificus are food-­poisoning bacteria frequently isolated from seawater and shellfish. Among halophilic vibrios, V. alginolyticus, V. fluvialis, and V. metschnikovii are also pathogenic for humans. These human pathogens have distinct clinical features, pathogenic mechanisms, epidemiologic characteristics, and ecological positions. However, in recent years, microorganisms from the genera Aeromonas and Plesiomonas have also been implicated in a number of enterotoxic episodes. In 1998, shellfish were incriminated in 7.1% of the 84 reported foodborne disease outbreaks in Italy. The European Directive No. 492/91 prescribes that bivalve mollusks marketed for human consumption should be raised in waters that meet certain microbiological requirements defined as “Type A” or, if originating elsewhere, should be run through a depuration procedure. In order for bivalve mollusks to be deemed fit for human consumption, legislation requires that they contain

<300 fecal coliforms (or <230 E. coli cells per 100 g of shellfish meat and intervalve water) and no Salmonella species in 25 g of meat. In the United States, the fecal coliform median (or geometric mean) of the water sample should not exceed 14 per 100 ml, and the estimated 90th percentile shall not exceed (i) 43 MPN per 100 ml for a five-tube decimal dilution test or (ii) 49 MPN per 100 ml for a three-tube decimal dilution test. However, several studies have revealed there is no close correlation between the presence in these products of microorganisms of fecal origin and the presence of vibrios that are potentially pathogenic for humans. The vibrios constitute a considerable part of marine halophilic bacterial populations. Environmental research has revealed that different ecological parameters such as nutrient availability, temperature, and salinity influence the presence and persistence of different Vibrio species in the sea. Studies have also revealed that V. vulnificus is not detected in seawater, oysters, or suspended particulate matter (SPM) samples during the cold winter months but can be detected at low levels in sediment samples collected during this time period. Increased levels of this microbe are first observed in early spring in the sediment, then in SPM, and finally in oysters. A major increase in V. vulnificus populations occurs only after the seawater temperature increases to above 20°C and the salinity decreases below 16 ppt due to winter and spring rainfall. The highest V. vulnificus levels recorded were associated with SPM. These results suggest that V. vulnificus (i) overwinters in a floc zone present at the sedimentwater interface; (ii) is resuspended into the water ­column in early spring following changes in climatic conditions; (iii) colonizes the surfaces of zooplankton that is also blooming during early spring; and (iv) is ingested by oysters during their normal feeding process (311). In coastal environments, a direct relationship exists between the consumption of raw or partially cooked shellfish and the occurrence of episodes of human intestinal and extraintestinal infection attributed to some Vibrio species. Mollusks are theoretically an excellent vehicle of vibrios because they concentrate these bacteria in their soft tissues by filter feeding. It has been estimated that 1 in 2,000 meals of raw shellfish results in disease, giving shellfish a reputation as one of the most hazardous foods. The most studied mollusk species are the Pacific oyster (Crassostrea gigas) and the Eastern oyster (Crassostrea virginica).

Oysters

Oysters are filter feeders that efficiently concentrate microorganisms; because oysters are consumed raw, they

6.  Meat, Poultry, and Seafood pose a health risk to consumers. The consumption of raw oysters has been linked to outbreaks of acute gastroenteritis in several communities in both the Eastern and Western hemispheres. Pathogenic Vibrio spp., in particular V. vulnificus, V. parahaemolyticus, and V. cholerae non-O1, are of concern in oysters. Nearly all V. vulnificus strains found in oysters appear to be virulent, and genetic tests have not been able to distinguish fully virulent from less virulent strains or environmental from clinical strains. The prevalence of V. vulnificus and V. parahaemolyticus in the contents of in-shell oysters (Crassostrea virginica and Crassostrea gigas) sampled from 370 lots, coming from 275 different establishments (71% restaurants or oyster bars, 27% retail seafood markets, and 2% wholesale seafood markets) in coastal and inland regions of the United States over a 1-year period was determined by Cook et al. (43). The oysters were harvested from the Gulf of Mexico (49%), Pacific (14%), Mid-Atlantic (18%), and North Atlantic (11%) coasts and from Canada (8%). Results of the percentage of positive lots from each region and the V. vulnificus and V. parahaemolyticus MPN counts within each of eight specified density ranges are provided in Tables 6.9 and 6.10. The concentrations of these bacteria in market oysters from all harvest regions followed a seasonal distribution, with the highest densities observed during the summer months. The highest densities of both microbes were observed in oysters harvested from the Gulf Coast, where densities often exceeded 10,000 MPN/g. Most (78%) of the lots harvested from the North Atlantic, Pacific, and Canadian coasts had V. vulnificus densities below the detectable level of 0.2 MPN/g; none exceeded 100 MPN/g. V. parahaemolyticus densities were greater than those of V. vulnificus in lots from the same areas, with some lots exceeding 1,000 MPN/g for V. parahaemolyticus. Overall, there was a significant correlation with salinity. Storage time significantly affected both V. vulnificus (10% decrease per day) and V. parahaemolyticus (7% decrease per day) densities in the market oysters. Intertidal harvest is practiced extensively in some Pacific Northwest estuaries. This involves hand picking the shellfish after the tide recedes from the harvest area and placing them in large baskets. The baskets are left in the harvest area until the tide rises to a depth sufficient for a vessel to retrieve and transport them to the processing plant. Intertidal harvesting potentially exposes oysters to favorable conditions for growth of vibrios as well as other bacteria, especially on sunny days. Nordstrom et al. (207) determined that

151 when using the intertidal harvest method, the mean V. parahaemolyticus densities in oysters were generally four to eight times greater at maximum exposure (when the next high tide occurred) than at the corresponding first exposure (when the initial low tide occurred). While pathogen counts were generally low (£10 CFU/g) at first exposure, counts as high as 160 CFU/g were found at maximum exposure. Pathogenic V. parahaemolyticus was detected in 21% of the oyster samples at maximum exposure and in 26% of sediment samples. Hence, summer conditions permit rapid multiplication of the bacterium in oysters exposed by a receding tide. The incidence of Salmonella infections has increased in recent years, and many cases are associated with seafood (37), particularly with the consumption of shellfish. Harvesting areas have become more populated in recent years, resulting in more human ­sewage discharged into coastal waters with an accompanying increase in pathogens; hence, a higher incidence of foodborne disease from shellfish has occurred. In oysters harvested from 36 bays in the United States (12 each from the West, East, and Gulf coasts in the summer and 12 bays, four per coast in the winter), Salmonella was isolated from each coast; 7.4% of all oysters sampled tested positive for the bacterium. Isolates tended to be bay specific, with some bays having a high prevalence of Salmonella, whereas other bays had none. A difference in the percentage of oysters from which Salmonella was isolated was observed between the summer and winter months; the prevalence during the winter months was much lower, probably due to a variety of weather-related factors. The vast majority (78/101) of Salmonella isolates from oysters were Salmonella enterica serovar Newport, a major human pathogen, confirming the potential hazard of raw oyster consumption. In contrast to previous findings, there was no relationship between the isolation of fecal coliforms from the water and the presence of Salmonella in oysters. Many viral illnesses have also been associated with the consumption of contaminated shellfish. Viruses likely to be transmitted by this route are enteric viruses, which are capable of persisting in the environment and can be concentrated by shellfish. In the 1960s, hepatitis A was the predominant viral disease resulting from the consumption of raw oysters, but more recently, acute viral gastroenteritis has emerged as the most prevalent. However, hepatitis A virus is still a pathogen of concern when raw oysters are consumed. In 2005, hepatitis A virus was confirmed as causing illness among restaurant patrons in four

Microbial Spoilage and Public Health Concerns

152

Table 6.9  Prevalence and cell densities of Vibrio vulnificus in oysters at retail by harvest region Harvest state, country, or region

% of samples with V. vulnificus densities in MPN/g range of:

No. of samples

0 (none detected)

<1

>1–10

Connecticut

17

52.9

17.6

29.4

Massachusetts

12

75.0

8.3

8.3

8.3

Rhode Island

12

91.7

8.3

Canada, Atlantic

19

73.7

10.5

10.5

5.3

  All North Atlantic

60

71.7

11.7

13.3

3.3

Washington

44

81.8

6.8

4.5

6.8

California

1

100.0

Oregon

2

100.0

Canada, Pacific

5

100.0

  All Pacific Coast

52

90.4

3.8

1.9

3.8

Delaware

2

50.0

Maryland

25

56.0

12.0

16.0

4.0

New Jersey

11

45.5

27.2

18.2

>10–102

>102–103

>103–104

>104–105

>105

50.0 4.0

8.0 9.1

Virginia

26

42.3

3.8

7.7

15.4

15.4

7.7

7.7

  All Mid-Atlantic

64

48.4

6.3

14.1

4.7

9.4

10.9

3.1

3.1

Florida

47

8.5

8.5

10.6

10.6

14.9

36.2

8.5

2.1

Mississippi

4

25.0

25.0

25.0

25.0

Louisiana

80

3.8

1.3

13.8

21.3

27.5

12.5

Texas

35

2.9

  All Gulf Coast

166

4.8

Mexico

1

Unknown

2

15.0

6.3

2.9

8.6

28.6

8.6

31.4

17.1

3.6

12.1

12.1

13.2

27.7

19.8

100.0 50.0

states who ate Gulf Coast oysters. Many cases of gastroenteritis are caused by small round-structured viruses. A small round-structured virus was implicated in an outbreak caused by oysters (Crassostrea virginica) harvested in Louisiana. Molecular characterization has revealed that these viruses are human caliciviruses and that they are genetically related to the Norwalk virus.

Mussels

7.2

Samples of mussels (Mytilus galloprovincialis) were collected from approved coastal sites located on the Adriatic Sea (Central Italy) to be examined for the presence of V. parahaemolyticus and the occurrence of pathogenic strains. From 144 samples, 35 V. parahaemolyticus strains were isolated. Genes coding for the major virulence factor of V. parahaemolyticus

50.0

were detected in four isolates. Another Italian study of the same mussel species obtained from retail stores in the Puglia region revealed that V. parahaemolyticus and V. vulnificus were in 7.83% and 2.83% of the samples, respectively. A third study of mussel farms in Italy revealed that a total of 125 vibrios from 152 isolates were identified as Vibrio fluvialis (55 strains), V. alginolyticus (40 strains), V. parahaemolyticus (11 strains), and V. mimicus (9 strains). The remaining 27 isolates were not identified. A fourth study of mussels and water collected from 30 sampling sites in the Ionian Sea revealed that V. alginolyticus was the predominant species of the total culturable vibrios. Some Vibrio species such as V. mediterranei, V. parahaemolyticus, V. diazotrophicus, V. nereis, and V. splendidus were present in both the water and the mussels. Selective retention in mussels, however, was

6.  Meat, Poultry, and Seafood

153

Table 6.10  Prevalence and cell densities of Vibrio parahaemolyticus in oysters at retail by harvest region % of samples with V. parahaemolyticus densities in MPN/g range of: Harvest state, country, or region

0 No. of samples (none detected)

<1

>1–10

>10–102

>102–103

>103–104

Connecticut

17

47.0

11.7

11.7

29.4

Massachusetts

12

50.0

8.3

8.3

8.3

16.6

8.3

Rhode Island

12

91.7

8.3

Canada, Atlantic

20

50.0

10.0

15.0

10.0

5.0

10.0

  All North Atlantic

61

57.4

9.8

9.8

13.2

4.9

4.9

Washington

44

65.9

11.4

6.8

6.8

4.5

4.5

California

1

100.0

Oregon

2

50.0

Canada, Pacific

5

80.0

  All Pacific Coast

52

67.3

3.8

3.8

Delaware

2

Maryland

25

16.0

New Jersey

11

54.4

Virginia

26

26.9

15.4

  All Mid-Atlantic

64

26.6

17.1

Florida

49

4.1

>104–105

>105

50.0 20.0 9.6

5.8

9.6 50.0

28.0

32.0

50.0

20.0

4.0

18.2

9.1

15.4

11.5

11.5

7.7

11.5

18.8

17.1

7.8

4.7

7.8

2.0

16.3

16.3

40.8

12.2

6.1

2.0

28.2

Mississippi

4

25.0

50.0

25.0

Louisiana

80

2.6

1.3

11.3

7.5

11.3

30.0

27.5

8.8

Texas

34

5.8

5.8

7.4

29.4

11.8

23.5

5.8

2.9

  All Gulf Coast

167

3.6

2.4

13.2

14.9

20.9

23.4

16.2

5.4

12.9

13.5

9.2

2.6

Mexico

1

Unknown

2

50.0

347

27.1

  Total

100.0 50.0 7.5

observed for other vibrios (V. vulnificus, V. cincinnatiensis, V. orientalis, V. anguillarum, V. marinus, and V. hollisae). A fifth study, performed over a 2-year period in which 726 bacterial strains were isolated, revealed that 46.9% were of the Vibrio genus, 29% of the Aeromonas genus, and the remaining 23.3% of the Pseudomonas, Flavobacterium, Pasteurella, Agrobacterium, and Ochrobacterium genera (Table 6.11). A Norwegian study of 885 blue mussel (Mytilus edulis) samples revealed that V. parahaemolyticus, V. cholerae, and V. vulnificus were isolated at rates of 10.3%, 1.0%, and 0.1%, respectively. Four of the V. parahaemolyticus samples contained one of the major virulent genes. A study of Campylobacter contamination of mussels (species not provided) in The Netherlands re-

12.6

14.4

vealed that most of the contamination was caused by Campylobacter lari (over 90%). Other Campylobacter species isolated were C. jejuni, C. coli, C. upsaliensis, and C. hyointestinalis. The study was initiated because Campylobacter is a common bacterial pathogen that causes enteritis in humans worldwide. Specifically, C. jejuni and C. coli ­account for most Campylobacter enteric infections in humans. C. lari appears to be widely present in the environment but is rarely reported as a human pathogen. During recent years, a very small number of cases of C. lari have been described.

Crawfish

The microflora associated with live crawfish reflects the microbial populations of the harvest water and sediments in which the crawfish were harvested (91). A study of

Microbial Spoilage and Public Health Concerns

154

Table 6.11  Prevalence of genera and species of bacteria isolated from mussels and seawater samples

Genus

No. of isolates in genus

% of total isolates

Vibrio

340

46.9

Aeromonas

Pseudomonas

Flavobacterium

216

50

43

29.8

6.9

5.9

No. of isolates in species

% of isolates in same genus

% of total isolates

V. alginolyticus

231

68.0

31.8

Photobacterium damselae

37

10.9

5.1

V. parahaemolyticus

34

10

4.7

V. vulnificus

34

10

4.7

V. cholerae non-O1

4

1.2

0.6

A. salmonicida

125

57.9

17.2

A. hydrophila/caviae

84

38.9

11.6

A. sobria

7

3.2

1.0

P. paucimobilis

25

50

3.4

P. putrefaciens

15

30

2.1

P. vesicularis

10

20

1.4

F. indologenes

19

44.2

2.6

F. multivorum

16

37.2

2.2

F. odoratum

8

18.6

1.1

Pasteurella spp.

23

82.1

3.2

Species

Pasteurella

28

3.9

P. multocida

5

17.9

0.7

Ochrobacterium

28

3.9

O. anthropi

28

100

3.9

A. radiobacter

21

100

2.9

726

100

100

Agrobacterium

21

2.9

TOTAL

726

100

health-related bacteria in fresh crawfish (Procambarus clarkii) revealed the presence of coliforms (100%), E. coli (92.6%), fecal streptococci (94.1%), ­ coagulasepositive staphylococci (3.0%), and Salmonella (3.0%). Clostridium botulinum type E was not found in any of the samples analyzed (66 collections from 22 sites). Staphylococcus aureus, Streptococcus faecalis, E. coli, and Salmonella serovar Typhimurium grew well in cooked tail meat at 25 and 37°C. None of these four bacteria grew at 5°C. Clostridium botulinum type E produced toxin in the tail meat within 48 to 72 h at 30°C and after 33 days at 5°C. Toxin was not produced at 55 days when meats were stored in packed ice. A subsequent study identified the bacteria responsible for spoilage in crawfish tail meat stored at 0 and 5°C. The results (Tables 6.12 and 6.13) showed substantial differences in the bacterial species responsible for spoilage. Pseudomonas was the major bacterial genus responsible for spoilage at 0°C, whereas the Achromobacter group (including Moraxella and Acinetobacter) was the major group of bacteria ­responsible for spoilage at 5°C.

A survey of raw whole crawfish, cooked tail meat, and environmental samples was conducted in crawfish processing facilities. Of 337 samples collected, 31 contained Listeria spp. Although Listeria innocua was the predominant Listeria species obtained, four samples were positive for L. monocytogenes. L. monocytogenes was detected in three raw material samples and one environmental sample. Listeria spp. were found in 29.5% of raw whole crawfish, in 4.4% of environmental samples, and in none of the finished product samples. Among the environmental samples, Listeria spp. were isolated from 15.4% of the drains and 5.1% of the employee contact surfaces (gloves and aprons) but in none of the samples from food contact surfaces.

Crab

Until recent years, blue crab meat was marketed as a fresh product. The major spoilage microorganism was Pseudomonas. Outbreaks of V. parahaemolyticus and S. aureus foodborne illnesses have been associated with crab meat consumption. L. monocytogenes has been

6.  Meat, Poultry, and Seafood

155

Table 6.12  Bacteria isolated from crayfish tail meat, fresh or stored at 0°C for up to 24 days % of isolatesa after indicated storage period (days) Genus

0 (fresh)

8

12

24 (spoilage)

Achromobacter

12.5

82.5

17.5

22.5

Alcaligenes

15.0

2.5

<1.0

7.5

Flavobacterium

10.0

5.0

5.0

<1.0

Micrococcus

30.0

<1.0

<1.0

<1.0

Pseudomonas

<1.0

7.5

45.0

65.0

Staphylococcus

17.5

2.5

<1.0

<1.0

Other genera

15.0

<1.0

32.5

5.0

b

   

Percentage of total isolates at the indicated time of storage. b Other genera included Aerobacter, Bacillus, Lactobacillus, Proteus, and Sarcina. a

detected in 6% of fresh blue crab meat. Most blue or swimming crab meat is marketed pasteurized, both to extend shelf life and to eliminate pathogenic micro­ organisms. Properly pasteurized and stored crab meat can have a shelf life of several years; however, the normal storage times range from 6 months to one year. Pasteurized meat should be stored under refrigerated conditions at 3°C, because pathogens such as psychrotrophic strains of B. cereus, C. botulinum E, nonproteolytic C. botulinum B, and nonproteolytic C. botulinum F have the potential to grow under refrigerated storage conditions. Depending upon the pasteurization process and the conditions before and after pasteurization, several ­microorganisms have been identified in the meat. These include Bacillus, Micrococcus, Alcaligenes, Coryne­ bacterium, Lactobacillus, Flavobacterium, Acineto­

bacter, and Brevebacterium. When spoiled pasteurized crab meat was examined, the following microorganisms were identified: Bacillus, Carnobacterium, Acinetobacter, Aerococcus, Rhodococcus, Providencia, Brevebacterium, Clostridium (but not C. botulinum or C. perfringens), Kocuria, and Pseudomonas. Many of these bacteria were viable in pasteurized crab meat stored for over 20 years. The growth that appeared indicated that some microorganisms were surviving in a dormant state and when exposed to favorable growth conditions still retained the capacity to reproduce. All of the microorganisms found in stored crab meat were not strict anaerobes and could be classified as facultative anaerobes. The other major crabs of commerce, king and snow crabs, are primarily marketed in the frozen state. As such, spoilage would only occur subsequent to home or institutional preparation.

Table 6.13  Bacteria isolated from crayfish tail meat, fresh or stored at 5°C for up to 24 days % of isolatesa after indicated storage period (days) Genus

0 (fresh)

8

12

24 (spoilage)

Achromobacter

12.5

47.5

65.0

62.5

Alcaligenes

15.0

15.0

15.0

<1.0

Flavobacterium

10.0

7.5

<1.0

2.5

Micrococcus

30.0

2.5

<1.0

<1.0

Pseudomonas

<1.0

10.0

17.5

32.5

Staphylococcus

17.5

12.5

<1.0

<1.0

Other genera

15.0

5.0

2.5

2.5

b

   

a b

Percentage of total isolates at the indicated time of storage. Other genera included Aerobacter, Bacillus, Lactobacillus, Proteus, and Sarcina.

156

CONCLUDING REMARKS Contamination of raw muscle foods with spoilage microbes and pathogens is natural and unavoidable but should be minimized and kept as low as possible at all stages of the food chain. Contamination control is needed, even at preharvest, because it may lead to a reduced probability that errors occurring in subsequent parts of the food chain will lead to foodborne illness associated with muscle or related food products. Decontamination processes are applied to animals and carcasses through a variety of physical and chemical interventions. Additional interventions to help enhance food safety are applied during processing and include heating, chilling, freezing, drying, fermentation, use of chemicals as acidulants or antimicrobials, packaging, proper storage and distribution, and appropriate handling and preparation for consumption. Food safety assurance involves activities and responsibilities ­throughout the food chain. Contamination control efforts early in food production allow subsequent food processes to be effective. Since microbial hazards and associated issues remain major challenges to muscle food safety, it is important to realize that management of meat safety risks should be based on an integrated effort and approach that applies to all sectors, from the producer through the processor, distributor, packer, retailer, food service worker and consumer.

References 1. Adam, K. H., S. H. Flint, and G. Brightwell. 2010. Psychrophilic and psychrotrophic clostridia: sporulation and germination processes and their role in the spoilage of chilled, vacuum-packaged beef, lamb and venison. Int. J. Food Sci. Technol. 45:1539–1544. 2. Allen, V. M., J. E. L. Corry, C. H. Burton, R. T. Whyte, and G. C. Mead. 2000. Hygienic aspects of modern poultry chilling. Int. J. Food Microbiol. 58:39–48. 3. Allen, V. M., H. Weaver, A. M. Ridley, J. A. Harris, M. Sharma, J. Emery, N. Sparks, M. Lewis, and S. Edge. 2008. Sources and spread of thermophilic Campylobacter spp. during partial depopulation of broiler chicken flocks. J. Food Prot. 71:264–270. 4. American Meat Institute. 2003. Sanitary Design Checklist and Fact Sheet. Equipment Design Task Force, American Meat Institute, Washington, DC. 5. Ammor, M. S., A. Argyri, and G.-J. E. Nychas. 2009. Rapid monitoring of the spoilage of minced beef stored under conventionally and active packaging conditions using Fourier transform infrared spectroscopy in tandem with chemometrics. Meat Sci. 81:507–514. 6. Anang, D. M., G. Rusul, F. H. Ling, and R. Bhat. 2010. Inhibitory effects of lactic acid and lauricidin on spoilage

Microbial Spoilage and Public Health Concerns organisms of chicken breast during storage at chilled temperature. Int. J. Food Microbiol. 144:152–159. 7. Argyri, A. A., E. Z. Panagou, P. A. Tarantilis, M. Polysiou, and G.-J. E. Nychas. 2010. Rapid qualitative and quantitative detection of beef fillets spoilage based on Fourier transform infrared spectroscopy data and artificial neural networks. Sensors Actuators B 145:146–154. 8. Arnold, J. W., and S. Silvers. 2000. Comparison of poultry processing equipment surfaces for susceptibility to bacterial attachment and biofilm formation. Poult. Sci. 79:1215–1221. 9. Arthur, T. M., D. M. Brichta-Harhay, J. M. Bosilevac, N. Kalchayanand, S. D. Shackelford, T. L. Wheeler, and M. Koohmaraie. 2010. Super shedding of Escherichia coli O157:H7 by cattle and the impact on beef carcass contamination. Meat Sci. 86:32–37. 10. Avens, J. S., S. N. Albright, A. S. Morton, B. E. Prewitt, P. A. Kendall, and J. N. Sofos. 2002. Destruction of microorganisms on chicken carcasses by steam and boiling water immersion. Food Control 13:445–450. 11. Aymerich, T., P. A. Picouet, and J. M. Monfort. 2008. Decontamination technologies for meat products. Meat Sci. 78:114–129. 12. Bacon, R. T., and J. N. Sofos. 2003. Food hazards: biological food; characteristics of biological hazards in foods, p. 157–195. In R. H. Schmidt and G. Rodrick (ed.), Food Safety Handbook. Wiley Interscience, New York, NY. 13. Bailey, J.-D., and P. Guerre. 2009. Mycotoxins in meat and processed meat products, p. 83–124. In F. Toldra (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 14. Bailey, J. S., J. E. Thomson, and N. E. Cox. 1987. Contamination of poultry during processing, p. 193–211. In F. E. Cunningham and N. A. Cox (ed.), The Microbiology of Poultry Meat Products. Academic Press, Orlando, FL. 15. Balamatsia, C. C., E. K. Paleologos, M. G. Kontominas, and I. N. Savvaidis. 2006. Correlation between microbial flora, sensory changes and biogenic amines formation in fresh chicken meat stored aerobically or under modified atmosphere packaging at 4 degrees C: possible role of biogenic amines as spoilage indicators. Antonie van Leeuwenhoek Int. J. G 89:9–17. 16. Barbut, S. 2002. Poultry Products Processing, an Industry Guide. CRC Press, New York, NY. 17. Barbut, S., L. F. Moza, F. Nattress, B. Dilts, and C. O. Gill. 2009. The microbiological conditions of air or water chilled carcasses at the same poultry packing plant. J. Appl. Poult. Res. 18:501–507. 18. Barker, G. C., N. Gomez, and J. Smid. 2009. An introduction to biotracing in food chain systems. Trends Food Sci. Technol. 20:220–226. 19. Baron, W. L., L. E. Dawson, and R. V. Lechowich. 1973. Influence of chlorine dioxide water treatment on numbers of bacteria associated with processed turkey. Poult. Sci. 52:1053–1058. 20. Bauermeister, L. J., J. W. J. Bowers, J. C. Townsend, and S. R. Mckee. 2008. Validating the efficacy of peracetic acid mixture as an antimicrobial in poultry chillers. J. Food Prot. 71:1119–1122.

6.  Meat, Poultry, and Seafood 21. Bazemore, R., S. G. Fu, Y. Yoon, and D. Marshall. 2003. Major causes of shrimp spoilage and methods for assessment, p. 223–234. In A. M. Rimando and K. K. Schrader (ed.), Off-Flavors in Aquaculture. ACS Symposium Series no. 848, American Chemical Society, Washington, DC. 22. Beef Industry Food Safety Council. 2006. Best practices: pathogen control during tenderizing/enhancing of whole muscle cuts. Available at: http://www.bifsco. org/CMDocs/BIFSCO/Best%20Practices/03_29_06non­intactbestpractices.pdf. Accessed 28 February 2011. 23. Berrang, M. E., J. K. Northcutt, D. L. Fletcher, and N. A. Cox. 2003. Role of dump cage fecal contamination in the transfer of Campylobacter to carcasses of previously negative broilers. J. Appl. Poult. Res. 12:190–195. 24. Bjorkroth, J. 2005. Microbiological ecology of marinated meat products. Meat Sci. 70:477–480. 25. Bolder, R. D. 1997. Decontamination of meat and poultry carcasses. Trends Food Sci. Technol. 8:221–227. 26. Brewer, M. S. 2009. Irradiation effects on meat flavor: a review. Meat Sci. 81:1–14. 27. Broda, D. M., D. J. Saul, R. G. Bell, and D. R. Musgrave. 2000. Clostridium algidixylanolyticum sp. nov., a psychrotolerant, xylan-degrading, spore-forming bacterium. Int. J. Syst. Evol. Microbiol. 50:623–631. 28. Buhr, R. J., D. V. Bourassa, J. K. Northcutt, A. Hinton, K. D. Ingram, and J. A. Cason. 2005. Bacteria recovery from genetically feathered and featherless broiler carcasses after immersion chilling. Poult. Sci. 84:1499–1504. 29. Bunkova, L., F. Bunka, P. Klcovska, V. Mrkvicka, M. Dolezalova, and S. Kracmar. 2010. Formation of biogenic amines by Gram-negative bacteria isolated from poultry skin. Food Chem. 121:203–206. 30. Burgess, C. M., L. Rivas, M. J. McDonnell, and G. Duffy. 2008. Biocontrol of pathogens in the meat chain, p. 253–288. In F. Toldra (ed.), Meat Biotechnology. Springer, New York, NY. 31. Byelashov, O. A., and J. N. Sofos. 2009. Strategies for on-line decontamination of carcasses, p. 149–182. In F. Toldrá (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 32. Cannarsi, M., A. Baiano, R. Sinigaglia, and M. A. Del Nobile. 2005. Use of biodegradable films for fresh cut beef steaks packaging. Meat Sci. 70:259–265. 33. Carlson, B. A., K. K. Nightingale, G. L. Mason, J. R. Ruby, W. T. Choat, G. H. Loneragan, G. C. Smith, J. N. Sofos, and K. E. Belk. 2009. Escherichia coli O157: H7 strains that persist in feedlot cattle are genetically related and demonstrate an enhanced ability to adhere to intestinal epithelial cells. Appl. Environ. Microbiol. 75:5927–5937. 34. Carroll, C. D., and C. Z. Alvarado. 2008. Comparison of air and immersion chilling on meat quality and shelf life of marinated broiler breast fillets. Poult. Sci. 87:368–372. 35. Cason, J. A., N. A. Cox, and J. S. Bailey. 1994. Transmission of Salmonella typhimurium during hatching of broiler chicks. Avian Dis. 38:583–588. 36. Cason, J. A., A. D. Whittemore, and A. D. Shackelford. 1999. Aerobic bacteria and solids in a three-tank, two pass counterflow scalder. Poult. Sci. 78:144–147.

157 37. Centers for Disease Control and Prevention. 2000. Surveillance for foodborne disease outbreaks—United States, 1993−1997. MMWR Surveill. Summ. 49:1–51. 38. Cerveny, J., J. D. Meyer, and P. D. Hall. 2010. Microbiological spoilage of meat and products, p. 69–86. In W. H. Sperber and M. P. Doyle (ed.), Compendium of the Microbiological Spoilage of Foods and Beverages. Springer, New York, NY. 39. Charles, N., S. K. Williams, and G. E. Rodrick. 2006. Effects of packaging systems on the natural microflora and acceptability of chicken breast meat. Poult. Sci. 85:1798–1801. 40. Chiesa, F., and L. Vercellotti. 1992. Controllo igienicosanitario e ambientale in un macello per la selvaggina allevata. Ingegneria Aliment. 4:24. 41. Cocolin, L., P. Dolci, and K. Rantsiou. 2008. Molecular methods for identification of microorganisms in traditional meat products, p. 91–127. In F. Toldrá (ed.), Meat Biotechnology. Springer, New York, NY. 42. Coma, V. 2008. Bioactive packaging technologies for extended shelf life of meat-based products. Meat Sci. 78:90–103. 43. Cook, D. W., P. O’Leary, J. C. Hunsucker, E. M. Sloan, J. C. Boweers, R. J. Blodgett, and A. DePaola. 2002. Vibrio vulnificus and Vibrio parahaemolyticus in U.S. retail shell oysters: a national survey from June 1998 to July 1999. J. Food Prot. 65:79–87. 44. Cornforth, D. P., and M. C. Hunt. 2008. Low-oxygen packaging of fresh meat with carbon monoxide: meat quality, microbiology, and safety. American Meat Science Association White Paper Series, January. American Meat Science Association, Savoy, IL. 45. Cotton, L. N., and D. L. Marshall. 1998. Rapid impediometric method to determine crustacean food freshness, p. 147–160. In M. H. Tunick, S. A. Palumbo, and P. M. Fratamico (ed.), New Techniques in the Analysis of Foods. Plenum Publishing Corp., New York, NY. 46. Cox, N. A., J. S. Bailey, J. M. Mauldin, and L. C. Blankenship. 1990. Presence and impact of Salmonella contamination in commercial broiler hatcheries. Poult. Sci. 69:1606–1609. 47. Cox, N. A., L. J. Richardson, J. S. Bailey, D. E. Cosby, J. A. Cason, and M. T. Musgrove. 2005. Bacterial contamination of poultry as a risk to human health, p. 21– 43. In J. Mead (ed.), Food Safety Control in the Poultry Industry. CRC Woodhead Publishing Limited, New York, NY. 48. Cox, N. E., M. E. Berrang, and J. A. Cason. 2000. Salmonella penetration of egg shells and proliferation in broiler hatching eggs—a review. Poult. Sci. 79:1571–1574. 49. Crowley, K. M., D. M. Prendergast, J. J. Sheridan, and D. A. McDowell. 2010. Survival of Pseudomonas fluorescens on beef carcass surfaces in a commercial abattoir. Meat Sci. 85:550–554. 50. Dainty, R. H. 1996. Chemical/biochemical detection of spoilage. Int. J. Food Microbiol. 33:19–34. 51. Dainty, R. H., R. A. Edwards, and C. M. Hibbard. 1985. Time course of volatile compound formation ­ during

Microbial Spoilage and Public Health Concerns

158

52. 53.

54.

55. 56.

57.

58.

59.

60.

61.

62. 63.

64. 65.

r­ efrigerated storage of naturally contaminated beef in air. J. Appl. Bacteriol. 59:303–309. Dainty, R. H., R. A. Edwards, and C. M. Hibbard. 1989. Spoilage of vacuum-packed beef by a Clostridium sp. J. Sci. Food Agric. 49:473–486. Dalgaard, P. 2000. Fresh and lightly preserved seafood, p. 110–139. In C. M. D. Man and A. A. Jones (ed.), Shelf-life Evaluation of Foods. Aspen Publishers, London, United Kingdom. Dalsgaard, A., H. H. Huss, A. H-Kittikum, and J. L. Larsen. 1995. Prevalence of Vibrio cholerae and Salmonella in a major shrimp production area in Thailand. Int. J. Food Microbiol. 28:101–113. Dalvit, C., M. De Marchi, and M. Cassandro. 2007. Genetic traceability of livestock products: a review. Meat Sci. 77:437–449. Davies, A. R. 1995. Advances in modified-atmosphere packaging, p. 304–320. In G. W. Gould (ed.), New Methods of Food Preservation. Blackie Academic and Professional, London, United Kingdom. De Boer, E., A. Zwartkruis-Nahuis, A. E. Heuvelink, C. Harmanus, and E. J. Kuijper. 2011. Prevalence of Clostridium difficile in retailed meat in The Netherlands. Int. J. Food Microbiol. 144:561–564. De Boer, E., J. T. M. Zwartkruis-Nahuis, B. Wit, X. W. Huijsdens, A. J. de Neeling, T. Bosch, R. A. A. van Oosterom, A. Vila, and A. E. Heuvelink. 2009. Prevalence of methicillin-resistant Staphylococcus aureus in meat. Int. J. Food Microbiol. 134:52–56. Del Rio, E., R. Capita, M. Prieto, and C. AlonsoCalleja. 2006. Comparison of pathogenic and spoilage bacterial levels on refrigerated poultry parts following treatment with trisodium phosphate. Food Microbiol. 23:195–198. Del Rio, E., M. Panizo-Moran, M. Prieto, C. AlonsoCalleja, and R. Capita. 2007. Effect of various chemical decontamination treatments on natural microflora and sensory characteristics of poultry. Int. J. Food Microbiol. 115:268–280. Del Rio, E., R. Muriente, M. Prieto, C. Alonso-Calleja, and R. Capita. 2007. Effectiveness of trisodium ­phosphate, acidified sodium chlorite, citric acid, and peroxyacids against pathogenic bacteria on poultry during refrigerated storage. J. Food Prot. 70:2063–2071. De Vuyst, L., G. Falony, and F. Leroy. 2008. Probiotics in fermented sausages. Meat Sci. 80:75–78. Diez-Valcarce, M., and K. Kovac. 2010. High hydrostatic pressure as emergent technology for the elimination of foodborne viruses: use of HHP in food industry. Trends Food Sci. Technol. 21:558–568. Dolezalova, M., Z. Molatova, F. Bunka, P. Brezina, and M. Marounek. 2010. Effect of organic acids on growth of chilled chicken skin microflora. J. Food Saf. 30:353–365. Dominguez, S. A., and D. W. Schaffner. 2007. Development and validation of a mathematical model to describe the growth of Pseudomonas spp. in raw poultry stored under aerobic conditions. Int. J. Food Microbiol. 120:287–295.

66. Doulgeraki, A. I., S. Paramithiotis, D. M. Kagkli, and G.-J. E. Nychas. 2010. Lactic acid bacteria population dynamics during minced beef storage under aerobic or modified atmosphere packaging conditions. Food Microbiol. 27:1028–1034. 67. Doyle, M. P., and M. C. Erickson. 2006. Emerging microbiological food safety issues related to meat. Meat Sci. 74:98–112. 68. Drosinos, E. H., and G.-J. E. Nychas. 1997. Production of acetic acid in relation to the content of glucose during modified atmosphere storage of gilt-head seabream (Sparus aurata) at 0 ± 1°C. Food Res. Int. 30:711–717. 69. Drosinos, E. H., and R. G. Board. 1994. Metabolic activities of pseudomonads in batch cultures in extract of minced lamb. J. Appl. Bacteriol. 77:613–620. 70. Drosinos, E. H., and R. G. Board. 1995. Attributes of microbial associations of meat growing as xenic batch cultures in a meat juice at 4°C. Int. J. Food Microbiol. 26:279–293. 71. Eblen, D. R. 2007. Public health importance of non-O157 Shiga toxin- producing Escherichia coli (non-O157 STEC) in the US food supply. U.S. Department of Agriculture, Food Safety and Inspection Service. Available from http:// www.fsis.usda.gov/PDF/STEC_101207.pdf. Accessed 24 February 2011. 72. Edwards, R. A., R. H. Dainty, and C. M. Hibbard. 1987. Volatile compounds produced by meat pseudomonads and related reference strains during growth in air at chill temperatures. J. Appl. Bacteriol. 62:403–412. 73. Ellebroek, L., J. A. Lienau, T. Alter, and D. Schlichting. 2007. Effectiveness of different chemical decontamination methods on the Campylobacter load of poultry carcasses. Fleischwirtschaft 87:224–227. 74. Ellerbroek, L. I., J. A. Lienau, and G. Klein. 2010. Campylobacter spp. in broiler flocks at the farm level and the potential for cross-contamination during slaughter. Zoonoses Public Health 57:E81–E88. 75. Eltholth, M. M., V. R. Marsh, S. Van Winden and F. J. Guitian. 2009. Contamination of food products with Mycobacterium avium paratuberculosis: a systematic review. J. Appl. Microbiol. 107:1061–1071. 76. Emborg, J., B. G. Laursen, and P. Dalgaard. 2005. Significant histamine formation in tuna (Thunnus albacares) at 2°C effect of vacuum- and modified atmosphere-packaging on psychrotolerant bacteria. Int. J. Food Microbiol. 101:263–279. 77. Emirog˘lu, Z. K., G. P. Yemis¸, B. K. Cos¸kun, and K. Candog˘an. 2010. Antimicrobial activity of soy edible films incorporated with thyme and oregano essential oils on fresh ground beef patties. Meat Sci. 86:283–288. 78. Ercolini, D., F. Russo, E. Torrieri, P. Masi, and F. Villani. 2006. Changes in the spoilage-related microbiota of beef during refrigerated storage under different packaging conditions. Appl. Environ. Microbiol. 72:4663–4671. 79. Ercolini, D., F. Russo, A. Nasi, P. Ferranti, and F. Villani. 2009. Mesophilic and psychrotrophic bacteria from meat and their spoilage potential in vitro and in beef. Appl. Environ. Microbiol. 75:1990–2001.

6.  Meat, Poultry, and Seafood 80. Ercolini, D., I. Ferrocino, A. La Storia, G. Mauriello, S. Gigli, P. Masi, and F. Villani. 2010. Development of spoilage microbiota in beef stored in nisin activated packaging. Food Microbiol. 27:137–143. 81. Erickson, M. C., and M. P. Doyle. 2007. Food as a vehicle for transmission of Shiga toxin–producing Escherichia coli. J. Food Prot. 70:2426–2449. 81a. The European Commission. Food Hygiene - Community Directives. http://ec.europa.eu/food/food/biosafety/hygienelegislation/directives_en.print.htm. Accessed 11 September 2012. 82. European Food Safety Authority. 2010. Analysis of the baseline survey of the prevalence of Campylobacter in broiler batches and of Campylobacter and Salmonella on broiler carcasses in the EU, 2008. Part A: Campylobacter and Salmonella prevalence estimates. EFSA J. 8(3):1503. 83. European Food Safety Authority. 2010. Decontamina­ tion of carcasses. Available at: http://www.efsa.europa. eu/en/ahawtopics/topic/decontamination.htm. Accessed November 2010. 84. Evans, S., and A. R. Sayers. 2000. A longitudinal study of Campylobacter infection of broiler flocks in Great Britain. Prevent. Vet. Med. 46:209–223. 85. Fabrizio, K. A., R. R. Sharma, A. Demirci, and C. N. Cutter. 2002. Comparison of electrolyzed oxidizing water with various antimicrobial interventions to reduce Salmonella species on poultry. Poult. Sci. 81:1598–1605. 86. Farkas, J., and C. Mohacsi-Farkas. 2011. History and future of food irradiation. Trends Food Sci. Technol. 22:121–126. 87. Feller, M. K., K. Huwiler, R. Stephan, E. Altpeter, A. Shang, H. Furrer, E. Pfyffer, T. Jemmi, A. Baumgartner, and M. Egger. 2007. Mycobacterium avium subspecies paratuberculosis and Crohn’s disease: a systematic review and meta-analysis. Lancet Infect. Dis. 7:617–613. 88. Fletcher, D. L. 2002. Poultry meat quality. Worlds Poult. Sci. 58:131–145. 89. Flick, G. J., Jr. 2007. Pathogenic vibrios in shellfish. Global Aquaculture Advocate 10(6):46, 48. 90. Flick, G. J., Jr. 2008. Microbiological safety of farmed fish. Global Aquaculture Advocate 11(2):33–34. 91. Flick, G. J., Jr. 2008. Solutions to microbiological challenges in crawfish production. Global Aquaculture Advocate 11(4):30, 32–33. 92. Flick, G. J., Jr. 2010. Post-harvest quality of freshwater prawns. Part I. Shelf-life. Global Aquaculture Advocate 13(6):51–52. 93. Flick, G. J., Jr. 2011. Post-harvest quality of freshwater prawns. Part II. Microbial composition, food safety. Global Aquaculture Advocate 14(1):61–62. 94. Flick, G. J., Jr. 2011. Post-harvest quality of freshwater prawns. Part III. Microflora, composition, frozen storage. Global Aquaculture Advocate 14(2):44–46. 95. Food and Agriculture Organization. 2010. Database Faostat. Available at: http://faostat.fao.org/. Accessed December 2010.

159 96. Food Safety and Inspection Service, U.S. Department of Agriculture. 1996. Pathogen reduction; hazard analysis and critical control point (HACCP) systems: final rule, 9CFR Part 304, et al. Fed. Regist. 61:38805–38989. 97. Food Safety and Inspection Service, U.S. Department of Agriculture. 1999. Irradiation of meat and meat products. Fed. Regist. 4:9089–9105. 98. Food Safety and Inspection Service, U.S. Department of Agriculture. 1999. FSIS policy on non-intact raw beef products contaminated with E. coli O157:H7. Available at: http://www.fsis.usda.gov/OA/background/ O157policy.htm. Accessed 31 March 2010. 99. Food Safety and Inspection Service, U.S. Department of Agriculture. 2002. Comparative risk assessment for intact (non-tenderized) and non-intact (tenderized) beef: technical report. Available at: http://www.fsis.usda.gov/ PDF/Beef_Risk_Assess_Report_Mar2002.pdf. Accessed 31 March 2010. 100. Food Safety and Inspection Service, U.S. Department of Agriculture. 2002. Comparative risk assessment for intact (non-tenderized) and non-intact (tenderized) beef: executive summary. Available at: http://www.fsis.usda. gov/PDF/Beef_Risk_Assess_ExecSumm_Mar2002.pdf. Accessed 31 March 2010. 101. Food Safety and Inspection Service, U.S. Department of Agriculture. 2003. Control of Listeria monocytogenes in ready-to-eat meat and poultry products; final rule. Fed. Regist. 68:34208–34254. 102. Food Safety and Inspection Service, U.S. Department of Agriculture. 2005. HACCP plan reassessment for mechanically tenderized beef products. Fed. Regist. 70:30331–30334. 103. Food Safety and Inspection Service, U.S. Department of Agriculture. 2006. Compliance guidelines to control Listeria monocytogenes in post-lethality exposed readyto-eat meat and poultry products. Available at: http:// www.fsis.usda.gov/oppde/rdad/FRPubs/97-013F/LM_ Rule_Compliance_Guidelines_May_2006.pdf. Accessed 31 March 2010. 104. Fratamico, P. M., A. A. Bhagwat, L. Injaian, and P. J. Fedorka-Cray. 2008. Characterization of shiga toxin– producing Escherichia coli strains isolated from swine feces. Foodborne Pathog. Dis. 5:827–838. 105. Gale, P., T. Drew, L. P. Phipps, G. David, and M. Wooldridge. 2009. The effect of climate change on the occurrence and prevalence of livestock diseases in Great Britain: a review. J. Appl. Microbiol. 106:1409–1423. 106. Gariga, M., and T. Aymerich. 2009. Advanced decontamination technologies: high hydrostatic pressure on meat products, p. 183–208. In F. Toldra (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 107. Geesey, G. G. 2001. Bacterial behavior at surfaces. Curr. Microbiol. 4:296–300. 108. Geornaras, I., and J. N. Sofos. 2005. Combining physical and chemical decontamination interventions for meat, p. 433– 460. In J. N. Sofos (ed.), Improving the Safety of Fresh Meat. CRC/Woodhead Publishing Limited, Cambridge, United Kingdom.

160 109. Geornaras, I., G. A. Dykes and A. von Holy. 1995. Biogenic amine formation by poultry-associated spoilage and pathogenic bacteria. Lett. Appl. Microbiol. 21:164–166. 110. Ghaly, A. E., D. Dave, S. Budge, and M. S. Brooks. 2010. Fish spoilage mechanisms and preservation techniques: review. Am. J. Appl. Sci. 7:859–877. 111. Gil, L., J. M. Barat, D. Baigts, R. Martínez-Máñez, J. Soto, E. Garcia-Breijo, M. C. Aristoy, F. Toldrá, and E. Llobet. 2011. Monitoring of physical–chemical and microbiological changes in fresh pork meat under cold storage by means of a potentiometric electronic tongue. Food Chem. 126:1261–1268. 112. Gill, A., and C. Gill. 2010. Packaging and the shelf life of fresh red and poultry meats, p. 259–280. In L. Robertson (ed.), Food Packaging and Shelf Life: A Practical Guide. CRC Press, Boca Raton, FL. 113. Gill, A., and C. Gill. 2010. Non-O157 verotoxigenic Escherichia coli and beef: a Canadian perspective. Can. J. Vet. Res. 74:161–169. 114. Gill, C. O. 1982. Microbial interaction with meats, p. 225–264. In M. H. Brown (ed.), Meat Microbiology. Applied Science Publishers, Inc., New York, NY. 115. Gill, C. O. 1983. Meat spoilage and evaluation of the potential storage life of fresh meat. J. Food Prot. 46:444–452. 116. Gill, C. O. 1986. The control of microbial spoilage in fresh meats, p. 49–88. In A. M. Pearson and T. R. Dutson (ed.), Advances in Meat Research, vol. 2. Meat and Poultry Microbiology. AVI Publishing Co., Westport, CT. 117. Gill, C. O. 2005. Sources of microbial contamination at slaughtering plants, p. 231–243. In J. N. Sofos (ed.), Improving the Safety of Fresh Meat. CRC/Woodhead Publishing Limited, Cambridge, United Kingdom. 118. Gill, C. O., and K. G. Newton. 1979. Spoilage of vacuum-packaged dark, firm, dry meat at chill temperatures. Appl. Environ. Microbiol. 37:362–364. 119. Gill, C. O., and K. G. Newton. 1978. The ecology of bacterial spoilage of fresh meat at chill temperatures. Meat Sci. 2:207–217. 120. Gill, C. O., J. C. L. Harrison, and N. Penny. 1990. The storage life of chicken carcasses stored under carbon dioxide. Int. J. Food Microbiol. 11:151–158. 121. Gloria, M. B. A., M. A. Daeschel, C. Craven, and K. S. Hilderbrand, Jr. 1999. Histamine and other biogenic amines in Albacore tuna. J. Aquatic Food Product Technol. 8:55–69. 122. Gonzalez-Fandos, E., B. Herrera, and N. Maya, N. 2009. Efficacy of citric acid against Listeria monocytogenes attached to poultry skin during refrigerated storage. Int. J. Food Sci. Technol. 44:262–268. 123. Gould, G., P. Franken, P. Hammer, B. Mackey, and F. Shanahan. 2005. Mycobacterium avium subsp. paratuberculosis (MAP) and the food chain. Food Prot. Trends 25:268–297. 124. Goulter, R. M., I. R. Gentle, and G. A. Dykes. 2009. Issues in determining factors influencing bacterial attachment: a

Microbial Spoilage and Public Health Concerns review using the attachment of Escherichia coli to abiotic surfaces as an example. Lett. Appl. Microbiol. 49:1–7. 125. Gram, L., and H. H. Huss. 2000. Fresh and processed fish and shellfish, p. 472–506. In B. M. Lund, A. C. BairdParker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Foods. Aspen Publishers, Gaithersburg, MD. 126. Gram, L., and P. Dalgaard. 2002. Fish spoilage bacteria—problems and solutions. Curr. Opin. Biotechnol. 13:262–266. 127. Grant, M. A., C. Hedberg, R. Johnson, J. Harris, C. M. Logue, J. Meng, J. N. Sofos, and J. S. Dickson. 2011. Executive summary: the significance of non-O157 shiga toxin-producing Escherichia coli in food. Food Prot. Trends 31:33–45. 128. Greaser, M. L. 1986. Conversion of muscle to meat, p. 37–102. In P. J. Bechtel (ed.), Muscle as Food. Academic Press, New York, NY. 129. Gunderson, M. F., H. W. McFadden, and T. S. Kyle. 1954. Bacteriology of Commercial Poultry Processing, p. 59–63. Burgess Publishing, Minneapolis, MN. 130. Handley, J. A., I. Hanning, S. C. Ricke, M. G. Johnson, F. T. Jones, and R. O. Apple. 2010. Temperature and bacterial profile of post chill poultry carcasses stored in processing combo held at room temperature. J. Food Sci. 75:M515–M520. 131. Hanninen, M. L., P. Perko-Makela, A. Pitkala, and H. Rautelin. 2000. A three-year study of Campylobacter jejuni genotypes in humans with domestically acquired infections and in chicken samples from the Helsinki area. J. Clin. Microbiol. 38:1998–2000. 132. Head, M., J. R. Saunders, and R. W. Pickup. 1998. Microbial evolution, diversity and ecology: a decade of ribosomal RNA analysis of uncultivated microorganisms. Microb. Ecol. 35:1–21. 133. Herrero, A. M. 2008. Raman spectroscopy, a promising technique for quality assessment of meat and fish: a review. Food Chem. 107:1642–1651. 134. Hinton, A., J. A. Cason, and K. D. Ingram. 2004. Tracking spoilage bacteria in commercial poultry processing and refrigerated storage of poultry carcasses. Int. J. Food Microbiol. 91:155–165. 135. Hinton, A., J. K. Northcutt, D. P. Smith, M. T. Musgrove, and K. D. Ingram. 2007. Spoilage microflora of broiler carcasses washed with electrolyzed oxidizing or chlorinated water using an inside-outside bird washer. Poult. Sci. 86:123–127. 136. Holzapfel, W. H. 1998. The Gram-positive bacteria associated with meat and meat products, p. 35–84. In R. G. Board and A. R. Davies (ed.), The Microbiology of Meat and Poultry. Blackie Academic and Professional, London, United Kingdom. 137. Hoorfar, J., M. Wagner, K. Jordan, S. L. Boquin, and J. Skiby. 2011. Towards biotracing in food chains. Int. J. Food Microbiol. 145:S1–S4. 138. Huezo, R., D. P. Smith, J. K. Northcutt, and D. L. Fletcher. 2007. Effect of immersion or dry air chilling on broiler carcass moisture retention and breast fillet functionality. J. Appl. Poult. Res. 16:438–447.

6.  Meat, Poultry, and Seafood 139. Huffman, N. M. 2002. Current and future technologies for the decontamination of carcasses and fresh meat. Meat Sci. 62:285–294. 140. Hugas, M., and E. Tsagarida. 2008. Pros and cons of carcass decontamination: the role of the European Food Safety Authority. Meat Sci. 78:43–52. 141. Hugenholtz, P., and N. R. Pace. 1996. Identifying microbial diversity in the natural environment: a molecular phylogenetic approach. Trends Biotechnol. 14:90–97. 142. Hunt, M. C., R. A. Mancini, K. A. Haachmeister, D. H. Kropf, M. Merriman, G. Delduca, and G. Milliken. 2004. Carbon monoxide in modified atmosphere packaging affects color, shelf life, and microorganisms of beef steaks and ground beef. J. Food Sci. 69:FCT45–FCT52. 143. Huss, H. H. 1995. Assurance of Seafood Quality, FAO Fisheries Technical Paper, No. 334. United Nations Food and Agriculture Organization, Rome, Italy. 144. Hussein, H. S. 2007. Prevalence and pathogenicity of Shiga toxin-producing Escherichia coli in beef cattle and their products. J. Anim. Sci. 85(E. Suppl.):E63–E72. 145. Ingram, M., and B. Simonsen. 1980. Meats and meat products, p. 333–409. In J. H. Silliker, R. P. Elliot, A. C. Baird-Parker, F. L. Bryan, J. H. B. Christian, D. S. Clark, J. C. Olsen, Jr., and T. A. Roberts (ed.), Microbial Ecology of Foods, vol. 2. Food Commodities. Academic Press, New York, NY. 146. Institute of Food Technologists. 2006. Antimicrobial Resistance—Implications for the Food System. Institute of Food Technologists, Chicago, IL. 147. International Commission for Microbiological Specifications in Foods. 1996. Microorganisms in Foods 5: Characteristics of Microbial Pathogens. Blackie Academic & Professional, London, United Kingdom. 148. International Commission for Microbiological Specifications in Foods. 2005. Microorganisms in Foods 6. Microbial Ecology of Food Commodities. Blackie Academic and Professional, London, United Kingdom. 149. Investnow.lk.  2010.  Available  at:  http://investnow. lk/2010/06/21/global-poultry-trends-asia-leads-the-worldin-chicken-demand/1732/. Accessed December 2010. 150. Ismail, S. A. S., T. Deak, H. A. Abd El-Rahman, M. A. M. Yassien, and L. R. Beuchat. 2001. Effectiveness of immersion treatments with acids, trisodium phosphate, and herb decoctions in reducing populations of Yarrowia lipolytica and naturally occurring aerobic microorganisms on raw chicken. Int. J. Food Microbiol. 64:13–19. 151. James, C., E. O. Göksoy, J. E. L. Corry and S. J. James. 2000. Surface pasteurisation of poultry meat using steam at atmospheric pressure, J. Food Eng. 45:111–117. 152. James, C., S. J. James, N. Hannay, G. Purnell, C. BarbedoPinto, H. Yaman, M. Araujo, M. L. Gonzalez, J. Calvo, M. Howell, and J. E. L. Corry. 2007. Decontamination of poultry carcasses using steam or hot water in combination with rapid cooling, chilling or freezing of carcass surfaces, Int. J. Food Microbiol. 114:195–203. 153. Jeong, J. Y., and J. R. Claus. 2010. Color stability and reversion in carbon monoxide packaged ground beef. Meat Sci. 85:525–530.

161 154. Jeong, J. Y., and J. R. Claus. 2011. Color stability of ground beef packaged in a low carbon monoxide atmosphere or vacuum. Meat Sci. 87:1–6. 155. Jhung, M. A., A. D. Thompson, G. E. Killgore, W. E. Zukowski, G. Songer, M. Warny, S. Johnson, D. N. Gerding, L. C. McDonald, and B. M. Limbago. 2008. Toxinotype V Clostridium difficile in humans and food animals. Emerg. Infect. Dis. 14:1039–1045. 156. Jin, T., L. Liu, H. Zhang, and K. Hicks. 2009. Antimicrobial activity of nisin incorporated in pectin and polylactic acid composite films against Listeria monocytogenes. Int. J. Food Sci. Technol. 44:322–329. 157. Jöbstl, M., S. Heuberger, A. Indra, R. Nepf, J. Köfer, and M. Wagner. 2010. Clostridium difficile in raw products of animal origin. Int. J. Food Microbiol. 138:172–175. 158. Jones, J. M., G. C. Mead, N. M. Griffiths, and B. W. Adams. 1982. Influence of packaging on microbiological, chemical and sensory changes in chill-stored poultry portions. Br. Poult. Sci. 23:25–40. 159. Juneja, V. K., and J. N. Sofos. 2002. Control of Foodborne Microorganisms. Marcel Dekker, New York, NY. 160. Juneja, V. K., and J. N. Sofos. 2009. Pathogens and Toxins in Foods: Challenges and Interventions. ASM Press, Washington, DC. 161. Kakouri, A., E. Drosinos, and G.-J. E. Nychas. 1997. Storage of Mediterranean fresh fish (Boops boops, and Sparus aurata) under modified atmospheres or vacuum at 3 and 10°C, p. 171–178. In J. B. Luten, T. Borresen, and J. Oehlenschlager (ed.), Development in Food Science, vol. 38. Seafood From Producer to Consumer, Integrated Approach to Quality. Elsevier, Amsterdam. The Netherlands. 162. Kalac, P. 2006. Biologically active polyamines in beef, pork and meat products: a review. Meat Sci. 73:1–11. 163. Kalchayanand, N., B. Ray, and R. A. Field. 1993. Characteristics of psychrotrophic Clostridium laramie causing spoilage of vacuum-packaged refrigerated fresh and roasted beef. J. Food Prot. 56:13–17. 164. Kargiotou, C., E. Katsanidis, J. Rhoades, M. Kontominas, and K. Koutsoumanis. 2011. Efficacies of soy sauce and wine base marinades for controlling spoilage of raw beef. Food Microbiol. 28:158–163. 165. Karmali, M. A., V. Gannon, and J. M. Sargeant. 2010. Verocytotoxin-producing Escherichia coli (VTEC). Vet. Microbiol. 140:360–370. 166. Kemp, G. K., M. L. Aldrich, and A. L. Waldroup. 2000. Acidified sodium chlorite antimicrobial treatment of broiler carcasses. J. Food Prot. 63:1087–1092. 167. Kerry, J. P., M. N. O’Grady, and S. A. Hogan. 2006. Past, current and potential utilisation of active and intelligent packaging systems for meat and muscle-based products: a review. Meat Sci. 74:113–130. 168. Kouakou, P., H. Ghalfi, C. Dortu, P. Evrard, and P. Thonart. 2010. Combined use of bacteriocin-producing strains to control Listeria monocytogenes regrowth in raw pork meat. Int. J. Food Sci. Technol. 45:937–943.

162 169. Koutsoumanis, K. P., and J. N. Sofos. 2004. Microbial contamination of carcasses and cuts, p. 727–737. In W. K. Jensens (ed.), Encyclopedia of Meat Sciences. Elsevier Academic Press, Amsterdam, The Netherlands. 170. Koutsoumanis, K. P., K. Lambropoulou, and G.-J. E. Nychas. 1999. Biogenic and sensory changes associated with the microbial flora of Mediterranean gilt-head seabream (Sparus aurata) stored aerobically at 0, 8, and 15°C. J. Food Prot. 62:392–402. 171. Koutsoumanis, K. P., I. Geornaras, and J. N. Sofos. 2006. Microbiology of land muscle food, vol. 1, p. 52.1–52.43. In Y. H. Hui (ed.), Handbook of Food Science. CRC Press, Taylor & Francis Group, Boca Raton, FL. 172. Kovac, K., M. Diez-Valcarce, M. Hernandez, P. Raspor, and D. Rodrıguez-Lazaro. 2010. High hydrostatic pressure as emergent technology for the elimination of foodborne viruses. Trends Food Sci. Technol. 21:558–568. 173. Lalitha, K. V., and P. K. Surendran. 2006. Microbiological changes in farmed reared freshwater prawn (Macrobrachium rosenbergii de Man) in ice. Food Control 17:802–807. 174. Lee, E. J., and D. U. Ahn. 2009. Advanced decontamination technologies: irradiation, p. 209–228. In F. Toldra (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 175. Leistner, L., and G. Gould. 2002. Hurdle Technologies. Kluwer Academic/Plenum Publishers, New York, NY. 176. Leroi, F. 2010. Occurrence and role of lactic acid bacteria in seafood products. Food Microbiol. 27:698–709. 177. Leuschner, R. G., M. Heidel, and W. P. Hammes. 1998. Histamine and tyramine degradation by food fermenting microorganisms. Int. J. Food Microbiol. 39:1–10. 178. Lillard, H. S. 1980. Effect on broiler carcasses and water of treating chiller water with chlorine or chlorine dioxide. Poult. Sci. 59:1761–1766. 179. Lillard, H. S. 1989. Incidence and recovery of salmonellae and other bacteria from commercially processed poultry carcasses at selected pre-evisceration and postevisceration steps. J. Food Prot. 52:88–91. 180. Lillard, H. S. 1990. The impact of commercial processing procedures on the bacterial-contamination and cross-contamination of broiler carcasses. J. Food Prot. 53:202–204. 181. Linam, W. M., and M. A. Gerber. 2007. Changing epidemiology and prevention of Salmonella infections. Pediatr. Infect. Dis. J.  26:747–748. 182. Loretz, M., R. Stephan, and C. Zweifel. 2010. Antimicrobial activity of decontamination treatments for poultry carcasses: a literature survey. Food Control 21:791–804. 183. Loretz, M., R. Stephan, and C. Zweifel. 2011. Antibacterial activity of decontamination treatments for cattle hides and beef carcasses. Food Control 22:347–359. 184. Lucker, E. 2009. Methodologies for the detection of BSE risk material in meat and meat products, chapt. 19, p. 499–514. In F. Toldra (ed.), Safety of Meat and Processed Meat. Springer, New York, NY.

Microbial Spoilage and Public Health Concerns 185. Mastromatteo, M., M. Mastromatteo, A. Contea, and M. A. Del Nobile. 2010. Advances in controlled release devices for food packaging applications. Trends Food Sci. Technol. 21:591–598. 186. McArdle, R. A., B. Marcos, J. P. Kerry, and A. M. Mullen. 2011. Influence of HPP conditions on selected beef quality attributes and their stability during chilled storage. Meat Sci. 87:274–281. 187. McArdle, R., B. Marcos, J. P. Kerry, and A. Mullen. 2010. Monitoring the effects of high pressure processing and temperature on selected beef quality attributes. Meat Sci. 86:629–634. 188. McMeekin, T. A. 1982. Microbial spoilage of meats, p. 1–40. In R. Davies (ed.), Developments in Food Microbiology—1. Applied Science Publishers, London, United Kingdom. 189. McMillin, K. W. 2008. Where is MAP going? A review and future potential of modified atmosphere packaging for meat. Meat Sci. 80:43–65. 190. Mead, G. C. 1989. Hygiene problems and control of process contamination, p. 183–220. In G. C. Mead (ed), Processing of Poultry. Elsevier Applied Science, London, United Kingdom. 191. Mead, G. C. 2004. Microbiological quality of poultry meat: a review. Braz. J. Poult. Sci. 6:135–142. 192. Milillo, S. R., and S. C. Ricke. 2010. Synergistic reduction of Salmonella in a model raw chicken media using a combined thermal and acidified organic acid salt intervention treatment. J. Food Sci. 75:M121–M125. 193. Monaghan, T., T. Boswell, and Y. R. Mahida. 2008. Recent advances in Clostridium difficile-associated disease.http://gut.bmj.com/cgi/content/abstract/gut. 2007.128157v1. 194. Morgan, A. I., N. Goldberg, E. R. Radewonuk, and O. J. Scullen. 1996. Surface pasteurisation of raw poultry meat by steam. Lebensm. Wissensch. Technol. 29:447–451. 195. Mor-Mur, M., and J. Yuste. 2010. Emerging bacterial pathogens in meat and poultry: an overview. Food Bioprocess Technol. 3:24–35. 196. Morris, M., A. L. Brody, and L. Wicker. 2007. Nonthermal food processing/preservation technologies: a review with packaging implications. Packag. Technol. Sci. 20:275–286. 197. Moschonas, G., D. J. Bolton, J. J. Sheridan, and D. A. McDowell. 2009. Effect of storage temperature and inoculum level on the time of onset of “blown pack” spoilage. J. Appl. Microbiol. 108:532–539. 198. Moschonas, G., D. J. Bolton, J. J. Sheridan, and D. A. McDowell. 2009. Isolation and sources of “blown pack” spoilage clostridia in beef abattoirs. J. Appl. Microbiol. 107:616–624. 199. Moschonas, G., D. J. Bolton, J. J. Sheridan, D. A. McDowell. 2011. The effect of heat shrink treatment and storage temperature on the time of onset of “blown pack” spoilage. Meat Sci. 87:115–118. 200. Mutharia, L. M., M. D. Klassen, J. Fairles, S. Barbut, and C. O. Gill. 2010. Mycobacterium avium subsp. paratuberculosis in muscle, lymphatic and organ tissues

6.  Meat, Poultry, and Seafood from cows with advanced Johne’s disease. Int. J. Food Microbiol. 136:340–344. 201. Nam, K. C., K. Y. Ko, B. R. Min, H. Ismail, E. J. Lee, J. Cordray, and D. U. Ahn. 2006. Influence of rosemary– tocopherol/packaging combination on meat quality and the survival of pathogens in restructured irradiated pork loins. Meat Sci. 74:380–387. 202. National Advisory Committee on Microbiological Criteria for Foods. 2002. Final report: E. coli O157:H7 in blade tenderized beef. Available at http://www.fsis. usda.gov/OPHS/NACMCF/2002/rep_blade1.htm. 203. National Meat Association. 1999. Guidelines for Developing Good Manufacturing Practices (GMPs), Standard Operating Procedures (SOPs) and Environmental Sampling/Testing Recommendations (ESTRs) in Ready to Eat (RTE) Products. National Meat Association, Oakland, CA. 204. Newell, D. G., and C. Fearnley. 2003. Sources of Campylobacter colonization in broiler chickens. Appl. Environ. Microbiol. 69:4343–4351. 205. Newton, K. G., and C. O. Gill. 1980. The microbiology of DFD fresh meats: a review. Meat Sci. 5:223–232. 206. Nieminen, T. T., E. Vihavainen, A. Paloranta, J. Lehto, L. Paulin, P. Auvinen, M. Solismaa, and K. J. Björkroth. 2011. Characterization of psychrotrophic bacterial communities in modified atmosphere-packed meat with terminal restriction fragment length polymorphism. Int. J. Food Microbiol. 144:360–366. 207. Nordstrom, J. L., C. A. Kaysner, G. M. Blackstone, M. C. L. Vickery, J. C. Bowers, and A. De Paola. 2004. Effect of intertidal exposure on Vibrio parahaemolyticus levels in Pacific oysters. J. Food Prot. 67:2178–2182. 208. Nørrung, B., J. K. Andersen, and S. Buncic. 2009. Main concerns of pathogenic microorganisms in meat, p. 3– 29. In F. Toldra (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 209. Northcutt, J., D. Smith, K. D. Ingram, A. Hinton, and M. Musgrove. 2007. Recovery of bacteria from broiler carcasses after spray washing with acidified electrolyzed water or sodium hypochlorite solutions. Poult. Sci. 86:2239–2244. 210. Northcutt, J. K., D. P. Smith, M. T. Musgrove, K. D. Ingram, and A. Hinton. 2005. Microbiological impact of spray washing broiler carcasses using different chlorine concentrations and water temperatures. Poult. Sci. 84:1648–1652. 211. Norton, T., and D. W. Sun. 2008. Recent advances in the use of high pressure as an effective processing technique in the food industry. Food Bioprocess Technol. 1:2–34. 212. Nottingham, P. M. 1982. Microbiology of carcass meats, p. 13–65. In M. H. Brown (ed.), Meat Microbiology. Applied Science Publishers Ltd., New York, NY. 213. Nottingham, P. M., C. O. Gill, and K. G. Newton. 1981. Spoilage at fat surfaces of meat, p. 183–190. In T. A. Roberts, G. Hobbs, J. H. B. Christian, and N. Skovgaard (ed.), Psychrotrophic Microorganisms in Spoilage and Pathogenicity. Academic Press, New York, NY.

163 214. Ntzimani, A. G., E. K. Paleologos, I. N. Savvaidis, and M. G. Kontominas. 2008. Formation of biogenic amines and relation to microbial flora and sensory changes in smoked turkey breast fillets stored under various packaging conditions at 4 degrees C. Food Microbiol. 25:509–517. 215. Nychas, G.-J. E. 1994. Modified atmosphere packaging of meats, p. 417–436. In R. P. Singh and F. A. R. Oliveira (ed.), Minimal Processing of Foods and Process Optimization, An Interface. CRC Press, London, United Kingdom. 216. Nychas, G.-J. E., D. L. Marshall, and J. N. Sofos. 2007. Meat, poultry, and seafood, p. 105–140. In M. P. Doyle and L. R. Beuchat (ed.), Food Microbiology, Fundamentals and Frontiers, 3rd ed. ASM Press, Washington, DC. 217. Nychas, G.-J. E., E. H. Drosinos, and R. G. Board. 1998. Chemical changes in stored meat, p. 288–326. In R. G. Board and A. R. Davies (ed.), The Microbiology of Meat and Poultry. Blackie Academic and Professional, London, United Kingdom. 218. Nychas, G.-J. E., and P. Skandamis. 2005. Fresh meat spoilage and modified atmosphere packaging (MAP), p. 461–502. In J. N. Sofos (ed.), Improving the Safety of Fresh Meat. CRC/Woodhead Publishing Limited, Cambridge, United Kingdom. 219. Ogilvy, W. S., and J. C. Ayres. 1951. Post-mortem changes in stored meats. II. The effect of atmospheres containing carbon dioxide in prolonging the storage life of cut-up chicken. Food Technol. 5:97–102. 220. O’Grady, M. N., and J. P. Kerry. 2008. Smart packaging technologies and their application in conventional meat packaging systems, p. 425–451. In F. Toldra (ed.), Meat Biotechnology. Springer, New York, NY. 221. Olaoye, O. A, and O. A. Idowu. 2009. Features and functional properties of lactic acid bacteria used as biological preservatives of meat processing. J. Agric. Technol. 6:449–460. 222. Olsen, K. N., M. Lund, J. Skov, L. S. Christensen, and J. Hoorfar. 2009. Towards real-time monitoring of broiler flocks: detection of Campylobacter in air samples for continuous monitoring of Campylobacter colonization in broiler flocks. Appl. Environ. Microbiol. 75:2074–2078. 223. Omer, M. K., O. Alvseike, A. Holck, L. Axelsson, M. Prieto, E. Skjerve, and E. Heir. 2010. Application of high pressure processing to reduce verotoxigenic E. coli in two types of dry-fermented sausage. Meat Sci. 86:1005–1009. 224. Osés, S. M., K. Rantsiou, L. Cocolin, I. Jaime, and J. Rovira. 2010. Prevalence and quantification of Shigatoxin producing Escherichia coli along the lamb food chain by quantitative PCR. Int. J. Food Microbiol. 141: S163–S169. 225. Over, K. F., N. Hettiarachchy, M. G. Johnson, and B. Davis. 2009. Effect of organic acids and plant extracts on Escherichia coli O157:H7, Listeria monocytogenes, and Salmonella Typhimurium in broth ­culture model and chicken meat systems. J. Food Sci. 74: M515–M521.

164 226. Ozdemir, H., A. Gucukoglu, and A. Koluman. 2006. Acidified sodium chlorite, trisodium phosphate and populations of Campylobacter jejuni on chicken breast skin. J. Food Proc. Preserv. 30:608–615. 227. Panagou, E. Z., F. R. Mohareb, A. A. Argyri, C. M. Bessant, and G.-J. E. Nychas. 2010. A comparison of artificial neural networks and partial least squares modelling for the rapid detection of the microbial spoilage of beef fillets based on Fourier transform infrared spectral fingerprints. Food Microbiol. 27:1–9. 228. Paramithiotis, S., E. H. Drosinos, J. N. Sofos, and G.-J. E. Nychas. 2010. Fermentation: microbiology and biochemistry, p. 185–198. In F. Toldrá (ed.), Handbook of Meat Processing. Wiley-Blackwell, Ames, IA. 229. Paramithiotis, S., P. N. Skandamis, and G.-J. E. Nychas. 2009. Insights into fresh meat spoilage, p. 55–82. In F. Toldrá (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 230. Park, S. Y., C. L. Woodward, L. F. Kubena, D. J. Nisbet, S. G. Birkhold, and S. C. Ricke. 2008. Environmental dissemination of foodborne Salmonella in preharvest poultry production: reservoirs, critical factors, and research strategies. Crit. Rev. Environ. Sci. Technol. 38:73–111. 231. Pennacchia, C., D. Ercolini, and F. Villani. 2011. Spoilagerelated microbiota associated with chilled beef stored in air or vacuum pack. Food Microbiol. 28:84–93. 232. Petrak, T., Z. Kalod-era, P. Novkovic’, and L. Gumhalter Karolyi. 1999. Bacteriological comparison of parallel and counter flow water chilling of poultry meat. Meat Sci. 53:269–271. 233. Prendergast, D. M., K. M. Crowley, D. A. McDowell, and J. J. Sheridan. 2009. Survival of Escherichia coli O157:H7 and non-pathogenic E. coli on irradiated and non-irradiated beef surfaces. Meat Sci. 83:468–473. 234. Prinz, S., G. Vanoijen, W. Bessei, F. Ehinger, and A. M. L. Cienen. 2010. Effects of water bath stunning on the electroencephalograms and physical reflexes of broilers using a pulsed direct current. Poult. Sci. 89:1275–1284. 235. Purnell, G., K. Mattick, and T. Humphrey. 2004. The use of ‘hot wash’ treatments to reduce the number of pathogenic and spoilage bacteria on raw retail poultry. J. Food Eng. 62:29–36. 236. Raines, C. R., and M. C. Hunt. 2010. Headspace volume and percentage of carbon monoxide affects carboxymyoglobin layer development of modified atmosphere packaged beef steaks. J. Food Sci. 75:C62–C65. 237. Ramamoorthi, L., S. Toshkov, and M. S. Brewer. 2009. Effects of carbon monoxide-modified atmosphere packaging and irradiation on E. coli K12 survival and raw beef quality. Meat Sci. 83:358–365. 238. Rantsiou, K., and L. Cocolin. 2006. New developments in the study of the microbiota of naturally fermented sausages as determined by molecular methods: a review. Int. J. Food Microbiol. 108:255–267. 239. Rantsiou, K., and L. Cocolin. 2008. Fermented meat products, p. 91–118. In L. Cocolin and D. Ercolini

Microbial Spoilage and Public Health Concerns (ed.), Molecular Techniques in the Microbial Ecology of Fermented Foods. Springer, New York, NY. 240. Rasschaert, G., K. Houf, C. Godard, C. Wildemauwe, M. Pastuszczak-Frak, and L. De Zutter. 2008. Contamination of carcasses with Salmonella during poultry slaughter. J. Food Prot. 71:146–152. 241. Rendueles, E., M. K. Omer, O. Alvseike, C. Alonso-Calleja, R. Capita, and M. Prieto. 2010. Microbiological food safety assessment of high hydrostatic pressure processing: a review. LWT Food Sci. Technol. J. 43:1–10. 242. Rhoades, J. R., G. Duffy, and K. Koutsoumanis. 2009. Prevalence and concentration of verocytotoxigenic Escherichia coli, Salmonella enterica and Listeria monocytogenes in the beef production chain: a review. Food Microbiol. 26:357–376. 243. Riedel, C. T., L. Brondsted, H. Rosenquist, S. N. Haxgart, and B. B. Christensen. 2009. Chemical decontamination of Campylobacter jejuni on chicken skin and meat. J. Food Prot. 72:1173–1180. 244. Rivas-Cañedo, A., C. Juez-Ojeda, M. Nuñez, and E. Fernández-García. 2011. Volatile compounds in ground beef subjected to high pressure processing: a comparison of dynamic headspace and solid-phase microextraction. Food Chem. 124:1201–1207. 245. Rivas-Cañedo, A., E. Fernández-García, and M. Nuñez. 2009. Volatile compounds in fresh meats subjected to high pressure processing: effect of the packaging material. Meat Sci. 81:321–328. 246. Rodriguez, A., P. Pangloli, H. A. Richards, J. R. Mount, and F. A. Draughon. 2006. Prevalence of Salmonella in diverse environmental farm samples. J. Food Prot. 69:2576–2580. 247. Rulis, A. M. 21 February 2002. Agency response letter, re: GRAS notice no. GRN 000083. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. http://vm.cfsan.fda.gov/~rdb/opa-g083.html. Accessed 7 December 2007. 248. Rupnick, M. 2007. Is Clostridium difficile-associated infection a potentially zoonotic and foodborne disease? Clin. Microbiol. Infect. 13:457–459. 249. Russell, S. M., and S. P. Axtell. 2005. Monochloramine versus sodium hypochlorite as antimicrobial agents for reducing populations of bacteria on broiler chicken carcasses. J. Food Prot. 68:758–763. 250. Sakhare, P. Z., N. M. Sachindra, K. P. Yashoda, and D. N. Rao. 1999. Efficacy of intermittent decontamination treatments during processing in reducing the microbial load on broiler chicken carcass. Food Control 10:189–194. 251. Samelis J., and J. N. Sofos. 2003. Strategies to control stress-adapted pathogens, p. 303–351. In A. E. Yousef and V. K. Juneja (ed.), Microbial Stress Adaptation and Food Safety. CRC Press, Boca Raton, FL. 252. Sams, A. R. 2010. First processing: slaughter through chilling, p. 19–34. In C. M. Owens, C. Z. Alvarado, and A. R. Sams (ed.), Poultry meat processing. CRC Press, New York, NY.

6.  Meat, Poultry, and Seafood 253. Sanchez, M. X., W. M. Fluckey, M. M. Brashears, and S. R. McKee. 2002. Microbial profile and antibiotic susceptibility of Campylobacter spp and Salmonella spp. in broilers processed in air-chilled and immersion-chilled environments. J. Food Prot. 65:948–956. 254. Sandusky, C. L., G. S. Reynhout, and T. S. Jones. May 2006. Method of extending color life of modified atmosphere packaged fresh red meat using Labiatae plant extracts. U.S. patent 7,037,543. 255. Scallan, E., P. M. Griffin, F. J. Angulo, R. V. Tauxe, and R. M. Hoekstra. 2011. Foodborne illness acquired in the United States—unspecified agents. Emerg. Infect. Dis. 17:16–22. 256. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M.-A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. 257. Schirmer, B. C., and S. Langsrud. 2010. A dissolving CO2 headspace combined with organic acids prolongs the shelf-life of fresh pork. Meat Sci. 85:280–284. 258. Sebranek, J. 2008. The political saga of carbon monoxide packaging. Scientists speak. April 20, Marco Island Marriott Beach Resort Annual Convention & Expo. IARW-WFLO-IRTA. Available at http://www.WFLO. org. 259. Selby, W., P. Pavli, B. Crotty, T. Florin, G. RadfordSmith, P. Gibson, B. Mitchell, W. Connell, R. Read, M. Merrett, H. Ee, and D. Hetzel. 2007. Two year combination antibiotic therapy with clarithromycin, rifabutin, and colfazimine for Crohn’s disease. Gastroenterology 132:2313–2319. 260. Seyfert, M., R. A., Mancini, and M. C. Hunt. 2004. Internal premature browning in cooked ground beef patties packaged from high oxygen modified atmosphere packaging. J. Food Sci. 69:142–146. 261. Seyfert, M., R. A., Mancini, and M. C. Hunt. 2004. Internal premature browning in cooked steaks from enhanced beef round muscles packaged in high oxygen and ultra-low oxygen modified atmospheres. J. Food Sci. 69:721–725. 262. Shackell, G. H. 2008. Traceability in the meat industry— the farm to plate continuum. Int. J. Food Sci. Technol. 43:2134–2142. 263. Shackell, G. H., and K. G. Dodds. 2008. DNA-based traceability of meat, p. 61–88. In F. Toldra (ed.), Meat Biotechnology. Springer, New York, NY. 264. Sheldon, B. W., and A. L. Brown. 1986. Efficacy of ozone as a disinfectant for poultry carcasses and chill water. J. Food Sci. 51:305–309. 265. Simoes, M., L. C. Simoes, and M. J. Vieira. 2010. A review of current and emergent biofilm control strategies. LWT Food Sci. Techonol. 43:573–583. 266. Simpson, R., and S. Acevedo. 2009. Mass transfer of CO2 in MAP systems: advances for non-respiring foods. J. Food Eng. 92:233–239. 267. Sinhamahapatra, M., S. Biswas, A. K. Das, and D. Bhattacharyya. 2004. Comparative study of different

165 surface decontaminants on chicken quality. Br. Poult. Sci. 45:624–630. 268. Skandamis, P. N., G.-J. E. Nychas, and J. N. Sofos. 2010. Meat decontamination, p. 43–85. In F. Toldrá (ed.), Hand­ book of Meat Processing. Wiley-Blackwell, Ames, IA. 269. Smith, G. C., D. L. Pendell, K. E. Belk, J. D. Tatum, J. N. Sofos, and D. L. Morris. 2009. Opinions of those in cattle, swine, and sheep slaughtering and rendering sectors regarding aspects of the National Animal Identification System. Prof. Anim. Sci. 25:641–653. 270. Smith, G. C., J. D. Tatum, K. E. Belk, J. A. Scanga, T. Grandin, and J. N. Sofos. 2005. Traceability from a US perspective. Meat Sci. 71:174–193. 271. Smith, J. S., and S. Pillai. 2004. Irradiation and food safety. Food Technol. 58(11):48–55. 272. Smith, K. E., C. Medus, S. D. Meyer, D. J. Boxrud, F. Leano, C. W. Hedberg, K. Elfering, C. Braymen, J. B. Bender, and R. N. Danila. 2008. Outbreaks of salmonellosis in Minnesota (1998 through 2006) associated with frozen, microwaveable, breaded, stuffed chicken products. J. Food Prot. 71:2153–2160. 273. Sofos, J. N. 1994. Microbial growth and its control in meat poultry and fish, p. 359–403. In A. M. Pearson and T. R. Dutson (ed.), Quality Attributes and Their Measurements in Meat, Poultry and Fish Products. Blackie Academic and Professional, Glasgow, United Kingdom. 274. Sofos, J. N. 2002. Approaches to pre-harvest food safety assurance, p. 23–48. In F. J. M. Smulders and J. D. Collins (ed.), Food Safety Assurance and Veterinary Public Health, vol. 1. Food Safety Assurance in the Pre-Harvest Phase. Wageningen Academic Publishers, Wageningen, The Netherlands. 275. Sofos, J. N. 2004. Pathogens in animal products: major biological hazards, p. 698–700. In W. Pond and A. Bell (ed.), Encyclopedia of Animal Science. Marcel Dekker, New York, NY. 276. Sofos, J. N. 2004. Pathogens in animal products: sources and control, p. 701–703. In W. Pond and A. Bell (ed.), Encyclopedia of Animal Science. Marcel Dekker, New York, NY. 277. Sofos, J. N. 2005. Improving the Safety of Fresh Meat. CRC/Woodhead Publishing, Limited, Cambridge, United Kingdom. 278. Sofos, J. N. 2006. Field data availability and needs for use in microbiological risk assessment, p. 57–74. In F. J. M. Smulders (ed.), Food Safety Assurance and Veterinary Public Health. vol. 4. Towards a Risk-based Chain Control. Wageningen Academic Publishers, Wageningen, The Netherlands. 279. Sofos, J. N. 2008. Challenges to meat safety in the 21st century. Meat Sci. 78:3–13. 280. Sofos, J. N. 2009. ASAS centennial paper: developments and future outlook for postslaughter food safety. J. Am. Sci. 87:2448–2457. 281. Sofos, J. N. 2009. Biofilms: our constant enemies. Food Safety Magazine 38(Feb-Mar):40–41. 282. Sofos, J. N. 2009. Preparing safe nonintact meat products—a review. Fleischwirtschaft Int. 24:68–73.

166 283. Sofos, J. N. 2010. Biofilms—importance and need for control. Fleischwirtschaft Int. 25:27–30. 284. Sofos, J. N., and G. C. Smith. 1998. Non-acid meat decontamination technologies: model studies and commercial applications. Int. J. Food Microbiol. 44:171–188. 285. Sofos, J. N., and I. Geornaras. 2010. Overview of current meat hygiene and safety risks and summary of recent studies on biofilms, and control of Escherichia coli O157:H7 in nonintact, and Listeria monocytogenes in ready-to-eat, meat products. Meat Sci. 86:2–14. 286. Sofos, J. N., I. Geornaras, K. E. Belk, and G. C. Smith. 2008. Nonintact whole muscle food safety: the problem and research needs. Proceedings of the 61st Reciprocal Meat Conference, American Meat Science Association, 22 to 25 June 2008, Gainesville, FL. American Meat Science Association, Champaign, IL. 287. Sofos, J. N., L. R. Beuchat, P. M. Davidson, and E. A. Johnson. 1998. Naturally Occurring Antimicrobials in Food. Council for Agricultural Science and Technology, Ames, IA. 288. Songer, J. G., H. T. Trinh, G. E. Killgore, A. D. Thompson, L. C. McDonald, and B. M. Limbago. 2009. Clostridium difficile in retail meat products, USA, 2007. Emerg. Infect. Dis. 15:819–821. 289. Stanbridge, L. H., and A. R. Davis. 1998. The microbiology of chill-stored meat, p. 174–219. In R. G. Board and A. R. Davies (ed.), The Microbiology of Meat and Poultry. Blackie Academic and Professional, London, United Kingdom. 290. Stern, N. J., M. R. S. Clavero, J. S. Bailey, N. A. Cox, and M. C. Robach. 1995. Campylobacter spp. in broilers on the farm and after transport. Poult. Sci. 74:937–941. 291. Stopforth, J. D., and J. N. Sofos. 2006. Recent advances in pre- and post-slaughter intervention strategies for control of meat contamination, p. 66–86. In V. K. Juneja, J. P. Cherry, and M. H. Tunick (ed.), Advances in Microbial Food Safety, ACS Symposium 931. Recent Advances in Intervention Strategies To Improve Food Safety. American Chemical Society, Oxford University Press, Washington, DC. 292. Stopforth, J. D., R. O’Connor, M. Lopes, W. E. Hill, and M. Samadpour. 2007. Validation of individual and multiple sequential interventions for reduction of microbial populations during processing of poultry carcasses and parts. J. Food Prot. 70:1393–1401. 293. Sutherland, J. 2003. Modelling food spoilage, p. 451– 474. In P. Zeuthen and L. Bogh-Sorensen (ed.), Food Preservations Techniques. CRC Woodhead Publishing Limited, Cambridge, United Kingdom. 294. Takahashi, H., B. Kimura, M. Yoshikawa, S. Gotou, I. Watanabe, and T. Fujii. 2004. Direct detection and identification of lactic acid bacteria in a food processing plant and in meat products using denaturing gradient gel electrophoresis. J. Food Prot. 67:2515–2520. 295. Talon, R., S. Leroy, and I. Lebert. 2007. Microbial ecosystems of traditional fermented meat products: the importance of indigenous starters. Meat Sci. 77:55–62.

Microbial Spoilage and Public Health Concerns 296. Tamblyn, K. C., D. E. Conner, and S. F. Bilgili. 1997. Utilization of the skin attachment model (SAM) to determine the antibacterial activity of potential carcass treatments. Poult. Sci. 76:1318–1323. 297. Thomas, C. J., and T. A. McMeekin. 1980. Contamination of broiler carcass skin during commercial processing procedures: an electron microscopic study. Appl. Environ. Microbiol. 19:133–144. 298. Thonart, P., P. Kouakou, D. I. Coulibaly, A. Yao, H. Ghalfi, M. Diop, A. Aguilar, F. Delvigne, and J. Destain. 2009. Lactic bacteria as protective starters. Bull. UASVM Agric. 66(2):478–481. 299. Timoney, J. F., H. L. Shivaprasad, R. C. Baker, and B. Rowe, 1989. Egg transmission after infection of hens with Salmonella enteritidis phage type 4. Vet. Rec. 125:600–601. 300. Toldra, F. (ed.). 2009. Safety of Meat and Processed Meat. Springer, New York, NY. 301. Toldra, F. (ed.). 2010. Handbook of Meat Processing. Wiley-Blackwell, Ames, IA. 302. Tompkin, R. B. 1986. Microbiology of ready-to-eat meat and poultry products, p. 89–121. In A. M. Pearson and T. R. Dutson (ed.), Advances in Meat Research, vol. 2. Meat and Poultry Microbiology. AVI Publishing Co., Westport, CT. 303. Tompkin, R. B. 2002. Control of Listeria monocytogenes in the food-processing environment. J. Food Prot. 65:709–725. 304. Tompkin, R. B., A. M. McNamara, and G. R. Acuff. 2001. Meat and poultry products, p. 463–471. In F. P. Downes and K. Ito (ed.), Compendium of the Methods for the Microbiological Examination of Foods. American Public Health Association, Washington, DC. 305. Tsigarida, E., and G.-J. E. Nychas. 2001. Ecophysiologi­ cal attributes of a Lactobacillus sp. and a Pseudomonas sp. on sterile beef fillets in relation to storage temperature and film permeability. J. Appl. Microbiol. 90:696–705. 306. Tsigarida, E., I. S. Boziaris, and G.-J. E. Nychas. 2003. Bacterial synergism or antagonism in a gel cassette system. Appl. Environ. Microbiol. 69:7204–7209. 307. Tsutsui, T., and F. Kasuga. 2009. Quantitative risk assessment of bovine spongiform encephalopathy, p. 615– 629. In F. Toldrá (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 307a. U.S. Food and Drug Administration. 2009. CPG Sec. 540.525. Decomposition and histamine raw, frozen tuna and mahi-mahi; canned tuna; and related species.­ http://www.fda.gov/ICECI/ComplianceManuals/ Comp­liancePolicyGuidanceManual/ucm074506.htm. Accessed 11 September 2012. 308. Uttaro, B., M. Badoni, S. Zawadski, and C. O. Gill. 2011. Effects of the pressure, flow rate and delivered volume of brine on the distributions of brine and bacteria in brine-injected meat. Food Control 22:180–185. 309. Valnegri, L., M. Franzoni, L. Vercllotti, and G. Soncini. 2010. Microbiological evaluation of the dry-slaughter process for small scale poultry meat production. Ital. J. Food Sci. 22:217–221.

6.  Meat, Poultry, and Seafood 310. Van Boekel, M. A. J. S. 2008. Kinetic modeling of food quality: a critical review. Comp. Rev. Food Sci. Safety 7:141–158. 311. Vanoy, R. W., M. L. Tamplin, and J. R. Schwarz. 1992. Ecology of Vibrio vulnificus in Galveston Bay oysters, suspended particulate matter, sediment and seawater: detection by monoclonal antibody-immunoassaymost probable number procedures. J. Ind. Microbiol. 9:219–223. 312. Vettorato, M. P., A. F. P. De Castro, M. C. CergoleNovella, F. L. L. Camargo, K. Irino, and B. E. C. Guth. 2009. Shiga toxin-producing Escherichia coli and atypical enteropathogenic Escherichia coli strains isolated from healthy sheep of different populations in São Paulo, Brazil. Lett. Appl. Microbiol. 49:53–59. 313. Viljoen, B. C., I. Geonaras, A. Lamprecht, and A. von Holy. 1998. Yeast populations associated with processed poultry. Food Microbiol. 15:133–117. 314. Villarreal, M. E., R. C. Baker, and J. M. Regenstein. 1990. The incidence of Salmonella on poultry carcasses following the use of slow release chlorine dioxide (Alcide). J. Food Prot. 53:464−467, 510. 315. Vinci, G. M., and M. L. Antonelli. 2002. Biogenic amines: quality index of freshness in red and white meat. Food Control 13:519–524. 316. Vizzier-Thaxton, Y., K. D. Christensen, M. W. Schilling, R. J. Buhr, and J. P. Thaxton. 2010. A new humane method of stunning broilers using low atmospheric pressure. J. Appl. Poult. Res. 19:341–348. 317. Volkova, V. V., R. H. Bailey, M. L. Rybolt, K. DazoGalarneau, S. A. Hubbard, D. Magee, J. A. Byrd, and R. W. Wills. 2010. Interrelationships of Salmonella status of flock and grow-out environment at sequential segments in broiler production and processing. Zoonoses Public Health 57:463–475. 318. Ward, H. J. T., and R. S. G. Knight. 2009. Transmissible spongiform encephalopathy and meat safety, p. 125–146. In F. Toldra (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 319. Weese, J. S., B. P. Avery, J. Rousseau, and R. J. ReidSmith. 2009. Detection and enumeration of Clostridium difficile spores in retail beef and pork. Appl. Environ. Microbiol. 75:5009–5011. 320. Weese, J. S., R. J. Reid-Smith, B. P. Avery, and J. Rousseau. 2010. Detection and characterization of Clostridium difficile in retail chicken. Lett. Appl. Microbiol. 50:362–365. 321. Weiss, J., M. Gibis, V. Schuh, and H. Salminen. 2010. Advances in ingredient and processing systems for meat and meat products. Meat Sci. 86:196–213.

167 322. White, D. G., S. Zhao, R. Sudler, S. Ayers, S. Friedman, S. Chen, P. F. McDermott, S. McDermott, D. D. Wagner, and J. Meng. 2001. The isolation of antibiotic-resistant Salmonella from retail ground meats. New Engl. J. Med. 345:1147–1154. 323. Whyte, P., K. McGill, and J. D. Collins. 2003. An assessment of steam pasteurization and hot water immersion treatments for the microbiological decontamination of broiler carcasses. Food Microbiol. 20:111–117. 324. Wilson, A. D., and M. Baietto. 2009. Applications and advances in electronic-nose technologies. Sensors 9:5099–5148. 325. Wilson, D. R., L. Dabrowski, S. Stringer, R. Moezelaar, and T. F. Brocklehurst. 2008. High pressure in combination with elevated temperature as a method for the sterilisation of food. Trends Food Sci. Technol. 19:289–299. 326. Xia, X., J. Meng, P. F. McDermott, S. Ayers, K. Blickenstaff, T. Tran, J. Abbott, J. Zheng, and S. Zhao. 2010. Presence and characterization of Shiga toxinproducing Escherichia coli and other potentially diarrheagenic E. coli strains in retail meats. Appl. Environ. Microbiol. 76:1709–1717. 327. Xiong, H., Y. B. Li, M. F. Slavik, and J. T. Walker. 1998. Spraying chicken skin with selected chemicals to reduce attached Salmonella Typhimurium. J. Food Prot. 61:272–275. 328. Yang, X., S. Balamurugan, and C. O. Gill. 2009. Substrate utilization by Clostridium estertheticum cultivated in meat juice medium. Int. J. Food Microbiol. 128:501–505. 329. Yang, Z. P., Y. B. Li, and M. Slavik. 1998. Use of antimicrobial spray applied with an inside-outside birdwasher to reduce bacterial contamination on prechilled chicken carcasses. J. Food Prot. 61:829–832. 330. Zhao, S., S. R. Young, E. Tong, J. W. Abbott, N. Womack, S. L. Friedman, and P. F. McDermott. 2010. Antimicrobial resistance of Campylobacter isolates from retail meat in the United States between 2002 and 2007. Appl. Environ. Microbiol. 76:7949–7956. 331. Zhou, G. H., X. L. Xu, and Y. Liu. 2010. Preservation technologies for fresh meat—a review. Meat Sci. 86:119–128. 332. Zhuang, H., E. M. Savage, D. P. Smith, and M. E. Berrang. 2009. Effect of dry-air chilling on sensory descriptive profiles of cooked broiler breast meat deboned four hours after the initiation of chilling. Poult. Sci. 88:1282–1291.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch7

Obianuju N. Nsofor Joseph F. Frank

Milk and Dairy Products

Being both highly perishable and nutritious, milk has, since prehistoric times, been subject to a variety of preservation treatments. Modern dairy processing utilizes pasteurization, heat sterilization, fermentation, dehydration, refrigeration, thermization, pulsed electric field, high pressure processing, bactofugation, biopreservation, and freezing as preservation treatments. The result, when combined with component separation processes (e.g., membrane filtration or coagulation), is an assortment of dairy foods having vastly different tastes and textures that may or may not be amenable to microbial growth. Some defects of milk and cheese caused by microorganisms are listed in Tables 7.1 and 7.2. In this chapter, the discussion of spoilage is based on the types of microorganisms associated with various defects. These include gram-negative psychrotrophic microorganisms, gram-positive bacteria including lactic acid bacteria and spore-forming bacteria, yeasts, and molds. The major objective of this chapter is to describe the interactions of these microorganisms with dairy foods that lead to commonly encountered product defects.

7

MILK AND DAIRY PRODUCTS AS GROWTH MEDIA

Milk

Milk is a good growth medium for many microorganisms because of its high water content, near-neutral pH, and available nutrients. Milk, however, is not an ideal growth medium for some microorganisms due to insufficient amounts of amino acids, for example, lysine, arginine, isoleucine, and glutamic acid. Milk supplementation with yeast extract or protein hydrolysates increases bacterial growth rates. Table 7.3 lists the major nutritional components of milk and their normal concentrations. These components consist of lactose, fat, protein, minerals, and various nonprotein nitrogenous compounds. Many micro­organisms cannot utilize lactose and therefore must rely on proteolysis or lipolysis to obtain carbon and energy. Microbial lipolysis occurs faster in milk stored at a temperature of 39°F than in milk stored at 45°F because during this cold storage, psychrotrophic bacteria have a greater growth rate and produce more lipases (135).

Obianuju N. Nsofor, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740. Joseph F. Frank, Department of Food Science and Technology, University of Georgia, Athens, GA 30602-7610.

169

Microbial Spoilage and Public Health Concerns

170

Table 7.1  Some defects of fluid milk that result from microbial growth Defect

Associated microorganisms

Type of enzyme

Bitter flavor

Psychrotrophic bacteria, Bacillus

Protease peptidase

Bitter peptides

86, 92

Rancidity flavor

Psychrotrophic bacteria

Lipase

Free fatty acids

114

Fruity flavor

Psychrotrophic bacteria

Esterase

Ethyl esters

99

Coagulation

Bacillus spp.

Protease

Casein destabilization

86

Sour flavor

Lactic acid bacteria

Glycolytic

Lactic, acetic acids

41, 113

Malty flavor

Lactic acid bacteria

Oxidase

3-Methyl butanal

89

Ropy texture

Lactic acid bacteria

Polymerase

Exopolysaccharides

16

On the other hand, proteolysis occurs faster in milk stored at 45°F than at 39°F. These proteolytic and lipolytic enzymes have major effects on the quality of milk and dairy products.

Carbon and Nitrogen Availability

Carbon sources in milk include lactose, protein, and fat. The citrate in milk can be utilized by many microorganisms but is not present in sufficient amount to support significant growth. A sufficient amount of glucose is present in milk to allow initiation of growth by some microorganisms, but the ability to metabolize lactose and proteins in milk varies widely among many microorganisms. Few spoilage microorganisms utilize milk fat as a carbon or energy source. This is because the milk fat globules are surrounded by a protective membrane composed of glycoproteins, lipoproteins, and phospholipids. Milk fat is available for microbial metabolism only if the globule membrane is physically damaged or enzymatically degraded (2). Caseins are present in the form of highly hydrated micelles and are readily susceptible to proteolysis. Whey proteins (b-lactoglobulin, a-lactalbumin, serum albumin, and immunoglobulins) remain soluble in the milk after precipi-

Metabolic product(s)

Reference(s)

tation of casein. They are less susceptible than caseins to microbial proteolysis. Milk contains nonprotein nitrogenous compounds such as urea, peptides, and amino acids that are readily available for microbial utilization (Table 7.3). These compounds are present in insufficient quantities to support the extensive growth required for spoilage.

Minerals and Micronutrients

Milk is a good source of certain B vitamins and minerals. The major mineral salt cations and anions present in milk are listed in Table 7.3. Although milk contains many trace mineral nutrients such as iron, cobalt, copper, and molybdenum, some of these, such as iron, may not be present in a readily usable form. Supplementation of milk with trace elements may be necessary to achieve maximum microbial growth rates. Milk also contains a growth stimulant, orotic acid (a metabolic precursor for pyrimidines).

Natural Inhibitors

The major microbial inhibitors in raw milk are lactoferrin and the lactoperoxidase system. Natural inhibitors of lesser importance include lysozyme, specific immunoglobulins, and folate and vitamin B12 binding

Table 7.2  Some defects of cheese that result from microbial growth Defect

Associated microoganisms

Metabolic product

Reference

Open texture, fissures

Heterofermentative lactobacilli

Carbon dioxide

68

Early gas

Coliforms, yeasts

Carbon dioxide, hydrogen

80

Late gas

Clostridium spp.

Carbon dioxide, hydrogen

29

Rancidity

Psychrotrophic bacteria

Free fatty acids

71

Fruity flavor

Lactic acid bacteria

Ethyl esters

9

White crystalline surface deposits

Lactobacillus spp.

Excessive D-lactate

101

Pink discoloration

Lactobacillus delbrueckii subsp. bulgaricus

High redox potential

109

7.  Milk and Dairy Products

171

Table 7.3  Approximate concentrations of some nutritional

components of milka Component

Amtb

Water

87.3

Lactose

4.8

Fat

3.7

Casein

2.6

Whey protein

0.6

Salt cations   Sodium

58

  Potassium

140

  Calcium

118

  Magnesium

12

Salt anions   Citrate

176

  Chloride

104

  Phosphorus

74

NPN

c

  Total NPN

296

  Urea N

142

  Peptide N

32

  Amino acid N

44

  Creatine N

25

Adapted from reference 58. The amounts for the first five components are in grams per 100 g, those for the salt cations and anions are in milligrams per 100 g, and those for the nitrogen are in milligrams per liter. c NPN, nonprotein nitrogen. a

b

systems. Lactoferrin, a glycoprotein, acts as an antimicrobial agent by binding iron. Human milk contains over 2 mg/ml lactoferrin, but it is of lesser importance in cow milk, which contains only 20 to 200 μg/ml (81). Psychrotrophic aerobes are inhibited by lactoferrin, but the presence of citrate in cow’s milk limits its effectiveness, as the citrate competes with lactoferrin for binding the iron (8). The most effective microbial inhibitor in cow’s milk is the lactoperoxidase system. Lactoperoxidase is heat stable and catalyzes the oxidation of thiocyanate and simultaneous reduction of hydrogen peroxide, resulting in the accumulation of a highly reactive oxidant, hypothiocyanite (OSCN–), which is responsible for the a­ntibacterial activity (134). The heat stability of lactoperoxidase is reduced under acidic conditions (pH, £5.3) possibly due to an increase in calcium ion concentra-

tion. Two mechanisms for this reaction are illustrated in Fig. 7.1. Hypothiocyanite oxidizes sulfhydryl groups of proteins, resulting in enzyme inactivation and structural damage to the microbial cytoplasmic membrane (132). Lactoperoxidase and thiocyanate are present in milk during synthesis, whereas hydrogen peroxide is formed in milk when oxygen is metabolized by lactic acid bacteria. Hydrogen peroxide is the limiting substrate of the reaction, so the effective use of this inhibitor system for preserving milk relies on adding a hydrogen peroxidegenerating system such as sodium percarbonate or, less effectively, hydrogen peroxide to the milk (31). The level of thiocyanate in milk is low and highly variable; therefore, it is also recommended that it should be added in milk for the lactoperoxidase system to be highly effective. Lactic acid bacteria, coliforms, and various pathogens are inhibited by this system (132), provided that hydrogen peroxide is in sufficient amount. As a result, indigenous lactoperoxidase may not contribute significantly to pathogen control in fresh raw milk just by itself. In developing countries, the lactoperoxidase system with added hydrogen peroxide and thiocyanate has been utilized to extend the shelf life of raw milk when there is a lack of refrigeration during on-farm storage and transportation to the processing facility.

Effect of Heat Treatments

Pasteurization of milk was introduced as a public health measure in order to eliminate or reduce microbial pathogens or spoilage microorganisms. The minimum required heat treatment during pasteurization of milk to be sold for fluid consumption in the United States is 145°F (63°C) for 30 minutes or 161°F (72°C) for 15 s, though most processors use higher temperatures and l­onger holding times. Normal milk pasteurization at 72°C for 15 s does not inactivate lactoperoxidase in milk and therefore does not destroy its ability to catalyze the

Figure 7.1  Two mechanisms for generation of hypothiocyanite (OSCN–) inhibitor in milk. Adapted from reference 132. doi:10.1128/9781555818463.ch7f1

Microbial Spoilage and Public Health Concerns

172 reaction between hydrogen peroxide and thiocyanate (77). Complete inactivation of lactoperoxidase occurs when cow milk is heated at 80°C for 2.5 s or 78°C for 15 s. Partial inactivation of lactoperoxidase may occur during short-time pasteurization at 74°C (110).

Dairy Products

Dairy products provide substantially different growth environments from those of fluid milk because these products have had nutrients removed or concentrated or have lower pH or water activity (aw). The composition, pH, and aw of selected dairy products are presented in Table 7.4. Butter is a water-in-oil emulsion, and microorganisms can be trapped within serum droplets if the micelles are sufficiently large. If butter is salted, the mean salt content of the water droplets will be 6 to 8%, which is sufficient to inhibit gram-negative spoilage organisms that could grow during refrigeration. However, individual droplets will have significantly higher or lower salt content if the salt is not uniformly distributed during manufacture. This can result in psychrotrophic bacteria growing in droplets of low salt content. Unsalted butter is usually made from acidified or fermented cream and relies on low pH and refrigeration for preservation.

PUBLIC HEALTH SIGNIFICANCE OF MILK AND DAIRY PRODUCTS As much as milk and dairy products are important components of a healthy diet, the presence of pathogenic microorganisms poses a potential health hazard to the consumers. Consumption of raw or inadequately pasteurized milk and milk products has played significant roles in foodborne outbreaks including salmonellosis, listeriosis, hemolytic uremic syndrome associated with Escherichia coli strain O157:H7, staphylococcal enterotoxin poisoning, campylobacteriosis, brucellosis,

yersiniosis, and tuberculosis (73). Human pathogens such as Listeria monocytogenes, Salmonella species, and Escherichia coli serotype O157:H7 have been found in raw milk and milk products. Although the prevalence of these pathogenic microorganisms might be low in raw milk compared to other microorganisms, the illnesses associated with these pathogens are severe. For ex­ample, invasive listeriosis typically has high case fatality rates for susceptible populations (60, 61, 96). L. monocytogenes can persist in food-processing environments under poor sanitary conditions and is also capable of growing in foods stored at refrigeration temperatures, thus making it a significant threat to public health. Salmonella enterica is one of the most frequently linked Salmonella species to foodborne illnesses. Antimicrobial resistance of Salmonella serotypes has become a concern. Outbreaks due to m­ultidrugresistant Salmonella enterica serotype Newport have been associated with Mexican-style soft cheese made from unpasteurized milk (15). Consumption of unpasteurized dairy products is linked to enterohemorrhagic E. coli strain O157:H7 outbreaks. The presence of these pathogens in milk and milk products has resulted in product recalls by regulatory agencies in various countries. These pathogenic microorganisms pose a great concern to public health agencies and the food industry because the infectious dose can be low in some cases. Furthermore, improper handling of milk or milk products, especially those manufactured from raw milk, may lead to bacterial survival and/or growth, creating a potential health risk to consumers of milk and milk products (28, 57, 61, 126). U.S. food regulations require that these pathogens be absent in food, and the Food and Drug Administration mandates that all fluid milk and milk products moved in interstate commerce must be pasteurized. Dairy manufacturing plants should empha-

Table 7.4  Approximate composition, pH, and aw of selected dairy productsa Component (g/100 g) Product

Water

Fat

Butter

16.0

81.0

3.6

0.06

Cheddar cheese

37.0

32.8

24.9

1.3

Swiss cheese

37.2

27.4

28.4

3.4

Nonfat dry milk

Protein

Carbohydrate

aw

6.3 0.90–0.95

3.2

0.8

36.2

52.0

0.2

79.4

0.2

7.5

11.3

0.93–0.98

Yogurt

89

1.7

3.5

5.1

Compiled from references 5, 7, 23, and 62.

5.2 5.6

Evaporated skim milk

a

pH

4.3

7.  Milk and Dairy Products

173

size the prevention of products from being contaminated with these pathogens by having a detailed food safety plan. It has been shown that an inverse relationship exists between preventive measures and the incidence of foodborne illnesses. Consumption of pasteurized milk and milk products remains a key factor in the prevention of foodborne illnesses.

initial Pseudomonas population. These strains also exhibited 1,000-fold-greater proteolytic activity and 280fold-greater lipolytic activity than the initial Pseudomonas population. Psychrotrophic bacteria commonly found in raw milk are inactivated by pasteurization, but spores from aerobic bacteria such as Bacillus cereus could survive pasteurization temperatures and potentially pose a health risk.

PSYCHROTROPHIC SPOILAGE

Sources of Psychrotrophic Bacteria in Milk

Preservation of fluid milk relies on effective sanitation, timely marketing, heat treatment, and refrigeration. Raw milk is rapidly cooled after collection and is kept cold until pasteurized, after which it is kept cold until consumption. There is often sufficient time between milk collection and consumption for psychrotrophic bacteria to grow. Flavor defects can result from this growth. Pasteurized milk is expected to have a shelf life of 16 to 22 days, so contamination of the contents of a container with even one rapidly growing psychrotrophic micro­organism can lead to spoilage. Growth of psychrotrophic bacteria in raw milk can lead to defects in products made from that milk because of residual enzyme activity.

Psychrotrophic Bacteria in Milk

Psychrotrophic bacteria that spoil raw and pasteurized milk are primarily aerobic gram-negative rods in the family Pseudomonadaceae. It is typical that 65 to 70% of psychrotrophic isolates from raw milk are in the genus Pseudomonas (44). Although representatives of other genera, such as Aeromonas, Bacillus, Listeria, Staphylococcus, and Enterococcus, and of the family Enterobacteriaceae may be present in raw milk and increase in number during storage, they are usually outgrown by the gram-negative obligate aerobes when milk is held at a storage temperature of 3 to 7°C (67). An exception is when bacterial spores are also present in pasteurized milk. The majority of aerobic sporeforming bacteria commonly found in milk belong to the genus Bacillus. The psychrotrophic spoilage microflora of milk is generally proteolytic, with many isolates able to produce extracellular lipases, phospholipase, and other hydrolytic enzymes but unable to utilize lactose. The bacterium most often associated with flavor defects in refrigerated milk is Pseudomonas fluorescens (34), with Pseudomonas fragi, Pseudomonas putida, and Pseudomonas lundensis (119) also commonly encountered. Jaspe et al. (56) observed that incubating raw milk for 3 days at 7°C selected for a population of Pseudomonas spp. that had a 10-fold-higher growth rate at 7°C and a lower growth rate at 21°C than the

Soil, water, animals, and plant material constitute the natural habitat of psychrotrophic bacteria and spores found in milk (24). Plant materials, such as grass and hay used for animal feed, may contain over 108 psychrotrophs per gram (120). Levels of aerobic spores found in silage range from 10 to >105 CFU per gram. Psychrotrophic bacteria can be isolated in low numbers from water used on the dairy farms (121). Use of this water to clean and rinse milking equipment provides a direct means for their entry into milk. Psychrotrophic bacteria isolated from water are often active producers of extracellular enzymes and grow rapidly in refrigerated milk (24). Consequently, water is an important source of milk spoilage bacteria. The teat and udder area of the cow can harbor high levels of psychrotrophic bacteria, even after washing and sanitizing (90). These psychrotrophs probably originate from soil. Milking equipment, utensils, and storage tanks are the major source of psychrotrophic contamination of raw milk (24). Proper cleaning and sanitizing procedures can effectively reduce contamination from these sources. Milk residues on unclean equipment provide a growth niche for psychrotrophic bacteria that enter milking machines, pipelines, and holding tanks with water rinses or milk. Milking equipment is generally fabricated from stainless steel; however, for some parts, rubber or other nonmetal materials must be used. Rubber materials are difficult to sanitize, since only moderate use results in the formation of microscopic cracks. Bacteria attached to these parts are difficult to inactivate by chemical s­anitization (91). Pasteurized milk products become contaminated with psychrotrophic bacteria by exposure to contaminated equipment or air. Schröder (108) determined that the filling equipment was most often the source of psychrotrophs in packaged milk. Although only low levels of psychrotrophic bacteria are found in air, only one viable cell per container is required to spoil the product. Aseptic packaging eliminates this low level of contamination and therefore is used to extend the shelf life of pasteurized milk.

174

Growth Characteristics and Defects

In milk, the generation times of the most rapidly growing psychrotrophic Pseudomonas spp. isolated from raw milk are 8 to 12 h at 3°C and 5.5 to 10.5 h at 3 to 5°C (117). These growth rates are sufficient to cause spoilage within 5 days if the milk initially contains only one cell/ml. However, most psychrotrophic pseudomonads present in raw milk grow much more slowly, causing refrigerated milk to spoil in 10 to 20 days. Defects of fluid milk associated with the growth of psychrotrophic bacteria are related to the production of extracellular enzymes. Sufficient enzyme to cause defects is usually present when the population of psychrotrophs reaches 106 to 107 CFU/ml (35). Bitter and putrid flavors and coagulation result from proteolysis. Rancid and fruity flavors result from lipolysis. The production of extracellular enzymes by psychrotrophic bacteria in raw milk also has implications for the quality of products produced from that milk.

Proteases Factors Affecting Protease Production

Pseudomonas fluorescens and other psychrotrophs found in milk generally produce extracellular proteases during the late exponential and stationary phases of growth (45, 105, 127), probably due to the release of preformed enzyme from the cells (45). The temperature for optimum production of protease by psychrotrophic Pseudomonas spp. is lower than the temperature for optimum rate of growth (83). Relatively high amounts of protease are produced at temperatures as low as 5°C (82). Raw milk held at 2°C exhibits little proteolysis after 10 days of storage (48). Since Pseudomonas spp. are obligate aerobes, it is expected that oxygen is required for protease synthesis. Myhara and Skura (94) reported optimum protease production for P. fragi in a medium containing 7.4 μg/ml dissolved oxygen, slightly less than the 8.4 μg/ml contained in saturated water. The effect of calcium and iron ions on protease production by Pseudomonas spp. is relevant to dairy spoilage. McKellar and Cholette (84) reported that in the absence of ionic calcium, an inactive precursor to proteinase was produced by P. fluorescens. This precursor could not be activated, indicating that calcium is required to stabilize the enzyme. Iron, which may be at a growth-limiting concentration in milk, represses protease production by Pseudomonas spp. when added to milk (36). Maximum protease production will occur only if iron is growth limiting, though this may be an indirect effect of reduced cytochrome synthesis and de-

Microbial Spoilage and Public Health Concerns creased energy levels (83). Pyoverdine, an iron-chelating pigment produced by P. fluorescens, stimulates protease production (85). Most evidence indicates that P. fluorescens regulates protease production as a means to provide carbon rather than amino acids to the cell for protein synthesis (35). Protease production is induced by various low-molecularweight protein degradation products and is subject to end product and catabolite repression. Asparagine is the most effective inducer, and citric acid is an effective inhibitor of synthesis (83). Protease production by P. fluorescens in raw milk is preceded by the depletion of glucose, galactose, lactate, glutamine, and glutamic acid. Supplementation of milk with these compounds delays or inhibits protease production (55).

Properties of Proteases from Psychrotrophic Pseudomonas

Fox et al. (39) summarized the properties of proteinases produced by P. fluorescens. The properties most relevant to dairy product spoilage include temperature optima from 30 to 45°C, with significant activity at 4°C. The pH optima are near neutrality or alkaline. All the proteinases are metalloenzymes containing either Zn2+ or Ca2+. Perhaps the most important technological characteristic of these enzymes is their heat stability. Decimal reduction times at 140°C range from 50 to 200 s, which is sufficient to allow enzymes to retain significant activity after ultrahigh temperature (UHT) milk processing (64). This heat stability is surprising, considering that the enzymes are produced and active at refrigeration temperatures. Another unexpected property of these enzymes is their susceptibility to autodegradation at 55 to 60°C, apparently due to an unfolding of the protein chain into a more sensitive conformation (115).

Protease-Induced Product Defects

Proteases of psychrotrophic bacteria cause product defects either at the time they are produced in the product or as a result of the enzymes surviving a heat process. Most investigators have observed that these proteases preferentially hydrolyze k-casein, although some show preference for b- or as-1-casein (25). Degradation of casein results in the liberation of bitter peptides. Bitterness is a common off flavor in pasteurized milk that has been subject to postpasteurization contamination with psychrotrophic bacteria. Continued proteolysis results in putrid off flavors associated with lower-molecularweight degradation products such as ammonia, amines, and sulfides. Bitterness in UHT (commercially sterile) milk develops when sufficient psychrotrophic bacterial growth occurs in raw milk (estimated at 105 to 107

7.  Milk and Dairy Products CFU/ml) to leave residual enzyme after heat treatment (92). Low-level protease activity in UHT milk can also result in coagulation or sediment formation. UHT milk appears to be more sensitive to protease-induced defects than raw milk, probably as a result of either heatinduced changes in casein micelle structure or heat i­nactivation of protease inhibitors (100). The effect of proteases of psychrotrophic origin on the quality of cheese and cultured products is minimal, because the combination of low pH and low storage temperature inhibits their activity (69). In addition, some proteases may be removed with the whey fraction during cheese manufacture. However, growth of proteolytic bacteria in raw milk lowers the cheese yield because proteolytic products of casein degradation are lost to the whey rather than becoming part of the cheese (133). Proteases can also alter the functional characteristics of milk powders (19).

Lipases

Psychrotrophic P. fluorescens isolates from milk often produce extracellular lipase in addition to protease. Other commonly found lipase-producing psychrotrophs include P. fragi and P. aeruginosa.

Factors Affecting Lipase Production

Lipases of psychrotrophic pseudomonads, like proteases, are produced in the late log or stationary phase of growth (1, 83). As with protease, optimal synthesis of lipase generally occurs below the optimum temperature for growth. For example, Andersson (3) reported optimum production of lipase at 8°C by a P. fluorescens isolate that exhibited optimum growth at 20°C. Milk is an excellent medium for lipase production by pseudomonads. Lipase production requires an organic nitrogen source; however, some amino acids repress lipase production, especially those that serve as nitrogen but not carbon sources (36). Ionic calcium is required for lipase activity. Lipase activity is inhibited by EDTA, which chelates the calcium, and is reversed by addition of Ca2+ (1). Pseudomonas fluore­ scens can produce lipase in calcium-free media, although the amount produced is less than in the presence of calcium (84). Lipase production by P. fluorescens, to a greater degree than protease production, is stimulated by limiting iron availability (85). Supplementation of milk with iron delays the onset of lipase production by P. fluorescens in raw milk (37). Although there is conflicting evidence, most reports indicate that lipase production is subject to catabolite repression (83, 114). Lipases are produced by P. fluorescens and P. fragi in the absence of triglycerides. The presence of

175 triglycerides can either inhibit or stimulate production of lipase, depending on the specific strain, triglyceride concentration, and growth conditions. Surface-active agents such as polysaccharides and lecithin often stimulate the release of lipase from the cell surface (114).

Properties of Lipase from Psychrotrophic Pseudomonas spp.

Properties of lipases produced by Pseudomonas spp. have been summarized by Stead (114) and Fox et al. (39). Temperatures for optimal activity of these lipases range from 22 to 70°C with most between 30 and 40°C. The optimal pH for activity is from 7.0 to 9.0, with most having optima between 7.5 and 8.5. P. fluorescens lipase in milk is active at refrigeration temperatures, and significant activity remains at subfreezing temperatures and at low aw (4). The heat stability of these lipases is similar to that of the proteases. Decimal reduction times at 140°C range from 48 to 437 seconds (64), which is sufficient to provide residual activity after UHT treatment. Most of these lipases are also subject to accelerated irreversible inactivation at temperatures of less than 100°C (65). Histidine and MgCl2 are required for the low-temperature inactivation of a heat-stable lipase isolated from P. fluorescen­s (22).

Lipase-Induced Product Defects

The triglycerides in raw milk are present in globules that are protected from enzymatic degradation by a membrane. Milk becomes susceptible to lipolysis if this membrane is disrupted by excessive shear force (from pumping, agitation, and freezing). Raw milk contains a mammalian lipase (milk lipase) that will rapidly act on the fat if the globule membrane is disrupted. Most cases of rancidity in milk are a result of this process, rather than from the growth of lipase-producing micro­organisms. Phospholipase C and protease produced by psychrotrophic bacteria can degrade the fat globule membrane, resulting in the enhancement of milk lipase activity (2, 21). Milk lipase is heat labile, so most milk products (other than raw-milk cheeses) will not have residual activity. Sufficient bacterial lipase can be produced in raw milk to cause defects in products manufactured from that milk. Since residual activities are usually low and the reaction environment less than optimum, usually only products with long storage times or high storage temperatures are affected. This includes UHT milk, some cheeses, butter, and whole-milk powder. Lipase-­producing bacteria can also recontaminate pasteurized milk and cream and grow in these ­products during refrigerated

176 storage. The rancid flavor and odor resulting from lipase action are usually from the liberation of C4 to C8 fatty acids. Fatty acids of higher molecular weight produce a flavor described as soapy. Low levels of unsaturated fatty acids liberated by enzymatic activity may be oxidized to ketones and aldehydes to produce an oxidized or “cardboardy” off flavor (30). P. fragi produces a fruity off flavor in milk by esterifying free fatty acids with ethanol (99). Ethyl butyrate and ethyl hexanoate are the esters formed at highest amounts, with low levels of ethyl esters of acetate, propionate, and isovalerate also produced. Low levels of ethanol in milk, present as a result of microbial activity, stimulate ester production. Residual activity from heatstable microbial lipases can cause off flavors in UHT milk, but lipase-induced defects are not as common as those resulting from microbial protease (92). Rancidity in butter may result from the growth of lipolytic microorganisms during storage, residual heatstable microbial lipase originating from the growth of psychrotrophic bacteria in the milk or cream, or native milk lipase activity in the raw milk or cream. When butter is manufactured from rancid cream, low-molecular-weight free fatty acids are removed with the watery portion of the cream (buttermilk), so the resulting butter will not have a typical rancid flavor but a less pronounced soapy off flavor associated with C10 to C12 free fatty acids. However, the typical odor of rancid butter is associated with lower-molecular-weight fatty acids (C4 to C8). Microbial lipases present in butter will exhibit activity even if the product is stored at −10°C (95). Growth of psychrotrophic bacteria in butter occurs only if the product is made from sweet rather than ripened (sour) cream. Sweet cream butter is preserved by salt and refrigeration. Butter is a water-in-fat emulsion, so moisture and salt may not equilibrate during storage. Consequently, if salt and moisture are not evenly distributed in the product during manufacture, then lipolytic psychrotrophs will have pockets of high aw in which to grow (30). Cheese is more susceptible to defects caused by bacterial lipases than to those caused by proteases because lipases, unlike most proteases, are concentrated along with the fat in the curd. The acidic environment of most cheeses limits lipase activity. Some cheeses, such as Camembert and Brie, increase in pH to near neutrality during ripening. Camembert is, in fact, susceptible to defects associated with microbial lipase (33). Moreacidic cheeses, e.g., Cheddar, are susceptible if cured for several months or if high amounts of lipase are present (70). Law et al. (71) reported that Cheddar cheese made from milk containing over 106 CFU/ml of P. fluorescens

Microbial Spoilage and Public Health Concerns before pasteurization developed a rancid flavor in 6 to 8 months. Whole-milk powder containing bacterial lipase may develop rancidity and low-fat-milk-derived powders may contain residual lipase, which becomes active when these products are used in fat-containing food formulations (19, 114).

Control of Product Defects Associated with Psychrotrophic Bacteria Raw Milk

Preventing product defects that result from growth of psychrotrophic bacteria in raw milk involves limiting contamination levels, rapid cooling immediately after milking, and maintenance of cold storage temperatures. Limiting populations of bacteria primarily involves cleaning, sanitizing, and drying cows’ teats and udders before milking and the use of cleaned and sanitized equipment. Removal of residual milk solids from milk contact surfaces is critical to psychrotroph control, since these residues protect cells from the action of chemical sanitizers and provide nutrients for growth. Subsequent growth over a period of days results in a biofilm that in addition to containing high numbers of bacteria is highly resistant to chemical sanitizers (40). Rapid cooling of milk after collection is important because contamination of the product with psychrotrophic bacteria is unavoidable. As previously indicated, psychrotrophic activity in milk is inhibited at 2°C, but freezing of milk causes disruption of the fat globule membrane, making it highly susceptible to lipolysis. Therefore, the challenge of farm storage systems is to rapidly cool milk to as low a temperature as possible while avoiding ice formation. Proteolysis in raw milk can be inhibited by addition of carbon dioxide, which hypothetically decreases microbial protease activity or decreases endogenous protease activity, resulting in a lower pH of milk (76).

Pasteurized Products

Preventing contamination of pasteurized dairy products with psychrotrophic bacteria is primarily a matter of equipment cleaning and sanitation, although airborne psychrotrophs may also limit product shelf life. Even when filling equipment is effectively cleaned and sanitized, it can still become a source of psychrotrophic microorganisms that accumulate during normal hours of continuous use (108). These microorganisms probably enter the filler through the vacuum system or from containers. Complete elimination of psychrotrophic microorganisms from products is best achieved by using aseptic packaging technologies.

7.  Milk and Dairy Products SPOILAGE BY FERMENTATIVE NONSPOREFORMERS Spoilage of milk and dairy products resulting from growth of acid-producing fermentative bacteria occurs when storage temperatures are sufficiently high for these microorganisms to outgrow psychrotrophic bacteria or when product composition is inhibitory to gramnegative aerobic organisms. For example, the presence of lactic acid in fluid milk is a good indication that the product was exposed to an unacceptably high storage temperature that allowed growth of lactic acid bacteria. Fermented dairy foods, though manufactured using lactic acid bacteria, can be spoiled by the growth of “wild” lactic acid bacteria that produce unwanted gas, off flavors, or appearance defects. Fluid milk, cheese, and c­ultured milks are the major dairy products susceptible to spoilage by non-spore-forming fermentative bacteria. Non-spore-forming bacteria responsible for fermentative spoilage of dairy products are mostly associated with lactic acid bacteria or coliform (lactose-fermentin­g) species within the family Enterobacteriaceae. Genera of lactic acid bacteria involved in spoilage of milk and fermented products include Lactococcus, Lactobacillus, Leuconostoc, Enterococcus, Pediococcus, and Strep­ tococcus (18). While Enterobacteriaceae can spoil milk, this is seldom a problem since they are usually outgrown by either the lactic acid or psychrotrophic bacteria. Coliform-induced spoilage is more common in some cheese varieties. Members of the Enterobacter and Klebsiella genera are often the Enterobacteriaceae associated with spoilage.

Sources of Spoilage Bacteria

Lactic acid-producing bacteria are normal inhabitants of the cow’s teat. Lactic acid bacteria are also associated with silage and other animal feeds and feces. Coliform bacteria are present on udder skin as a result of fecal contamination, so ineffective cleaning of this area before milking will contribute to high coliform populations in milk. Coliform bacteria in raw milk are also associated with inadequately cleaned milking equipment (10).

Defects in Fluid Milk Products

The most common fermentative defect in fluid milk products is souring caused by the growth of lactic acid bacteria. Lactic acid by itself has a clean, pleasant acid flavor and no odor. The unpleasant “sour” odor and taste of spoiled milk are the result of small amounts of acetic and propionic acids (113). Sour odor can be detected before a noticeable acid flavor develops. For a discussion of lactic acid production in milk, see chap-

177 ter 32. Other defects may occur in combination with acid production. A malty flavor results from growth of Lactococcus lactis subsp. lactis var. maltigenes. This strain is unique among lactococci in its ability to produce 2-methylpropanal, 3-methylbutanal, and the corresponding alcohols (89). The aldehydes are produced by decarboxylation of a-ketoisocaproic and a-ketoisovaleric acids. These keto acids are also concurrently used to synthesize leucine and valine by transamination with glutamic acid. Alcohols corresponding to the aldehydes are formed by the action of alcohol dehydrogenase in the presence of NADH. Malty flavor is primarily from 3-methylbutanal. Another defect associated with growth of lactic acid bacteria in milk is “ropy” texture. Most dairy-associated species of lactic acid bacteria have strains that produce exocellular polymers that cause the ropy defect (16). Some of these strains are used to produce high-viscosity fermented products such as yogurt and Scandinavian ropy milk (e.g., vilia, skyr). The defect in noncultured fluid milk products is usually caused by growth of specific strains of lactococci. The polymer produced by these organisms is a polysaccharide containing glucose and galactose with small amounts of mannose, rhamnose, and pentose (17).

Defects in Cheese Lactic Acid Bacteria

Some strains of lactic acid bacteria produce flavor and appearance defects in cheese. Lactobacilli are a normal part of the dominant microflora of aged Cheddar cheese. If heterofermentative lactobacilli predominate, the cheese is prone to develop an “open” texture or fissures, a result of gas production during aging (68). Off flavors are also associated with the growth of these organisms (125). Gassy defects in aged Cheddar cheese are more often associated with growth of lactobacilli than with growth of coliforms, yeasts, or sporeformers. The use of elevated ripening temperatures for Cheddar cheese, e.g., 15 rather than 8°C, encourages growth of heterofermentative lactobacilli but not that of non-lactic acid bacteria (27). This phenomenon limits the use of high-temperature storage to accelerate ripening. Lactobacillus brevis and Lactobacillus casei subsp. pseudoplantarum have been associated with gas production in retail Mozzarella cheese (53). Lactobacillus casei subsp. casei produces a soft body defect in Mozzarella cheese (54). The softened cheese cannot be readily sliced or grated and does not melt properly. Some cheese varieties occasionally exhibit a pink discoloration. Pink spots in Swiss-type varieties result

178 from the growth of pigmented strains of propionibac­ teria. In Italian cheese varieties a pink discoloration may occur either in a band near the surface or throughout the whole cheese. This defect is associated with strains of Lactobacillus delbrueckii subsp. bulgaricus that fail to lower the redox potential of the cheese (109). Another common defect of aged Cheddar cheese is the appearance of white crystalline deposits on the surface. Although they do not affect flavor, these deposits reduce consumer acceptability. Rengipat and Johnson (101) observed an atypical strain of a facultatively heterofermentative Lactobacillus associated with the deposits. This strain produces an unusually high amount of d-lactic acid during cheese aging, resulting in the formation of insoluble calcium lactate crystals, the primary component of the white deposits. Lactobacillus casei subsp. alactosus and subsp. rhamnosus have been associated with the development of a phenolic flavor in Cheddar cheese, described as being similar to horse urine (53). The flavor develops after 2 to 6 months of aging. Fruity off flavor in Cheddar cheese is usually not caused by growth of psychrotrophic bacteria, as it is in milk, but is rather a result of growth of lactic acid bacteria (usually Lactococcus spp.) that produce esterase. Fruity-flavored cheeses contain high levels of ethanol, a substrate for esterification (9). The major esters contributing to fruity flavor in cheese are ethyl hexanoate and ethyl butyrate.

Coliform Bacteria

Coliform bacteria were identified as causing gassy defects in Cheddar and related cheese varieties as early as 1885 (106). If present, they grow during the cheese manufacture process or shortly thereafter, producing “early gas” defect. In hard cheeses, such as Cheddar, this defect occurs when lactic acid fermentation fails to rapidly lower the pH or when highly contaminated raw milk is used. Cheese varieties in which acid production is purposely delayed by washing the curds are highly susceptible to coliform growth (43). Soft, mold-ripened cheeses, such as Camembert, increase in pH during ripening, with a resulting susceptibility to coliform growth (42, 107). Gas formation in retail Mozzarella cheese has been associated with growth of Klebsiella pneumoniae (80). Coliform growth in retail cheese is often manifested as a swelling of the plastic package. Approximately 107 CFU/g of coliform are needed to produce a gassy defect.

Defects in Fermented Milk Products

Fermented milk products such as cultured buttermilk, sour cream, and cottage cheese rely on diacetyl pro-

Microbial Spoilage and Public Health Concerns duced during fermentation for their typical “buttery” flavor and aroma. These products have reduced consumer appeal when this flavor is lost due to reduction of diacetyl to acetoin and 2,3-butanediol (41). Lactococci capable of growing at 7°C may produce sufficient diacetyl reductase to destroy diacetyl in cultured milks (51). Other psychrotrophic contaminants in cultured milks, including yeasts and coliforms, may also be involved in diacetyl reduction (128).

Control of Defects Caused by Lactic Acid and Coliform Bacteria

Defects in fluid milk caused by coliforms and lactic acid bacteria are controlled by good sanitation practices during milking, maintaining raw milk at temperatures below 7°C, pasteurization, and refrigeration of pasteurized products. These microorganisms seldom grow to significant levels in refrigerated pasteurized milk because of their slow growth rates compared to psychrotrophic bacteria. Control of coliform growth in cheese is achieved by using pasteurized milk, encouraging rapid fermentation of lactose, temperature and salt control (87), and good sanitation during manufacture. Controlling defects produced by undesirable lactic acid bacteria in cheese and fermented milks is more difficult, since growth of lactic acid bacteria must be encouraged during manufacture and the products often provide suitable growth environments. The use of frozen starter cultures for direct inoculation into cheese vats during cheese manufacturing and the use of bacteriocin-producing lactic acid bacterial strains are some of the ways to ensure the integrity of starter cultures and the quality of fermented milk products. Undesirable strains of lactic acid bacteria are readily isolated from the manufacturing environment, so their control requires attention to plant cleanliness and to protection of the product during manufacture.

SPORE-FORMING BACTERIA Spoilage by spore-forming bacteria can occur in low-acid fluid milk products that are preserved by substerilization heat treatments and packaged with little chance for recontamination with vegetative cells. Products in this category include aseptically packaged milk and cream and sweetened and unsweetened concentrated canned milks. Nonaseptically packaged refrigerated fluid milk may spoil due to growth of psychrotrophic Bacillus cereus, Bacillus mycoides, Bacillus weihenstephanensis, and Bacillus poly­ myxa in the absence of more rapidly growing gram-negative psychrotrophs (47, 119). Hard cheeses, especially those with low interior salt concentrations, are also susceptible to spoilage by spore-forming bacteria.

7.  Milk and Dairy Products Spore-forming bacteria that spoil dairy products usually originate in the raw milk. Populations present in raw milk are generally quite low (<5,000 CFU/ml), and the occurrence of sporeformer-induced defects does not always correlate with initial numbers of sporeformers in the raw product (88). This is because products prone to support sporeformer growth are stored for sufficiently long periods of time that outgrowth of small numbers of cells can eventually cause a defect. Spore-forming bacteria in raw milk are predominantly Bacillus spp., including Bacillus licheniformis, B. pallidus, B. cereus, B. subtilis, and B. megaterium (79, 112). Clostridium spp. are present in raw milk at such low levels that enrichment and most-probable-number techniques must be used for quantification (103). Populations of sporeforming bacteria in raw milk vary seasonally. Growth of these bacteria in silage contributes to high spore loads in raw milk (118).

Defects in Fluid Milk Products

Pasteurized milk packaged under conditions that limit recontamination can spoil due to the growth of psychrotrophic B. cereus. This topic has been reviewed by Meer et al. (86). Psychrotrophic B. cereus is present in over 80% of raw milk samples. There is also evidence that psychrotrophic Bacillus spp. are introduced into the milk at the processing plant as postpasteurization contaminants (46). Psychrotrophic B. cereus can reach populations exceeding 106 CFU/ml in milk held for 14 days at 7°C, although slower growth is more common (88). The defect is described as sweet curdling, since it first ­appears as coagulation without significant acid or off ­flavor being formed. Coagulation is caused by a ­chymosin-like protease (20). Eventually the enzyme degrades ­casein sufficiently to produce a bitter-flavored product. Growth may become visible as “buttons” at the bottom of the carton; these buttons are actually bacterial colonies. Psychrotrophic Bacillus spp. other than B. cereus are also capable of spoiling heat-treated milk. Cromie et al. (26) observed that psychrotrophic Bacillus circulans was the predominant spoilage organism in aseptically packaged heat-treated milk. This microorganism produces acid from lactose, giving the milk a sour flavor. Bacillus mycoides is another frequently isolated psychrotrophic sporeformer in milk (97). Most bacterial spores present in raw milk are moderately heat labile and destroyed by UHT treatments. The major heat-resistant species in milk is Geobacillus stea­ rothermophilus (formerly Bacillus stearothermophilus) (93). Bacillus sporothermodurans and Paenibacillus lac­ tis have been isolated from UHT milk (111).

179

Defects in Canned Condensed Milk

Canned condensed milk may be either sweetened with sucrose and glucose to lower the aw or left unsweetened. The unsweetened product must be sterilized by heat treatment, whereas the sweetened product has sufficiently low aw to inhibit spore germination. Defects associated with growth of surviving spore-forming organisms in this product have been described by Cari (13). “Sweet coagulation” is caused by growth of B. coagulans, G. stearothermophilus, or B. cereus. This defect is similar to the sweet curdling defect caused by psychrotrophic B. cereus in pasteurized milk. Protein destruction, in addition to curdling, can also occur and is usually caused by growth of B. subtilis or B. licheniformis. Swelling or bursting of cans can be caused by growth of Clostridium sporogenes. “Flat sour” defect (acidification without gas production) can result from growth of G. stearothermophilus, B. licheniformis, B. co­ agulans, B. macerans, and B. subtilis (59).

Control of Sporeformer-Associated Defects in Fluid Products

Methods for controlling growth of sporeformers in fluid products mainly involve the use of appropriate heat treatments. UHT treatments yield products microbiologically stable at room temperature. However, when sub-UHT heat treatments are more severe than that required for pasteurization, the shelf life of cream and milk can actually decrease, a phenomenon attributed to spore activation (93). Microfiltration or bactofugation techniques can be used to remove bacteria and spores from raw milk before pasteurization. These techniques help to extend the shelf of milk or milk products.

Defects in Cheese

The major defect in cheese caused by spore-forming bacteria is gas formation resulting from growth of clostridia, particularly Clostridium tyrobutyricum and Clostridium beijerinckii, and occasionally Clostridium sporogenes and Clostridium butyricum (72). This defect is often called “late gas” because it occurs after the cheese has aged for several weeks. Emmental, Swiss, Gouda, and Edam cheeses are most often affected because of their relatively high pH and moisture content and low interior salt levels. Late gas defect results from the fermentation of lactate to butyric acid, acetic acid, carbon dioxide, and hydrogen gas. Populations of C. ty­ robutyricum spores of less than one per ml of milk can produce the defect, because the spores are concentrated in the cheese curd during manufacture (104). The number of spores required to cause late gas in 9-kg wheels of rinded Swiss cheese was estimated at >100 per liter of

180 raw milk (29). The presence of C. tyrobutyricum spores in milk has been traced to the consumption of contaminated silage, which increases levels in the cow’s feces (29). Contaminated silage generally has a high pH that allows growth of clostridia.

Control of Sporeformer-Associated Defects in Cheese

Ideally, control of late gas defect would occur at the farm by instituting feeding and management practices that would reduce the number of spores entering the milk supply (49). In practice, this approach has not achieved the required results, so cheese manufacturers have tried to control the defect by removing spores from the milk at the plant or by inhibiting their growth in the cheese (116). Numbers of bacterial spores can be reduced in milk by a centrifugation process known as bactofugation (62). Bacteriocins produced by lactic acid bacteria may provide a highly specific means of inhibiting anaerobic spore germination (122).

YEASTS AND MOLDS Growth of yeasts and molds is a common cause of spoilage of fermented dairy products, because these microorganisms are able to grow well at low pH. Yeast spoilage is manifested as fruity or yeasty odor and/or gas formation. Cured cheeses, when properly made, have low amounts of lactose, thus limiting the potential for yeast growth. Cultured milks, such as yogurt and buttermilk, and fresh cheeses, such as cottage cheese, normally contain sufficient lactose to support yeast growth. A “fermented/yeasty” flavor was observed in Cheddar cheese spoiled by growth of a Candida sp. and was associated with elevated ethanol, ethyl acetate, and ethyl butyrate levels (52). The affected cheese had a high moisture content (associated with low starter activity, and therefore high residual lactose) and low salt content, which contributed to allowing yeast growth. Yeast spoilage can also occur in dairy foods with low aw, such as sweetened condensed milk and butter. The most common yeasts present in dairy products are Kluyveromyces marxianus and Debaromyces hansenii (the teleomorph) and their asporegenous counterparts (the anamorph), Candida species, and Zygosaccharomyces mieroellipsoides (12, 38, 63). Kluyveromyces and Saccharomyces genera are galactose-fermenting yeasts and cause spoilage in cheeses made with thermophilic and mesophilic starters such as Streptococcus thermophilus and Lactococcus lactis, respectively. This phenomenon is due in part to the inability of these lactic acid bacteria

Microbial Spoilage and Public Health Concerns to metabolize the galactose part of lactose (75). Also prevalent are Rhodotorula mucilaginosa, Yarrowia lipolytica, Torulospora, and Pichia spp. (102, 130). Y. lipolytica causes deleterious brown discoloration in cheeses such as Camembert and Gorgonzola-type cheeses. Although the mechanism of discoloration is not fully known, the pigment is thought to be produced as a result of tyrosine degradation to melanin via the oxidation and polymerization of intermediates (e.g., homoprotocatechuic (3,4-dihydroxyphenylacetic) acid and homogentisic (2,5-dihydroxyphenylacetic) acid (14, 131). Fermented dairy products provide a highly specialized ecological niche for yeasts, selecting for those that can utilize lactose or lactic acid and that tolerate high salt concentrations (38). Yeasts able to produce proteolytic or lipolytic enzymes may also have a selective advantage for growth in dairy products. Growth of spoilage molds on cheese is a problem that still has significance, though it dates back to prehistory. Some molds could produce toxins in the cheese that could present a possible health risk to the consumer. The presence of molds in raw milk is rare, but molds could contribute to cheese contamination during processing and storage. Airborne mold in the processing environment may enter through the ventilation system or via the clothes of production workers, although there is no clear correlation between airborne molds and molds isolated from products (66). Further, packaged cheeses stored under refrigeration temperature are susceptible to mold growth, especially by psychrotolerant species that require very little oxygen for growth. The most common molds found on cheese are Penicillium spp. (11, 123), with others occasionally found, including Aspergillus, Alternaria, Mucor, Fusarium, Cladosporium, Geotrichum, and Hormodendrum. Mold species commonly isolated from processed cheese include Penicillium spp. P. roqueforti, P. cyclopium, P. viridicatum, and P. crustosum (124). Vacuum-packaged cured cheese supports the growth of Cladosporium cladosporioides, Penicillium commune, Cladosporium herbarum, Penicillium glabrum, and Phoma glomerata (6, 50).

Controlling Fungal Spoilage

Yeasts and molds that spoil dairy products can usually be isolated in the processing plant on packaging equipment, in the air, in salt brines, on manufacturing equipment, and in the general environment (floors, walls, ­ventilation ducts, etc.). Successful control efforts must start with limiting the exposure of products to these sources. Mold spores do not survive pasteurization (32). If the

7.  Milk and Dairy Products initial contamination level is limited, strategies to inhibit growth are more likely to succeed. These include packaging to reduce oxygen (and/or increase carbon dioxide), cold storage, and the use of antimycotic chemicals such as sorbate, propionate, and natamycin (pimaracin). Added liquid smoke is also a potent mold inhibitor (129). None of these control measures is completely effective. Vacuum-packaged cheese is susceptible to thread mold defect, whereby the fungi grow in the wrinkles of the plastic film (50). Some molds are resistant to antimycotic additives. Sorbate-resistant molds are commonly isolated from sorbate-treated cheese, but not from untreated cheese (74). Some Penicillium spp. not only are resistant to sorbate but will also degrade it by decarboxylation, producing 1,3-pentadiene (78). This imparts a k­erosene-­like odor to the cheese. Some Mucor spp. degrade sorbate to 4-hexenol, and some Geotrichum spp. degrade it to 4-hexenoic acid. Sorbate can also be used as a carbon source or be oxidized to carbon dioxide and water (74). The ability of some molds to degrade sorbate explains why cheeses with high levels of mold contamination are not effectively preserved by this additive. The incorporation of antimicrobial agents such as malic acid, nisin, and natamycin in edible films for wrapping cheese may provide a hurdle against spoilage by Y. lipo­ lytica and Penicillium spp. on cheese surfaces (98).

References 1. Abad, P., A. Villafafila, J. D. Frias, and C. RodriguezFernandez. 1993. Extracellular lipolytic activity from Pseudomonas fluorescens biovar I (Pseudomonas fluore­ scens NC1). Milchwissenschaft 48:680–683. 2. Alkanhal, H. A., J. F. Frank, and G. L. Christen. 1985. Microbial protease and phospholipase C stimulate lipolysis of washed cream. J. Dairy Sci. 68:3162–3170. 3. Andersson, R. E. 1980. Lipase production, lipolysis and formation of volatile compounds by Pseudomonas fluore­ scens in fat containing media. J. Food Sci. 45:1694–1701. 4. Andersson, R. E. 1980. Microbial lipolysis at low temperatures. Appl. Environ. Microbiol. 39:36–40. 5. Banwart, G. J. 1981. Basic Food Microbiology. AVI Publishing Co., Westport, CT. 6. Basilico, J. C., M. Z. deBasilico, C. Chiericatti, and C. G. Vinderola. 2001. Characterization and control of thread mould in cheese. Lett. Appl. Microbiol. 32:419–423. 7. Bassette, R., and J. S. Acosta. 1988. Composition of milk products, p. 39–79. In N. P. Wong (ed.), Fundamentals of Dairy Chemistry. Van Nostrand Reinhold Co., New York, NY. 8. Batish, V. K., H. Chander, K. C. Zumdegni, K. L. Bhatia, and R. S. Singh. 1988. Antibacterial activity of lactoferrin against some common foodborne pathogenic organisms. Aust. J. Dairy Technol. 43:16–18.

181 9. Bills, D. D., M. E. Morgan, L. M. Reddy, and E. A. Day. 1965. Identification of compounds responsible for fruit flavor defect of experimental Cheddar cheeses. J. Dairy Sci. 48:1168–1170. 10. Bramley, A. J., and C. H. McKinnon. 1990. The microbiology of raw milk, p. 163–208. In R. K. Robinson (ed.), Dairy Microbiology, vol. 1. Elsevier Applied Science, New York, NY. 11. Bullerman, L. B., and F. J. Olivigni. 1974. Mycotoxin producing potential of molds isolated from Cheddar cheese. J. Food Sci. 39:1166–1168. 12. Cappa, F., and P. F. Cocconcilli. 2001. Identification of fungi from dairy products by means of 18S rRNA analysis. Int. J. Food Microbiol. 69:157–169. 13. Cari´c, M. 1994. Concentrated and Dried Dairy Products. VCH Publishers, Inc., New York, NY. 14. Carreira A., L. M. Ferreira, and V. Loureiro. 2001. Production of brown tyrosine pigments by the yeast Yarrowia lipolytica. J. Appl. Microbiol. 90: 372–379. 15. Centers for Disease Control and Prevention. 2008. Outbreak of multidrug-resistant Salmonella enterica sero­ type Newport infection associated with the consumption of unpasteurized Mexican-style aged cheese—Illinois, March 2006–April 2007. MMWR Morb. Mortal. Wkly. Rep. 57:432–435. 16. Cerning, J. 1990. Exocellular polysaccharides produced by lactic acid bacteria. FEMS Microbiol. Rev. 87:113–130. 17. Cerning, J., C. Bouillanne, M. Landon, and M. Desmazeaud. 1992. Isolation and characterization of exopolysaccharides from slime-forming mesophilic lactic acid bacteria. J. Dairy Sci. 75:692–699. 18. Chapman, H. R., and M. E. Sharpe. 1990. Microbiology of cheese, p. 203–289. In R. K. Robinson (ed.), Dairy Microbiology, vol. 2. Elsevier Applied Science, New York, NY. 19. Chen, L., R. M. Daniel, and T. Coolbear. 2003. Detection and impact of protease and lipase activities in milk and milk powders. Int. Dairy J. 13:255–275. 20. Choudhery, A. K., and E. M. Mikolajcik. 1971. Activity of Bacillus cereus proteinases in milk. J. Dairy Sci. 53:363–366. 21. Chrisope, G. L., and R. T. Marshall. 1976. Combined action of lipase and microbial phospholipase C on a model fat emulsion and raw milk. J. Dairy Sci. 59:2024–2030. 22. Christen, G. L., and R. T. Marshall. 1985. Effect of histidine on thermostability of lipase and protease of Pseudomonas fluorescens 27. J. Dairy Sci. 68:594–604. 23. Christian, J. H. B. 1980. Reduced water activity, p. 70– 91. In International Commission on Microbiological Specifications for Foods (ed.), Microbial Ecology of Foods, vol. 1. Academic Press, New York, NY. 24. Cousin, M. A. 1982. Presence and activity of psychrotrophic microorganisms in milk and dairy products: a review. J. Food Prot. 45:172–207.

182 25. Cousin, M. A. 1989. Physical and biochemical effects of milk components, p. 205–225. In R. C. McKellar (ed.), Enzymes of Psychrotrophs in Raw Food. CRC Press, Inc., Boca Raton, FL. 26. Cromie, S. J., T. W. Dommett, and D. Schmidt. 1989. Changes in the microflora of milk with different pasteurization and storage conditions and aseptic packaging. Aust. J. Dairy Technol. 44:74–77. 27. Cromie, S. J., J. E. Giles, and J. R. Dulley. 1987. Effect of elevated ripening temperatures on the microflora of Cheddar cheese. J. Dairy Res. 54:69–76. 28. D’Amico, D. J., E. Groves, and C.W. Donnelly. 2008. Low incidence of foodborne pathogens of concern in raw milk utilized for farmstead cheese production. J. Food Prot. 71:1580–1589. 29. Dasgupta, A. R., and R. R. Hull. 1989. Late blowing of Swiss cheese: incidence of Clostridium tyrobutyricum in manufacturing milk. Aust. J. Dairy Technol. 44:82–87. 30. Deeth, H. C., and C. H. Fitz-Gerald. 1983. Lipolytic enzymes and hydrolytic rancidity in milk and milk products, p. 195–239. In P. F. Fox (ed.), Developments in Dairy Chemistry, Part II. Applied Science, London, United Kingdom. 31. Dionysius, D. A., P. A. Grieve, and A. C. Vos. 1992. Studies on the lactoperoxidase system: reaction kinetics and antibacterial activity using two methods for hydrogen peroxide generation. J. Appl. Bacteriol. 72:146–153. 32. Doyle, M. P., and E. H. Marth. 1975. Thermal inactivation of conidia from Aspergillus flavus and Aspergillus parasiticus. I. Effects of moist heat, age of conidia, and sporulation medium. J. Milk Food Technol. 38:678–682. 33. Dumont, J. P., G. Delespaul, B. Miquot, and J. Adda. 1977. Influence des bactéries psychrotrophes sur les qualités organoleptiques de fromages à pâte molle. Lait 57:619–630. 34. Ewings, K. N., R. E. O’Conner, and G. E. Mitchell. 1984. Proteolytic microflora of refrigerated raw milk in South East Queensland. Aust. J. Dairy Technol. 39:65–68. 35. Fairbairn, D. J., and B. A. Law. 1987. The effect of nitrogen and carbon sources on proteinase production by Pseudomonas fluorescens. J. Appl. Bacteriol. 62:105–113. 36. Fernandez, L., J. A. Alvarez, P. Palacios, and C. San Jose. 1992. Proteolytic and lipolytic activities of Pseudomonas fluorescens grown in raw milk with variable iron content. Milchwissenschaft 47:160–163. 37. Fernandez, L., C. San Jose, and R. C. McKellar. 1990. Repression of Pseudomonas fluorescens extracellular lipase secretion by arginine. J. Dairy Res. 57:69–78. 38. Fleet, G. H. 1990. Yeasts in dairy products. J. Appl. Bacteriol. 68:199–211. 39. Fox, P. F., P. Power, and T. M. Cogan. 1989. Isolation and molecular characteristics, p. 57–120. In R. C. McKellar (ed.), Enzymes of Psychrotrophs in Raw Food. CRC Press, Inc., Boca Raton, FL. 40. Frank, J. F., and R. A. Koffi. 1990. Surface-adherent growth of Listeria monocytogenes is associated with

Microbial Spoilage and Public Health Concerns

41.

42.

43.

44.

45.

46.

47. 48.

49. 50.

51. 52. 53. 54.

55.

increased resistance to surfactant sanitizers and heat. J. Food Prot. 53:560–564. Frank, J. F., and E. H. Marth. 1988. Fermentations, p. 656–738. In N. P. Wong (ed.), Fundamentals of Dairy Chemistry, 3rd ed. Van Nostrand Reinhold Co., New York, NY. Frank, J. F., E. H. Marth, and N. F. Olson. 1977. Survival of enteropathogenic and non-pathogenic Escherichia coli during the manufacture of Camembert cheese. J. Food Prot. 40:835–842. Frank, J. F., E. H. Marth, and N. F. Olson. 1978. Behavior of enteropathogenic Escherichia coli during manufacture and ripening of brick cheese. J. Food Prot. 41:111–115. Garcia, M. L., B. Sanz, P. Garcia-Collia, and J. A. Ordonez. 1989. Activity and thermostability of the extracellular lipases and proteinases from pseudomonads isolated from raw milk. Milchwissenschaft 44:547–560. Griffiths, M. W. 1989. Effect of temperature and milk fat on extracellular enzyme synthesis by psychrotrophic bacteria during growth in milk. Milchwissenschaft 44:539–543. Griffiths, M. W., and J. D. Phillips. 1990. Incidence, source and some properties of psychrotrophic Bacillus spp. found in raw and pasteurized milk. J. Soc. Dairy Technol. 43:62–70. Hanson, M. L., W. L. Wendorff, and K. B. Houck. 2005. Effect of heat treatment of milk on activation of Bacillus spores. J. Food Prot. 68:1484–1486. Haryani, S., N. Datta, A. J. Elliot, and H. C. Deeth. 2003. Production of proteinases by psychrotrophic bacteria in raw milk stored at low temperature. Aust. J. Dairy Technol. 58:15–20. Herlin, A. H., and A. Christansson. 1993. Cheese-blowing anaerobic spores in bulk milk from loose-housed and tied dairy cows. Milchwissenschaft 48:686–689. Hocking, A. D., and M. Faedo. 1992. Fungi causing thread mould spoilage of vacuum packaged Cheddar cheese during maturation. Int. J. Food Microbiol. 16: 123–130. Hogarty, S. L., and J. F. Frank. 1982. Low-temperature activity of lactic streptococci isolated from cultured buttermilk. J. Food Prot. 43:1208–1211. Horwood, J. F., W. Stark, and R. R. Hull. 1987. A “fermented, yeasty” flavour defect in Cheddar cheese. Aust. J. Dairy Technol. 42:25–26. Hull, R., S. Toyne, I. Haynes, and F. Lehman. 1992. Thermoduric bacteria: a re-emerging problem in cheesemaking. Aust. J. Dairy Technol. 47:91–94. Hull, R. R., A. V. Roberts, and J. J. Mayes. 1983. The association of Lactobacillus caseii with a soft-body defect in commercial Mozzarella cheese. Aust. J. Dairy Technol. 22:78–80. Jaspe, A., P. Palacios, P. Matias, L. Fernandez, and C. Sanjose. 1994. Proteinase activity of Pseudomonas fluorescens grown in cold milk supplemented with nitrogen and carbon sources. J. Dairy Sci. 77: 923–929.

7.  Milk and Dairy Products 56. Jaspe, A., P. Oviedo, L. Fernandez, P. Palacios, and C. Sanjose. 1995. Cooling raw milk: change in the spoilage potential of contaminating Pseudomonas. J. Food Prot. 58:915–921. 57. Jayarao, B. M., S. C. Donaldson, B. A. Straley, A. A. Sawant, N. V. Hegdeand, and J. L. Brown. 2006. A survey of foodborne pathogens in bulk tank milk and raw milk consumption among farm families in Pennsylvania. J. Dairy Sci. 89:2451–2458. 58. Jenness, R. 1988. Composition of milk, p. 1–38. In N. P. Wong (ed.), Fundamentals of Dairy Chemistry. Van Nostrand Reinhold Co., New York, NY. 59. Kalogridou-Vassiliadou, D. 1992. Biochemical activities of Bacillus species isolated from flat sour evaporated milk. J. Dairy Sci. 75:2681–2686. 60. Karns, J. S., J. S. Van Kessel, B. J. McCluskey, and M. L. Perdue. 2005. Prevalence of Salmonella enterica in bulk tank milk from US dairies as determined by polymerase chain reaction. J. Dairy Sci. 88:3475–3479. 61. Karns, J. S., J. S. Van Kessel, B. J. McClusky, and M. L. Perdue. 2007. Incidence of Escherichia coli O157:H7 and E. coli virulence factors in US bulk tank milk as determined by polymerase chain reaction. J. Dairy Sci. 90:3212–3219. 62. Kosikowski, F. V. 1982. Cheese and Fermented Milk Foods, 2nd ed. F. V. Kosikowski and Associates, Brook­ tondale, NY. 63. Kosse, D., H. Seiler, R. Amann, W. Ludwig, and S. Scherer. 1997. Identification of yoghurt-spoiling yeasts with 18s rRNA oligonucleotide probes. Syst. Appl. Microbiol. 20:468–480. 64. Kroll, S. 1989. Thermal stability, p. 121–152. In R. C. McKellar (ed.), Enzymes of Psychrotrophs in Raw Food. CRC Press, Inc., Boca Raton, FL. 65. Kumura, H., K. Mikawa, and Z. Saito. 1993. Influence of milk proteins on the thermostability of the lipase from Pseudomonas fluorescens 33. J. Dairy Sci. 76:2164–2167. 66. Kure, C. F., E. Borch, I. Karlsson, J. P. Homleid, and S. Langsrud. 2008. Use of selective agar medium CREAD for monitoring the level of airborne spoilage moulds in cheese production. Int. J. Food Microbiol. 122:29–34. 67. Lafarge, V., J.-C. Ogier, V. Girard, V. Maladen, J.-Y. Leveau, A. Gruss, and A. Delacroix-Buchet. 2004. Raw cow milk bacterial population shifts attributable to refrigeration. Appl. Environ. Microbiol. 70:5644–5650. 68. Lalaye, L. C., R. E. Simard, B.-H. Lee, R. A. Holley, and R. N. Giroux. 1987. Involvement of heterofermentative lactobacilli in development of open texture in cheeses. J. Food Prot. 50:1009–1012. 69. Law, B. A. 1979. Reviews of the progress of dairy science: enzymes of psychrotrophic bacteria and their effects on milk and milk products. J. Dairy Res. 46:573–588. 70. Law, B. A., C. M. Cousins, M. E. Sharpe, and F. L. Davies. 1979. Psychrotrophs and their effects on milk and dairy products, p. 137–152. In A. D. Russell and R. Fuller (ed.), Cold Tolerant Microbes in Spoilage and the Environment. Academic Press, New York, NY.

183 71. Law, B. A., M. E Sharpe, and H. R. Chapman. 1976. Effect of lipolytic Gram negative psychrotrophs in stored milk on the development of rancidity in Cheddar cheese. J. Dairy Res. 43:459–468. 72. Le Bourhis, A.-G., K. Saunier, J. Doré, J.-P. Carlier, J.-F. Chamba, M.-R. Popoff, and J.-L. Tholozan. 2005. Development and validation of PCR primers to assess the diversity of Clostridium spp. in cheese by temporal temperature gradient gel electrophoresis. Appl. Environ. Microbiol. 71:29–38. 73. LeJeune, J. T., and P. J. Rajala-Schultz. 2009. Unpasteurized milk: a continued public health threat. Clin. Infect. Dis. 48:93–100. 74. Liewen, M. B., and E. H. Marth. 1985. Growth and inhibition of microorganisms in the presence of sorbic acid: a review. J. Food Prot. 48:364–375. 75. Liu, S., and M. Tsao. 2009. Inhibition of spoilage yeasts in cheese by killer Williopsis saturnus var. saturnus. Int. J. Food Microbiol. 131:280–282. 76. Ma, Y., D. M. Barbano, and M. Santos. 2003. Effect of CO2 addition to raw milk on proteolysis and lipolysis at 4°C. J. Dairy Sci. 86:1616–1631. 77. Marks, N. E., A. S. Grandison, and M. J. Lewis. 2001. Challenge testing of the lactoperoxidase system in pasteurized milk. J. Appl. Microbiol. 91:735–741. 78. Marth, E. H., C. M. Capp, L. Hasenzahl, H. W. Jackson, and R. V. Hussong. 1966. Degradation of potassium sorbate by Penicillium species. J. Dairy Sci. 49: 1197–1205. 79. Martin, J. H., D. P. Stahly, W. J. Harper, and I. A. Gould. 1962. Sporeforming microorganisms in selected milk supplies, p. 295–304. Proceedings of the XVIth International Dairy Congress, vol. C. International Dairy Federation, Brussels, Belgium. 80. Massa, S., F. Gardini, M. Sinigaglia, and M. E. Guerzoni. 1992. Klebsiella pneumoniae as a spoilage organism in Mozzarella cheese. J. Dairy Sci. 75:1411–1414. 81. Masson, P. L., and J. F. Heremans. 1971. Lactoferrin in milk from different species. Comp. Biochem. Physiol. 39B:119–129. 82. McKellar, R. C. 1982. Factors influencing the production of extracellular proteinase by Pseudomonas fluore­ scens. J. Appl. Bacteriol. 53:305–316. 83. McKellar, R. C. 1989. Regulation and control of synthesis, p. 153–172. In R. C. McKellar (ed.), Enzymes of Psychrotrophs in Raw Foods. CRC Press, Inc., Boca Raton, FL. 84. McKellar, R. C., and H. Cholette. 1986. Possible role of calcium in the formation of active extracellular proteinase by Pseudomonas fluorescens. J. Appl. Bacteriol. 60:37–44. 85. McKellar, R. C., K. Shamsuzzaman, C. San Jose, and H. Cholette. 1987. Influence of iron (iii) and pyoverdine, a siderophore produced by Pseudomonas fluorescens B52, on its extracellular proteinase and lipase production. Arch. Microbiol. 147:225–230. 86. Meer, R. R., J. Baker, F. W. Bodyfelt, and M. W. Griffiths. 1991. Psychrotrophic Bacillus spp. in fluid milk products: a review. J. Food Prot. 54:969–979.

184 87. Melilli, C., D. M. Barbano, M. Cacamo, M. A. Calvo, G. Schembari, and G. Licitra. 2004. Influence of brine concentration, brine temperature, and presalting on early gas defect in raw milk pasta filata cheese. J. Dairy Sci. 87:3648–3657. 88. Mikolojcik, E. M., and N. T. Simon. 1978. Heat resistant psychrotrophic bacteria in raw milk and their growth at 7°C. J. Food Prot. 41:93–95. 89. Morgan, M. E. 1976. The chemistry of some microbially induced flavor defects in milk and dairy foods. Biotechnol. Bioeng. 18:953–965. 90. Morse, P. M., H. Jackson, C. H. McNaughton, A. G. Leggatt, G. B. Landerkin, and C. K. Johns. 1968. Investigation of factors contributing to the bacteria count of bulk tank milk. II. Bacteria in milk from individual cows. J. Dairy Sci. 51:1188–1191. 91. Mosteller, T. M., and J. R. Bishop. 1993. Sanitizer efficacy against attached bacteria in milk biofilm. J. Food Prot. 56:34–41. 92. Mottar, J. F. 1989. Effect on the quality of dairy products, p. 227–243. In R. C. McKellar (ed.), Enzymes of Psychrotrophs in Raw Food. CRC Press, Inc., Boca Raton, FL. 93. Muir, D. D. 1989. The microbiology of heat treated fluid milk products, p. 209–270. In R. K. Robinson (ed.), Dairy Microbiology, vol. 1. Elsevier Applied Science, New York, NY. 94. Myhara, R. M., and B. Skura. 1990. Centroid search optimization of cultural conditions affecting the production of extracellular proteinase by Pseudomonas fragi ATCC 4973. J. Appl. Bacteriol. 69:530–538. 95. Nashif, S. A., and F. E. Nelson. 1953. The lipase of Pseudomonas fragi. III. Enzyme action in cream and butter. J. Dairy Sci. 36:481–488. 96. Oliver, S. P., K. J. Boor, S. C. Murphy, and S. E. Murinda. 2009. Food safety hazards associated with consumption of raw milk. Foodborne Pathog. Dis. 6:793–806. 97. Phillips, J. D., and M. W. Griffiths. 1986. Factors contributing to the seasonal variation of Bacillus species in pasteurized products. J. Appl. Bacteriol. 61:275–285. 98. Pintado, C. M. B. S., M. A. S. S. Ferreira, and I. Sousa. 2010. Control of pathogenic and spoilage microorganisms from cheese surface by whey protein films containing malic acid, nisin and natamycin. Food Control 21:240–246. 99. Reddy, M. C., D. D. Bills, R. C. Lindsay, and L. M. Libbey. 1968. Ester production by Pseudomonas fragi. I. Identification and quantification of some esters produced in milk cultures. J. Dairy Sci. 51:656–659. 100. Reimerdes, E. H. 1982. Changes in the proteins of raw milk during storage, p 271. In P. F. Fox (ed.), Developments in Dairy Chemistry, part I. Applied Science, London, United Kingdom. 101. Rengipat, S., and E. A. Johnson. 1989. Characterization of a Lactobacillus strain producing white crystals on Cheddar cheese. Appl. Environ. Microbiol. 56:2579–2582.

Microbial Spoilage and Public Health Concerns 102. Rohm, H., F. Eliskases-Lechner, and M. Bräuer. 1992. Diversity of yeasts in selected dairy products. J. Appl. Bacteriol. 72:370–376. 103. Rosen, B., U. Merin, and I. Rosenthal. 1989. Evaluation of clostridia in raw milk. Milchwissenschaft 44:356–357. 104. Rosen, B., G. Popel, and I. Rosenthal. 1990. The affinity of Clostridium tyrobutyricum to casein in raw milk. Milchwissenschaft 45:152–154. 105. Rowe, M. T. 1990. Growth and extracellular enzyme production by psychrotrophic bacteria in raw milk stored at low temperature. Milchwissenschaft 45:495–499. 106. Russell, H. L. 1885. Gas producing bacteria and the relation of the same to cheese, p. 139–150. In Wisconsin Agricultural Experimental Station 12th Annual Report. 107. Rutzinski, J. L., E. H. Marth, and N. F. Olson. 1979. Behavior of Enterobacter aerogenes and Hania species during the manufacture and ripening of Camembert cheese. J. Food Prot. 42:790–793. 108. Schröder, M. J. A. 1984. Origins and levels of post pasteurization contamination of milk in the dairy and their effects of keeping quality. J. Dairy Res. 51:59–67. 109. Shannon, E. L., N. F. Olson, and J. H. von Elbe. 1969. Effect of lactic starter culture on pink discoloration and oxidation-reduction potential in Italian cheese. J. Dairy Sci. 52:1557–1561. 110. Seifu, E., E. M. Buys, and E. F. Donkin. 2005. Significance of the lactoperoxidase system in the dairy industry and its potential applications: a review. Trends Food Sci. Technol. 16:137–154. 111. Sheldeman, P., K. Goossens, M. Rodriguez-Diaz, A. Pil, J. Goris, L. Herman, P. de Vos, N. A. Logan, and M. Heyndrickx. 2004. Paenibacillus lactis sp. nov. isolated from raw and heat treated milk. Int. J. Syst. Evol. Microbiol. 54:885–891. 112. Sheldeman, P., A. Pil, L. Herman, P. de Vos, and M. Hendrickx. 2005. Incidence and diversity of potentially highly heat resistant spores isolated at dairy farms. Appl. Environ. Microbiol. 71:1480–1494. 113. Shipe, W. F., R. Bassette, D. D. Deane, W. L. Dinkley, E. G. Hammond, W. J. Harper, D. H. Klein, M. E. Morgan, J. H. Nelson, and R. A. Scanlan. 1978. Off flavors in milk: nomenclature, standards, and bibliography. J. Dairy Sci. 61:856–869. 114. Stead, D. 1986. Microbial lipases: their characteristics, role in food spoilage and industrial uses. J. Dairy Res. 53:481–505. 115. Stepaniak, L., E. Zakrzewski, and T. Sorhaug. 1991. Inactivation of heat-stable proteinase from Pseudomonas fluorescens P1 at pH 4.5 and 56EC. Milchwissenshaft 46:139–142. 116. Su, Y. C., and S. C. Ingham. 2000. Influence of milk centrifugation, brining and ripening conditions in preventing gas formation by Clostridium spp. in Gouda cheese. Int. J. Food Microbiol. 54:147–154. 117. Suhren, G. 1989. Producer microorganisms, p. 3–34. In R. C. McKellar (ed.), Enzymes of Psychrotrophs in Raw Foods. CRC Press, Inc., Boca Raton, FL.

7.  Milk and Dairy Products 1 18. te Giffel, M. C. T., A. Wagendorp, A. Herrewegh, and F. Driehuis. 2002. Bacterial spores in silage and raw milk. Antonie van Leeuwenhoek 81:625–630. 119. Ternstrom, A., M. A. Lindberg, and G. Molin. 1993. Classification of the spoilage flora of raw and pasteurized bovine milk, with special reference to Pseudomonas and Bacillus. J. Appl. Bacteriol. 75:25–34. 120. Thomas, S. B. 1966. Sources, incidence, and significance of psychrotrophic bacteria in milk. Milchwissenschaft 21:270–275. 121. Thomas, S. B., and B. F. Thomas. 1973. Psychrotrophic bacteria in refrigerated bulk-collected raw milk. Part I. Dairy Ind. 38:11–15. 122. Thualt, D., E. Beliard, J. Je Guern, and C.-M. Bourgeois. 1991. Inhibition of Clostridium tyrobutyricum by bacteriocin-like substances produced by lactic acid bacteria. J. Dairy Sci. 74:1145–1150. 123. Torrey, G. S., and E. H. Marth. 1977. Isolation and toxicity of molds from foods stored in homes. J. Food Prot. 40:187–190. 124. Tsai, W.-Y. J., M. B. Liewen, and L. Bullerman. 1988. Toxicity and sorbate sensitivity of molds isolated from surplus commodity cheese. J. Food Prot. 51:457–462. 125. Turner, K. W., and T. D. Thomas. 1980. Lactose fermentation in Cheddar cheese and the effect of salt. New Zealand J. Dairy Sci. Technol. 15:265–276. 126. Van Kessel, J. S., J. S. Karns, L. Gorski, B. J. McCluskey, and M. L. Perdue. 2004. Prevalence of Salmonellae, Listeria monocytogenes, and fecal coliforms in bulk tank milk on U.S. dairies. J. Dairy Sci. 87:2822–2830. 127. Vilafafila, A., J. D. Frias, P. Abad, and C. RodriguezFernandez. 1993. Extracellular proteinase activity from psychrotrophic Pseudomonas fluorescens biovar 1 (Ps. fluorescens NC1). Milchwissenschaft 48:435–438.

185 128. Wang, J. J., and J. F. Frank. 1981. Characterization of psychrotrophic bacterial contamination of commercial buttermilk. J. Dairy Sci. 64:2154–2160. 129. Wendorff, W. L., W. E. Riha, and E. Muehlenkamp. 1993. Growth of molds on cheese treated with heat or liquid smoke. J. Food Prot. 56:963–966. 130. Westall, S., and O. Liltenborg. 1998. Spoilage yeasts of decorated soft cheese packed in modified atmosphere. Food Microbiol. 15:243–249. 131. Williams, A. G., and S. E. Withers. 2007. Tyrosine metabolism in pigment-forming Yarrow lipolytica strains isolated from English and European speciality mouldripened cheese exhibiting a brown discolouration defect. Int. J. Dairy Technol. 60:165–174. 132. Wolfson, L. M., and S. S. Sumner. 1993. Antibacterial activity of the lactoperoxidase system: a review. J. Food Prot. 56:887–892. 133. Yan, L., B. E. Langlois, J. O’Leary, and C. Hicks. 1983. Effect of storage conditions of grade A raw milk on proteolysis and cheese yield. Milchwissenschaft 38: 715–719. 134. Zelent, B., K. A. Sharp, and J. M. Vanderkooi. 2010. Differential scanning calorimetry and fluorescence study of lactoperoxidase as a function of guanidiniumHCl, urea, and pH. Biochim. Biophys. Acta 1804: 1508–1515. 135. Zimmerman, S., I. J. Jeon, J. E. Shirley, L. McVay, E. Ferdinand, D. Sukup, and K. A. Schmidt. 2008. Bacterial degradation of milk components is affected by storage temperature and time. Report of Progress 881. Kansas State University Agricultural Experimental Station and Cooperative Extension Service. http://en.engormix.com/ MA-dairy-cattle/articles/bacterial-degradation-milkcomponents-t1098/p0.htm

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch8

8

Frédéric Carlin

Fruits and Vegetables

Fresh fruits and vegetables are an extraordinary dietary source of nutrients, micronutrients, vitamins, and fibers for humans and an essential basic raw material for the food industry (23, 26). They are living organs detached from their parent plants and have a high water content, which contributes to their natural fragility. In addition, fruits and vegetables are widely exposed to microbial contamination through contact with soil, dust, and water and by handling at harvest or during postharvest processing, thereby establishing conditions that may lead to spoilage and loss of quality. A single produce item such as tomato is cultivated under many different climates and at many latitudes from its tropical area of origin to the colder Nordic countries. Giving an exhaustive account of microbial spoilage of tomatoes, as well as a multitude of other types of produce, presents some difficulties. Moreover, fruits and vegetables cover many different species and include many different plant organs at varied stages of physiological maturity in their consumed forms. Reviewing microbial spoilage of fruits and vegetables is a real challenge, and the literature on this topic is particularly rich. This chapter focuses on the origin, description, and control of bacterial and fungal spoil-

age of fruits and vegetables. Table 8.1 presents examples of some important spoilage molds, their host vegetables or fruits, and symptoms of infection. More extensive descriptions of fruit and vegetable spoilage can be found in book chapters, on Internet websites, and in many food microbiology, plant pathology, and postharvest biology scientific journals (3, 99, 123, 124; The Commercial Storage of Fruits, Vegetables, and Florist and Nursery Stocks, available at http://www. ba.ars.usda.gov/hb66).

MAIN CHARACTERISTICS OF FRUITS AND VEGETABLES AND THEIR ROLES IN MICROBIAL SPOILAGE Fruits and vegetables are the edible parts of plant organs of very diverse nature: leaves (lettuce and cabbage, for instance), stems (leek, asparagus), flowers (artichoke, cauliflower, broccoli), roots (beet, carrot, turnip), bulbs (garlic, onion), tubers (potato), and fruits in their botanical meaning, e.g., simple fruit such as tomato, cucumber, and pepper, stone fruit such as peach, seed fruit such as apple, multiple fruit such as pineapple, aggregate fruit such as raspberry, and fruits consumed in their

Frédéric Carlin, INRA, UMR408, Sécurité et Qualité des Produits d’Origine Végétale, Avignon, F-84914, France.

187

SMP_Food microbiology_CH08.indd

187

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

188

Table 8.1  Important microbial agents of postharvest spoilage of fruits and vegetablesa Type of postharvest disease or spoilage

Spoilage agent Alternaria alternata and Alternaria spp.

Black rot, black spots, dark lesions

Produce affected Cucurbits, solanaceous fruits and vegetables, green bean, brassica, potato, citrus, persimmon, mango, pome fruits

Biology Stem end pathogen. Penetration by flower or stem scars. Infection may remain quiescent.

Botrytis cinerea

Soft rot covered with gray mold

Cucurbits, solanaceous fruits and vegetables, green bean, peas, brassica, artichoke, celery, lettuce and chicory, onion, garlic, carrot, citrus, apple, strawberry, raspberry

Wide spectrum. Infection before or after harvest through damaged or senescent tissue. Favored by wet conditions. May spread into neighboring fruits, causing “nesting.” Possible growth even at low temperatures.

Colletotrichum musae

Anthracnose. Dark circular spots on ripening fruits.

Banana

Quiescent infection until fruit ripening.

Colletotrichum gloeosporioides and other Colletotrichum spp.

Anthracnose, lesion on the skin. Dark spots, sunken lesions.

Cucurbits, solanaceous fruits and vegetables, green bean, avocado, apple, mango

Quiescent infections. May form appressoria on the plant cuticle. Development of decay during fruit maturation.

Geotrichum candidum

Sour rot

Cucurbits, carrot, citrus, tomato

Soil pathogen. Transmission by insects. Wound pathogen.

Monilinia spp.

Brown rot. Brown spots, white molds in concentric circles.

Apple, stone fruits

Survival in winter on mummified fruits. Infection may remain quiescent on immature fruits.

Penicillium spp.

Blue mold, green mold covering lesions or rot

Cucurbits, onion, garlic, grape, apple, citrus

Wound pathogen. Colored spores at the center of the lesions. Slow development at low temperature. May spread from fruit to fruit.

Rhizopus spp.

Soft, very wet rot. Development of profuse mycelium with spore heads turning black.

Cucurbits, solanaceous fruits and vegetables, green bean, peas, sweet potato, stone fruits, papaya

Ubiquitous. Infection at or after harvest by wound or by contact with soil or infected produce. Rapid decay above 20°C.

Pectobacterium (Erwinia) carotovora

Soft rot

Cucurbits, solanaceous fruit and vegetables, brassica, asparagus, celery, lettuce and chicory, carrot

Infection by wounds, scars, and lenticels before or after harvest and by contact with decaying vegetable. Favored by wet conditions and temperatures of 24–30°C.

a

Adapted from references 3, 99, and 124.

immature form such as green beans. Plant tissues in fruits and vegetables consist of an assemblage of cells surrounded by a pectic and cellulosic cell wall organized in a network and a middle lamella rich in pectin cementing together cell walls (14, 54, 140). Both the water content of the cell vacuole and the cell wall organization and composition contribute to edible plant tissue firmness. The outer parts of fruits and vegetables are

SMP_Food microbiology_CH08.indd

188

characterized by layers of varying thickness according to the type of produce and include a hydrophobic cuticle consisting of cutin and wax covering an epidermis made of a layer of cells and eventually a layer of cork cells (44). Fruit and vegetable surfaces can be interrupted by natural openings, e.g., stomata or lenticels involved in the respiration and transpiration of the plant organs; trichomes, cracks, wounds caused by insects, mechani-

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

189

cal injury, and other stress assaults; and scars resulting from detachment from plants. Once detached from the plant, the physiological activity of fruits and vegetables continues and is negatively correlated with shelf life. Commodities such as apple, citrus fruit, potato, onion, garlic, carrot, cantaloupe, and watermelon with a low respiration rate (<10 ml CO2/kg/h at 5°C) can be stored for longer periods than those such as strawberry, raspberry, Brussels sprouts, spinach, broccoli, asparagus, or mushroom with a high respiration rate (>20 ml CO2/kg/h at 5°C) (63). For “climacteric” fruits (e.g., apple, tomato, avocado, and banana), several biochemical changes associated with natural respiration occur and, triggered by autocatalytic production of the plant hormone ethylene, lead to horticultural maturity (5, 118). Nonclimacteric commodities (e.g., strawberry, citrus, grape, and cherry) are picked at their horticultural maturity, a stage of development corresponding to the prerequisites for utilization by consumers. The senescence process leads, after horticultural maturity, to irreversible changes in structure and

metabolism of the organs, and finally to deterioration (75). Postharvest moisture loss due to respiration causes structural damage in fruits such as apples and wilting in leafy vegetables. A decrease in acidity and softening is followed by cell wall degradation (135). Degradation of phenolic compounds during ripening of climacteric fruits is among the physiological changes that may favor the invasion and growth of spoilage microorganisms. Fruits and vegetables have a high water content, often exceeding 95% (fresh weight), and contain significant amounts of nutrients essential for microbial growth (Table 8.2) (102). The main limitation preventing growth of most bacteria is the low pH of fruits (as low as 2.0 in some Citrus species). Spoilage of low-pH fruits is restricted to that caused by molds and yeasts, which are more tolerant than bacteria to high acidity. The favorable effects of basic nutrients in produce tissues may be balanced by compounds known for their antimicrobial activity, such as phenolic compounds and tannins in many fruits and vegetables, e.g., tomatine (a saponine in tomato), sulfur-derived compounds in

Table 8.2  Approximate pH values and water, protein, and sugar contents of some fresh fruits and vegetablesa pH

Water (g/100 g)

Protein (g/100 g, fresh wt)

Sugars (g/100 g, fresh wt)

Asparagus

5.0–6.1

93.2

2.2

1.9

Beans (lima)

5.4–6.5

70.2

6.8

1.5

Fruit or vegetable

Broccoli

6.5

89.3

2.8

1.7

Carrot

4.9–6.3

88.3

0.93

4.5

Cauliflower

6.0–6.7

91.9

2.0

2.4

Corn (sweet)

5.9–7.3

76.0

3.2

3.2

Lettuce

6.0–6.4

95.6

0.9

1.7

Onion

5.0–5.8

88.5

0.9

4.3

Pepper (red)

5.3–5.8

92.0

1.0

4.2

Potato tuber

5.6–6.2

81.6

1.7

1.2

Spinach

5.1–6.8

91.4

2.9

0.4

Squash

5.0–5.4

94.6

1.2

2.2

Tomato (ripe)

3.4–4.7

94.5

0.88

2.6

Apple

2.9–3.3

85.6

0.26

10.4

Banana

4.5–5.2

74.9

1.1

2.4

Grape

3.4–4.5

80.5

0.72

15.0

Lime

1.8–2.0

88.3

0.70

1.7

Melon (cantaloupe)

6.2–6.5

90.2

0.84

7.8

Orange

3.6–4.3

86.8

0.94

9.4

Data from reference 84 and from U.S. Department of Agriculture, A.R.S. 2005, National Nutrient Database for Standard Reference, Release 18. Nutrient Data Laboratory Home Page, available at http://www.nal.usda.gov/fnic/foodcomp. a

SMP_Food microbiology_CH08.indd

189

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

190 the Alliaceae (onion and garlic), or terpenoids in carrot (3, 12, 108, 122).

ORIGIN OF CONTAMINATION Fruits and vegetables harbor a wide range of microbial contaminants, which may be present in a wide range of population densities. Populations of the aerobic mesophilic bacteria range from 102 to 108 CFU/g of produce, and bacterial contamination at levels of 105 to 106 CFU/g on fruits has been reported (85, 99). Numbers of yeasts and molds (reported in CFU, although CFU quantification is probably not appropriate because of the possible presence of extensive mycelia or fruiting structures) ranging between 101 and 106 CFU/g on vegetables and between 103 and 107 CFU/g on fruits have been reported (25, 85, 99). These variations reflect the diversity of the proper characteristics of the plants (aboveground or subterranean organs that expose microorganisms to different water, nutrient, and UV conditions), of the conditions prevailing during cultivation and postharvest storage, and to some extent, of the techniques implemented for enumeration (80, 99). With few exceptions, gram-negative species of bacteria are dominant on vegetables, and basidiomycetous yeasts are among the major species most frequently found on fruits (25, 99). Postharvest spoilage microorganisms do not seem to include a dominating species in sound fresh fruits and vegetables (99). The dominating species of bacteria and yeasts are generally not known for their ability to cause decay on fresh produce, and when detected, spoilage microorganisms represent only a low proportion of the normal microflora. This general view has been obtained with culture-­dependent methods, and the application of culture-independent methods based on direct analysis of DNA (or RNA) extracted from the microbial communities may result in a different representation of that diversity and of the structure of the microbial communities present on fruits and vegetables (136). However, the application of these molecular methods has, up to now, not commonly been used on fruits and vegetables. Postharvest spoilage microorganisms may take many different routes to contaminate fruits and vegetables. Seeds, including vegetable seeds, have been shown to be the primary source of postharvest diseases such as Colletotrichum infections on pepper, onion bulb neck rot caused by Botrytis spp., potato tuber soft rot caused by Erwinia spp., and potato tuber gangrene caused by Phoma spp. (99). Rainwater is a significant vehicle for microorganisms from plant to plant, plant to soil, and soil to plant through splashing (79). Soil and its components (rhizosphere and plant debris) are the natu-

SMP_Food microbiology_CH08.indd

190

ral reservoirs of spoilage bacteria such as Bacillus and Clostridium (83) and molds such as Sclerotinia spp. (68) and Rhizoctonia solani (123) and facilitate survival, in particular in plant debris or on fruits and vegetables in contact with the rhizosphere, as shown for Erwinia carotovora (105). Insects may be the vectors and the hosts of postharvest pathogens such as Botrytis cinerea or Pectobacterium carotovora (formerly Erwinia carotovora) (98, 137). Air and wind disperse spores or fruiting bodies of molds (37), leaves and microorganisms adhering to leaves, and aerosol particles that may contain bacteria, presuming the acquisition of some resistance to desiccation (79). Postharvest handling of produce has been shown to be a cause of contamination with spoilage microorganisms (99). Immersion tank solutions have been suspected to be a cause of redistribution of Phialophora malorum on pears (128). Wooden boxes previously used to store carrots may contain many species of molds pathogenic to carrot, including Rhizoctonia carotae, Sclerotinia sclerotiorum, and Botrytis cinerea, which can cause lesions on sound carrots (67). The common postharvest spoilage molds B. cinerea and Penicillium spp. can be found at multiple sites in production areas, e.g., on fruits, in orchard litter and soil, in orchard air, on packing lines, and in cold storage air; however, in these cases, contamination on the fruit surface is critical to further decay (74). The inoculum can consist of spores or conidia, mycelium, or sclerotia. Mycelium is generally infectious. Germination of spores and sclerotia generally depends on water; on the presence of nutrients, at least in small quantities, that may have leached from fruit and vegetable tissues; or on juice released from damaged tissues in wounds (3, 99). The time elapsed between contact of microorganisms with the surface of the plant organ and initiation of spoilage is extremely variable. This process presumes the establishment of some sort of colonization on the organ surface. Fluorescent pseudomonads, including pectolytic strains, form bacterial communities aggregated in a matrix of exopolymers and assimilated to form biofilms (13, 96). Adhesion of molds to plant surfaces involves very specific interactions that can involve lectins, hydrophobic contact with the plant cuticle, or secreted adhesives (133). Fungal agents of storage diseases may then colonize a few cells in a limited area of plant tissue, followed by a delay before becoming active under specific circumstances. This period without growth is known as quiescence and may vary among plant pathogens and hosts. Infection during the quiescent period can be symptomless, as that of germinated spores of Colletotrichum spp. on various types of produce, or can result in visible but nonexpanding symp-

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

191

toms, as in ghost spot of tomato, caused by B. cinerea (111). The pathogen experiences a biotrophic mode of nutrition, in which nutrients are released by the living cells of the host. The quiescent period can be observed while fruits are still attached to the parent plant. For instance, the initial infection of gray mold on strawberry or grape occurs during flowering and remains dormant until fruit formation or during postharvest storage (3). Spoilage involves to some extent the penetration of microbial cells into plant tissue. The cuticle barrier can be compromised by many postharvest spoilage microorganisms, e.g., B. cinerea on cucumber and tomato fruits, Colletotrichum spp. on tomato and bell pepper (99), or S. sclerotiorum on carrot (68). Despite evidence of cutinase activity of some fungal pathogens, the actual contribution of these enzymes to tissue penetration remains controversial, some mutants unable to produce cutinases still being pathogenic (3, 116). The formation at the tip of the germ tube is an organ called the appressorium, a structure used by fungal pathogens to attach to and then penetrate plant surfaces by mechanical pressure and likely also enzymatic activity as a means of initiating infection (50, 111). Infections induced by appressorium formation contribute, for example, to the development of ghost spots caused by B. cinerea on tomato fruits, to anthracnose caused by Colletotrichum sp. in pepper and Colletotrichum musae in banana (3, 99), or to spoilage of carrot caused by S. sclerotiorum (68). In the case of bananas, appressoria formed on preharvest fruits remain quiescent until harvest and fruit maturation to eventually penetrate into tissue and cause spoilage. Germination and appressorium formation by C. musae and Colletotrichum gloeosporioides may be induced by ethylene produced by the infected commodity (42). Fruits and vegetables also offer a large diversity of natural or accidental openings that can serve as ports of entry for penetration of postharvest spoilage microorganisms. Adverse conditions (e.g., wind, frost, and contact between fruits and limbs) in orchards, vineyards, and fields, as well as harvesting and postharvest handling involving mechanical or human interventions, can result in wounding of fruits caused by stems of other fruits or abrasion and shocks during transport (83). Up to 14% of hand-picked pears and up to 30% of apples may present wounds after harvest (1, 126). Wounds are a common site of penetration of postharvest spoilage microorganisms and are critical, for instance, for infection of tomatoes by Rhizopus stolonifer, carrots by a range of postharvest pathogens, and plums by Monilinia fructicola (53, 81, 99). Natural openings such as dead tissues at the blossom end, stem scars on apple and cit-

SMP_Food microbiology_CH08.indd

191

rus, and the calyx of tomato, apple, eggplant, and bell pepper are also a potential site for microorganisms to penetrate tissues (91, 99, 101). In a general way, bacteria lack the ability to directly penetrate plant epidermis. Pathogenic bacteria are able to counter the stomata closure of plants, increasing therefore the possibility of entry into the plant tissue (143). A higher density of lenticels on various apple cultivars is correlated with higher susceptibility to infection with Penicillium spp. (1). The presence of free water in which bacteria may be suspended can favor infiltration into plant tissues and their internalization. This process can be enhanced by hydrostatic pressure created by immersing warm fruits in cold wash water. Internalization and infiltration have been well described for human pathogens but also occur for spoilage microorganisms (91). Because of an increased association of vegetables with outbreaks of foodborne poisonings, research about the contamination routes (soil to the aerial parts of the plants, contamination through natural and accidental openings, interactions with the plant tissue, etc.) of enteric pathogenic bacteria such as Escherichia coli O157:H7 and Salmonella has been strongly stimulated (24, 130). Some of the mechanisms described for foodborne pathogens could also be relevant for the specific plant pathogens.

CAUSES OF DISEASE OR SPOILAGE Specific factors associated with virulence of microorganisms are the primary mechanisms at the origin of postharvest spoilage of fruits and vegetables. Their role is usually evident when comparing, within species, certain strains able to cause spoilage to those unable to cause spoilage. A comparison of different strains of Pseudomonas fluorescens has shown, for instance, that the production of a biosurfactant (a peptidolipid named viscosin) is a key factor in the decay of broccoli florets caused by the bacterium, while pectolytic activity is necessary but not sufficient to cause spoilage (51, 52). The aggressiveness of Mycocentrospora acerina is related to the production of pectinase and glucanase (73). Molecular biology techniques enable the production of mutants with specific characteristics and tests to determine if these characteristics (and their genes) are necessary for microorganisms to cause spoilage. This approach has enabled researchers to demonstrate the role of pectinases in pathogenesis of soft-rot bacteria. Mutants of the soft-rot bacterium Erwinia chrysanthemi that do not produce pectin methylesterase or a specific isoenzyme of pectate lyase have a reduced ability to macerate potato tubers, whereas the production of another pectate lyase does not seem to be necessary for tissue maceration

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

192 (4). These enzymes also have different implications in pathogenesis. Pectic enzymes are clearly involved in postharvest spoilage of fruits and vegetables by bacteria as well as by molds. Purified pectin-degrading enzymes are able to macerate plant tissue and cause cell death without involving other enzymes or toxic factors. Pectinases break down pectic components in the middle lamella and cell wall, resulting in tissue maceration, loss of rigidity of plant tissue, and nonreversible cell damage. Polygalacturonase, pectin esterase, and pectin lyase are the main types of pectinases that act at different sites on the d-galacturonic acid chain constituting the basic structure of pectic compounds (47). Pectolytic bacteria and molds generally produce several types of pectinases, not always playing the same role (4). Eight isozymes of polygalacturonase are produced by Phomopsis cucurbitae, a quiescent infection mold, during postharvest infection and decay of cantaloupe, and their activity changes during maturation of the fruit (144). An endopolygalacturonase mutant of Alternaria citri loses its ability to cause black rot on citrus, while a mutant of Alternaria alternata for the same function is still able to cause brown spot on citrus (56). Pectic enzymes are necessary for A. citri to progress from the pedicel in the central axis of the citrus fruit to the sac juice containing nutrients (57). There is a fine regulation of the (at least) six endopolygalacturonase genes of B. cinerea, implicating complex signaling pathways that allow adaptation to a wide range of possible hosts (137). Some postharvest spoilage bacteria and molds also produce toxins. Some molds, in particular Alternaria spp., have very specific host-parasite interactions implicating toxins (82). Mycotoxins are also known for their toxicity to humans or animals. Implications of the nonspecific toxins produced by molds causing postharvest spoilage are highly variable (3). For instance, production of patulin by Penicillium expansum does not appear to be involved in postharvest spoilage of apples during storage. In contrast, the production of oxalic acid is strongly implicated in the infection process of a number of postharvest pathogens such as B. cinerea, Sclerotium rolfsii, or S. sclerotiorum (50, 112, 137). Oxalic acid reduces pH to values optimal for enzymatic activity and sequesters Ca2+ ions from the cell wall, thereby stimulating pectin hydrolysis. Oxalic acid may also suppress the oxidative burst and prevent the progression of an infection by a local accumulation of reactive oxygen species (ROS). In some instances, spoilage can be fully opportunistic. The physiological activity of minimally processed vegetables is markedly changed by processing and stor-

SMP_Food microbiology_CH08.indd

192

age conditions, e.g., modified atmosphere packaging (100, 113). The lactic acid bacterium Leuconostoc mesenteroides is not known as a plant pathogen but has been shown to be strongly associated with the spoilage of shredded carrots. In this situation, spoilage is thought to be due to a shift toward anaerobic metabolism in modified atmospheres with concentrations of CO2 in excess and/or to low concentrations of O2 that induce toxicity to carrot cells and leakage of electrolytes and nutrients used by the saprophyte (17). On minimally processed green leafy salads, in contrast, spoilage can be explained by the development of pectolytic fluorescent pseudomonads.

DEFENSE REACTIONS Fruits and vegetables offer a range of barriers to infection by postharvest spoilage microorganisms. Some are preformed or constitutive in the plant organ. The cuticle barrier is the most external barrier to penetration (115). Removal of waxes from the cuticle has been shown to increase the vulnerability of pepper fruits to infection by Colletotrichum capsici and C. gloeosporioides and of cabbage by B. cinerea (99). The cuticle thickness is correlated with the resistance of tomato fruit or grape berries to B. cinerea and of peaches to Monilinia fructigena (3, 43). A “delayed-deterioration” tomato mutant with an increased accumulation of cutin shows a higher resistance to B. cinerea (115). In many types of produce, the barrier effect of the cuticle is reinforced by epidermis or peridermis tissues, which can be relatively thick structures such as the rind of citrus and melons. Enhanced resistance to infection linked to these structures can be explained by their higher resistance to crack formation and therefore to penetration of spoilage microorganisms, higher amounts of protective material to be degraded, and lower diffusion or access to water and nutrients required for the infection process (3). Preformed antimicrobial compounds, designated “phytoanticipins,” may also be involved in plant resistance, although demonstration of their actual effects on resistance is relatively hard to achieve because of difficulties in assessing inhibitory activity and in correlating changes in concentrations with decay development (7, 109). These compounds can be extremely diverse, as fruits and vegetables cover a wide range of plant families. The presence of the alkaloidal sap­ onin tomatine has a role in the resistance of tomato to B. cinerea (3). Lower concentrations of 5,12-cis­heptadecenyl resorcinol and 5-pentadecenyl resorcinol in the skin of mango fruit during ripening are related to an increase in susceptibility to C. gloeosporioides and

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

193

A. alternata, higher concentrations occurring in the most resistant cultivars (49, 109). Quiescent infections of C. gloeosporioides may be regulated by preformed epicatechin acting as an inhibitor of lipoxygenase activity and consequently delaying the degradation of an antifungal diene present in unripe avocado fruit (46). A monoterpene aldehyde, citral, present in particular in the oil cavities of citrus albedo, is thought to be involved in the resistance of young mature green lemons to Penicillium (109). A wide range of defensive barriers may also be formed in reaction to an infection and constitute a plant immune system (61). Pathogens produce pathogen-associated molecular pattern elicitors, which elicit plant defense reactions. Plant tissues may produce small molecules of varied nature, known as phytoalexins (87). Bramley’s Seedling apples produce benzoic acid in response to the infection by Nectria galligena (129), and the resistance of carrot to B. cinerea has been attributed to a coumarin,6-methoxymellein (45). Structural changes such as accumulation of lignin or suberin or development of callus have also been observed as defensive reactions (117), particularly in carrot, potato tuber, and pear (6, 77, 126). Polygalacturonase-inhibiting proteins produced by plants and acting against endopolygalacturonases of plant-pathogenic molds that cause wall degradation and tissue maceration have been detected in a range of fruits and vegetables, including apple, pear, grape, raspberry, onion, and pepper (27). An esterase produced during the interaction between mature peppers and C. gloeosporioides has been shown to inhibit the formation of the fungal appressoria and therefore decay (65). A production of ROS (including H2O2) occurs with the oxidative burst that follows infection. In pepper infected by C. gloeosporioides, the direct inhibitory effect of ROS on the pathogen still remains unclear, but ROS likely are at the origin of activation of the phenylpropanoid pathway and accumulation of the antifungal compound diene, both implicated in resistance (8). Other mechanisms implicating enzymic activities (chitinases and lipoxygenases) or accumulation of hydroxyproline-rich proteins have been proposed (3, 99). Interactions between plants and pathogens are a very fertile research area, and novel defense mechanisms are regularly discovered.

CONTROLLING SPOILAGE Postharvest control of temperature, relative humidity, and composition of the gaseous atmosphere can reduce the physiological activity of fruits and vegetables, thereby delaying ripening and senescence and consequently prolonging shelf life. These environmental fac-

SMP_Food microbiology_CH08.indd

193

tors may act by giving less opportunity for the pathogen to develop by retaining the integrity of the plant organ and by directly inhibiting microbial growth. The most suitable temperature, relative humidity, and modified atmosphere for preserving the quality of most fruits and vegetables are now relatively well established and have been extensively reviewed (3, 97, 119). A decrease in temperature by 10°C reduces the respiratory activity 2- to 4-fold (63), and temperature close to 0°C is recommended for most commodities, with the exception of those of tropical origin and a few temperate produce items that suffer from physiological disorders (chilling injuries) when stored at refrigeration temperatures. For these products, the optimal storage temperature is close to 10°C. While growth of many postharvest spoilage microorganisms is still possible at very low refrigeration temperature, their rate of development is reduced (18, 106). In addition, some major spoilage microorganisms are inhibited at refrigeration temperatures at which produce is often stored. The molds Rhizopus stolonifer, Phytophthora infestans, and Aspergillus niger cannot grow below 2 to 5°C, 4°C, and 11°C, respectively (3, 70). Modified atmospheres, used in combination to chill storage, also reduce the physiological activity of fresh produce. Recommended CO2 concentrations for produce storage rarely exceed 10%, and 1 to 5% O2 is tolerated (119). Exposure to higher CO2 (lower O2) levels may result in physiological disorders, leading to loss in quality. As reviewed by El Goorani and Sommer (36), either reduction in O2 or increase in CO2 delays in vitro growth of many postharvest pathogens, without complete inhibition. Modified atmospheres reduce microbial spoilage of fruits and vegetables in many instances, although some spoilage microorganisms are not directly inhibited. For example, controlled atmosphere storage of apples at 1°C in an atmosphere containing 5% CO2 and 3% O2 prevents the development of lesions due to Pezicula alba, while in vitro this gaseous atmosphere has no effect (11). Generally, high concentrations of CO2 or low concentrations of O2, often less than concentrations tolerated by the plant organs, are needed for a significant reduction of in vitro growth. The lower susceptibility to postharvest pathogens of fruits and vegetables stored under controlled atmospheres is mainly due to delayed senescence (36). In contrast, however, modified atmospheres can increase the extent of diseases in potato tubers, carrot, and other root crops. Reduced physiological activity slows wound healing, giving pathogens additional time to establish infections. Water loss is a consequence of respiratory activity and is highly detrimental to quality. A 10% weight loss

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

194 (much less for leafy vegetables) makes produce unacceptable to consumers (9). Storage at relative humidity higher than 90% is recommended for most commodities, with a few exceptions, e.g., garlic and onion (97). High humidity also increases the availability of nutrients to spoilage microorganisms, favoring their survival and their germination (127). Despite this, storage under high humidity is generally favorable for produce quality. Storage of cabbage, celery, leek, and carrot at 98 to 100% relative humidity instead of 90 to 95% relative humidity results in lower losses caused by decay (134). Thorne (132) observed that spoilage due to R. stolonifer only occurs on carrots that have lost more than 3 to 8% of their fresh weight. The possibility of using ionizing radiation to extend the shelf life of fruits and vegetables has been studied since the 1950s (3). Postharvest spoilage bacteria and fungi are sensitive to ionizing radiation. Doses lower than 4 kGy reduce 1,000-fold the germination of major postharvest pathogens such as Penicillium spp., Monilinia fructicola, R. stolonifer, and Alternaria spp. Decimal reduction doses for Pseudomonas and other gram-negative bacteria are approximately 0.2 kGy (3, 104). However, it appears that some fruits and vegetables are adversely affected (by tissue softening, for instance) by doses necessary to inactivate some postharvest pathogens (71). Treatment with ionizing radiation is consequently useful to lower initial contamination or to inhibit the growth of postharvest pathogens and to delay the development of disease, but not for complete elimination of microbial contaminants. In the case of strawberry, a 2-kGy dose prolongs shelf life by several days (3). Prestorage heat treatment has been used with some success to reduce postharvest spoilage of a wide range of fresh produce, including green pepper, apple, and citrus (40, 121). Two main types of applications can be distinguished: short-term exposure to heat, from a few seconds at 60 to 62°C to 60 to 120 min at about 45°C, by immersion in or rinsing and spraying with hot water; or long-term exposure, also known as “curing” (for a few hours to several days), mainly in hot air (121). Two kinds of effects can be observed, either direct effects on microbial contaminants or indirect effects on the treated fruit or vegetable. Treatment of tomatoes for 3 days at 38°C before storage at 20°C for 7 days results in inhibition of B. cinerea without alteration of quality (41). Short-time exposure at 56°C delays germination of Penicillium digitatum, and treatment at 59°C and 62°C inhibits germination (107). A similar level of sensitivity to heat has been observed for M. fructicola, B. cinerea, Cladosporium herbarum, R. stolonifer, A. alternata,

SMP_Food microbiology_CH08.indd

194

and Penicillium expansum (3). These treatments also reduce 1,000-fold the natural epiphytic microflora on citrus. On whole citrus, hot water treatment of fruits artificially inoculated with P. digitatum has been shown to result in less than 20% decay after 4 days at 24°C, while all control fruits are spoiled (107).

Chemical Treatments Fungicides

Synthetic antimicrobial chemicals are still widely applied to fruits and vegetables after harvest, with substantial differences between countries, according to approval by legal authorities, as complements to modifications of storage environments or when modifications are not possible. Their spectrum of activity and possibilities of applications have been reviewed elsewhere (31, 32, 89), and the list of authorized molecules is updated by periodic regulation reviews with the objectives of applying “reduced-risk” fungicides. These chemical compounds have different modes of action. Imazalil, a molecule widely used against Penicillium on citrus, belongs to a family of systemic fungicides that inhibit the biosynthesis of ergosterol, an essential compound in the membrane of fungal cells (3). Dicarboximides alter the osmotic adaptation capabilities of molds (141). The recently authorized azoxystrobin and fludioxonil act on fungal spore germination (64). Concerns about toxicity limit official authorization for use of fungicides on defined commodities and create some negative attitudes among consumers. Application strategies combining, for instance, heat and fungicide treatments are being increasingly developed to use fewer products and leave fewer residues (120). In addition, the occurrence of fungal resistance reduces the efficiency and restricts possible applications of fungicides. Alternative chemicals such as Generally Recognized As Safe or natural compounds, e.g., acetaldehyde, hexanal, and essential oils, have also shown antimicrobial properties suitable for postharvest control of phytopathogens (72). The presence of the essential oil carvone in the storage atmosphere has been shown to reduce, for instance, decay of potato tubers inoculated with Phoma exigua and Fusarium sulphureum but not with Fusarium solani (48).

Decontamination

Decontamination aims at reducing the number of microbial contaminants on the surface of fruits and vegetables, thereby prolonging the time required to develop spoilage. Various chemicals have been tested for possible application to fruits and vegetables. Hypochlorous acid (HClO), chlorine dioxide (ClO2), ozone, and hydrogen

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

195

peroxide are among the most extensively studied chemicals, while peroxyacetic acid and electrolyzed water are receiving an emerging interest (2, 10). Limitations in their action, in particular that of chlorine, may be due to various factors: the natural hydrophobicity of plant surfaces, inaccessibility of microorganisms within plant tissue, and possible neutralization of the decontaminating agent upon contact with fruit and vegetable tissue components. Chlorine at concentrations that will cause several decimal reductions of pathogens in pure water is often slightly more efficient than washing produce with pure water (100). Reduction in microbial populations is generally 1 to 2 log CFU/g. However, despite these limitations, application of disinfectants may result in better retention of quality during storage. For example, prestorage chlorine application at suitable concentrations and for appropriate exposure times results in significant reductions of black spot disease caused by A. alternata on persimmon (110) and spoilage of nectarines and plums caused by Monilinia laxa (88). Sanitizers are also useful to prevent buildup of microorganisms on equipment and in wash water and to avoid dissemination from contaminated material.

Triggering Defense Reactions

As stated above, fruits and vegetables, as any plant organ, have intrinsic defense mechanisms that respond to postharvest microbial infection. The induction of these mechanisms may be systemic, as shown for carrot roots, in which inoculation with B. cinerea and S. sclerotiorum near the root tip reduces infection after subsequent inoculation with these molds near the crown (92). A range of biotic and abiotic factors that induce defense reactions have been shown to improve quality retention during storage. This induction of a beneficial plant response by sublethal or low doses of chemical or physical elicitors is sometimes designated “plant hormesis” (131). Appropriate fertilization of plants with calcium or direct application of a calcium solution to fruits and vegetables to be stored increases the calcium content in tissue. These treatments have been successful in controlling the growth of P. expansum on apples and soft rot caused by E. carotovora var. atroseptica on potatoes. The effect of calcium is attributed to a reinforcement of pectin bonds, making the cell wall more resistant to pectic enzymes, or to a direct inhibitory effect of calcium on the enzymes themselves (22). Accelerating healing of wounds by suberization of parenchyma cells as a result of exposure at moderately high, ambient, or low temperatures (a process also known as curing) is a common procedure to reduce postharvest spoilage of carrot, potato, and sweet potato (38). Heat treatment

SMP_Food microbiology_CH08.indd

195

may also induce structural changes in the epicuticular wax, resulting in fewer cracks, thereby providing a better mechanical barrier to spoilage microorganisms (121). Heat treatments may also delay ripening, inhibit antifungal activity, and stimulate phytoalexin production or chitinase and glucanase activities, which play a role in degradation (121). Chitosan, a compound derived from chitin, has antimicrobial properties when used, for instance, to coat fresh fruits and vegetables for the purpose of regulating gas and moisture exchange (28, 138). In addition to damaging fungal hyphae, chitosan also reduces the ability of fungal pectic enzymes to macerate plant tissue through a direct effect on enzymes and through indirect effects, by inhibiting the production of oxalic and fumaric acids by pathogens. Chitosan may also inhibit host-specific toxin production, phytoalexin production, and structural changes in the cell wall, as shown in tomatoes infected by the black rot mold, A. alternata, or on bell pepper fruit infected by B. cinerea (33, 34, 114). Stimulation of antifungal hydrolyses by chitinases and glucanases has also been reported (138). Plant regulators are involved in the development and response of produce to environmental stresses. Jasmonates, in particular, may play a role in plant defense responses to microbial attacks, and their postharvest application on fruits is effective in controlling spoilage molds such as M. fructicola and P. expansum on peaches and P. digitatum on grapefruit (29, 142). This positive effect has been attributed to induced resistance of the fruit, in particular to higher synthesis and activity of pathogenesis-related ­proteins such as chitinase, glucanase, phenylalanine ammonia lyase, and peroxidase (142). Similar activities or accumulations of antimicrobial compounds related to ­ disease resistance (phytoalexins) are induced in peach, grapefruit, and carrot after exposure to UV-C (19). However, UV-C has little activity against B. cinerea inoculated on wounded bell pepper, indicating that wounded tissues are protected against exposure to irradiation, while complete inhibition has been observed in vitro (93). In this case and in grapefruit inoculated by P. digitatum, the induction of defense reactions is the likely cause of a lower rate of decay.

Resistance of Cultivars

Cultivars within a plant species may exhibit differences in susceptibility to infection by postharvest pathogens. This is a basis for plant breeding for genotypes resistant to postharvest pathogens, a common practice, particularly with apples (62). Tests using artificial inoculation of produce show some differences in susceptibility among cultivars, e.g., susceptibility of broccoli to Pseudomonas

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

196 marginalis (16), onion to Aspergillus niger and Botrytis allii (66, 78), pepper to C. gloeosporioides (86), and sweet potato to Rhizopus spp. (20). Spoilage of these resistant cultivars is usually only delayed, not prevented (99). The resistance of cultivars to postharvest spoilage microorganisms may be the result of several complex factors and interactions. A low number of pores and a thick external grape berry skin, for example, are positively correlated with resistance to B. cinerea, both factors likely giving a general protection to fungal attack (43). In contrast, among 12 apple cultivars tested for resistance against B. cinerea, P. expansum, Mucor piriformis, and Pezicula malicorticis, no single cultivar was the most resistant to all pathogens and each cultivar that was the most resistant to one pathogen was also the most susceptible to one of the other pathogens. It has been suggested that Golden Delicious cultivar apples, despite their fragile epidermis, have a lower probability of wounding than Granny Smith cultivar apples because of the particularity of their pedicels, stiffer and shorter in the latter case (125). Genetic engineering of fruit, particularly targeting ethylene metabolism, which controls ripening, and cell wall components, which determine fruit firmness, or cuticle, which determines skin integrity, may limit microbial infection (90). For example, increased firmness in transgenic tomatoes with reduced levels of polygalacturonase results in increased resistance to Geotrichum candidum and R. stolonifer, while tomatoes with suppressed expression of ripening-related expansin, with a similar increased firmness, are not more resistant to B. cinerea and A. alternata (15, 69).

Biological Control

Biological control refers to the application of microbial antagonists of postharvest spoilage microorganisms, in particular on wounds, which are a natural site of entry for pathogens. Biological control is based on the selection of naturally occurring microorganisms particularly adapted to surviving or growing on fruit and vegetable surfaces. Many mechanisms are probably involved in interactions between antagonists and pathogens (30, 59). These may include antibiosis, such as the production of pyrrolnitrin by Pseudomonas cepacia, which has been shown to control blue mold rot caused by P. expansum and gray mold rot caused by B. cinerea on pome fruits (59). Competition for space and nutrients is a seductive hypothesis to account for the antagonistic effect of some biological agents, in particular on wound sites, which are often rich in nutrients. For instance the yeast-like Aureobasidium pullulans depletes amino acids in vitro and inhibits germination of P. expansum in apple juice (60). In addition to this possible competition for space

SMP_Food microbiology_CH08.indd

196

and nutrients, the antagonist A. pullulans induces apple -1,3-glucanase, chitinase, and peroxydase activity, which controls decay caused by B. cinerea and P. expansum (55). A similar induction process has been shown with the antagonist yeast Candida saitoana (35). Induction of these defense reactions could explain the effect of antagonists against postharvest pathogens. Some yeasts also show a strong attachment to the hyphae of postharvest spoilage molds and production of cell wall-degrading enzymes (139). Some of these biocontrol agents, in particular, strains of the yeasts Metschnikowia fructicola and Candida oleophila and of the bacterium Pseudomonas syringae, are registered as biopesticides in different countries, with varying degrees of success (30). The diversity of biocontrol methods illustrates the diversity of possibilities based on direct control of pathogen contamination or of its development, indirect control through delayed senescence and preserved integrity of the organ host, or induced defense mechanisms. Under practical conditions, the delay of spoilage depends on combinations of several factors: prevention of wounding during harvest and postharvest handling, cold storage together with modified atmospheres, and application of fungicides for many commodities, e.g., apples and pears, which can be successfully stored for several months and therefore offered for sale to the consumer long after harvest. Recent research has also revealed the possibility of other combinations, such as postharvest heat treatment and use of antagonists on apples followed by cold storage to control P. expansum and Colletotrichum acutatum (21, 58, 76). In the near future, integrated approaches, combining control along the production chain at various preharvest stages (to control the initial inoculum for instance) and postharvest stages (to protect natural defenses, to stimulate defense reactions of the produce, and/or to reduce the pathogen development), will probably be increasingly used. This has been successfully implemented for the management of gray mold caused by B. cinerea on kiwi fruit in New Zealand (30, 94). Similarly, the sequential application of steam, an aqueous decontamination solution, and an antagonistic yeast synergistically reduced the incidence of the fungal black root rot on carrot (39).

CONCLUSION Spoilage of fruits and vegetables is the result of complex interactions between a living plant organ and its microflora and therefore deals with plant pathology and plant physiology as much as with food microbiology. Control of postharvest spoilage microorganisms largely accounts for these interactions, which could be

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

197

additionally affected by the global climate change (95, 103). Improving the quality retention of fruits and vegetables will raise for a long time some exciting and difficult questions about microbial ecology impacting the survival and development of microorganisms in the environment, in particular during production, and on the plant organ to be stored and later consumed. The future global exchange of fruits and vegetables will likely not decrease. At the present time, millions of tons of fruits and vegetables are crossing seas, oceans, and continents, from the Southern to the Northern hemisphere and from tropical to temperate zones. Developing countries increasingly play a role in this world market. Further development of minimally processed fruits and vegetables will bring new questions as to how to maintain the quality of processed produce, which requires prevention of spoilage, when the produce is often heavily stressed and naturally occurring defenses of the intact tissues have been overwhelmed. Consumer demand for high-quality fruits and vegetables produced under environmentally friendly conditions will probably not decrease. Finding solutions to historical problems associated with the preservation of fruits and vegetables against infection and spoilage by bacteria and fungi will be an ongoing future challenge.

References 1. Amiri, A., and G. Bompeix. 2005. Diversity and population dynamics of Penicillium spp. on apples in pre- and postharvest environments: consequences for decay development. Plant Pathol. 54:74–81. 2. Artés, F., P. Gómez, E. Aguayo, V. Escalona, and F. Artés-Hernández. 2009. Sustainable sanitation techniques for keeping quality and safety of fresh-cut plant commodities. Postharv. Biol. Technol. 51:287–296. 3. Barkai-Golan, R. 2001. Postharvest Diseases of Fruits and Vegetables. Development and Control. Elsevier, Amsterdam, The Netherlands. 4. Barras, F., F. Vangijsegem, and A. K. Chatterjee. 1994. Extracellular enzymes and pathogenesis of soft-rot Erwinia. Annu. Rev. Phytopathol. 32:201–234. 5. Barry, C. S., and J. J. Giovannoni. 2007. Ethylene and fruit ripening. J. Plant Growth Regul. 26:143–159. 6. Bartz, J. A., and J. W. Eckert. 1987. Bacterial diseases of vegetable crops after harvest, p. 351–376. In J. Weichmann (ed.), Postharvest Physiology of Vegetables. Marcel Dekker, New York, NY. 7. Bednarek, P., and A. Osbourn. 2009. Plant-microbe interactions: chemical diversity in plant defense. Science 324:746–748. 8. Beno-Moualem, D., and D. Prusky. 2000. Early events during quiescent infection development by Colletotrichum gloeosporioides in unripe avocado fruits. Phytopathology 90:553–559.

SMP_Food microbiology_CH08.indd

197

9. Ben-Yehoshua, S. 1987. Transpiration, water stress, and gas exchange, p. 113–170. In J. Weichmann (ed.), Postharvest Physiology of Vegetables. Marcel Dekker, New York, NY.   10. Beuchat, L. R. 1998. Surface decontamination of fruits and vegetables eaten raw. WHO/FSF/FOS/98.2. http:// www.who.int/foodsafety/publications/fs_management/ surfac_decon/en/index.html.   11. Bompeix, G. 1978. The comparative development of Pezicula alba and P. malicortis on apples and in vitro (air and controlled atmosphere). Phytopathol. Z. 91:97–109.   12. Boudet, A. M. 2007. Evolution and current status of research in phenolic compounds. Phytochemistry 68:2722–2735.   13. Boureau, T., M. A. Jacques, R. Berruyer, Y. Dessaux, H. Dominguez, and C. E. Morris. 2004. Comparison of the phenotypes and genotypes of biofilm and solitary epiphytic bacterial populations on broad-leaved endive. Microb. Ecol. 47:87–95.   14. Bowes, B. G., and J. D. Mauseth. 2008. Plant Structure— A Colour Guide, 2nd ed. Manson Publishing Ltd., London, United Kingdom.   15. Brummell, D. A., W. J. Howie, C. Ma, and P. Dunsmuir. 2002. Postharvest fruit quality of transgenic tomatoes suppressed in expression of a ripening-related expansin. Postharv. Biol. Technol. 25:209–220.   16. Canaday, C. H., J. E. Wyatt, and J. A. Mullins. 1991. Resistance in broccoli to bacterial soft rot caused by Pseudomonas marginalis and fluorescent Pseudomonas species. Plant Dis. 75:715–720.   17. Carlin, F., C. Nguyen-the, Y. Chambroy, and M. Reich. 1990. Effects of controlled atmospheres on microbial spoilage, electrolyte leakage and sugar content of ‘freshready-to-use’ grated carrots. Int. J. Food Sci. Technol. 25:110–119.   18. Casals, C., I. Vinas, R. Torres, C. Griera, and J. Usall. 2010. Effect of temperature and water activity on in vitro germination of Monilinia spp. J. Appl. Microbiol. 108:47–54.   19. Charles, M. T., and J. Arul. 2007. UV treatment of fesh fruits and vegetables for improved quality: a status report. Stewart Postharv. Rev. 3:1–8.   20. Clark, C. A., and M. W. Hoy. 1994. Identification of resistance in sweetpotato to Rhizopus soft rot using two inoculation methods. Plant Dis. 78:1078–1082.   21. Conway, W. S., B. Leverentz, W. J. Janisiewicz, R. A. Saftner, and M. J. Camp. 2005. Improving biocontrol using antagonist mixtures with heat and/or sodium bicarbonate to control postharvest decay of apple fruit. Postharv. Biol. Technol. 36:235–244.   22. Conway, W. S., C. E. Sams, and A. Kelman. 1994. Enhancing the natural resistance of plant tissues to postharvest diseases through calcium applications. HortScience 29:751–754.   23. Cook, R. L. 2002. The U.S. fresh produce industry: an industry in transition, p. 5–30. In A. A. Kader (ed.), Postharvest Technology of Horticultural Crops. University of California, Agriculture and Natural Resources, Oakland, CA.

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

198   24. Critzer, F. J., and M. P. Doyle. 2010. Microbial ecology of foodborne pathogens associated with produce. Curr. Opin. Biotechnol. 21:125–130.   25. Deak, T. 1991. Foodborne yeasts. Adv. Appl. Microbiol. 36:179–278.   26. de Kok, T. M., S. G. van Breda, and M. M. Manson. 2008. Mechanisms of combined action of different chemopreventive dietary compounds. Eur. J. Nutr. 47:51–59.   27. De Lorenzo, G., R. D’Ovidio, and F. Cervone. 2001. The role of polygalacturonase-inhibiting proteins (PGIPs) in defense against pathogenic fungi. Annu. Rev. Phytopathol. 39:313–335.   28. Devlieghere, F., A. Vermeulen, and J. Debevere. 2004. Chitosan: antimicrobial activity, interactions with food components and applicability as a coating on fruit and vegetables. Food Microbiol. 21:703–714.   29. Droby, S., R. Porat, L. Cohen, B. Weiss, B. Shapiro, S. Philosoph-Hadas, and S. Meir. 1999. Suppressing green mold decay in grapefruit with postharvest jasmonate application. J. Am. Soc. Hort. Sci. 124:184–188.   30. Droby, S., M. Wisniewski, D. Macarisin, and C. Wilson. 2009. Twenty years of postharvest biocontrol research: is it time for a new paradigm? Postharv. Biol. Technol. 52:137–145.   31. Eckert, J. W., and J. M. Ogawa. 1988. The chemical control of postharvest diseases: deciduous fruits, berries, vegetables, and root/tuber crops. Annu. Rev. Phytopathol. 26:433–469.   32. Eckert, J. W., and J. M. Ogawa. 1985. The chemical control of postharvest diseases: subtropical and tropical fruits. Annu. Rev. Phytopathol. 23:421–454.   33. El Ghaouth, A., J. Arul, C. Wilson, and N. Benhamou. 1997. Biochemical and cytochemical aspects of the interactions of chitosan and Botrytis cinerea in bell pepper fruit. Postharv. Biol. Technol. 12:183–194.   34. El Ghaouth, A., J. Arul, C. Wilson, and N. Benhamou. 1994. Ultrastructural and cytochemical aspects of the effect of chitosan on decay of bell pepper fruit. Physiol. Mol. Plant Pathol. 44:417–432.   35. El Ghaouth, A., C. L. Wilson, and M. Wisniewski. 2003. Control of postharvest decay of apple fruit with Candida saitoana and induction of defense responses. Phytopathology 93:344–348.   36. El Goorani, M. A., and N. F. Sommer. 1981. Effects of modified atmospheres on postharvest pathogens of fruits and vegetables. Hort. Rev. 3:412–461.   37. Elliot, M. A., and N. J. Talbot. 2004. Building filaments in the air: aerial morphogenesis in bacteria and fungi. Curr. Opin. Microbiol. 7:594–601.   38. Eshel, D. 2011. Non-chemical approaches for postharvest quality management of underground vegetables. Stewart Postharv. Rev. 7:1–7.   39. Eshel, D., R. Regev, J. Orenstein, S. Droby, and S. GanMor. 2009. Combining physical, chemical and biological methods for synergistic control of postharvest diseases: a case study of Black Root Rot of carrot. Postharv. Biol. Technol. 54:48–52.

SMP_Food microbiology_CH08.indd

198

  40. Fallik, E. 2004. Prestorage hot water treatments (immersion, rinsing and brushing). Postharv. Biol. Technol. 32:125–134.   41. Fallik, E., J. Klein, S. Gringerg, E. Lomaniec, S. Lurie, and A. Lalazar. 1993. Effect of postharvest heat treatment of tomatoes on fruit ripening and decay caused by Botrytis cinerea. Plant Dis. 77:985–988.   42. Flaishman, M. A., and P. E. Kolattukudy. 1994. Timing of fungal invasion using hosts ripening hormone as a signal. Proc. Natl. Ac. Sci. USA 91:6579–6583.   43. Gabler, F. M., J. L. Smilanick, M. Mansour, D. W. Ramming, and B. E. Mackey. 2003. Correlations of morphological, anatomical, and chemical features of grape berries with resistance to Botrytis cinerea. Phytopathology 93:1263–1273.   44. Glenn, G. M., B.-S. Chiou, S. Imam, D. Wood, and W. Orts. 2005. Role of cuticles in produce quality and preservation, p. 19–53. In O. Lamikanra, S. Iman, and D. Ukuku (ed.), Produce Degradation—Pathways and Prevention. Taylor and Francis Group, Boca Raton, FL.   45. Goodliffe, J. P., and J. B. Heale. 1978. The role of 6methoxy mellein in the resistance and susceptibility of carrot root tissue to the cold-storage pathogen Botrytis cinerea. Physiol. Plant Pathol. 12:27–43.   46. Guetsky, R., I. Kobiler, X. Wang, N. Perlman, N. Gollop, G. Avila-Quezada, I. Hadar, and D. Prusky. 2005. Metabolism of the flavonoid epicatechin by laccase of Colletotrichum gloeosporioides and its effect on pathogenicity on avocado fruits. Phytopathology 95:1341–1348.   47. Gummadi, S. N., and T. Panda. 2003. Purification and biochemical properties of microbial pectinases—a review. Process Biochem. 38:987–996.   48. Hartmans, K. J., P. Diepenhorst, W. Bakker, and L. G. M. Gorris. 1995. The use of carvone in agriculture: sprout suppression of potatoes and antifungal activity against potato tuber and other plant diseases. Ind. Crops Prod. 4:3–13.   49. Hassan, M. K., E. K. Dann, D. E. Irving, and L. M. Coates. 2007. Concentrations of constitutive alk(en)ylresorcinols in peel of commercial mango varieties and resistance to postharvest anthracnose. Physiol. Mol. Plant Pathol. 71:158–165.   50. Hegedus, D. D., and S. R. Rimmer. 2005. Sclerotinia sclerotiorum: when “to be or not to be” a pathogen? FEMS Microbiol. Lett. 251:177–184.   51. Hernandez-Anguiano, A. M., T. V. Suslow, L. Leloup, and C. I. Kado. 2004. Biosurfactants produced by Pseudomonas fluorescens and soft-rotting of harvested florets of broccoli and cauliflower. Plant Pathol. 53:596–601.   52. Hildebrand, P. D., P. G. Braun, K. B. McRae, and X. Lu. 1998. Role of the biosurfactant viscosin in broccoli head rot caused by a pectolytic strain of Pseudomonas fluorescens. Can. J. Plant Pathol. 20:296–303.   53. Hong, C. X., T. J. Michailides, and B. A. Holtz. 1998. Effects of wounding, inoculum density, and biological

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

  54.

  55.

  56.

  57.

  58.

  59.   60.

  61.   62.

63.

  64.

  65.

  66.

199

control agents on postharvest brown rot of stone fruits. Plant Dis. 82:1210–1216. Imam, S., J. Shey, D. Wood, G. Glenn, B.-S. Chiou, M. Inglesby, C. Ludvik, A. Klamczynski, and W. Orts. 2005. Structure and function of complex carbohydrates in produce and their degradation process, p. 563–597. In O. Lamikanra, S. Iman, and D. Ukuku (ed.), Produce Degradation—Pathways and Prevention. Taylor and Francis Group, Boca Raton, FL. Ippolito, A., A. El Ghaouth, C. L. Wilson, and M. Wisniewski. 2000. Control of postharvest decay of apple fruit by Aureobasidium pullulans and induction of defense responses. Postharv. Biol. Technol. 19:265–272. Isshiki, A., K. Akimitsu, M. Yamamoto, and H. Yamamoto. 2001. Endopolygalacturonase is essential for citrus black rot caused by Alternaria citri but not brown spot caused by Alternaria alternata. Mol. PlantMicrobe Interact. 14:749–757. Isshiki, A., K. Ohtani, M. Kyo, H. Yamamoto, and K. Akimitsu. 2003. Green fluorescent detection of fungal colonization and endopolygalacturonase gene expression in the interaction of Alternaria citri with citrus. Phytopathology 93:768–773. Janisiewicz, W. J., and W. S. Conway. 2010. Combining biological control with physical and chemical treatments to control fruit decay after harvest. Stewart Postharv. Rev. 6:1–16. Janisiewicz, W. J., and L. Korsten. 2002. Biological control of postharvest diseases of fruits. Annu. Rev. Phytopathol. 40:411–441. Janisiewicz, W. J., T. J. Tworkoski, and C. Sharer. 2000. Characterizing the mechanism of biological control of postharvest diseases on fruits with a simple method to study competition for nutrients. Phytopathology 90:1196–1200. Jones, J. D., and J. L. Dangl. 2006. The plant immune system. Nature 444:323–329. Jurick, W. M., W. J. Janisiewicz, R. A. Saftner, I. Vico, V. L. Gaskins, E. Park, P. L. Forsline, G. Fazio, and W. S. Conway. 2011. Identification of wild apple germplasm (Malus spp.) accessions with resistance to the postharvest decay pathogens Penicillium expansum and Colletotrichum acutatum. Plant Breed. 130:481–486. Kader, A. A. 2002. Postharvest biology and technology: an overview, p. 39–48. In A. A. Kader (ed.), Postharvest Tech­ nology of Horticultural Crops. University of California, Agriculture and Natural Resources, Oakland, CA. Kanetis, L., H. Forster, and J. E. Adaskaveg. 2007. Comparative efficacy of the new postharvest fungicides azoxystrobin, fludioxonil, and pyrimethanil for managing citrus green mold. Plant Dis. 91:1502–1511. Kim, Y. S., H. H. Lee, M. K. Ko, C. E. Song, C. Y. Bae, Y. H. Lee, and B. J. Oh. 2001. Inhibition of fungal appressorium formation by pepper (Capsicum annuum) esterase. Mol. Plant-Microbe Interact. 14:80–85. Ko, S. S., J. W. Huang, J. F. Wang, S. Shanmugasundaram, and W. N. Chang. 2002. Evaluation of onion cultivars

SMP_Food microbiology_CH08.indd

199

  67.   68.

  69.

  70.   71. 72.   73.

  74.

  75.   76.

  77.   78.

79.

  80.   81.

for resistance to Aspergillus niger, the causal agent of black mold. J. Am. Soc. Hort. Sci. 127:697–702. Kora, C., M. R. McDonald, and G. J. Boland. 2005. Occurrence of fungal pathogens of carrots on wooden boxes used for storage. Plant Pathol. 54:665–670. Kora, C., M. R. McDonald, and G. J. Boland. 2003. Sclerotinia rot of carrot—an example of phenological adaptation and bicyclic development by Sclerotinia sclerotiorum. Plant Dis. 87:456–470. Kramer, M., R. Sanders, H. Bolkan, C. Waters, R. E. Sheeny, and W. R. Hiatt. 1992. Postharvest evaluation of transgenic tomatoes with reduced levels of polygalacturonase: processing, firmness and disease resistance. Postharv. Biol. Technol. 1:241–255. Lacey, J. 1989. Pre- and post-harvest ecology of fungi causing spoilage of foods and other stored products. J. Appl. Bacteriol. Symp. Suppl. 67:11S–25S. Lacroix, M., and C. Vigneault. 2007. Irradiation treatment for increasing fruit and vegetable quality. Stewart Postharv. Rev. 3:1–8. Lazar, E. E., J. J. Jobling, and N. Benkeblia. 2010. Postharvest disease mangaement of horticultural produce using essential oils: today’s propects. Stewart Postharv. Rev. 6:1–9. Le Cam, B., P. Massiot, and F. Rouxel. 1994. Cell-wall polysaccharide-degrading enzymes produced by isolates of Mycocentrospora acerina differing in aggressiveness on carrot. Physiol. Mol. Plant Pathol. 44:187–198. Lennox, C. L., R. A. Spotts, and L. A. Cervantes. 2003. Populations of Botrytis cinerea and Penicillium spp. on pear fruit, and in orchards and packinghouses, and their relationship to postharvest decay. Plant Dis. 87:639–644. Lers, A., and S. Burd. 2007. The potential to retard postharvest senescence using biotechnology. Stewart Postharv. Rev. 3:10. Leverentz, B., W. S. Conway, W. J. Janisiewicz, R. A. Saftner, and M. J. Camp. 2003. Effect of combining MCP treatment, heat treatment, and biocontrol on the reduction of postharvest decay of “Golden Delicious” apples. Postharv. Biol. Technol. 27:221–233. Lewis, B. G., and B. Garrod. 1983. Carrots, p. 103–124. In C. Dennis (ed.), Post-Harvest Pathology of Fruits and Vegetables. Academic Press, London, United Kingdom. Lin, M. W., J. F. Watson, and J. R. Baggett. 1995. Inheritance of resistance to neck-rot disease incited by Botrytis allii in bulb onions. J. Am. Soc. Hort. Sci. 120:297–299. Lindow, S. E. 1996. Role of immigration and other processes in determining epiphytic bacterial populations. Implications for disease management, p. 155–168. In C. E. Morris, P. C. Nicot, and C. Nguyen-the (ed.), Aerial Plant Surface Microbiology. Plenum Press, New York, NY. Lindow, S. E., and M. T. Brandl. 2003. Microbiology of the phyllosphere. Appl. Environ. Microbiol. 69:1875–1883. Logan, C. 1983. Potatoes, p. 179–217. In C. Dennis (ed.), Post-Harvest Pathology of Fruits and Vegetables. Academic Press, London, United Kingdom.

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

200   82. Logrieco, A., A. Moretti, and M. Solfrizzo. 2009. Alternaria toxins and plant diseases: an overview of origin, occurrence and risks. World Mycotox. J. 2:129–140.   83. Lund, B. M. 1983. Bacterial spoilage, p. 219–257. In C. Dennis (ed.), Post-Harvest Pathology of Fruits and Vegetables. Academic Press, London, United Kingdom.   84. Lund, B. M. 1992. Ecosystems in vegetable foods. J. Appl. Bacteriol. Symp. Suppl. 73:115S–126S.   85. Lund, B. M., and A. L. Snowdon. 2000. Fresh and processed fruits, p. 738–758. In B. M. Lund, A. C. BairdParker, and G. W. Gould (ed.), The Microbiological Quality and Safety of Food, vol. I. Aspen Publishers, Gaithersburg, MD. 86. Manandhar, J. B., G. L. Hartman, and T. C. Wang. 1995. Anthracnose development on pepper fruits inoculated with Colletotrichum gloeosporioides. Plant Dis. 79:380–383. 87. Mansfield, J. W. 1983. Antimicrobial compounds, p. 237– 265. In J. A. Callow (ed.), Biochemical Plant Pathology. John Wiley & Sons, Chichester, United Kingdom.   88. Mari, M., T. Cembali, E. Baraldi, and L. Casalini. 1999. Peracetic acid and chlorine dioxide for postharvest control of Monilinia laxa in stone fruits. Plant Dis. 83:773–776.   89. Mastovska, K. 2005. Role of pesticides in produce production, preservation quality, and safety, p. 341–378. In O. Lamikanra, S. Iman, and D. Ukuku (ed.), Produce Degradation—Pathways and Prevention. Taylor and Francis Group, Boca Raton, FL.   90. Matas, A. J., N. E. Gapper, M. Y. Chung, J. J. Giovannoni, and J. K. Rose. 2009. Biology and genetic engineering of fruit maturation for enhanced quality and shelf-life. Curr. Opin. Biotechnol. 20:197–203.   91. Mendonca, A. 2005. Bacterial infiltration and internalization in fruits and vegetables, p. 441–482. In O. Lamikanra, S. Iman, and D. Ukuku (ed.), Produce Degradation—Pathways and Prevention. Taylor and Francis Group, Boca Raton, FL.   92. Mercier, J., and J. Arul. 1993. Induction of systemicdisease resistance in carrot roots by pre-inoculation with storage pathogens. Can. J. Plant Pathol. 15:281–283.   93. Mercier, J., M. Baka, B. Reddy, R. Corcuff, and J. Arul. 2001. Shortwave ultraviolet irradiation for control of decay caused by Botrytis cinerea in bell pepper: induced resistance and germicidal effects. J. Am. Soc. Hort. Sci. 126:128–133.   94. Michailides, T. J., and P. A. G. Elmer. 2000. Botrytis gray mold of kiwifruit caused by Botrytis cinerea in the United States and New Zealand. Plant Dis. 84:208–223.   95. Moretti, C. L., L. M. Mattos, A. G. Calbo, and S. A. Sargent. 2010. Climate changes and potential impacts on postharvest quality of fruit and vegetable crops: a review. Food Res. Int. 43:1824–1832.   96. Morris, C. E., and J. M. Monier. 2003. The ecological significance of biofilm formation by plant-associated bacteria. Annu. Rev. Phytopathol. 41:429–453. 97. Morris, J. R., and P. L. Brady. 2005. Temperature effects on produce degradation, p. 599–647. In O. Lamikanra, S. Iman, and D. Ukuku (ed.), Produce Degradation—

SMP_Food microbiology_CH08.indd

200

Pathways and Prevention. Taylor and Francis Group, Boca Raton, FL.   98. Nadarasah, G., and J. Stavrinides. 2011. Insects as alternative hosts for phytopathogenic bacteria. FEMS Microbiol. Rev. 35:555–575. 99. Nguyen-the, C., and F. Carlin. 2000. Fresh and processed vegetables, p. 620–684. In B. M. Lund, A. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Quality and Safety of Food. Aspen Publishers, Gaithersburg, MD. 100. Nguyen-the, C., and F. Carlin. 1994. The microbiology of minimally processed fresh fruits and vegetables. Crit. Rev. Food Sci. Nutr. 34:371–401. 101. Niem, J., I. Miyara, Y. Ettedgui, M. Reuveni, M. Flaishman, and D. Prusky. 2007. Core rot development in Red Delicious apples is affected by susceptibility of the seed locule to Alternaria alternata colonization. Phytopathology 97:1415–1421. 102. Paliyath, G., and D. P. Murr. 2008. Biochemistry of fruits, p. 19–50. In G. Paliyath, D. P. Murr, A. K. Handa, and S. Lurie (ed.), Postharvest Biology and Technology of Fruits, Vegetables and Flowers. Wiley-Blackwell, Ames, IA. 103. Pangga, I. B., J. Hanan, and S. Chakraborty. 2011. Pathogen dynamics in a crop canopy and their evolution under changing climate. Plant Pathol. 60:70–81. 104. Patterson, M. F., and P. Loaharanu. 2000. Irradiation, p. 65–88. In B. M. Lund, A. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Quality and Safety of Food. Aspen Publishers, Gaithersburg, MD. 105. Perombelon, M. C. M., and A. Kelman. 1980. Ecology of the soft rot erwinias. Annu. Rev. Phytopathol. 18:361–387. 106. Plaza, P., J. Usall, N. Teixido, and I. Vinas. 2003. Effect of water activity and temperature on germination and growth of Penicillium digitatum, P. italicum and Geotrichum candidum. J. Appl. Microbiol. 94:549–554. 107. Porat, R., A. Daus, B. Weiss, L. Cohen, E. Fallik, and S. Droby. 2000. Reduction of postharvest decay in organic citrus fruit by a short hot water brushing treatment. Postharv. Biol. Technol. 18:151–157. 108. Pourcel, L., J.-M. Routaboul, V. Cheynier, L. Lepiniec, and I. Debeaujon. 2007. Flavonoid oxidation in plants: from biochemical properties to physiological functions. Trends Plant Sci. 12:29–36. 109. Prusky, D. 1996. Pathogen quiescence in postharvest diseases. Annu. Rev. Phytopathol. 34:413–434. 110. Prusky, D., D. Eshel, I. Kobiler, N. Yakoby, D. BenoMoualem, M. Ackerman, Y. Zuthji, and R. Ben Arie. 2001. Postharvest chlorine treatments for the control of the persimmon black spot disease caused by Alternaria alternata. Postharv. Biol. Technol. 22:271–277. 111. Prusky, D., and A. Lichter. 2007. Activation of quiescent infections by postharvest pathogens during transition from the biotrophic to the necrotrophic stage. FEMS Microbiol. Lett. 268:1–8. 112. Punja, Z. K. 1985. The biology, ecology and control of Sclerotium rolfsii. Annu. Rev. Phytopathol. 23:97–127. 113. Ragaert, P., F. Devlieghere, and J. Debevere. 2007. Role of microbiological and physiological spoilage mecha-

Manila Typesetting Company

11/08/2012 06:54AM

8.  Fruits and Vegetables

201

nisms during storage of minimally processed vegetables. Postharv. Biol. Technol. 44:185–194. 114. Reddy, M. V. B., P. Angers, F. Castaigne, and J. Arul. 2000. Chitosan effects on black mold rot and pathogenic factors produced by Alternaria alternata in postharvest tomatoes. J. Am. Soc. Hort. Sci. 125:742–747. 115. Reina-Pinto, J. J., and A. Yephremov. 2009. Surface lipids and plant defenses. Plant Physiol. Biochem. 47: 540–549. 116. Reis, H., S. Pfiffi, and M. Hahn. 2005. Molecular and functional characterization of a secreted lipase from Botrytis cinerea. Mol. Plant. Pathol. 6:257–267. 117. Ride, J. P. 1983. Cell walls and other structural barriers in defence, p. 215–236. In J. A. Callow (ed.), Biochemical Plant Pathology. John Wiley & Sons, Chichester, United Kingdom. 118. Saltveit, M. E. 2004. Ethylene effects. In K. C. Gross, C. Y. Yang, and M. Saltveit (ed.), The Commercial Storage of Fruits, Vegetables, and Florist and Nursery Stocks. Agricultural Research Service, Beltsville, MD, 2004, draft version of revised USDA Agriculture Handbook. Available at http://www.ba.ars.usda.gov/hb66. Accessed July 2011. 119. Saltveit, M. E. 2003. Is it possible to find an optimal controlled atmosphere? Postharv. Biol. Technol. 27:3–13. 120. Schirra, M., S. D’Aquino, P. Cabras, and A. Angioni. 2011. Control of postharvest diseases of fruit by heat and fungicides: efficacy, residue levels, and residue persistence. A review. J. Agric. Food Chem. 59:8531–8542. 121. Schirra, M., G. D’hallewin, S. Ben-Yehoshua, and E. Fallik. 2000. Host-pathogen interactions modulated by heat treatment. Postharv. Biol. Technol. 21:71-85. 122. Serrano, J., R. Puupponen-Pimia, A. Dauer, A. M. Aura, and F. Saura-Calixto. 2009. Tannins: current knowledge of food sources, intake, bioavailability and biological effects. Mol. Nutr. Food Res. 53:S310–S329. 123. Snowdon, A. L. 2010. A Colour Atlas of Post-Harvest Diseases and Disorders of Fruits and Vegetables, vol. 1. General Introduction and Fruits. Manson Publishing, London, United Kingdom. 124. Snowdon, A. L. 2010. A Colour Atlas of Post-Harvest Diseases and Disorders of Fruits and Vegetables, vol. 2. Vegetables. Manson Publishing, London, United Kingdom. 125. Spotts, R. A., L. A. Cervantes, and E. A. Mielke. 1999. Variability in postharvest decay among apple cultivars. Plant Dis. 83:1051–1054. 126. Spotts, R. A., P. G. Sanderson, C. L. Lennox, D. Sugar, and L. A. Cervantes. 1998. Wounding, wound healing and staining of mature pear fruit. Postharv. Biol. Technol. 13:27–36. 127. Sugar, D. 2009. Influence of temperature and humidity in management of postharvest decay. Stewart Postharv. Rev. 5:1–5. 128. Sugar, D., and R. A. Spotts. 1993. Dispersal of inoculum of Phialophora malorum in pear orchards and inoculum redistribution in pear immersion tanks. Plant Dis. 77:47–49.

SMP_Food microbiology_CH08.indd

201

129. Swinburne, T. R., and A. E. Brown. 1975. The biosynthesis of benzoic acid in Bramley’s seedling apples infected by Nectria galligena Bres. Physiol. Plant Pathol. 6:259–264. 130. Teplitski, M., K. Warriner, J. Bartz, and K. R. Schneider. 2011. Untangling metabolic and communication networks: interactions of enterics with phytobacteria and their implications in produce safety. Trends Microbiol. 19:121–127. 131. Terry, L. A., and D. C. Joyce. 2004. Elicitors of induced disease resistance in postharvest horticultural crops: a brief review. Postharv. Biol. Technol. 32:1–13. 132. Thorne, S. N. 1972. Studies of the behaviour of stored carrots with respect to their invasion by Rhizopus stolonifer Lind. J. Food Technol. 7:139–151. 133. Tucker, S. L., and N. J. Talbot. 2001. Surface attachment and pre-penetration stage development by plant pathogenic fungi. Annu. Rev. Phytopathol. 39:385–417. 134. van den Berg, L. 1987. Water vapor pressure, p. 203– 230. In J. Weichmann (ed.), Postharvest Physiology of Vegetables. Marcel Dekker, New York, NY. 135. Vicente, A. R., M. Saladie, J. K. C. Rose, and J. M. Labavitch. 2007. The linkage between cell wall metabolism and fruit softening: looking to the future. J. Sci. Food Agric. 87:1435–1448. 136. Whipps, J. M., P. Hand, D. Pink, and G. D. Bending. 2008. Phyllosphere microbiology with special reference to diversity and plant genotype. J. Appl. Microbiol. 105:1744–1755. 137. Williamson, B., B. Tudzynski, P. Tudzynski, and J. A. van Kan. 2007. Botrytis cinerea: the cause of grey mould disease. Mol. Plant Pathol. 8:561–580. 138. Wilson, C. L., A. Elghaouth, E. Chalutz, S. Droby, C. Stevens, J. Y. Lu, V. Khan, and J. Arul. 1994. Potential of induced resistance to control postharvest diseases of fruits and vegetables. Plant Dis. 78:837–844. 139. Wisniewski, M., C. Biles, S. Droby, R. McLaughlin, C. Wilson, and E. Chalutz. 1991. Mode of action of the postharvest biocontrol yeast, Pichia guilliermondii.1. Characterization of attachment to Botrytis cinerea. Physiol. Mol. Plant Pathol. 39:245–258. 140. Wood, D., S. Imam, G. P. Sabellano, W. Orts, and G. M. Glenn. 2005. Microstructure of produce degradation, p. 529–561. In O. Lamikanra, S. Iman, and D. Ukuku (ed.), Produce Degradation—Pathways and Prevention. Taylor and Francis Group, Boca Raton, FL. 141. Yamaguchi, I., and M. Fujimura. 2005. Recent topics on action mechanisms of fungicides. J. Pestic. Sci. 30:67–74. 142. Yao, H. J., and S. P. Tian. 2005. Effects of a biocontrol agent and methyl jasmonate on postharvest diseases of peach fruit and the possible mechanisms involved. J. Appl. Microbiol. 98:941–950. 143. Zeng, W., M. Melotto, and S. Y. He. 2010. Plant stomata: a checkpoint of host immunity and pathogen virulence. Curr. Opin. Biotechnol. 21:599–603. 144. Zhang, J. X., B. D. Bruton, and C. L. Biles. 1997. Poly­ galacturonase isozymes produced by Phomopsis cucurbitae in relation to postharvest decay of cantaloupe fruit. Phytopathology 87:1020–1025.

Manila Typesetting Company

11/08/2012 06:54AM

SMP_Food microbiology_CH08.indd

202

Manila Typesetting Company

11/08/2012 06:54AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch9

Linda J. Harris, Joseph R. Shebuski, Michelle D. Danyluk, Mary S. Palumbo, and Larry R. Beuchat

9

Nuts, Seeds, and Cereals

The culinary definition of nuts is very broad and includes botanically defined nuts (e.g., acorn, chestnut, and filbert), seeds (e.g., Brazil nut, cashew, pignolia or pine nut, and pumpkin, sesame, and sunflower seeds), legumes (e.g., peanut), and drupes (e.g., almond, coconut, macadamia nut, pecan, pistachio, and walnut) (3, 65, 98). Nuts are grown all over the world in temperate, subtropical, and tropical zones. The most economically important nut worldwide is the peanut or groundnut (Arachis hypogaea L.). Other economically important nut crops include almond (Prunus dulcis), Brazil nut (Bertholletia excelsa), cashew (Anacardium occidentale L.), chestnut (Castanea spp.), coconut (Cocos nucifera L.), filbert or hazelnut (Corylus spp.), macadamia (Macadamia spp.), pecan (Carya illinoinensis), pistachio (Pistacia vera L.), and walnut (Juglans spp.). Nuts may be sold in the shell or shelled. In addition to being eaten out-of-hand, nuts are used as ingredients in a wide variety of baked goods and confectionary, dairy, and snack foods and are sold whole, chopped, diced, slivered, ground, or as a flour or paste. Detailed

descriptions of various nuts and their production, harvesting, processing, composition, and microbiology can be found in a number of published works (2, 98, 123). Cereal grains are the most important agricultural products in the world. Since their first cultivation in prehistoric times, cereal grains have sustained the development of civilizations. Over the course of millennia, agricultural practices have been refined in order to protect the security, safety, and quality of grain supplies. The most important cereal crops, based on 2008 Food and Agriculture Organization production data, are corn, rice, and wheat (www.faostat.fao.org). Corn is produced in the greatest quantity, while rice and wheat are produced in approximately equal amounts. Most of the world’s grain supply is consumed by humans. The major intrinsic factor that influences the survival and growth of spoilage- and disease-causing micro­ organisms on a wide range of nuts and cereals is water activity (aw). At aw values above which growth can occur in these low-moisture foods, temperature is the major extrinsic factor influencing proliferation as well

Linda J. Harris, Department of Food Science and Technology, University of California, One Shields Ave., Davis, CA 95616-8598. Joseph R. Shebuski, Corporate Food Safety and Regulatory Affairs, Cargill, Incorporated, 15407 McGinty Rd. West, Wayzata, MN 55391. Michelle D. Danyluk, Department of Food Science and Human Nutrition, Citrus Research and Education Center, University of Florida, 700 Experiment Station Road, Lake Alfred, FL 33850-2290. Mary S. Palumbo, Western Institute for Food Safety and Security, University of California, One Shields Ave., Davis, CA 95616-8598. Larry R. Beuchat, Center for Food Safety, University of Georgia, 1109 Experiment St., Griffin, GA 30223-1797.

203

SMP_Food Microbiology_CH09.indd

203

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

204 as toxin production. This chapter presents an overview of the behavior of microorganisms on nuts, cereals, and products produced from them, with particular emphasis on describing conditions that permit or inhibit growth and treatments that can be used for microbial control or elimination.

MICROBIOLOGY OF NUTS

Harvesting, Processing, and Storage of Nuts

The edible portion of a nut may be referred to as a seed, kernel, or meat. Seed coats or pellicles are present on all nuts and develop from tissues originally surrounding the ovule. The seed coats may be as thin as paper (e.g., peanuts) or thick and hard (e.g., coconuts). All nuts have a relatively rigid outer casing or shell. Some nuts (e.g., almond, coconut, pecan, pistachio, and walnut) also have a hull, which is the pulpy tissue that provides a protective outer covering for the shell. Worldwide, the production, harvesting, and processing techniques for nuts range from highly mechanized to labor intensive, and methods vary significantly among various types of nuts (2, 98, 123). Many tree nuts are harvested by knocking them to the ground or onto a tarp, either by hand or mechanically. Some nuts are partially dried on the tree and then on the orchard floor or elsewhere on the ground, taking advantage of solar heat and low environmental humidity. In other cases, mechanical drying is used alone or in conjunction with natural drying. For example, almonds may be dried in the orchard after they have been shaken to the ground. They are harvested by sweeping from the orchard floor, and the hull and often the shell are removed in a facility that is usually separate from areas used for further processing steps. At the processing facility, almond kernels are sized and then stored from refrigerated to ambient temperatures for various times prior to final processing and packaging. Pecans are often mechanically dried before storing in the shell; they are wetted before cracking to soften the shell and kernel, thereby minimizing breakage and facilitating clean removal of kernel halves. The hull of the walnut is removed in a wet process prior to low-temperature (<43°C) drying for 8 to 48 h and in-shell storage at refrigeration to ambient temperatures. Pistachios, particularly those grown in the United States, are harvested directly into trailers; the hulls are removed shortly thereafter in a wet process before the nuts are partially dried in forced-air dryers (70 to 90°C) from approximately 50% to 9 to 14% moisture and then

SMP_Food Microbiology_CH09.indd

204

further dried in storage silos at ambient temperatures to a stable moisture content (<7%).

Natural Microbiota

Although production and processing vary for each type of nut, nuts are generally dried to an aw of less than 0.70, with corresponding moisture contents ranging from 3.8% (macadamia) to 7.0% (pine nut) (13). The low aw prevents the growth of microorganisms, resulting in microbiologically stable products. Growth of mycotoxigenic molds and bacterial pathogens may occur in nutmeats at high aw. Salmonella can grow on highmoisture pecan halves, pieces, and granules (aw, 0.96 to 0.99), reaching populations as high as 6.4 log CFU/g at 37°C within 24 h (17). Populations of bacteria, yeasts, and molds present on raw nuts vary greatly and depend on growing, harvesting, and handling practices. Because nuts often come in contact with the soil during harvest (tree nuts) or growth (peanuts), their microbiotas (type and population) are greatly influenced by the microorganisms present in the soil. When almonds are harvested onto canvas tarps, aerobic plate and yeast and mold counts are significantly lower than counts on almonds harvested from the ground (66). Higher microbial populations are also associated with the amount of foreign material present, the amount of insect damage, and shell integrity. The thick shells of most tree nuts provide an important and effective barrier to microbial penetration, and the presence of a hull further reduces the risk of microbial invasion. The internal surface of a dry, intact nut picked from the tree is thought to be virtually sterile (36, 61, 80). In contrast to these findings, Hanlin (55) reported that as pecans mature during the growing season, the level of internal fungi approaches 100%. Hull or shell dehiscence (splitting) can occur on the tree or after harvest. Within a single nut type, different varieties may have widely differing shell thicknesses. Thin and porous shells are easily damaged, and any damage to the shell during harvest—caused either naturally, by birds, other vertebrates, or insects, or by other means—will reduce the protection the shell provides (16, 66). The shell may crack along the suture during wetting (79) or drying, providing another means for microbial entry (66, 80). Coconut water leaked from cracked shells will support the growth of microorganisms, which can lead to contamination of the meat prior to processing (100). Cover crops may be used between rows of trees in orchards. The presence of grass or other cover crops in pecan orchards may increase the likelihood of invasion of the kernel by molds. Allowing domestic animals to

Manila Typesetting Company

11/08/2012 06:54AM

9.  Nuts, Seeds, and Cereals

205

graze on these cover crops increases the risk of nut contamination by fecal microorganisms, especially if the nuts drop to the ground before or during harvest (79, 100). However, there are no published data available that evaluate or quantify the food safety risks of these practices. The moisture content of nuts in the field, particularly during harvest, will have a direct influence on the microbiota. Nuts may come in contact with water in the orchard when they drop prematurely to the ground while the trees are still being irrigated or when rain falls during or after harvest. Nuts grown in humid areas are more likely to become contaminated with molds than nuts grown in drier regions (126), and rainfall just prior to or during harvest can cause molding of almonds (64) and pecans (12) on the tree as well as on the ground. Dry almond hulls swell upon exposure to water, and Salmonella can grow in wet hulls (39, 115). Movement of Salmonella through intact almond hulls and shells to the kernel has been demonstrated in intact hulls and shells immersed in water (39). Pecan shells crack along the suture line when held in water for 48 h or more, and the cracks remain after nuts have dried (79). Cracked and damaged pecan shells facilitate rapid infiltration of water and any microorganisms the water may contain (16). Walnuts with deteriorated hulls and shells are contaminated with Escherichia coli to a greater extent than those with intact hulls and shells when soaked in a buffered suspension of E. coli (80). The initial microbiota of peanuts, which develop beneath the soil surface, originates from the soil. Mold contamination of the pods is inevitable. Growing peanut crops in the same field year after year has been shown to increase mold populations in peanuts (88). Pod cracking and a higher incidence of molds occur following drought stress, which may be caused either by low rainfall or overplanting, and are more common in sandy soils. Systemic invasion by Aspergillus flavus throughout the peanut plant has been demonstrated under experimental conditions (89) and may provide a means by which peanut kernels can become contaminated. Peanut lines resistant to fungal invasion have been identified (82). Resistance factors include increased tannin content, amino acid composition, waxy surface, permeability, and cell structure arrangement (82). The aspergilli, especially A. flavus, are ubiquitous invaders of nuts (89) and can produce the mycotoxin aflatoxin. Additional fungal invaders include Alternaria, other species of Aspergillus, Acremonium, Chaetomium, Cladosporium, Fusarium, Eurotium, Mucor, Paecilomyces, Penicillium, Phialophora, Phomopsis, Rhizopus, Tricothecium, and Trichosporon (12, 36,

SMP_Food Microbiology_CH09.indd

205

51, 54, 64, 89, 90, 125). Bacteria isolated from nuts include the genera Achromobacter, Acinetobacter, Bacillus, Brevibacterium, Clostridium, Corynebacterium, Escherichia, Flavobacterium, Lactobacillus, Leuconostoc, Microbacterium, Micrococcus, Proteus, Pseudomonas, Staphylococcus, Streptococcus, and Xanthamonas as well as members of the Enterobacteriaceae (36, 58, 61, 66).

Microbiological Spoilage

Spoilage of nuts is primarily due to oxidative rancidity of the fats. Microbial spoilage, which is less common, can be completely controlled by maintaining an appropriate aw. Water activity values of less than 0.70 essentially eliminate bacterial and most fungal growth in nuts. Molds may grow on nuts that are damp, and in almonds, bacteria are capable of growing in wet hulls and shells (115) and in wet areas of the processing environment (46). Condensate may form on nuts that are cooled too quickly or are removed from refrigerated storage and placed in a warm humid environment. Every effort should be made to protect nuts from moisture in areas in which nuts are processed and stored.

Safety Considerations

Data on the prevalence of bacterial foodborne pathogens in nuts (Table 9.1) are mostly limited to Salmonella. A 7-year survey of 12,972 100-g samples of almond kernels (after hulling and shelling but before further processing) revealed a low but consistent background level of Salmonella (average incidence of 0.98%) (10, 41). A single 25-g sample of raw almonds (out of 60) received at three processors in Australia over a period of 3 years was positive for Salmonella enterica serovar Fremantle subsp. II, a serovar that has been found in a number of Australian environmental samples (48). Surveys of ready-to-eat nuts and seeds (25-g samples) obtained from retail locations in the United Kingdom and Australia revealed a low prevalence of Salmonella (Table 9.1); Salmonella was isolated from Brazil nuts (77), pistachios (76), and sesame, melon, linseed, sunflower, alfalfa, and mixed seeds (128). Salmonella has also been isolated from sesame seed products such as halva (a sweet), tahini (paste), and hummus (tahini with chickpeas) (26, 92). Riyaz-Ul-Hassan et al. (96) isolated Salmonella from 1 in 50 10-g samples of walnut kernels collected in postharvest processing units and villages in regions of India (Jammu and Kashmir). Salmonella has also been isolated in macadamia nuts (111), cashew and Brazil nut kernels (51), and dried coconut meat (100).

Manila Typesetting Company

11/08/2012 06:54AM

Nut or seed

Collection locationa

Sample size (g)

No. of No. positive samples for Salmonella % Positive tested (n) (n+) (if n > 50)

Salmonella serotype

206

Table 9.1  Microbial pathogen prevalence on naturally contaminated nuts and dried seeds

Salmonella levels

Other pathogens assayed

Reference(s)

Nuts SMP_Food Microbiology_CH09.indd

Processor receiving, California

100

12,972

127

Almond, raw in-shell

Processor receiving, California

100

455

7

1.5

Muenchen, Typhimurium, Newport, Thompson, Give, IIIa:18:z32

10

Almond, raw kernel

Processor receiving, Australia

25

60

1

1.7

Fremantle subsp. II

48

Almond, treated (roasted and unknown)

Retail, UK

25

359

0

0

Almond, treated

RTE packages at processor, Australia

25

42

0

Brazil nut, shelled and whole in-shell

Processor

50

20

0

Brazil nut

Processor receiving, Australia

25

60

0

Brazil nut

RTE packages at processor, Australia

25

40

0

Brazil nut

Retail, UK

25

469

Brazil nut

Retail, Brazil

206

Almond, raw kernel

11/08/2012 06:54AM

2 Not given

10, 41

77

0/42 positive for Listeria monocytogenes, coagulase-positive staphylococci

47

7

0

0.4

48

0/40 positive for Listeria monocytogenes, ­coagulasepositive staphylococci

47

Staphylococcus aureus

51

Senftenberg, Tennessee 9, 23 MPN/100 g

Not given Typhimurium

77

Microbial Spoilage and Public Health Concerns

Manila Typesetting Company

Not given Two 2-kg samples, subsample size not given

0.97 ± 0.34 Montevideo, Thompson, 1.2–15.5 MPN/ Enteritidis, 100 g Typhimurium, Senftenberg, and 30 others

207 Manila Typesetting Company 11/08/2012 06:54AM

Retail, UK

25

459

0

0

77

Cashew

Processor receiving, Australia

25

100

0

0

48

Cashew

RTE packages at processor, Australia

25

45

0

Coconut

Husked nuts from 5 countries

25 ml of lactose broth rinse (from 100 ml/ nut)

15

<4/shell

Hazelnut

Retail, UK

25

195

0

Hazelnut

Processor receiving, Australia

25

48

0

Hazelnut

RTE packages at processor, Australia

25

51

0

Macadamia

RTE packages, retail, UK

25

65

0

Peanut

Processor receiving, Australia

25

653

0

Peanut (raw whole Retail, 2; shelled 2) Scotland

25

4

0

Peanut

RTE packages at processor, Australia

25

343

0

0

Peanut

RTE packages, retail, UK

25

148

0

0

Pecan

Retail, UK

25

151

0

0/45 positive for Listeria monocytogenes, coagulasepositive staphylococci

47

0/15 positive for coagulase-positive staphylococci

61

0

9.  Nuts, Seeds, and Cereals

SMP_Food Microbiology_CH09.indd

Cashew

77 48

0

0/51 positive for Listeria monocytogenes, coagulasepositive staphylococci

47

77

0

48

Also tested negative for Listeria spp. and S. aureus

29

0/343 positive for L. monocytogenes, coagulase-positive staphylococci

47

77

77

207

(Continued )

Nut or seed

Collection locationa

Sample size (g)

No. of No. positive samples for Salmonella % Positive tested (n) (n+) (if n > 50)

Salmonella serotype

208

Table 9.1  Microbial pathogen prevalence on naturally contaminated nuts and dried seeds (Continued)

Salmonella levels

Other pathogens assayed S. aureus, 350 CFU/g

Reference(s)

SMP_Food Microbiology_CH09.indd 208

Pistachio, raw whole

Retail, Scotland

25

2

0

29

Pistachio (kernels only 73; with shells 111)

Retail, UK

25

184

0

0

Walnut

Retail, UK

25

441

0

0

Walnut

India (Kashmir and Jammu)

10

50

1

Mixed nuts (almonds, Brazils, cashews, peanuts, walnuts)

Retail, UK

25

105

1

0.95

Anatum

<1 MPN/100 g

77

Alfalfa

Retail, UK

25

58

1

1.7

Not given

—b

128

Hemp

Retail, UK

25

121

0

0

Flax (linseed)

Retail, UK

25

284

1

0.4

Not given

Melon

Retail, UK

25

47

4

Poppy

Retail, UK

25

202

0

0

Pumpkin

Retail, UK

25

886

0

0

Sesame

Retail, UK

25

771

13

1.7

Sesame

Retail, Germany

25

16

2

Tennessee, Offa

26

Sesame paste (tahini)

Retail, Germany

25

12

1

Typhimurium DT 104

26

Halvah (helva or Retail, halawa; sesame Germany paste plus heated acidified glucose syrup and flavor)

25

71

8

Typhimurium DT 104, monophasic B-strain, Poona, Typhimurium DT 134

26

Sunflower

25

976

1

77

77 Staphylococcus spp. (5/50 positive) and Bacillus cereus (3/50 positive)

96

Edible seeds

11/08/2012 06:54AM

b

Retail, UK

0.1

—

128

Unnamed (47:z4,z23:-) —

b

128 128 128

Drypool, unnamed (47:z4,z23:-)

Not given

—

b

—b

RTE, ready to eat; UK, United Kingdom. Salmonella cells were enumerated in six samples in this study, but it is not clear which seeds were represented. In 4/6 samples, MPN < 0.1/g; in 2/6, MPN = 0.1/g and 0.2/g.

128

128

Microbial Spoilage and Public Health Concerns

Manila Typesetting Company

a

128 b

9.  Nuts, Seeds, and Cereals

209

Very few surveys also include an evaluation of pathogen levels, but they are typically less than one cell per gram. Levels of Salmonella determined for 99 of 127 Salmonella-positive almond samples were 0.0079 (for most samples) to 0.012 to 0.15 most probable number (MPN)/g (10, 41; L. J. Harris, unpublished data); Salmonella levels were 0.09 and 0.23 MPN/g in two Brazil nut samples (77), <0.01 MPN/g in one sample of mixed nuts (77), and <0.1 MPN/g (four samples), 0.1, and 0.2 MPN/g in six edible-seed samples (128) (Table 9.1). Nuts are not commonly associated with foodborne illness; however, salmonellosis has been associated with consumption of a wide variety of nuts and nut products (Table 9.2). In some outbreaks of salmonellosis, microbial analyses during traceback investigations have identified high numbers of positive samples (usually 25-g samples) but low numbers of Salmonella per gram of product tested. Recalled, unopened packages of in-shell peanuts (“dry-flavored” and roasted) were positive for Salmonella in, on average, 28 of 72 samples (38%) tested in Australia, Canada, and the United Kingdom (67). Population levels of Salmonella detected in the same products in Australia ranged from <0.03 to approximately 2 MPN/g (67). Populations of fewer than three Salmonella cells per gram (three samples) and four Salmonella cells per gram (one sample) were reported for opened and unopened jars of peanut butter involved in an outbreak of salmonellosis (102). Two large outbreaks of salmonellosis associated with peanut butter occurred in the United States in 2006–2007 and in 2008–2009. In the earlier outbreak, strains of Salmonella serovar Tennessee matching patient isolates were obtained from opened and unopened jars of peanut butter and from environmental swabs of the production facility (31). In the latter outbreak, Salmonella serovar Typhimurium matching the outbreak strain was isolated from unopened food service containers of peanut butter; however, many cases had not eaten the identified vehicle, leading to the discovery that peanut paste and peanut granules, produced in the same facility and used as ingredients in products such as peanut butter crackers, were continuing to cause illness (32). The recall of over 3,900 products that used the associated peanut paste, peanut butter, and peanut granules makes this one of the largest recalls in the United States (119). Contamination associated with poor plant construction and maintenance, particularly in the production and packaging areas, may have contributed to contamination of the peanut products following roasting (91, 119). Recalled lots of almonds from the 2000–2001 serovar Enteritidis outbreak were positive in 64 to 84% of 100-g

SMP_Food Microbiology_CH09.indd

209

samples tested; positive lots sampled 4 to 6 months after the outbreak peak had populations of 6 to 9 MPN/100 g (40). Danyluk et al. (40) used these levels, 95% MPN confidence intervals, and survivor curves (114) to predict levels as high as 130 to 2,400 MPN/100 g for a portion of the outbreak. Populations detected during outbreak investigations suggest that low numbers of Salmonella in these products may have been sufficient to cause illness. Low infectious doses of salmonellae have also been reported for a number of other dried foods (43, 104). Although salmonellosis is most often associated with low-moisture foods, an outbreak of E. coli O157:H7 was associated with consumption of in-shell hazelnuts in 2011 (33); eight persons from three states were impacted over a 2-month period. A matching E. coli O157: H7 strain was isolated from three separate samples of in-shell hazelnuts or from mixed nuts containing hazelnuts taken from patients’ homes. Currently, very little is known about the prevalence or survival of E. coli O157: H7 in low-moisture foods and nuts. Inadequate processing was a possible reason for outbreaks of botulism implicating canned peanuts (37). A combination of high pH (5.0 to 5.5), a formulation change from sugar to aspartame, and insufficient heating enabled survival and outgrowth of Clostridium botulinum spores in a hazelnut conserve used for flavoring yogurt (84). C. botulinum is capable of growing and producing toxin in aerobically stored peanut spread (due to mold growth on the product surface) but not in anaerobically stored spread, when the aw is 0.96 or above (38). An aw of 0.94 or less prevented growth of C. botulinum, regardless of atmospheric conditions; conventionally formulated peanut butter has an aw of less than 0.40. Mycotoxigenic molds may be introduced into nuts on the tree, in the ground, or during harvest and storage. The presence of these molds is not a significant problem unless sufficient moisture becomes available to permit growth. Once the nuts are dried to an aw of 0.70 or below, growth is inhibited and toxin production is prevented. Aflatoxins, produced by Aspergillus species A. flavus, A. parasiticus, and A. nomius, are the most common mycotoxins found in nuts and nut products. Aflatoxins are produced mainly under warm conditions and thus are more likely to be present in nuts grown in countries with warm temperate, subtropical, or tropical climates. Agricultural practices can also influence contamination levels. Susceptibility to contamination is greatest for peanuts and tree nuts that remain on the ground after maturity (which allows for wetting and drying cycles), split hulls or shells that expose the kernel, and insect-infested and mechanically damaged nuts

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

210 (89, 103). See chapter 23 (mycotoxins) for more detailed information on aflatoxins.

Effects of Postharvest Processing and Storage

Current processes for removing hulls and shells differ for each kind of nut and lead to varying risks of crosscontamination between the hull and/or shell and the kernel. Soils picked up with nuts during harvest are partially removed by forced air or sifting, but significant amounts may be brought into storage and processing facilities and can contaminate equipment and nuts during processing. Kokal and Thorpe (69) determined the incidence of E. coli on soft-shelled almonds at growth, production, harvest, hulling, and shelling stages. A low percentage (1%)

of almonds on the tree were positive. The proportion of positive almonds gradually increased with each successive harvest step (shaking trees to detach nuts, sweeping, pickup, storage, weighing, and precleaning), up to 40% prior to hulling. During subsequent separate hulling and shelling steps, the percentage of E. coli-positive almonds decreased to 13 to 17%. Most almonds are currently hulled and shelled in a single-step process that exposes the kernel directly to the outer part of the hull, further increasing the potential for soil contamination at this stage. When hulls, shells, and seed coats of nuts are mechanically separated from the kernel under dry conditions, large volumes of fine particulate matter are generated. For almonds, this “dust,” comprised of soil, hulls, and shells, is extremely difficult to eliminate from the

Table 9.2  Outbreaks of foodborne illness associated with the consumption of nuts and oilseeds

Nut or seed Almond

Coconut

Product

Yr(s)

No. of ­confirmed cases

Isolated from product?

Outbreak location(s)

Reference(s)

2000–2001

168

Yes

Canada, USA

34, 59

Salmonella serovar and strain or other organism

Raw whole

Enteritidis PT30

Raw whole

Enteritidis PT9c

Raw whole

Enteritidis

Desiccated

2004

47

No

Canada, USA

30

2005–2006

15

No

Sweden

72

Typhi, Senftenberg, and possibly others

1953

>50 (estimated from epia curve)

Yes

Australia

129

Desiccated

Java PT Dundee

1999

18

Yes

United Kingdom

122

Hazelnut

In-shell

E. coli O157:H7

2011

8

Yes

USA

33

Peanut

Savory snack containing corn and peanuts

Agona PT15

1994–1995

71

Yes

Peanut butter

Mbandaka

1996

15

Yes

Australia

102

Flavored or Stanley, Newport roasted in-shell

2001

97 (Stanley), 12 (Newport)

Yes

Australia, Canada, United Kingdom

67

Boiled peanuts

Thompson

2006

100

Yes

South Carolina

93

Peanut butter

Tennessee

2006–2007

425

Yes

USA

31

Peanut butter, peanut butter-containing products

Typhimurium

2008–2009

684

Yes

USA, one case in Canada

32

Halva

Typhimurium DT104

2001

17 (Australia) 18 (Norway) 27 (Sweden)

Yes

Australia, Germany, Norway, Sweden, United Kingdom

1, 25, 44, 75, 92

Tahini

Montevideo

2002

55

Yes

Australia

116

Tahini

Montevideo

2003

3

Yes

Australia

116

Tahini and halva

Montevideo

2003

10

Yes

New Zealand

116

Sesame seed

a

Israel, United Kingdom, USA

epi, epidemic.

SMP_Food Microbiology_CH09.indd

210

Manila Typesetting Company

11/08/2012 06:54AM

63, 106, 112

9.  Nuts, Seeds, and Cereals

211

­ rocessing environment and can contribute significantly p to the contamination of kernels. The addition of water or aqueous quaternary ammonium compounds (200 to 1,000 mg/ml) to almond dust can result in rapid growth of Salmonella and significant increases in the aerobic plate counts (46). As with all dry-processing environments, appropriate care should be used when cleaning and sanitizing equipment to ensure that rapid and complete drying occurs. The use of isopropyl alcohol-based sanitizers on almond hulling and shelling equipment has the dual benefit of decreasing microbial populations in the presence of a high organic load and drying rapidly on equipment surfaces (46). A wet process is used to remove the hull of some nuts (e.g., pecan, pistachio, and walnut). The shells of pecans and some other nuts are conditioned or softened by soaking in water or by humidifying with water sprays, steaming, or holding under high relative humidity. In some cases, the water may be chlorinated, but in dump or soak tanks the high organic load makes it difficult to maintain residual free chlorine at microbicidal concentrations. Water can be a significant source of cross-contamination of nuts if the microbial populations are not controlled by chemical or physical means. However, there are few published data on these potential sources of contamination. Pecan meats from nuts

with visibly intact shells were not contaminated when soaked in lactose broth containing E. coli (79). Other studies have revealed that Salmonella infiltrates in-shell pecans when immersed in a suspension of the pathogen (15, 16). Although liquid can be absorbed at the base and apex of intact pecan shells, the pecan packing tissue (material surrounding the kernel) can either inhibit growth or cause a reduction in the number of Salmonella cells, depending upon the concentration of the tissue in broth (15). The pathogen is inactivated upon contact with water-saturated middle septum tissue in pecans (17). The authors hypothesized that this effect was due to tannins and polyphenolic compounds present in the packing and middle septum tissues. Kokal (68) similarly hypothesized that tannins present in walnut skin might be responsible for observed reductions of E. coli when inoculated nuts were stored at room temperature. Shelled or in-shell nuts may be fumigated for insect control and stored for up to a year before further processing and shipping (60). Long-term storage temperatures range from 4°C to ambient, but after processing and packaging, nuts are generally distributed and displayed at the retail level at ambient temperature. Consumers may store purchased almonds at ambient, refrigerated, or frozen temperatures for an additional 1 to 2 years (73).

Figure 9.1  Survival of a six-strain mixture of Salmonella inoculated onto raw almond kernels, pistachios (in-shell), pecans (in-shell), and raw, shelled peanuts at a population of approximately 6 log CFU/g and stored at 23°C after initial drying of the inoculum. LOD, limit of detection (L. J. Harris and M. D. Danyluk, unpublished data). doi:10.1128/9781555818463.ch9f1

SMP_Food Microbiology_CH09.indd

211

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

212 Few data are available on the survival of microorganisms on nuts or nut products, but in general, survival is remarkably good. E. coli or Salmonella inoculated into nuts or nut products was detected for several months to years (15, 16, 28, 66, 68, 70, 113). Significantly higher numbers of Salmonella survived on pecan halves, inshell pecans (15, 16), walnuts (21), and almonds (113) and in peanut butter (28, 86) stored under refrigerated conditions than at ambient or elevated temperatures. A temperature effect was not observed when Salmonella was inoculated into halva (70). Salmonella Enteritidis phage type (PT) 30 or a cocktail of six serovars of Salmonella did not decline on almonds over 1 or more years of storage at −20 ± 2°C and 4 ± 2°C (40, 113; Harris, unpublished). At 23 ± 3°C storage, average calculated reductions were 0.23 or 0.22 log CFU/ almond per month, respectively. Similar survival patterns were also observed on in-shell raw pistachios (Harris, unpublished) and on shelled raw peanuts and pecans (M. D. Danyluk, unpublished data) (Fig. 9.1). Long-term survival was independent of the initial inoculum level, but survival during initial drying was impacted significantly by the methods used to culture cells to prepare the inoculum (114; L. Wang and L. J. Harris, unpublished data). Salmonella retains high viability on in-shell pecans for 18 months and on pecan halves and pieces stored for 1 year at −20 and 4°C (16). Significant reductions occurred on in-shell nuts and nutmeats stored at 21 and 37°C. Collectively, these data indicate that Salmonella can survive on contaminated nuts and nut products throughout their typical 1- to 2-year shelf life.

Thermal and Nonthermal Processing

Processing steps that involve a thermal treatment often play a dual role of altering the texture and appearance of nuts and reducing microbial contamination. Some products are exposed to several sequential thermal-processing steps. For example, to prepare roasted blanched almond slices or slivers, the almonds are first exposed to hot water or steam (blanching) to loosen the pellicle. After the pellicle is mechanically removed, the almonds are dried by exposure to hot air. To prepare the almonds for slicing, they may be soaked in water and then heated to increase pliability; after slicing or chopping, the pieces may be exposed to dry heat to roast or darken the kernel. In addition to hot air roasting, many nuts are roasted by immersion in hot oil. Alternatively, nuts may be moistened by water or steam or sprayed with a saturated salt solution prior to roasting. Each of these pretreatments will have an impact on microbial reduction during heating, although published nut-specific data are generally lacking in this area.

SMP_Food Microbiology_CH09.indd

212

Microorganisms in dry environments and under dry-processing conditions are significantly more heat resistant than when suspended in liquids or high-aw foods (95). The D-value for serovar Enteritidis PT 30 on inoculated almond kernels immersed in hot water at 88°C was 0.39, and the calculated z-value was 35 Celsius degrees (56). A 2-min exposure to 88°C hot water is conservatively recommended to achieve a 5-log reduction of Salmonella in almonds (4). Typical industry practices for blanching almonds are to expose the nuts to 88 to 99°C for 5 to 1 min, respectively, which is sufficient to achieve greater than a 5-log CFU/g reduction of Salmonella. Treatment of in-shell pecans in water at 99°C for 2 min reduced Salmonella counts by 5.8 log CFU/g (15). Reductions of greater than 5 log CFU/g can be achieved by treating pecans in hot (90°C) water for 80 s (18). A reduction of several log of Salmonella on coconut pieces heated to 80°C for 5 min has been observed without deleterious effects on the sensory quality of shredded product (100). Coconut meat is often pasteurized before or after shredding (123). For oil roasting, exposure at 138 to 177°C for 3 to 15 min is required to obtain the desired level of roasting in almonds (5); this exposure is predicted to far exceed a 5-log reduction of Salmonella (45). Exposure of pecan pieces to peanut oil at 127°C for 1.5 min or 132°C for 1.0 min reduced Salmonella by 5 log CFU/g (19). Treatment of pecan halves at 138°C for 2.0 min was necessary to reduce Salmonella by 5 log. As with almonds, treatment temperatures and times typically used to oil roast pecan nutmeats appear to be sufficient to provide a 5-log reduction. Salmonellae are exceptionally heat resistant when embedded in peanut butter. Thermal treatment at 90°C for less than 30 min is insufficient to kill 5 log CFU/g (78, 105). The D60°C values for Listeria monocytogenes in peanut butter (aw, 0.32) and chocolate-peanut spread (aw, 0.46) are 26 and 37.5 min, respectively (62). Dry air heating is significantly less efficient than oil roasting in reducing microbial contamination. Several factors influence this efficacy, including bed depth, air temperature and velocity, chamber humidity, and nut moisture. Almonds tolerate higher temperatures than do many other nuts such as coconut (100), walnut, pecan, macadamia nut, and Brazil nut. These nuts darken at a significantly faster rate than almonds, and it is not known what level of microbial reduction is achieved during normal roasting processes. Substantial reductions of Salmonella on almonds by use of steam (35), infrared heat (24, 130), or high hydrostatic pressure (127) have also been determined.

Manila Typesetting Company

11/08/2012 06:54AM

9.  Nuts, Seeds, and Cereals

213

Propylene oxide (C3H6O), also known as methyloirane or propene oxide, is currently registered in the United States as a fumigant to reduce the populations of bacteria, yeasts, and molds in a variety of dry foods including cocoa, processed spices, edible gums, and tree nuts. It is used routinely to reduce microbial populations in bulk tree nuts and their products without adversely affecting sensory quality. The lethality of propylene oxide treatment is dependent upon concentration, exposure time, exposure temperature, and humidity (11, 42). Under commercial conditions, a 5-log or greater reduction of serovar Enteritidis PT 30 is achieved when inoculated almonds are prewarmed to 30°C and exposed to propylene oxide at a concentration of 0.5 kg/m3 for 4 h (42). However, this level of reduction was only observed consistently when almonds were held for 5 days postexposure. Aerobic plate counts declined by 0 to 1.1 log CFU/g under the same treatment conditions. Treatment of pecan halves with propylene oxide gave 80 to 92% reductions in the surface microbiota and at least 64% reductions in the internal microbiota, but neither bacteria nor fungi could be eliminated, even with high doses (20).

MICROBIOLOGY OF GRAINS AND MILLED CEREAL GRAINS Microbiological contamination of cereal grains occurs while the grains are growing in the field. These contaminants can increase in number while the grains are actively growing and after harvesting. However, postharvest microbial growth is limited by proper drying of the grains and good storage practices, and microbial contamination is further reduced during the grain milling process. Microbial growth in prepared cereal products can result in spoilage or foodborne illness if appropriate control measures are not followed.

Spoilage of Pre- and Postharvest Grains

Cereal grains accumulate a large and varied microbiota during growth in the field. Molds are among the common contaminants of grains during this time, and some of these molds are toxigenic. It is common to divide these fungal contaminants into two categories (81). Molds that infect the grain before harvest are referred to as “field fungi” and consist primarily of the genera Alternaria, Cladosporium, Fusarium, and Helminthosporium. These molds commonly grow on grains with aw of 0.90 or higher, a level that corresponds to a moisture content of about 20% or higher. Climatic conditions can greatly impact the amount of mold growth that occurs on c­ereal grains. Below-normal temperatures in combination with above-normal

SMP_Food Microbiology_CH09.indd

213

precipitation and relative humidity will foster excessive mold growth, particularly in wheat and barley crops. Under these conditions, field fungi can d­amage the grain, even to the point of total crop loss (99). Less-invasive mold growth can reduce grain quality and may result in the production of mycotoxins. The second category of molds, “storage fungi,” infect grains postharvest and consist primarily of the genera Aspergillus, Rhizopus, Mucor, Wallemia, and Penicillium. Growth of the storage fungi is supported in inadequately dried grains, and as a result mycotoxin levels can increase during storage (85, 99). Modern grain harvesting and storage practices typically prevent further mold growth. Crops are cut, threshed, and winnowed by mechanical harvesters in order to separate the grain from the chaff, and then the crop is mechanically dried before storage. Grain crops can be stored for 1 year or longer before being processed. During this storage period, the grain should be stored in a manner to ensure the exclusion of water, birds, insects, and rodents (74, 99). Grains that are properly dried to moisture levels of 12 to 14% will not mold during storage, provided these moisture levels are maintained. Short-term storage of high-moisture grains (>15%) intended for animal feed can minimize this concern if grains are treated with f­ungistatic agents such as propionic acid, formaldehyde, and acetic acid, or combinations thereof to prevent mold growth (74, 99). Future concerns related to mycotoxins may change in some geographies if mold contamination of grains increases as a result of changes to climate (50, 87). Consequently, regulatory scrutiny of mycotoxin levels in grains is likely to increase. Future regulatory limits for mycotoxins in grains will need to be based on risk but must also be practical and flexible when warranted to ensure that an adequate supply of grain continues to be available for human and animal consumption. In addition to fungi, a wide variety of bacteria, including lactic acid bacteria, micrococci, bacilli, and some enteric bacteria, can grow on cereal crops. Some species of enteric bacteria are normal plant saprophytes, and as such their presence in grain or milled products is not related to fecal contamination. Additionally, crops and stored grains can be contaminated by microorganisms from dust, birds, rodents, insects, and other environmental sources. Insects or insect fragments and rodent hairs and excreta are almost certain to be present in the grain supply. The U.S. Food and Drug Administration (FDA) establishes “maximum levels of natural or unavoidable defects in foods for human use that present no health

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

214 hazard” (118). These defect action levels have been established for wheat, wheat flour, corn meal, macaroni and noodle products, and popcorn.

Effect of Wheat Milling on Microbiological Quality

Before milling, wheat is cleaned through aspiration and screening, which reduces the number of microorganisms present. The grain is then typically sprayed with water and stored in tempering bins for 6 to 18 h before milling. The water added is sufficient to moisten the surface of the grain. This tempering step simultaneously toughens the bran and softens the endosperm, thereby facilitating bran removal and separation of the endosperm, making the latter available for crushing in the first milling step (9). The removal of the bran results in a significant further reduction in microbial counts. Historically, many millers added up to 300 mg calcium hypochlorite/ml to the temper water; however, the function of this practice is unknown. The hypochlorite has no apparent anti­microbial effect because it is quickly neutralized by the grain and other organic material present (71). In wheat tempered for 16 h, with and without 200 mg calcium hypochlorite/ml in the temper water, no reduction in total plate or coliform counts occurred in hypochlorite-treated samples compared to the untreated samples (W. H. Sperber, unpublished data). The conversion of wheat into flour during the drymilling process typically reduces the overall microbial load by 1 to 2 log CFU/g from that originally present on the grain. This is achieved by the physical removal of the more heavily contaminated components of the wheat and not by the actual destruction of the microbes present. Greater microbial reduction is generally observed in flours with the lower bran or ash content. An aerobic plate count (geometric mean) of 4.9 log CFU/g was observed for 54 wheat samples before milling (97). The flour produced from this wheat had a mean aerobic plate count of 3.6 log CFU/g. Hesseltine (57) provided information on the use of heat as a means to further reduce these microbial levels. Five wheat samples were heated to 60°C for periods ranging from 1 to 4 h before milling. The heated wheat samples had a mean aerobic plate count of 4.6 log CFU/g, and the derived flours from those samples had a mean aerobic plate count of 2.4 log CFU/g. The dry-milling process, in a similar manner, can also affect the levels of mycotoxins, if present. Mycotoxins are not destroyed during the milling process, but the cleaning process and the physical separation of the components of the wheat kernel reduce the mycotoxin levels in some components and increase them

SMP_Food Microbiology_CH09.indd

214

in others. One component in which the mycotoxin levels increase is the bran fraction (27). Recently, there has been an increasing desire to improve the healthfulness of food products by increasing the amount of fiber in the form of bran into those products. This increase in bran consumption should be tempered by the understanding that this practice may also result in an increase in the amount of mycotoxins consumed. Other than aspiration, screening, and removal of the bran, there are no steps in a typical dry-milling process that provide a targeted reduction in the microbial load of milled cereal grains. Therefore, milled grains should be considered to be minimally processed agricultural commodities. The mean counts of several indicator micro­ organisms detected in a large number of wheat flour samples indicate the excellent microbiological quality of flour, despite its status as a minimally processed agricultural commodity (Table 9.3). At normal moisture levels of 12 to 14%, flours will not support microbial growth. The dry-milling industry considers excessive mold or yeast counts to be indicative of elevated moisture levels in the flour handling system. Excessive counts trigger equipment inspections focused on detecting and eliminating sources of moisture before levels become high enough to permit bacterial growth (110). It is important to understand the significance of typical yeast and mold levels for different products because yeast and mold counts vary considerably, depending upon the type of flour (Table 9.4). Coliform counts in wheat flour average approximately 450 CFU/g (110) (Table 9.3). To some, higher coliform counts represent an insanitary condition, and many food processors expect their flour suppliers not to exceed a specified maximum coliform limit. However, the coliform group of bacteria was originally adopted as an indicator of fecal contamination in drinking water and pasteurized dairy products, but it is not suited for a similar use for milled cereal grains. A number of coliform species in the genera Erwinia, Pantoea, Serratia, Klebsiella, and Enterobacter grow predominantly on

Table 9.3  Microbiological profile of North American wheat

floura

Microorganism

No. of samples

Geometric mean (log CFU/g)

Aerobic plate count

6,598

3.8

Yeasts

6,573

1.3

Molds

6,869

2.4

Coliforms (Petrifilm)

2,467

2.6

a

Adapted from Sperber, 2007 (110).

Manila Typesetting Company

11/08/2012 06:54AM

9.  Nuts, Seeds, and Cereals

215

Table 9.4  Geometric mean values of combined yeast and

mold counts in several miller cereal grainsa No. of samples

Yeast and mold counts (log CFU/g)

Wheat

48

3.2

Rye

24

3.6

Buckwheat

24

4.3

Corn

48

5.1

Type of flour/meal

a

Adapted from Beuchat, 1992 (14).

plants and are usually not found in fecal matter. Growth of these particular coliforms on wheat plants will ensure their carryover into wheat flour, thereby negating any potential utility of a coliform specification for wheat flour or other milled cereal grains (110). The prevalence of salmonellae in 4,358 samples of wheat flour was about 0.14% (110). The occasional presence of salmonellae in flour is not necessarily considered a public health hazard, as most flour is used in products that are baked, fried, or cooked, i.e., processes that kill these bacteria. If flour is used in a nonconventional manner or added raw to a ready-to-eat product, then the presence of Salmonella in the flour should be considered to be a potential safety concern. In some cases, flour may not be baked before consumption. In 2005, an outbreak of salmonellosis was associated with the consumption of cake batter ice cream. Since other ice cream flavors produced by the same manufacturer were not involved in the reported cases, it was thought some component of the cake batter may have been responsible for the illnesses. The cake batter had not been baked or pasteurized before its addition to the ice cream mix. As a result, the FDA issued a warning against the use of ingredients intended to be cooked in ready-to-eat foods (118). Flour, however, was never specifically linked to this outbreak. Recently, raw cookie dough was epidemiologically linked to an outbreak of foodborne illness caused by E. coli O157:H7. Although the source of the pathogen was not determined, the flour used in this product was considered a possible source of the pathogen. Due to the nature of the wheat growing process and the lack of a step in the milling process that can eliminate pathogens, this cannot be ruled out. However, testing has indicated the incidence of this pathogen in flour appears to be extremely low (0.003%), far lower than that observed for Salmonella (J. R. Shebuski, presented at the 2010 International Association for Food Protection Annual Meeting, Anaheim, CA). Attempts have been made to reduce or eliminate salmonellae in flours by dry heating at temperatures

SMP_Food Microbiology_CH09.indd

215

as high as 60°C. However, prolonged heating above 55°C may degrade flour protein functionality and nutritional attributes (9, 22). At aw values of 0.40 to 0.60, D-values for S. enterica serovar Weltevreden inactivation averaged about 1 h in the temperature range of 69 to 77°C, with a z-value of about 30 Celsius degrees (8). Storing corn flour at 49°C for 24 h resulted in a 3-log CFU/g reduction of salmonellae (122). Microwave treatment also reduces salmonellae in dry corn-soy milk blends at temperatures of 61 and 67°C, with little change in functionality or nutritional content, but nutrient degradation does occur at temperatures in excess of ~78°C (23). Milled cereal grains are produced by a dry-milling process. The resulting flours can be further wet processed, or “hydroprocessed,” in order to separate the gluten and starch components. Lactobacilli dominate the wet-processing system, substantially reducing the pH and thereby preventing the growth of nonlactic microbiota originally present in the flour. Microorganisms such as salmonellae will not increase during wet processing. Therefore, products such as wheat gluten and some starch products will have essentially the same microbiological pathogen profile as the original wheat flour from which they are derived (Sperber, unpublished).

Spoilage of Cereal Products

The baking of dough products simultaneously reduces the microbial load and moisture content, thereby limiting the types of microorganisms that can cause spoilage. Even though most mold spores and vegetative microbial cells are easily killed during baking, the predominant cause of baked product spoilage is mold growth. The surfaces of baked products may be recontaminated with airborne mold spores during the relatively long period of product cooling between baking and packaging. Many bakery products contain one or more fungistatic agents that retard mold growth. One of the most effective food-grade fungistatic agents is potassium sorbate, typically used at concentrations of 1,000 to 3,000 mg/g. However, because potassium sorbate inhibits yeast growth and metabolism, its use is limited to chemically leavened products. Yeast-leavened products are typically protected from mold spoilage by the use of calcium propionate at concentrations of 2,000 to 8,000 mg/g. Bakery products are relatively easily stabilized against rapid mold spoilage by the interaction of a number of preservative factors. Reduced aw and pH values and pasteurization during baking enhance the effectiveness of chemical preservatives (53, 74).

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

216 Confectionery components of baked goods, e.g., fruit or cream fillings and icings, are vulnerable to yeast spoilage. This spoilage can be effectively controlled by use of one or more of the following preservative techniques: aw reduction, pH reduction, and the addition of a chemical preservative. Potassium sorbate is particularly effective in this application when used at 1,000 to 2,000 mg/g (74). Bakery products with a moist interior are vulnerable to “rope” spoilage. Rope is the result of extracellular capsular material production by Bacillus subtilis, which was originally referred to as Bacillus mesentericus (124). Endospores that survive baking will grow under favorable conditions to very high numbers of vegetative cells and produce rope spoilage. Ropy products have a melon-like odor and a stringy, mucilaginous appearance when pulled apart. Early bakers were encouraged to avoid rope spoilage by adding acetic acid to the dough before baking (124). Later research revealed that calcium acid phosphate was a more effective inhibitor than acetic acid (49). Propionates currently used in bakery products to retard mold spoilage have some bacteriostatic activity against B. subtilis and help in the prevention of rope formation (74). Refrigerated dough products are made by packing unbaked, chemically leavened dough into containers that become hermetically sealed as the dough proofs. This raw dough is then baked by the consumer. There are two general types of refrigerated dough products: breadstuffs that have aw values above 0.90, and cookie doughs that have an aw of approximately 0.80. The only microorganisms that can grow in refrigerated cookie dough are osmotolerant yeasts and then only after abuse at ambient or higher temperatures. Bacterial growth is prevented by refrigeration and the reduced aw of the dough. Xerophilic mold growth is prevented by the carbon dioxide content in the dough (Sperber, unpublished). In the past several decades, considerable progress has been made to minimize rope spoilage in baked goods and lactic acid bacteria spoilage of refrigerated dough products. The principal factors for improvement in both cases are the improved sanitary design and improved cleaning and sanitation of the dough-handling equipment (74; Sperber, unpublished). These advances have proven to be more practical and effective than earlier management attempts to control spoilage and safety principally by the use of microbiological specifications on the raw materials. Mold spoilage of some types of specialty bakery products is retarded by the use of modified atmosphere packaging or oxygen scavengers (107). The use of headspace gases

SMP_Food Microbiology_CH09.indd

216

to inhibit mold growth requires the use of packaging materials with very low oxygen permeability, and the packages must be well sealed to prevent leakage of the gases.

Food Safety Considerations in Cereal Products

The reduced aw (0.94 or less) of most cereal products prevents the growth of many microorganisms that can cause foodborne illness. Specifically, C. botulinum, Clostridium perfringens, E. coli, and salmonellae are prevented from growing at this reduced aw (109). Although proteolytic strains of C. botulinum have a minimum aw of 0.94 for growth, these bacteria have been a concern in bakery products, because of the severity of the illness they may create and because their spores survive the baking process (108). The product category of cooked, refrigerated pasta products can sometimes present a substantial safety challenge. Produced both with and without a variety of fillings, the pasta and filling components sometimes have an aw above 0.95. These products are packaged in hermetically sealed containers with very low headspace oxygen to preclude mold growth. The resulting anaerobic condition favors the growth of C. botulinum. Additionally, cooking eliminates much of the competitive microflora, thereby facilitating C. botulinum growth if temperature abuse occurs. Glass and Doyle (52) determined that C. botulinum cannot produce toxin in refrigerated pasta, even when held at 30°C, when the aw is 0.94 or below; however, toxin was produced at that temperature when the aw was 0.96. A survey of commercially available refrigerated, filled pasta products revealed that some are capable of supporting C. botulinum growth and toxin production when held at 30°C (101). Some of these commercial products had aw values as high as 0.983. Refrigerated pasta products must be formulated to prevent food safety hazards should temperature abuse occur. This product design activity must be an integral part of the manufacturer’s hazard analysis and critical control points plan. If practical food safety measures such as the control of pH and aw limits cannot be incorporated to protect products from food safety hazards during extended temperature abuse, the products should not be produced. Analogous situations have been determined to exist with cream- or custard-filled pastries, nonfruit pies, filled breads, and very moist cakes. These products can present opportunities for Staphylococcus aureus to grow to populations sufficient to cause illness. Use of the “hurdle effect,” i.e., combinations of preservative factors such as aw reduction, pH reduction, and preservatives, can successfully eliminate

Manila Typesetting Company

11/08/2012 06:54AM

9.  Nuts, Seeds, and Cereals

217

the potential hazards posed by this pathogen when products are marketed at ambient temperatures. Historically, commercially produced dried pasta products were vulnerable to the growth of S. aureus and/or salmonellae during dough mixing, extrusion, and drying. Some strains of S. aureus are capable of growth in pasta doughs with aw as low as 0.86, although enterotoxin production is generally inhibited at aw of 0.90 to 0.94 (120). Staphylococci die within several weeks in dried pasta, whereas salmonellae can persist for more than 1 year (94). However, the potential safety and public health hazards presented by these two pathogens in contaminated dry pasta are the opposite at the time of pasta preparation and consumption. Surviving salmonellae in pasta would be easily killed during cooking, presenting no public health hazard. In contrast, the dead S. aureus cells could leave behind their thermostable entero­toxins, which would not be destroyed by cooking, thereby remaining as a potential cause of illness (6). Therefore, properly drying the pasta by rapidly reducing the water activity through the range in which these enterotoxins are produced is critical to avoiding this concern. Many outbreaks of illness caused by Bacillus cereus have been associated with consumption of fried rice (83). In these cases, rice was cooked, stored at ambient temperature for up to 24 h before being further prepared, and then served. Under these conditions, B. cereus can grow rapidly and produce a heat-stable emetic toxin. This potential hazard can be minimized by storing the cooked rice at refrigeration temperature or eliminating storage of the cooked rice at ambient temperature.

References 1. ­ Aavitsland, P., O. Alvseike, P. J. Guérin, and T. L. Stavnes. 16 August 2001. International outbreak of Salmonella Typhimurium DT104—update from Norway. Eurosurveill. Wkly. 5(33):1701. 2. Alasalvar, C., and F. Shahidi (ed.). 2009. Tree Nuts: Composition, Phytochemicals, and Health Effects. CRC Press, Boca Raton, FL. 3. Alden, L. 2005. The Cook’s Thesaurus: Nuts. http://www. foodsubs.com/Nuts.html. 4. Almond Board of California. 2007. Guidelines for validation of blanching processes, v1.0, April 13, 2007. http://www.almondboard.com/Handlers/Food QualitySafety/Pasteurization/PasteurizationProgram/ ValidationGuidelines/. 5. Almond Board of California. 2007. Guidelines for validation of oil roasting processes, v1.0, April 13, 2007. http://www.almondboard.com/Handlers/Food QualitySafety/Pasteurization/PasteurizationProgram/ ValidationGuidelines/.

SMP_Food Microbiology_CH09.indd

217

6. Anderson, J. E. 1996. Survival of the serological and biological activities of staphylococcal enterotoxin A in canned mushrooms. Doctoral dissertation. Pennsylvania State University, University Park. 7. Arrus, K., G. Blank, R. Clear, R. A. Holley, and D. Abramson. 2005. Microbiological and aflatoxin evaluation of Brazil nut pods and the effects of unit processing operations. J. Food Prot. 68:1060–1065. 8. Archer, J., E. T. Jervis, J. Bird, and J. E. Gaze. 1998. Heat resistance of Salmonella weltevreden in low-moistur­e environments. J. Food Prot. 61:969–973. 9. Atwell, W. A. 2001. Wheat Flour. Eagan Press, St. Paul, MN. 10. Bansal, A., T. M. Jones, S. J. Abd, M. D. Danyluk, and L. J. Harris. 2010. Most-probable-number determination of Salmonella levels in naturally contaminated raw almonds using two sample preparation methods. J. Food Prot. 73:1986–1992. 11. Beuchat, L. R. 1973. Escherichia coli on pecans: survival under various storage conditions and disinfection with propylene oxide. J. Food Sci. 38:1063–1066. 12. Beuchat, L. R. 1975. Incidence of molds on pecan nuts at different points during harvesting. Appl. Microbiol. 29:852–854. 13. Beuchat, L. R. 1978. Relationship of water activity to moisture content in tree nuts. J. Food Sci. 43:754–755, 758. 14. Beuchat, L. R. 1992. Enumeration of fungi in grain flours and meals as influenced by settling time in diluent and by the recovery medium. J. Food Prot. 55:899–901. 15. Beuchat, L. R., and E. K. Heaton. 1975. Salmonella survival on pecans as influenced by processing and storage conditions. Appl. Microbiol. 29:795–801. 16. Beuchat, L. R., and D. A. Mann. 2010. Factors affecting infiltration and survival of Salmonella on in-shell pecans and pecan nutmeats. J. Food Prot. 73:1257–1268. 17. Beuchat, L. R., and D. A. Mann. 2010. Survival and growth of Salmonella in high-moisture pecan nutmeats, in-shell pecans, inedible nut components, and orchard soil. J. Food Prot. 73:1975–1985. 18. Beuchat, L. R., and D. A. Mann. 2011. Inactivation of Salmonella on in-shell pecans during conditioning treatments preceding cracking and shelling. J. Food Prot. 74:588–602. 19. Beuchat, L. R., and D. A. Mann. 2011. Inactivation of Salmonella on pecan nutmeats by hot air treatment and oil roasting. J. Food Prot. 74:1441–1450. 20. Blanchard, R. O., and R. T. Hanlin. 1973. Effect of propylene oxide treatment on the microflora of pecans. Appl. Microbiol. 26:768–772. 21. Blessington, T., E. J. Mitcham, and L. J. Harris. 2012. Survival of Salmonella enterica, Escherichia coli O157: H7, and Listeria monocytogenes on inoculated walnut kernels during storage. J. Food Prot. 75:245–254. 22. Bookwalter, G. N., R. J. Bothast, W. F. Kwolek, and M. R. Gumbmann. 1980. Nutritional stability of

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

218

23.

24.

25.

26. 27. 28. 29.

30.

31.

32.

33.

34.

35.

c­orn-soy-milk blends after dry heating to destroy salmonellae. J. Food Sci. 45:975–980. Bookwalter, G. N., T. P. Shukla, and W. F. Kwolek. 1982. Microwave processing to destroy Salmonella in corn-soy-milk blends and effect on product quality. J. Food Sci. 47:1683–1686. Brandl, M., Z. Pan, S. Huynh, Y. Zhu, and T. H. McHugh. 2008. Reduction of Salmonella Enteritidis population sizes on almond kernels with infrared heat. J. Food Prot. 71:897–902. Brockmann, S. 16 August 2001. International outbreak of Salmonella Typhimurium DT104 due to contaminated sesame seed products—update from Germany (BadenWürttemberg). Eurosurveill. Wkly. 5(33):1699. Brockmann, S. O., I. Piechotowski, and P. Kimmig. 2004. Salmonella in sesame seed products. J. Food Prot. 67:178–180. Bullerman, L. B., and A. Bianchini. 2007. Stability of mycotoxins during food processing. Int. J. Food Microbiol. 119:140–146. Burnett, S. L., E. R. Gehm, W. R. Weissinger, and L. R. Beuchat. 2000. Survival of Salmonella in peanut butter and peanut butter spread. J. Appl. Microbiol. 89:472–477. Candlish, A. A. G., S. M. Pearson, K. E. Aidoo, J. E. Smith, B. Kelly, and H. Irvine. 2001. A survey of ethnic foods for microbial quality and aflatoxin content. Food Addit. Contam. 18:129–136. Centers for Disease Control and Prevention. 2004. Outbreak of Salmonella serotype Enteritidis infections associated with raw almonds—United States and Canada, 2003–2004. MMWR Morb. Mortal. Wkly. Rep. 53(22):484–487. Centers for Disease Control and Prevention. 2007. Multistate outbreak of Salmonella serotype Tennessee infections associated with peanut butter—United States, 2006–2007. MMWR Morb. Mortal. Wkly. Rep. 56(21):521–524. Centers for Disease Control and Prevention. 2009. Multistate outbreak of Salmonella infections associated with peanut butter and peanut butter-containing products—United States, 2008–2009. MMWR Morb. Mortal. Wkly. Rep. 58(4):85–90. Centers for Disease Control and Prevention. 2011. Multistate outbreak of E. coli O157:H7 infections associated with in-shell hazelnuts. (Final update: April 7, 2011). http://www.cdc.gov/ecoli/2011/hazelnuts0157/ index.html. Chan, E. S, J. Aramini, B. Ciebin, D. Middleton, R. Ahmed, M. Howes, I. Brophy, I. Mentis, F. Jamieson, F. Rodgers, M. Nazarowec-White, S. C. Pichette, J. Farrar, M. Gutierrez, W. J. Weis, L. Lior, A. Ellis, S. Isaacs. 2002. Natural or raw almonds and an outbreak of a rare phage type of Salmonella enteritidis infection. Can. Commun. Dis. Rep. 28:97–99. Chang, S.-S., A. R. Han, J. I. Reyes-De-Corcuera, J. R. Powers, and D.-H. Kang. 2010. Evaluation of steam pasteurization in controlling Salmonella serotype Enteritidis on raw almond surfaces. Lett. Appl. Microbiol. 50:393–398.

SMP_Food Microbiology_CH09.indd

218

36. Chipley, R. J., and E. K. Heaton. 1971. Microbial flora of pecan meat. Appl. Microbiol. 22:252–253. 37. Chou, J. H., P. H. Hwang, and M. D. Malison. 1988. An outbreak of type A foodborne botulism in Taiwan due to commercially preserved peanuts. Int. J. Epidemiol. 17:899–902. 38. Clavero, M. R. S., R. E. Brackett, L. R. Beuchat, and M. P. Doyle. 2000. Influence of water activity and storage conditions on survival and growth of proteolytic Clostridium botulinum in peanut spread. Food Microbiol. 17:53–61. 39. Danyluk, M. D., M. T. Brandl, and L. J. Harris. 2008. Migration of Salmonella Enteritidis phage type 30 through almond hulls and shells. J. Food Prot. 71:397–401. 40. Danyluk, M. D., L. J. Harris, and D. W. Schaffner. 2006. Monte Carlo simulations assessing the risk of salmonellosis from consumption of almonds. J. Food Prot. 69:1594–1599. 41. Danyluk, M. D., T. M. Jones, S. J. Abd, F. SchlittDittrich, M. Jacobs, and L. J. Harris. 2007. Prevalence and amounts of Salmonella found on raw California almonds. J. Food Prot. 70:820–827. 42. Danyluk, M. D., A. R. Uesugi, and L. J. Harris. 2005. Survival of Salmonella Enteritidis PT 30 on inoculated almonds after commercial fumigation with propylene oxide. J. Food Prot. 68:1613–1622. 43. D’Aoust, J.-Y., J. Maurer, and J. S. Bailey. 2001. Salmonella species, p. 141–178. In M. P. Doyle, L. R. Beuchat, and T. J. Montville (ed.), Food Microbiology: Fundamentals and Frontiers, 2nd ed. ASM Press, Washington, DC. 44. de Jong, B., Y. Andersson, J. Giesecke, L. Hellström, U. Stamer, and R. Wollin. 19 July 2001. Salmonella Typhimurium outbreak in Sweden from contaminated jars of helva (or halva). Eurosurveill. Wkly. 5(29):1715. 45. Du, W.-X., S. J. Abd, K. L. McCarthy, and L. J. Harris. 2010. Reduction of Salmonella on inoculated almonds exposed to hot oil. J. Food Prot. 73:1238–1246. 46. Du, W.-X., M. D. Danyluk, and L. J. Harris. 2010. Efficacy of aqueous and alcohol-based quaternary a­mmonium sanitizers for reducing Salmonella in dusts generated in almond hulling and shelling facilities. J. Food Sci. 75:M7–M13. 47. Eglezos, S. 2010. The bacteriological quality of retaillevel peanut, almond, cashew, hazelnut, Brazil, and mixed nut kernels produced in two Australian nutprocessin­g facilities over a period of 3 years. Foodborne Pathog. Dis. 7:863–866. 48. Eglezos, S., B. Huang, and E. Stuttard. 2008. A survey of the bacteriological quality of preroasted peanut, almond, cashew, hazelnut, and Brazil nut kernels received into three Australian nut-processing facilities over a period of 3 years. J. Food Prot. 71:402–404. 49. Fisher, E. A., and P. Halton. 1928. A study of “rope” in bread. Cereal Chem. 5:192–208. 50. Food and Agriculture Organization. 2008. Climate Change: Implications for Food Safety. Food and Agriculture Organization, Rome, Italy.

Manila Typesetting Company

11/08/2012 06:54AM

9.  Nuts, Seeds, and Cereals

219

51. Freire, F. C. O., and L. Offord. 2002. Bacterial and yeast counts in Brazilian commodities and spices. Brazilian J. Microbiol. 33:145–148. 52. Glass, K. A., and M. P. Doyle. 1991. Relationship between water activity of fresh pasta and toxin production by proteolytic Clostridium botulinum. J. Food Prot. 54:162–165. 53. Guynot, M. E., A. J. Ramos, V. Sanchis, and S. Marin. 2005. Study of benzoate, propionate, and sorbate salts as mould spoilage inhibitors on intermediate moisture bakery products of low pH (4.5-5.5). Int. J. Food Microbiol. 101:161–168. 54. Hao, D. Y.-Y., E. K. Heaton, and L. R. Beuchat. 1989. Microbial, compositional, and other quality characteristics of pecan kernels stored at −20°C for twenty-five years. J. Food Sci. 54:472–474. 55. Hanlin, R. T. 1971. Fungi associated with young pecans. Proc. Georgia Pecan Growers’ Assoc. 2: 20–26. 56. Harris, L. J., A. R. Uesugi, S. Abd, and K. L. McCarthy. 2011. Survival of Salmonella Enteritidis PT 30 on inoculated almond kernels in hot water treatments. Food Res. Int. 54:1093–1098. 57. Hesseltine, C. W. 1968. Flour and wheat: research on their microbiological flora. Baker’s Digest 42:40–46. 58. Hyndman, J. B. 1963. Comparison of enterococci and coliform microorganisms in commercially produced pecan nut meats. Appl. Microbiol. 11:268–272. 59. Isaacs, S., J. Aramini, B. Ceibin, J. A. Farrar, R. Ahmed, D. Middleton, A. U. Chandran, L. J. Harris, M. Howes, E. Chan, A. S. Pichette, K. Campbell, A. Gupta, L. J. Lior, M. Pearce, C. Clark, F. Rodgers, F. Jamieson, I. Brophy, and A. Ellis. 2005. An international outbreak of salmonellosis associated with raw almonds contaminated with a rare phage type of Salmonella Enteritidis. J. Food Prot. 68:191–198. 60. Kader, A. A. 1996. In-plant storage, p. 274–277. In W. C. Micke (ed.), Almond Production Manual. Division of Agriculture and Natural Resources, University of California, Oakland. 61. Kajs, T. M., R. Hagenmaier, C. Vanderzant, and K. F. Mattil. 1976. Microbiological evaluation of coconut and coconut products. J. Food Sci. 41:352–356. 62. Kenney, S. J., and L. R. Beuchat. 2004. Survival, growth, and thermal resistance of Listeria monocytogenes in products containing peanut and chocolate. J. Food Prot. 67:2205–2211. 63. Killalea, D., L. R. Ward, D. Roberts, J. de Louvois, F. Sufi, J. M. Stuart, P. G. Wall, M. Susman, M. Schwieger, P. J. Sanderson, I. S. T. Fisher, P. S. Mead, O. N. Gill, C. L. R. Bartlett, and B. Rowe. 1996. International epidemiological and microbiological study of outbreak of Salmonella Agona infection from a ready to eat savoury snack. I. England and Wales and the United States. Br. Med. J. 313:1105–1107. 64. King, A. D., Jr., W. U. Halbrook, G. Fuller, and L. C. Whitehand. 1983. Almond nutmeat moisture and water activity and its influence on fungal flora and seed composition. J. Food Sci. 48:615–617.

SMP_Food Microbiology_CH09.indd

219

65. King, A. D., Jr., and T. Jones. 2001. Nut meats, p. 561– 563. In F. P. Downes and K. Ito (ed.), Compendium of Methods for the Microbiological Examination of Foods. American Public Health Association, Washington, DC. 66. King, A. D., Jr., M. J. Miller, and L. C. Eldridge. 1970. Almond harvesting, processing, and microbial flora. Appl. Microbiol. 20:208–214. 67. Kirk, M. D., C. L. Little, M. Lem, M. Fyfe, D. Genobile, A. Tan, J. Threlfall, A. Paccagenella, D. Lightfoot, H. Lyi, L. McIntyre, L. Ward, D. J. Brown, S. Surnam, and I. S. T. Fisher. 2004. An outbreak due to peanuts in their shell caused by Salmonella enterica serotypes Stanley and Newport—sharing molecular information to solve international outbreaks. Epidemiol. Infect. 132:571–577. 68. Kokal, D. 1965. Viability of Escherichia coli on English walnut meats (Juglans regia). J. Food Sci. 30:325–332. 69. Kokal, D., and D. W. Thorpe. 1969. Occurrence of Escherichia coli in almonds of Nonpareil variety. Food Technol. 23:227–232. 70. Kotzekidou, P. 1998. Microbial stability and fate of Salmonella Enteritidis in halva, a low-moisture confection. J. Food Prot. 61:181–185. 71. Kurtzman, C. P., and C. W. Hesseltine.1970. Chlorine tolerance of microorganisms found in wheat and flour. Cereal Chem. 47:244–246. 72. Ledet Müller, L., M. Hjertqvist, L. Payne, H. Pettersson, A. Olsson, L. Plym Forshell, and Y. Andersson. 2007. Cluster of Salmonella Enteritidis in Sweden 20052006—suspected source: almonds. Eurosurveillance 12(6):153–155. 73. Lee, L. E., D. Metz, M. Giovanni, and C. M. Bruhn. 2011. Consumer knowledge and handling of tree nuts: food safety implications. Food Prot. Trends 31(1):18–27. 74. Legan, J. D. 2000. Cereals and cereal products, p. 759– 783. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Food. Aspen Publishers, Inc., Gaithersburg, MD. 75. Little, C. 16 August 2001. International outbreak of Salmonella Typhimurium DT104—update from the United Kingdom. Eurosurveill. Wkly. 5(33):1700. 76. Little, C. L., W. Jemmott, S. Surman-Lee, L. Hucklesby, and E. de Pinna. 2009. Assessment of microbiological safety of edible roasted nut kernels on retail sale in England, with a focus on Salmonella. J. Food Prot. 72:853–855. 77. Little, C. L., N. Rawal, E. de Pinna, and J. McLauchlin. 2010. Survey of Salmonella contamination of edible nut kernels on retail sale in the UK. Food Microbiol. 27:171–174. 78. Ma, L., G. Zhang, P. Gerner-Smidt, V. Mantripragada, I. Ezeoke, and M. P. Doyle. 2009. Thermal inactivation of Salmonella in peanut butter. J. Food Prot. 72:1596–1601. 79. Marcus, K. A., and H. J. Amling. 1973. Escherichia coli field contamination of pecan nuts. Appl. Microbiol. 26:279–281. 80. Meyer, M. T., and R. H. Vaughn. 1969. Incidence of Escherichia coli in black walnut meats. Appl. Microbiol. 18:925–931.

Manila Typesetting Company

11/08/2012 06:54AM

Microbial Spoilage and Public Health Concerns

220 81. Miller, J. D. 1995. Fungi and mycotoxins in grains: implications for stored product research. J. Stored Prod. Res. 31:1–16. 82. Mixon, A. C. 1980. Potential for aflatoxin contamination in peanuts (Arachis hypogaea L.) before and after harvest—a review. J. Environ. Qual. 9:344–349. 83. Mortimer, P. R., and G. McCann. 1974. Food-poisoning episodes associated with Bacillus cereus in fried rice. Lancet 303:1043–1045. 84. O’Mahony, M., E. Mitchell, R. J. Gilbert, N. D. Hutchinson, N. T. Begg, J. C. Rodhouse, and J. E. Morris. 1990. An outbreak of foodborne botulism associated with contaminated hazelnut yoghurt. Epidemiol. Infect. 104:385–395. 85. Ominski, K. H., R. R. Marquardt, R. N. Sinha, and D. Abramson. 1994. Ecological aspects of growth and mycotoxin production by storage fungi, p. 287–312. In J. D. Miller and H. L. Trenholm (ed.), Mycotoxins in Grain: Compounds Other than Aflatoxin. Eagan Press, St. Paul, MN. 86. Park, E.-J., S.-W. Oh, and D.-H. Kang. 2008. Fate of Salmonella Tennessee in peanut butter at 4 and 22°C. J. Food Sci. 73:M82–M86. 87. Paterson, R. R. M., and N. Lima. 2010. How will climate change affect mycotoxins in food? Food Res. Int. 43:1902–1914. 88. Pettit, R. E., and R. A. Taber. 1968. Factors influencing aflatoxin accumulation in peanut kernels and the associated mycoflora. Appl. Microbiol. 16:1230–1234. 89. Pitt, J. I., S. K. Dyer, and S. McCammon. 1991. Systemic invasion of developing peanut plants by Aspergillus flavus. Lett. Appl. Microbiol. 13:16–20. 90. Pitt, J. I., A. D. Hocking, K. Bhudhasamai, B. F. Miscamble, K. A. Wheeler, and P. Tanboon-Ek. 1993. The normal microflora of commodities from Thailand. 1. Nuts and oilseeds. Int. J. Food Microbiol. 20:211–216. 91. Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott. 2010. Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. J. Food Prot. 73:1919–1936. 92. ProMED-mail. 2001. Salmonella Typhimurium DT104—Australia, Europe. Pro-MED-mail 2001; 22 August: 20010822.1980. http://www.promedmail. org. 93. ProMED-mail. 2006. Salmonellosis, boiled peanuts— USA (South Carolina). Pro-MED-mail 2006; 2 Nov: 20061102.3135. http://www.promedmail.org. 94. Rayman, M. K., J.-Y. D’Aoust, B. Aris, C. Maishment, and R. Wasik. 1979. Survival of microorganisms in stored pasta. J. Food Prot. 42:330–334. 95. Riemann, H. 1968. Effect of water activity on the heat resistance of Salmonella in “dry” materials. Appl. Microbiol. 16:1621–1622. 96. Riyaz-Ul-Hassan, S., V. Verma, A. Malik, and G. N. Qazi. 2003. Microbiological quality of walnut kernels and apple juice concentrate. World J. Microbiol. Biotechnol. 19:845–850.

SMP_Food Microbiology_CH09.indd

220

97. Rogers, R. F., and C. W. Hesseltine. 1978. Microflora of wheat and wheat flour from six areas of the United States. Cereal Chem. 55(6):889–898. 98. Rosengarten, F., Jr. 1984. The Book of Edible Nuts. Dover Publications, Inc., Mineola, NY. 99. Sauer, D. B., R. A. Meronuck, and C. M. Christensen. 1992. Microflora, p. 313–340. In D. B. Sauer (ed.), Storage of Cereal Grains and Their Products, 4th ed. American Association of Cereal Chemists, St. Paul, MN. 100. Schaffner, C. P., K. Mosbach, V. C. Bibit, and C. H. Watson. 1967. Coconut and Salmonella infection. Appl. Microbiol. 15:471–475. 101. Schebor, C., and J. Chirife. 2000. A survey of water activity and pH values in fresh pasta packed under modified atmosphere manufactured in Argentina and Uruguay. J. Food Prot. 63:965–969. 102. Scheil, W., S. Cameron, C. Dalton, C. Murray, and D. Wilson. 1998. A South Australian Salmonella Mbandaka outbreak investigation using a database to select controls. Aust. N.Z. J. Public Health 22:536–539. 103. Schroeder, H. W., and J. B. Storey. 1976. Development of aflatoxin in ‘Stuart’ pecans as affected by shell integrity. HortScience 11:53–54. 104. Scott, V. N., Y. U. H. Chen, T. A. Freier, J. Kuehm, M. Moorman, J. Meyer, T. Morille-Hinds, L. Post, L. Smoot, S. Hood, J. Shebuski, and J. Banks. 2009. Control of Salmonella in low-moisture foods. I. Minimizing entry of Salmonella into a processing facility. Food Prot. Trends 29:342–353. 105. Shachar, D., and S. Yaron. 2006. Heat tolerance of Salmo­nella enterica serovars Agona, Enteritidis, and Typhimurium in peanut butter. J. Food Prot. 69: 2687–2691. 106. Shohat, T., M. S. Green, D. Merom, O. N. Gill, A. Reisfeld, A. Matas, D. Blau, N. Gal, and P. E. Slater. 1996. International epidemiological and microbio­ logical study of outbreak of Salmonella agona infection from a ready to eat savoury snack. II. Israel. Br. Med. J. 313:1107–1109. 107. Smith, J. P., B. Ooraikul, W. J. Koersen, E. D. Jackson, and R. A. Lawrence. 1986. Novel approach to oxygen control in modified atmosphere packaging of bakery products. Food Microbiol. 3:315–320. 108. Sperber, W. H. 1982. Requirements of Clostridium botulinum for growth and toxin production. Food Technol. 36:89–94. 109. Sperber, W. H. 1983. Influence of water activity on foodborne bacteria—a review. J. Food Prot. 46:142–150. 110. Sperber, W. H. 2007. Role of microbiological guidelines in the production and commercial use of milled cereal grains: a practical approach for the 21st century. J. Food Prot. 70:1041–1053. 111. St. Clair, V. J., and M. M. Klenk. 1990. Performance of three methods for the rapid identification of Salmonella in naturally contaminated foods and feeds. J. Food Prot. 53:961–964. 112. Threlfall, E. J., M. D. Hampton, L. R. Ward, and B. Rowe. 1996. Application of pulsed-field gel electropho-

Manila Typesetting Company

11/08/2012 06:54AM

9.  Nuts, Seeds, and Cereals

113.

114.

115.

116.

117.

118.

119.

120.

221

resis to an international outbreak of Salmonella agona. Emerg. Infect. Dis. 2:130–132. Uesugi, A. R., M. D. Danyluk, and L. J. Harris. 2006. Survival of Salmonella Enteritidis phage type 30 on inoculated almonds stored at −20, 4, 23 and 35°C. J. Food Prot. 69:1851–1857. Uesugi, A. R., M. D. Danyluk, R. E. Mandrell, and L. J. Harris. 2007. Isolation of Salmonella Enteritidis phage type 30 from a single almond orchard over a five-year period. J. Food Prot. 70:1784–1789. Uesugi, A. R., and L. J. Harris. 2006. Growth of Salmonella Enteritidis phage type 30 in almond hull and shell slurries and survival in drying almond hulls. J. Food Prot. 69:712–718. Unicomb, L. E., G. Simmons, T. Merritt, J. Gregory, C. Nicol, P. Jelfs, M. Kirk, A. Tan, R. Thomson, J. Adamopoulos, C. L. Little, A. Currie, and C. B. Dalton. 2005. Sesame seed products contaminated with Salmonella: three outbreaks associated with tahini. Epidemiol. Infect. 133:1065–1072. U.S. Food and Drug Administration. 2005. Bulletin to the food service and retail food store industry regarding cake batter ice cream and similar products. (Revised August 19, 2005). http://www.fda.gov/Food/ FoodSafety/RetailFoodProtection/IndustryandRegul atoryAssistanceandTrainingResources/ucm114975. htm. U.S. Food and Drug Administration. 2005. The food defect action levels. http://www.fda.gov/food/guidance complianceregulatoryinformation/guidancedocuments/ sanitation/ucm056174.htm. U.S. Food and Drug Administration. 2009. Peanut Corporation of America Form 483 issued 2/05/2009. http://www.fda.gov/AboutFDA/CentersOffices/ORA/ ORAElectronicReadingRoom/ucm109818.htm. Valik, L., and F. Görner. 1993. Growth of Staphylococcus aureus in pasta in relation to its water activity. Int. J. Food Microbiol. 20:45–48.

SMP_Food Microbiology_CH09.indd

221

121. VanCauwenberge, J. E., R. J. Bothast, and W. F. Kwolek. 1981. Thermal inactivation of eight Salmonella serotypes in dry corn flour. Appl. Environ. Microbiol. 42:688–691. 122. Ward, L., S. Brusin, G. Duckworth, and S. O’Brien. 18 March 1999. Salmonella java phage type Dundee— rise in cases in England: update. Eurosurveill. Wkly. 3(12):1435. 123. Wareing, P. W., L. Nicolaides, and D. R. Twiddy. 2000. Nuts and nut products, p. 919–940. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Food. Aspen Publishers, Inc., Gaithersburg, MD. 124. Watkins, E. J. 1906. Ropiness in flour and bread and its detection and prevention. J. Soc. Chem. Ind. 25:350–357. 125. Wells, J. M. 1980. Toxigenic fungi isolated from lateseason pecans. J. Food Saf. 4:213–220. 126. Wells, L. 2007. Southeastern Pecan Growers’ Handbook. Bulletin 1327. University of Georgia Cooperative Extension Service, College of Agricultural and Environmental Sciences, Athens, GA. 127. Willford, J., A. Mendonca, and L. D. Goodridge. 2008. Water pressure effectively reduces Salmonella enterica serovar Enteritidis on the surface of raw almonds. J. Food Prot. 71:825–829. 128. Willis, C., C. L. Little, S. Sagoo, E. de Pinna, and J. Threlfall. 2009. Assessment of the microbiological safety of edible dried seeds from retail premises in the United Kingdom with a focus on Salmonella spp. Food Microbiol. 26:847–852. 129. Wilson, M. M., and E. F. Mackenzie. 1955. Typhoid fever and salmonellosis due to consumption of infected desiccated coconut. J. Appl. Bacteriol. 18:510–521. 130. Yang, J., G. Bingol, Z. Pan, M. T. Brandl, T. H. McHugh, and H. Wang. 2010. Infrared heating for dryroasting and pasteurization of almonds. J. Food Eng. 101:273–280.

Manila Typesetting Company

11/08/2012 06:54AM

SMP_Food Microbiology_CH09.indd

222

Manila Typesetting Company

11/08/2012 06:54AM

Foodborne Pathogenic Bacteria

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch10

Haiping Li Hua Wang Jean-Yves D’Aoust John Maurer

10

Salmonella Species†

In the early 19th century, clinical pathologists in France first documented the association of human intestinal ulceration with a contagious agent, with the disease later being identified as typhoid fever. Further investigations by European scientists led to the isolation and characterization of the typhoid bacillus responsible for typhoid fever and to the development of a serodiagnostic test for the detection of this serious human disease agent (137). Differential clinical and serological traits were used subsequently to identify the closely related paratyphoid organisms. In the United States, contemporary work by Salmon and Smith in 1885 led to the isolation of Bacillus cholerae-suis, now known as Salmonella enterica serovar Choleraesuis, from swine suffering from hog cholera (137). During the first quarter of the 20th century, great advances occurred in the serological detection of somatic (O) and flagellar (H) antigens within the Salmonella group, a generic term coined by Lignières in 1900 (137). An antigenic scheme for the classificaThis chapter is based on chapter 10 by Jean-Yves d’Aoust and John Maurer in the 3rd edition of Food Microbiology: Fundamentals and Frontiers. †

tion of salmonellae was first proposed by White in 1926 and subsequently expanded by Kauffmann in 1941 into the Kauffmann-White scheme, which presently includes more than 2,579 serovars (97).

CHARACTERISTICS OF THE ORGANISM

Taxonomy

Salmonella spp. are facultatively anaerobic gramnegative rod-shaped bacteria belonging to the family Enterobacteriaceae. Although members of this genus are motile by peritrichous flagella, nonflagellated variants such as Salmonella Pullorum and Salmonella Gallinarum and nonmotile strains resulting from dysfunctional flagella do occur. Salmonellae are chemo-organotrophic with an ability to metabolize nutrients by both respiratory and fermentative pathways. The bacteria grow optimally at 37°C and catabolize d-glucose and other carbohydrates, with the production of acid and gas. Salmonellae are oxidase negative and catalase positive, grow on citrate as a sole carbon source, generally produce hydrogen sulfide (H2S), decarboxylate lysine, and ornithine, and do not

Haiping Li, Division of Food Processing Science and Technology, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 6502 S. Archer Rd., Bedford Park, IL 60501. Hua Wang, Division of Microbiology, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740. Jean-Yves D’Aoust, Food Directorate, Health Products & Food Branch, Health Canada, Ottawa, Ontario, Canada. John Maurer, Department of Population Health, University of Georgia, Athens, GA 30602.

225

226 hydrolyze urea. Many of these traits formed the basis for the presumptive biochemical identification of Salmonella isolates. According to a contemporary definition, a typical Salmonella isolate would produce acid and gas from glucose in triple sugar iron (TSI) agar medium and would not utilize lactose (Lac−) or sucrose (Suc−) in TSI or in differential plating media such as brilliant green, xylose lysine desoxycholate (XLD), and Hektoen enteric (HE) agars. Typical Salmonella spp., which do not ferment lactose, form colorless or faint pink colonies on brilliant green agar, easily differentiated from the deep red colonies by a pink halo that is produced by lactosefermenting coliforms. Typical colonies on XLD agar plates appear pink with black centers and appear blue-green with black centers on HE agar plates due to H2S production. Many Salmonella strains may produce colonies with large, glossy black centers or appear almost completely black. The addition of lactose and sucrose on XLD and HE agars produces acid in excess by non-Salmonella H2S producers, which prevents their colonies from blackening. Additionally, typical salmonellae readily produce an acid reaction from glucose fermentation in TSI and an alkaline reaction from the decarboxylation of lysine to cadaverine in lysine iron agar, generate hydrogen sulfide gas in both media, and fail to hydrolyze urea (8, 63). The genetic variability arising from bacterial mutations and conjugative intra- and intergeneric exchange of plasmids encoding determinant biochemical traits continue to reduce the proportion of typical Salmonella biotypes. From the early studies of Le Minor and colleagues confirming that Salmonella utilization of lactose and sucrose was plasmid mediated (139, 140), many studies have since emphasized the occurrence of Lac+ and/or Suc+ biotypes in clinical specimens and food materials. This situation is of public health concern because biochemically atypical salmonellae could easily escape detection on disaccharidedependent plating media, which are commonly used in hospital and food industry laboratories. Bismuth sulfite (BS) agar remains a medium of choice for isolating salmonellae because, in addition to its high level of selectivity, it responds solely and most effectively to the production of extremely low levels of hydrogen sulfide gas (60). Those salmonellae that are lactose or sucrose utilizing will produce atypical colonies on XLD and HE agars but will produce typical colonies on BS agar plates. Typical colonies appear brown, gray, or black, sometimes with a metallic sheen and black halo. The diagnostic hurdles engendered by the changing patterns of disaccharide utilization by Salmonella are being further confounded by the increasing occurrence of biotypes that cannot decarboxylate lysine, that possess urease activity, that produce indole, and that readily grow in the

Foodborne Pathogenic Bacteria presence of potassium cyanide. Clearly, the recognition of Salmonella as a biochemically homogeneous group of microorganisms is becoming obsolete. The “one serovar, one species” concept is untenable because most serovars cannot be separated by biochemical tests. The situation has led to a reassessment of the diagnostic value of these and other biochemical traits and to their likely replacement with molecular technologies targeted at the identification of stable genetic loci and/or their products that are unique to the Salmonella genus. Nomenclature of the Salmonella group has progressed through a succession of taxonomical schemes based on biochemical and serological characteristics and on principles of numerical taxonomy and DNA homology (Table 10.1). In the early development of taxonomic schemes, determinant biochemical reactions were used to separate salmonellae into subgroups. The KauffmannWhite scheme was the first attempt to systematically classify salmonellae by using these scientific parameters. This major undertaking culminated in the identification of five biochemically defined subgenera (I to V) wherein individual serovars were afforded species status (122). Subgenus III included members of the Arizona-species (S. arizonae). Subsequently, a three-species nomenclatural system was proposed using 16 discriminating tests to identify S. typhi (single serovar), S. choleraesuis (single serovar), and S. enteritidis (all other salmonella serovars). The latter scheme recognized members of the Arizona group as a distinct genus (75). A defining development in Salmonella taxonomy occurred in 1973 when Crosa et al. (55) determined by DNA-DNA hybridization that all serotypes and subgenera of Salmonella and Arizona were related at the species level. The single exception to this observation was the subsequent description of S. bongori, previously known as subgenus V, as a distinct species. Another system based on numerical taxonomy and DNA relatedness proposed a single species (S. choleraesuis) consisting of seven subspecies (141). In numerical taxonomy, a statistical comparison of morphological and biochemical attributes of strains (phenetic analysis) measures the taxonomical proximity of test strains and allows for their separation into distinct taxa. In DNA homology, a high degree of hybridization of Salmonella reference DNA with extracts of test strains confirms the genetic relatedness of nucleic acid reactants and supports the inclusion of the test micro­ organisms into the Salmonella genus. A subsequent modification of the former scheme, while retaining the name of the seven recognized subspecies, changed the type species from S. choleraesuis to S. enterica (138). The following nomenclature is now officially accepted by the international community

10.  Salmonella Species

227

Table 10.1  Taxonomical schemes for Salmonella spp. Diagnostic basis

Salient features

Serovar designation

Reference

Biochemical

Five subgenera (I–V) Serovar = species status S. arizonae

S. typhimurium

122

Biochemical

Three species (S. typhi, S. choleraesuis, S. enteritidis) Arizona = separate genus

S. enteritidis serovar Typhimurium

75

Phenetic/DNA homology

Single species (S. choleraesuis) Seven subspecies (choleraesuis, salamae, arizonae, diarizonae, houtenae, bongori, indica) Type strain = S. choleraesuis

S. choleraesuis subsp. choleraesuis serovar Typhimurium

141

Phenetic/DNA homology

Single species (S. enterica) Seven subspecies (see above) Type strain = S. typhimurium LT 2

S. enterica subsp. enterica serovar Typhimurium

138

Multilocus enzyme electrophoresis

Two species (S. enterica [six subspecies] and S. bongori)

(Judicial Commission of the International Committee on Systematics of Prokaryotes, 2005). The genus Salmonella consists of two species, each of which contains multiple serovars (Table 10.2). The two species are S. enterica, the type species, and S. bongori, which was formerly subspecies V. S. enterica is divided into six subspecies, which are referred to by a Roman numeral and a name (I, S. enterica subsp. enterica; II, S. enterica subsp. salamae; IIIa, S. enterica subsp. arizonae; IIIb, S. enterica subsp. diarizonae; IV, S. enterica subsp.

Table 10.2  Species within the Salmonella genusa Salmonella species and subspecies S. enterica subsp. enterica (I) S. enterica subsp. salamae (II) S. enterica subsp. arizonae (IIIa) S. enterica subsp. diarizonae (IIIb)

No. of serovars 1,531 505 99 336

S. enterica subsp. houtenae (IV)

73

S. enterica subsp. indica (VI)

13

S. bongori (V) Total a

From references 34 and 97.

22 2,579

183

houtenae; and VI, S. enterica subsp. indica) (34, 183). S. enterica subspecies are differentiated on the basis of biochemical traits and genomic relatedness. The biochemical identification of foodborne and clinical Salmonella spp. isolates is generally coupled to serological confirmation, a complex and labor-intensiv­e technique involving the agglutination of bacterial surface antigens with Salmonella-specific antibodies. These include somatic (O) lipopolysaccharides (LPS) on the external surface of the bacterial outer membrane, flagellin (H) antigens associated with the peritrichous flagella, and the capsular (Vi) antigen which occurs only in S. Typhi, S. Paratyphi C, and S. Dublin (137). The heat-stable somatic (O) antigens are classified as major or minor antigens. The former category consists of antigens such as the somatic factors O:4 and O:3, which are specific determinants for the somatic groups B and E, respectively. In contrast, minor somatic antigenic components such as O:12 are nondiscriminatory, as evidenced by their presence in different somatic groups. Smooth (S) variants relate to strains with welldeveloped serotypic LPS that readily agglutinate with specific antibodies, whereas rough (R) variants exhibit incomplete LPS antigens resulting in weak or no agglutination with Salmonella somatic antibodies. Flagellar (H) antigens are heat-labile proteins, and individual Salmonella strains may produce one (monophasic) or

228 two (diphasic) sets of flagellar antigens. Although serovars such as S. Dublin produce a single set of flagellar (H) antigen, most serovars can alternatively elaborate two sets of antigens, i.e., phase 1 and phase 2 antigens. These homologous surface antigens are chromosomally encoded by the H1 (phase 1) and H2 (phase 2) genes and transcribed under the control of the vh2 locus (137). Capsular (K) antigens, commonly encountered in members of Enterobacteriaceae, are limited to the Vi antigen in the Salmonella genus. Thermal solubilization of the Vi antigen is necessary for the immunological identification of underlying serotypic LPS. Serological testing procedures seek to derive the complete antigenic formula of individual Salmonella i­solates, expressed as Salmonella (species) serotype (O antigens; H antigens, motile; H antigens, nonmotile). We shall use S. Infantis (6,7:r:1,5) as a working example. Commercially available polyvalent somatic antisera each consist of a mixture of antibodies specific for a limited number of major antigens, e.g., polyvalent B (Poly B) antiserum (BD Difco, Becton, Dickinson and Company, Sparks, MD) recognizes somatic (O) groups C1 (O6 and 7), C2 (O8), F (O11), G (O13 and 22), and H (flagellar). Following a positive agglutination with Poly B antiserum, single-group antisera representing the five somatic groupings included in the Poly B reagent would be used to identify the serogroup of the isolate. The test isolate would react with the C1 group antiserum, indicating that antigens 6 and 7 are present. Flagellar (H) antigens would then be determined by broth agglutination reactions using Poly H antisera or the Spicer-Edwards series of antisera. In the former assay, a positive agglutination reaction with one of the five polyvalent antisera (poly A to E; Becton, Dickinson and Company) would lead to testing with single-factor antisera to specifically identify the phase 1 and/or phase 2 flagellar antigens present. Agglutination in Poly C flagellar antiserum and subsequent reaction of the isolate with single-group H antisera would confirm the presence of the r antigen (phase 1). The empirical antigenic formula of the isolate would then be 6,7:r. Phase reversal in semisolid agar supplemented with r antiserum would immobilize phase 1 salmonellae at or near the point of inoculation, thereby facilitating the recovery of phase 2 cells from the edge of the zone of migration. Serological testing of phase 2 cells with Poly E and 1-complex antisera would confirm the presence of the flagellar 1 factor. Confirmation of the flagellar 5 antigen with single-factor antiserum would yield the final antigenic formula 6,7:r:1,5, which corresponds to S. Infantis. A similar analytical approach would be used with the Spicer-Edwards polyvalent H antisera, whereby the identification of flagellar (H) anti-

Foodborne Pathogenic Bacteria gens would arise from the pattern of agglutination reactions among the four Spicer-Edwards antisera and with three additional polyvalent antisera including the L, 1, and e,n complexes. Detection and isolation of Salmonella from foods by culture are based on three widely used reference methods: the U.S. Food and Drug Administration’s (FDA’s) Bacteriological Analytical Manual, the U.S. Department of Agriculture’s (USDA’s) Microbiology Laboratory Guidebook, and the International Organization for Standardization’s (ISO) Salmonella method ISO 6579: 2002 (E). All three reference methods include five steps including nonselective preenrichment of the sample h­omogenate, selective enrichment, selective plating, bio­ chemical screening, and biochemical/serological confirmation. Different media are used for each of the steps. Preenrichment is the initial and critical step that is necessary to allow injured Salmonella cells to resuscitate and proliferate to detectable levels. Selective enrichment suppresses the growth of competitive microflora while allowing uninjured Salmonella cells to proliferate. Direct selective enrichment or direct plating of the food homogenate may not allow the detection of low numbers of Salmonella cells that are injured and susceptible to the selective agents in the broth media or agars, so it is necessary to precede selective/differential plating with nonselective and selective enrichments. Plating onto selective/differential agars allows for the growth of distinct well-isolated Salmonella colonies while suppressing the growth of competitors. Various selective/differential plating media that produce discrete and well-isolated colonies have been developed. Use of more than one selective plating agar enhances the sensitivity of the method, since each agar uses a different selective/differential strategy, thus increasing the odds of producing a pure colony. It is recommended to use an agar that is capable of isolating atypical colonies, such as BS agar. All the presumptive-positive Salmonella colonies should be confirmed both biochemically and serologically. Biochemical screening is recommended by the Bacteriological Analytical Manual, Microbiology Laboratory Guidebook, and ISO culture methods to differentiate presumptive Salmonella spp. from nonSalmonella. The Salmonella culture method can take up to 6 days to obtain definitive negative results. Many commercial rapid methods for high-throughput Salmonella screening can produce a result in 2 days or less, whereas molecular assays using PCR technology generally require no more than a single day for results from a 24hour preenrichment medium. PCR technology has made significant advances in diagnostic microbiology, becoming an important tool for

10.  Salmonella Species the rapid detection of bacterial, viral, and protozoan pathogens. While it is unlikely that PCR techniques will ever completely replace culture-based methods for detection, this novel technology when appropriately utilized can help make important and timely decisions within the food microbiology laboratory. Important developments in the processing of samples for PCR analysis, increased sensitivity of PCR techniques, shortened detection time, and standardized use in foods have been reported. With recent advances in PCR technologies, real-time PCR and PCR-enzyme-linked immunosorbent assay (PCR-ELISA) may provide the user with a qualitative, yes/no answer as well as a quantitative assessment of pathogen load in foods (147). When the cell number of pathogens in the assay sample is low (<100 CFU/g), the ability to accurately quantitate may be limited because of the need for an enrichment step. Most PCR-based tests for Salmonella are targeted towards the detection of invA, a gene that is conserved and unique to this genus (180). Other genes with a more limited distribution in S. enterica have also been examined for their potential use in identifying specific Salmonella serovars (e.g., S. Enteritidis [66]), phage types (105), or specific, multidrug-resistant (MDR) Salmonella strains (42). PCR has also been examined as a possible tool for sero­typing S. enterica by targeting gene(s), gene combinations, or sequences responsible for the antigenic variability in S. enterica (106, 150). Several PCR-based molecular subtyping techniques, such as enterobacterial repetitive intergenic consensus (ERIC)-PCR, repetitive extragenic palindromic (REP)-PCR, and random amplified polymorphic DNA (RAPD) or arbitrarily primed (AP)-PCR, have also proven useful for the identification of strain differences and sources of foodborne outbreaks (80). In these subtyping approaches, primers target repetitive elements (ERIC-PCR or REP-PCR) (146, 224) or short 10-mer oligonucleotides (RAPD or AP-PCR) (146) that randomly bind within the bacterial genome, resulting in distinctive DNA patterns. The major advantages of these PCR subtyping methods are that they generate results quickly with a relatively high discriminatory value and they are inexpensive and practical compared to pulsedfield gel electrophoresis (PFGE) (6, 79, 80). Their drawback is that they are less reproducible than PFGE, since variations in amplification can lead to distinct changes in the fingerprinting profiles (79). Multilocus sequence typing (MLST) is another DNA sequence-based typing tool that targets ~7 genes for PCR amplification followed by sequencing to identify allele differences within a bacterial population (74). Genetic relatedness among clinical isolates can be inferred from collective comparison of all genes typed by this method. The advantages of

229 this methodology over PFGE are that they incur lower costs associated with DNA sequencing; that no additional, specialized equipment or software is required for a laboratory already set up for PCR; and that a database containing more than 48 microbial taxa is publicly available (http://pubmlst.org/databases3.shtml) (229). The challenges are to identify the set of genes that exhibit enough sequence diversity to be included in MLST for organism X and to obtain agreement from the scientific community on the genes selected for MLST. In recent years, the rapid advance of DNA/RNA sequencing technology has made whole-genomic sequencing a very promising tool in the investigation of foodborne outbreaks, as it can be used to assess the population structure of highly clonal, outbreak-related pathogens at a single-base resolution and can help identify temporal, geographical, and evolutionary origins of outbreaks. A study using a whole-genome single nucleotide polymorphism-based assay has demonstrated its discriminatory power by being able to distinguish between outbreakrelated and non-outbreak-related cases that were associated with a multistate Salmonella Montevideo outbreak originating from salami made with contaminated red and black peppers (67). This whole-genome single nucleotide polymorphism typing assay has provided resolution and accuracy levels for outbreak investigations that are unattainable by current subtyping methods. The rapid generation of whole-genome data of potential outbreak strains will help characterize and predict their phenotypic characteristics, so as to allow the development of rapid molecular assays for large-scale rapid screening in support of outbreak investigations (145).

Physiology

Growth

Salmonella spp. are resilient microorganisms that can adapt to conditions of temperature, pH, and water activity (aw) beyond their normal growth range, posing great risks to food safety. Although salmonellae are generally considered mesophilic in nature, some Salmonella strains can grow at elevated temperatures of up to 54°C, and others are able to grow in foods stored at 2°C to 4°C. The physiological adaptability of Salmonella spp. is further demonstrated by their ability to proliferate at pH values ranging from 3.99 to 9.5 with an optimum pH for growth of 6.5 to 7.5. Studies have revealed that Salmonella can grow in rehydrated dry soups with aw as low as 0.93 after 3 days’ incubation at 30°C (Table 10.3). Preconditioning of Salmonella cells to high or low temperature or acidic conditions can markedly increase their ability to grow under the respective conditions.

Foodborne Pathogenic Bacteria

230

Table 10.3  Physiological limits for the growth of Salmonella spp. in foods and bacteriological media Limit Parameter Temp (°C)

Minimum

Maximum

2 (24 h)

Minced beef

2 (2 days) 4 (£10 days) d

4.05e 9.5 aw

0.93

43

Minced chicken

Typhimurium

23

Shell eggsb

Enteritidis

126

Agar medium

Typhimurium

72

Tomatoes

Infantis Anatum Tennessee Senftenberg

19

Typhimurium

109

Oranienburg

215

Liquid medium

f

Reference

Typhimurium b

54.0 3.99

Serovar

a

c

pH

Product

Egg wash waterb Rehydrated dried soup

b

48

Naturally contaminated. Artificially contaminated. Mutants selected to grow at elevated temperature. d Growth within 24 h at 22°C. e Acidified with HCl or citric acid; growth within 24 h at 30°C. f Growth within 3 days at 30°C. a

b c

The prolonged exposure of S. Typhimurium to thermal stress conditions results in mutants capable of growth at 48°C (ttl) and 54°C (mth). Similarly, preconditioning of cells to low temperatures can markedly increase the growth of salmonellae in refrigerated food products. Salmonella preconditioned on pH gradient plates could grow in liquid and solid media at considerably lower pH values than the same strains without preconditioning. These findings indicate that the microbial growth controls provided by refrigerated storage, thermal processing, or fermentation may present environments for the proliferation of adapted salmonellae. The dynamic interaction among various extrinsic growth factors (e.g., salt concentration, pH, and temperature) further expands conditions for adaptation by Salmonella spp. Although Salmonella is generally inhibited in the presence of 3 to 4% NaCl, tolerance to salt can increase with increasing temperature in the range of 10 to 30°C. At 30°C, Salmonella was able to grow in the presence of 6% NaCl, albeit with a protracted lag phase and a decreased rate of growth. Similarly, studies on the interrelationship between pH and NaCl have underscored the dominant impact of pH on the growth of Salmonella; however, the presence of low concentrations of salts in acidified foods reduced the inhibitory effect of organic acids and stimulated the growth of S. Enteritidis in broth medium acidified to pH 5.19 with acetic acid (179). Such findings, as well as other reports of the enhanced salt-dependent survival of salmonellae in rennet whey (pH 4.8 to 5.6) and mayonnaise, indicate that low levels of salt can undermine the preservative action of

organic acids and potentially compromise the safety of fermented and acidified foods. The complex interaction of inhibitory growth factors is also illustrated by the proliferation of salmonellae in inoculated raw minced beef and cooked crab meat stored at 8°C to 11°C under modified atmospheres containing low levels of CO2 (20 to 50%, vol/vol) (113). Normally, gaseous mixtures con­ sisting of 60 to 80% (vol/vol) CO2 with varying proportions of N2 and/or O2 can inhibit the growth of aerobic spoilage microorganisms such as Pseudomonas spp. without promoting the growth of Salmonella spp.; thus, refrigerated storage of foods packaged under vacuum or modified atmosphere is widely used to extend shelf life. The proliferation of salmonellae at lower levels of CO2/low temperature warrants caution in the general application of this technology (22). The studies of interactive forces generated by temperature, pH, and salt on the growth and survival of Salmonella have led to the development and application of mathematical models to predict the fate of salmonellae in foods. In this approach, the growth, survival, or inactivation of a target microorganism under various conditions of pH, NaCl, temperature, or other environmental factors of interest is laboriously characterized in laboratory media. The generated data are then used to derive mathematical models that depict the response of the microorganism under different combinations of environmental factors. Predictive models are not without limitations. For example, the use of a model to predict the growth of salmonellae in a food whose salt content lies beyond the range of values originally studied for the

10.  Salmonella Species derivation of the mathematical model would likely lead to erroneous conclusions. Moreover, models are generally based on the behavior of a few Salmonella strains under selected environmental conditions. The physiological diversity among the large number of serovars may unduly challenge the reliability of models in predicting bacterial growth responses. Additionally, the lot-to-lot variations in the composition of a given food and in the types and numbers of background microflora are variables that could seriously undermine the predictive capability of mathematical models. Developments in predictive modeling are ongoing, and current modeling approaches for Salmonella and other foodborne bacterial pathogens from various aspects have been compiled (120).

Survival

The resistance and survival of foodborne salmonellae to inactivation processes and hostile environments are often the reasons underlying many food-associated Salmonella outbreaks. The resistance that salmonellae demonstrate to heat, chemical sanitizers or preservatives, low pH, and aw ultimately plays an important role in causing human disease. Moreover, Salmonella, like many other pathogens, has demonstrated a cross-protectio­n against multiple stresses, in which the exposure to one stress is able to induce resistances in Salmonella against other subsequent stresses. For example, the growth of S. Typhimurium at pH 5.8 engendered an increased thermal resistance at 50°C, an enhanced tolerance to high osmotic stress (2.5 M NaCl), a greater surface hydrophobicity, and an increased resistance to the antibacterial lactoperoxidase system and surface-active agents such as crystal violet and polymyxin B (143). Similarly, Salmonella cells surviving from long-term starvation and desiccation stresses demonstrate significant resistance to thermal, chemical, and other intervention processes. The attributes of Salmonella to resist physical and chemical stresses and its widespread occurrence in nature present a great food safety challenge for the food industry.

Desiccation Resistance

Low-moisture foods are traditionally characterized as low-risk foods, since the low aw is inhibitory to pathogen growth; however, an increasing number of multistate Salmonella outbreaks associated with dry foods have occurred (45, 173). Regardless of the route of the contamination in each case, the survival of Salmonella under these dry conditions has been determined as the cause of these outbreaks. In the 2008-2009 Salmonella outbreak, S. Typhimurium was isolated from an un-

231 opened 5-pound container of King Nut Brand peanut butter with a production date of ~5 months earlier, showing that Salmonella remained viable in the lowmoisture environment for at least 5 months. Numerous laboratory studies have shown that Salmonella can survive for months and up to a few years under certain dry conditions or in low-moisture food matrices. The aw of food matrices, product formulation, and storage temperature critically affect the survival of Salmonella in dry food matrices (207). A study of surviva­l of Salmonella on dry surfaces at different equilibrated relative humidities from 11% to 97% demonstrated an inverse correlation between the desiccation survival rate and aw. Comparing the impacts of food formulation on the survival of salmonellae in real food matrices is challenging because it is difficult to create different formulations without affecting their aw. Studies have shown that the survival of Salmonella on dried paper disks and dried squids is drastically increased by the addition of sucrose, in comparison to corresponding conditions without sucrose (108). Lower storage temperature enhances the survival of Salmonella in foods stored in dry conditions. S. Enteritidis is able to survive in almonds for more than 550 days with no significant log reduction at 4°C, but with approximately a 3- to 4-log reduction at 23°C (209). Similarly, different serotypes of Salmonella cells survived on dried paper disks for 35 to 70 days when stored at 25°C and 35°C but survived for 22 to 24 months when stored at 4°C (108). When a composite of Salmonella strains was inoculated at 1.5 log CFU/g, six of the seven products evaluated were positive for the pathogen at 5°C, but all products (except one peanut butter spread) were negative for Salmonella after storage at 21°C for 24 weeks (37). The mechanism by which Salmonella survives on a dry surface or in low-moisture foods remains elusive. Changes in outer membrane LPS, thin aggregative fimbriae (curli), cellulose formation, and filaments are observed in Salmonella grown in low-aw aqueous solutions or during prolonged growth on agar plates. The formation of multicellular rdar morphology involving thin aggregative fimbriae and cellulose increases the long-term survival of Salmonella, and the defective LPS capsule formation reduces the resistance of Salmonella to desiccation. When colonies that had formed rdar morphology after prolonged growth on agar were transferred to plastic microtiter plate wells, the cells remained viable for months (227). Although the formation of multicellular filamentous cells is a major morphological change induced in Salmonella by low-aw aqueous solution, its role in desiccation resistance has not been fully illustrated (196). It is noteworthy that the nutrient content

232 and aw environment of the agar plate or broth provided the conditions for inter- and intracellular physiological adaptation by Salmonella. Nonetheless, Salmonella cells associated with dry processing environments and extremely low-aw foods are often exposed to more stringent desiccation and may totally lack the nutrient and water for the development of the rdar morphology and filamentous cells. It is not clear how bacteria respond immediately to the desiccation stresses associated with dry foods and processing environments. Transcriptome microarray analysis has shown that fatty acid metabolism, accumulation of osmoprotective compatible solutes, and transporters were the major functional groups upregulated more than fivefold in cells stored under extremely dry conditions (equilibrated relative humidity, 11%). Studies showed that low pH and high growth temperature change the ratios among unsaturated, saturated, and branched-chain fatty acids and reduce the membrane fluidity in Salmonella (3). The change in fatty acid metabolism under desiccation conditions likely affects the fluidity of the cell membrane and thus the ability to protect the cell from loss of cytoplasmic water. Accumulation of ionic and nonionic compatible solutes is considered to be the major mechanism by which bacteria respond to environmental osmotic pressure shifts. In S. Tennessee and S. Typhimurium LT2, transporter genes associated with the nonionic solutes betaine glycine, choline, and proline betaine are upregulated more than fivefold when Salmonella is exposed to air drying. The proteins form transport machineries, encoded by opuBB, proXVW, STM1492, and STM 1493, located primarily on the inner membrane and spanning into the periplasmic and cytoplasmic compartments, indicating that solute transport via the above transporters occurs to a greater extent between the two compartments than into the extracellular space. Salmonellae surviving long-term starvation and desiccation stresses demonstrate significant resistance to thermal, chemical, and other intervention processes. Inactivation of Salmonella in many low-moisture foods and processing environments proves to be a challenge. Preventing Salmonella from contaminating the dry food products is a key step to improve the safety of dry foods.

Thermal Resistance

Heat is widely used in food manufacturing processes to control the microbiological quality and safety of products. Salmonellae inoculated on the surface of chicken breast, in nugget meat blend, and in broth with high aw do not exhibit unusually high thermal resistance when normal cooking conditions are applied; however, stress

Foodborne Pathogenic Bacteria factors, such as low aw and acidic conditions, present in food matrices drastically increase the thermal resistance of salmonellae. Thermal tolerance of Salmonella in lowmoisture foods currently remains a significant challenge for the dry food industry. Numerous studies have reported that D values of Salmonella thermal inactivation are drastically increased in various dry foods compared to high-aw aqueous systems. The D value represents the amount of time required to reach a 90% kill (1.0-log10 reduction) in the number of viable cells upon heating at a specified constant temperature. Low aw in those food matrices is the decisive factor leading to the increased D value. It was reported that the D value of S. Anatum in molten chocolate was up to 1,200 min at 71.1°C; however, 1 and 4% additional water reduced it to 520 and 210 min, respectively (45). A similar study showed that D65.5°C of S. enterica serovar Typhimurium in molten chocolate was 396 min; in contrast, D65.5°C in chocolate syrup (aw  = 0.84) was only 2.7 min (45). The high D values of Salmonella reflect the low efficiency of thermal inactivation in dry foods, such as flour, nuts, butter, dry milk, and chocolate. The mechanisms of Salmonella thermal resistance under low-moisture conditions are largely unknown. One of the theories regarding bacterial thermal inactivation is “water vibration” such that the heat causes the vibration of water molecules, which in turn breaks the disulfide and hydrogen bonds in the adjacent proteins, leading to changes in the tertiary structure; thus, the bacterial cell loses viability. In Salmonella cells exposed to desiccation stress, a limited amount of water is present, which reduces the vibration and denaturation of the proteins; thus, cells demonstrate thermal resistance (73). Following exposure of S. Tennessee to 11% relative humidity for 5 days, trehalose production was significantly increased, suggesting that trehalose may provide a protective effect on Salmonella surviving under dry conditions. The “water replacement” theory suggests that trehalose can replace water molecules to form hydrogen bonds with the polar residues of lipids and proteins, resulting in a more stable structure with reduced vibrations from heat; hence, the desiccated Salmonella cells become resistant to heat. As a general rule, the heat resistance of Salmonella spp. increases as the aw of the heating menstruum decreases; however, exceptions are observed, indicating that additional factors affect Salmonella thermal resistance. In aqueous systems at intermediate aw levels, the solutes used to alter the aw of the heating menstruum play a determining role in the level of acquired heat re-

10.  Salmonella Species sistance. For example, the heating of S. Typhimurium in menstrua adjusted to an aw of 0.90 with sucrose and glycerol conferred different levels of heat resistance, as evidenced by D57.2 values of 40 to 55 min and 1.8 to 8.3 min, respectively (95). This indicates that ingredients in a food matrix also affect the thermal resistance of Salmonella in ways other than simple changes in aw. Other important features associated with adaptive responses include the greater heat resistance of salmonellae grown in nutritionally rich than in minimal media, with cells derived from stationary- rather than logarithmicphase cultures, and cells exposed to the sublethal temperatures. The ability of Salmonella to acquire greater heat resistance following exposure to elevated sublethal temperatures is a universal biological response and is most intensively studied. The phenomenon stems from a rapid adaptation of the bacterium to rising temperatures in the microenvironment to a level of enhanced thermotolerance quite distinct from that described in conventional time-temperature curves of thermal lethality. This adaptive response has potentially serious implications in the safety of thermal processes that expose or maintain food products at marginally lethal temperatures. Exposure of salmonellae to sublethal temperatures induces the translation of heat shock proteins (HSPs), such as DnaK and GroEL. HSPs are a group of proteins that function mainly as chaperones and proteases to control the proper folding of proteins or promote their degradation. Nonetheless, most of the sublethal heat shock studies are carried out at temperatures below 50°C, which cannot simulate the thermal stresses in food manufacturing processes, which are often above 70°C and up to 130°C. It is questionable whether cells exposed to high inactivation temperature have sufficient time or are under appropriate conditions to synthesize HSPs. The likelihood that other protective cellular functions are triggered by heat shock stimuli cannot be discounted.

Acid Resistance

Brief exposure of S. Typhimurium to mildly acidic environments of pH 5.5 to 6.0 (preshock) followed by exposure of the adapted cells to pH 4.5 (acid shock) triggers a complex acid tolerance response (ATR) that potentiates the survival of the microorganism under extremely acid environments (pH 3.0 to 4.0). The response translates into an induced synthesis of 50 acid shock and outer membrane proteins (OMPs), reduced growth rate, and pH homeostasis as demonstrated by the bacterial maintenance of internal pH values of 7.0 to 7.1 and 5.0 to 5.5 upon sequential exposure of cells to external pHs of 5.0 and 3.3, respectively. In the ATR,

233 bacterial Mg2+-dependent proton translocating ATPase encoded by the atp operon plays an important role in maintaining cellular pH at ³5.0 through an energydependent transport of intracellular protons to the cell exterior. Other transmembrane mechanisms that putatively operate in pH homeostasis include the H+-coupled ion transport systems (antiport) for the intra- and extracellular transfer of K+, Na+, and H+ ions, the electron transport chain-dependen­t efflux of H+, and transport systems committed to the symport of H+ and solutes (168). The Fe2+-binding regulatory protein encoded by the fur (ferric uptake regulator) gene also impacts bacterial acid tolerance, as evidenced by the inability of fur mutant strains to survive under highly acidic conditions (88). Acid-induced activation of amino acid decarboxy­ lases in Salmonella provides an additional protective mechanism whereby cadaverine and putrescine from the enzymic breakdown of lysine and ornithine, respectively, potentiate acid neutralization and enhanced bacterial survival (168). Further characterization of the Salmonella response to acid stress has led to the identification of two additional protective mechanisms that operate in salmonellae in the stationary phase of growth and are distinct from the previously discussed ATR, which prevails in log-phase cells (86, 141). One of these pH-dependent responses, designated stationary-phase ATR, provides greater acid resistance than the log-phase ATR, is induced at pHs of <5.5, and functions maximally at pH 4.3. This s­tationary-phase ATR induces the synthesis of only 15 shock proteins, is not affected by mutations in the atp and fur genes, and is rpoS independent (135). The induction of the remaining acid protective mechanism associated with Salmonella in the stationary phase is independent of external pH and dependent on the alternative sigma factor (ss) encoded by the rpoS locus. The mechanism seemingly reinforces the ability of stationary-phase cells to survive under hostile environmental conditions. We have thus seen that three possibly overlapping cellular systems confer acid tolerance in Salmonella spp. These include (i) the pH-dependent, rpoS-independent log-phase ATR; (ii) the pH-dependent and rpoS-independent stationary-phase ATR; and (iii) the pH-independent, rpoS-dependent stationary-phase acid resistance. These systems likely operate in the acidic environments that prevail in fermented and in acidified foods and in phagocytic cells of the infected host. Acid stress can also trigger enhanced bacterial resistance to other adverse environmental conditions. The growth of S. Typhimurium at pH 5.8 engendered an increased thermal resistance at 50°C, an enhanced tolerance to high osmotic stress (2.5 M NaCl) ascribed to the

234 induced synthesis of the OmpC OMPs, a greater surface hydrophobicity, and an increased resistance to the antibacterial lactoperoxidase system and surface-active agents such as crystal violet and polymyxin B (143).

INTERVENTIONS AND PREVENTIVE CONTROLS IN FOODS The widespread occurrence of Salmonella in the natural environment, the intensive husbandry practices used in the meat, fish, and shellfish industries, and the recycling of offal and inedible raw materials into animal feeds have contributed to the persistence of this human bacterial pathogen in the global food chain (60). Poultry meat and eggs are predominant reservoirs of Salmonella in many countries, and they overshadow the importance of other meats such as pork, beef, and mutton as potential vehicles of infection (60). In an effort to actively address the problem of Salmonella in meat products, the USDA Food Safety Inspection Service (FSIS) published in July 1996 the final rule on pathogen reduction and hazard analysis and critical control point (HACCP) systems. This final rule requires the meat and poultry industry to implement HACCP plans in all plants and requires the systematic sampling and testing of final products for the indicator organism Escherichia coli biotype I. For the poultry industry, HACCP-based intervention strategies include the implementation of various hurdles in the whole process from breeder flocks to the final meat and egg production. Careful management of vaccination and testing of breeder flocks is the first major step to reduce the incidence of Salmonella in poultry. Since salmonellae can be transmitted from hen to egg, prevention of colonization by vaccination is a key control point. Minimizing the exposure and carriage of Salmonella in the hatchery and primary production environments are approached by various disinfection techniques, biosecurity rules, and litter/water/feed management schemes. HACCP principles are applied through transportation and meat processing to minimize cross-contamination of the pathogen, for example, from chicken to chicken, and from the viscera to the carcass. Baseline studies prior to the implementation of the HACCP rule and FSIS testing in 1996 indicated a Salmonella contamination rate of about 24% (1996) in U.S. broiler chickens. In 1999, the rate of Salmonella contamination on processed broiler chickens had been reduced to about 11%, and it further dropped to 6.7% in 2010 (221). The continuing pandemic of human S. Enteritidis phage type 4 (Europe) and phage type 8 (North America) infections associated with the consumption of raw or lightly cooked shell eggs and egg-containing products

Foodborne Pathogenic Bacteria further emphasizes the importance of poultry as vehicles of human salmonellosis and the need for sustained and stringent bacteriological control of poultry husbandry practices. This egg-related pandemic is of particular concern because the problem arises from transovarian transmission of the infective agent into the interior of the egg prior to shell deposition. The viability of these internalized S. Enteritidis cells remains unaffected by the egg surface sanitizing practices currently applied in egggrading stations. Studies on intact egg pasteurization using thermal, microwave, and ionization irradiation technologies have been conducted. Thermal processes using different temperature-time combinations have been effective; for example, 57°C for 25 min in a water bath followed by 55°C for 57 min in a dry oven resulted in a >5-log reduction of salmonellae with no destruction of the overall quality of the eggs (133). The USDA pasteurization standard is 60°C for 3.5 min. The limitation of water bath pasteurization is the high water demand, long come-up time to achieve effective temperature, and slow turnover rate, all of which hinder the potential for using water bath as an in-line processing technique. Microwave pasteurization of intact eggs is considered to be superior to the water bath method in terms of efficiency; however, different microwave equipment designs have led to variable inactivation rates, ranging from 2to more than 5-log reductions of Salmonella in the eggs (133). In 2000, the FDA approved the application of up to a 3-kGy dose of irradiation to inactivate the organism in whole fresh eggs. Studies have shown that the irradiation efficacy is very effective overall and the inactivation efficacy is dose dependent as well as Salmonella serovar specific. The major barrier to using ionizing irradiation as a control measure is consumer resistance and the costly implementation; thus, it is not the long-sought silver bullet for eliminating Salmonella contamination in eggs. Preventive control should be regarded as the front line of defense for controlling salmonellae in eggs. In 2009, the FDA issued the Egg Rule, which requires shell egg producers to implement stringent measures to systematically monitor S. Enteritidis, from pullets to the environment, as preventive measures during production, storage, and transportation. The producers also need to register with the FDA and to maintain records documenting their compliance with the rule (222). Rapid depletion of feral stocks of fish and shellfish in recent years has greatly increased the importance of the international aquaculture industry as an alternate source for these popular food items. The high-density farming conditions required to maximize biological yields and to satisfy growing market demands open gateways to the widespread infection of species reared

10.  Salmonella Species in earthen ponds and other unprotected facilities that are continuously exposed to environmental contamination. It is noteworthy that much of the currently available aquacultural products originate from the Asian, African, and South American continents. The use of raw meat scraps and offal, of night soil potentially contaminated with typhoid and paratyphoid salmonellae, and of Salmonella-contaminated animal feeds and feces is not uncommon in these geographic areas. Clearly, such husbandry practices favor widespread bacterial contamination of rearing ponds. Moreover, aquaculture products from developing countries could be contaminated with highly resistant salmonellae arising from the use of subtherapeutic levels of antibiotics in rearing ponds to prevent disease and to promote the growth of aquaculture species. The human health risks associated with the consumption of lightly cooked aquaculture fish or shellfish or raw sushi dishes potentially contaminated with Salmonella spp. cannot be minimized. Chlorine has been used in washing to decontaminate pathogenic microorganisms in seafoods, but the interference of organic matter with chlorine’s bactericidal efficacy and the formation of carcinogenic by-products limit its use for ready-to-eat (RTE) seafoods. Alternative treatments, such as ionizing radiation, have been evaluated, and 2.0 kGy of X-ray and 3.0 kGy of gamma radiation are able to inactivate more than 5 to 6 log of Salmonella in RTE shrimp and oysters (115, 151). Fresh produce has gained notoriety in recent years as a vehicle of human salmonellosis. On the one hand, the situation has developed from the increased public demand for fresh produce as part of a healthy diet. On the other hand, greatly improved detection methods and traceback abilities facilitate the identification of food sources in general. The increase in global trade of fresh produce from developing countries further adds to the safety risk of produce, since the prevailing hygienic conditions during the production, harvesting, and distribution of products in these countries do not always meet minimum standards and thus facilitate product contamination. More specifically, the fertilization of crops with untreated sludge or sewage effluents potentially contaminated with Salmonella, the irrigation of garden plots and fields and the washing of fruits and vegetables with contaminated waters, the repeated handling of product by local workers, and the propensity for environmental contamination of spices and other condiments during drying in unprotected facilities constitute weak production links that undermine food safety. Nonetheless, the lack of effective control measures is the ultimate challenge for keeping fresh produce free from Salmonella contamination, even in developed countries. Therefore,

235 the implementation of preventive control strategies such as HACCP plays a remarkably important role in enhancing the bacterial quality and safety of this group of products. Each commodity needs a specific analysis to define the critical points in the entire process that potentially introduce pathogens into the final products. FDA, USDA, and the fresh produce industry have been working together to develop and improve guidelines to minimize microbial hazards in tomatoes, leafy greens, and other commodities. The basic concept underlying all of the guidelines is to minimize the chance of introducing pathogen to the produce. Good agricultural practices or good manufacturing practices (GMP) are required for each step from preharvest to postharvest, e.g., field irrigation with treated effluents, fertilization of soils with fully composted animal wastes, limiting intrusion of wild animals into growing areas, washing of fruits and vegetables with potable and/or bactericidal wash waters, prevention of cross-contamination during the postharvest washing steps, education of local workers on the hygienic handling of fresh produce, and greater protection of products from environmental contamination during all phases of production and marketing. Since no single treatment is fully effective, a combination of strategies may prove to be the best available approach to reduce microbial contamination in fresh produce. Low-moisture foods, such as peanut butter, flour, nuts, alfalfa seeds for sprouting, cereals, spices, and chocolate, have caused a large number of cases of human salmonellosis in the past decade. Although lowmoisture conditions are not favorable to the growth of Salmonella, its ability to survive under these conditions has been responsible for the outbreaks. A particular challenge for controlling salmonellae in this group of foods is the low heat transfer rate and the thermal resistance of the pathogen in dry foods. Another major challenge is cross-contamination in the final products after an effective kill step. In the latter, lack of GMP can introduce the pathogen from the environment, raw ingredients, and workers. In both cases, a good HACCP plan will greatly reduce Salmonella contamination in dry foods. In 2009, the Grocery Manufacturers Association established a low-moisture food task force, which recommended seven elements of manufacturing practices to minimize the risk of Salmonella contamination. The overall principles are to prevent the presence, spread, and growth of Salmonella through the design and maintenance of building and equipment, hygiene/pest control, strict zoning, and strict sanitation, as well as the implementation of validation, verification, and corrective action of control measures. Thermal inactivation of salmonellae on almonds is one of the few successful

236 control measures. The Almond Board of California’s Technical Expert Review Panel has recommended oil roasting at or above 260°F for 2 min to reach a 5-log reduction of Salmonella. To inactivate Salmonella under low-moisture conditions, other thermal and nonthermal processes that have been validated in the laboratory include dry- and moist-air convection heating, propylene oxide or ethylene oxide fumigation, low-energy X-ray irradiation, and cold plasma. The efficacy of each technique is largely affected by the food matrix, aw, and the Salmonella strain and surrogate used; a process can be fully effective in one commodity but not feasible in another food product. Moreover, cleaning and sanitation of the processing equipment and environment are prominent issues, since Salmonella’s heat resistance under dry conditions renders dry cleaning ineffective; nevertheless, dry cleaning is still the best available option in order to avoid introducing water into the environment. In all, the development of effective processes to inactivate Salmonella under dry conditions continues to be one of the top priorities in battling foodborne salmonellae.

FOODBORNE OUTBREAKS National annual disease statistics continue to underscore the importance of Salmonella spp. as a leading cause of foodborne bacterial illnesses in humans. The number of reported incidents of foodborne salmonellosis tends to dwarf those of illnesses associated with other foodborne pathogens. Although the incidence of human foodborne salmonellosis seemingly is considerably greater in some countries, national disease statistics should be considered with caution because of differences in the inclusivity and refinement of national reporting systems, in the ability of health agencies to conduct comprehensive and timely epidemiological investigations, and in the availability and reliability of clinical and food analysis data. Worldwide trends in the incidence of human foodborne salmonellosis are revealing. Several countries have reported decreases in the annual number of incidents in recent years, whereas other countries report little change in the occurrence of foodborne salmonellosis. Major outbreaks of foodborne salmonellosis in the last few decades are of interest because they underline the multiplicity of foods and Salmonella serovars that have been implicated in human illness (see Tables 10.5 through 10.9). Dairy products have been incriminated in large outbreaks of human salmonellosis (Table 10.4). The outbreaks of salmonellosis in Australia and Scotland r­eveal the persistence of bacterial disease outbreaks in the United States from the consumption of raw fluid

Foodborne Pathogenic Bacteria milk, which many consider a health fetish (178). In 1984, Canada experienced a large outbreak of no fewer than 2,700 confirmed cases of S. Typhimurium PT 10 infections from the consumption of Cheddar cheese manufactured from heat-treated and pasteurized milk. Manual override of the flow diversion valve reportedly led to the entry of raw milk into vats of thermized and pasteurized cheese milk (59). The following year witnessed the largest outbreak of salmonellosis in the United States, in which 16,284 confirmed cases of illness were associated with the consumption of pasteurized fluid milk (134, 189). Although the cause of this outbreak was never ascertained, a cross-connection between raw and pasteurized milk lines seemingly was at fault. In 1994, a major outbreak of foodborne salmonellosis in the United States was associated with ice cream contaminated with S. Enteritidis. The transportation of pasteurized ice cream mix in an unsanitized truck that had previously carried raw eggs was identified as the source of contamination (10, 104). In 1998, a nationwide outbreak of S. Enteritidis PT 8 in Canada was associated with contaminated Cheddar cheese included in retail luncheon packages consumed primarily by children. From 1999 to 2011, serovars such as S. Newport, S. Typhimurium, S. Dublin, S. Montevideo, and S. Java were reported to have caused outbreaks associated with consumption of milk, cheese, and their products in the United States. Most of these instances of salmonellosis were related to the consumption of unpasteurized cheese. In 2008, a rare serotype, S. Kedougou, isolated from a commercial, presumably milk-based powdered infant formula, was the likely vehicle of transmission in this Salmonella outbreak in Spain. A more recent S. Oranienburg outbreak, associated with milk powder, sickened 13 infants, 1 child, and 2 adults in Russia from November 2011 to January 2012. In the last decade, fresh fruits and vegetables figured prominently as vehicles of human salmonellosis (Table 10.5). Many factors contribute to this situation including the perishable nature of these products and consumption as an RTE food without any intervention step (i.e., cooking) prior to consumption. Moreover, the growing popularity of convenience foods such as precut and prepackaged produce introduces new public health concerns because damaged plant tissues release nutrients and provide a favorable matrix for bacterial proliferation. Repeated outbreaks of human salmonellosis from fresh tomatoes, lettuce, mixed salads, bean and alfalfa sprouts, cantaloupes, papayas, and orange juice, as well as recent large-scale outbreaks involving jalapeño and Serrano peppers (Table 10.5), underline the major challenge to the produce industry and government reg-

10.  Salmonella Species

237

Table 10.4  Examples of major foodborne outbreaks of human salmonellosis from dairy products No. of: Year(s)

Country

Vehicle

Serovar

Cases

a

Deaths

Reference(s)

1973

Trinidad

Milk powder

Derby

3,000b

NSc

225

1976

Australia

Raw milk

Typhimurium PT 9

>500

NS

192

1981

Scotland

Raw milk

Typhimurium PT 204

1984

Canada

Cheddar cheese

Typhimurium PT 10

1985

United States

Pasteurized milk

Typhimurium

1994

United States

Ice cream

1998

Canada

Cheddar cheese

2001

United States

Multiple cheeses, unpasteurized

Newport

2002

United States

2% Milk, pasteurized

2003

United States

2004 2006

654

2

49

2,700

0

59

16,284

7

134

Enteritidis PT 8

740

0

10, 104

Enteritidis PT 8

700

0

182

27

0

215

Typhimurium

116

0

215

Queso fresco

Typhimurium

50

0

215

United States

Pasteurized other milk

Newport

100

0

215

United States

Cheese

Dublin

4

0

215

2006–2007

United States

Unpasteurized Mexicanstyle aged cheese

Newport

85

0

214

2007

United States

Shredded cheese

Montevideo

20

0

215

2008

United States

Cheddar cheese

Java

70

0

215

2008

Spain

Infant formula

Kedougou

42

0

186

2011–2012

Russia

Powdered milk

Oranienburg

16

0

27

a b c

Confirmed cases unless stated otherwise. Estimated number of cases. NS, not specified.

ulatory agencies in the implementation and consistent application of stringent on-farm pathogen control measures. Salmonella contamination of fruits and vegetables could arise from the entry of pathogens through scar tissues, from the natural uptake of pathogens through root systems, from surface contamination of flowering plants and subsequent entrapment of the pathogen during embryogenesis of the fruit or vegetable, and from the transfer of surface contaminants onto edible plant tissues during slicing or into the juice of freshly pressed fruits (101, 148, 160). The human health risk is further increased by the propensity of salmonellae to actively grow on cut tomatoes and melons, in precut salad mixes, and on fresh produce moistened during retail display at ambient temperature (64). Worldwide reports on the presence of Salmonella spp. in sesame seed products, notably tahineh (tahini) and halawa (halva) and attendant recalls of adulterated products imported primarily from Middle-East countries in the last 2 decades (64, 231) finally culminated in confirmed international outbreaks of S. Typhimurium DT 104 and S. Montevideo

in 2001 and 2002, respectively (Table 10.5). Scant reports of human salmonellosis from the consumption of tahineh or halawa likely stem from single or few cases of illness among family members that are not reported to health agencies. The rash of bacterial disease outbreaks from the consumption of sprouted seeds in recent years (23, 161) likely resulted from the germination of internally contaminated seeds at ambient temperatures under high-moisture conditions and concurrent growth of salmonellae on the leguminous sprouts. Salmonella spp. are more likely localized under the seed coat as suggested by the difficulty in isolating salmonellae from intact seeds and the need to germinate seeds for the successful cultural isolation of the pathogen. The inability of bactericidal agents to effectively eliminate salmonellae from intact seeds further points to an internal localization of Salmonella spp. (23, 161). The ubiquity of Salmonella spp. in the natural environment, the intense husbandry practices associated with the rearing of meat animals, the provision of Salmonellacontaminated animal feeds to reared d­omestic animals,

Foodborne Pathogenic Bacteria

238

Table 10.5  Examples of major foodborne outbreaks of human salmonellosis from fruits and vegetables No. of: Year(s)

Country

Vehicle

Serovar

1981

The Netherlands

Salad base

Indiana

1984

United States

Salad bars

Typhimurium

1991

United States and Canada

Cantaloupes

Poona

1991

Germany

Fruit soup

1993

United States

1994

Finland / Sweden

1995

Reference

Casesa

Deaths

600

0

25

b

751

0

205

>400

NSc

87

Enteritidis

600

NS

89

Tomatoes

Montevideo

100

0

102

Alfalfa sprouts

Bovismorbificans

492

0

175

United States

Orange juice

Hartford Gaminara

62

0

171

1996

United States

Alfalfa sprouts

Montevideo Meleagridis

481

1

158

1999

Australia

Orange juice

Typhimurium

427

NS

21

1999

Canada

Alfalfa sprouts

Paratyphi B (Java)

>53

NS

96

1999

United States and Canada

Orange juice

Muenchen

>220

0

31

1999

United States

Tomatoes

Baildon

86

3

57

1999

United States and Brazil

Mangoes

Newport

78

2

193

2000

United States

Orange juice

Enteritidis

>74

0

40

2000

United States

Mung bean sprouts

Enteritidis

>45

0

12

2000

The Netherlands

Bean sprouts

Enteritidis PT 4b

2000

Europe (5 countries)

Lettuce

Typhimurium PT 204b

2000–2002

United States and Canada

Cantaloupe

Poona

2001

United States

Alfalfa sprouts

Kottbus

2001

Europe and Australia

Halva (halawa)

Typhimurium DT 104

2002

United States

Roma tomatoes

Javiana

159

0

206

2002–2003

Australia and New Zealand

Tahini (tahineh)

Montevideo

68

NS

213

2004

United States and Canada

Roma tomatoes

Braenderup Javiana Typhimurium Anatum Thompson Muenchen

125 390 27 5 4 4

0 0 0 0 0 0

44

2004

United Kingdom

Lettuce

Newport

>372

0

91

2005

Austria

Mixed salad

Enteritidis PT 21

85

0

186

2005

Finland

Iceberg lettuce

Typhimurium var. Copenhagen DT 104b

60

0

200

2006

United States

Tomatoes

Berta Typhimurium Newport

16 218 115

0 0 0

129, 215

2007

Australia

Papayas

Litchfield

26

0

90

2009

United States

Alfalfa sprouts

Saintpaul

235

0

217

12

0

223

>392

0

54

155

1

7

31

NS

230

>70

NS

14, 15

(Continued)

10.  Salmonella Species

239

Table 10.5  Examples of major foodborne outbreaks of human salmonellosis from fruits and vegetables (Continued) No. of: Year(s)

Country

Vehicle

Serovar

2010

United Kingdom

Bean sprouts

Bareilly

2011

Denmark, Germany, Austria

Tomatoes

Strathcona

2011

United States

Papayas

Agona

a b c

Casesa

Deaths

Reference

197

0

211

40 14 1

0 0 0

197

106

0

220

Confirmed cases unless stated otherwise. Estimated number of cases. NS, not specified.

Table 10.6  Examples of major foodborne outbreaks of human salmonellosis from meats and meat products No. of: Year(s)

Cases

Deaths

Reference(s)

1984

France and England

Country

Liver pâté

Goldcoast

756

0

33, 170

1984

International

Aspic glaze

Enteritidis PT4

866

2

38

1997

United States

Stuffed ham

Heidelberg

746

1

25

2000

United States

Hamburger buns/ infected worker

Thompson

55

0

127

2001

United States

Pork

Uganda

24

NSb

119

2003

United States

Ground beef`

Typhimurium DT 104

58

0

65

2004

United States

Ground beef

Typhimurium

>31

0

44

2004

Germany

Raw minced pork

Give

115

NS

117

2004

United States

Roast beef

Salmonella spp.

28

NS

56

2004–2005

Germany

Raw pork

Bovismorbificans PT 24

402

1

92

2005

Canada

Deli meats

Typhimurium PT U302

55

0

17

2005

The Netherlands

Imported raw beef

Typhimurium DT 104

165

0

128

2005

Honduras

Cooked chicken

Salmonella spp.

>600

NS

84

2005

Spain

Cooked chicken

Hadar

2,138

1

142

2005

United States

Cooked turkey

Enteritidis

>304

1

82

2005

Canada

Roast beef

Salmonella spp.

155

0

16

2005

England

Kebab

Enteritidis PT 1

195

NS

5

2007

United States

Banquet pot pies (chicken or turkey)

Salmonella I 4,[5], 12:i:-

272

0

218

2008

Ireland

Pre-cooked meat products

Agona

163

0

163

2010

Denmark

Salami

Typhimurium

20

0

132

2010

France

Dried pork sausage

Salmonella I 4, 12:i:-

69

0

32

2011

United States

Ground turkey

Heidelberg

136

1

216

Confirmed cases unless stated otherwise. b NS, not specified. a

Vehicle

Serovar

a

Foodborne Pathogenic Bacteria

240

Table 10.7  Examples of major foodborne outbreaks of human salmonellosis from fish and fish products No. of: Year

Country

Vehicle

Serovar

1988

Japan

Cuttlefish

Champaign

1998

Germany

Smoked eel

Blockey

1999

Japan

Dried squid

Salmonella spp.

1999

Japan

Cuttlefish chips

Chester Oranienburg

2000

United States

Salmon, seafood dish

2001

Norway and Sweden

2001 2003

Deaths

Reference

330

0

166 76

Casesa 28

0

>453

0

³1,500

NS

208

Enteritidis

36

1

215

Fish

Livingstone

60

3

99

United States

Crab cake, lobster

Enteritidis

14

0

215

United States

Raw oysters

Typhi

6

0

215

2004

United States

Shrimp cocktail

Enteritidis

18

0

215

2004

United States

Tuna, unspecified

Weltevreden

63

NS

215

2006

United States

Scallop, sea cucumber

Agona, Heidelberg

32

0

215

2007

United States

Fish, ahi

Paratyphi B

44

0

215

2008

United States

Hybrid striped bass

Barranquilla, Bovismorbificans, Barenderup, Javiana, Rubislaw, Thompson

45

0

215

2009

United States

Red snapper

Salmonella spp.

7

0

215

11 b

Confirmed cases unless stated otherwise. b NS, not specified. a

and the propensity for cross-contamination of animals in feed lots, in holding pens, and during animal slaughter and processing of carcasses are factors that could contribute to the presence of salmonellae in raw meat products (64) (Table 10.6). In 1984, liver paté manufactured in France was incriminated as the cause of no fewer than 506 and 250 cases of infections with S. Goldcoast in France and in England and Wales, respectively. Localization of the pathogen within the external gelatin layer of the paté suggested that product adulteration resulted from product handling by plant workers (33). In the same year, gelatin glaze was incriminated in another international outbreak of 866 cases of S. Enteritidis PT4 among airline passengers, flight crews, ground personnel, and food catering staff who had consumed glazed canapés and appetizers prepared for a major European airline. The incident was tentatively ascribed to cross-contamination of the aspic glaze by infected food handlers in the flight kitchen. Fortunately, no serious airline mishaps resulted from this episode other than flight delays and the grounding of a Concorde aircraft in Washington, DC, because the entire crew was incapacitated by salmonellosis and a replacement crew was not readily available. In 1997, numerous participants at a church fundraising

dinner were infected with S. Heidelberg when undercooking and slow cooling of stuffed ham and the use of an unsanitized mechanical meat slicer contributed to the magnitude of the outbreak. A large protracted outbreak of S. Bovismorbificans PT24 in Germany was putatively ascribed to the consumption of raw minced pork and fermented raw pork sausages. In 2005, undercooked chicken was incriminated as the vehicle of infection of 600 participants in a political campaign in Honduras. In the same year, more than 2,000 human cases of S. Hadar PT2 infections resulted from the consumption of a single nationally distributed brand of roasted, vacuumpackaged chicken in Spain; the incident resulted in the hospitalization of more than 200 infected cases and one fatality. In 2005, undercooked turkey served to restaurant patrons resulted in more than 300 cases of S. Enteritidis infections in the United States. In May 2010, a nationwide outbreak of salmonellosis with the specific monophasic variant Salmonella enterica serotype 4,12:i:- associated with a dried pork sausage was reported in France. This serotype has been identified in a variety of foodstuffs, but most frequently in pork delicatessen. In 2011, a multistate outbreak associated with the consumption of ground turkey contaminated

10.  Salmonella Species

241

Table 10.8  Examples of major foodborne outbreaks of human salmonellosis from eggs and egg products No. of Year

Country

Vehicle

Serovar

Casesa

Deaths

Reference

3,400

0

111

1974

United States

Potato salad

Newport

1976

Spain

Egg salad

Typhimurium

702

6

9

1977

Sweden

Mustard dressing

Enteritidis PT 4

2,865

0

103

1987

China (P.R.)

Egg drink

Typhimurium

1,113

NS

232

1988

Japan

Cooked eggs

Salmonella spp.

10,476

NS

157

1993

France

Mayonnaise

Enteritidis

751

0

184

2001

United States

Tuna salad with eggs

Enteritidis

688

0

71

2001

Latvia

Cake/raw egg sauce

Enteritidis PT 4

19

0

116

2002

Spain

Custard-filled pastry

Enteritidis PT 6

1435

0

41

2002

England

Bakery products

Enteritidis PT 14b

>150

1

165

2003

England/ Wales and Scotland

Egg sandwiches

Bareilly

186

NS

52

2003

Australia

Raw egg mayonnaise

Salmonella spp.

>106

1

83

2003

United States

Egg salad kit

Typhimurium

0

124

2004

China (P.R.)

Cake/raw egg topping

Enteritidis

NS

149

2005

England

Imported shell eggs

Enteritidis PT 6

68

0

50

2006

United States

Raw egg mayonnaise

Heidelberg

22

0

215

2007

United States

Eggs, scrambled

Enteritidis

81

0

215

2008

United States

Eggnog

Enteritidis

18

0

215

2009

Australia

Raw egg mayonnaise

Typhimurium

71

0

20

2009

United States

Eggs, scrambled

Enteritidis

59

0

215

2009

United Kingdom

Eggs and poultry

Enteritidis PT 14b

2010

United States

Shell eggs

Enteritidis

2010

Ireland

Duck eggs

Typhimurium DT8

a b

b

18 197

136 1,939 24

2

212

NS

215

0

81

Confirmed cases unless stated otherwise. Estimated number of cases.

with a multidrug-resistant S. Heidelberg was reported in the United States. A total of 136 persons were infected with the outbreak strain, and one death was reported. Major outbreaks of salmonellosis associated with fish and shellfish occurred mainly in Japan, where marine species are a food staple (Table 10.7). The consumption of cuttlefish, a cephalopod mollusk, that had been left to thaw at room temperature for up to 30 hours and then was boiled for only a short time was incriminated in 330 infections of S. Champaign in children. In 1999, more than 400 cases of salmonellosis in Japan were associated with dried squid from a processing plant whose supply of well water was contaminated. In the same year, cuttlefish chip snacks contaminated with S. Chester and S. Oranienburg were implicated in an outbreak involving

more than 1,500 human cases of illness in Japan. An investigation of an unusual outbreak, in Germany in 1998, of salmonellosis caused by S. Blockley in smoked eel indicated that hot smoke processing may be insufficient to kill the Salmonella cells on contaminated fish. Consumption of fish gratin manufactured in Sweden and marketed as a frozen product in Norway and Sweden resulted in 60 cases of S. Livingstone infections and three fatalities in medically challenged elderly persons. Indirect evidence suggested that the egg powder ingredient for the fish gratin was the source of contamination. Although seafood is not a common vehicle for salmonellosis, outbreaks associated with Salmonellacontaminated seafood were frequently reported in the United States (Table 10.7). The Salmonella outbreaks associated with fish and other seafood products could

Foodborne Pathogenic Bacteria

242

Table 10.9  Examples of major foodborne outbreaks of human salmonellosis from other products No. of: Year(s)

Country

Vehicle

Serovar

Casesa

Deaths

Reference

1973

Canada and United States

Chocolate

Eastbourne

217

0

53

1987

Norway and Finland

Chocolate

Typhimurium

361

0

121

1993

Germany

Paprika chips

Saintpaul Javiana Rubislaw

>670

0

136

1998

United States

Toasted oat cereal

Agona

209

0

215

1999

Japan

Peanut sauce

Enteritidis PT 1

2000–2001

Germany and International

Chocolate

Oranienburg

2000–2001

Canada and United States

Raw almonds

2001

Romania

2001

644

NS

118

>439

NSb

226

Enteritidis PT 30

168

0

114

Pastries

Salmonella spp.

>250

0

85

Canada and Australia

Shandong peanuts

Stanley

93

NS

13

2003

Germany

Aniseed herbal tea

Agona

42

0

130

2003–2004

United States and Canada

Raw almonds

Enteritidis

29

0

123

2006

United Kingdom

Chocolates

Montevideo

37

0

210

2006

United States

Peanut butter

Tennessee

715

0

215

2006

United States

Peanuts

Thompson

100

0

215

2008–2009

United States

Peanut butter, peanut paste

Typhimurium

714

9

215

2008

United States

Cereal, puffed rice and wheat

Agona

35

0

215

2008

United States

Ground white pepper

Rissen

87

NS

215

2010

United States

Black and red pepper

Montevideo

272

NS

215

2011

United States

Turkish pine nuts

Enteritidis

43

NS

215

Confirmed cases unless stated otherwise. b NS, not specified. a

be from cross-contamination during farming, processing, preparation, and transportation (70, 201). The ongoing pandemic of eggborne S. Enteritidis continues to impact the incidence of foodborne outbreaks worldwide (Table 10.8). The large Swedish outbreak of S. Enteritidis PT4 in 1977 was attributed to the consumption of a mayonnaise dressing prepared in a central kitchen and distributed to school cafeterias in Stockholm. Although no fatalities were associated with this episode, 50 of the 80 cases requiring hospitalization suffered from reactive arthritis sequelae. In

1993, school cafeterias in Douai, France, were implicated in a large outbreak of S. Enteritidis infections among 751 school children, teachers, and support staff who had consumed tuna salad prepared with raw egg mayonnaise. In 2001, consumption of tuna salad and hard-boiled eggs contaminated with S. Enteritidis PT2, 13a, and 23 infected 688 inmates in four prison facilities in South Carolina. In the following year, pastries filled with vanilla cream were incriminated as the vehicle of 1,435 cases of S. Enteritidis PT6 infections in the province of Gerona, Spain. Cross-contamination

10.  Salmonella Species of the supply bakery by fresh shell eggs was identified as the probable cause of the outbreak. In 2002, the use of fresh shell eggs in the preparation of uncooked bakery products in Cheshire, England, led to more than 150 cases of infection with S. Enteritidis PT14b (165). In 2009, an outbreak in the United Kingdom involving 136 cases and 2 deaths was the result of a nalidixic acid-resistant strain of S. Enteritidis PT14b. This outbreak was linked to the consumption of egg-containing food (211). In 2004, a cake pastry topped with a mixture containing raw eggs resulted in 197 human cases of S. Enteritidis infections in the People’s Republic of China. In 2005, several catering services in the northeast of England were associated with a total of 68 cases of S. Enteritidis PT6 infections. Shell eggs imported from The Netherlands were suspected as the source of infection. Other serovars have been involved in eggborne outbreaks. In 1974, temperature abuse of eggcontaining potato salad served at an outdoor barbecue led to an estimated 3,400 human cases of S. Newport infection, where cross-contamination of the salad by an infected food handler was suspected. In 1987–1988, large outbreaks of S. Typhimurium and of Salmonella spp. linked to an egg-based drink and to a cooked egg dish were reported in China and in Japan, respectively. Prepackaged egg sandwiches were identified as the vehicle of S. Bareilly infection in 186 consumers from England, Wales, and Scotland. Meat rolls prepared with raw eggs were implicated in a restaurant outbreak of more than 100 cases of salmonellosis in Melbourne, Australia. More recently, a nationwide shell egg outbreak linked to S. Enteritidis infections caused 1,939 illnesses from May to November 2010 in the United States. This outbreak showed that S. Enteritidis is still an important cause of human illness in the United States. In 2010, an outbreak of S. Typhimurium DT8 linked to the consumption of duck eggs was reported as the largest food poisoning outbreak of salmonellosis in recent years in Ireland. Good hygiene and thorough cooking of eggs are crucial to reducing the numbers of egg-related salmonellosis outbreaks. Chocolates, pastries, spices, nuts, and cereals have all been involved in salmonellosis outbreaks (Table 10.9). In 1973, milk chocolate manufactured in Canada was implicated in more than 200 cases of S. Eastbourne infections in Canada and the United States. Dry roasting of imported cocoa beans and environmental contamination of the manufacturing plant with dust from the raw cocoa bean storage rooms contributed to the adulteration of finished products. Additional episodes of human salmonellosis involving chocolate products are noteworthy. In 1982, 245 cases of S. Napoli infec-

243 tions resulted from the consumption of chocolate bars imported from Italy. In 1987, Norway experienced an outbreak of 349 cases of S. Typhimurium infections associated with a variety of domestic chocolate products manufactured by a single plant in Trondheim. Chocolate bars manufactured by this company and exported to Finland were also incriminated as the cause of an additional 12 cases of salmonellosis. In 2000 and 2001, chocolate products from a large firm in Germany were identified as the vehicle of 439 cases of S. Oranienburg infections in Germany and of additional cases in several European countries and probably Canada. In 1993, paprika imported from South America was incriminated as the contaminated ingredient used in the manufacture of potato chips distributed in Germany. In 2006, chocolates contaminated with Salmonella sickened 37 people in the United Kingdom. In 2001, more than 250 cases of salmonellosis, including 194 children in northern Romania, were infected following the consumption of Salmonella-tainted pastries. In 2008, an outbreak of S. Rissen infections, associated with ground white pepper, sickened 87 victims in five states. S. Rissen is a rare strain in the United States, but it is fairly common in Southeast Asia. In 2010, a nationwide outbreak of S. Montevideo sickened 272 people in the United States. The vector was salami that contained contaminated imported black and red pepper. More recently, epidemiological and environmental investigations indicated that cross-contamination in the manufacturing plants probably accounted for the repeated outbreaks of lowmoisture foods such as chocolate, peanut butter, toasted cereal, and nuts (Table 10.9).

CHARACTERISTICS OF DISEASE

Symptoms and Treatment

Human Salmonella infections can lead to several clinical conditions, including enteric (typhoid) fever, uncomplicated enterocolitis, and systemic infections by non­typhoid microorganisms. Enteric fever is a serious human disease associated with S. Typhi and S. Paratyphi, which are mainly transmitted from human to human via the fecal-oral route and are particularly well adapted for invasion and survival within host tissues. Clinical manifestations of enteric fever appear after a period of incubation ranging from 7 to 28 days and may include diarrhea, prolonged and spiking fever, abdominal pain, headache, and prostration (61). Diagnosis of the disease relies on the isolation of the infective agent from blood or urine samples in the early stages of the disease or from stools after the onset of clinical symptoms. An asymp­tomatic chronic carrier state commonly follows

244 the acute phase of the disease. The treatment of enteric fever is based on supportive therapy and/or the administration of antibiotics. In the United States, ciprofloxacin, ceftriaxone, or cefotaxime is prescribed for nonpregnant adults. Ceftriaxone, an injectable antibiotic, is an alternative for women who are pregnant and for children who may not be candidates for ciprofloxacin. In developing countries, amoxicillin, ciprofloxacin, chloramphenicol, ampicillin, or trimethoprim-sulfamethoxazole is commonly employed. Marked global increases in the resistance of typhoid and paratyphoid microorganisms to these antibacterial drugs have greatly undermined their efficacy in human therapy. The problem is particularly serious in developing countries, where m­ultipleantibiotic-resistan­t salmonellae are frequently implicated in outbreaks of enteric fever and are the cause of unusually high fatality rates. The widespread use of fluoroquinolones in these countries is hampered by the high cost of these drugs. In response to the current era of antibiotic-resistant salmonellae, there is a growing interest worldwide in the use of phage therapy for the clinical management of severe human infections with multipleantibiotic-resistant bacterial pathogens and for veterinary purposes (152). Nontyphoid Salmonella is mainly transmitted through food (or water) vehicles and commonly results in enterocolitis, which appears 8 to 72 h after contact with the invasive pathogen. The clinical condition is generally self-limiting, and remission of the characteristic nonbloody diarrheal stools and abdominal pain usually occurs within 5 days of onset of symptoms. The successful treatment of uncomplicated cases of enterocolitis may require only supportive therapy such as fluid and electrolyte replacement. The use of antibiotics in such episodes is contraindicated because it tends to prolong the carrier state and the intermittent excretion of salmonella­e (60). This asymptomatic persistence of salmonellae in the gut likely results from a marked antibiotic-dependen­t repression of native gut microflora members that normally compete with Salmonella for nutrients and intestinal binding sites. Human infections with nontyphoid strains can also degenerate into systemic infections and precipitate various chronic conditions. In addition to S. Dublin and S. Choleraesuis, which exhibit a predilection towards septicemia, similarly high levels of virulence have been observed with other nontyphoid strains. Preexisting physiological, anatomical, and immuno­ logical disorders in human hosts can also favor severe and protracted illness through the inability of host defense mechanisms to respond effectively to the presence of invasive salmonellae (60). More frequent reports, in recent years, on the chronic and debilitating sequelae

Foodborne Pathogenic Bacteria of nontyphoid systemic infections are of concern because this emerging medical pattern may be linked to increased levels of virulence among nontyphoid salmonellae, increased susceptibility of human populations to chronic bacterial diseases, or synergy between the two factors (47). Salmonella-induced chronic conditions such as aseptic reactive arthritis, Reiter’s syndrome, and ankylosing spondylitis are noteworthy. Bacterial prerequisites for the onset of these chronic diseases include the ability of the bacterial strain to infect mucosal surfaces, the presence of outer membrane LPS, and a propensity to invade host cells (36, 194). Recent evidence suggests that these arthropathies may be linked to a genetic predisposition in individuals that carry the class I HLA-B27 histocompatibility antigen. Other antigens encoded by the HLA-B locus such as the B7, B22, B40, and B60 antigens, which serologically cross-react with B27 antiserum, have also been associated with reactive arthritis (36, 153, 202). Therapeutic eradication of a human Salmonella infection in the intestinal tract does not preclude the subsequent onset of chronic rheumatoid diseases in a distal, noninfected limb (202).

Antibiotic Resistance

Antibiotic resistance in Salmonella spp. has been reported since the early 1960s, when most of the reported resistance was to a single antibiotic (35). However, in the mid-1970s MDR Salmonella emerged. In 1979, a clone of S. Typhimurium DT204 with chromosomally encoded and integron-mediated resistance to chloramphenicol, streptomycin, sulfonamides, and tetracycline moved through the food chain from calves to humans (188). Other MDR clones were subsequently identified; some of these clones appeared to be geographically localized, whereas others, such as S. Typhimurium DT204 and DT193, were encountered in several countries (188). The appearance of S. Typhimurium DT 104 in the late 1980s raised major concerns because of its common pentaresistance (ACSSuT) to ampicillin (A), chloramphenicol (C), streptomycin (S), sulfonamides (Su), and tetracycline (T). Resistances of this phagovar to additional antibiotics, such as trimethoprim, nalidixic acid, and spectinomycin, coupled with a reduced sensitivity to ciprofloxacin, have been reported (203, 204). While MDR S. Typhimurium DT 104 continues to cause serious foodborne outbreaks worldwide, a new MDR S. enterica serovar, S. Newport MDR-AmpC, has emerged; it is resistant to ACSSuT and amoxicillinclavulanic acid, cephalothin, cefoxitin, and ceftiofur and is less susceptible to ceftriaxone (1). Its resistance to the expanded-­spectrum cephalosporins is alarming,­

10.  Salmonella Species since cephalosporins are currently prescribed to treat Salmonella infections in children. In 2009, the top three most frequently isolated Salmonella serovars from human infections were Enteritidis, Typhimurium, and Newport (162), and the nontyphoid Salmonella serovars with the highest ratios of isolates resistant to ACSSuTAuCx were Newport and Typhimurium. The rate of non-Typhi Salmonella isolates from humans resistant to at least one antibiotic has been found to be decreasing, from 33.8% in 1996 to 22.5% in 2003 and 16.8% in 2009 (162). To examine susceptibility trends among foodborne pathogens and commensal bacteria isolated from both food animals and humans, the National Antimicrobial Resistance Monitoring System began Salmonella surveillance by sampling chicken, turkey, cattle, and swine carcasses in 1997, and in 2002 sampling expanded to retail meats, including chicken breast, ground turkey, ground beef, and pork chops. Antibiotic resistances of Salmonella isolates from ground turkey and turkey samples have been rising from 60% (turkey) and 70% (ground turkey) in 2002 to approximately 80% for both in 2009 (162). The high rate of antibiotic resistance of isolates from turkey and turkey products invokes the need to improve our current practices in turkey farms and processing plants. Ground beef, cattle, and chicken samples have shown the least proportion of antibiotic resistance among isolates from all meat and food animal groups since 2002; however, the ratio of Salmonella isolates from the processed chicken breast resistant to at least one antibiotic has been at least 10% higher than that from chicken carcass during the same period. The difference between the prevalences of antibioticresistant Salmonella in chicken breast and chicken carcass may reflect the high portion of antibiotic-resistant salmonellae in viscera and the cross-contamination associated with current poultry processing practices; thus, preventive mitigation and modification of processing strategies are very important. Four major mechanisms are involved in antibiotic resistance in Salmonella: enzymatic degradation (b-lactams), chemical modification (aminoglycoside), drug-binding site gene mutation (quinolones/fluoroquinolones), and activation of the efflux pump (phenol/tetracycline). The b-lactam antibiotics are able to interfere with a group of penicillin-binding proteins to disrupt the bacterial cell wall formation, including four major groups: penicillins, cephalosporins, monobactams, and carbapenems. The resistance against b-lactams in Salmonella is conferred by b-lactamase enzymes to cleave the b-lactam ring structure (39). Aminoglycosides, such as kanamycin, gentamicin, and streptomycin, are able to

245 bind the 16s rRNA to inhibit or mislead the translation process in bacteria. The resistances to aminoglycosides are carried out by modifying the aminoglycoside structure via phosphorylation, acetylation, or adenylation. Quinolones and fluoroquinolones are a group of drugs targeting the DNA gyrase alone, or both the gyrase and other topoisomerases. The resistance of quinolones in Salmonella is primarily associated with the mutation of two genes respectively encoding the subunits of gyrase and the chromosome replication enzyme topoisomerase. The reduced quinolone sensitivity in Salmonella is also conferred by another mechanism, the overexpressed efflux system. Tetracycline and chloramphenicol are drugs targeting the protein synthesis steps, the former inhibiting the tRNA binding to the A site of 30S ribosome subunits and the latter preventing the extension of peptides. Resistance toward tetracycline and chloramphenicol is attributed mainly to the activity of efflux pumps to expel the drug out of the bacterial cells, along with other enzymatic inactivation and target modification mechanisms (1). These antibiotic resistance genes often reside on mobile genetic elements, i.e., plasmids, transposons, and integrons, that can potentially ferry the resistance from commensal to pathogen (125). One genetic element in particular, the integron, appears to be especially culpable in the development and dissemination of MDR for many microorganisms, including Salmonella. An integron is a genetic element capable of capturing, combining, or swapping a large assortment of antibiotic resistance genes through recognition of a 59-bp element present in “selected” genes by its recombinase, IntI, and integrating the captured gene into its resident integration site, attI. This g­enetic element can create a tandem of antibiotic resistance genes (198). There are at least 8 classes of integrons in nature, with the two newest classes coming from analyses of environmental DNA (164). Of the 8 classes of integrons, class 1 integrons are the most studied and prevalent of the integron classes among human and veterinary pathogens. Class 1 integrons have been reported in various S. enterica serovars isolated from humans, animals, and food, and they possess a vast assortment of drug resistance genes (181, 228). The resistance genes associated with class 1 integrons confer resistances to a diverse array of antibiotics and disinfectants, including aminoglycosides, b-lactams, chloramphenicol, macrolides, quaternary ammonium, and trimethoprim (78). Tetracycline resistance genes are the only drug resistance genes that have not been identified among the myriad of class 1 integron cassettes.

Foodborne Pathogenic Bacteria

246

Infectious Dose

It is well established that newborns, infants, the elderly, and immunocompromised individuals are more susceptible to Salmonella infections than healthy adults (60). The incompletely developed immune system in newborns and in infants, the frequently weak and/or delayed immunological responses in the elderly and debilitated persons, and the generally low gastric acid production in infants and seniors facilitate the intestinal colonization and systemic spread of salmonellae in these segments of the population (30, 61). Antibiotic treatment of subjects before their encounter with Salmonella enhances bacterial virulence through an antibiotic-mediated clearance of native gut microflora, which reduces the level of bacterial competition for nutrients and attachment sites in the intestinal tract of the host (110, 195). Detailed investigations of foodborne outbreaks have indicated that the ingestion of only a few Salmonella cells can be infectious (Table 10.10). Despite early reports that large numbers of salmonellae inoculated into eggnog and fed to human volunteers produced overt disease (153), more recent evidence suggests that 1 to 10 cells can constitute a human infectious dose (59, 118). Determinant factors in salmonellosis are not limited to the immunological hetero-

Table 10.10  Human infectious doses of Salmonellaa Food

Salmonella serovar

Infectious dose (CFU)

Eggnog

Meleagridis Anatum

104–107 105–107

Goat cheese

Zanzibar

105–1011

Carmine dye

Cubana

104

Imitation ice cream

Typhimurium

104

Chocolate

Eastbourne

102

Hamburger

Newport

101–102

Cheddar cheese

Heidelberg

102

Chocolate

Napoli

101–102

Cheddar cheese

Typhimurium

100–101

Chocolate

Typhimurium

£101

Paprika potato chips

Saintpaul Javiana Rubislaw

£4.5 × 101

Alfalfa sprouts

Newport

£4.6 × 102

Ice cream

Enteritidis

£2.8 × 101

a

Adapted from reference 60.

geneity within human populations and to the virulence of infecting strains but may also include the chemical composition of incriminated food vehicles. A common denominator of the foods associated with low infectious doses (Table 10.10) is a high fat content, as seen in chocolate (cocoa butter), cheese (milk fat), and meat (animal fat). This suggests that entrapment of salmonellae within hydrophobic lipid micelles would afford protection against the bactericidal action of gastric acidity. Following a bile-mediated dispersion of the lipid moieties in the duodenum, the viable salmonellae would resume their infectious course in search of suitable points of attachment in the lower portion of the small intestine (colonization). The rapid emptying of gastric contents could also provide an alternate mechanism for the successful infection of susceptible hosts. The swift passage of a liquid bolus through an empty stomach would minimize bacterial exposure to gastric acidity and sustain the migration of viable salmonellae in the intestinal tract (30). Publications on the dynamics of human Salmonella infections are of singular interest (93, 94). An in-depth epidemiologic study of a large outbreak of S. Typhimurium involving chicken served to delegates at a medical conference revealed that the clinical course in patients was directly related to the number of ingested salmonellae (94). The incubation period for the onset of symptoms was inversely related to the infectious dose. Patients with short (£22-h) periods of incubation suffered more frequent diarrheal bowel movements, higher maximum body temperatures, greater persistence of clinical symptoms, and greater frequency of hospitalization. Interestingly, no association between the age of infected individuals and the length of the incubation period was noted. Similar findings were reported in retrospective dose-response studies of foodborne salmonellosis (94, 156). The compelling evidence that ingestion of few Salmo­ nella cells can develop into a variety of clinical conditions and deteriorate into septicemia and even death underlines the unpredictable pathogenicity of this large and heterogeneous group of human bacterial pathogens. Food producers, processors, and distributors should recognize that low levels of salmonellae in a finished food product could lead to serious public health consequences (60).

PATHOGENICITY AND VIRULENCE FACTORS

Specific and Nonspecific Human Responses

The presence of viable salmonellae in the human intestinal tract confirms the successful evasion of ingested bacteria from nonspecific host defenses. Antibacterial lactoperoxidase in saliva, gastric acidity, mucoid secretions from intestinal goblet cells, intestinal peristalsis, and sloughing

10.  Salmonella Species of luminal epithelial cells synergistically oppose bacterial colonization of the intestinal mucosa. In addition to these constitutive hurdles to bacterial infection, the antibacterial action of nonspecific phagocytic cells (neutrophils, macrophages, and monocytes) coupled with the immune responses associated with specific T and B lymphocytes, the epitheliolymphoid tissues (Peyer’s patches), and the classical or alternative pathways for complement inactivation of invasive pathogens mount a formidable defense against the systemic spread of Salmonella. The human diarrheagenic response to foodborne salmonellosis results from the migration of the pathogen in the oral cavity to intestinal tissues and mesenteric lymph follicles (enterocolitis). The event coincides with extensive leukocyte influx into the infected tissues, increased mucus secretion by goblet cells, and mucosal inflammation triggered by the leukocytic release of prostaglandins. The last occurrence also activates the adenyl cyclase in intestinal epithelial cells, resulting in increased fluid secretion into the intestinal lumen (77, 174). The failure of host defense systems to hold the invasive Salmonella in check can degenerate into septicemia and other chronic clinical conditions.

Salmonella Pathogenicity Islands and Type III Secretion System

Genomic sequencing has provided a global view of the genetics of Salmonella virulence. The complete and annotated sequence of the S. Typhimurium genome was first published in 2001 (154), and since that time, 1 Salmonella bongori and 20 S. enterica serovar genomes have been completely annotated and published for Salmonella serovars (http://www.ncbi.nlm.nih.gov/genomes/lproks.cgi, 2011). In genome comparisons between S. Typhimurium and E. coli K-12, ~25% of the 4,552 genes of Salmonella are absent from the E. coli genome, and 2 to 8% of the S. Typhimurium LT2 complement of 4,500 genes are absent in one or more S. enterica serovars. Further genomic comparisons between E. coli and Salmonella reveal that many of these “Salmonella-specifi­c” genes are clustered together at focal points or loci within the bacterial chromosome into “islands” (Fig. 10.1 and 10.2). These genomic islands are called Salmonella pathogenicity islands (Fig. 10.1). They vary in size, generally map next to tRNA genes, often have remnants of phage genes at their borders, and usually have a GC content lower than the cognizant GC content for most metabolic and housekeeping genes, suggesting horizontal gene transfer occurred (190). Many of these genomic islands in Salmonella encode important functions that are essential to the pathogen’s virulence, and these loci define this genus and species. While the five genomic islands, SPI-1 through SPI-5, are conserved in S. enterica, there are genetic variabili-

247 ties within several of these loci among the serovars (4) (Fig. 10.2). Recent completion of the S. Typhi genome has identified five additional pathogenicity islands, SPI-6 through SPI-10 (Fig. 10.2), which appear to be unique to S. Typhi and other serovars associated with enteric fever in humans (177). Type III secretion systems (T3SS) are the major virulence devices encoded by genes on SPIs. T3SS encoded by SPI-1 (inv) and SPI-2 in S. Typhimurium are involved in causing host gastroenteritis via invasion/attachment and macrophage survival, respectively. The driving force of Salmonella pathogenesis is the delivery of over 30 effector proteins into the host cell via SPI-1 and SPI-2 T3SSs. The invasion-associated T3SS apparatus structurally resembles the flagellar basal body of gram-negative bacteria and forms a secretion channel spanning the inner and outer membranes responsible for the escort and insertion of Salmonella invasion proteins, such as SipA, SipB, SipC, SptP, SopE2, and SopB, into the epithelial cell (112) (Fig. 10.3). The T3SS encoded by Salmonella enterica SPI-2 transports effector proteins that interfere with the intracellular transport process and manipulate the formation of host cell actin and microtubule cytoskeleton. In addition, the SPI-2 T3SS promotes bacterial cell survival and replication and represses the host’s innate and adaptive immune responses. A third pathogenicity island, SPI-3, is required for intracellular growth of Salmonella within a macrophage (28). The genes mgtCB, present in SPI3, are required for growth under magnesium-limiting conditions and are essential for intracellular growth (28). The other genes identified in this locus do not contribute to Salmonella pathogenesis in a mouse typhoid model. Their functions are currently unknown (29).

Virulence Plasmids

The Salmonella virulence plasmids are 50 to 90 kb in size and occur with a frequency of one to two copies per chromosome. The presence of virulence plasmids within the Salmonella genus is limited in certain strains of S. enterica subspecies I and has been confirmed in S. Typhimurium, S. Dublin, S. Gallinarum-Pullorum, S. Enteritidis, S. Cholerae-suis, and S. Abortusovis but not in the highly infectious S. Typhi. All Salmonella virulence plasmids share a 7.8-kb region, spv, which encodes products for the prolific growth of salmonellae in host reticuloendothelial tissues (187). Transcription of spv is induced by the hostile environment within host phagocytes, iron limitation, elevated temperatures, low pH, and nutrient deprivation associated with the stationary phase of growth under the regulation of sigma (sS) factor (rpoS). The spv regulon contains at least two

Foodborne Pathogenic Bacteria

248

f64 sinR

SGI1

phoN ssb

thdF cspA

100 SPI 6 SPI 10 S 94SPI 7 92 SPI 4 fim

82 SPI 3 l 79 lpf rfa 4 Mb

argU 15

23

Chromosome Chro

pncB

FELS-1 GIFSY-2

SPI 5 25

putA,P 67.2 SPI 8 mutY

63.9 40 SPI 1 SPI 2 SPI 9 mutS ttr 62 2 44 57 45.5 rfb recA his valW(tRNA) FEL-2

cps lepA cp

Virulence Plasmid

pef rsk spv

GIFSY-1

Fimbrialoperon Prophage PAI

S. Typhi-specific prophage64 S. Typhi-specific PAI

SGI1: DT104 MDR locus

Figure 10.1  Genetic map of Salmonella enterica virulence loci. Genomic islands are depicted on the chromosome relative to their position to housekeeping genes or specific tRNA. All 10 SPIs including SPI-6 through -10, which are only present in Salmonella serovars associated with enteric fever, are illustrated as circles with S. Typhi genomic islands attached to the map via arrows with dotted lines. Prophage genomes are also included in the map and are designated by “lollipop” symbols. Squares denote adhesions/fimbrial operons. More-detailed genetic organization of SPI1-5 is given in Figure 10.2. doi:10.1128/9781555818463.ch10f1

operons, one for the positive regulator spvR locus and another for spvABCD. The nucleotide sequences of the spv loci are highly conserved within the Salmonella genus (199). Mutations in spvR severely affect the bacterium’s ability to proliferate in intestinal tissues and at extraintestinal sites (144). Minor differences in the nucleotide sequence of the spvR locus in S. Dublin and S. Typhimurium markedly alter the capacity of the spvR gene to induce the spvA promoter and transcription

of the spvABCD operon. These findings provide some insight into the molecular basis for the comparatively greater virulence of S. Dublin in humans (199). SpvB functions as an ADP-ribosyltransferase (169) that ADPribosylates F-actin and blocks its polymerization into filaments. Mutations in this region strongly attenuate or inactivate the ability of salmonellae to establish deepseated infections. The Salmonella virulence plasmid also contains a fimbrial operon (pef) and a conjugal transfer

10.  Salmonella Species

249

SPI 1

avrspr orgprg orf hil iagspt iac

sit

sip

sic

inv

spa

A BCD A BA A KILIH X A B P PA D C BA A S R QP O N MC B A E G F H

SPI 2 ssa

sse

U T SRQPO N

ssc sse

sscsse

ssa

V M L K J I H G F BE D C A B A E D C B

ssr

spi

A B CA B

SPI 3 selC sugR rhuM rmbA

misL

marT slsA cigR

fidL

mgtB

mgtC orf307

SPI 4 A

B C DE F

G H I

SPI 5

J

cop R

S

Virulence Plasmid spv R

A

B

C D

Regulation

K

LM N

pip

sop

D

B

O

P Q R

pip C

B

A

pef B A

rck C

D

I

Type III Secretion System

Pilus assembly

Virulence factor

Type III Secretion

Chaperone

Autotransporter

Chaperone

Usher

Ion Transporter

Invasion Appendage

Pilin subunit

Type I Secretion System ABC Transporter Membrane Fusion Protein Outermembrane Protein

Outer Membrane Secretion Channel Type III Effector

Figure 10.2  Genetic organization of virulence genes present in SPI-1 through -5 and virulence plasmid. The genes and their organization into operons are shown with arrows demarcating operons and direction of transcription. The known functions of genes are depicted in this illustration with shading or pattern designating its function as described in the key below the schemas. Those genes with no known function are shown with white foreground. Type III effectors are bacterial proteins injected to host cell cytoplasm via T3SS. doi:10.1128/9781555818463.ch10f2

250

Foodborne Pathogenic Bacteria In addition to enterotoxin, Salmonella strains generally elaborate a thermolabile cytotoxic protein that is localized in the bacterial outer membrane (18, 61). The cytotoxin is not inactivated with antisera raised against Shiga toxin or E. coli Shiga toxins 1 and 2 as determined in green monkey kidney (Vero) cell and HeLa cell bioassays (18, 68). Hostile environments such as acidic pH and elevated (42°C) temperature induce an extracellular release of toxin, possibly as a result of induced bacterial lysis (68). The virulence attribute of cytotoxin stems from its inhibition of protein synthesis and lysis of host cells, thereby promoting the dissemination of viable salmonellae into host tissues (61, 131). It is not clear whether this cytotoxin may actually be one of the Salmonella invasion proteins reported to activate apoptosis in epithelial and phagocytic cell types (185).

Other Factors Contribute to Virulence

Figure 10.3  Complex needle and base structure of the T3SS in Salmonella. (A) Transmission electron microscope image of the T3SS needle complex in S. Typhimurium. (B) A cut-away view and description of individual substructures of the needle complex. OR, outer ring; IR, inner ring. Bar = 10 nm. (Adapted from reference 191.) doi:10.1128/9781555818463.ch10f3

gene, traT. Pef encodes adhesin involved in colonization of the small intestine and enteropathogenicity of the bacterium, and the presence of a complete traT operon supports the fact that spv plasmids are found to be selftransmissible in certain strains (187).

Toxin

The diarrheagenic enterotoxin produced in Salmonella is structurally and immunologically related to cholera toxin and the heat-labile enterotoxin in E. coli (LT-I). The release of toxin into the cytoplasm of infected host cells precipitates an activation of adenyl cyclase localized in the epithelial cell membrane and a marked increase in the cytoplasmic concentration of cyclic AMP in host cells. The concurrent fluid exsorption into the intestinal lumen results from a net secretion of Cl– ions in the crypt regions of the intestinal mucosa and depressed Na+ absorption at the level of the intestinal villi (61). Enterotoxigenicity is a virulence phenotype that prevails in Salmonella serovars, including S. Typhi, and is expressed within hours following bacterial contact with the targeted host cells (60).

The virulence of Salmonella spp. as reflected in the ability to cause acute and chronic diseases in humans and in a variety of animal hosts stems from bacterial structural and physiological attributes that act synergistically or independently in promoting bacterial adhesion and invasiveness (61). In a broad sense, bacterial surface structures, such as capsular layer and outer membrane, and cellular responses to adverse environmental conditions, such as iron acquisition, concurrently potentiate greater bacterial survival and prolific growth, which directly or indirectly contribute to virulence in hosts. Capsular polysaccharide Vi (virulence), porin, and LPS are three additional virulence determinants located within or on the external surface of the Salmonella outer membrane. The capsular polysaccharide Vi antigen occurs in most strains of S. Typhi, in a few strains of S. Paratyphi C, and rarely in S. Dublin (107). The Vi antigen significantly increases virulence by inhibiting the opsonization of the C3b host complement factor to bacterial surface LPS, a critical event in the induction of macrophage phagocytosis of invasive salmonellae (61). Porins are OMPs that function as transmembrane (outer membrane) channels in regulating the influx of nutrients, antibiotics, and other small-molecular species in response to environmental signals such as osmolarity, nutrient availability, and temperature. To date, four porins, encoded by the ompF, ompC, ompD, and phoE genes of S. Typhimurium, have been described. EnvZ activates OmpR through a phosphorylation-dependent mechanism, whereas the activated ompR locus regulates transcription of ompF and ompC. Mutations in the ompR envZ (designated ompB) regulon or in both ompC and ompF significantly attenuate the virulence of strains. Purified porins from various mutants induced long-term immunological responses in mice. In the ma-

10.  Salmonella Species

251

jority of serotypes, LPS forms the outermost layer protecting Salmonella from the environment in repelling the potentially lytic attack of the host complement system. LPS determines the bacteria’s virulence in an inverse relationship to the rates by which cells are phagocytosed and by which they activate the serum complement pathway. Salmonella with a smooth (long-LPS) phenotype is more virulent than Salmonella with the rough (shortLPS) phenotype (61). The long LPS of smooth variants sterically hinders the stable insertion of the C5b-9 complement factor into the inner cytoplasmic membrane that would precipitate bacteriolysis. The primary carbohydrate composition of LPS determines the serogroups, whereby serogroup B activates serum complement at a lower rate than group C. Correspondingly, serogroup B exhibits greater virulence than does serogroup C (61). Salmonella produces siderophores to compete with host transferrin, lactoferrin, and ferritin ligands for utilizing available iron, which is needed to drive key cellular functions such as the electron transport chain and enzymes associated with iron cofactors. In response to the limited availability of Fe3+ in host tissue, Salmonella sequesters Fe3+ ions by means of a high-affinity phenolate enterochelin (also designated enterobactin) and a low-affinity hydroxamate aerobactin chelator (159, 176). The fur (ferric uptake regulator) gene regulates the synthesis of these bacterial siderophores. Salmonella binding of trivalent iron begins with the interaction of a ferri-siderophore complex with an OMP receptor that was induced in response to the limiting concentrations of intracellular iron. The complex is then transposed into the bacterial cytoplasm, where the ferric moiety is reduced to the ferrous state. The low affinity of the side­rophore for Fe2+ results in the release of the divalent ion into the bacterial cytoplasm for subsequent use in key metabolic functions. It is notable that the degree of Salmonella virulence is directly related to the level of enterochelin in the infecting strain (60). Siderophores are not the only mechanism by which Salmonella can acquire iron from its host. A new fur-regulated operon that can compete with chelators like 2,2¢-dipyridyl for iron and influence Salmonella growth in vivo was identified within SPI-1 (233).

PhoQ regulates groups of genes essential to the survival of the bacterial cell within the macrophages and resistance to defensin and acid, as well as the causing of typhoid fever in mice. Signals from the hostile phagolysosomal environment trigger the autophosphorylation of the kinase sensor PhoQ, followed by the phosphorylation of PhoP at an aspartate residue by PhoQ (167). The phosphorylated PhoP protein activates the transcription of several phoPactivated genes (pag genes), including pagA, pagB, pagC, psiD, and the phoN locus, which encodes the periplasmic nonspecific acid phosphatase, as well as 13 more positively regulated loci (pagD to pagP). Among them, pagC encodes an OMP that promotes survival within macrophages, and pagB is part of an operon that includes the polymyxin resistance locus pmrA and pmrB genes, which are involved in regulation of multiple factors including virulence (69, 155). The pmrAB locus appears to regulate the pmrHFIJKLM operon, genes that are necessary for synthesis of 4-amino­arabinose lipid A. This modified LPS confers to the bacterium resistance to polymyxin and defensins (100). The rest of the pag genes (pagD through pagP) do not contribute to the phoP- and phoQ-dependent resistance to defensins, whereas the pagD, pagJ, pagK, and pagM genes participate in mouse virulence and in bacterial survival in macrophages (26). Mutations in this regulon (pagC, D, J, K, and M) result in phenotypes exhibiting decreased survival within macrophages and increased susceptibility to acidic pH, serum complement (pagC), and defensins. In contrast to the pag genes that are expressed under adverse environmental conditions such as low pH and nutrient deficiency and at stationary phase, the phoP-phoQ system also regulates the expression of phoP-repressed genes (prg), which are induced under nonstress conditions (25). The prgH operon (SPI-1) plays a determining role in the export of bacterial proteins necessary for invasion of epithelial cells (25, 172). Other proteins arising from the transcription of prg genes are required for the diffusion membrane ruffling of macrophages, increased pinocytosis, and formation within macrophages of spacious phagosomes, wherein slow acidification favors greater bacterial survival and attendant virulence (2).

Virulence Gene Regulation

FINAL REMARKS

A number of gene regulator families in Salmonella, such as CRP-cAMP, LysR/MetR, AraC-type, and twocomponen­t systems, tightly regulate the expression of virulence genes in response to environmental signals (46). Among them, the role of the two-component regulatory system PhoP/PhoQ in Salmonella virulence regulation has been characterized in the most detail. PhoP/

Salmonella spp. continue to be a leading cause of foodborne illness. The situation has persisted because of the widespread occurrence of salmonellae in the natural environment and their prevalence in many sectors of the global food chain (60). Raw poultry, meats, and meat-derived products are important v­ehicles of human salmonellosis; however, increasingly, illnesses

252 are associated with the consumption of fresh produce and dry food products. Sanitary practices during the harvesting, processing, and distribution of raw foods and food ingredients are critical, since there are no effective controls for many of those food categories (58, 62). The occurrence of Salmonella in low-moisture foods is a good example of current issues and future tasks. First, it is clear that inactivation of Salmonella under dry conditions is very difficult because of the enhanced resistance of the desiccated cells to other stresses such as heat, acid, and detergents (98). The enhanced resistance presents a great challenge to the effectiveness of both of the inactivation steps, i.e., food processing and the overall sanitation of the processing environment and equipment. The need for continued research on developing effective control measures is great. Second, the main sources of cross-contamination of Salmonella into final products are the processing environments and raw ingredients. Salmonella has been detected in the processing environment in multiple outbreak investigations. Contaminated raw ingredients used in products without a final kill step or stored near the final products after an effective kill step have been implicated in previous outbreaks. For example, contaminated spices used in children’s snack foods and potato chips have been shown to be a source of contamination (173). In the 2008–2009 peanut butter outbreak investigation, violation of zoning in the manufacturer’s facility was cited; raw peanuts were found stored in a room next to the peanut butter packaging area. Lack of GMPs is the root problem for cross-contamination of salmonellae in the final products. Systematic preventive control measures must be reinforced in the low-moisture-food industry. With the hope of achieving reductions in Salmonella incidence similar to that associated with poultry products, implementation of a HACCP plan may be successfully applied; however, the technical barriers in processing and sanitation will hinder the gains from these preventive efforts. Joint efforts are needed from academia, industry, and the government to improve the control of salmonellae in dry foods, through both the development of effective inactivation processes and the implementation of stringent preventive risk management strategies. We thank Mary Lou Tortorello, Donald H. Burr, and Thomas S. Hammack for their critical reviews of this chapter.

References 1. Alcaine, S. D., L. D. Warnick, and M. Wiedmann. 2007. Antimicrobial resistance in nontyphoidal Salmonella. J. Food Prot. 70:780–790.

Foodborne Pathogenic Bacteria 2. Alpuche-Aranda, C. M., J. A. Swanson, W. P. Loomis, and S. I. Miller. 1992. Salmonella Typhimurium activates virulence gene transcription within acidified macrophage phagosomes. Proc. Natl. Acad. Sci. USA 89:10079–10083. 3. Alvarez-Ordonez, A., A. Fernandez, M. Lopez, R. Arenas, and A. Bernardo. 2008. Modifications in membrane fatty acid composition of Salmonella Typhimurium in response to growth conditions and their effect on heat resistance. Int. J. Food Microbiol. 123:212–219. 4. Amavisit, P., D. Lightfoot, G. F. Browning, and P. F. Markham. 2003. Variation between pathogenic serovars within Salmonella pathogenicity islands. J. Bacteriol. 185:3624–3635. 5. Anaraki, S., I. Giraudon, and S. Cathcart. 2005. Large outbreak of Salmonella Enteritidis in north east London. Eurosurveill. Wkly. Rep. 10(11). 6. Anderson, P. N., M. E. Hume, J. A. Byrd, C. Hernandez, S. M. Stevens, K. Stringfellow, and D. J. Caldwell. 2010. Evaluation of repetitive extragenic palindromicpolymeras­e chain reaction and denatured gradient gel electrophoresis in identifying Salmonella serotypes isolated from processed turkeys. Poult. Sci. 89: 1293–1300. 7. Anderson, S. M., L. Verchick, R. Sowadsky, B. Sun, R. Civen, J. C. Mohle-Boetani, S. B. Werner, M. Starr, S. Abbott, M. Gutierrez, M. Palumbo, J. Farrar, P. Shillam, E. Umland, M. Tanuz, M. Sewell, J. Cato, W. Keene, M. Goldoft, J. Hofmann, P. Waller, C. Braden, G. Djomand, M. Reller, and W. Chege. 2002. Multistate outbreaks of Salmonella serotype Poona infections associated with eating cantaloupe from Mexico—United States and Canada, 2000-2002. MMWR Morb. Mortal. Wkly. Rep. 51:1044–1047. 8. Andrews, W. H., V. R. Bruce, G. A. June, P. Sherrod, T. S. Hammack, and R. M. Amaguana. 1995. Salmonella (chapt. 5). Bacteriological Analytical Manual, 8th ed. AOAC International, Arlington, VA. 9. Anonymous. 1977. Aviation Catering. Report of Working Group. World Health Organization (WHO) Regional Office for Europe, Copenhagen, Denmark. 10. Anonymous. 1995. Ice-cream firm reaches tentative Salmonella case agreement. Food Chem. News 36:53. 11. Anonymous. 1999. Salmonellosis, dried squid—Japan (Tokyo). Mainichi Daily News, 25 April 1999. 12. Anonymous. 2000. Salmonellosis outbreak associated with raw mung bean sprouts. News Release (2200), April. California Department of Health Services, Sacramento, CA. 13. Anonymous. 2001. Salmonella Stanley and Salmonella Newport in imported peanuts—international outbreak. WHO Newsl. 70:5. 14. Anonymous. 16 August 2001. International outbreak of Salmonella Typhimurium DT 104 due to contaminated sesame seed products. Update from Germany, United Kingdom and Norway. Eurosurveill. Wkly. Rep. 5(33). 15. Anonymous. 16 August 2001. Salmonella Typhimurium definitive type 104 in halva. Commun. Dis. Wkly. Rep. 11(33).

10.  Salmonella Species 16. Anonymous. 2005. Roast beef source of Salmonella that sickened 155 at Mother’s Day brunch. Canadian Press Wire (7 June 2005). 17. Anonymous. 2006. Outbreak of Salmonella Typhimurium phage type U302 in Ontario, Spring 2005. Canada Commun. Dis. Rep. 32(07):75–82. 18. Ashkenazi, S., T. G. Cleary, B. E. Murray, A. Wanger, and L. K. Pickering. 1988. Quantitative analysis and partial characterization of cytotoxin production by Salmonella strains. Infect. Immun. 56:3089–3094. 19. Asplund, K., and E. Nurmi. 1991. The growth of salmonellae in tomatoes. Int. J. Food Microbiol. 13:177–182. 20. Australian Department of Health and Ageing. 2011. Monitoring the incidence and causes of diseases potentially transmitted by food in Australia: annual report of the OzFoodNet network, 2009. Australian Department of Health and Ageing, Canberra, Australia. http://www. health.gov.au/internet/main/publishing.nsf/Content/cdacdi3404b.htm. Accessed January 2012. 21. Australian Health Commission. 1999. End to Salmonella outbreak. Press release (Adelaide), 25 March 1999. 22. Baker, R. C., R. A., Qureshi, and J. H. Hotchkins. 1986. Effect of an elevated level of carbon dioxide containing atmosphere on the growth of spoilage and pathogenic bacteria. Poult. Sci. 65:729–737. 23. Beales, N. 2004. Review of the microbiological risks associated with sprouted seeds. Report no. 41. Campden and Chorleywood Food Research Association Group, Campden, United Kingdom. 24. Beckers, H. J., M. S. M. Daniels-Bosman, A. Ament, J. Daenen, A. W. J. Hanekamp, P. Knipschild, A. H. H. Schuurmann, and H. Bijkerk. 1985. Two outbreaks of salmonellosis caused by Salmonella Indiana. A survey of the European Summit outbreak and its consequences. Int. J. Food Microbiol. 2:185–195. 25. Beers, A. 1997. Maryland church dinner Salmonella outbreak blamed on faulty food preparation. Food Chem. News 39:19–20. 26. Belden, W. J., and S. I. Miller. 1994. Further characterization of the PhoP regulon: identification of new PhoP-activated virulence loci. Infect. Immun. 62: 5095–5101. 27. Belgian Federal Agency for the Safety of the Food Chain. 2011. Russia: likely contamination of milk powder with Salmonella. Belgian Federal Agency for the Safety of the Food Chain, Brussels, Belgium. http://www.fasfc.be/press release/_documents/2012-01-23_salmonella-oranienburg_ en.pdf. Accessed January 2012. 28. Blanc-Potard, A. B., and E. A. Groisman. 1997. The Salmonella selC locus contains a pathogenicity island mediating intramacrophage survival. EMBO J. 16:5376–5385. 29. Blanc-Potard, A. B., F. Solomon, J. Kayser, and E. A. Groisman. 1999. The SPI-3 pathogenicity island of Salmonella enterica. J. Bacteriol. 181:998–1004. 30. Blaser, M. J., and L. S. Newman. 1982. A review of human salmonellosis. 1. Infective dose. Rev. Infect. Dis. 4:1096–1106.

253 31. Boase, J., S. Lipsky, P. Simani, S. Smith, C. Skilton, S. Greenman, S. Harrison, J. Duchin, M. Samadpour, R. Gautom, S. Lankford, T. Harris, K. Ly, D. Green, J. Kobyashi, E. DeBess, T. McGivern, S. Mauvais, V. Balan, D. Fleming, K. Sanchez, P. D. Vertz, J. C. Mohle-Boetani, D. Seuring, J. H. Goddard, S. A. Bidol, J. Bender, C. M. Sewell, I. N. Vold, L. Marengo, and J. Archer. 1999. Outbreak of Salmonella serotype Muenchen infections associated with unpasteurized orange juice—United States and Canada. MMWR Morb. Mortal. Wkly. Rep. 48:582–585. 32. Bone, A., H. Noel, S. Le. Hello, N. Pihier, C. Danan, M. E. Raguenaud, S. Salah, H. Bellali, V. Vaillant, F. X. Weill, and N. Jourdan-da Silva. 2010. Nationwide outbreak of Salmonella enterica serotype 4,12:i:- infections in France, linked to dire pork sausage, MarchMay 2010. Euro Surveill. 15:pii=19592. http://www. eurosurveillance.org/ViewArticle.aspx?ArticleId= 19592. Accessed January 2012. 33. Bouvet, E., C. Jestin, and R. Ancelle. 1986. Importance of exported cases of salmonellosis in the revelation of an epidemic, p. 303. In Proceedings of the Second World Congress on Foodborne Infections and Intoxications. Institute of Veterinary Medicine—Robert von OstertagInstitute, Berlin, Germany. 34. Brenner, F. W., R. G. Villar, F. J. Angulo, R. Tauxe, and B. Swaminathan. 2000. Salmonella nomenclature. J. Clin. Microbiol. 38:2465–2467. 35. Bulling, E., R. Stephan, and V. Sebek. 1973. The development of antibiotic resistance among Salmonella bacteria of animal origin in the Federal Republic of Germany and West Berlin: 1st communication: a comparison between the years of 1961 and 1970-1971. Zentralbl. Bakteriol. Mikrobiol. Hyg. 1. Abt. Orig. A 225:245–256. 36. Bunning, V. K., R. B. Raybourne, and D. L. Archer. 1988. Foodborne enterobacterial pathogens and rheumatoid disease. J. Appl. Bacteriol. Symp. Suppl. 17:87S–107S. 37. Burnett, S. L., E. R. Gehm, W. R. Weissinger, and L. R. Beuchat. 2000. Survival of Salmonella in peanut butter and peanut butter spread. J. Appl. Microbiol. 89:472–477. 38. Burslem, C. D., M. J. Kelly, and F. S. Preston. 1990. Food poisoning—a major threat to airline operations. J. Soc. Occup. Med. 40:97–100. 39. Bush, K. 2001. New beta-lactamases in gram-negative bacteria: diversity and impact on the selection of anti­ microbial therapy. Clin. Infect. Dis. 32:1085–1089. 40. Butler, M. A. 2000. Salmonella outbreak leads to juice recall in western states. Food Chem. News 42:19–20. 41. Camps, N., A. Dominguez, M. Cy, M. Perez, J. Pardos, T. Llobet, M. A. Usera, L. Salleras, and the Working Group for the Investigation of the Outbreak of Salmonellosis in Torroella de Montgri. 2005. A foodborne outbreak of Salmonella infection due to overproduction of eggcontaining foods for a festival. Epidemiol. Infect. 133:817–822.

254 42. Carlson, S. A., L. F. Bolton, C. E. Briggs, H. S. Hurd, V. K. Sharma, P. J. Fedorka-Cray, and B. D. Jones. 1999. Detection of multiresistant Salmonella typhimurium DT104 using multiplex and fluorogenic PCR. Mol. Cell. Probes 13:213–222. 43. Catsaras, M., and D. Grebot. 1984. Multiplication des Salmonella dans la viande hachée. Bull. Acad. Vét. France 57:501–512. 44. Centers for Disease Control and Prevention. 2006. Multistate outbreak of Salmonella Typhimurium infections associated with eating ground beef—United States, 2004. MMWR Morb. Mortal. Wkly. Rep. 55:180–182. 45. Chen, Y., T. Freier, J. Kuehm, M. Moorman, J. Scott, J. Meyer, T. Morille-Hinds, L. Post, L. Smoot, L. Hood, J. Shebuski, and J. Banks. 2009. Control of Salmonella in Low Moisture Foods. Grocery Manufacturers Association, Washington, DC. 46. Clements, M., S. Eriksson, D. Tezcan-Merdol, J. C. Hinton, and M. Rhen. 2001. Virulence gene regulation in Salmonella enterica. Ann. Med. 33:178–185. 47. Christmann, D., T. Staub, and Y. Hansmann. 1992. Manifestations extra-digestives des salmonelloses. Méd. Mal. Infect. 22:289–298. 48. Chung, K. C., and J. M. Goepfert. 1970. Growth of Salmonella at low pH. J. Food Sci. 35:326–328. 49. Cohen, D. R., I. A. Porter, T. M. S. Reid, J. C. M. Sharp, G. I. Forbes, and G. M. Paterson. 1983. A cost benefit study of milk-borne salmonellosis. J. Hyg. 91:17–23. 50. Communicable Disease Report Weekly. 2005. Outbreaks of infection with Salmonella Enteritidis PT6 infection in the north east of England associated with eggs. CDR Wkly. 15:41. 51. Corby, S., V. Lanni, V. Kistler, V. Dato, A. Weltman, C. Yozviak, K. Waller, K. Nalluswami, M. Moll, J. Lockett, S. Montgomery, M. Lynch, C. Braden, S. K. Gupta, and A. DuBois. 2005. Outbreaks of Salmonella infections associated with eating Roma tomatoes—United States and Canada, 2004. MMWR Morb. Mortal. Wkly. Rep. 54:325–328. 52. Cowden, J., S. O’Brien, B. Adak, et al. 2003. Outbreak of Salmonella Bareilly in Great Britain. Results from the case-control study. Eurosurveill. Wkly. Rep. 7:44. 53. Craven, P. C., D. C. Mackel, W. B. Baine, W. H. Barker, E. J. Gangarosa, M. Goldfield, H. Rosenfeld, R. Altman, G. Lachapelle, J. W. Davies, and R. C. Swanson. 1975. International outbreak of Salmonella Eastbourne infection traced to contaminated chocolate. Lancet i:788–793. 54. Crook, P. D., J. F. Aguilera, E. J. Threlfall, S. J. O’Brien, G. Sigmunddsdottir, D. Wilson, I. S. T. Fisher, A. Ammon, H. Briem, J. M. Cowden, M. E. Locking, H. Tschäpe, W. van Pelt, L. R. Ward and M. A. Widdowson. 2003. A European outbreak of Salmonella enterica serotype Typhimurium definitive phage type 204b in 2000. Clin. Microbiol. Infect. 9:839–845. 55. Crosa, J. H., D. J. Brenner, W. H. Ewing, and S. Falkow. 1973. Molecular relationships among the salmonellae. J. Bacteriol. 115:307–315.

Foodborne Pathogenic Bacteria 56. Cukan, A. 2005. Caregiving: food safety is a big concern. United Press International (October 13, 2005). http://www.extension.iastate.edu/foodsafety/news/ fsnews.cfm?newsid=10645. Accessed August, 2012. 57. Cummings, K., E. Barrett, J. C. Mohle-Boetani, J. T. Brooks, J. Farrar, T. Hunt, A. Fiore, K. Komatsu, S. B. Werner, and L. Slutsker. 2001. A multistate outbreak of Salmonella enterica serotype Baildon associated with domestic raw tomatoes. Emerg. Infect. Dis. 7:1046–1048. 58. D’Aoust, J.-Y. 1977. Salmonella and the chocolate industry. J. Food Prot. 40:718–727. 59. D’Aoust, J.-Y., D. W. Warburton, and A. M. Sewell. 1985. Salmonella Typhimurium phage-type 10 from cheddar cheese implicated in a major Canadian foodborne outbreak. J. Food Prot. 48:1062–1066. 60. D’Aoust, J.-Y. 1989. Salmonella, p. 327-445, In M.P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY. 61. D’Aoust, J.-Y. 1991. Pathogenicity of foodborne Salmonella. Int. J. Food Microbiol. 12:17–40. 62. D’Aoust, J.-Y. 1994. Salmonella and the international food trade. Int. J. Food Microbiol. 24:11–31. 63. D’Aoust, J.-Y., and U. Purvis. 1998. Isolation and Identification of Salmonella from Foods. MFHPB-20. Health Protection Branch, Health Canada, Ottawa, Ontario, Canada. 64. D’Aoust, J.-Y. 2000. Salmonella, p. 1233–1299. In B. M. Lund, A. C. Baird-Parker and G. W. Gould (ed.), The Microbiological Safety and Quality of Food. Aspen Publishers, Gaithersburg, MD. 65. Dechet, A. M., E. Scallan, K. Gensheimer, R. Hoekstra, J. Gunderman-King, J. Lockett, D. Wrigley, W. Chege, J. Sobel, and the Multistate Working Group. 2006. Outbreak of multidrug-resistant Salmonella enterica serotype Typhimurium definitive type 104 infection linked to commercial ground beef, northeastern United States, 2003-2004. Clin. Infect. Dis. 42:747–752. 66. De Medici, D., L. Croci, E. Delibato, S. Di Pasquale, E. Filetici, and L. Toti. 2003. Evaluation of DNA extraction methods for use in combination with SYBR green I realtime PCR to detect Salmonella enterica serotype enteritidis in poultry. Appl. Environ. Microbiol. 69:3456–3461. 67. den Bakker, H. C., A. I. M. Switt, C. A. Cummings, K. Hoelzer, L. Degoricija, L. D. Rodriguez-Rivera, E. M. Wright, R. Fang, M. Davis, T. Root, D. SchoonmakerBopp, K. A. Musser, E. Villamil, H. Waechter, L. Kornstein, M. R. Furtado, and M. Wiedmann. 2011. A whole-genome single nucleotide polymorphism-based approach to trace and identify outbreaks linked to a common Salmonella enterica subsp. enterica serovar Montevideo pulsed-field gel electrophoresis type. Appl. Environ. Microbiol. 77:8648–8655. 68. Dewanti, R., and M. P. Doyle. 1992. Influence of cultural conditions on cytotoxin production by Salmonella Enteritidis. J. Food Prot. 55:28–33. 69. Dorman C. J., S. Chatfield, C. F. Higgins, C. Hayward, and G. Dougan. 1989. Characterization of porin and ompR mutants of a virulent strain of Salmonella

10.  Salmonella Species

70.

71.

72.

73.

74. 75. 76.

77. 78. 79.

80.

81.

82. 83.

Typhimurium: ompR mutants are attenuated in vivo. Infect. Immun. 57:2136–2140. Doyle, M. E., C. Kaspar, J. Archer, and R. Klos. 2009. White paper on human illness caused by Salmonella from all food and non-food vectors. Food Research Institute, University of Wisconsin-Madison, Madison, WI. http://fri.wisc.edu/docs/pdf/FRI_Brief_Salmonella_ Human_Illness_6_09.pdf. Accessed January 2012. Drociuk, D., S. Carnasale, G. Elliot, L. J. Bell, J. J. Gibson, L. Wolf, D. Briggs, B. Jenkins, J. M. Maillard, M. Huddle, F. Virgin, C. Braden, P. Srikantiah, A. Stoica, and T. Chiller. 2003. Outbreaks of Salmonella serotype Enteritidis infection associated with eating shell eggs—United States, 1999-2001. MMWR Morb. Mortal. Wkly. Rep. 51:1149–1151. Droffner, M. L., and N. Yamamoto. 1992. Procedure for isolation of Escherichia, Salmonella, and Pseudomonas mutants capable of growth at the refractory temperature of 54oC. J. Microbiol. Methods 14:201–206. Earnshaw, R. G., J. Appleyard, and R. M. Hurst. 1995. Understanding physical inactivation processes: combined preservation opportunities using heat, ultrasound and pressure. Int. J. Food Microbiol. 28:197–219. Enright, M. C., and B. G. Spratt. 1999. Multilocus sequence typing. Trends Microbiol. 7:482–487. Ewing, W. H. 1972. The nomenclature of Salmonella, its usage, and definitions for the three species. Can. J. Microbiol. 18:1629–1637. Fell, G., O. Hamouda, R. Lindner, S. Rehmet, A. Liesegang, R. Prager, B. Gericke, and L. Petersen. 2000. An outbreak of Salmonella Blockley infections following smoked eel consumption in Germany. Epidemiol. Infect. 125:9–12. Ferreira, M. A. S. S., and B. M. Lund. 1987. The influence of pH and temperature on initiation of growth of Salmonella spp. Lett. Appl. Microbiol. 5:67–70. Fluit, A. C., and F. J. Schmitz. 1999. Class 1 integrons, gene cassettes, mobility, and epidemiology. Eur. J. Clin. Microbiol. Infect. Dis. 18:761–770. Foley, S. L., D. G. White, P. F. McDermott, R. D. Walker, B. Rhodes, P. J. Fedorka-Cray, S. Simjee, and S. Zhao. 2006. Comparison of subtyping methods for differentiating Salmonella enterica serovar Typhimurium isolates obtained from food animal sources. J. Clin. Microbiol. 44:3569–3577. Foley, S. L., S. Zhao, and R. D. Walker. 2007. Comparison of molecular typing methods for the differentiation of Salmonella foodborne pathogens. Foodborne Pathog. Dis. 4:253–276. Food Safety Authority of Ireland. 2010. Largest outbreak in recent years linked to Salmonella in duck eggs. Food Safety Authority of Ireland, Dublin, Ireland. http://www .fsai.ie/14092010.html. Accessed January 2012. Food Safety Net—Canada. 2005. Salmonellosis, foodborne, fatal—USA (South Carolina): turkey (3 June 2005). Food Safety Net—Canada. 2003. Gastro cases top 100 as likely cause found (17 January 2003).

255 84. Food Safety Net—Canada. 2005. Salmonellosis, political gathering—Honduras, Atlantida (28 September 2005). 85. Food Safety Net—Canada. 2001. Romanian kids sickened by Salmonella (27 October 2001). 86. Foster, J. W., and H. K. Hall. 1991. Inducible pH homeostasis and the acid tolerance response of Salmonella Typhimurium. J. Bacteriol. 173:5129–5135. 87. Francis, B. J., J. V. Altamirano, M. G. Stobierski, W. Hall, B. Robinson, S. Dietrich, R. Martin, F. Downes, K. R. Wilcox, C. Hedberg, R. Wood, M. Osterholm, G. Genese, M. J. Hung, S. Paul, K. C. Spitalny, C. Whalen, and J. Spika. 1991. Multistate outbreak of Salmonella Poona infections—United States and Canada, 1991. MMWR Morb. Mortal. Wkly. Rep. 40:549–552. 88. Garcia-del Portillo, F., J. W. Foster, and B. B. Finlay. 1993. Role of acid tolerance response genes in Salmonella Typhimurium virulence. Infect. Immun. 61: 4489–4492. 89. Geiss, H. K., I. Ehrhard, A. Rösen-Wolff, H. G. Sonntag, J. Pratsch, A. Wirth, D. Krüger, I. KnollmannSchanbacher, H. Kühn, and C. Treiber-Klötzer. 1993. Foodborne outbreak of a Salmonella Enteritidis epidemic in a major pharmaceutical company. Gesundh.Wes. 55:127–132. 90. Gibbs, R., N. Pingault, T. Mazzucchelli, L. O’Reilly, B. MacKenzie, J. Green, R. Mogyorosy, R. Stafford, R. Bell, L. Hiley, K. Fullerton, and P. V. Buynder. 2009. An outbreak of Salmonella enterica serotype Litchfield infection in Australia linked to consumption of contaminated papaya. J. Food Prot. 72:1094–1098. 91. Gillispie, I. 2004. Outbreak of Salmonella Newport infection associated with lettuce in the U.K. Eurosurveill. Rep. 8:41. 92. Gilsdorf, A., A. Jansen, K. Alpers, H. Dieckmann, U. van Treeck, A. M. Hauri, G. Fell, M. Littmann, P. Rautenberg, R. Prager, W. Rabsch, P. Roggentin, A. Schroeter, A. Miko, E. Bartelt, J. Bräunig, and A. Ammon. 2005. A nationwide outbreak of Salmonella Bovismorbificans PT24, Germany, December 2004– March 2005. Eurosurveill. Wkly. Rep. 10:12. 93. Glynn, J. R., and D. J. Bradley. 1992. The relationship of infecting dose and severity of disease in reported outbreaks of Salmonella infections. Epidemiol. Infect. 109:371–388. 94. Glynn, J. R., and S. R. Palmer. 1992. Incubation period, severity of disease, and infecting dose: evidence from a Salmonella outbreak. Am. J. Epidemiol. 136:1369–1377. 95. Goepfert, J. M., I. K. Iskander, and C. H. Amundson. 1970. Relation of the heat resistance of salmonellae to the water activity of the environment. Appl. Microbiol. 19:429–433. 96. Gordenker, A. 1999. Common seed source identified in Canadian outbreak. Food Chem. News 41:23. 97. Grimont, P. A. D., and F. Weill. 2007. Antigenic formulae of the Salmonella serovars, 9th ed. WHO Collaborating Centre for Reference and Research on Salmonella, Institut Pasteur, Paris, France.

256 98. Gruzdev, N., R. Pinto, and S. Sela. Effect of desiccation on tolerance of Salmonella Enterica to multiple stresses. Appl. Environ. Microbiol. 77:1667–1673. 99. Guerin, P. J., B. De Jong, E. Heir, V. Hasseltvedt, G. Kapperud, K. Styrmo, B. Gondrosen, J. Lassen, Y. Andersson, and P. Aavitsland. 2004. Outbreak of Salmonella Livingstone infection in Norway and Sweden due to contaminated processed fish products. Epidemiol. Infect. 132:889–895. 100. Gunn, J. S., S. S. Ryan, J. C. Van Velkinburgh, R. K. Ernst, and S. I. Miller. 2000. Genetic and functional analysis of a Pmr-A-PmrB-regulated locus necessary for lipopolysaccharide modification, antimicrobial peptide resistance, and oral virulence of Salmonella enterica serovar Typhimurium. Infect. Immun. 68:6139–6146. 101. Guo, X., J. Chen, R. E. Brackett, and L. R. Beuchat. 2001. Survival of Salmonella on and in tomato plants from the time of inoculation at flowering and early stages of fruit development through fruit ripening. Appl. Environ. Microbiol. 76:4760–4764. 102. Hedberg, C. W., F. J. Angulo, K. E. White, C. W. Langkop, W. L. Schell, M. G. Stobierski, A. Schuchat, J. M. Besser, S. Dietrich, L. Helsel, P. M. Griffin, J. W. McFarland, M. T. Osterholm, et al. 1999. Outbreaks of salmonellosis associated with eating uncooked tomatoes: implications for public health. Epidemiol. Infect. 122:385–393. 103. Hellström, L. 1980. Food-transmitted S. Enteritidis epidemic in 28 schools, p. 397–400. In Proceedings of the World Congress on Foodborne Infections and Intoxications. Institute of Veterinary Medicine—Robert von Ostertag-Institute, Berlin, Germany. 104. Henkel, J. 1995. Ice cream linked to Salmonella outbreak. FDA Consumer 29:30–31. 105. Hermans, A. P., T. Abee, M. H. Zwietering, and H. J. Aarts. 2005. Identification of novel Salmonella enterica serovar Typhimurium DT104-specific prophage and nonprophage chromosomal sequences among serovar Typhimurium isolates by genomic subtractive hybridization. Appl. Environ. Microbiol. 71:4979–4985. 106. Herrera-Leon, S., J. R. McQuiston, M. A. Usera, P. I. Fields, J. Garaizar, and M. A. Echeita. 2004. Multiplex PCR for distinguishing the most common phase-1 flagellar antigens of Salmonella spp. J. Clin. Microbiol. 42:2581–2586. 107. Hickey, E. W., and I. N. Hirshfield. 1990. Low pH-induce­d effects of patterns of protein synthesis and on internal pH in Escherichia coli and Salmonella Typhimurium. Appl. Environ. Microbiol. 56:1038–1045. 108. Hiramatsu, R., M. Matsumoto, K. Sakae, and Y. Miyazaki. 2005. Ability of Shiga toxin-producing Escherichia coli and Salmonella spp. to survive in a desiccation model system and in dry foods. Appl. Environ. Microbiol. 71:6657–6663. 109. Holley, R. A., and M. Proulx. 1986. Use of egg washwater pH to prevent survival of Salmonella at moderate temperatures. Poult. Sci. 65:922–928. 110. Holmberg, S. D., M. T. Osterholm, K. A. Senger, and M. L. Cohen. 1984. Drug-resistant Salmonella from animal fed antimicrobials. N. Engl. J. Med. 311:617–622.

Foodborne Pathogenic Bacteria 111. Horwitz, M. A., R. A. Pollard, M. H. Merson, and S. M. Martin. 1977. A large outbreak of foodborne salmonellosis on the Navajo nation Indian reservation, epidemiology and secondary transmission. Am. J. Public Health 67:1071–1076. 112. Hueck, C. J. 1998. Type III secretion systems in bacterial pathogens of animals and plants. Microbiol. Mol. Biol. Rev. 62:379–433. 113. Ingham, S. C., R. A. Alford, and A. P. McCown. 1990. Comparative growth rates of Salmonella Typhimurium and Pseudomonas fragi on cooked crab meat stored under air and modified atmosphere. J. Food Prot. 53:56. 114. Isaacs, S., J. Arimini, B. Cieben et al. 2005. An international outbreak of salmonellosis associated with raw almonds contaminated with a rare phage type of Salmonella Enteritidis. J. Food Prot. 68:191–198. 115. Jakabi, M., D. S. Gelli, J. C. Torre, M. A. Rodas, B. D. Franco, M. T. Destro, and M. Landgrafi. 2003. Inactivation by ionizing radiation of Salmonella Enteritidis, Salmonella Infantis, and Vibrio parahaemolyticus in oysters (Crassostrea brasiliana). J. Food Prot. 66:1025–9. 116. Jalava, K., J. Perevoscikovs, A. Siitonen, et al. 2002. Salmonella Enteritidis PT 4 infections among a group of Finns visiting Riga: effective collaboration between Latvian and Finnish authorities resolved an outbreak. EpiNorth 2 (1). 117. Jansen, A., C. Frank, R. Prager, H. Oppermann, and K. Stark. 2005. Nation-wide outbreak of Salmonella Give in Germany, 2004. Z. Gastroenterol. 43:707–713. 118. Japan Ministry of Health and Welfare. 1999. Departmental Health Report. Japan Ministry of Health and Welfare, Tokyo. 119. Jones, R. C., V. Reddy, L. Kornstein, J. R. Fernandez, F. Stavinsky, A. Agasan, and S. I. Gerber. 2004 Salmonella enterica serotype Uganda infection in New York City and Chicago. Emerg. Infect. Dis. 10:1665–1667. 120. Juneja, V. J., and D. Schaffner. 2011. Predictive modelling in foods. Food Microbiol. 28:629–838. 121. Kapperud, G., S. Gustavsen, I. Hellesnes, A. H. Hansen, J. Lassen, J. Hirn, M. Jahkola, M. A. Montenegro, and R. Helmuth. 1990. Outbreak of Salmonella Typhimurium infection traced to contaminated chocolate and caused by a strain lacking the 60-megadalton virulence plasmid. J. Clin. Microbiol. 28:2597–2601. 122. Kauffmann, F. 1966. The Bacteriology of Enterobacteria­ ceae. Munksgaard, Copenhagen, Denmark. 123. Keady, S., G. Briggs, J. Farrar, J. C. Mohle-Boetani, J. O’Connell, S. B. Werner, D. Anderson, L. Tenglesen, S. Bidols, B. Albanese, C. Gordan, E. DeBess, J. Hatch, W. E. Keene, M. Plantenga, J. Tierheimer, A. L. Hackmann, C. E. Rinehardt, C. H. Sandt, A. Ingram, S. Hansen, S. Hurt, M. Poulson, R. Pallipamu, J. Wiclund, C. Braden, J. Lockett, S. Van Duyne, A. Dechet, and C. Smelser. 2004. Outbreak of Salmonella serotype Enteritidis infections associated with raw almonds-United States and Canada, 2003–2004. MMWR Morb. Mortal. Wkly. Rep. 53:484–487. 124. Keene, W. E., K. Hedberg, P. Cieslak, S. Schafer, and E. Dechet. 2004. Salmonella serotype Typhimurium

10.  Salmonella Species outbreak associated with commercially processed egg salad—Oregon, 2003. MMWR Morb. Mortal. Wkly. Rep. 53:1132–1134. 125. Keyes, K., M. D. Lee, and J. J. Maurer. 2003. Antibiotics: mode of action, mechanism of resistance and transfer, p. 45–46. In M. Torrence and R. Isaacson (ed.), Current Topics in Food Safety in Animal Agriculture. Iowa State University Press, Ames, IA. 126. Kim, C. J., D. A. Emery, H. Rinke, K. V. Nagaraja, and D. A. Halvorson. 1989. Effect of time and temperature on growth of Salmonella Enteritidis in experimentally inoculated eggs. Avian Dis. 33:735–742. 127. Kimura, A. C., M. S. Palumbo, H. Meyers, S. Abbott, R. Rodriguez, and S. B. Werner. 2005. A mutli-state outbreak of Salmonella serotype Thompson infection from commercially distributed bread contaminated by an ill food handler. Epidemiol. Infect. 133:823–828. 128. Kivi, M., W. van Pelt, D. Notermans, A. van de Giessen, W. Wannet, and A. Bosman. 2005. Large outbreak of Salmonella Typhimurium DT 104, The Netherlands, September-November, 2005. Eurosurveill. Wkly. 10:12. 129. Klontz, K. C., J. C. Klontz, and R. M. Hoekstra. 2010. Analysis of tomato and jalapeno and serrano pepper imports into the United States from Mexico before and during a national outbreak of Salmonella serotype Saintpaul infections in 2008. J. Food Prot. 73:1967–1974. 130. Koch, J., A. Schrauder, K. Alpers, D. Werber, C. Frank, R. Prager, W. Rabsch, S. Broll, F. Feil, P. Roggentin, J. Bockemuhl, H. Tschape, A. Ammon, and K. Stark. 2005. Salmonella Agona outbreak from contaminated aniseed, Germany. Emerg. Infect. Dis. 11:1124–1127. 131. Koo, F. C. W., J. W. Peterson, C. W. Houston, and N. C. Molina. 1984. Pathogenesis of experimental salmonellosis: inhibition of protein synthesis by cytotoxin. Infect. Immun. 43:93–100. 132. Kuhn, K. G., M. Torpdahl, C. Frank, K. Sigsgaard, and S. Ethelberg. 2011. An outbreak of Salmonella Typhimurium traced back to salami, Denmark, April to June 2010. Eurosurveillance 16:13–16. 133. Lakins, D. G., C. Z. Alvarado, L. D. Thompson, M. T. Brashears, J. C. Brooks, and M. M. Brashears. 2008. Reduction of Salmonella Enteritidis in shell eggs using directional microwave technology. Poult. Sci. 87:985–991. 134. Lecos, C. 1986. Of microbes and milk: probing America’s worst Salmonella outbreak. Dairy Food Sanitat. 6:136–140. 135. Lee, I. S., J. L. Slonczewski, and J. W. Foster. 1994. A low-pH-inducible, stationary phase acid tolerance response in Salmonella Typhimurium. J. Bacteriol. 176:1422–1426. 136. Lehmacher, A., J. Bockemühl, and S. Aleksic. 1995. A nationwide outbreak of human salmonellosis in Germany due to contaminated paprika and paprika-powdered potato chips. Epidemiol Infect. 115:501–511. 137. Le Minor, L. 1981. The Genus Salmonella, p. 1148– 1159. In M. P. Starr, H. Stolp, H. G. Truper, A. Balows,

257 and H. G. Schlegel (ed.), The Prokaryotes. SpringerVerlag, New York, NY. 138. Le Minor, L., and M. Y. Popoff. 1987. Request for an opinion. Designation of Salmonella enterica sp. nov., nom. rev., as the type and only species of the genus Salmonella. Int. J. Syst. Bacteriol. 37:465–468. 139. Le Minor, L., C. Coynault, and G. Pessoa. 1974. Déterminisme plasmidique du caractère atypique “lactose positif” de souches de S. Typhimurium et de S. oranienburg isolées au Brésil lors d’épidémies de 1971 à 1973. Ann. Microbiol. 125A:261–285. 140. Le Minor, L., C. Coynault, R. Rhode, B. Rowe, and S. Aleksic. 1973. Localisation plasmidique de déterminant génétique du caractère atypique “saccharose +” des Salmonella. Ann. Microbiol. 124B:295–306. 141. Le Minor, L., M. Y. Popoff, B. Laurent, and D. Hermant. 1986. Individualisation d’une septième sous-espèce de Salmonella: S. choleraesuis subsp. indica subsp. nov. Ann. Inst. Pasteur Microbiol. 137B:211–217. 142. Lenglet, A. 2005. E-alert 9 August: over 2000 cases so far in Salmonella Hadar outbreak in Spain associated with consumption of pre-cooked chicken, July-August, 2005. Eurosurveill. Wkly. Rep. 10:8. 143. Leyer, G. J., and E. A. Johnson. 1993. Acid adaptation induces cross-protection against environmental stresses in Salmonella Typhimurium. Appl. Environ. Microbiol. 59:1842–1847. 144. Libby, S. J., L. G. Adams, T. A. Ficht, C. Allen, H. A. Whitford, N. A. Buchmeier, S. Bossie, and D. G. Guiney. 1997. The spv genes on the Salmonella Dublin virulence plasmid are required for severe enteritis and systemic infection in the natural host. Infect. Immun. 65:1786–1792. 145. Lienau, E. K., E. Strain, C. Wang, J. Zheng, A. R. Ottesen, C. E. Keys, T. S. Hammack, S. M. Musser, E. W. Brown, M. W. Allard, G. Cao, J. Meng, and R. Stones. Identification of a salmonellosis outbreak by means of molecular sequencing. N. Engl. J. Med. 364:981–982. 146. Lim, H., K. H. Lee, C.-H. Hong, G.-J. Bahk, and W. S. Choi. 2005. Comparison of four molecular typing methods for the differentiation of Salmonella spp. Int. J. Food Microbiol. 105:411–418. 147. Liming, S. H., and A. A. Bhagwat. 2004. Application of a molecular beacon-real-time PCR technology to detect Salmonella species contaminating fruits and vegetables. Int. J. Food Microbiol. 95:177–187. 148. Lin, C., and C. Wei. 1997. Transfer of Salmonella Montevideo onto the interior surfaces of tomatoes by cutting. J. Food Prot. 60:858–863. 149. Liu, L., H. F. He, C. F. Dai , L. H. Liang, T. Li, L. H. Li, H. M. Luo, and R. Fontain. 2006. Salmonellosis outbreak among factory workers—Huizhou, Guangdong Province, China, July 2004. MMWR Morb. Mortal. Wkly. Rep. 55(Suppl. 01):35–38. 150. Luk, J. M., U. Kongmuang, P. R. Reeves, and A. A. Lindberg. 1993. Selective amplification of abequose and paratose synthase genes (rfb) by polymerase chain reaction for identification of Salmonella major serogroups (A, B, C2, and D). J. Clin. Microbiol. 31:2118–2123.

258 151. Mahmoud, B. S. 2009. Effect of X-ray treatments on inoculated Escherichia coli O157: H7, Salmonella enterica, Shigella flexneri and Vibrio parahaemolyticus in ready-to-eat shrimp. Food Microbiol. 26:860–864. 152. Matsuzaki, S., M. Rashel, J. Uchiyama, S. Sakurai, T. Ujihara, M. Kuroda, M. Ikeuchi, T. Tani, M. Fujieda, H. Wakiguchi, and S. Imai. 2005. Bacteriophage therapy: a revitalized therapy against bacterial infectious diseases. J. Infect. Chemother. 11:211–219. 153. Mattila, L. M. Leirisalo-Repo, S. Koskimies, K. Granfors, and A. Sütonen. 1994. Reactive arthritis following an outbreak of Salmonella infection in Finland. Br. J. Rheumat. 33:1136–1141. 154. McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L. Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S. Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan, H. Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston, and R. K. Wilson. 2001. Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413:852–856. 155. Miller, S. I., A. M. Kukral, and J. J. Mekalanos. 1989. A two-component regulatory system (phoP/phoQ) controls Salmonella Typhimurium virulence. Proc. Natl. Acad. Sci. USA 86:5054–5058. 156. Mintz, E. D., M. L. Cartter, J. L. Hadler, J. T. Wassell, J. A. Zingeser, and R. V. Tauxe. 1994. Dose-response effects in an outbreak of Salmonella Enteritidis. Epidemiol. Infect. 112:13–23. 157. Miyagawa, S., and A. Miki. 1992. The epidemiological data of food poisoning in 1991. Food Sanit. Res. 42:78–104. 158. Mouzin, E., S. B. Werner, R. G. Bryant, et al. 1997. When a health food becomes a hazard: a large outbreak of salmonellosis associated with alfalfa sprouts—California. Epidemic Intelligence Service Conference, Centers for Disease Control and Prevention, Atlanta, GA. 159. Nassif, X., and P. Sansonetti. 1987. Les systèmes bactériens de captation du fer: leur rôle dans la virulence. Bull. Inst. Pasteur 85:307–327. 160. National Advisory Committee on Microbiological Criteria for Foods. 1999. Microbiological safety evaluations and recommendations on fresh produce. Food Control 10:117–143. 161. National Advisory Committee on Microbiological Criteria for Foods. 1999. Microbiological safety evaluations and recommendations on sprouted seeds. Int. J. Food Microbiol. 52:123–153. 162. National Antimicrobial Resistance Monitoring System. 2009. NARMS 2009 Executive Report. U.S. Food and Drug Administration/Centers for Disease Control and Prevention/U.S. Department of Agriculture. http://www. fda.gov/downloads/AnimalVeterinary/SafetyHealth/ AntimicrobialResistance/NationalAntimicrobialResistance MonitoringSystem/UCM268954.pdf. Accessed January 2012. 163. Nicolay, N., L.Thornton, S. Cotter, P. Garvery, O.Bannon, P. McKeown, M. Cormican, I. Fisher, C. Little, N. Boxall, E. Depinna, T. M. Peters, J. Cowden, R. Salmon, B. Mason, N. Irvine, P. Rooney, and D.

Foodborne Pathogenic Bacteria O’Flanagan. 2011. Salmonella enterica serovar Agona European outbreak associated with a food company. Epidemiol. Infect. 139:1272–1280. 164. Nield, B. S., A. J. Holmes, M. R. Gillings, G. D. Recchia, B. C. Mabbutt, K. M. Nevalainen, and H. W. Stokes. 2001. Recovery of new integron classes from environmental DNA. FEMS Microbiol. Lett. 195: 59–65. 165. O’Brien, S., and L. Ward. 2002. Outbreak of Salmonella Enteritidis PT 14b in the United Kingdom—second update. Euro Surveill. 6:pii=1934. http://www.eurosurveil lance.org/ViewArticle.aspx?ArticleId=1934. Accessed January 2012. 166. Ogawa, H., H. Tokunou, M. Sasaki, T. Kishimoto, and K. Tamura. 1991. An outbreak of bacterial food poisoning caused by roast cuttlefish “yaki-ika” contaminated with Salmonella spp. Champaign. Jpn. J. Food Microbiol. 7:151–157. 167. Ohl, M. E., and S. I. Miller. 2001. Salmonella: a model for bacterial pathogenesis. Annu. Rev. Med. 52:259–274. 168. Olson, E. R. 1993. Influence of pH on bacterial gene expression. Mol. Microbiol. 8:5–14. 169. Otto, H. D. Tezcan-Merdol, R. Girisch, F. Haag, M. Rhen, and F. Koch-Nolte. 2000. The spvB geneproduct of the Salmonella enterica virulence plasmid is a mono (ADP-ribosyl) transferase. Mol. Microbiol. 37:1106–1115. 170. Palmer, S. R., and B. Rowe, et al. 1986. Trends in Salmonella infections. Microbiol. Dig. 3:18–21. 171. Parish, M. E. 1998. Coliforms, Escherichia coli and Salmonella serovars associated with a citrus-processing facility implicated in a salmonellosis outbreak. J. Food Prot. 61:280–284. 172. Pengues, D. A., M. J. Hantman, I. Behlau, and S. I. Miller. 1995. PhoP/PhoQ transcriptional repression of Salmonella typhimurium invasion genes: evidence for a role in protein secretion. Mol. Microbiol. 17:169–181. 173. Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott. Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. J. Food Prot. 73:1919–1936. 174. Polotsky. Y., E. Dragunsky, and T. Khavkin. 1994. Morphologic evaluation of the pathogenesis of bacterial enteric infections. Crit. Rev. Microbiol. 20:161–208. 175. Pönkä, A., Y. Andersson, A. Sütonen, B. de Jong, M. Jahkola, O. Haikala, A. Kuhmonen, and P. Pakkala. 1995. Salmonella in alfalfa sprouts. Lancet 345:462–463. 176. Popoff, M. Y., and F. Norel. 1992. Bases moléculaires de la pathogénicité des Salmonella. Méd. Mal. Infect. 22:310–324. 177. Porwollik, S., E. F. Boyd, C. Choy, P. Cheng, L. Florea, E. Proctor, and M. McClelland. 2004. Characterization of Salmonella enterica subspecies I genovars by use of microarrays. J. Bacteriol. 186:5883–5898. 178. Potter, M. E., A. F. Kaufmann, P. A. Blake, and R. A. Feldman. 1984. Unpasteurized milk. The hazards of a health fetish. JAMA 252:2048–2052.

10.  Salmonella Species 179. Radford, S. A., and R. G. Board. 1995. The influence of sodium chloride and pH on the growth of Salmonella Enteritidis PT 4. Lett. Appl. Microbiol. 20:11–13. 180. Rahn, K., S. A. Grandis, R. C. Clarke, S. A. McEwen, J. E. Galan, C. Ginocchio, R. Curtiss III, and C. L. Gyles. 1992. Amplification of invA gene sequences of Salmonella Typhimurium by polymerase chain reaction as a specific method of detection of Salmonella. Mol. Cell Probes 6:271–279. 181. Randall, L. P., S. W. Cooles, M. K. Osborn, L. J. Piddock, and M. J. Woodward. 2004. Antibiotic resistance genes, integrons and multiple antibiotic resistance in thirtyfive serotypes of Salmonella enterica isolated from humans and animals in the UK. J. Antimicrob. Chemother. 53:208–216. 182. Ratman, S., F. Stratton, C. O’Keefe, A. Roberts, R. Coates, M. Yetman, S. Squires, R. Khakhria, and J. Hockin. 1999. Salmonella Enteritidis outbreak due to contaminated cheese—Newfoundland. Can. Commun. Dis. Rep. 25:17–20. 183. Reeves, M. W., G. M. Evins, A. A.Heiba, B. D. Plikaytis, and J. J. Farmer III. 1989. Clonal nature of Salmonella typhi and its genetic relatedness to other salmonellae as shown by multilocus enzyme electrophoresis, and proposal of Salmonella bongori comb. nov. J. Clin. Microbiol. 27:313–320. 184. Richard, F., E. Pons, B. Lelore, V. Bleuze, B. Grandbastien, C. Collinet, R. Mathis, J.-F. Diependale, and P. Legrand. 1994. Toxi-infection alimentaire collective du 8 juin 1993 à Douai. Bull. Epidemiol. Hebd. 3:9–11. 185. Richter-Dahlfors, A., A. M. J. Buchan, and B. B. Finlay. 1997. Murine salmonellosis studied by confocal microscopy: Salmonella Typhimurium resides intracellularly inside macrophages and exerts a cytotoxic effect on phagocytes in vivo. J. Exp. Med. 186:569–580. 186. Rodríguez-Urrego, J., S, Herrera-León, A. EcheitaSarriondia, P. Soler, F. Simon, and S. Mateo. 2010. National outbreak of Salmonella serotype Kedougou associated with infant formula, Spain, 2008. Euro Surveill. 15:pii=19582. http://www.eurosurveillance.org/View Article.aspx?ArticleId=19582. Accessed January 2012. 187. Rotger, R., and J. Casadesus. 1999. The virulence plasmids of Salmonella. Int. Microbiol. 2:177–184. 188. Rowe, B., E. J. Threlfall, L. R. Ward, and A. S. Ashley. 1979. International spread of multiresistant strains of Salmonella Typhimurium phage types 204 and 193 from Britain to Europe. Vet. Rec. 105:468–469. 189. Ryan, C. A., M. K. Nickels, N. T. Hargrett-Bean, M. E. Potter, T. Endo, L. Mayer, C. W. Langkop, C. Gibson, R. C. McDonald, R. T. Kenney, N. D. Puhr, P. J. McDonnell, R. J. Martin, M. L. Cohen, and P. A. Blake. 1987. Massive outbreak of antimicrobial-resistant salmonellosis traced to pasteurized milk. JAMA 258:3269–3274. 190. Schmidt, H., and M. Hensel. 2004. Pathogenicity islands in bacterial pathogenesis. Clin. Microbiol. Rev. 17:14–56. 191. Schraidt, O., M. D. Lefebre, M. J. Brunner, W. H. Schmied, A. Schmidt, J. Radics, K. Mechtler, J. E. Galán,

259 and T. C. Marlovits. 2010. Topology and organization of the Salmonella typhimurium type iii secretion needle complex components. PLoS Pathog. 6(4):e1000824. doi:10.1371/journal.ppat.1000824 192. Seglenieks, Z., and S. Dixon. 1977. Outbreak of milkborne Salmonella gastroenteritis—South Australia MMWR Morb. Mortal. Wkly. Rep. 26:127. 193. Sivapalasingam, S., E. Barrett, A. Kumura, S. Van Duyne, W. De Witt, M. Ying, A. Frisch, Q. Phan, E. Gould, P. Shillam, V. Reddy, T. Cooper, M. Hoekstra, C. Higgins, J. P. Sanders, R. V. Tauxe, and L. Slutsker. 2003. A multistate outbreak of Salmonella enterica Serotype Newport infection linked to mango consumption: impact of water-dip disinfestations technology. Clin. Infect. Dis. 37:1585–1590. 194. Smith, J. L. 1994. Arthritis and foodborne bacteria. J. Food Prot. 57:935–941. 195. Spika, J. S., S. H. Waterman, G. W. Soo Hoo, M. E. St.Louis, R. E. Pacer, S. M. James, M. L. Bissett, L. W. Mayer, J. Y. Chiu, B. Hall, K. Greene, M. E. Potter, M. L. Cohen, and P. A. Blake. 1987. Chloramphenicol-resistant Salmonella Newport traced through hamburger to dairy farms. N. Engl. J. Med. 316:565–570. 196. Stackhouse, R. R., N. G. Faith, C. W. Kaspar, C. J. Czuprynski, and A. C. Wong. 2012. Survival and virulence of Salmonella enterica serovar Enteritidis filaments induced by reduced water activity. Appl. Environ. Microbiol. 78:2213–2220. 197. Statens Serum Institut. 2011. Salmonella outbreak associated with imported tomatoes. Statens Serum Institut, Copenhagen, Denmark. http://www.ssi.dk/English/ News/News/2011/Salm%20imported%20tomatoes. aspx. Accessed January 2012. 198. Stokes, H. W., and R. M. Hall. 1989. A novel family of potentially mobile DNA elements encoding site-specific gene integration functions: integrons. Mol. Microbiol. 3:1669–1683. 199. Taira, S., P. Heiskanen, R. Hurme, H. Heikkilä, P. Riikonen, and M. Rhen. 1995. Evidence for functional polymorphism of the spv R gene regulating virulence gene expression in Salmonella. Mol. Gen. Genet. 246: 437–444. 200. Takkinen, J., U. M. Nakari, T. Johansson, T. Niskanen, A. Siitonen, and M. Kuusi. 2005. A nationwide outbreak of multiresistant Salmonella Typhimurium var Copenhagen DT 104B infection in Finland due to contaminated lettuce from Spain, May 2005. Eurosurveill. Rep. 10:26. 201. Tanaka, N. 1993. Food hygiene in Japan—Japanese food hygiene regulations and food poisoning incidence. Dairy Food Environ. Sanit. 13:152–156. 202. Thomson, G. T. D., D. A. DeRubeis, M. A. Hodge, C. Rajanayagam, and R. D. Inman. 1995. Post-Salmonella reactive arthritis: late clinical sequelae in point source cohort. Am. J. Med. 98:13–21. 203. Threlfall, E. J., F. J. Angulo, and P. G. Wall. 1998. Ciprofloxacin-resistant Salmonella Typhimurium DT104. Vet. Rec. 142:255.

260 204. Threlfall, E. J., L. R. Ward, J. A. Skinner, and B. Rowe. 1997. Increase in multiple antibiotic resistance in nont­yphoidal salmonellas from humans in England and Wales: a comparison of data for 1994 and 1996. Microb. Drug Resist. 3:263–266. 205. Torok, T. J., R. V. Tauxe, R. P. Wise J. R. Livengood, R. Sokolow, S. Mauvais, K. A. Birkness, M. R. Skeels, J. M. Horan, and L. R. Foster. 1997. A large community outbreak of salmonellosis caused by intentional contamination of restaurant salad bars. JAMA 278:389–395. 206. Toth, B., D. Bodager, R. M. Hammond S. Stenzel, J. K. Adams, T. Kass-Hout, R. M. Hoekstra, P. S. Mead, and P. Srikantiah. 2002. Outbreak of Salmonella serotype Javiana infections—Orlando, Florida, June 2002. MMWR Morb. Mortal. Wkly. Rep. 51:31. 207. Troller, J. A. 1986. Water relations of foodborne bacterial pathogens—an updated review. J. Food Prot. 49:656–670. 208. Tsujii, H., and K. Hamada. 1999. Outbreak of salmonellosis caused by ingestion of cuttlefish chips contaminated by both Salmonella Chester and Salmonella Oranienburg. Jpn. J. Infect. Dis. 52:138–139. 209. Uesugi, A. R., M. D. Danyluk, and L. J. Harris. 2006. Survival of Salmonella Enteritidis phage type 30 on inoculated almonds stored at -20, 4, 23, and 35 degrees C. J. Food Prot. 69:1851–1857. 210. UK Food Standards Agency. 2006. Cadbury recall update 1 August 2006. UK Food Standards Agency, London, United Kingdom. http://www.food.gov.uk/news/news archive/2006/aug/cadbury. Accessed January 2012. 211. UK Health Protection Agency. 2010. Outbreaks of Salmonella cases linked to bean sprouts continue. UK Health Protection Agency, London, United Kingdom. http://www.hpa.org.uk/NewsCentre/NationalPress Releases/2010PressReleases/101028SBareillyoutbreak continues/. Accessed January 2012. 212. UK Health Protection Agency. 2010. S. Enteritidis infections in England in 2009: national case control study report. UK Health Protection Agency, London, United Kingdom. http://www.hpa.org.uk/hpr/archives/2010/ news0610.htm#pt14b. Accessed January 2012. 213. Unicomb, L. E., G. Simmons, T. Merritt, J. Gregory, C. Nicol, P. Jelfs, M. Kirk, A. Tan, R. Thomson, J. Adamopoulos, C. L. Little, A. Currie, and C. B. Dalton. 2005. Sesame seed products contaminated with Salmonella: three outbreaks associated with tahini. Epidemiol. Infect. 133:1065–1072. 214. U.S. Centers for Disease Control and Prevention. 2008. Outbreak of multidrug-resistant Salmonella enterica serotype Newport infections associated with consumption of unpasteurized Mexican-style aged cheese—Illinois, March 2006–April 2007. MMWR Morb. Mortal. Wkly. Rep. 57:432–435. 215. U.S. Centers for Disease Control and Prevention. 2011. CDC foodborne outbreak online database (food). U.S. Centers for Disease Control and Prevention, Atlanta, GA. http://wwwn.cdc.gov/foodborneoutbreaks/. Accessed January, 2012.

Foodborne Pathogenic Bacteria 216. U.S. Centers for Disease Control and Prevention. 2011. Investigation announcement: multistate outbreak of human Salmonella Heidelberg infections. U.S. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/ salmonella/heidelberg/080111/. Accessed January 2012. 217. U.S. Centers for Disease Control and Prevention. 2011. Investigation of an outbreak of Salmonella Saintpaul infections linked to raw alfalfa sprout. U.S. Centers for Disease Control and Prevention, Atlanta, GA. http://www. cdc.gov/salmonella/saintpaul/alfalfa/. Accessed January, 2012. 218. U.S. Centers for Disease Control and Prevention. 2011. Investigation of outbreak of human infections caused by Salmonella I 4,[5],12:i:-. U.S. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc. gov/salmonella/4512eyeminus.html. Accessed January, 2012. 219. U.S. Centers for Disease Control and Prevention. 2011. Investigation update: multistate outbreak of human Salmonella I 4,[5],12:i:-infections linked to alfalfa sprouts. U.S. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/salmonella/i4512i-/021011/ index.html. Accessed January, 2012. 220. U.S. Centers for Disease Control and Prevention. 2011. Multistate outbreak of human Salmonella Agona infections linked to whole, fresh imported papayas. U.S. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/salmonella/agona-papayas/index. html. Accessed January, 2012. 221. U.S. Department of Agriculture Food Safety Inspection Service. 2010. Progress report on Salmonella testing of raw meat and poultry products, 1998-2010. USDA/FSIS, Washington, DC. 222. U.S. Food and Drug Administration. 2009. Prevention of Salmonella Enteritidis in shell eggs during production, storage, and transportation. Final rule. Fed. Reg. 74:33029–33101. 223. Van Duynhoven, Y. T. H. P., M. Widdowson, C. M. de Jager, T. Fernandes, S. Neppelenbroek, W. van den Brandhof, W. J. B. Wannet, J. A. van Kooij, H. J. M. Rietveld, and W. van Pelt. 2002. Salmonella enterica serotype Enteritidis phage type 4b outbreak associated with bean sprouts. Emerg. Infect. Dis. 8:440–443. 224. Weigel, R. M., B. Qiao, B. Teferedegne, D. K. Suh, D. A. Barber, R. E. Isaacson, and B. A. White. 2004. Comparison of pulsed field gel electrophoresis and repetitive sequence polymerase chain reaction as genotyping methods for detection of genetic diversity and inferring transmission of Salmonella. Vet. Microbiol. 100:205–217. 225. Weissman, J. B., R. M. A. D. Deen, M. Williams, N. Swanton, and S. Ali. 1977. An island-wide epidemic of salmonellosis in Trinidad traced to contaminated powdered milk. West Indies Med. J. 26:135–143. 226. Werber, D., J. Dreesman, F. Feil, U. van Treeck, G. Fell, S. Ethelberg, A. M. Hauri, P. Roggentin, R. Prager, I. S. Fisher, S. C. Behnke, E. Bartelt, E. Weise, A. Ellis, A. Siitonen, Y. Andersson, H. Tschape, M. H. Kramer, and A. Ammon. 2005. International outbreak of Salmonella Oranienburg due to German chocolate. BMC Infect. Dis. 5:7.

10.  Salmonella Species 227. White, A. P., and M. G. Surette. 2006. Comparative genetics of the rdar morphotype in Salmonella. J Bacteriol. 188:8395–8406. 228. White, D. G., S. Zhao, R. Sudler, S. Ayers, S. Friedman, S. Chen, P. F. McDermott, S. McDermott, D. D. Wagner, and J. Meng. 2001. The isolation of antibiotic-resistant Salmonella from retail ground meats. N. Engl. J. Med. 345:1147–1154. 229. Wilson, M. K., A. B. Lane, B. F. Law, W. G. Miller, L. A. Joens, M. E. Konkel, and B. A. White. 2009. Analysis of the pan genome of Campylobacter jejuni isolates recovered from poultry by pulse-field gel electrophoresis, multilocus sequence typing (MLST), and repetitive sequence polymerase chain reaction (rep-PCR) reveals different discriminatory capabilities. Microb. Ecol. 58: 843–855.

261 230. Winthrop, K. L., M. S. Palumbo, J. A. Farra, J. C. MohleBoetani, S. Abbot, M. E. Beatty, G. Inami, and S. B. Serner. 2003. Alfalfa sprouts and Salmonella Kottbus infection: a multistate outbreak following inadequate seed disinfection with heat and chlorine. J. Food Prot. 66:13–17. 231. World Health Organization. 2004. German study looks at prevalence of Salmonella in sesame seed products. WHO Newsl. No. 80, June 2004. 232. Ye, X. L., C. C. Yan, H. H. Xie, X. P. Tan, Y. Z. Wang, and L. M. Ye. 1990. An outbreak of food poisoning due to Salmonella Typhimurium in the People’s Republic of China. J. Diarrheal Dis. Res. 8:97–98. 233. Zhou, D., W. D. Hardt, and J. E. Galan. 1999. Salmonella Typhimurium encodes a putative iron transport system within the centisome 63 pathogenicity island. Infect. Immun. 67:1974–1981.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch11

Ihab Habib Lieven De Zutter Mieke Uyttendaele

11

Campylobacter Species

Campylobacter is regarded as a leading cause of bacterial foodborne infection in many areas of the world. One of the first links between Campylobacter and human diarrheal illness occurred in 1938, when Campylobacter was isolated in association with a milkborne outbreak in the United States. In the 1970s, with the development of suitable selective media, scientists were able to isolate Campylobacter spp. from blood and feces of a young (previously healthy) woman with acute febrile hemorrhagic enteritis and later from a baby with febrile diarrhea. Since then, it has been established that Campylobacter jejuni, and to a lesser extent Campylobacter coli, are important causes of human diarrheal illnesses, even surpassing Salmonella in importance in many countries (36, 88). Although human illnesses are usually self-limiting, the associated morbidity and cost are significant. Campylobacter is estimated to cause 1.4 million infections, 13,000 hospitalizations, and 100 deaths annually in the United States (65). In Europe, the burden of human campylobacteriosis is estimated to be between 8 and 100 times higher than the

annually reported number of cases, which has approximated 200,000 cases in recent years (6). Such a high incidence of Campylobacter-related diarrhea causes significant social-economic impacts, making this pathogen a priority for public health and food safety researchers. This chapter will highlight the important bacteriological and epidemiologic features related to thermotolerant Campylobacter spp. (C. jejuni and C. coli, and to a lesser extent C. lari) contamination in the human food chain. Since no single review can be completely comprehensive, the reader is encouraged to refer to several excellent recent books and reviews for additional information (36, 88, 104, 139).

THE GENUS CAMPYLOBACTER The genus Campylobacter belongs to the family Campylobacteraceae together with the genera Arcobacter and Sulfurospirillum. The genus was established in 1963, and its taxonomic structure has been ­ revised over the

Ihab Habib, Department of Food Safety and Food Quality, Ghent University, Faculty of Bioscience Engineering, Coupure Links 653, B-9000, Ghent, Belgium, and Division of Food Hygiene and Control, Alexandria University, High Institute of Public Health, El-Horrya Avenue 165, Alexandria, Egypt. Lieven De Zutter, Department of Veterinary Public Health and Food Safety, Ghent University, Faculty of Veterinary Medicine, Salisburylaan 133, B-9820 Merelbeke, Belgium. Mieke Uyttendaele, Department of Food Safety and Food Quality, Ghent University, Faculty of Bioscience Engineering, Coupure Links 653, B-9000, Ghent, Belgium.

263

SMP_Food Microbiology_CH11.indd

263

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

264 years (178, 200). At present, the genus Campylobacter contains 25 species, 4 of which have been further divided into eight subspecies (42). These species can be traced to a variety of animal and environmental sources. In veterinary medicine, C. fetus subsp. venerealis is an important cause of epizootic bovine infertility, and C. fetus subsp. fetus is most often associated with spontaneous abortion in cattle and sheep. In human oral medicine, the species C. concisus, C. curvus, C. hominis, C. sputorum, C. rec­ tus, C. showae, and C. gracilis are predominantly important (149, 200). Nonetheless, from a wider public health and food safety perspective, C. jejuni, C. coli, C. lari, and C. upsaliensis are the most important species because of their significant roles in foodborne and zoonotic illnesses. These four species are often referred to as thermotolerant campylobacters, as most strains of these species exhibit optimal growth at 41 to 43°C. Relevant biochemical properties for some of the important Campylobacter species are summarized in Table 11.1. The species C. jejuni is comprised of two subspecies: the first, C. jejuni subsp. jejuni, is one of the most common causes of human gastroenteritis worldwide (151, 200), whereas the second, C. jejuni subsp. doylei, is less well defined as a human pathogen. It is isolated infrequently from human clinical samples and has been associated with infants’ bacteremia and pediatric diarrhea in developing countries (59). The distinction between the two subspecies is rarely considered in public health and

food microbiology laboratories. However, a multiplex PCR method has been described for the differentiation between the two C. jejuni subspecies (126). For simplicity, throughout this chapter, C. jejuni will be used to refer mainly to C. jejuni subsp. jejuni. Campylobacter coli is regarded as the second most frequently isolated Campylobacter species from human diarrheal samples. It was first isolated from pigs with infectious dysentery and is still the most frequently isolated Campylobacter species from pork and swine farms (88). Nevertheless, C. coli can also be found in poultry, cattle, and dogs (149). This species is closely related to C. jejuni, having a large set of genes in common with C. jejuni. This likely explains the problematic differentiation between C. jejuni and C. coli that occurs in ­biochemical and phenotypic tests (40). To circumvent this problem, a number of PCR tests have been developed that can differentiate C. jejuni and C. coli (43, 147). Despite being closely related, recent genetic evidence indicates that C. coli strains are less diverse and more clonal than C. jejuni strains (45, 116, 127). Campylobacter lari strains have been isolated from the intestinal tracts of sea gulls, wild birds, poultry, cattle, and shellfish and from untreated water. In humans, C. lari has been occasionally isolated from cases with diarrhea. C. lari strains are primarily distinguished from C. jejuni and C. coli by their resistance to nalidixic acid (88); however, Endtz et al. (53) have described nalidixic-

Table 11.1  Biochemical and growth characteristics of Campylobacter spp.a Characteristic Growth at 25°C Growth at 35–37°C Growth at 42°C Nitrate reduction 3.5% NaCl H2S, lead acetate strip H2S, TSI Catalase Oxidase MacConkey’s agar Motility (wet mount) Growth in 1% glycine Glucose utilization Hippurate hydrolysis Resistance to naladixic acid Resistance to cephalothin

C. jejuni

C. jejuni subsp. doylei

C. coli

C. lari

C. fetus subsp. fetus

C. hyointestinalis

C. upsaliensis

− + + + − + − + + + + + − + Sc

± + ± − − + − + + + + + − + S

− + + + − + D + + + + + − − S

− + + + − + − + + + + + − − R

+ + D + − + − + + + + + − − R

D + + + − + +b + + + + + − − R

0 + + + − + − − + − + + − − S

R

R

R

R

Sd

S

S

Symbols: +, 90% or more of strains are positive; −, 90% or more of strains are negative; D, 11 to 89% of strains are positive; R, resistant; S, susceptible; TSI, triple sugar iron. b Small amount of H2S on fresh (<3 days) TSI slants. c Nalidixic acid-resistant C. jejuni strains have been reported. d Cephalothin-resistant C. fetus subsp. fetus strains have been reported. a

SMP_Food Microbiology_CH11.indd

264

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

265

susceptible variant groups. C. upsaliensis is a catalasenegative thermotolerant Campylobacter species (149, 200), and its natural hosts appear to be pet dogs and cats (212), although it has been isolated from sporadic and outbreak cases of human enteric illness (22).

BACTERIOLOGICAL ASPECTS OF CAMPYLOBACTER

Characteristics and Growth Conditions

Campylobacters are gram-negative, oxidase-positive, non-spore-forming bacteria. The name Campylobacter is derived from the Greek word “kampylos,” which means curved, reflecting to a certain extent its morphology under the microscope. Log-phase cells have a characteristic spiral, curved, or occasionally straight rod shape, with pleomorphic size ranging from 0.2 to 0.8 mm in width and 0.5 to 8 mm in length. As Campylobacter cells begin to age, they tend to be coccoid in shape. The cells are highly motile by means of a single flagellum or occasionally multiple flagella at one or both ends, which gives them a very characteristic darting, corkscrewlike motility and may serve as an important feature in pathogenesis. Campylobacters are relatively inactive biochemically, obtaining their energy from amino acids or tricarboxylic acid cycle intermediates (rather than carbohydrates) (88, 181, 200). These microaerobic organisms grow best in an atmosphere containing 5 to 10% carbon dioxide and 3 to 5% oxygen. Most Campylobacter strains grow at 37°C; C. jejuni and C. coli optimum growth temperature ranges from 42 to 45°C, and the organisms do not grow at <30°C. These two characteristics restrict the ability of campylobacters to grow outside the host, and consequently, unlike most bacterial foodborne pathogens, campylobacters are not normally capable of multiplication in food during processing or storage (88, 200). Adding to that, campylobacters are sensitive to exogenous superoxides and peroxides (which can be easily generated in growth media by, for example, exposure to light). Therefore, it is necessary to use a controlled microaerobic atmosphere, preferably with hydrogen, since some Campylobacter strains tend to grow better in the presence of hydrogen (33). Experienced researchers are generally able to recognize Campylobacter by the typical appearance of colonies as flat, glossy, effuse, and thinly spreading if the agar is moist. However, colonies are sometimes atypical, especially if the agar plates are rather dry or old. Colony morphology can vary according to the basal medium but also according to the bacterial strain and the length and temperature of incubation (Fig. 11.1).

SMP_Food Microbiology_CH11.indd

265

Responses to Unfavorable Environments

There is debate as to the extent to which Campylobacter is sensitive to unfavorable environmental stresses. Despite being widely prevalent in the human food chain and diarrheal cases, numerous well-established bacterial stress response genes are noticeably absent from the sequenced C. jejuni, C. coli, and C. lari genomes, for example, those encoding the stationary-phase and stress response sigma factor RpoS, the oxidative stress response regulatory elements SoxRS and OxyR, and osmotic shock regulators BetAB (136). Campylobacter spp. survive poorly under dry or acid conditions and in sodium chloride concentrations above 2%. Survival in foods is better at chilled (4°C) than ambient (25°C) temperatures; for instance, survival has been recorded in milk and water at 4°C after several weeks. Freezing reduces the cell numbers of Campylobacter; but even after freezing (−20°C) for several weeks, low levels of Campylobacter can be still detected on contaminated poultry (2, 88, 96). Adding to that, Campylobacter species are relatively sensitive to heat (D55, 2.5 to 6.6 min) and irradiation and can readily be inactivated during proper cooking (162, 174). Under prolonged storage and starvation, the organism changes to a viable but nonculturable (VBNC) form, which cannot be grown in routine media. Published articles provide contradictory results on the infectivity of VBNC C. jejuni. Some authors determined that VBNC C. jejuni strains were unable to infect chicks, mice, or human volunteers (17, 122, 201), whereas other investigators found that VBNC C. jejuni strains were infectious in chicks and mice (26, 100, 189). The inconsistency between these publications might be attributed to differences in challenge levels, inoculation routes, strain variations, and differences in methods used to determine the VBNC status. A recent study revealed that culturability and adhesion/invasion of C. jejuni are linearly related (207). Hence, the plate count number of viable C. jejuni cells can be a reliable surrogate for the exposure/infection risk. Valuable insights on mechanisms by which C. jejuni overcomes stressful environments are based on findings from laboratory stress models. The tolerance response of C. jejuni to several stressors such as aerobic exposure and acid stresses after mild laboratory adaptation has been reported (30, 134). Adding to that, it has been determined that challenging C. jejuni strains under aerobic conditions can enhance their acid tolerance response (135). Such crossprotection response from one stress to another has also been determined, as genes involved in heat shock response in C. jejuni NCTC 11168 were upregulated in response to acid stress (165). Additionally, it has been revealed that starved C. jejuni cells are able to withstand heat stress

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

266

Figure 11.1  Colony morphology of campylobacters on different growth media. doi:10.1128/9781555818463.ch11f1

(106). Some studies also showed that strain-to-strain differences in stress response may be a reflection of the genotypic diversity within the Campylobacter population (18, 75). A wide variety of subtyping methods have been applied to Campylobacter isolates and revealed its extensive population diversity. The genotypic diversity of Campylobacter is an important feature that contributes to its survival and prevalence in food sources and human illness. More on this topic is described in the following section.

CAMPYLOBACTER DIVERSITY AND SUBTYPING

Genome Insight over Diversity

The publication of the complete genome sequence of C. jejuni strain NCTC 11168 was crucial in understanding the molecular basis behind the genotypic diversity of this pathogen (153). This opened the opportunity for in-depth molecular investigations. For instance, comparative genomic hybridization analysis of 11 C. jejuni clinical isolates (of various heat-stable serotypes) revealed low levels of genome plasticity among these strains; approximately 84% of the 1,654 genes analyzed were common to all strains tested. These “core genes” appeared to encode housekeeping functions such as biosynthesis, metabolic, cellular, and regulatory processes. On the other hand,

SMP_Food Microbiology_CH11.indd

266

strain-specific gene differences were involved in the biosynthesis and modification of cell surface structures, including flagella, lipooligosaccharide (LOS), and capsular polysaccharide (CPS) (47). In line with this study, Leonard et al. (114) utilized C. jejuni DNA microarray technology to investigate genetic diversity among five clusters of epidemiologically related isolates. The authors observed a similar variability in genes involved in surface modification such as the LOS, capsule, and flagellin loci. Hence, the presence of several dozen phase-variable genes in surface antigen biosynthesis loci provides C. jejuni with exceptional diversity when interacting with different niches, including immune evasion in the human host and survival in other reservoirs (3). This can be further elaborated from the study of On et al. (148), in which a whole-genome microarray was used to detect C. jejuni cytolethal distending toxin (CDT) and hemolysin activities, along with survival under aerobic conditions. The strains tested were grouped into two clusters, of which one showed better viability during aerobic incubation. Sixty-seven genes were present in this cluster but were missing from the other one. Most of these variable genes were localized within the gene associated with surface structures, including flagella, LOS, and membrane transport protein. These studies provide a molecular basis for the correlation between C. jejuni genomic content, particularly in surface-coding regions, and its diversity and environmental survival potential.

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

267

Other elements that contribute to C. jejuni diversity are its natural competence and recombination ability. It has been determined that C. jejuni is naturally competent, as it can take up DNA from the environment. This leads to recombination between strains, which allows the generation of even more genetic diversity. In vitro, C. jejuni has a marked preference for DNA from C. jejuni strains, as opposed to DNA from other species (215, 219). Additionally, all sequenced C. jejuni genomes have chromosomal loci that are hot spots for intra- and intergenomic recombination (3). Frequent recombination and intrachromosomal rearrangements in C. jejuni can affect the entire chromosome as well as individual loci, such as flagellin genes (153). Another feature of the C. jejuni genome that contributes to its diversity is the presence of homopolymeric tracts (short DNA intervals in which the same base is tandemly repeated) that can provide a window to genes that slip in and out of frame (47). According to van Belkum et al. (199), such phasevariable genes can enable key determinates to switch on and off depending on the environmental selective pressure to which C. jejuni is exposed.

Overview of Selected Molecular Subtyping Methods

As discussed in the previous section, C. jejuni has considerable genetic diversity. According to Wassenaar and Newell (209), a combination of subtyping methods might be required for studying C. jejuni because diversity and recombination levels might render the use of a single typing method of limited value when applied in an epidemiologic context. Some of the early phenotypic methods (e.g., biotyping and serotyping) are still used today for subtyping of C. jejuni. However, a number of disadvantages are associated with these methods, such as lack of standardization, poor discriminatory ability, and lack of commercially available high-quality reagents (e.g., antisera) (209). Since the 1980s, subtyping of C. jejuni for public health and food safety purposes has benefited from the emergence of various genotypic typing methods. Table 11.2 highlights the main features of some of the commonly used genotypic typing methods for C. jejuni.

Contribution of Subtyping to Understanding Campylobacter Epidemiology

Over the last 2 decades, molecular subtyping of Campylobacter spp. has provided valuable clinical and epidemiologic contributions, of which some are described below.

Outbreak Investigations

The method of pulsed-field gel electrophoresis (PFGE) has been useful in including or excluding cases of cam-

SMP_Food Microbiology_CH11.indd

267

pylobacter enteritis from outbreaks (61, 112). The most extensive use of PFGE to support outbreak investigation of campylobacter enteritis has been within USA PulseNet, in which more than 2,500 profiles have been identified using standardized methods (69). Applying multilocus sequence typing (MLST) to Campylobacter isolates from several outbreaks has been less discriminatory than PFGE in identifying which cases are outbreak related and which are not (171, 124). However, a combination of MLST and highly variable gene subtyping methods should increase discrimination to enable this function, thereby providing discrimination of isolates at least equivalent to that achieved with PFGE. This approach of combining MLST with variable gene sequences typing is being developed (46).

Host Association

Recent findings from MLST-based studies have provided evidence for possible association of certain C. jejuni subtypes with specific hosts. Interestingly, the association between Campylobacter subtypes and certain hosts has shown a robustness to even large geographic distances. For example, relationships have been observed between clonal complex 61 (CC-61) and cattle, CC-257 and poultry, and CC-403 and pigs; and also, perhaps less stringently, between CC-45 and poultry and wild birds, and CC-42 and sheep (64, 88, 160, 179). Despite the fact that some C. jejuni clonal complexes show consistent associations with certain hosts, others such as CC21 are widely distributed in multiple hosts (e.g., farm animals, humans, pets, wild birds, and environmental samples).

Source Attribution

The philosophy behind the source attribution approach is that control of Campylobacter in the reservoir will prevent subsequent human exposure, regardless of the transmission route or vehicle identity. DNA-based subtyping studies, especially those considering regional and temporal aspects in the study design, provided important insight over sources of clinical strains of C. jejuni. In The Netherlands, using amplification fragment length polymorphism (AFLP) typing, Duim et al. (49) observed an overlap between the genotypes of C. jejuni from poultry flocks and those from human clinical strains within the same locality. An analogous finding was reported in a study in New Zealand, where C. jejuni PFGE subtypes of human origin matched closely with subtypes from chicken meat, compared to isolates from other veterinary and environmental sources (44). In Finland, based on PFGE typing, Hänninen et al. (80) concluded that over three subsequent years chicken meat was the most important source of human infection in the Helsinki

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

268 Table 11.2  Molecular subtyping assays for Campylobacter Method

Discriminatory power

Convenience

Practical considerations

Flagellin gene (flaA)RFLP

Almost 100%

Typeability

Fair; can be enhanced by combining patterns of restriction enzymes (e.g., DdeI with HinfI).

Quick (<24 h) and simple PCR-based (equipment becomes widely available). High-throughput analysis

PFGE

95–100%; DNA methylation and DNase activity in some isolates can limit the action of restriction enzymes.

Good, and better than flaA-RFLP Using KpnI in addition to SmaI enhances PFGE discriminatory power.

Time-consuming (3 to 5 days; however, a 24-h protocol has been published). Low-throughput analysis Need specialized and expensive (initial setup) equipment.

AFLP

Almost 100%

Single-enzyme AFLP had discriminatory ability similar to that of PFGE with SmaI. Some studies conclude that AFLP is more discriminatory than PFGE.

Depends on the protocol used: fluorescent AFLP is very demanding and expensive; single-enzyme AFLP is less expensive and less labor-intensive than PFGE. Requires 2 to 3 days.

MLST

100%

High

Automatable and high throughput. Needs highly skilled operator and sequencing facility (failure on first run is frequently encountered). Costly; however, the price is rapidly declining.

Targets one polymorphic gene; thus, the technique is widely criticized for being vulnerable to genetic instability. Of important value in outbreak investigations, and could be used to predict MLST clonal complexes. Internationally standardized protocol is available. Analysis of results is difficult and includes the use of specialized software in combination with visual examination of bands on gel. Some reports indicate that PFGE is also vulnerable to C. jejuni genetic instability, including point mutation, transformation, and plasmid acquisition or loss, which can lead to minor changes in band profile. Internationally standardized protocols are available. Interpretation of AFLP band patterns can be a complex matter; method is fully dependent on specialized software. Compared to PFGE and flaA-RFLP, the technique is the least susceptible to genetic instability. Internationally standardized protocols are not yet available. Shows good correlation with MLST identification of phylogenetic relationships. Involves the analysis of the DNA sequence for each isolate at a standardized panel of genes (seven for Campylobacter) that are chosen because they have an essential function (housekeeping genes) and are present in the vast majority of isolates. Web-based free software is available for ease of result analysis and interpretation. Freely accessible online database is available. Direct comparison between different laboratories and different isolate collections is possible, which emphasizes its epidemiological importance.

region.­ In the same study, certain genotypes (PFGE based and confirmed by AFLP) persisted throughout the 3-year-time period, suggesting that common Finnish C. jejuni genotypes from selected lineages are able to colonize both humans and chickens and that these genotypes persist from one year to another. Furthermore, findings of other DNA-based typing studies suggest a role of

SMP_Food Microbiology_CH11.indd

268

farm animals and wildlife hosts in Campylobacter infection of humans and broiler flocks (24, 67, 156). Despite the value of DNA-based subtyping studies, conclusions from these studies are limited to a crude and qualitative indication of an overlap between clinical cases and infection source strains of Campylobacter. Adding to that, lack of a unified nomenclature for

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

269

g­ enotype assignment makes it difficult to compare results from different studies. Some model-based approaches are increasingly being considered to quantify source attributions of C. jejuni clinical strains based on MLST data. For example, Wilson et al. (214) developed an evolutionary probability model to trace the sources of 1,231 human campylobacteriosis cases in England based on MLST data. The assignment probability estimated that chicken is the source of infection in most (56.5%) of the cases, followed by cattle (35.0%) and sheep (4.3%). Additionally, Mullner et al. (133) used MLST data as prior information in a Bayesian modeling approach to estimate the contribution of C. jejuni from different food animals to human illness in New Zealand. They estimated that of the 474 human cases, 379 (80%) were attributed to poultry, 48 cases (10%) to bovines, 44 cases (9%) to ovines, and 4 cases (1%) to the environment. These models highlight the fact that source contribution can vary widely depending on the geographical location and modeling approach. In addition, these models reveal that MLST provides a good degree of subtyping resolution that can be used in source attribution of campylobacter enteritis. Reservoir attribution of human campylobacter enteritis can provide guidance to risk managers and policy makers on the implementation and evaluation of control strategies for major reservoirs of Campylobacter (132).

CAMPYLOBACTER ASSOCIATION WITH FOODS Campylobacter enteritis in humans is characterized by a large number of sporadic cases rather than single-source outbreaks. Infection can be acquired by a number of routes. However, Campylobacter enteritis in humans is considered to be mainly foodborne. As a common inhabitant of the gastrointestinal tract of warm-blooded animals, Campylobacter contamination of foods can occur in different ways. Campylobacter contaminates meat when carcasses (e.g., poultry and beef) contact intestinal contents during slaughter and evisceration (31, 151). Other foods such as milk can be contaminated with Campylobacter as a result of fecal contamination on the farm. Most cases of campylobacter enteritis attributed to milk have involved unpasteurized products, because Campylobacter cannot survive proper pasteurization. Additionally, postpasteurization contamination of milk might be a source of Campylobacter contamination. For example, pecking of doorstep-delivered pasteurized milk by birds has been strongly implicated in a number of cases of campylobacter enteritis in the United Kingdom (88). However, dairy products (e.g., hard

SMP_Food Microbiology_CH11.indd

269

cheese and powdered milk) pose a limited threat due to the low resistance of Campylobacter to conditions of reduced pH and water activity. Also, eggs do not appear to be an important source of Campylobacter, despite its frequent occurrence in poultry (35). The dry surface of the egg does not favor Campylobacter survival, and egg albumen has been shown to exhibit bactericidal properties. Results of some recent surveys on the presence of thermotolerant Campylobacter strains in different food categories are presented in Table 11.3. These results reveal that much of the world’s poultry meat is contaminated with campylobacters. Evidence for poultry meat being the prime source of human campylobacter enteritis has been partially supported by indirect epidemiologic evidence, because most of the human cases are sporadic and difficult to trace to a specific source. In Belgium, poultry was withdrawn from the market during the summer of 1999 in response to dioxin contamination in the preharvest sector. Subsequently, there was a 40% reduction of human cases of campylobacter enteritis (206). Furthermore, in The Netherlands, during an avian influenza outbreak in 2003, 1,300 commercial and >17,000 noncommercial poultry flocks (more than 30 million birds) were culled. Concurrently, there was a 40% reduction in human campylobacter enteritis cases during the months of culling compared to the same period in the years before (205). Campylobacters colonize extensively the poultry cecum, large intestine, and cloaca but are generally restricted to the mucous layer in the crypts of the intestinal epithelium at these locations (14). The nature of the association between Campylobacter and poultry can be attributed to certain adaptation mechanisms. First, it has been suggested that the microaerobic nature of Campylobacter is probably a reflection of the limited oxygen concentration encountered in the avian gut (151). Second, the optimal growth temperature of the thermotolerant campylobacters (42°C) mirrors that of the avian gastrointestinal tract, which differs considerably from that of the mammalian gut (37°C) (151, 219). Third, the motility and chemotactic behavior of C. jejuni play a role in colonizing the chick gastrointestinal tract, because the organisms are preferentially attracted towards fucose components of the chick’s intestinal mucin. Fourth, campylobacters might have evolved mechanisms to overcome nutritional limitations while colonizing the avian gastrointestinal system. One of these mechanisms is the ability to acquire iron from both the avian host and the gastrointestinal flora. C. jejuni is able to utilize host-derived hemin and hemoglobin, and some C. jejuni strains have transport

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

270

Table 11.3  Examples of recent surveys on the prevalence of thermotolerant Campylobacter in different foods Food source Poultry meat

Raw milk

Red meat

Country (reference)

Sampling

Belgium (74)

Chicken meat preparations/processing Chicken carcass/retail Chicken carcass/retail Poultry meat/retail Chicken fillet/retail Chicken carcass/slaughter Chicken meat and products/retail Poultry meat/retail Poultry carcass/slaughter Broiler breast skin/slaughter Chicken products/retail Chicken carcass/retail Broiler carcass/slaughter Raw cow’s milk/bulk tank sampling Raw cow’s milk/bulk tank sampling Raw cow’s milk/bulk tank sampling Raw goat’s milk/bulk tank sampling Raw cow’s milk/bulk tank sampling Beef/retail Pork/retail Beef/retail Pork/retail Lamb and mutton/retail Beef/retail Pork/retail Lamb (liver)/retail Beef/retail Pork/retail Minced pork/retail Raw oyster/retail Raw oyster/retail Raw mussel/retail Raw oyster/coastal farms Raw mussel/coastal farms Ready-to-eat vegetables Vegetables Vegetables Ready-to-eat vegetables

England (101) Ethiopia (56) France (43) Germany (119) Iran (161) Japan (172) New Zealand (216) Poland (121) The Netherlands (198) Turkey (175) United Kingdom (123) United States (190) Ireland (213) Pakistan (90) Spain (208) Switzerland (131) United States (97) Italy (173) New Zealand (216)

South Korea (86) United Kingdom (111) United States (48, 220)

Seafood

Ireland (213) The Netherlands (53) Thailand (186)

Fresh produce

Canada (146) Ireland (213) Malaysia (28) United Kingdom (170)

systems that enable them to scavenge siderophores (high-affinity­ iron compounds) generated by other members of the gastrointestinal flora (152, 157). Campylobacter is primarily a surface contaminant and can be retained with water on the poultry skin, thereby entering openings and crevices of skin over time (29). Campylobacters that enter skin crevices and feather follicles may be difficult to remove because of capillary action or irreversible attachment to

SMP_Food Microbiology_CH11.indd

270

No. positive/no. tested (%) 315/656 (48.02) 199/241 (83.0) 160/220 (72.7) 53/70 (75.7) 87/100 (87.0) 186/336 (55.4) 110/170 (64.7) 204/230 (89.1) 464/625 (74.2) 1,447/3,680 (39.6) 106/127 (83.4) 616/877 (70.2) 1,094/4,200 (26) 1/62 (1.6) 13/127 (10.2) 35/98 (36) 0/344 (0.0) 5/248 (2.2) 20/142 (14) 6/106 (5.7) 8/230 (3.5) 21/230 (9.1) 16/231 (6.9) 3/250 (1.2) 4/250 (1.6) 70/96 (72.9) 1/182 (0.5) 3/181 (1.7) 5/348 (1.3) 3/129 (2.3) 11/41 (27) 41/59 (69) 24/30 (80) 22/30 (73.3) 0/361 (0.0) 2/279 (0.72) 28/309 (9.1) 0/3852 (0.0)

the skin tissue (94). These sites can provide a suitable microenvironment for the survival of Campylobacter in chicken skin. Moreover, some components within the surface structure of chicken skin, such as proteins, fatty acids, and oils, can enhance the survival of campylobacters by hindering the formation of ice crystals in frozen poultry meat (37, 184). The reported Campylobacter counts on fresh poultry carcasses and products (e.g., fillet) can vary from 1 to 6 log10 CFU/g,

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

271

depending on the study settings and testing methodologies (7, 16, 74, 88, 93). Hence, once poultry contaminated with Campylobacter is introduced into the kitchen, it can serve as a focal point for crosscontamination (63). In addition, irrespective of the transfer route, Campylobacter cells initially present on chicken breast fillets can be transferred efficiently to prepared foods (63, 88, 119). This efficient transfer potential emphasizes the importance of taking measures to prevent cross-contamination during food handling as long as the incidence of Campylobacter on chicken products remains high.

PATHOGENICITY AND DISEASEMEDIATING FACTORS Despite the importance of C. jejuni as a human pathogen, little is known about its pathogenicity-associated factors and their role in mediating disease. This is, in part, due to the fact that, except for the CDT and CPS, campylobacters lack homologues of virulence factors that are common to other pathogens (73). The invasion ability of C. jejuni strains is commonly used as a surrogate of virulence potential. This is supported by clinical evidence on the presence of intracellular C. jejuni in tissue biopsy specimens from patients. In addition, several tissue culture models have revealed the invasive ability of C. jejuni (72, 109), and results from many animal model studies revealed that mucosal damage is a result of invasion of C. jejuni in colonic epithelial cells (8, 60, 87, 88). However, strains of C. jejuni vary in their invasive ability (77, 143, 203), probably reflecting the involvement of multiple bacterial structures and mechanisms in this process. Some of the well-characterized factors implicated in Campylobacter’s host cell invasion and disease pathogenesis are the flagellar apparatus, a variety of genes and proteins linked to adhesion and invasion, LOS, capsule, and CDT.

Flagellar Apparatus

Flagella provide Campylobacter with a high degree of motility, which is necessary to overcome peristalsis and entry into the mucous layer. In addition, flagella and flagellar motility are vital to many aspects of C. jejuni biology, including host colonization, virulence in ferret models, and secretion and host cell invasion factors (210, 219). Konkel et al. (108) determined that C. jejuni uses its flagella as an export apparatus to secrete both flagellar and nonflagellar proteins, some of which are required for adhesion and invasion.

SMP_Food Microbiology_CH11.indd

271

Furthermore, flagella play a prime role in C. jejuni chemotactic behavior (218). It has been shown that C. jejuni might be able to exhibit chemotactic motility towards amino acids that are present at high levels in the chick gastrointestinal tract and towards components of mucus (109).

Adhesion- and Invasion-Related Genes and Proteins

The identification of several putative adhesion factors in C. jejuni was based mainly on experiments in culture cell lines. Fauchere et al. (58) identified a correlation between the severity of clinical symptoms in infected humans and the degree to which C. jejuni adheres to cultured cells. Among a variety of cell lines, human intestinal epithelial (INT407) and human colon (Caco2) cell lines were good models, mimicking conditions encountered by C. jejuni in vivo (109). Fibronectinbinding outer membrane protein CadF (110), the periplasmic binding protein PEB1 (155), and the ­surface-exposed lipoprotein J1pA (98) are examples of putative adhesion factors associated with C. jejuni. Studies demonstrating a putative role for these factors are based mainly on a significant decline in adhesion and invasion by isogenic mutant strains in which the putative virulence gene is inactivated. Protein CiaB is another putative pathogenicity factor of C. jejuni (107). It has been shown that CiaB mutants deficient in secretion of a number of bacterial proteins were less invasive than wild-type strains. It has been proposed that CiaB and other secreted proteins require a functional flagellar export apparatus for their secretion, as no type III secretion system has been found in C. jejuni (108). Comparative genomics of non-jejuni Campylobacter spp. (NJC) indicate that CadF is predicted to be encoded by all of the NJC genomes, as is the protein CiaB. However, the lipoprotein J1pA is encoded only by C. coli RM2228 and C. upsaliensis RM3195, and not by the C. lari genome. Similarly, the binding protein PEB1 is not encoded by C. lari RM2100 (125).

Plasmid pVir

Plasmids of wide ranges in size have been observed in the genomes of C. jejuni, C. coli, and C. upsalien­ sis. Plasmids are considered a primary marker for Campylobacter resistance to antibiotics (e.g., tetracycline and chloramphenicol). However, a type IV secretion apparatus was described on a large plasmid of C. jejuni 81176, known as plasmid pVir (11). The pVir plasmid was identified in 17% of 104 C. jejuni clinical

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

272 isolates studied and was significantly associated with the occurrence of blood in patient stools, a marker of invasive infection (195). Nonetheless, other studies questioned the role of pVir in invasion and enteric illnesses, as they revealed no clear association between the plasmid pVir and development of bloody diarrhea in patients infected with C. jejuni (117, 177).

C. coli strains. In one study, the quantification of active CDT levels produced by the isolates revealed that C. coli isolates produced little or no toxin (57). The limited CDT activity in C. coli strains might be a reflection of changes in critical amino acids within the CDT subunits of C. coli compared to C. jejuni.

Lipooligosaccharides and Capsular Polysaccharides

CLINICAL ASPECTS OF HUMAN ILLNESSES

Some other bacterial factors that may play a role in the invasion process of C. jejuni are the CPSs and the LOS outer core. Experiments performed with C. jejuni 81-176 CPSs determined the role of capsule in serum resistance, epithelial invasion, and diarrheal disease (10). In addition, Bachtiar et al. (9) determined that capsule-deficient C. jejuni strain 81116 exhibits reduced adherence to human intestinal epithelia, but there was no decline in colonization of the chicken gut, suggesting a possible role of CPSs in species­dependent interactions. It has been shown that mutations in several genes involved in LOS biosynthesis affect serum resistance, as well as adherence to and invasion of INT407 cells (66, 219). Recent studies have revealed that sialylation of LOS enhances C. je­ juni invasion of Caco-2 cells (77, 118), and this is also supported by clinical data that associate sialylation of C. jejuni LOS with severity and prolongation of human diarrhea (130).

Toxins

Campylobacters can produce a tripartite complex toxin designated cytolethal distending toxin, which is considered an important virulence-associated factor (99). This toxin can induce cell distension in different mammalian cell lines. Inactivating the genes of C. je­ juni 11168 encoding the toxin subunits decreased the strains’ ability to invade HeLa cells (human epithelial cells from a fatal cervical carcinoma) more than 10-fold (19). It appears that CdtA and CdtC remain at the host cell membrane while CdtB is translocated into the host cell cytoplasm and transported via the Golgi apparatus to the endoplasmic reticulum and from there reaches the nucleus (84). CdtB is the toxic component and is thought to act as DNase I-like proteins, causing DNA damage and arresting the mitosis cycle of the cell. In addition, CDT of C. jejuni may play a role in invasiveness and modulation of immune response (159). In vitro cell toxicity assays of a variety of C. jejuni and C. coli strains have revealed that most C. jejuni strains produce significantly higher CDT titers than do

SMP_Food Microbiology_CH11.indd

272

High rates in the incidence of campylobacter enteritis translate into substantial social and economic costs. Gellynck et al. (68) estimated that the cost for 2004 of campylobacter enteritis and sequelae in Belgium was €27.3 million. In The Netherlands, the incidence of campylobacter enteritis for 2000 was estimated to be 500 cases per 100,000 people, resulting in a total social cost of €21 million (202). In the United States, estimates of foodborne campylobacter enteritis in the 1990s were 2.5 million cases annually, with an estimated annual cost of €3.52 billion. Foodborne campylobacter enteritis in the United Kingdom costs at least €96 million annually (88). A careful estimation of campylobacter enteritis disease burden should consider the impact of both intestinal illness and extraintestinal illness and its complications.

Intestinal Illness

More than 90% of human cases of campylobacter enteritis cases could be attributed to C. jejuni and, to a lesser extent, C. coli (31, 151). However, enteric illnesses from C. jejuni and C. coli are clinically indistinguishable. Gastroenteritis is the most common clinical presentation seen in humans infected with C. jejuni. Following an incubation period typically in the order of 24 to 72 hours, an acute diarrheal illness develops and is commonly followed by a nonspecific syndrome of fever, chills, myalgia, and headache, in addition to abdominal cramping. Patients typically have 8 to 10 bowel movements a day at the peak of illness. The diarrhea can range from loose and watery to bloody, and leukocytes and erythrocytes are present in most cases, whether or not blood is voided in the stool (95, 158, 210). A few reports are available on the infective dose of C. jejuni, which is believed to be as low as a few hundred bacteria. In two studies of experimental infections in humans, C. jejuni caused illness with an oral dose of 500 CFU (167) and 800 CFU (20). The enteric illness was self-limiting and resolved within a week in most individuals. However, extended illness can occur, particularly in individuals with underlying illness (95, 158). In many areas of the world, human campylobacter enteritis cases are reported throughout the year but

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

273

peak during the summertime. The mechanisms behind this phenomenon are poorly understood; however, ambient temperature is hypothesized to play a role in this seasonal trend. A study in England revealed a linear relationship between mean weekly temperature and reported campylobacter enteritis cases, with a 1°C increase corresponding to a 5% increase in the number of reports up to a threshold of 14°C (193). In addition, some studies attribute seasonality of campylobacter enteritis cases to variation in human behavior during the summer months, such as increased outdoor activities (e.g., barbecuing and swimming). Seasonal variation in the occurrence of Campylobacter in poultry flocks is also suggested to play a role in the seasonality of human cases (193). The numbers of Campylobacter in the small intestines and ceca of broilers have a seasonal pattern, with a distinct seasonal peak in the summer (92, 164). Also, it is hypothesized that temporal distribution of human cases of campylobacter enteritis could be a function of the growth kinetics of one or more fly species that become more active in the summer months (51, 144). The development of protective immunity has been documented in volunteer infection studies in which individuals challenged with a particular C. jejuni strain were protected from disease (but not reinfection) when rechallenged with the same strain (20). Studies of poultry abattoir workers have revealed that they develop antibody responses during their employment. Longterm workers may asymptomatically fecally shed C. je­ juni, unlike new workers, who develop clinical signs of C. jejuni-associated gastroenteritis (27). Recently, Tam et al. (192) determined in a case-control study that the risk for illness associated with recent chicken consumption was much lower for persons who regularly ate chicken than for those who did not, suggesting that partial immunological protection against enteric illness may follow regular chicken preparation or consumption. Hence, it may be necessary to assess the relative importance of immunity and behavioral factors when determining the risk of acquiring a Campylobacter ­infection (82).

Extraintestinal Illness

Antimicrobial-Resistant Campylobacter

Antimicrobial intervention is recommended only in campylobacter enteritis cases with severe and/or prolonged enteritis, septicemia, or extraintestinal complications (1). Ternhag et al. (194) determined through a meta-analysis of clinical trials that antimicrobial therapy shortened the carrier state of Campylobacter species in the human intestinal flora. For confirmed C. jejuni infections, the macrolide antibiotic erythromycin is the first choice of treatment. Fluoroquinolones are also recommended, but not in children with campylobacter enteritis because lesions in the articular cartilage have been observed in laboratory animals exposed to ciprofloxacin (71). Many studies have revealed a higher frequency of erythromycin resistance in C. coli than in C. jejuni (1). Table 11.4  Prevalence of production of antibody to

Campylobacter jejuni in patients with GBS Study reference

Occasionally, individuals infected with C. jejuni may present with a pseudoappendicitis syndrome and do not have diarrhea. Furthermore, campylobacter enteritis may play a role in the induction of irritable bowel syndrome and inflammatory bowel disease (187). Other organ systems can also be affected following C. jejuni infection; however, the incidence of extraintesti-

SMP_Food Microbiology_CH11.indd

nal manifestations associated with C. jejuni infection is very low compared to incidence of the enteric disease (95, 182). To date, Guillain-Barré syndrome (GBS) is the most extensively studied extraintestinal manifestation associated with C. jejuni infection. GBS is the most common form of acute neuromuscular paralysis since the nearworldwide eradication of poliomyelitis, with an annual incidence of 1 to 2/100,000 population in many parts of the world (204). Symptoms generally begin with motor and sensory deficits of the lower extremities, which can subsequently spread to upper extremities and the trunk; this can lead to the need for ventilatory support. Most GBS patients recover completely, but others can be left with severe neurological impairment (113, 128). It is now known that in most cases in which a specific pathogen is identified, C. jejuni is the most frequently reported etiologic agent associated with GBS cases (Table 11.4). It has been hypothesized that patients who suffer GBS following C. jejuni infection develop antibodies against certain classes of the bacterial LOS that cross-react with peripheral nerve cell surface gangliosides (137, 210).

273

Islam et al. (91) Barzegar et al. (12) Nachamkin et al. (138) Takahashi et al. (191) Sinha et al. (180) Hadden et al. (78)

Manila Typesetting Company

No. positive/no. tested (%) 54/97 (56) 23/48 (48) 32/78 (41) 113/1,049 (11) 19/80 (26) 53/229 (23)

11/08/2012 06:55AM

Detection procedure Serology testing Serology testing Serology testing Stool culture testing Serology testing Serology testing

Foodborne Pathogenic Bacteria

274 Development of resistance to macrolides in C. jejuni and C. coli is mainly attributed to mutations at positions A2074 and A2075 of the 23S rRNA of the erythromycin binding site (70, 154). An A2075G transition is the most frequent mutation observed in clinical strains, and this can confer high values of MICs (>128 mg/ml) (70). The effects of macrolide-resistant Campylobacter on human health have been determined only in a small number of studies. For instance, a Danish cohort study estimated an increase in the risk of invasive illness and death following infection with macrolide-­resistant Campylobacter strains compared to susceptible strains (83). Development of resistance to macrolides in Campylobacter during therapy has not been documented in humans, and the origin of erythromycinresistant strains has been linked to veterinary use of the macrolide-lincosamide group (15, 79). Ciprofloxacin is widely regarded as a valuable antimicrobial therapy in severe cases of community­acquired bacterial diarrhea (1). Mutation in the gyrase gene (gyrA), most commonly at position threonine-86isoleucine (ACT®ATT), is the principal mechanism conferring quinolone resistance in C. jejuni. Luo et al. (120) determined that C. jejuni strains gain increased fitness after acquiring this resistance-conferring mutation. There is a growing concern regarding the increase in resistance of C. jejuni to ciprofloxacin and other fluoroquinolones (129, 142, 183). Several studies have revealed that infections with fluoroquinolone-­resistant Campylobacter in humans are associated with prolonged diarrhea (55, 103, 196). Hence, in areas of high endemicity for quinolone-resistant organisms, fluoroquinolones are not recommended for treating community-acquired bacterial diarrhea because campylobacters are often among the predominant causes (197). The use of fluoroquinolones (mainly enrofloxacin) in veterinary medicine is correlated with an increase in quinolone resistance in food animals, especially in poultry and, most importantly, in human Campylobacter infections (142). A temporal association between the emergence of quinolone resistance and its increase in animals and humans following the introduction of enrofloxacin in animal production has been determined by several investigators (52, 54). These findings, along with risk assessment modeling, prompted the U.S. Food and Drug Administration to propose the withdrawal of fluoroquinolone use in poultry in 2000; a lengthy legal hearing concluded with an order to withdraw enrofloxacin from use in poultry (effective in September 2005). However, it is important to note that the veterinary use of fluoroquinolones is not the only selective pressure that acts on C. jejuni to select for quinolone resistance.

SMP_Food Microbiology_CH11.indd

274

A number of studies have revealed that fluoroquinolone use in humans can lead to the emergence of quinoloneresistant campylobacters (55, 103, 211). In addition, some studies revealed significantly higher rates of quinolone resistance in travel-related Campylobacter infections than in domestically acquired infections (55).

DETECTION OF CAMPYLOBACTER IN FOODS Campylobacters can be difficult to grow because of their sensitivity to oxygen and its reactive derivatives. However, most of the selective media used today include ingredients that quench the toxic effect of oxygen derivatives on campylobacters, such as lysed blood, charcoal, hemin or hematin, and a combination of ferrous sulfate, sodium metabisulfite, and sodium pyruvate (33). For enumeration of campylobacters, direct plating on modified charcoal cefoperazone deoxycholate agar (mCCDA) is a reliable alternative to the most-probablenumber method (176). In addition, Rosenquist et al. (168) determined through a collaborative study that direct plating on mCCDA is an acceptable protocol for enumeration of thermotolerant Campylobacter on chicken meat. mCCDA is currently the medium recommended by the International Organization for Standardization (ISO) for enumeration of thermotolerant Campylobacter in food (5). Detection of Campylobacter after enrichment is recommended for recovery of a low number of sublethally damaged cells in food samples. The formulation of enrichment broth media has been modified to avoid inhibitory effects on Campylobacter by components in the broth formula. Corry et al. (33) reviewed the historical developments in culture-based procedures for detection of campylobacters. In 2006, the ISO method for detection of thermotolerant Campylobacter in food (Fig. 11.2) recommended the use of Bolton broth (BB) medium, instead of Preston broth, for selective enrichment. Enrichment procedures are recommended to begin at 37°C for 4 h followed by 44 h at 41.5°C. Enrichment cultures are subsequently plated on mCCDA and a second solid medium left to the choice of the researcher (4). BB is also now recommended by the U.S. Food and Drug Administration (89). There is debate regarding the efficiency of methods for isolating Campylobacter from foods and water. It is likely that no single method is ideal for the entire range of foods to be tested. In addition, it is important to understand the test characteristics (sensitivity and specificity), detection limit, and limitations of the method used. mCCDA, the most widely used selective medium for Campylobacter isolation from food, has been evaluated

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

275

Enrichment A: for food with low background count of non-campylobacters and/or with stressed campylobacters (e.g. frozen products) Test portion (x g or x ml) + (9x g or 9x ml) Bolton broth

Enrichment B(1): for food with high background count of non-campylobacters (e.g. raw chicken, raw meat and raw milk) Test portion (x g or x ml) + (9x g or 9x ml) Preston broth

4-6 h at 37°C + 40-48 h at 41.5°C in microaerobic atmosphere

24 h at 41.5°C in microaerobic atmosphere

Isolation of selective medium Plate on mCCDA and another selective medium (The second medium should be based on a principle different from mCCDA) 40-48 h at 41.5°C in microaerobic atmosphere Pick up to five presumptive Campylobacter colonies from each agar plate and subculture on Blood agar 24-48 h at 41.5°C in microaerobic atmosphere Confirmation(2)

Microscopic examination Morphology + motility

Biochemical tests Growth at 25°C in microaerobic atmosphere, growth at 41.5°C in aerobic atmosphere and oxidase activity

Identification (optional)(3) Catalase activity, sensitivity to nalidixic acid and to cephalothin, hippurate hydrolysis, indoxyl acetate hydrolysis

1

Modification proposed by the ad Hoc group Campylobacter for revision of ISO 10272-1/2. Alternative confirmation using PCR and immunological tests is more convenient and became widely accepted for Campylobacter. 3 Antibiotic sensitivity tests for nalidixic acid and to cephalothin can be omitted. 2

Figure 11.2  Methods for detecting and isolating thermotolerant Campylobacter in food. doi:10.1128/9781555818463.ch11f2

in a number of studies. De Boer et al. (38) determined that mCCDA and Karmali agars performed best for isolation of campylobacters from chicken meat. For testing rinsates of naturally contaminated chicken carcass, Stern and Line (188) obtained a higher Campylobacter isolation rate when using mCCDA than when using Campy-

SMP_Food Microbiology_CH11.indd

275

Cefex agar. Oyarzabal et al. (150) determined that mCCDA and modified Campy-Cefex agar yielded comparable results for enumerating Campylobacter spp. from poultry carcass rinses. However, limited ­ comparisons have been made to verify the performance of BB versus other enrichment media. The most extensive study was

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

276 conducted by Baylis et al. (13), comparing BB, Preston broth, and Campylobacter enrichment broth. This study revealed that BB was the best compromise between recovery of Campylobacter from naturally contaminated food samples and inhibition of competitors. However, the same study also revealed that Escherichia coli and Pseudomonas spp. were frequently isolated from BB. The growth of these competing bacteria, among other factors, can hinder the isolation of Campylobacter. In another study, Kim et al. (105) evaluated the performance of eight enrichment broths for isolating C. jejuni from inoculated suspensions and ground pork. They determined that Hunt broth and BB had the highest efficacy and most rapid enrichment time for the cell suspensions and ground pork, respectively. Hence, use of the appropriate broth is important for the rapid and efficacious enrichment culture of C. jejuni, especially with heat-injured cells, which require a longer cultivation time and a suitable, non­inhibitory enrichment broth. The current routine and standard culture detection method is not a “gold standard” method per se, as its specificity and sensitivity are far from optimum. There are considerable false-negative findings associated with campylobacter testing of food. This issue has been highlighted in a recent study conducted in the United States based on the analysis of Campylobacter proficiency testing data. This study determined that the rate of false negatives for Campylobacter results from food samples was 13.6%, and specifically for C. coli the false-negative rate was 24.0% (50). This rate of false-negative results is considerably higher than the 5.9% rate for Salmonella spp. and 7.2% rate for Listeria monocytogenes. Hence, there is a need to improve the performance of detection methods for Campylobacter in foods. A new generation of Campylobacter isolation media was recently introduced (e.g., CampyFood agar from bioMérieux SA, France; and Brilliance CampyCount agar from Oxoid, England). These media are chromogenically based and have a transparent surface to provide easy and precise enumeration. The performances of these media are similar to that of mCCDA for enumerating Campylobacter on chicken meat (76) (Table 11.5); however, more studies are needed to evaluate the performances of these new-generation media. In recent years, alternative methods have been developed for detection of Campylobacter spp. in foods. Conventional PCR is now routinely used for the detection and identification of Campylobacter, and numerous methods with various levels of accuracy and sensitivity, depending on the specificity of the targeted regions of the genome, have been published (41, 147). For detection­ of the genus Campylobacter, a highly conserved region of the 16S rRNA is the target for the PCR, whereas more specific

SMP_Food Microbiology_CH11.indd

276

Table 11.5  Results of new selective-differential media for

enumeration of campylobacters on chicken meata Medium

Countable results

Meanb (log10 CFU/g) ± SD

mCCDA CFA BCC

15/49 17/49 14/49

1.92 ± 0.77 1.92 ± 0.88 2.15 ± 0.69

a Abbreviations: mCCDA, modified charcoal cefoperazone deoxycholate agar; CFA, CampyFood agar; BCC, Brilliance CampyCount. Data are from reference 76. b Means and SD were calculated for countable results.

loci are used for the detection of particular Campylobacter species (39, 163, 217). Real-time (RT)-PCR has been developed for qualitative and quantitative detection of campylobacters in food samples, and some of these methods have been commercialized. The quantitative dimension of RT-PCR is based on correlating the number of cycles required to reach the threshold value with the initial number of campylobacters in the sample. However, the threshold limit of quantification of these RT-PCR methods ranges from 2 to 3 log CFU/g, which could result in falsenegative results (21). Another drawback of PCR-based detection methods is that many research groups have developed specific protocols that work well in their own laboratory but do not allow for comparison of results between laboratories using different protocols. Hence, interlaboratory proficiency tests, collaborative trials, and standardized protocols are much needed for these alternative methods. In contrast to classical culture methods that detect living bacteria that can grow, genome-based methods detect DNA from live and dead bacteria. This issue might present interpretation difficulties when used for analysis of food and environmental samples. It has been shown that a propidium monoazide sample treatment step prior to RT-PCR could ensure the quantification of only viable cells with intact membranes. Propidium monoazide can intercalate into the double-helical DNA available from dead cells with compromised membranes, and upon extensive visible light exposure, cross-linking of the two strands of DNA occurs, leaving it unavailable for PCR amplification (102). However, this limitation of molecular methods is not restricted to campylobacters and is also relevant for other microorganisms.

CAMPYLOBACTER CONTROL IN THE FOOD CHAIN As described previously, direct and indirect epidemiologic data indicate that poultry meat is the most important­ source of human cases of campylobacter enteritis. This evidence provides the justification to focus control measures primarily along the poultry meat chain. Poultry-

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

277

related interventions at primary production have been proposed; aspects related to enhancing biosecurity at the farm level can be accomplished at relatively low cost (e.g., hand washing, fly screens, pest-proof buildings, and foot dipping). However, applying biosecurity interventions at poultry production sites has resulted in different levels of success in different countries (34, 62). Such variation may be attributed to differences in the Campylobacter loads in the poultry chain and environment. Hence, the effectiveness of biosecurity-related interventions in primary production should be based on a good understanding of the regional risk factors at the farm level. In addition, there are several intervention options for feed and water additives (e.g., organic acids, probiotics, short- and medium-chain fatty acids) (115). However, the effectiveness of feed and water additives is variable for different products and is often difficult to reproduce, even in slightly different settings. The integration of multiple hurdles and management options at the farm level is necessary. Additional measures to be considered in the longer term include vaccination, phage therapy, and bacteriocins (25, 34, 115, 141). However, considerable additional research on these strategies is needed to ensure the practicality, reproducibility, and efficacy of such approaches under field conditions. In addition to primary interventions at the farm level, there is a need to apply interventions at the processing level in order to reduce contamination in products meant for human consumption. Highly contaminated samples have been associated with a higher probability of causing human illness (119, 140). Rosenquist et al. (169), using a risk assessment model, estimated that a 2-log reduction of Campylobacter counts on broiler carcasses would result in a significant reduction in cases of human campylobacter enteritis associated with exposure to broiler meat. Freezing of contaminated poultry carcasses is a reliable intervention to achieve a 2-log reduction of Campylobacter counts. Compulsory freezing of processed broilers from Campylobacter-positive broiler flocks in Iceland resulted in substantially reducing the number of human cases of campylobacter enteritis and is currently being used on a voluntary basis in Norway, Sweden, and Denmark (81, 85, 169). However, worldwide, many consumers prefer to buy fresh poultry meat with no change in product quality. In addition, freezing meat from all Campylobacter-positive broiler flocks might not be a feasible option in many countries, as it would limit the marketing of domestically produced chilled meat and increase dependence on imported product. This dilemma highlights the need to apply multiple hurdles during postharvest in order to achieve low counts of Campylobacter on chicken meat (162, 174).

SMP_Food Microbiology_CH11.indd

277

Some alternative physical decontamination technologies may also achieve a reduction in Campylobacter numbers that is comparable to that obtained with freezing. For example, Corry et al. (32) determined that crust freezing of chicken carcasses (based on rapid ice crystallization on the “meat surface” that results in a thin frozen crust, followed by temperature equalization) could reduce Campylobacter numbers by ³2 log CFU. However, Boysen and Rosenquist (23) reported that crust freezing of broiler carcasses provided only a 0.42-log CFU reduction in Campylobacter counts. Hence, the application of crust freezing needs to be optimized before it is widely adopted as a Campylobacter intervention. Another temperature-related intervention is the application of a steam-ultrasound treatment. Recent studies in Denmark revealed that this technology could reduce Campylobacter counts by ³2.5 log CFU on broiler carcasses (23). However, treated carcasses had the appearance of being slightly boiled (23). Chemical decontamination can also be an effective intervention for reducing the microbial load on carcasses. Chlorine, chlorine dioxide, acidified sodium chlorite, trisodium phosphate, and peroxyacid are typically used in poultry processing in the United States either as sprays or washes for online reprocessing or added to the chill water tank to reduce microbial contamination and to limit the potential for microbial cross-contamination. Trisodium phosphate solutions of 8 to 12% can reduce Campylobacter counts on chicken carcasses by 1.0 to 2.0 log CFU (166, 185). Treatment of chicken carcasses with chlorine compounds has also been extensively studied but with varying results, depending on the compound and treatment regime used in the processing plant. The use of electrolyzed water, of which hypochlorous acid is the principal active antimicrobial agent, has shown some degree of promise under experimental conditions in reducing numbers of Campylobacter on broiler carcasses but needs additional evaluation under processing facility conditions (68, 145). Lactic acid (2.5%) has been highlighted as a cost-effective intervention strategy in a Dutch risk assessment study (140). However, lower concentrations might be required, as the use of 2.5% lactic acid causes a yellow discoloration of the skin of chicken carcasses. Detailed research is still needed on appropriate treatment time and temperature and the effects of the food matrix on the antimicrobial activity of chemicals. In addition, more research may be needed on the toxicological, environmental, and food sensory aspects of chemical applications to carcasses. Results of current risk assessment models are in agreement in showing that reducing the numbers of Campylobacter on broiler meat is highly effective in

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

278 reducing the burden of illness (140). Because campylobacter enteritis is a leading foodborne bacterial infection in many parts of the world, there is a need for setting targets (e.g., process hygiene criteria) for Campylobacter in the broiler meat chain. The setting of such targets would ideally be based on an associated risk reduction; however, it is seemingly not possible to consider a zero-tolerance policy or complete elimination of risk of campylobacter enteritis with respect to consumption of broiler meat.

CONCLUSIONS Campylobacter species, and notably C. jejuni, are among the most common causes of human bacterial enteric illnesses worldwide. In the last decade, and after completion of the genome-sequencing project for multiple Campylobacter strains, more fundamental information became known regarding the diversity and pathogenesis of these intriguing bacteria. In addition, the application of discriminatory molecular subtyping tools (e.g., MLST and PFGE) was useful in clarifying some aspects of host association and source attribution. Evidence from epidemiologic studies and molecular subtyping investigations has identified poultry meat as a major vehicle for foodborne transmission of campylobacter enteritis. Future research and scientific collaborations among the medical, food, and veterinary professions are needed to substantially reduce Campylobacter contamination in the poultry meat chain. Research directions should focus on practical control options that would be appealing to stakeholders in the farm, slaughterhouse, and processing sectors. In addition, there are opportunities for the development of enhanced Campylobacter detection and quantification methods. Methods able to identify highly contaminated samples through online detection would be very useful, as this could help in identifying and excluding highly contaminated samples from the human food chain.

References 1. Alfredson, D. A., and V. Korolik. 2007. Antibiotic resistance and resistance mechanisms in Campylobacter jejuni and Campylobacter coli. FEMS Microbiol. Lett. 277:123–132. 2. Alter, T., and K. Scherer. 2006. Stress response of Campylobacter spp. and its role in food processing. J. Vet. Med. B 53:351–357. 3. Ambur, O. H., T. Davidsen, S. A. Frye, S. V. Balasingham, K. Lagesen, T. Rognes, and T. Tonjum. 2009. Genome dynamics in major bacterial pathogens. FEMS Microbiol. Rev. 33:453–470.

SMP_Food Microbiology_CH11.indd

278

4. Anonymous. 2006. ISO 10272. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for Detection and Enumeration of Campylobacter spp.— Part 1: Detection Method. International Organization for Standardization, Geneva, Switzerland. 5. Anonymous. 2006. ISO 10272. Microbiology of Food and Animal Feeding Stuffs—Horizontal Method for Detection and Enumeration of Campylobacter spp.— Part 2: Enumeration Method. International Organi­ zation for Standardization, Geneva, Switzerland. 6. Anonymous. 2009. The community summary report on trends and sources of zoonoses, zoonotic agents, anti­ microbial resistance and foodborne outbreaks in the European Union in 2007. EFSA J. 223. 7. Atanassova, V., F. Reich, L. Beckmann, and G. Klein. 2007. Prevalence of Campylobacter spp. in turkey meat from a slaughterhouse and in turkey meat retail products. FEMS Immunol. Med. Microbiol. 49:141–145. 8. Babakhani, F. K., G. A. Bradley, and L. A. Joens. 1993. Newborn piglet model for campylobacteriosis. Infect. Immun. 61:3466–3475. 9. Bachtiar, B. M., P. J. Coloe, and B. N. Fry. 2007. Knockout mutagenesis of the kpsE gene of Campylobacter jejuni 81116 and its involvement in bacterium-host interactions. FEMS Immunol. Med. Microbiol. 49:149–154. 10. Bacon, D. J., C. M. Szymanski, D. H. Burr, R. P. Silver, R. A. Alm, and P. Guerry. 2001. A phase-variable capsule is involved in virulence of Campylobacter jejuni 81-176. Mol. Microbiol. 40:769–777. 11. Bacon, D. J., R. A. Alm, L. Hu, T. E. Hickey, C. P. Ewing, R. A. Batchelor, T. J. Trust, and P. Guerry. 2002. DNA sequence and mutational analyses of the pVir plasmid of Campylobacter jejuni 81-176. Infect. Immun. 70:6242–6250. 12. Barzegar, M., A. Alizadeh, V. Toopchizadeh, S. Dastgiri, and J. Majidi. 2008. Association of Campylobacter jejuni infection and Guillain-Barre syndrome: a cohort study in the northwest of Iran. Turk. J. Pediatr. 50:443–448. 13. Baylis, C. L., S. MacPhee, K. W. Martin, T. J. Humphrey, and R. P. Betts. 2000. Comparison of three enrichment media for the isolation of Campylobacter spp. from foods. J. Appl. Microbiol. 89:884–891. 14. Beery, J. T., M. B. Hugdahl, and M. P. Doyle. 1988. Colonization of gastrointestinal tracts of chicks by Campylobacter jejuni. Appl. Environ. Microbiol. 54:2365–2370. 15. Belanger, A. E., and T. R. Shryock. 2007. Macrolideresistant Campylobacter: the meat of the matter. J. Antimicrob. Chemother. 60:715–723. 16. Berndtson, E., M. Tivemo, and A. Engvall. 1992. Distribution and numbers of Campylobacter in newly slaughtered broiler chickens and hens. Int. J. Food Microbiol. 15:45–50. 17. Beumer, R. R., J. de Vries, and F. M. Rombouts. 1992. Campylobacter jejuni non-culturable coccoid cells. Int. J. Food Microbiol. 15:153–163.

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

279

18. Bhaduri, S., and B. Cottrell. 2004. Survival of coldstressed Campylobacter jejuni on ground chicken and chicken skin during frozen storage. Appl. Environ. Microbiol. 70:7103–7109. 19. Biswas, D., U. M. Fernando, C. D. Reiman, P. J. Willson, H. G. Townsend, A. A. Potter, and B. J. Allan. 2007. Correlation between in vitro secretion of virulenceassociated proteins of Campylobacter jejuni and colonization of chickens. Curr. Microbiol. 54:207–212. 20. Black, R. E., M. M. Levine, M. L. Clements, T. P. Hughes, and M. J. Blaser. 1988. Experimental Campylobacter jejuni infection in humans. J. Infect. Dis. 157:472–479. 21. Botteldoorn, N., E. Van Coillie, V. Piessens, G. Rasschaert, L. Debruyne, M. Heyndrickx, L. Herman, and W. Messens. 2008. Quantification of Campylo­ bacter spp. in chicken carcass rinse by real-time PCR. J. Appl. Microbiol. 105:1909–1918. 22. Bourke, B., V. L. Chan, and P. Sherman. 1998. Campylobacter upsaliensis: waiting in the wings. Clin. Microbiol. Rev. 11:440–449. 23. Boysen, L., and H. Rosenquist. 2009. Reduction of thermotolerant Campylobacter species on broiler carcasses following physical decontamination at slaughter. J. Food Prot. 72:497–502. 24. Broman, T., J. Waldenstrom, D. Dahlgren, I. Carlsson, I. Eliasson, and B. Olsen. 2004. Diversities and similarities in PFGE profiles of Campylobacter jejuni isolated from migrating birds and humans. J. Appl. Microbiol. 96:834–843. 25. Buckley, A. M., J. Wang, D. L. Hudson, A. J. Grant, M. A. Jones, D. J. Maskell, and M. P. Stevens. 2009. Evaluation of live-attenuated Salmonella vaccines expressing Campylobacter antigens for control of C. je­ juni in poultry. Vaccine 28:1094–1105. 26. Cappelier, J. M., J. Minet, C. Magras, R. R. Colwell, and M. Federighi. 1999. Recovery in embryonated eggs of viable but nonculturable Campylobacter jejuni cells and maintenance of ability to adhere to HeLa cells after resuscitation. Appl. Environ. Microbiol. 65:5154–5157. 27. Cawthraw, S. A., L. Lind, B. Kaijser, and D. G. Newell. 2000. Antibodies, directed towards Campylobacter je­ juni antigens, in sera from poultry abattoir workers. Clin. Exp. Immunol. 122:55–60. 28. Chai, L. C., T. Robin, U. M. Ragavan, J. W. Gunsalam, F. A. Bakar, F. M. Ghazali, S. Radu, and M. P. Kumar. 2007. Thermophilic Campylobacter spp. in salad vegetables in Malaysia. Int. J. Food Microbiol. 117:106–111. 29. Chantarapanont, W., M. Berrang, and J. F. Frank. 2003. Direct microscopic observation and viability determination of Campylobacter jejuni on chicken skin. J. Food Prot. 66:2222–2230. 30. Chynoweth, R. W., J. A. Hudson, and K. Thom. 1998. Aerobic growth and survival of Campylobacter jejuni in food and stream water. Lett. Appl. Microbiol. 27:341–344.

SMP_Food Microbiology_CH11.indd

279

31. Corry, J. E., and H. I. Atabay. 2001. Poultry as a source of Campylobacter and related organisms. Symp. Ser. Soc. Appl. Microbiol. 2001(30):96S–114S. 32. Corry, J. E., C. James, D. O’Neill, H. Yaman, and A. Kendall. 2003. Physical methods, readily adapted to existing commercial processing plants, for reducing numbers of campylobacters, on raw poultry. Int. J. Med. Microbiol. 293:S32. 33. Corry, J. E., D. E. Post, P. Colin, and M. J. Laisney. 1995. Culture media for the isolation of Campylobacter. Int. J. Food Microbiol. 26:43–76. 34. Cox, J. M., and A. Pavic. 2009. Advances in enteropathogen control in poultry production. J. Appl. Microbiol. 108:745–755. 35. Cox, N. A., L. J. Richardson, R. J. Buhr, and P. J. Fedorka-Cray. 2009. Campylobacter species occurrence within internal organs and tissues of commercial caged Leghorn laying hens. Poult. Sci. 88:2449–2456. 36. Dasti, J. I., A. M. Tareen, R. Lugert, A. E. Zautner, and U. Gross. 2009. Campylobacter jejuni: a brief overview on pathogenicity-associated factors and disease-mediating mechanisms. Int. J. Med. Microbiol. 300:205–211. 37. Davis, M. A., and D. E. Conner. 2007. Survival of Campylobacter jejuni on poultry skin and meat at varying temperatures. Poult. Sci. 86:765–777. 38. De Boer, E., P. van Beek, and K. Pelgrom. 1998. Comparison of culture media for the isolation of campylobacters from chicken meat, p. 370–372. In A. J. Lastovica, D. G. Newell, and E. E. Lastovica (ed.), Proceedings of the 9th International Workshop on Campylobacter, Helicobacter and Related Organisms, 1997. University of Cape Town, Cape Town, South Africa. 39. Debretsion, A., T. Habtemariam, S. Wilson, D. Nganwa, and T. Yehualaeshet. 2007. Real-time PCR assay for rapid detection and quantification of Campylobacter jejuni on chicken rinses from poultry processing plant. Mol. Cell. Probes 21:177–181. 40. Debruyne, L., D. Gevers, and P. Vandamme. 2008. Taxonomy of the Family Campylobacteraceae, p. 3–25. In I. Nachamkin, C. M. Szymanski, and M. J. Blaser (ed.), Campylobacter, 3rd ed. ASM Press, Washington, DC. 41. Debruyne, L., E. Samyn, E. De Brandt, O. Vandenberg, M. Heyndrickx, and P. Vandamme. 2008. Comparative performance of different PCR assays for the identification of Campylobacter jejuni and Campylobacter coli. Res. Microbiol. 159:88–93. 42. Debruyne, L., S. L. On, E. De Brandt, and P. Vandamme. 2009. Novel Campylobacter lari-like bacteria from humans and molluscs: description of Campylobacter pe­ loridis sp. nov., Campylobacter lari subsp. concheus subsp. nov. and Campylobacter lari subsp. lari subsp. nov. Int. J. Syst. Evol. Microbiol. 59:1126–1132. 43. Denis, M., J. Refregier-Petton, M. J. Laisney, G. Ermel, and G. Salvat. 2001. Campylobacter contamination in French chicken production from farm to consumers. Use of a PCR assay for detection and identification

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

280

44.

45.

46.

47.

48.

49.

50.

51. 52.

53.

54.

55.

of Campylobacter jejuni and Camp. coli. J. Appl. Microbiol. 91:255–267. Devane, M. L., C. Nicol, A. Ball, J. D. Klena, P. Scholes, J. A. Hudson, M. G. Baker, B. J. Gilpin, N. Garrett, and M. G. Savill. 2005. The occurrence of Campylobacter subtypes in environmental reservoirs and potential transmission routes. J. Appl. Microbiol. 98:980–990. Dingle, K. E., F. M. Colles, D. Falush, and M. C. Maiden. 2005. Sequence typing and comparison of population biology of Campylobacter coli and Campylobacter je­ juni. J. Clin. Microbiol. 43:340–347. Dingle, K. E., N. D. McCarthy, A. J. Cody, T. E. Peto, and M. C. Maiden. 2008. Extended sequence typing of Campylobacter spp., United Kingdom. Emerg. Infect. Dis. 14:1620–1622. Dorrell, N., J. A. Mangan, K. G. Laing, J. Hinds, D. Linton, H. Al Ghusein, B. G. Barrell, J. Parkhill, N. G. Stoker, A. V. Karlyshev, P. D. Butcher, and B. W. Wren. 2001. Whole genome comparison of Campylobacter jejuni human isolates using a low-cost microarray reveals extensive genetic diversity. Genome Res. 11:1706–1715. Duffy, E. A., K. E. Belk, J. N. Sofos, G. R. Bellinger, A. Pape, and G. C. Smith. 2001. Extent of microbial contamination in United States pork retail products. J. Food Prot. 64:172–178. Duim, B., T. M. Wassenaar, A. Rigter, and J. Wagenaar. 1999. High-resolution genotyping of Campylobacter strains isolated from poultry and humans with amplified fragment length polymorphism fingerprinting. Appl. Environ. Microbiol. 65:2369–2375. Edson, D. C., S. Empson, and L. D. Massey. 2009. Pathogen detection in food microbiology laboratories: an analysis of qualitative proficiency test data, 1999– 2007. J. Food Saf. 29:521–530. Ekdahl, K., B. Normann, and Y. Andersson. 2005. Could flies explain the elusive epidemiology of campylobacteriosis? BMC Infect. Dis. 5:11. Endtz, H. P., G. J. Ruijs, B. van Klingeren, W. H. Jansen, T. van der Reyden, and R. P. Mouton. 1991. Quinolone resistance in Campylobacter isolated from man and poultry following the introduction of fluoro­ quinolones in veterinary medicine. J. Antimicrob. Chemother. 27:199–208. Endtz, H. P., J. S. Vliegenthart, P. Vandamme, H. W. Weverink, N. P. van den Braak, H. A. Verbrugh, and A. van Belkum. 1997. Genotypic diversity of Campylobacter lari isolated from mussels and oysters in The Netherlands. Int. J. Food Microbiol. 34:79–88. Engberg, J., F. M. Aarestrup, D. E. Taylor, P. GernerSmidt, and I. Nachamkin. 2001. Quinolone and macrolide resistance in Campylobacter jejuni and C. coli: resistance mechanisms and trends in human isolates. Emerg. Infect. Dis. 7:24–34. Engberg, J., J. Neimann, E. M. Nielsen, F. M. Aerestrup, and V. Fussing. 2004. Quinolone-resistant Campylobacter infections: risk factors and clinical consequences. Emerg. Infect. Dis. 10:1056–1063.

SMP_Food Microbiology_CH11.indd

280

56. Ewnetu, D., and A. Mihret. 2010. Prevalence and antimicrobial resistance of Campylobacter isolates from humans and chickens in Bahir Dar, Ethiopia. Foodborne Pathog. Dis. 7:667–670. 57. Eyigor, A., K. A. Dawson, B. E. Langlois, and C. L. Pickett. 1999. Detection of cytolethal distending toxin activity and cdt genes in Campylobacter spp. isolated from chicken carcasses. Appl. Environ. Microbiol. 65:1501–1505. 58. Fauchere, J. L., A. Rosenau, M. Veron, E. N. Moyen, S. Richard, and A. Pfister. 1986. Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infect. Immun. 54:283–287. 59. Fernandez, H., N. U. Fagundes, and S. Ogatha. 1997. Acute diarrhea associated with Campylobacter jejuni subsp. doylei in Sao Paulo, Brazil. Pediatr. Infect. Dis. J. 16:1098–1099. 60. Field, L. H., V. L. Headley, J. L. Underwood, S. M. Payne, and L. J. Berry. 1986. The chicken embryo as a model for campylobacter invasion: comparative virulence of human isolates of Campylobacter jejuni and Campylobacter coli. Infect. Immun. 54:118–125. 61. Fitzgerald, C., L. O. Helsel, M. A. Nicholson, S. J. Olsen, D. L. Swerdlow, R. Flahart, J. Sexton, and P. I. Fields. 2001. Evaluation of methods for subtyping Campylobacter jejuni during an outbreak involving a food handler. J. Clin. Microbiol. 39:2386–2390. 62. Fraser, R. W., N. T. Williams, L. F. Powell, and A. J. Cook. 2009. Reducing Campylobacter and Salmonella infection: two studies of the economic cost and attitude to adoption of on-farm biosecurity measures. Zoonoses Public Health 57:e109–e115. 63. Fravalo, P., M. J. Laisney, M. O. Gillard, G. Salvat, and M. Chemaly. 2009. Campylobacter transfer from naturally contaminated chicken thighs to cutting boards is inversely related to initial load. J. Food Prot. 72:1836–1840. 64. French, N. P., A. Midwinter, B. Holland, J. CollinsEmerson, R. Pattison, F. Colles, and P. Carter. 2009. Molecular epidemiology of Campylobacter jejuni isolates from wild-bird fecal material in children’s playgrounds. Appl. Environ. Microbiol. 75:779–783. 65. Friedman, C. R., R. M. Hoekstra, M. Samuel, R. Marcus, J. Bender, B. Shiferaw, S. Reddy, S. D. Ahuja, D. L. Helfrick, F. Hardnett, M. Carter, B. Anderson, and R. V. Tauxe. 2004. Risk factors for sporadic Campylobacter infection in the United States: a casecontrol study in FoodNet sites. Clin. Infect. Dis. 38(Suppl. 3):S285–S296. 66. Fry, B. N., S. Feng, Y. Y. Chen, D. G. Newell, P. J. Coloe, and V. Korolik. 2000. The galE gene of Campylobacter jejuni is involved in lipopolysaccharide synthesis and virulence. Infect. Immun. 68:2594–2601. 67. Garrett, N., M. L. Devane, J. A. Hudson, C. Nicol, A. Ball, J. D. Klena, P. Scholes, M. G. Baker, B. J. Gilpin, and M. G. Savill. 2007. Statistical comparison of Campylobacter jejuni subtypes from human cases and environmental sources. J. Appl. Microbiol. 103:2113–2121.

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

281

68. Gellynck, X., W. Messens, D. Halet, K. Grijspeerdt, E. Hartnett, and J. Viaene. 2008. Economics of reducing Campylobacter at different levels within the Belgian poultry meat chain. J. Food Prot. 71:479–485. 69. Gerner-Smidt, P., K. Hise, J. Kincaid, S. Hunter, S. Rolando, E. Hyytia-Trees, E. M. Ribot, and B. Swaminathan. 2006. PulseNet USA: a five-year update. Foodborne Pathog. Dis. 3:9–19. 70. Gibreel, A., V. N. Kos, M. Keelan, C. A. Trieber, S. Levesque, S. Michaud, and D. E. Taylor. 2005. Macrolide resistance in Campylobacter jejuni and Campylobacter coli: molecular mechanism and stability of the resistance phenotype. Antimicrob. Agents Chemother. 49:2753–2759. 71. Goodman, L. J., G. M. Trenholme, R. L. Kaplan, J. Segreti, D. Hines, R. Petrak, J. A. Nelson, K. W. Mayer, W. Landau, and G. W. Parkhurst. 1990. Empiric antimicrobial therapy of domestically acquired acute diarrhea in urban adults. Arch. Intern. Med. 150:541–546. 72. Grant, C. C., M. E. Konkel, W. Cieplak, Jr., and L. S. Tompkins. 1993. Role of flagella in adherence, internalization, and translocation of Campylobacter jejuni in nonpolarized and polarized epithelial cell cultures. Infect. Immun. 61:1764–1771. 73. Guerry, P. 2007. Campylobacter flagella: not just for motility. Trends Microbiol. 15:456–461. 74. Habib, I., I. Sampers, M. Uyttendaele, D. Berkvens, and L. De Zutter. 2008. Baseline data from a Belgiumwide survey of Campylobacter species contamination in chicken meat preparations and considerations for a reliable monitoring program. Appl. Environ. Microbiol. 74:5483–5489. 75. Habib, I., M. Uyttendaele, and L. De Zutter. 2010. Survival of poultry-derived Campylobacter jejuni of multilocus sequence type clonal complexes 21 and 45 under freeze, chill, oxidative, acid and heat stresses. Food Microbiol. 27:829–834. 76. Habib, I., M. Uyttendaele, and L. De Zutter. 2011. Evaluation of ISO 10272:2006 standard versus alternative enrichment and plating combinations for enumeration and detection of Campylobacter in chicken meat. Food Microbiol. 28:1117–1123. 77. Habib, I., R. Louwen, M. Uyttendaele, K. Houf, O. Vandenberg, E. E. Nieuwenhuis, W. G. Miller, A. van Belkum, and L. De Zutter. 2009. Correlation between genotypic diversity, lipooligosaccharide gene locus class variation, and Caco-2 cell invasion potential of Campylobacter jejuni isolates from chicken meat and humans: contribution to virulotyping. Appl. Environ. Microbiol. 75:4277–4288. 78. Hadden, R. D., H. Karch, H. P. Hartung, J. Zielasek, B. Weissbrich, J. Schubert, A. Weishaupt, D. R. Cornblath, A. V. Swan, R. A. Hughes, and K. V. Toyka. 2001. Preceding infections, immune factors, and outcome in Guillain-Barre syndrome. Neurology 56:758–765. 79. Han, F., S. Pu, F. Wang, J. Meng, and B. Ge. 2009. Fitness cost of macrolide resistance in Campylobacter jejuni. Int. J. Antimicrob. Agents 34:462–466. 80. Hänninen, M. L., P. Perko-Makela, A. Pitkala, and H. Rautelin. 2000. A three-year study of Campylobacter

SMP_Food Microbiology_CH11.indd

281

81.

82.

83.

84. 85.

86.

87. 88. 89.

90.

91.

92.

93.

94.

jejuni genotypes in humans with domestically acquired infections and in chicken samples from the Helsinki area. J. Clin. Microbiol. 38:1998–2000. Hansson, I., L. P. Forshell, P. Gustafsson, S. Boqvist, J. Lindblad, E. O. Engvall, Y. Andersson, and I. Vagsholm. 2007. Summary of the Swedish Campylobacter program in broilers, 2001 through 2005. J. Food Prot. 70:2008–2014. Havelaar, A. H., W. van Pelt, C. W. Ang, J. A. Wagenaar, J. P. van Putten, U. Gross, and D. G. Newell. 2009. Immunity to Campylobacter: its role in risk assessment and epidemiology. Crit. Rev. Microbiol. 35:1–22. Helms, M., J. Simonsen, K. E. Olsen, and K. Molbak. 2005. Adverse health events associated with antimicrobial drug resistance in Campylobacter species: a registrybased cohort study. J. Infect. Dis. 191:1050–1055. Heywood, W., B. Henderson, and S. P. Nair. 2005. Cytolethal distending toxin: creating a gap in the cell cycle. J. Med. Microbiol. 54:207–216. Hofshagen, M., and H. Kruse. 2005. Reduction in flock prevalence of Campylobacter spp. in broilers in Norway after implementation of an action plan. J. Food Prot. 68:2220–2223. Hong, J., J. M. Kim, W. K. Jung, S. H. Kim, W. Bae, H. C. Koo, J. Gil, M. Kim, J. Ser, and Y. H. Park. 2007. Prevalence and antibiotic resistance of Campylobacter spp. isolated from chicken meat, pork, and beef in Korea, from 2001 to 2006. J. Food Prot. 70:860–866. Humphrey, C. D., D. M. Montag, and F. E. Pittman. 1985. Experimental infection of hamsters with Campylobacter jejuni. J. Infect. Dis. 151:485–493. Humphrey, T., S. O’Brien, and M. Madsen. 2007. Campylobacters as zoonotic pathogens: a food production perspective. Int. J. Food Microbiol. 117:237–257. Hunt, J. M., C. Abeyta, and T. Tran. 1998. Campylobacter. In U.S. Food and Drug Administration, Bacteriological Analytical Manual (BAM), Edition 8, Revision A. http://www.fda.gov/Food/ScienceResearch/ LaboratoryMethods/BacteriologicalAnalyticalManual BAM/default.htm. Accessed 14 August 2012. Hussain, I., M. M. Shahid, M. Akhtar, and A. Khan. 2007. Prevalence of Campylobacter species in meat, milk and other food commodities in Pakistan. Food Microbiol. 24:219–222. Islam, Z., A. van Belkum, A. J. Cody, H. Tabor, B. C. Jacobs, K. A. Talukder, and H. P. Endtz. 2009. Campylobacter jejuni HS:23 and Guillain-Barre syndrome, Bangladesh. Emerg. Infect. Dis. 15:1315–1317. Jacobs-Reitsma, W. F., N. M. Bolder, and R. W. Mulder. 1994. Cecal carriage of Campylobacter and Salmonella in Dutch broiler flocks at slaughter: a oneyear study. Poult. Sci. 73:1260–1266. Jacobs-Reitsma, W., U. Lyhs, and J. Wagenaar. 2008. Campylobacter in the food supply, p. 627–644. In I. Nachamkin, C. M. Szymanski, and M. J. Blaser (ed.), Campylobacter, 3rd ed. ASM Press, Washington, DC. Jang, K. I., M. G. Kim, S. D. Ha, K. S. Kim, K. H. Lee, D. H. Chung, C. H. Kim, and K. Y. Kim. 2007. Morphology and adhesion of Campylobacter jejuni to

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

282

95.

96.

97.

98.

99. 100. 101.

102.

103.

104. 105.

106.

chicken skin under varying conditions. J. Microbiol. Biotechnol. 17:202–206. Janssen, R., K. A. Krogfelt, S. A. Cawthraw, W. van Pelt, J. A. Wagenaar, and R. J. Owen. 2008. Hostpathogen interactions in Campylobacter infections: the host perspective. Clin. Microbiol. Rev. 21:505–518. Jasson, V., M. Uyttendaele, A. Rajkovic, and J. Debevere. 2007. Establishment of procedures provoking sub-lethal injury of Listeria monocytogenes, Campylobacter jejuni and Escherichia coli O157 to serve method performance testing. Int. J. Food Microbiol. 118:241–249. Jayarao, B. M., S. C. Donaldson, B. A. Straley, A. A. Sawant, N. V. Hegde, and J. L. Brown. 2006. A survey of foodborne pathogens in bulk tank milk and raw milk consumption among farm families in Pennsylvania. J. Dairy Sci. 89:2451–2458. Jin, S., A. Joe, J. Lynett, E. K. Hani, P. Sherman, and V. L. Chan. 2001. JlpA, a novel surface-exposed lipo­ protein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol. Microbiol. 39:1225–1236. Johnson, W. M., and H. Lior. 1988. A new heatlabile cytolethal distending toxin (CLDT) produced by Campylobacter spp. Microb. Pathog. 4:115–126. Jones, D. M., E. M. Sutcliffe, and A. Curry. 1991. Recovery of viable but non-culturable Campylobacter jejuni. J. Gen. Microbiol. 137:2477–2482. Jørgensen, F., R. Bailey, S. Williams, P. Henderson, D. R. Wareing, F. J. Bolton, J. A. Frost, L. Ward, and T. J. Humphrey. 2002. Prevalence and numbers of Salmonella and Campylobacter spp. on raw, whole chickens in relation to sampling methods. Int. J. Food Microbiol. 76:151–164. Josefsen, M. H., C. Löfström, T. B. Hansen, L. S. Christensen, J. E. Olsen, and J. Hoorfar. 2010. Rapid quantification of viable Campylobacter bacteria on chicken carcasses, using real-time PCR and propidium monoazide treatment, as a tool for quantitative risk assessment. Appl. Environ. Microbiol. 76:5097–5104. Kassenborg, H. D., K. E. Smith, D. J. Vugia, T. RabatskyEhr, M. R. Bates, M. A. Carter, N. B. Dumas, M. P. Cassidy, N. Marano, R. V. Tauxe, and F. J. Angulo. 2004. Fluoroquinolone-resistant Campylobacter infections: eating poultry outside of the home and foreign travel are risk factors. Clin. Infect. Dis. 38(Suppl. 3): S279–S284. Ketley, J. M., and M. E. Konkel. 2005. Campylobacter: Molecular and Cellular Biology. Horizon Bioscience, Norfolk, United Kingdom. Kim, S. A., Y. M. Lee, I. G. Hwang, D. H. Kang, G. J. Woo, and M. S. Rhee. 2009. Eight enrichment broths for the isolation of Campylobacter jejuni from inoculated suspensions and ground pork. Lett. Appl. Microbiol. 49:620–626. Klancnik, A., B. Guzej, P. Jamnik, D. Vuckovic, M. Abram, and S. S. Mozina. 2009. Stress response and pathogenic potential of Campylobacter jejuni cells exposed to starvation. Res. Microbiol. 160:345–352.

SMP_Food Microbiology_CH11.indd

282

107. Konkel, M. E., B. J. Kim, V. Rivera-Amill, and S. G. Garvis. 1999. Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol. Microbiol. 32:691–701. 108. Konkel, M. E., J. D. Klena, V. Rivera-Amill, M. R. Monteville, D. Biswas, B. Raphael, and J. Mickelson. 2004. Secretion of virulence proteins from Campylo­ bacter jejuni is dependent on a functional flagellar export apparatus. J. Bacteriol. 186:3296–3303. 109. Konkel, M. E., M. R. Monteville, V. Rivera-Amill, and L. A. Jones. 2001. The pathogenesis of Campylobacter jejuni-mediated enteritis. Curr. Issues Intest. Microbiol. 2:55–71. 110. Konkel, M. E., S. G. Garvis, S. L. Tipton, D. E. Anderson, Jr., and W. Cieplak, Jr. 1997. Identification and molecular cloning of a gene encoding a fibronectinbinding protein (CadF) from Campylobacter jejuni. Mol. Microbiol. 24:953–963. 111. Kramer, J. M., J. A. Frost, F. J. Bolton, and D. R. Wareing. 2000. Campylobacter contamination of raw meat and poultry at retail sale: identification of multiple types and comparison with isolates from human infection. J. Food Prot. 63:1654–1659. 112. Kuusi, M., J. P. Nuorti, M. L. Hanninen, M. Koskela, V. Jussila, E. Kela, I. Miettinen, and P. Ruutu. 2005. A large outbreak of campylobacteriosis associated with a municipal water supply in Finland. Epidemiol. Infect. 133:593–601. 113. Kuwabara, S., S. Misawa, S. Sakamoto, and T. Hattori. 2004. Isolated posterior interosseous nerve palsy subsequent to Campylobacter jejuni enteritis. Eur. Neurol. 52:63–64. 114. Leonard, E. E., T. Takata, M. J. Blaser, S. Falkow, L. S. Tompkins, and E. C. Gaynor. 2003. Use of an open-reading frame-specific Campylobacter jejuni DNA microarray as a new genotyping tool for studying epidemiologically related isolates. J. Infect. Dis. 187:691–694. 115. Lin, J. 2009. Novel approaches for Campylobacter control in poultry. Foodborne Pathog. Dis. 6:755–765. 116. Litrup, E., M. Torpdahl, and E. M. Nielsen. 2007. Multilocus sequence typing performed on Campylobacter coli isolates from humans, broilers, pigs and cattle originating in Denmark. J. Appl. Microbiol. 103:210–218. 117. Louwen, R. P., A. van Belkum, J. A. Wagenaar, Y. Doorduyn, R. Achterberg, and H. P. Endtz. 2006. Lack of association between the presence of the pVir plasmid and bloody diarrhea in Campylobacter jejuni enteritis. J. Clin. Microbiol. 44:1867–1868. 118. Louwen, R., A. Heikema, A. van Belkum, A. Ott, M. Gilbert, W. Ang, H. P. Endtz, M. P. Bergman, and E. E. Nieuwenhuis. 2008. The sialylated lipooligosaccharide outer core in Campylobacter jejuni is an important determinant for epithelial cell invasion. Infect. Immun. 76:4431–4438. 119. Luber, P., S. Brynestad, D. Topsch, K. Scherer, and E. Bartelt. 2006. Quantification of Campylobacter species cross-contamination during handling of contaminated fresh chicken parts in kitchens. Appl. Environ. Microbiol. 72:66–70.

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

283

120. Luo, N., S. Pereira, O. Sahin, J. Lin, S. Huang, L. Michel, and Q. Zhang. 2005. Enhanced in vivo fitness of fluoroquinolone-resistant Campylobacter jejuni in the absence of antibiotic selection pressure. Proc. Natl. Acad. Sci. USA 102:541–546. 121. Mac¢kiw, E., J. Popowski, and L. Szponar. 2007. Thermotolerant Campylobacter spp.—report on monitoring studies performed in 2004–2005 in Poland. Food Control. 19:219–222. 122. Medema, G. J., F. M. Schets, A. W. van de Giessen, and A. H. Havelaar. 1992. Lack of colonization of 1 day old chicks by viable, non-culturable Campylobacter jejuni. J. Appl. Bacteriol. 72:512–516. 123. Meldrum, R. J., and I. G. Wilson. 2007. Salmonella and Campylobacter in United Kingdom retail raw chicken in 2005. J. Food Prot. 70:1937–1939. 124. Mellmann, A., J. Mosters, E. Bartelt, P. Roggentin, A. Ammon, A. W. Friedrich, H. Karch, and D. Harmsen. 2004. Sequence-based typing of flaB is a more stable screening tool than typing of flaA for monitoring of Campylobacter populations. J. Clin. Microbiol. 42:4840–4842. 125. Miller, W. G. 2008. Comparative genomics of Campylobacter species other than Campylobacter je­ juni, p. 73–95. In I. Nachamkin, C. M. Szymanski, and M. J. Blaser (ed.), Campylobacter, 3rd ed. ASM Press, Washington, DC. 126. Miller, W. G., C. T. Parker, S. Heath, and A. J. Lastovica. 2007. Identification of genomic differences between Campylobacter jejuni subsp. jejuni and C. jejuni subsp. doylei at the nap locus leads to the development of a C. jejuni subspeciation multiplex PCR method. BMC. Microbiol. 7:11. 127. Miller, W. G., M. D. Englen, S. Kathariou, I. V. Wesley, G. Wang, L. Pittenger-Alley, R. M. Siletz, W. Muraoka, P. J. Fedorka-Cray, and R. E. Mandrell. 2006. Identification of host-associated alleles by multi­locus sequence typing of Campylobacter coli strains from food animals. Microbiology 152:245–255. 128. Mishu, B., and M. J. Blaser. 1993. Role of infection due to Campylobacter jejuni in the initiation of GuillainBarre syndrome. Clin. Infect. Dis. 17:104–108. 129. Moore, J. E., M. D. Barton, I. S. Blair, D. Corcoran, J. S. Dooley, S. Fanning, I. Kempf, A. J. Lastovica, C. J. Lowery, M. Matsuda, D. A. McDowell, A. McMahon, B. C. Millar, J. R. Rao, P. J. Rooney, B. S. Seal, W. J. Snelling, and O. Tolba. 2006. The epidemiology of anti­ biotic resistance in Campylobacter. Microbes Infect. 8:1955–1966. 130. Mortensen, N. P., M. L. Kuijf, C. W. Ang, P. Schiellerup, K. A. Krogfelt, B. C. Jacobs, A. van Belkum, H. P. Endtz, and M. P. Bergman. 2009. Sialylation of Campylobacter jejuni lipo-oligosaccharides is associated with severe gastro-enteritis and reactive arthritis. Microbes Infect. 11:988–994. 131. Muehlherr, J. E., C. Zweifel, S. Corti, J. E. Blanco, and R. Stephan. 2003. Microbiological quality of raw goat’s and ewe’s bulk-tank milk in Switzerland. J. Dairy Sci. 86:3849–3856.

SMP_Food Microbiology_CH11.indd

283

132. Mullner, P., G. Jones, A. Noble, S. E. Spencer, S. Hathaway, and N. P. French. 2009. Source attribution of food-borne zoonoses in New Zealand: a modified Hald model. Risk Anal. 29:970–984. 133. Mullner, P., S. E. Spencer, D. J. Wilson, G. Jones, A. D. Noble, A. C. Midwinter, J. M. Collins-Emerson, P. Carter, S. Hathaway, and N. P. French. 2009. Assigning the source of human campylobacteriosis in New Zealand: a comparative genetic and epidemiological approach. Infect. Genet. Evol. 9:1311–1319. 134. Murphy, C., C. Carroll, and K. N. Jordan. 2003. Induction of an adaptive tolerance response in the foodborne pathogen, Campylobacter jejuni. FEMS Microbiol. Lett. 223:89–93. 135. Murphy, C., C. Carroll, and K. N. Jordan. 2005. The effect of different media on the survival and induction of stress responses by Campylobacter jejuni. J. Microbiol. Methods 62:161–166. 136. Murphy, C., C. Carroll, and K. N. Jordan. 2006. Environmental survival mechanisms of the foodborne pathogen Campylobacter jejuni. J. Appl. Microbiol. 100:623–632. 137. Nachamkin, I., B. M. Allos, and T. Ho. 1998. Campylobacter species and Guillain-Barre syndrome. Clin. Microbiol. Rev. 11:555–567. 138. Nachamkin, I., B. P. Arzarte, H. Ung, C. Lobato, R. A. Gonzalez, P. Rodriguez, B. A. Garcia, L. M. Cordero, P. L. Garcia, J. C. Perez, M. Ribera, P. C. Aldama, G. D. Guiterrez, L. F. Sarnat, M. R. Garcia, J. Veitch, C. Fitzgerald, D. R. Cornblath, P. M. Rodriguez, J. W. Griffin, H. J. Willison, A. K. Asbury, and G. M. McKhann. 2007. Patterns of Guillain-Barre syndrome in children: results from a Mexican population. Neurology 69:1665–1671. 139. Nachamkin, I., C. M. Szymanski, and M. J. Blaser. 2008. Campylobacter, 3rd ed. ASM Press, Washington, DC. 140. Nauta, M., A. Hill, H. Rosenquist, S. Brynestad, A. Fetsch, P. van der Logt, A. Fazil, B. Christensen, E. Katsma, B. Borck, and A. Havelaar. 2009. A comparison of risk assessments on Campylobacter in broiler meat. Int. J. Food Microbiol. 129:107–123. 141. Nazef, L., Y. Belguesmia, A. Tani, H. Prévost, and D. Drider. 2008. Identification of lactic acid bacteria from poultry feces: evidence on anti-Campylobacter and antiListeria activities. Poult. Sci. 87:329–334. 142. Nelson, J. M., T. M. Chiller, J. H. Powers, and F. J. Angulo. 2007. Fluoroquinolone-resistant Campylobacter species and the withdrawal of fluoroquinolones from use in poultry: a public health success story. Clin. Infect. Dis. 44:977–980. 143. Newell, D. G., H. McBride, F. Saunders, Y. Dehele, and A. D. Pearson. 1985. The virulence of clinical and environmental isolates of Campylobacter jejuni. J. Hyg. 94:45–54. 144. Nichols, G. L. 2005. Fly transmission of Campylobacter. Emerg. Infect. Dis. 11:361–364. 145. Northcutt, J., D. Smith, K. D. Ingram, A. Hinton, Jr., and M. Musgrove. 2007. Recovery of bacteria from

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

284

146.

147.

148.

149.

150.

151. 152.

153.

154.

155.

156.

157.

broiler carcasses after spray washing with acidified electrolyzed water or sodium hypochlorite solutions. Poult. Sci. 86:2239–2244. Odumeru, J. A., S. J. Mitchel, D. M. Alves, J. A. Lynch, A. J. Yee, S. L. Wang, S. Styliadis, and J. M. Farber. 1997. Assessment of the microbiological quality of ready-to-use vegetables for health-care food services. J. Food Prot. 60:954–960. On, S. L., and P. J. Jordan. 2003. Evaluation of 11 PCR assays for species-level identification of Campylobacter jejuni and Campylobacter coli. J. Clin. Microbiol. 41:330–336. On, S. L., N. Dorrell, L. Petersen, D. D. Bang, S. Morris, S. J. Forsythe, and B. W. Wren. 2006. Numerical analysis of DNA microarray data of Campylobacter jejuni strains correlated with survival, cytolethal distending toxin and haemolysin analyses. Int. J. Med. Microbiol. 296:353–363. On, S. L., N. McCarthy, W. G. Miller, and B. Gilpin. 2008. Molecular epidemiology of Campylobacter Species, p. 191– 211. In I. Nachamkin, C. M. Szymanski, and M. J. Blaser (ed.), Campylobacter, 3rd ed. ASM Press, Washington, DC. Oyarzabal, O. A., K. S. Macklin, J. M. Barbaree, and R. S. Miller. 2005. Evaluation of agar plates for direct enumeration of Campylobacter spp. from poultry carcass rinses. Appl. Environ. Microbiol. 71:3351–3354. Park, S. F. 2002. The physiology of Campylobacter species and its relevance to their role as foodborne pathogens. Int. J. Food Microbiol. 74:177–188. Park, S. F., and P. T. Richardson. 1995. Molecular characterization of a Campylobacter jejuni lipoprotein with homology to periplasmic siderophore-binding proteins. J. Bacteriol. 177:2259–2264. Parkhill, J., B. W. Wren, K. Mungall, J. M. Ketley, C. Churcher, D. Basham, T. Chillingworth, R. M. Davies, T. Feltwell, S. Holroyd, K. Jagels, A. V. Karlyshev, S. Moule, M. J. Pallen, C. W. Penn, M. A. Quail, M. A. Rajandream, K. M. Rutherford, A. H. van Vliet, S. Whitehead, and B. G. Barrell. 2000. The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403:665–668. Payot, S., L. Avrain, C. Magras, K. Praud, A. Cloeckaert, and E. Chaslus-Dancla. 2004. Relative contribution of target gene mutation and efflux to fluoroquinolone and erythromycin resistance, in French poultry and pig isolates of Campylobacter coli. Int. J. Antimicrob. Agents 23:468–472. Pei, Z., and M. J. Blaser. 1993. PEB1, the major cellbinding factor of Campylobacter jejuni, is a homolog of the binding component in gram-negative nutrient transport systems. J. Biol. Chem. 268:18717–18725. Petersen, L., E. M. Nielsen, J. Engberg, S. L. On, and H. H. Dietz. 2001. Comparison of genotypes and serotypes of Campylobacter jejuni isolated from Danish wild mammals and birds and from broiler flocks and humans. Appl. Environ. Microbiol. 67:3115–3121. Pickett, C. L., T. Auffenberg, E. C. Pesci, V. L. Sheen, and S. S. Jusuf. 1992. Iron acquisition and hemolysin

SMP_Food Microbiology_CH11.indd

284

158.

159.

160.

161.

162.

163.

164.

165.

166.

167. 168.

169.

170.

production by Campylobacter jejuni. Infect. Immun. 60:3872–3877. Poly, F., T. D. Read, Y. H. Chen, M. A. Monteiro, O. Serichantalergs, P. Pootong, L. Bodhidatta, C. J. Mason, D. Rockabrand, S. Baqar, C. K. Porter, D. Tribble, M. Darsley, and P. Guerry. 2008. Characterization of two Campylobacter jejuni strains for use in volunteer experimental-infection studies. Infect. Immun. 76:5655–5667. Purdy, D., C. M. Buswell, A. E. Hodgson, K. McAlpine, I. Henderson, and S. A. Leach. 2000. Characterisation of cytolethal distending toxin (CDT) mutants of Campylobacter jejuni. J. Med. Microbiol. 49:473–479. Ragimbeau, C., F. Schneider, S. Losch, J. Even, and J. Mossong. 2008. Multilocus sequence typing, pulsed-field gel electrophoresis, and fla short variable region typing of clonal complexes of Campylobacter jejuni strains of human, bovine, and poultry origins in Luxembourg. Appl. Environ. Microbiol. 74:7715–7722. Rahimi, E., H. Momtaz, M. Ameri, H. GhasemianSafaei, and M. Ali-Kasemi. 2010. Prevalence and antimicrobial resistance of Campylobacter species isolated from chicken carcasses during processing in Iran. Poult. Sci. 89:1015–1020. Rajkovic, A., N. Smigic, M. Uyttendaele, H. Medic, L. De Zutter, and F. Devlieghere. 2009. Resistance of Listeria monocytogenes, Escherichia coli O157:H7 and Campylobacter jejuni after exposure to repetitive cycles of mild bactericidal treatments. Food Microbiol. 26:889–895. Rantsiou, K., C. Lamberti, and L. Cocolin. 2010. Survey of Campylobacter jejuni in retail chicken meat products by application of a quantitative PCR protocol. Int. J. Food Microbiol. 141(Suppl. 1):S75–S79. Refrégier-Petton, J., N. Rose, M. Denis, and G. Salvat. 2001. Risk factors for Campylobacter spp. contamination in French broiler-chicken flocks at the end of the rearing period. Prev. Vet. Med. 50:89–100. Reid, A. N., R. Pandey, K. Palyada, L. Whitworth, E. Doukhanine, and A. Stintzi. 2008. Identification of Campylobacter jejuni genes contributing to acid adaptation by transcriptional profiling and genome-wide mutagenesis. Appl. Environ. Microbiol. 74:1598–1612. Riedel, C. T., L. Brøndsted, H. Rosenquist, S. N. Haxgart, and B. B. Christensen. 2009. Chemical decontamination of Campylobacter jejuni on chicken skin and meat. J. Food Prot. 72:1173–1180. Robinson, D. A. 1981. Infective dose of Campylobacter jejuni in milk. Br. Med. J. 282:1584. Rosenquist, H., A. Bengtsson, and T. B. Hansen. 2007. A collaborative study on a Nordic standard protocol for detection and enumeration of thermotolerant Campylobacter in food (NMKL 119, 3rd. ed., 2007). Int. J. Food Microbiol. 118:201–213. Rosenquist, H., L. Boysen, C. Galliano, S. Nordentoft, S. Ethelberg, and B. Borck. 2009. Danish strategies to control Campylobacter in broilers and broiler meat: facts and effects. Epidemiol. Infect. 137:1742–1750. Sagoo, S. K., C. L. Little, and R. T Mitchell. 2003. Microbiological quality of open ready-to-eat salad veg-

Manila Typesetting Company

11/08/2012 06:55AM

11.  Campylobacter Species

171.

172. 173.

174.

175. 176.

177.

178. 179.

180.

181. 182.

183.

184.

285

etables: effectiveness of food hygiene training of management. J. Food Prot. 66:1581–1586. Sails, A. D., B. Swaminathan, and P. I. Fields. 2003. Utility of multilocus sequence typing as an epidemio­ logical tool for investigation of outbreaks of gastroenteritis caused by Campylobacter jejuni. J. Clin. Microbiol. 41:4733–4739. Sallam, K. I. 2007. Prevalence of Campylobacter in chicken and chicken by-products retailed in Sapporo area, Hokkaido, Japan. Food Control 18:1113–1120. Sammarco, M. L., G. Ripabelli, I. Fanelli, G. M. Grasso, and M. Tamburro. 2010. Prevalence and biomolecular characterization of Campylobacter spp. isolated from retail meat. J. Food Prot. 73:720–728. Sampers, I., I. Habib, L. De Zutter, A. Dumoulin, and M. Uyttendaele. 2010. Survival of Campylobacter spp. in poultry meat preparations subjected to freezing, refrigeration, minor salt concentration, and heat treatment. Int. J. Food Microbiol. 28:147–153. Savasci, M., and H. Ozdemir. 2006. Prevalence of thermophilic Campylobacter spp. in retail chicken meat in Ankara. J. Food Saf. 26:244–250. Scherer, K., E. Bartelt, C. Sommerfeld, and G. Hildebrandt. 2006. Quantification of Campylobacter on the surface and in the muscle of chicken legs at retail. J. Food Prot. 69:757–761. Schmidt-Ott, R., F. Brass, C. Scholz, C. Werner, and U. Gross. 2005. Improved serodiagnosis of Campylobacter jejuni infections using recombinant antigens. J. Med. Microbiol. 54:761–767. Sebald, M., and M. Veron. 1963. Base DNA content and classification of vibrios. Ann. Inst. Pasteur. (Paris) 105:897–910. (In French.) Sheppard, S. K., J. F. Dallas, M. MacRae, N. D. McCarthy, E. L. Sproston, F. J. Gormley, N. J. Strachan, I. D. Ogden, M. C. Maiden, and K. J. Forbes. 2009. Campylobacter genotypes from food animals, environmental sources and clinical disease in Scotland 2005/6. Int. J. Food Microbiol. 134:96–103. Sinha, S., K. N. Prasad, S. Pradhan, D. Jain, and S. Jha. 2004. Detection of preceding Campylobacter jejuni infection by polymerase chain reaction in patients with Guillain-Barre syndrome. Trans. R. Soc. Trop. Med. Hyg. 98:342–346. Smibert, R. M. 1978. The genus Campylobacter. Annu. Rev. Microbiol. 32:673–709. Smith, J. L. 2002. Campylobacter jejuni infection during pregnancy: long-term consequences of associated bacteremia, Guillain-Barre syndrome, and reactive arthritis. J. Food Prot. 65:696–708. Smith, K. E., J. M. Besser, C. W. Hedberg, F. T. Leano, J. B. Bender, J. H. Wicklund, B. P. Johnson, K. A. Moore, M. T. Osterholm, et al. 1999. Quinoloneresistant Campylobacter jejuni infections in Minnesota, 1992-1998. N. Engl. J. Med. 340:1525–1532. Solow, B. T., O. M. Cloak, and P. M. Fratamico. 2003. Effect of temperature on viability of Campylobacter je­ juni and Campylobacter coli on raw chicken or pork skin. J. Food Prot. 66:2023–2031.

SMP_Food Microbiology_CH11.indd

285

185. Somers, E. B., J. L. Schoeni, and A. C. Wong. 1994. Effect of trisodium phosphate on biofilm and planktonic cells of Campylobacter jejuni, Escherichia coli O157: H7, Listeria monocytogenes and Salmonella typhimurium. Int. J. Food Microbiol. 22:269–276. 186. Soonthornchaikul, N., and H. Garelick. 2009. Antimicrobial resistance of Campylobacter species isolated from edible bivalve molluscs purchased from Bangkok markets, Thailand. Foodborne Pathog. Dis. 6:947–951. 187. Spiller, R. C. 2007. Role of infection in irritable bowel syndrome. J. Gastroenterol. 42(Suppl. 17):41–47. 188. Stern, N. J., and J. E Line. 1992. Comparison of three methods for recovery of Campylobacter spp. from broiler carcasses. J. Food Prot. 55:663–666. 189. Stern, N. J. 1994. Mucosal competitive exclusion to diminish colonization of chickens by Campylobacter jejuni. Poult. Sci. 73:402–407. 190. Stern, N. J., and S. Pretanik. 2006. Counts of Campylobacter spp. on U.S. broiler carcasses. J. Food Prot. 69:1034–1039. 191. Takahashi, M., M. Koga, K. Yokoyama, and N. Yuki. 2005. Epidemiology of Campylobacter jejuni isolated from patients with Guillain-Barre and Fisher syndromes in Japan. J. Clin. Microbiol. 43:335–339. 192. Tam, C. C., C. D. Higgins, K. R. Neal, L. C. Rodrigues, S. E. Millership, and S. J. O’Brien. 2009. Chicken consumption and use of acid-suppressing medications as risk factors for Campylobacter enteritis, England. Emerg. Infect. Dis. 15:1402–1408. 193. Tam, C. C., L. C. Rodrigues, S. J. O’Brien, and S. Hajat. 2006. Temperature dependence of reported Campylobacter infection in England, 1989-1999. Epidemiol. Infect. 134:119–125. 194. Ternhag, A., T. Asikainen, J. Giesecke, and K. Ekdahl. 2007. A meta-analysis on the effects of antibiotic treatment on duration of symptoms caused by infection with Campylobacter species. Clin. Infect. Dis. 44:696–700. 195. Tracz, D. M., M. Keelan, J. Ahmed-Bentley, A. Gibreel, K. Kowalewska-Grochowska, and D. E. Taylor. 2005. pVir and bloody diarrhea in Campylobacter jejuni enteritis. Emerg. Infect. Dis. 11:838–843. 196. Travers, K., and M. Barza. 2002. Morbidity of infections caused by antimicrobial-resistant bacteria. Clin. Infect. Dis. 34(Suppl. 3):S131–S134. 197. Tribble, D. R., S. Baqar, L. W. Pang, C. Mason, H. S. Houng, C. Pitarangsi, C. Lebron, A. Armstrong, O. Sethabutr, and J. W. Sanders. 2008. Diagnostic approach to acute diarrheal illness in a military population on training exercises in Thailand, a region of campylobacter hyperendemicity. J. Clin. Microbiol. 46:1418–1425. 198. van Asselt, E. D., W. F. Jacobs-Reitsma, R. van Brakel, H. van der Foet, and H. J. van der Fels-Klerx. 2008. Campylobacter prevalence in the broiler supply chain in the Netherlands. Poult. Sci. 87:2166–2172. 199. van Belkum, A., S. Scherer, L. van Alphen, and H. Verbrugh. 1998. Short-sequence DNA repeats in prokaryotic genomes. Microbiol. Mol. Biol. Rev. 62: 275–293.

Manila Typesetting Company

11/08/2012 06:55AM

Foodborne Pathogenic Bacteria

286 200. Vandamme, P., and S. L. On. 2001. Recommendations of the subcommittee on the taxonomy of Campylobacter and related bacteria. Int. J. Syst. Evol. Microbiol. 51:719–721. 201. van de Giessen, A. W., C. J. Heuvelman, T. Abee, and W. C. Hazeleger. 1996. Experimental studies on the infectivity of non-culturable forms of Campylobacter spp. in chicks and mice. Epidemiol. Infect. 117:463–470. 202. van den Brandhof, W. E., G. A. De Wit, M. A. de Wit, and Y. T. van Duynhoven. 2004. Costs of gastroenteritis in The Netherlands. Epidemiol. Infect. 132:211–221. 203. Van Deun, K., F. Haesebrouck, M. Heyndrickx, H. Favoreel, J. Dewulf, L. Ceelen, L. Dumez, W. Messens, S. Leleu, F. Van Immerseel, R. Ducatelle, and F. Pasmans. 2007. Virulence properties of Campylobacter jejuni isolates of poultry and human origin. J. Med. Microbiol. 56:1284–1289. 204. van Doorn, P. A., L. Ruts, and B. C. Jacobs. 2008. Clinical features, pathogenesis, and treatment of Guillain-Barre syndrome. Lancet Neurol. 7:939–950. 205. Van Pelt, W., A. Havelaar, P. Westra, and J. A. Wagenaar. 2009. Strong regional reduction of campylobacteriosis during and after avian influenza poultry farm culling. A model for future intervention studies at primary production?, abstr. P25, p. 83. In Proceedings of the 15th International Workshop on Campylobacter, Helicobacter and Related Organisms, 2 to 5 September 2009, Niigata, Japan. 206. Vellinga, A., and F. Van Loock. 2002. The dioxin crisis as experiment to determine poultry-related campylobacter enteritis. Emerg. Infect. Dis. 8:19–22. 207. Verhoeff-Bakkenes, L., W. C. Hazeleger, R. de Jonge, and M. H. Zwietering. 2009. Campylobacter jejuni: a study on environmental conditions affecting culturability and in vitro adhesion/invasion. J. Appl. Microbiol. 106:924–931. 208. Vilar, M. J., F. J. Pena, I. Perez, F. J. Dieguez, M. L. Sanjuan, J. L. Rodriguez-Otero, and E. Yus. 2010. Presence of Listeria, Arcobacter, and Campylobacter spp. in dairy farms in Spain. Berl. Munch. Tierarztl. Wochenschr. 123:58–62. 209. Wassenaar, T. M., and D. G. Newell. 2000. Genotyping of Campylobacter spp. Appl. Environ. Microbiol. 66:1–9. 210. Wassenaar, T. M., and M. J. Blaser. 1999. Pathophysiology of Campylobacter jejuni infections of humans. Microbes Infect. 1:1023–1033.

SMP_Food Microbiology_CH11.indd

286

211. Wassenaar, T. M., M. Kist, and A. de Jong. 2007. Reanalysis of the risks attributed to ciprofloxacin-resistant Campylobacter jejuni infections. Int. J. Antimicrob. Agents 30:195–201. 212. Westgarth, C., C. J. Porter, L. Nicolson, R. J. Birtles, N. J. Williams, C. A. Hart, G. L. Pinchbeck, R. M. Gaskell, R. M. Christley, and S. Dawson. 2009. Risk factors for the carriage of Campylobacter upsaliensis by dogs in a community in Cheshire. Vet. Rec. 165:526–530. 213. Whyte, P., K. McGill, D. Cowley, R. H. Madden, L. Moran, P. Scates, C. Carroll, A. O’Leary, S. Fanning, J. D. Collins, E. McNamara, J. E. Moore, and M. Cormican. 2004. Occurrence of Campylobacter in retail foods in Ireland. Int. J. Food Microbiol. 95:111–118. 214. Wilson, D. J., E. Gabriel, A. J. Leatherbarrow, J. Cheesbrough, S. Gee, E. Bolton, A. Fox, P. Fearnhead, C. A. Hart, and P. J. Diggle. 2008. Tracing the source of campylobacteriosis. PLoS Genet. 4:e1000203. 215. Wilson, D. L., J. A. Bell, V. B. Young, S. R. Wilder, L. S. Mansfield, and J. E. Linz. 2003. Variation of the natural transformation frequency of Campylobacter jejuni in liquid shake culture. Microbiology 149:3603–3615. 216. Wong, T. L., L. Hollis, A. Cornelius, C. Nicol, R. Cook, and J. A. Hudson. 2007. Prevalence, numbers, and subtypes of Campylobacter jejuni and Campylobacter coli in uncooked retail meat samples. J. Food Prot. 70:566–573. 217. Yang, C., Y. Jiang, K. Huang, C. Zhu, and Y. Yin. 2003. Application of real-time PCR for quantitative detection of Campylobacter jejuni in poultry, milk and environmental water. FEMS Immunol. Med. Microbiol. 38:265–271. 218. Yao, R., D. H. Burr, P. Doig, T. J. Trust, H. Niu, and P. Guerry. 1994. Isolation of motile and non-motile insertional mutants of Campylobacter jejuni: the role of motility in adherence and invasion of eukaryotic cells. Mol. Microbiol. 14:883–893. 219. Young, K. T., L. M. Davis, and V. J. Dirita. 2007. Campylobacter jejuni: molecular biology and patho­ genesis. Nat. Rev. Microbiol. 5:665–679. 220. Zhao, C., B. Ge, V. J. De, R. Sudler, E. Yeh, S. Zhao, D. G. White, D. Wagner, and J. Meng. 2001. Prevalence of Campylobacter spp., Escherichia coli, and Salmonella serovars in retail chicken, turkey, pork, and beef from the Greater Washington, D.C., area. Appl. Environ. Microbiol. 67:5431–5436.

Manila Typesetting Company

11/08/2012 06:55AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch12

Jianghong Meng Jeffrey T. LeJeune Tong Zhao Michael P. Doyle

Enterohemorrhagic Escherichia coli

Escherichia coli is a facultatively anaerobic gram­ egative bacterium that is primarily present in the n ­gastrointestinal tract of humans and warm-blooded animals. Although most of these commensal E. coli strains are harmless, many are pathogenic and cause a variety of diseases in humans and animals (32). Specific virulence attributes that have been acquired by such strains enable them to cause three principal types of infections in humans including intestinal gastroenteritis, urinary tract infections, and neonatal sepsis/meningitis. E. coli isolates can be serologically or genetically differentiated based on three major surface antigens or their encoding genes, which enable serotyping: the “O” (somatic), “H” (flagella), and “K” (capsule) antigens (1, 86). At present, more than 700 serotypes of E. coli have been identified based on “O,” “H,” and “K” antigens (23). It is considered necessary to determine only the O and the H antigens, not the K antigens, to serotype strains of E. coli associated with diarrheal disease. The O antigen identifies the serogroup of a strain, and the H antigen identifies its serotype. The application of serotyping to isolates associated with diarrheal disease has revealed that particular serogroups often fall into

12 one category of pathogenic E. coli. However, some serogroups such as O55, O111, O126, and O128 are in more than one category. Pathogenic E. coli strains are categorized into specific groups (pathotypes) based on their virulence determinants. These virulence determinants include those ­controlling adhesions (CFAI/CFAII, type 1 fimbriae, P fimbriae, S fimbriae, and intimin), invasions (hemolysins, siderophores, siderophore uptake systems, and Shigellalike invasins), motility (flagella), toxins (heat-stable and heat-labile enterotoxins, Shiga toxins [Stxs], cytotoxins, and endotoxins), antiphagocytic surface structures (capsules, K antigens, lipopolysaccharides [LPS]), and genetic characteristics (genetic exchange through transduction or conjugation, transmissible plasmids, R factors, and drug resistance and virulence plasmids). The five categories of gastrointestinal pathogenic E. coli include enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), enteroinvasive E. coli (EIEC), enteroaggregative E. coli (EAEC), and enterohemorrhagic E. coli (EHEC). This chapter will focus largely on the EHEC group, which among the E. coli strains that cause foodborne illness is the most significant group based on the

Jianghong Meng, Department of Nutrition and Food Science, University of Maryland, College Park, MD 20742. Jeffrey T. LeJeune, Food Animal Health Research Program, Ohio Agricultural Research and Development Center, Wooster, OH 44691. Tong Zhao, Center for Food Safety, University of Georgia, Griffin, GA 30223. Michael P. Doyle, Center for Food Safety, University of Georgia, Griffin, GA 30223.

287

Foodborne Pathogenic Bacteria

288 frequency of foodborne illness in the United States and the severity of illness. More information on other diarrheagenic E. coli is available in several review articles (55, 78).

EPEC EPEC was the first pathotype of E. coli described and can cause watery diarrhea like ETEC, but these organisms do not possess the same colonization factors as ETEC and do not produce LT or ST toxins. The major O serogroups associated with illness include O55, O86, O111ab, O119, O125ac, O126, O127, O128ab, and O142. Humans are an important reservoir. The original definition of EPEC is “diarrheagenic E. coli belonging to serogroups epidemiologically incriminated as pathogens but whose pathogenic mechanism have not been proven to be related to either enterotoxins, or Shigella-like invasiveness” (78). However, EPEC have been determined to induce attaching and effacing (A/E) lesions in cells to which they adhere and can invade epithelial cells. Some types of EPEC are referred to as enteroadherent E. coli (EAEC), based on specific patterns of adherence. EAEC are an important cause of traveler’s diarrhea in Mexico and in North Africa.

ETEC ETEC are a major cause of infantile diarrhea in developing countries or regions with poor sanitation. They are also the agents most frequently responsible for traveler’s diarrhea but do not cause disease in the local adults because of developed immunity. ETEC colonize the proximal small intestine by fimbrial colonization factors (e.g., CFA/I and CFA/II) and produce LT or ST enterotoxin that elicits fluid accumulation and a diarrheal response. The LT enterotoxin is similar to cholera toxin in both structure and mode of action, and the ST enterotoxin is a peptide that causes an increase in cyclic GMP in host cell cytoplasm, leading to the same effects as those that occur with an increase in cyclic AMP. The most frequently isolated ETEC serogroups include O6, O8, O15, O20, O25, O27, O63, O78, O85, O115, O128ac, O148, O159, and O167. Humans are the principal reservoir of ETEC strains that cause human illness.

EIEC EIEC cause nonbloody diarrhea and dysentery similar to that caused by Shigella spp. by invading and multiplying within colonic epithelial cells. As for Shigella, the invasive capacity of EIEC is associated with the pres-

ence of a large plasmid (ca. 140 MDa) that encodes several outer membrane proteins involved in invasiveness. The antigenicity of these outer membrane proteins and the O antigens of EIEC are closely related. They do not produce LT, ST, or Stx. The principal site of bacterial localization is the colon, where EIEC invade and proliferate in epithelial cells, causing widespread cell death. Humans are a major reservoir, and the serogroups most frequently associated with illness include O28ac, O29, O112, O124, O136, O143, O144, O152, O164, and O167. Among these serogroups, O124 is the serogroup most commonly encountered.

EAEC EAEC recently have been associated with persistent diarrhea in infants and children in several countries worldwide. These E. coli strains are uniquely different from the other types of pathogenic E. coli because of their ability to produce a characteristic pattern of aggregative adherence on HEp-2 cells. EAEC adhere to the surface of HEp-2 cells in an appearance of stacked bricks. Serogroups associated with EAEC include O3, O15, O44, O77, O86, O92, O111, and O127. A distinctive heat-labile, plasmidencoded toxin has been isolated from these strains, called the EAST (enteroaggregative ST) toxin. EAEC also produce a hemolysin related to the hemolysin produced by uropathogenic strains of E. coli. However, the role of the toxin and the hemolysin in virulence has not been proven. More epidemiologic information is needed to elucidate the significance of EAEC as an agent of diarrheal disease. Although EAEC has seldom been implicated in major foodborne disease incidents, a large outbreak that occurred in 2011 was centered in Germany but affected various other countries in the European Union (40). This outbreak, suspected to have been caused by contaminated sprouts, infected over 3,700 people, had a high hemolytic-uremic syndrome (HUS) rate (~24%), and resulted in more than 50 fatalities (93). The causative agent was identified as E. coli O104:H4, which produced Stx2a and therefore was considered a Stx-producing E. coli (STEC) strain. However, whole-genome sequencing of the pathogen revealed that it shared 93% genomic homology with EAEC strain 55589 and also carried the aggR gene, which is a transcriptional activator essential for the expression of AAF I and found on an EAEC virulence plasmid. Hence, genetic analyses revealed that the causative pathogen was a multiantibiotic-resistant EAEC strain that had acquired the ability to produce Stx via phage conversion.

12.  Enterohemorrhagic Escherichia coli

289

Table 12.1  Serotypes of non-O157 Stx-producing E. coli recovered from patients with hemorrhagic colitis and/or HUSa Sero­ group O1 O2

O4

O5 O6

O8

O9 O11 O14 a b

H type — 7 6 7 29 — 5 10 — — 2 4 2 9 19 21 — 2 —

b

Sero­ group O15 O18 O20 O22 O23

O25 O26 O45 O46 O48 O50 O55

H type — — 7 19 5 8 0 7 16 2 — 11 2 31 21 — 7 6 10

Sero­ group O68 O69 O70 O73 O75 O76 O77 O79 O83 O84 O86 O91

O92 O98 O100

H type 4 — 35 34 5 7 — 7 1 — — — 10 21 40 — 33 — 25

Sero­ group O100 O101 O103

O104

O105 O105ac O107 O111

O111ac O112ab

H type 32 — 2 18 21 25 — 2 21 18 18 27 — 2 7 8 11 — 2

Sero­ group O112ac O113 O118

O119

O121 O125 O126 O127 O128

O128ab

H type — 21 — 12 16 30 2 5 6 10 19 — 27 21 — 2 7 2 45

Sero­ group O134 O137 O145

O146

O153 O163 O165 O168 O172 O173 O174

H type 25 41 — 25 28 8 21 28 2 25 19 — 25 — — 2 — 2 21

Sero­ group O? OR

OX3

OX174 OX177 OX181

H type — 11 — 4 9 11 16 25 49 — 2 21 2 21 — 49

Data from reference 7. —, nonmotile.

EHEC EHEC were first recognized as human pathogens in 1982 when E. coli O157:H7 was identified as the cause of two outbreaks of hemorrhagic colitis. Since then, many other serogroups of E. coli, such as O26, O111, and sorbitol-fermenting O157:NM, also have been associated with cases of hemorrhagic colitis and have been classified as EHEC. However, serotype O157:H7 is the predominant cause of EHEC-associated disease in the United States and many other countries. All EHEC produce factors cytotoxic to African green monkey kidney (Vero) cells, factors which are hence named verotoxins or Stxs because of their similarity to the Stx produced by Shigella dysenteriae type 1 (81). Production of Stxs by E. coli O157:H7 was first reported in 1983 (54) and was subsequently associated with a severe and sometimes fatal condition, HUS (58). E. coli organisms of many different serotypes are capable of producing Stxs and hence are named Shiga toxin-producing E. coli (STEC). More than 600 serotypes of STEC have been identified, including approximately 160 O serogroups and 50 H types, and the list is still growing (7). However, only those strains that cause hemorrhagic colitis are considered to be EHEC, and there are at least 130 EHEC serotypes that have been recovered from human patients (Table 12.1). Major non-O157 EHEC serogroups identified in the United States include O26, O45, O103, O111, O121, and O145 (12).

CHARACTERISTICS OF E. COLI O157:H7 AND NON-O157 EHEC Escherichia coli O157:H7 was first identified as a foodborne pathogen in 1982. There had been prior isolation of the organism, identified retrospectively among isolates at the CDC; the isolate was obtained from a California woman with bloody diarrhea in 1975 (46). In addition to production of Stx(s), most strains of E. coli O157:H7 also possess several characteristics uncommon to most other E. coli strains: inability to grow well at temperatures ³44.5°C in E. coli broth, inability to ferment sorbitol within 24 hours, inability to produce b-glucuronidase (i.e., inability to hydrolyze 4-methylumbelliferyl-d-glucuronide [MUG]), possession of a pathogenicity island known as the locus of enterocyte effacement (LEE), and carriage of a 60-MDa (92-kbp) plasmid. Non-O157 EHEC do not share the previously described growth and metabolic characteristics, although they all produce Stx(s) and many contain LEE and the large plasmid. EHEC can produce pediatric diarrhea, copious bloody discharge, i.e., hemorrhagic colitis, and intense inflammatory response and may be complicated by HUS. There is tremendous genetic diversity among EHEC isolates. O157 EHEC infections are more likely than non-O157 EHEC infections to result in bloody diarrhea (80% versus 45%), hospitalization (34% versus 8%), and HUS (6% versus <2%) (48). Of the non-O157

290 EHEC cases in the United States, 74% were represented by just five serotypes, including O26 (27%), O103 (21%), O111 (19%), O145 (5%), and O45 (4%). At present, E. coli O157:H7 is the dominant EHEC isolate in the United States, Canada, the United Kingdom, and Japan, and non-O157 EHEC are dominant among isolates in Europe, Argentina, Australia, Chile, and South Africa.

Acid Resistance

Foodborne pathogens must pass through an acidic gastric barrier with pH values as low as 1.5 to 2.5 to cause infections in humans. Some enteric pathogens such as Vibrio cholerae use an “assault tactic” that involves large numbers of infecting cells, in the hope that a few will survive and gain entrance into the intestine. E. coli O157:H7, however, has effective mechanisms in tolerating extreme acid stress. Three systems in EHEC are involved in acid resistance, including an acid-induced oxidative, an acid-induced arginine-dependent, and a glutamate-dependent system (66). The oxidative system is less effective in protecting the organism from acid stress than the arginine-dependent and glutamate-dependent systems. The alternate sigma factor RpoS is required for oxidative acid tolerance but is only partially involved with the other two systems. Once induced, the acid resistance state can persist for a prolonged time (³28 days) at refrigeration temperature. More detailed information on acid resistance can be found in a review article by Foster (39). The minimum pH for E. coli O157:H7 growth is 4.0 to 4.5, but growth is dependent upon the interaction of pH with other factors. Studies on inactivation of E. coli O157:H7 with organic acid sprays on beef using acetic, citric, or lactic acid at concentrations of up to 1.5% revealed that E. coli O157:H7 populations were not appreciably affected by any of the treatments (9). E. coli O157:H7, when inoculated at high populations, survived fermentation, drying, and storage in fermented sausage (pH 4.5) for up to 2 months at 4°C (45), in mayonnaise (pH 3.6 to 3.9) for 5 to 7 weeks at 5°C and for 1 to 3 weeks at 20°C (114), and in apple cider (pH 3.6 to 4.0) for 10 to 31 days or 2 to 3 days at 8 or 25°C (115). Induction of acid resistance in E. coli O157:H7 also can increase tolerance to other environmental stresses, such as heat, radiation, and some antimicrobials. Studies (3) have compared the survival characteristics of E. coli O157:H7 and other EHEC (O26:H11 and O111:NM) in chocolate and confectionery products during storage at different temperatures. Results revealed that all three serotypes survived storage at 38°C for up to 43 days, but after 90 days, only E. coli O26:

Foodborne Pathogenic Bacteria H11 and O111 were recovered. However, E. coli O157: H7 was recovered after O26 and O111 were no longer detected when a similar study was conducted with biscuit cream and mallow. The determination of the desiccation tolerance with 15 strains of E. coli O157:H7, 15 strains of E. coli O26:H11, and 5 strains of E. coli O111:NM revealed that all of them survived on paper disks after 24 h of drying at 35°C, showing no difference among serotypes (49).

Antimicrobial Resistance

Initially, when E. coli O157:H7 was first associated with human illness, the pathogen was susceptible to most antimicrobials affecting gram-negative bacteria (59). Several studies revealed a trend toward increasing resistance to antimicrobials among E. coli O157: H7 isolates (74, 100). Overall, antimicrobial resistance among E. coli O157:H7 clinical isolates is low compared to other enteric pathogens and had no significant change from 1998 to 2007 (http://www.cdc. gov/narms/annual/2007/NARMSAnnualReport2007. pdf). However, resistance to clinically important antimicrobials has been reported; 2.1% (4/190) of E. coli O157 clinical isolates in 2007 were resistant to nalidixic acid, and a single (0.5%) isolate was resistant to ciprofloxacin. Some E. coli O157:H7 strains isolated from humans, animals, and food have developed resistance to multiple antimicrobials, with streptomycin-sulfisoxazole-tetracycline being the most common resistance profile. Approximately 2% of the E. coli O157 clinical isolates in 2007 were resistant to three or more classes. Non-O157 EHEC strains isolated from humans and animals also have acquired antimicrobial resistance, and some are resistant to multiple antimicrobials commonly used in human and veterinary medicine (99). However, antimicrobial resistance among EHEC/STEC was low compared to non-STEC E. coli strains (100).

Inactivation by Heat and Irradiation

Studies on the thermal sensitivity of E. coli O157:H7 in ground beef revealed that the pathogen has no unusual resistance to heat, with D values at 57.2, 60, 62.8, and 64.3°C of 270, 45, 24, and 9.6 seconds, respectively (33). Heating ground beef sufficiently to kill typical strains of Salmonella will also kill E. coli O157:H7 (Table 12.2). The presence of fat protects E. coli O157:H7 in ground beef, with D values for lean (2.0% fat) and fatty (30.5% fat) ground beef of 4.1 and 5.3 min at 57.2°C, respectively, and 0.3 and 0.5 min at 62.8°C, respectively (67). Pasteurization of milk (at 72°C for 16.2 s) is an effective treatment that will kill more than 104

12.  Enterohemorrhagic Escherichia coli Table 12.2  Comparison of D values for E. coli O157:H7

and Salmonella spp. in ground beef

D value (min) E. coli O157:H7 Temp (°C) 51.7 57.2 62.8 a

30.5% fat

17–20% fat

Salmonella spp.

115.5 5.3 0.47

NDa 4.5 0.40

54.3 5.43 0.54

ND, not determined.

E. coli O157:H7 cells per ml (27). Proper heating of foods of animal origin, e.g., heating foods to an internal temperature of at least 68.3°C for several seconds, is an important critical control point to ensure inactivation of E. coli O157:H7. The use of irradiation to eliminate foodborne pathogens in food has been approved by many countries. Unlike many other processing technologies, irradiation at dosages that kill enteric foodborne pathogens still maintains the raw character of foods. In the United States, an irradiation dose of 4.5 kGy is approved for refrigerated, and 7.5 kGy for frozen, raw ground beef. D10 values for E. coli O157:H7 in raw ground beef patties range from 0.241 to 0.307 kGy, depending on temperature, with D10 values significantly higher for patties irradiated at −16°C than at 4°C (24). Hence, an irradiation dose of 1.5 kGy should be sufficient to eliminate E. coli O157:H7 at the cell numbers likely to occur in ground beef. At present, there is no reason to believe that current interventions used in foods for mitigating Salmonella and E. coli O157:H7 contamination would not be effective against non-O157 EHEC.

Comparative Genomics of EHEC

The chromosome of E. coli O157:H7 consists of 4.1Mb backbone sequences shared by E. coli K-12, and 1.4 Mb O157-specific sequences encoding many virulence determinants, such as Stx genes (stx) and LEE (83). Genomic comparison between EHEC strains of serotypes O26, O111, and O103 reveals that similar to O157, all EHEC have larger genomes (5.5 to 5.9 Mb) than E. coli K-12 (4.6 Mb) and contain a large number of mobile elements such as prophages and integrative elements. However, the chromosomal backbone regions are highly conserved as well among non-O157 EHEC strains of O26, O111, and O103 (82). Virulence genes, especially those for non-LEE effectors and nonfimbrial adhesions, are well conserved in non-O157 EHEC in addition to the stx genes and LEE island. They have a great similarity to their whole gene repertoire and share many genes that are specific to

291 EHEC or rarely present in other pathotypes (83). These genes are directly or indirectly related to virulence, thus conferring a similar virulence potential among EHEC strains. It is noteworthy that, despite carrying the same or similar virulence genes, mobile elements that are commonly present in EHEC (multiple lambdoid PPs, several types of integrative elements, and virulence plasmids) have remarkably divergent genomic structures. This property suggests that EHEC strains have complex and independent evolutionary pathways and that mobile elements are the primary driving force for the parallel evolution of EHEC (83).

RESERVOIRS OF E. COLI O157:H7 AND NON-O157 EHEC

Cattle

Initially, foods of bovine origin, notably undercooked ground beef, and less frequently unpasteurized milk, were the vehicles most frequently associated with outbreaks of E. coli O157:H7 infection. Subsequently, cattle were identified as important sources, or reservoirs, of this pathogen. Since that time, cattle have been the focus of many studies on their role as a reservoir of E. coli O157:H7. A number of other vehicles have since been implicated in E. coli O157 infections, including fresh produce, contaminated water, and direct contact with animals or their environment at livestock exhibitions and by petting. The sources and reservoirs of nonO157 STEC infections are not as clearly defined. Unlike E. coli O157, most major outbreaks of non-O157 STEC have not been directly associated with beef products. Instead, outbreaks have been traced to a wider variety of sources, including vegetables, water, and unpasteurized milk (70). Given that non-O157 STEC also colonize live cattle, it is probable that cattle play an important role in the epidemiology of human non-O157 STEC as well. However, epidemiologic data to pinpoint the primary routes of human exposure are lacking. Recently, meat or meat products have been implicated in human outbreaks (36–38). As isolation and detection methods for non-O157 STEC are improved, the understanding of the vehicles and routes of non-O157 STEC should improve.

Detection of E. coli O157:H7 and STEC on Farms

The first reported isolation of E. coli O157:H7 from cattle was from a <3-week-old calf with colibacillosis in Argentina in 1977 (85). However, this presentation was

292 atypical, as the clonal genotypic group of E. coli O157 most frequently associated with human disease rarely causes bovine illness. Instead, most cattle harbor these bacteria without outward signs of illness or loss of productivity (8). The prevalence of fecal excretion of E. coli O157:H7 varies by age, with higher prevalence values reported in younger animals (2 to 24 months of age) than adults in field studies. The reasons for age-related differences are unclear, but it may be due to ruminal development differences, differences in microbial flora in gastrointestinal tract, or management differences such as dietary factors (75). Nevertheless, older animals, including those at the time of harvest and during ­lactation, may also shed these bacteria in their feces asymptomatically. The presence of more than one strain of E. coli O157:H7 on a single farm on a single sample date has been described (92). Shedding of E. coli O157 at the individual animal level typically lasts for a few days to several weeks following exposure (6). However, there is an association with the excretion of larger numbers of bacteria and for longer periods of time if the bacteria are intimately attached to the intestinal mucosa, a phenomenon that occurs predominately, if not exclusively, at the rectoanal junction in cattle (79). Cattle may excrete E. coli O157 in the feces at cell numbers that are so low as to be detectable only through sensitive enrichment culture methods or are as high as 106 CFU/gram. Cattle shedding high numbers of E. coli O157 (>103 CFU/g) can contribute substantially to contamination of carcasses at harvest, the environment, and cattle-to-cattle transmission and are considered to be “supershedding” (22, 29). The factors that govern bovine supershedding of E. coli O157 (and whether the phenomenon occurs for non-O157 STEC) are currently poorly understood and under investigation. At the herd level, most bovine populations are positive for E. coli O157 and non-O157 STEC at some time or another (47). Prevalence, however, is variable, and peaks in prevalence are sporadic and currently unpredictable. Fecal excretion of E. coli O157:H7 by cattle occurs in a seasonal pattern, with higher prevalence occurring in the summertime or early fall, which coincides with the seasonal variation in disease incidence seen in humans, with higher rates also observed during the summer months. In nine herds sampled for approximately 1 year, the prevalence of E. coli O157: H7 during the months of June through October was several times that observed in December through March. Observed seasonal effects of E. coli O157:H7 excretion in cattle could be due to confounding factors, such as differences in the microbial flora of the

Foodborne Pathogenic Bacteria gastrointestinal tract in cattle during the summer and during the winter months, due to changes in diet, or related to conditions conducive for multiplication of the bacteria in environmental niches. Non-O157 STEC are suspected to follow a similar seasonal pattern of colonization in cattle, but this has not been extensively documented (2). Herd prevalence rates of STEC fluctuate between 0 and 100% (63). Although total STEC prevalence in a herd may average 60 to 70%, the fraction of these strains that actually pose a threat to public health is undetermined. Many STEC isolated from cattle carry only the stx gene and no ancillary virulence genes typically found in cases of human disease (91, 116). A study on STEC carriage by dairy cattle on farms in Canada revealed 36% of cows and 57% of calves were STEC positive in all of the 80 herds tested (59). Of these, only seven animals (0.45%) on four farms (5%) were positive for E. coli O157:H7. A 2002 U.S. Department of Agriculture national study revealed that 38.5% of dairy farms had at least one E. coli O157:H7-positive cow and that 4.3% of individual cows were E. coli O157 positive.

Factors Associated with Bovine Carriage of E. coli O157:H7

E. coli O157:H7 has been isolated from cattle feces from most regions of the world in which studies have been conducted (47). There are, however, other regions, such as Scandinavia, Africa, and Norway, that report lower prevalence rates than others. This may be due to climate factors, or farm management practices less conducive to cattle being colonized with E. coli O157:H7, or the possibility that E. coli O157 has not yet been introduced into these regions.

Domestic Animals and Wildlife

Although cattle contribute significantly to STEC contamination of the food chain, either directly through meat or milk or indirectly through contamination of water and the food production environment, STEC are also frequently isolated from many other domestic and wild ruminants such as sheep, goats, deer, and water buffalo (8, 108). In addition, E. coli O157 and other non-O157 STEC can colonize a number of other animals including dogs, horses, swine, wild birds, and rodents (80), albeit less commonly than occurs in ruminants. Foodborne disease outbreaks of STEC infections have been associated with food products derived from sheep, goats, and deer. Moreover, nonruminant species may play a role in transmission of STEC between cattle farms and contamination of the environment (and

12.  Enterohemorrhagic Escherichia coli crops) and may constitute a source for direct transmission routes of exposure.

Possibility of Control of STEC in Food Animals

Despite 25 years of research on the topic and their potential for enhancing food safety, very few effective control measures for STEC in live animals have been identified. Several tentative associations between fecal shedding of E. coli O157:H7 and feed or environmental factors have been made from epidemiologic studies of dairy herds. For example, some calf starter feed regimes or environmental factors and feed components such as whole cottonseed were associated with reduced prevalence of E. coli O157:H7. Feeding of distillers’ grains or barley results in increased E. coli O157 shedding compared to corn-fed cattle, but the mechanisms driving these differences are unknown (51, 52). Likewise, grouping calves before weaning is associated with increased carriage of E. coli O157:H7 (25). Given that E. coli O157 can survive for several months to years in environmental niches on the farm, food and water hygiene and manure handling have been researched (95). Several studies have revealed that E. coli O157:H7 can survive for weeks and months in bovine feces and water (62). The pathogen was frequently isolated from water troughs on farms. Commercial feeds often contained detectable E. coli, indicating widespread fecal contamination, although E. coli O157:H7 was only infrequently detected (31, 69). Despite the gaps in understanding the factors influencing E. coli O157 carriage in cattle, preharvest interventions have been applied to live cattle with mixed results (64). Feeding of specific probiotic bacteria has repeatedly resulted in decreased prevalence of E. coli O157 in feedlot cattle (10, 107). Likewise, the administration of sodium chlorate in the feed or water may provide a control method immediately prior to harvest (17). Vaccination of cattle to control E. coli O157 colonization has also been studied (71, 102). The last strategy for control is promising, but sufficient data demonstrating the efficacy of current vaccines are lacking. Other possible control measures include bacteriophage therapy and washing the hides of cattle. Achieving a better understanding of the factors influencing the exposure and colonization of cattle with E. coli O157 and other STEC will enhance the development of novel control strategies. Many interventions have also been applied at the time of harvest to mitigate the potential negative impacts of cattle entering beef processing facilities carrying STEC. These include strict attention to slaughter and processing hygiene as well as postharvest interventions such as surface steam pasteurization and application of acid rinses

293 on carcasses (64). Although preliminary studies clearly identified a direct correlation between bovine prevalence in the live animal and contamination rates of carcasses, modern processing techniques can reduce and mitigate the impacts E. coli O157 and other STEC carriage among animals presented for harvest (34). Presently, it is thought that most animals can be processed safely, and it is only when the prevalence is extremely high or the magnitude of shedding is large (supershedding cattle) that these factors overwhelm the current system and contamination persists on carcasses or in product.

Humans

Fecal shedding of E. coli O157:H7 by patients with hemorrhagic colitis or HUS usually lasts for no more than 13 to 21 days following onset of symptoms (56). However, in some instances, the pathogen can be excreted in feces for many weeks. A child infected during a day care center-associated outbreak continued to ­ excrete the pathogen for 62 days after the onset of diarrhea (84). Studies of persons living on dairy farms, aiming to determine carriage of E. coli O157:H7 by farm families, revealed elevated antibody titers against the surface antigens of E. coli O157; however, the pathogen was not isolated from feces (112). An asymptomatic long-term carrier state has not been identified. The significance of fecal carriage of E. coli O157:H7 by humans is the potential for person-to-person dissemination of the pathogen, a situation which has been observed repeatedly in outbreak settings. A contributing factor to person-toperson transmission of the pathogen is its extraordinarily low infectious dose, estimated at <100 cells, and possibly as few as 10 cells can produce illness in highly susceptible populations (105). Inadequate attention to personal hygiene, especially after using the bathroom, can transfer the pathogen to other persons through contaminated hands, resulting in secondary transmission.

CHARACTERISTICS OF DISEASE The spectrum of human illness of E. coli O157:H7 infection includes nonbloody diarrhea, hemorrhagic colitis, and HUS. Some persons may be infected but asymptomatic, but typically for a short time (<3 weeks). Ingestion of the bacteria is followed typically by a 3- to 4-day incubation period (range, 2 to 8 days), during which colonization of the large bowel occurs. Illness typically begins with severe abdominal cramps and nonbloody diarrhea for 1 to 2 days, which then progresses in the second or third day of illness to bloody diarrhea that lasts for 4 to 10 days (5, 103). Many outbreak investigations revealed that more than 80% of microbiologically

Foodborne Pathogenic Bacteria

294

INFECTIOUS DOSE Retrospective analyses of foods associated with outbreaks of EHEC infection revealed that the infectious dose is very low. For example, between 0.3 and 15 CFU of E. coli O157:H7 per gram was enumerated in lots of frozen ground beef patties associated with a 1993 multistate outbreak in the western United States. Similarly, 0.3 to 0.4 CFU of E. coli O157:H7 per gram was detected in several intact packages of salami that were associated with a foodborne outbreak. These data suggest that the infectious dose of E. coli O157:H7 may be fewer than 100 cells. In an outbreak of E. coli O111:NM infection in Australia, the implicated salami was estimated to contain less than one cell per 10 g. Additional evidence for a low infectious dose is the capability for person-to-­person and waterborne transmission of EHEC infection.

DISEASE OUTBREAKS

Geographic Distribution

E. coli O157:H7 has been the cause of many major outbreaks of severe illness worldwide. At least 30 countries

on six continents have reported E. coli O157:H7 infection in humans. In the United States, 264 outbreaks of E. coli O157:H7 infection associated with food were documented between 2000 and 2010 (Table 12.3). These outbreaks contributed to 5,875 illnesses during this time. More importantly, most cases of STEC infections occur as sporadic cases. Over the last 10 years, the number of actual cases reported each year in the United States averaged about 4,000 (Fig. 12.1) (http://www. cdc.gov/outbreaknet/outbreaks.html). The precise incidence of E. coli O157:H7 foodborne illness in the United States is not known because infected persons presenting mild or no symptoms and persons with nonbloody diarrhea are less likely to seek medical attention than patients with bloody diarrhea; hence, such cases would not be reported. The CDC reports that the annual averages of laboratory-confirmed cases of E. coli O157:H7 and non-O157 infections are 3,704 and 1,579, respectively (96). It also estimates that E. coli O157:H7 causes 63,153 illnesses each year in the United States, and non-O157 STEC account for an additional 112,752 cases. Sixty-eight percent of O157 and 82% of non-157 cases are attributed to foodborne transmission. Large outbreaks of E. coli O157:H7 infections involving hundreds of cases also have been reported in Canada, Japan, and the United Kingdom. The largest outbreak occurred in May to December 1996 in Japan, involving more than 9,000 reported cases. In the same year, 21 elderly people died in a large outbreak involving 501 cases in central Scotland. Although E. coli O157:H7 is still the predominant serotype of EHEC in the United States, Canada, the United Kingdom, and Japan, an increasing number of outbreaks and sporadic cases related to EHEC of serotypes other than O157:H7 have been reported. A large epidemic involving several thousand cases of E. coli O157:NM infection occurred in Swaziland and South Africa following 6000 No. Reported Cases

confirmed cases of diarrhea caused by E. coli O157:H7 showed frank blood in the stools, but in some outbreaks there have been reports of 30% of cases with nonbloody diarrhea. Symptoms usually resolve after a week, but about 6% of patients progress to HUS, one-half of whom require dialysis, and 75% require transfusions of erythrocytes and/or platelets. The case-fatality rate from E. coli O157:H7 infection is about 1%. Similar but less severe symptoms have been observed in infections with non-O157 EHEC: only 45% of cases develop bloody diarrhea, and fewer than 2% progress to HUS. HUS largely affects children, among whom it is the leading cause of acute renal failure. The risk that a child younger than 10 years of age with a diagnosed E. coli O157:H7 infection will develop HUS is about 15% (103). The syndrome is characterized by a triad of features: acute renal insufficiency, microangiopathic hemolytic anemia, and thrombocytopenia. Significant pathological changes include swelling of endothelial cells, widened subendothelial regions, and hypertrophied mesangial cells between glomerular capillaries. These changes combine to narrow the lumina of the glomerular capillaries and afferent arterioles and result in thrombosis of the arteriolar and glomerular microcirculation. Complete obstruction of renal microvessels can produce glomerular and tubular necrosis, with an increased probability of subsequent hypertension or renal failure.

5000 4000 3000 2000 1000 0

00

01

02

03

04

05 06 Year

07

08

09

10

Figure 12.1  Number of Stx-producing Escherichia coli O157:H7 cases in the United States by year, 2000 to 2010. doi:10.1128/9781555818463.ch12f1

12.  Enterohemorrhagic Escherichia coli

295

Table 12.3  Vehicles of foodborne outbreaks and associated cases of E. coli O157 infections in the United States

between 2000 and 2010a Transmission route

Ground beef Unknown food vehicle Produce Other beef Other food vehicle Dairy product Ground beef; other beef Ground beef; other food vehicle Ground beef; produce Ground beef; produce; other beef Other food vehicle; produce Total a

No. of outbreaks

% of total outbreaks

No. of cases associated with outbreaks

% of total cases

61 93 29 18 45 11 1 2 1 1 2 264

23.1 35.2 11.0 6.8 17.1 4.2 0.38 0.76 0.38 0.38 0.76

1,022 1,549 832 174 1,772 430 8 24 5 3 56 5,875

17.4 26.4 14.2 3.0 30.2 7.3 0.14 0.41 0.09 0.05 0.95

Data from the CDC, http://www.cdc.gov/outbreaknet/outbreaks.html.

c­ onsumption of contaminated surface water. In continental Europe, Australia, and Latin America, non-O157 EHEC infections are more common than E. coli O157: H7 infections. Details of many reported foodborne and waterborne outbreaks of EHEC infections are provided in Table 12.4. There are no distinguishing biochemical phenotypes for non-O157 EHEC, making screening for these bacteria problematic and labor-intensive, and for this reason only a limited number of clinical laboratories test for them. Therefore, the prevalence of non-O157 EHEC infections may be underestimated.

Seasonality of E. coli O157:H7

Outbreaks and clusters of E. coli O157:H7 infections peak during the warmest months of the year. Approximately 89% of the outbreaks reported in the United States occurred from May to November (90). FoodNet data indicate the same trend. The reasons for this seasonal pattern are unknown but may include (i) an increased prevalence of contaminated product on the market (due to increased prevalence in cattle), (ii) increased exposure to E. coli O157 from nonbeef sources during the summer (111), and (iii) increased exposure from consumption of vegetables or from recreational contact with the environment.

Age of Patients

All age groups can be infected by E. coli O157:H7, but the very young and the elderly most frequently experience severe illness with complications (61). HUS usually occurs in children. Population-based studies have suggested that the highest age-specific incidence of E. coli O157:H7 infection occurs in children 2 to 10 years of age. In addition to naïve or incompletely developed immune responses,

the high rate of infection in this age group may be attributable to more frequent exposure to contaminated environments, infected animals, and more opportunities for person-to-person spread between infected children with relatively undeveloped hygiene skills.

Transmission of E. coli O157:H7

Food remained the predominant transmission route, accounting for 52% of 350 outbreaks and 61% of 8,598 outbreak-related cases from 1982 to 2002 (90). A variety of foods have been identified as vehicles of E. coli O157:H7 infections, although ground beef is one of the most frequent food vehicles. Examples of other foods that have been implicated in outbreaks include roast beef, cooked meats, venison meat and jerky, salami, raw milk, pasteurized milk, yogurt, cheese, ice cream bars, lettuce, prepackaged spinach, unpasteurized apple cider/juice, cantaloupe, potatoes, radish sprouts, alfalfa sprouts, fruit/vegetable salad, cookie dough, pepperoni pizzas, and cake (73). Among 264 foodborne outbreaks, the food vehicle in 61 (23%) was ground beef, in 93 (35%) was unknown, in 29 (11%) was produce, in 18 (7%) was other beef, in 45 (17%) was other foods, and in 11 (4%) was dairy products (Table 12.3). The route of E. coli O157:H7 transmission for many outbreaks was unknown. Outbreaks attributed to transmission by person-to-person contact, water, animal contact, and laboratory exposure have also been reported. An outbreak investigation reported that a petting zooassociated E. coli O157:H7 infection subsequently caused secondary transmission, asymptomatic infection, and prolonged shedding in the classroom (28). In contrast to E. coli O157:H7 outbreaks, in which a food is most often identified as a vehicle, the modes of transmission of most

Foodborne Pathogenic Bacteria

296

Table 12.4  Representative foodborne and waterborne outbreaks of E. coli O157:H7 and other EHEC infectionsa Yr

Month

Locationb

1982 1982 1985 1987

2 5

1988 1989 1990 1991 1991 1992c 1992 1993

10 12 7 11 7

1993 1993 1994d 1994 1995e 1995 1995 1995 1995 1996f 1996 1996 1996

7 8 2 11 2 10 11 7 9

1996 1997 1997 1997 1998 1998 1998 1998 1998 1998 1999g 1999h 1999 1999 2002 2005 2006 2006i 2007j 2007

11 5 6 11 6 6 7 7 8 9 7 7 8 10 8 9, 10 8, 9 2, 4 2, 5 6

Minnesota Missouri North Dakota Massachusetts Oregon France Oregon California, Idaho, Nevada, and Washington Washington Oregon Montana Washington, California Adelaide, Australia Kansas Oregon Montana Maine Komatsu, Japan Connecticut, Illinois Japan California, Washington, Colorado Central Scotland, UK Illinois Michigan, Virginia Wisconsin Wisconsin Wyoming North Carolina California New York California Texas Connecticut New York Ohio, Indiana Washington Minnesota 26 states Norway Denmark United Kingdom

2008e 2009k

8 2

Oklahoma France

6

12 1

5,6 7 10

Oregon Michigan Canada Utah

No. of cases/no. of deaths 26 21 73/17 51

Setting

Vehicle/transmission mode

54 243 65 23 21 >4 9 732/4

Community Community Nursing home Custodial institution School Community Community Community Community Community Community Restaurant

Ground beef Ground beef Sandwiches Ground beef/ person-to-person Precooked ground beef Water Roast beef Apple cider Swimming water Goat cheese Raw milk Ground beef

16 27 18 19 >200 21 11 74 37 126 47 9,451/12 71/1

Church picnic Restaurant Community Home Community Wedding Home Community Camp School Community Community Community

Pea salad Cantaloupe Milk Salami Semidry sausage Punch/fruit salad Venison jerky Leaf lettuce Lettuce Luncheon Mesclun lettuce White radish sprouts Apple juice

501/21 3 108 13 63 114 142 28 11 20 56 11 900/2 47 32 23 199/3 17/1 20 12

Community School Community Church banquet Community Community Restaurant Prison Deli Church Camp Community Fair Community Camp Community Community Community Community Community

341/1 2

Community Home

Cooked meat Ice cream bar Alfalfa sprouts Meatballs/coleslaw Cheese curds Water Coleslaw Milk Macaroni salad Cake Salad bar Lake water Well water Lettuce Romaine lettuce Prepackaged lettuce Prepackaged spinach Mutton Fermented beef sausage Ready-to-eat chicken wrap Food handler Ground beef

(Continued)

12.  Enterohemorrhagic Escherichia coli

297

Table 12.4  Representative foodborne and waterborne outbreaks of E. coli O157:H7 and other EHEC infectionsa (Continued) Yr

Month

Locationb

2010l

3, 5

2011m

5, 6

Michigan, New York, Ohio, Pennsylvania, and Tennessee Germany

No. of cases/no. of deaths

Setting

Vehicle/transmission mode

26

Community/food service

Romaine lettuce

>3,700

Community/food service

Sprouts

E. coli O157:H7 unless otherwise noted. State of the United States unless otherwise noted. c E. coli O119. d E. coli O104:H21. e E. coli O111:NM. f E.coli O118:H2. g E. coli O111:H8. h E. coli O121:H19. i E. coli O103:H25. j E. coli O26:H11. k E. coli O123:H-. l E. coli O145. m E. coli O104:H4. a

b

outbreaks caused by non-O157 EHEC are unknown (12, 70). Only a few outbreaks of non-O157 EHEC have been clearly associated with foods/water (Table 12.4).

Examples of Foodborne and Waterborne Outbreaks The Original Outbreaks

The first documented outbreak of E. coli O157:H7 infection occurred in the state of Oregon in 1982, with 26 cases and 19 persons hospitalized (110). All patients had bloody diarrhea and severe abdominal pain. The median age was 28 years, with a range of 8 to 76 years. The duration of illness ranged from 2 to 9 days, with a median of 4 days. This outbreak was associated with eating undercooked hamburgers from fast-food restaurants of a specific chain. E. coli O157:H7 was recovered from stools of patients. A second outbreak followed 3 months later and was associated with the same fast-food restaurant chain in Michigan, with 21 cases and 14 persons hospitalized. The median age was 17 years, with a range of 4 to 58 years. Contaminated hamburgers again were implicated as the vehicle, and E. coli O157:H7 was isolated both from patients and from a frozen ground beef patty. That E. coli O157:H7, a heretofore unknown human pathogen, was the causative agent was established by its association with the food and recovery of the bacterium with identical microbiologic characteristics from both the patients and the meat from the implicated supplier.

1993 Multistate Outbreak

A large multistate outbreak of E. coli O157:H7 infection in the United States occurred in Washington, Idaho,

California, and Nevada in early 1993 (4). Approximately 90% of primary cases were associated with eating at a single fast-food restaurant chain (chain A), from which E. coli O157:H7 was isolated from hamburger patties. Transmission was amplified by secondary spread (48 patients in Washington alone) via person-to-person transmission. In total, 731 cases were identified, with 629 in Washington, 13 in Idaho, 57 in Las Vegas, NV, and 34 in Southern California. The median age of patients was 11 years, with a range of 4 months to 88 years. One hundred seventy-eight persons were hospitalized, 56 developed HUS, and 4 children died. Because neither specific laboratory testing nor surveillance for E. coli O157:H7 was carried out for earlier cases in Nevada, Idaho, and California, the outbreak went unrecognized until a sharp increase in cases of HUS was identified and investigated in the state of Washington. The outbreak resulted from insufficient cooking of hamburgers by chain A restaurants. Epidemiologic investigation revealed that 10 of 16 hamburgers cooked according to chain A’s cooking procedures in Washington State had internal temperatures below 60°C, which was substantially less than the minimum internal temperature of 68.3°C required by the state of Washington. Cooking patties to an internal temperature of 68.3°C would have been sufficient to kill the low populations of E. coli O157: H7 detected in the contaminated ground beef.

Outbreaks Associated with Produce

Produce-associated outbreaks of E. coli O157:H7 infection were first reported in 1991, and produce has remained a prominent food vehicle. Raw vegetables,

298 particularly lettuce and alfalfa and vegetable sprouts, have been implicated in several outbreaks of E. coli O157:H7 infection in North America, Europe, and Japan. In May 1996, a mesclun mix of organic lettuce was associated with a multistate outbreak in which 47 cases were identified in Illinois and Connecticut. A large multistate (26 states) outbreak of E. coli O157:H7 occurred in the summer of 2006 (19). A total of 199 persons were infected, with 102 (51%) patients hospitalized, 31 (16%) cases of HUS, and three deaths. E. coli O157 was isolated from 13 packages of spinach supplied by patients living in 10 states. Between May and December 1996, multiple outbreaks of E. coli O157:H7 infection occurred in Japan, involving 9,451 cases and 12 deaths (76). The largest outbreak affected 7,470 schoolchildren, teachers, and staff in Osaka in July 1996. Epidemiologic investigations revealed that white radish sprouts were the vehicle of transmission.

Apple Cider/Juice Outbreaks

The first confirmed outbreak of E. coli O157:H7 infection associated with apple cider occurred in Massachusetts in 1991, involving 23 cases. In 1996, three outbreaks of E. coli O157:H7 infection associated with unpasteurized apple juice/cider were reported in the United States. The largest of the three occurred in three western states (California, Colorado, and Washington) and in British Columbia, Canada, with 71 confirmed cases and one death. E. coli O157:H7 was isolated from the implicated apple juice. An outbreak also occurred in Connecticut, with 14 confirmed cases. Manure contamination of apples was the suspected source of E. coli O157:H7 in several of the outbreaks. Using apple drops (i.e., apples picked up from the ground) for making apple cider is a common practice, and apples can become contaminated by resting on soil contaminated with manure. Apples can also become contaminated if they are transported or stored in areas that contain manure or are treated with contaminated water. Investigation of the 1991 outbreak in Massachusetts revealed that the implicated cider press processor also raised cattle that grazed in a field adjacent to the cider mill. Fecal droppings from deer also were found in the orchard where apples used to make the cider were harvested.

Waterborne Outbreaks

Reported waterborne outbreaks of E. coli O157:H7 infection have increased substantially in recent years, being associated with swimming water, drinking water, well water, and ice. Investigations of lake-associated outbreaks revealed that in some instances the water was

Foodborne Pathogenic Bacteria likely contaminated with E. coli O157:H7 by toddlers defecating while swimming and that swallowing lake water was subsequently identified as the risk factor. A 1995 outbreak in Illinois involved 12 children ranging in age from 2 to 12 years (20). Although E. coli O157: H7 was not recovered from water samples, high levels of E. coli were detected, indicating likely fecal contamination. A large waterborne outbreak of E. coli O157:H7 among attendees of a county fair in New York occurred in August 1999 (21, 65). More than 900 persons were infected, of which 65 were hospitalized. Campylobacter jejuni also was identified in some patients. Two persons died including a 3-year-old girl, from HUS, and a 79year-old man, from HUS/thrombotic thrombocytopenic purpura. Unchlorinated well water used to make beverages and ice was the vehicle. Recreational water exposure is responsible for many cases of E. coli O157:H7 infections (30). Waterborne outbreaks of E. coli O157 infections have also been reported in other locations of the world. Drinking water, which was probably contaminated with bovine feces, was implicated in outbreaks in Scotland (65) and southern Africa (77). E. coli O157:NM was isolated from water associated with the latter outbreak. Walkerton, Ontario, Canada, was the site of one of the largest waterborne disease outbreaks associated with E. coli O157:H7 (50). In this community, 2,300 people were infected and 7 of them died.

Outbreaks of Non-O157 EHEC

Several outbreaks of non-O157 EHEC infections have been reported worldwide. An outbreak in early 1995 in South Australia was associated with E. coli O111: NM and involved 23 cases of HUS after consumption of an uncooked, semidry fermented sausage product. In June 1999, an outbreak of E. coli O111:H8 ­ involving 58 cases occurred at a teenage cheerleading camp in Texas. Contaminated ice was the implicated vehicle. More recently, a restaurant-associated E. coli O111: NM outbreak in Oklahoma during late August and early September 2008 was reported (16, 89). The outbreak caused 341 cases, 70 hospitalizations, and one death. The exact source of the contamination was undetermined, but contamination by a food handler was suspected. Several other outbreaks caused by non-O157 EHEC have also been reported. EHEC O103:H25 was the cause of an outbreak associated with fermented sausage in Norway in 2006 (101). EHEC O123:H- was identified as the causative agent in a family outbreak associated with eating undercooked ground beef in France in 2009 (60), whereas EHEC O26 sickened several

12.  Enterohemorrhagic Escherichia coli individuals in Maine and New York in 2010, leading to a large recall of ground beef (http://www.fsis.usda. gov/News_&_Events/Recall_050_2010_Release/index. asp). A multistate outbreak of EHEC O145 infections occurred in May 2010 in the United States, with more than 30 cases reported from five states (http://www. cdc.gov/ecoli/2010/ecoli_o145/index.html). Shredded romaine lettuce from one processing facility was identified as a source of infection in this outbreak. An outbreak associated with multiple EHEC serotypes (O26, O84, and O121) occurred in a Colorado prison in 2007, involving 135 cases and 10 hospitalizations. Pasteurized cheese and margarine were the food vehicles of the outbreak. A foodborne outbreak in Germany in May and June 2011 sickened more than 3,700 people and caused more than 50 deaths (40, 93). The causative agent was identified as an E. coli O104:H4 that produced Stx2a. Approximately 24% of the patients developed HUS, which was much higher than previously reported rates for patients infected with EHEC. The outbreak spread quickly over northern Germany with some cases in other European countries, the United States, and Canada and has become one of the largest outbreaks of E. coli infections reported to date. Six confirmed cases of O104: H4 infections were identified in the United States. An Arizona resident who traveled to Germany before becoming ill died. HUS is most commonly triggered by EHEC. However, according to the whole-genome sequencing analysis, the outbreak strain was genetically more related to EAEC, which is associated with cases of acute or persistent diarrhea worldwide in children and adults (see “EAEC” above).

MECHANISMS OF PATHOGENICITY Significant virulence factors associated with the pathogenicity of EHEC have been identified based on histopathology of tissues of HUS and hemorrhagic colitis patients, studies with tissue culture and animal models, and studies using cell biology and molecular genetics approaches. A general body of knowledge of the pathogenicity of EHEC has been developed and indicates that the bacteria cause disease by their ability to adhere to the host cell membrane and colonize the large intestine and then produce one or more Stxs.

Attaching and Effacing

Numerous studies on the pathogenesis of EHEC have focused on elucidating the mechanisms of adherence and colonization. By adhering to intestinal epithelial cells, EHEC subvert cytoskeletal processes to produce a histopathological feature known as an attaching-and-effacing

299

(A) EHEC Type III Secretion Machinery EspA

Tir

Tir

EspD/B

EspF Others

Map EsoG

Host cell

(B)

EHEC Pedestal

Intimin

Tir

Tir

WASP WASP

Tir

Arp2/3 Arp2/3

F-actin & Cytoskeletal proteins

Host cell Figure 12.2  Schematic illustration of A/E lesion formation in EHEC, modified from reference 18. (A) A/E translocation of effector proteins through T3SS that forms a pore through the membranes of EHEC. EHEC translocate a number of proteins: EspB and EspD, which form a translocon in the plasma membrane; the cytoplasmic proteins EspF, G, and Map; the translocated intimin receptor Tir, which inserts into the plasma membrane; and other unidentified effectors. (B) Formation of EHEC pedestal. EHEC intimately attaches to the host cell through intimin-Tir binding. The binding triggers the formation of actin-rich pedestals beneath adherent bacteria after Wiskott-Aldrich syndrome protein (WASP) and the heptameric actin-related protein Arp2/3 are recruited to the pedestal tip. doi:10.1128/9781555818463.ch12f2

(A/E) lesion (Fig. 12.2). E. coli O157:H7 produces an A/E lesion in the large intestine similar to that induced by EPEC, which in contrast occurs predominantly in the small intestine. The A/E lesion is characterized by intimate attachment of the bacteria to the plasma membranes of the host epithelial cells, localized destruction of the brush border microvilli, and assembly of highly organized pedestallike actin structures (44). Most EHEC strains contain a ca. 43-kb pathogenicity island, LEE (Fig. 12.3). LEE is

Foodborne Pathogenic Bacteria

300

Figure 12.3  Genetic organization of the EHEC LEE and EHEC prophages CP-933U, CP933K, and CP-933P, reproduced from reference 44. doi:10.1128/9781555818463.ch12f3

organized into five major perons, LEE1 to LEE5, encoding a type III secretion system (T3SS), secreted proteins, chaperones, and regulators (55). The secreted proteins consist of effectors that are translocated into the host cell by the T3SS and translocators required for delivering the effectors.

Type III Secretion System

T3SS is associated with the virulence of many gramnegative bacterial pathogens. The T3SS apparatus (Fig. 12.4) is a complex “needle and syringe” structure that is assembled from the products of approximately 20 genes in LEE (35, 44). The system is used by EHEC to directly translocate virulence factors from the bacteria into the targeted host cells in a single step. The genes encoding structural proteins of the T3SS are largely conserved, whereas genes encoding effector proteins display substantial variability (98). The conserved T3SS gene cluster in LEE is likely acquired by horizontal gene transfer, while genes encoding secreted proteins are more diverse and might have been obtained by distinct events.

Intimin

Intimin is a 94-kDa outer membrane protein encoded by eae (E. coli attaching and effacing). The eae genes of pathogenic E. coli present a considerable heterogeneity in their 3¢ end that encodes the C-terminal 280 amino acids (Int280) involved in binding to the enterocytes and transmembrane intimin receptor (Tir) (see below), and the corresponding changes in the amino acid sequence also represent antigenic variations. Based on the sequence and antigenic differences, more than 10 distinct intimin types have been identified and classified, with a, b, e, and g being the main intimin types (88). Intimin a is generally found in EPEC, whereas e and g are closely associated with EHEC, and b is present in both EPEC and EHEC. E. coli O157:H7 produces intimin g. Intimin is exported via the general secretory pathway to the periplasm, where it is inserted into the outer membrane. Intimin has two functional regions: the highly conserved N-terminal region is inserted into the bacterial outer membrane, forming a b-barrel-like structure, and mediates dimerization; the variable Int280 extends from the bacterium and interacts with receptors in the host

12.  Enterohemorrhagic Escherichia coli

~ 10 nm

EspD/B

~ 260 nm

EspA

~ 50 nm

EscF

~ 10 nm

EscC

~ 10 nm

EscJ

~ 7 nm

Host cell plasma membrane

Bacterial outer membrane Periplasmic space Bacterial inner membrane

EscR, S, T, U, V SepD ?

EscN Tir CesT dimer

Map

SepL ?

ATP ADP

Figure 12.4  T3SS apparatus of EHEC. The basal body of the T3SS is composed of the secretin EscC, the inner membrane proteins EscR, EscS, EscT, EscU, and EscV, and the EscJ lipoprotein, which connects the inner and outer membrane ring structures. EscF constitutes the needle structure, whereas EspA subunits polymerize to form the EspA filament. EspB and EspD form the translocation pore in the host cell plasma membrane, connecting the bacteria with the eukaryotic cell via EspA filaments. The cytoplasmic ATPase EscN provides the energy to the system by hydrolyzing ATP molecules into ADP. SepD and SepL have been represented as cytoplasmic components of the T3SS. (Reproduced from reference 44.) doi:10.1128/9781555818463.ch12f4

cell plasma membrane. Interaction of intimin with host cells stimulates production of microvilli-like processes.

Effector Proteins

Numerous effector proteins have been identified in EHEC and are translocated into the host cell via the LEEencoded T3SS (44), including Tir, Map (mitochondrionassociated protein), EspF, EspG, EspH, SepZ, and EspB, which are encoded by LEE, whereas others such as Cif (cycle inhibiting factor), EspI, EspJ, and TccP (Tir-cytoskeleton coupling protein) are in prophages. Tir localizes to the host cell plasma membrane. It contains two membrane-spanning transmembrane domains and forms a hairpin-like structure with both its C and N termini located within the host cell and the region between the two transmembrane domains forming an extracellular loop, exposed on the surface of the cell, which interacts with intimin. Like intimin in the bacterial outer membrane, plasma membrane-bound Tir is a

301 dimer. Tir intracellular amino and carboxy termini interact with a number of focal adhesion and cytoskeletal proteins, linking the extracellular bacterium to the host cell cytoskeleton. These interactions lead to the formation of actin-rich pedestals beneath adherent bacteria after Wiskott-Aldrich syndrome protein (WASP) and the heptameric actin-related protein Arp2/3 are recruited to the pedestal tip (Fig. 12.2B). Effector proteins are delivered to the host cell cytoplasm from the extremity of the EspA filament through a translocation pore formed in the plasma membrane of the host cell by the translocator proteins EspB and EspD (Fig. 12.2A) (44). Additional proteins, SepL (“Sep” is an acronym for secretion of EPEC protein) and SepD, also play a role in the formation of the translocation apparatus. SepL is a soluble cytoplasmic protein that interacts with SepD. These proteins could be involved in the “switch” from secretion of translocator proteins to secretion of effector proteins through the type III machinery.

Virulence Plasmids

The primary virulence determinants of EHEC strains are chromosomally encoded. However, plasmids may play an important role in EHEC pathogenesis as well. An F-like 92-kb plasmid, pO157, found in most clinical isolates of E. coli O157:H7, shares sequence similarities with plasmids present in other EHEC serotypes. Based on DNA sequence analysis, pO157 contains 100 open reading frames (15). Genes coding for putative virulence factors in pO157 include those coding for enterohemolysin (ehxA), the general secretory pathway (etpC to etpO), serine protease (espP), catalase-peroxidase (katP), a potential adhesin (toxB), a Cl esterase inhibitor (stcE), and A/E gene-positive conserved fragments (ecf ). Nonetheless, the role of pO157 in bacterial virulence and survival is largely unknown. Toxicity results from the insertion of EhxA into the cytoplasmic membrane of target mammalian cells, thereby disrupting permeability. The EHEC catalase-peroxidase, a bifunctional periplasmic enzyme, protects the bacterium against oxidative stress, a possible defense strategy of mammalian cells during bacterial infection. Large hemolysin-encoding plasmids are also found in the majority of non-157 EHEC strains (53). A large plasmid of E. coli O113 (pO113) shares ehx, espP, and iha genes present in pO157. It also contains genes sharing similarity with the IncI1 transfer region and several putative adhesins and toxins but lacks the toxB region found in pO157. Analysis of ehxA and repA (a replication gene) of the RepFIB replicon revealed the evolutionary divergence of plasmids pO157 and pO113 from a common ancestor. Phylogenetic analyses of ehxA and

Foodborne Pathogenic Bacteria

302 repA were incongruent. These findings indicate differences in selective pressures between virulence genes and constitutive genes and point to the difficulties in examining the phylogeny of plasmid genomes due to their high degree of plasticity and mobility.

Shiga Toxins

Enterohemorrhagic E. coli produce one or two Stxs. The nomenclature of the Stx family and their important characteristics are listed in Table 12.5. Molecular studies on Stx1 from different E. coli strains revealed that Stx1a either is completely identical to the Stx of Shigella dysenteriae type 1 or differs by only one amino acid. However, during the last decade, several variants of Stx1 have been described. Some minor variants have 99% nucleotide sequence homology with stx1 of phage 933J. A more substantial deviation of Stx1 was observed in an ovine strain, OX3:H8 131/3, and subsequently among human isolates (113). It differs from Stx1a of phage 933J by 9 amino acids within the A subunit and 3 amino acids within the B subunit and is designated Stx1c. Another Stx1 variant, Stx1d, was identified in STEC ONT:H19, of bovine origin, showing difference from Stx1a by 20 amino acids in the A subunit and by 7 amino acids in the B subunit (14). Unlike Stx1, toxins of the Stx2 group are not neutralized by antiserum raised against Stx and do not crosshybridize with Stx1-specific DNA probes. There is sequence and antigenic variation within toxins of the

Stx2 family produced by E. coli O157:H7 and other STEC. At least 11 variants of Stx2 have been identified, including Stx2a, Stx2b, Stx2c (Stx2vh-a and Stx2vh-b), Stx2d (Stx2d-OX3a and Stx2d-Ount), Stx2e, Stx2f, and Stx2g (11, 41, 43). The Stx2c subgroup is approximately 97% related to the amino acid sequence of the B subunits of Stx2a, whereas the A subunit of Stx2c shares 98 to 100% amino acid sequence homology with Stx2a. Stx2e is associated with edema disease that principally occurs in piglets and shares 93% and 84% amino acid sequence homology with the A and B subunits, respectively, of Stx2a. Hence, the Stx2-related toxins have only partial serological reactivity with anti-Stx2 serum. Stx2f and Stx2g of STEC strains isolated from feral pigeons and cattle wastewater have also been described (43, 97).

Structure of the Stx Family

Stxs are a holotoxin comprised of a single enzymatic A subunit of approximately 32 kDa in association with a pentamer of receptor-binding B subunits of 7.7 kDa (81). The Stx A subunit can be split by trypsin into an enzymatic A1 fragment (approximately 27 kDa) and a carboxyl-terminal A2 fragment (approximately 4 kDa) that links A1 to the B subunits. The A1 and A2 subunits remain linked by a single disulfide bond until the enzymatic fragment is released and enters the cytosol of a susceptible mammalian cell. Each B subunit is comprised of six antiparallel strands forming a closed barrel capped by a single helix between strands 3 and 4. The A

Table 12.5  Nomenclature and biological characteristics of Stxsa Biological characteristics % Nucleotide sequence homology to stx

% Nucleotide sequence homology to stx2

Genetic loci

A subunit

B subunit

A subunit

Stx

Chromosome

NA

NA

Gb3

No

Stx1a

Phage

99

100

Gb3

No

Stx1c Stx1d Stx2a

Chromosome Unknown Phage

97 93

96 92

Unknown Unknown No

Stx2b

Nomenclature

B subunit

Receptor

Activated by intestinal mucus

NA

NA

Unknown Unknown Gb3

Phage

95

87

Gb3

No

Stx2c

Phage

100

97

Gb3

No

Stx2d

Phage

99

97

Gb3

Yes

Stx2e Stx2f Stx2g

Chromosome Unknown Unknown

93 63 94

84 57 91

Gb4 Unknown Unknown

No Unknown Unknown

a

From references 11, 41, 72, and 81. Abbreviations: HC, hemorrhagic colitis; HUS, hemolytic-uremic syndrome; NA, not applicable.

Disease Human diarrhea, HC, HUS Human diarrhea, HC, HU Human and sheep? Cattle? Human diarrhea, HC, HUS Human diarrhea, HC, HUS Human diarrhea, HC, HUS Human diarrhea, HC, HUS Pig edema disease Pigeon Bovine

12.  Enterohemorrhagic Escherichia coli subunit lies on the side of the B subunit pentamer, nearest to the C-terminal end of the B-subunit helices. The A subunit interacts with the B-subunit pentamer through a hydrophobic helix that extends to half of the 2.0-nm length of the pore in the B pentamer. This pore is lined by the hydrophobic side chains of the B-subunit helices. The A subunit also interacts with the B subunit via a four-stranded mixed sheet composed of residues of both the A2 and A1 fragments.

Genetics of Stxs

While most stx1 operons share a great deal of homology, there is considerable heterogeneity in the stx2 ­family. Unlike the genes of other Stx2 that are located on bacteriophage that integrate into the chromosome, the Stx of S. dysenteriae type 1 Stx1c and Stx2e are encoded by chromosomal genes (109, 113). A sequence comparison of the growing stx2 family indicates that genetic recombination among the B-subunit genes, rather than base substitutions, has given rise to the variants of Stx2 present in human and animal strains of E. coli (42). However, the operons for every member of the Stx subgroups are organized identically; the A and B subunit genes are arranged in tandem and separated by a 12- to 15-nucleotide gap in between. The operons are transcribed from a promoter that is located 5¢ to the A-subunit gene, and each gene is preceded by a putative ribosome-binding site. The existence of an independent promoter for the B-subunit genes has been suggested. The holotoxin stoichiometry suggests that expression of the A- and B-subunit genes is differentially regulated, permitting overproduction of the B polypeptides.

Receptors

All members of the Stx family bind to globoseries glycolipids on the eukaryotic cell surface; Stx, Stx1a, Stx2a, Stx2b, Stx2c, and Stx2d bind to glycolipid globotriaosylceramide (Gb3), whereas Stx2e primarily binds to glycolipid globotetraosylceramide (Gb4) (26). The alteration of binding specificity between Stx2e and the rest of the Stx family is related to carbohydrate specificity of receptors (68). The amino acid composition of B subunits of Stx2a and Stx2e differs at only 11 positions, yet Stx2e binds primarily to Gb4, whereas Stx2a binds only to Gb3. High-affinity binding also depends on multivalent presentation of the carbohydrate, as would be provided by glycolipids in a membrane. The affinity of Stx1 for Gb3 isoforms is influenced by fatty acyl chain length and by its level of saturation. Stx1 binds preferentially to Gb3 containing C20:1 fatty acid, whereas Stx2c prefers Gb3 containing C18:1 fatty acid. The basis for these findings may be related to the ability of different Gb3 isoforms

303 to present multivalent sugar-binding sites in the optimal orientation and position at the membrane surface. It is also possible that different fatty acyl groups affect the conformation of individual receptor epitopes on the sugar. Bovine cells do not express high numbers of the Gb3 receptors on their surface, and hence, cattle are not adversely affected by the toxin.

Mode of Action of the Stxs

Stxs act by inhibiting protein synthesis. Each of the B subunits is capable of binding with high affinity to an unusual disaccharide linkage (galactose 1-4 galactose) in the terminal trisaccharide sequence of Gb3 (or Gb4) (104). Following binding to the glycolipid receptor, the toxin is endocytosed from clathrin-coated pits and transferred first to the trans-Golgi network and subsequently to the endoplasmic reticulum and nuclear envelope. While it appears that transfer of the toxin to the Golgi apparatus is essential for intoxication, the mechanism of entry of the A subunit from the endosome to the cytosol, particularly the role of the B subunit in the process, remains unclear. In the cytosol, the A subunit undergoes partial proteolysis and splits into a 27-kDa active intracellular enzyme (A1) and a 4-kDa fragment (A2) bridged by a disulfide bond. Although the entire toxin is necessary for its toxic effect on whole cells, the A1 subunit is capable of cleaving the N-glycoside bond in one adenosine position of the 28S rRNA that comprises 60S ribosomal subunits (94). This elimination of a single adenine nucleotide inhibits the elongation factor-dependent binding to ribosomes of aminoacyl-bound transfer RNA molecules. Peptide chain elongation is truncated, and overall protein synthesis is suppressed, resulting in cell death.

The Role of Stxs in Disease

The role of Stxs in mediating colonic disease, HUS, and neurological disorders has been investigated in numerous studies. However, there is no satisfactory animal model for hemorrhagic colitis or HUS, and the severity of disease precludes study of experimental infections in humans. Therefore, the present understanding of the role of Stxs in causing disease is obtained from a combination of studies, including histopathology of diseased human tissues, animal models, and endothelial tissue culture cells. Results of recent studies support the concept that Stxs contribute to pathogenesis by directly damaging vascular endothelial cells in certain organs, thereby disrupting the homeostatic properties of these cells. The involvement of Stx in enterocolitis is demonstrated when fluid accumulation and histological damage occur after purified Stx is injected into ligated

304 rabbit intestinal loops. The fluid secretion may be due to the selective killing of absorptive villus tip intestinal epithelial cells by Stx. However, intravenous administration of Stx to rabbits can produce nonbloody diarrhea, suggesting other mechanisms for triggering diarrhea are possible. Studies with genetically mutated STEC strains also indicate that Stx has a role in intestinal disease, but the significance of Stx in provoking a diarrheal response differs depending upon the animal model used. Epidemiologic studies have identified a correlation between enteric infection with E. coli O157:H7 and ­development of HUS in humans. Histopathologic examination of kidney tissue from HUS patients revealed profound structural alterations in the glomeruli, the basic filtration unit of the kidney (72). The damage caused by Stxs is often not limited to the glomeruli. Arteriolar damage, involving internal cell proliferation, fibrin thrombus deposition, and perivascular inflammation, occurs (106). Cortical necrosis also occurs in a small number of HUS cases. In addition, human glomerular endothelial cells are sensitive to the direct cytotoxic action of bacterial endotoxin. Endotoxin in the presence of Stxs can also activate macrophage and polymorphonuclear neutrophils to synthesize and release cytokines, superoxide radicals, or proteinases and amplify endothelial cell damage. Neurological symptoms in patients and experimental animals infected with E. coli O157:H7 have also been described and may be caused by secondary neuron disturbances that result from endothelial cell damage by Stxs. Studies involving mice perorally administered an E. coli O157:H strain revealed that Stx2 impaired the blood-brain barrier and damaged neuron fibers, resulting in death. The presence of the toxin in neurons was verified by immunoelectron microscopy. Epidemiologic and laboratory studies have revealed that stx genotype and host factors such as age, preexisting immunity, and the use of antibiotics are important in the development of HUS (41, 57, 87). Stx2a and Stx2c are associated with high virulence and the ability to cause HUS, whereas Stx2b, Stx2d, Stx2e, Stx1a, and Stx1c occur in milder or asymptomatic infection (41, 87). However, in cell culture studies, Stx2a, Stx2d, and elastase-cleaved Stx2d were at least 25 times more potent than Stx2b and Stx2c. In vivo, in mice, potency of Stx2b and Stx2c was similar to that of Stx1, while Stx2a, Stx2d, and elastase-cleaved Stx2d were 40 to 400 times more potent than Stx1 (40).

Foodborne Pathogenic Bacteria Concluding Remarks The serious nature of the symptoms of hemorrhagic colitis and HUS caused by E. coli O157:H7 places this pathogen in a category apart from other foodborne pathogens, which typically cause only mild symptoms. The severity of the illness it causes combined with its apparent low infectious dose (<100 cells) qualifies E. coli O157:H7 to be among the most serious of known foodborne pathogens. E. coli O157: H7 causes disease by its ability to adhere to the host cell membrane and colonize the large intestine, after which it produces one or more Stxs. Although the pathogen has been isolated from a variety of domestic animals and wildlife, cattle are a major reservoir of E. coli O157:H7, with undercooked ground beef being among the most frequently implicated vehicles of transmission. Very few effective control measures for E. coli O157:H7 and other STEC in live animals have been identified. The number of cases associated with fresh produce such as lettuce, sprouts, and spinach has increased substantially in recent years. An important feature of this pathogen is its acid tolerance. Outbreaks have been associated with consumption of contaminated high-acid foods, including apple juice and fermented dry salami. Recreational and drinking waters also have been identified as vehicles of transmission of E. coli O157:H7 infections. Stx-producing E. coli strains other than O157: H7 have been increasingly associated with cases of HUS. More than 130 non-O157 STEC serotypes have been isolated from humans, but not all of these serotypes have been shown to cause illness. Although genomic analyses reveal that virulence genes are well conserved in many non-O157 STEC, in addition to the stx genes and the LEE island, some STEC may have a low potential to cause HUS; other non-O157 STEC isolates, including many found in healthy individuals, may not be pathogens. E. coli O157:H7 is still by far the most important serotype of STEC in North America. Isolation of non-O157 STEC requires techniques not generally used in clinical laboratories; hence, these bacteria are infrequently sought or detected in routine practice. Recognition of non-O157 STEC in foodborne illness necessitates identification of serotypes of EHEC other than O157:H7 in persons with bloody diarrhea and/or HUS and preferably in implicated food. The increased availability in clinical laboratories of techniques such as testing for Stxs or their genes and identification of other virulence markers unique for EHEC will continue to enhance the detection of disease attributable to non-O157 EHEC.

12.  Enterohemorrhagic Escherichia coli

305

References   1. Ballmer, K., B. M. Korczak, P. Kuhnert, P. Slickers, R. Ehricht, and H. Hachler. 2007. Fast DNA serotyping of Escherichia coli by use of an oligonucleotide microarray. J. Clin. Microbiol. 45:370–379.   2. Barkocy-Gallagher, G. A., T. M. Arthur, M. RiveraBetancourt, X. Nou, S. D. Shackelford, T. L. Wheeler, and M. Koohmaraie. 2003. Seasonal prevalence of Shiga toxin-producing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. J. Food Prot. 66:1978–1986.   3. Baylis, C. L., S. MacPhee, A. J. Robinson, R. Griffiths, K. Lilley, and R. P. Betts. 2004. Survival of Escherichia coli O157:H7, O111:H- and O26:H11 in artificially contaminated chocolate and confectionery products. Int. J. Food Microbiol. 96:35–48.   4. Bell, B. P., M. Goldoft, P. M. Griffin, M. A. Davis, D. C. Gordon, P. I. Tarr, C. A. Bartleson, J. H. Lewis, T. J. Barrett, J. G. Wells, et al. 1994. A multistate outbreak of Escherichia coli O157:H7-associated bloody diarrhea and hemolytic uremic syndrome from hamburgers. The Washington experience. JAMA 272:1349–1353.   5. Besser, R. E., P. M. Griffin, and L. Slutsker. 1999. Escherichia coli O157:H7 gastroenteritis and the hemolytic uremic syndrome: an emerging infectious disease. Annu. Rev. Med. 50:355–367.   6. Besser, T. E., D. D. Hancock, L. C. Pritchett, E. M. McRae, D. H. Rice, and P. I. Tarr. 1997. Duration of detection of fecal excretion of Escherichia coli O157:H7 in cattle. J. Infect. Dis. 175:726–729.   7. Bettelheim, K. A. 2007. The non-O157 Shiga-toxigenic (verocytotoxigenic) Escherichia coli; under-rated pathogens. Crit. Rev. Microbiol. 33:67–87.   8. Beutin, L., D. Geier, H. Steinruck, S. Zimmermann, and F. Scheutz. 1993. Prevalence and some properties of verotoxin (Shiga-like toxin)-producing Escherichia coli in seven different species of healthy domestic animals. J. Clin. Microbiol. 31:2483–2488.   9. Brackett, R., Y. Hao, and M. Doyle. 1994. Ineffectiveness of hot acid sprays to decontaminate Escherichia coli O157:H7 on beef. J. Food Prot. 57:198–203.   10. Brashears, M. M., A. Amezquita, and D. Jaroni. 2005. Lactic acid bacteria and their uses in animal feeding to improve food safety. Adv. Food Nutr. Res. 50:1–31.   11. Brett, K. N., M. A. Hornitzky, K. A. Bettelheim, M. J. Walker, and S. P. Djordjevic. 2003. Bovine non-O157 Shiga toxin 2-containing Escherichia coli isolates commonly possess stx2-EDL933 and/or stx2vhb subtypes. J. Clin. Microbiol. 41:2716–2722.   12. Brooks, J. T., E. G. Sowers, J. G. Wells, K. D. Greene, P. M. Griffin, R. M. Hoekstra, and N. A. Strockbine. 2005. Non-O157 Shiga toxin-producing Escherichia coli infections in the United States, 1983–2002. J. Infect. Dis. 192:1422–1429.   13. Reference deleted.   14. Burk, C., R. Dietrich, G. Acar, M. Moravek, M. Bulte, and E. Martlbauer. 2003. Identification and characterization of a new variant of Shiga toxin 1 in Escherichia

  15.

  16.

  17.

  18.

  19.

  20.   21.

  22.

  23.

  24.

  25.

  26.

coli ONT:H19 of bovine origin. J. Clin. Microbiol. 41:2106–2112. Burland, V., Y. Shao, N. T. Perna, G. Plunkett, H. J. Sofia, and F. R. Blattner. 1998. The complete DNA sequence and analysis of the large virulence plasmid of Escherichia coli O157:H7. Nucleic Acids Res. 26:4196–4204. Calderon, V. E., Q. Chang, M. McDermott, M. B. Lytle, G. McKee, K. Rodriguez, D. A. Rasko, V. Sperandio, and A. G. Torres. 2010. Outbreak caused by cadnegative Shiga toxin-producing Escherichia coli O111, Oklahoma. Foodborne Pathog. Dis. 7:107–109. Callaway, T. R., R. C. Anderson, K. J. Genovese, T. L. Poole, T. J. Anderson, J. A. Byrd, L. F. Kubena, and D. J. Nisbet. 2002. Sodium chlorate supplementation reduces E. coli O157:H7 populations in cattle. J. Anim. Sci. 80:1683–1689. Campellone, K. G., and J. M. Leong. 2003. Tails of two Tirs: actin pedestal formation by enteropathogenic E. coli and enterohemorrhagic E. coli O157:H7. Curr. Opin. Microbiol. 6:82–90. Centers for Disease Control and Prevention. 2006. Importance of culture confirmation of Shiga toxinproducing Escherichia coli infection as illustrated by outbreaks of gastroenteritis—New York and North Carolina, 2005. MMWR Morb. Mortal. Wkly. Rep. 55:1042–1045. Centers for Disease Control and Prevention. 1996. Lake-associated outbreak of Escherichia coli O157: H7—Illinois, 1995. JAMA 275:1872–1873. Centers for Disease Control and Prevention. 1999. Outbreak of Escherichia coli O157:H7 and Campylobacter among attendees of the Washington County Fair—New York, 1999. MMWR Morb. Mortal. Wkly. Rep. 48:803–805. Cernicchiaro, N., D. L. Pearl, S. A. McEwen, H. N. Zerby, F. L. Fluharty, S. C. Loerch, M. D. Kauffman, J. L. Bard, and J. T. LeJeune. 2010. A randomized controlled trial to assess the impact of dietary energy sources, feed supplements, and the presence of supershedders on the detection of Escherichia coli O157:H7 in feedlot cattle using different diagnostic procedures. Foodborne Pathog. Dis. 7:1071–1081. Chaudhuri, R. R., N. J. Loman, L. A. Snyder, C. M. Bailey, D. J. Stekel, and M. J. Pallen. 2008. xBASE2: a comprehensive resource for comparative bacterial genomics. Nucleic Acids Res. 36:D543–D546. Clavero, M., J. Monk, L. Beuchat, M. Doyle, and R. Brackett. 1994. Inactivation of Escherichia coli O157: H7, salmonellae, and Campylobacter jejuni in raw ground beef by gamma irradiation. Appl. Environ. Microbiol. 60:2069–2075. Cobbold, R., and P. Desmarchelier. 2000. A longitudinal study of Shiga-toxigenic Escherichia coli (STEC) prevalence in three Australian diary herds. Vet. Microbiol. 71:125–137. Cohen, A., G. E. Hannigan, B. R. Williams, and C. A. Lingwood. 1987. Roles of globotriosyl- and galabiosylceramide in verotoxin binding and high affinity interferon receptor. J. Biol. Chem. 262:17088–17091.

306   27. D‘Aoust, J., C. Park, R. Szabo, E. Todd, B. Emmons, and R. McKellar. 1988. Thermal inactivation of Campylobacter species, Yersinia enterocolitca, and hemorrhagic Escherichia coli O157:H7 in fluid milk. J. Dairy Sci. 71:3230–3236.   28. David, S. T., L. MacDougall, K. Louie, L. McIntyre, A. M. Paccagnella, S. Schleicher, and A. Hamade. 2004. Petting zoo-associated Escherichia coli O157:H7—secondary transmission, asymptomatic infection, and prolonged shedding in the classroom. Can. Commun. Dis. Rep. 30:173–180.   29. Davis, M. A., D. H. Rice, H. Sheng, D. D. Hancock, T. E. Besser, R. Cobbold, and C. J. Hovde. 2006. Comparison of cultures from rectoanal-junction mucosal swabs and feces for detection of Escherichia coli O157 in dairy heifers. Appl. Environ. Microbiol. 72:3766–3770.   30. Denno, D. M., W. E. Keene, C. M. Hutter, J. K. Koepsell, M. Patnode, D. Flodin-Hursh, L. K. Stewart, J. S. Duchin, L. Rasmussen, R. Jones, and P. I. Tarr. 2009. Tri-county comprehensive assessment of risk factors for sporadic reportable bacterial enteric infection in children. J. Infect. Dis. 199:467–476.   31. Dodd, C. C., M. W. Sanderson, J. M. Sargeant, T. G. Nagaraja, R. D. Oberst, R. A. Smith, and D. D. Griffin. 2003. Prevalence of Escherichia coli O157 in cattle feeds in Midwestern feedlots. Appl. Environ. Microbiol. 69:5243–5247.   32. Donnenberg, M. S., and T. S. Whittam. 2001. Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J. Clin. Investig. 107:539–548.   33. Doyle, M. P., and J. L. Schoeni. 1984. Survival and growth characteristics of Escherichia coli associated with hemorrhagic colitis. Appl. Environ. Microbiol. 48:855–856.   34. Elder, R. O., J. E. Keen, G. R. Siragusa, G. A. BarkocyGallagher, M. Koohmaraie, and W. W. Laegreid. 2000. Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proc. Natl. Acad. Sci. USA 97:2999–3003.   35. Elliott, S. J., L. A. Wainwright, T. K. McDaniel, K. G. Jarvis, Y. K. Deng, L. C. Lai, B. P. McNamara, M. S. Donnenberg, and J. B. Kaper. 1998. The complete sequence of the locus of enterocyte effacement (LEE) from enteropathogenic Escherichia coli E2348/69. Mol. Microbiol. 28:1–4.   36. Espie, E., F. Grimont, V. Vaillant, M. P. Montet, I. Carle, C. Bavai, H. de Valk, and C. Vernozy-Rozand. 2006. O148 Shiga toxin-producing Escherichia coli outbreak: microbiological investigation as a useful complement to epidemiological investigation. Clin. Microbiol. Infect. 12:992–998.   37. Ethelberg, S., B. Smith, M. Torpdahl, M. Lisby, J. Boel, T. Jensen, and K. Molbak. 2007. An outbreak of Verocytotoxin-producing Escherichia coli O26:H11 caused by beef sausage, Denmark 2007. Euro Surveill. 12:E070531.4.

Foodborne Pathogenic Bacteria   38. Ethelberg, S., B. Smith, M. Torpdahl, M. Lisby, J. Boel, T. Jensen, E. M. Nielsen, and K. Molbak. 2009. Outbreak of non-O157 Shiga toxin-producing Escherichia coli infection from consumption of beef sausage. Clin. Infect. Dis. 48:e78–e81.   39. Foster, J. W. 2004. Escherichia coli acid resistance: tales of an amateur acidophile. Nat. Rev. Microbiol. 2:898–907.   40. Frank, C., D. Werber, J. P. Cramer, M. Askar, M. Faber, M. A. Heiden, H. Bernard, A. Fruth, R. Prager, A. Spode, M. Wadl, A. Zoufaly, S. Jordan, K. Stark, and G. Krause. 2011. Epidemic profile of Shiga toxin-­producing Escherichia coli O104:H4 outbreak in Germany—­preliminary report. N. Engl. J. Med. http:// www.nejm.org/doi/full/10.1056/NEJMoa1106483. Accessed 16 July 2011.   41. Fuller, C. A., C. A. Pellino, M. J. Flagler, J. E. Strasser, and A. A. Weiss. 2011. Shiga toxin subtypes display dramatic differences in potency. Infect. Immun. 79:1329–1337.   42. Gannon, V. P., C. Teerling, S. A. Masri, and C. L. Gyles. 1990. Molecular cloning and nucleotide sequence of another variant of the Escherichia coli Shiga-like toxin II family. J. Gen. Microbiol. 136:1125–1135.   43. Garcia-Aljaro, C., M. Muniesa, J. E. Blanco, M. Blanco, J. Blanco, J. Jofre, and A. R. Blanch. 2005. Characterization of Shiga toxin-producing Escherichia coli isolated from aquatic environments. FEMS Microbiol. Lett. 246:55–65.   44. Garmendia, J., G. Frankel, and V. F. Crepin. 2005. Enteropathogenic and enterohemorrhagic Escherichia coli infections: translocation, translocation, translocation. Infect. Immun. 73:2573–2585.   45. Glass, K., J. Loeffelholz, J. Ford, and M. Doyle. 1992. Fate of Escherichia coli O157:H7 as affected by pH or sodium chloride in fermented, dry sausage. Appl. Environ. Microbiol. 58:2513–2516.   46. Griffin, P. M., and R. V. Tauxe. 1991. The epidemiology of infections caused by Escherichia coli O157:H7, other enterohemorrhagic E. coli, and the ­associated hemolytic uremic syndrome. Epidemiol. Rev. 13:60–98.   47. Hancock, D., T. Besser, J. Lejeune, M. Davis, and D. Rice. 2001. The control of VTEC in the animal reservoir. Int. J. Food Microbiol. 66:71–78.   48. Hedican, E. B., C. Medus, J. M. Besser, B. A. Juni, B. Koziol, C. Taylor, and K. E. Smith. 2009. Characteristics of O157 versus non-O157 Shiga toxin-producing Escherichia coli infections in Minnesota, 2000–2006. Clin. Infect. Dis. 49:358–364.   49. Hiramatsu, R., M. Matsumoto, K. Sakae, and Y. Miyazaki. 2005. Ability of Shiga toxin-producing Escherichia coli and Salmonella spp. to survive in a desiccation model system and in dry foods. Appl. Environ. Microbiol. 71:6657–6663.   50. Hrudey, S. E., P. Payment, P. M. Huck, R. W. Gillham, and E. J. Hrudey. 2003. A fatal waterborne disease epidemic in Walkerton, Ontario: comparison with other waterborne outbreaks in the developed world. Water Sci. Technol. 47:7–14.

12.  Enterohemorrhagic Escherichia coli   51. Jacob, M. E., J. T. Fox, J. S. Drouillard, D. G. Renter, and T. G. Nagaraja. 2008. Effects of dried distillers grain on fecal prevalence and growth of Escherichia coli O157 in batch culture fermentations from cattle. Appl. Environ. Microbiol. 74:38–43.   52. Jacob, M. E., Z. D. Paddock, D. G. Renter, K. F. Lechtenberg, and T. G. Nagaraja. 2010. Inclusion of dried or wet distillers’ grains at different ­ levels in diets of feedlot cattle affects fecal shedding of Escherichia coli O157:H7. Appl. Environ. Microbiol. 76:7238–7242.   53. Johnson, T. J., and L. K. Nolan. 2009. Pathogenomics of the virulence plasmids of Escherichia coli. Microbiol. Mol. Biol. Rev. 73:750–774.   54. Johnson, W. M., H. Lior, and G. S. Bezanson. 1983. Cytotoxic Escherichia coli O157:H7 associated with haemorrhagic colitis in Canada. Lancet i:76.   55. Kaper, J. B., J. P. Nataro, and H. L. Mobley. 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol. 2:123–140.   56. Karch, H., H. Russmann, H. Schmidt, A. Schwarzkopf, and J. Heesemann. 1995. Long-term shedding and clonal turnover of enterohemorrhagic Escherichia coli O157 in diarrheal diseases. J. Clin. Microbiol. 33:1602–1605.   57. Karmali, M. A. 2009. Host and pathogen determinants of verocytotoxin-producing Escherichia coliassociated hemolytic uremic syndrome. Kidney Int. Suppl. 2009(112):S4–S7.   58. Karmali, M. A., M. Petric, C. Lim, P. C. Fleming, G. S. Arbus, and H. Lior. 1985. The association between idiopathic hemolytic uremic syndrome and infection by verotoxin-producing Escherichia coli. J. Infect. Dis. 151:775–782.   59. Kim, H. H., M. Samadpour, L. Grimm, C. R. Clausen, T. E. Besser, M. Baylor, J. M. Kobayashi, M. A. Neill, F. D. Schoenknecht, and P. I. Tarr. 1994. Characteristics of antibiotic-resistant Escherichia coli O157:H7 in Washington State, 1984–1991. J. Infect. Dis. 170:1606–1609.   60. King, L. A., I. Filliol-Toutain, P. Mariani-Kurkidjian, V. Vaillant, C. Vernozy-Rozand, S. Ganet, N. Pihier, P. Niaudet, and H. de Valk. 2010. Family outbreak of Shiga toxin-producing Escherichia coli O123:H-, France, 2009. Emerg. Infect. Dis. 16:1491–1493.   61. Koehler, K. M., T. Lasky, S. B. Fein, S. M. Delong, M. A. Hawkins, T. Rabatsky-Ehr, S. M. Ray, B. Shiferaw, E. Swanson, and D. J. Vugia. 2006. Population-based incidence of infection with selected bacterial enteric pathogens in children younger than five years of age, 1996–1998. Pediatr. Infect. Dis. J. 25:129–134.   62. LeJeune, J. T., T. E. Besser, and D. D. Hancock. 2001. Cattle water troughs as reservoirs of Escherichia coli O157. Appl. Environ. Microbiol. 67:3053–3057.   63. LeJeune, J. T., D. Hancock, Y. Wasteson, E. Skjerve, and A. M. Urdahl. 2006. Comparison of E. coli O157 and Shiga toxin-encoding genes (stx) prevalence between Ohio, USA and Norwegian dairy cattle. Int. J. Food Microbiol. 109:19–24.

307   64. LeJeune, J. T., and A. N. Wetzel. 2007. Preharvest control of Escherichia coli O157 in cattle. J. Anim. Sci. 85: E73–E80.   65. Licence, K., K. R. Oates, B. A. Synge, and T. M. Reid. 2001. An outbreak of E. coli O157 infection with evidence of spread from animals to man through contamination of a private water supply. Epidemiol. Infect. 126:135–138.   66. Lin, J., M. Smith, K. Chapin, H. Baik, G. Bennett, and J. Foster. 1996. Mechanisms of acid resistance in enterohemorrhagic Escherichia coli. Appl. Environ. Microbiol. 62:3094–3100.   67. Line, J., A. Fain, A. Moran, L. Martin, R. Lechowich, J. Carosella, and W. Brown. 1991. Lethality of heat to Escherichia coli O157:H7: D-value and z-value determinations in ground beef. J. Food Prot. 54:762–766.   68. Lingwood, C. A. 1996. Role of verotoxin receptors in pathogenesis. Trends Microbiol. 4:147–153.   69. Lynn, T. V., D. D. Hancock, T. E. Besser, J. H. Harrison, D. H. Rice, N. T. Stewart, and L. L. Rowan. 1998. The occurrence and replication of Escherichia coli in cattle feeds. J. Dairy Sci. 81:1102–1108.   70. Mathusa, E. C., Y. Chen, E. Enache, and L. Hontz. 2010. Non-O157 Shiga toxin-producing Escherichia coli in foods. J. Food Prot. 73:1721–1736.   71. McNeilly, T. N., M. C. Mitchell, A. J. Nisbet, S. McAteer, C. Erridge, N. F. Inglis, D. G. Smith, J. C. Low, D. L. Gally, J. F. Huntley, and A. Mahajan. 2010. IgA and IgG antibody responses following systemic immunization of cattle with native H7 flagellin differ in epitope recognition and capacity to neutralise TLR5 signalling. Vaccine 28:1412–1421.   72. Melton-Celsa, A., and A. O‘Brien. 1998. Structure, biology, and relative toxicity of Shiga toxin family members for cells and animals, p. 121–128. In J. Kaper and A. O’Brien (ed.), Escherichia coli O157:H7 and Other Shiga ToxinProducing E. coli Strains. ASM Press, Washington, DC.   73. Meng, J., and M. P. Doyle. 1998. Microbiology of Shiga toxin-producing Escherichia coli in foods, p. 92–111. In J. Kaper and A. O’Brien (ed.), Escherichia coli O157:H7 and Other Shiga Toxin-Producing E. coli Strains. ASM Press, Washington, DC.   74. Meng, J., S. Zhao, M. Doyle, and S. Joseph. 1998. Antibiotic resistance of Escherichia coli O157:H7 and O157:NM isolated from animals, food and humans. J. Food Prot. 61:1511–1514.   75. Meyer-Broseta, S., S. N. Bastian, P. D. Arne, O. Cerf, and M. Sanaa. 2001. Review of epidemiological surveys on the prevalence of contamination of healthy cattle with Escherichia coli serogroup O157:H7. Int. J. Hyg. Environ. Health 203:347–361.   76. Michino, H., K. Araki, S. Minami, S. Takaya, N. Sakai, M. Miyazaki, A. Ono, and H. Yanagawa. 1999. Massive outbreak of Escherichia coli O157:H7 infection in schoolchildren in Sakai City, Japan, associated with consumption of white radish sprouts. Am. J. Epidemiol. 150:787–796.   77. Muller, E. E., M. M. Ehlers, and W. O. Grabow. 2001. The occurrence of E. coli O157:H7 in South African water sources intended for direct and indirect human consumption. Water Res. 35:3085–3088.

308   78. Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142–201.   79. Naylor, S. W., J. C. Low, T. E. Besser, A. Mahajan, G. J. Gunn, M. C. Pearce, I. J. McKendrick, D. G. Smith, and D. L. Gally. 2003. Lymphoid follicle-dense mucosa at the terminal rectum is the principal site of colonization of enterohemorrhagic Escherichia coli O157:H7 in the bovine host. Infect. Immun. 71:1505–1512.   80. Nielsen, E. M., M. N. Skov, J. J. Madsen, J. Lodal, J. B. Jespersen, and D. L. Baggesen. 2004. Verocytotoxinproducing Escherichia coli in wild birds and rodents in close proximity to farms. Appl. Environ. Microbiol. 70:6944–6947.   81. O’Brien, A. D., V. L. Tesh, A. Donohue-Rolfe, M. P. Jackson, S. Olsnes, K. Sandvig, A. A. Lindberg, and G. T. Keusch. 1992. Shiga toxin: biochemistry, genetics, mode of action, and role in pathogenesis. Curr. Top. Microbiol. Immunol. 180:65–94.   82. Ogura, Y., T. Ooka, Asadulghani, J. Terajima, J. P. Nougay­ rede, K. Kurokawa, K. Tashiro, T. Tobe, K. Nakayama, S. Kuhara, E. Oswald, H. Watanabe, and T. Hayashi. 2007. Extensive genomic ­ diversity and selective conservation of virulence-­determinants in enterohemorrhagic Escherichia coli strains of O157 and non-O157 serotypes. Genome Biol. 8:R138.   83. Ogura, Y., T. Ooka, A. Iguchi, H. Toh, M. Asadulghani, K. Oshima, T. Kodama, H. Abe, K. Nakayama, K. Kurokawa, T. Tobe, M. Hattori, and T. Hayashi. 2009. Comparative genomics reveal the mechanism of the parallel evolution of O157 and non-O157 enterohemorrhagic Escherichia coli. Proc. Natl. Acad. Sci. USA 106:17939–17944.   84. Orr, P., D. Milley, D. Coiby, and M. Fast. 1994. Prolonged fecal excretion of verotoxin-producing Escherichia coli following diarrheal illness. Clin. Infect. Dis. 19:796–797.   85. Orskov, F., I. Orskov, and J. A. Villar. 1987. Cattle as reservoir of verotoxin-producing Escherichia coli O157: H7. Lancet ii:276.   86. Orskov, I., F. Orskov, B. Jann, and K. Jann. 1977. Serology, chemistry, and genetics of O and K antigens of Escherichia coli. Bacteriol. Rev. 41:667–710.   87. Orth, D., K. Grif, A. B. Khan, A. Naim, M. P. Dierich, and R. Wurzner. 2007. The Shiga toxin genotype rather than the amount of Shiga toxin or the cytotoxicity of Shiga toxin in vitro correlates with the appearance of the hemolytic uremic syndrome. Diagn. Microbiol. Infect. Dis. 59:235–242.   88. Oswald, E., H. Schmidt, S. Morabito, H. Karch, O. Marches, and A. Caprioli. 2000. Typing of intimin genes in human and animal enterohemorrhagic and enteropathogenic Escherichia coli: characterization of a new intimin variant. Infect. Immun. 68:64–71.   89. Piercefield, E. W., K. K. Bradley, R. L. Coffman, and S. M. Mallonee. 2010. Hemolytic uremic syndrome after an Escherichia coli O111 outbreak. Arch. Intern. Med. 170:1656–1663.   90. Rangel, J. M., P. H. Sparling, C. Crowe, P. M. Griffin, and D. L. Swerdlow. 2005. Epidemiology of Escherichia coli O157:H7 outbreaks, United States, 1982–2002. Emerg. Infect. Dis. 11:603–609.

Foodborne Pathogenic Bacteria   91. Renter, D. G., V. Bohaychuk, J. Van Donkersgoed, and R. King. 2007. Presence of non-O157 Shiga toxinproducing Escherichia coli in feces from feedlot cattle in Alberta and absence on corresponding beef carcasses. Can. J. Vet. Res. 71:230–235.   92. Rice, D. H., K. M. McMenamin, L. C. Pritchett, D. D. Hancock, and T. E. Besser. 1999. Genetic subtyping of Escherichia coli O157 isolates from 41 Pacific Northwest USA cattle farms. Epidemiol. Infect. 122:479–484.   93. Robert Koch Institut. 2011. Technical report: EHEC/HUS O104:H4 outbreak, Germany, May/June 2011. http:// www.rki.de/cln_160/nn_217400/EN/Home/EHEC__ Report,templateId=raw,property=publicationFile.pdf/ EHEC_Report.pdf. Accessed 16 July 2011.   94. Sandvig, K., and B. van Deurs. 1996. Endocytosis, intracellular transport, and cytotoxic action of Shiga toxin and ricin. Physiol. Rev. 76:949–966.   95. Sargeant, J. M., M. W. Sanderson, D. D. Griffin, and­ R. A. Smith. 2004. Factors associated with the presence of Escherichia coli O157 in feedlot-cattle water and feed in the Midwestern USA. Prev. Vet. Med. 66:207–237.   96. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States-major pathogens. Emerg. Infect. Dis. 17:7–15.   97. Schmidt, H., J. Scheef, S. Morabito, A. Caprioli, L. H. Wieler, and H. Karch. 2000. A new Shiga toxin 2 variant (Stx2f) from Escherichia coli isolated from pigeons. Appl. Environ. Microbiol. 66:1205–1208.   98. Schmidt, M. A. 2010. LEEways: tales of EPEC, ATEC and EHEC. Cell Microbiol. 12:1544–1552.   99. Schroeder, C. M., J. Meng, S. Zhao, C. DebRoy, J. Torcolini, C. Zhao, P. F. McDermott, D. D. Wagner, R. D. Walker, and D. G. White. 2002. Antimicrobial resistance of Escherichia coli O26, O103, O111, O128, and O145 from animals and humans. Emerg. Infect. Dis. 8:1409–1414. 100. Schroeder, C. M., C. Zhao, C. DebRoy, J. Torcolini, S. Zhao, D. G. White, D. D. Wagner, P. F. McDermott, R. D. Walker, and J. Meng. 2002. Antimicrobial resistance of Escherichia coli O157 isolated from humans, cattle, swine, and food. Appl. Environ. Microbiol. 68:576–581. 101. Sekse, C., K. O‘Sullivan, P. E. Granum, L. M. Rorvik, Y. Wasteson, and H. J. Jorgensen. 2009. An outbreak of Escherichia coli O103:H25— bacteriological investigations and genotyping of isolates from food. Int. J. Food Microbiol. 133:259–264. 102. Smith, D. R., R. A. Moxley, T. J. Klopfenstein, and G. E. Erickson. 2009. A randomized longitudinal trial to test the effect of regional vaccination within a cattle feedyard on Escherichia coli O157:H7 rectal colonization, fecal shedding, and hide contamination. Foodborne Pathog. Dis. 6:885–892. 103. Tarr, P. I., C. A. Gordon, and W. L. Chandler. 2005. Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 365:1073–1086.

12.  Enterohemorrhagic Escherichia coli 104. Tesh, V. L., and A. D. O‘Brien. 1991. The pathogenic mechanisms of Shiga toxin and the Shiga-like toxins. Mol. Microbiol. 5:1817–1822. 105. Teunis, P., K. Takumi, and K. Shinagawa. 2004. Dose response for infection by Escherichia coli O157:H7 from outbreak data. Risk Anal. 24:401–407. 106. van Setten, P. A., L. A. Monnens, R. G. Verstraten, L. P. van den Heuvel, and V. W. van Hinsbergh. 1996. Effects of verocytotoxin-1 on nonadherent human monocytes: binding characteristics, protein synthesis, and induction of cytokine release. Blood 88:174–183. 107. Vasconcelos, J. T., N. A. Elam, M. M. Brashears, and M. L. Galyean. 2008. Effects of increasing dose of live cultures of Lactobacillus acidophilus (strain NP 51) combined with a single dose of Propionibacterium freudenreichii (strain NP 24) on performance and carcass characteristics of finishing beef steers. J. Anim. Sci. 86:756–762. 108. Vu-Khac, H., and N. A. Cornick. 2008. Prevalence and genetic profiles of Shiga toxin-producing Escherichia coli strains isolated from buffaloes, cattle, and goats in central Vietnam. Vet. Microbiol. 126:356–363. 109. Weinstein, D. L., M. P. Jackson, J. E. Samuel, R. K. Holmes, and A. D. O’Brien. 1988. Cloning and sequencing of a Shiga-like toxin type II variant from Escherichia coli strain responsible for edema disease of swine. J. Bacteriol. 170:4223–4230. 110. Wells, J. G., B. R. Davis, I. K. Wachsmuth, L. W. Riley, R. S. Remis, R. Sokolow, and G. K. Morris. 1983. Laboratory investigation of hemorrhagic colitis out-

309 breaks associated with a rare Escherichia coli serotype. J. Clin. Microbiol. 18:512–520. 111. Williams, M. S., J. L. Withee, E. D. Ebel, N. E. Bauer, Jr., W. D. Schlosser, W. T. Disney, D. R. Smith, and R. A. Moxley. 2010. Determining relationships between the seasonal occurrence of Escherichia coli O157:H7 in live cattle, ground beef, and humans. Foodborne Pathog. Dis. 7:1247–1254. 112. Wilson, J. B., R. C. Clarke, S. A. Renwick, K. Rahn, R. P. Johnson, M. A. Karmali, H. Lior, D. Alves, C. L. Gyles, K. S. Sandhu, S. A. McEwen, and J. S. Spika. 1996. Vero cytotoxigenic Escherichia coli infection in dairy farm families. J. Infect. Dis. 174:1021–1027. 113. Zhang, W., M. Bielaszewska, T. Kuczius, and H. Karch. 2002. Identification, characterization, and distribution of a Shiga toxin 1 gene variant (stx1c) in Escherichia coli strains isolated from humans. J. Clin. Microbiol. 40:1441–1446. 114. Zhao, T., and M. Doyle. 1994. Fate of enterohemorrhagic Escherichia coli O157:H7 in commercial mayonnaise. J. Food Prot. 57:780–783. 115. Zhao, T., M. Doyle, and R. Besser. 1993. Fate of enterohemorrhagic Escherichia coli O157:H7 in apple cider with and without preservatives. Appl. Environ. Microbiol. 59:2526–2530. 116. Zweifel, C., N. Giezendanner, S. Corti, G. Krause, L. Beutin, J. Danuser, and R. Stephan. 2010. Characteristics of Shiga toxin-producing Escherichia coli isolated from Swiss raw milk cheese within a 3-year monitoring program. J. Food Prot. 73:88–91.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch13

Franco J. Pagotto Kahina Abdesselam

Cronobacter Species

The first report of a “yellow-pigmented coliform” as the causative agent in a case of septicemia in an infant dates back to 1929 (124). In 1961, a yellow-pigmented coliform was reported as the causative agent in two cases of fatal neonatal meningitis that occurred in 1958 in St. Albans, England (154). Another report described a case of meningitis caused by yellow-pigmented Enterobacter cloacae in a child from Denmark who had presented with severe mental and neurological sequelae (65). The name Enterobacter sakazakii was recorded for the first time in 1977 (12). In 1980, E. sakazakii was distinguished from Enterobacter cloacae and became a new species named after the Japanese microbiologist Riichi Sakazaki, in honor of his contribution to the current understanding of enteric bacteriology (32). In 2008, E. sakazakii was moved to a new genus, Cronobacter, consisting of five species (61). The first documented link of Cronobacter to powdered infant formula (PIF) as a causative factor of infection was reported in 1990 by Clark et al. (17), who confirmed the link between Cronobacter isolated from patients and from PIF in two hospitals, using plasmid analysis, antibiograms, chromosomal restriction endonuclease analysis, ribotyping, and multilocus enzyme

13 electrophoresis. To date, there have been approximately 120 reported cases of Cronobacter infections, resulting in 19 deaths, described in 11 different countries and often associated with neonates and children from 3 days to 4 years of age (163). A surveillance study was conducted in the United Kingdom and Philippines, where an additional 85 infections were reported in infants and young children. The incidence of Cronobacter infections in infants worldwide is relatively low compared to other infectious diseases, estimated to be around 4 to 5 cases per year; however, it has increased over the years (163). There have been at least 20 recorded cases of adult infections from Cronobacter species (9a, 26, 39a, 47, 63, 81, 137a). Cronobacter species is an emerging foodborne pathogen that has increasingly raised interest among the scientific community, health care providers, and the food industry since the early 1980s. However, there is still a lack of information on this bacterium, including its natural habitat, mechanisms of virulence, and dose response for humans. In 2002, the International Commission for Microbiological Specifications for Foods (53) classified Cronobacter species as a severe hazard for restricted populations, causing life-threatening or substantial

Franco J. Pagotto and Kahina Abdesselam, Bureau of Microbial Hazards, Health Products and Food Branch, Food Directorate, Health Canada, Sir F.G. Banting Research Centre, P/L2204E, Ottawa, Ontario, K1A 0K9, Canada.

311

312 chronic sequelae or illness of long duration, with the risk populations being newborns and immunocompromised infants. In 2004, the Food and Agriculture Organization of the United Nations (FAO) and the World Health Organization (WHO) jointly held an expert meeting on Cronobacter species and other microorganisms of concern in PIF (34), aiming to gather information for revising the 1979 Recommended International Code of Hygienic Practice for Foods for Infants and Children. In 2005, a Proposed Draft Recommended International Code of Hygienic Infant Formula was released, accepted in 2007, with amendments made in 2008, all intended to replace the 1979 Code (20). The WHO/FAO expert meeting of 2004 agreed on a list of recommendations to the scientific community and infant formula manufacturers focusing on the need for a better understanding of Cronobacter species and potentially other microorganisms that could be found in infant formula, including (i) the use of internationally validated detection and molecular typing methods for Cronobacter species; (ii) investigation and reporting of sources and vehicles of infection by Cronobacter species including the establishment of a laboratory-based network; and (iii) a better understanding of the ecology, taxonomy, virulence, and other characteristics of this emerging pathogen. The aim of this chapter is to provide a review of the basic biology, ecology, pathogenicity, and epidemiology of this emerging opportunistic foodborne pathogen.

CHARACTERISTICS OF THE ORGANISM Cronobacter species is categorized in the family Enterobacteriaceae and like most species in this family is considered an opportunistic pathogen. Cronobacter species are gram-negative, oxidase-negative, non-sporeforming, non-acid-fast, straight, rod-shaped bacteria, having dimensions of 0.3 to 1.0 by 1.0 to 6.0 µm. It is motile by peritrichous flagella, nonhalophilic, and facultatively anaerobic and grows over a wide range of temperatures (6 to 45°C). The biochemical characteristics of Cronobacter species, according to the second edition of Bergey’s manual, are summarized in Table 13.1. Prior to the early 1980s, when it became a new species, Cronobacter was referred to as a yellowpigmente­d strain of E. cloacae. DNA-DNA hybridization results showed that Cronobacter sakazakii was 53 to 54% related to two distinct genera: Enterobacter and Citrobacter (11, 12, 61). However, other pheno-

Foodborne Pathogenic Bacteria and genotypic tests confirmed that they were closer to E. cloacae than Citrobacter freundii; thus, the new species was included in the Enterobacter genus. After the pathogen was classified, 15 biogroups were further included, with suggestions that they represented multiple species of Enterobacter (32, 61). Iversen et al. investigated several different strains of Cronobacter and, based on fluorescent amplified fragment length polymorphism fingerprints, ribopatterns, and full-length 16S rRNA gene analyses, proposed a reclassification of this species. Phenotypic profiling, using 14 biochemical characteristics, permitted better differentiation of the species (Table 13.2). Joseph et al. recently reported a new species, C. condimenti, and reclassified C. genomespecies 1 as C. universalis, while further evaluating Cronobacter strains 1330 (formerly sakazakii), NCTC 9529, 731 (formerly genomespecies1) and 96, 1435 (formerly t­uricensis) isolated from a wide range of sources that includes spiced meat and leg infections (66). The phenotypic differentiation of the seven Cronobacter species is presented in Table 13.2. Colonies of Cronobacter on standard laboratory growth media such as trypticase soy agar (TSA) measure 2 to 3 mm and 1 to 1.5 mm in diameter at 36°C and 25°C, respectively, after overnight incubation. Cronobacter spp. produce colonies with distinct morphologies (88); one colony type is described as being dry, matte, and leathery or rubbery, retracting to the agar when touched, and having little biomass adhering to an inoculation loop. The second type is moist, glossy, and easy to remove from the agar with a loop (43). Conditions influencing the production of an exopolysaccharide (EPS) by Cronobacter have been studied (137), with distinguishing features being its high viscosity and gel formation. Maximum amounts of EPS produced by Cronobacter were obtained with a carbon/nitrogen ratio of 20.2:1 at 27°C in media supplemented with glucose. Other bacteria such as Enterobacter aerogenes, Xanthomonas campestris, and Arthrobacter viscosus have also been previously shown to have enhanced EPS production in the presence of glucose. Harris and Oriel (45) have registered a patent for the production of gums from Cronobacter, hypothesizing that in addition to being a frictional drag reduction agent in aqueous systems, the heteropolysaccharide could be used as a suspending, thickening, or stabilizing agent (45). Typing methods often focus on phenotypic aspects related to antigenic structures. Information on fine structures is limited in Cronobacter species; however,

13.  Cronobacter Species

313

Table 13.1  Biochemical characterization of Cronobacter speciesa Characteristic

Utilization of:

Acid from:

Motility (36°C) Yellow pigment Urea hydrolysis

+ + –

Adonitol l-Arabinose d-Arabitol

– + –

cis-Aconitate trans-Aconitate Adonitol

+ d –

Indole production

d +

Cellobiose Dulcitol

+ –

4-Aminobutyrate 5-Aminovalerate

+ –

d-Arabitol Benzoate Citrate m-Coumarate Dulcitol l-Fucose Gentisate

– – + – d – –

b-Xylosidase test Methyl red Voges-Proskauer Growth in KCN Gelatin hydrolysis (22°C) DNase (25°C) Lysine decarboxylase Arginine dehydrolase

– + + – (+) – +

Ornithine decarboxylase Phenylalanine deaminase Glucose dehydrogenase Gluconate dehydrogenase Growth at 41°C Esculin hydrolysis Acetate

+ d + – + + –

meso-Erythritol Glycerol myo-Inositol Maltose d-Mannitol Melibiose a-Methylglucosidase Mucate Raffinose l-Rhamnose Salicin d-Sorbitol Sucrose Trehalose d-Xylose Lactose

– – (+) + + + + + + + + – + + + +

Histamine 3-Hydroxybenzoate 4-Hydroxybenzoate 3-Hydroxybutyrate myo-Inositol 5-Ketogluconate 2-Ketoglutarate Lactose Lactulose d-Lyxose d-Malate Malonate

– – – – d – – + + – (d) (d)

Maltitol d-Melibiose 1-O-Methyl-a-galactoside 1-O-Methyl-d-glucose 1-O-Methyl-a-d-glucoside Mucate Palatinose Phenylacetate l-Proline Protocatechuate Putrescine Quinate d-Raffinose l-Rhamnose d-Saccharate d-Sorbitol Sucrose d-Tagatose meso-Tartrate Tricarballylate Tryptamine d-Turanose l-Tyrosine

+ + + – + – + – + – + – + + – – + – – – – d –

a +, 90 to 100% positive in 1 to 2 days; (+), 90 to 100% positive in 1 to 4 days; −, 90 to 100% negative in 4 days; d, positive or negative in 1 to 4 days; (d), positive or negative in 3 to 4 days. Modified from Garrity et al. (38).

studies have been under way to examine the structure of the O polysaccharide of different strains of Cronobacter, including the species sakazakii, muytjensii, malonaticus, and turicensis (95-99). Our group was the first to report the presence of legionaminic acid in Cronobacter lipopolysaccharide (LPS) (99). This work will be useful in further refining the diversity of the O-antigen structures and help develop an O-antigenbased serotyping scheme for all Cronobacter species (62, 149), as well as in providing unique and specific structurally defined markers for epidemiological tracking and typing. Interestingly, the two morphologies described for Cronobacter cells may be lost upon repeated subculturing. The yellow pigment production is also unstable and, similar to colony morphology, can revert from one state to another upon subculturing. Other features that may be related to the heteropolysaccharide production are the sediment formation when grown in static liquid media and the increased growth medium viscosity when grown with agitation.

INITIAL DATA ON PHYLOGENETIC CHARACTERIZATION Because it is an emerging pathogen, there is a large amount of information that remains unknown about Cronobacter species. Nevertheless, advances have been made in the molecular characterization of Cronobacter species using amplification and sequencing of the 16S rRNA gene, pulsed-field gel electrophoresis (PFGE), ribotyping, and plasmid typing (116). Clementino et al. (18) were one of the first research groups to use tRNA intergenic spacer-PCR (tDNAPCR) and 16S-23S internal transcribed spacer-PCR (ITS-PCR) for the characterization of E. cloacae. In their study, specific and reproducible patterns were obtained for Cronobacter strain ATCC 29004. However, the study aimed at differentiating E. cloacae, and only a single Cronobacter strain was used in their analyses. The first PCR-based application for Cronobacter species was described by Keyser et al. (71). They used the full-length 16S rRNA gene from a Cronobacter type

Foodborne Pathogenic Bacteria

314

Table 13.2  Biochemical differentiation of Cronobacter species belonging to seven different biogroupsa Cronobacter biogroup Characteristic Indole production Dulcitol Lactulose Malonate Maltitol Palatinose Putrescine Melezitose Turanose myo−Inositol cis−Aconitate trans-Aconitate 1-0-Methyl a-d-glucopyranoside 4-Aminobutyrate a

sakazakii

malonaticus

turicensis

muytjensii

dublinensis

condimenti

universalis

− − ++ − ++ ++ ++ − ++ + ++ − ++

− − ++ ++ ++ ++ + − ++ + ++ ++ ++

− ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ − ++

++ ++ ++ ++ − + ++ − + ++ + + −

++* − * * * ++ ++* −* * ++ ++ ++ ++

+ − − + − − − − − − − − +

− ++ ++ ++ ++ − − ++ − ++ − − ++

++

++

++

++

++

−

−

+, 20 to 80% positive; ++, >90% positive; −, <10% positive; *, variation within subspecies. Adapted from Healy et al., Iversen et al., and Joseph et al. (49, 61, 66).

strain to develop a detection system for upflow anaerobic sludge blankets. Subsequently, Lehner et al. (89) reevaluated this approach, overcame some of the specificity aspects, and generated more 16S rRNA sequences, which led to the description of two strain lineages. They then developed a 16S rRNA gene-specific PCR method to distinguish Cronobacter species belonging to both lineages. Iversen et al. (60) used partial 16S rRNA gene, along with hsp60 sequences, to investigate the phylogenetic relationship among 126 Cronobacter strains. Their study reinforced the polyphyletic nature of Enterobacter spp., as did a previous study based on gyrB gene (25). Iversen et al. (60) identified four clusters and reported that a substantial amount of taxonomic heterogeneity existed within the species. They were able to group most strains in one cluster and hypothesized that the other three clusters could possibly represent new lineages within the species. Interestingly, they reported similarity among a Cronobacter type strain and Citrobacter koseri. Equally important, their work demonstrated that current methods based on biochemical profiling (e.g., API 20E or ID 32E) did not always align with the 16S rRNA gene-based approach; we have reached a similar conclusion in our laboratory. Liu et al. (94) reported on the use of a PCR-oligonucleotide array for the detection of Cronobacter species from infant formula. Using phylogenetic analyses of the ITS region of C. sakazakii, the group developed two pairs of species-specific primers that amplified a 282and 251-bp fragment of the ITS G-operon and ITS IA-

operon, respectively. Ten oligonucleotide features (i.e., probes) were designed to target these ITS products using a nylon membrane array. A 5¢ nuclease real-time PCR for the specific identification of Cronobacter species isolates has been developed (100). In this method, the 16S rRNA gene was targeted, along with a coamplified internal amplification control (IAC) in order to monitor falsenegative results. Researchers concluded that based on the use of an IAC, their assay may be potentially suitable for the rapid detection of Cronobacter species in foods, and they reinforced the importance of using an IAC in diagnostic PCRs to decrease false negatives caused by PCR inhibitors, technical errors, or equipment malfunction. Their proposed method is 3 to 4 days faster than culture and isolation methods. If one were to start with a putative positive on a plate, their method would yield a result in approximately 2 h compared to 2 days using the traditional biochemical methods. In an epidemiological study done in 1990, Clark et al. (17) used plasmid profiling to analyze 32 Cronobacter species isolates, 27 from a hospital outside the United States and 5 from a hospital in the United States. Among the 27 non-U.S. isolates, 24 were from infant formulae and 3 were from patients. The sources of the 5 U.S. isolates were as follows: 1 from an infant formula, 1 from a blender used for infant formula preparation, and 3 from patients. Of the 27 isolates, 26 (3 from patients and 23 from infant formula) had the same plasmid profile with four bands of approximately 3.2, 42, 70, and 85 MDa in

13.  Cronobacter Species size. The five isolates from the United States had the same plasmid profile, giving three bands of 3.2, 29, and 75 MDa. We have undertaken a comprehensive phenotypic and genotypic characterization of over 200 strains of Cronobacter species collected from around the world (R. Lenati, K. Hébert, Y. Kou, S. McIlwham, K. Tyler, J. Farber, and F. Pagotto, unpublished data). Using 1 kb of PCR-amplified 16S rRNA gene sequences, phylogenetic analyses indicated that there is a >95% sequence identity among isolates from environmental, food, and clinical sources. These same isolates were typed with PFGE and automated ribotyping. Using the restriction endonuclease XbaI, 119 different pulsotypes were generated from 145 strains, of which the two largest clusters comprised five isolates each. The five strains forming one of the clusters had the same ribotype. In addition to one cluster in common with PFGE, the ribotyping technique generated two other clusters containing five strains each. These three larger clusters were part of the 110 ribotypes generated with the restriction enzyme EcoRI. While both PFGE and ribotyping are techniques presently used for epidemiological studies during interlaboratory network investigations of outbreaks or isolated cases of bacterial infections, the congruence between the two approaches was found to be only 20% among the strains analyzed in our laboratory. PFGE seemed to better discriminate strains according to their sources, grouping strains isolated from the same outbreak and separating some that were known to be of different origin.

SUSCEPTIBILITY TO PHYSICAL AND CHEMICAL TREATMENTS

Temperature

Studies focusing on thermotolerance have reported on D values and z values for Cronobacter species grown in reconstituted PIF (10, 59, 118). Most studies report D54°C values varying from 10.2 to 16.4 min, D58°C from 2.4 min to 4.2 min, D60°C values of 1.1 min, and D62°C ranging from 0.2 to 0.4 min. The z values fall in the range of 5.6 to 5.8°C, with one study citing a z value of 3.1°C (118). Extrapolating from these collective data, the decimal reduction time at 71.2°C would be 0.7 s and the minimum high-temperature-shorttreatment pasteurization process of 15 s at 71.7°C should theoretically result in a 21-log reduction in viable counts of the organism, indicating that pasteurized milk should be free of Cronobacter species

315 after a pasteurization step. This would suggest that PIF testing positive for this bacterium are most likely contaminated after the various heating steps that are used in the process. Williams et al. (161) used a top-down proteomics approach to look at the proteins expressed by 12 strains of Cronobacter species with the aim of identifying proteins expressed in a particular group of strains. Using strains resistant to higher temperatures, based on their previous work on the thermal inactivation of Cronobacter in reconstituted PIF (27), they looked for markers of thermal tolerance. Applying liquid chromatography mass spectrometry, they obtained protein expression profiles for the 12 strains and found a putative marker common among thermotolerant Cronobacter strains. Interestingly, this protein was homologous to a protein in the bacteria Methylobacillus flagellatus KT, an obligate methylotroph that grows at relatively high temperatures. This protein has not been found in genera closely related to Cronobacter species, such as Escherichia coli and Salmonella enterica. The effects of microwave heating on Cronobacter strains have also been evaluated. In a study by Kindle et al. (76), five strains of Cronobacter were suspended in reconstituted PIF at an initial concentration of 5 log CFU/ml and then placed in a microwave oven until the first signs of boiling, reaching temperatures of approximately 82 to 93°C. After the microwave treatment, four of the five samples of reconstituted infant formula tested negative for Cronobacter, while one of the samples contained 20 CFU/ml. The researchers hypothesized that differences in infant formula composition could possibly account for different rates of inactivation of Cronobacter species. There has been little work done on the ability of Cronobacter species to survive in frozen foods. Our laboratory has shown that the organism can be frozen in reconstituted PIF for over 6 months without any decrease of viable cell counts (R. Lenati, K. Hébert, J. Farber, and F. Pagotto, unpublished data). Guidelines for preparing and manipulating reconstituted infant formula and fortified breast milk in the household and health care settings in order to manage the risk of Cronobacter species growth are starting to be released by groups such as the American Dietetic Association (ADA) (125), the European Society for Pediatric Gastroenterology, Hepatology and Nutrition (3), the U.S. Food and Drug Administration (FDA) (156), Health Canada (48), the New Zealand Ministry of Health (119), and the FAO/WHO (20). In addition, in order to evaluate current recommendations regarding

Foodborne Pathogenic Bacteria

316 “hang times” (i.e., the amount of time a formula is kept at room temperature in the feeding bag and accompanying lines during enteral tube feeding), Telang et al. (150) assessed total bacterial growth in fortified breast milk. Most guidelines and recommendations are aimed for specific high-risk groups. A summary of some other current recommendations with regard to the handling of PIF and fortified breast milk can be found in Table 13.3.

Water Activity

The major food product linked to outbreaks of Cronobacter infection is PIF. It is known that the organism can survive for a long time in PIF (28) as well

as in spices (57). However, the exact mechanism(s) this organism uses to survive in these very dry environments (e.g., PIF average water activity [aw] is 0.2) is unknown. It has been previously demonstrated that some bacteria increase osmolarity by the intracellular accumulation of ions (e.g., K+) and compatible solutes such as proline, glycine, betaine, and trehalose (70). When the drying process causes osmotic stress for the bacteria, polyhydroxyl compounds such as trehalose can be used by the organism to replace the shell of water around macromolecules, thus protecting cells from damage. The role of trehalose accumulation in the cells during the drying process and the osmotic

Table 13.3  Guidelines for reconstituted PIF and BMFa preparation and handling Reconstituted PIF Refrigerate prepared formula to 4°C (40°F) within 1 h of preparation. Single-use bottles and nipples (except specialty products not available as single-use). Use only chilled, sterile water for IF preparation. Use equipment that keeps PIF chilled at 4°C (40°F) to safely transport it to the patient unit. Autoclaving or a thermal process is recommended for PIF preparation equipment and utensils. Discard any formula left in the bottle after the feed. Reconstitute small volume of formula for each feeding to reduce the quantity and “hang time” at room temperature before consumption. Due to differences in PIF preparation among hospitals, each facility should identify and follow procedures appropriate to minimize microbial growth in reconstituted PIF. Minimize the hang time (i.e., the amount of time a PIF is at room temperature in the feeding bag and accompanying lines during enteral tube feeding), with no hang time exceeding 4 h. Preparation of PIF in a laminar flow hood. Rehydrate PIF with sterile water. Follow recommendations of ADA. Use PIF powder within 4 weeks of opening the can. Sterilize bottles. Prepare only the amount for baby’s next feed, and prepare it close to feeding time. Use warmed formula within 20 min. Discard any reconstituted formula left in the bottle after the feed. Discard reconstituted formula out of the fridge (4°C) for more than 4 h.

BMF Store expressed breast milk at £20°C (−4°F). Each mother’s expressed breast milk must be stored in a separate container. BMF should be stored in the refrigerator at 2–4°C (35–40°F) and should be used within 24 h. Fortifiers must be measured accurately, using aseptic techniques.

BMF, human breast milk plus powdered breast milk fortifiers.

125

125, 156

48 119

Colony counts of aerobic mesophilic bacteria and E. sakazakii did not increase significantly in either unfortified or fortified breast milk samples over 6 h. Supplementing breast milk fortifier with up to 1.44 mg iron/14 kcal does not compromise antibacterial activity of human milk held at room temperature for 6 h. Support ADA recommendations. a

Reference(s)

150

13.  Cronobacter Species tolerance of Cronobacter were studied by Breeuwer et al. (10). They demonstrated that stationary-phase cells of Cronobacter are more resistant to drying than are exponential-phase cells. Further work in this area done with E. coli has demonstrated that the trehalose synthesis genes are induced by stationary-phase sigma factor rpoS (50, 82, 83) and that the exposure of the organism to dry conditions results in permeability changes caused by the influx of trehalose. Nonstressed stationary-phase E. coli cells did not accumulate trehalose, similar to the situation observed with nonstressed stationary-phase cells of Cronobacter. The trehalose concentration in nonstressed Cronobacter stationary-phase cells was ca. 0.040 µmol/mg protein, while in dried stationary-phase cells, the concentration increased to 0.23 µmol/mg protein. It would appear that dry stress of Cronobacter stationary-phase cells could be a prerequisite for its accumulation of trehalose. The higher synthesis of trehalose and the induction of periplasmic trehalose in stationary-phase cells may stimulate the transport of nonfunctional trehalose to the periplasm, where it is converted to glucose. This cycle was described in a study with E. coli, in which mutant strains defective in periplasmic trehalose accumulated large amounts of trehalose in the medium (148). Breeuwer et al. (10) demonstrated that after air drying Cronobacter stationary-phase cell cultures in an incubator at 25°C and air humidity of 20.7%, initial populations of 9 log CFU/ml decreased by only 1 to 1.5 log when resuspended from the air-dried state 46 days later and by 1.5 to 2.5 log when air dried and maintained for the same period of time at 45°C. Similar r­esults were obtained for air-dried cells maintained in desiccators with saturated solutions of LiCl (aw = 0.113), potassium acetate (aw = 0.225), or magnesium nitrate (aw = 0.529). Not surprisingly, exponential-phase cells of Cronobacter were more sensitive to drying and decreased in greater numbers than stationary-phase cells when exposed to a low aw. In the same study, Breeuwer et al. (10) demonstrated that Cronobacter survived better in dry environments than did Salmonella and E. coli. Therefore, the colonization of postpasteurization environments by Cronobacter species could represent a greater concern in dry products that become contaminated during ingredient mixing, filling, or packaging. The greater survival of air-dried Cronobacter species at elevated temperatures such as 45°C to 47°C represents another advantage of this bacterium in the warm and dry environments surrounding the areas of drying equipment in food factories.

317 Edelson-Mammel and Buchanan (28) studied the long-term survival of Cronobacter species in PIF by inoculation of approximately 6 log CFU/ml of Cronobacter species into PIF. The inoculated samples were stored at room temperature in a closed screw-cap bottle for >1.5 years. Levels of Cronobacter species cells declined 2.5 log during the initial 5 months of storage, at a rate of approximately 0.5 log per month, with another 0.5-log decline observed in the following year, reaching a final count of 3 log CFU/ml at the end of the 1.5-year study. This study demonstrated that Cronobacter can survive for extended periods in a dry environment such as PIF. Further studies are required to address the survival of Cronobacter in infant formula stored at higher temperatures and moisture levels such as in tropical countries, preferably with naturally contaminated product.

Biological Inactivation

Control of Cronobacter species through the use of bacteriophages has been investigated by Kim et al. (75). Using sewage and UV irradiation of pure cultures, six bacteriophages were isolated. Two bacteriophages specific to the species C. sakazakii were isolated to demonstrate that they could be used to control growth both in media and in reconstituted infant formula at 12, 24, and 37°C. Phages at 107, 108, or 109 PFU/ml were used to demonstrate that higher temperatures (24 and 37°C) and high titers (109) were required for contamination levels equivalent to 73,000 CFU per 100 gram of PIF. Zuber et al. (165) isolated 67 phages and tested their lytic abilities on a 40-strain set of C. sakazakii. They reported that a cocktail of 5 phages was successful in preventing growth of 35 (of the 40) strains in artificially contaminated formula. It will be interesting to see how this field progresses with respect to isolating phages that have broad-spectrum activity against the genus while being able to manipulate them to be effective at targeting low levels of Cronobacter in PIF.

Chemical Inactivation

Cronobacter has been isolated from acidic food products such as fermented bread and beverages (average pH of 4.0) (22, 39, 115). Nair et al. (112) studied the antibacterial properties of caprylic acid on Cronobacter in reconstituted PIF. Caprylic acid is a natural eight-carbon chain fatty acid present in human breast milk and has generally recognized as safe status in the United States. In their study, Nair et al. (112) inoculated reconstituted PIF with 6 log CFU/ml of Cronobacter containing monocaprylin at a final concentration of 25 or 50 mM, with samples being incubated at 4, 8, 23, or 37°C. At concentrations of 50 mM, monocaprylin rapidly

318 i­nactivated Cronobacter at 37°C, reducing its population to undetectable levels even after enrichment, while at 25 mM monocaprylin, Cronobacter populations were reduced by 4.5 log after 24 h. At room temperature, the number of Cronobacter species decreased by 5 and 4 log in the presence of 50 and 25 mM monocaprylin, respectively, in 1 h. Cronobacter species remained at <10 CFU/ ml after 24 h in the presence of 50 mM monocaprylin but increased in number from 2 to 4.5 log CFU/ml at the end of 24 h in the presence of 25 mM monocaprylin. At 4 and 8°C, both concentrations of monocaprylin were effective in inactivating Cronobacter species in the reconstituted PIF. In the United States, monoglycerides are currently added to infant formula as emulsifiers. Therefore, the incorporation of monocaprylin as an antimicrobial ingredient in PIF is potentially feasible. Recently, Yemis¸ et al. (164) investigated the antibacterial activity of vanillin, ethyl vanillin, and vanillic acid as possible preservatives in the intervention strategy for control of Cronobacter species in foods. Seven strains of Cronobacter species were used to evaluate the effectiveness of the compounds on the growth and heat resistance. Vanillin, ethyl vanillin, and vanillic acid were shown to have bactericidal effects, prevent the growth of the organism, and reduce the heat resistance in microbiological media.

Competitive Exclusion/Probiotics

Telang et al. (150) evaluated the growth of resident aerobic mesophilic flora on artificially inoculated C. sakazakii in fresh human breast milk, fresh human breast milk with fortifiers, and reconstituted PIF. Samples were inoculated with 2 or 3 log CFU/ml of C. sakazakii and were maintained at 22°C with sampling every 2 h until hour 6. Total counts of C. sakazakii in both fortified and nonfortified human breast milk as well as in infant formula increased less than 1 log and were not significantly different over the 6-h period. In contrast, Chan (14) demonstrated a greater increase of total bacterial growth in fortified human breast milk than in nonfortified breast milk. Chan suggested that the high iron content in breast milk fortifier formulations could play a role in enhancing the growth of bacteria in the fortified breast milk, due to the inhibition of lactoferrin in highiron environments. Lactoferrin is a well known ironbinding glycoprotein that is present in human breast milk and is believed to have antibacterial properties. The survival and growth of C. sakazakii in infant rice cereal was investigated by Richards et al. (132). Water, apple juice, milk, or liquid infant formula was used to reconstitute infant rice cereal prior to inoculation with a 10-strain cocktail of C. sakazakii at populations of

Foodborne Pathogenic Bacteria 0.27, 0.93, and 9.3 CFU/ml, followed by incubation at 4, 12, 21, or 30°C for up to 72 h. Growth was not observed in the samples reconstituted with apple juice at any temperature or in the samples rehydrated with water, milk, or formula incubated at 4°C. The lag time observed for growth in cereal reconstituted with water, milk, or formula increased as the incubation temperature decreased (lag time12°C > lag time21°C > lag time30°C). In addition, it was also found that once the bacterial population reached its maximum cell density (8 log CFU/ml), numbers decreased, in some cases to nondetectable levels during subsequent storage, concurrent with a decrease in pH. Our laboratory completed a study on the growth of Cronobacter in reconstituted PIF and human breast milk with and without fortification at 10, 23, and 37°C (90). In this study, reconstituted PIF, human breast milk, or human breast milk with fortifiers was inoculated with approximately 10 CFU/ml of clinical, environmental, or food isolates of Cronobacter, incubated at 10, 23, or 37°C, and sampled periodically. A total of 105 growth curves were modeled using the modified Gompertz model, from which estimated lag and generation time parameters were derived (90). At 10°C, growth was observed in 14 of the 23 curves, and interestingly, Cronobacter did not grow in fortified breast milk at 10°C. At 23 and 37°C, growth occurred in all experiments, with longer generation times being observed at 23°C. Lower generation time parameters were seen in PIF than in breast milk or fortified breast milk.

NICHES AND RESERVOIRS The natural habitat of Cronobacter species remains unknown. In recent years, efforts have focused on detecting this organism in a wide variety of environments. Not surprisingly, like other members of the family Enterobacteriaceae, Cronobacter has been isolated from environmental samples such as water, dust, soil, plant materials, mud, and even household vacuum cleaner bags, indicating that its ecological niche is quite diverse. This diversity also supports the notion that its reservoir is likely an environmental source. Interestingly, new vectors and sources of contamination such as flies and rodents (44, 68, 80) have been recently reported. Although Muytjens and Kollee (107) were not able to isolate Cronobacter from cattle, cattle milk, domesticated animals, rodents, bird dung, grain, rotting wood, mud, soil, or surface water, several other reports have described the isolation of Cronobacter from various sources such as food manufacturing facilities and food products (e.g., powdered milk, cereal, chocolate, potato flour, pasta,

13.  Cronobacter Species and spices), households, water samples, dust, and grass silage (36, 43, 67, 68, 106, 158). In a survey of water springs in Spain that included waters from hypothermal (<30°C), mesothermal (30 to 40°C), and hyperthermal (>40°C) sources, 665 strains were isolated, and among them some were identified as Cronobacter species (106). Cronobacter has been recovered from clinical specimens such as cerebrospinal fluid (CSF), blood, sputum, throat, nose, stool, gut, skin, wounds, bone marrow, eye, ear, stomach aspirates, anal swabs, and the breast abscess of infected patients (8, 35, 37, 47, 77, 78, 85, 105, 109, 111, 130, 135, 144, 151, 157, 162). Farmer et al. (32) reported the isolation of Cronobacter from the respiratory tract of 29 patients in one hospital over a 7-month time period. Cronobacter has also been shown to survive in clinical settings over extended periods of time. For example, based on ribotyping, NazarowecWhite and Farber (116) reported the persistence of one isolate of Cronobacter in the same hospital over an 11year period. Foods such as milk powders, cheese products, and dry food ingredients such as herbs, spices, and rice seeds (21) have also been reported to contain Cronobacter; however, these foods have never been linked to human illnesses. In addition to its ability to survive in low-aw environments, Cronobacter can also survive in acidic food products. For example, it has been isolated from a Saudi Arabian fermented bread called khamir (pH 3.8) (39), from two traditional drinks consumed in Jordan, one with a final pH of 8.6 and the other with a pH of 2.8 (39), and from a fermented cassava product (pH 4.4) known as attiéké, popular in African countries (22). The major food commodity associated with Cronobacter infection in neonates is PIF (121, 127, 157). This bacterium was isolated for the first time from an unopened can of dried milk by Farmer et al. in 1980 (32). Muytjens et al. (110) isolated Enterobacteriaceae from 52.5% of 141 infant formula cans from 35 countries, with Cronobacter being detected in 20 of 141 (14.2%) samples from 13 of the 35 countries. Leuschner and Bew (93) isolated Cronobacter from 8 of 58 (13.8%) samples of infant formula from 11 countries. Iversen and Forsythe (57) surveyed 82 samples of powdered PIF and 404 other food products for the presence of Enterobacteriaceae. Cronobacter was isolated from two of the formulas, 5 of 49 (10.2%) dried infant foods, 3 of 72 (4.1%) milk powders, 2 of 62 (3.2%) cheese products, and various dry food ingredients, including 40 of 122 (37.8%) herbs and spices. Soriano et al. (146) iso-

319 lated Cronobacter from raw lettuce from a survey of products consumed in restaurants in Spain. Leclercq et al. (84), in a study done to compare fecal coliform agar and violet red bile lactose agar for the enumeration of fecal coliform in foods, isolated Cronobacter from cheese, minced beef, sausage meat, and vegetables. Sprouts (alfafa and mung bean) have been reported twice as a source of Cronobacter (23, 136). In our laboratory, we isolated this bacterium from crab meat (Lenati et al., unpublished) using buffered peptone water (BPW) and modified lauryl sulfate tryptose broth (mLST) as the enrichment and selective enrichment broths, respectively. Using the most-probable-number (MPN) method to enumerate the microorganism, we estimated a Cronobacter load of >4 log CFU/g. The detection of Cronobacter in PIF that meets the current microbiological standards has led policy makers in many countries to propose new microbiological criteria for PIF products (Table 13.4). A list of the various foods from which Cronobacter has been isolated can be seen in Table 13.5. PIF was introduced as a replacement for human breast milk more than 50 years ago. The powder form constitutes over 80% of the infant formula used worldwide. In the United States, the consumption of PIF in 2005 was greater than 6 billion 8-oz bottles in addition to 133 million 8-oz bottles of follow-on powder (A. C. Nielsen, personal communication). The powdered form has advantages over the liquid form, in terms of both cost and storage. However, while the liquid form is sterile, the powder may contain low levels of microorganisms. The understanding of the behavior of Cronobacter cells in dry products is a key element to Table 13.4  Microbiological criteria to be applied to PIFa Bacterium Mesophilic aerobic bacteriac Enterobacteriaceaed Salmonella Cronobacter

M

Class planb

10/g

10/g

3

0/g 0/25 g 0/10 g

0/g 0/g 0/g

2 2 2

n

c

m

5

2

10 60 30

0 0 0

a ISO methods are to be used for all determinations listed. n, number of sample units examined from a lot; c, number of samples being accepted between m and M; m, microbial limit separating good quality from marginally acceptable quality; M, microbial limit separating marginally acceptable quality from defective quality. Data from reference 20. b Class plan 2, product is classified into one of 2 groups, i.e., above or below M; class plan 3, product is classified into one of 3 groups, i.e., below m, at M, or between M and m. c The proposed criteria for mesophilic aerobic bacteria are reflective of GMPs and do not include nonpathogenic microorganisms that may be intentionally added, such as probiotics. d Reductions in the levels of Enterobacteriaceae in PIF will lead to lower populations of Cronobacter.

Foodborne Pathogenic Bacteria

320 Table 13.5  Foods from which Cronobacter spp. have been isolated Food type Powdered milk PIF PIF

Fermentation of sorghum bread (khamir)

Incidence

<1 CFU/g from 14.2% of the products from 13 of the 35 countries 1 × 104 CFU/g prefermentation), decreasing (<1 CFU/ml) after 2 × 24-h fermentation

PIF

PIF

PIF PIF

PIF

Raw lettuce Raw rice PIF Cheese, minced beef, sausage meat, and vegetables Sous and tamarind drinks Fermented cassava (attiéké)

PIF, dried infant foods, milk powders, cheese products, dry food ingredients (herbs and spices)

Mung bean sprouts

Alfalfa sprouts Crab meat

a

1/40 samples 20 Cronobacter/one lot

6.3 × 102 CFU/ml in sous drink; 2 CFU/ml in tamarind drink Inoculum (2.3 × 106 CFU/g); grated pulp (1.8 × 105 CFU/ml); fermented pulp (2.3 × 102) 2/82 (infant formula), 5/49 (dried infant foods), 3/72 (milk powders), 2/62 (cheese products), 40/122 (herbs and spices) 10/74 + vea for E. sakazakii using 25-g samples for testing 2 × 102 to 2 × 107 CFU/g Cronobacter, among Enterobacter spp. in 25% of 300 sprouts tested 8% + ve (5/60) >1.1 × 104 CFU/g

+ ve, positive results for food matrices tested for E. sakazakii.

Comments First reported isolation from powdered milk Former Czechoslovakia Survey of 141 milk substitutes obtained in 35 countries

Reference(s) 32

Prefermentation pH, 6.77; final product pH, 3.93

39

Plasmid profiles done by Clark et al. (17) showed a similar profile for infant formula and patient isolates 22 infant formula isolates had biotypes, antibiograms, and plasmid profiles identical to isolates from 4 infants, reported by Clark et al. (17) Cronobacter was isolated from a blender used to prepare formula Isolates from stomach aspirate, anal swab, and/or blood sample for 6 of the 12 neonates were reported to be similar to isolate from infant formula (PCR) First outbreak associated with infant formula. There was a total of 9 cases, 2 infected and 7 colonized infants in a neonatal intensive care unit Restaurant in Spain Farms in Philippines Isolated from blender after an infant case in a hospital in Jerusalem France

17, 139

Sous drink pH, 8.6; tamarind drink pH, 2.8 Inoculum pH, 5.0; grated pulp pH, 6.2; fermented pulp pH, 4.4

115

127 110

8, 17

121 157

159

146 21 9 84

21

57

Indonesia

31

136

Mexico City MPN method used for enumeration

23 Lenati, Hébert, Farber, et al., unpublished

13.  Cronobacter Species be considered in the evaluation of potential treatments for inactivating Cronobacter and other pathogens in PIF (34). Current processing technology is unable to completely eliminate the potential for microbial contamination in PIF without affecting its organoleptic and nutritional requirements. Based on currently available knowledge, sterilization of the final product in its dry form using cans or sachets seems possible only by using irradiation. However, the doses that are likely to be required to inactivate Cronobacter do not appear to be feasible due to organoleptic deterioration of the product. A number of other technologies, such as ultrahigh pressure and magnetic fields, combined with other potential hurdles, may be potential candidates in the future.

ISOLATION METHODS One of the most commonly used methods for isolation of Cronobacter from PIF was described by the FDA (155). Developed in 2002, it is based on the isolation method described by Muytjens et al. in 1988 (110). The FDA method was developed for the isolation and enumeration of Cronobacter from PIF. It consists of preenrichment of the IFP in sterile distilled water overnight at 36°C, followed by selective enrichment in Enterobacteriaceae enrichment (EE) broth overnight at 36°C. Samples are surface plated (100 µl) onto violet red bile glucose (VRBG) agar and incubated overnight at 36°C. Presumptive colonies are selected and streaked onto TSA, and after 48 to 72 h of incubation at 25°C, yellow colonies are selected for biochemical tests using API 20E test strips. Kandhai et al. (68) analyzed 152 environmental samples using isolation methods with and without an enrichment step. Approximately 65% of the samples were tested without being enriched, 35% were tested after enrichment, and 18% were tested with both methods for comparative purposes. The enrichment used was BPW, incubated for 20 to 24 h at 36°C, followed by streaking onto VRBG agar media. When the enrichment step was not used, samples were streaked directly onto VRBG. Presumptive Cronobacter colonies were selected and streaked onto TSA media and incubated for 48 h at room temperature. Yellow colonies were put through the API 20E kit as well as tested for a-glucosidase activity. Ribotyping was also done on presumptive positive isolates. From this study, the researchers concluded that Cronobacter could be isolated from environmental samples with or without the enrichment step and that, combined, yellow pigmentation and a-glucosidase-positive activity would be sufficient for Cronobacter identification. The method

321 proposed by Kandhai et al. (68) is shorter than that of the FDA and could be useful in routine screening of environmental samples from food manufacturing sites. Much work has been done in looking into the development of rapid, sensitive, and accurate methods for isolation of Cronobacter. In 2004, Leuschner and Bew (93) described a modified version of the FDA method for Cronobacter, which was validated in 16 laboratories from 8 European countries. The modification suggested was the use of nutrient agar supplemented with 4-methyl-umbelliferyl-a-d-glucosidase (NA-aMUG). The preenrichment and selective enrichment steps were similar to those in previously described methods. Selective enrichment broths were streaked in parallel onto VRBG and NA-a-MUG plates. On NA-a-MUG plates, Cronobacter colonies are yellow under normal light and show blue-violet fluorescence under UV light. The authors recommended the use of both agar media for the presumptive detection of Cronobacter. Another recently developed chromogenic medium (Oh-Kang [OK] medium) uses a-MUG as a selective marker in a differential medium for Cronobacter (122). The fluorogenic substrate was added to VRBG, tryptone bile agar, and TSA media. On OK medium, Cronobacter shows a strong fluorogenic characteristic, which clearly distinguishes it from other microorganisms under UV light. All presumptive Cronobacter-positive fluorescent colonies were confirmed to be Cronobacter by API 20E strips. Iversen et al. (58) also described a new chromogenic medium named Druggan-Forsythe-Iversen (DFI), based on detecting the presence of the enzyme a-glucosidase using the substrate 5-bromo-4-chloro-3-indolyl-a-dglucopyranoside (XaGlc). Cronobacter hydrolyzes this substrate to an indigo pigment, producing blue-green colonies. The authors reported on a comparative study of the new medium with the FDA method and showed that 95 clinical and food isolates of Cronobacter were detected on DFI agar 2 days sooner than by the current FDA method. The characteristics of 148 strains representing 17 genera of the Enterobacteriaceae family other than Cronobacter were also compared using the two methods. Only a few isolates of Escherichia vulneris, Pantoea spp., and C. koseri strains gave falsepositive results on DFI agar. A few a-glucosidase-positive strains were identified as Pantoea spp. based on API 20E biochemical profiles but had higher percentage identification as Cronobacter with the ID32E biochemical test strip. It was concluded that the DFI medium enables the detection of Cronobacter within mixed cultures of Enterobacteriaceae, something not

322 feasible with VRBG agar, a general selective medium for Enterobacteriaceae. A new promising selective agar, the Enterobacter sakazakii Plating Medium (ESPM), has recently been described (131). Its innovative characteristic is the use of two chromogenic substrates that are Cronobacter specific, carbohydrates that are not metabolized by Cronobacter, and a pH indicator to indicate the presence of other bacteria that ferment these sources of sugar, diminishing the possibility of false-positive results. A rapid 6-h screening medium consisting of a bisugar medium containing sucrose and melibiose was also developed. Presumptive colonies on ESPM can be streaked to the bisugar plate (131). Guillaume-Gentil et al. (41) described a method for the detection and identification of Cronobacter from environmental samples. Their method includes enrichment using mLST broth and an incubation temperature of 45°C for 22 to 24 h, followed by streaking onto TSA containing bile salts (TSAB). The group suggested exposure to light during incubation of the TSAB plates at 37°C, in order to observe the characteristic yellow colonies, which were confirmed using API 20E and aglucosidase activity tests. This method was superior (40% positives) to the “reference” method (enrichment in BPW followed by isolation on VRBG agar), which yielded 26% positives from a total of 192 environmental samples. A real-time PCR method for the detection of Cronobacter in infant formula in which the specific target sequence was within the macromolecular synthesis operon has been developed (138). The macromolecular synthesis operon consists of three genes, fpsU, dnaG, and rpoD, which are involved in the synthesis of DNA, RNA, and protein. The dnaG-rpoD intergenic sequence differs in length and primary sequence between species. The 5¢-nuclease-based amplifications are often more specific than standard PCR amplifications due to a requirement for 100% homology between probe and template, in addition to primer template specificity. The assay detected 100 CFU/ml in pure culture and in reconstituted infant formula using 50 cycles of PCR without the need for enrichment. Gurtler and Beuchat (42) compared the ability of spiral plating and ecometric techniques to recover stressed cells of Cronobacter using a variety of selective and differential agars. Five stress conditions were used to test the survival of Cronobacter cells on TSA supplemented with 0.1% pyruvate (TSAP, a nonselective control medium); Leuschner, Baird, Donald, and Cox (LBDC) agar (a differential, nonselective medium); OK agar; fecal coliform agar (FCA); DFI medium; VRBG agar; and EE

Foodborne Pathogenic Bacteria agar. Cronobacter cells were stressed using heat (55°C for 5 min), freeze-thawin­g (−20°C for 24 h, thawed, frozen again at −20°C for 2 h, thawed), acidification at pH 3.54, alkaline conditions (pH 11.25), and desiccation in PIF (aw, 0.25; 21°C for 31 days). The study revealed the following order for spiral plating recovery for heat-, freeze-thawing-, acid-, and alkaline-stressed cells, as well as unstressed control cells: TSAP > LBDC > FCA > OK > VRBG > DFI > EE. Desiccation stress, however, gave different results: TSAP = LBDC = FCA = OK > DFI = VRBG = EE. ESPM agar was included in the recovery study using the ecometric technique, with better recovery being observed. From best to worst in recovering stressed cells of Cronobacter, the following order was observed: TSAP = LBDC > FCA > ESPM = VRBG = OK > DFI = EE. The study by Gurtler and Beuchat (42) highlights the fact that while some of the newer differential and chromogenic media are useful when using pure cultures of Cronobacter, variability in recovery does occur. Interestingly, in this study, spiral plating was considered better than the ecometric technique when attempting to evaluate the performance of the recovery media. Equally noteworthy is the relatively poor performance of EE agar, further emphasizing the need for better isolation methodologies that are aimed at detection and/or isolation in the absence of any preenrichment or selective enrichment. Table 13.6 summarizes some of the current methods used to detect and/or enumerate Cronobacter. The International Organization for Standardization (ISO) (54) developed a methodology for the isolation of Cronobacter from milk and milk products that includes the use of the selective enrichment broth mLST incubated at 45°C and containing vancomycin (10 µg/ml) as well as a higher concentration of NaCl (34 g/liter), in addition to a chromogenic agar media.

SYMPTOMS, AT-RISK POPULATIONS, AND FOODBORNE OUTBREAKS Cronobacter has been mainly associated with necrotizing enterocolitis (NEC), septicemia, and meningitis. Neurological sequelae are commonly reported and include brain abscess and infarction, ventricle compartmentalization due to necrosis of brain tissue and l­iquefaction of white cerebral matter, and cranial cystic changes, as well as hemorrhagic and nonhemorrhagic intercerebral infarctions leading to cystic encephalomalacia (13, 78, 147, 162). While Cronobacter has caused disease in all age groups, on the basis of the age distribution of reported

13.  Cronobacter Species

323

Table 13.6  Summary of isolation and enumeration methods for Cronobacter spp.a Format Biochemical

Selective broth

Selective agar

EE broth, at 37°C mLST broth, at 45°C

VRBG Chromogenic Cronobacter media (5-bromo-4-chloro3-indolyl-a,dglucopyranosid­e + dimethylformamide)

BPW, at 37°C

VRBG

EE broth, at 37°C

mLST broth, at 45°C

VRBG, NA-a-MUG (fluorescence under UV light) OK media (with 4-MUG) DFI (Oxoid) (XaGlc) Cronobacter-specific plating medium (ESPM), pH indicator for fermentation of sugars not fermented by Cronobacter ESIA chromogenic medium TSA + bile salts

DNA (16s rRNA gene) DNA (real-time PCR)

NA EE broth

NA VRBG agar

DNA (PCR-oligonucleotide array) DNA (16s rDNA and IAC) DNA (BAX PCR)

mLST + van, at 44°C, 20 h NA

BHI, 5 h, followed by DNA extraction NA

mLST + van (45°C, 20–22 h)

BHI, 3 h, 37°C

NAb NA NA

ESSB, at 37°C

a b

Confirmation method

Durationb

Reference or source

7 days 7 days

155 54

5–6 days

68

4 days

93

NA NA NA

122 58 131

2 days

AES Labs

Yellow colonies confirmed using API 20E and a-glucosidase activity NA MMS operon (fpsU, dnaG, and rpoD) ITS G and ITS IA-operon NA

3–5 days

41

4h 2 days

89 138

2 days

94

2h

100

PCR results on machine, isolation from enrichment broths to nutrient agar

2 days

API 20E Oxidase, l-ornithine, l-arginine, d-sorbitol, l-rhamnose, d-sucrose, d-melibiose, amygdaline, and citrate API 20E, a-glucosidase API 20E

API 20E API 20E, ID 32E NA

NA

Dupont Qualicon

Abbreviations: NA, not available; van, vancomycin. Duration is approximate and includes confirmation using biochemical tests such as the API 20E.

cases, it was deduced that the group at particular risk is infants less than 1 year old (34). Among infants, the immunocompromised and neonates younger than 28 days are considered to be at greatest risk; lowbirth-weight (LBW) neonates, weighing less than 2.5 kg at birth, could possibly be at even greater risk.

However, a survey by Stoll et al. (147) identified only one case of Cronobacter sepsis among 10,660 LBW infants, suggesting that outside an epidemic situation, LBW infants are likely not a group at high risk for Cronobacter infection. Sondheimer et al. (145) found that premature or term neonates secrete less gastric

324 acid than older infants, a potentially important factor contributing to the increased survival of Cronobacter during its passage through the stomach and then into the intestine. Globally, while there are few data on active surveillance of Cronobacter spp.-related illness, available data suggest that among infants, neonates and infants less than 2 months of age are at greatest risk for infection (19). An additional issue in developing countries is infants of HIV-positive mothers, considering that they may be fed solely PIF in order to prevent the documented transmission of the virus from HIV-positive mothers to infants through breastfeeding (34). The first reported cases of Cronobacter occurred in 1958 in England, and since then, cases of neonatal and infant infections associated with this bacterium have been reported and further described in many regions of the world including Belgium, Canada, Denmark, France, Germany, Greece, Iceland, Israel, The Netherlands, Spain, United Kingdom, New Zealand, and nine states in the United States. A summary of the reported cases of Cronobacter infections reported from 1958 until 2011, including symptoms and outcomes, is given in Table 13.7. Most of the Cronobacter infections that have been reported have occurred in developed nations. The lack of reports from developing countries may be due to lack of awareness of the organism and methods to isolate it, rather than the absence of illness related to this microorganism. Because of this general lack of awareness of this organism, combined with poor existing methodologies, there are likely to be a number of unreported or misdiagnosed cases, even in developed countries. Cases within adult populations have been reported, although with a much smaller incidence. Twenty cases of Cronobacter infection have been reported among adults, all of whom were immunocompromised (26, 30, 47, 63, 81, 128).

VIRULENCE FACTORS AND PATHOGENICITY In general, the most frequently cited risk factors for the acquisition of a Cronobacter infection are severe debilitating underlying illness and the prior use of antimicrobial agents, factors that facilitate the colonization of Cronobacter spp. in the human intestinal tract. The immature neonatal immune system may increase the risk of acquiring a Cronobacter infection. However, it is not known exactly which host and environmental factors need to be present in order to cause infection in neonates.

Foodborne Pathogenic Bacteria Since Cronobacter species are a “recently” recognized human pathogen, not much research has been conducted on this organism, and therefore little is known about its pathogenesis. Keller et al. (69) have indicated that the virulence factors of E. sakazakii (exotoxins, aerobactin, and hemagglutinin) may be similar to those of E. cloacae. However, in the last decade much progress has been made. Figure 13.1 shows a summary model of the relationship between the pathogen and the host. No additional data on dose-response models for C. sakazakii in humans have become available since the FAO/WHO meeting in 2004 (34). Studies involving suckling mice suggest that a large number of cells may be required to cause infection in healthy neonates (123). Iversen and Forsythe (55, 56) hypothesized a minimum infectious dose of 1,000 CFU, based on a comparison with E. coli O157:H7 and Listeria monocytogenes serotype 4b. They further surmised that based on an average number of 0.36 to 66 CFU of C. sakazakii/100 g, previously reported (110, 117) to be found in unopened cans of PIF, and the recommended average portion of 18 g of powder reconstituted for one single feeding, 14 generations would be needed to reach 1,000 CFU. This would take average times of 7 h at 37°C, 17.9 h at 21°C, 1.7 days at 18°C, and 7.9 days at 10°C, according to the generation times they observed in growth studies. In summary, therefore, PIF containing the levels of C. sakazakii previously reported is unlikely to cause illness unless there is temperature abuse, contamination during handling (17), long periods of storage, and/or a combination of these factors. The main route of entry for Cronobacter species appears to be through ingestion; therefore, factors related to contamination should be addressed. The vehicle of Cronobacter infections in infants and neonates is PIF. Major determinants that contribute to the presence and growth of the organism in PIF include temperature, equipment, and biofilms. PIF is normally stored at room temperature, which is approximately 23°C (15). The optimal temperature for Cronobacter species is 37°C; however, it has the ability to grow between 6 and 45°C (15). PIF are nonsterile; therefore, the composition of PIF and storage temperature of reconstituted PIF will enhance the growth of the organism. Another factor that has been found in the constitution of PIF is LPS (151a). LPS appear to play a role in compromising the intestinal barrier integrity, by allowing the organism to cross the gastrointestinal tract and become bloodborne (151a). Other sources of contamination include the equipment used to produce and reconstitute PIF.

13.  Cronobacter Species Occurrences of extrinsic contamination from utensils such as blenders, brushes, and spoons have been documented (6, 15, 49, 139). Biofilms have been defined as a group of bacterial cells that attach to a surface and form an exopolysaccharide that surrounds the group of bacteria and are also a major concern for contamination of PIF. Cronobacter species have been shown to attach to and form biofilms on abiotic surfaces (46, 64, 72, 73, 87). Abiotic surfaces used to demonstrate Cronobacter’s ability to form biofilms are materials commonly associated with PIF-feedin­g equipment and surfaces such as glass, stainless steel, polyvinyl chloride, polycarbonate, silicone, and enteral feeding tubes (46, 52, 64, 72, 73, 87). The ability of gram-negative bacteria to coordinate colonization and association with other cells by intercellular communication systems triggered by low-m­olecular-weight signaling compounds has been previously investigated (141, 160). N-Acyl derivatives of homoserine lactone (acyl HSLs) are examples of signaling molecules previously described in gramnegative bacteria. Modulation of the physiological processes controlled by acyl HSLs as well as by nonacyl HSL-mediated systems depends on the cell density and on the growth phase of the organisms. Gram et al. (40) have studied the production of acylated homoserine lactones (AHLs) in Enterobacteriaceae strains in naturally contaminated as well as artificially inoculated foods at concentrations of 5 to 7 log CFU/g, suggesting that AHLs could be implicated in regulating phenotypes important in food spoilage and thus could possibly play a role in food quality as well as food safety. While Escherichia and Salmonella do not synthesize quorum-sensing signaling molecules of the AHL type, they can detect the AHLs produced by other species (142, 143). The role of AHLs and Sdi for their possible involvement in, for example, biofilm formation for Cronobacter species remains to be elucidated. There are three main focuses in addressing the biofilm issue: (i) understanding the physiology and genetic basis of biofilms (i.e., identifying key genes); (ii) determining whether biofilm formation is a virulence factor for Cronobacter; and (iii) evaluating and determining whether there are proper treatments efficient in reducing biofilm formation (4, 24, 46, 64, 72, 73, 87). Amalaradjou and Venkitanaravanan have recently shown that trans-cinnamaldehyd­e, an ingredient in cinnamon oil, has the ability to inhibit and inactivate C. sakazakii biofilms in the presence or absence of PIF on polystyrene plates, stainless steel coupons, feeding bottle coupons, and enteral feeding

325 tube coupons at both 12 and 24°C (4). Hartmann et al. screened a library of mutants of strain ES5 (clinical isolate) using the crystal violet microtiter assay and found four mutants defective in the cellulose biosynthesis, three mutants in flagellar structure, three in basic functions (such as cell division, energy metabolism, and acid fermentation), one in virulence, and four in unknown functions (46). Of the 14 mutants, two hypothetical proteins, ESA_00281 and ESA_00282, proved to be important in biofilm formation structure, these data being further supported by confocal laser scanning microscopy (46). Hurrell et al. (52) reported that C. sakazakii strain ATCC 12868 was able to form a biofilm with a cell density of 107 CFU/cm. In addition to demonstrating that biofilm production was not correlated to capsule formation, they described how silver-impregnated enteral tubing would not be inhibitory to the organism, suggesting that silver may not be an adequate antibacterial against this organism. Recent research in Health Canada demonstrated that both diguanylate cyclase and threonine synthase play an important role in biofilm production (K. Abdesselam and F. J. Pagotto, unpublished data). The disruption of these genes greatly altered biofilm production, and the lack of biofilms was further supported by the visualization of the scanning electron microscope. Diguanylate cyclase’s major role in most gram-negative bacteria is to produce c-di-GMP, which has been referred to as the biofilm regulator. One of the first papers describing putative virulence factors studied was conducted in our lab by Pagotto et al., where they demonstrated, using a suckling mouse assay, Cronobacter species’ ability to produce an enterotoxin (123). This ability was further supported by in vitro assays using CHO, Vero, and Y1 cells, in which many Cronobacter strains exhibited cytopathic effects (123). Raghav and Aggarwal purified and identified biochemical characteristics of the enterotoxin. The purified toxin confirmed the Pagotto et al. study, since it demonstrated cytopathic effects in the suckling mouse assay (129). The toxin was found to have a mass of 66 kDa and optimal activity at pH 6 and was stable at 90°C when exposed for 30 minutes (123). Pagotto et al. also described the dose-response relationships in Cronobacter infection using mice that were intraperitoneally injected with levels of bacteria as low as 105 CFU per mouse, which were found to be lethal (123). Studies using neonatal mice as models for the purposes of investigating Cronobacter infection have been done using BALB/c, C57BL/6, and CD-1 mice (133).

Foodborne Pathogenic Bacteria

326

Table 13.7  Summary of Cronobacter species cases and outbreaks in infants reported in the literature Date and location

No. of cases and outcome

1958, St. Albans, England

2 infants

1965, Denmark

1 infant, 2 days old. Recovery at 4 months of age, yet experienced extreme mental impairment.

1979, Macon, GA

1 infant, 7 days old

1981, Indianapolis, IN

1 infant, 5 weeks old; recovery, after treatment. Severely developmentally delayed

1983, The Netherlands

8 cases in a 6-year period

1985, Athens, Greece

1 infant, 3 days old. Other infants were colonized but not infected by E. sakazakii. 3 infants. E. sakazakii recovered from urine, groin, and anal swabs of a 3-day-old asymptomatic child.

1986–1987, Reykjavik, Iceland

1987, Boston, MA, and New Orleans, LA

2 infants, 4 weeks and 8 days old

Disease symptoms and organs affected White cranial matter resulting in degeneration into soft hemorrhagic mass. Cronobacter (E. sakazakii) isolated from brain, cerebrospinal fluid (CSF), liver, and marrow swabs. Meningitis. E. sakazakii isolated from CSF. Blood, feces and throat cultures tested negative.

First reported case of nonmeningital bacteremia caused by E. sakazakii Increase of head’s circumference. Meningoencephalitis and cerebral ventricular compartmentation. Bulging fontanels and grand mal seizures. Meningitis. 2 of 8 patients experiencing necrotizing enterocolitis. E. sakazakii isolated from the blood and CSF of all patients. Septicemia caused by E. sakazakii commingled with K. pneumoniae First infant became mentally retarded and quadriplegic after recovery. Second child had Down’s syndrome and died of complications from the E. sakazakii i­nfection 5 days after birth. Third child developed a seizure disorder and was moderately delayed in all developmental areas. Meningitis, necessitating ventricular shunts. Cerebral destruction, developmental damage, and severe neurologic complications.

Comments

Reference

Isolate sensitive to chloramphenicol and streptomycin

154

Treatment with streptomycin, chloramphenicol, ampicillin, sulfadiazine, sulfadimidine, sulfamerazine, and sulfacombin Ampicillin treatment

65

Ampicillin and gentamicin treatments

77

E. sakazakii CSF isolates indistinguishable from isolates recovered from prepared infant formula and utensils used to prepare the formula Resistant to ampicillin, netilmicin, cefotaxime, and amikacin

111

105

5

Infants were fed PIF. E. sakazakii was recovered from five packages of infant formula. 22 of 23 isolates from infant formula were identical in biotype, antibiotic profile, and plasmid profile to the 3 neonatal strains.

8

The patients were never geographically proximate

162

(Continued)

13.  Cronobacter Species

327

Table 13.7  Summary of Cronobacter species cases and outbreaks in infants reported in the literature (Continued) Date and location

No. of cases and outcome

Disease symptoms and organs affected

Comments

Reference

Blender tested positive for E. sakazakii and E. cloacae. Isolates from infant formula and isolates from infants had the same plasmid and multilocus enzyme profile. Resistant to vancomycin and ampicillin. Blender used to rehydrate the PIF was heavily contaminated with both bacteria.

139

Ampicillin and cefotaxime treatment

37

Gentamicin, cefotaxime, ceftazidime, cefuroxime axetil, and clindamycin treatment

81

All patients were fed infant formula prior to the illness. A survey of unopened cans of PIF yielded 14 Cronobacter isolates.

157

1988, Memphis, TN

4 preterm neonates

Bacteremia, septicemia, urinary tract infection, abdominal distension, and bloody diarrhea or stool

1990, Baltimore, MD

1 infant, 6 months old

1990, Cincinnati, OH

1 infant, 2 days old

1995–1996, Boston, MA

5 cases within a 1-year period. Individuals of 3, 39, 73, 76, and 82 years of age.

1998, Belgium

12 infants; 4 patients required operative treatment; 2 patients (twins) died within 3 weeks of each other

Septicemia following small bowel complications (exploratory laparotomy and a gastrostomy tube). Blood cultures were positive for both E. sakazakii and Leuconostoc mesenteroides. Stool cultures were negative for both bacteria. Blood culture tested positive for the presence of Cronobacter. Hemorrhage, abscess, and brain infarction. All patients had potentially immunosuppressive illnesses under treatment. The 2 youngest patients survived infection. Blood, anal swabs, and stomach aspirates of 6 of the 12 patients were positive for E. sakazakii. 2 strains with two differing morphologies were isolated from 1 patient.

2000, Winston-Salem, NC

1 premature infant, 6 days old. No apparent neurological or developmental deficits after 5 weeks.

Brain abscess, high fever, irritability, and seizure activity. Abnormal cerebritis-like indicators in the frontal lobe of the brain. Draining of purulent fluid via craniotomy. Cronobacter was isolated from blood and CSF but not from urine.

Ampicillin, cefotaxime, and bactrim treatment. 1st reported case of E. sakazakii being isolated directly from a drained cranial abscess.

13

1993, 1995, 1997–2000, Israel

2 underweight infants, one being premature. 4 and 9 days of age, recovery after treatment. 3 infants colonized but not infected.

Meningitis, seizures, infarction, liquefaction, and cavitation of the brain. Ventriculoperitoneal shunt. E. sakazakii was recovered from the CSF and blood.

Cefotaxime treatment. Infants fed PIF. Antibiotic treatment did not eliminate E. sakazakii from colonized asymptomatic patients. Blender used to prepare infant formula tested positive for E. sakazakii.

6

121

(Continued)

Foodborne Pathogenic Bacteria

328

Table 13.7  Summary of Cronobacter species cases and outbreaks in infants reported in the literature (Continued) Date and location

No. of cases and outcome

Disease symptoms and organs affected

2001, Knoxville, TN

49 infants in a NICU screened; 10 infants tested E. sakazakii positive

2003, Brazil

14-day-old breast-fed infant girl

2004, France

9 infants

2 deaths

2004, New Zealand

1 premature infant. Waikato neonatal intensive care unit Two 5-month-old infants

Meningitis and death

2011, Mexico

7 of the 10 Cronobacterpositive neonates were colonized, but not infected. Source of the bacterium was traced to PIF. Failed ampicillin and ceftriaxone treatment. Meningitis and death.

Bloody diarrhea

Using a single strain, MNW2, Richardson et al. inoculated pups on postnatal day 3.5 with various concentrations in reconstituted PIF at levels ranging from 102 to 1011. C. sakazakii was isolated from brains, livers, and ceca in all three mouse models. It appeared that the CD-1 mouse was the most susceptible, having the lowest infectious and lethal dose of 102. In a follow-up study using CD-1 mice, Richardson et al. (134) compared the virulence of three Cronobacter strains, MNW2 (food isolate), SK81 (clinical isolate), and 3290 (clinical isolate) at doses ranging from 102.8 to 1010.5. While all strains were isolated from tissues examined, strain 3290 was more invasive in brains than MNW2 or SK81. However, strain SK81 demonstrated a higher mortality rate than MNW2 or 3290, suggesting that invasiveness and mortality were not directly linked in this animal model. Recently, Lee et al. (86) used pathogen-free, time-pregnant ICR mice and on postnatal day 3.5, 107 CFU of strain 3439 (isolated from PIF) was administered orally to pups. In six pups, death occurred at 3 days postinoculation. Among the five surviving pups, meningitis and gliosis were detected in the brain in three pups, including inflammatory cell infiltration and cellular degeneration in the liver. Similar findings were observed in the

Comments

Reference 159

Appears to be first report of transmission of E. sakazakii via human breast milk Linked to infant formula. Recall of the implicated infant formula by manufacturer. Common source and distribution of infant formula. Linked to infant formula

7

Linked to consumption of rehydrated PIF

33

1

119

cecum, leading the authors to conclude that the ICR model may be an effective model for human neonatal infections. The dose-response behavior in the suckling mouse assay may not necessarily be a good representation of the dose-response curve for human neonates, for which the oral infectious dose has been estimated by the World Health Organization to range from 103 to 108 CFU (15, 49, 163). The gastrointestinal tract is a primary target for Cronobacter pathogenesis, and its success is dependent on the bacterium’s ability to adhere to the intestinal epithelial layer (29, 51, 74, 101). Hunter et al. reported that Cronobacter species binds to enterocytes in rat pups, and a cascade of host responses are triggered, which include tumor necrosis factor alpha, nitric oxide, and interleukin-6 (IL-6), also observed when Cronobacter attached to the enterocytes of rat pups (51). They also observed an increase of enterocyte apoptosis. This increase in apoptosis results in a decrease in the barrier integrity, which appears to facilitate bacterial translocation. Adhesion to intestinal cells appears to be a necessary step in colonization of the intestinal tract, which induces NEC and invasion in the animal model and in IEC-6 cells (rat intestinal cell line) (51). A recent study conducted by Emami et

13.  Cronobacter Species

Biofilm

329

Equipment

Temperature

Protease unique to Cronobacter seems to contribute to crossing BBB

OmpA binds to fibronectin, facilitating invasion of brain endothelial cells Can cross the BBB

At-risk populations

Powdered Infant Formula (PIF)

Ingested Enterotoxin Higher stomach pH (>4) may assist in survival

Can cross GI barrier

Figure 13.1  A current model of Cronobacter pathogenicity. GI, gastrointestinal. doi:10.1128/9781555818463.ch13f1

al. used a mouse model and Caco-2 cells to show that Cronobacter sakazakii damages the intestinal epithelial cells by recruiting a large amount of dendritic cells in the intestine and suppressing their maturation, which increases the tumor growth factor b production, resulting in NEC in the host (29). Adhesion and invasion of the blood-brain barrier (BBB) and the gastrointestinal tract are crucial for infections of Cronobacter species in the host, and a better understanding of these processes would help in unraveling the mechanism(s) of pathogenesis. Mammalian tissue culture assays are a popular approach to investigate the adhesive and invasive properties. The most common cell lines used to study the adherence and invasion are HEp-2, Caco-2, and human brain microvascular endothelial (HBMEC) (51, 101, 104, 152). Mange et al. screened 50 Cronobacter strains and found two distinctive adherence patterns in all three cell lines tested. They were described as diffuse adhesion and/or formation of localized clusters of bacteria on the cell surface. Some strains were able to show both patterns (101). Mange et al. also demonstrated that adherence was optimal during late exponential phase of bacterial

growth with a 10-fold increase in adherence cells and also concluded that adhesion to epithelial and endothelial cells was predominantly nonfimbrially based (101). Townsend et al. studied the attachment and invasion properties of seven different Cronobacter strains a­ssociated with o­utbreaks of NEC, bacteremia, and meningitis using Caco-2 cells and rat brain capillary endothelial cells (152). All seven strains attached to and invaded Caco-2 cells after 3 h. These particular isolates have the ability to replicate and adapt in macrophage cells (U937) (152). Townsend et al. (153) demonstrated that the ratio of IL-10 to IL-12 was increased following phagocytocis by macrophages, suggesting that a type II immune response may not be sufficient in the absence of antimicrobial therapy. This group used human U937 macrophages to demonstrate that Cronobacter is able to persist once phagocytized, with strains differing in their ability to do so. The mechanism(s) of intracellular survival remains to be determined. Bacterial translocation from the gastrointestinal barrier is critical in the pathogenesis of Cronobacter meningitis (74, 101, 103, 104, 113, 140). Trans­ location from the GI tract provides the ­bacteria with

330 access to the bloodstream, which may lead to sepsis in the host and ultimately provide access to the central nervous system. Cronobacter species are able to overcome host defensive mechanisms, penetrate the BBB, and survive in the CSF (74, 101, 103, 113, 140). One virulence factor that has been shown to be essential in Cronobacter pathogenesis is outer membrane protein A (OmpA). Interestingly, OmpA plays a role in the suppression of the maturation of dendritic cells (29). OmpA expression is also required for invasion of mammalian host cells and is necessary for Cronobacter resistance to blood and serum killing in newborn rats (74, 102, 104, 140). Mittal et al. have shown that Cronobacter isolates that are positive for OmpA can successfully cross the intestinal barrier in newborn rats, multiply in the blood, and traverse the BBB, whereas isolates with ompA deletion mutations fail to bind to intestinal epithelial cells both in vivo and in vitro (103). Recent studies from Kim et al. demonstrated that OmpX is also required for translocation. They were the first to conduct in-frame deletion mutants for OmpA and OmpX and showed that translocation does not occur in the host when both proteins are not expressed (74). When the Cronobacter BAA894 genome and that of E. cloacae were compared using a BLAST search, ompX was found to be similar (81% identity) (74). OmpX in other bacteria, including E. cloacae, plays an important role in virulence since it assists in the invasion of the host cells and helps to overcome the host’s defenses (74). Kim et al. have demonstrated using Caco-2 and INT-407 cells and rat pups that both OmpA and OmpX need to be expressed in Cronobacter in order for translocation and basolateral invasion and adhesion of mammalian cells to occur (74). Singamsetty et al. noticed that several meningitiscausing gram-negative bacteria require the expression of OmpA to invade HBMEC. They demonstrated, using HBMEC, that OmpA is required in Cronobacter species to invade the BBB by inducing microtubule condensation and phosphatidylinositol 3-kinase and protein kinase C a activation (140). Both Nair et al. and Mittal et al. showed that OmpA binds to fibronectin, thereby facilitating the invasion of brain endothelial cells (103, 113). Mittal et al. were the first to demonstrate that OmpA expression affects the onset of meningitis in newborn rats (103). Mortality rate was 100% when newborn rats were infected with Cronobacter isolates from OmpA-positive strains, and no pathological symptoms were observed when using ompA deletion mutant strains (103).

Foodborne Pathogenic Bacteria Kothary et al. examined 135 different Cronobacter strains and identified a cell-bound zinc-containing metalloprotease encoded by a zpx gene (79). The protease was found to be active against azocasein, resulting in the rounding of Chinese hamster ovarian cells, and may be responsible for allowing the organism to cross the BBB or perhaps help in cellular destruction in neonates with NEC (79). A virulence mechanism that seems to be common with species belonging to the family of Enterobacteriaceae is their active efflux (49). This active efflux ejects a range of xenobiotic compounds from the cell, including bile salts, antibiotics, disinfectants, and dyes (49). Current studies are under way to fully understand the role of the transporters of the active efflux in virulence of Cronobacter isolates. Currently, we are looking at a number of nonprimate animal models to gain a better understanding of the mechanisms of pathogenesis and dose-response behavior of Cronobacter species. The neonatal gerbil model appears to be quite promising and would be used to investigate the dose-response relationship, as well as to identify the pathogenic potential of strains, such as growth in major organs and time of death.

ANTIBIOTIC RESISTANCE Exposing microorganisms to a wide array and concentrations of antimicrobials potentially can lead to an increase in the resistance to antibiotics available for human treatment. Several studies have described the resistance of Enterobacter isolates to quinolones, b-lactams, and trimethoprim-sulfamethoxazole (43, 126). Bacterial meningitis demands prompt and effective treatment, and knowledge of antibiotic susceptibility is of great importance regarding neonatal Cronobacter infections and their treatments. Cronobacter has been reported as being more sensitive than other Enterobacter spp. to some antibiotics including the aminoglycosides, ureidopenicillins, ampicillin, and carboxypenicillins (2, 43, 116, 162). Muytjens and van der Ros-van de Repe (108) tested the susceptibilities of 195 Cronobacter isolates from various sources such as the respiratory tract, digestive tract, CSF, superficial wounds, urine, upper respiratory tract, blood, and utensils against 29 antimicrobials and reported Cronobacter to be the most susceptible among the eight Enterobacter spp. tested. Concentrations of 24 of the 29 antibiotics necessary to inhibit at least 90% of the strains were £8 μg/ml, which was twofold lower than that required for inhibiting

13.  Cronobacter Species E. cloacae. Antibiotics having MICs greater than 8 μg/ml were chloramphenicol (16 μg/ml), cefaloridin (16 μg/ml), cefsulodin (32 μg/ml), cephalothin (>128 μg/ml), and sulfamethoxazole (>128 μg/ml). Lai (81) reported successful treatment of the first known cases of Cronobacter infections with ampicillin in combination with gentamicin or chloramphenicol. Willis and Robinson (162) described the ampicillin-gentamicin treatment as the “gold standard” for Cronobacter treatments; however, recent reports have described resistance of Cronobacter to the blactams, as well as to the gold standard antibiotics. Enterobacter organisms are known to be prolific in terms of their ability to inactivate broad-spectrum penicillins and cephalosporins, and this ability appears to be on the increase (16, 30, 81). Pitout et al. (126) tested eight strains of Cronobacter for susceptibility against ampicillin, ampicillin-sulbactam, amoxicillinclavulani­c acid, ticarcillin, ticarcillin-clavulanic acid, piperacillin, piperacillin-tazobactam, aztreonam, cephalothin, cefazolin, cefoxitin, cefotaxime, ceftriaxone, ceftazidime, cefepime, and imipenem. In this study, derepressed mutant phenotype strains were resistant to all antimicrobials except to imipenem, whereas the wild-type strain (possessing an inducible Bush group 1 b-lactamase) or basal strain (not expressing or expressing in low levels a Bush group 1 blactamase) was sensitive to all antimicrobials tested. One wild-type strain was resistant to cefoxitin. Block et al. (9) have also reported b-lactamase activity in Cronobacter isolates recovered from six neonatal and childhood infections. The enzyme was believed to be most likely part of Bush group 1 b-lactamase. Lai (81) and Weir (159) indicated the possibility of using carbapenems or the newer broad-spectrum cephalosporins together with an aminoglycoside or trimethoprim-sulfamethoxazole, to treat Cronobacter meningitis. Interestingly, Muytjens et al. (111) reported that the Cronobacter strains isolated from their study were susceptible in vitro to ampicillin, gentamicin, chloramphenicol, and kanamycin. However, six of eight patients responded poorly and died after being treated with these antimicrobials. Arseni et al. (5) also reported resistance to ampicillin treatment, among others (netilmicin, cefotaxime, and amikacin), in a neonatal case of septicemia caused by Cronobacter and possibly Klebsiella pneumoniae. In contrast, Naqvi et al. (114) successfully treated a Cronobacter infection in one patient by using cefotaxime, and Willis and Robinson (162) were able to effectively treat two

331 cases of Cronobacter-induced neonatal meningitis with moxalactam after observing that ampicillin-gentamicin therapy was unresponsive. Kleiman et al. (77) reported the resistance of a Cronobacter isolate from a 5-week-old female with meningoencephalitis to cephalothin (MIC = 16 μg/ml). Nazarowec-White and Farber (116) tested the antibiotic resistance of 17 strains of Cronobacter and found four antibiotic susceptibility patterns (antibiograms). The largest cluster contained five strains that were resistant to sulfisoxazole and cephalothin and susceptible to ampicillin, cefotaxime, chloramphenicol, gentamicin, kanamycin, polymyxin B, trimethoprimsulfamethoxazol­e, tetracycline, and streptomycin. The authors concluded that when compared to ribotyping, PFGE and randomly amplified polymorphic DNA, antibiogram patterns were the least discriminatory to distinguish bacterial strains. Farmer et al. (32) reported on the MICs of 13 antibiotics, namely, ampicillin, carbenicillin, cephalothin, amikacin, gentamicin, kanamycin, tobramycin, chloramphenicol, sulfisoxazole, tetracycline, nalidixic acid, nitrofurantoin, and trimethoprim-sulfamethoxazole, against 10 Cronobacter strains. They also tested the antibiotic resistance of 24 strains of Cronobacter against 12 antibiotics, showing that 100% of the strains tested were susceptible to gentamicin, kanamycin, chloramphenicol, and ampicillin, while none of them were susceptible to penicillin and only 13% were susceptible to cephalothin. While vancomycin is not commonly used to treat infections caused by Cronobacter, it is used at a concentration of 10 μg/ml in the selective broth mLST.

CONCLUSIONS Cronobacter species has become a growing concern for government regulatory agencies, health care providers (especially those in neonatal intensive care units), and PIF manufacturers. In the past few years, great efforts have been undertaken by regulatory agencies and the scientific community to acquire new knowledge on this emerging opportunistic pathogen. With the push to sequencing whole genomes as a result of increases in technology and decreases in cost, it is expected that in the next 3 to 5 years, our understanding of this organism will increase tremendously and, as a result, science/e­vidence-based policies and guidelines will likely be developed to better control this organism and reduce the number of Cronobacter outbreaks related to PIF.

332

References 1. Acoignard, B., V. Vaillant, J. P. Vincent, A. Leflèche, P. Mariani-Kurkdjian, C. Bernet, F. L’Hériteau, H. Sénéchal, P. Grimont, E. Bingen, and J. C. Desenclos. 2006. Infections sévères à Enterobacter sakazakii chez des nouveau-nés ayant consommé une préparation en poudre pour nourrissons, France, octobre-décembre 2004. Bull. Epidemiol. Hebd. 2–3:10–13. 2. Adamson, D. H., and J. R. Rogers. 1981. Enterobacter sakazakii meningitis with sepsis. Clin. Microbiol. Newsl. 3:19–20. 3. Agostoni, C., I. Axelsson, O. Goulet, B. Koletzko, K. F. Michaelsen, J. W. Puntis, J. Rigo, R. Shamir, H. Szajewska, D. Turck, Y. Vandenplas, and L. T. Weaver. 2004. Preparation and handling of powdered infant formula: a commentary by the ESPGHAN Committee on Nutrition. J. Pediatr. Gastroenterol. Nutr. 39:320–322. 4. Amalaradjou, M. A., and K. Venkitanarayanan. 2011. Effect of trans-cinnamaldehyde on inhibition and inactivation of Cronobacter sakazakii biofilm on abiotic surfaces. J. Food Prot. 74:200–208. 5. Arseni, A., E. Malamou-Ladas, C. Koutsia, M. Xanthou, and E. Trikka. 1987. Outbreak of colonization of neonates with Enterobacter sakazakii. J. Hosp. Infect. 9:143–150. 6. Bar-Oz, B., A. Preminger, O. Peleg, C. Block, and I. Arad. 2001. Enterobacter sakazakii infection in the newborn. Acta Paediatr. 90:356–358. 7. Barreira, E. R., D. Costa de Souza, P. de Freitas Góis, and J. C. Fernandes. 2003. Meningite por Enterobacter sakazakii em recém-nascido: relato de caso. Pediatria (São Paulo) 25:65–70. 8. Biering, G., S. Karlsson, N. C. Clark, K. E. Jonsdottir, P. Ludvigsson, and O. Steingrimsson. 1989. Three cases of neonatal meningitis caused by Enterobacter sakazakii in powdered milk. J. Clin. Microbiol. 27:2054–2056. 9. Block, C., O. Peleg, N. Minster, B. Bar-Oz, A. Simhon, I. Arad, and M. Shapiro. 2002. Cluster of neonatal infections in Jerusalem due to unusual biochemical variant of Enterobacter sakazakii. Eur. J. Clin. Microbiol. Infect. Dis. 21:613–616. 9a. Bowen, A. B., and C. R. Braden. 2006. Invasive Enterobacter sakazakii disease in infants. Emerg. Infect. Dis. 12:1185–1189. 10. Breeuwer, P., A. Lardeau, M. Peterz, and H. M. Joosten. 2003. Desiccation and heat tolerance of Enterobacter sakazakii. J. Appl. Microbiol. 95:967–973. 11. Brenner, D. J. 1974. DNA reassociation for the clinical differentiation of enteric bacteria. Public Health Lab. 32:118–130. 12. Brenner, D. J., J. J. Farmer III, F. W. Hickman, M. A. Asbury, and A. G. Steigerwalt. 1977. Taxonomic and Nomenclature Changes in Enterobacteriaceae. Centers for Disease Control and Prevention, Atlanta, GA. 13. Burdette, J. H., and C. Santos. 2000. Enterobacter sakazakii brain abscess in the neonate: the importance of neuroradiologic imaging. Pediatr. Radiol. 30:33–34. 14. Chan, G. M. 2003. Effects of powdered human milk fortifiers on the antibacterial actions of human milk. J. Perinatol. 23:620–623.

Foodborne Pathogenic Bacteria 15. Chenu, J. W., and J. M. Cox. 2009. Cronobacter (‘Enterobacter sakazakii’): current status and future prospects. Lett. Appl. Microbiol. 49:153–159. 16. Chow, J. W., M. J. Fine, D. M. Shlaes, J. P. Quinn, D. C. Hooper, M. P. Johnson, R. Ramphal, M. M. Wagener, D. K. Miyashiro, and V. L. Yu. 1991. Enterobacter bacteremia: clinical features and emergence of antibiotic resistance during therapy. Ann. Intern. Med. 115:585–590. 17. Clark, N. C., B. C. Hill, C. M. O’Hara, O. Steingrimsson, and R. C. Cooksey. 1990. Epidemiologic typing of Enterobacter sakazakii in two neonatal nosocomial outbreaks. Diagn. Microbiol. Infect. Dis. 13:467–472. 18. Clementino, M. M., I. de Filippis, C. R. Nascimento, R. Branquinho, C. L. Rocha, and O. B. Martins. 2001. PCR analyses of tRNA intergenic spacer, 16S-23S internal transcribed spacer, and randomly amplified polymorphic DNA reveal inter- and intraspecific relationships of Enterobacter cloacae strains. J. Clin. Microbiol. 39:3865–3870. 19. Codex Alimentarius Commission. 2008. Joint FAO/WHO Food Standards Programme CODEX Committee on Food Hygiene, 40th Session. ftp://ftp.fao.org/codex/ccfh40/fh40_ 03e.pdf. 20. Codex Alimentarius Commission. 2008. Code of Hygienic Practice for Powdered Formulae for Infants and Young Children. CAC/RCP 66-2008. http://www.codexalimentarius.net/download/standards/11026/CXP_066e.pdf. 21. Cottyn, B., E. Regalado, B. Lanoot, M. De Cleene, T. W. Mew, and J. Swings. 2001. Bacterial populations associated with rice seed in the tropical environment. Phytopathology 91:282–292. 22. Coulin, P., Z. Farah, J. Assanvo, H. Spillmann, and Z. Puhan. 2005. Characterisation of the microflora of attieke, a fermented cassava product, during traditional small-scale preparation. Int. J. Food Microbiol. 106:131–136. 23. Cruz, A. C., E. Fernandez, E. Salinas, P. Ramirez, C. Montiel, and C. A. Eslaval. 2004. Characterization of Enterobacter sakazakii isolated from different sources, abstr. Q-051. Abstr. 104th Gen. Mtg. Am. Soc. Microbiol. American Society for Microbiology, Washington, DC. 24. Dancer, G. I., J. H. Mah, and D. H. Kang. 2009. Influences of milk components on biofilm formation of Cronobacter spp. (Enterobacter sakazakii). Lett. Appl. Microbiol. 48:718–725. 25. Dauga, C. 2002. Evolution of the gyrB gene and the molecular phylogeny of Enterobacteriaceae: a model molecule for molecular systematic studies. Int. J. Syst. Evol. Microbiol. 52:531–547. 26. Dennison, S. K., and J. Morris. 2002. Multiresistant Enterobacter sakazakii wound infection in an adult. Infect. Med. 19:533–535. 27. Edelson-Mammel, S. G. and R. L. Buchanan. 2004. Thermal inactivation of Enterobacter sakazakii in rehydrated infant formula. J. Food Prot. 67:60–63. 28. Edelson-Mammel, S. G., M. K. Porteous, and R. L. Buchanan. 2005. Survival of Enterobacter sakazakii in a dehydrated powdered infant formula. J. Food Prot. 68:1900–1902. 29. Emami, C. N., R. Mittal, L. Wang, H. R. Ford, and N. V. Prasadarao. 2011. Recruitment of dendritic cells

13.  Cronobacter Species

30.

31.

32.

33.

34.

35.

36. 37. 38.

39.

39a.

40.

41.

42.

is responsible for intestinal epithelial damage in the pathogenesis of necrotizing enterocolitis by Cronobacter sakazakii. J. Immunol. 186:7067–7079. Emery, C. L., and L. A. Weymouth. 1997. Detection and clinical significance of extended-spectrum betalactamases in a tertiary-care medical center. J. Clin. Microbiol. 35:2061–2067. Estuningsih, E. 2004. Present Work and Future Research Needs on Enterobacter sakazakii. FAO/WHO Regional Conference on Food Safety for Asia and Pacific, Seremban, Malaysia, 24 to 27 May 2004. FAO/WHO, Rome, Italy. ftp://ftp.fao.org/docrep/fao/meeting/006/ad731e.pdf. Farmer, J. J., M. A. Asbury, F. W. Hickman, D. J. Brenner, and the Enterobacteriaceae Study Group. 1980. Enterobacter sakazakii, new species of Enterobacteriaceae isolated from clinical specimens. Int. J. Syst. Bacteriol. 30:569–584. Flores, J. P., S. A. Medrano, J. S. Sánchez, and E. Fernndez-Escartín. 2011. Two cases of hemorrhagic diarrhea caused by Cronobacter sakazakii in hospitalized nursing infants associated with consumption of powdered infant formula. J. Food Prot. 74:2177–2181. Food and Agriculture Organization/World Health Organization. 2004. Joint FAO/WHO workshop on Enterobacter sakazakii and other microorganisms in powdered infant formula, Geneva, Switzerland, 2 to 5 February 2004. Food and Agriculture Organization/ World Health Organization, Geneva, Switzerland. http:// www.who.int/foodsafety/publications/micro/mra6/en/. Foreman, S. D., E. E. Smith, N. J. Ryan, and G. R. Hogan. 1984. Neonatal Citrobacter meningitis: pathogenesis of cerebral abscess formation. Ann. Neurol. 16: 655–659. Friedemann, M. 2007. Enterobacter sakazakii in food and beverages (other than infant formula and milk powder). Int. J. Food Microbiol. 116:1–10. Gallagher, P. G., and W. S. Ball. 1991. Cerebral infarctions due to CNS infection with Enterobacter sakazakii. Pediatr. Radiol. 21:135–136. Garrity, G., D. J. Brenner, N. Krigg, and J. T. Staley. 2005. Bergey’s Manual of Systematic Bacteriology, vol. II. The Protobacteria, Part B. The Gammaproteobacteria. Springer, New York, NY. Gassem, M. A. 1999. Study of the micro-organisms associated with the fermented bread (khamir) produced from sorghum in Gizan region, Saudi Arabia. J. Appl. Microbiol. 86:221–225. Gosney, M. A., M. V. Martin, A. E. Wright, and M. Gallagher. 2006. Enterobacter sakazakii in the mouths of stroke patients and its association with aspiration pneumonia. Eur. J. Intern. Med. 17:185–188. Gram, L., A. B. Christensen, L. Ravn, S. Molin, and M. Givskov. 1999. Production of acylated homoserine lactones by psychrotrophic members of the Enterobacteriaceae isolated from foods. Appl. Environ. Microbiol. 65:3458–3463. Guillaume-Gentil, O., V. Sonnard, M. C. Kandhai, J. D. Marugg, and H. Joosten. 2005. A simple and rapid cultural method for detection of Enterobacter sakazakii in environmental samples. J. Food Prot. 68:64–69. Gurtler, J. B., and L. R. Beuchat. 2005. Performance of media for recovering stressed cells of Enterobacter saka-

333

43. 44. 45. 46.

47. 48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

zakii as determined using spiral plating and ecometric techniques. Appl. Environ. Microbiol. 71:7661–7669. Gurtler, J. B., J. L. Kornacki, and L. R. Beuchat. 2005. Enterobacter sakazakii: a coliform of increased concern to infant health. Int. J. Food Microbiol. 104:1–34. Hamilton, J. V., M. J. Lehane, and H. R. Braig. 2003. Isolation of Enterobacter sakazakii from midgut of Stomoxys calcitrans. Emerg. Infect. Dis. 9:1355–1356. Harris, L. S., and P. J. Oriel. 1989. Heteropolysaccharide produced by Enterobacter sakazakii. U.S. patent 4,806,636. Hartmann, I., P. Carranza, A. Lehner, R. Stephan, L. Eberl, and K. Riedel. 2010. Genes involved in Crono­ bacter sakazakii biofilm formation. Appl. Environ. Microbiol. 76:2251–2261. Hawkins, R. E., C. R. Lissner, and J. P. Sanford. 1991. Enterobacter sakazakii bacteremia in an adult. South. Med. J. 84:793–795. Health Canada. 2010. Preparing and handling ­powdered infant formula. http://www.hc-sc.gc.ca/fn-an/securit/ kitchen-cuisine/pif-ppn-eng.php. Accessed October 23, 2012. Healy, B., S. Cooney, S. O’Brien, C. Iversen, P. Whyte, J. Nally, J. J. Callanan, and S. Fanning. 2009. Cronobacter (Enterobacter sakazakii): an opportunistic foodborne pathogen. Foodborne Pathog. Dis. 4:339–350. Hengge-Aronis, R., W. Klein, R. Lange, M. Rimmele, and W. Boos. 1991. Trehalose synthesis genes are controlled by the putative sigma factor encoded by rpoS and are involved in stationary-phase thermotolerance in Escherichia coli. J. Bacteriol. 173:7918–7924. Hunter, C. J., V. K. Singamsetty, N. K. Chokshi, P. Boyle, V. Camerini, A. V. Grishin, J. S. Upperman, H. R. Ford, and N. V. Prasadarao. 2008. Enterobacter sakazakii enhances epithelial cell injury by inducing apoptosis in a rat model of necrotizing enterocolitis. J. Infect. Dis. 198:586–593. Hurrell, E., E. Kucerova, M. Loughlin, J. CaubillaBarron, and S. J. Forsythe. 2009. Biofilm formation on enteral feeding tubes by Cronobacter sakazakii, Salmonella serovars and other Enterobacteriaceae. Int. J. Food Microbiol. 136:227–231. International Commission on Microbiological Specifi­ cation for Foods/Micro-Organisms in Foods. 2002. Microbiological Testing in Food Safety Management, vol. 7, p. 128–130. Kluwer Academic/Plenum Publishers, New York, NY. International Organization for Standardization. 2006. ISO/TS 22964 Milk and Milk Products—Detection of Enterobacter sakazakii (TC 34/SC; ISO Standards). International Organization for Standardization, Geneva, Switzerland. Iversen, C., and S. Forsythe. 2003. Risk profile of Enterobacter sakazakii, an emergent pathogen associated with infant milk formula. Trends Food Sci. Technol. 14:443–454. Iversen, C., N. Lazar Adler, and S. J. Forsythe. 2004. Virulence factors of Enterobacter sakazakii, abstr. P-108. Abstr. 104th Gen. Mtg. Am. Soc. Microbiol. American Society for Microbiology, Washington, DC. Iversen, C., and S. Forysthe. 2004. Isolation of Enterobacter sakazakii, and other Enterobacteriaceae

Foodborne Pathogenic Bacteria

334

58.

59.

60.

61.

62.

63. 64.

65. 66.

67.

68.

69.

from powdered infant formula milk and related products. Food Microbiol. 21:771–776. Iversen, C., P. Druggan, and S. Forsythe. 2004. A selective differential medium for Enterobacter sakazakii, a preliminary study. Int. J. Food Microbiol. 96: 133–139. Iversen, C., M. Lane, and S. J. Forsythe. 2004. The growth profile, thermotolerance and biofilm formation of Enterobacter sakazakii grown in infant formula milk. Lett. Appl. Microbiol. 38:378–382. Iversen, C., M. Waddington, S. L. On, and S. Forsythe. 2004. Identification and phylogeny of Enterobacter sakazakii relative to Enterobacter and Citrobacter species. J. Clin. Microbiol. 42:5368–5370. Iversen, C., N. Mullane, B. McCardell, B. D. Tall, A. Lehner, S. Fanning, R. Stephan, and H. Joosten. 2008. Cronobacter gen. nov., a new genus to accommodate the biogroups of Enterobacter sakazakii, and proposal of Cronobacter sakazakii gen. nov., comb. nov., Cronobacter malonaticus sp. nov., Cronobacter turicensis sp. nov., Cronobacter muytjensii sp. nov., Cronobacter dublinensis sp. nov., Cronobacter genomospecies 1, and of three subspecies, Cronobacter dublinensis subsp. dublinensis subsp. nov., Cronobacter d­ublinensis subsp. lausannensis subsp. nov. and Cronobacter dublinensis subsp. lactaridi subsp. nov. Int. J. Syst. Evol. Microbiol. 58:1442–1447. Jarvis, K. G., C. J. Grim, A. A. Franco, G. Gopinath, V. Sathyamoorthy, L. Hu, J. A. Sadowski, C. S. Lee, and B. D. Tall. 2011. Molecular characterization of Cronobacter lipopolysaccharide O-antigen gene clusters and development of serotype-specific PCR assays. Appl. Environ. Microbiol. 77:4017–4026. Jimenez, E. B., and C. Gimenez. 1982. Septic shock due to Enterobacter sakazakii. Clin. Microbiol. Newsl. 4:30. Jo, S. H., S. B. Baek, J. H. Ha, and S. D. Ha. 2010. Maturation and survival of Cronobacter biofilms on silicone, polycarbonate, and stainless steel after UV light and ethanol immersion treatments. J. Food Prot. 73:952–956. Joker, R. N., T. Norholm, and K. E. Siboni. 1965. A case of neonatal meningitis caused by a yellow Enterobacter. Danish Med. Bull. 12:128–130. Joseph, S., E. Cetinkaya, H. Drahovska, A. Levican, M. J. Figueras, and S. J. Forsythe. 2011. Cronobacter condimenti sp. nov., isolated from spiced meat and Cronobacter universalis sp. nov., a novel species designation for Cronobacter sp. genomospecies 1, recovered from a leg infection, water, and food ingredients. Int. J. Syst. Evol. Microbiol. 62:1277–1283. Kandhai, M. C., M. W. Reij, L. G. Gorris, O. GuillaumeGentil, and M. van Schothorst. 2004. Occurrence of Enterobacter sakazakii in food production environments and households. Lancet 363:39–40. Kandhai, M. C., M. W. Reij, K. van Puyvelde, O. Guillaume-Gentil, R. R. Beumer, and M. van Schothorst. 2004. A new protocol for the detection of Enterobacter sakazakii applied to environmental samples. J. Food Prot. 67:1267–1270. Keller, R., M. A. Pedroso, R. Ritchman, and R. M. Silva. 1998. Occurrence of virulence associated properties in Enterobacter cloacae. Infect. Immun. 66:645–649.

70. Kempf, B., and E. Bremer. 1998. Uptake and synthesis of compatible solutes as microbial stress responses to highosmolality environments. Arch. Microbiol. 170:319–330. 71. Keyser, M., R. C. Witthuhn, L. C. Ronquest, and T. J. Britz. 2003. Treatment of winery effluent with upflow anaerobic sludge blanket (UASB) granular sludges enriched with Enterobacter sakazakii. Biotechnol. Lett. 25:1893–1898. 72. Kim, H., J. Bang, L. R. Beuchat, and J. H. Ryu. 2008. Fate of Enterobacter sakazakii attached to or in biofilms on stainless steel upon exposure to various temperatures or relative humidities. J. Food Prot. 71:940–945. 73. Kim, H., J. H. Ryu, and L. R. Beuchat. 2006. Attachment of and biofilm formation by Enterobacter sakazakii on stainless steel and enteral feeding tubes. Appl. Environ. Microbiol. 72:5846–5856. 74. Kim, K., K. P. Kim, J. Choi, J. A. Lim, J. Lee, S. Hwang, and S. Ryu. 2010. Outer membrane proteins A (OmpA) and X (OmpX) are essential for basolateral invasion of Cronobacter sakazakii. Appl. Environ. Microbiol. 76:5188–5198. 75. Kim, K. P., J. Klumpp, and M. J. Loessner. 2007. Enterobacter sakazakii bacteriophages can prevent bacterial growth in reconstituted infant formula. Int. J. Food Microbiol. 115:195–203. 76. Kindle, G., A. Busse, D. Kampa, U. Meyer-Koenig, and F. D. Daschner. 1996. Killing activity of microwaves in milk. J. Hosp. Infect. 33:273–278. 77. Kleiman, M. B., S. D. Allen, P. Neal, and J. Reynolds. 1981. Meningoencephalitis and compartmentalization of the cerebral ventricles caused by Enterobacter sakazakii. J. Clin. Microbiol. 14:352–354. 78. Kline, M. W. 1988. Pathogenesis of brain abscess caused by Citrobacter diversus or Enterobacter sakazakii. Pediatr. Infect. Dis. J. 7:891–892. 79. Kothary, M. H., B. A. McCardell, C. D. Frazar, D. Deer, and B. D. Tall. 2007. Characterization of the zinc-containing metalloprotease encoded by zpx and development of a species-specific detection method for Enterobacter sakazakii. Appl. Environ. Microbiol. 73:4142–4151. 80. Kuzina, L. V., J. J. Peloquin, D. C. Vacek, and T. A. Miller. 2001. Isolation and identification of bacteria associated with adult laboratory Mexican fruit flies, Anastrepha ludens (Diptera: Tephritidae). Curr. Microbiol. 42:290–294. 81. Lai, K. K. 2001. Enterobacter sakazakii infections among neonates, infants, children, and adults. Case reports and a review of the literature. Medicine (Baltimore) 80:113–122. 82. Lange, R., and R. Hengge-Aronis. 1991. Growth phaseregulated expression of bolA and morphology of stationary-phase Escherichia coli cells are controlled by the novel sigma factor sS. J. Bacteriol. 173:4474–4481. 83. Lange, R., and R. Hengge-Aronis. 1991. Identification of a central regulator of stationary-phase gene expression in Escherichia coli. Mol. Microbiol. 5:49–59. 84. Leclercq, A., C. Wanegue, and P. Baylac. 2002. Comparison of fecal coliform agar and violet red bile lactose agar for fecal coliform enumeration in foods. Appl. Environ. Microbiol. 68:1631–1638. 85. Lecour, H., A. Seara, J. Cordeiro, and M. Miranda. 1989. Treatment of childhood bacterial meningitis. Infection 17:343–346.

13.  Cronobacter Species 86. Lee, H.-A, S. Hong, H. Park, H. Kim, and O. Kim. 2011. Cronobacter sakazakii infection induced fatal clinical sequels including meningitis in neonatal ICR mice. Lab Animal Res. 27:59–62. 87. Lehner, A., K. Riedel, L. Eberl, P. Breeuwer, B. Diep, and R. Stephan. 2005. Biofilm formation, extracellular polysaccharide production, and cell-to-cell signaling in various Enterobacter sakazakii strains: aspects promoting environmental persistence. J. Food Prot. 68:2287–2294. 88. Lehner, A., and R. Stephan. 2004. Microbiological, epidemiological, and food safety aspects of Enterobacter sakazakii. J. Food Prot. 67:2850–2857. 89. Lehner, A., T. Tasara, and R. Stephan. 2004. 16S rRNA gene based analysis of Enterobacter sakazakii strains from different sources and development of a PCR assay for identification. BMC Microbiol. 4:43. 90. Lenati, R. F., D. L. O’Connor, K. C. Hebert, J. M. Farber, and F. J. Pagotto. 2008. Growth and survival of Enterobacter sakazakii in human breast milk with and without fortifiers as compared to powdered infant formula. Int. J. Food Microbiol. 122:171–179. 91. Reference deleted. 92. Reference deleted. 93. Leuschner, R. G., and J. Bew. 2004. A medium for the presumptive detection of Enterobacter sakazakii in infant formula: interlaboratory study. J. AOAC Int. 87:604–613. 94. Liu, Y., Q. Gao, X. Zhang, Y. Hou, J. Yang, and X. Huang. 2006. PCR and oligonucleotide array for detection of Enterobacter sakazakii in infant formula. Mol. Cell. Probes 1:11–17. 95. MacLean, L. L., F. Pagotto, J. M. Farber, and M. B. Perry. 2009. Structure of the antigenic repeating pentasaccharide unit of the LPS O-polysaccharide of Cronobacter sakazakii implicated in the Tennessee outbreak. Biochem. Cell. Biol. 87:459–465. 96. MacLean, L. L., F. Pagotto, J. M. Farber, and M. B. Perry. 2009. The structure of the O-antigen in the endotoxin of the emerging food pathogen Cronobacter (Enterobacter) muytjensii strain 3270. Carbohydr Res. 344:667–671. 97. MacLean, L. L., E. Vinogradov, F. Pagotto, J. M. Farber, and M. B. Perry. 2009. Characterization of the O-antigen in the lipopolysaccharide of Cronobacter (Enterobacter) malonaticus 3267. Biochem. Cell. Biol. 87:927–932. 98. MacLean, L. L., E. Vinogradov, F. Pagotto, J. M. Farber, and M. B. Perry. 2010. The structure of the O-antigen of Cronobacter sakazakii HPB2855 isolate involved in a neonatal infection. Carbohydr Res. 345:1932–1937. 99. MacLean, L. L., E. Vinogradov, F. Pagotto, and M. B. Perry. Characterization of the lipopolysaccharide O-antige­n of Cronobacter turicensis HPB3287 as a polysaccharide containing a 5,7-diacetamido-3,5,7,9-­tetradeoxyD-glycero-D-galacto-non-2-ulosonic acid (legionaminic acid) residue. Carbohydr Res. 346:2589–2594. 100. Malorny, B., and M. Wagner. 2005. Detection of Enterobacter sakazakii strains by real-time PCR. J. Food Prot. 68:1623–1627. 101. Mange, J. P., R. Stephan, N. Borel, P. Wild, K. S. Kim, A. Pospischil, and A. Lehner. 2006. Adhesive properties of Enterobacter sakazakii to human epithelial and brain microvascular endothelial cells. BMC Microbiol. 6:58.

335 102. Mittal, R., S. Bulgheresi, C. Emami, and N. V. Prasadarao. 2009. Enterobacter sakazakii targets DC-SIGN to induce immunosuppressive responses in dendritic cells by modulating MAPKs. J. Immunol. 183:6588–6599. 103. Mittal, R., Y. Wang, C. J. Hunter, I. Gonzalez-Gomez, and N. V. Prasadarao. 2009. Brain damage in newborn rat model of meningitis by Enterobacter sakazakii: a role for outer membrane protein A. Lab. Investig. 89:263–277. 104. Mohan Nair, M. K., and K. Venkitanarayanan. 2007. Role of bacterial OmpA and host cytoskeleton in the invasion of human intestinal epithelial cells by Enterobacter sakazakii. Pediatr. Res. 62:664–669. 105. Monroe, P. W., and W. L. Tift. 1979. Bacteremia associated with Enterobacter sakazakii (yellow, pigmented Enterobacter cloacae). J. Clin. Microbiol. 10: 850–851. 106. Mosso, M. A., M. C. de la Rosa, C. Vivar, and M. R. Medina. 1994. Heterotrophic bacterial populations in the mineral waters of thermal springs in Spain. J. Appl. Bacteriol. 77:370–381. 107. Muytjens, H. L., and L. A. Kollee. 1990. Enterobacter sakazakii meningitis in neonates: causative role of formula? Pediatr. Infect. Dis. J. 9:372–373. 108. Muytjens, H. L., and J. van der Ros-van de Repe. 1986. Comparative in vitro susceptibilities of eight Enterobacter species, with special reference to Enterobacter sakazakii. Antimicrob. Agents Chemother. 29:367–370. 109. Muytjens, H. L., and L. A. Kollee. 1982. Neonatal meningitis due to Enterobacter sakazakii. Tijdschr. Kindergeneeskd. 50:110–112. 110. Muytjens, H. L., H. Roelofs-Willemse, and G. H. Jaspar. 1988. Quality of powdered substitutes for breast milk with regard to members of the family Enterobacteriaceae. J. Clin. Microbiol. 26:743–746. 111. Muytjens, H. L., H. C. Zanen, H. J. Sonderkamp, L. A. Kollee, I. K. Wachsmuth, and J. J. Farmer III. 1983. Analysis of eight cases of neonatal meningitis and sepsis due to Enterobacter sakazakii. J. Clin. Microbiol. 18:115–120. 112. Nair, M. K., J. Joy, and K. S. Venkitanarayanan. 2004. Inactivation of Enterobacter sakazakii in reconstituted infant formula by monocaprylin. J. Food Prot. 67:2815–2819. 113. Nair, M. K., K. Venkitanarayanan, L. K. Silbart, and K. S. Kim. 2009. Outer membrane protein A (OmpA) of Cronobacter sakazakii binds fibronectin and contributes to invasion of human brain microvascular endothelial cells. Foodborne Pathog. Dis. 6:495–501. 114. Naqvi, S. H., M. A. Maxwell, and L. M. Dunkle. 1985. Cefotaxime therapy of neonatal gram-negative bacillary meningitis. Pediatr. Infect. Dis. 4:499–502. 115. Nassereddin, R. A., and M. I. Yamani. 2005. Micro­ biological quality of sous and tamarind, traditional drinks consumed in Jordan. J. Food Prot. 68:773–777. 116. Nazarowec-White, M., and J. M. Farber. 1999. Phenotypic and genotypic typing of food and clinical isolates of Enterobacter sakazakii. J. Med. Microbiol. 48:559–567. 117. Nazarowec-White, M., and J. M. Farber. 1997. Enterobacter sakazakii: a review. Int. J. Food Microbiol. 34:103–113.

336 118. Nazarowec-White, M., and J. M. Farber. 1997. Thermal resistance of Enterobacter sakazakii in reconstituted dried-infant formula. Lett. Appl. Microbiol. 24:9–13. 119. New Zealand Ministry of Health. 5 April 2005. News and issues: Cronobacter meningitis to become a notifiable disease. www.moh.govt.nz/moh.nsf/bfc540e5ac1abe02 cc256e7d0082eede/d6081bec1d116521cc256fd90074d 556?OpenDocument#1. Accessed 1 February 1 2006. 120. Reference deleted. 121. Noriega, F. R., K. L. Kotloff, M. A. Martin, and R. S. Schwalbe. 1990. Nosocomial bacteremia caused by Enterobacter sakazakii and Leuconostoc mesenteroides resulting from extrinsic contamination of infant formula. Pediatr. Infect. Dis. 9:447–449. 122. Oh, S. W., and D. H. Kang. 2004. Fluorogenic selective and differential medium for isolation of Enterobacter sakazakii. Appl. Environ. Microbiol. 70:5692–5694. 123. Pagotto, F. J., M. Nazarowec-White, S. Bidawid, and J. M. Farber. 2003. Enterobacter sakazakii: infectivity and enterotoxin production in vitro and in vivo. J. Food Prot. 66:370–375. 124. Pangalos, G. 1929. Sur un bacille chromogene isole par hemoculture. C. R. Soc. Biol. 100:1097. 125. Pediatric Nutrition Practice Group of the American Dietetic Association. 2004. Infant Feedings: Guidelines for Preparation of Formula and Breastmilk in Health Care Facilities. http://www.eatright.org/cps/rde/xchg/ ada/hs.xsl/nutrition_5441_ENU_HTML.htm Accessed 14 February 2006. 126. Pitout, J. D., E. S. Moland, C. C. Sanders, K. S. Thomson, and S. R. Fitzsimmons. 1997. b-Lactamases and detection of b-lactam resistance in Enterobacter spp. Antimicrob. Agents Chemother. 41:35–39. 127. Postupa, R., and E. Aldova. 1984. Enterobacter sakazakii: a Tween 80 esterase-positive representative of the genus Enterobacter isolated from powdered milk specimens. J. Hyg. Epidemiol. Microbiol. Immunol. 28:435–440. 128. Pribyl, C., R. Salzer, J. Beskin, R. J. Haddad, B. Pollock, R. Beville, B. Holmes, and W. J. Mogabgab. 1985. Aztreonam in the treatment of serious orthopaedic i­nfections. Am. J. Med. 78:51–56. 129. Raghav, M., and P. K. Aggarwal. 2007. Purification and characterization of Enterobacter sakazakii enterotoxin. Can. J. Microbiol. 53:750–755. 130. Reina, J., F. Parras, J. Gil, F. Salva, and P. Alomar. 1989. Human infections caused by Enterobacter sakazakii. Microbiologic considerations. Enferm. Infecc. Microbiol. Clin. 7:147–150. 131. Restaino, L., E. W. Frampton, W. C. Lionberg, and R. J. Becker. 2006. A chromogenic plating medium for the isolation and identification of Enterobacter sakazakii from foods, food ingredients, and environmental sources. J. Food Prot. 69:315–322. 132. Richards, G. M., J. B. Gurtler, and L. R. Beuchat. 2005. Survival and growth of Enterobacter sakazakii in infant rice cereal reconstituted with water, milk, liquid infant formula, or apple juice. J. Appl. Microbiol. 99:844–850. 133. Richardson, A. N., S. Lambert, and M. A. Smith. 2009. Neonatal mice as models for Cronobacter sakazakii infection in infants. J. Food Prot. 72:2363–2367.

Foodborne Pathogenic Bacteria 134. Richardson, A. N., L. B. Beuchat, S. Lambert, D. Williams, and M. A. Smith. 2010. Comparison of virulence of three strains of Cronobacter sakazakii in neonatal CD-1mice. J. Food Prot. 73:849–854. 135. Ries, M., D. Harms, and J. Scharf. 1994. Multiple cerebral infarcts with resulting multicystic encephalomalacia in a premature infant with Enterobacter sakazakii meningitis. Klin. Padiatr. 206:184–186. 136. Robertson, L. F., G. S. Johannessen, B. K. Gjerde, and S. Loncarevic. 2002. Microbiological analysis of seed sprouts in Norway. Int. J. Food Microbiol. 75:119–126. 137. Scheepe-Leberkuhne, M., and F. Wagner. 1986. Opti­ mization and preliminary characterization of an exopolysaccharide synthesized by Enterobacter sakazakii. Biotechnol. Lett. 8:695–700. 137a. See, K. C., H. A. Than, and T. Tang. 2007. Enterobacter sakazakii bacteremia with multiple splenic abscesses in a 75-year-old woman: a case report. Age Ageing 36: 595–596. 138. Seo, K. H., and R. E. Brackett. 2005. Rapid, specific detection of Enterobacter sakazakii in infant formula using a real-time PCR assay. J. Food Prot. 68:59–63. 139. Simmons, B. P., M. S. Gelfand, M. Haas, L. Metts, and J. Ferguson. 1989. Enterobacter sakazakii infections in neonates associated with intrinsic contamination of a powdered infant formula. Infect. Control Hosp. Epidemiol. 10:398–401. 140. Singamsetty, V. K., Y. Wang, H. Shimada, and N. V. Prasadarao. 2008. Outer membrane protein A expression in Enterobacter sakazakii is required to induce microtubule condensation in human brain microvascular endothelial cells for invasion. Microb. Pathog. 45:181–191. 141. Smith, J. L., P. M. Fratamico, and S. J. Novak. 2004. Quorum sensing: a primer for microbiologists. J. Food Prot. 67:1063–1070. 142. Smith, J. L., P. M. Fratamico, and X. Yan. 2011. Eavesdropping by bacteria: the role of SdiA in Escherichia coli and Salmonella enterica serovar Typhymurium quorum sensing. Foodborne Pathog. 8:169–178. 143. Soares, J. A., and B. M. Ahmer. 2011. Detection of acylhomoserine lactones by Escherichia and Salmonella. Curr. Opin. Microbiol. 14:188–193. 144. Sogaard, P., and P. Kjaeldgaard. 1986. Two isolations of enteric group 69 from human clinical specimens. Acta Pathol. Microbiol. Immunol. Scand. B 94:365–367. 145. Sondheimer, J., D. Clark, and E. Gervaise. 1985. Contin­ uous gastric pH measurement in young and older healthy preterm infants receiving formula and clear liquid feedings. J. Ped. Gastr. Nutr. 4:352–355. 146. Soriano, J. M., H. Rico, J. C. Molto, and J. Manes. 2001. Incidence of microbial flora in lettuce meat and Spanish potato omelette from restaurants. Food Microbiol. 18:159–163. 147. Stoll, B. J., N. Hansen, A. A. Fanaroff, and J. A. Lemons. 2004. Enterobacter sakazakii is a rare cause of neonatal septicemia or meningitis in VLBW infants. J. Pediatr. 144:821–823. 148. Styrvold, O. B., and A. R. Strom. 1991. Synthesis, accumulation, and excretion of trehalose in osmotically stressed Escherichia coli K-12 strains: influence of amber suppressors and function of the periplasmic trehalase. J. Bacteriol. 173:1187–1192.

13.  Cronobacter Species 149. Sun, Y., M. Wang, H. Liu, J. Wang, X. He, J. Zeng, X. Guo, K. Li, B. Cao, and L. Wang. 2011. Development of an O-antigen serotyping scheme for Cronobacter sakazakii. Appl. Environ. Microbiol. 77:2209–2214. 150. Telang, S., C. L. Berseth, P. W. Ferguson, J. M. Kinder, M. Deroin, and B. W. Petschow. 2005. Fortifying fresh human milk with commercial powdered human milk fortifiers does not affect bacterial growth during 6 hours at room temperature. J. Am. Diet. Assoc. 105:1567–1572. 151. Tift, W. L. 1977. Group B streptococcal infections in the neonate. J. Med. Assoc. Ga. 66:703–705. 151a. Townsend, S., J. Caubilla-Barron, C. Loc-Carrillo, and S. Forsythe. 2007. The presence of endotoxin in powdered infant formula milk and the influence of endotoxin and Enterobacter sakazakii on bacterial translocation in the infant rat. Food Microbiol. 24:67–74. 152. Townsend, S., E. Hurrell, and S. Forsythe. 2008. Virulence studies of Enterobacter sakazakii isolates associated with a neonatal intensive care unit outbreak. BMC Microbiol. 8:64. 153. Townsend, S. M., E. Hurrell, I. Gonzalez-Gomez, J. Lowe, J. G. Frye, S. Forsythe, and J. L. Badger. 2007. Enterobacter sakazakii invades brain capillary endothelial cells, persists in human macrophages influencing cytokine secretion and induces severe brain pathology in the neonatal rat. Microbiology 153:3538–3547. 154. Urmenyi, A. M. C., and A. W. Franklin. 1961. Neonatal death from pigmented coliform infection. Lancet i:313-315. 155. U.S. Food and Drug Administration. 2002. Isolation and enumeration of Enterobacter sakazakii from dehydrated powdered infant formula. http://www.fda.gov/ food/scienceresearch/laboratorymethods/ucm114665. htm. Accessed 23 October 2012. 156. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition Office of Nutritional Products, Labeling and Dietary Supplements. 2002. April 11, 2002; Revised October 10, 2002. Health Professionals Letter on Enterobacter sakazakii Infections

337 Associated with Use of Powdered (Dry) Infant Formulas in Neonatal Intensive Care Units. U.S. Food and Drug Administration, Rockville, MD. 157. van Acker, J., F. de Smet, G. Muyldermans, A. Bougatef, A. Naessens, and S. Lauwers. 2001. Outbreak of necrotizing enterocolitis associated with Enterobacter sakazakii in powdered milk formula. J. Clin. Microbiol. 39:293–297. 158. Van Os, M., P. G. Van Wikeselaar, and S. F. Spoelstra. 1996. Formation of biogenic amines in well fermented grass silages. J. Agric. Sci. Cambridge 127:97–107. 159. Weir, E. 2002. Powdered infant formula and fatal infection with Enterobacter sakazakii. CMAJ 166:1570. 160. Whitehead, N. A., A. M. L. Barnard, H. Slater, N. J. L. Simpson, and G. P. C. Salmond. 2001. Quorum-sensin­g in gram-negative bacteria. FEMS Microbiol. Rev. 25: 365–404. 161. Williams, T. L., S. R. Monday, S. Edelson-Mammel, R. Buchanan, and S. M. Musser. 2005. A top-down prote­omics approach for differentiating thermal resistant strains of Enterobacter sakazakii. Proteomics 5: 4161–4169. 162. Willis, J., and J. E. Robinson. 1988. Enterobacter sakazakii meningitis in neonates. Pediatr. Infect. Dis. J. 7: 196–199. 163. World Health Organization. 2008. Enterobacter sakazakii and Salmonella in Powdered Infant Formula: Meet­ ing Report, Microbial Risk Assessment Series 10. Food Safety. http://www.who.int/foodsafety/publications/micro/ mra10/en/index.html. Accessed 17 February 2012. 164. Yemis¸, G. P., F. Pagotto, S. Bach, and P. Delaquis. 2011. Effect of vanillin, ethyl vanillin, and vanillic acid on the growth and heat resistance of Cronobacter species. J. Food Prot. 74:2062–2069. 165. Zuber, S., C. Boissin-Delaporte, L. Michot, C. Iversen, B. Diep, H. Brüssow, and P. Breeuwer. 2008. Decreasing Enterobacter sakazakii (Cronobacter spp.) food contamination level with bacteriophages: prospects and problems. Microbiol. Biotechnol. 1:532–543.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch14

14

Roy M. Robins-Browne

Yersinia enterocolitica

CHARACTERISTICS OF THE ORGANISM In 1883, Malassez and Vignal reported a disease in guinea pigs they called “tuberculose zoogloeique” (178). Their description matched that of an epizootic tuberculosislike disease which Eberth subsequently termed “pseudo tuberculosis” (83). From 1889, the suspected causative organism of this condition was assigned a series of name changes that reflected its uncertain taxonomic status. These included “Bacillus pseudotuberculosis rodentium” (227) and “Bacillus parapestis” (304), the latter reflecting the biological similarity of this organism to the plague bacillus. In 1944, van Loghen indicated that “Pasteurella” pseudotuberculosis, as it was then known, and the closely related “Pasteurella” pestis were sufficiently distinct from the pasteurellae of hemorrhagic septicemia to warrant their own generic status (311). He proposed the name Yersinia after Alexandre Yersin, who first described the plague bacillus and had named it in honor of Louis Pasteur. Whereas Yersinia pseudotuberculosis first attracted attention as a pathogen of animals, Yersinia enteroco­ litica was initially associated with human infection. In 1934, McIver and Pike identified a novel bacterium in pus from the facial skin of a New York farm worker (185). They named this bacterium “Flavobacterium pseudo­

mallei Whitmore,” but in retrospect it is likely that this was Y. enterocolitica. Five years later, Schleifstein and Coleman recorded the similarity of McIver and Pike’s isolate and four others they had obtained to Pasteurella (Yersinia) pseudotuberculosis (257). Later, they proposed the name “Bacterium enterocoliticum” for this organism (258). Today, the genus Yersinia comprises 13 species within the family Enterobacteriaceae (Table 14.1) (20, 327) and includes three well-characterized pathogens of mammals and one of fish and several other species whose etiologic role in disease is uncertain (for a review of the last, see reference 284). Two species that have been added to the genus since the previous edition of this book are Y. nurmii and Y. pekkanenii. Y. nurmii was discovered in packaged broiler meat and most closely resembles Y. ruckeri (199). Y. pekkanenii is based on three isolates from water, soil, and lettuce and can be mistaken for Y. pseudotuberculosis, although 16S rRNA sequence analysis reveals that it is more closely related to Y. aldo­ vae and Y. mollaretii (200). Neither of these species has been associated with disease. The four known pathogenic Yersinia species are Y. pestis, the causative agent of bubonic and pneumonic

Roy M. Robins-Browne, Department of Microbiology and Immunology, The University of Melbourne, Victoria 3010, and Murdoch Childrens Research Institute, Royal Children’s Hospital, Parkville, Victoria 3052, Australia.

339

SMP_Food Microbiology_CH14.indd

339

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52� 340

SMP_Food Microbiology_CH14.indd

Table 14.1  Some biochemical tests used to differentiate Yersinia speciesa Biochemical test result

Test

340

a b

Y. bercovieri 1 through 4

5

Y. Y. Y. Y. Y. Y. frederiksenii intermedia kristensenii mollaretii nurmii pekkanenii

Y. pestis

Y. pseudotu­ berculosis

Y. Y. rohdei ruckeri

− + D

− − −

D + −

− +b −

+ D D

+ + +

D − −

− − −

− + ND

− − ND

− − −

− − −

− − +

− − −

+ D + − − + − − D

+ + + + + − − − +

+ − D + + − − D D

− − − d + − − D −

+ D + + + + − + +

+ D + + + + + + D

+ − +b − + − − + D

+ + + + + − − + −

+ ND ND + +b − − − ND

− ND ND − + − − − ND

− − − − − − D − ND

− − −b − − + + − −

+ − + + + − D ND ND

+ − ND − − − − ND ND

Data are from references 199, 200, and 328. +, positive; −, negative; D, different reactions; ND, not determined. Some reactions may be delayed or weakly positive.

11/08/2012 06:58AM

Foodborne Pathogenic Bacteria

Manila Typesetting Company

Indole Voges-Proskauer Citrate (Simmons) l-Ornithine Mucate, acid Pyrazinamidase Sucrose Cellobiose l-Rhamnose Melibiose l-Sorbose l-Fucose

Y. aldovae

Y. enterocolitica biotype(s):

14.  Yersinia enterocolitica

341

plague; Y. pseudotuberculosis, a rodent pathogen that occasionally causes mesenteric lymphadenitis, septicemia, and immunomediated diseases in humans; Y. ruck­ eri, the cause of enteric redmouth disease in salmonids and other freshwater fish; and Y. enterocolitica, a versatile intestinal pathogen. When Y. enterocolitica first emerged as a human pathogen, it was considered an oddity, but now it is by far the most prevalent Yersinia species among humans. However, Y. pseudotuberculo­ sis, which was once quite common in Europe, is now comparatively rare (12). Y. pestis is transmitted to its host via flea bites or respiratory aerosols, whereas Y. pseudotuberculosis and Y. enterocolitica are foodborne pathogens. These three species share a number of essential virulence determinants that enable them to overcome the innate defenses of their hosts. Analogs of these virulence determinants occur in several other enterobacteria, such as enteropathogenic and enterohemorrhagic Escherichia coli and Salmonella and Shigella species, as well as in various other pathogens of animals (e.g., Pseudomonas aeruginosa and Bordetella species) and plants (e.g., Erwinia amylovora, Xanthomonas campestris, and Pseudomonas syringae), hence providing evidence for horizontal transfer of virulence genes between diverse bacterial pathogens. Given that Y. pestis is incapable of infecting the intestinal tract directly and is not pathogenic when ingested and that the role of most other Yersinia species in disease is uncertain, this chapter will focus on Y. enteroco­ litica and Y. pseudotuberculosis.

non-spore-forming, facultative anaerobes that ferment glucose. They are smaller than most other enterobacteria and when grown at 37°C often appear as coccobacilli in stained smears. Y. enterocolitica shares between 10 and 30% DNA homology with other genera in the Enterobacteriaceae and is approximately 50% related to Y. pseudotuberculosis and Y. pestis. The last two species share greater than 90% DNA homology, with genetic analysis revealing that Y. pestis is a clone of Y. pseudotuberculosis that evolved some 1,500 to 20,000 years ago, shortly before the first known pandemics of human plague (2). Y. pseudotuberculosis is a relatively homogenous species, which is subdivided into serotypes according to its lipopolysaccharide (LPS) O antigens. Currently, 11 major serotypes, designated 1 to 11 (or I to XI), have been reported (9, 305). Serotypes 1 and 2 are further subdivided into 3 subgroups, A, B, and C; serotypes 4 and 5 are each divided into subgroups A and B (9, 305). Y. enterocolitica is far more heterogenous than Y. pseudotuberculosis, being divisible into a large number of subgroups according to biochemical activity and LPS O antigens (Tables 14.2 and 14.3). Biotyping is based on the ability of Y. enterocolitica to metabolize selected organic substrates and provides a convenient means to subdivide the species into subtypes with various levels of clinical and epidemiologic significance (Tables 14.2 and 14.3) (329). Most pathogenic strains of humans and domestic animals occur within biotypes 1B, 2, 3, 4, and 5. The most frequent biotype associated with human disease worldwide is biotype 4. By contrast, Y. enterocolitica strains of biotype 1A are commonly obtained from terrestrial and freshwater ecosystems. For this reason, they are often referred to as environmental

Classification

As members of the family Enterobacteriaceae, yersiniae are gram-negative, oxidase-negative, rod-shaped,

Table 14.2  Biotyping scheme of Y. enterocoliticaa Reaction of biotype: Biochemical test

1A

1B

2

3

4

5

Lipase (Tween hydrolysis) Esculin hydrolysis Indole production d-Xylose fermentation Voges-Proskauer reaction Trehalose fermentation Nitrate reduction Pyrazinamidase

+ D + + + + + + +

+ − + + + + + − −

− − (+) + + + + − −

− − − + + + + − −

− − − − + + + − −

− − − D (+) − − − −

D

−

−

−

−

−

b-d-Glucosidase Proline peptidase a

+, positive; (+), delayed positive; −, negative; d, different reactions. Adapted from reference 329.

SMP_Food Microbiology_CH14.indd

341

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

342 Table 14.3  Relationship between O serotype and

pathogenicity of Y. enterocolitica and related species Yersinia species and biotype Y. enterocolitica 1A

1B 2 3 4 5 Y. bercovieri Y. frederiksenii Y. intermedia Y. kristensenii

Y. mollaretii

Serotype(s)a O:4; O:5; O:6,30; O6,31; O:7,8; O:7,13; O:10; O:14; O:16; O:21; O:22; O:25; O:37; O:41,42; O:46; O:47; O:57; NT O:4,32; O:8; O:13a,13b; O:16; O:18; O:20; O:21; O:25; O:41,42; NT O:5,27; O:9; O:27 O:1,2,3; O:3; O:5,27 O:3 O:2,3 O:8; O:10; O;58,16; NT O:3; O:16; O:35; O:38; O:44: NT O:17; O:21,46; O:35; O:37; O:40; O:48; O:52; O:55; NT O:11; O:12,25; O:12,26; O:16; O:16,29; O:28,50; O:46; O:52; O:59; O:61; NT O:3, O:6,30; O:7,13; O:59: O:62,22; NT

a NT, not typeable. Serotypes that include strains considered to be primary pathogens are in bold.

strains, although some of them may be responsible for human infection (293). Biotype 1B bacteria are relatively more frequent in the United States and are sometimes referred to as “American” or “New World” strains, although they also occur in Europe, Africa, Asia, and Australasia. Although not common anywhere, biotype 1B yersiniae are inherently more virulent for mice (and probably for humans) than strains in the other pathogenic categories and have been identified as the cause of several foodborne outbreaks of yersiniosis in the United States. Sequence analysis of Y. enterocolitica 16S rRNA and whole genomes supports the division of Y. entero­ colitica into 3 broad categories that correspond with biotype and virulence: biotype 1B (the most virulent), biotypes 2 to 5 (intermediate virulence), and biotype 1A (the least virulent) (137, 205). The sequence data led to Y. enterocolitica being divided into two subspecies: Y. enterocolitica subsp. enterocolitica and Y. en­ terocolitica subsp. palearctica, comprising biotype 1B strains and biotype 2 to 5 strains, respectively (205). A third subspecies comprising biotype 1A has been proposed (137). Serotyping of Y. enterocolitica, based on LPS surface O antigens, also coincides to some extent with biotype

SMP_Food Microbiology_CH14.indd

342

(Table 14.3) and provides a useful additional tool to subdivide this species in a way that relates to pathologic significance (326). Serotype O:3 is the variety most frequently isolated from humans, and almost all of these isolates are biotype 4. Other serotypes commonly obtained from humans, particularly in northern Europe, include O:9 (biotype 2) and O:5,27 (biotype 2 or 3). The usefulness of serotyping is limited to some extent by the fact that the overwhelming majority of human infections are due to strains of serotype O:3 and by the presence of cross-reacting O antigens in Y. enterocolitica strains of various levels of pathological and epidemiologic significance. In addition, some bacteria that were originally allocated as O serotypes of Y. enterocolitica were later reclassified as separate species (326). At least 18 flagellar (H) antigens of Y. enterocolitica, designated by lowercase letters (a,b; b,c; b,c,e,f,k; m, etc.), have also been identified. Although there is little overlap between the H antigens of Y. enterocolitica sensu stricto and those of related species, antigenic characterization of isolates by complete O and H serotyping is seldom attempted (326). Other schemes for subtyping Yersinia species include bacteriophage typing, pulsed-field gel electrophoresis, multienzyme electrophoresis, multilocus sequence typing, and the demonstration of restriction fragment length polymorphisms of chromosomal and plasmid DNA (147, 163). These techniques can be used to determine the relatedness of different isolates and to facilitate epidemiologic investigations of outbreaks or to trace the source of sporadic infections (96, 321).

Cold Adaptation

The enteropathogenic Yersinia species are unusual among pathogenic enterobacteria of humans in being psychrophilic. A major reason for storing food at refrigeration temperature is to minimize bacterial replication; however, Y. enterocolitica and Y. pseudotuberculosis can grow at this temperature. The cold shock response in bacteria is a stress response that follows a reduction in ambient temperature by 10°C or more. Although the response to cold shock has been extensively studied in E. coli, E. coli is a mesophile and therefore does not necessarily respond to cold shock in the same way as the pathogenic yersiniae (218). For example, in E. coli exposure to low temperatures induces both the stationary-phase sigma factor, sS (RpoS), and the envelope stress factor, sE. In Y. enterocolitica, however, sS does not appear to be required at low temperatures (15), and the control of expression of sE in Y. enterocolitica differs markedly from that in E. coli (131).

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

343

Bacteria can sense alterations in temperature in different ways, e.g., by temperature-induced changes to membrane fluidity and by DNA supercoiling, but the way by which yersiniae sense cold is unknown. Like E. coli, Y. enterocolitica responds to cold shock by the transient expression of csp genes, which encode lowmolecular-weight, cold shock proteins (Csp), which prepare the bacteria for growth at low temperature. These proteins are highly conserved among bacteria, with more than 45% amino acid sequence similarity between gram-negative and gram-positive bacteria (107). The period of cold acclimatization that is characterized by the production of Csp is followed by growth at low temperature, during which a number of cold acclimatization proteins (Caps) are produced more or less continuously until the temperature rises. A number of Caps are isoenzymes that function relatively more efficiently at low temperatures. To determine the processes involved in cold adaptation by Y. enterocolitica, Bresolin et al. (36) constructed a mutant library of Y. enterocolitica strain W22703, using a luxCDABE transposon as a reporter. Investigation of 5,700 transposon mutants led to the identification of 42 transcriptional units with strongly enhanced or reduced promoter activity at 10°C compared to 30°C. Investigation of transcription over time revealed a specific order of activation of cold response genes. Soon after a reduction in temperature from 30 to 10°C, two almost identical coding regions, cspA1 and cspA2, are expressed (11). The presence of two copies of csp may enhance the ability of Y. enterocolitica to adapt to cold, compared with E. coli, which carries only one copy of this gene (206). Other genes that are expressed during the acclimatization phase include cspB, gltP (which encodes a glutamate-aspartate symporter), and ymoA, the gene for a histone-like protein that is involved in temperature-regulated virulence of yersiniae (56). Bresolin et al. (36) also determined that during early and mid-exponential growth at low temperature, genes for environmental sensors and regulators involved in signal transduction were activated. Other genes expressed preferentially at low temperature included those for flagellar synthesis and chemotaxis, as well the genes for the tc complex for insecticidal toxins (294). During the late exponential and stationary phases of growth by Y. en­ terocolitica at 10°C, a number of genes involved in biodegradation are induced. These include urease, alkaline serine protease, histidine-ammonia lyase, and amylase. Increased production of these enzymes may compensate for the overall diminution in bacterial metabolic activity at low temperatures by ensuring that sufficient nutrients are available for growth (218).

SMP_Food Microbiology_CH14.indd

343

Interestingly, during the course of their work, Bresolin et al. (36) did not identify any essential genes that Y. enterocolitica requires to grow at low temperature. However, they investigated only about 70% of all known genes or operons of Y. enterocolitica, suggesting that they may not have identified some essential genes. In this regard, Bresolin et al. did not identify pnp, the gene for polynucleotide phosphorylase, which Goverde et al. (103) had previously determined is required by the same strain of Y. enterocolitica investigated by Bresolin et al. to grow at low temperature. Clearly, much remains to be learned about the mechanisms of cold adaptation by Y. enterocolitica and Y. pseudotuberculosis.

Growth, Susceptibility, and Tolerance

The optimum growth temperature of both Y. enteroco­ litica and Y. pseudotuberculosis is approximately 28 to 30°C. The doubling time of Y. enterocolitica at this temperature is approximately 34 min, which increases to 1 h at 22°C, 5 h at 7°C, and approximately 40 h at 1°C (256). Y. enterocolitica readily withstands freezing and can survive in frozen foods for extended periods even after repeated freezing and thawing (299). Studies of the ability of Y. enterocolitica to survive and grow in artificially contaminated foods under different conditions of storage have revealed it generally survives better at room and refrigeration temperatures than at intermediate temperatures. Y. enterocolitica persists longer in cooked foods than in raw foods, probably due to an increased availability of nutrients in cooked foods and the fact that the presence of other psychrotrophic bacteria in unprocessed food may restrict bacterial growth (256). The number of viable Y. enterocolitica cells may increase more than 1 million-fold on cooked beef or pork within 24 h at 25°C, or within 10 days at 7°C (117). Y. pseudotuberculosis can grow in raw ground beef at temperatures ranging from 0 to 30°C, with growth rates of 0.023, 0.622, and 0.236 CFU/hour at 0, 25, and 30°C, respectively (22). Y. enterocolitica can grow at refrigeration temperature in vacuum-­packaged meat, boiled eggs, boiled fish, pasteurized liquid eggs, pasteurized whole milk, cottage cheese, and tofu (soybean curd) (88, 256). Proliferation also occurs in refrigerated seafood, such as oysters, raw shrimp, and cooked crab meat, but at a slower rate than in pork or beef (220). Bacteria may also persist for extended periods in refrigerated vegetables and cottage cheese, particularly in the presence of chicken meat (268). The psychrophilic nature of Y. enterocolitica also poses problems for the blood transfusion industry, largely because of the pathogen’s ability to proliferate and release endotoxin

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

344 in blood products stored at 4°C without manifestly altering their appearance (13). Y. enterocolitica and Y. pseudotuberculosis are able to grow over a pH range of approximately pH 4 to 10, with an optimum pH of ca. 7.6 (252). They can tolerate alkaline conditions extremely well, but their acid tolerance is less pronounced and depends on the environmental temperature, the composition of the medium, the growth phase of the bacteria, and the acidulant used (4, 322). Under most conditions studied, Y. enterocolitica is more susceptible to lactic acid than to citric acid (322). Acid tolerance of Y. enterocolitica is enhanced by the production of urease, which hydrolyzes urea to release ammonia that elevates the cytoplasmic pH (75). Y. enterocolitica and Y. pseudotuberculosis are susceptible to heat and are readily inactivated by pasteurization at 71.8°C for 18 s or 62.8°C for 30 min (70, 299). Exposure of surface-contaminated meat to hot water (80°C) for 10 to 20 s reduces Yersinia viability by at least 99.9% (279). However, the resistance to heat of pathogenic Y. enterocolitica can vary with growth temperature and serotype (122). Y. enteroco­ litica is also susceptible to ionizing and UV irradiation (42, 80), high-pressure homogenization (336), and sodium nitrate and nitrite when added to food (72). It is relatively resistant to sodium nitrate and nitrite in solution, however, and can also tolerate NaCl at concentrations of up to 5% (72, 282). Y. enterocolitica is generally susceptible to chlorine, although some resistance is displayed by yersiniae grown under conditions that approximate natural aquatic environments or when they are cocultivated with predatory aquatic protozoa (85, 120, 160, 323).

CHARACTERISTICS OF INFECTION The clinical manifestations of infections with Y. entero­ colitica and Y. pseudotuberculosis overlap considerably and are indistinguishable from one other. Infections with Y. enterocolitica are far more common than those with Y. pseudotuberculosis; hence, this section will focus on the former. Infection with the enteropathogenic yersiniae typically manifests as nonspecific, self-limiting ­diarrhea but may produce a variety of suppurative and autoimmune complications (Table 14.4), the risk of which is determined partly by host factors, in particular age and underlying immune status.

Acute Infection

Y. enterocolitica and Y. pseudotuberculosis enter the gastrointestinal tract after ingestion in contaminated food or water. The median infective dose for humans is

SMP_Food Microbiology_CH14.indd

344

Table 14.4  Clinical manifestations of infections with

Y. enterocolitica

Common manifestations Diarrhea (“gastroenteritis”), especially in young children Enterocolitis Pseudoappendicitis syndrome due to terminal ileitis, acute mesenteric lymphadenitis Pharyngitis Postinfection autoimmune sequelae Arthritis, especially associated with HLA-B27 Erythema nodosum Uveitis, associated with HLA-B27 Glomerulonephritis (uncommon) Myocarditis (uncommon) Thyroiditis (uncertain) Less common manifestations Septicemia Visceral abscesses; for example, in liver, spleen, lung Skin infections: pustules, wound infection, pyomyositis Pneumonia Endocarditis Osteomyelitis Peritonitis Meningitis Intussusception Eye infections: conjunctivitis, panophthalmitis

not known but is likely to exceed 104 CFU. Gastric acid appears to be a significant barrier to infection with Y. enterocolitica, and in individuals with gastric hypoacidity, the infectious dose may be lower (75, 90). Most symptomatic infections with Y. enterocolitica occur in children, especially in those less than 5 years of age. In these patients, yersiniosis presents as diarrhea, often accompanied by low-grade fever and abdominal pain (136, 180). The character of the diarrhea varies from watery to mucoid. A small proportion of children (generally less than 10%) have frankly bloody stools. Children with Y. enterocolitica-induced diarrhea often complain of abdominal pain and headache. Sore throat is a frequent accompaniment and may dominate the clinical picture in older patients (287). The illness typically lasts from a few days to 3 weeks, although some patients develop chronic enterocolitis, which may persist for several months (249). Occasionally, acute enteritis progresses to intestinal ulceration and perforation or to ileocolic intussusception, toxic megacolon, or mesenteric vein thrombosis (65). On rare occasions, patients may present with peritonitis in the absence of intestinal perforation (235).

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

345

In children older than 5 years of age and adolescents, acute yersiniosis often presents as a pseudoappendicular syndrome due to acute inflammation of the terminal ileum or the mesenteric lymph nodes. The usual features of this syndrome are abdominal pain and tenderness localized to the right lower quadrant. These symptoms are usually accompanied by fever, with little or no diarrhea. The importance of this form of the disease lies in its close resemblance to appendicitis (65). Of those patients with this syndrome who undergo surgical treatment, approximately 60 to 80% have terminal ileitis, with or without mesenteric adenitis, and a normal or slightly inflamed appendix (230, 312). Y. enterocolitica may be cultured from the distal ileum and the mesenteric lymph nodes. The pseudoappendicular syndrome appears to be more frequent in patients infected with the relatively more virulent strains of Y. enterocolitica, notably strains of biotype 1B and with Y. pseudotuberculosis. Y. enterocolitica is rarely present in patients with true appendicitis (312). Although Y. enterocolitica is seldom isolated from extraintestinal sites, there is no known tissue in which it will not grow. In adults, pharyngitis, sometimes with cervical lymphadenitis, may dominate the clinical presentation (287). Focal disease, in the absence of obvious bacteremia, may present as cellulitis, subcutaneous abscess, pyomyositis, suppurative lymphadenitis, septic arthritis, osteomyelitis, urinary tract infection, renal abscess, sinusitis, pneumonia, lung abscess, or empyema (65). Bacteremia is a rare complication of infection, except in patients who are immunocompromised or in an iron-overloaded state (90). Factors that predispose to the ­development of Yersinia bacteremia include immunosuppression, blood dyscrasias, malnutrition, chronic renal failure, cirrhosis, alcoholism, diabetes mellitus, and acute and chronic iron overload states, particularly when managed by chelation therapy with desferrioxamine B (65). Bacteremic dissemination of Y. enterocolitica may lead to various manifestations, including splenic, hepatic, and lung abscesses, catheter-associated infections, osteomyelitis, panophthalmitis, endocarditis, mycotic aneurysm, and meningitis (65). Yersinia bacteremia is reported to have a case-fatality rate of between 30 and 60%. Bacteremia may also result from direct inoculation of Y. enterocolitica into the circulation during blood transfusion (34, 167). Indeed, Y. enterocolitica is one of the most important causes of fatal bacteremia following transfusion with packed red blood cells or platelets (34). Patients infused with contaminated blood may develop symptoms of a severe transfusion reaction minutes to hours after exposure, depending on the number of bacteria and the amount of endotoxin administered with the blood (13). The varieties of Y. enterocolitica responsible

SMP_Food Microbiology_CH14.indd

345

for transfusion-acquired yersiniosis are the same serobiotypes as those associated with enteric infections. The probable sources of these infections are blood donors with low-grade, subclinical bacteremia. A small number of bacteria in donated blood can increase during storage at refrigeration temperatures without manifestly altering the appearance of the blood (296).

Autoimmune Complications

Although most episodes of yersiniosis remit spontaneously without long-term sequelae, infections with the enteropathogenic yersiniae are noteworthy for a variety of immunological complications, such as reactive arthritis, erythema nodosum, uveitis, glomerulonephritis, carditis, and thyroiditis, which may follow acute infection (65). Of these, reactive arthritis is the most widely recognized (6, 28, 170). This manifestation of yersiniosis is infrequent before the age of 10 years and occurs most often in Scandinavian countries, where serotype O:3 strains and the human leukocyte antigen HLA-B27 are especially prevalent. Men and women are affected equally. Arthritis typically follows the onset of diarrhea or the pseudoappendicular syndrome by 1 to 2 weeks. The joints most commonly involved are the knees, ankles, toes, tarsal joints, fingers, wrists, and elbows. Synovial fluid from affected joints contains large numbers of inflammatory cells, principally polymorphonuclear leukocytes (PMNs), and is invariably sterile, although it generally contains bacterial antigens (104). The duration of arthritis is typically less than 3 months, and the long-term prognosis is good, although some patients may have symptoms that persist for several years (53, 130). Many patients with arthritis also have extra-articular symptoms, including urethritis, uveitis, and erythema nodosum (28). Yersinia-induced erythema nodosum occurs predominantly in women and is not associated with HLA-B27. Other autoimmune complications of yersiniosis, including uveitis, acute proliferative glomerulonephritis, collagenous colitis, and rheumatic-like carditis, have been reported, mostly from Scandinavian countries (166). Yersiniosis has also been linked to various thyroid disorders, including Graves’ disease hyperthyroidism, nontoxic goiter, and Hashimoto’s thyroiditis, although the causative role of yersiniae in these conditions is uncertain (297). Y. pseudotuberculosis has also been implicated in the etiology of some cases of Kawasaki’s disease (320).

RESERVOIRS Infections with Yersinia species are zoonotic. Y. en­ terocolitica and Y. pseudotuberculosis occur in a broad range of environments and have been isolated from the

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

346 intestinal tract of many different mammalian species as well as from birds, frogs, fish, flies, fleas, crabs, and oysters (30, 65, 306). The subgroups of Y. enterocolitica that commonly occur in humans are also present in domesticated animals, whereas Y. pseudotuberculosis and the subgroups of Y. enterocolitica that are infrequent in humans generally infect wild rodents (98). Nevertheless, Y. pseudotuberculosis does also infect and cause disease in food and companion animals (97, 277, 278). Foods that may harbor Y. enterocolitica include pork, beef, lamb, poultry, and dairy products such as milk, cream, and ice cream (88, 155, 256). Recently reported outbreaks of infection with Y. pseudotuber­ culosis were linked to the consumption of lettuce and carrots (151, 211). Yersiniae are also commonly found in a variety of terrestrial and freshwater ecosystems, including soil, vegetation, lakes, rivers, wells, and streams, and can persist for extended periods in soil, vegetation, streams, lakes, wells, and spring water, especially at low environmental temperatures (49, 155). Most isolates of Y. enterocolitica obtained from these sources, however, lack markers of bacterial virulence and are of uncertain significance for human or animal health (195). Although Y. enterocolitica has been recovered from a wide variety of wild and domesticated animals, pigs are the only animal species from which Y. enterocolitica of biotype 4 serotype O:3 (the variety most commonly associated with human disease) has been isolated with any degree of frequency (155). Pigs may also carry Y. en­ terocolitica of serotypes O:9 and O:5,27, particularly in regions where human infections with these varieties are relatively common. In countries with a high incidence of human yersiniosis, Y. enterocolitica is frequently isolated from pigs at slaughterhouses (10, 94, 113). The tissue most frequently culture positive at slaughter is the tonsils, which appears to be the preferred site of Y. en­ terocolitica infection in pigs. Other sites that frequently yield yersiniae include the tongue, cecum, rectum, feces, and gut-associated lymphoid tissue. Known pathogenic serotypes of Y. enterocolitica are seldom isolated from meat offered for retail sale, however, apart from pork tongue (155, 256), although standard methods of bacterial isolation and detection may underestimate the true incidence of contamination (95, 204). Some domesticated farm animals, notably sheep and cattle, may suffer symptoms as a result of infection with Y. enterocolitica and Y. pseudotuberculosis (276–278), but in most cases the biotypes and serotypes of these bacteria differ from those responsible for human infection, indicating a lack of transmission of these particular bacteria between animals and humans. By contrast, individual isolates of Y. enterocolitica from pigs and hu-

SMP_Food Microbiology_CH14.indd

346

mans are indistinguishable from each other in terms of serotype, biotype, restriction fragment length polymorphism of chromosomal and plasmid DNA, and carriage of virulence determinants (155). Further evidence that pigs are a significant reservoir of human infections is provided by epidemiologic studies pointing to the ingestion of raw or undercooked pork as a major risk factor for the acquisition of yersiniosis (215, 292). Infection also occurs after handling of contaminated chitterlings (pig intestine), particularly by children (153, 168). Food animals are seldom contaminated with biotype 1B strains of Y. enterocolitica (Y. enterocolitica subsp. enterocolitica), the reservoir of which remains unknown (256) but appears likely to be wild rodents (123). Wild rodents are also the probable reservoir of Y. pseudo­ tuberculosis (98, 319). The relatively low incidence of human yersiniosis caused by Y. pseudotuberculosis and Y. enterocolitica subsp. enterocolitica, despite their comparatively high virulence, points to a lack of significant contact between their reservoir and humans.

INCIDENCE OF INFECTIONS Considering the widespread occurrence of yersiniae in nature and their ability to colonize food animals, to persist within animals and the environment, and to proliferate at refrigeration temperatures, infections with Y. enterocolitica and Y. pseudotuberculosis are surprisingly uncommon. In the United States, active surveillance by FoodNet for laboratory-confirmed cases of yersiniosis from 1996 to 2007 determined that the annual average incidences of Y. enterocolitica and Y. pseudotuberculo­ sis were 0.35 and 0.004 per 100,000, respectively (177). For comparison, in 2007, the incidences of infections with Campylobacter, Salmonella, Shigella, Shiga toxinproducing E. coli, and Yersinia were 12.79, 14.92, 6.26, 1.77, and 0.36 per 100,000, respectively (48). The incidence of infections with Y. enterocolitica in northern Europe, in particular in Belgium, The Netherlands, Denmark, Sweden, Norway, and Finland is far higher than in the United States but appears to be decreasing, at least in Belgium (214, 314). Although there are no comparable published data for Y. pseudotuberculosis, infections with this species appear to be relatively common in Japan (306) and may be increasing in some parts of Europe (319).

FOODBORNE OUTBREAKS Most foodborne outbreaks of Y. enterocolitica infection in which a source was identified have been traced to milk (Table 14.5). As Y. enterocolitica is rapidly de-

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

347

Table 14.5  Selected foodborne outbreaks of infections with Y. enterocolitica Location Canada New York Japan New York Washington Pennsylvania Southern United States Hungary Georgia, United States Northeastern United States a

Yr

Mo

No. of cases

1976 1976 1980 1981 1981 1982 1982 1983 1989 1995

April September April July December February June December November October

138 38 1,051 159 50 16 172 8 15 10

Serotype O:5,27 O:8 O:3 O:8 O:8 O:8 O:13a,13b O:3 O:3 O:8

Source

a

Raw milk (?) Chocolate-flavored milk Milk Powdered milk/chow mein Tofu/spring water Bean sprouts/well water Milk (?) Pork “cheese” (sausage) Pork chitterlings Pasteurized milk (?)

Reference 156 26 182 264 286 65 288 179 169 3

(?), the bacteria were not isolated from the incriminated source.

stroyed by pasteurization, infection results from the consumption of raw milk or milk that is contaminated after pasteurization (82, 254). During the mid-1970s, two outbreaks of yersiniosis caused by Y. enterocolit­ ica O:5,27 occurred among 138 Canadian schoolchildren who had consumed raw milk, but the organism was not recovered from the suspected source (156). In 1976, Y. enterocolitica serotype O:8 biotype 1B was responsible for an outbreak in New York State that affected 217 people, 38 of whom were culture positive (26). The source of infection was chocolateflavored milk, which evidently became contaminated after pasteurization. In 1981, an outbreak of Y. enterocolitica O:8 infection affected 35% of 455 individuals at a diet camp in New York State (196). Seven patients were hospitalized as a result of infection, five of whom underwent appendectomies. The source of the infection was reconstituted powdered milk and/or chow mein, which probably became contaminated during preparation by an infected food handler. In 1982, 172 cases of infection with Y. enterocolitica O:13a,13b occurred in an area that included parts of Tennessee, Arkansas, and Mississippi (288). The suspected source was pasteurized milk that may have become contaminated with pig manure during transport (82). More recently, three unrelated outbreaks of infection with Y. enterocolitica O:3 affecting infants and children in Atlanta, Chicago, and the state of Tennessee between 1989 and 2002 were attributed to the transmission of yersiniae from raw chitterlings on the hands of food handlers to affected children (47, 153, 168). Other foods that have been responsible for outbreaks of yersiniosis include “pork cheese” (a type of sausage prepared from chitterlings), bean sprouts, and tofu (65, 179, 286). In the outbreaks associated with bean sprouts and tofu,

SMP_Food Microbiology_CH14.indd

347

contaminated well or spring water was the probable source of the bacteria. Water was also the putative source of infection in a case of sporadic Y. enterocolitica bacteremia in a 75-year-old man in the state of New York (157) and in a small family outbreak in Ontario, Canada (295). Several outbreaks of presumed foodborne Y. enterocolitica O:3 infection have also been reported from the United Kingdom and Japan, but in most cases the sources of these outbreaks were not identified. In 1989, Tsubokura et al. (306) reported the features of Y. pseudotuberculosis infection in Japan. Their paper included a summary of 12 outbreaks, which altogether affected more than 1,200 people, mostly children. In eight outbreaks, in which the probable source was identified, water, vegetable juice, sandwiches, barbecue, and school lunches were the suspected vehicles. More recently, two outbreaks of foodborne Y. pseudotuberculosis infection have been reported in Finland. One outbreak of Y. pseudotuberculosis O:3 infection in 1998 affected 47 people, 1 of whom died and 5 of whom underwent appendectomies (211). A case-control study associated the consumption of locally grown iceberg lettuce with infection, and while no lettuce was cultured at the time of the outbreak, subsequent culture of soil, irrigation water, and lettuce yielded Y. pseudotuberculosis that was indistinguishable from the outbreak strain. In 2003, a second outbreak that was caused by Y. pseudotuberculosis O:1 occurred (151). This outbreak presented as a cluster of gastrointestinal illness and erythema nodosum in schoolchildren who had eaten lunches prepared in the same kitchen. A case-control study identified raw grated carrots as the likely source of infection. This was confirmed by microbiological studies, which revealed Y. pseudotuberculosis O:1 isolates with the same pulsed-field gel electrophoresis subtype obtained

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

348 from patients’ feces, from soil samples that contained carrot residue, and from peeling and washing equipment at the production farm.

epithelial cells in moderate numbers and resist killing by macrophages to a greater extent than the second category, i.e., biotype 1A strains obtained from nonclinical sources (105, 106, 269). Clinical and environmental biotype 1A strains of the same O serogroup also differ from each other genetically (248). In addition, products of a gene complex resembling the insecticidal toxin complex genes first discovered in Photorhabdus luminescens were determined to contribute to the virulence of some strains of biotype 1A by facilitating their persistence in vivo (294). Because the mechanisms by which biotype 1A strains cause disease are largely unknown, the remainder of this section will focus chiefly on the virulence determinants of the classical pathogenic, i.e., pYV-bearing, highly invasive strains of Y. enterocolitica.

MECHANISMS OF PATHOGENICITY Y. enterocolitica is an invasive enteric pathogen whose v­ irulence determinants have been the subject of intensive investigation (57, 121, 316), but not all strains of Y. enterocolitica are equally virulent (Table 14.6) Y. enterocolitica strains of biotypes of 1B, 2, 3, 4, and 5 possess a panoply of virulence determinants, including a chromosomally encoded invasin and a ca. 70-kb virulence plasmid, termed pYV (acronym for plasmid for Yersinia virulence) (57, 229, 316). In addition, strains of Y. enterocolitica subsp. enterocolitica (i.e., biotype 1B strains) harbor two pathogenicity islands that are associated with enhanced virulence. All pYV-bearing clones of Y. enterocolitica have the capacity to invade epithelial cells in large numbers in vitro, a feature that distinguishes them from clones that never carry pYV (105, 238). Paradoxically, however, this highly invasive phenotype is not specified by genes within pYV and is maximally expressed by bacteria from which pYV has been cured. Until recently, weakly invasive, pYV-negative strains of Y. enterocolitica, most of which belong to biotype 1A, were regarded as avirulent, because they never carry pYV or any of the other well-characterized virulence-­associated genes of this species (see below). However, there is now persuasive clinical, epidemiologic, and experimental evidence indicating that at least some of these strains may cause gastrointestinal symptoms clinically indistinguishable from those caused by pYV-bearing strains (293). This is supported by laboratory results that separate biotype 1A strains into two categories, the first consisting of isolates recovered from symptomatic patients that can penetrate

Pathological Changes

Examination of surgical specimens from patients with yersiniosis has confirmed that Y. enterocolitica is an invasive pathogen that displays a tropism for lymphoid tissue, with the distal ileum, in particular the epithelium overlying the intestinal lymphoid follicles (Peyer’s patches), bearing the brunt of the infection (40, 54). The mesenteric lymph nodes are frequently enlarged and contain focal areas of necrosis. As investigations in volunteers are precluded by the risk of autoimmune sequelae, most information regarding the pathogenesis of yersiniosis in vivo has been obtained from animal models, in particular mice and rabbits (124, 173, 309). Although these animals are not the natural hosts of the serotypes of Y. enterocolitica that commonly infect humans, they have provided valuable insights into the probable pathogenesis of human disease. Nevertheless, some data derived from animal studies should be interpreted with caution, particularly when death is used as the end point of infection, since this is not the usual outcome of human infection (222).

Table 14.6  Characteristics of pathogenic subgroups of Y. enterocolitica Subgroup Classical

Atypical a

Capacity to invade epithelial cells in vitro

Biotype(s)a

High (lower if pYV present)

2, 3, 4, 5

High (lower if pYV present)

1B

Low to moderate

1A

Virulence-associated determinants Invasin; Ail; fibrillae (Myf); urease; heat-stable enterotoxin (Yst); LPS; and a virulence plasmid (pYV) that encodes YadA, the Ysc T3SS, and translocator and effector proteins (Yops) All of the above, plus the HPI; the Ysa T3SS; the Yts1 T2SS pathway, and phospholipase A (YplA) Products of genes homologous to those for an insecticidal toxin complex

Biotypes 1A, 2, 3, 4, and 5 are also known as Y. enterocolitica subsp. palearctica; biotype 1B is also known as Y. enterocolitica subsp. enterocolitica.

SMP_Food Microbiology_CH14.indd

348

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

349

After oral inoculation of mice with a virulent strain of Y. enterocolitica, most bacteria remain within the intestinal lumen, whereas a small number adhere to the mucosal epithelium, having no particular preference for any cell type (111). However, invasion of the epithelium occurs almost exclusively through M cells (microfold cells) (Fig. 14.1) (111). These are specialized epithelial cells that overlie Peyer’s patches, where they play a major role in antigen sampling (152, 164, 251). Studies in experimentally infected mice, rabbits, and pigs have revealed that after penetrating the epithelium, Y. enterocolitica traverses the basement membrane to reach the gut-associated lymphoid tissue and the lamina propria, where it causes localized tissue destruction and the formation of microabscesses (Fig. 14.2) (174, 212, 244). These lesions occur chiefly within intestinal crypts but may extend as far as the crypt-villus junction. Y. enterocolitica often spreads via the lymph to the draining mesenteric lymph nodes, where it may also lead to microabscess formation. If the bacteria circumvent the lymph nodes to enter the bloodstream, they can disseminate to any organ but continue to show a tropism for lymphoid tissue by preferentially localizing in the ­reticuloendothelial tissues of the liver and spleen. Although Y. enterocolitica is often regarded as a facultative intracellular pathogen

because of its innate resistance to killing by macrophages (74, 229), most of the bacteria observed in histological sections are located extracellularly (119). Nevertheless, macrophages may provide a niche for bacterial replication during the early stages of infection and serve as a vehicle for their dissemination throughout the body (229). In mice, highly virulent (biotype 1B) strains of Y. entero­ colitica can bypass the Peyer’s patches and disseminate via the bloodstream directly (44).

Virulence Determinants Chromosomal Determinants of Virulence Invasin All strains of Y. enterocolitica that carry pYV also produce a 91-kDa surface-expressed protein termed invasin. This outer membrane protein was first identified in Y. pseudotuberculosis as a 102-kDa protein product of the chromosomal inv gene (144). When introduced into an innocuous laboratory strain of E. coli, such as E. coli K-12, inv imbues the recipient with the ability to penetrate mammalian cells, including epithelial cells and macrophages (74, 144). Despite differences in the sizes of invasins from Y. enterocolitica and

Figure 14.1  Transmission electron micrograph showing the initial interaction (black ­arrowhead) and transport (white arrow) of Y. enterocolitica through an intestinal M cell, 60 min after inoculation into mouse ileum. (Reprinted with permission from reference 111.) doi:10.1128/9781555818463.ch14f1

SMP_Food Microbiology_CH14.indd

349

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

350

Figure 14.2  Light micrograph of a section through the colon of a gnotobiotic piglet 3 days after inoculation with a virulent strain of Y. enterocolitica O:3. Note the microabscess, comprising mostly bacteria, the surrounding inflammatory cells (arrows), and the disrupted epithelium with vacuolated and necrotic cells. Epoxy section, methylene blue stain. (Reprinted with permission from reference 308.) doi:10.1128/9781555818463.ch14f2

Y. pseudotuberculosis, these two proteins are functionally highly conserved. The amino terminus of invasin is inserted in the bacterial outer membrane, while the carboxyl terminus is exposed on the surface, where it mediates binding to host cell integrins (250). The latter are heterodimeric transmembrane proteins that communicate extracellular signals to the cytoskeleton. Integrins comprise a and b subunits, which form the basis of their classification into families. The b1 class integrins, which are the principal

SMP_Food Microbiology_CH14.indd

350

receptors for invasin, occur on many cell types including epithelial cells, macrophages, T lymphocytes, and Peyer’s patch M cells (52). Their physiological role is to act as receptors for fibronectin, laminin, and related host proteins, which may bear conformational similarities to invasin (115). However, the affinity of invasin for integrins a3 b1, a4 b1, a5 b1, and a6 b1 is much greater than that of fibronectin. Accordingly, when invasin binds to these integrins, it causes them to cluster and initiate a sequence of events, including the activation of focal adhesion kinase

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

351

(Fak) and the small GTPase Rac1, which results in reorganization of the host cell cytoskeleton and internalization of the bacteria (143, 334). The internalization process is governed entirely by the host cell, because nonviable bacteria and even latex particles coated with invasin are internalized in the same way as living bacteria (145). Although DNA sequences homologous to inv occur in all Yersinia species (except Y. ruckeri), this gene is functional only in Y. pseudotuberculosis and the classical pathogenic biotypes (1B through 5) of Y. enterocolit­ ica, suggesting that invasin plays a key role in virulence (191). Nevertheless, although inv mutants of Y. entero­ colitica have a pronounced reduction in their ability to invade epithelial cells in vitro, their virulence for orally inoculated mice is barely affected (222). There is evidence, however, that invasin contributes to gastrointestinal tract colonization of mice, as inv-negative mutants of Y. enterocolitica present diminished translocation into Peyer’s patches and lymph nodes compared with wild-type strains (222, 302).

Ail Classical pathogenic strains of Y. enterocolitica produce an outer membrane protein unrelated to invasin, which also confers invasive ability on E. coli (189). This 17-kDa peptide is specified by a chromosomal ail (attachmentinvasion) locus, so called because it mediates bacterial attachment to some cultured epithelial cell lines and invasion of others. In concert with YadA, a pYV-encoded protein, Ail may also enable yersiniae to persist extracel-

lularly by protecting them from complement-mediated killing (25, 190). Ail and YadA achieve this by binding C4b-binding protein, which is an inhibitor of both the classical and lectin pathways of complement (161). The ail gene occurs only in the classical pathogenic varieties of Y. enterocolitica (238) and in vitro is optimally expressed at 37°C (unlike inv, which is optimally expressed at 25°C, unless the pH is reduced to 5.5 [221]). Further circumstantial evidence supporting a role for Ail in virulence stems from the finding that ail mutants fail to adhere to or invade cultured cells when the inv gene is not expressed (142). Surprisingly, however, an ail-negative mutant of Y. enterocolitica showed no reduction in virulence for perorally inoculated mice, indicating that Ail is not required to establish infection or even to cause systemic infection in these animals (324). However, as Y. enterocolitica is inherently resistant to complement-mediated killing by murine serum (in contrast to its pronounced susceptibility to killing by human serum [216]), the mouse model is not well suited to investigate the contribution of anticomplement factors to virulence.

Heat-stable enterotoxins When first isolated from clinical material, most strains of Y. enterocolitica secrete a heat-stable enterotoxin, known as Yst (or Yst-a), which is reactive in infant mice (217). Yst is comprised of 30 amino acids (Fig. 14.3). Its carboxyl terminus is homologous to those of heat-stable enterotoxins from enterotoxigenic E. coli, Citrobacter

Figure 14.3  Amino acid sequences of the mature heat-stable enterotoxins produced by Y. enterocolitica (138, 234, 291), enterotoxigenic E. coli of human (STh) and porcine (STp) subtypes (8, 289), C. freundii (112), V. cholerae non-O1 (290), and the intestinal hormone guanylin (68). Amino acid residues that are shaded are common to all seven peptides. The first 23 amino acids at the N terminus of the Yst-c mature toxin (denoted by superscript “a”) are not included in the sequence alignments. doi:10.1128/9781555818463.ch14f3

SMP_Food Microbiology_CH14.indd

351

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

352 freundii, and non-O1 serotypes of Vibrio cholerae, and also to guanylin, an intestinal paracrine hormone (Fig. 14.3) (8, 68, 112, 138, 234, 289–291). These polypeptides share a common mechanism of action that involves binding to and activation of cell-associated guanylate cyclase, with subsequent elevation of intracellular concentrations of cyclic GMP (242). This in turn causes perturbation of fluid and electrolyte transport in intestinal absorptive cells, which may result in diarrhea. Despite its similarity to known virulence factors and the fact that production of Yst is by and large restricted to the classical pathogenic biotypes of Y. enterocolitica (77), the contribution of Yst to the pathogenesis of diarrhea associated with yersiniosis is uncertain. Doubts regarding the role of Yst in virulence stem from the following observations: (i) the toxin is generally not detectable in bacterial cultures incubated at temperatures above 30°C, (ii) production of Yst has not been observed in vivo, and (iii) strains of Y. enterocolitica that have spontaneously lost the ability to produce Yst retain full virulence in experimental animals (244, 253). However, Delor and Cornelis determined that a ystnegative mutant of a strain of Y. enterocolitica serotype O:9 caused milder diarrhea in infant rabbits than did the wild-type strain (76). This observation suggests that the reason why yst is not normally expressed at 37°C is that the conditions used to study expression of this gene in vitro do not necessarily reflect those to which the bacteria are exposed in vivo. In this regard, the finding that Y. enterocolitica can produce Yst at 37°C if the bacteria are grown in media with an osmolarity and pH resembling those in the intestinal lumen is significant, although the precise stimulus to Yst synthesis in vivo is not yet known (188). After repeated passage or prolonged storage, Ystsecreting strains of Y. enterocolitica frequently become Yst negative. This phenomenon is due to silencing of yst by YmoA (acronym for Yersinia modulator). The latter is an 8-kDa, Hha-like protein, which downregulates gene expression in yersiniae by binding to DNA and altering its topology (62). Expression of yst is also regulated by sS (RpoS), an alternative sigma factor of RNA polymerase, which is involved in regulating the expression of a number of stationary-phase genes in enterobacteria (202). Toxins that resemble Yst in terms of heat stability and reactivity in infant mice, but with a different structure, molecular weight, and/or mechanism of action, have been detected in various Yersinia species, including biotype 1A strains of Y. enterocolitica and “avirulent” Yersinia species, such as Y. bercovieri and Y. mollaretii (234, 243, 285, 339). Some of these toxins have been

SMP_Food Microbiology_CH14.indd

352

characterized, and the genes encoding their production have been cloned and sequenced (Fig. 14.3). The availability of diagnostic probes for the genes encoding Yst-b and Yst-c has revealed that biotype 1A Y. enterocolitica strains commonly carry the genes for Yst-b, whereas Yst-c, which was originally identified in a serotype O:3 strain of biotype 4, is rare (105, 234, 339). Some strains of Y. enterocolitica elaborate Yst or other enterotoxins over a wide range of temperatures, from 4°C to 37°C (154, 243). Since these toxins are relatively acid stable, they could resist inactivation by stomach acid and thus conceivably cause food poisoning if they were ingested preformed in food. In artificially inoculated foods, however, these toxins are synthesized optimally at 25°C during the stationary phase of bacterial growth (255). Accordingly, the storage conditions required for their production in food would generally result in severe spoilage, thereby reducing the likelihood of ingestion of preformed toxin.

Myf fibrillae Most intestinal pathogens possess distinctive colonization factors on their surface, which mediate their ­adherence to the intestinal epithelium. In noninvasive, enterotoxin-secreting bacteria, such as enterotoxigenic E. coli, these factors frequently take the form of surface fimbriae, which allow the bacteria to deliver their toxins close to epithelial cells while resisting removal by peristalsis (99). In enteroinvasive bacteria, surface adhesins may augment virulence by enabling the bacteria to adhere to cells, such as M cells, which they preferentially invade. The key intestinal colonization factors of Y. enterocolitica are invasin and YadA (see below), which mediate binding to M cells. In addition, some strains of Y. enterocolitica produce a fimbrial adhesin, named Myf (acronym for mucoid Yersinia fibrillae), because it bestows a mucoid appearance on bacterial colonies that express it (141). Myf are narrow flexible fimbriae that resemble CS3, an essential colonization factor of some human strains of enterotoxigenic E. coli. MyfA, the major structural subunit of Myf, has some homology to the PapG protein of pyelonephritis-associated strains of E. coli and is 44% identical at the DNA level to the pH 6 antigen of Y. pseudotuberculosis and Y. pestis, which also has a fibrillar structure and mediates thermoinducible binding of Y. pseudotuberculosis to tissue culture cells (175, 338). Like ail and yst, myf occurs predominantly in Y. enterocolitica strains of the classical pathogenic varieties commonly associated with disease (141). The pH 6 antigen of Y. pseudotuberculosis is synthesized within the acidic phagolysozomes of macrophages and may be involved in the interaction between bacteria

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

353

and phagocytic cells, although it does not appear to contribute to bacterial survival in these cells. Its main role in virulence may relate to its ability to mediate binding of bacteria to intestinal mucus before the bacteria make contact with epithelial cells (181). Although Myf may contribute to the colonizing ability of Y. enterocolitica, direct proof of this is lacking.

LPS As with other enterobacteria, Y. enterocolitica can be classified as smooth or rough depending on the amount of O-side chain polysaccharide attached to the inner core region of the cell wall LPS. Synthesis of the O-side chain by Y. enterocolitica is regulated by temperature, such that colonies are smooth when grown at temperatures below 30°C but rough at 37°C (273). O-antigen-­negative mutants of a Y. enterocolitica serotype O:8 strain have impaired ability to colonize Peyer’s patches, spleen, and liver of mice infected via different routes (19). These mutants also have altered expression or function of other virulence-associated determinants, including YadA, Ail, phospholipase A (YplA), and flagellin, indicating that LPS may be required for the normal expression or function of other surface molecules or secretory systems. In addition, the outer core region of LPS plays a role in maintaining outer membrane integrity and may contribute to the resistance of Y. enterocolitica to bactericidal peptides in host tissues and macrophages (271, 273). Smooth LPS may also enhance virulence by increasing bacterial hydrophilicity and thus facilitate their passage through the mucous secretions that line the intestinal epithelium. Flagellum and biofilm formation When grown in vitro, Y. enterocolitica is motile at 25°C but not at 37°C. Expression of the genes encoding flagellin and invasin appears to be coordinately regulated in that mutants of Y. enterocolitica that are defective in the expression of invasin are hypermotile (16). In addition, a strain bearing a mutation in the master regulatory operon for flagellum biosynthesis, flhDC, had enhanced production of pYV-encoded virulence proteins (27). Flagella also have an essential role in biofilm formation by Y. enterocolitica (159), which may enhance the bacterium’s ability to persist in some environments. Nevertheless, motility does not appear to contribute to the virulence of Y. enterocolitica for mice (140), and Y. enterocolitica does require formation in a biofilm to kill the nematode Caenorhabditis elegans (281). However, biofilm formation is a prerequisite for the virulence and transmission of Y. pestis from fleas to mammals (71, 132). Biofilm formation

SMP_Food Microbiology_CH14.indd

353

by Y. pestis in the flea foregut involves the formation of an extracellular matrix that is synthesized by products of Yersinia hms genes. Y. pseudotuberculosis possesses hms genes identical to those of Y. pestis and can form biofilms in vitro (132). Biofilm formation may enable Y. pseudotuberculosis to persist on certain foods, such as lettuce, during washing, but this has not been investigated.

Phospholipase Some isolates of Y. enterocolitica are hemolytic due to the production of phospholipase A (YplA). A strain of Y. enterocolitica in which the yplA gene encoding this enzyme was deleted had diminished virulence for perorally inoculated mice (260). Interestingly, this mutant induced less inflammation and necrosis in intestinal and lymphoid tissues than did the wild-type strain, suggesting that phospholipase contributes to microabscess formation by Y. enterocolitica. YplA is secreted by Y. enterocolitica via the same type III export apparatus as that used for flagellar proteins (325). Y. pseudotuber­ culosis also produces a phospholipase. Although this enzyme is highly homologous to Y. enterocolitica YlpA, the two proteins differ in their activity, regulation, and secretion (186). Iron acquisition and the high-pathogenicity island (HPI) Iron is an essential micronutrient of almost all bacteria. Despite the nutrient-rich environment provided to bacteria by mammalian tissues, the availability of iron in many extracellular locations is limited (330). This is because most extracellular iron is bound to high-affinity transport glycoproteins such as transferrin and lactoferrin or is incorporated into organic molecules. Several species of pathogenic bacteria produce low-molecular-weight, high-affinity iron chelators known as siderophores (203). These compounds are secreted by the bacteria into the surrounding medium, where they form a complex with ferric iron. The resultant iron-siderophore complex then binds to specific receptors on the bacterial surface to be taken up by the cell. The observation that patients suffering from iron overload have an increased susceptibility to severe Y. enterocolitica infections suggested that the availability of iron in tissues may determine the severity of the outcome of yersiniosis (241). Investigations into the relationship of yersiniae to iron have revealed that these bacteria employ a wide array of processes to acquire iron from inorganic and organic sources (224, 225). The fact that most clinical isolates of Y. enterocolitica do not produce siderophores accounts for their reliance on abnormally high

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

354 concentrations of iron for growth in mammalian tissue. Interestingly, however, the highly virulent biotype 1B strains, as well as Y. pestis and Y. pseudotuberculosis, harbor genes for the biosynthesis, transport, and regulation of a 482-Da, catechol-containing siderophore known as yersiniabactin. The ca. 40-kbp ybt locus that contains these genes has a higher G+C content (57.5 mol%) than that of the Y. enterocolitica chromosome (47 mol%), is flanked on one side by an asn tRNA gene, and carries the gene for a putative integrase (43). These features, which are typical of a pathogenicity island, have led to the ybt locus also being described as the Yersinia high-pathogenicity island (HPI). The designation “high” alludes to the observation that bacteria that carry this locus are more virulent for mice infected perorally (median lethal dose, <103 CFU) than strains that lack it (median lethal dose is typically >106 CFU). Interestingly, the Yersinia HPI is also present in a variety of other enterobacteria, such as E. coli, Klebsiella, Citrobacter, and Salmonella spp. Lateral transfer of the HPI can occur between selected strains of Y. pseu­ dotuberculosis and between Y. pseudotuberculosis and Y. pestis (171). The HPI of Y. enterocolitica serotype O:8 contains 22 open reading frames within 43.4 kb, approximately 30.5 kb of which are conserved between Y. enteroco­ litica, Y. pestis, and Y. pseudotuberculosis (Fig. 14.4) (232). Synthesis of yersiniabactin requires irp (iron-regulated protein) genes irp1 to irp5, as well as ybtA, which encodes an Ara-C like regulator, whereas irp6 and irp7 are required for utilization of the iron-yersiniabactin complex (35). The major receptor for this complex is a 65-kDa outer membrane protein, named FyuA, which also serves as a receptor for pesticin, a bacteriocin produced by Y. pestis (125). Transport of iron-yersiniabactin complexes across the cell wall of Y. enteroco­

litica resembles the analogous pathway in E. coli in that it is an energy-dependent process that requires TonB. The TonB protein couples energy provided by inner membrane metabolism to outer membrane protein receptors, such as FyuA. TonB-, FyuA-, and yersiniabactin-deficient mutants of Y. enterocolitica all have reduced virulence in mice, presumably because of their limited capacity to acquire sufficient iron to grow in tissues (17). The fact that aroA is also required for yersiniabactin synthesis may partly explain the attenuation of virulence of aroA mutants of Y. enterocolitica in mice (32). Although biotypes of Y. enterocolitica other than 1B do not produce yersiniabactin, they are able to acquire iron from a number of sources, including ironsiderophore complexes, in which the siderophore, such as desferrioxamine B, is synthesized by another microorganism (17, 240). The iron-desferrioxamine complex (known as ferrioxamine) binds to FoxA, a 76-kDa outer membrane protein, which shares 33% amino acid homology with FhuA, the high-affinity ferrichrome receptor of E. coli (18). The ability of Y. enterocolitica to acquire iron from ferrioxamine may have important clinical implications, because desferrioxamine B is used therapeutically to reduce iron overload in patients with hemosiderosis and other forms of iron intoxication. When administered to patients, desferrioxamine B forms an iron-siderophore complex, which Y. enterocolitica can utilize as a growth factor (239). Accordingly, if patients undergoing iron chelation therapy with desferrioxamine B become infected with Y. enterocolitica, the bacteria may be able to proliferate in tissues where under normal circumstances, the poor availability of iron would limit their growth. Apart from its effects on microbial iron metabolism, desferrioxamine B may also increase susceptibility to

Figure 14.4  A representation of the HPI of Y. enterocolitica O:8 strain WA-C. Arrows indicate the positions of the open reading frames and the direction of transcription. The region that is conserved in Y. pestis and Y. pseudotuberculosis is indicated by a double-headed arrow. (Adapted from reference 231.) doi:10.1128/9781555818463.ch14f4

SMP_Food Microbiology_CH14.indd

354

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

355

systemic yersiniosis by interfering with host immune responses (14).

host tissues, but the mechanism by which this occurs is not known (109).

The YSA pathogenicity island Strains of Y. enterocolitica biotype 1B, but not other yersiniae, possess another pathogenicity island, comprised of approximately 30 open reading frames, which encodes a type III secretion system (T3SS) known as the Yersinia secretion apparatus (Ysa) (341). This apparatus is distinct from the well-characterized pYV-encoded Ysc T3SS discussed below. The Ysa T3SS is required for the export of a set of 15 proteins called Yersinia-secreted proteins (Ysps) (183), including YspA, YspB, YspC, YspD, YspE, YspF, YspI, YspK, YspL, YspN, YspP, YspY, YopE, YopN, and YopP. The last three proteins are also secreted by the plasmid-encoded Ysc T3SS discussed below (340). Although the precise contribution of Ysps to virulence is not known, Ysa-defective mutants have a reduced ability to colonize the mouse intestine (313).

The Virulence Plasmid (pYV)

T2SS Strains of Y. enterocolitica biotype 1B also possess a type II protein secretion system (T2SS), known as Yts1, which is absent from less virulent strains of this species (148). The genes for Yts1 are located within a region of the genome, which encodes several virulence-associated factors, including the Ysa T3SS (148). Yts1 contributes to the virulence of biotype 1B Y. enterocolitica for mice but is not required for initial infection of Peyer’s patches (148). Three proteins are known to be secreted by this T2SS. They are ChiY, EngY (YE2830), and YE3650 (266). Both ChiY and EngY bind to chitin, and YE3650 has features of oligosaccharide-binding enzymes, but the way in which they contribute to virulence is not known. Genomic analysis of Y. enterocolitica biotype 1B strains has revealed a second T2SS, known as Yts2, which is present in all isolates of Y. enterocolitica, as well as in Y. pseudotuberculosis, Y. pestis, Y. bercovieri, Y. frederiksenii, and Y. intermedia (148). The role of Yts2 in virulence, if any, and the protein(s) secreted by this apparatus are not known. Urease All enteric pathogens must negotiate the acid barrier of the stomach to cause disease. In Y. enterocolitica, acid tolerance relies on the production of urease, which catalyzes the release of ammonia from urea and allows the bacteria to resist pH as low as 2.5 (23, 75, 342). Urease also contributes to the survival of Y. enterocolitica in

SMP_Food Microbiology_CH14.indd

355

All fully virulent strains of Y. enterocolitica biotypes 1B to 5 carry pYV, a ca. 70-kb plasmid, which is highly conserved among Y. enterocolitica, Y. pestis, and Y. pseu­ dotuberculosis (for reviews, see references 58 and 60). pYV has functions that interfere with innate immune response, such as phagocytosis, complement activation, and the production of proinflammatory cytokines, thus allowing yersiniae to proliferate extracellularly in tissues (Table 14.7). In addition, some factors encoded by pYV may act on T and B cells directly to modify adaptive immune responses. Yersiniae that carry pYV exhibit a distinctive phenotype, known as “calcium dependency” or “the low calcium response” because it manifests when the bacteria are grown in media containing low concentrations of Ca2+. The principal features of this response are the cessation of bacterial growth after one or two generations and the appearance of at least 12 new proteins (Yops) on the bacterial surface or in the culture medium. Yops are characterized by their common mode of secretion and their regulation by a pYV-encoded DNA-binding protein, known as VirF. Yops are so named because they were once thought to be outer membrane proteins, but they are now known to be secreted by the bacteria via the Ysc T3SS, encoded by pYV. The expression of pYV-encoded proteins in vitro imbues Y. enterocolitica with novel properties, such as autoagglutination, resistance to killing by human serum, and an ability to bind Congo red and crystal violet (31). Some of these characteristics have been exploited in the design of culture media, such as calcium-depleted agar containing Congo red (21, 87), to facilitate the isolation and identification of pYV-bearing yersiniae. Several pYV plasmids have been sequenced (for an example, see reference 280). The genes carried on these plasmids include those for (i) an outer membrane protein adhesin, YadA; (ii) the Ysc T3SS that transports Yops across the bacterial cell wall; (iii) at least six distinct antihost effector Yops; (iv) a translocation apparatus, comprising certain Yops, which the effector Yops require to gain access to the cytosol of host cells; and (v) factors that regulate Yop biosynthesis, secretion, and translocation (Table 14.7). Genes for the effector Yops are scattered around pYV, whereas those required for Yop secretion and translocation are clustered together (Fig. 14.5). Although Yops are highly conserved between Yersinia species, there is little homology between individual Yops. pYV also encodes YlpA, a 29-kDa

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

356

Table 14.7  Major pYV-encoded determinants of Y. enterocolitica and their role in virulence Determinant(s)

Contribution to virulence

YadA................................................................... Bacterial adhesion and invasion; reduction of opsonization   by interfering with binding of complement proteins Ysc complex (product of virC,           T3SS virG, virA, virB) Yops Effector Yops: YopH.......................................................... Protein tyrosine phosphatase, which interferes with   phagocytosis and other immunological responses by   dephosphorylating proteins in focal adhesion complexes YopE, YopT, YopO/YpkAa......................... Interfere with phagocytosis by disrupting Rho GTPases and the host cell cytoskeleton YopP/YopJa................................................. Reduces inflammation by suppressing signaling via MAPK and NF-kB YopM, YopQ/Ka......................................... Suppression of innate immune responses (?) YopB, YopD, LcrV...................................... Yop translocation; antihost effects; regulation of yop gene expression Syc proteins SycE, H, D, N, T............................. Yop chaperones VirF.................................................................... Regulation of expression of yop, virC, and yadA genes a

Designation in Y. pseudotuberculosis.

lipoprotein, related to the TraT protein of plasmids in various enterobacteria, and, in Y. enterocolitica strains of biotypes 2 to 5 (but not biotype 1B or Y. pseudotu­ berculosis), an operon that specifies resistance to arsenic (208).

YadA, a pYV-Encoded Adhesin and Antihost Factor

YadA, formerly known as Yop1 or P1, is a 44- to 47kDa outer membrane protein that belongs to the family of autotransporter proteins (127). These proteins have a characteristic arrangement of functional domains, including an N-terminal signal sequence, an internal passenger domain, and a C-terminal translocator domain. YadA represents a subfamily of autotransporter proteins, termed oligomeric coiled-coil adhesins or trimeric autotransporters because they possess a short trimeric translocator domain at their C terminus (64, 127). Individual YadA monomers aggregate in solution to form oligomers with an apparent molecular mass of around 200 kDa. On the bacterial surface, however, YadA forms trimers that appear as lollipop-shaped structures that envelop the entire outer membrane as a densely packed array (5). YadA mediates bacterial adhesion to intestinal mucus and to certain extracellular matrix proteins, including collagen, laminin, and cellular fibronectin (274). However, YadA proteins from Y. enterocolitica and Y. pseudotuberculosis show differences in the specificity of their binding to extracellular matrix proteins, which ac-

SMP_Food Microbiology_CH14.indd

356

counts for the observation that YadA from Y. pseudotu­ berculosis, but not that from Y. enterocolitica, promotes efficient uptake into mammalian human cells (84). Heise and Dersch (126) have identified a unique N-terminal amino acid sequence of YadA from Y. pseudotuber­ culosis that acts as an “uptake domain” by mediating tight binding to fibronectin bound to a5b1 integrin in a way that induces bacterial internalization. However, YadA from Y. enterocolitica promotes adhesion mainly to collagen and laminin and plays only a minor role in invasion. Both YadA and invasin induce cell signaling via b1 integrins to kinases, such as small GTPases and mitogenactivated protein kinases (MAPKs) (including p38, MEK1, and Jun N-terminal protein kinase [JNK]) leading to the production of interleukin-8 (IL-8) (259). IL-8 is a proinflammatory cytokine that promotes chemotaxis and activation of PMNs. Apart from its role as an adhesin and invasin, YadA also contributes to virulence by conveying resistance to complement-mediated opsonization. It achieves this by binding factor H and C4b-binding protein, thereby reducing deposition of complement on the bacterial surface (24, 161). As a result, YadA is associated with resistance of Y. enterocolitica to complement-mediated lysis and phagocytosis and an ability to inhibit the respiratory burst of PMNs (50, 51). Given the pluripotential capacity of YadA to enhance bacterial virulence, it is not surprising that YadA mutants of Y. enterocolitica and Y. pseudotubercu­ losis are attenuated for mice (223, 246). This is in contrast

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

357

Figure 14.5  Map of the virulence plasmid pYVe of Y. enterocolitica serogroup O:9 showing the location and direction of transcription (arrows) of the genes encoding (i) YadA; (ii) YlpA; (iii) Yops B, D, E, H, M, N, O, P, Q, T, and LcrV; (iv) specific Yop chaperones Syc D, E, H, and T; (v) secretion elements VirA, -B, -C, -G; and the regulatory element VirF (adapted from reference 139). doi:10.1128/9781555818463.ch14f5

to Y. pestis, which is extremely virulent for mice but is naturally defective in YadA production due to a singlebase-pair deletion resulting in a shift of the reading frame of the gene (116). This apparent paradox can be explained by the finding that YadA binds to collagen within neutrophil extracellular traps, which increases the susceptibility of YadA-bearing bacterial killing by PMNs (46).

Yop Secretion via the Ysc Secretion Apparatus

Secretion of Yops from yersiniae is mediated by the Ysc T3SS. Similar systems are utilized by several species of pathogenic bacteria to transfer proteins directly from their own cytoplasm to that of a host cell via a syringeand-­needle-like complex, known as an injectisome (102). T3SSs integrate transport across the inner and outer membranes in a temporally linked fashion (102). In contrast to proteins exported via the Sec-dependent, general secretory pathway, the N termini of T3SS proteins have no resemblance to each other and are not cleaved during export. Another feature of T3SS is the presence of structurally conserved chaperone proteins, which bind to individual

SMP_Food Microbiology_CH14.indd

357

secreted proteins and guide them to the secretion apparatus while preventing their premature interaction with other proteins in the bacterial cytosol and injectisome. Chaperone proteins for Yops are denoted by the prefix Syc (acronym for specific Yop chaperone) and include SycE (the chaperone for YopE), SycH (for YopH), SycD (for YopB and YopD), SycN (for YopN), and SycT (for YopT). The genes encoding these chaperones are located on pYV close to the corresponding yop gene (Fig. 14.5). Although they share no significant homology with each other, all Yop chaperones identified to date are low-­molecular-mass (14- to 19-kDa) proteins, with a C-terminal amphipathic a-helix and a pI of approximately 4.5 (60). The passage of proteins through the T3SS is tightly controlled to ensure that only selected proteins can exit the bacterial cell via this pathway. Nevertheless, the signal by which the T3SS recognizes proteins destined for export is not known. Evidence has been presented for two apparently contradictory scenarios: (i) that the secretion signal is contained within the noncleaved N terminus of the protein and (ii) that it is embodied within the 5¢ end of the mRNA encoding the protein (233). Although the

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

358 weight of current opinion favors the idea that the signal for initial secretion is in the protein, it is possible that secretion may be cotranslational during the later stages of infection (58). Whichever is correct, it is clear that the secretion signal is contained within the first 15 to 20 amino acids or codons of the secreted protein. The Ysc secretion apparatus itself is a paradigm of all T3SS machines (60). It comprises a base that spans the inner and outer membranes in a manner analogous to that of the basal body of flagella and a needle-like structure that protrudes from the cell and is reminiscent of the flagellar hook (Fig. 14.6). The 29 genes encoding the Ysc apparatus are contained within four contiguous loci, called virC (comprising yscABCDEFGHIJKLM), virG (which encodes YscW), virA (encoding YopN, TyeA, SycN, YscX, YscY, YscV, and YscR/LcrD), and

virB (comprising yscNOPQRSTU) (Fig. 14.5) (139). Ten Ysc proteins (YscD, -J, -L, -N, -Q, -R, -S, -T, -U, and -V) have counterparts in almost every T3SS apparatus, including the T3SS for flagella, from which the virulence-associated T3SSs probably evolved. YscD, -J, -R, -S, -T, -U, and -V appear to span the inner membrane (Fig. 14.6) (44). YscN has ATP-binding motifs (Walker boxes A and B) resembling the catalytic subunit of F0F1 proton translocase and related ATPases and is predicted to energize the secretory machine (333). The results of yeast two- and three-hybrid experiments suggest that YscN forms a complex with YscK, YscQ, and YscL (149). YscJ is a lipoprotein whose counterpart in enteropathogenic E. coli, EscJ, forms a large ring structure with extensive grooves and ridges. YscC also has a counterpart in most T3SS, apart from that for fla-

Figure 14.6  Schematic representation of Yop secretion and translocation by Y. ente­ rocolitica. The major structural proteins of the secretory apparatus are shown in relation to their known or deduced location in the cell wall. The effector Yop chaperone (Syc) and translocation pore comprising YopB and YopD are also depicted. Not to scale. doi:10.1128/9781555818463.ch14f6

SMP_Food Microbiology_CH14.indd

358

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

359

gella. It is an outer membrane protein of the secretin family and is envisaged to form a ring-like structure with an external diameter of 20 nm and a central pore of around 5 nm (162). It is stabilized in the outer membrane by YscW, a lipoprotein product of the virG gene (162). Several Ysc proteins, including YscO, -P, and -X, are secreted into the medium together with Yops. YscX binds to YscY and contributes to the regulation of secretion, whereas YscP determines the length of the needle. At least one Yop, YopN, appears to serve as a constituent protein of the T3SS rather than as one of its substrates. Its role is evidently to plug the secretory channel together with TyeA, a surface protein, which binds to YopN, to prevent the uncontrolled release of translocated Yops from the secretory channel (92). Electron microscopic studies have revealed that the distal portion of the Ysc T3SS comprises a “needle” formed by the polymerization of the 6-kDa YscF ­protein (133). Isolated needles have a length of approximately 60 to 80 nm, a width of 6 to 7 nm, and a hollow center of around 2 nm. Apart from its role in Yop protein secretion, YscF may also contribute to the cell-contactdependent regulation of the T3SS (300). Studies by Diepold et al. (79) have revealed that assembly of the Ysc apparatus is initiated by the formation of the YscC outer membrane ring and that it then progresses inwards through the inner membrane into the cytosol. Little is known about the actual mechanism of protein secretion via the T3SS, but it is envisioned that the Ysc apparatus serves as a hollow channel through which exported proteins traverse the inner membrane, peptidoglycan layer, and outer membrane in a single step. Whether proteins travel folded or unfolded is not known, but given the dimensions of the channel, it is likely that they travel, at least partially, unfolded (58).

Yop Translocation

The principal function of the Ysc T3SS is to inject (translocate) proteins into the cytosol of eukaryotic cells, where they act on specific targets (58, 261, 303, 316). Yops are classified into those that are translocated into host cells to exert an antihost action (effector Yops) and those that are primarily involved in the translocation process. There is evidence, however, that some Yops required for translocation may also serve as antihost effectors. The transport of Yops from the bacterial cytoplasm via Ysc and the translocation apparatus into the host cell cytosol is believed to occur in one step from bacteria that are closely bound to the host cell (169). Two of the proteins required for translocation are YopB (42 kDa) and YopD (33 kDa,) both of which have

SMP_Food Microbiology_CH14.indd

359

hydrophobic domains, suggesting that they interact directly with host cell membranes (114, 207). YopB resembles members of the RTX toxin family and can form a pore approximately 2 nm in diameter in the plasma membrane of eukaryotic cells (207). YopD associates with YopB to contribute to the formation of the pore (207). Another effect of YopD is to negatively regulate Yop synthesis and secretion (332, 335). YopB also has additional functions, including that of stimulating proinflammatory signaling in host cells (318). YopB and YopD are encoded by the lcrGV-sycDyopBD operon, which also encodes LcrV, LcrG, and SycD, the chaperone for YopB and YopD (Fig. 14.5). LcrG is a negative regulator of Yop secretion. This effect is negated when LcrG binds to LcrV (210). LcrV, also known as the V antigen, is a versatile Yop that is involved in regulating the expression of Ysc and Yops (58). Unlike other Yops, LcrV is secreted into the extracellular milieu in significant concentrations by Y. pestis, at least, both in vitro and in vivo (66). Some LcrV coats the bacterial surface, while free LcrV can exert profound immunological effects: inhibiting chemotaxis of PMNs and stimulating the release of the immunosuppressive cytokine IL-10 (66). Secreted LcrV also binds to the tip of the needle complex, where it assists YopB and YopD in forming the translocation pore (197). LcrV is not part of the pore itself, however, because it is hydrophilic, whereas both YopB and YopD are hydrophobic. The key role of LcrV in virulence is evidenced by the fact that antibodies to LcrV protect mice from infection by all three pathogenic Yersinia species (33, 66, 69). The mechanism of action of anti-LcrV antibodies is uncertain, but there is evidence that they opsonize the bacteria in preparation for phagocytosis by PMNs and that they interfere with formation of the translocation pore (66). The protective role of anti-LcrV antibodies has created considerable interest in LcrV as a potential vaccine for plague and the enteropathogenic yersiniae (63).

Mechanism of Action of the Effector Yops

The purpose of the Ysc T3SS is to transfer the six effector Yops, i.e., YopH, YopM, YopE, YopT, YopO/YpkA, and YopP/J, from yersiniae into host cells (60). The collective role of these Yops is to overcome the host’s innate immune response, which is the first line of defense against invasive pathogens. Five of the effector Yops are enzymes (261, 303, 316). The exception is YopM, whose mechanism of action is not entirely known. The collective action of the effector Yops is to inhibit bacterial uptake and killing by phagocytic cells, inhibit

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

360 cytokine production, and induce apoptosis, hence enabling yersiniae to persist extracellularly in tissues. Yops achieve these outcomes by interfering with invasin- and YadA-induced rearrangements of the actin cytoskeleton required for phagocytosis and by disrupting proinflammatory signaling pathways that are activated in response to stimulation by LPS and other pathogen-associated molecular patterns. Four Yops, YopH, YopE, YopT, and YopO/YpkA, act synergistically to prevent phagocytosis (110). YopH (51 kDa) is a potent protein tyrosine phosphatase. In HeLa cells, YopH dephosphorylates Fak and the proteins paxillin and p130Cas, which form the focal adhesion complex (226). Another target of YopH is a second adhesion complex, comprised of Syc-phosphorylated p130Cas, the adaptor protein Crk, and Pyk2, another tyrosine-phosphorylated kinase (41). Both of these adhesion complexes signal integrin-mediated activation of actin polymerization via Rac1 (316). By targeting these and other complexes involved in b1-integrin-mediated endocytosis, YopH interferes with the uptake of yersiniae by epithelial cells and macrophages. YopH also affects other immune response pathways by suppressing the oxidative burst in macrophages and neutrophils, by preventing the production of macrophage chemoattractant protein 1 (MCP-1), by counteracting T and B cell activation by antigens, and by reducing the production of IL-2 by T cells (316). YopE is a 25-kDa GTPase-activating protein (261, 303). It targets small RhoA-like G-proteins, such as RhoA, Rac1, and Cdc42, all of which are GTPases that control specific cytoskeletal elements in host cells (316). YopE alters the conformation of these proteins by inducing their GTPase activity, hence converting them from an active GTP-bound state to an inactive GDP-bound state, which impairs their signaling function (29). This leads to disruption of the actin cytoskeleton, which manifests as the rounding and detachment of affected cells in culture. Through its action on Rho GTPases, YopE also appears to inhibit the production of pro­inflammatory cytokines by epithelial cells infected with yersiniae (318). In addition to mimicking a host enzyme, YopE (like YopT) evidently plays a role in maintaining cell integrity after the Yop effectors have been injected. Cells infected by wild-type Yersinia die as a result of apoptosis, which does not cause the release of cellular contents. However, cells infected with a YopE-deficient strain of Y. pseu­ dotuberculosis take up extracellular molecules and release cytosolic proteins, such as lactate dehydrogenase (LDH) (315). These results suggest that activation of Rho GTPase is required for pore formation in infected

SMP_Food Microbiology_CH14.indd

360

host cells and that inhibition of Rho GTPases by YopE and YopT counteracts this. YopE may also play a role in regulating how much effector is delivered into a target cell, because infection with strains that lack YopE results in a “hypertranslocation” phenotype and a failure to decrease synthesis of effectors in the presence of host cells (7). YopT resembles YopE in targeting Rho GTPases, but its action depends upon cysteine protease activity, which removes a lipid modification from RhoA, Rac1, and Cdc42 that causes the GTPases to dissociate from the membrane (262, 263). The morphological changes induced by YopT on epithelial cells resemble those caused by YopE. YopT is not produced by serotype O:3 strains of Y. enterocolitica, and YopT knockout mutants of serotype O:8 Y. enterocolitica and Y. pseu­ dotuberculosis are no less virulent that the wild-type strains; hence, the contribution of YopT to virulence is uncertain (316). However, there is evidence that YopT can fulfill some functions of YopE when the latter is absent (317). YopO is an 82-kDa protein, which is known as YpkA (Yersinia protein kinase A) in Y. pseudotuber­ culosis. YpkA mutants of Y. pseudotuberculosis are avirulent for mice, indicating that YpkA is an essential virulence factor. YopO/YpkA is a serine/threonine kinase which induces rounding of HeLa cells without causing them to detach. Interestingly, mutational analysis of YpkA has revealed that its effect on epithelial cells is unrelated to its kinase activity. Further studies revealed that the C terminus of YpkA, which does not contain the kinase domain, interacts with host RhoA/Rac but not with Cdc42. This region of YpkA inhibits nucleotide exchange in RhoA/Rac by mimicking the host Rho GDP dissociation inhibitor, RhoGDI (228). Nevertheless, the kinase activity of YpkA is also required for virulence (331). Its target is the G protein, Gaq, which it phosphorylates on S47 in the conserved N-terminal, diphosphate-binding region of Ga, hence impairing activation of Gaq (201). As Gaq controls RhoA-mediated actin stress fiber formation, its inactivation leads to cytoskeletal changes that are independent of the effect of YpkA on RhoA/Rac. Together, these findings indicate that YpkA is a bifunctional effector, with an N-terminal domain that phosphorylates Ga and a C-terminal domain that acts directly on the small G proteins RhoA/Rac (261). Another intriguing feature of YpkA is that it is produced by yersiniae as an inactive kinase, which becomes activated only after it has bound to monomeric G actin in host cells (301). The contribution of YpkA to virulence is evidently due to its inhibi-

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

361

tory action on phagocytosis and its capacity to induce apoptosis of epithelial and immune cells (219). YopP (known as YopJ in Y. pseudotuberculosis) is a 32-kDa protein that is encoded by the same operon as YopO/YpkA. YoP/J is a cysteine protease that downregulates inflammatory responses by inhibiting signaling pathways that are controlled by MAPK and NF-kB. YopP/J binds directly to MAPK kinases and prevents their activation (213). The action of YopP/J on NF-kB is due to its ability to bind to and inhibit IkB kinase b (IKKb), which is required for the activation of NF-kB (146, 213). The resultant inhibition of NF-kB, a transcription factor of central importance in activating inflammatory responses, has profound effects on cytokine production and induces macrophage apoptosis (343). The enzymic activity of YopP/J appears to involve the removal of ubiquitin or ubiquitin-like modifications from proteins (213), but it is not clear how this relates to its effects of MAPK, as ubiquitination is not known to play a direct role in MAPK signaling. The finding that YopP/J has a second catalytic activity that causes the acetylation of Ser/Thr residues on MAPK kinase and IKKb and prevents their activation (193, 198) provides an explanation for the action of YopP/J on multiple pathways involving MAPK and NF-kB (261). YopM is produced by all pYV-bearing Yersinia species, but its size and sequence differ in different strains and serotypes. It is the only effector Yop that does not exhibit any enzymic activity. YopM protein contains from 12 to 20 repeats of a 19-residue leucine-rich repeat motifs (172). These motifs are common modules of protein-protein interaction in eukaryotic proteins and are also present in some virulence-associated proteins of other pathogenic bacteria, including Salmonella and Shigella (261). After YopM is translocated into cells, it traffics to the nucleus by means of a vesicle-associated, microtubule­dependent pathway (270). YopM is the only Yop that enters the nucleus, but its intranuclear role is not known. YopM also forms a complex with two cytoplasmic kinases, RSK1 and PRK2/PKN2, and activates them (184). Although the precise mechanism of action of YopM is unknown, its contribution to virulence is evident from the finding that yopM mutants of yersiniae are significantly attenuated. This may be partly due to the role that YopM plays in depleting NK cells and the proinflammatory cytokines IL-12, IL-18, and gamma interferon (158). YopK (known as YopQ in Y. enterocolitica) has long been recognized as an essential virulence determinant of Y. pseudotuberculosis (135). YopQ/K is translocated

SMP_Food Microbiology_CH14.indd

361

into host cells, but it was thought that its only role in virulence was to regulate the T3SS by limiting the quantity of effector Yops that can enter target cells (78, 134). Recently, however, Brodsky et al. (39) reported that YopK also performs an antihost function in that it prevents recognition of the T3SS by inflammasomes. This would enhance bacterial survival in tissues, because the inflammasome activation is necessary for activation of caspase-1 and host defense (93). Confirmation of these findings would indicate that YopQ/K is a previously unrecognized Yop effector of Yersinia species.

The Antihost Action of pYV

Invading bacteria carry identifiable markers, known as pathogen-associated molecular patterns, which are recognized by specific protein receptors, termed pattern recognition receptors, on host cells. One of the key responses to bacterial invasion is the mobilization and activation of macrophages to ingest and kill the invaders. This involves upregulation of a number of ­pathways leading to the rearrangement of the actin cytoskeleton and the production of proinflammatory cytokines, such as interleukins and tumor necrosis factor (TNF). Cytoskeletal rearrangements are controlled by the Rho family of small GTPases, including RhoA, Rac, and Cdc42, whereas cytokine production is regulated by signal transduction pathways, such as the MAPK pathway, and signaling via NF-kB. As discussed above, several of the Yop effectors encoded by pYV act by subverting these responses. Strains of Y. enterocolitica that have been “cured” of pYV can colonize the intestinal tract of experimental animals, penetrate Peyer’s patches, and even travel to the mesenteric lymph nodes, indicating that pYV-encoded determinants are not required during the early stages of infection (229). After a period, however, pYV-negative strains are eliminated from the body, whereas pYVbearing strains proliferate and spread to other tissues. Macrophages and PMNs are major participants in the first line of defense against invading microbes by ingesting and destroying bacteria, producing and releasing proinflammatory cytokines, and priming T and B cells. When Y. enterocolitica first penetrates the gut epithelium, before pYV-encoded proteins are expressed, the bacteria may be taken up by macrophages. They are able to survive within these cells, possibly due to the combined actions of OmpR and GrsA (229). OmpR is a regulator that mediates changes in gene expression in response to changes in osmolarity, whereas GrsA (also known as HtrA or DegP) is a heat shock-induced serine protease. Although Yops are not synthesized while the bacteria are located within host cells (150), yersiniae that survive their initial ­ encounter

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

362 with phagocytes and escape from these cells will express and translocate Yops when they reestablish contact with phagocytes. The translocated Yops then inhibit bacterial uptake, impair the respiratory burst, interfere with the production and release of several proinflammatory cytokines, and induce apoptosis. If pYV-bearing bacteria are preopsonized by antibodies, resistance to phagocytosis is mediated largely by YadA, whereas unopsonized bacteria resist ingestion by phagocytes due to the action of YopH, acting in concert with YopE, YopO, and YopT. YopH, -E, -O, and -T are also involved in inhibition of the respiratory burst. In concert with YopP/J, some of the effectors also contribute to the induction of apoptosis and the inhibition of cytokine release. pYV-bearing strains of Y. enterocolitica also resist ingestion and killing by neutrophils, whereas plasmidless strains are killed (50). The principal virulence determinants involved in this process are YadA, which interferes with opsonization, and YopH and YopE, which inhibit the respiratory burst and retard phagocytosis (51). Should pYV-bearing bacteria be ingested by PMNs, they can resist killing due to their relatively low susceptibility to the antimicrobial peptides produced by these cells. This resistance is attributable to YadA and smooth LPS (275).

Regulation of Yop Production and Secretion

All Yops and YlpA are produced in vitro at 37°C (but not below 30°C) when the concentration of Ca2+ is sufficiently low to induce bacteriostasis. The mechanism of regulation by temperature involves VirF and DNA supercoiling (247). VirF is a pYV-encoded, DNA-binding protein of the AraC family of transcriptional regulators and is active only at 37°C (55). Other members of this family include VirF from Shigella flexneri, Rns from enterotoxigenic E. coli, ToxT from V. cholerae, and RegA from Citrobacter rodentium, all of which are involved in the regulation of expression of virulence in their respective bacteria (337). VirF plays a central role in the virulence of Y. entero­ colitica by governing the transcriptional activation of yadA, ylpA, all of the yop genes, and the virC operon. Because these genes are coregulated, they have been named the Yop regulon (59). Transcription of virF itself is also regulated by temperature. In vitro transcription of the yop genes, but not the ysc genes, yadA, or virF, is repressed by Ca2+ or by mutations in the secretion apparatus, Ysc. Repression due to mutations in Ysc is due to feedback inhibition from the closed secretion apparatus (44). Repression by Ca2+ may occur in a similar manner, as supported by the observation that Ca2+ affects the interaction between YopN, TyeA, and LcrG, permitting them to bind together and plug the Ysc channel (89).

SMP_Food Microbiology_CH14.indd

362

There is no doubt that Yops are produced in vivo because animals and humans infected with virulent Yersinia species develop antibodies to these proteins during the course of infection (Fig. 14.7). Although host temperature (37°C) is likely to be a key stimulus for the production of Yops in vivo, the signal equating to low Ca2+ is not known. Forsberg et al. (91) have determined that extracellular yersiniae produce Yops and are cytotoxic for HeLa cells even in the presence of millimolar concentrations of Ca2+, provided the bacteria can attach to the target cells. This observation indicates that in tissues, contact between Y. enterocolitica and host cells permits Yop release in the same way that Ca2+ depletion does in vitro.

Pathogenesis of Yersinia-Induced Autoimmunity Arthritis Following an acute infection with pYV-bearing Y. en­ terocolitica, some patients develop autoimmune (reactive) arthritis. A similar syndrome may also occur after infections with Campylobacter, Salmonella, Shigella, or Chlamydia species (236, 310). The overwhelming majority of individuals with postinfective reactive arthritis caused by Y. enterocolitica carry the class I human leukocyte antigen HLA-B27 (267). In addition, HLA-B27positive individuals have more severe arthritic symptoms and a more prolonged course than individuals with different HLA antigens. The pathogenesis of reactive arthritis is poorly understood (310). The synovial fluid from affected joints of patients with Yersinia-induced arthritis is culture negative but generally contains bacterial antigens, such as LPS and heat shock proteins, within inflammatory cells (101, 104). The presence of these antigens may reflect an impaired ability of HLA-B27-bearing phagocytes to eliminate them. LPS in synovial fluid may stimulate the local production of TNF-a, which plays a key role in the pathogenesis of reactive arthritis (236). Explanations for the link between yersiniosis and ­autoimmunity include antigen persistence, molecular mimicry, impaired immune responsiveness, and infection-induced presentation of normally cryptic cellular antigens (128, 297, 310). Although patients with reactive arthritis display higher levels of serum immunoglobulin A antibodies to Yersinia antigens than individuals without arthritis, these antibodies are unlikely to contribute to the development of arthritis. Instead, they probably reflect enhanced stimulation of the mucosal immune system due to the persistence of bacterial antigens in the intestine or other tissues (73).

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

363

Figure 14.7  Antibody response of sheep infected with Y. enterocolitica or Y. pseudotu­ berculosis to Yops. Yops were prepared from Y. enterocolitica serogroup O:3, separated by polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, and subjected to reaction with preimmune (lanes 1 and 3) or immune (lanes 2 and 4) sera from lambs with naturally acquired infection with pYV-bearing Y. enterocolitica (lanes 1 and 2) or Y. pseudotuberculosis (lanes 3 and 4). (Reprinted with permission from reference 237.) doi:10.1128/9781555818463.ch14f7

The observation that rats transgenic for HLA-B27 show a weaker cytotoxic T-cell response against Y. pseudotuberculosis than that of nontransgenic syngeneic rats provides a link between prolonged infection, antigen persistence, and HLA-B27 (86). Furthermore, HLA-B27-expressing monocytes have an impaired capacity to limit the intracellular replication of salmonellae (310). Support for the molecular mimicry hypothesis comes from the demonstration of autoreactive T cells in the synovial tissue of patients with reactive arthritis (129). Among the bacterial antigens that may provoke autoreactivity are heat shock proteins, which are somewhat conserved among bacteria and mammals and share a number of antigenic determinants. Accordingly, an immune response to selected epitopes on bacterial heat shock proteins may lead to an autoimmune response at sites where bacterial antigens accumulate, including in the joints of patients with a predisposition to reactive arthritis. In keeping with this suggestion is the observation that synovial fluid from patients with reactive ar-

SMP_Food Microbiology_CH14.indd

363

thritis may contain CD4+ major histocompatibility class II restricted T lymphocytes, which recognize epitopes that are shared by a 60-kDa Yersinia heat shock protein and its human counterpart (129). Further evidence for the molecular mimicry hypothesis comes from the finding that an immunodominant epitope from GroEL, a heat shock protein of Salmonella enterica serovar Typhimurium (which is almost identical to the homologous protein in Y. enterocolitica), can be presented by a class I major histocompatibility antigen in mice and can then be recognized by CD8+ cytotoxic T-cells that crossreact with a peptide from a murine heat shock protein (176). Other bacterial antigens that may contribute to autoimmunity via specific interactions with the immune system are YadA and the b subunit of urease (108, 272). However, early suggestions that YadA shared epitopes with the peptide-binding groove of the HLA-B27 antigen have been discounted (165). The subunit of urease is a cationic protein that (i) is recognized by CD4+ T cells from patients with Yersinia-induced reactive arthritis

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

364 and (ii) induces arthritis when injected into the joints of rats (272). The finding that a urease-deficient mutant of Y. enterocolitica retains its capacity to induce arthritis in a rat model, however, casts doubt on the role of urease in this condition (109). There is also a suggestion that HLA-B27 itself may serve as an autoantigen in the pathogenesis of arthritis (236). Y. enterocolitica and Y. pseudotuberculosis may also induce polyclonal T-cell stimulation by virtue of their ability to secrete toxins that resemble superantigens (1, 283). The best characterized of these are the Y. pseudotubercu­ losis-derived mitogen YPMa and its variants, YPMb and YPMc, which are 14- to 15-kDa proteins produced by approximately 20% of Y. pseudotuberculosis strains (81). These proteins activate human T cells of Vb phenotypes 3, 9, 31.1, and 13.2 (45, 194). Their structure is unlike that of other bacterial superantigens and most closely resembles that of virus capsid proteins and members of the TNF superfamily (81). The Y. enterocolitica superantigen is poorly characterized but evidently can stimulate T cells with Vb phenotypes 3, 7, 8.1, 9, and 11 (283). Yersiniae may also provoke nonspecific immune stimulation when invasin binds to b1 integrins on T lymphocytes, hence providing a costimulatory signal to these cells (37).

Thyroid Diseases Y. enterocolitica has been implicated in the etiology of various thyroid disorders, including autoimmune thyroiditis and Graves’ disease hyperthyroidism (297, 298). The latter is an immunologic disorder mediated by autoantibodies to the thyrotropin receptor. The chief link between Y. enterocolitica and thyroid diseases is that patients with these disorders frequently have elevated titers of serum agglutinins to Y. enterocolitica O:3 (265). There is no clear relationship between the incidence or geographic distribution of yersiniosis and that of autoimmune thyroid diseases; hence, the presence of circulating antibodies to Y. enterocolitica in such patients is likely to reflect a fortuitous cross-reaction between Yersinia and thyroid antigens rather than a causal relationship (297). In addition, followup of patients many years after infection with Y. entero­ colitica or Y. pseudotuberculosis has shown no increased frequency of thyroid disease, nor is there any evidence that hyperthyroidism is exacerbated by infection with Y. en­ terocolitica (297). Two published studies that investigated the relationship between yersiniosis and Graves’ disease in twins reached opposing conclusions (38, 118). Although few studies into the pathogenesis of Yersiniainduced erythema nodosum have been ­conducted, experience with other infective agents indicates that this disorder is caused by the deposition of immune complexes in affected organs.

SMP_Food Microbiology_CH14.indd

364

SUMMARY AND CONCLUSIONS The enteropathogenic Yersinia strains are versatile foodborne pathogens with a remarkable ability to adapt to a wide range of environments within and outside their mammalian hosts. Y. enterocolitica and Y. pseudotu­ berculosis typically access their hosts via food or water in which they will have grown to stationary phase at ambient temperature. Under these circumstances, Y. enterocolitica expresses factors such as urease, flagella, and smooth LPS, which facilitate its passage through the stomach and the mucus layer of the small intestine. Bacteria in this state may also carry Myf fibrillae and invasin, which may promote adherence to and penetration of the dome epithelium overlying the Peyer’s patches. The higher infectivity of Y. enterocolitica when grown at ambient temperature compared with 37°C may account for the small number of reports of human-tohuman transmission of yersiniosis (209). Once Y. enterocolitica begins to replicate in the body at 37°C, LPS becomes rough, and Ail and YadA appear on the bacterial surface. These factors may promote further invasion while protecting the bacteria from ­complementmediated opsonization. After a period in the host, which may include some time spent within macrophages, yersiniae make contact with host cells in lymphoid tissue, where they synthesize and translocate the effector Yops YopH, YopE, YopT, YopO/YpkA, and YopP/J, which further frustrate the efforts of phagocytes to ingest and remove them. Subsequent bacterial replication may lead to tissue damage and the formation of microabscesses. If strains of Y. pseudotuberculosis and Y. enterocolitica that harbor the HPI gain access to tissues where iron supplies would normally be growth limiting, they may produce yersiniabactin, which will enable them to acquire iron and allow bacterial replication to proceed. Eventually, the cycle is completed when the bacteria rupture through microabscesses in intestinal crypts to reenter the intestine and regain access to the environment. This well-defined life cycle of enteropathogenic yersiniae, with its distinctive temperature-­induced phases, is reminiscent of the flea-rat-flea cycle of Y. pestis. Although much remains to be learned about Y. en­ terocolitica, investigations into the pathogenesis of yersiniosis to date have provided fascinating new insights into bacterial pathogenesis as a whole and its genetic control. Y. enterocolitica and Y. pseudotuberculosis were the first invasive human pathogens in which plasmid-mediated virulence was documented (344), from which internalins (invasin, Ail, and YadA) were cloned and characterized (142, 192), in which the relationship between iron limitation and iron-­siderophore uptake

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

365

assumed clinical significance (240), in which type III protein secretion was discovered (187), and in which a virulence-related, eukaryotic-like Ser/Thr kinase, YopO/ YpkA, was first identified (100). Future research in this area will no doubt lead to new and unexpected discoveries of bacterial strategies to evade immune responses that will further advance our understanding of the interface between microbes and their hosts.

References Abe, J., M. Onimaru, S. Matsumoto, S. Noma, K. Baba, Y. Ito, T. Kohsaka, and T. Takeda. 1997. Clinical role for a superantigen in Yersinia pseudotuberculosis infection. J. Clin. Investig. 99:1823–1830. 2. Achtman, M., K. Zurth, G. Morelli, G. Torrea, A. Guiyoule, and E. Carniel. 1999. Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc. Natl. Acad. Sci. USA 96:14043–14048. 3. Ackers, M. L., S. Schoenfeld, J. Markman, M. G. Smith, M. A. Nicholson, W. DeWitt, D. N. Cameron, P. M. Griffin, and L. Slutsker. 2000. An outbreak of Yersinia enterocolitica O:8 infections associated with pasteurized milk. J. Infect. Dis. 181:1834–1837. 4. Adams, M. R., C. L. Little, and M. C. Easter. 1991. Modelling the effect of pH, acidulant and temperature on the growth rate of Yersinia enterocolitica. J. Appl. Bacteriol. 71:65–71. 5. Aepfelbacher, M., R. Zumbihl, K. Ruckdeschel, C. A. Jacobi, C. Barz, and J. Heesemann. 1999. The tranquilizing injection of Yersinia proteins: a pathogen’s strategy to resist host defense. Biol. Chem. 380:795–802. 6. Ahvonen, P., K. Sievers, and K. Aho. 1969. Arthritis associated with Yersinia enterocolitica. Acta Rheumatol. Scand. 15:232–255. 7. Aili, M., E. L. Isaksson, S. E. Carlsson, H. Wolf-Watz, R. Rosqvist, and M. S. Francis. 2008. Regulation of Yersinia Yop-effector delivery by translocated YopE. Int. J. Med. Microbiol. 298:183–192. 8. Aimoto, S., T. Takao, Y. Shimonishi, S. Hara, T. Takeda, Y. Takeda, and T. Miwatani. 1982. Aminoacid sequence of a heat-stable enterotoxin produced by human enterotoxigenic Escherichia coli. Eur. J. Biochem. 129:257–263. 9. Aleksic, S., G. Suchan, J. Bockemuhl, and V. Aleksic. 1991. An extended antigenic scheme for Yersinia pseudotuberculosis. Contrib. Microbiol. Immunol. 12:235–238. 10. Andersen, J. K., R. Sorensen, and M. Glensbjerg. 1991. Aspects of the epidemiology of Yersinia enterocolitica: a review. Int. J. Food Microbiol. 13:231–237. 11. Annamalai, T., and K. Venkitanarayanan. 2005. Expression of major cold shock proteins and genes by Yersinia enterocolitica in synthetic medium and foods. J. Food Prot. 68:2454–2458. 12. Anonymous. 1983. Yersiniosis, p. 1–31. World Health Organization, Copenhagen, Denmark. 1.

SMP_Food Microbiology_CH14.indd

365

13. Arduino, M. J., L. A. Bland, M. A. Tipple, S. M. Aguero, M. S. Favero, and W. R. Jarvis. 1989. Growth and endotoxin production of Yersinia enterocolitica and Enterobacter agglomerans in packed erythrocytes. J. Clin. Microbiol. 27:1483–1485. 14. Autenrieth, I. B., R. Reissbrodt, E. Saken, R. Berner, U. Vogel, W. Rabsch, and J. Heesemann. 1994. Desferrioxamine-promoted virulence of Yersinia entero­ colitica in mice depends on both desferrioxamine type and mouse strain. J. Infect. Dis. 169:562–567. 15. Badger, J. L., and V. L. Miller. 1995. Role of RpoS in survival of Yersinia enterocolitica to a variety of environmental stresses. J. Bacteriol. 177:5370–5373. 16. Badger, J. L., and V. L. Miller. 1998. Expression of invasin and motility are coordinately regulated in Yersinia enterocolitica. J. Bacteriol. 180:793–800. 17. Baumler, A., R. Koebnik, I. Stojiljkovic, J. Heesemann, V. Braun, and K. Hantke. 1993. Survey on newly characterized iron uptake systems of Yersinia enterocolitica. Int. J. Med. Microbiol. Virol. Parasitol. Infect. Dis. 278:416–424. 18. Baumler, A. J., and K. Hantke. 1992. Ferrioxamine uptake in Yersinia enterocolitica: characterization of the receptor protein FoxA. Mol. Microbiol. 6:1309–1321. 19. Bengoechea, J. A., H. Najdenski, and M. Skurnik. 2004. Lipopolysaccharide O antigen status of Yersinia entero­ colitica O:8 is essential for virulence and absence of O antigen affects the expression of other Yersinia virulence factors. Mol. Microbiol. 52:451–469. 20. Bercovier, H., and H. H. Mollaret. 1984. Genus XIV. Yersinia Van Loghem 1944, 15AL, p. 498–506. In N. R. Krieg and J. G. Holt (ed.), Bergey’s Manual of Systematic Bacteriology, vol. 1. Williams & Wilkins, Baltimore, MD. 21. Bhaduri, S., B. Cottrell, and A. R. Pickard. 1997. Use of a single procedure for selective enrichment, isolation, and identification of plasmid-bearing virulent Yersinia enterocolitica of various serotypes from pork samples. Appl. Environ. Microbiol. 63:1657–1660. 22. Bhaduri, S., and J. G. Phillips. 2011. Growth model of a plasmid-bearing virulent strain of Yersinia pseudotu­ berculosis in raw ground beef. Zoonoses Public Health 58:77–84. doi:10.1111/j.1863-2378.20009.0127.x. 23. Bhagat, N., and J. S. Virdi. 2009. Molecular and biochemical characterization of urease and survival of Yersinia enterocolitica biovar 1A in acidic pH in vitro. BMC Microbiol. 9:262. 24. Biedzka-Sarek, M., S. Salmenlinna, M. Gruber, A. N. Lupas, S. Meri, and M. Skurnik. 2008. Functional mapping of YadA- and Ail-mediated binding of human factor H to Yersinia enterocolitica serotype O:3. Infect. Immun. 76:5016–5027. 25. Biedzka-Sarek, M., R. Venho, and M. Skurnik. 2005. Role of YadA, Ail, and lipopolysaccharide in serum resistance of Yersinia enterocolitica serotype O:3. Infect. Immun. 73:2232–2244. 26. Black, R. E., R. J. Jackson, T. Tsai, M. Medvesky, M. Shayegani, J. C. Feeley, K. I. E. MacLeod, and A. M. Wakelee. 1978. Epidemic Yersinia enterocolitica

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

366

27.

28. 29. 30. 31. 32.

33.

34. 35.

36.

37.

38.

39.

40. 41.

i­nfection due to contaminated chocolate milk. N. Engl. J. Med. 298:76–79. Bleves, S., M. N. Marenne, G. Detry, and G. R. Cornelis. 2002. Up-regulation of the Yersinia enterocolitica yop regulon by deletion of the flagellum master operon flhDC. J. Bacteriol. 184:3214–3223. Borg, A. A., J. Gray, and P. T. Dawes. 1992. Yersiniarelated arthritis in the United Kingdom. A report of 12 cases and review of the literature. Q. J. Med. 84:575–582. Bos, J. L., H. Rehmann, and A. Wittinghofer. 2007. GEFs and GAPs: critical elements in the control of small G proteins. Cell 129:865–877. Bottone, E. J. 1977. Yersinia enterocolitica: a panoramic view of a charismatic microorganism. Crit. Rev. Microbiol. 5:211–241. Bottone, E. J. 1997. Yersinia enterocolitica: the charisma continues. Clin. Microbiol. Rev. 10:257–276. Bowe, F., P. O’Gaora, D. Maskell, M. Cafferkey, and G. Dougan. 1989. Virulence, persistence, and immunogenicity of Yersinia enterocolitica O:8 aroA mutants. Infect. Immun. 57:3234–3236. Branger, C. G., A. Torres-Escobar, W. Sun, R. Perry, J. Fetherston, K. L. Roland, and R. Curtiss III. 2009. Oral vaccination with LcrV from Yersinia pestis KIM delivered by live attenuated Salmonella enterica serovar Typhimurium elicits a protective immune response against challenge with Yersinia pseudotuberculosis and Yersinia enterocolitica. Vaccine 27:5363–5370. Brecher, M. E., and S. N. Hay. 2005. Bacterial contamination of blood components. Clin. Microbiol. Rev. 18:195–204. Brem, D., C. Pelludat, A. Rakin, C. A. Jacobi, and J. Heesemann. 2001. Functional analysis of yersiniabactin transport genes of Yersinia enterocolitica. Microbiology 147:1115–1127. Bresolin, G., K. Neuhaus, S. Scherer, and T. M. Fuchs. 2006. Transcriptional analysis of long-term adaptation of Yersinia enterocolitica to low-temperature growth. J. Bacteriol. 188:2945–2958. Brett, S. J., A. V. Mazurov, I. G. Charles, and J. P. Tite. 1993. The invasin protein of Yersinia spp. provides costimulatory activity to human T cells through interaction with beta 1 integrins. Eur. J. Immunol. 23:1608–1614. Brix, T. H., P. S. Hansen, L. Hegedus, and B. E. Wenzel. 2008. Too early to dismiss Yersinia enterocolitica infection in the aetiology of Graves’ disease: evidence from a twin case-control study. Clin. Endocrinol. (Oxford) 69:491–496. Brodsky, I. E., N. W. Palm, S. Sadanand, M. B. Ryndak, F. S. Sutterwala, R. A. Flavell, J. B. Bliska, and R. Medzhitov. 2010. A Yersinia effector protein promotes virulence by preventing inflammasome recognition of the type III secretion system. Cell Host Microbe 7:376–387. Brubaker, R. R. 1991. Factors promoting acute and chronic diseases caused by yersiniae. Clin. Microbiol. Rev. 4:309–324. Bruce-Staskal, P. J., C. L. Weidow, J. J. Gibson, and A. H. Bouton. 2002. Cas, Fak and Pyk2 function in

SMP_Food Microbiology_CH14.indd

366

42.

43. 44.

45.

46.

47.

48.

49. 50.

51.

52.

53. 54.

55.

­diverse signaling cascades to promote Yersinia uptake. J. Cell Sci. 115:2689–2700. Butler, R. C., V. Lund, and D. A. Carlson. 1987. Susceptibility of Campylobacter jejuni and Yersinia en­ terocolitica to UV radiation. Appl. Environ. Microbiol. 53:375–378. Carniel, E. 2001. The Yersinia high-pathogenicity island: an iron-uptake island. Microbes Infect. 3:561–569. Carniel, E., I. Autenrieth, G. Cornelis, H. Fukushima, F. Guinet, R. Isberg, J. Pham, M. Prentice, M. Simonet, M. Skurnik, and G. Wauters. 2006. Y. enteroco­ litica and Y. pseudotuberculosis, p. 270–398. In M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt (ed.), The Prokaryotes, vol. 6, 3rd ed. Springer, New York, NY. Carnoy, C., H. Müeller-Alouf, S. Haentjens, and M. Simonet. 1998. Polymorphism of ypm, Yersinia pseu­ dotuberculosis superantigen encoding gene. Zentralbl. Bakteriol. 29(Suppl.):397–398. Casutt-Meyer, S., F. Renzi, M. Schmaler, N. J. Jann, M. Amstutz, and G. R. Cornelis. 2010. Oligomeric coiledcoil adhesin YadA is a double-edged sword. PLoS ONE 5:e15159. Centers for Disease Control and Prevention. 2003. Yersinia enterocolitica gastroenteritis among infants exposed to chitterlings—Chicago, Illinois, 2002. MMWR Morb. Mortal. Wkly. Rep. 52:956–958. Centers for Disease Control and Prevention. 2008. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food— 10 states, 2007. MMWR Morb. Mortal. Wkly. Rep. 57:336–370. Chao, W. L., R. J. Ding, and R. S. Chen. 1988. Survival of Yersinia enterocolitica in the environment. Can. J. Microbiol. 34:753–756. China, B., B. T. N’Guyen, M. de Bruyere, and G. R. Cornelis. 1994. Role of YadA in resistance of Yersinia enterocolitica to phagocytosis by human polymorphonuclear leukocytes. Infect. Immun. 62:1275–1281. China, B., M. P. Sory, B. T. N’Guyen, M. de Bruyere, and G. R. Cornelis. 1993. Role of the YadA protein in prevention of opsonization of Yersinia enterocolitica by C3b molecules. Infect. Immun. 61:3129–3136. Clark, M. A., B. H. Hirst, and M. A. Jepson. 1998. M-cell surface beta1 integrin expression and invasin-­ mediated targeting of Yersinia pseudotuberculo­ sis to mouse Peyer’s patch M cells. Infect. Immun. 66:1237–1243. Colmegna, I., and L. R. Espinoza. 2005. Recent advances in reactive arthritis. Curr. Rheumatol. Rep. 7:201–207. Cornelis, G., Y. Laroche, G. Balligand, M. P. Sory, and G. Wauters. 1987. Yersinia enterocolitica, a primary model for bacterial invasiveness. Rev. Infect. Dis. 9:64–87. Cornelis, G., C. Sluiters, C. L. de Rouvroit, and T. Michiels. 1989. Homology between virF, the transcriptional activator of the Yersinia virulence regulon, and

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

56.

57. 58. 59.

60.

61. 62.

63. 64. 65. 66.

67. 68.

69.

70.

71. 72.

367

AraC, the Escherichia coli arabinose operon regulator. J. Bacteriol. 171:254–262. Cornelis, G., C. Sluiters, I. Delor, D. Geib, K. Kaninga, C. Lambert de Rouvroit, M.-P. Sory, J.-C. Vanooteghem, and T. Michiels. 1991. ymoA, a Yersinia enterocolitica chromosomal gene modulating the expression of virulence functions. Mol. Microbiol. 5:1023–1034. Cornelis, G. R. 2002. The Yersinia Ysc-Yop ‘type III’ weaponry. Nat. Rev. Mol. Cell Biol. 3:742–752. Cornelis, G. R. 2002. Yersinia type III secretion: send in the effectors. J. Cell Biol. 158:401–408. Cornelis, G. R., T. Biot, C. Lambert de Rouvroit, T. Michiels, B. Mulder, C. Sluiters, M. P. Sory, M. Van Bouchaute, and J. C. Vanooteghem. 1989. The Yersinia yop regulon. Mol. Microbiol. 3:1455–1459. Cornelis, G. R., A. Boland, A. P. Boyd, C. Geuijen, M. Iriarte, C. Neyt, M.-P. Sory, and I. Stainier. 1998. The virulence plasmid of Yersinia, an antihost genome. Microbiol. Mol. Biol. Rev. 62:1315–1352. Reference deleted. Cornelis, G. R., C. Sluiters, I. Delor, D. Geib, K. Kaniga, C. Lambert de Rouvroit, M. P. Sory, J. C. Vanooteghem, and T. Michiels. 1991. ymoA, a Yersinia enterocolitica chromosomal gene modulating the expression of virulence functions. Mol. Microbiol. 5:1023–1034. Cornelius, C., L. Quenee, D. Anderson, and O. Schneewind. 2007. Protective immunity against plague. Adv. Exp. Med. Biol. 603:415–424. Cotter, S. E., N. K. Surana, and J. W. St Geme III. 2005. Trimeric autotransporters: a distinct subfamily of autotransporter proteins. Trends Microbiol. 13:199–205. Cover, T. L., and R. C. Aber. 1989. Yersinia enteroco­ litica. N. Engl. J. Med. 321:16–24. Cowan, C., A. V. Philipovskiy, C. R. Wulff-Strobel, Z. Ye, and S. C. Straley. 2005. Anti-LcrV antibody inhibits delivery of Yops by Yersinia pestis KIM5 by directly promoting phagocytosis. Infect. Immun. 73:6127–6137. Reference deleted. Currie, M. G., K. F. Fok, J. Kato, R. J. Moore, F. K. Hamra, K. L. Duffin, and C. E. Smith. 1992. Guanylin: an endogenous activator of intestinal guanylate cyclase. Proc. Natl. Acad. Sci. USA 89:947–951. Daniel, C., F. Sebbane, S. Poiret, D. Goudercourt, J. Dewulf, C. Mullet, M. Simonet, and B. Pot. 2009. Protection against Yersinia pseudotuberculosis infection conferred by a Lactococcus lactis mucosal delivery vector secreting LcrV. Vaccine 27:1141–1144. D’Aoust, J. Y., C. E. Park, R. A. Szabo, E. C. Todd, D. B. Emmons, and R. C. McKellar. 1988. Thermal inactivation of Campylobacter species, Yersinia entero­ colitica, and hemorrhagic Escherichia coli O157:H7 in fluid milk. J. Dairy Sci. 71:3230–3236. Darby, C. 2008. Uniquely insidious: Yersinia pestis biofilms. Trends Microbiol. 16:158–164. de Giusti, M., and E. de Vito. 1992. Inactivation of Yersinia enterocolitica by nitrite and nitrate in food. Food Addit. Contam. 9:405–408.

SMP_Food Microbiology_CH14.indd

367

73. de Koning, J., J. Heesemann, J. A. Hoogkamp-Korstanje, J. J. Festen, P. M. Houtman, and P. L. van Oijen. 1989. Yersinia in intestinal biopsy specimens from patients with seronegative spondyloarthropathy: correlation with specific serum IgA antibodies. J. Infect. Dis. 159:109–112. 74. de Koning-Ward, T. F., T. Grant, F. Oppedisano, and R. M. Robins-Browne. 1998. Effect of bacterial invasion of macrophages on the outcome of assays to assess bacterium-macrophage interactions. J. Immunol. Methods 215:39–44. 75. de Koning-Ward, T. F., and R. M. Robins-Browne. 1995. Contribution of urease to acid tolerance in Yersinia en­ terocolitica. Infect. Immun. 63:3790–3795. 76. Delor, I., and G. R. Cornelis. 1992. Role of Yersinia en­ terocolitica Yst toxin in experimental infection of young rabbits. Infect. Immun. 60:4269–4277. 77. Delor, I., A. Kaeckenbeeck, G. Wauters, and G. R. Cornelis. 1990. Nucleotide sequence of yst, the Yersinia enterocolitica gene encoding the heat-stable enterotoxin, and prevalence of the gene among pathogenic and nonpathogenic yersiniae. Infect. Immun. 58:2983–2988. 78. Dewoody, R., P. M. Merritt, A. S. Houppert, and M. M. Marketon. 2011. YopK regulates the Yersinia pestis type III secretion system from within host cells. Mol. Microbiol. 79:1445–1461.  doi:10.1111/j.13652958.2011.07534.x. 79. Diepold, A., M. Amstutz, S. Abel, I. Sorg, U. Jenal, and G. R. Cornelis. 2010. Deciphering the assembly of the Yersinia type III secretion injectisome. EMBO J. 29:1928–1940. 80. Dion, P., R. Charbonneau, and C. Thibault. 1994. Effect of ionizing dose rate on the radioresistance of some food pathogenic bacteria. Can. J. Microbiol. 40:369–374. 81. Donadini, R., and B. A. Fields. 2007. Yersinia pseudo­ tuberculosis superantigens. Chem. Immunol. Allergy 93:77–91. 82. Doyle, M. P. 1990. Pathogenic Escherichia coli, Yersinia enterocolitica, and Vibrio parahaemolyticus. Lancet 336:1111–1115. 83. Eberth, C. J. 1885. Zwei Mykosen des Meerschweinchens. Archiv. Pathol. Anat. Physiol. Klin. Med. 100:15–27. 84. Eitel, J., and P. Dersch. 2002. The YadA protein of Yersinia pseudotuberculosis mediates high-efficiency uptake into human cells under environmental conditions in which invasin is repressed. Infect. Immun. 70:4880–4891. 85. Escudero, M. E., L. Velazquez, M. S. Di Genaro, and A. M. de Guzman. 1999. Effectiveness of various disinfectants in the elimination of Yersinia enterocolitica on fresh lettuce. J. Food Prot. 62:665–669. 86. Falgarone, G., H. S. Blanchard, B. Riot, M. Simonet, and M. Breban. 1999. Cytotoxic T-cell-mediated response against Yersinia pseudotuberculosis in HLA-B27 transgenic rat. Infect. Immun. 67:3773–3779. 87. Farmer, J. J., III, G. P. Carter, V. L. Miller, S. Falkow, and I. K. Wachsmuth. 1992. Pyrazinamidase, CR-MOX agar, salicin fermentation-esculin hydrolysis, and D-xylose fermentation for identifying pathogenic serotypes of Yersinia enterocolitica. J. Clin. Microbiol. 30:2589–2594.

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

368 88. Feng, P., and S. D. Weagant. 1993. Yersinia, p. 427– 460. In Y. H. Hui, J. R. Gorham, K. D. Murrell, and D. O. Cliver (ed.), Foodborne Disease Handbook, vol. 1. Marcel Dekker, New York, NY. 89. Ferracci, F., F. D. Schubot, D. S. Waugh, and G. V. Plano. 2005. Selection and characterization of Yersinia pestis YopN mutants that constitutively block Yop secretion. Mol. Microbiol. 57:970–987. 90. Foberg, U., A. Fryden, E. Kihlstrom, K. Persson, and O. Weiland. 1986. Yersinia enterocolitica septicemia: clinical and microbiological aspects. Scand. J. Infect. Dis. 18:269–279. 91. Forsberg, A., R. Rosqvist, and H. Wolf-Watz. 1994. Regulation and polarized transfer of the Yersinia outer proteins (Yops) involved in antiphagocytosis. Trends Microbiol. 2:14–19. 92. Forsberg, A., A. M. Viitanen, M. Skurnik, and H. Wolf-Watz. 1991. The surface-located YopN protein is involved in calcium signal transduction in Yersinia pseu­ dotuberculosis. Mol. Microbiol. 5:977–986. 93. Franchi, L., J. H. Park, M. H. Shaw, N. Marina-Garcia, G. Chen, Y. G. Kim, and G. Núñez. 2008. Intracellular NOD-like receptors in innate immunity, infection and disease. Cell. Microbiol. 10:1–8. 94. Fredriksson-Ahomaa, M., M. Bucher, C. Hank, A. Stolle, and H. Korkeala. 2001. High prevalence of Yersinia en­ terocolitica 4:O3 on pig offal in southern Germany: a slaughtering technique problem. Syst. Appl. Microbiol. 24:457–463. 95. Fredriksson-Ahomaa, M., and H. Korkeala. 2003. Low occurrence of pathogenic Yersinia enterocolitica in clinical, food, and environmental samples: a methodological problem. Clin. Microbiol. Rev. 16:220–229. 96. Fredriksson-Ahomaa, M., A. Stolle, and H. Korkeala. 2006. Molecular epidemiology of Yersinia enterocolitica infections. FEMS Immunol. Med. Microbiol. 47:315–329. 97. Fukushima, H., M. Gomyoda, S. Ishikura, T. Nishio, S. Moriki, J. Endo, S. Kaneko, and M. Tsubokura. 1989. Cat-contaminated environmental substances lead to Yersinia pseudotuberculosis infection in children. J. Clin. Microbiol. 27:2706–2709. 98. Fukushima, H., M. Gomyoda, and S. Kaneko. 1990. Mice and moles inhabiting mountainous areas of Shimane Peninsula as sources of infection with Yersinia pseudotuberculosis. J. Clin. Microbiol. 28:2448–2455. 99. Gaastra, W., and A. M. Svennerholm. 1996. Colonization factors of human enterotoxigenic Escherichia coli (ETEC). Trends Microbiol. 4:444–452. 100. Galyov, E. E., S. Hakansson, A. Forsberg, and H. WolfWatz. 1993. A secreted protein kinase of Yersinia pseu­ dotuberculosis is an indispensable virulence ­determinant. Nature 361:730–732. 101. Gaston, J. S., C. Cox, and K. Granfors. 1999. Clinical and experimental evidence for persistent Yersinia infection in reactive arthritis. Arthritis Rheum. 42:2239–2242. 102. Ghosh, P. 2004. Process of protein transport by the type III secretion system. Microbiol. Mol. Biol. Rev. 68:771–795.

SMP_Food Microbiology_CH14.indd

368

103. Goverde, R. L., J. H. J. Huis in’t Veld, J. G. Kusters, and F. R. Mooi. 1998. The psychrotrophic bacterium Yersinia enterocolitica requires expression of pnp, the gene for polynucleotide phosphorylase, for growth at low temperature (5ºC). Mol. Microbiol. 28:555–569. 104. Granfors, K., S. Jalkanen, R. von Essen, R. LahesmaaRantala, O. Isomaki, K. Pekkola-Heino, R. MerilahtiPalo, R. Saario, H. Isomaki, and A. Toivanen. 1989. Yersinia antigens in synovial-fluid cells from patients with reactive arthritis. N. Engl. J. Med. 320:216–221. 105. Grant, T., V. Bennett-Wood, and R. M. RobinsBrowne. 1998. Identification of virulence-associated characteristics in clinical isolates of Yersinia enteroco­ litica lacking classical virulence markers. Infect. Immun. 66:1113–1120. 106. Grant, T., V. Bennett-Wood, and R. M. Robins-Browne. 1999. Characterization of the interaction between Yersinia enterocolitica biotype 1A and phagocytes and epithelial cells in vitro. Infect. Immun. 67:4367–4375. 107. Graumann, P. L., and M. A. Marahiel. 1998. A superfamily of proteins that contain the cold-shock domain. Trends Biochem. Sci. 23:286–290. 108. Gripenberg-Lerche, C., M. Skurnik, L. Zhang, K.-O. Söderström, and P. Toivanen. 1994. Role of YadA in arthritogenicity of Yersinia enterocolitica serotype O:8: experimental studies with rats. Infect. Immun. 62:5568–5575. 109. Gripenberg-Lerche, C., L. Zhang, P. Ahtonen, P. Toivanen, and M. Skurnik. 2000. Construction of urease-negative mutants of Yersinia enterocolitica serotypes O:3 and O:8: role of urease in virulence and arthritogenicity. Infect. Immun. 68:942–947. 110. Grosdent, N., I. Maridonneau-Parini, M. P. Sory, and G. R. Cornelis. 2002. Role of Yops and adhesins in resistance of Yersinia enterocolitica to phagocytosis. Infect. Immun. 70:4165–4176. 111. Grützkau, A., C. Hanski, H. Hahn, and E. O. Riecken. 1990. Involvement of M cells in the bacterial invasion of Peyer’s patches: a common mechanism shared by Yersinia enterocolitica and other enteroinvasive bacteria. Gut 31:1011–1015. 112. Guarino, A., R. Giannella, and M. R. Thompson. 1989. Citrobacter freundii produces an 18-amino-acid heat-stable enterotoxin identical to the 18-amino-acid Escherichia coli heat-stable enterotoxin (STIa). Infect. Immun. 57:649–652. 113. Gurtler, M., T. Alter, S. Kasimir, M. Linnebur, and K. Fehlhaber. 2005. Prevalence of Yersinia enterocolitica in fattening pigs. J. Food Prot. 68:850–854. 114. Håkansson, S., T. Bergman, J. C. Vanooteghem, G. Cornelis, and H. Wolf-Watz. 1993. YopB and YopD constitute a novel class of Yersinia Yop proteins. Infect. Immun. 61:71–80. 115. Hamburger, Z. A., M. S. Brown, R. R. Isberg, and P. J. Bjorkman. 1999. Crystal structure of invasin: a bacterial integrin-binding protein. Science 286:291–295. 116. Han, Y. W., and V. L. Miller. 1997. Reevaluation of the virulence phenotype of the inv yadA double mu-

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

369

tants of Yersinia pseudotuberculosis. Infect. Immun. 65:327–330. 117. Hanna, M. O., J. C. Stewart, D. L. Zink, Z. L. Carpenter, and C. Vanderzant. 1977. Development of Yersinia en­ terocolitica on raw and cooked beef and pork at different temperatures. J. Food Sci. 42:1180–1184. 118. Hansen, P. S., B. E. Wenzel, T. H. Brix, and L. Hegedus. 2006. Yersinia enterocolitica infection does not confer an increased risk of thyroid antibodies: evidence from a Danish twin study. Clin. Exp. Immunol. 146:32–38. 119. Hanski, C., U. Kutschka, H. P. Schmoranzer, M. Naumann, A. Stallmach, H. Hahn, H. Menge, and E. O. Riecken. 1989. Immunohistochemical and electron microscopic study of interaction of Yersinia enterocolitica serotype O8 with intestinal mucosa during experimental enteritis. Infect. Immun. 57:673–678. 120. Harakeh, M. S., J. D. Berg, J. C. Hoff, and A. Matin. 1985. Susceptibility of chemostat-grown Yersinia en­ terocolitica and Klebsiella pneumoniae to chlorine dioxide. Appl. Environ. Microbiol. 49:69–72. 121. Hartland, E. L., and R. M. Robins-Browne. 1998. Infections with enteropathogenic Yersinia species: paradigms of bacterial pathogenesis. Rev. Med. Microbiol. 9:191–205. 122. Hayashidani, H., Y. Hara-Kudo, S. Kinoshita, K. Saeki, A. T. Okatani, Y. Nomura, and S. Kumagai. 2005. Differences in heat resistance among pathogenic Yersinia enterocolitica depended on growth temperature and serotype. J. Food Prot. 68:1081–1082. 123. Hayashidani, H., Y. Ohtomo, Y. Toyokawa, M. Saito, K. Kaneko, J. Kosuge, M. Kato, M. Ogawa, and G. Kapperud. 1995. Potential sources of sporadic human infection with Yersinia enterocolitica serovar O:8 in Aomori Prefecture, Japan. J. Clin. Microbiol. 33:1253–1257. 124. Heesemann, J., K. Gaede, and I. B. Autenrieth. 1993. Experimental Yersinia enterocolitica infection in rodents: a model for human yersiniosis. APMIS 101:417–429. 125. Heesemann, J., K. Hantke, T. Vocke, E. Saken, A. Rakin, I. Stojiljkovic, and R. Berner. 1993. Virulence of Yersinia enterocolitica is closely associated with side rophore production, expression of an iron-repressible outer membrane polypeptide of 65,000 Da and pesticin sensitivity. Mol. Microbiol. 8:397–408. 126. Heise, T., and P. Dersch. 2006. Identification of a domain in Yersinia virulence factor YadA that is crucial for extracellular matrix-specific cell adhesion and uptake. Proc. Natl. Acad. Sci. USA 103:3375–3380. 127. Henderson, I. R., F. Navarro-Garcia, M. Desvaux, R. C. Fernandez, and D. Ala’Aldeen. 2004. Type V protein secretion pathway: the autotransporter story. Microbiol. Mol. Biol. Rev. 68:692–744. 128. Hermann, E. 1993. T cells in reactive arthritis. APMIS 101:177–186. 129. Hermann, E., D. T. Yu, K. H. Meyer zum Buschenfelde, and B. Fleischer. 1993. HLA-B27-restricted CD8 T cells derived from synovial fluids of patients with re-

SMP_Food Microbiology_CH14.indd

369

active arthritis and ankylosing spondylitis. Lancet 342:646–650. 130. Herrlinger, J. D., and J. U. Asmussen. 1992. Long term prognosis in yersinia arthritis: clinical and serological findings. Ann. Rheum. Dis. 51:1332–1334. 131. Heusipp, G., M. A. Schmidt, and V. L. Miller. 2003. Identification of rpoE and nadB as host responsive elements of Yersinia enterocolitica. FEMS Microbiol. Lett. 226:291–298. 132. Hinnebusch, B. J., and D. L. Erickson. 2008. Yersinia pestis biofilm in the flea vector and its role in the transmission of plague. Curr. Top. Microbiol. Immunol. 322:229–248. 133. Hoiczyk, E., and G. Blobel. 2001. Polymerization of a single protein of the pathogen Yersinia enterocolitica into needles punctures eukaryotic cells. Proc. Natl. Acad. Sci. USA 98:4669–4674. 134. Hölmstrom, A., J. Petterson, R. Rosqvist, S. Hakansson, F. Tafazoli, M. Fallman, K. E. Magnusson, H. WolfWatz, and A. Forsberg. 1997. YopK of Yersinia pseu­ dotuberculosis controls translocation of Yop effectors across the eukaryotic cell membrane. Mol. Microbiol 24:73–91. 135. Hölmstrom, A., R. Rosqvist, H. Wolf-Watz, and A. Forsberg. 1995. Virulence plasmid-encoded YopK is essential for Yersinia pseudotuberculosis to cause systemic infection in mice. Infect. Immun. 63:2269–2276. 136. Hoogkamp-Korstanje, J. A. A., and V. M. M. StolkEngelaar. 1995. Yersinia enterocolitica infection in children. Pediatr. Infect. Dis. J. 14:771–775. 137. Howard, S. L., M. W. Gaunt, J. Hinds, A. A. Witney, R. Stabler, and B. W. Wren. 2006. Application of comparative phylogenomics to study the evolution of Yersinia enterocolitica and to identify genetic differences relating to pathogenicity. J. Bacteriol. 188:3645–3653. 138. Huang, X., K. Yoshino, H. Nakao, and T. Takeda. 1997. Nucleotide sequence of a gene encoding the novel Yersinia enterocolitica heat-stable enterotoxin that includes a pro-region-like sequence in its mature toxin molecule. Microb. Pathog. 22:89–97. 139. Iriarte, M., and G. R. Cornelis.1999. Identification of SycN, YscX, and YscY, three new elements of the Yersinia yop virulon. J. Bacteriol. 181:675–680. 140. Iriarte, M., I. Stainier, A. V. Mikulskis, and G. R. Cornelis. 1995. The fliA gene encoding s28 in Yersinia enterocolitica. J. Bacteriol. 177:2299–2304. 141. Iriarte, M., J. C. Vanooteghem, I. Delor, R. Diaz, S. Knutton, and G. R. Cornelis. 1993. The Myf fibrillae of Yersinia enterocolitica. Mol. Microbiol. 9:507–520. 142. Isberg, R. R. 1990. Pathways for the penetration of enteroinvasive Yersinia into mammalian cells. Mol. Biol. Med. 7:73–82. 143. Isberg, R. R., and P. Barnes. 2001. Subversion of integrins by enteropathogenic Yersinia. J. Cell Sci. 114:21–28. 144. Isberg, R. R., and S. Falkow. 1985. A single genetic locus encoded by Yersinia pseudotuberculosis permits

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

370

145. 146. 147.

148.

149.

150.

151.

152. 153. 154. 155. 156. 157. 158.

159. 160.

invasion of cultured animal cells by Escherichia coli K12. Nature 317:262–264. Isberg, R. R., and G. T. Van Nhieu. 1994. Two mammalian cell internalization strategies used by pathogenic bacteria. Annu. Rev. Genet. 28:395–422. Israël, A. 2010. The IKK complex, a central regulator of NF-kB activation. Cold Spring Harb. Perspect. Biol. 2: a000158. Iteman, I., A. Guiyoule, and E. Carniel. 1996. Comparison of three molecular methods for typing and subtyping pathogenic Yersinia enterocolitica strains. J. Med. Microbiol. 45:48–56. Iwobi, A., J. Heesemann, E. Garcia, E. Igwe, C. Noelting, and A. Rakin. 2003. Novel virulence-associated type II secretion system unique to high-pathogenicity Yersinia enterocolitica. Infect. Immun. 71:1872–1879. Jackson, M. W., and G. V. Plano. 2000. Interactions between type III secretion apparatus components from Yersinia pestis detected using the yeast two-hybrid system. FEMS Microbiol. Lett. 186:85–90. Jacobi, C. A., A. Roggenkamp, A. Rakin, R. Zumbihl, L. Leitritz, and J. Heesemann. 1998. In vitro and in vivo expression studies of yopE from Yersinia entero­ colitica using the gfp reporter gene. Mol. Microbiol. 30:865–882. Jalava, K., M. Hakkinen, M. Valkonen, U. M. Nakari, T. Palo, S. Hallanvuo, J. Ollgren, A. Siitonen, and J. P. Nuorti. 2006. An outbreak of gastrointestinal illness and erythema nodosum from grated carrots contaminated with Yersinia pseudotuberculosis. J. Infect. Dis. 194:1209–1216. Jepson, M. A., and M. A. Clark. 1998. Studying M cells and their role in infection. Trends Microbiol. 9:359–365. Jones, T. F. 2003. From pig to pacifier: chitterlingassociated yersiniosis outbreak among black infants. Emerg. Infect. Dis. 9:1007–1009. Kapperud, G. 1982. Enterotoxin production at 4o, 22o, and 37o among Yersinia enterocolitica and Y. enteroco­ litica-like bacteria. APMIS 90B:185–189. Kapperud, G. 1991. Yersinia enterocolitica in food hygiene. Int. J. Food Microbiol. 12:53–65. Kasatiya, S. S. 1976. Yersinia enterocolitica gastroenteritis outbreak—Montreal. Can. Dis. Wkly. Rep. 2:73–74. Keet, E. E. 1974. Yersinia enterocolitica septicemia: source of infection and incubation period identified. N. Y. State J. Med. 74:2226–2229. Kerschen, E. J., D. A. Cohen, A. M. Kaplan, and S. C. Straley. 2004. The plague virulence protein YopM targets the innate immune response by causing a global depletion of NK cells. Infect. Immun. 72:4589–4602. Kim, T. J., B. M. Young, and G. M. Young. 2008. Effect of flagellar mutations on Yersinia enterocolitica biofilm formation. Appl. Environ. Microbiol. 74:5466–5474. King, C. H., E. B. Shotts, R. E. Wooley, and K. G. Porter. 1988. Survival of coliforms and bacterial pathogens within protozoa during chlorination. Appl. Environ. Microbiol. 54:3023–3033.

SMP_Food Microbiology_CH14.indd

370

161. Kirjavainen, V., H. Jarva, M. Biedzka-Sarek, A. M. Blom, M. Skurnik, and S. Meri. 2008. Yersinia entero­ colitica serum resistance proteins YadA and Ail bind the complement regulator C4b-binding protein. PLoS Pathog. 4:e1000140. 162. Koster, M., W. Bitter, H. de Cock, A. Allaoui, G. R. Cornelis, and J. Tommassen. 1997. The outer membrane component, YscC, of the Yop secretion machinery of Yersinia enterocolitica forms a ring-shaped multimeric complex. Mol. Microbiol. 26:789–797. 163. Kotetishvili, M., A. Kreger, G. Wauters, J. G. Morris, Jr., A. Sulakvelidze, and O. C. Stine. 2005. Multilocus sequence typing for studying genetic relationships among Yersinia species. J. Clin. Microbiol. 43:2674–2684. 164. Kraehenbuhl, J. P., and M. R. Neutra. 2000. Epithelial M cells: differentiation and function. Annu. Rev. Cell Dev. Biol. 16:301–332. 165. Lahesmaa, R., M. Skurnik, K. Granfors, T. Mottonen, R. Saario, A. Toivanen, and P. Toivanen. 1992. Molecular mimicry in the pathogenesis of spondyloarthropathies. A critical appraisal of cross-reactivity between microbial antigens and HLA-B27. Br. J. Rheumatol. 31:221–229. 166. Larsen, J. H. 1980. Yersinia enterocolitica infection and rheumatic diseases. Scand. J. Rheumatol. 9:129–137. 167. Leclercq, A., L. Martin, M. L. Vergnes, N. Ounnoughene, J. F. Laran, P. Giraud, and E. Carniel. 2005. Fatal Yersinia enterocolitica biotype 4 serovar O:3 sepsis after red blood cell transfusion. Transfusion 45:814–818. 168. Lee, L. A., A. R. Gerber, D. R. Lonsway, J. D. Smith, G. P. Carter, N. D. Puhr, C. M. Parrish, R. K. Sikes, R. J. Finton, and R. V. Tauxe. 1990. Yersinia enteroco­ litica O:3 infections in infants and children, associated with the household preparation of chitterlings. N. Engl. J. Med. 322:984–987. 169. Lee, V. T., D. M. Anderson, and O. Schneewind. 1998. Targeting of Yersinia Yop proteins into the cytosol of HeLa cells: one-step translocation of YopE across bacterial and eukaryotic membranes is dependent on SycE chaperone. Mol. Microbiol. 28:593–601. 170. Leirisalo-Repo, M. 1987. Yersinia arthritis. Acute clinical picture and long-term prognosis. Contrib. Microbiol. Immunol. 9:145–154. 171. Lesic, B., and E. Carniel. 2005. Horizontal transfer of the high-pathogenicity island of Yersinia pseudotuber­ culosis. J. Bacteriol. 187:3352–3358. 172. Leung, K. Y., and S. C. Straley. 1989. The yopM gene of Yersinia pestis encodes a released protein having homology with the human platelet surface protein GPIb alpha. J. Bacteriol. 171:4623–4632. 173. Lian, C. J., W. S. Hwang, J. K. Kelly, and C. H. Pai. 1987. Invasiveness of Yersinia enterocolitica lacking the virulence plasmid: an in-vivo study. J. Med. Microbiol. 24:219–226. 174. Lian, C. J., W. S. Hwang, J. K. Kelly, and C. H. Pai. 1987. Penetration of the intestinal mucosa by Yersinia enterocolitica lacking the virulence plasmid. Contrib. Microbiol. Immunol. 9:239–242. 175. Lindler, L. E., and B. D. Tall. 1993. Yersinia pestis pH 6 antigen forms fimbriae and is induced by intracel-

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

371

lular association with macrophages. Mol. Microbiol. 8:311–324. 176. Lo, W. F., A. S. Woods, A. DeCloux, R. J. Cotter, E. S. Metcalf, and M. J. Soloski. 2000. Molecular mimicry mediated by MHC class Ib molecules after infection with gram-negative pathogens. Nat. Med. 6:215–218. 177. Long, C., T. F. Jones, D. J. Vugia, J. Scheftel, N. Strockbine, P. Ryan, B. Shiferaw, R. V. Tauxe, and L. H. Gould. 2010. Yersinia pseudotuberculosis and Y. enterocolitica infections, FoodNet, 1996-2007. Emerg. Infect. Dis. 16:566–567. 178. Malassez, L., and W. Vignal. 1883. Tuberculose zoogloeique (forme ou espèce de tuberculose sans bacille). Arch. Physiol. Normale Pathol. Ser. 3 2:369–412. 179. Marjai, E., M. Kalman, I. Kajary, A. Belteky, and M. Rodler. 1987. Isolation from food and characterization by virulence tests of Yersinia enterocolitica associated with an outbreak. Acta Microbiol. Hung. 34:97–109. 180. Marks, M. I., C. H. Pai, L. Lafleur, L. Lackman, and O. Hammerberg. 1980. Yersinia enterocolitica gastroenteritis: a prospective study of clinical, bacteriologic, and epidemiologic features. J. Pediatr. 96:26–31. 181. Marra, A., and R. R. Isberg. 1997. Invasin-dependent and invasin-independent pathways for translocation of Yersinia pseudotuberculosis across the Peyer’s patch intestinal epithelium. Infect. Immun. 65:3412–3421. 182. Maruyama, T. 1987. Yersinia enterocolitica infection in humans and isolation of the microorganism from pigs in Japan. Contrib. Microbiol. Immunol. 9:48–55. 183. Matsumoto, H., and G. M. Young. 2006. Proteomic and functional analysis of the suite of Ysp proteins exported by the Ysa type III secretion system of Yersinia enterocolitica Biovar 1B. Mol. Microbiol. 59:689–706. 184. McDonald, C., P. O. Vacratsis, J. B. Bliska, and J. E. Dixon. 2003. The Yersinia virulence factor YopM forms a novel protein complex with two cellular kinases. J. Biol. Chem. 278:18514–18523. 185. McIver, M. A., and R. M. Pike. 1934. Chronic ­glanderslike infection of the face caused by an organism resembling Flavobacterium pseudomallei Whitmore, p. 16–21. In Clinical Miscellany, Mary Imogene Bassett Hospital, vol. 1. Cooperstown, New York, NY. 186. Meysick, K. C., J. Seidman, and J. R. Falconio. 2009. The Yersinia pseudotuberculosis YplA phospholipase differs in its activity, regulation and secretion from that of the Yersinia enterocolitica YplA. Microb. Pathog. 47:24–32. 187. Michiels, T., P. Wattiau, R. Brasseur, J. M. Ruysschaert, and G. Cornelis. 1990. Secretion of Yop proteins by yersiniae. Infect. Immun. 58:2840–2849. 188. Mikulskis, A. V., I. Delor, V. H. Thi, and G. R. Cornelis. 1994. Regulation of the Yersinia enterocolitica enterotoxin Yst gene. Influence of growth phase, temperature, osmolarity, pH and bacterial host factors. Mol. Microbiol. 14:905–915. 189. Miller, V. L. 1992. Yersinia invasion genes and their products. ASM News 58:26–33.

SMP_Food Microbiology_CH14.indd

371

190. Miller, V. L., K. B. Beer, G. Heusipp, B. M. Young, and M. R. Wachtel. 2001. Identification of regions of Ail required for the invasion and serum resistance phenotypes. Mol. Microbiol. 41:1053–1062. 191. Miller, V. L., and S. Falkow. 1988. Evidence for two genetic loci in Yersinia enterocolitica that can promote invasion of epithelial cells. Infect. Immun. 56:1242–1248. 192. Miller, V. L., B. B. Finlay, and S. Falkow. 1988. Factors essential for the penetration of mammalian cells by Yersinia. Curr. Top. Microbiol. Immunol. 138:15–39. 193. Mittal, R., S. Y. Peak-Chew, and H. T. McMahon. 2006. Acetylation of MEK2 and IkB kinase (IKK) activation loop residues by YopJ inhibits signaling. Proc. Natl. Acad. Sci. USA 103:18574–18579. 194. Miyoshi-Akiyama, T., W. Fujimaki, X. J. Yan, J. Yagi, K. Imanishi, H. Kato, K. Tomonari, and T. Uchiyama. 1997. Identification of murine T cells reactive with the bacterial superantigen Yersinia pseudotuberculosisderived mitogen (YPM) and factors involved in YPM-induced toxicity in mice. Microbiol. Immunol. 41:345–352. 195. Mollaret, H. H., H. Bercovier, and J. M. Alonso. 1979. Summary of the data received at the WHO Reference Centre for Yersinia enterocolitica. Contrib. Microbiol. Immunol. 5:174–184. 196. Morse, D. L., M. Shayegani, and R. J. Gallo. 1984. Epidemiologic investigation of a Yersinia camp outbreak linked to a food handler. Am. J. Public Health 74:589–592. 197. Mueller, C. A., P. Broz, S. A. Muller, P. Ringler, F. Erne-Brand, I. Sorg, M. Kuhn, A. Engel, and G. R. Cornelis. 2005. The V-antigen of Yersinia forms a distinct structure at the tip of injectisome needles. Science 310:674–676. 198. Mukherjee, S., G. Keitany, Y. Li, Y. Wang, H. L. Ball, E. J. Goldsmith, and K. Orth. 2006. Yersinia YopJ acetylates and inhibits kinase activation by blocking phosphorylation. Science 312:1211–1214. 199. Murros-Kontiainen, A. E., M. Fredriksson-Ahomaa, H. Korkeala, P. Johansson, R. Rahkila, and J. Bjorkroth. 2011. Yersinia nurmii sp. nov. Int. J. Syst. Evol. Microbiol. 61:2368–2372. 200. Murros-Kontiainen, A. E., P. Johansson, T. Niskanen, M. Fredriksson-Ahomaa, H. Korkeala, and J. Bjorkroth. 2010. Yersinia pekkanenii sp. nov. Int. J. Syst. Evol. Microbiol. 61:2363–2367. 201. Navarro, L., A. Koller, R. Nordfelth, H. Wolf-Watz, S. Taylor, and J. E. Dixon. 2007. Identification of a molecular target for the Yersinia protein kinase A. Mol. Cell 26:465–477. 202. Navarro Llorens, J. M., A. Tormo, and E. MartinezGarcia. 2010. Stationary phase in gram-negative bacteria. FEMS Microbiol. Rev. 34:476–495. 203. Neilands, J. B. 1981. Microbial iron compounds. Annu. Rev. Biochem. 50:715–731. 204. Nesbakken, T., G. Kapperud, K. Dommarsnes, M. Skurnik, and E. Hornes. 1991. Comparative study of a DNA hybridization method and two isolation procedures for detection of Yersinia enterocolitica O:3 in

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

372 naturally contaminated pork products. Appl. Environ. Microbiol. 57:389–394. 205. Neubauer, H., S. Aleksic, A. Hensel, E. J. Finke, and H. Meyer. 2000. Yersinia enterocolitica 16S rRNA gene types belong to the same genospecies but form three homology groups. Int. J. Med. Microbiol. 290:61–64. 206. Neuhaus, K., K. P. Francis, S. Rapposch, A. Gorg, and S. Scherer. 1999. Pathogenic Yersinia species carry a novel, cold-inducible major cold shock protein tandem gene duplication producing both bicistronic and monocistronic mRNA. J. Bacteriol. 181:6449–6455. 207. Neyt, C., and G. R. Cornelis. 1999. Insertion of a Yop translocation pore into the macrophage plasma membrane by Yersinia enterocolitica: requirement for translocators YopB and YopD, but not LcrG. Mol. Microbiol. 33:971–981. 208. Neyt, C., M. Iriarte, V. H. Thi, and G. R. Cornelis. 1997. Virulence and arsenic resistance in yersiniae. J. Bacteriol. 179:612–619. 209. Nilehn, B. 1969. Studies on Yersinia enterocolitica with special reference to bacterial diagnosis and occurrence in human acute enteric disease. Acta Pathol. Microbiol. Scand. Suppl. 206:1–48. 210. Nilles, M. L., A. W. Williams, E. Skrzypek, and S. C. Straley. 1997. Yersinia pestis LcrV forms a stable complex with LcrG and may have a secretion-related regulatory role in the low-Ca2+ response. J. Bacteriol. 179:1307–1316. 211. Nuorti, J. P., T. Niskanen, S. Hallanvuo, J. Mikkola, E. Kela, M. Hatakka, M. Fredriksson-Ahomaa, O. Lyytikainen, A. Siitonen, H. Korkeala, and P. Ruutu. 2004. A widespread outbreak of Yersinia pseudotuber­ culosis O:3 infection from iceberg lettuce. J. Infect. Dis. 189:766–774. 212. Oellerich, M. F., C. A. Jacobi, S. Freund, K. Niedung, A. Bach, J. Heesemann, and K. Trulzsch. 2007. Yersinia enterocolitica infection of mice reveals clonal invasion and abscess formation. Infect. Immun. 75:3802–3811. 213. Orth, K. 2002. Function of the Yersinia effector YopJ. Curr. Opin. Microbiol. 5:38–43. 214. Ostroff, S. 1995. Yersinia as an emerging infection: epidemiologic aspects of yersiniosis. Contrib. Microbiol. Immunol. 13:5–10. 215. Ostroff, S. M., G. Kapperud, L. C. Hutwagner, T. Nesbakken, N. H. Bean, J. Lassen, and R. V. Tauxe. 1994. Sources of sporadic Yersinia enterocolitica infections in Norway: a prospective case-control study. Epidemiol. Infect. 112:133–141. 216. Pai, C. H., and L. De Stephano. 1982. Serum resistance associated with virulence in Yersinia enterocolitica. Infect. Immun. 35:605–611. 217. Pai, C. H., V. Mors, and S. Toma. 1978. Prevalence of enterotoxigenicity in human and nonhuman isolates of Yersinia enterocolitica. Infect. Immun. 22:334–338. 218. Palonen, E., M. Lindstrom, and H. Korkeala. 2010. Adaptation of enteropathogenic Yersinia to low growth temperature. Crit. Rev. Microbiol. 36:54–67.

SMP_Food Microbiology_CH14.indd

372

219. Park, H., K. Teja, J. J. O’Shea, and R. M. Siegel. 2007. The Yersinia effector protein YpkA induces apoptosis independently of actin depolymerization. J. Immunol. 178:6426–6434. 220. Peixotto, S. S., G. Finne, M. O. Hanna, and C. Vanderzant. 1979. Presence, growth and survival of Yersinia enterocolitica in oyster, shrimp and crab. J. Food Prot. 42:974–981. 221. Pepe, J. C., J. L. Badger, and V. L. Miller. 1994. Growth phase and low pH affect the thermal regulation of the Yersinia enterocolitica inv gene. Mol. Microbiol. 11:123–135. 222. Pepe, J. C., and V. L. Miller. 1993. Yersinia enterocolit­ ica invasin: a primary role in the initiation of infection. Proc. Natl. Acad. Sci. USA 90:6473–6477. 223. Pepe, J. C., M. R. Wachtel, E. Wagar, and V. L. Miller. 1995. Pathogenesis of defined invasion mutants of Yersinia enterocolitica in a BALB/c mouse model of infection. Infect. Immun. 63:4837–4848. 224. Perry, R. D. 1993. Acquisition and storage of inorganic iron and hemin by the yersiniae. Trends Microbiol. 1:142–147. 225. Perry, R. D., and J. D. Fetherston. 1997. Yersinia pes­ tis—etiologic agent of plague. Clin. Microbiol. Rev. 10:35–66. 226. Persson, C., N. Carballeira, H. Wolf-Watz, and M. Fallman. 1997. The PTPase YopH inhibits uptake of Yersinia, tyrosine phosphorylation of p130Cas and FAK, and the associated accumulation of these proteins in peripheral focal adhesions. EMBO J. 16:2307–2318. 227. Pfeiffer, A. 1889. Ueber die bacillare Pseudotuberculose bei Nagethieren. Verlag von Georg Thieme, Leipzig, Germany. 228. Prehna, G., M. I. Ivanov, J. B. Bliska, and C. E. Stebbins. 2006. Yersinia virulence depends on mimicry of host Rho-family nucleotide dissociation inhibitors. Cell 126:869–880. 229. Pujol, C., and J. B. Bliska. 2005. Turning Yersinia pathogenesis outside in: subversion of macrophage function by intracellular yersiniae. Clin. Immunol. 114:216–226. 230. Puylaert, J. B., R. J. Vermeijden, S. D. van der Werf, L. Doornbos, and R. K. Koumans. 1989. Incidence and sonographic diagnosis of bacterial ileocaecitis masquerading as appendicitis. Lancet ii:84–86. 231. Rakin, A., C. Noelting, S. Schubert, and J. Heesemann. 1999. Common and specific characteristics of the highpathogenicity island of Yersinia enterocolitica. Infect. Immun. 67:5265–5274. 232. Rakin, A., C. Noelting, S. Schubert, and J. Heesemann. 1999. Common and specific characteristics of the highpathogenicity island of Yersinia enterocolitica. Infect. Immun. 67:5265–5274. 233. Ramamurthi, K. S., and O. Schneewind. 2003. Substrate recognition by the Yersinia type III protein secretion machinery. Mol. Microbiol. 50:1095–1102. 234. Ramamurthy, T., K. Yoshino, X. Huang, G. B. Nair, E. Carniel, T. Maruyama, H. Fukushima, and T. Takeda. 1997. The novel heat-stable enterotoxin subtype gene (ystB)

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

235. 236. 237.

238.

239. 240. 241.

242. 243.

244.

245. 246.

247.

248.

249.

373

of Yersinia enterocolitica: nucleotide sequence and distribution of the yst genes. Microb. Pathog. 23:189–200. Reed, R. P., R. M. Robins-Browne, and M. L. Williams. 1997. Yersinia enterocolitica peritonitis. Clin. Infect. Dis. 25:1468–1469. Reveille, J. D., and F. C. Arnett. 2005. Spondyloarthritis: update on pathogenesis and management. Am. J. Med. 118:592–603. Robins-Browne, R. M., A. M. Bordun, and K. J. Slee. 1993. Serological response of sheep to plasmid-encoded proteins of Yersinia species following natural infection with Y. enterocolitica and Y. pseudotuberculosis. J. Med. Microbiol. 39:268–272. Robins-Browne, R. M., M. D. Miliotis, S. Cianciosi, V. L. Miller, S. Falkow, and J. G. Morris, Jr. 1989. Evaluation of DNA colony hybridization and other techniques for detection of virulence in Yersinia species. J. Clin. Microbiol. 27:644–650. Robins-Browne, R. M., and J. K. Prpic. 1985. Effects of iron and desferrioxamine on infections with Yersinia enterocolitica. Infect. Immun. 47:774–779. Robins-Browne, R. M., J. K. Prpic, and S. J. Stuart. 1987. Yersiniae and iron. A study in host-parasite relationships. Contrib. Microbiol. Immunol. 9:254–258. Robins-Browne, R. M., A. R. Rabson, and H. J. Koornhof. 1979. Generalised infection with Yersinia enterocolitica and the role of iron. Contrib. Microbiol. Immunol. 5:277–282. Robins-Browne, R. M., C. S. Still, M. D. Miliotis, and H. J. Koornhof. 1979. Mechanism of action of Yersinia enterocolitica enterotoxin. Infect. Immun. 25:680–684. Robins-Browne, R. M., T. Takeda, A. Fasano, A. M. Bordun, S. Dohi, H. Kasuga, G. Fang, V. Prado, R. L. Guerrant, and J. G. Morris, Jr. 1993. Assessment of enterotoxin production by Yersinia enterocolitica and identification of a novel heat-stable enterotoxin produced by a noninvasive Y. enterocolitica strain isolated from clinical material. Infect. Immun. 61:764–767. Robins-Browne, R. M., S. Tzipori, G. Gonis, J. Hayes, M. Withers, and J. K. Prpic. 1985. The pathogenesis of Yersinia enterocolitica infection in gnotobiotic piglets. J. Med. Microbiol. 19:297–308. Reference deleted. Roggenkamp, A., H.-R. Neuberger, A. Flugel, T. Schmoll, and J. Heesemann. 1995. Substitution of two histidine residues in YadA protein of Yersinia entero­ colitica abrogates collagen binding, cell adherence and mouse virulence. Mol. Microbiol. 16:1207–1219. Rohde, J. R., J. M. Fox, and S. A. Minnich. 1994. Thermoregulation in Yersinia enterocolitica is coincident with changes in DNA supercoiling. Mol. Microbiol. 12:187–199. Sachdeva, P., and J. S. Virdi. 2004. Repetitive elements sequence (REP/ERIC)-PCR based genotyping of clinical and environmental strains of Yersinia enterocolitica biotype 1A reveal existence of limited number of clonal groups. FEMS Microbiol. Lett. 240:193–201. Saebo, A., and J. Lassen. 1992. Acute and chronic gastrointestinal manifestations associated with Yersinia

SMP_Food Microbiology_CH14.indd

373

enterocolitica infection. A Norwegian 10-year ­ followup study on 458 hospitalized patients. Ann. Surg. 215:250–255. 250. Saltman, L. H., Y. Lu, E. M. Zaharias, and R. R. Isberg. 1996. A region of the Yersinia pseudotuber­ culosis invasin protein that contributes to high affinity binding to integrin receptors. J. Biol. Chem. 271:23438–23444. 251. Sansonetti, P. J., and A. Phalipon. 1999. M cells as ports of entry for enteroinvasive pathogens: mechanisms of interaction, consequences for the disease process. Semin. Immunol. 11:193–203. 252. Schiemann, D. A. 1980. Yersinia enterocolitica: observations on some growth characteristics and response to selective agents. Can. J. Microbiol. 26:1232–1240. 253. Schiemann, D. A. 1981. An enterotoxin-negative strain of Yersinia enterocolitica serotype O:3 is capable of producing diarrhea in mice. Infect. Immun. 32:571–574. 254. Schiemann, D. A. 1987. Yersinia enterocolitica in milk and dairy products. J. Dairy Sci. 70:383–391. 255. Schiemann, D. A. 1988. Examination of enterotoxin production at low temperatures by Yersinia spp. in culture media and foods. J. Food Prot. 51:571–573. 256. Schiemann, D. A. 1989. Yersinia enterocolitica and Yersinia pseudotuberculosis, p. 601–672. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY. 257. Schleifstein, J., and M. B. Coleman. 1939. An unidentified microorganism resembling A. lignieri and Past. pseudotuberculosis and pathogenic for man. N. Y. State J. Med. 39:1749–1753. 258. Schleifstein, J., and M. B. Coleman. 1943. Bacterium en­ terocoliticum, p. 56. In Annual Report of the Division of Laboratories and Research. New York State Department of Health, Albany, NY. 259. Schmid, Y., G. A. Grassl, O. T. Buhler, M. Skurnik, I. B. Autenrieth, and E. Bohn. 2004. Yersinia enterocolitica adhesin A induces production of interleukin-8 in epithelial cells. Infect. Immun. 72:6780–6789. 260. Schmiel, D. H., E. Wagar, L. Karamanou, D. Weeks, and V. L. Miller. 1998. Phospholipase A of Yersinia enterocolitica contributes to pathogenesis in a mouse model. Infect. Immun. 66:3941–3951. 261. Shao, F. 2008. Biochemical functions of Yersinia type III effectors. Curr. Opin. Microbiol. 11:21–29. 262. Shao, F., P. M. Merritt, Z. Bao, R. W. Innes, and J. E. Dixon. 2002. A Yersinia effector and a Pseudomonas avirulence protein define a family of cysteine proteases functioning in bacterial pathogenesis. Cell 109:575–588. 263. Shao, F., P. O. Vacratsis, Z. Bao, K. E. Bowers, C. A. Fierke, and J. E. Dixon. 2003. Biochemical characterization of the Yersinia YopT protease: cleavage site and recognition elements in Rho GTPases. Proc. Natl. Acad. Sci. USA 100:904–909. 264. Shayegani, M., D. Morse, I. DeForge, T. Root, L. M. Parsons, and P. S. Maupin. 1983. Microbiology of a major foodborne outbreak of gastroenteritis caused by

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

374 Yersinia enterocolitica serogroup O:8. J. Clin. Microbiol. 17:35–40. 265. Shenkman, L., and E. J. Bottone. 1976. Antibodies to Yersinia enterocolitica in thyroid disease. Ann. Intern. Med. 85:735–739. 266. Shutinoski, B., M. A. Schmidt, and G. Heusipp. 2010. Transcriptional regulation of the Yts1 type II secretion system of Yersinia enterocolitica and identification of secretion substrates. Mol. Microbiol. 75:676–691. 267. Simonet, M. L. 1999. Enterobacteria in reactive arthritis: Yersinia, Shigella, and Salmonella. Rev. Rhum. Engl. Ed. 66:14S–18S. 268. Sims, G. R., D. A. Glenister, T. F. Brocklehurst, and B. M. Lund. 1989. Survival and growth of food poisoning bacteria following inoculation into cottage cheese varieties. Int. J. Food Microbiol. 9:173–195. 269. Singh, I., and J. S. Virdi. 2005. Interaction of Yersinia enterocolitica biotype 1A strains of diverse origin with cultured cells in vitro. Jpn. J. Infect. Dis. 58:31–33. 270. Skrzypek, E., C. Cowan, and S. C. Straley. 1998. Targeting of the Yersinia pestis YopM protein into HeLa cells and intracellular trafficking to the nucleus. Mol. Microbiol. 30:1051–1065. 271. Skurnik, M. 2003. Molecular genetics, biochemistry and biological role of Yersinia lipopolysaccharide. Adv. Exp. Med. Biol. 529:187–197. 272. Skurnik, M., S. Batsford, A. Mertz, E. Schiltz, and P. Toivanen. 1993. The putative arthritogenic cationic 19kilodalton antigen of Yersinia enterocolitica is a urease beta-subunit. Infect. Immun. 61:2498–2504. 273. Skurnik, M., and J. A. Bengoechea. 2003. The biosynthesis and biological role of lipopolysaccharide O-antigens of pathogenic yersiniae. Carbohydr. Res.338:2521–2529. 274. Skurnik, M., Y. el Tahir, M. Saarinen, S. Jalkanen, and P. Toivanen. 1994. YadA mediates specific binding of enteropathogenic Yersinia enterocolitica to human intestinal submucosa. Infect. Immun. 62:1252–1261. 275. Skurnik, M., R. Venho, J. A. Bengoechea, and I. Moriyon. 1999. The lipopolysaccharide outer core of Yersinia enterocolitica serotype O:3 is required for virulence and plays a role in outer membrane integrity. Mol. Microbiol. 31:1443–1462. 276. Slee, K. J., and C. Button. 1990. Enteritis in sheep and goats due to Yersinia enterocolitica infection. Aust. Vet. J. 67:396–398. 277. Slee, K. J., and C. Button. 1990. Enteritis in sheep, goats and pigs due to Yersinia pseudotuberculosis infection. Aust. Vet. J. 67:320–322. 278. Slee, K. J., and N. W. Skilbeck. 1992. Epidemiology of Yersinia pseudotuberculosis and Y. enterocolitica infections in sheep in Australia. J. Clin. Microbiol. 30:712–715. 279. Smith, M. G. 1992. Destruction of bacteria on fresh meat by hot water. Epidemiol. Infect. 109:491–496. 280. Snellings, N. J., M. Popek, and L. E. Lindler. 2001. Complete DNA sequence of Yersinia enterocolitica serotype O:8 lowcalcium-response plasmid reveals a new virulence plasmidassociated replicon. Infect. Immun. 69:4627–4638.

SMP_Food Microbiology_CH14.indd

374

281. Spanier, B., M. Starke, F. Higel, S. Scherer, and T. M. Fuchs. 2010. Yersinia enterocolitica infection and tcaAdependent killing of Caenorhabditis elegans. Appl. Environ. Microbiol 76:6277–6285. 282. Stern, N. J., M. D. Pierson, and A. W. Kotula. 1980. Effects of pH and sodium chloride on Yersinia entero­ colitica growth at room and refrigeration temperatures. J. Food Sci. 45:64–67. 283. Stuart, P. M., and J. G. Woodward. 1992. Yersinia en­ terocolitica produces superantigenic activity. J. Immunol. 148:225–233. 284. Sulakvelidze, A. 2000. Yersiniae other than Y. enteroco­ litica, Y. pseudotuberculosis, and Y. pestis: the ignored species. Microbes Infect. 2:497–513. 285. Sulakvelidze, A., A. Kreger, A. Joseph, R. M. RobinsBrowne, A. Fasano, G. Wauters, N. Harnett, L. DeTolla, and J. G. Morris, Jr. 1999. Production of enterotoxin by Yersinia bercovieri, a recently identified Yersinia enterocolitica-like species. Infect. Immun. 67:968–971. 286. Tacket, C. O., J. Ballard, N. Harris, J. Allard, C. Nolan, T. Quan, and M. L. Cohen. 1985. An outbreak of Yersinia enterocolitica infections caused by contaminated tofu (soybean curd). Am. J. Epidemiol. 121:705–711. 287. Tacket, C. O., B. R. Davis, G. P. Carter, J. F. Randolph, and M. L. Cohen. 1983. Yersinia enterocolitica pharyngitis. Ann. Intern. Med. 99:40–42. 288. Tacket, C. O., J. P. Narain, R. Sattin, J. P. Lofgren, C. Konigsberg, Jr., R. C. Rendtorff, A. Rausa, B. R. Davis, and M. L. Cohen. 1984. A multistate outbreak of infections caused by Yersinia enterocolitica transmitted by pasteurized milk. JAMA 251:483–486. 289. Takao, T., T. Hitouji, S. Aimoto, Y. Shimonishi, S. Hara, T. Takeda, Y. Takeda, and T. Miwatani. 1983. Amino acid sequence of a heat-stable enterotoxin isolated from enterotoxigenic Escherichia coli strain 18D. FEBS Lett. 152:1–5. 290. Takao, T., Y. Shimonishi, M. Kobayashi, O. Nishimura, M. Arita, T. Takeda, T. Honda, and T. Miwatani. 1985. Amino acid sequence of heat-stable enterotoxin produced by Vibrio cholerae non-01. FEBS Lett. 193:250–254. 291. Takao, T., N. Tominaga, S. Yoshimura, Y. Shimonishi, S. Hara, T. Inoue, and A. Miyama. 1985. Isolation, primary structure and synthesis of heat-stable enterotoxin produced by Yersinia enterocolitica. Eur. J. Biochem. 152:199–206. 292. Tauxe, R. V., J. Vandepitte, G. Wauters, S. M. Martin, V. Goossens, P. de Mol, R. Van Noyen, and G. Thiers. 1987. Yersinia enterocolitica infections and pork: the missing link. Lancet i:1129–1132. 293. Tennant, S. M., T. H. Grant, and R. M. Robins-Browne. 2003. Pathogenicity of Yersinia enterocolitica biotype 1A. FEMS Immunol. Med. Microbiol. 38:127–137. 294. Tennant, S. M., N. A. Skinner, A. Joe, and R. M. Robins-Browne. 2005. Homologues of insecticidal toxin complex genes in Yersinia enterocolitica biotype

Manila Typesetting Company

11/08/2012 06:58AM

14.  Yersinia enterocolitica

375

1A and their contribution to virulence. Infect. Immun. 73:6860–6867. 295. Thompson, J. S., and M. J. Gravel. 1986. Family outbreak of gastroenteritis due to Yersinia enterocolitica serotype 0:3 from well water. Can. J. Microbiol. 32:700–701. 296. Tipple, M. A., L. A. Bland, J. J. Murphy, M. J. Arduino, A. L. Panlilio, J. J. Farmer III, M. A. Tourault, C. R. Macpherson, J. E. Menitove, and A. J. Grindon. 1990. Sepsis associated with transfusion of red cells contaminated with Yersinia enterocolitica. Transfusion 30:207–213. 297. Toivanen, P., and A. Toivanen. 1994. Does Yersinia induce autoimmunity? Int. Arch. Allergy Immunol. 104:107–111. 298. Tomer, Y., and T. F. Davies. 1993. Infection, thyroid disease, and autoimmunity. Endocr. Rev. 14:107–120. 299. Toora, S., E. Budu-Amoako, R. F. Ablett, and J. Smith. 1992. Effect of high-temperature short-time pasteurization, freezing and thawing and constant freezing, on the survival of Yersinia enterocolitica in milk. J. Food Prot. 55:803–805. 300. Torruellas, J., M. W. Jackson, J. W. Pennock, and G. V. Plano. 2005. The Yersinia pestis type III secretion needle plays a role in the regulation of Yop secretion. Mol. Microbiol. 57:1719–1733. 301. Trasak, C., G. Zenner, A. Vogel, G. Yuksekdag, R. Rost, I. Haase, M. Fischer, L. Israel, A. Imhof, S. Linder, M. Schleicher, and M. Aepfelbacher. 2007. Yersinia protein kinase YopO is activated by a novel G-actin binding process. J. Biol. Chem. 282:2268–2277. 302. Trcek, J., K. Berschl, and K. Trulzsch. 2010. In vivo analysis of Yersinia enterocolitica infection using luxCDABE. FEMS Microbiol. Lett. 307:201–206. 303. Trosky, J. E., A. D. Liverman, and K. Orth. 2008. Yersinia outer proteins: Yops. Cell. Microbiol. 10:557–565. 304. Truche, C., and J. Bauche. 1933. La pseudotuberculose chez la poule et le faisan. Bull. Acad. Vet. France 1:43–46. 305. Tsubokura, M., S. Aleksic, H. Fukushima, G. Schulze, K. Someya, T. Sanekata, K. Otsuki, T. Nagano, Y. Kuratani, M. Inoue, et al. 1993. Characterization of Yersinia pseudotuberculosis serogroups O9, O10 and O11; subdivision of O1 serogroup into O1a, O1b, and O1c subgroups. Int. J. Med. Microbiol. Virol. Parasitol. Infect. Dis. 278:500–509. 306. Tsubokura, M., K. Otsuki, K. Sato, M. Tanaka, T. Hongo, H. Fukushima, T. Maruyama, and M. Inoue. 1989. Special features of distribution of Yersinia pseudotuberculosis in Japan. J. Clin. Microbiol. 27:790–791. 307. Reference deleted. 308. Tzipori, S., R. Robins-Browne, and J. K. Prpic. 1987. Studies on the role of virulence determinants of Yersinia enterocolitica in gnotobiotic piglets. Contrib. Microbiol. Immunol. 9:233–238.

SMP_Food Microbiology_CH14.indd

375

309. Une, T. 1977. Studies on the pathogenicity of Y. entero­ colitica. I. Experimental infection in rabbits. Microbiol. Immunol. 21:349–363. 310. Vahamiko, S., M. A. Penttinen, and K. Granfors. 2005. Aetiology and pathogenesis of reactive arthritis: role of non-antigen-presenting effects of HLA-B27. Arthritis Res.Ther. 7:136–141. 311. Van Loghem, J. J. 1944. The classification of the plague bacillus. Antonie van Leeuwenhoek J. Microbiol. Serol. 10:15–16. 312. Van Noyen, R., R. Selderslaghs, J. Bekaert, G. Wauters, and J. Vandepitte. 1991. Causative role of Yersinia and other enteric pathogens in the appendicular syndrome. Eur. J. Clin. Microbiol. Infect. Dis. 10:735–741. 313. Venecia, K., and G. M. Young. 2005. Environmental regulation and virulence attributes of the Ysa type III secretion system of Yersinia enterocolitica biovar 1B. Infect. Immun. 73:5961–5977. 314. Verhaegen, J., J. Charlier, P. Lemmens, M. Delmee, R. van Noyen, L. Verbist, and G. Wauters. 1998. Surveillance of human Yersinia enterocolitica infections in Belgium: 1967–1996. Clin. Infect. Dis. 27:59–64. 315. Viboud, G. I., and J. B. Bliska. 2001. A bacterial type III secretion system inhibits actin polymerization to prevent pore formation in host cell membranes. EMBO J. 20:5373–5382. 316. Viboud, G. I., and J. B. Bliska. 2005. Yersinia outer proteins: role in modulation of host cell signaling responses and pathogenesis. Annu. Rev. Microbiol. 59:69–89. 317. Viboud, G. I., E. Mejia, and J. B. Bliska. 2006. Comparison of YopE and YopT activities in counteracting host signalling responses to Yersinia pseudotubercu­ losis infection. Cell. Microbiol. 8:1504–1515. 318. Viboud, G. I., S. S. So, M. B. Ryndak, and J. B. Bliska. 2003. Proinflammatory signalling stimulated by the type III translocation factor YopB is counteracted by multiple effectors in epithelial cells infected with Yersinia pseudo­ tuberculosis. Mol. Microbiol. 47:1305–1315. 319. Vincent, P., A. Leclercq, L. Martin, J. M. Duez, M. Simonet, and E. Carniel. 2008. Sudden onset of pseudotuberculosis in humans, France, 2004–05. Emerg. Infect. Dis. 14:1119–1122. 320. Vincent, P., E. Salo, M. Skurnik, H. Fukushima, and M. Simonet. 2007. Similarities of Kawasaki disease and Yersinia pseudotuberculosis infection epidemiology. Pediatr. Infect. Dis. J. 26:629–631. 321. Virdi, J. S., and P. Sachdeva. 2005. Molecular heterogeneity in Yersinia enterocolitica and ‘Y. enterocoliticalike’ species: implications for epidemiology, ­typing and taxonomy. FEMS Immunol. Med. Microbiol. 45:1–10. 322. Virto, R., D. Sanz, I. Alvarez, Condon, and J. Raso. 2005. Inactivation kinetics of Yersinia enterocolitica by citric and lactic acid at different temperatures. Int. J. Food Microbiol. 103:251–257. 323. Virto, R., D. Sanz, I. Alvarez, S. Condon, and J. Raso. 2005. Comparison of the chlorine inactivation of Yersinia enterocolitica in chlorine demand and demandfree systems. J. Food Prot. 68:1816–1822.

Manila Typesetting Company

11/08/2012 06:58AM

1� 2� 3� 4� 5� 6� 7� 8� 9� 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

1� 2� 3� 4� 5� 6� 7� 8� 9� 10� 11� 12� 13� 14� 15� 16� 17� 18� 19� 20� 21� 22� 23� 24� 25� 26� 27� 28� 29� 30� 31� 32� 33� 34� 35� 36� 37� 38� 39� 40� 41� 42� 43� 44� 45� 46� 47� 48� 49� 50� 51� 52�

Foodborne Pathogenic Bacteria

376 324. Wachtel, M. R., and V. L. Miller. 1995. In vitro and in vivo characterization of an ail mutant of Yersinia en­ terocolitica. Infect. Immun. 63:2541–2548. 325. Warren, S. M., and G. M. Young. 2005. An aminoterminal secretion signal is required for YplA export by the Ysa, Ysc, and flagellar type III secretion systems of Yersinia enterocolitica biovar 1B. J. Bacteriol. 187:6075–6083. 326. Wauters, G., S. Aleksic, J. Charlier, and G. Schulze. 1991. Somatic and flagellar antigens of Yersinia enterocolitica and related species. Contrib. Microbiol. Immunol. 12:239–243. 327. Wauters, G., M. Janssens, A. G. Steigerwalt, and D. J. Brenner. 1988. Yersinia mollaretii sp. nov. and Yersinia bercovieri sp. nov., formerly called Yersinia enterocolitica biogroups 3A and 3B. Int. J. Syst. Bacteriol. 38:424–429. 328. Reference deleted. 329. Wauters, G., K. Kandolo, and M. Janssens. 1987. Revised biogrouping scheme of Yersinia enterocolitica. Contrib. Microbiol. Immunol. 9:14–21. 330. Weinberg, E. D. 1984. Iron withholding: a defense against infection and neoplasia. Physiol. Rev. 64:65–102. 331. Wiley, D. J., R. Nordfeldth, J. Rosenzweig, C. J. DaFonseca, R. Gustin, H. Wolf-Watz, and K. Schesser. 2006. The Ser/Thr kinase activity of the Yersinia protein kinase A (YpkA) is necessary for full virulence in the mouse, mollifying phagocytes, and disrupting the eukaryotic cytoskeleton. Microb. Pathog. 40:234–243. 332. Williams, A. W., and S. C. Straley. 1998. YopD of Yersinia pestis plays a role in negative regulation of the low-calcium response in addition to its role in translocation of Yops. J. Bacteriol. 180:350–358. 333. Woestyn, S., A. Allaoui, P. Wattiau, and G. R. Cornelis. 1994. YscN, the putative energizer of the Yersinia Yop secretion machinery. J. Bacteriol. 176:1561–1569. 334. Wong, K. W., and R. R. Isberg. 2005. Emerging views on integrin signaling via Rac1 during invasin-promoted bacterial uptake. Curr. Opin. Microbiol. 8:4–9. 335. Wulff-Strobel, C. R., A. W. Williams, and S. C. Straley. 2002. LcrQ and SycH function together at

SMP_Food Microbiology_CH14.indd

376

336.

337.

338.

339.

340.

341. 342.

343.

344.

the Ysc type III secretion system in Yersinia pestis to impose a hierarchy of secretion. Mol. Microbiol. 43:411–423. Wuytack, E. Y., A. M. Diels, and C. W. Michiels. 2002. Bacterial inactivation by high-pressure homogenisation and high hydrostatic pressure. Int. J. Food Microbiol. 77:205–212. Yang, J., M. Tauschek, and R. M. Robins-Browne. 2011. Control of bacterial virulence by AraC-like regulators that respond to chemical signals. Trends Microbiol. 19:128–135. doi:10.1016/j.tim.2010.12.00. Yang, Y., and R. R. Isberg. 1997. Transcriptional regulation of the Yersinia pseudotuberculosis pH6 antigen adhesin by two envelope-associated components. Mol. Microbiol. 24:499–510. Yoshino, K., T. Takao, X. Huang, H. Murata, H. Nakao, T. Takeda, and Y. Shimonishi. 1995. Characterization of a highly toxic, large molecular size heat-stable enterotoxin produced by a clinical isolate of Yersinia enteroco­ litica. FEBS Lett. 362:319–322. Young, B. M., and G. M. Young. 2002. Evidence for targeting of Yop effectors by the chromosomally encoded Ysa type III secretion system of Yersinia enterocolitica. J. Bacteriol. 184:5563–5571. Young, G. M. 2007. The Ysa type 3 secretion system of Yersinia enterocolitica biovar 1B. Adv. Exp. Med. Biol. 603:286–297. Young, G. M., D. Amid, and V. L. Miller. 1996. A bifunctional urease enhances survival of pathogenic Yersinia enterocolitica and Morganella morganii at low pH. J. Bacteriol. 178:6487–6495. Zhang, Y., A. T. Ting, K. B. Marcu, and J. B. Bliska. 2005. Inhibition of MAPK and NF-kappa B pathways is necessary for rapid apoptosis in macrophages infected with Yersinia. J. Immunol. 174:7939–7949. Zink, D. L., J. C. Feeley, J. G. Wells, C. Vanderzant, J. C. Vickery, W. D. Roof, and G. A. O’Donovan. 1980. Plasmid-mediated tissue invasiveness in Yersinia entero­ colitica. Nature 283:224–226.

Manila Typesetting Company

11/08/2012 06:58AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch15

15

Rachel Binet Keith A. Lampel

Shigella Species

Foodborne infections with Shigella species are an important source of illness in both economically developed and developing countries. A considerable amount of research has been conducted on Shigella since shigellosis (bacillary dysentery) was first differentiated from amoebic dysentery in 1887 and the etiologic agent was isolated and described by Shiga during an outbreak in Japan 10 years later. Most cases of shigellosis are not transmitted by contaminated food, and as such the foodborne aspects of shigellosis are often a neglected area of study across the international food safety community. Shigella spp. are endemic worldwide because of their ease of transmission. Fecally contaminated food and water can lead to rapid onset of large-scale epidemics of shigellosis with a high mortality rate in underdeveloped countries. Compounding the extent of this pathogen’s effect on global health are the current lack of a protective vaccine and the extraordinary ease with which these bacteria acquire antibiotic resistance. The globalization and importation of the food supply present additional risks that are being measured through the international cooperation among the World Health Organization (WHO) and public regulatory agencies (63).

This chapter presents to food microbiologists important features of Shigella spp., the disease they cause, and the impact that these pathogens have with respect to food safety. Diagnosis, epidemiology, ecology, modes of transmission, and examples of recent foodborne outbreaks are presented, along with the current understanding of the genetics of Shigella pathogenesis, the genes involved in causing disease, and how they are regulated. No single review can be completely comprehensive, so the reader is encouraged to refer to several excellent recent reviews for additional information (62, 76, 83, 129, 132, 137, 146, 169).

CHARACTERISTICS OF THE ORGANISM

Classification and Biochemical Characteristics

Shigella is a gram-negative, nonsporulating, nonmotile, facultative anaerobic, rod-shaped bacterium. Adopted as a genus in 1950, Shigella is subgrouped into four species based on biochemical, serological, and clinical phenotypic differences: S. dysenteriae (group A, 15 serotypes); S. flexneri (group B, 14 classical serotypes and subserotypes); S. boydii (group C, 19 serotypes); and S.

Rachel Binet and Keith A. Lampel, Center for Food Safety and Applied Nutrition, Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740.

377

SMP_Food Microbiology_CH15.indd

377

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

378 sonnei (group D, 1 serotype). S. boydii 13 was removed from the Shigella genus in 2005 (26), but a new serotype, S. boydii 20 serovar nov., was added in the same year (174). Serologic typing is based only on differences in the O (lipopolysaccharide [LPS]) antigen because Shigella cells lack the H (flagellar) and K (capsular) antigens. Consequently, Shigella serotyping is affected by the bacterial transition from the smooth form to the rough form, a phenotype frequently seen with S. sonnei, which harbors the LPS-encoding genes on a plasmid that is relatively unstable during bacterial culture. Plasmidencoded genes are also necessary for synthesis of the LPS O side chain in S. dysenteriae 1 (19). S. sonnei, S. flexneri 2a, and S. dysenteriae type 1 are the most common species isolated from cases of shigellosis, the latter being responsible for most of the deadly cases due to the unique expression of the cytotoxic Shiga toxin common to Shiga toxin-producing pathogenic Escherichia coli, including the enterohemorrhagic E. coli O157:H7 strain. As members of the family Enterobacteriaceae, Shigella spp. are nearly identical genetically to organisms of the genus Escherichia and are closely related to the salmonellae. Nevertheless, they are biochemically distinguished from other enterics by their inability to ferment acetate, mucate, and lactose (although some strains of S. sonnei may ferment mucate or lactose upon prolonged incubation); lack of citric acid, inositol, salicin, or adonitol utilization as a sole carbon source; and inability to synthesize lysine decarboxylase (51). The four species require nicotinic acid for growth in a minimal synthetic medium, are oxidase negative, do not produce H2S, and do not produce gas from glucose, except for S. flexneri 6 and S. boydii 14. S. dysenteriae strains have the additional property of not being able to ferment mannitol. S. dysenteriae type 1 expresses an active b-galactosidase but does not produce catalase, an extremely rare feature among Enterobacteriaceae. Although S. sonnei strains have only one serotype, they can be subdivided into biovars on the basis of their ability to hydrolyze O-nitrophenyl b-d-galactyl pyranoside (ONPG), xylose, and rhamnose. The traditional classification of shigellae has been challenged by the advent of comparative genomic techniques. Nucleotide sequencing of multiple conserved genes has indicated that Shigella emerged from E. coli. Most Shigella strains can be separated into three main clusters that appear to have originated from multiple E. coli ancestors. Cluster 3 contains S. boydii 12 and all S. flexneri serotypes except serotype 6, and cluster 2 contains S. dysenteriae type 2 and seven different serotypes of S. boydii. Cluster 1 can be further subdivided into

SMP_Food Microbiology_CH15.indd

378

three subclusters that contain strains from mainly one serogroup: S. dysenteriae for subgroup 1 and S. boydii for subgroups 2 and 3. Whereas S. sonnei and S. dysenteriae types 1, 8, and 10 are isolated clones with E. coli, S. dysenteriae type 1 also appears to share an ancestor with E. coli O157:H7. Thus, while modern genotyping does not perfectly match with traditional serotyping, genotyping and serogrouping are correlated (176). Modern taxonomy also revealed that enteroinvasive E. coli (EIEC) strains were more closely related to Shigella than to commensal E. coli strains. EIEC strains cause dysentery like Shigella and may share identical O antigens with Shigella, but traditional classification places them as a subgroup of E. coli because many are motile and are lactose, mucate, and/or acetate fermenters. It is now commonly accepted that the EIEC phenotype arose several times independently from different E. coli ancestors and represents an intermediate stage in the convergent evolution leading towards the more contagious and virulent Shigella (87) (Fig.15.1).

Diagnosis

Shigella spp. are not particularly fastidious in their growth requirements, and in most cases, the organisms are routinely cultivated in the laboratory on artificial media. Cultures of Shigella are easily isolated and grown from analytical samples including water and clinical samples. In the latter case, Shigella are present in the feces of patients in concentrations ranging from 103 to1010 viable bacteria per gram of stool during the acute phase of infection. Identification is readily accomplished using culture media, biochemical analysis, and serological typing. Shigella spp. are shed in the feces and continue to be isolated from convalescent patients (102 to 103 cells per gram of stool) for weeks or longer after the initial infection. They have the reputation of being more fastidious and difficult to cultivate than other enteric bacteria, particularly during later stages of the illness. However, this relates to the lack of a selective enrichment broth for Shigella, leading to it being outgrown by the resident bacterial fecal population. Use of the semiselective agar media Hektoen, SalmonellaShigella (which is unfortunately inhibitory for S. dysenteriae type 1), and xylose-lysine-deoxycholate or recently developed chromogenic agars can increase the chance of isolating Shigella (181). Because the time from the initial clinical diagnosis indicating a potential outbreak to the start of an investigation to determine the cause of that outbreak can be considerable, identifying the causative agent in a timely manner can be challenging. This time lag can be exacerbated when the etiological agent has a narrow

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

379 Farfan et al. (45) recently identified two markers that specifically differentiate between S. flexneri, S. sonnei, and other diarrheagenic E. coli.

Epidemiology

Figure 15.1  Genetic steps that contribute to the formation of pathogenic Shigella species from an E. coli ancestor. These include the acquisition of the virulence plasmid, several pathogenicity islands (SHI-1 and SHI-2 are shown), and the loss of ancestral traits incompatible with virulence (e.g., cadA and avl; often referred to as antivirulence genes) (103, 132). doi:10.1128/9781555818463.ch15f1

e­ nvironmental niche and survivability band, as does Shigella. Despite it being saprophytic, Shigella has a very specific host range, essentially residing only in the intestine of humans and certain higher primates. In the latter case, it is unclear whether this is a natural reservoir or if the infections are acquired from humans. Once excreted, Shigella cells are very sensitive to environmental conditions and die rapidly, especially when exposed to direct sunlight or dried. Isolation of Shigella from foods is therefore a challenge. Many additional characteristics of the food including its composition, physical parameters such as pH and salt, and the natural resident microbial flora members that compete for nutrients may affect the successful recovery of shigellae. Considering that Shigella may be present in low numbers or in poor physiological state in the suspected food samples, special enrichment procedures are often required for successful detection of foodborne Shigella (5). Although molecular detection does not generally prove that an organism is alive and competent to cause disease, PCR techniques may be considered a more sensitive and specific technique than conventional culture techniques. Several single, nested, and multiplex PCR assays have been designed to routinely amplify marker(s) present in single or multiple copies (i.e., ipaH) in Shigella and EIEC bacterial genomes (36). However, they generally do not differentiate between the four Shigella spp., EIEC, and/or serotypes. Targeting the Oserotype-specific gene(s) using PCR (64) or microarrays (95) can bring more-discriminatory power to the tests.

SMP_Food Microbiology_CH15.indd

379

Approximately 164.7 million Shigella diarrheal episodes and 1.1 million deaths occur each year throughout the world, with 99% of those occurring in developing countries (80% in Asia) (85). Although most cases of shigellosis are due to person-to-person transmission, outbreaks commonly result from food and/or water contamination. Episodes of shigellosis seem to follow seasonal variations in certain countries. In arid countries such as Egypt, transmission peaks in the hot dry season, mainly due to the consumption of contaminated water and decreased personal hygiene in times of water shortage (51). Conversely, the peak is in the rainy season in China and Thailand (9) as a result of water-washedrelated transmission during heavy rains. These reflect the association of shigellosis with unsanitary conditions that foster fecal transmission. In developed countries, the highest incidence of shigellosis generally occurs during the warmer months of the year, when consumption of raw foods and fresh fruits and vegetables and the use of recreational facilities are the highest. The distribution of Shigella spp. varies in different parts of the world. For example, S. boydii is not frequently encountered outside the Indian subcontinent, where it was first identified. Ten years ago, the proportions of S. flexneri, S. sonnei, S. boydii, and S. dysenteriae were estimated to be respectively 60% ­(predominantly serotype 2a, followed by 1b, 3a, 4a, and 6), 15%, 6%, and 6% (30% of S. dysenteriae cases were type 1), respectively, in developing countries; and 16% (predominantly serotype 3a, followed by 1b, 1c, 2a, and Y variant [4]), 77%, 2%, and 1% in industrialized countries, respectively, with nearly one-half of the cases reported among travelers (85). Travelers returning from India and neighboring countries and from East, West, or North Africa were at a higher risk for shigellosis (42). Importantly, the distribution of Shigella species also seems to evolve with time and with the economy of a region (115). S. dysenteriae dominated in early parts of the 20th century but was replaced by S. flexneri in the 1930s and 1940s in the absence of epidemics and more recently by S. sonnei in developed countries. When the economy of a country improves, S. sonnei becomes responsible for the majority of shigellosis outbreaks, as seen since 2000 in Thailand (9) and Iran (139). While studying diarrheal disease caused by Shigella in six Asian countries, von Seidlein et al. (167) found a common link between socioeconomic factors with particular Shigella spp., e.g.,

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

380 in less-resource-rich countries the predominant species was S. flexneri. In addition, they observed that the variation of serotype differed temporally and with regard to the geographic location, implying that the change in distribution may affect the effectiveness of any potential vaccine to use in areas of the world that have such shifts in serotypes. Shigella spp. are capable of causing disease in otherwise healthy individuals. Certain populations, however, may be more predisposed to infection and disease due to the nature of transmission of the organism. Shigellae are transmitted by the fecal-oral route, via direct personto-person contact (fingers), food, water, and inanimate objects. More than two-thirds of all episodes of shigellosis are seen in children between 1 and 5 years old, who tend to explore their environment with their mouths (7, 85). Sustained endemic transmission and epidemics of bacillary dysentery are facilitated when crowding and poor sanitation conditions create an environment for direct fecal-oral contamination, such as day care centers, custodial institutions, mental hospitals, and nursing homes, or mass displacement and gathering such as refugee camps in time of war or political turmoil. In the past 3 decades, major outbreaks have occurred in Africa, South Asia, and Central America, causing up to 20,000 deaths in only 1 month (90). Oral/anal and digital/anal sexual practices favor the transmission of Shigella and contribute to outbreaks of shigellosis among the homosexual community (33). Additionally, HIV-positive individuals may present more severe and persistent forms of shigellosis. The disease is communicable as long as an infected person excretes the organism in the stool. Secondary attack rates following exposure to the primary case can be as high as 40% among household contacts (7). The ease with which bacillary dysentery is transmitted from person to person is likely due to the very low infectious inoculum required.

Ecology

Shigella and EIEC are considered the only obligate pathogenic strains of E. coli and have an extremely narrow host range. Only humans and higher primates can develop shigellosis, but there is no evidence that the disease occurs naturally in the wild in nonhuman primates without prior contact with humans (78). While Shigella is usually excreted for a few weeks after the illness, more than 10% of infected individuals, notably children, excrete shigellae for longer than 10 weeks (99). In areas of high endemicity, up to 50% of all Shigella infections may be asymptomatic, and asymptomatic carriers of Shigella may exacerbate the maintenance and spread

SMP_Food Microbiology_CH15.indd

380

of this pathogen in developing countries. Two studies, one in Bangladesh (30) and the other in Mexico (55), showed that Shigella was isolated from stool samples collected from asymptomatic children under the age of 5 years. The presence and persistence of Shigella in the environment are not well documented compared to other Enterobacteriaceae. While the E. coli/Shigella ancestors gained genes that allowed them to access a new niche, i.e., the intracellular host milieu, positive selection subsequently fashioned the genome to allow for virulence factors to be maintained (43) (Fig. 15.1). Accordingly, the evolution of Shigella was shaped by a massive loss of genes: 543 in S. dysenteriae Sd197, 347 to 371 in S. flexneri, 366 in S. boydii Sb227, and 255 in S. sonnei Ss046 (60). The majority of the lost phenotypes include surface appendages, cell motility, transport, and bacterial metabolism, reflecting the organism’s specialization to the intracellular pathogenic lifestyle (31, 103, 162). Because Shigella cells lack flagella, fimbriae, and curli appendages and do not produce poly-b-1,6-N-acetylglucosamine or colonic acid polysaccharides, they are not expected to form biofilms to protect them from damaging environmental factors (12). Nevertheless, their incorporation into preestablished mixed-species biofilm communities has been suggested (56). Ingestion of contaminated water is a recognized mode of transmission for shigellosis. Nevertheless, S. dysenteriae does not survive in water for more than 2 to 3 days. S. flexneri and S. sonnei can survive from 6 to 47 days and from 35 to 39 days, respectively (137). The community of organisms within fertile soil is biologically diverse and includes many bacteriovorous organisms, including amoebae and nematodes. Over the last few years, these environmental predators have gained increasing attention as their roles as hosts for pathogenic bacteria, including Legionella and Chlamydia, are being uncovered (62). S. sonnei and S. dysenteriae can also survive phagocytosis by Acanthamoebae and grow in this group of ubiquitous free-living amoebae (70, 144). Considering that EIEC can paralyze and kill the nematode Caenorhabditis elegans (6), agents of bacillary dysentery may well use amoebae and worms as environmental reservoirs.

Shigella in foods

Food Contamination

Shigella can be transmitted by consumption of raw or processed food. Generally, poor personal hygiene practices by food workers at the point of final preparation and food service are the major factor for food con-

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

381

tamination (97). It has been recommended that infected personnel should be monitored until stool samples are negative for Shigella (7). An extensive epidemiological investigation, for example, traced the source of a 2000 S. sonnei outbreak involving 406 diarrheal patients across 10 states in the United States to an employee participating in the preparation of the contaminated commercial five-layer dip (81). On at least one occasion, food (muffin and doughnut) deliberately contaminated with S. dysenteriae type 2 was used as a biological weapon, leading to a dozen cases of shigellosis in Texas in 1996 (38). Shigellae from infected individuals can spread by several routes including food, fingers, feces, flies, and fomites. S. sonnei can survive for over 3 h on fingers, and S. dysenteriae type 1 can be recovered for up to 1 h (65, 68). S. flexneri can survive in feces for 12 days at 25°C (120). Flies can transmit the bacteria from fecal matter to foods. Utensils used in food preparation can also act as a vehicle for food contamination. For example, an early study by Nakamura (119) showed that S. sonnei could survive on metal utensils for more than 2 to 28 days at 15°C and up to 13 days at 37°C. Improper storage of contaminated foods is the second most common factor accounting for foodborne outbreaks due to Shigella. Other contributing factors are inadequate cooking, contaminated equipment, and food obtained from unsafe sources. Consumption of raw vegetables harvested in fields where sewage is used as fertilizer or wastewater is used for irrigation can cause contamination­ (20).

Survival and Growth in Foods

A number of raw or undercooked foods have been linked to shigellosis outbreaks including lettuce, parsley, bean dip, cold sandwiches, potato salad, tofu salad, egg salad, hamburgers, tomatoes, and oysters (57). Establishments that served these foods ranged from homes to restaurants, camps, picnics, schools, airlines, sorority houses, and military mess halls. In many cases, the food source was not determined (53). Because Shigella spp. are not generally associated with specific foods, routine microbiological examinations of foods to identify these pathogens are not usually performed. Many outbreaks are attributed to foods that require extensive manipulation during preparation and service, and investigation of food handling practices is generally more productive than microbiological examination of foods. As discussed previously, the phylosphere (the total above-ground surfaces of plants) is not the natural habitat for Shigella. Yet, rapid growth of S. sonnei has been documented in shredded lettuce held at room tempera-

SMP_Food Microbiology_CH15.indd

381

ture (34). Similarly, viable S. flexneri 2a survived for at least 72 h on damaged tomatoes (140), while S. sonnei did not survive desiccation on intact tomato surfaces (170). Islam et al. (67) showed that S. flexneri could grow on the surface of cucumbers for up to 24 h at 25°C and 37°C and survived for more than 72 h at 5°C. Cucumber skin is more textured than the waxy tomato skin and may provide sites where the bacteria are more protected from desiccation (20). S. flexneri grew extremely well at 25°C and 37°C in sterilized milk, cooked rice, lentil soups, and cooked beef or fish and can persist for at least 3 days in all these foods but fish when stored at 5°C (67). Additionally, survival of S. sonnei in potato salad and raw ground beef at cold temperature has been observed for at least 28 days and 9 days, respectively (170). Shigella cells can survive a temperature range of −20°C to room temperature; however, they survive for a longer time in foods stored frozen or at refrigeration temperatures than at room temperature. For example, two strains of S. boydii 18 were inoculated into bean salad and stored at 4 or 23°C (2). No growth was observed at the lower temperatures, but both strains survived. At 23°C, an increase of 2 orders of magnitude was noted over 2 days, after which the number of viable cells declined. Shigella tolerance to the low pH of the stomach contributes to its low infectious dose and its ability to survive in moderately acidic foods such as fruits and vegetables. S. flexneri and S. sonnei can grow in the laboratory in rich media acidified to pH 4.75 and 4.5, respectively, and can survive in media with a pH range of 2 to 3 for a few hours. S. sonnei and S. flexneri could be recovered from tomato juice (pH 3.9 to 4.1) and apple juice (pH 3.3 to 3.4) after a few days at room temperature (8). Lowering the temperature increases the survival rate of Shigella in response to low-pH and high-salt stress. S. flexneri can grow at 37°C at pH 6 in the presence of 6% NaCl or at pH 5 with only up to 2% salt (179). The tolerance of S. flexneri to salt suggests that it may survive in salty foods such as pickled vegetables, caviar, pickled herring, cured ham, and certain cheeses over a long period of time. Whether S. sonnei survives better than S. flexneri in foods is still a matter of debate (180).

Examples of Foodborne Outbreaks

Foodborne outbreaks are not always easily recognizable when symptoms are mild and cases go unreported (57). As an example, a shigellosis outbreak caused by S. sonnei in Spain led to the infection of 60 preschool pupils and 28 family members over a 2-month period because it was undetected during the first month of occurrence (74). Even when epidemiological methods may strongly

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

382 implicate a common food source, rapid turnover of produce may impair the ability to trace it back for analysis. Moreover, as discussed earlier, Shigella spp. are often not recovered and identified from foods by standard bacteriological methods. It is estimated that the number of foodborne cases in the United States is 6 to 81 million annually (111) and that the number of foodborne outbreaks may be significantly underreported because smaller outbreaks and geographically dispersed outbreaks may not reach the attention of public health officials. Based on the results of their active and passive surveillance systems (http://www.cdc.gov/ncidod/dbmd/ foodnet) (24), the CDC estimated that there were annually 448,000 cases of shigellosis in the United States, making it the third leading cause of foodborne outbreaks by bacterial pathogens (111). They also estimated that annually 600 deaths are caused by Shigella. From 1998 to 2004, 90 foodborne outbreaks of shigellosis were responsible for 5,324 illnesses reported to CDC by state and local health departments (57). Compared to previous years, foodborne shigellosis has been following a decreasing trend in the United States since 2000. Of the 17,468 laboratory-confirmed cases of foodborne infection in the United States in 2009, Shigella was identified in 1,849 cases, a 22% decrease compared to the period from 2006 to 2008 (28). A 2011 CDC publication on foodborne illnesses caused by 25 pathogens in the United States (147), which updated data from the Mead et al. (111) report, was based on laboratory-confirmed cases from 5 surveillance programs, some of which are indicated above. For illnesses caused by Shigella, most of the information cited in this new report used data from the surveillance programs covering the years 2005 to 2008. The estimated number of episodes attributed to this pathogen was given as 14,864 per year, of which nearly one-third were foodborne pathogen related. This report also indicated that the numbers of hospitalizations and deaths acquired via foods were 1,456 and 10, respectively. One of the striking features about foodborne outbreaks caused by shigellae is that contamination of foods usually is not at the processing plant; rather, the source is often traced to workers during the final preparation of the food. As evident from the examples below, these incidents can occur by improper food handling in homes, small town gatherings and picnics, and large-scale food service operations such as those on cruise ships and institutional feeding facilities. Disease is caused by the ingestion of these contaminated foods and can subsequently lead to rapid person-to-person dissemination. S. sonnei has caused the majority of foodborne shigellosis in developed countries; however, S. flexneri and S. dys-

SMP_Food Microbiology_CH15.indd

382

enteriae have been involved too, generally via the import of contaminated produce.

1989 and 1994—Shigellosis aboard Cruise Ships

In October 1989, 14% of the passengers and 3% of the crew members aboard a cruise ship reported having gastrointestinal symptoms (91). A multiple-antibiotic-resistant strain of S. flexneri 4a was isolated from several ill passengers and crew members. The source of this outbreak was identified as German potato salad. Contamination was introduced by infected food handlers, first in the country where the food was originally prepared and second by a member of the galley crew on the cruise ship. Another outbreak of shigellosis occurred in August 1994 on the cruise ship SS Viking Serenade (23). A total of 586 passengers (37%) and 30 crew members (4%) reported having diarrhea, and one death occurred. In this outbreak, S. flexneri 2a was isolated from patients. The suspected source of contamination was spring onions.

1990—Operation Desert Shield

Diarrheal diseases during a military operation can be a major factor in reducing troop readiness. In Operation Desert Shield, enteric pathogens were isolated from 214 U.S. soldiers, and of those, 113 cases were diagnosed with Shigella, S. sonnei being the most prevalent species­ isolated (66). Shigellosis accounted for more time lost from military duties and was responsible for more­severe morbidity than enterotoxigenic E. coli, the most common enteric pathogen isolated from U.S. troops in Saudi Arabia (66). The suspected source was contaminated fresh vegetables, notably lettuce.

2000—Five-Layer Bean Dip

An outbreak of shigellosis caused by the ingestion of contaminated five-layer (bean, salsa, guacamole, nacho cheese, and sour cream) party dip occurred in three West Coast states (25). The causative agent was S. sonnei and was isolated from at least 30 patients. The pathogen was found in just one layer (cheese) of the dip and was initially detected by a PCR assay targeting shigellae. Eventually, the pathogen was isolated by enrichment followed by plating on selective agar (81).

2001—Tomato-Linked Outbreak

Several restaurants in the New York area purchased tomatoes that were overripe and bruised from one distributor. These “special” tomatoes were identified as the source of an outbreak based on epidemiological data including the fact that only those restaurants that received

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

383

these special tomatoes were linked to the illnesses. Over 880 people reported being ill. S. flexneri 2a was isolated from several patrons of the restaurant as well as ill and asymptomatic employees from these restaurants. None of the workers from the restaurants reported being ill before the tomatoes arrived. A pulsed-field gel electrophoresis analysis of all the isolated S. flexneri strains linked all the restaurant patrons to this one outbreak (140).

2004—Unpasteurized Milk Curds

In October 2004, 41 persons from Vilnius, Lithuania, became sick after consuming unpasteurized milk curd produced in a local dairy. In addition to the patients, S. sonnei was isolated from the dairy owner and two of her household members (178).

2004—Airline Food

In August 2004, in-flight meals, likely raw carrot salad, contaminated with S. sonnei were served on 12 different flights departing from Honolulu International Airport in Hawaii and led to 116 cases of shigellosis in two foreign countries, a U.S. territory, and nearly one-half of the U.S. states. Between 300 and 1,500 passengers were potentially affected. The source of the outbreak was not identified with certainty.

2007—Raw Baby Corn

In August 2007, two shigellosis outbreaks in Denmark and Australia involving 215 and 20 cases, respectively, had a similar multidrug-resistant S. sonnei strain (93, 154). Epidemiological investigations linked the outbreaks to the consumption of raw baby corn imported from a common supply chain in Thailand. Inspection of a Thai collecting house revealed that the baby corn was placed directly on the ground and processed manually by locals without strict hygiene measures. Importantly, the water used to wash the corn contained only one-half of the recommended chlorine dose for disinfection (92).

2009—Sugar Peas

In April through June 2009, 37 persons became infected with S. sonnei in Denmark (118) and Norway (59) after consuming sugar peas imported from Kenya. S. sonnei was detected on the contaminated vegetable by PCR but could not be culture confirmed (59). During the same period, 35 to 47 persons got contaminated with S. dysenteriae type 2, presumably after consuming sugar peas imported from Kenya (96). However, the pathogen could not be identified from the leftover vegetable recovered at a patient’s house.

SMP_Food Microbiology_CH15.indd

383

CHARACTERISTICS OF DISEASE

Clinical Presentation

Disease caused by Shigella spp. is distinguished from disease caused by most of the other foodborne pathogens described in this volume in at least two important aspects: the production of bloody diarrhea or dysentery and the low infectious dose required to cause clinical symptoms. Bloody diarrhea refers to diarrhea in which the stools contain visible red blood. Dysentery has the same meaning, but the passage of bloody mucoid stools is accompanied by severe abdominal and rectal pain, cramps, and fever. While abdominal pain and diarrhea are experienced by nearly all patients with shigellosis, fever occurs only in about one-third, and frank blood in the stools in about 40% of the cases (156). The clinical picture of shigellosis ranges from a mild watery diarrhea to severe dysentery. All Shigella spp. can cause acute bloody diarrhea. The dysentery stage of the disease caused by Shigella spp. may or may not be preceded by watery diarrhea. This stage reflects the transient multiplication of bacteria as they pass through the small bowel. Jejunal secretions probably are not effectively reabsorbed in the colon due to transport abnormalities caused by bacterial invasion and destruction of the colonic mucosa. The dysentery stage of disease correlates with extensive bacterial colonization of the colonic mucosa. The bacteria invade the epithelial cells of the colon and spread from cell to cell but penetrate only as far as the lamina propria. Foci of individually infected cells produce microabscesses that coalesce, forming large abscesses and mucosal ulcerations. As the infection progresses, dead cells of the mucosal surface slough off, thus leading to the presence of blood, pus, and mucus in the stools. The incubation period for shigellosis is 1 to 7 days, but the illness usually begins within 3 days. Strains of S. dysenteriae type 1 cause the most severe disease, whereas S. sonnei strains produce the mildest. S. flexneri and S. boydii infections can be either mild or severe. Despite the severity of the disease, shigellosis is generally selflimiting. If left untreated, clinical illness usually persists for 1 to 2 weeks (although it may be as long as 1 month) and the patient recovers. In some cases, there can be protracted asymptomatic shedding of the pathogen.

Infectious Dose

As mentioned previously, an important aspect of Shigella pathogenesis is the extremely low 50% infective dose (ID50), i.e., the oral dose required to cause disease in 50% of healthy adult volunteers challenged with a virulent strain of the pathogen. The ID50 for S. flexneri, S.

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

384 sonnei, and S. dysenteriae varies, ranging from 10 CFU for virulent strains of S. dysenteriae and 140 CFU for S. flexneri to <500 CFU for S. sonnei (see reference 161 for a review on infective doses of pathogens). Volunteers became ill when doses as low as 200 cells were given (41), and this number is generally considered an average infectious dose for Shigella spp. One of the host’s innate defense factors is the low pH of the stomach, in which the gastric fluids can effectively kill microbial pathogens in 15 minutes (52). However, the influence of the food matrix that harbors bacterial pathogens, such as Shigella, may be significant in regard to surviving this harsh environment and indirectly affect the infectious dose. An example of this is when 180 CFU of a wild-type S. flexneri 2a strain was administered with bicarbonate buffer instead of milk, which increased the shigellosis rate to 43%, indicating that resistance to gastric juice partially accounts for the high bacterial infectivity (84). The low infectious dose of Shigella underscores the high communicability of bacillary dysentery and gives the disease great explosive potential for person-to-person spread as well as foodborne and waterborne outbreaks.

Complications

Shigellosis can be a very painful and incapacitating disease and is more likely to require hospitalization than many bacterial diarrheas. It is not usually life-threatening, and mortality is rare, except in malnourished children, immunocompromised individuals, and the elderly. However, complications arising from the disease include severe dehydration, intestinal perforation, toxic megacolon, septicemia, seizures, hemolytic uremic syndrome (HUS), and reactive arthritis (13). HUS is a rare but potentially fatal complication associated with infection by S. dysenteriae 1 (148). The syndrome is characterized by hemolytic anemia, thrombocytopenia, and acute renal failure. Epidemiologic studies suggest that Shiga toxin produced by S. dysenteriae 1 is the cause of HUS. This hypothesis is supported by the fact that HUS is also caused by strains of enterohemorrhagic E. coli, which produce high levels of Shiga toxin (see chapter 12). Shiga toxins cause HUS by entering the bloodstream and damaging vascular endothelial cells, particularly those in the kidney. Reactive arthritis is a postinfection sequela to shigellosis that is strongly associated with individuals of the HLA-B27 histocompatibility group (152). The syndrome is comprised of three symptoms, urethritis, uveitis, and reactive arthritis, with the last being the most dominant symptom. Infections caused by several other gram-negative enteric pathogens also can lead to this type of sterile inflammatory polyarthropathy (22).

SMP_Food Microbiology_CH15.indd

384

Treatment and Prevention

Although stool fluid losses are not as massive as with other bacterial diarrheas (e.g., cholera), the diarrhea associated with shigellosis combined with water loss from fever and the decreased water intake due to anorexia may result in severe dehydration (13). Oral intake of liquids can generally replace fluid losses, although intravenous rehydration may be required in very young and elderly patients. Antidiarrheal drugs are not recommended because they decrease intestinal mobility and may worsen shigellosis. On the other hand, antibiotics shorten the duration of dysentery and bacterial excretion in feces. However, there is some controversy regarding the use of antibiotics in treating shigellosis. Since the infection is self-limited in normally healthy patients and full recovery occurs without the use of antibiotics, drug therapy is usually not indicated. Additionally, treatment of shigellosis has been confounded by a widespread increase in antibiotic resistance among shigellae. Although clonal expansion of particular resistant strains may participate in the increase of treatment failure, horizontal transfer has been proposed as the main mechanism of dissemination of antibiotic resistance determinants among Shigella spp. via mobile genetic elements, such as plasmids and transposons. Additionally, these genetic elements may carry integrons than can assimilate and exchange exogenous DNA cassettes, usually antibiotic resistance genes, by site-specific recombination and therefore can easily accumulate antibiotic resistance traits. Class 1 integrons have been detected in S. flexneri and S. dysenteriae (40, 50) but are less frequent in other Shigella spp. than class 2 integrons isolated from developed countries (e.g., Japan, South Korea, and Ireland) (3, 73, 168) and developing countries (e.g., Brazil), especially among S. sonnei isolates. The choice of antimicrobial drugs has changed over the years as resistance to antibiotics has occurred, with different patterns of resistance being reported around the world. Ampicillin and a combination of trimetho­ prim and sulfamethoxazole (Sxt), once the treatments of choice against shigellosis, long ago lost efficacy in most of the world due to the development of drug resistance (167). In Belgium, resistance to Sxt progressed from 38% to 67% of S. sonnei isolates between 1990 and 2007 (168). In the United States, among the 552 Shigella isolates received by the National Antimicrobial Resistance Monitoring System at the CDC in 2008 (89.9% S. sonnei and 8.7% S. flexneri), 62.5% were resistant to ampicillin, 41.1% to Sxt, and 22.8% to both treatments (27). In sub-Saharan Africa, among the 109 Shigella isolates isolated in Mozambique from 2001 to

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

385

2003 (86% S. flexneri and 14% S. sonnei), 56% were resistant to ampicillin, 84% to Sxt, and 25% to both treatments (100). Subsequently, narrow-spectrum quinolones came as a solution for the last 2 decades, but resistance to nalidixic acid is now common in Asian countries (69) and frequently found in Africa. For example, the majority of the S. dysenteriae type I isolates that were responsible for the devastating epidemics between 1993 and 1995 in refugee camps in Rwanda, Tanzania, and Democratic Republic of the Congo were resistant to all the commonly used antibiotics including nalidixic acid (79). Nalidixic acid has therefore been replaced by fluoroquinolones, but ciprofloxacin resistance is now increasingly common in Asian countries (167), in particular in India, where a 48% increase has been reported between 2002 and 2007 (153). Quinolone resistance remains rare in the United States compared to Asian countries, but fluoroquinolone resistance has been reported in New York City, with one-half of the cases coming from travelers from Asia or people who may have come into contact with travelers from the Indian subcontinent (173). The emergence of ciprofloxacin resistance in Shigella has shifted the attention to broad-spectrum cephalosporins such as ceftriaxone, but their cost is often too prohibitive for developing countries. The discovery of extended-spectrum b-lactamases that confer resistance to broad-spectrum cephalosporins in S. flexneri and S. sonnei isolates in the United States (47), Vietnam (99, 124), France, Lebanon (143), and India (99) compromises treatment options for shigellosis and is extremely concerning, particularly for isolates displaying additional resistance to nalidixic acid (69). The macrolide azithromycin is recommended by the American Academy of Pediatrics for treatment of shigellosis in children and by the WHO as a second-line treatment in adults. A macrolide-resistant clinical variant of S. sonnei was isolated in France in 2007 (18). Extensive use of antibiotics selects for drug-resistant organisms, and therefore, many believe that antimicrobial therapy for shigellosis should be reserved only for the most severely ill patients. On the other hand, there are persuasive public health arguments for the antibiotic management of shigellosis. Antibiotic treatment limits the duration of disease and shortens the period of fecal excretion of this pathogen (58). Since an infected person or asymptomatic carrier can be an index case for person-to-person, food, and waterborne spread, antibiotic treatment of these individuals can be a significant public health tool to contain the spread of shigellosis. However, antibiotics are not a substitute for improved hygienic conditions to contain secondary spread of shig-

SMP_Food Microbiology_CH15.indd

385

ellosis. The single most effective means of preventing secondary transmission is hand washing with soap and chlorination of water. Food handling and preparation are important processes that also deserve attention, and persons with diarrhea should be excluded from handling food. Recently, Bardhan et al. (9a) proposed that the 80% reduction in shigellosis-associated deaths observed in Asia in the past 10 years was due to measles vaccination, vitamin A supplementation, and improved nutrition in that part of the world. Breastfeeding has also been shown to provide protection during the first 3 years of life, especially in malnourished children. Despite many years of intensive effort, an effective vaccine against shigellosis has not been developed yet. Natural infections offer about 72% protection against a second episode of shigellosis due to the homologous serotype but only offer less than 30% protection against a heterogenous serotype, suggesting Shigella O antigen is the key antigen for protection (90). A vaccine active against S. sonnei, S. dysenteriae 1, S. flexneri 2a, S. flexneri 3, and S. flexneri 6 would cover more than 80% of the strains currently causing morbidity and mortality in both developed and developing countries. Two main approaches have been undertaken initially: (i) attenuated strains of Shigella were used as live oral vaccines, and (ii) O polysaccharides of Shigella covalently linked to carrier proteins were used in a parenteral conjugate vaccine (76). Whereas the first approach generated some degree of protection, administration of the live attenuated Shigella variants—for example, S. flexneri 2a SC602, S. sonnei WRSS1, or S. dysenteriae 1 strain WRSd1—was followed by fever and mild diarrhea. The second approach was relatively successful and safe when tested in adults and young children, but the S. sonnei conjugate had no efficacy in children less than 2 years old (131). New vaccine candidates include the use of new conjugate vaccines that use synthetic oligosaccharides or a combination of invasion proteins and LPS named Invaplex (76).

VIRULENCE FACTORS

Hallmarks of Virulence

Shigella spp. and EIEC are the principal agents of bacillary dysentery and as such belong to the group of enteric pathogens that cause disease by overt invasion of epithelial cells in the large intestine. The clinical symptoms of shigellosis can be directly attributed to the hallmarks of Shigella virulence, which include the ability to induce diarrhea, invade epithelial cells of the intestine, multiply

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

386 intracellularly, spread from cell to cell, and induce an inflammatory response. The overall picture for Shigella pathogenesis parallels that of other enteric bacterial pathogens. After ingestion, the organism transits the stomach and eventually enters M cells that overlay lymphoid nodules. Exploiting the arsenal provided by the type III secretory system (T3SS), the bacteria induce their uptake by colonic M cells and, after transcytosis, are released in the intraepithelial pocket. Macrophages engulf the pathogens that survive phagocytosis and use effector proteins to kill the macrophages. After macrophage cell death, proinflammatory cytokines are released, thereby causing an acute inflammatory response. The current thought is that the host response provides the pathogen with the ability to evade the host innate and adaptive immune response and that controlling the host response may limit the severity of the disease. Bacterial cells released after the destruction of macrophages enter the basolateral membrane of adjacent epithelial cells by endocytosis. Inside the host cell cytoplasm, the pathogen multiplies. Although Shigella spp. lack flagella, these pathogens are intracellularly motile because of their ability to polymerize actin. This allows shigellae to spread from cell to cell via the formation of “actin rockets,” similar to those observed with Listeria monocytogenes (see chapter 20). Shigella colonizes the small intestine only transiently and causes little tissue damage (142). Production of enterotoxins by the bacteria while in the small bowel likely results in the diarrhea that often precedes the onset of dysentery (122), the first hallmark of Shigella virulence. The jejunal secretions elicited by these toxins may facilitate passage of the bacteria through the small intestine and into the colon, where they colonize and invade the epithelium. Formal and colleagues established the essential role of epithelial cell invasion in Shigella pathogenesis in a landmark study that employed both in vitro tissue culture assays for invasion and animal models (86). They found that spontaneous colonial variants of S. flexneri 2a that are unable to invade epithelial cells in tissue culture do not cause disease in monkeys. Gene transfer studies using E. coli K-12 donors and S. flexneri 2a recipients established the third hallmark, intracellular multiplication, of Shigella virulence. An S. flexneri 2a recipient that inherits the xyl-rha region of the E. coli K-12 chromosome retains the ability to invade epithelial cells but has a reduced ability to multiply within these cells. This hybrid strain fails to cause a fatal infection in the opium-treated guinea pig model and is unable to cause disease when fed to rhesus monkeys (49).

SMP_Food Microbiology_CH15.indd

386

It is necessary but not sufficient for Shigella to be able to multiply within the host epithelial cell after invasion. The bacterium must also be able to spread through the epithelial layer of the colon by cell-to-cell spread that does not require the bacterium to leave the intracellular environment and be reexposed to the intestinal lumen or the host’s immune response. Mutants of Shigella that are competent for invasion and multiplication but unable to spread between cells in this fashion have been isolated. These mutants established intracellular spread as the fourth hallmark of Shigella virulence and will be discussed further below. Along with the ability to colonize and cause disease, an intrinsic part of a bacterium’s pathogenicity is its mechanism for regulating expression of the genes involved in virulence. Virulence in Shigella spp. is regulated by growth temperature. After growth at 37°C, virulent strains of Shigella are able to invade mammalian cells, but when cultivated at 30°C, they are noninvasive. This noninvasive phenotype is reversible by shifting the growth temperature to 37°C. The temperature change enables the bacteria to reexpress their virulence properties (105). Temperature regulation of virulence gene expression is a characteristic that Shigella shares with other human pathogens, such as E. coli, Salmonella enterica serovar Typhimurium, Bordetella pertussis, Yersinia spp., and L. monocytogenes (see reference 82 for a review). Regulation of gene expression in response to environmental temperature is a useful bacterial strategy. By sensing the ambient temperature of the mammalian host (e.g., 37°C for humans) to trigger gene expression, this strategy permits Shigella to economize energy that would be expended on the synthesis of virulence products when the bacteria are outside the host. The system also permits the bacteria to coordinately regulate expression of multiple unlinked genes that are required for the full-virulence phenotype. Temperature regulation in S. flexneri 2a operates at the level of gene transcription and is mediated by both positive and negative transcription factors. A chromosomal gene, virR(hns), encodes a repressor of virulence gene expression (107), while two plasmid-borne genes, virF and virB, encode positive activators (1). These genes will be discussed in a later section. A more thorough treatment of virulence gene regulation in Shigella can be found in several review articles (39, 135, 149).

Genetics Virulence-Associated Plasmid Genes

Given the complexity of the interactions between host and pathogen, it is not surprising that Shigella virulence

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

387

is multigenic, involving both chromosomal and plasmid-borne genes (Table 15.1). Another landmark report on the pathogenicity of Shigella was the demonstration of the indispensable role of a large plasmid in invasion. A 180-kbp plasmid in S. sonnei and a 220-kbp plasmid in S. flexneri are essential for invasion (104). Other Shigella spp., as well as strains of EIEC, possess homologous plasmids that are functionally interchangeable and share significant degrees of DNA homology. The DNA sequence of the virulence plasmid of S. flexneri was the first to be published (21, 72, 165), and the virulence plasmids of S. sonnei, S. dysenteriae 1, and S. boydii have also been sequenced (71, 175). The virulence plasmid from EIEC has been sequenced and can be accessed from GenBank (accession number CP001064). A 37-kb region of the invasion plasmid of S. flexneri 2a contains all of the genes necessary to permit the bacteria to penetrate into tissue culture cells. This DNA segment was identified as the minimal region of virulence plasmid necessary to allow a plasmid-cured derivative of S. flexneri (and E. coli K-12) to invade tissue culture cells (81). The region carries about 33 genes contained in two groups of genes transcribed in opposite orientations (Fig. 15.2). These genes encode proteins that are components of the T3SS apparatus, effectors, translocators, transcription activators, and chaperones. Although a precise transcription map of these genes has not been defined, available evidence and the DNA sequence of the region suggest a multiple operon organization. The genes comprising the ipaBCDA (where ipa is an acronym for invasion plasmid antigens) cluster encode the immunodominant antigens detected with sera from convalescent patients and experimentally challenged monkeys (125). The genes in ipaBCD have been experimentally demonstrated to be required for invasion of mammalian cells (114). The Ipa products act as translocators by interacting with the host cell membrane subsequent to being shunted through the bacterial membrane via the T3SS. IpaB and IpaC (and probably IpaA) form a complex on the bacterial cell surface and are responsible for transducing the signal that leads to entry of Shigella into host cells via bacterium-directed phagocytosis (128). IpaD is also required for insertion of IpaBC into the membrane (133). However, when IpaB and IpaC are coated onto latex beads, they form a complex that promotes uptake of the beads by HeLa cells (112). This complex of Ipa proteins also binds to cell surface receptors such as α5-β1 integrins (171). Purified IpaC induces cytoskeletal reorganization via actin polymerization and depolymerization, including formation of filopodia and lamellipodial extensions on permeabilized cells (163). IpaA binds to vinculin and promotes F-actin depolymer-

SMP_Food Microbiology_CH15.indd

387

ization (138). This step is thought to facilitate reorganization of the host cell surface structures induced by contact with Shigella and modulate bacterial entry (37). Although the Ipa proteins have no typical signal sequence for recognition by the usual gram-negative bacterial transport system, these proteins are secreted into the extracellular medium via a T3SS (16, 32). The T3SS pathway requires a dedicated apparatus composed of gene products from the mxi and spa (membrane expression of invasion plasmid antigens/surface presentation of Ipa antigens) loci (Fig. 15.2). Secretion through T3SS is extremely orchestrated, as recently reviewed (35). Contact of the bacterium with epithelial cells induces secretion of the cytoplasmic pool of Ipa products (172). Fibronectin, laminin, collagen type IV, Congo red, bile salts, and fetal bovine serum have also been reported to induce Ipa secretion in the laboratory (155). IpaD forms an antisecretion complex, or plug, with IpaB. Consequently, ipaD mutants are hypersecretors of the Ipa products (113). MxiC plays a similar role in controlling the secretion of effectors, acting as a gatekeeper for secretion (17, 102). ipgC (invasion plasmid gene) is required for invasion and acts as a cytoplasmic chaperone, which prevents IpaB and IpaC from forming complexes while in the bacterial cytoplasm (128). In the absence of IpgC, IpaB and IpaC are rapidly degraded. IpgC also plays a role as a coactivator in virulence gene regulation (see below). The product of ipaB has also been postulated to be the “contact hemolysin” that is responsible for lysis of the phagocytic vacuole minutes after entry of the bacterium into the host cell (61). The ability of Shigella to induce apoptosis in infected macrophages is an additional property assigned to IpaB (182). The secretion apparatus consists of inner membrane proteins MxiG and MxiJ and an outer membrane protein, MxiD, which, along with MxiM, is proposed to form a bridge across the periplasmic space to contact MxiJ and MxiG (150). The secretion “needle” is composed of MxiH and possibly MxiI (15, 157). Spa47 plays a critical role in facilitating the transfer of proteins through the needle complex (75), while Spa32 is involved in controlling needle length (98, 158). Spa33 associates with MxiG and MxiJ in the putative C ring of the secretion apparatus and controls secretion of IpaB and IpaC (116). Interestingly, expression of both spa32 and spa33, but not that of other T3SS genes, is reduced under anaerobic conditions prevalent in the lumen of the large intestine, suggesting that effective secretion of TTSS effectors would occur only when the bacteria reach the surface of intestinal epithelial cells, where oxygen is more abundant (10, 101). Nonpolar null mutations in all of the mxi and spa genes tested

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

388 Table 15.1  Virulence-associated loci of Shigella Locus

Product

Role in virulence

Chromosomal loci cpxR

Response regulator of CpxA-CpxR

Activator of virF two-component system

iuc

Synthesis of aerobactin and receptor

Acquisition of iron in the host

rfa; rfb

Enzymes for core and O antigen

Correct polar localization of IcsA biosynthesis

set

SHET1

Enterotoxin

she

Putative hemagglutinin and mucinase

Unknown

sodB

Superoxide dismutase

Inactivation of superoxide radicals; defense against oxygen-dependent killing in host

stxb

Shiga toxin

Destruction of vascular tissue

vacB (rnr)

Exoribonuclease RNase R

Posttranscriptional regulation of virulence gene expression

virR(hns)

Histone-like protein

Repressor of virulence gene expression

icsA(virG)

Cell-bound and secreted protein

Actin polymerization for intracellular motility and intercellular spread

ipaA

Secreted effector

Efficient invasion; binds to vinculin and promotes F-actin

ipaB

Secreted effector

Invasion; lysis of vacuole; induction of apoptosis

ipaC

Secreted effector

Invasion; induces cytoskeletal reorganization

ipaD

Secreted effector

Invasion; antisecretion plug (with IpaB) depolymerization

a

Plasmid-borne loci

Ubiquitin ligases (?)c

ipaH ipgC

17-kDa protein

Cytoplasmic chaperone for IpaB and IpaC; coactivator of MxiE

ipgB1

Secreted effector

Promotes membrane ruffling by activation of Rac1 and Cdc42

mxi/spa

20 proteins

T3SS for secretion of Ipa and other virulence proteins

mxiE

Transcriptional activator

AraC family postinvasion activator

opsC2, opsC3

Secreted effector

Unknown

ospD1

Secreted effector

Antiactivator of MxiE

ospD2

Secreted effector

Unknown

ospE

Secreted effector

Maintains bacterial cell integrity postinvasion

ospF

Secreted effector

Phosphothreonine lyase

ospG

Secreted effector

Targets ubiquitin-conjugating enzymes; downregulates host innate immune response

icsB

Secreted effector

Shields IcsA from binding autophagy protein Atg5; protects against autophagy

sen

SHET2

Enterotoxin

virB

Transcriptional activator

Temperature regulation of virulence genes

virF

Transcriptional activator

Temperature regulation of virulence genes

The set locus and production of SHET1 are found almost exclusively in S. flexneri. The stx locus and production of Shiga toxin are observed only in S. dysenteriae 1. For ipaH, there are four loci located on the chromosome and five loci located on the virulence plasmid. The role of the ipaH genes appears to be to encode E3 ubiquitin ligases (141). Since there are nine genetic loci, the ipaH genes have been a common target in many PCR-based assays. a

b

c

SMP_Food Microbiology_CH15.indd

388

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

389

Figure 15.2  S. flexneri 2a virulence plasmid map. A SalI restriction map of the 220-kb plasmid is shown in the center. Sections of SalI fragments B and P (upper map) and fragments P, H, and D (lower map) are expanded to illustrate the virulence loci carried in these regions. The expanded regions are contiguous and cover 32 kb. The open reading frame for icsB is separated from that of ipgD by 314 bp. doi:10.1128/9781555818463.ch15f2

to date result in loss of ability to secrete the Ipas and loss of invasive capacity for cultured cells. Hence, T3SS is an essential component of Shigella virulence. The exception to this statement, i.e., loss of invasive capability, is mxiE, which encodes a transcriptional regulator of genes expressed ­ postinvasion (77). MxiE is addressed below in “Virulence Gene Regulation.” The spa genes encode proteins that share significant homologies with proteins involved in flagellar ­synthesis in E. coli, Salmonella Typhimurium, Bacillus subtilis, and Caulobacter crescentus (149). Included among these genes is spa47, which encodes a ­protein that probably functions as the energy-generating component of the secretion apparatus, since it shows sequence similarities with ATPases of the flagellar assembly ­machinery of other bacteria. In addition, there is an even more striking similarity both in gene organization and predicted

SMP_Food Microbiology_CH15.indd

389

protein sequence between the mxi and spa region of the Shigella virulence plasmid and a virulence-­associated chromosomal region of Salmonella Typhimurium (54). The Salmonella spa region encodes homologues of the Shigella spa genes in the same gene order. Sequence identities between the protein homologues reach levels as high as 86% (Spa9 and SpaQ). The relatedness of the spa regions strongly suggests that these two human pathogens evolved similar mechanisms for secretion of the virulence proteins required for signal transduction with the mammalian host. Plant pathogens such as Pectobacterium carotovorum, Xanthomonas campestris, and Ralstonia solanacearum also contain genes that encode homologues of the mxi- and spa-encoded proteins (164). It is now recognized that the T3SS for transport of virulence proteins is a critical element of both plant and animal bacterial pathogenesis.

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

390 A new class of exported proteins was identified from the sequence of the virulence plasmid. Designated Osp (outer Shigella proteins), the function of these proteins is only just beginning to be determined (21). OspD1 negatively controls virulence gene expression by acting as an antiactivator (see below), and OspG is involved in downregulation of the host innate immune response through interference with activation of the NF-kB pathway (80). The expressions of the genes for the effector proteins OspB, OspC1, and OspF are increased after T3SS activation. OspF has been identified as a phosphothreonine lyase and plays a role in the polymorphonuclear leukocyte recruitment, possibly as a modulator of inflammation (94). A plasmid-borne virulence gene that is unlinked to the 37-kb region shown in Fig. 15.1 is not required for invasion but is crucial for intra- and intercellular motility. This gene, known as virG or icsA (intracellular spread), encodes a protein that catalyzes the polymerization of actin in the cytoplasm of the infected host cell (14). The IcsA protein is unusual in that it is expressed asymmetrically on the bacterial surface, being found only at one pole. The polymerization of actin monomers by IcsA forms a tail leading from the pole and provides the force that propels the bacterium through the cytoplasm. Hence, unipolar expression of IcsA imparts directionality of movement to the bacterium. A ­plasmid-encoded protease, SopA/IcsP, cleaves IcsA and is proposed to play a role in unipolar localization of IcsA (151). Synthesis of a complete LPS is also crucial for correct unipolar localization of IcsA (145). The use of translational fusions between IcsA and green fluorescent protein defined the amino acids that are sufficient for polar localization and confirmed correct polar localization of IcsA when expressed in E. coli as well as Salmonella Typhimurium, Yersinia pseudotuberculosis, and Vibrio cholerae (29). Surface-expressed IcsA is a trigger for autophagy. IcsB is a protein secreted via the Shigella T3SS that acts to shield IcsA from binding the autophagy protein Atg5 and thus protects the bacterium from being trapped and degraded by autophagy within the infected host cell (127).

Chromosomal Virulence Loci

Although Shigella and E. coli are very closely related at the genetic level, there are significant differences beyond the presence of the virulence plasmid in Shigella. Pathogenicity islands, i.e., clusters of genes that have a role in virulence, have been identified in several virulent bacteria including Shigella. These regions are typically large (20 to 200 kb) and in addition to virulence genes may contain transposable elements and remnants of mo-

SMP_Food Microbiology_CH15.indd

390

bility plasmid or bacteriophage genes, all indicating possible horizontal transfer of genetic information. Two pathogenicity islands have been identified in the chromosome of S. flexneri (Table 15.1). SHI-1 (Shigella pathogenicity island 1) contains the set gene, which encodes an enterotoxin (46). It is contained within the open reading frame of another gene, she, which encodes a protein with putative hemagglutinin and mucinase activity (123, 136). The iuc locus, which contains the genes for aerobactin synthesis and transport, is present in SHI-2 (117, 166). Aerobactin is a hydroxamate siderophore that S. flexneri uses to scavenge iron. When the iuc locus is inactivated, the aerobactin-deficient mutants retain their capacity to invade host cells but are altered in virulence as measured in animal models. These results suggest that aerobactin synthesis is important for bacterial growth within the mammalian host (88, 121). S. dysenteriae type 1, like E. coli O157:H7, produces Shiga toxin, which is responsible for some of the symptoms of HUS. Acquisition of the Shiga toxin genes, stxA and stxB, by S. dysenteriae has been postulated to be a consequence of a lysogenic bacteriophage. Over time, the phage became stably integrated due to loss of genetic function of essential phage genes, most likely as a result of rearrangement and transposition events. Shiga toxin inhibits protein synthesis by inactivating rRNA (28S rRNA) in the 60S subunit of mammalian ribosomes and preventing elongation factor 2 from interacting with the ribosome (126). Shiga toxin is not found in other Shigella spp. The Shiga toxin produced by S. dysenteriae and the Shiga toxin elaborated by E. coli O157:H7 are nearly identical, differing from each other by just one amino acid. A mutation in the stx locus does not alter the ability of the bacterium to invade epithelial cells or cause keratoconjunctivitis in the Sereny test. However, when tested in macaque monkeys, the mutant strain causes less vascular damage in the colonic tissue than does the toxin-producing parent (48). Hence, production of Shiga toxin may account for the generally more severe infections caused by S. dysenteriae 1 than by the other species of Shigella. Two other enterotoxins are produced by Shigella. The gene for Shigella enterotoxin 1 (SHET1) is chromosomally borne and is present in S. flexneri but is not usually in other Shigella strains. SHET2 (Shigella enterotoxin 2) is encoded by the virulence plasmid in many but not all Shigella serotypes. The early genetic studies by Formal et al. (49), and reports that select genetic loci are not present in the Shigella chromosome but are present in the nonpathogenic E. coli chromosome (106), suggest that the ­absence of particular genes in Shigella have had a direct consequence on its pathogenesis. The absence of these genetic

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

391

determinants in Shigella spp. occurred through either ­ eletions or mutations and is referred to as a “Black d Hole” (Fig. 15.1) (103). When a gene complementary to a genetic locus that is absent in Shigella is introduced into the pathogen, the presence of that newly acquired gene product in Shigella may inhibit the action of a particular virulence factor. For instance, although lysine decarboxylase activity is present in >85% of E. coli strains, it is missing in all strains of Shigella spp. and enteroinvasive E. coli. Lysine decarboxylase (encoded by the cadA gene) produces cadaverine from lysine. Cadaverine inhibits the action of the Shigella enterotoxins that are believed to be responsible for the diarrheal symptoms associated with shigellosis. Therefore, cadA is considered an “antivirulence gene” for Shigella (103). Thus, as part of the evolution of Shigella to a human pathogen from a nonpathogenic E. coli lineage, the acquisition of the large virulence plasmid may have also involved the loss of specific incompatible genetic loci. Recently, Yang et al. reviewed the role of insertion sequence elements in the evolution of Shigella genomes from an ancestral E. coli strain (177). In contrast to the genes of the virulence plasmid that are responsible for invasion of mammalian tissues, most of the chromosomal loci associated with Shigella virulence are involved in regulation or survival within the host.

Virulence Gene Regulation

An important feature of Shigella pathogenesis is the ability of the bacterium to modulate expression of its virulence genes in response to growth temperature (149). Several activators and a repressor control the virulence regulon of Shigella such that genes are turned on at 37°C and turned off at 30°C. The product of the chromosomal virR(hns) locus is a histone-like protein, H-NS, which behaves as a repressor of Shigella virulence gene expression in response to growth temperature (107). Mutations in virR(hns) cause deregulation of temperature control and result in expression of genes in the virulence regulon even at the nonpermissive temperature of 30°C. The virR(hns) locus is allelic with regulatory loci in other enteric bacteria, and like virR(hns), these alleles act as repressors of their respective regulons (see references cited in reference 39). Several different models to explain how VirR/H-NS acts as a transcriptional repressor have been proposed (39). However, because VirR/H-NS is involved in gene regulation in response to diverse environmental stimuli such as osmolarity, pH, and temperature, a comprehensive model to explain its activity has been elusive. H-NS binds to the promoter regions of two transcriptional activators, virF and virB, and blocks their tran-

SMP_Food Microbiology_CH15.indd

391

scription at 30°C (11, 44). At 37°C, the H-NS binding sites undergo a conformational transition and no longer bind H-NS, thus leading to increased transcription of the activator genes (149). Expression of genes in the ipa and mxi and spa clusters is dependent on VirB, and mutations in virB abolish the bacterium’s ability to invade tissue culture cells. Transcription of virB is dependent on growth temperature and VirF (159). VirB is a DNAbinding protein and shares homology with the plasmidpartitioning proteins ParB of bacteriophage P1 and SopB of plasmid F. Purified VirB shows preferential binding to the intergenic icsB-ipgD region (Fig. 15.2) and displays a similar preference for binding to the spa and virA gene promoter regions (110). The product of the virF locus is a key element in temperature regulation of the Shigella virulence regulon. A helix-turn-helix motif in the carboxyl-terminal portion of VirF is characteristic of members of the AraC family of transcriptional activators. Consistent with its predicted role as a DNA-binding protein, VirF binds to sequences upstream of virB (160). The binding of VirF may act as an antagonist to binding by VirR/H-NS and thereby provides a mechanism for responding to temperature. Expression of virF is subject to temperature regulation by H-NS. Repression of virF occurs at the critical temperature of <32°C and takes place through the binding of H-NS at two sites within the virF promoter (44). Activation of the T3SS leads to increased expression of a subset of virulence genes encoding secreted effectors. Expression of these genes is regulated by MxiE, a transcriptional activator of the AraC family (77). MxiEregulated promoters are preceded by a 17-bp MxiE box 33 to 49 bp upstream of the transcription start site (109). Activity of MxiE requires a coactivator, IpgC, the chaperone of IpaB and IpaC (108, 134). Under nonsecretion conditions, IpgC is associated with IpaB and IpaC, whereas MxiE is associated with the T3SS substrate OspD1, which acts as an antiactivator. Hence, both activator (MxiE) and coactivator (IpgC) are titrated and unavailable to interact with each other. When the secretion apparatus is activated, IpaB and IpaC are secreted and release IpgC. After OspD1 is secreted, MxiE is released and is free to bind IpgC and activate MxiE-controlled promoters (130). Expression profile analysis defined three classes of T3SS secreted substrates: (i) those that are controlled by VirB (expressed independently of secretion activity); (ii) those that are controlled by MxiE (expressed only under conditions of secretion); and (iii) those that are controlled by both VirB and MxiE (expressed under conditions of nonsecretion and induced under conditions of secretion) (89).

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

392

CONCLUSIONS While foodborne infections due to members of the Shigella spp. may not be as frequently reported as those caused by other foodborne pathogens, they have the potential for explosive spread due to the low infectious dose that can cause overt clinical disease. In addition, cases of bacillary dysentery frequently require medical attention (even hospitalization) and cause lost time from work as the severity and duration of symptoms can be incapacitating. There is no effective vaccine against dysentery caused by Shigella. These features, coupled with the wide geographical distribution of the strains and sensitivity of the human population, make Shigella a formidable public health threat.

References   1. Adler, B., C. Sasakawa, T. Tobe, S. Makino, K. Komatsu, and M. Yoshikawa. 1989. A dual transcriptional activation system for the 230 kb plasmid genes coding for virulence-associated antigens of Shigella flexneri. Mol. Microbiol. 3:627–635.   2. Agle, M. E., S. E. Martin, and H. P. Blaschek. 2005. Survival of Shigella boydii 18 in bean salad. J. Food Prot. 68:838–840.   3. Ahmed, A. M., K. Furuta, K. Shimomura, Y. Kasama, and T. Shimamoto. 2006. Genetic characterization of multidrug resistance in Shigella spp. from Japan. J. Med. Microbiol. 55:1685–1691.   4. Al-Nimri, S., W. A. Miller, B. A. Byrne, G. Guibert, and L. Chen. 2009. A unified approach to molecular epidemiology investigations: tools and patterns in California as a case study for endemic shigellosis. BMC Infect. Dis. 9:184.   5. Andrews, W. H. 1989. Methods for recovering injured classical entric pathogenic bacteria (Salmonella, Shigella and enteropathogenic Escherichia coli) from foods, p. 55–113. In B. Ray (ed.), Injured Index and Pathogenic Bacteria: Occurrence and Detection in Foods, Water and Feeds. CRC Press, Boca Raton, FL.   6. Anyanful, A., J. M. Dolan-Livengood, T. Lewis, S. Sheth, M. N. Dezalia, M. A. Sherman, L. V. Kalman, G. M. Benian, and D. Kalman. 2005. Paralysis and killing of Caenorhabditis elegans by enteropathogenic Escherichia coli requires the bacterial tryptophanase gene. Mol. Microbiol. 57:988–1007.   7. Arvelo, W., C. J. Hinkle, T. A. Nguyen, T. Weiser, N. Steinmuller, F. Khan, S. Gladbach, M. Parsons, D. Jennings, B. P. Zhu, E. Mintz, and A. Bowen. 2009. Transmission risk factors and treatment of pediatric shigellosis during a large daycare center-associated outbreak of multidrug resistant Shigella sonnei. Implications for the management of shigellosis outbreaks among children. Pediatr. Infect. Dis. J. 28:976–980.   8. Bagamboula, C. F., M. Uyttendaele, and J. Debevere. 2002. Acid tolerance of Shigella sonnei and Shigella flexneri. J. Appl. Microbiol. 93:479–486.

SMP_Food Microbiology_CH15.indd

392

  9. Bangtrakulnonth, A., A. R. Vieira, D. M. A. L. Wong, S. Pornreongwong, C. Pulsrikarn, P. Sawanpanyalert, R. S. Hendriksen, and F. M. Aarestrup. 2008. Shigella from humans in Thailand during 1993 to 2006: ­spatial-time trends in species and serotype distribution. Foodborne Pathog. Dis. 5:773–784.   9a. Bardhan, P., A. S. G. Faruque, A. Naheed, and D. A. Sack. 2010. Decreasing shigellosis-related deaths without Shigella spp.-specific interventions. Asia Emerg. Infect. Dis. 16:1718–1723.   10. Baxt, L. A., and M. B. Goldberg. 2010. Anaerobic environment of the intestine primes pathogenic Shigella for infection. Exp. Rev. Anti Infect. Ther. 8:1225–1229.   11. Beloin, C., and C. J. Dorman. 2003. An extended role for the nucleoid structuring protein H-NS in the virulence gene regulatory cascade of Shigella flexneri. Mol. Microbiol. 47:825–838.   12. Beloin, C., A. Roux, and J. M. Ghigo. 2008. Escherichia coli biofilms. Curr. Top. Microbiol. Immunol. 322:249–289.   13. Bennish, M. L. 1991. Potentially lethal complications of shigellosis. Rev. Infect. Dis. 13(Suppl. 4):S319–S324.   14. Bernardini, M. L., J. Mounier, H. D’Hauteville, M. Coquis-Rondon, and P. J. Sansonetti. 1989. Identification of icsA, a plasmid locus of Shigella flexneri that governs bacterial intra- and intercellular spread through interaction with F-actin. Proc. Natl. Acad. Sci. USA 86:3867–3871.   15. Blocker, A., N. Jouihri, E. Larquet, P. Gounon, F. Ebel, C. Parsot, P. Sansonetti, and A. Allaoui. 2001. Structure and composition of the Shigella flexneri “needle complex”, a part of its type III secretion. Mol. Microbiol. 39:652–663.   16. Blocker, A. J., J. E. Deane, A. K. Veenendaal, P. Roversi, J. L. Hodgkinson, S. Johnson, and S. M. Lea. 2008. What’s the point of the type III secretion system needle? Proc. Natl. Acad. Sci. USA 105:6507–6513.   17. Botteaux, A., M. P. Sory, L. Biskri, C. Parsot, and A. Allaoui. 2009. MxiC is secreted by and controls the substrate specificity of the Shigella flexneri type III secretion apparatus. Mol. Microbiol. 71:449–460.   18. Boumghar-Bourtchai, L., P. Mariani-Kurkdjian, E. Bingen, I. Filliol, A. Dhalluin, S. A. Ifrane, F. X. Weill, and R. Leclercq. 2008. Macrolide-resistant Shigella sonnei. Emerg. Infect. Dis. 14:1297–1299.   19. Brahmbhatt, H. N., A. A. Lindberg, and K. N. Timmis. 1992. Shigella lipopolysaccharide: structure, genetics, and vaccine development. Curr. Top. Microbiol. Immunol. 180:45–64.   20. Brandl, M. T. 2006. Fitness of human enteric pathogens on plants and implications for food safety. Annu. Rev. Phytopathol. 44:367–392.   21. Buchrieser, C., P. Glaser, C. Rusniok, H. Nedjari, H. D’Hauteville, F. Kunst, P. Sansonetti, and C. Parsot. 2000. The virulence plasmid pWR100 and the repertoire of proteins secreted by the type III secretion apparatus of Shigella flexneri. Mol. Microbiol. 38:760–771.   22. Bunning, V. K., R. B. Raybourne, and D. L. Archer. 1988. Foodborne enterobacterial pathogens and rheumatoid disease. Soc. Appl. Bacteriol. Symp. Ser. 17:87S–107S.

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

393

  23. Centers for Disease Control and Prevention. 1994. Outbreak of Shigella flexneri 2a infections on a cruise ship. MMWR Morb. Mortal. Wkly. Rep. 43:657.   24. Centers for Disease Control and Prevention. 1997. Foodborne Diseases Active Surveillance Network (FoodNet). Emerg. Infect. Dis. 3:581–583.   25. Centers for Disease Control and Prevention. 2000. Outbreak of Shigella sonnei infections associated with eating a nationally distributed dip—California, Oregon, and Washington, January 2000. MMWR Morb. Mortal. Wkly. Rep. 49:60–61.   26. Centers for Disease Control and Prevention. 2008. Shigella Surveillance: Annual Summary, 2006. U.S. Department of Health and Human Services, Altanta, GA.   27. Centers for Disease Control and Prevention. 2010. National Antimicrobial Resistance Monitoring System for Enteric Bacteria (NARMS): Human Isolates Final Report, 2008. U.S. Department of Health and Human Services, Atlanta, GA.   28. Centers for Disease Control and Prevention. 2010. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food - 10 states, 2009. MMWR Morb. Mortal. Wkly. Rep. 59:418–422.   29. Charles, M., M. Perez, J. H. Kobil, and M. B. Goldberg. 2001. Polar targeting of Shigella virulence factor IcsA in Enterobacteriaceae and Vibrio. Proc. Natl. Acad. Sci. USA 98:9871–9876.   30. Chart, H., R. M. A. Daniel, and T. Cheasty. 2009. The expression of lipopolysaccharide by strains of Shigella dysenteriae, Shigella flexneri and Shigella boydii and their cross-reacting strains of Escherichia coli. FEMS Microbiol. Lett. 292:21–26.   31. Chattopadhyay, S., S. J. Weissman, V. N. Minin, T. A. Russo, D. E. Dykhuizen, and E. V. Sokurenko. 2009. High frequency of hotspot mutations in core genes of Escherichia coli due to short-term positive selection. Proc. Natl. Acad. Sci. USA 106:12412–12417.   32. Cornelis, G. R. 2010. The type III secretion injectisome, a complex nanomachine for intracellular ‘toxin’ delivery. Biol. Chem. 391:745–751.   33. Daskalakis, D. C., and M. J. Blaser. 2007. Another perfect storm: Shigella, men who have sex with men, and HIV. Clin. Infect. Dis. 44:335–337.   34. Davis, H., J. P. Taylor, J. N. Perdue, G. N. Stelma, J. M. Humphreys, R. Rowntree, and K. D. Greene. 1988. A shigellosis outbreak traced to commercially distributed shredded lettuce. Am. J. Epidemiol. 128:1312–1321.   35. Deane, J. E., P. Abrusci, S. Johnson, and S. M. Lea. 2010. Timing is everything: the regulation of type III secretion. Cell. Mol. Life Sci. 67:1065–1075.   36. Deer, D. M., and K. A. Lampel. 2010. Development of a multiplex real-time PCR assay with internal amplification control for the detection of Shigella species and enteroinvasive Escherichia coli. J. Food Prot. 73:1618–1625.   37. Demali, K. A., A. L. Jue, and K. Burridge. 2006. IpaA targets β1 integrins and rho to promote actin cytoskeleton rearrangements necessary for Shigella entry. J. Biol. Chem. 281:39534–39541.

SMP_Food Microbiology_CH15.indd

393

  38. Dembek, Z. F., M. G. Kortepeter, and J. A. Pavlin. 2007. Discernment between deliberate and natural infectious disease outbreaks. Epidemiol. Infect. 135:353–371.   39. Dorman, C. J., S. McKenna, and C. Beloin. 2001. Regulation of virulence gene expression in Shigella flexneri, a facultative intracellular pathogen. Int. J. Med. Microbiol. 291:89–96.   40. Dubois, V., M. P. Parizano, C. Arpin, L. Coulange, M. C. Bezian, and C. Quentin. 2007. High genetic stability of integrons in clinical isolates of Shigella spp. of worldwide origin. Antimicrob. Agents Chemother. 51:1333–1340.   41. Dupont, H. L., M. M. Levine, R. B. Hornick, and S. B. Formal. 1989. Inoculum size in shigellosis and implications for expected mode of transmission. J. Infect. Dis. 159:1126–1128.   42. Ekdahl, K., and Y. Andersson. 2005. The epidemiology of travel-associated shigellosis—regional risks, seasonality and serogroups. J. Infect. 51:222–229.   43. Escobar-Paramo, P., O. Clermont, A. B. Blanc-Potard, H. Bui, B. C. Le, and E. Denamur. 2004. A specific genetic background is required for acquisition and expression of virulence factors in Escherichia coli. Mol. Biol. Evol. 21:1085–1094.   44. Falconi, M., B. Colonna, G. Prosseda, G. Micheli, and C. O. Gualerzi. 1998. Thermoregulation of Shigella and Escherichia coli EIEC pathogenicity. A temperature­dependent structural transition of DNA modulates accessibility of virF promoter to transcriptional repressor H-NS. EMBO J. 17:7033–7043.   45. Farfan, M. J., T. A. Garay, C. A. Prado, I. Filliol, M. T. Ulloa, and C. S. Toro. 2010. A new multiplex PCR for differential identification of Shigella flexneri and Shigella sonnei and detection of Shigella virulence determinants. Epidemiol. Infect. 138:525–533.   46. Fasano, A., F. R. Noriega, D. R. Maneval, Jr., S. Chanasongcram, R. Russell, S. Guandalini, and M. M. Levine. 1995. Shigella enterotoxin 1: an enterotoxin of Shigella flexneri 2a active in rabbit small intestine in vivo and in vitro. J. Clin. Investig. 95:2853–2861.   47. Folster, J. P., G. Pecic, A. Krueger, R. Rickert, K. Burger, A. Carattoli, and J. M. Whichard. 2010. Identification and characterization of CTX-M-producing Shigella isolates in the United States. Antimicrob. Agents Chemother. 54:2269–2270.   48. Fontaine, A., J. Arondel, and P. J. Sansonetti. 1988. Role of Shiga toxin in the pathogenesis of bacillary dysentery, studied by using a Tox- mutant of Shigella dysenteriae 1. Infect. Immun. 56:3099–3109.   49. Formal, S. B., E. H. LaBrec, T. H. Kent, and S. Falkow. 1965. Abortive intestinal infection with an Escherichia coliShigella flexneri hybrid strain. J. Bacteriol. 89:1374–1382. 50. Gassama Sow, A., A. Aidara-Kane, O. Barraud, M. Gatet, F. Denis, and M. C. Ploy. 2010. High prevalence of trimethoprim-resistance cassettes in class 1 and 2 integrons in Senegalese Shigella spp isolates. J. Infect. Dev. Ctries. 4:207–212.   51. Germani, Y., and P. Sansonetti. 2006. The Genus Shigella, p. 99–122. In M. Dworkin, S. Falkow, E.

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

394 Rosenberg, K. H. Schleifer, and E. Stackebrandt (ed.), The Prokaryotes. Springer, New York, NY.   52. Giannella, R. A., S. A. Broitman, and N. Zamcheck. 1972. Gastric acid barrier to ingested microorganisms in man: studies in vivo and in vitro. Gut 13:251–256.   53. Gikas, A., J. Pediaditis, Z. Giti, J. Papadakis, and Y. Tselentis. 1996. Shigellosis on an Italian cruise ship. Lancet 348:1593–1594.   54. Groisman, E. A., and H. Ochman. 1993. Cognate gene clusters govern invasion of host epithelial cells by Salmonella typhimurium and Shigella flexneri. EMBO J. 12:3779–3787.   55. Guerrero, L., J. J. Calva, A. L. Morrow, F. R. Velazquez, F. Tuzdzib, Y. Lopezvidal, H. Ortega, H. Arroyo, T. G. Cleary, L. K. Pickering, and G. M. Ruizpalacios. 1994. Asymptomatic Shigella infections in a cohort of mexican children younger than 2 years of age. Pediatr. Infect. Dis. J. 13:597–602.   56. Gunduz, G. T., and G. Tuncel. 2006. Biofilm formation in an ice cream plant. Antonie Van Leeuwenhoek 89:329–336.   57. Haley, C. C., K. L. Ong, K. Hedberg, P. R. Cieslak, E. Scallan, R. Marcus, S. Shin, A. Cronquist, J. Gillespie, T. F. Jones, B. Shiferaw, C. Fuller, K. Edge, S. M. Zansky, P. A. Ryan, R. M. Hoekstra, and E. Mintz. 2010. Risk factors for sporadic shigellosis, FoodNet 2005. Foodborne. Pathog. Dis. 7:741–747.   58. Haltalin, K. C., J. D. Nelson, R. Ring III, M. Sladoje, and L. V. Hinton. 1967. Double-blind treatment study of shigellosis comparing ampicillin, sulfadiazine, and placebo. J. Pediatr. 70:970–981.   59. Heier, B. T., K. Nygard, G. Kapperud, B. A. Lindstedt, G. S. Johannessen, and H. Blekkan. 2009. Shigella sonnei infections in Norway associated with sugar peas, May-June 2009. Euro Surveill. 14:pii 19243.   60. Hershberg, R., H. Tang, and D. A. Petrov. 2007. Reduced selection leads to accelerated gene loss in Shigella. Genome Biol. 8(8):R164.   61. High, N., J. Mounier, M. C. Prevost, and P. J. Sansonetti. 1992. IpaB of Shigella flexneri causes entry into epithelial cells and escape from the phagocytic vacuole. EMBO J. 11:1991–1999.   62. Hilbi, H., S. S. Weber, C. Ragaz, Y. Nyfeler, and S. Urwyler. 2007. Environmental predators as models for bacterial pathogenesis. Environ. Microbiol. 9:563–575.   63. Hird, S., C. Stein, T. Kuchenmuller, and R. Green. 2009. Meeting report: second annual meeting of the World Health Organization initiative to estimate the global burden of foodborne diseases. Int. J. Food Microbiol. 133:210–212.   64. Houng, H. S., O. Sethabutr, and P. Echeverria. 1997. A simple polymerase chain reaction technique to detect and differentiate Shigella and enteroinvasive Escherichia coli in human feces. Diagn. Microbiol. Infect. Dis. 28:19–25.   65. Hutchinson, R. I. 1956. Some observations on the method of spread of Sonne dysentery. Mon. Bull. Minist. Health Public Health Lab Serv. 15:110–118.

SMP_Food Microbiology_CH15.indd

394

  66. Hyams, K. C., A. L. Bourgeois, B. R. Merrell, P. Rozmajzl, J. Escamilla, S. A. Thornton, G. M. Wasserman, A. Burke, P. Echeverria, K. Y. Green, et al. 1991. Diarrheal disease during Operation Desert Shield. N. Engl. J. Med. 325:1423–1428.   67. Islam, M. S., M. K. Hasan, and S. I. Khan. 1993. Growth and survival of Shigella flexneri in common Bangladeshi foods under various conditions of time and temperature. Appl. Environ. Microbiol. 59:652–654.   68. Islam, M. S., M. Z. Hossain, S. I. Khan, A. Felsenstein, R. B. Sack, and M. J. Albert. 1997. Detection of nonculturable Shigella dysenteriae 1 from artificially contaminated volunteers’ fingers using fluorescent anti­body and PCR techniques. J. Diarrhoeal Dis. Res. 15:65–70.   69. Izumiya, H., Y. Tada, K. Ito, T. Morita-Ishihara, M. Ohnishi, J. Terajima, and H. Watanabe. 2009. Characterization of Shigella sonnei isolates from travel-associated cases in Japan. J. Med. Microbiol. 58:1486–1491.   70. Jeong, H. J., E. S. Jang, B. I. Han, K. H. Lee, M. S. Ock, H. H. Kong, D. Il Chung, S. Y. Seol, D. T. Cho, and H. S. Yu. 2007. Acanthamoeba: could it be an environmental host of Shigella? Exp. Parasitol. 115:181–186.   71. Jiang, Y., F. Yang, X. Zhang, J. Yang, L. Chen, Y. Yan, H. Nie, Z. Xiong, J. Wang, J. Dong, Y. Xue, X. Xu, Y. Zhu, S. Chen, and Q. Jin. 2005. The complete sequence and analysis of the large virulence plasmid pSS of Shigella sonnei. Plasmid 54:149–159.   72. Jin, Q., Z. Yuan, J. Xu, Y. Wang, Y. Shen, W. Lu, J. Wang, H. Liu, J. Yang, F. Yang, X. Zhang, J. Zhang, G. Yang, H. Wu, D. Qu, J. Dong, L. Sun, Y. Xue, A. Zhao, Y. Gao, J. Zhu, B. Kan, K. Ding, S. Chen, H. Cheng, Z. Yao, B. He, R. Chen, D. Ma, B. Qiang, Y. Wen, Y. Hou, and J. Yu. 2002. Genome sequence of Shigella flexneri 2a: insights into pathogenicity through comparison with genomes of Escherichia coli K12 and O157. Nucleic Acids Res. 30:4432–4441.   73. Jin, Y. H., Y. H. Oh, J. H. Jung, S. J. Kim, J. A. Kim, K. Y. Han, M. Y. Kim, S. G. Park, and Y. K. Lee. 2010. Antimicrobial resistance patterns and characterization of integrons of Shigella sonnei isolates in Seoul, 1999– 2008. J. Microbiol. 48:236–242.   74. Jonsson, J., M. C. Alvarez-Castillo, J. C. Sanz, R. Ramiro, E. Ballester, M. Fernanez, M. A. Echeita, and N. F. Martinez. 2005. Late detection of a shigellosis outbreak in a school in Madrid. Euro Surveill. 10:268–270.   75. Jouihri, N., M. P. Sory, A. L. Page, P. Gounon, C. Parsot, and A. Allaoui. 2003. MxiK and MxiN interact with the Spa47 ATPase and are required for transit of the needle components MxiH and MxiI, but not of Ipa proteins, through the type III secretion apparatus of Shigella flexneri. Mol. Microbiol. 49:755–767.   76. Kaminski, R. W., and E. V. Oaks. 2009. Inactivated and subunit vaccines to prevent shigellosis. Exp. Rev. Vaccines 8:1693–1704.   77. Kane, C. D., R. Schuch, W. A. Day, Jr., and A. T. Maurelli. 2002. MxiE regulates intracellular expression of factors secreted by the Shigella flexneri 2a type III secretion system. J. Bacteriol. 184:4409–4419.

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

395

  78. Kennedy, F. M., J. Astbury, J. R. Needham, and T. Cheasty. 1993. Shigellosis due to occupational contact with non-human primates. Epidemiol. Infect. 110:247–251.   79. Kerneis, S., P. J. Guerin, L. von Seidlein, D. Legros, and R. F. Grais. 2009. A look back at an ongoing problem: Shigella dysenteriae type 1 epidemics in refugee settings in Central Africa (1993–1995). PLoS One 4(2):e4494.   80. Kim, D. W., G. Lenzen, A. L. Page, P. Legrain, P. J. Sansonetti, and C. Parsot. 2005. The Shigella flexneri effector OspG interferes with innate immune responses by targeting ubiquitin-conjugating enzymes. Proc. Natl. Acad. Sci. USA 102:14046–14051.   81. Kimura, A. C., K. Johnson, M. S. Palumbo, J. Hopkins, J. C. Boase, R. Reporter, M. Goldoft, K. R. Stefonek, J. A. Farrar, T. J. Van Gilder, and D. J. Vugia. 2004. Multistate shigellosis outbreak and commercially prepared food, United States. Emerg. Infect. Dis. 10:1147–1149.   82. Konkel, M. E., and K. Tilly. 2000. Temperature-regulated expression of bacterial virulence genes. Microbes Infect. 2:157–166.   83. Kosek, M., P. P. Yori, and M. P. Olortegui. 2010. Shigellosis update: advancing antibiotic resistance, investment empowered vaccine development, and green bananas. Curr. Opin. Infect. Dis. 23:475–480.   84. Kothary, M. H., and U. S. Babu. 2001. Infective dose of foodborne pathogens in volunteers: a review. J. Food Saf. 21:49–73.   85. Kotloff, K. L., J. P. Winickoff, B. Ivanoff, J. D. Clemens, D. L. Swerdlow, P. J. Sansonetti, G. K. Adak, and M. M. Levine. 1999. Global burden of Shigella infections: implications for vaccine development and implementation of control strategies. Bull. W. H. O. 77:651–666.   86. LaBrec, E. H., H. Schneider, T. J. Magnani, and S. B. Formal. 1964. Epithelial cell penetration as an essential step in the pathogenesis of bacillary dysentery. J. Bacteriol. 88:1503–1518.   87. Lan, R., M. C. Alles, K. Donohoe, M. B. Martinez, and P. R. Reeves. 2004. Molecular evolutionary relationships of enteroinvasive Escherichia coli and Shigella spp. Infect. Immun. 72:5080–5088.   88. Lawlor, K. M., P. A. Daskaleros, R. E. Robinson, and S. M. Payne. 1987. Virulence of iron transport mutants of Shigella flexneri and utilization of host iron compounds. Infect. Immun. 55:594–599.   89. Le, G. T., M. Mavris, M. C. Martino, M. L. Bernardini, E. Denamur, and C. Parsot. 2005. Analysis of virulence plasmid gene expression defines three classes of effectors in the type III secretion system of Shigella flexneri. Microbiology 151:951–962.   90. Levine, M. M., K. L. Kotloff, E. M. Barry, M. F. Pasetti, and M. B. Sztein. 2007. Clinical trials of Shigella vaccines: two steps forward and one step back on a long, hard road. Nat. Rev. Microbiol. 5:540–553.   91. Lew, J. F., D. L. Swerdlow, M. E. Dance, P. M. Griffin, C. A. Bopp, M. J. Gillenwater, T. Mercatante, and R. I.

SMP_Food Microbiology_CH15.indd

395

Glass. 1991. An outbreak of shigellosis aboard a cruise ship caused by a multiple-antibiotic-resistant strain of Shigella-flexneri. Am. J. Epidemiol. 134:413–420.   92. Lewis, H. C., S. Ethelberg, K. E. P. Olsen, E. M. Nielsen, M. Lisby, S. B. Madsen, J. Boel, R. Stafford, M. Kirk, H. V. Smith, S. Tikumrum, A. Wisetrojana, A. Bangtrakulnonth, J. Vithayarungruangsri, P. Siriarayaporn, K. Ungchusak, J. Bishop, and K. Molbak. 2009. Outbreaks of Shigella sonnei infections in Denmark and Australia linked to consumption of imported raw baby corn. Epidemiol. Infect. 137:326–334.   93. Lewis, H. C., M. Kirk, S. Ethelberg, R. Stafford, K. Olsen, E. M. Nielsen, M. Lisby, S. B. Madsen, and K. Molbak. 2007. Outbreaks of shigellosis in Denmark and Australia associated with imported baby corn, August 2007—final summary. Euro Surveill. 12:E071004.   94. Li, H., H. Xu, Y. Zhou, J. Zhang, C. Long, S. Li, S. Chen, J. M. Zhou, and F. Shao. 2007. The phosphothreonine lyase activity of a bacterial type III effector family. Science 315:1000–1003.   95. Li, Y. Y., B. Y. Cao, B. Liu, D. Liu, Q. L. Gao, X. Peng, J. L. Wu, D. A. Bastin, L. Feng, and L. Wang. 2009. Molecular detection of all 34 distinct O-antigen forms of Shigella. J. Med. Microbiol. 58:69–81.   96. Lofdahl, M., S. Ivarsson, S. Andersson, J. Langmark, and L. Plym-Forshell. 2009. An outbreak of Shigella dysenteriae in Sweden, May-June 2009, with sugar snaps as the suspected source. Euro Surveill. 14(28):pii 19268.   97. Lynch, M. F., R. V. Tauxe, and C. W. Hedberg. 2009. The growing burden of foodborne outbreaks due to contaminated fresh produce: risks and opportunities. Epidemiol. Infect. 137:307–315.   98. Magdalena, J., A. Hachani, M. Chamekh, N. Jouihri, P. Gounon, A. Blocker, and A. Allaoui. 2002. Spa32 regulates a switch in substrate specificity of the type III secreton of Shigella flexneri from needle components to Ipa proteins. J. Bacteriol. 184:3433–3441.   99. Mandal, J., N. Mondal, S. Mahadevan, and S. C. Parija. 2010. Emergence of resistance to third-generation cephalosporin in Shigella—a case report. J. Trop. Pediatr. 56:278–279. 100. Mandomando, I., D. Jaintilal, M. J. Pons, X. Valles, M. Espasa, L. Mensa, B. Sigauque, S. Sanz, J. Sacarlal, E. Macete, F. Abacassamo, P. L. Alonso, and J. Ruiz. 2009. Antimicrobial susceptibility and mechanisms of resistance in Shigella and Salmonella isolates from children under five years of age with diarrhea in rural Mozambique. Antimicrob. Agents Chemother. 53:2450–2454. 101. Marteyn, B., N. P. West, D. F. Browning, J. A. Cole, J. G. Shaw, F. Palm, J. Mounier, M. C. Prevost, P. Sansonetti, and C. M. Tang. 2010. Modulation of Shigella virulence in response to available oxygen in vivo. Nature 465:355–358. 102. Martinez-Argudo, I., and A. J. Blocker. 2010. The Shigella T3SS needle transmits a signal for MxiC release, which controls secretion of effectors. Mol. Microbiol. 78:1365–1378.

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

396 103. Maurelli, A. T. 2007. Black holes, antivirulence genes, and gene inactivation in the evolution of bacterial pathogens. FEMS Microbiol. Lett. 267:1–8. 104. Maurelli, A. T., B. Baudry, H. D’Hauteville, T. L. Hale, and P. J. Sansonetti. 1985. Cloning of plasmid DNA sequences involved in invasion of HeLa cells by Shigella flexneri. Infect. Immun. 49:164–171. 105. Maurelli, A. T., B. Blackmon, and R. Curtiss III. 1984. Temperature-dependent expression of virulence genes in Shigella species. Infect. Immun. 43:195–201. 106. Maurelli, A. T., R. E. Fernandez, C. A. Bloch, C. K. Rode, and A. Fasano. 1998. “Black holes” and bacterial pathogenicity: a large genomic deletion that enhances the virulence of Shigella spp. and enteroinvasive Escherichia coli. Proc. Natl. Acad. Sci. USA 95:3943–3948. 107. Maurelli, A. T., and P. J. Sansonetti. 1988. Identification of a chromosomal gene controlling temperature­regulated expression of Shigella virulence. Proc. Natl. Acad. Sci. USA 85:2820–2824. 108. Mavris, M., A. L. Page, R. Tournebize, B. Demers, P. Sansonetti, and C. Parsot. 2002. Regulation of transcription by the activity of the Shigella flexneri type III secretion apparatus. Mol. Microbiol. 43:1543–1553. 109. Mavris, M., P. J. Sansonetti, and C. Parsot. 2002. Identification of the cis-acting site involved in activation of promoters regulated by activity of the type III secretion apparatus in Shigella flexneri. J. Bacteriol. 184:6751–6759. 110. McKenna, S., C. Beloin, and C. J. Dorman. 2003. In vitro DNA-binding properties of VirB, the Shigella flexneri virulence regulatory protein. FEBS Lett. 545:183–187. 111. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5:607–625. 112. Menard, R., M. C. Prevost, P. Gounon, P. Sansonetti, and C. Dehio. 1996. The secreted Ipa complex of Shigella flexneri promotes entry into mammalian cells. Proc. Natl. Acad. Sci. USA 93:1254–1258. 113. Menard, R., P. Sansonetti, and C. Parsot. 1994. The secretion of the Shigella flexneri Ipa invasins is activated by epithelial cells and controlled by IpaB and IpaD. EMBO J. 13:5293–5302. 114. Menard, R., P. J. Sansonetti, and C. Parsot. 1993. Nonpolar mutagenesis of the ipa genes defines IpaB, IpaC, and IpaD as effectors of Shigella flexneri entry into epithelial cells. J. Bacteriol. 175:5899–5906. 115. Miller, M. A., J. Sentz, M. A. Rabaa, and E. D. Mintz. 2008. Global epidemiology of infections due to Shigella, Salmonella serotype Typhi, and enterotoxigenic Escherichia coli. Epidemiol. Infect. 136:433–435. 116. Morita-Ishihara, T., M. Ogawa, H. Sagara, M. Yoshida, E. Katayama, and C. Sasakawa. 2006. Shigella Spa33 is an essential C-ring component of type III secretion machinery. J. Biol. Chem. 281:599–607. 117. Moss, J. E., T. J. Cardozo, A. Zychlinsky, and E. A. Groisman. 1999. The selC-associated SHI-2 patho-

SMP_Food Microbiology_CH15.indd

396

genicity island of Shigella flexneri. Mol. Microbiol. 33:74–83. 118. Muller, L., T. Jensen, R. F. Petersen, K. Molbak, and S. Ethelberg. 2009. Imported fresh sugar peas as suspected source of an outbreak of Shigella sonnei in Denmark, April-May 2009. Euro Surveill. 14:pii 19241. 119. Nakamura, M. 1962. The survival of Shigella sonnei on cotton, glass, wood, paper, and metal at various temperatures. J. Hyg. (London) 60:35–39. 120. Nakamura, M., and B. C. Taylor. 1965. Survival of Shigella in biological materials. Health Lab Sci. 2:220–226. 121. Nassif, X., M. C. Mazert, J. Mounier, and P. J. Sansonetti. 1987. Evaluation with an iuc::Tn10 mutant of the role of aerobactin production in the virulence of Shigella flexneri. Infect. Immun. 55:1963–1969. 122. Nataro, J. P., J. Seriwatana, A. Fasano, D. R. Maneval, L. D. Guers, F. Noriega, F. Dubovsky, M. M. Levine, and J. G. Morris, Jr. 1995. Identification and cloning of a novel plasmid-encoded enterotoxin of enteroinvasive Escherichia coli and Shigella strains. Infect. Immun. 63:4721–4728. 123. Navarro-Garcia, F., J. Gutierrez-Jimenez, C. GarciaTovar, L. A. Castro, H. Salazar-Gonzalez, and V. Cordova. 2010. Pic, an autotransporter protein secreted by different pathogens in the Enterobacteriaceae family, is a potent mucus secretagogue. Infect. Immun. 78:4101–4109. 124. Nguyen Thi, K. N., V. Ha, T. N. Tran Vu, R. Stabler, T. D. Pham, T. M. Le, V, H. van Doorn, A. CerdenoTarraga, N. Thomson, J. Campbell, M. H. Nguyen Van, T. N. Tran Thi, V. M. Pham, T. T. Cao, B. Wren, J. Farrar, and S. Baker. 2010. The sudden dominance of bla(CTX-M) harbouring plasmids in Shigella spp. circulating in southern Vietnam. PLoS Negl. Trop. Dis. 4. 125. Oaks, E. V., T. L. Hale, and S. B. Formal. 1986. Serum immune response to Shigella protein antigens in rhesus monkeys and humans infected with Shigella spp. Infect. Immun. 53:57–63. 126. O’Brien, A. D., V. L. Tesh, A. Donohue-Rolfe, M. P. Jackson, S. Olsnes, K. Sandvig, A. A. Lindberg, and G. T. Keusch. 1992. Shiga toxin: biochemistry, genetics, mode of action, and role in pathogenesis. Curr. Top. Microbiol. Immunol. 180:65–94. 127. Ogawa, M., T. Yoshimori, T. Suzuki, H. Sagara, N. Mizushima, and C. Sasakawa. 2005. Escape of intracellular Shigella from autophagy. Science 307:727–731. 128. Page, A. L., H. Ohayon, P. J. Sansonetti, and C. Parsot. 1999. The secreted IpaB and IpaC invasins and their cytoplasmic chaperone IpgC are required for intercellular dissemination of Shigella flexneri. Cell. Microbiol. 1:183–193. 129. Parsot, C. 2009. Shigella type III secretion effectors: how, where, when, for what purposes? Curr. Opin. Microbiol. 12:110–116. 130. Parsot, C., E. Ageron, C. Penno, M. Mavris, K. Jamoussi, H. D’Hauteville, P. Sansonetti, and B. Demers. 2005. A secreted anti-activator, OspD1, and its chaperone, Spa15, are involved in the control of transcription by the

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

397

type III secretion apparatus activity in Shigella flexneri. Mol. Microbiol. 56:1627–1635. 131. Passwell, J. H., S. Ashkenzi, Y. Banet-Levi, R. RamonSaraf, N. Farzam, L. Lerner-Geva, H. Even-Nir, B. Yerushalmi, C. Chu, J. Shiloach, J. B. Robbins, and R. Schneerson. 2010. Age-related efficacy of Shigella Ospecific polysaccharide conjugates in 1-4-year-old Israeli children. Vaccine 28:2231–2235. 132. Peng, J. P., J. Yang, and Q. Jin. 2009. The molecular evolutionary history of Shigella spp. and enteroinvasive Escherichia coli. Infect. Genet. Evol. 9:147–152. 133. Picking, W. L., H. Nishioka, P. D. Hearn, M. A. Baxter, A. T. Harrington, A. Blocker, and W. D. Picking. 2005. IpaD of Shigella flexneri is independently required for regulation of Ipa protein secretion and efficient insertion of IpaB and IpaC into host membranes. Infect. Immun. 73:1432–1440. 134. Pilonieta, M. C., and G. P. Munson. 2008. The chaperone IpgC copurifies with the virulence regulator MxiE. J. Bacteriol. 190:2249–2251. 135. Prosseda, G., M. Falconi, M. Nicoletti, M. Casalino, G. Micheli, and B. Colonna. 2002. Histone-like proteins and the Shigella invasivity regulon. Res. Microbiol. 153:461–468. 136. Rajakumar, K., C. Sasakawa, and B. Adler. 1997. Use of a novel approach, termed island probing, identifies the Shigella flexneri she pathogenicity island which encodes a homolog of the immunoglobulin A protease-like family of proteins. Infect. Immun. 65:4606–4614. 137. Ram, P. K., J. A. Crump, S. K. Gupta, M. A. Miller, and E. D. Mintz. 2008. Part II. Analysis of data gaps pertaining to Shigella infections in low and medium human development index countries, 1984–2005. Epidemiol. Infect. 136:577–603. 138. Ramarao, N., C. C. Le, T. Izard, R. Bourdet-Sicard, E. Ageron, P. J. Sansonetti, M. F. Carlier, and N. G. Tran Van. 2007. Capping of actin filaments by vinculin activated by the Shigella IpaA carboxyl-terminal domain. FEBS Lett. 581:853–857. 139. Ranjbar, R., M. M. S. Dallal, M. Talebi, and M. R. Pourshafie. 2008. Increased isolation and characterization of Shigella sonnei obtained from hospitalized children in Tehran, Iran. J. Health Popul. Nutr. 26:426–430. 140. Reller, M. E., J. M. Nelson, K. Molbak, D. M. Ackman, D. J. Schoonmaker-Bopp, T. P. Root, and E. D. Mintz. 2006. A large, multiple-restaurant outbreak of infection with Shigella flexneri serotype 2a traced to tomatoes. Clin. Infect. Dis. 42:163–169. 141. Rohde, J. R., A. Breitkreutz, A. Chenal, P. J. Sansonetti, and C. Parsot. 2007. Type III secretion effectors of the IpaH family are E3 ubiquitin ligases. Cell Host Microbe 1:77–83. 142. Rout, W. R., S. B. Formal, R. A. Giannella, and G. J. Dammin. 1975. Pathophysiology of Shigella diarrhea in the rhesus monkey: intestinal transport, morphological, and bacteriological studies. Gastroenterology 68:270–278. 143. Sabra, A. H., G. F. Araj, M. M. Kattar, R. Y. Abi-Rached, M. T. Khairallah, J. D. Klena, and G. M. Matar. 2009.

SMP_Food Microbiology_CH15.indd

397

Molecular characterization of ESBL-producing Shigella sonnei isolates from patients with bacilliary dysentery in Lebanon. J. Infect. Dev. Ctries. 3:300–305. 144. Saeed, A., H. Abd, B. Edvinsson, and G. Sandstrom. 2009. Acanthamoeba castellanii an environmental host for Shigella dysenteriae and Shigella sonnei. Arch. Microbiol. 191:83–88. 145. Sandlin, R. C., K. A. Lampel, S. P. Keasler, M. B. Goldberg, A. L. Stolzer, and A. T. Maurelli. 1995. Avirulence of rough mutants of Shigella flexneri: requirement of O antigen for correct unipolar localization of IcsA in the bacterial outer membrane. Infect. Immun. 63:229–237. 146. Sasakawa, C. 2010. A new paradigm of bacteria-gut interplay brought through the study of Shigella. Proc. Jpn. Acad. B 86:229–243. 147. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. 148. Scheiring, J., A. Rosales, and L. B. Zimmerhackl. 2010. Clinical practice. Today’s understanding of the haemolytic uraemic syndrome. Eur. J. Pediatr. 169:7–13. 149. Schroeder, G. N., and H. Hilbi. 2008. Molecular pathogenesis of Shigella spp.: controlling host cell signaling, invasion, and death by type III secretion. Clin. Microbiol. Rev. 21:134–156. 150. Schuch, R., and A. T. Maurelli. 1999. The mxi-Spa type III secretory pathway of Shigella flexneri requires an outer membrane lipoprotein, MxiM, for invasin translocation. Infect. Immun. 67:1982–1991. 151. Shere, K. D., S. Sallustio, A. Manessis, T. G. D’Aversa, and M. B. Goldberg. 1997. Disruption of IcsP, the major Shigella protease that cleaves IcsA, accelerates actinbased motility. Mol. Microbiol. 25:451–462. 152. Simon, D. G., R. A. Kaslow, J. Rosenbaum, R. L. Kaye, and A. Calin. 1981. Reiter’s syndrome following epidemic shigellosis. J. Rheumatol. 8:969–973. 153. Srinivasa, H., M. Baijayanti, and Y. Raksha. 2009. Magnitude of drug resistant shigellosis: a report from Bangalore. Indian J. Med. Microbiol. 27:358–360. 154. Stafford, R., M. Kirk, C. Selvey, D. Staines, H. Smith, C. Towner, and M. Salter. 2007. An outbreak of multiresistant Shigella sonnei in Australia: possible link to the outbreak of shigellosis in Denmark associated with imported baby corn from Thailand. Euro Surveill. 12: E070913. 155. Stensrud, K. F., P. R. Adam, C. D. La Mar, A. J. Olive, G. H. Lushington, R. Sudharsan, N. L. Shelton, R. S. Givens, W. L. Picking, and W. D. Picking. 2008. Deoxycholate interacts with IpaD of Shigella flexneri in inducing the recruitment of IpaB to the type III secretion apparatus needle tip. J. Biol. Chem. 283:18646–18654. 156. Sur, D., T. Ramamurthy, J. Deen, and S. K. Bhattacharya. 2004. Shigellosis : challenges & management issues. Indian J. Med. Res. 120:454–462. 157. Tamano, K., S. Aizawa, E. Katayama, T. Nonaka, S. Imajoh-Ohmi, A. Kuwae, S. Nagai, and C. Sasakawa. 2000. Supramolecular structure of the Shigella type III secretion machinery: the needle part is changeable in

Manila Typesetting Company

11/08/2012 06:59AM

Foodborne Pathogenic Bacteria

398 length and essential for delivery of effectors. EMBO J. 19:3876–3887. 158. Tamano, K., E. Katayama, T. Toyotome, and C. Sasakawa. 2002. Shigella Spa32 is an essential secretory protein for functional type III secretion machinery and uniformity of its needle length. J. Bacteriol. 184:1244–1252. 159. Tobe, T., S. Nagai, N. Okada, B. Adler, M. Yoshikawa, and C. Sasakawa. 1991. Temperature-regulated expression of invasion genes in Shigella flexneri is controlled through the transcriptional activation of the virB gene on the large plasmid. Mol. Microbiol. 5:887–893. 160. Tobe, T., M. Yoshikawa, T. Mizuno, and C. Sasakawa. 1993. Transcriptional control of the invasion regulatory gene virB of Shigella flexneri: activation by virF and repression by H-NS. J. Bacteriol. 175:6142–6149. 161. Todd, E. C., J. D. Greig, C. A. Bartleson, and B. S. Michaels. 2008. Outbreaks where food workers have been implicated in the spread of foodborne disease. Part 4. Infective doses and pathogen carriage. J. Food Prot. 71:2339–2373. 162. Touchon, M., C. Hoede, O. Tenaillon, V. Barbe, S. Baeriswyl, P. Bidet, E. Bingen, S. Bonacorsi, C. Bouchier, O. Bouvet, A. Calteau, H. Chiapello, O. Clermont, S. Cruveiller, A. Danchin, M. Diard, C. Dossat, M. E. Karoui, E. Frapy, L. Garry, J. M. Ghigo, A. M. Gilles, J. Johnson, C. Le Bouguenec, M. Lescat, S. Mangenot, V. Martinez-Jehanne, I. Matic, X. Nassif, S. Oztas, M. A. Petit, C. Pichon, Z. Rouy, C. S. Ruf, D. Schneider, J. Tourret, B. Vacherie, D. Vallenet, C. Medigue, E. P. C. Rocha, and E. Denamur. 2009. Organised genome dynamics in the Escherichia coli species results in highly diverse adaptive paths. PLoS Genet. 5:e1000344. 163. Tran Van Nhieu, G., E. Caron, A. Hall, and P. J. Sansonetti. 1999. IpaC induces actin polymerization and filopodia formation during Shigella entry into epithelial cells. EMBO J. 18:3249–3262. 164. Van, G. F., S. Genin, and C. Boucher. 1993. Conservation of secretion pathways for pathogenicity determinants of plant and animal bacteria. Trends Microbiol. 1:175–180. 165. Venkatesan, M. M., M. B. Goldberg, D. J. Rose, E. J. Grotbeck, V. Burland, and F. R. Blattner. 2001. Complete DNA sequence and analysis of the large virulence plasmid of Shigella flexneri. Infect. Immun. 69:3271–3285. 166. Vokes, S. A., S. A. Reeves, A. G. Torres, and S. M. Payne. 1999. The aerobactin iron transport system genes in Shigella flexneri are present within a pathogenicity island. Mol. Microbiol. 33:63–73. 167. von Seidlein, L., D. R. Kim, M. Ali, H. Lee, X. Wang, V. D. Thiem, d. G. Canh, W. Chaicumpa, M. D. Agtini, A. Hossain, Z. A. Bhutta, C. Mason, O. Sethabutr, K. Talukder, G. B. Nair, J. L. Deen, K. Kotloff, and J. Clemens. 2006. A multicentre study of Shigella ­diarrhoea in six Asian countries: disease burden,­ clinical manifestations, and microbiology. PLoS Med. 3:e353. 168. Vrints, M., E. Mairiaux, E. Van Meervenne, J. M. Collard, and S. Bertrand. 2009. Surveillance of antibi-

SMP_Food Microbiology_CH15.indd

398

otic susceptibility patterns among Shigella sonnei strains isolated in Belgium during the 18-year period 1990 to 2007. J. Clin. Microbiol. 47:1379–1385. 169. Warren, B. R., M. E. Parish, and K. R. Schneider. 2006. Shigella as a foodborne pathogen and current methods for detection in food. Crit. Rev. Food Sci. Nutr. 46:551–567. 170. Warren, B. R., H. G. Yuk, and K. R. Schneider. 2007. Survival of Shigella sonnei on smooth tomato surfaces, in potato salad and in raw ground beef. Int. J. Food Microbiol. 116:400–404. 171. Watarai, M., S. Funato, and C. Sasakawa. 1996. Interaction of Ipa proteins of Shigella flexneri with α5β1 integrin promotes entry of the bacteria into mammalian cells. J. Exp. Med. 183:991–999. 172. Watarai, M., T. Tobe, M. Yoshikawa, and C. Sasakawa. 1995. Contact of Shigella with host cells triggers release of Ipa invasins and is an essential function of invasiveness. EMBO J. 14:2461–2470. 173. Wong, M. R., V. Reddy, H. Hanson, K. M. Johnson, B. Tsoi, C. Cokes, L. Gallagher, L. Lee, A. Plentsova, T. Dang, A. Krueger, K. Joyce, and S. Balter. 2010. Antimicrobial resistance trends of Shigella serotypes in New York City, 2006-2009. Microb. Drug Resist. 16:155–161. 174. Woodward, D. L., C. G. Clark, R. A. Caldeira, R. Ahmed, G. Soule, L. Bryden, H. Tabor, P. Melito, R. Foster, J. Walsh, L. K. Ng, G. B. Malcolm, N. Strockbine, F. G. Rodgers, and the Canadian Public Health Lab Network. 2005. Identification and characterization of Shigella boydii 20 serovar nov., a new and emerging Shigella serotype. J. Med. Microbiol. 54:741–748. 175. Yang, F., J. Yang, X. B. Zhang, L. H. Chen, Y. Jiang, Y. L. Yan, X. D. Tang, J. Wang, Z. H. Xiong, J. Dong, Y. Xue, Y. F. Zhu, X. Y. Xu, L. L. Sun, S. X. Chen, H. Nie, J. P. Peng, J. G. Xu, Y. Wang, Z. H. Yuan, Y. M. Wen, Z. J. Yao, Y. Shen, B. Q. Qiang, Y. D. Hou, J. Yu, and Q. Jin. 2005. Genome dynamics and diversity of Shigella species, the etiologic agents of bacillary dysentery. Nucleic Acids Res. 33:6445–6458. 176. Yang, J., H. Nie, L. H. Chen, X. B. Zhang, F. Yang, X. Y. Xu, Y. F. Zhu, J. Yu, and Q. Jin. 2007. Revisiting the molecular evolutionary history of Shigella spp. J. Mol. Evol. 64:71–79. 177. Yang, J., V. Sangal, Q. Jin, and J. Yu. 2011. Shigella genomes: a tale of convergent evolution and specialization through IS expansion and genome reduction, p. 23–39. In P. Fratamico, Y. Liu, and S. Kathariou (ed.), Genomes of Foodborne and Waterborne Pathogens. ASM Press, Washington, DC. 178. Zagrebneviene, G., V. Jasulaitiene, B. Morkunas, S. Tarbunas, and J. Ladygaite. 2005. Shigella sonnei outbreak due to consumption of unpasteurised milk curds in Vilnius, Lithuania, 2004. Euro Surveill. 10: E051201. 179. Zaika, L. L. 2002. The effect of NaCl on survival of Shigella flexneri in broth as affected by temperature and pH. J. Food Prot. 65:774–779.

Manila Typesetting Company

11/08/2012 06:59AM

15.  Shigella Species

399

180. Zaika, L. L., and J. G. Phillips. 2005. Model for the combined effects of temperature, pH and sodium chloride concentration on survival of Shigella flexneri strain 5348 under aerobic conditions. Int. J. Food Microbiol. 101:179–187. 181. Zhang, G., and K. A. Lampel. 2010. Comparison of chromogenic Biolog Rainbow agar Shigella/Aeromonas

SMP_Food Microbiology_CH15.indd

399

with xylose lysine desoxycholate agar for isolation and detection of Shigella spp. from foods. J. Food Prot. 73:1458–1465. 182. Zychlinsky, A., B. Kenny, R. Menard, M. C. Prevost, I. B. Holland, and P. J. Sansonetti. 1994. IpaB mediates macrophage apoptosis induced by Shigella flexneri. Mol. Microbiol. 11:619–627.

Manila Typesetting Company

11/08/2012 06:59AM

SMP_Food Microbiology_CH15.indd

400

Manila Typesetting Company

11/08/2012 06:59AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch16

James D. Oliver Carla Pruzzo Luigi Vezzulli James B. Kaper

Vibrio Species

Whereas the 8th edition of Bergey’s Manual of Systematic Bacteriology listed five Vibrio species with two recognized as human pathogens, over 80 species have now been described (230) (http://www.bacterio. cict.fr/uw/vibrio.html), including at least 12 capable of causing infection in humans. Many reviews have been published on the pathogenic vibrios over the years (11, 109, 156, 166, 175, 178, 191), but with the exception of Vibrio cholerae, Vibrio parahaemolyticus, and Vibrio vulnificus, relatively little is known about the virulence mechanisms they employ. Of the 12 human pathogens, 8 have been directly associated with foods, and these are the subject of this review.

NATURAL HABITATS The presence and distribution of Vibrio species in coastal waters are dependent upon a variety of environmental factors such as temperature, salt concentration, pH, and nutrients. Vibrios tend to be more common in warmer waters, especially when temperatures rise above 17°C, and depending on the species, they tolerate a range of salinities. Vibrios that require small amounts of Na+ for

16 growth (e.g., V. cholerae and Vibrio mimicus) are also present in freshwater rivers and lakes (3). In marine and estuarine environments, vibrios are commonly isolated from sediment, the water column, and various plants and vertebrate and invertebrate animals. The association of vibrios with planktonic organisms, especially copepods, has been suggested as an important component of Vibrio ecology, especially for V. cholerae (129, 235, 240). Plankton represents an organically rich microenvironment where the high nutrient concentrations of the plankton microhabitat can selectively enrich heterotrophic bacteria, including vibrios (97, 129). Vibrio species are commonly isolated from filter-feeding shellfish, such as oysters, where they may be present at concentrations that are 100-fold higher than those in the surrounding water (68). For that reason, most foodborne vibrio infections are caused by consumption of oysters and other bivalves that are consumed frequently without processing or cooking.

INCIDENCE OF VIBRIOS IN SEAFOOD One of the most consistent features of human vibrio infections is a recent history of seafood consumption.

James D. Oliver, Department of Biology, University of North Carolina at Charlotte, 9201 University City Blvd., Charlotte, NC 28223. Carla Pruzzo and Luigi Vezzulli, Department of Biology, University of Genova, Corso Europa 26, 16132 Genova, Italy. James B. Kaper, Department of Microbiology and Immunology, Center for Vaccine Development, University of Maryland School of Medicine, 685 West Baltimore St., Baltimore, MD 21201.

401

402 Vibrios, which are generally the predominant bacterial genus in estuarine waters, are associated with a great variety of seafood. To cite a few studies, Gopal et al. (74) reported shrimp samples from India had a mean of up to 4.4 × 104 CFU of Vibrio spp. per gram, with Vibrio alginolyticus, V. parahaemolyticus, V. vulnificus, Vibrio fluvialis, V. mimicus, and occasionally V. cholerae identified. Buck (19) reported that 36 to 60% of finfish and shellfish sampled from supermarkets were contaminated with Vibrio spp., with V. parahaemolyticus and V. alginolyticus most commonly isolated. The recovery of vibrios from molluscan shellfish was substantially greater during the summer months. Das et al. (45) determined the prevalence of V. parahaemolyticus in 293 samples of finfish and shrimp obtained from wholesale markets in India and found that over 45% of shellfish and nearly 17% of finfish were positive. In a study of the distribution of vibrios in oysters (Crassostrea virgi­ nica) originating from the coast of Brazil, Matté et al. (142) determined the following order of prevalence: V. alginolyticus (81%), V. parahaemolyticus (77%), V. cholerae non-O1 (31%), V. fluvialis (27%), V. furnis­ sii (19%), V. mimicus (12%), and V. vulnificus (12%). Zimmerman et al. (261) determined the prevalence of V. parahaemolyticus in oysters obtained from two sites in the Gulf of Mexico and found pathogenic strains in 44 to 56% of samples. Elhadi et al. (58), examining 768 samples of a variety of seafoods from Malaysian markets, detected eight different pathogenic Vibrio spp., with all eight present in some shrimp and cockles. Their results revealed that seafoods in Malaysia were typically contaminated with potentially pathogenic vibrios regardless of the season. A study conducted in Thailand revealed that processed and ready-to-eat seafood were contaminated with at least one of the potentially pathogenic vibrios at significant frequencies (25 and 17.5% of samples, respectively), with cell populations as high as 103 to 104 per gram in some samples (34).

ISOLATION Available “official” or “reference” methods are based on traditional culture techniques, although they are generally incapable of reliably determining levels of Vibrio spp. in seafood or the environments where they are produced (52). Isolation of vibrios from foods is facilitated by the use of media formulated with an alkaline pH (52), and alkaline peptone water (APW) is used commonly for isolating several species of concern. This is frequently coupled with thiosulfate-citrate-bile salts-sucrose (TCBS) agar or other plating media selective for vibrios. Employing sucrose as a differentiating trait, the 12 human pathogenic vibrios

Foodborne Pathogenic Bacteria can be separated on TCBS into six species that are generally sucrose positive (V. cholerae, Vibrio metschnikovii, V. fluvialis, Vibrio furnissii, V. alginolyticus, and Vibrio carchariae) and five that are generally sucrose negative (V. mimicus, Vibrio hollisae, Vibrio damsela, V. parahae­ molyticus, and V. vulnificus). While most of the vibrios grow well on TCBS agar, V. hollisae exhibits very poor to no growth on this medium, which is also the case for V. damsela when incubated at 37°C. More-recent formulations for selective agars for the isolation of V. vulnificus have also proved effective. Among these are modified cellobiose polymyxin colistin (mCPC) (59) and cellobiose colistin (CC) (96) agars, formulated to differentiate V. vulnificus from other vibrios. V. cholerae strains, except the classical biotype, will grow on mCPC agar, whereas most V. parahaemolyticus strains and other species will not. Incubation at 40°C inhibits nontarget bacteria and maintains selectivity of mCPC. A chromogenic agar medium has been developed for the isolation and preliminary identification of several pathogenic Vibrio spp. (84). For reviews of enrichment and plating media for the vibrios, see Oliver (173) and Harwood et al. (87).

IDENTIFICATION The salient differentiating features of the eight human pathogenic vibrios associated with foods are shown in Table 16.1. Comments regarding the taxonomy of individual species are noted below. Accurate phenotypic identification of Vibrio species is problematic, largely because of the great variability in biochemical characteristics that they exhibit (226). Many isolates are not accurately identified by commercial methods, with the accuracy of systems ranging from 63.9% to 80.9% (169). Most members of the genus are halophilic, and the addition of NaCl is often required for isolation; however, the concentration of NaCl can affect the biochemical profile and lead to erroneous identification. The characteristics shared by Aeromonas and Vibrio sometimes also result into classification of isolates into the wrong genus. Additional drawbacks of biochemical methods are that they are time­consuming (i.e., reaction mixtures must incubate for an extended time period, often requiring 2 to 7 days to complete) and that the interpretation of results requires specialized training that may not be available to all ­laboratories (226). Some researchers have developed immunological assays in an attempt to overcome this problem, whereas most now employ molecular techniques as the primary tool to this end (87). Molecular methods that utilize the PCR and nucleotide sequence determination over-

16.  Vibrio Species

403

Table 16.1  Key differential traits of pathogenic food-associated Vibrio speciesa % of strains positive for testb Test

V. cholerae

V. mimicus

V. hollisae

V. fluvialisc

V. alginolyticus

V. parahaemolyticus

V. vulnificus

Voges-Proskauer (1% NaCl)

75

9

0

0

95

0

0

Motility

99

98

0

70–89

99

99

99

100

0

0

100

99

1

15

  d-Mannitol

99

99

0

97

100

100

45

  Cellobiose

8

0

0

30

3

4

99

  Salicin

1

0

0

0

4

1

95

Acid production from:   Sucrose

a b c

Adapted from Oliver and Kaper (180) and Abbott et al. (3). After 48 h of incubation at 36°C. Most of the positive reactions occurred during the first 24 h. Includes V. furnissii, which differs from V. fluvialis primarily by production of gas in d-glucose.

come many of the limitations of phenotypic methods. PCR-based methods can lead to identification of an isolate within hours as opposed to days and can be used on small quantities of cells, including those that are not viable or are otherwise unculturable. For example, DNA primers targeting the thermolabile hemolysin (tlh) and the thermostable direct hemolysin (tdh) genes are used to confirm total and pathogenic V. parahae­ molyticus, respectively. Similarly, the hemolysin gene (vvhA) is used as a target specific for V. vulnificus, as are the 16S-23S rRNA intergenic spacer region and ctx for total and toxin-producing strains of V. cholerae, respectively. Sequential and/or multiplex PCR systems were recently developed for the detection of the most frequent foodborne pathogenic Vibrio species (30, 60, 151, 243). To establish the relatedness of Vibrio isolates from disease epidemics, discriminate among strains with more or less potential to cause disease or epidemics, and explore the population biology of these bacteria, DNAbased methods, including whole-genome approaches, such as pulsed-field gel electrophoresis (PFGE), ribotyping, repetitive extragenic palindromic PCR, and both single- and multiple-gene targets (multilocus approaches), are often employed. Typing methods that target repetitive elements distributed throughout the genome, such as BOX-PCR and repetitive extragenic palindromic PCR, and DNA sequence-based methods such as multilocus sequence typing (MLST), are also highly discriminatory and, in some cases, superior to those previously listed for phylogenetic analysis and identification of strains with high epidemic or virulence potential (218).

VIBRIO ENUMERATION IN SEAFOOD Products such as seafood are generally homogenized in phosphate-buffered saline. Depending on the pathogen, the Food and Drug Administration (FDA) Bacteriological Analytical Manual (111) indicates three approaches for enumeration of vibrios. The first is a most-probable­number endpoint titration of replicate samples in enrichment broth cultures coupled with identification of suspect isolates using biochemical profiles, DNA probe colony hybridization, or PCR. The second is a membrane filtration procedure using hydrophobic grid membrane filters followed by identification of representative colonies. The third is a direct plating method using DNA probes for identification of the total population and pathogenic isolates. The ultimate goal for Vibrio detection and enumeration in oysters is the development of more rapid assays that do not require enrichment but still retain the required sensitivity (253). Most often, however, a brief enrichment period is required (185, 253).

EPIDEMIOLOGY The cell numbers of most Vibrio spp. in both surface waters and shellfish exhibit a seasonal response, generally being greatest during the warm weather months. Seasonality is most notable for the isolation of V. vul­ nificus and V. parahaemolyticus from environmental sources, as well as the infections they cause, whereas those of some vibrios, such as V. fluvialis, occur throughout the year. In a recent review (101), it was reported that in the United States during 1973 to 2006, vibrios were the most

Foodborne Pathogenic Bacteria

404 commonly reported cause of seafood-associated outbreaks (28% of all infectious disease outbreaks). Specifically, V. parahaemolyticus caused more outbreaks and illnesses than any other pathogen (24% of all infectious disease outbreaks) during the study period. Other Vibrio species included toxigenic V. cholerae, non-O1 and nonO139 V. cholerae, and V. vulnificus. Among outbreaks of Vibrio illness, 72.2% were associated with mollusks, particularly oysters, whereas 27.8% were associated with crustaceans. The number of reported outbreaks of Vibrio infections increased over the study period, with the greatest number reported during the last decade, 1997 to 2006. The largest reported seafood-associated outbreak occurred in 1998, when V. parahaemolyticus infections associated with oyster consumption caused 416 illnesses among persons in 13 states. During 1997 to 2006, about 4,700 sporadic cases of Vibrio illness were reported to the Cholera and Other Vibrio Illness Surveillance System (101). Of these, 71.6% were classified as foodborne infections. Outbreaks and sporadic cases of Vibrio illnesses were sharply seasonal, with most occurring during the warmer months, corresponding to warmer water temperatures. While there is considerable variation in the severity of the different vibrio diseases, and even with infections caused by the same species, the most severely ill patients generally have preexisting underlying illnesses, with chronic liver disease being one of the most common. An exception to this generalization is V. cholerae O1/O139, which can readily cause disease in noncompromised individuals. In almost all cases, there is a recent history of seafood consumption, especially raw oysters. V. chol­ erae O1/O139 is also exceptional in having a broader vehicle range of infection, although seafood remains important in its transmission.

CONTROL MEASURES AND VIBRIO SUSCEPTIBILITY TO PHYSICAL/CHEMICAL TREATMENTS Control strategies to prevent seafood-associated illnesses include consumer education, monitoring of harvest waters for vibrios, and postharvest Vibrio reduction treatments. Consumers should be aware of the potential health risks associated with eating raw or undercooked shellfish, particularly persons with medical conditions such as liver disease that predispose them to severe illness. Thoroughly cooking shellfish and preventing raw seafood from cross-contaminating other foods are effective measures for consumers to reduce risk. Like other microbes present in water, vibrios can be concentrated in the tissues of filter-feeding bivalve mol-

lusks. Regulatory control measures include monitoring of harvest waters and microbiological sampling and testing of oysters (101). Since Vibrio species are indigenous to the aquatic environment, their presence and numbers typically do not correlate with indicators of fecal human pathogens. Therefore, monitoring waters for fecal coliform bacteria only is not effective as an indicator of the presence of Vibrio in harvest environments (101). Several reports have indicated that Vibrio foodborne infections are often associated with consumption of shellfish harvested from warmer waters and have underscored the need to examine the emerging health risks posed by changing environmental factors, such as increasing water temperatures (66, 144). One of the most common processing practices for molluscan shellfish sanitation is depuration, which relies on the mollusk’s inherent ability to clean itself of contamination when supplied with a source of pathogenfree water, usually treated either by ozone or UV light. This process is of considerable value in removing contaminating bacteria such as Salmonella and Escherichia coli (20). However, many studies on depuration of oysters and clams revealed that this method, by itself, does not significantly reduce the naturally occurring Vibrio microflora present in these animals (105, 225). Postharvest processing methods, such as high-­pressure treatment, irradiation, quick-freezing, and pasteurization, are available to make oysters safer (254). In a survey of postharvest-processed oysters in the United States, the FDA collected oysters that had been subjected to such postharvest processing methods to determine V. vulni­ ficus and V. parahaemolyticus levels (51). The lowest frequency of isolation of either pathogen (<10%) was observed with the mild heat (pasteurization) process; however, all postharvest processing treatments greatly reduce the risk to consumers of pathogens by raw oyster consumption. With the exception of V. cholerae, V. para­ haemolyticus, and V. vulnificus, relatively little is known of the susceptibility of vibrios to various food preservation methods. Summarized here are some of the studies relevant to this point.

Cold

While reports generally indicate that vibrios are sensitive to cold temperatures, seafoods have also been reported to be protective for vibrios at refrigeration temperatures. Wong et al. (250) isolated several psychrotrophic strains of V. mimicus, V. fluvialis, and V. parahaemolyticus from frozen seafoods and found these to survive well at 10, 4, and −30°C. Although V. para­ haemolyticus populations were initially rapidly reduced (ca. 99%) when incubated on whole shrimp at 3, 7, 10,

16.  Vibrio Species or −18°C, survivors remained at the end of our 8-day study. V. parahaemolyticus can also survive storage in shellstock oysters for at least 3 weeks at 4°C and subsequently multiply when held at 35°C for 2 to 3 days. Similarly, numbers of cells of V. parahaemolyticus were reduced in cooked fish mince and surimi at 5°C for 48 hours, but growth occurred when the product was held at 25°C. V. vulnificus failed to grow in oysters held at 13°C and below, but significant growth occurred in oysters stored at 18°C and higher (41). These studies suggest that naturally occurring vibrios are able to multiply in unchilled shellstock oysters. Similarly, studies involving temperature abuse of octopus, cooked shrimp, and crab meat have revealed growth of V. parahaemo­ lyticus to very large populations when held for even short periods of time under improper refrigeration. The possibility of V. parahaemolyticus cells entering into the “viable but nonculturable state” when exposed to temperatures below 10oC may occur. See the “Reservoirs” sections for V. cholerae and V. vulnificus for a brief description of and references for this phenomenon. The persistence of V. vulnificus in oysters following freezing and storage at −20°C was dependent on the length of frozen storage time for cells packaged without vacuum, with a decrease from ca. 105 to ca. 101 CFU/ gram. Vacuum-packaged samples had significantly lower populations of V. vulnificus over a 70-day study period than did conventionally packaged samples (189). Further, “individually quick frozen” oyster technology has advanced to providing oysters a shelf life of over 18 months and a reduction of V. vulnificus populations to nondetectable levels (91).

Heat

All of the vibrios are sensitive to heat, although a wide range of thermal inactivation rates have been reported. Heating of V. parahaemolyticus cells at 60, 80, or 100°C for 1 minute is lethal for small (5 × 102) populations, although some cells survived heating at 60°C and even 80°C for 15 minutes when initial populations of 2 × 105 were used (238). Cook and Ruple (42) reported that decimal reduction times at 47°C averaged 78 seconds for the 52 strains of V. vulnificus examined. Thorough heating of shellfish to provide an internal temperature of at least 60°C for several minutes should be sufficient to kill the pathogenic vibrios (238), and Cook and Ruple (42) determined that heating oysters in water at 50°C for 10 minutes was sufficient to reduce V. vulnificus populations to nondetectable levels. Low-temperature pasteurization (e.g., 50°C for 10 minutes) of oysters reduces V. vulnificus by 6 log CFU (8), and this is the basis for one of the major commercially employed methods

405 of postharvest processing oysters to reduce their public health hazard. Nascumento et al. (162) determined that boiling an O1 strain of V. cholerae inoculated into white shrimp (Penaeus schmitti) completely inactivated the vibrios within 1 to 2 minutes. The FDA recommends steaming shellstock oysters, clams, and mussels for 4 to 9 min, frying shucked oysters for 10 min at 375°C, or baking oysters for 10 min at 450°C.

Irradiation

Doses of 3 kGy of gamma irradiation kill vibrios in frozen shrimp (196). High levels (>50 kilorads) of 60Co have been used to eliminate V. cholerae from both fresh and frozen frog legs (206). A dose of 1 kGy of ionizing radiation reduces V. vulnificus in shellstock oysters by more than a 5 log CFU, with no mortality of the oysters. Higher doses, e.g., 1.5 kGy, completely inactivated V. vulnificus but resulted in oyster mortalities of up to 16% (54). Similar studies on V. parahaemolyticus in oysters revealed doses of 3 kGy killed 6 log CFU of the pathogen, with no oyster mortalities. The effects of lowdose gamma irradiation on vibrios in oysters have also been described by Andrews et al. (9). More recently, the effects of X rays to inactivate pathogenic vibrios in shellstock oysters has been determined, with 1 to 5 kGy reducing V. parahaemolyticus to <10 CFU/g of oyster (137) and 3 kGy reducing V. vulnificus populations >6 log CFU without affecting oyster survival (136).

High Pressure

Reductions of up to 6 log CFU of V. vulnificus occur after 10 minutes of high hydrostatic pressure (30,000 to 50,000 psi) treatment, and up to 9 log CFU of V. para­ haemolyticus after 30 seconds (24). A 5-log reduction of V. parahaemolyticus occurred in shellstock oysters receiving ³350 MPa for 2 min (123), whereas a ca. 4-log reduction of V. parahaemolyticus occurred in Pacific oysters following treatment with 293 MPa for 2 min at 8°C (134).

Other Treatments

A large variety of dried spices, oils of several herbs, tomato sauce, and several organic acids are bactericidal to V. parahaemolyticus, with many being highly toxic at low levels. Similarly, V. vulnificus is inactivated by several fruit or vegetable juices and a variety of spice extracts. V. parahaemolyticus is highly sensitive to as little as 50 ppm butylated hydroxyanisole and is inhibited by 0.1% sorbic acid. Of ten generally recognized as safe compounds tested against V. vulnificus, only diacetyl was inhibitory to the pathogen when present in shellstock oysters (220). Vibrios are highly acid ­ sensitive,

406 although growth in media as low as pH 4.8 has been reported for V. parahaemolyticus. While the various postharvest treatments are effective at killing seafood-associated bacteria, most of these treatments also kill the mollusks. Because ­ consumer preferences for raw, live shellfish persist, biological approaches for reducing the load of shellfish-associated Vibrio spp. and promoting microbiological safety of live product (such as the use of probiotics, vibriophages, and inhibitors of quorum sensing [QS] and attachment) are also under study, as recently reviewed by Teplitski et al. (228).

V. CHOLERAE V. cholerae O1 is the causative agent of cholera, one of the few foodborne diseases with epidemic and pandemic potential. V. cholerae is a well-defined species on the basis of biochemical tests and DNA homology studies, but as reviewed elsewhere (109), this species is not homogeneous with regard to pathogenic potential. Specifically, important distinctions within the species are made on the basis of production of cholera enterotoxin (cholera toxin or Ctx), serogroup, and potential for epidemic spread. Until recently, the public health distinction was simple, that is, V. cholerae strains of the O1 serogroup that produced Ctx were associated with epidemic cholera and all other members of the species were either nonpathogenic or only occasional pathogens. However, with the emergence of cholera due to strains of the O139 serogroup (see below), such previous distinctions are no longer valid. There are two serogroups, O1 and O139, that have been associated with epidemic disease, but there are also strains of these serogroups that do not produce Ctx, do not cause cholera, and are not involved in human disease. Conversely, there are occasional strains of serogroups other than O1 or O139 that are clearly pathogenic, either by the production of Ctx or other virulence factors (see below); however, none of these other serogroups has caused large epidemics or pandemics. Therefore, in assessing the public health significance of an isolate of V. chol­ erae, there are two critical properties to be determined beyond the biochemical identification of the species V. cholerae. The first of these properties is production of cholera toxin, which is the toxin that is responsible for severe, cholera-like disease in epidemic and sporadic forms. The second property is possession of the O1 or O139 antigen, which, since the actual determinant of epidemic/pandemic potential is not known, is at least a marker of such potential. The finding that the O antigen gene cluster is mobile within the species V. cholerae

Foodborne Pathogenic Bacteria (36) suggests caution in using O serotyping as a reliable marker for diagnostic purposes in V. cholerae (35). To overcome this limitation, simple but reliable PCR-based methods targeting constant, nontransferable genomic regions unique to the current pandemic clade, defined as the seventh pandemic group (36), should be developed. Whole-genome sequences will likely provide the best foundation for such a task (35). The subject of cholera has been reviewed elsewhere, and readers are referred to these reviews for primary references, particularly from the older literature (109, 170). Unless otherwise stated, all information involving this species refers to V. cholerae O1 or O139 strains capable of causing cholera.

Classification V. cholerae O1

V. cholerae organisms of the O1 serogroup that produce a Ctx have long been associated with epidemic and pandemic cholera. Strains isolated from environmental samples in nonepidemic areas are usually Ctx negative and are considered to be nonpathogenic, based on volunteer studies (see “Reservoir” below). However, Ctx-negative V. cholerae O1 strains have been isolated from occasional cases of diarrhea or extraintestinal infections. This serogroup can be further subdivided into serotypes of the O1 serogroup called Ogawa and Inaba. V. cholerae O1 can also be divided into two biotypes, classical and El Tor, that differ in several phenotypic and genotypic characteristics, including pathogenic potential, survivability, and infection patterns in humans. For example, El Tor strains are associated more frequently with asymptomatic infections and fewer fatalities, whereas classical biotype isolates cause more severe clinical manifestation (203). Although cholera toxin is conserved among classical and El Tor strains, subtle sequence differences in CtxB exist between the two biotypes and serve as the basis for CtxB epityping and ctxB genotyping (13, 73). Both genetic and phenotypic diversity have arisen among strains of V. cholerae El Tor circulating in Asia and Africa. This diversity is reflected by the acquisition, loss, or alteration of mobile genetic elements, including CTX phage, which bears the genes encoding cholera toxin, genomic islands, and SXT family integrative and conjugative elements (encoding resistance to sulfamethoxazole, trimethoprim, and streptomycin) (33, 35). Single-nucleotide variations and insertions and deletions have also been detected in the core V. cholerae genome. V. cholerae O1 hybrids, which cannot be biotyped based on phenotypic tests and can produce cholera toxin of either biotype, and altered El

16.  Vibrio Species Tor variants, which produce cholera toxin of the classical biotype but can be biotyped as El Tor by conventional phenotypic assays, have been isolated repeatedly in Bangladesh and Mozambique (77, 197). These new variants have subsequently replaced the prototype seventh-pandemic V. cholerae O1 El Tor strains in Asia and Africa, with respect to frequency of isolation from clinical cases of cholera.

V. cholerae Non-O1/Non-O139

Nearly 200 serogroups of V. cholerae have been described. In recent years, until the emergence of the O139 serogroup, all isolates that were identified as V. cholerae on the basis of biochemical tests but that were negative for the O1 serogroup were referred to as “nonO1 V. cholerae.” In earlier years, non-O1 V. cholerae was referred to as NCV (noncholera vibrios) or NAG (nonagglutinable) vibrios. The basis for serotyping in V. cholerae is the lipopolysaccharide (LPS) somatic antigen; H antigens are not useful in serotyping. Most of these strains do not produce Ctx and are not associated with epidemic diarrhea (155, 157, 203). These strains are occasionally isolated from cases of diarrhea (usually associated with consumption of shellfish) and have been isolated from a variety of extraintestinal infections. These strains are regularly found in estuarine environments, and infections due to these strains are commonly of environmental origin. While most of these strains do not produce Ctx, some strains may produce other toxins (see below); however, for many strains of V. cholerae non-O1/non-O139 isolated from cases of gastroenteritis, the pathogenic mechanisms are unknown. Strains of the O141 serogroup have been isolated from sporadic cases of severe diarrhea and produce Ctx and the toxin coregulated pilus (TCP) colonization factor typical of O1 and O139 strains (43) but also produce a type III secretion system (TTSS) (see below).

V. cholerae O139 Bengal

The simple distinction between V. cholerae O1 and V. cholerae non-O1 was rendered obsolete in early 1993, when the first reports of a new epidemic of ­ severe, cholera-like disease emerging from eastern India and Bangladesh appeared (6). Further investigations re­vealed that this bacterium did not belong to the O serogroups previously described for V. cholerae but to a new serogroup, which was given the designation O139 and the synonym “Bengal,” in recognition of the origin of this strain. In important virulence characteristics, such as clinical manifestations and Ctx/TCP sequences, V. chol­ erae O139 is indistinguishable from typical El Tor V. cholerae O1 strains (6). However, this bacterium does

407 not produce the O1 LPS due to deletion of 22 kb of DNA necessary for production of the O1 antigen and insertion of a 35-kb region encoding the O139 antigen (203). Furthermore, and like many strains of non-O1 V. cholerae and unlike V. cholerae O1, it produces a polysaccharide capsule (see below). When it initially appeared, the O139 serogroup replaced the O1 serogroup in some parts of Southeast Asia and was feared to represent a new pandemic (the 8th pandemic) of cholera. However, few cases of O139 were reported beyond Southeast Asia, and O1 cases became dominant again in this part of the world (7).

Isolation and Identification

TCBS is the most commonly employed isolation media for V. cholerae, although this species can grow as ­lactosenegative colonies on MacConkey agar. Enrichment of specimens usually employs the nonselective APW, with plating after 6 to 8 hours (to prevent overgrowth by other species) or overnight incubation (111). Suspected V. cholerae­ isolates can be transferred from primary isolation plates to a standard series of biochemical media used for identification of Enterobacteriaceae and Vibrionaceae. Both conventional tube tests and commercially available enteric identification systems are suitable for identifying this species, although the accuracy of commercial identification kits can range from 50 to 97% (3). Several key characteristics for distinguishing V. cholerae from other species are given in Table 16.1. The key confirmation for identification of V. chol­ erae O1 is agglutination in polyvalent antisera raised against the O1 antigen. Polyvalent antiserum for V. cholerae O1 and O139 is commercially available and can be used in slide agglutination or coagglutination tests. Monoclonal antibody-based, coagglutination tests suitable for testing isolated colonies or diarrheal stool samples for O1 or O139 are also available commercially. Oxidase-positive bacteria (determined using colonies grown on nonselective media) that agglutinate in O1 or O139 antisera can be reported presumptively as V. cholerae O1 or O139 and then forwarded to a public health reference laboratory for confirmation. Antisera for serogroups other than O1 or O139 are not commercially available.

Molecular Techniques

Nucleic acid probes are not routinely employed for the identification of V. cholerae due to the ease of identifying this species by conventional methods. Where DNA probes and PCR techniques have been extremely useful is in distinguishing those strains of V. cholerae that contain genes (ctx) encoding cholera toxin from

408 those that do not contain these genes. This distinction is particularly important in examining environmental ­isolates of V. cholerae since most of these strains lack ctx sequences. A number of DNA fragment probes and synthetic oligonucleotide probes have been developed to detect ctx sequences in isolated colonies, although PCR techniques are now more commonly employed (78, 111, 122). PCR has been used to detect toxigenic V. cholerae O1 in food samples, including fruit, vegetables, and shellfish ­specimens (67). Subtyping of V. cholerae strains using a variety of techniques such as restriction fragment length polymorphism (RFLP) analysis, PFGE, ribotyping, and MLST has yielded significant insights into the molecular epidemiology of V. cholerae. For example, RFLP analysis revealed that a toxigenic O1 strain isolated from a cholera patient in Maryland was identical to isolates from Louisiana and Texas, which concurred with epidemiologic investigations showing that the crabs eaten by the Maryland patient were harvested along the Texas coast. PFGE has been used to differentiate strains of V. cholerae in many studies all over the world, and a study by the CDC revealed that PFGE could distinguish strains that were identical when examined by multi­ locus enzyme electrophoresis or ribotyping (25). MLST using only three housekeeping genes (gyrB, pgm, and recA) has discriminating ability superior to that of PFGE (117). Another approach is to use a single multiplex PCR assay to simultaneously amplify 95 “diagnostic” regions from V. cholerae and other Vibrio species (i.e., species/serogroup specific genes, toxin genes, etc.) that are hybridized to a microarray containing these genes (241). This approach allows rapid and definitive inter- and intraspecies ­discrimination that can be helpful in epidemiologic, environmental, and health risk assessment surveillance. Strain heterogeneity is most comprehensively captured by sequencing genomic DNA. In particular, third-generation single-molecule realtime sequencing is much faster and more productive than previously developed methods (209). Using this approach, it was recently reported that the 2010 Haitian epidemic is probably the result of the introduction of a V. cholerae strain from a distant geographic source (33).

Reservoirs Environment

V. cholerae is part of the normal, free-living (autochthonous) bacterial flora in estuarine areas. Non-O1/nonO139 strains are much more commonly isolated from the environment than are O1 strains, even in epidemic settings in which fecal contamination of the environment might be expected. Outside epidemic areas (and

Foodborne Pathogenic Bacteria away from areas that may have been contaminated by cholera patients), O1 environmental isolates are almost always Ctx negative. However, it is clear that Ctxproducing V. cholerae O1 can persist in the environment in the absence of known human disease and is most likely explained by the existence of environmental reservoirs, defined as locations out of the human body favoring bacterial persistence and replication in the environment, and pathogen transmission to susceptible hosts. V. cholerae strains are capable of colonizing the surfaces of zooplankton such as copepods with 104 to 105 V. cholerae cells attached to a single copepod (40, 129). Other aquatic biota, such as water hyacinths, filamentous green algae, and insects are also colonized by V. cholerae (22, 240). V. cholerae produces a chitinase and is able to bind to chitin, which is the principal component of crustacean shells; the bacterium can grow in media with chitin as the sole carbon source. Recent studies have revealed that interaction with chitin has a profound influence on the lifestyle of the bacterium, as it provides the microorganism with a number of advantages, including food availability, adaptation to environmental nutrient gradients, tolerance to stress, and protection from predators (192, 240). Chitin also induces nature competence in V. cholerae, hence suggesting that this species can acquire new genetic material by transformation during growth on chitin (148). The environmental reservoir is a major part of the V. cholerae life cycle, whereby bacteria discharged from the human host reside in association with aquatic life forms until they are once again ingested by humans via contaminated water or food (198). Persistence of V. cholerae within the environment may be facilitated by its ability to assume survival forms, including a “viable but nonculturable state,” biofilms, and a rugose survival form. The viable but nonculturable state is a dormant state in which V. cholerae is still viable but not culturable in conventional laboratory media (38, 172, 174, 176). In this dormant state, the cells are reduced in size and become ovoid. The continued viability of the nonculturable V. cholerae can be assessed by various direct viable count procedures (172). The viable but nonculturable state can be induced in the laboratory by incubating a culture of V. cholerae in phosphate-buffered saline at 4°C for several days. Although these cells are not culturable with nonselective enrichment broth or plates, nonculturable V. cholerae O1 cells injected into ligated rabbit ileal loops or ingested by volunteers have yielded culturable V. cholerae O1 in intestinal contents or stool specimens (39). A microarray study found that viable but nonculturable V. cholerae still produced transcripts of virulence factor genes (241). V. cholerae can also form biofilms that can enhance survival in the environment (246). A variety of factors,

16.  Vibrio Species including flagella, the type IV pilus mannose-sensitive hemagglutinin (MSHA), and QS (see below), are involved in the formation of biofilms. Biofilms allow the bacterium to persist in association with biotic and abiotic surfaces and help to prevent predation by grazing protozoa (143). Furthermore, V. cholerae cells present in biofilms are much more resistant to killing by acid shock than are planktonic cells that are not in biofilms (260). This increased acid resistance could help the bacterium survive stomach acidity after ingestion by the next human host. It has been also shown that cholera patients shed V. cholerae in complex, in vivo-formed biofilm-like aggregates (64). Hence, biofilm formation by V. cholerae likely plays a major role in its ecology and in the epidemiology of cholera (98, 163). A phenomenon that is closely related to biofilms is the formation of rugose or “wrinkled” colonial morphology due to the production of an exopolysaccharide (109). In this state, the cells are protected against adverse environmental conditions. Notably, rugose variants survive in the presence of chlorine and other disinfectants and have enhanced capacity to form biofilms. Such rugose forms are nonetheless still capable of causing diarrhea in volunteers. The rugose polysaccharide produced by O1 El Tor strains is called VPSETr, and genes required for synthesis are clustered in a 30-kb region called the vps locus (258). Microarray analysis of rugose and smooth variants of the same strain implicated 124 differentially regulated genes in this process encoding regulators, surface properties, and motility (257).

Humans and Animals

Long-term carriage of V. cholerae in humans is extremely rare and is not considered to be significant in transmission of disease. However, short-term carriage of V. cholerae by humans is quite important in transmission of disease. Persons with acute cholera excrete 107 to 108 CFU of V. cholerae per gram of stool; for patients who have 5 to 10 liters of diarrheal stool, the total output of V. cholerae can be in the range of 1011 to 1013 CFU. Even after cessation of symptoms, patients who have not been treated with antibiotics may continue to excrete vibrios for 1 to 2 weeks. Furthermore, a high percentage of persons infected with V. cholerae in endemic areas have inapparent illness and can still excrete the bacteria, although excretion generally lasts for less than a week. Asymptomatic carriers are most commonly identified among household members of persons with acute illness: in various studies, the rate of asymptomatic carriage in this group has ranged from 4% to almost 22%. There have also been studies indicating that V. cholerae O1 can be sporadically carried by house-

409 hold animals, including cows, dogs, and chickens, but no animal species consistently carries the bacterium. In the 1970s it was widely accepted that asymptomatic and convalescent human (and possibly, animal) carriers were the primary reservoir for cholera. With the recognition that V. cholerae can live and multiply in the environment, much greater attention has been given to ­identification and characterization of environmental reservoirs (38, 240). Nonetheless, Ctx-producing V. cholerae O1 (i.e., diseasecausing strains) continue to be isolated almost exclusively from areas that have been contaminated by human feces or sewage from persons or groups of persons known to have had cholera. Similarly, in areas of endemicity such as Lima, Peru, rates of isolation from the environment correlate primarily with the degree of sewage contamination. However, even areas where cholera is not endemic may be contaminated by ballast water from ships originating in areas of endemicity, such as was seen with oysters in Mobile Bay, AL, that contained toxigenic V. cholerae O1 strains with PFGE patterns that were identical to those of toxigenic V. cholerae strains isolated from ballast water from ships docked at Gulf of Mexico ports (160). A dynamic relationship between human and environmental sources of the bacterium is apparent, with carriage and amplification by human populations playing a critical role in the epidemic spread of Ctx-producing V. cholerae. Environmental bacteriophage capable of lysing V. cholerae O1 or O139 are also involved in this relationship, and a recent 3-year study in Bangladesh revealed that the presence of cholera phages was inversely correlated with the occurrence of viable V. cholerae in the environment and the number of cholera cases in the local community (63). Nelson et al. (163) determined that toxigenic strains of V. cholerae persist alongside nontoxigenic strains in the aquatic environment, aided by biofilm formation on biological surfaces. Susceptible hosts become infectious after consuming V. cholerae from an environmental source, sometimes accompanied by lytic choleraphages. Infected individuals recover through the actions of their immune systems and possibly those of lytic bacteriophages or succumb to the infection. Lytic phages and V. cholerae in a transient hyperinfectious state are shed in varying concentrations and can either rapidly passage to the next host or persist in the environment as culturable cells with unknown infectivity or transform into a viable but not culturable state.

Foodborne Outbreaks

The critical role of water in transmission of cholera has been recognized for more than a century, ever since the London physician and epidemiologist John Snow revealed in 1854 that illness was associated with

410 c­ onsumption of water from a water system that drew its water from the Thames at a point below major sewage inflows. In developing countries, ingestion of contaminated water and food are probably the major vehicles for transmission of cholera, whereas in developed countries, foodborne transmission is more important (202). Such distinctions are often difficult to make because contaminated water is frequently used in food preparation. For example, rice prepared with water contaminated with V. cholerae O1 has been implicated in outbreaks from Bangladesh as well as from the U.S. Gulf Coast. Fruit juices diluted with contaminated water and vegetables irrigated with untreated sewage have been associated with disease in South America. Seafood may acquire the bacterium from environmental sources and may serve as a vehicle in both endemic and epidemic disease, particularly if it is uncooked or only partially cooked. The role of food in transmitting V. cholerae O1 has been reviewed elsewhere (193, 222). The spectrum of food items implicated in transmission of cholera includes crabs, shrimp, raw fish, mussels, cockles, squid, oysters, clams, rice, raw pork, millet gruel, cooked rice, street vendor food, frozen coconut milk, and raw vegetables and fruit. One shared characteristic of the implicated foods is their neutral or nearly neutral pH. Hence, in investigating a suspected foodborne outbreak with many possible vehicles, one can eliminate the foods with an acid pH and concentrate on neutral or alkaline foods. This predilection for neutral foods was revealed in an epidemiologic study in West Africa, where boiled rice is commonly prepared in the morning, held unrefrigerated, and eaten with sauce at the midday and evening meals. In a case-control study of illness, tomato sauces with a pH of 4.5 to 5.0 were protective against illness, whereas less acidic sauces (pH 6.0 to 7.0) prepared from ground peanuts were associated with illness due to V. cholerae O1 (154). Survival and growth of V. cholerae O1 in foods are also enhanced by low temperatures, high organic content, high moisture, and absence of competing flora. Survival is increased when foods are cooked before contamination; cooking eliminates competing microbes and has also been suggested to destroy some heat-labile growth inhibitors and produce denatured proteins that the bacterium uses for growth. As noted below, food buffers V. cholerae O1 against killing by gastric acid. While many different food items can provide this buffering capacity, the protection provided by chitin is noteworthy because crustaceans are frequent vehicles of disease (192). In dilute hydrochloric acid solutions of approximately the same pH as human gastric acid, survival of V. cholerae absorbed to chitin was enhanced compared to its survival in the absence of chitin.

Foodborne Pathogenic Bacteria In the United States, both domestically acquired and imported cases of cholera occur (101, 231). For domestic cases, crabs, shrimp, and oysters have been the most frequently implicated vehicles, although the largest single outbreak (16 cases) was due to ingestion of contaminated rice (154). Although most domestic cases occur in states bordering the Gulf Coast, seafood shipped from this area has caused disease in both Maryland and Colorado (154). The risk of imported cholera has greatly increased since the establishment of endemic cholera in South America in 1991. The largest such outbreak (75 cases) involved crab salad served on an airplane flying from Peru to California. A smaller outbreak of 8 cases occurred in New Jersey due to crabs purchased in Ecuador and carried to the United States in an individual’s luggage. Importation of cholera from Asia can also occur, even in commercially imported food. A small outbreak of four cases in Maryland was attributed to frozen coconut milk imported from Thailand that was subsequently used for a rice pudding topping. Most cases of gastroenteritis caused by V. cholerae of serogroups other than O1 or O139 have been linked to the consumption of raw oysters (155). Both disease incidence and isolation rate of non-O1/O139 serogroups from oysters are highest in the summer. Such strains were isolated from up to 14% of freshly harvested oysters in one study conducted by the FDA. Outside the United States, outbreaks have also been linked to consumption of contaminated potatoes, chopped eggs, pre-prepared gelatin, vegetables, and meat samples (93, 155). As with V. cholerae O1, survival of non-O1/non-O139 V. chol­ erae is enhanced in foods of alkaline pH.

Characteristics of Disease

The explosive, potentially fatal dehydrating diarrhea that is characteristic of cholera is actually seen in only a minority of persons infected with Ctx-producing V. cholerae O1/O139. Most infections with V. cholerae O1 are mild or even asymptomatic. It has been estimated that 11% of patients with classical infections develop severe disease, compared with 2% of those with El Tor infections. An additional 5% of El Tor infections and 15% of classical infections result in moderate illness (109). Symptoms of persons infected with V. cholerae O139 Bengal appear to be virtually identical to those of persons infected with O1 strains. The incubation period of cholera can range from several hours to 5 days and is dependent in part on inoculum size. The onset of illness may be sudden, with profuse, watery diarrhea, or there can be premonitory symptoms such as anorexia, abdominal discomfort, and simple diarrhea. Initially, the stool is brown with fecal

16.  Vibrio Species matter, but soon the diarrhea develops a pale gray color with an inoffensive, slightly fishy odor. Mucus in the stool imparts the characteristic “rice water” appearance. Vomiting is often present, occurring a few hours after the onset of diarrhea. In its most severe form, termed cholera gravis, the rate of diarrhea may quickly reach 500 to 1,000 ml/h, leading rapidly to tachycardia, hypotension, and vascular collapse due to dehydration. Peripheral pulses may be absent, and blood pressure may be unobtainable. Skin turgor is poor, giving the skin a doughy consistency; the eyes are sunken; and hands and feet become wrinkled, as after long immersion in water (“washerwoman’s hands”). Such severe dehydration can lead to death within hours of the onset of symptoms unless fluids and electrolytes are rapidly replaced. While cholera gravis is a striking clinical entity, milder illnesses are not readily differentiated from other causes of gastroenteritis in areas of cholera endemicity. Gastroenteritis associated with V. cholerae nonO1/non-O139 is generally of mild to moderate severity, although severe, cholera-like illness has also been seen occasionally. Besides nonbloody and occasionally bloody diarrhea, symptoms can also include abdominal cramps, and fever with nausea and vomiting occurring in a minority of patients (155). Non-O1/O139 V. cholerae is also frequently isolated from extraintestinal infections such as septicemia, wound infections, and ear infections; these infections usually involve exposure to fresh or brackish water (157). The case-fatality rate of extraintestinal infections can exceed 50%, and individuals with preexisting liver disease are particularly at risk.

Infectious Dose and Susceptible Population

In healthy North American volunteers, doses of 1011 CFU of V. cholerae were required to consistently cause diarrhea when the inoculum was given in buffered saline (pH 7.2). When stomach acidity was neutralized with 2 grams of sodium bicarbonate immediately prior to administration of the inoculum, attack rates of 90% were seen with an inoculum of 106 CFU. Food has a buffering capacity comparable to that seen with sodium bicarbonate. Ingestion of 106 vibrios with food such as fish and rice resulted in the same high attack rate (100%) as when this inoculum was administered with buffer (127). Further studies revealed that most volunteers who received as few as 103 to 104 vibrios with buffer developed diarrhea, although lower inocula correlated with a longer incubation period and diminished severity. The incubation time in volunteers between ingestion of vibrios and onset of diarrhea ranged from 8 to 96 hours. Under

411 natural field circumstances, the inoculum size needed to cause cholera may be even lower, because attack rates are lower than in volunteer studies, and many patients have low gastric acid production (202). The volunteer data on the effect of buffer on infectious dose are consistent with epidemiologic data indicating that people who are achlorhydric because of surgery, medication (e.g., antacids), or other reasons are at increased risk for cholera. Individuals of blood group O are at increased risk of more severe cholera, which has been shown for natural infection as well as for experimental infection; the mechanism of this increased susceptibility is unknown. In addition to these factors, additional host factors, as yet poorly defined, play a role in disease susceptibility to cholera. The effect of host factors is illustrated by a study in which an identical inoculum caused 44 liters of diarrhea in one volunteer and caused little or no illness in other individuals (127). Strains of non-O1/non-O139 V. cholerae have also been studied in volunteers. Of three strains fed to volunteers that did not produce Ctx, only one strain caused diarrhea. This strain produced a heat-stable enterotoxin (ST)-like toxin (see below) and caused diarrhea in 6 of 8 volunteers at doses of 106 to 109 vibrios after neutralization of stomach acid with sodium bicarbonate (155). The severity of disease was generally mild, but in one volunteer, diarrheal stool volume exceeded 5 liters.

Virulence Mechanisms

Infection due to V. cholerae O1/O139 begins with the ingestion of food or water contaminated with the bacterium. After passage through the acid barrier of the stomach, vibrios colonize the epithelium of the small intestine by means of one or more adherence factors. Invasion into epithelial cells or the lamina propria does not occur. Production of cholera enterotoxin (and possibly other toxins) disrupts ion transport by intestinal epithelial cells. The subsequent loss of water and electrolytes leads to the severe diarrhea characteristic of cholera.

Ctx

Volunteer studies revealed that the massive, dehydrating diarrhea characteristic of cholera is induced by cholera enterotoxin, also referred to as cholera toxin (Ctx). Cholera toxin is among the best characterized of bacterial toxins and has been extensively reviewed (47, 205, 237). Ctx is a prototypic A-B subunit toxin in which the B subunit serves to bind the holotoxin to the eukaryotic cell receptor and the A subunit possesses specific enzymatic activities, ADP-ribosyltransferase and NADglycohydrolase, that act intracellularly. Ctx consists of

412 five identical B subunits and a single A subunit; neither of the subunits individually has significant secretogenic activity in animal or intact cell systems. The mature B subunit contains 103 amino acids with a subunit weight of 11.6 kDa. The mature A subunit has a mass of 27.2 kDa and is proteolytically cleaved to yield two polypeptide chains, a 195-residue A1 peptide of 21.8 kDa and a 45-residue A2 peptide of 5.4 kDa, which are linked by a disulfide bond. The ctxA and ctxB genes encoding the A and B subunits reside on a filamentous bacteriophage (CTXM) that is capable of transducing ctx genes into nontoxigenic strains (145). The receptor for Ctx is the ganglioside GM1, and binding of toxin to epithelial cells is enhanced by a neuraminidase produced by V. cholerae. This 83-kDa enzyme catalyzes the conversion of higher-order ­gangliosides to GM1, thereby enhancing the binding of Ctx and leading to greater fluid secretion. Binding of the CtxB pentamer to GM1 in lipid rafts of the plasma membrane triggers toxin internalization via endocytic vesicles. The internalized holotoxin traffics through the trans-Golgi network and to the endoplasmic reticulum (47, 205). Ultimately, the A1 peptide is translocated into the cell cytosol, where it irreversibly ADP-ribosylates the a subunit of the Gs protein. G proteins link many cell-surface receptors to effector proteins at the plasma membrane, thereby regulating an extensive set of metabolic pathways. Gs is involved in regulation of the adenylate cyclase complex, which mediates the transformation of ATP to cyclic AMP (cAMP), a crucial intracellular messenger for a variety of cellular pathways. The Ctx A1 peptide catalyzes the transfer of the ADP-ribose moiety of NAD to a specific arginine residue in the Gs a protein, resulting in the activation of adenylate cyclase and subsequent increases in intracellular levels of cAMP. cAMP activates a cAMP-dependent protein kinase (PKA), leading to protein phosphorylation of the major Cl− channel in epithelial cells, the cystic fibrosis transmembrane conductance regulator (CFTR) (Fig. 16.1). Increased Cl− secretion by intestinal crypt cells through CFTR and decreased NaCl-coupled absorption by villus cells (by mechanisms that are not well understood) result in a transepithelial osmotic gradient that causes water flow into the lumen of the intestine. The massive volume of water overwhelms the absorptive capacity of the intestine, resulting in diarrhea. The activation of adenylate cyclase leading to increased cAMP and activation of CFTR via protein kinase A is the “classic” mode of action of Ctx. However, there are additional effects of Ctx that could contribute to the secretory effects of cholera toxin, including production of prostaglandins and stimulation of the enteric nervous system (47,

Foodborne Pathogenic Bacteria 211). Ctx activates the platelet-­activating factor, leading to activation of phospholipase A2 and subsequent accumulation of arachidonic acid and prostaglandins. A substantial portion of the secretogenic potential of Ctx can be blocked by addition of platelet-activating factor receptor antagonists and phospholipase A2 inhibitors. Consistent with these observations, cholera patients in the active secretory disease stage have elevated jejunal concentrations of PGE2 compared to patients in the convalescent stage. The enteric nervous system (ENS) plays an important role in normal intestinal secretion and absorption and has been implicated in the diarrheal response to V. cholerae. Ctx can stimulate release of vasoactive intestinal peptide and serotonin (5-hydroxytryptamine or 5-HT), factors that can induce secretion via the ENS. A variety of studies using receptor antagonists, ganglionic or neurotransmitter blockers as well as direct measurements of increased levels of serotonin (5-HT) and vasoactive intestinal peptide in cholera patients supports the role of the ENS in cholera.

Other Toxins Produced by V. cholerae

When the first recombinant V. cholerae vaccine strains specifically deleted of genes encoding Ctx were tested in volunteers, it was somewhat surprising that mild-tomoderate diarrhea was still seen in ca. 50% of volunteers. The volume of diarrhea was not the severe, dehydrating diarrhea seen with wild-type strains, which can exceed 40 liters, but a much milder diarrhea that ranged from 0.3 to 2.1 liters. In addition, some volunteers also experienced abdominal cramps, anorexia, and low-grade fever when fed Dctx V. cholerae strains. These results prompted a search for additional toxins produced by V. cholerae, and it is now known that V. cholerae produces a variety of extracellular products that have deleterious effects on eukaryotic cells (69). The Zonula occludens toxin (Zot) increases the permeability of the small intestinal mucosa by affecting the structure of the intercellular tight junction, or zonula occludens (65, 210). Accessory cholera enterotoxin (Ace) causes fluid accumulation in rabbit ligated ileal loops and increases potential difference in intestinal tissue mounted in Ussing chambers (233). The ace and zot genes are located immediately upstream of the ctx genes and are also believed to be components of the CTXM filamentous phage-encoding cholera toxin (145). The soluble hemagglutinin/protease (HA/P) is a zinc-dependent metalloprotease that is capable of nicking and activating the A subunit of Ctx as well as cleaving mucin, fibronectin, and lactoferrin. HA/P can perturb the barrier function of epithelial cells by digesting occludin and rearranging the distribution of ZO-1 in tight junctions (256). Hemolysin (also called cytolysin or El Tor hemolysin) consists of two major toxin groups, namely, V. cholerae O1 (VCC1)

16.  Vibrio Species

Figure 16.1  Classic model of CT mode of action involving cAMP. More recent evidence indicates that prostaglandins and the enteric nervous system are also involved in the response to CT (see the text for details). (A) Adenylate cyclase, located in the basolateral membrane of intestinal epithelial cells, is regulated by G proteins. CT binds via the B subunit pentamer (shown as open circles with the A subunit as the inverted solid triangle) to the GM1 ganglioside receptor inserted into the lipid bilayer. (B) The toxin enters the cell via endosomes, and the A1 peptide ADP-ribosylates Gsa located in the basolateral membrane. (C) Increased cAMP activates protein kinase A, leading to protein phosphorylation. In crypt cells, the protein phosphorylation leads to increased Cl− secretion; in villus cells, it leads to decreased NaCl absorption. Adapted from reference 76. doi:10.1128/9781555818463.ch16f1

413

414 and V. cholerae non-O1 (VCC2), and is produced by all strains (46). The toxin, by forming anion channels on the apical membrane of enterocytes, triggers an outward trans­ cellular flux of chloride. Such ion movement, associated with the outward movement of Na+ and water, might be responsible for the diarrhea caused by the nontoxigenic strains of V. cholerae (46). The RtxA toxin is produced by O1 El Tor and O139 strains and is related to members of the RTX (repeats in toxin) toxin family, a group that includes the hemolysin of uropathogenic E. coli and adenylate cyclase of Bordetella pertussis. The huge RtxA toxin, with a predicted unprocessed size of ca. 500 kDa, causes depolymerization of actin stress fibers and covalent cross-linking of cellular actin and is responsible for the cytotoxicity of El Tor strains to Hep-2 cells (70, 121). The role of toxins other than cholera toxin in the pathogenesis of disease due to V. cholerae is largely unknown. These toxins clearly cannot cause cholera gravis because the diarrhea observed with Dctx strains presumably still producing these toxins is not the severe purging seen with wild-type V. cholerae strains. Dctx V. cholerae vaccine candidate strains lacking genes encoding Zot, Ace, hemolysin, or RtxA toxins still caused mild-to-moderate diarrhea as well as fever and abdominal cramps in volunteers. However, mutation of the hap gene encoding HA/P did reduce the reactogenicity observed with Dctx V. cholerae vaccine strains in volunteers (12). An El Tor strain deleted of genes encoding Ctx, hemolysin, HA/P and RtxA had reduced pulmonary inflammation when administered intranasally to mice (71), but the relevance to human intestinal infection is not clear. Toxins other than Ctx may contribute in part to the diarrhea and other symptoms seen with V. cholerae strains, perhaps serving as a secondary secretogenic mechanism when conditions for producing cholera toxin are not optimal.

Toxins of V. cholerae Non-O1 or Non-O139 Strains

Most non-O1/O139 strains do not contain ctx genes but usually possess genes encoding the hemolysin, RtxA, and HA/P. Some strains of V. cholerae non-O1/O139 (<10%) produce a 17-amino-acid heat-stable enterotoxin (designated NAG-ST for nonagglutinable Vibrio ST) that shares 50% sequence homology to the STa of enterotoxigenic E. coli. In a volunteer study, one subject who ingested a Ctx-negative V. cholerae non-O1 strain producing NAG-ST purged over 5 liters of diarrheal stool (155). The tdh gene encoding the thermostable direct hemolysin of V. parahaemolyticus (see below) has also been detected in some V. cholerae non-O1/nonO139 strains.

Foodborne Pathogenic Bacteria A recent genome sequence analysis of a clinical nonO1, non-O139 strain (O39 serogroup) that was particularly virulent in an animal model revealed the ­presence of a TTSS that is not present in O1/O139 strains (55, 56). This specialized secretion system mediates the translocation of toxins and other effector proteins from the bacterial cytoplasm into the mammalian cell and is an important virulence factor of several gram-negative pathogens, including Salmonella, Shigella, Yersinia, enteropathogenic E. coli, and Pseudomonas. The TTSS of V. cholerae was related to the TTSS2 system of V. para­ haemolyticus (see below) and was present in 6 of 12 V. cholerae non-O1/O139 serogroups tested, including all clinical isolates of the O141 serogroup.

Colonization Factors TCP

TCP is the best-characterized intestinal colonization factor of V. cholerae. The TCP pili consist of long filaments, 7 nm in diameter, that are laterally associated in bundles. The name of the pilus results from the fact that expression of the pilus is correlated with expression of cholera toxin (227). TCP belongs to the type IV family of pili and is the only colonization factor of V. cholerae whose importance in human disease has been proven. Volunteers ingesting V. cholerae strains specifically mutated in the tcpA gene did not experience diarrhea, and no vibrios were recovered from the stools of the volunteers. Direct binding of TCP to epithelial cells has not been observed, and the role of this factor may be to mediate interbacterial aggregation and thereby facilitate intestinal colonization. Synthesis of TCP is complex, and there are up to 15 open reading frames in the tcp gene cluster. The tcp gene cluster along with toxT (see below) are encoded on a 40-kb pathogenicity island called VPI (Vibrio pathogenicity island) that is present in all O1 and O139 clinical isolates and absent from nearly all environmental O1/O139 isolates and most clinical non-O1/O139 isolates (110).

Other Potential Colonization Factors

A number of other potential colonization factors have been described for V. cholerae, including a mannosefucose-resistant hemagglutinin, an MSHA, and several outer membrane proteins, but their role in human intestinal colonization remains unproven. The MSHA is a type IV pilus that has been proven in volunteer studies not to be necessary for human colonization but does appear to be involved in biofilm formation and adherence to zooplankton. A 53-kDa protein, called GbpA, was recently implicated in attachment to both zooplankton

16.  Vibrio Species and human epithelial cells by binding to a sugar present on both surfaces (115, 239).

Motility and Flagella

V. cholerae cells are motile by means of a single, polar, sheathed flagellum. Motility is an important virulence property, with nonmotile, fully enterotoxinogenic mutants being diminished in virulence. In several animal and in vitro models, motile V. cholerae cells rapidly enter the mucus gel overlying the intestinal epithelium and can be found in intervillus spaces within minutes to a few hours. The role of chemotaxis in V. cholerae infection is complex, and some chemotactic mutants have increased infectivity in infant mice. A model has been proposed wherein temporary downregulation of chemotaxis in wild-type V. cholerae enhances initial colonization specifically to the upper small intestine after that chemotaxis is upregulated to allow penetration through the mucus (22). Interestingly, it has been recently observed that V. cholerae flagellins can induce interkeukin-8 (IL-8) production in cultured intestinal epithelial cells (86).

LPS/Capsule

The LPS of V. cholerae O1 is the major protective antigen of this pathogen, and its importance in protective immunity greatly outweighs that of the cholera toxin. The importance of this antigen was seen in the initial outbreaks in India and Bangladesh when the O139 sero­ group caused widespread disease in individuals who were presumably immune to the O1 serogroup. V. chol­ erae O1 is unencapsulated, but strains of V. cholerae O139 produce a polysaccharide capsule, which has also been termed an O-antigen capsule. Both O1 LPS and O139 capsule have been implicated in intestinal colonization, probably by multiple mechanisms involving direct mucosal adherence and increased intestinal survival in the presence of bile and other factors (164).

Adherence Factors of Non-O1/Non-O139 V. cholerae

Intestinal colonization factors of non-O1/non-O139 V. cholerae strains are poorly characterized. Only a minority of clinical non-O1/non-O139 strains, such as those belonging to the O141 serogroup, possess tcp genes. A variety of fimbria hemagglutinins have been described for these strains, but their role in intestinal adherence is unclear. Most non-O1/non-O139 isolates produce a polysaccharide capsule that in addition to potentially aiding intestinal adherence could facilitate the septicemia that often occurs with these strains.

415

Regulation

Regulation of virulence gene expression in V. cholerae is highly complex, and many regulatory systems, both distinct and overlapping, have been described. The complex regulation is understandable in light of the dramatically different environments—the human intestine and the aquatic environment—that are involved in the V. cholerae life cycle. Expression of several virulence genes in V. cholerae O1 and O139 is coordinately regulated so that multiple genes respond in a similar fashion to environmental conditions. A wide variety of techniques have been used to study gene expression; microarray analysis has been applied to examine V. cholerae gene expression in rice water stools from cholera patients. Such an analysis suggests that vibrios in the stool are already modulating gene expression to prepare for entry into the aquatic environment and to create a hyperinfectious state that lowers the infectious dose for the next victim (150). The very active research in this area continues to yield the discovery of novel regulatory systems as well as detailed characterization of well-established regulatory systems in this species. For example, a recently described regulatory system of V. cholerae uses cyclic dinucleotide 3¢, 5¢-cyclic diguanylic acid (cdiGMP) as an intracellular signal to activate virulence genes and repress biofilm formation (26). The effect of iron on virulence and gene expression in this species has long been known, and recent microarray studies using a mutant in the iron-dependent negative regulator Fur have revealed the numerous genes involved in the response to this element (152). Space limitations do not permit a detailed discussion of regulation in V. cholerae, but two important systems that allow the pathogen to respond to its environment and regulate virulence gene expression will be briefly described.

The ToxR Regulon

The ToxR regulon controls expression of several critical virulence factors and has been the most extensively characterized regulatory system in V. cholerae (198, 120). The key regulator of Ctx and TCP expression in this system is ToxT, a member of the AraC/XylS family of transcription regulators that is encoded on the VPI. ToxT binds upstream of tcpA and ctxA to activate expression of these genes. Expression of ToxT itself is modulated by ToxR, a 32-kDa transmembrane protein that senses environmental conditions. The 19-kDa transmembrane protein ToxS associates with ToxR and helps to assemble or stabilize ToxR monomers into a dimeric form. ToxRS also directly regulates expression of the OmpT and OmpU outer membrane proteins ­independently of

Foodborne Pathogenic Bacteria

416 ToxT. The importance of ToxR in human disease was demonstrated in volunteer studies wherein a ToxR mutant V. cholerae O1 strain was fed to volunteers who subsequently suffered no diarrheal symptoms and did not shed the strain in their stools. Expression of toxT is also regulated by the TcpP and TcpH proteins that are encoded by genes located on the VPI upstream of toxT. TcpPH and ToxRS share sequence and functional similarities that permit the transmission of environmental signals to modulate expression of toxT. Expression of the tcpPH genes is in turn regulated by the AphAB proteins that are encoded outside the VPI. Microarray analysis of toxT, tcpPH, and toxRS mutants revealed 13, 27, and 60 genes, respectively, which were transcriptionally repressed in the mutants compared to the wild type (14). Hence, the major virulence factors of V. cholerae, cholera toxin and TCP, are regulated in a cascade fashion in that AphAB controls expression of TcpPH, which acts together with ToxRS to activate expression of ToxT, which then activates expression of Ctx and TCP.

QS

QS is a density-dependent regulatory process whereby small molecules termed autoinducers (AIs) act as signaling molecules to activate transcription of genes. AIs produced by bacteria are detected and responded to by other members of a population to coordinate gene expression, thereby allowing unicellular microbes to act as multicellular organisms. When the population density is high, the AI level is also high, leading to expression of genes that would not be expressed at low densities. The most common scenario for QS in bacterial pathogens is the activation of virulence genes at high population densities (108). V. cholerae also regulates expression of virulence genes by QS, but in this case, high densities lead to repression of virulence factors. V. cholerae has three distinct QS systems (126). The CAI-1 and AI-2 QS systems produce and respond to distinct AI molecules, but both end up in a common pathway involving the response regulator LuxO. Phosphorylation of LuxO in response to either CAI1 or AI-2 activates expression of four genes encoding small RNAs, ultimately leading to transcription of the master regulator HapR. HapR serves as a repressor of genes encoding Ctx, Tcp, and biofilm formation via the AphA regulator (see above) and activates expression of HA/P. At low population densities, little HapR is made, thereby allowing expression of Ctx, Tcp, and biofilms, but at high population densities, expression of HapR leads to repression of these factors. A third QS (VarS/ VarA-CsrA/BCD) has a distinct signaling pathway but

ultimately ties into the LuxO/HapR signaling cascade. A model has been proposed (82) wherein TCP and Ctx are expressed early in infection at low densities along with biofilm, which helps intestinal colonization. Later in the infection, when V. cholerae is at high densities, biofilm production ceases and HA/P protease production increases, thereby allowing the pathogen to exit the host and adapt to an environmental reservoir where expression of Ctx and TCP is not necessary. A similar QS system in which virulence factors are expressed at low densities and repressed at high densities is also seen with V. parahaemolyticus (see below).

V. PARAHAEMOLYTICUS Vibrio parahaemolyticus is the leading bacterial cause of intestinal infections due to ingestion of seafood, usually raw fish or shellfish. It has been implicated in numerous outbreaks of diarrheal disease throughout the world ever since the first description of its involvement in a major outbreak of food poisoning in 1950. V. para­ haemolyticus is also the Vibrio species most frequently isolated from clinical specimens in the United States (3, 44). CDC data indicate that vibrio infections, including those associated with V. parahaemolyticus, have increased since 2000, whereas the relative rates of infections from other major foodborne pathogens have decreased (27). An estimated 4,500 cases of V. para­ haemolyticus infection occur each year in the United States. However, the number of cases reported to CDC is much lower because surveillance is complicated by underreporting.

Classification

Biochemical characteristics that distinguish V. parahae­ molyticus from other Vibrio species are listed in Table 16.1. Commercial identification systems vary in their ability to correctly identify this and other Vibrio species, but addition of up to 2% NaCl to the test suspension diluent can improve the accuracy (3, 139). Conventional media for determining biochemical reactions should contain 2 to 3% NaCl (111). Although the core characteristics of V. parahaemolyticus are relatively consistent, as many previously described variant strains have been named as new species, strain variation can be seen with some traits, most notably with the production of urease by ca. 15% of strains. The ure gene is linked to the trh gene encoding a potential toxin of this bacterium (see below). In addition, some traits, such as H2S production, are dependent on the medium or assay method employed, and caution must be exercised in their determination.

16.  Vibrio Species V. parahaemolyticus is serotyped according to both its somatic O and capsular polysaccharide K antigens, and there are presently 13 O (LPS) antigens and over 71 K (acidic polysaccharide) antigens recognized (5, 111). Although many environmental and some clinical isolates are untypeable by the K antigen, most clinical strains can be classified to their O type. Until the description of serotype O3:K6, there appeared to be no correlation between serotype and virulence. This serotype, however, has recently been involved in epidemics of gastroenteritis throughout the world and is now referred to as the pandemic strain of V. parahaemolyticus. Antigenic (sero)variants belonging to serotypes O4:K68, O1:K25, and O1:KUT (untypeable) have also emerged and are largely indistinguishable from the O3:K6 pandemic strain with regard to ribotyping, PFGE patterns, and other subtyping methods (161). A study of 178 U.S. isolates from environmental, food, and clinical sources revealed 27 different serotypes and 28 ribotypes, with most clinical isolates from outside the Pacific Coast being serotype O3:K6 (49, 50). A special consideration in the classification of V. parahaemolyticus is the ability of certain strains to produce a hemolysin, termed the “thermostable direct hemolysin” (TDH) or “Kanagawa” hemolysin, which is correlated with virulence in this species (see “Virulence Mechanisms” below). The production of this hemolysin was originally established on a special blood agar called Wagatsuma agar, in which beta-hemolytic strains were termed “Kanagawa phenomenon-positive” (KP+) and nonhemolytic strains were KP−. However, this difficult-to-prepare medium (which requires freshly drawn human blood) has been supplanted by tdh gene probes and tdh-based PCR protocols. Detailed probe and PCR protocols are described in the online FDA Bacteriological Analytical Manual (111).

Reservoirs

V. parahaemolyticus occurs in estuarine waters throughout the world and is easily isolated from coastal waters of the entire United States, as well as from sediment, suspended particles, plankton, and a variety of fish and shellfish. The latter include at least 30 different species, among them clams, oysters, lobster, scallops, shrimp, and crab. In a study conducted by the FDA, 86% of the 635 seafood samples examined were positive for this species. Counts of V. parahaemolyticus have been reported to be as high as 1,300 CFU/g oyster tissue and 1,000 CFU/g crabmeat, although levels of 10 CFU/g are more typical for seafood products. Hackney et al. (81) determined, in a 3-year survey of 716 seafood samples obtained in North Carolina, that 46% of the samples

417 were V. parahaemolyticus positive. Notably, most V. parahaemolyticus-positive samples were unshucked oys­ ters (79% positive), unshucked clams (83% positive), unpeeled shrimp (60% positive), and live crabs (100% positive). Hackney et al. (81), as well as others, have observed no correlation with fecal coliforms or other indicator microorganisms, but V. parahaemolyticus cell numbers had a definite seasonal variation, with samples analyzed in January and February often free of V. parahaemolyticus. Others have also reported the ecology of V. parahaemolyticus to be heavily influenced by water temperature, salinity, and association with certain types of plankton, with highest populations occurring in warmer months and in waters of intermediate salinity (~23 ppt). The relationship between each of these parameters has been nicely determined by Kaneko and Colwell (107) in their pioneering study of the occurrence of this species in Chesapeake Bay waters. V. parahaemolyticus has been occasionally isolated from freshwater sites, but only at extremely low populations (<5 CFU/liter) and only during the warmest periods of the year. In a study of the impact of climate change on infectious diseases, increasing water temperatures in the Gulf of Alaska over the past 25 years have been implicated in causing a summertime outbreak of gastroenteritis among 62 cruise ship passengers associated with consumption of Alaskan oysters (144). Similarly, climate anomalies have been recently implicated in the expansion of the geographical and seasonal range of seafoodborne illnesses caused by V. parahaemolyticus in Peru (140). KP+ strains of V. parahaemolyticus are of prime importance in human disease, although occasional KP− strains have also been isolated from diarrheal stools. KP+ strains constitute a very small percentage (typically <1%) of the V. parahaemolyticus strains present in aquatic environments and seafoods. Hence, the simple isolation of this species from water or foodstuffs does not, in itself, indicate a health hazard (see “Virulence Mechanisms” below). The isolation of V. parahaemolyticus from foodstuffs generally involves a preenrichment step, the most common enrichment medium being APW. APW provides superior recovery of V. parahaemolyticus from a variety of fish and shellfish, even when the samples have been chilled or frozen (173). A new enrichment broth containing the bile salt sodium taurocholate (ST broth) has recently been described as superior to APW for detection and isolation of pathogenic V. parahaemolyticus from seafood (194). Of the many plating media suggested for this species, TCBS remains the most commonly employed (173). The

418 U.S. FDA Bacteriological Analytical Manual prescribes an overnight enrichment at 35 to 37°C in APW, from which a loopful is streaked onto TCBS agar to obtain isolated colonies of V. parahaemolyticus. TCBS is also commonly used for direct plating from stool samples (3). A chromogenic agar medium (CHROMagar Vibrio; CHROMagar Microbiology, Paris, France), on which V. parahaemolyticus produces purple colonies, has been described as yielding improved recovery of this species from seafood (84). Most-probable-number and membrane filtration procedures have been described for enumerating this species from seafood specimens (111). DNA probe and PCR procedures have been developed to detect the tlh gene (see below) to identify the isolate as V. parahaemolyticus and the tdh gene to identify pathogenic strains of this species (111).

Foodborne Outbreaks

Gastroenteritis with V. parahaemolyticus is almost exclusively associated with seafood that is consumed raw, inadequately cooked, or cooked but recontaminated. In Japan, V. parahaemolyticus is the major cause of foodborne illness, with as many as 70% of all bacterial food poisonings in that country attributable to this species in the 1960s (236). Outbreaks of V. parahaemolyticus gastroenteritis in the United States occurring between 1973 and 1987 have been summarized by Daniels et al. (44). Whereas most Japanese outbreaks involve fish, the U.S. outbreaks involved primarily crab, shrimp, lobster, and oysters. A very large outbreak of V. parahaemolyticus occurred during the summer of 1978 and affected 1,133 of 1,700 persons attending a dinner in Port Allen, LA (10). All stool isolates were KP+. The food implicated was boiled shrimp, which yielded positive cultures of V. parahaemolyticus. The raw shrimp had been purchased and shipped in standard wooden seafood boxes. They were boiled the morning of the dinner but were returned back to the same boxes in which they had been shipped. The warm shrimp were then transported 40 miles in an unrefrigerated truck to the site of the dinner and held an additional 7 to 8 hours until served that night. Another major outbreak (416 cases) was linked to consumption of raw oysters from Galveston Bay, TX (44). In their survey of four Gulf Coast states, Levine et al. (128) determined V. parahaemolyticus was the most common cause of gastroenteritis (37% of 71 cases) in that area. Similarly, Desenclos et al. (53) found V. para­ haemolyticus was the second leading cause (over 26%) of gastroenteritis cases in those persons who had consumed raw oysters in Florida. Recently, seafood consumption-related diarrhea became relevant in South America when the pandemic

Foodborne Pathogenic Bacteria strain of V. parahaemolyticus serotype O3:K6 reached a region in the south of Chile (Region de los Lagos) where approximately 80% of the country’s seafood is produced. Outbreaks in Region de los Lagos began in 2004 and reached a peak in 2005 with 3,725 clinical cases, all associated with the pandemic strain (72). After 2005, reported cases steadily decreased to a total of 477 cases in 2007 and 441 in 2009 (72). In 2005, pandemic V. parahaemolyticus was detected in Europe, with O3: K6 strains isolated from clinical cases in France, Spain, and Italy. These were determined to be associated with consumption of contaminated seafood (141, 184).

Characteristics of Disease

V. parahaemolyticus has a remarkable ability for rapid growth, with generation times as short as 8 to 9 minutes at 37°C reported. Even in seafoods, generation times of 12 to 18 minutes have been observed. As a result, V. parahaemolyticus has the ability to rapidly increase in cell numbers, both in vitro and in vivo, and this is evidenced in the characteristics of the disease it produces. Diarrhea, abdominal cramps, nausea, vomiting, headache, fever, and chills are symptoms of the disease. In the most severe cases, watery diarrhea is associated with mucus, blood, and tenesmus. The incubation period is 4 to 96 hours after ingestion of the bacteria, with a mean of 15 hours. The illness is usually mild or moderate, although some cases may require hospitalization. The median duration of the illness is 2.5 days. V. parahaemolyticus can also cause extraintestinal infections (177). In a study of U.S. infections from 1973 to 1998, 34% of all V. parahaemolyticus infections were wound infections and 5% were septicemia, compared to 59% for gastroenteritis (44). Of those patients with septicemia, 29% died, compared to 2% mortality associated with gastroenteritis.

Infectious Dose and Susceptible Population

Along with V. cholerae, V. parahaemolyticus is the only vibrio for which experimental evidence exists regarding the dosages required for initiation of gastroenteritis. Studies using human volunteers have revealed that ingestion of 2 × 105 to 3 × 107 CFU of KP+ cells led to the rapid development of gastrointestinal disease. Conversely, volunteers receiving as many as 1.6 × 1010 CFU of KP− cells did not exhibit signs of diarrheal disease (208). The KP+ and KP− strains used in the volunteer studies were not isogenic, and so other factors besides the TDH may have also contributed to the difference in infectious dose. It is not known whether the pandemic O3:K6 clone has a lower infectious dose than other sero­ types. Based on typical cell numbers of V. parahaemo­

16.  Vibrio Species lyticus present in fish and shellfish and the low prevalence of KP+ cells in these natural samples (see below), it would appear that a typical meal of raw shellfish would likely contain no greater than 104 CFU KP+ cells. Hence, for disease to result from consumption of contaminated food, it would appear that mishandling at temperatures allowing growth of the cells would be necessary (236). V. parahaemolyticus can increase by 3 log in shellstock oysters held at 26°C for 24 hours postharvest.

Virulence Mechanisms

The epidemiologic linkage between human virulence and the ability of V. parahaemolyticus isolates to produce the Kanagawa hemolysin (TDH) has long been established. In the initial study by Sakazaki et al. (204), 96.5% (2,655 of 2,720) of the strains isolated from human patients were KP+, whereas only 1% (7 of 650) of environmental isolates were KP+. Many subsequent studies have confirmed the low frequency of KP+ strains among environmental V. parahaemolyticus isolates (106), although some studies have shown higher frequencies (49). It is thought that a natural selection of KP+ strains occurs in the intestinal tract and that KP− strains survive better in the estuarine environment (106). The KP hemolysin is called thermostable direct hemolysin (TDH) because of its heat stability, with only partial inactivity observed after 30 min at 100°C, and its direct lysis of erythrocytes, which occurs without additional substituents. The lytic activity on erythrocytes from a wide variety of animals was the initial focus of investigations into TDH activity. TDH forms pores in erythrocytes and planar lipid bilayers (85), although it shares no significant homology to other pore-forming toxins. By constructing isogenic mutants in the tdh gene encoding TDH, Nishibuchi et al. (167) determined that only the TDH+ parent strain and not the tdh mutant was capable of inducing fluid accumulation in the rabbit ileal loop assay. Similar results were observed when culture supernatant fluids of these strains were tested on rabbit ileal tissue mounted in Ussing chambers, a more sensitive measure of secretory activity. In this assay, the ability of TDH to alter ion transport in the intestinal tract was observed at nanogram levels, with no histological changes. The purified TDH protein induces chloride ion secretion in rabbit and human intestinal tissue mounted in Ussing chambers (195, 224). However, unlike chloride ion secretion induced by cholera toxin, which acts through increased intracellular cAMP levels, TDH uses Ca2+ as an intracellular second messenger. This is the first bacterial enterotoxin for which the linkage between changes in intracellular calcium and secretory activity has been established.

419 KP+ isolates usually contain two nonidentical copies of the tdh gene, but many KP− or weakly positive isolates contained only one gene copy. The predicted protein products of the two gene copies (tdh1 and tdh2) differ in seven amino acid residues, although the proteins are immunologically indistinguishable. While both gene products contribute to the KP phenotype, >90% of the TDH protein can be attributed to a highlevel expression of the tdh2 gene (165). Expression of tdh2 but not tdh1 is under the control of a regulator similar to the ToxR of V. cholerae. The tdh genes are located on an 80-kb pathogenicity island on the small chromosome (138). During one outbreak of gastroenteritis, KP– isolates of V. parahaemolyticus produced a TDH-related hemolysin, termed TRH, but did not produce TDH. The trh gene shares 69% identity to the tdh2 gene, and its biological, immunological, and physicochemical characteristics are similar, but not identical, to those of TDH (168). A survey of 285 strains of V. parahaemolyticus revealed that not only were tdh-positive strains strongly associated with gastroenteritis but trh-positive strains were also thus associated. In one study, trh was detected in over 35% of 214 clinical strains, including 24% of those lacking the tdh gene (214). Purified TRH is less thermostable than TDH, being inactivated at 60°C for 10 min. Genes encoding a third unrelated hemolysin, called thermolabile hemolysin (tlh), are present in all strains of V. parahaemolyticus and are often used as a species-specific gene probe (111). Similar to TDH, purified TRH induces Cl− secretion in intestinal epithelial cells in a process involving increases in intracellular calcium (224). These results suggest that TRH may be an important virulence factor and possibly the cause of diarrhea in those patients from whom only KP− strains of V. parahaemolyticus are isolated from stools. The trh gene is colocated with the ure gene encoding urease on a 15.7-kb pathogenicity island with a lower G + C content (41%) than that of the total genome (46 to 47%) of this species (187). Hence, strains usually possess both trh and ure or lack both genes. As with tdh, environmental strains usually lack trh and ure (187). The function of urease in the pathogenesis of disease due to ure+ V. para­ haemolyticus strains is unknown, although it may assist in the survival of the pathogen during passage through the acid environment of the stomach. Putative virulenceassociated traits also include hemolysin, protease, motility, and biofilm formation (135). A surprising finding of the genome sequence analysis of a V. parahaemolyticus O3:K6 strain was the presence of two previously unknown TTSSs (138). Preliminary characterization of the functions of these

Foodborne Pathogenic Bacteria

420 two systems has revealed that the system encoded by the large chromosome, TTSS1, is involved in cytotoxicity to HeLa cells, with cell death being mediated by apoptosis (23, 183). TTSS1 is present in all V. para­ haemolyticus strains: clinical and environmental, KP+, and KP−. TTSS2 is present only in KP+ strains (138) and is encoded on the same pathogenicity island in the small chromosome that contains tdh. Mutation of TTSS2 genes resulted in diminished enterotoxigenicity in rabbit ileal loops (188). TTSS1 is regulated by quorum sensing (89), but similar to QS in V. cholerae, expression of the TTSS1 genes is repressed by QS at high densities. Since TTSS1 is present in all environmental and clinical strains, regulation by QS presumably plays a role in the environmental reservoir of V. parahaemolyticus. Comparison of the O3:K6 pandemic clone to nonpandemic clones of this species has resulted in the identification of a protein biomarker that is unique to the former (248). This protein is a histone-like DNAbinding protein that is encoded on a 16-kb region present in O3:K6 strains but absent from nonpandemic clinical isolates. The presence of such a protein could cause altered regulation of genes encoding virulence factors and could account for the enhanced transmission or virulence of the pandemic clone. Strains of the O3:K6 pandemic clone also contain tdh but not trh or ure (50). The genome sequence revealed many other potential virulence factors, including genes that encode homologues of E. coli cytotoxic necrotizing factor, Pseudomonas exoenzyme T, an RTX toxin, adherence factors, and other factors (138). The contribution of these factors to human disease caused by V. parahae­ molyticus is currently being investigated.

V. VULNIFICUS Of all of the pathogenic vibrios, V. vulnificus is the most serious disease-causing agent in the United States, alone responsible for 95% of all deaths caused by seafoodborne pathogens in that country. Indeed, in the state of Florida, this bacterium is the leading cause of reported deaths from all foodborne illness (95). Among that portion of the population that is at risk to infection by this bacterium, primary septicemia cases resulting from raw oyster consumption typically carry fatality rates of 50 to 60%. This is the highest death rate of any foodborne disease agent (232, 147). The CDC and FDA estimate there are approximately 50 cases of V. vulnificus foodborne illness per year in the United States that are serious enough to be recognized by

hospital personnel, although estimates of 17,500 to over 41,000 total cases have been calculated (232, 147). The bacterium is unusual in being able to produce wound, as well as food, infections. These carry a 20 to 25% fatality rate, are also seawater and/or shellfish associated, and frequently require surgical debridement or amputation of the affected tissue. The biology of V. vulnificus, as well as the clinical manifestations of both the primary septicemic and wound forms, has been reviewed (79, 104, 175, 177, 178, 219). Discussion here is limited to the foodborne (primary septicemic) form of infection caused by this bacterium.

Classification

The first detailed taxonomic study of this species was by researchers at the CDC, who in 1976 described 38 strains submitted by clinicians around the country. Originally termed the “lactose-positive” vibrio, its current name was suggested after a series of phenotypic and genetic studies by several laboratories. In 1982, a second biotype of V. vulnificus, that could easily be differentiated from biotype 1 by its negative indole reaction, was described. Biotype 2 strains are a major source of fatalities in eels but have been reported to lead to human infection in isolated instances. In 1999, a third biotype was described, which differs from biotype 1 and 2 isolates in being negative in the citrate and ONPG (o-nitrophenyl-b-d-galactopyranoside) tests and in lack of fermentation of salicin, cellobiose, and lactose. These biotype 3 strains appear to be a genetic hybrid of the biotype 1 and 2 forms (15) and to date have been isolated only in Israel, with all cases being wound infections. In 98% of the 62 cases, Tilapia spp. were implicated as the source of these infections. The key differential traits for the three biotypes of V. vul­ nificus are shown in Table 16.2. The discussions here focus on the originally described biotype 1, which is the major human pathogen. The isolation of V. vulnificus from blood samples is straightforward, as the bacterium grows readily on TCBS, MacConkey, and blood agars. Isolation and identification of environmental strains, on the other hand, have proven much more problematic. Vibrios tend to comprise 50% or more of estuarine bacterial populations, and many of these have not been characterized. While the phenotypic traits of this species have been fully described in a number of studies (3, 61, 111, 171), considerable variation exists in these traits, including lactose and sucrose fermentation, considered among the most important in identifying this species. This problem is exemplified by the isolation from a clinical sample of a bioluminescent strain of this species (182).

16.  Vibrio Species

421

Table 16.2  Differentiation of the three biogroups of V.

vulnificusa

Result for biogroup: Test

1

2

3

Ornithine decarboxylase Indole production Acid produced by:   d-Mannitol   d-Sorbitol   Cellobiose   Salicin

V +

− −

+ +

V − + +

− + + +

− − − −

a From Abbott et al. (3). Results were obtained after 48 h of incubation at 36°C. Most of the positive reactions occurred during the first 24 h. V, variable; +, positive; −, negative.

Various media have been proposed for isolating V. vulnificus, but the most commonly employed is ­colistinpolymyxin B-cellobiose (CPC) agar or one of several derivatives (173, 242). Many studies have employed this medium for the isolation of V. vulnificus from oysters and clams, and Sun and Oliver (221) determined that >80% of the 1,000 presumptive V. vulnificus colonies obtained from oyster homogenates could be identified as this species. A similar conclusion was reported by Sloan et al. (217) following their study of five selective enrichment broths and two selective agar media for isolating V. vulnificus from oysters. Currently, the FDA Bacteriological Analytical Manual (111) prescribes an 18- to 24-hour enrichment at 35 ± 2°C in APW, from which a loopful is streaked onto mCPC or CC agar (derivatives of CPC agar) to obtain isolated colonies of V. vulnificus. More recently, CHROMagar Vibrio has gained popularity, at least partially due to its ability to isolate and differentiate (by colony color) several pathogenic Vibrio spp., including V. vulnificus. The conclusive identification of this species is best accomplished by employing a probe against its hemolysin (vvhA) gene (159), and the latest FDA Bacteriological Analytical Manual (111) recommends the use of this probe. Several reports (e.g., references 243 and 244) have described the use of multiplex PCR for the detection of V. vulnificus in shellfish directed towards this and other gene targets.

Susceptibility to Control Methods

Cook and Ruple (42) determined that V. vulnificus cells naturally occurring in oysters underwent a time-dependent decrease in recovery when either shellstock oysters or shucked oyster meats were held at 4, 0, or −1.9°C. Cook (41) later revealed that, after harvest, V. vulnificus failed

to grow within 30 hours in shellstock oysters stored at 13°C or below. In oysters held at 18°C or higher for 12 or 30 hours, however, V. vulnificus numbers increased. Susceptibility of V. vulnificus to freezing, low-temperature pasteurization, high hydrostatic pressure, and ionizing radiation has been discussed in the introduction to this chapter.

Reservoirs

V. vulnificus is clearly a widespread inhabitant of estuarine environments, having now been described from the Gulf, East, and Pacific coasts of the United States and from around the world (111, 175, 178). V. vulnificus cells have been isolated from oysters, crabs, clams, ark shells, tarbos, plankton, and seawater samples (171). DePaola et al. (48) have isolated V. vulnificus from the intestinal tracts of a variety of bottom-feeding coastal fish and have suggested that such fish may represent a major reservoir of this species. Most studies have reported a lack of correlation between the presence of V. vulnificus and the presence of fecal coliforms, but this may be environment dependent (190). As exemplified in Fig. 16.2, a strong correlation between water temperature and isolation of Vibrio spp. has been observed by numerous investigators, agreeing with results of epidemiologic studies on V. vulnificus infections. Because of this seasonality and the inability to isolate V. vulnificus from water or oysters when water temperatures are low, there has been considerable investigation of the apparent die-off of this species during cold-weather months. This is now attributed to a coldinduced viable but nonculturable state, wherein the cells remain viable but are no longer culturable on the routine media normally employed for their isolation. This phenomenon, which has now been demonstrated in at least 25 genera, has been the subject of several reviews (172, 174, 176, 179, 180).

The Two Genotypes of V. vulnificus

In no case has more than one person developed V. vul­ nificus infection following consumption of the same lot of oysters. Indeed, the consumption of raw oysters and subsequent disease in one family member, while others exhibit no symptoms, is most common. Epidemiologic studies are complicated by the fact that raw oysters are usually eaten whole, and hence rarely do there exist any remains of the implicated oyster to sample. Additional raw oysters from the same lot, or even the same serving, may remain, but studies have indicated that two oysters taken from the same estuarine location may have vastly different numbers of V. vulnificus cells. Buchrieser et al.

422

Figure 16.2  Correlation between culturability of Vibrio spp. from estuarine environments and water temperature. From Pfeffer et al. (190). doi:10.1128/9781555818463.ch16f2

(18), employing clamped homogeneous electric field gel electrophoresis to examine 118 strains of V. vulnificus isolated from three oysters, reported that no two isolates had the same profile. Indeed, studies from other laboratories using arbitrarily primed PCR, ribotyping, PFGE, and amplified fragment length polymorphisms all indicate that no two V. vulnificus isolates have the same chromosomal arrangement (245). As noted by Buchrieser et al. (18), whether all strains are capable of causing infection or whether only certain strains are pathogenic remains to be determined. However, an answer to this question now appears to be developing. Studies reported as early as 1994 revealed that the 16S rRNA of V. vulnificus exhibits significant sequence variation and that two distinct genotypes may exist. Subsequent sampling studies revealed seasonal and environmental source variation that could be correlated to these two genotypes. The existence of several genotypes was further supported by Gutacker et al. (80), who used multilocus enzyme electrophoresis, randomly amplified polymorphic DNA-PCR, and sequence analysis of the recA and glnA genes to study the three biotypes of V. vulnificus. In a major study, Cohen et al. (37) were able to divide 63 of 67 V. vulnificus isolates into two main lineages using MLST of six housekeeping genes. Similarly, Sanjuán et al. (207) employed phenotypic and several genetic studies to characterize 111 V. vulnificus strains, with again two major lineages being observed. Such studies were extended by Rosche et al. (201), who identified significant differences not only in selected genes but also throughout the genome. Based on these findings, they developed a simple and rapid PCR procedure that strongly correlated the two genotypes with the source of their isolation. Most (90%) of the “C” genotype strains were

Foodborne Pathogenic Bacteria from clinical samples, whereas 93% of the strains isolated from a variety of environmental sources were classified as “E-type” (199). Hence, it is clear that two distinct genotypes of this pathogen exist, and it is likely that only one of these plays a significant role in initiating human infection. Interestingly, it was subsequently observed that while these two genotypes occur in approximately equal numbers in estuarine waters, oysters obtained from those waters contain ca. 84% of the E-genotype (243). Indeed, of 85 oysters harboring V. vulnificus, only two had more strains of the C genotype than of the E-type. This may help explain the low frequency of infections resulting from ingestion of raw oysters. Given such differential ability to cause human disease and to survive in oysters, Rosche et al. (201) suggested the two genotypes should be considered ecotypes.

Foodborne Outbreaks

Infections due to V. vulnificus have pronounced correlation with water temperature, with most cases occurring during warm months (Fig. 16.3). Furthermore, nearly all cases of V. vulnificus infection result from consumption of raw oysters, and most of these infections result in primary septicemias. In a study of 422 infections occurring in 23 states, Shapiro et al. (213) reported that 96% of those patients developing primary septicemia had consumed raw oysters. Similar findings were reported by Strom and Paranjpye (219) and by Oliver (175) in his review of over 500 cases. In the only prospective study performed to determine the incidence of vibrios in symptomatic or asymptomatic infections among persons eating raw shellfish, Lowry et al. (132) determined that 3 of the 479 persons tested had V. vulnificus present in their stools, although none was ill. Coupled with their finding that two-thirds of the raw oysters tested were culture positive for V. vulnificus, their study revealed that exposure to this species might be relatively high and underscores the need for “at-risk” persons to avoid raw seafood.

Characteristics of Disease

A number of major studies of infections caused by V. vulnificus have been reported. In a review of over 100 cases of primary septicemia, Oliver (175) determined the incubation period to range from 7 hours to 10 days, with most becoming symptomatic within 36 hours. The most significant symptoms included fever (94%), chills (86%), nausea (60%), and hypotension (systolic pressure, <85mm; 43%). While sometimes ­present, symptoms typical of gastroenteritis are not as common: abdominal pain (44%), ­ vomiting (35%), and diarrhea (30%). An unusual symptom that generally (69%) oc-

16.  Vibrio Species

423

Figure 16.3  Correlation between incidence of human V. vulnificus infection and seasonality. A total of 33 cases occurring during 2008 in the United States are shown. doi:10.1128/9781555818463.ch16f3

curs in these cases is the development of secondary lesions. These occur most frequently on the legs, frequently develop into necrotizing fasciitis or vasculitis, and often necessitate surgical debridement or limb amputation. Hlady and Klontz (94) determined that 94% of persons developing Vibrio infections in Florida were hospitalized, with stays of up to 43 days. Surprisingly, gastrointestinal disease with associated diarrhea is relatively rare. Johnston et al. (103) were the first to provide evidence of such a syndrome, reporting on three males who presented with abdominal cramps; V. vulnificus was isolated from their diarrheal stools. All three had a history of alcohol abuse, routinely took antacids, and had eaten raw oysters during the week prior to their illness. While symptoms in all three subsided without antibiotic treatment, diarrhea continued for a month in one case. Desenclos et al. (53), in their survey of vibrio infections in raw oyster eaters in Florida, reported eight cases of V. vul­ nificus-induced gastroenteritis, six of which involved consumption of raw oysters. The time to death in the fatal cases of V. vulnifi­ cus varies considerably, ranging from 2 hours after hospital admission to as long as 6 weeks (178). Most deaths occur within a few days. Hospital stays of several weeks are the norm for those surviving primary septicemic disease. To a great extent, survival depends on prompt antibiotic administration. Furthermore, V. vulnificus is susceptible to most antibiotics, and a large variety of antibiotics have been used clinically.

Infectious Dose and Susceptible Population

In almost all V. vulnificus infections that follow ingestion of raw oysters, the patient has an underlying chronic disease (173). The most common of these is a liver- or blood-related disorder, with liver cirrhosis secondary to alcoholism or alcohol abuse being the most typical. These diseases typically result in elevated serum iron levels, and laboratory studies of experimental V. vulnificus infections have revealed that elevated serum iron plays a major role in this disease (251). Other risk factors include hematopoietic disorders, chronic renal disease, gastric disease, use of immunosuppressive agents, and diabetes. The infectious dose of V. vulnificus is not known. Studies using mouse models, however, offer some insight on this point. Wright et al. (251) observed that, when mice were injected with 16 mg of iron to produce serum iron overload, the 50% lethal dose (LD50) decreased from ca. 106 CFU to a single cell. In a variation of these studies, the administration of small amounts of CCl4 to produce short-term liver necrosis was found to increase serum iron levels, with an inverse correlation between serum iron levels and LD50 observed. Such data agree with epidemiologic studies indicating that liver damage and immunocompromising diseases are major underlying factors in the development of V. vulnificus infections and suggest that ingestion of extremely low cell numbers of this pathogen may be sufficient to initiate potentially fatal infections. While cases in children have been reported, infections tend to be in males (86.5% of those reviewed by Oliver

Foodborne Pathogenic Bacteria

424 [173]) whose average age exceeds 50 years. A possible reason for this gender specificity has been suggested; Merkel et al. (149) found that estrogen protects against the V. vulnificus endotoxin, a virulence factor critical to infection with this bacterium (see below).

Virulence Mechanisms Capsule

There is no question that the polysaccharide capsule that is produced by nearly all strains of V. vulnificus is essential to its ability to initiate infection. Simpson et al. (216) examined 38 strains of V. vulnificus, including both clinical and environmental isolates and virulent and avirulent strains, and determined that all virulent strains were of the “opaque” (encapsulated) colony type, whereas isogenic cells obtained from “translucent” (acapsular) colonies, were avirulent (Fig. 16.4). It was further observed that encapsulated cells produce nonencapsulated cells and that this loss of capsule correlates with loss of virulence in otherwise isogenic strains. These authors subsequently determined that only the encapsulated cells were able to utilize transferrin-bound iron and that only these cells were iron responsive, i.e., were virulent at an inoculum of 103 CFU in iron-overloaded mice. The primary reason for the avirulence of translucent cells likely resides in the observation that the capsule allows these cells to resist phagocytosis. Capsular phase variation has been described (29, 252), with intermediate colony morphotypes reported (200) that may result from downregulation of capsular polysaccharide. A recent paper (69) also reported on the existence of a rugose phenotypic variant of V. vulnifi­ cus. The significance of intermediate and rugose cells in human virulence has yet to be fully elucidated. Simonson et al. (215) used polyclonal antibodies to study the capsule of V. vulnificus and identified at least 10 distinct serotypes. All 10 serotypes were associated with human infection, with 34.6% being of types 2 and 4. In contrast, 84.6% of the typeable environmental strains were of serotypes 3 or 5, but only 7.7% were of type 2 or 4. Similarly, Hayat et al. (88) reported that 19% of 21 clinical strains, but none of the 67 environmental isolates examined, agglutinated with antiserum prepared against a clinical isolate. Bush et al. (21) examined the sugars of these capsular polysaccharides in 120 V. vulnificus isolates and reported great diversity in capsular types, with no clear correlation of capsular types, genetic classification schemes, or pathogenic potential. Hence, whether capsular differences play a role in the epidemiology of V. vulnificus infections remains to be determined.

Iron

Elevated levels of serum iron appear to be essential for V. vulnificus to multiply in the human host (104, 219). The effect of elevated serum iron on reduction of LD50 in mice has been described above. Indeed, Wright et al. (251) have determined that normal human serum does not permit growth of V. vulnificus, suggesting that this bacterium may be able to produce septicemia only in those with elevated serum iron levels. Although V. vul­ nificus simultaneously produces both hydroxymate and phenolate siderophores, it is unable to compete with serum transferrin for iron, and this is likely a major factor in its inability to initiate infections in individuals with normal serum iron levels. A similar result has been reported for other iron-binding proteins, such as lactoferrin and ferritin. V. vulnificus is able to overcome the binding of haptoglobin to hemoglobin, however, and this may represent another aspect of the importance of iron in the pathogenesis of these infections. Recently, Bogard and Oliver (16) compared levels of resistance to human serum by C and E genotypes of V. vulnificus. It was determined that clinical genotype strains were significantly better able to survive in human serum than the environmental genotype strains, which might be explained by the possession by C strains of the side­ rophore-encoding gene, viuB.

LPS

The symptoms that occur during V. vulnificus septicemia, including fever, tissue edema, hemorrhage, and especially the significant hypotension, are those classically associated with gram-negative endotoxic shock. Hence, another product of V. vulnificus that may be critical to its virulence is the endotoxic LPS present in these cells. McPherson et al. (146) determined that intravenous injections of extracted and partially purified LPS from V. vulnificus caused decreased arterial blood pressure within 10 minutes in rats, which further declined, leading to decreased heart rate and death within 30 to 60 min. Subsequent studies revealed that an inhibitor of nitric oxide synthase (the LPS-induced enzyme responsible for release of nitric oxide and subsequent host tissue damage) administered 10 minutes after LPS injection reversed this lethal effect. These results indicate that the classic symptoms of endotoxic shock observed following V. vulnificus infection are likely due to the stimulation by LPS of nitric oxide synthase and that inhibition of this enzyme is a possible treatment for the endotoxic shock produced by this bacterium. As noted above, the gender-specificity of V. vulnificus infections appears to be due to an estrogeninduced protection of females against this endotoxin.

16.  Vibrio Species

425

Figure 16.4  Morphology of “opaque” (encapsulated) and “translucent” (nonencapsulated) colonies of V. vulnificus. doi:10.1128/9781555818463.ch16f4

Simonson et al. (215), employing monoclonal antibodies, have identified five LPS serological varieties in V. vulnificus. They found 25% of the clinical isolates expressed LPS antigens 1 and/or 5, whereas only 0.3% of environmental isolates were of these serotypes. While the chemical composition of the LPS extracted from one strain of V. vulnificus has been reported, nothing is known regarding the relative virulence of the different serotypes.

Other Toxins

In addition to the role of capsule, iron, and endotoxin in the pathogenesis of V. vulnificus infections, this bacterium produces a large number of extracellular compounds, including hemolysin, protease, elastase, collagenase, DNase, lipase, phospholipase, mucinase, chondroitin sulfatase, hyaluronidase, and fibrinolysin. To date, however, none of these putative virulence factors has definitively been shown to be involved in human pathogenesis. The most studied of these is a powerful heat-stable hemolysin/cytotoxin (product of the vvhA gene) that possesses cytolytic activity against a variety of mammalian erythrocytes, cytotoxic activity against CHO cells, vascular permeability activity against guinea pig skin, and lethal activity against mice (104). Production of the hemolysin, a metalloprotease, is regulated by a transmembrane virulence regulator homologous to ToxRS of V. cholerae. The toxin has been purified and has a molecular weight of ca. 56,000 and an LD50 for mice of ca. 3mg/kg of body weight following intravenous injection. It enhances vascular permeability

through the release of bradykinin, and can also specifically degrade type IV collagen, thereby destroying the basal membrane layer of capillary vessels. Interestingly, low-density lipoprotein inactivates this cytolysin through the oligomerization of the toxin monomer. However, mutants for the hemolysin/cytolysin do not reduce the LD50 for mice following intraperitoneal injection (212), and its role as a virulence factor remains unclear. The reader is directed to a recent review on the hemolysins of Vibrio species (259) for more in-depth information on these putative virulence factors. An elastolytic protease that lacks hemolytic activity but degrades albumin, immunoglobulin G, elastin, and complement factors C3 and 4 has been described. Minimum lethal doses in mice, regardless of route of injection, were reported to be ca. 25 mg, with extensive hemorrhagic necrosis, edema, and muscle tissue destruction occurring. A broad-specificity metalloprotease (product of the vvpE gene) is also produced (104), which cleaves several plasma proteins and in addition interferes with a number of blood clotting functions (28). Mutations in either the vvhA or vvpE genes, however, do not result in changes in either cytotoxicity or lethality compared to parent strains (102, 212). Rhee’s group (113) investigated the role of adenylate cyclase in the virulence of this species and reported that production of hemolysin and protease, motility, and cytotoxicity against HeLa cells were all decreased in a cya mutant. Furthermore, colony morphology (due to capsule production) was modified in this mutant, and LD50 in mice increased 100-fold.

Foodborne Pathogenic Bacteria

426 An RTX toxin, with high homology to that found in V. cholerae, has been studied in several labs (104). This toxin results in a rearrangement of the cytoskeletal structure, and mutants of this toxin exhibited decreased cellular damage compared to parent strains. It has been suggested that such cellular change might lead to cellular necrosis, allowing V. vulnificus to invade the bloodstream by crossing the intestinal epithelium (114). Interestingly, these authors also provide evidence that contact with host cells may result in increased toxin expression. There is increasing indication that flagella and motility may play a part in the pathogenesis of this species. Recently, Kim et al. (112) used in vivo-induced antigen technology to identify V. vulnificus genes induced in serum obtained from septicemic patients. They observed that proteins involved in chemotaxis, among other functions, stood out as likely being important virulence gene products. Increases in LD50 of V. vulnificus cells mutated in flagellar genes has also been reported (124). Similarly, strong evidence for a role of pili in attachment and virulence has been shown (186). While none of these many putative virulence factors is known to be essential to virulence of V. vulnificus, they likely play a role in the pathogenesis of foodborne infections or might be essential for the wound infections also produced by this species. Several excellent reviews of the molecular pathogenesis of this species have recently been published (79, 104), and the reader is directed to these and other recent reviews on disease production by this species (166, 173, 219).

V. FLUVIALIS

Classification

V. fluvialis was originally described by Lee et al. (125) and referred to as “Group F” Vibrio. Subsequently, it was determined that this group was identical to that referred to as “Group EF-6” Vibrio by the CDC, and further taxonomic study concluded that these isolates were a new species, V. fluvialis. Two “biogroups” of group F vibrios were originally described, of which biogroup I was anaerogenic and isolated from aquatic environments and diarrheal cases and biogroup II was aerogenic and not disease associated. Subsequent studies revealed that the aerogenic strains were a unique species, and these were reclassified (17) as V. furnissii (see “V. furnissii” below). The possibility of misidentifying V. fluvialis as Aero­ monas (especially A. hydrophila) exists, as both are arginine dihydrolase positive. The simplest differentiation is the inability of V. fluvialis, being halophilic, to grow

in media lacking NaCl. The lack of production of indole by V. fluvialis also differentiates this species from Aeromonas spp. (106).

Reservoirs

V. fluvialis has been frequently isolated from brackish and marine waters and sediments in the United States as well as other countries. It has also been isolated from fish and shellfish from the Pacific Northwest and Gulf Coast. Wong et al. (249) reported ca. 65 to 79% of the oysters, hard clams, and freshwater clams they assayed to harbor V. fluvialis, but only 25% of the crabs and 6% of the shrimp were V. fluvialis positive. The organism has only rarely been isolated from freshwaters.

Foodborne Outbreaks

V. fluvialis infections are common in areas that have high levels of fecal-contaminated water, food supplies and consumption of raw seafood, or contaminated seafood products (247). V. fluvialis was isolated from over 500 patients with diarrheal stools at the Cholera Research Laboratory of Bangladesh during a 9-month period (99). Approximately one-half of the patients were less than 5 years of age. Since that outbreak, however, V. fluvialis has only occasionally been reported to be an enteric pathogen. In the United States, Levine et al. (128) determined that V. fluvialis accounted for 10% of the clinical cases in their survey of Vibrio infections along the Gulf Coast. All seven of these cases had gastroenteritis, with three requiring hospitalization. Consumption of raw oysters was implicated in at least three of the seven cases, and shrimp in one other. In the largest study of clinical cases of V. fluvia­ lis cases occurring in the United States, Klontz and Desenclos (116) described 12 persons with gastro­enteritis in Florida from whom this species was recovered between 1982 and 1988. Eight of the 10 from whom the species was isolated reported eating seafood during the week before they became ill, with raw oysters implicated in five cases, shrimp in two, and cooked fish in one case. In a subsequent survey of vibriosis resulting from raw oyster consumption in Florida during an 8-year period, Desenclos et al. (53) reported 5.6% of 125 gastroenteritis cases to be caused by V. fluvialis.

Characteristics of Disease

V. fluvialis-related illness is characterized by gastro­ enteritis, nausea, loss of appetite, vomiting, and watery bloody diarrhea with abdominal cramps or significant fever (100). Diarrhea typically lasts from 16 hours to over 3 days. Moderate to severe dehydration, hypokale-

16.  Vibrio Species mia, metabolic acidosis, and occasionally, hypovolemic shock can occur in 4 to 12 hours if fluid losses are not replaced. Stools are colorless, with small flecks of mucus, and contain high concentrations of sodium, potassium, chloride, and bicarbonate. A notable difference from cholera is the frequent occurrence of bloody stools in infections due to V. fluvialis. Recently, V. fluvialis has been reported as causing necrotizing fasciitis and septicemia in the Gulf of Mexico and Southeast Asia, associated with minor trauma and exposure to fish, raw oysters, shellfish, crabs, or seawater, especially in the summer months (234).

Infectious Dose and Susceptible Population

The infectious dose for this species is not known. Persons developing gastroenteritis have been from 1 month old to >80 years old, and it does not appear that underlying disease plays a major role in the infections caused by this species. Klontz and Desenclos (116) reported that only 4 of the 10 patients they studied with V. flu­ vialis gastroenteritis had underlying medical conditions, which included diabetes, alcohol abuse, and ulcerative colitis. One patient had a history of cardiopulmonary disease and peptic ulcers and was taking antacids at the time of hospitalization. While successful resolution of infection has been reported without antibiotic therapy, such treatment, often along with intravenous fluids, is generally administered.

Virulence Mechanisms

V. fluvialis produces several extracellular products that may be important in pathogenesis including an enterotoxin-like substance (which is immunologically indistinguishable from cholera toxin), elastase, mucinase, protease, lipase, lecithinase, chondroitin sulfatase, hyaluronidase, DNase, fibrinolysin, and a hemolysin (100, 181). In addition, endotoxin activity of V. fluvialis has been observed in vitro using CHO cells (32). Han et al. (83) purified the hemolysin of V. fluvialis from culture supernatant fluids and determined it was hemolytic to a variety of mammalian erythrocytes. The nucleotide sequence of the vfh gene encoding this toxin, as well as the physical nature of the 740-amino-acid protein which results, was also characterized. Kothary et al. (119) subsequently reported on this hemolysin (although a different molecular weight was derived), noting that in addition to lysing erythrocytes of eight different animal species, it was cytotoxic against CHO cells and elicited fluid accumulation in suckling mice. Most interesting was the finding that 14 of the first 20 N-terminal amino acid residues were identical to the El Tor hemolysin of V. chol­ erae and the heat-labile hemolysin of V. mimicus. Finally,

427 Ahn et al. (4) identified an iron-regulated hemin-binding outer membrane protein that affected hemolytic activity and oxidative stress response, and a mutant of which had a significantly reduced LD50.

V. FURNISSII

Classification

The taxonomy of V. furnissii has been extensively described by Brenner et al. (17). Their studies included biochemical reactions and DNA-DNA hybridization relatedness to V. fluvialis, as well as to Aeromonas and Alteromonas species and other vibrios. This species is similar to Aeromonas hydrophila, from which it can easily be distinguished through its ability to grow in 6% NaCl. V. furnissii can be differentiated from V. fluvialis primarily by its production of gas from glucose (Table 16.1). Recently, the complete genome sequence of V. furnissii strain NCTC 11218 was determined using shotgun and pyrosequencing (133).

Reservoirs

V. furnissii has been isolated from river and estuarine waters and from marine mollusks and crustaceans from throughout the world. Wong et al. (249) determined that ca. 7 to 12% of the oysters, clams, shrimp, and crabs they assayed were V. furnissii positive.

Foodborne Outbreaks

The largest documented occurrences of V. furnissii infection were in 1969, when this species was isolated from two outbreaks of acute gastroenteritis in American tourists returning from Asia. In the first outbreak, 23 of 42 elderly passengers returning from Tokyo developed gastroenteritis; one woman died and two other persons required hospitalization. Food histories implicated shrimp and crab salad and/or the cocktail sauce served with the salads. V. furnissii was recovered from seven stool specimens, two of which also contained V. parahaemolyticus. The second outbreak affected 24 of 59 persons returning from Hong Kong. Nine persons were hospitalized. A food vehicle was not identified, but V. furnissii was isolated from at least five fecal specimens. However, because several other potentially enteropathogenic bacteria were also isolated from these stool samples, an absolute causal role of V. furnissii could not be documented.

Characteristics of Disease

Symptoms described by Brenner et al. (17) for the gastroenteritis outbreaks described above included diarrhea (91 to 100%), abdominal cramps (79 to 100%), nausea (65 to 89%), and vomiting (39 to 78%). Fever was not

428 reported. Onset of symptoms was between 5 and 20 hours, with patients recovering within 24 hours.

Infectious Dose and Susceptible Population

The infectious dose of V. furnissii and populations susceptible to this pathogen are not known.

Virulence Mechanisms

Putative virulence factors of V. furnissii include factors causing CHO cell elongation (32), factors cross-reacting with cholera toxin and causing fluid accumulation in rabbit ileal loops (130), and phosphomannomutase (11). Wu et al. (255) described an oligopeptide permease protein (Hly-OppA) from V. furnissii that has both a solute-binding function and an in vitro hemolytic activity, and they determined its virulent effect in mice. Despite reports of the pathogenicity of V. furnissii, the genome sequence that was recently determined does not contain homologues to a number of the genes associated with virulence identified in other Vibrio species, such as toxRT, zot, ctxAB, and genes encoding the TTSS (133).

V. HOLLISAE

Classification

V. hollisae was described as a new species in 1982 (90). Of the original 16 strains obtained and characterized by the CDC, 15 were from stool samples or intestinal contents, and many of the patients involved had diarrhea. The phenotypic traits of this species are described in the CDC paper and in Table 16.1. Based on 16S rRNA gene analysis, the placement of this species into a newly formed genus as Grimontia hollisae has recently been proposed (229). V. hollisae is unusual among the vibrios in its inability to grow on TCBS agar or MacConkey agar. As these two media are routinely employed for the examination of stool samples for vibrios, it is possible that infections caused by this species are not detected in clinical laboratories. V. hollisae grows well on blood agar and marine agar, however, and culture on xyloselysine-desoxycholate agar recovers this species. The use of the API-20E system has generally been reported to properly identify V. hollisae (1).

Reservoirs

The distribution of V. hollisae is not well documented, although it is likely a marine species, and it appears that this bacterium prefers warm waters (1).

Foodborne Pathogenic Bacteria

Foodborne Outbreaks

According to Abbott and Janda (1), 30 cases of V. holli­ sae have been reported in the literature since the original description of this species. Most (87%) were cases of gastroenteritis in adults, with 13% causing extraintestinal disease. Only four cases of septicemia associated with V. hollisae have been reported in the literature (92). Three cases occurred in patients with underlying liver disease. There appears to be a strong correlation between V. hollisae infections and consumption of raw seafood. Cases also have been reported following consumption of fried catfish and of dried and salted fish, suggesting an ability of the bacterium to survive these treatments. In addition, Abbott and Janda (1) cited a case of V. hollisae-associated diarrhea that involved a 61-year-old male who denied recent travel or seafood consumption. This suggests that additional vehicles may remain to be identified. Desenclos et al. (53), in a study of 333 adult cases of Vibrio illnesses in Florida over an 8-year period, determined that 32 cases were due to V. hollisae. Of these, 20 (62.5%) occurred following ingestion of raw oysters. Although no description of these cases was included in their study, 17 of these 20 cases developed gastroenteritis, while another 3 developed into septicemia. Morris et al. (158) described nine cases of diarrhea that were culture positive for V. hollisae, with no other enteric pathogen identified. All patients had diarrhea and abdominal pain, and all but one were hospitalized. The foods implicated were raw oysters or clams in six cases and raw shrimp in another. A seventh patient had eaten seafood but denied eating it raw. Hinestrosa et al. (92) reported a case of a 43-year-old Vietnamese male who, after shellfish consumption, suffered a more severe form of gastroenteritis, presenting with profound hypotension and acute renal failure secondary to hypovolemic shock. Although V. hollisae had previously been obtained from a septicemia case (158), Lowry et al. (131) were the first to describe a case of septicemia from which V. hollisae alone was isolated from blood cultures. A 65-year-old male was admitted to hospital with a 12hour history of night fever, vomiting, and abdominal pain. He had passed two loose stools. On the day before hospitalization, the patient ate fried Mississippi River catfish for lunch and again at dinner. His illness began at midnight, and on admission V. hollisae was isolated from a blood culture. The patient was given antibiotics and was discharged 8 days after admission. This case is unusual not only in being a septicemia but also in that the infection occurred following consumption of a

16.  Vibrio Species freshwater fish that had been fried. There was no history of exposure to saltwater or other seafood consumption by the patient. However, catfish are capable of adapting to low salinities, and the region of the Mississippi River where the fish was obtained often has salt concentrations up to 0.5%. It is also possible that incomplete cooking or recontamination of the fish occurred. An additional case of foodborne septicemia has been reported that was unusual in that the vehicle appeared to be dried and salted fish, obtained from a Southeast Asian food store, that was eaten uncooked. In the first case reported in Europe, Gras-Rouzet et al. (75) described a case of gastroenteritis and bacteremia in a previously healthy 76-year-old man who ate cockles from Brittany, France. In 2009, Edouard et al. (57) reported the first case of Grimontia hollisae bacteremia in the Mediterranean area, in an 81-year-old man with a severe gastroenteritis and hepatitis following the consumption of raw oysters.

Characteristics of Disease

Symptoms of gastroenteritis caused by V. hollisae are similar to those caused by non-O1 strains of V. cholerae (158) and typically include severe abdominal cramping, vomiting, fever, and watery diarrhea (1). In the cases reported by Morris et al. (158), the median duration of diarrhea was 1 day (range of 4 hours to 13 days), with occasional reports of bloody diarrhea. Eight of these patients were admitted to a hospital (median duration of 5 days, range of 2 to 9 days), but all recovered.

Infectious Dose and Susceptible Population

While six of the nine gastroenteritis cases described by Morris et al. (158) were male, with an average age of 35 years (range of 31 to 59 years), other reports do not suggest any significant preference for gender or age (1). There is no evidence of seasonality-associated infections, with cases appearing in the winter months as well as summer. Most cases of gastroenteritis are in otherwise healthy individuals, although abnormal liver function secondary to alcohol abuse was present in one case reported by Morris et al. (158). In the septicemia case described by Lowry et al. (131), the patient had consumed a fifth of wine daily for 20 years, and had previously undergone a pancreatectomy and splenectomy as well as other surgical procedures. Unlike cases of gastro­ enteritis, underlying disease appears to be the norm for septicemic cases (1, 158). The bacterium is highly susceptible to most antimicrobial agents (1), and tobramycin and cefamandole were successfully used in treatment of a septicemic case (131).

429

Virulence Mechanisms

Gene sequences in V. hollisae that are homologous to the TDH gene of V. parahaemolyticus have been reported, although strain-to-strain variation apparently exists (165). The hemolysin has been purified, partially characterized, and reported to be related to the V. para­ haemolyticus TDH. The two toxins have also been determined to have similar lethal toxicities in mice. Intragastric administration of V. hollisae cells into infant mice elicits intestinal fluid accumulation. An enterotoxin that elongates CHO cells and causes fluid accumulation in mice has been purified (118) and was detected in extracts of infected mice and in culture fluids from various growth media. A hydroxamate siderophore, identified as aerobactin, is produced by V. hollisae in response to iron limitation (223). This iron-binding protein may play a role in the ability of this species to cause infection, and the patient in one of the bacteremic cases had improperly dosed himself with ferrous sulfate as a supplement for chronic anemia, a treatment that may have exacerbated the disease process. Miliotis et al. (153) have described the ability of V. hollisae to adhere to and invade cultured epithelial cells, with internalization involving both eukaryotic and prokaryotic factors. The authors suggested that in addition to toxin production, the ability to invade epithelial cells is consistent with the invasive disease produced by V. hollisae in some patients (131).

V. ALGINOLYTICUS

Classification

V. alginolyticus was originally classified as a biotype of V. parahaemolyticus, and indeed the two are genetically quite similar. However, they can easily be differentiated phenotypically, most readily by the fermentation of sucrose by V. alginolyticus. Some differences in taxonomic traits between clinical and environmental isolates of V. alginolyticus have been suggested, but whether these are significant has not been demonstrated.

Reservoirs

V. alginolyticus has generally been reported to inhabit, often in high cell numbers, seawater and seafood obtained worldwide (106). It is easily isolated from fish, clams, crabs, oysters, mussels, and shrimp, as well as water. Indeed, many surveys have revealed this species is one of the most commonly isolated of the vibrios. Several studies have reported a temperature ­correlation

Foodborne Pathogenic Bacteria

430 with its isolation, with cell numbers greatest in the warm water months.

Foodborne Outbreaks

Most V. alginolyticus infections are associated with exposure to the marine environment and generally remain superficial (mainly ear soft tissue and wound infections). Only rarely has V. alginolyticus been implicated as a foodassociated pathogen. This species has been isolated from 0.5% of healthy people in Japan, with no clinically associated intestinal disease evident. V. alginolyticus has been isolated from the rice water diarrheal stool of a female patient with acute enterocolitis and also from the trout roe she had consumed. V. alginolyticus has also been isolated from blood of a leukemic, 49-year-old female who had consumed raw oysters 1 week earlier. Desenclos et al. (53) reported a case of V. alginolyticus-induced gastroenteritis in Florida during an 8-year survey but did not provide details. This case represented only one of the 333 bacteriologically confirmed Vibrio illnesses reported in that study. Chien et al. (31), however, have described a case of V. alginolyticus bacteremia in an immunocompromised patient who had eaten a large amount of seafood.

Characteristics of Disease

There are few reports in the literature describing the symptoms of gastroenteric disease caused by V. algino­ lyticus, and acute diarrhea has rarely been associated with these infections. A leukemic patient with V. alginolyticus infection exhibited confusion, shock, anemia, a temperature of 40°C, and a systolic blood pressure of 80 mm Hg. Despite antibiotic administration and treatment for shock, the patient died 12 days after admission. The exact role of V. alginolyticus in this case is unknown, however, as Pseudomonas aeruginosa was also isolated from the patient’s blood and the presence of the latter pathogen was very possibly a factor in the fatal outcome.

Infectious Dose and Susceptible Population

Whereas extraintestinal infections are typically selflimiting­ and relatively mild, systemic infections are generally severe. Most such cases appear to involve patients who are immunocompromised due to severe burns or cancer. A history of alcohol abuse may also be important in the development of V. alginolyticus infections.

Virulence Mechanisms

Virtually nothing is known regarding the pathogenic mechanisms of V. alginolyticus in humans. Oliver et al. (181) reported the production of lipase, lecithinase, chondroitin sulfatase, DNase, and hemolysin in the strain of V. algino­

lyticus they examined but saw no evidence for the production of elastase, mucinase, protease, or hyaluronidase, nor were culture filtrates cytotoxic for CHO cells.

References 1. Abbott, S. L., and J. M. Janda. 1994. Severe gastro­ enteritis associated with Vibrio hollisae infection: report of two cases and review. Clin. Infect. Dis. 18:310–312. 2. Reference deleted. 3. Abbott, S. L. J. M. Janda, J. A. Johnson, and J. J. Farmer III. 2007. Vibrio and related organisms Vibrio and related organisms, p. 723–733. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry, and M. A. Pfaller (ed.), Manual of Clinical Microbiology, 9th ed. ASM Press, Washington, DC. 4. Ahn, S.-H., J.-H. Han, J.-H. Lee, K.-J. Park, and I.-S. Kong. 2005. Identification of an iron-regulated heminbinding outer membrane protein, HupO, in Vibrio flu­ vialis: effects on hemolytic activity and the oxidative stress response. Infect. Immun. 73:722–729. 5. Alam, M., W. B. Chowdhury, N. A. Bhuiyan, A. Islam, N. A. Hasan, G. B. Nair, H. Watanabe, A. K. Siddique, A. Huq, R. B. Sack, M. Z. Akhter, C. J. Grim, K. M. Kam, C. K. Y. Lucy, H. P. Endtz, A. Cravioto, and R. R. Colwell. 2009. Serogroup, virulence, and genetic traits of Vibrio parahaemolyticus in the estuarine ecosystem of Bangladesh. Appl. Environ. Microbiol. 75:6268–6274. 6. Albert, M. J. 1994. Vibrio cholerae O139 Bengal. J. Clin. Microbiol. 32:2345–2349. 7. Albert, M. J., and G. B. Nair. 2005. Vibrio cholerae O139 Bengal—10 years on. Rev. Med. Microbiol. 16:135–143. 8. Andrews, L., D. L. Park, and Y.-P. Chen. 2000. Low temperature pasteurization to reduce the risk of Vibrio infections in raw-shellstock oysters. Proc. 25th Annu. Meet. Seafood Sci. Technol. Soc. Longboat Key, FL. 9. Andrews, L., M. Jahncke, and K. Mallikarjunan. 2003. Low dose gamma irradiation to reduce pathogenic vibrios in live oysters (Crassostrea virginica). J. Aquat. Food Prod. Tech. 121:71–82. 10. Anonymous. 1978. V. parahaemolyticus foodborne outbreak – Louisiana. MMWR Morb. Mortal. Wkly. Rep. 27:345–346. 11. Austin, B. 2010. Vibrios as causal agents of zoonoses. Vet. Microbiol. 140:310–317. 12. Benitez, J. A., L. Garcia, A. Silva, H. Garcia, R. Fando, B. Cedre, A. Perez, J. Campos, B. L. Rodriguez, J. L. Perez, T. Valmaseda, O. Perez, A. Perez, M. Ramirez, T. Ledon, M. D. Jidy, M. Lastre, L. Bravo, and G. Sierra. 1999. Preliminary assessment of the safety and immunogenicity of a new CTXF-negative, hemagglutinin/protease-defective El Tor strain as a cholera vaccine candidate. Infect. Immun. 67:539–545. 13. Bhuiyan, N. A., S. Nusrin, M. Alam, M. Morita, H. Watanabe, T. Ramamurthy, A. Cravioto, and G. B. Nair. 2009. Changing genotypes of cholera toxin (CT)

16.  Vibrio Species

14.

15. 16.

17.

18.

19.

20.

21.

22. 23.

24.

25.

26.

of Vibrio cholerae O139 in Bangladesh and description of three new CT genotypes. FEMS Immunol. Med. Microbiol. 57:136–141. Bina, J., J. Zhu, M. Dziejman, S. Faruque, S. Calderwood, and J. Mekalanos. 2003. ToxR regulon of Vibrio chol­ erae and its expression in vibrios shed by cholera patients. Proc. Natl. Acad. Sci. USA 100:2801–2806. Bisharat, N., D. I. Cohen, R. M. Harding, D. Falush, D. W. Crook, T. Peto, and M. C. Maiden. 2005. Hybrid Vibrio vulnificus. Emerg. Infect. Dis. 11:30–35. Bogard, R., and J. D. Oliver. 2007. Role of iron in human serum resistance of the clinical and environmental Vibrio vulnificus genotypes. Appl. Environ. Microbiol. 73:7501–7505. Brenner, D. J., F. W. Hickman-Brenner, J. V. Lee, A. G. Steigerwalt, G. R. Fanning, D. G. Hollis, J. J. Farmer III, R. E. Weaver, S. W. Joseph, and R. J. Seidler. 1983. Vibrio furnissii (formerly aerogenic biogroup of Vibrio fluvialis), a new species isolated from human feces and the environment. J. Clin. Microbiol. 18:816–824. Buchrieser, C., V. V. Gangar, R. L. Murphree, M. L. Tamplin, and C. W. Kaspar. 1995. Multiple Vibrio vul­ nificus strains in oysters as demonstrated by clamped homogeneous electric field gel electrophoresis. Appl. Environ. Microbiol. 61:1163–1168. Buck, J. D. 1998. Potentially pathogenic Vibrio spp. in market seafood and natural habitats from Southern New England and Florida. J. Aquat. Food Prod. Technol. 7:53–62. Burkhardt, W., III, S. R. Rippey, and W. D. Watkins. 1992. Depuration rates of Northern quahogs, Merce­ naria mercenaria (Linnaeus, 1758) and Eastern oysters Crassostrea virginica (Gmelin, 1791) in ozone- and ultraviolet light-disinfected seawater systems. J. Shellfish Res. 11:105–109. Bush, C. A., P. Patel, S. Gunawardena, J. Powell, A. Joseph, J. A. Johnson, and J. G. Morris. 1997. Classification of Vibrio vulnificus strains by the carbohydrate composition of their capsular polysaccharides. Anal. Biochem. 250:186–195. Butler, S. M., and A. Camilli. 2005. Going against the grain: chemotaxis and infection in Vibrio cholerae. Nat. Rev. Microbiol. 3:611–620. Caburlotto, G., M. M. Lleo, T. Hilton, A. Huq, R. R. Colwell, and J. B. Kaper. 2010. Effect on human cells of environmental Vibrio parahaemolyticus strains carrying type III secretion system 2. Infect. Immun. 78:3280–3287. Calik, H., M. T. Morrissey, P. W. Reno, and H. An. 2002. Effect of high-pressure processing on Vibrio parahaemolyticus strains in pure culture and Pacific oysters. J. Food Sci. 67:1506–1510. Cameron, D. N., F. M. Khambaty, I. K. Wachsmuth, R. V. Tauxe, and T. J. Barrett. 1994. Molecular characterization of Vibrio cholerae O1 by pulsed-field gel electrophoresis. J. Clin. Microbiol. 32:1685–1690. Camilli, A., and B. L. Bassler. 2006. Bacterial smallmolecule signaling pathways. Science 311:1113–1116.

431 27. Centers for Disease Control and Prevention. 2009. Centers for Disease Control and Prevention (CDC), Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food – 10 states, 2008, MMWR Morb. Mortal. Weekly Rep. 58:333–337. 28. Chang, A. L, H. Y. Kim, J. E. Park, P. Acharya, I.-S. Park, S. M Yoon, H. J. You, K.-S. Ham, J. K. Park, and J. S. Lee. 2005. Vibrio vulnificus secretes a broadspecificity metalloprotease capable of interfering with blood homeostasis through prothrombin activation and fibrinolysis. J. Bacteriol. 187:6909–6916. 29. Chatzidaki-Livanis, M., M. K. Jones, and A. C. Wright. 2006. Genetic variation in the Vibrio vulnificus group 1 capsular polysaccharide operon. J. Bacteriol. 188:1987–1998. 30. Chen, W., S. Yu, C. Zhang, J. Zhang, C. Shi, Y. Hu, B. Suo, H. Cao, and X. Shi. 2011. Development of a single base extension-tag microarray for the detection of pathogenic Vibrio species in seafood. Appl. Microbiol. Biotechnol. 89:1979–1990. 31. Chien, J. Y., J. T. Shih, P. R. Hsueh, P. C. Yang, and K. T. Luh. 2002. Vibrio alginolyticus as the cause of pleural empyema and bacteremia in an immuno­ compromised patient. Eur. J. Clin. Microbiol. Infect. Dis. 21:401–403. 32. Chikahira, M., and K. Hamada. 1988. Enterotoxigenic substances and other toxins produced by Vibrio fluvia­ lis and Vibrio furnissii. Jpn. J. Vet. Sci. 50:865–873. 33. Chin, C. S., J. Sorenson, J. B. Harris, W. P. Robins, R. C. Charles, R. R. Jean-Charles, J. Bullard, D. R. Webster, A. Kasarskis, P. Peluso, E. E. Paxinos, Y. Yamaichi, S. B. Calderwood, J. J. Mekalanos, E. E. Schadt, and M. K. Waldor. 2011. The origin of the Haitian cholera outbreak strain. N. Engl. J. Med. 364:33–42. 34. Chitov, T., S. Wongdao, W. Thatum, T. Puprae, and P. Sisuwan. 2009. Occurrence of potentially pathogenic Vibrio species in raw, processed, and ready-to-eat seafood and seafood products. Maejo Int. J. Sci. Technol. 3:88–98. 35. Cho, Y. J., H. Yi, J. H. Lee, D. W. Kim, and J. Chun. 2010. Genomic evolution of Vibrio cholerae. Curr. Opin. Microbiol. 13:646–651. 36. Chun, J., C. J. Grim, N. A. Hasan, J. H. Lee, S. Y. Choi, B. J. Haley, E. Taviani, Y. S. Jeon, D. W. Kim and T. S. Brettin, D. C. Bruce, J. F. Challacombe, J. C. Detter, C. S. Han, A. C. Munk, O. Chertkov, L. Meincke, E. Saunders, R. A. Walters, A. Huq, G. B. Nair, and R. R. Colwell. 2009. Comparative genomics reveals mechanism for short-term and long-term clonal transitions in pandemic Vibrio cholerae. Proc. Natl. Acad. Sci. USA 106:15442–15447. 37. Cohen, A. L. V., J. D. Oliver, A. DePaola, E. J. Feil, and E. F. Boyd. 2007. Emergence of a virulent clade of Vibrio vulnificus and correlation with the presence of a 33-kilobase genomic island. Appl. Environ. Microbiol. 73:5553–5565. 38. Colwell, R. R., and A. Huq. 1994. Vibrios in the environment: viable but nonculturable Vibrio cholerae, p. 117–133. In K. Wachsmuth, P. A. Blake, and O.

Foodborne Pathogenic Bacteria

432

39.

40.

41.

42.

43.

44.

45. 46.

47. 48.

49.

50.

Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, DC. Colwell, R. R., M. L. Tamplin, P. R. Brayton, A. L. Gauzens, B. D. Tall, D. Herrington, M. M. Levine, S. Hall, A. Huq, and D. A. Sack. 1990. Environmental aspects of Vibrio cholerae in transmission of cholera, p. 327–343. In R. B. Sack and Y. Zinnaka (ed.), Advances in Research on Cholera and Related Diarrheas, vol. 7. KTK Scientific Publishers, Tokyo, Japan. Colwell, R. R., A. Huq, M. S. Islam, K. M. A. Aziz, M. Yunus, N. H. Khan, A. Mahmud, R. B. Sack, G. B. Nair, J. Chakraborty, D. A. Sack, and E. RussekCohen. 2003. Reduction of cholera in Bangladeshi villages by simple filtration. Proc. Natl. Acad. Sci. USA 100:1051–1055. Cook, D. W. 1994. Effect of time and temperature on multiplication of Vibrio vulnificus in postharvest Gulf Coast shellstock oysters. Appl. Environ. Microbiol. 60:3483–3484. Cook, D. W., and A. D. Ruple. 1992. Cold storage and mild heat treatment as processing aids to reduce the numbers of Vibrio vulnificus in raw oysters. J. Food Prot. 55:985–989. Dalsgaard, A., O. Serichantalergs, A. Forslund, W. Lin, J. Mekalanos, E. Mintz, T. Shimada, and J. G. Wells. 2001. Clinical and environmental isolates of Vibrio cholerae serogroup O141 carry the CTX phage and the genes encoding the toxin-coregulated pili. J. Clin. Microbiol. 39:4086–4092. Daniels, N. A., L. MacKinnon, R. Bishop, S. Altekruse, B. Ray, R. M. Hammond, S. Thompson, S. Wilson, N. H. Bean, P. M. Griffin, and L. Slutsker. 2000. Vibrio parahaemolyticus infections in the United States, 1973– 1998. J. Infect. Dis. 181:1661–1666. Das, B., S. K. Manna, P. Sarkar, and K. Batabyal. 2009. Occurrence of Vibrio parahaemolyticus in different finfish and shellfish species. J. Food Saf. 29:118–125. Debellis, L., A. Diana, D. Arcidiacono, R. Fiorotto, P. Portincasa, D. F. Altomare, C. Spirlì, and M. de Bernard. 2009. The Vibrio cholerae cytolysin promotes chloride secretion from intact human intestinal mucosa. PLoS One 4:e5074. de Haan, L., and T. R. Hirst. 2004. Cholera toxin: a paradigm for multi-functional engagement of cellular mechanisms. Mol. Membr. Biol. 21:77–92. DePaola, A., G. M. Capers, and D. Alexander. 1994. Densities of Vibrio vulnificus in the intestines of fish from the U.S. Gulf Coast. Appl. Environ. Microbiol. 60:984–988. DePaola, A., J. L. Norstrom, J. C. Bowers, J. G. Wells, and D. W. Cook. 2003. Seasonal abundance and total and pathogenic V. parahaemolyticus in Alabama oysters. Appl. Environ. Microbiol. 69:1521–1526. DePaola, A., J. Ulaszek, C. A. Kaysner, B. J. Tenge, J. L. Nordstrom, J. Wells, N. Puhr, and S. M. Gendel. 2003. Molecular, serological, and virulence characteristics of Vibrio parahaemolyticus isolated from environmental, food, and clinical sources in North America and Asia. Appl. Environ. Microbiol. 69:3999–4005.

51. DePaola, A., J. L. Jones, K. E. Noe, R. H. Byars, and J. C. Bowers. 2009. Survey of postharvest-processed oysters in the United States for levels of Vibrio vulnificus and Vibrio parahaemolyticus. J. Food Prot. 72:2110–2113. 52. DePaola, A., Jr., and C. A. Kaysner. 2004. Vibrio, chapter 9. In U.S. Food and Drug Administration (ed.), Bacteriological Analytical Manual Online. U.S. Food and Drug Administration, College Park, MD. 53. Desenclos, J.-C. A., K.C. Klontz, L. E. Wolfe, and S. Hoecherl. 1991. The risk of Vibrio illness in the Florida raw oyster eating population, 1981–1988. Am. J. Epidemiol. 134:290–297. 54. Dixon, W. D. 1992. The effects of gamma radiation (60Co) upon shellstock oysters in terms of shelf life and bacterial reduction, including Vibrio vulnificus levels. M.S. thesis. University of Florida, Gainesville, FL. 55. Dziejman, M., D. Serruto, V. C. Tam, D. Sturtevant, P. Diraphat, S. M. Faruque, M. H. Rahman, J. F. Heidelberg, J. Decker, L. Li, K. T. Montgomery, G. Grills, R. Kucherlapati, and J. J. Mekalanos. 2005. Genomic characterization of non-O1, non-O139 Vibrio cholerae reveals genes for a type III secretion system. Proc. Natl. Acad. Sci. USA 102:3465–3470. 56. Dziejman, M., E. Balon, D. Boyd, C. M. Frase, J. F. Heidelberg, and J. J. Mekalanos. 2002. Comparative genomic analysis of Vibrio cholerae: Genes that correlate with cholera endemic and pandemic disease. Proc. Natl. Acad. Sci. USA 99:1556–1561. 57. Edouard, S., A. Daumas, S. Branger, J. M. Durand, D. Raoult, and P. E. Fournier. 2009. Grimontia hollisae, a potential agent of gastroenteritis and bacteraemia in the Mediterranean area. Eur. J. Clin. Microbiol. Infect. Dis. 28:705–707. 58. Elhadi, N., S. Radu, C.-H. Chen, and M. Nishibuchi. 2004. Prevalence of potentially pathogenic Vibrio species in the seafood marketed in Malaysia. J. Food. Prot. 67:1469–1475. 59. Elliot, E. L., C. A. Kaysner, L. Jackson, and M. L. Tamplin. 1995. Vibrio cholerae, V. parahaemolyticus, V. vulnificus and other Vibrio spp, p. 9.01–9.27. In FDA Bacteriological Analytical Manual, 8th ed. AOAC International, Gaithersburg, MD. 60. Espiñeira, M., M. Atanassova, J. M. Vieites, and F. J. Santaclara. 2010. Validation of a method for the detection of five species, serogroups, biotypes and virulence factors of Vibrio by multiplex PCR in fish and seafood. Food Microbiol. 27:122–131. 61. Farmer, J. J., III, F. W. Hickman-Brenner, and M. T. Kelly. 1985. Vibrio, p. 282–301. In Manual of Clinical Microbiology, 4th ed. American Society for Microbiology, Washington, DC. 62. Reference deleted. 63. Faruque, S. M., I. B. Naser, M. J. Islam, A. S. Faruque, A. N. Ghosh, G. B. Nair, D. A. Sack, and J. J. Mekalanos. 2005. Seasonal epidemics of cholera inversely correlate with the prevalence of environmental cholera phages. Proc. Natl. Acad. Sci. USA 102:1702–1707. 64. Faruque, S. M., K. Biswas, S. M. Nashir Udden, Q. S. Ahmad, D. A. Sack, G. B. Nair, and J. J. Mekalanos.

16.  Vibrio Species

65.

66.

67.

68. 69.

70. 71.

72.

73.

74.

75.

76.

77.

2006. Transmissibility of cholera: in vivo formed biofilms and their relationship to infectivity and persistence in the environment. Proc. Natl Acad. Sci. USA 103:6350–6355. Fasano, A., B. Baudry, D. W. Pumplin, S. S. Wasserman, B. D. Tall, J. M. Ketley, and J. B. Kaper. 1991. Vibrio cholerae produces a second enterotoxin, which affects intestinal tight junctions. Proc. Natl. Acad. Sci. USA 88:5242–5246. FDA. 2005. Quantitative Risk Assessment on the Public Health Impact of Pathogenic Vibrio parahaemolyticus in Raw Oysters. Food and Drug Administration, College Park, MD. Fedio, W., G. M. Blackstone, L. Kikuta-Oshima, C. Wendakoon, T. H. McGrath, and A. DePaola. 2007. Rapid detection of the Vibrio cholerae ctx gene in food enrichments using real-time polymerase chain reaction. J. AOAC. Int. 90:1278–1283. Feldhusen, F. 2000. The role of seafood in bacterial foodborne diseases. Microbes Infect. 2:1651–1660. Fullner, K. J. 2003. Toxins of Vibrio cholerae: consensus and controversy, p. 481–502. In G. A. Hecht (ed.), Microbial Pathogenesis and the Intestinal Epithelial Cell. ASM Press, Washington, DC. Fullner, K. J., and J. J. Mekalanos. 2000. In vivo covalent cross-linking of cellular actin by the Vibrio chol­ erae RTX toxin. EMBO J. 19:5315–5323. Fullner, K. J., J. C. Boucher, M. A. Hanes, G. K. Haines III, B. M. Meehan, C. Walchle, P. J. Sansonetti, and J. J. Mekalanos. 2002. The contribution of accessory toxins of Vibrio cholerae O1 El Tor to the proinflammatory response in a murine pulmonary cholera model. J. Exp. Med. 195:1455–1462. Garcia, K., R. Torres, P. Uribe, C. Hernandez, M.L. Rioseco, J. Romero, and R.T. Espejo. 2009. Dynamics of clinical and environmental Vibrio parahaemolyticus strains during seafood-related summer diarrhea outbreaks in Southern Chile. Appl. Environ. Microbiol. 75:7482–7487. Goel, A. K., M. Jain, P. Kumar, S. Bhadauria, D. V. Kmboj, and L. Singh. 2008. A new variant of Vibrio cholerae O1 El Tor causing cholera in India. J. Infect. 57:280–281. Gopal, S., S. K. Otta, S. Kumar, I. Karunasagar, M. Nishibuchi, and I. Karunasagar. 2005. The occurrence of Vibrio species in tropical shrimp culture environments; implications for food safety. Int. J. Food Microbiol. 102:151–159. Gras-Rouzet, S., P. Y. Donnio, F. Juguet, P. Plessis, J. Minet, and J. L. Avril. 1996. First European case of gastroenteritis and bacteremia due to Vibrio hollisae. Eur. J. Clin. Microbiol. Infect. Dis. 15:864–866. Grau, B. L., M. C. Henk, and G. S. Pettis. 2005. Highfrequency phase variation of Vibrio vulnificus 1003: isolation and characterization of a rugose phenotypic variant. J. Bacteriol. 187:2519–2525. Grim, C. J., N. A. Hasan, E. Taviani, B. Haley, J. Chun, T. S. Brettin, D.C. Bruce, J. C. Detter, C. S. Han, O. Chertkov, J. Challacombe, A. Huq, G. B. Nair, and

433

78. 79. 80.

81.

82. 83.

84.

85.

86.

87.

88.

89. 90.

91.

R. R. Colwell. 2010. Genome sequence of hybrid Vibrio cholerae O1 MJ-1236, B-33, and CIRS101 and comparative genomics with V. cholerae. J. Bacteriol. 192:3524–3533. Gubala, A. J., and D. F. Proll. 2006. Molecular-beacon multiplex real-time PCR assay for detection of Vibrio cholerae. Appl. Environ. Microbiol. 72:6424–6428. Gulig, P. A., K. L. Bourdage, and A. M. Starks. 2005. Molecular pathogenesis of Vibrio vulnificus. J. Microbiol. 43:118–131. Gutacker, M., N. Conza, C. Benagli, A. Pedroli, M. V. Bernasconi, L. Permin, R. Aznar, and J.-C. Piffaretti. 2003. Population genetics of Vibrio vulnificus: identification of two divisions and a distinct eel-pathogenic clone. Appl. Environ. Microbiol. 69:3203–3212. Hackney, C. R., B. Ray, and M. L. Speck. 1980. Incidence of Vibrio parahaemolyticus in and the microbiological quality of seafood in North Carolina. J. Food Prot. 43:769–773. Hammer, B. K., and B. L. Bassler. 2003. Quorum sensing controls biofilm formation in Vibrio cholerae. Mol. Microbiol. 50:101–104. Han, H.-H., J.-H. Lee, Y.-H. Choi, J.-H. Park, T.-J. Choi, and I.-S. Kong. 2002. Purification, characterization and molecular cloning of Vibrio fluvialis hemolysin. Biochim. Biophys. Acta 1599:106–114. Hara-Kudo, Y., T. Nishina, H. Nakagawa, H. Konuma, J. Hasegawa, and S. Kumagai. 2001. Improved method for detection of Vibrio parahaemolyticus in seafood. Appl. Environ. Microbiol. 67:5819–5823. Hardy, S. P., M. Nakano, and T. Iida. 2004. Single channel evidence for innate pore-formation by Vibrio parahaemolyticus thermostable direct haemolysin (TDH) in phospholipid bilayers. FEMS Microbiol. Lett. 240:81–85. Harrison, L. M., P. Rallabhandi, J. Michalski, X. Zhou, S. R. Steyert, S. N. Vogel, and J. B. Kaper. 2008. Vibrio cholerae flagellins induce toll-like receptor 5-mediated interleukin-8 production through mitogen-activated protein kinase and NF-kappa B activation. Infect. Immun. 76:5524–5534. Harwood, V. J., J. P. Gandhi, and A. C. Wright. 2004. Methods for isolation and confirmation of Vibrio vulni­ ficus from oysters and environmental sources: a review. J. Microbiol. Methods 59:301–316. Hayat, U., G. P. Reddy, C. A. Bush, J. A. Johnson, A. C. Wright, and J. G. Morris, Jr. 1993. Capsular types of Vibrio vulnificus: an analysis of strains from clinical and environmental sources. J. Infect. Dis. 168:758–762. Henke, J. M., and B. L. Bassler. 2004. Quorum sensing regulates type III secretion in Vibrio harveyi and Vibrio parahaemolyticus. J. Bacteriol. 186:3794–3805. Hickman, F. W., J. J. Farmer III, D. G. Hollis, G. R. Fanning, A. G. Steigerwalt, R. E. Weaver, and D. J. Brenner. 1982. Identification of Vibrio hollisae sp. nov. from patients with diarrhea. J. Clin. Microbiol. 15:395–401. Hillman, C. 2000. Low temperature pasteurization to reduce the risk of Vibrio infections in raw-shellstock

Foodborne Pathogenic Bacteria

434

92.

93. 94. 95.

96.

97.

98.

99. 100.

101.

102.

103. 104. 105.

106.

oysters. Proc. 25th Annu. Meet. Seafood Sci. Technol. Soc. Longboat Key, FL. Hinestrosa, F., R. G. Madeira, and P. P. Bourbeau. 2007. Severe gastroenteritis and hypovolemic shock caused by Grimontia (Vibrio) hollisae infection. J. Clin. Microbiol. 45:3462–3463. Hlady, W. G. 1997. Vibrio infections associated with raw oyster consumption in Florida, 1981–94. J. Food Prot. 60:353–357. Hlady, W. G., and K. C. Klontz. 1996. The epidemiology of Vibrio infections in Florida, 1981–1993. J. Infect. Dis. 173:1176–1183. Hlady, W. G., R. C. Mullen, and R. S. Hopkin. 1993. Vibrio vulnificus from raw oysters. Leading cause of reported deaths from foodborne illness in Florida. J. Fla. Med. Assoc. 80:536–538. Hoi, L., I. Dalsgaard, and A. Dalsgaard. 1998. Improved isolation of Vibrio vulnificus from seawater and sediment with cellobiose-colistin agar. Appl. Environ. Microbiol. 64:1721–1724. Huq, A., E. B. Small, P. A. West, M. I. Huq, R. Rahman, and R. R. Colwell. 1983. Ecological relationships between Vibrio cholerae and planktonic crustacean copepods. Appl. Environ. Microbiol. 45:275–283. Huq, A., C. A. Whitehouse, C. J. Grim, M. Alam, and R. R. Colwell. 2008. Biofilms in water, its role and impact in human disease transmission. Curr. Opin. Biotechnol. 19:244–247. Huq, M. I., A. K. M. J. Alam, D. J. Brenner, and G. K. Morris. 1980. Isolation of Vibrio-like group, EF-6, from patients with diarrhea. J. Clin. Microbiol. 11:621–624. Igbinosa, E. O., and A. I. Okoh. 2010. Vibrio fluvia­ lis: an unusual enteric pathogen of increasing public health concern. Int. J. Environ. Res. Public Health 7:3628–3643. Iwamoto, M., T. Ayers, B. E. Mahon, and D. L. Swerdlow. 2010. Epidemiology of seafood-associated infections in the United States. Clin. Microbiol. Rev. 23:399–411. Jeong, K. C., H. S. Jeong, J. H. Rhee, S. E. Lee, S. S. Chung, A. M. Starks, G. M. Escudero, P. A. Gulig, and S. H. Choi. 2000. Construction and phenotypic evaluation of a Vibrio vulnificus vvpE mutant for elastolytic protease. Infect. Immun. 68:5096–5106. Johnston, J. M., S. F. Becker, and L. M. McFarland. 1986. Gastroenteritis in patients with stool isolations of Vibrio vulnificus. Am. J. Med. 80:336–338. Jones, M. K., and J. D. Oliver. 2009. Vibrio vul­ nificus: disease and pathogenesis. Infect. Immun. 77:1723–1733. Jones, S. H., T. L. Howell, and K. R. O’Neill. 1991. Differential elimination of indicator bacteria and pathogenic Vibrio sp. from Eastern oysters (Crassostrea virgi­ nica gmelin, 1791) in a commercial purification facility in Maine. J. Shellfish Res. 10:105–112. Joseph, S. W., R. R. Colwell, and J. B. Kaper. 1982. Vibrio parahaemolyticus and related halophilic vibrios. CRC Crit. Rev. Microbiol. 10:77–124.

107. Kaneko, T., and R. R. Colwell. 1978. The annual cycle of Vibrio parahaemolyticus in Chesapeake Bay. Microb. Ecol. 4:135–155. 108. Kaper, J. B., and V. Sperandio. 2005. Bacterial cellto-cell signaling in the gastrointestinal tract. Infect. Immun. 73:3197–3209. 109. Kaper, J. B., J. G. Morris, Jr., and M. M. Levine. 1995. Cholera. Clin. Microbiol. Rev. 8:48–86. 110. Karaolis, D. K. R., J. A. Johnson, C. C. Bailey, E. C. Boedeker, J. B. Kaper, and P. R. Reeves. 1998. A Vibrio cholerae pathogenicity island associated with epidemic and pandemic strains. Proc. Natl. Acad. Sci. USA 95:3134–3139. 111. Kaysner, C. A., and A. DePaola, Jr. 2004. Vibrio. In Bacteriological Analytical Manual Online, chapt. 9. U.S. Food and Drug Administration, Rockville, MD. http://www.cfsan.fda.gov/~ebam/bam-9.html. 112. Kim, R. K., S. E. Lee, C. M. Kim, S. Y. Kim, E. K. Shin, D. H. Shin, S. S. Chung, H. E. Choy, A. Progulske-Fox, J. D. Hillman, M. Handfield, and J. H. Rhee. 2003. Characterization and pathogenic significance of Vibrio vulnificus antigens preferentially expressed in septicemic patients. Infect. Immun. 71:5461–5471. 113. Kim, Y. R., S. Y. Kim, C. M. Kim, S. E. Lee, and J. H. Rhee. 2005. Essential role of an adenylate cyclase in regulating Vibrio vulnificus virulence. FEMS Microbiol. Lett. 243:497–503. 114. Kim, Y. R., S. E. Lee, H. Kook, J. A. Yeom, H. S. Nu, S. Y. Kim, S. S. Chung, H. E. Choy, and J. H. Rhee. 2008. Vibrio vulnificus RTX toxin kills host cells only after contact of the bacteria with host cells. Cell Microbiol. 10:848–862. 115. Kirn, T. J., B. A. Jude, and R. K. Taylor. 2005. A colonization factor links Vibrio cholerae environmental survival and human infection. Nature 438:863–866. 116. Klontz, K. C., and J.-C. A. Desenclos. 1990. Clinical and epidemiological features of sporadic infections with Vibrio fluvialis in Florida USA. J. Diarrh. Dis. Res. 8:1–2. 117. Kotetishvili, M., O. C. Stine, Y. Chen, A. Kreger, A. Sulakvelidze, S. Sozhamannan, and J. G. Morris, Jr. 2003. Multilocus sequence typing has better discriminatory ability for typing Vibrio cholerae than does pulsed-field gel electrophoresis and provides a measure of phylogenetic relatedness. J. Clin. Microbiol. 41:2191–2196. 118. Kothary, M. H., E. F. Claverie, M. D. Miliotis, J. M. Madden, and S. H. Richardson. 1995. Purification and characterization of a Chinese hamster ovary cell elongation factor of Vibrio hollisae. Infect. Immun. 63:2418–2423. 119. Kothary, M. H., H. Lowman, B. A. McCardell, and B. D. Tall. 2003. Purification and characterization of enterotoxigenic El Tor-like hemolysin produced by Vibrio fluvialis. Infect. Immun. 71:3213–3220. 120. Krukonis, E. S., and V. J. DiRita. 2003. From motility to virulence: sensing and responding to environmental signals in Vibrio cholerae. Curr. Opin. Microbiol. 6:186–190.

16.  Vibrio Species 121. Kudryashov, D. S., C. L. Cordero, E. Reisler and K. J. Satchell. 2008. Characterization of the enzymatic activity of the actin cross-linking domain from the Vibrio cholerae MARTX Vc toxin. J. Biol. Chem. 283:445–452. 122. Kumar, P., W. A. Peter, and S. Thomas. 2010. Rapid detection of virulence-associated genes in environmental strains of Vibrio cholerae by multiplex PCR. Curr. Microbiol. 60:199–202. 123. Kural, A. G., A. E. H. Shearer, D. H. Kingsley and C. Haiqiang. 2008. Conditions for high pressure inactivation of Vibrio parahaemolyticus in oysters. Int. J. Food. Microbiol. 127:1–5. 124. Lee, J. H., J. B. Rho, K. J. Park, C. B. Kim, Y. S. Han, S. H. Choi, K. H. Lee, and S. J. Park. 2004. Role of flagellum and motility in pathogenesis of Vibrio vulnificus. Infect. Immun. 72:4905–4910. 125. Lee, J. V., P. Shread, and A. L. Furniss. 1978. The taxonomy of group F organisms: relationships to Vibrio and Aeromonas. J. Appl. Bacteriol. 45:ix. 126. Lenz, D. H., M. B. Miller, J. Zhu, R. V. Kulkarni, and B. L. Bassler. 2005 . CsrA and three redundant small RNAs regulate quorum sensing in Vibrio cholerae. Mol. Microbiol. 58:1186–1202. 127. Levine, M. M., R. E. Black, M. L. Clements, D. R. Nalin, L. Cisneros, and R. A. Finkelstein. 1981. Volunteer studies in development of vaccines against cholera and enterotoxigenic Escherichia coli: a review, p. 443–459. In T. Holme, J. Holmgren, M. H. Merson, and R. Mollby (ed.), Acute Enteric Infections in Children. New prospects for Treatment and Prevention. Elsevier/ North-Holland Biomedical Press, Amsterdam, The Netherlands. 128. Levine, W. C., P. M. Griffin, and the Gulf Coast Vibrio Working Group. 1993. Vibrio infections on the Gulf Coast: results of first year of regional surveillance. J. Infect. Dis. 167:479–483. 129. Lipp, E. K., A. Huq, and R. R. Colwell. 2002. Effects of global climate on infectious disease: the cholera model. Clin. Microbiol. Rev. 15:757–770. 130. Lockwood, D. E., A. S. Kreger, and S. H. Richardson. 1982. Detection of toxins produced by Vibrio fluvialis. Infect. Immun. 35:702–708. 131. Lowry, P. W., L. M. McFarland, and H. K. Threefoot. 1986. Vibrio hollisae septicemia after consumption of catfish. J. Infect. Dis. 154:730–731. 132. Lowry, P. W., L. M. McFarland, B. H. Peltier, N. C. Roberts, H. B. Bradford, J. L. Herndon, D. F. Stroup, J. B. Mathison, P. A. Blake, and R. A. Gunn. 1989. Vibrio gastroenteritis in Louisiana: a prospective study among attendees of a scientific congress in New Orleans. J. Infect. Dis. 160:978–984. 133. Lux, T. M., R. Lee, and J. Love. 2011. The complete genome sequence of a free-living Vibrio furnissii sp. nov. strain (NCTC 11218). J. Bacteriol. 193:1487–1488. 134. Ma, L., and Y. C. Su. 2011. Validation of high pressure processing for inactivating Vibrio parahaemolyti­ cus in Pacific oysters (Crassostrea gigas). Int. J. Food Microbiol. 144:469–474.

435 135. Mahoney, J. C., M. J. Gerding, S. H. Jones, and C. A. Whistler. 2010. Comparison of the pathogenic potentials of environmental and clinical Vibrio parahaemo­ lyticus strains indicates a role for temperature regulation in virulence. Appl. Environ. Microbiol. 76:7459–7465. 136. Mahmoud, B. S. M. 2009. Reduction of Vibrio vulni­ ficus in pure culture, half shell and whole shell oysters (Crassostrea virginica) by X-ray. Int. J. Food Microbiol. 130:135–139. 137. Mahmoud, B. S. M., and D. D. Burrage. 2009. Inactivation of Vibrio parahaemolyticus in pure culture, half shell and whole shell oysters (Crassostrea virginica) byX-ray. Lett. Appl. Microbiol. 48:572–578. 138. Makino, K., K. Oshima, K. Kurokawa, K. Yokoyama, T. Uda, K. Tagomori, Y. Iijima, M. Najima, M. Nakano, A. Yamashita, Y. Kubota, S. Kimura, T. Yasunaga, T. Honda, H. Shinagawa, M. Hattori, and T. Iida. 2003. Genome sequence of Vibrio parahaemolyticus: a pathogenic mechanism distinct from that of V. cholerae. Lancet 361:743–749. 139. Martinez-Urtaza, J., A. Lozano-Leon, A. Vina-Feas, J. de Novoa, and O. Garcia-Martin. 2006. Differences in the API 20E biochemical patterns of clinical and environmental Vibrio parahaemolyticus isolates. FEMS Microbiol. Lett. 255:75–81. 140. Martinez-Urtaza, J., J. C. Bowers, J. Trinanes, and A. DePaola. 2010. Climate anomalies and the increasing risk of Vibrio parahaemolyticus and Vibrio vulnificus illnesses. Food Res. Int. 43:1780–1790. 141. Martinez-Urtaza, J. L., D. Simental, A. Velasco, A. DePaola, M. Ishibashi, Y. Nakaguchi, M. Nishibuchi, D. Carrera-Flores, C. Rey-Alvarez, and A. Pousa. 2005. Pandemic Vibrio parahalemolyticus O3: K6, Europe. Emerg. Infect. Dis. 11:1319–1320. 142. Matté, G. R., M. H. Matté, I. G. Rivera, and M. T. Martins. 1994. Distribution of pathogenic vibrios in oysters from a tropical region. J. Food Prot. 57:870–873. 143. Matz, C., D. McDougald, A. M. Moreno, P. Y. Yung, F. H. Yildiz, and S. Kjelleberg. 2005. Biofilm formation and phenotypic variation enhance predation-driven persistence of Vibrio cholerae. Proc. Natl. Acad. Sci. USA 102:16819–16824. 144. McLaughlin, J. B., A. DePaola, C. A. Bopp, K. A. Martinek, N. P. Napolilli, C. G. Allison, S. L. Murray, E. C. Thompson, M. M. Bird, and J. P. Middaugh. 2005. Outbreak of Vibrio parahaemolyticus gastro­ enteritis associated with Alaskan oysters. N. Engl. J. Med. 353:1463–1470. 145. McLeod, S. M., H. H. Kimsey, B. M. Davis, and M. K. Waldor. 2005. CTXphi and Vibrio cholerae: exploring a newly recognized type of phage-host cell relationship. Mol. Microbiol. 57:347–356. 146. McPherson, V. L., J. A. Watts, L. M. Simpson, and J. D. Oliver. 1991. Physiological effects of the lipopolysaccharide of Vibrio vulnificus on mice and rats. Microbios 67:141–149. 147. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. B. V. Tauxe. 1999. Emerging infectious diseases (serial online) 5, 26

Foodborne Pathogenic Bacteria

436

148. 149.

150.

151.

152.

153. 154.

155. 156.

157. 158.

159.

160.

161.

screens available at URL: http://www.cdc.gov/ncidod/ edi/vol5no5/mead.htm. Meibom, K. L., M. Blokesch, N. A. Dolganov, C. Y. Wu, and G. K. Schoolnik. 2005. Chitin induces natural competence in Vibrio cholerae. Science 310:1824–1827. Merkel, S. M., S. Alexander, J. D. Oliver, and Y. M. Huet-Hudson. 2001. Essential role for estrogen in protection against Vibrio vulnificus induced endotoxic shock. Infect. Immun. 69:6119–6122. Merrell, D. S., S. M. Butler, F. Qadri, N. A. Dolganov, A. Alam, M. B. Cohen, S. B. Calderwood, G. K. Schoolnik, and A. Camilli. 2002. Host-induced epidemic spread of the cholera bacterium. Nature 417:642–645. Messelhäusser, U., J. Colditz, D. Thärigen, W. Kleih, C. Höller, and U. Busch. 2010. Detection and differentiation of Vibrio spp. in seafood and fish samples with cultural and molecular methods. Int. J. Food Microbiol. 142:360–364. Mey, A. R., E. E. Wyckoff, V. Kanukurthy, C. R. Fisher, and S. M. Payne. 2005. Iron and fur regulation in Vibrio cholerae and the role of fur in virulence. Infect. Immun. 73:8167–8178. Miliotis, M. D., B. D. Tall, and R. T. Gray. 1995. Adherence to and invasion of tissue culture cells by Vibrio hollisae. Infect. Immun. 63:4959–4963. Mintz, E. D., T. Popovic, and P. A. Blake. 1994. Transmission of Vibrio cholerae O1, p. 345–356, In I. K. Wachsmuth, P. A. Blake, and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, DC. Morris, J. G., Jr. 1990. Non-O group 1 Vibrio cholerae: a look at the epidemiology of an occasional pathogen. Epidemiol. Rev. 12:179–191. Morris, J. G., Jr. 1995. “Noncholera” Vibrio species. p. 671–685. In M. J. Blaser, P. D. Smith, J. I. Ravdin, H. B. Greenberg, and R. L. Guerrant (ed.), Infections of the Gastrointestinal Tract. M. Raven Press, Ltd., New York, NY. Morris, J. G., Jr. 2003. Cholera and other types of vibriosis: a story of human pandemics and oysters on the half shell. Clin. Infect. Dis. 37:272–280. Morris, G. J., R. Wilson, D. G. Hollis, R. E. Weaver, H. G. Miller, C. O. Tacket, F. W. Hickman, and P. A. Blake. 1982. Illness caused by Vibrio damsela and Vibrio hollisae. Lancet i:1294–1296. Morris, J. G., Jr., A. C. Wright, D. M. Roberts, P. K. Wood, L. M. Simspon, and J. D. Oliver. 1987. Identification of environmental Vibrio vulnificus isolates with a DNA probe for the cytotoxin-hemolysin gene. Appl. Environ. Microbiol. 53:193–195. Motes, M., A. DePaola, S. Zywno-Van Ginkel, and M. McPhearson. 1994. Occurrence of toxigenic Vibrio cholerae O1 in oysters in Mobile bay, Alabama: an ecological investigation. J. Food Prot. 57:975–980. Nair, G. B., T. Ramamurthy, S. K. Bhattacharya, B. Dutta, Y. Takeda, and D. A. Sack. 2007. Global dissemination of Vibrio parahaemolyticus serotype O3:K6 and its serovariants. Clin. Microbiol. Rev. 20:39–48.

162. Nascumento, D. R., R. H. Vieira, H. B. Almeida, T. R. Patel, and S. T. Iaria. 1998. Survival of Vibrio cholerae O1 strains in shrimp subjected to freezing and boiling. J. Food Prot. 61:1317–1320. 163. Nelson, E. J., J. B. Harris, J. G. Morris, Jr., S. B. Calderwood, and A. Camilli. 2009. Cholera transmission: the host, pathogen and bacteriophage dynamic. Nat. Rev. Microbiol. 7:693–702. 164. Nesper, J., S. Schild, C. M. Lauriano, A. Kraiss, K. E. Klose, and J. Reidl. 2002. Role of Vibrio cholerae O139 surface polysaccharides in intestinal colonization. Infect. Immun. 70:5990–5996. 165. Nishibuchi, M., and J. B. Kaper. 1995. Thermostable direct hemolysin gene of V. parahaemolyticus: a virulence gene acquired by a marine bacterium. Infect. Immun. 63:2093–2099. 166. Nishibuchi, M., and A. DePaola. 2005. Vibrio species, p. 251–271. In P. M. Fratamico, A. K. Bhunia, and J. L. Smith (ed.), Food-Borne Pathogens: Microbiology and Molecular Biology. Caister Academic Press, Norfolk, United Kingdom. 167. Nishibuchi, M., A. Fasano, R. G. Russell, and J. B. Kaper. 1992. Enterotoxigenicity of Vibrio parahaemo­ lyticus with and without genes encoding thermostable direct hemolysin. Infect. Immun. 60:3539–3545. 168. Nishibuchi, M., T. Taniguchi, T. Misawa, V. Khaeomanee-iam, T. Honda, and T. Miwatani. 1989. Cloning and nucleotide sequence of the gene (trh) encoding the hemolysin related to the thermostable direct hemolysin of Vibrio parahaemolyticus. Infect. Immun. 57:2691–2697. 169. O’Hara, C. M., E. G. Sowers, C. A. Bopp, S. B. Duda, and N. A. Strockbine. 2003. Accuracy of six commercially available systems for identification of members of the family Vibrionaceae. J. Clin. Microbiol. 41:5654–5659. 170. Okeke, I. N., J. Eardley, C. C. Bailey, and J. B. Kaper. 2001. Vibrio cholerae, p. 1191–1236. In M. Sussman (ed.), Molecular Medical Microbiology. Academic Press, London, United Kingdom. 171. Oliver, J. D. 1989. Vibrio vulnificus, p. 569–600. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY. 172. Oliver, J. D. 2000. Public health significance of viable but nonculturable bacteria, p. 277–300. In R. R. Colwell and D. J. Grimes (ed.), Nonculturable Microorganisms in the Environment. ASM Press, Washington, DC. 173. Oliver, J. D. 2000. Culture media for the isolation and enumeration of pathogenic Vibrio species in foods and environmental samples, p. 377–402. In J. E. L. Corry, G. D. W. Curtis, and R. M. Baird (ed.), Handbook of Culture Media for Food Microbiology, 3rd ed. Royal Society of Chemistry, Cambridge, United Kingdom. 174. Oliver, J. D. 2005. Viable but nonculturable bacteria in food environments. In P. M. Fratamico, A. K. Bhunia, and J. L. Smith (ed.), Food-Borne Pathogens: Microbiology and Molecular Biology. Caister Academic Press, Norfolk, United Kingdom.

16.  Vibrio Species 175. Oliver, J. D. 2005. Vibrio vulnificus, p. 253–276. In S. Belkin and R. R. Colwell (ed.), Oceans and Health: Pathogens in the Marine Environment. Springer Science, New York, NY. 176. Oliver, J. D. 2005. The viable but nonculturable state in bacteria. J. Microbiol. 43:93–100. 177. Oliver, J. D. 2005. Wound infections caused by Vibrio vulnificus and other marine bacteria. Epidemiol. Infect. 133:383–391. 178. Oliver, J. D. 2006. Vibrio vulnificus, p. 349–366. In F. L. Thompson, B. Austin, and J. Swing (ed.), Biology of Vibrios. ASM Press, Washington, DC. 179. Oliver, J. D. 2009. Recent findings on the viable but nonculturable state in pathogenic bacteria. 2009. FEMS Microbiol. Rev. 34:415–425. 180. Oliver, J. D., and J. Kaper. 2001. Vibrio species, p. 263– 300. In M. P. Doyle et al. (ed.), Food Microbiology: Fundamentals and Frontiers, 2nd ed. ASM Press. 181. Oliver, J. D., M. B. Thomas, and J. Wear. 1986. Production of extracellular enzymes and cytotoxicity by Vibrio vulnificus. Diagn. Microbiol. Infect. Dis. 5:99–111. 182. Oliver, J. D., D. M. Roberts, V. K. White, M. A. Dry, and L. M. Simpson. 1986. Bioluminescence in a strain of the human bacterial pathogen, Vibrio vulnificus. Appl. Environ. Microbiol. 52:1209–1211. 183. Ono, T., K. S. Park, M. Ueta, T. Iida, and T. Honda. 2006. Identification of proteins secreted via Vibrio para­ haemolyticus type III secretion system 1. Infect. Immun. 74:1032–1042. 184. Ottaviani, D., F. Leoni, E. Rocchegiani, C. Canonico, S. Potenziani, S. Santarelli, L. Masini, S. Scuota, and A. Carraturo. 2010. Vibrio parahaemolyticus-associated gastroenteritis in Italy: persistent occurrence of O3:K6 pandemic clone and emergence of O1:KUT serotype. Diagn. Microbiol. Infect. Dis. 66:452–455. 185. Panicker, G., and A. K. Bej. 2005. Real-time PCR detection of Vibrio vulnificus in oysters: comparison of oligonucleotide primers and probes targeting vvhA. Appl. Environ. Microbiol. 71:5702–5709. 186. Paranjpye, R. N., and M. S. Strom. 2005. A Vibrio vulnificus type IV pilin contributes to biofilm formation, adherence to epithelial cells, and virulence. Infect. Immun. 73:1411–1422. 187. Park, K. S., T. Iida, Y. Yamaichi, T. Oyagi, K. Yamamoto, and T. Honda. 2000. Genetic characterization of DNA region containing the trh and ure genes of Vibrio para­ haemolyticus. Infect. Immun. 68:5742–5748. 188. Park, K. S., T. Ono, M. Rokuda, M. H. Jang, K. Okada, T. Iida, and T. Honda. 2004. Functional characterization of two type III secretion systems of Vibrio parahae­ molyticus. Infect. Immun. 72:6659–6665. 189. Parker, R. W., E. M. Maurer, A. B. Childers, and D. H. Lewis. 1994. Effect of frozen storage and vacuum-packaging on survival of Vibrio vulnificus in Gulf Coast oysters (Crassostrea virginica). J. Food Prot. 57:604–606. 190. Pfeffer, C. S., M. F. Hite, and J. D. Oliver. 2003. The ecology of Vibrio vulnificus in estuarine waters of

437

191.

192. 193. 194.

195.

196.

197.

198. 199.

200. 201. 202. 203. 204.

205.

eastern North Carolina. Appl. Environ. Microbiol. 69:3526–3531. Pruzzo, C., A. Huq, R. R. Colwell, and G. Donelli. 2005. Pathogenic Vibrio species in marine and estuarine environment, p. 217–252. In R. Colwell and S. Belkin (ed.), Oceans and Health: Pathogens in the Marine Environment. Kluwer Academic/Plenum Publishers, Dordrecht, The Netherlands. Pruzzo, C., L. Vezzulli, and R. R. Colwell. 2008. Global impact of Vibrio cholerae interactions with chitin. Environ. Microbiol., 10:1400–1410. Rabbani, G. H., and W. B. Greenough III. 1999. Food as a vehicle of transmission of cholera. J. Diarrh. Dis. Res. 17:1–9. Raghunath, P., I., Karunasagar, and I. Karunasagar. 2009. Improved isolation and detection of pathogenic Vibrio parahaemolyticus from seafood using a new enrichment broth. Int. J. Food Microbiol. 129:200–203. Raimondi, D., J. P. Y. Kao, J. B. Kaper, S. Guandalini, and A. Fasano. 1995. Calcium-dependent intestinal chloride secretion by Vibrio parahaemolyticus thermostable direct hemolysin in a rabbit model. Gastroenterology 109:381–386. Rashid, H. O., H. Ito, and I. Ishigaki. 1992. Distribution of pathogenic vibrios and other bacteria in imported frozen shrimps and their decontamination by gammairradiation. World J. Microbiol. Biotechnol. 8:494–499. Raychoudhuri, A., A. K. Mukhopadhyay, T. Ramamurthy, R. K. Nandy, Y. Takeda, and G. B. Nair. 2008. Biotyping of Vibrio cholerae O1: time to redefine the scheme. Indian J. Med. Res. 128:695–698. Reidl, J., and K. E. Klose. 2002. Vibrio cholerae and cholera: out of the water and into the host. FEMS Microbiol. Rev. 26:125–139. Rosche, T. M., Y. Yano, and J. D. Oliver. 2005. A rapid and simple PCR analysis indicates there are two subgroups of Vibrio vulnificus which correlate with clinical or environmental isolation. Microbiol. Immunol. 49:381–389. Rosche, T. M., B. Smith, and J. D. Oliver. 2006. Evidence for an intermediate colony morphology of Vibrio vulni­ ficus. Appl. Environ. Microbiol. 72:6430–6430. Rosche, T. M., E. A. Binder, and J. D. Oliver. 2010. Vibrio vulnificus genome suggests two distinct ecotypes. Environ. Microbiol. Rep. 2:128–132. Sack, D. A., R. B. Sack, G. B. Nair, and A. K. Siddique. 2004. Cholera. Lancet 17:223–233. Safa, A., G. B. Nair, and R. Y. Kong. 2010. Evolution of new variants of Vibrio cholerae O1. Trends Microbiol. 18:46–54. Sakazaki, R., K. Tamura, T. Kato, Y. Obara, S. Yamai, and K. Hobo. 1968. Studies on the enteropathogenic, facultatively halophilic bacteria, Vibrio parahaemo­ lyticus. III. Enteropathogenicity. Jpn. J. Med. Sci. Biol. 21:325–331. Sanchez, J., and J. Holmgren. 2008. Cholera toxin structure, gene regulation and pathophysiological and immuno­ logical aspects. Cell. Mol. Life Sci. 65:1347–1360.

Foodborne Pathogenic Bacteria

438 206. Sang, F. C., M. E. Hugh-Jones, and H. V. Hagstad. 1987. Viability of Vibrio cholerae 01 on frog legs under frozen and refrigerated conditions and low dose radiation treatment. J. Food Prot. 50:662–664. 207. Sanjuán, E., J. D. Oliver, and C. Amaro. 2009. Evaluation of genotypic and phenotypic methods to distinguish clinical from environmental Vibrio vulnificus strains. Appl. Environ. Microbiol. 75:1594–1598. 208. Sanyal, S. C., and P. C. Sen. 1974. Human volunteer study on the pathogenicity of Vibrio parahaemolyticus, p. 227–230. In T. Fujino, G. Sakaguchi, R. Sakazaki, and Y. Takeda (ed.), International Symposium on Vibrio parahaemolyticus. Saikon Publishing Co., Ltd., Tokyo, Japan. 209. Schadt, E. E., M. D. Linderman, J. Sorenson, L. Lee, and G. P. Nolan. 2010. Computational solutions to large-scale data management and analysis. Nat. Rev. Genet. 11:647–657. 210. Schmidt, E., S. M. Kelly, and C. F. van der Walle. 2007. Tight junction modulation and biochemical characterisation of the zonula occludens toxin C- and N-termini. FEBS Lett. 581:2974–2980. 211. Sears, C. L., and J. B. Kaper. 1996. Enteric bacterial toxins: mechanisms of action and linkage to intestinal secretion. Microbiol. Rev. 60:167–215. 212. Shao, C.-P., and L.-I. Hor. 2000. Metalloprotease is not essential for Vibrio vulnificus virulence in mice. Infect. Immun. 68:3569–3573. 213. Shapiro, R. L., S. Altekruse, L. Hutwagner, R. Bishop, R. Hammond, S. Wilson, B. Ray, S. Thompson, R. V. Tauxe, and P. M. Griffin. 1998. The role of Gulf Coast oysters harvested in warmer months in Vibrio vulnificus infections in the United States, 1988–1996. J. Infect. Dis. 178:752–759. 214. Shirai, H., H. Ito, T. Kirayama, Y. Nakamoto, N. Nakabayashi, K. Kumagni, Y. Takeda, and M. Nishibuchi. 1990. Molecular epidemiologic evidence for association of thermostable direct hemolysin (TDH) and TDH-related hemolysin of Vibrio para­ haemolyticus with gastroenteritis. Infect. Immun. 58:3568–3573. 215. Simonson, J. G., P. Danieu, A. B. Zuppardo, R. J. Siebeling, R. L. Murphree, and M. L. Tamplin. 1995. Distribution of capsular and lipopolysaccharide antigens among clinical and environmental Vibrio vulnificus isolates. Abstr. B-286, p. 215. Abstr. Annu. Meet. Am. Soc. Microbiol. American Society for Microbiology, Washington, DC. 216. Simpson, L. M., V. K. White, S. F. Zane, and J. D. Oliver. 1987. Correlation between virulence and colony morphology in Vibrio vulnificus. Infect. Immun. 55:269–272. 217. Sloan, E. M., C. J. Hagen, G. A. Lancette, J. T. Peeler, and J. N. Sofos. 1992. Comparison of five selective enrichment broths and two selective agars for recovery of Vibrio vulnificus from oysters. J. Food Prot. 55:356–359. 218. Staley, C., and V. J. Harwood. 2010. The use of genetic typing methods to discriminate among strains of Vibrio

219. 220. 221. 222. 223.

224.

225.

226.

227.

cholerae, V. parahaemolyticus, and V. vulnificus. J. AOAC Int. 93:1553–1569. Strom, M. S., and R. N. Paranjpye. 2000. Epidemiology and pathogenesis of Vibrio vulnificus. Microbes Infect. 2:177–188. Sun, Y., and J. D. Oliver. 1994. Effects of GRAS compounds on natural Vibrio vulnificus populations in oysters. J. Food Prot. 57:921–923. Sun, Y., and J. D. Oliver. 1995. The value of CPC agar for the isolation of Vibrio vulnificus from oysters. J. Food Prot. 58:439–440. Suzita, R., A. S. Abdulamir, F. A. bu Bakar, and R. Son. 2009. Cholera outbreak via shellfish. Am. J. Infect. Dis. 5:40–47. Suzuki, K., T. Tababe, Y. H. Moon, T. Funahashi, H. Nakao, S. Narimatsu, and S. Yamamoto. 2006. Identification and transcriptional organization of aerobactin transport and biosynthesis cluster genes of Vibrio hollisae. Res. Microbiol. 157:730–740. Takahashi, A., Y. Sato, Y. Shiomi, V. V. Cantarelli, T. Iida, M. Lee, and T. Honda. 2000. Mechanisms of chloride secretion induced by thermostable direct haemolysin of Vibrio parahaemolyticus in human colonic tissue and a human intestinal epithelial cell line. J. Med. Microbiol. 49:801–810. Tamplin, M. L., and G. M. Capers. 1992. Persistence of Vibrio vulnificus in tissues of Gulf Coast oysters, Crassostrea virginica, exposed to seawater disinfected with UV light. Appl. Environ. Microbiol. 58:1506–1510. Tarr, C. L., J. S. Patel, N. D. Puhr, E. G. Sowers, C. A. Bopp, and N. A. Strockbine. 2007. Identification of Vibrio isolates by a multiplex PCR assay and rpoB sequence determination. J. Clin. Microbiol. 45:134–140. Taylor, R. K., V. L. Miller, D. B. Furlong, and J. J. Mekalanos. 1987. Use of phoA gene fusions to identify a pilus colonization factor coordinately regulated with cholera toxin. Proc. Natl. Acad. Sci. USA 84:2833–2837.

228. Teplitski, M., A. C. Wright, and G. Lorca. 2009. Biological approaches for controlling shellfish-­associated pathogens. Curr. Opin. Biotechnol. 20:185–190. 229. Thompson, F. L., B. Hoste, K. Vandemeulebroecke, and J. Swings. 2003. Reclassification of Vibrio hollisae as Grimontia hollisae gen. nov., comb. nov. Int. J. Syst. Evol. Microbiol. 53:1615–1617. 230. Thompson, F. L., T. Iida, and J. Swings. 2004. Biodiversity of vibrios. Microbiol. Mol. Biol. Rev. 68:403–431. 231. Tobin-D’Angelo, M., A. R. Smith, S. N. Bulens, S. Thomas, M. Hodel, H. Izumiya H, E. Arakawa, M. Morita, H. Watanabe, C. Marin, M. B. Parsons, K. Greene, K. Cooper, D. Haydel, C. Bopp, P. Yu, and E. Mintz. 2008. Severe diarrhea caused by cholera toxinproducing Vibrio cholerae serogroup O75 infections acquired in the southeastern United States. Clin. Infect. Dis. 47:1035–1040.

16.  Vibrio Species 232. Todd, E. C. D. 1989. Preliminary estimates of costs of foodborne disease in the United States. J. Food Prot. 52:595–601. 233. Trucksis, M., J. E. Galen, J. Michalski, A. Fasano, and J. B. Kap5er. 1993. Accessory cholera enterotoxin (Ace), the third toxin of a Vibrio cholerae virulence cassette. Proc. Natl. Acad. Sci. USA 90:5267–5271. 234. Tsai, Y. H., R. W. W. Hsu, and K. C. Huang. 2004. Systemic Vibrio infection presenting as necrotising fasciitis and sepsis—a series of thirteen cases. J. Bone Joint Surg. 86:2497–2502. 235. Turner, J. W., B. Good, D. Cole, and E. K. Lipp. 2009. Plankton composition and environmental factors contribute to Vibrio seasonality. ISME J. 3:1082–1092. 236. Twedt, R. M. 1989. Vibrio parahaemolyticus, p. 543– 568. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY. 237. Vanden Broeck D., C. Horvath, and M. J. S. De Wolf. 2007. Pathogens in focus Vibrio cholerae: cholera toxin. Int. J. Biochem. Cell Biol. 39:1771–1775. 238. Vanderzant, C., and R. Nickelson. 1972. Survival of Vibrio parahaemolyticus in shrimp tissue under various environmental conditions. Appl. Microbiol. 23:34–37. 239. Vezzulli, L., C. A. Guzmán, R. R. Colwell, and C. Pruzzo. 2008. Dual role colonization factors connecting Vibrio cholerae’s lifestyles in human and aquatic environments open new perspectives for combating infectious diseases. Curr. Opin. Biotechnol. 19:254–259. 240. Vezzulli, L., C. Pruzzo, A. Huq, and R. R. Colwell. 2010. Environmental reservoirs of Vibrio cholerae and their role in cholera. Environ. Microbiol. Rep. 2:27–33. 241. Vora, G. J., C. E. Meador, M. M. Bird, C. A. Bopp, J. D. Andreadis, and D. A. Stenger. 2005. Microarraybased detection of genetic heterogeneity, antimicrobial resistance, and the viable but nonculturable state in human pathogenic Vibrio spp. Proc. Natl. Acad. Sci. USA 102:19109–19114. 242. Warner, E., and J. D. Oliver. 2007. Refined medium for direct isolation of Vibrio vulnificus from oyster tissue and sea water. Appl. Environ. Microbiol. 73:3098–3100. 243. Warner, E. B., and J. D. Oliver. 2008. Population structure of two genotypes of Vibrio vulnificus in oysters (Crassostrea virginica) and sea water. Appl. Environ. Microbiol. 74:80–85. 244. Warner, E. B., and J. D. Oliver. 2009. Multiplex PCR assay for detection and simultaneous differentiation of genotypes of Vibrio vulnificus biotype 1. Foodborne Pathog. Dis. 5:691–693. 245. Warner, J. M., and J. D. Oliver. 1999. Randomly amplified polymorphic DNA analysis of clinical and environmental isolates of Vibrio vulnificus and other Vibrio species. Appl. Environ. Microbiol. 65:1141–1144. 246. Watnick, P. I., and R. Kolter. 1999. Steps in the development of a Vibrio cholerae El Tor biofilm. Mol. Microbiol. 34:586–595. 247. World Health Organization. 2006. Guidelines for Drinking-Water Quality Incorporating First Addendum. http://www.who.int/water_sanitation_health/. Accessed 17 August 2009.

439 248. Williams, T. L., S. M. Musser, J. L. Nordstrom, A. DePaola, and S. R. Monday. 2004. Identification of a protein biomarker unique to the pandemic O3:K6 clone of Vibrio parahaemolyticus. J. Clin. Microbiol. 42:1657–1665. 249. Wong, H.-C., S.-H. Ting, and W.-R. Shieh. 1992. Incidence of toxigenic vibrios in foods available in Taiwan. J. Appl. Bacteriol. 73:197–202. 250. Wong, H.-C., L.-L. Chen, and C.-M. Yu. 1994. Survival of psychrotrophic Vibrio mimicus, Vibrio fluvialis and Vibrio parahaemolyticus in culture broth at low temperatures. J. Food Prot. 57:607–610. 251. Wright, A. C., L. M. Simpson, and J. D. Oliver. 1981. Role of iron in the pathogenesis of Vibrio vulnificus infections. Infect. Immun. 34:503–507. 252. Wright, A. C., J. L. Powell, M. K. Tanner, L. A. Ensor, A. B. Karpas, J. G. Morris, Jr., and M. B. Sztein. 1999. Differential expression of Vibrio vulnificus capsular polysaccharide. Infect. Immun. 67:2250–2257. 253. Wright A. C., V. Garrido, G. Debuex, M. Farrell-Evans, A. A. Mudbidri, and W. S. Otwell. 2007. Evaluation of post-harvest processed oysters using PCR-based most probable number for Vibrio vulnificus. Appl. Environ. Microbiol. 73:7477–7481. 254. Wright, A. C., M. D. Danyluka, and W. S. Otwella. 2009. Pathogens in raw foods: what the salad bar can learn from the raw bar. Curr. Opin. Biotechnol. 20:172–177. 255. Wu, T. K., Y. K. Wang, Y. C. Chen, J. M. Feng, Y. H. Liu, and T. Y. Wang. 2007. Identification of a Vibrio furnissii oligopeptide permease and characterization of its in vitro hemolytic activity. J. Bacteriol. 189:8215–8223. 256. Wu, Z., P. Nbom, and K.-E. Magnusson. 2000. Distinct effects of Vibrio cholerae haemagglutinin/protease on the structure and localization of the tight junction­associated proteins occludin and ZO-1. Cell. Microbiol. 2:11–17. 257. Yildiz, F. H., X. S. Liu, A. Heydorn, and G. K. Schoolnik. 2004. Molecular analysis of rugosity in a Vibrio cholerae O1 El Tor phase variant. Mol. Microbiol. 53:497–515. 258. Yildiz, F. H., and G. K. Schoolnik. 1999. Vibrio chol­ erae O1 El Tor: identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation. Proc. Natl. Acad. Sci. USA 96:4028–4033. 259 Zhang, X.-H., and B. Austin. 2005. Haemolysins in Vibrio species. J. Appl. Microbiol. 98:1011–1019. 260. Zhu, J., M. B. Miller, R. E. Vance, M. Dziejman, B. L. Bassler, and J. J. Mekalanos. 2002. Quorumsensing regulators control virulence gene expression in Vibrio cholerae. Proc. Natl. Acad. Sci. USA 99:3129–3134. 261. Zimmerman, A. M., A. DePaola, J. C. Bowers, J. A. Krantz, J. L. Nordstrom, C. N. Johnson, and D. J. Grimes. 2007. Variability of total and pathogenic Vibrio parahaemolyticus densities in northern Gulf of Mexico water and oysters. Appl. Environ. Microbiol. 73:7389–7596.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch17

17

Eric A. Johnson

Clostridium botulinum

Botulism is a neuroparalytic disease in humans and animals resulting from the actions of botulinum neurotoxins produced by Clostridium botulinum and rare strains of Clostridium butyricum and Clostridium baratii (19, 42, 51, 64). Botulism can be life threatening, generally due to respiratory paralysis and failure and occasionally due to secondary infections and cardiac arrest (65). Although botulism is considered an acute intoxication, the syndrome can affect all muscles in the body and the duration of paralysis can last for weeks to several months. Foodborne botulism occurs following ingestion of botulinum neurotoxin preformed in foods. Botulism can also result from the growth of and toxin production by C. botulinum in the intestine (infant botulism and adult intestinal botulism) and in wounds (wound botulism) (5, 19, 42, 121). An “undetermined classification” refers to cases of diagnosed botulism for which no plausible food vehicle or intestinal colonization by C. botulinum can be determined (5, 121). Inhalational botulism is extremely rare, but a small number of human cases have been reported, and it has been determined to occur in nonhuman primates and animal models (7, 107, 117). Intentional botulism (bioterrorism) occurred in the mid-1900s, and concerns have recently reemerged

that it could be again associated with terrorist activities (7, 17), particularly through foods (1, 142). Botulism outbreaks can have a dramatic impact on the populations in which they occur (7, 30), and outbreaks of animal botulism have periodically caused devastating losses of domestic and wild animals (30, 37, 82, 94). Since the early 1900s, botulism has been a serious concern of the food industry and regulatory agencies because of the resistance properties of the endospores of C. botulinum (chapter 3), the ability of C. botulinum to grow in many foods and form botulinum neurotoxins, the stability of botulinum neurotoxins in foods, and the severity of the disease (12, 19, 29, 40, 90). Resistant endospores produced by C. botulinum are widely distributed in soils and contaminate many foods (28, 29, 53, 54, 64, 91, 127). In improperly processed and preserved foods, the endospores can germinate and vegetative cells proliferate to form botulinum neurotoxins, which cause botulism on ingestion. Botulinum neurotoxins are the most poisonous toxins known and are highly toxic by the oral route (64, 118, 129). Consequently, a major goal of the food industry and of regulatory agencies is to prevent the occurrence of viable spores in foods, their germination, and their proliferation with ­ formation of

Eric A. Johnson, Department of Bacteriology, Botulinum Toxins Laboratory, Food Research Institute, University of Wisconsin, 1550 Linden Dr., Madison, WI 53706.

441

SMP_Food Microbiology_CH17.indd

441

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

442 botulinum neurotoxin (47, 100). Certain food regulations and industry preservation practices have been designed specifically to prevent growth and toxin formation by C. botulinum (12, 54, 64). The importance of C. botulinum and its neurotoxins in food safety has contributed to unique research approaches in food microbiology (47, 111).

HISTORICAL FEATURES OF C. BOTULINUM AND BOTULISM Although anecdotal evidence suggests that botulism was recognized as a dreaded food poisoning over 1,000 years ago during the reign of Emperor Leo VI of Byzantium (886 to 911 AD) (30), the disease in humans was first definitively described in Germany by Müller (1735–1793) and by Justinius Kerner (1786–1862) (27, 30, 62, 90). During the 1800s in Germany, raw “blood” and “liver sausages” were associated with a disease characterized by muscle paralysis and suffocation. The disease was referred to as sausage “botulus” poisoning. Kerner determined experimentally that poison developed within the sausage and that exclusion of air was required for the poison formation (62). Botulism was often referred to as Kerner’s disease following these investigations (62, 90). Subsequent to the investigations in Germany, a disease with similar symptoms was recorded in Russia and Denmark from the consumption of fish, termed “ichthyism” (30). Botulism has since been recognized as a disease that occurs worldwide from the consumption of a variety of foods containing botulinum neurotoxins (42, 47, 53, 64, 109). The highest prevalence of foodborne botulism is in Eastern European regions such as Hungary, the Republic of Georgia, and Poland (28, 29, 34, 64, 89, 135), whereas the highest global incidence of infant botulism has occurred in the United States, Argentina, Australia, Canada, and Japan (75). Although many theories were proposed for the cause of botulism, its etiology remained obscure until a series of remarkable experiments were conducted by the Belgian microbiologist Emile Pierre van Ermengem in the late 1890s (62, 139). In 1895, van Ermengem published a classic treatise describing the isolation of an anaerobic bacillus from a raw salted ham implicated in a botulism outbreak affecting 34 individuals, which occurred at a funeral wake in Belgium (139). He determined that the bacillus produced a very potent toxin that was released into the medium. van Ermengem established the etiology of botulism by isolation of “Bacillus botulinus” from the ham and the spleens and large intestines of persons who had died from the food poisoning. His success in isolating the anaerobe and determining the extracellular

SMP_Food Microbiology_CH17.indd

442

nature of the toxin was a triumph in food microbiology. In 1897, Kempner (74) subsequently determined that van Ermengem’s cultures produced a substance that on injection in an inactive form gave rise to antitoxin in the blood of goats, which prevented death in animals exposed to the toxin. This provided the first evidence that antitoxin to botulinum neurotoxin could neutralize toxicity and prevent illness and death (62). van Ermengem’s and Kempner’s landmark investigations established several principles of botulism that remain valid today and form the cornerstone for understanding and control of the disease: (i) foodborne botulism is a true toxemia caused by botulinum neurotoxins produced by C. botulinum; (ii) the toxin is produced in foods by a specific organism, “Bacillus botulinus”; (iii) the toxin is active by the oral route; (iv) the toxin is inactivated by heat and alkali but is stable under acidic conditions; (v) the toxin is not produced in food containing sufficient salt or acid; (vi) C. botulinum produces heat-resistant endospores; (vii) animals vary in their susceptibility to botulinum neurotoxins; and (viii) animals can develop immunity to the botulinum neurotoxins by exposure to inactive toxin or toxoid (62, 139). Investigation of subsequent outbreaks established that there were several types of C. botulinum. Certain of these types produce paralytic diseases similar to human botulism such as “limberneck” in chickens and wild birds, and flaccid paralysis in monkeys, lions, cats, horses, cattle, dogs, and many other animals (30, 37, 82, 127). Strains isolated from certain outbreaks were more proteolytic than the van Ermengem isolate in digesting meat and milk protein and formed a toxin that was not neutralized by antitoxin to the van Ermengem strain (62). In the mid-to-late 1900s, it was established that botulism could also result from wound and intestinal infections in humans (64, 65, 106). Since the early 1900s, seven serologically distinguishable types of botulinum neurotoxins (A to G) have been identified, although evidence has revealed that variations of the toxins (subtypes) occur within a primary serotype (4, 46, 50, 56, 128).

BACTERIAL SOURCES OF BOTULINUM NEUROTOXINS Botulinum neurotoxins are produced by a heterogeneous group of Eubacteria that differ widely in gene­ tic and metabolic characteristics (16, 35, 42, 52, 67). Known species of Clostridium that produce botulinum neurotoxins are C. botulinum (which produces serotypes A through F), C. argentinense (type G), and rare strains of C. butyricum (serotype E) and C. baratii (sero­ type F). The genus Clostridium belongs in the phylum­

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

443

Firmicutes, order Bacillales, family Clostridiaceae based primarily on the DNA sequences of genes encoding small-subunit rRNA (83). Interestingly, the Bacillales order also includes genera of food-related, non-spore-forming pathogenic and spoilage bacteria, including Staphylococcus, Listeria, Streptococcus, and Lactobacillus, and it has been hypothesized that these organisms have lost the ability to sporulate during evolution (97). Seven neurotoxin serotypes (A, B, C1, D, E, F, and G) are currently recognized and are distinguished by toxicity neutralization using serotype-specific antitoxins raised against purified neurotoxins (46, 50, 51). The neurotoxigenic clostridia are generally classified into the serotypes A through G according to the hosts for which they cause disease or in which they reside environmentally and by the genetic elements that harbor the genes encoding botulinum neurotoxins (Table 17.1). Most strains of neurotoxigenic clostridia produce only one serotype of botulinum neurotoxin, but C. botulinum strains that produce more than one serotype of toxin or that contain “silent” or unexpressed genes have been isolated with increasing frequency (41, 42, 46, 63, 64). Neurotoxin formation in C. botulinum has been reported to be unstable in certain serotypes and strains (63, 64, 118), and the genes of the neurotoxin gene complex can be associated with mobile elements in certain serotypes (42, 52, 63, 86, 87, 124, 127). Organisms resembling C. baratii and C. butyricum and producing botulinum-like neurotoxins were originally isolated from infants with botulism (42, 45, 51, 52), and neurotoxigenic C. butyricum and C. baratii have recently been isolated from wider

geographical regions and associated with foodborne as well as infant botulism (42, 45, 51, 63, 64, 137). The heterogeneous nature of the neurotoxigenic clostridia suggests that genes for botulinum neurotoxin have been laterally transferred by plasmid exchange or bacteriophage infection to distinct clostridial species throughout evolution and possibly in contemporary times, since many of the new toxigenic species have been isolated from the human intestine (62). Recently, genes encoding botulinum neurotoxins have been detected on plasmids (43, 44, 86, 87), and the horizontal gene transfer of genes encoding botulinum neurotoxins has been demonstrated (86). It seems plausible that horizontal gene transfer could lead to new bacterial strains or species that form botulinum neurotoxins. Genomic analyses have indicated that various mobile genetic elements are present in strains of C. botulinum (14, 101).

ECOLOGY OF BOTULINOGENIC CLOSTRIDIA Botulinogenic clostridia are widely dispersed in nature by virtue of their ability to form resistant endospores (28, 54, 127). They have been isolated mainly from two principal habitats: soils, including sediments within lakes and oceans, and the intestinal tracts of animals (but not healthy humans) (28, 52, 54, 64). The prevalence of C. botulinum worldwide varies according to the geographical region, which is probably related to the physical and chemical composition of the soil and the microbiota present (37, 54, 90). Spores are readily dispersed in dust and aerosols and hence frequently

Table 17.1  Primary hosts and gene location of toxin genes in Clostridium botulinum

serotypes A through G Serotype

Suspectible species

Location of neurotoxin gene

A

Humans, horses

Chromosome, large plasmids in some strains

B

Humans, horses, swine, primates

Chromosome, large plasmids in some strains

C

Birds, horses, cattle, minks, foxes, dogs, turtles

Pseudolysogenic bacteriophage, plasmid

D

Cattle

Pseudolysogenic bacteriophage, plasmid

E

Fish, waterfowl

Chromosome, large plasmids in some strains

F

Humans (rare)

Chromosome, large plasmids in some strains

G

None known

Large plasmid

SMP_Food Microbiology_CH17.indd

443

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

444 occur in many ­environments and in foods (28, 29, 52, 64). Neurotoxigenic clostridia are saprophytic and do not have an obligatory relationship with an animal host (52). Geographic regions with high prevalence of C. bot­ ulinum spores generally experience a higher incidence of botulism than regions containing fewer spores (28, 29, 52, 64). The highest number of cases of recorded botulism has been reported from Russia, Poland, China, France, and the United States, and these countries have regions with high levels of C. botulinum spores (28, 29, 52, 64). C. botulinum spores have a relatively low prevalence in several countries including Great Britain, Sweden, The Netherlands, Switzerland, Austria, Greece, Australia, New Zealand, Mexico, and countries in the South American and African continents (28, 54, 64). Type B (nonproteolytic) and type E spores are relatively common in regions of Europe and Scandinavia. Type E, associated with marine and freshwater foods, is the predominant cause of botulism in cooler aquatic regions, including coastal regions of Canada, Alaska, and northern Japan; Scandinavia; and regions of Russia (53, 54, 64). C. botulinum serotype A is most prevalent in the Western continental United States, and type A spores also have a high prevalence in regions of China and Argentina. C. botulinum type B (proteolytic) is most prevalent in eastern U.S. soils (53, 54, 64). Twenty-four percent of the soils tested in the United States were found to harbor C. botulinum spores (28). The prevalence of spores was higher in sediments and soils in and near Lake Michigan and along the North American Pacific coast (28). The relative prevalence of C. botulinum spores in soils and sediments is similar in analogous regions of Canada, Central and South America, Europe, and Asia (28, 54). C. botulinum spores have also been associated with a variety of foods, generally at low concentrations (29, 54, 64). The prevalence and type of spores in various raw foods usually reflect the prevalence of the spores in the geographical origin of the foods. For example, fish are often contaminated with C. botulinum type E due to the prevalence of spores in many marine and freshwater coastal environments. Foods harvested from continental soils such as vegetables and fruits often contain type A and B spores (28, 54, 64). Certain foods are rarely contaminated with C. botulinum spores, including many types of meats such as raw poultry, beef, pork, and dairy products (29, 54). However, spores can be inadvertently added, particularly through formulation with fish, vegetables, and dry ingredients such as dehydrated vegetables and spices. Processing conditions, plant hygiene, and human error also impact the presence of viable C. botulinum spores in

SMP_Food Microbiology_CH17.indd

444

foods (105, 115). When new processing techniques are implemented in food production, the process must be carefully evaluated to verify that it will not increase the risk of C. botulinum growth in foods (47, 105). In addition to processing considerations, the formulation of foods, taking into consideration intrinsic and extrinsic factors (including ­packaging), is critical for controlling C. botulinum growth and toxin production in raw and minimally processed low-acid foods (47). This aspect is discussed in more detail below.

MICROBIOLOGICAL CHARACTERISTICS OF BOTULINOGENIC CLOSTRIDIA The defining feature of neurotoxigenic clostridia is that they produce the characteristic botulinum neurotoxin. Clostridia are strict anaerobes and obtain energy by fermentation (16, 35, 64). In toxigenic species, the spores are generally wider than the vegetative organisms in which they are formed, imparting spindle shapes, the characteristic clostridial forms (52, 57, 67, 70). In culture, neurotoxigenic clostridia typically grow as large rod-shaped bacteria and often form filaments or chains. The vegetative cells are often curved, their sides parallel, and their ends rounded (52, 57, 67, 70). Certain features are valuable in initial evaluation of C. botulinum cultures in addition to demonstration of botulinal neurotoxin production. All strains are anaerobic, gram-positive spore-forming rods (50, 52, 57, 67, 70). The Gram reaction may be weak or appear negative in 24-h or older cultures (57, 67, 70). Spores are generally present when cells are grown in rich media, but their microscopic presence may require several days to 1 to 2 weeks of incubation. The lipase reaction, which appears as a pearly film surrounding the colonies on egg yolk agar, is a characteristic of all serotypes of C. botu­ linum, except type G and C. baratii and C. butyricum (50, 70). Botulinogenic clostridia are extremely heterogeneous in their physiology, and significant differences occur in several characteristics, including heat stability of spores, temperature range of growth, acid and salt tolerance, proteolytic activity, and substrate utilization (Table 17.2). Studies from a number of laboratories have revealed that C. botulinum can be separated into four physiological groups (groups I to IV), and this grouping has been widely accepted (16, 42, 50, 57, 64, 99). These groups have been determined to be genetically distinct by several methods, including DNA hybridization, pulsed-field gel electrophoresis, multilocus enzyme electrophoresis, sequencing of genes encoding rRNA, and gene arrangements within the toxin gene clusters (42, 62). With the

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

445

Table 17.2  Groupings and relevant growth and resistance properties of botulinogenic clostridiaa Group (C. botulinum) Property Neurotoxin type(s)

I

II

III

IV (C. argentinense)

V (C. butyricum)

VI (C. baratii)

A, B, F

B, E, F

C, D

G

E

F

Minimum

10°C

3°C

15°C

12°C

10°C

20°C

Optimum

35–40°C

18–25°C

35–40°C

35–40°C

30–37°C

30–40°C

48°C

45°C

NA

45°C

~40°C

NA

Minimum pH

4.6

5.0

NA

NA

~3.6

NA

Inhibitory aw

0.94

0.97

NA

NA

0.97

NA

Inhibitory NaCl concn (%)

10%

5%

3%

>3%

~5%

5%

D100°C of spores (min)

~25 min

<0.1 min

NA

NA

NA

NA

D121°C of spores (min)

0.21 min

<0.005 min

Growth temp (°C)

Maximum

Data are for a limited number of strains. Inhibitory factors interact, and hence, the inhibitory values may be ­affected by sublethal combinations of factors. C. sporogenes, often considered a surrogate for group I C. botulinum, has D121°C values ranging from 0.5 to 6 min. NA, insufficient data. a

isolation of toxigenic C. butyricum and C. baratii, the addition of groups V and VI has been suggested (42, 50). These physiological groupings have also been related to metabolic and physiological characteristics, including nutritional requirements (50, 52, 57, 145), resistance to salt, acidity, and other environmental and food components (9, 47, 54, 64, 96), spore heat resistance and germination properties (38, 42, 47, 51, 54, 60, 99; see also chapter 3), tolerance of high-pressure treatment (110), resistance to chemicals and sanitizers (54, 60, 74), resistance to tolerance to air and modified atmospheres (47, 54, 64, 78, 144), minimum growth temperature (36, 54, 99), end product formation (19, 42, 50, 57), neurotoxin gene cluster composition and arrangement, and neurotoxin expression (13, 63). The four groups also have distinctive surface antigen relationships (50, 52) and differ in the host range of bacteriophages and antibacterial activity of bacteriocins (boticins) (62, 72, 127). Characteristics of botulinogenic clostridia, particularly pertaining to growth and survival in foods, are summarized in Table 17.2. Botulinal neurotoxin production is affected by a variety of factors, including temperature, amino acid and peptone composition and concentration, glucose concentration, and several other factors. Botulinal neurotoxin in type A strains is formed during late exponential phase and into early stationary phase (13). Following synthesis, the single-chain neurotoxin is activated to the dichain form by proteolysis in late exponential and stationary phases in the growth of proteolytic strains (13, 63). The

SMP_Food Microbiology_CH17.indd

445

genetic regulation of botulinum toxin production is a relatively new area of study, largely due to the recent development of genetic methods for its study. In group I C. botulinum, arginine and glucose decrease toxin production and activation, whereas in C. ­ botulinum type E (group II), tryptophan decreases toxin formation (reviewed in reference 63). A putative q­uorum-sensing agrBD system, which is known to regulate gene expression in other gram-positive bacteria, may also regulate neurotoxin production and sporulation in proteolytic C. botulinum (25). Genetic tools for elucidating genes involved in toxin formation and activation as well as other traits in C. botulinum and related neurotoxigenic clostridia are being developed, particularly the use of a type II intron to inactivate genes as originally developed by the Lambowitz group (77) and modified for clostridia (55). This gene inactivation system has been used to elucidate mechanisms of botulinum neurotoxin and spore formation (25) and to develop isogenic but nontoxic strains for food challenge studies (15). An interesting feature of the genetic regulation of botulinal toxin formation is that it is positively regulated by a potentially new class of RNA polymerase sigma factors (for a review, see reference 35). Toxin gene regulation in C. botulinum and other toxin-forming clostridia has been reviewed (35, 63). As genetic tools become available, much more information will be elucidated regarding the genes involved in botulinal toxin biosynthesis, proteolytic activation, complex formation, and other properties such as stability, secretion, and function in

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

446 C. botulinum, which could provide target genes for prevention of toxin synthesis through a bioinformatic and genetic approach. The resistant endospores of neurotoxigenic clostridia are widely distributed in the biosphere and are disseminated in dust, waters, vapors, sewage, and various fomites such as insects, from which they readily contaminate most environments, including soils, waters, humans and animals, households, and buildings, as well as commodities such as foods, with honey being the prototypic example (5, 31, 52, 67, 75). Unlike many other pathogenic bacteria, neurotoxigenic clostridia are saprophytic and do not have an obligatory relationship with an animal host (52, 62).

BIOCHEMISTRY AND PHARMACOLOGY OF BOTULINAL NEUROTOXINS The exceptional feature of the neurotoxigenic clostridia is the production of a characteristic neurotoxin of extraordinary potency for humans and animals (116, 132). Botulinum neurotoxins are the most potent toxins known, with an estimated lethal human intravenous dose of 0.1 to 1 ng per kg of body weight (7, 116, 118). Botulinal neurotoxin has the prominent feature among protein toxins of being extremely poisonous by the oral route, with an estimated lethal dose of 0.1 to 1 mg per kg (7; A. E. Larson and E. A. Johnson, unpublished data). The oral toxicity varies depending upon the serotypes of botulinum neurotoxin, the toxin complex, the toxic food consumed, and the presence of food and alcohol in the gastrointestinal tract, as well as other factors (7, 20; Larson and Johnson, unpublished). The assay generally used for detection of botulinum neurotoxin in foods and clinical samples is the mouse bioassay. The intraperitoneal 50% lethal dose (LD50) for a 20-g Swiss-Webster (18- to 22-g) mouse is ~3 to 10 pg of purified botulinum neurotoxins (116). The mouse bioassay is an extremely sensitive method in interlaboratory studies and is the gold standard for measuring the active levels of botulinal neurotoxins in foods and clinical samples (19, 118, 131, 138). Due to known drawbacks of the mouse bioassay, particularly the need for large numbers of mice, other assays have been developed such as neuronal cell-based assays with increased sensitivity and specificity compared to the mouse bioassay. Certain neuronal cell-based assays are more sensitive than the mouse bioassay and also measure each cellular step in the intoxication process (73, 102–104), unlike certain other screening methods such as enzyme-linked immunosorbent assays (19, 51, 52, 81, 131) and in vitro cleavage assays of botulinum neuronal substrates (3, 48, 68, 71).

SMP_Food Microbiology_CH17.indd

446

Botulinum neurotoxins are proteins of ~150 kDa that exist naturally as components of progenitor toxin complexes (13, 116, 132), whereby the neurotoxin component is associated with nontoxic proteins and RNA (118, 132). The nontoxic proteins in the complexes provide protection during experimental manipulations, food processing, and passage through the gastrointestinal tract (116, 118). Botulinal neurotoxins are produced as single-chain molecules of ca. 150 kDa that achieve their characteristic high toxicities of 107 to 108 mouse LD50s (MLD50) per mg by posttranslational proteolytic cleavage to form a dichain molecule composed of a light (L) chain (~50 kDa) and a heavy (H) chain (~100 kDa) linked by a disulfide bond (92, 118, 132). Botulinum neurotoxin consists of three basic functional domains (76, 92, 93, 120): (i) L chain, the catalytic domain that has endopeptidase activity on ­neuronal substrates; (ii) HN, the translocation domain residing in the N-terminal region of the H chain; and (iii) HC, the receptor-binding domain located in the C-terminal region of the H chain. The gene and amino acid sequences of botulinum neurotoxins have been determined for a number of C. botulinum strains (4, 14, 56, 128). Recent studies have revealed that subtypes of botulinum neurotoxins occur within specific serotypes, in which the 150-kDa proteins differ by 3 to 12% in amino acid sequence and associated properties such as immunogenicity and neutralization by antitoxins (4, 61, 128). These findings reveal that different evolutionary lineages of botulinal neurotoxins exist within this toxin family, with differing properties of importance to microbiology and medicine. Botulinum neurotoxin enters the circulation from the intestinal tract in foodborne and intestinal botulism or from wound infections in cases of wound botulism (7, 93, 113). Once in the circulation, the neurotoxin binds to and is internalized within target nerves, primarily in cholinergic nerve endings at the neuromuscular junction (21, 93, 120). Botulinum neurotoxins enter nerves by a multistep process: (i) binding to receptor proteins and lipid gangliosides; (ii) internalization with vesicles by endocytosis; (iii) translocation of the catalytic portion (Lc) into the nerve cytosol; and (iv) cleavage of neuronal substrates (components of the soluble NSF attachment protein receptor apparatus) by the Lc (92, 93, 120). This ultimately results in the inhibition of release of acetylcholine at the neuromuscular junction, preventing muscle activation and causing the characteristic flaccid paralysis of botulism (7, 21, 26, 93). The cellular mechanisms of botulinum neurotoxin intoxication have been an area of intense investigation, and excellent reviews of this intricate process are available (93, 120).

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

447

CLINICAL ASPECTS Since botulism is a true toxemia and botulinum neurotoxin is solely responsible for the illness, foodborne, infant, and wound botulism are clinically similar. The hallmark clinical symptoms of botulism are a bilateral and descending weakening and paralysis of skeletal muscles (7, 21) (Fig. 17.1). Certain of the serotypes, especially type B, cause pronounced autonomic symptoms (136). The incubation time for onset of symptoms varies with the type of botulism, the botulinum neurotoxin serotype, and the quantity of neurotoxin that reaches target nerves. Foodborne botulism occurs following the consumption of food contaminated with preformed botulinum neurotoxin; the vast majority of cases are caused by types A, B, and E, and cases are rarely due to type F (42, 50, 65, 121). In foodborne botulism cases, the onset time is usually 12 to 36 h following consumption of the toxic food. The incubation period can be as short as 2 h when large concentrations of toxin are ingested or as long as 2 to 14 days with serotypes B, E, or F or ingestion of small amounts of botulinum neurotoxin (50, 64, 129). Wound botulism usually has a relatively long incubation period of 4 to 14 days, reflecting the time needed for neurotoxigenic clostridia to colonize the wound and produce neurotoxin (129). Infant botulism has an incubation time from 6 to 8 hours to several days (5), although a more

Figure 17.1  Portrayal of a person with the flaccid paralysis symptoms characteristic of botulism. Drawing prepared by James K. Archer, Centers for Disease Control and Prevention, Atlanta, GA. doi:10.1128/9781555818463.ch17f1

SMP_Food Microbiology_CH17.indd

447

rapid onset has occasionally been reported. It is controversial whether rapid onset of fulminant botulism is a cause of sudden infant death syndrome (6). Intestinal botulism has also occurred concurrently with other clostridial intestinal infections such as that caused by C. difficile (31). In most cases of botulism, cranial nerves are first affected, particularly those innervating the eyes, and the first symptoms are blurred and double vision, dilated pupils, and drooping eyelids (Fig. 17.1) (7, 17, 21, 65, 121). The eyes respond slowly to light in a darkened room (5). These abnormalities are followed by other symptoms in the cranium, including difficulty in swallowing, weakness of the neck and mouth, dysphagia (drooling and difficulty swallowing), and problems in speaking (7, 17, 21, 121). As the paralysis descends, weakness of the upper limbs and the torso occurs, and in severe cases muscles affecting respiration are weakened, and mechanical ventilation to prevent fatality by suffocation is required (7, 21). Generally, the patient’s hearing remains normal, consciousness is not lost, and the victim is cognizant of the progression of the disease. Other symptoms may occur, including nausea or vomiting, dizziness or vertigo, diarrhea or constipation, dry mucous membranes in the mouth and throat, sore throat, and parasthesias, which may be related to botulinal neurotoxin or other toxins or pathogens in the contaminated foods. Infant botulism has certain distinct features, including constipation, weakness and hypotonia, poor suck and feeding, weak cry, lack of head control, and cardiovascular abnormalities (hypotension and tachycardia) (5). Certain cases of botulism have been associated with abnormalities of autonomic and sensory functions such as constipation, dry mouth, and difficulty in urination (21). A patient’s awareness of weakening of muscle activity and ensuing paralysis can lead to considerable emotional distress, including anxiety and depression (22, 24). The most severe and longlasting foodborne botulism generally occurs with type A (7, 121). In general, the duration of foodborne botulism illness follows the pattern “A > B > F > E” in animal models and in humans (5, 7, 8, 121). In infant botulism, nearly all cases are caused by types A and B, and rarely type F. Recently, a case of type E infant botulism was documented (84). The fatality rate from foodborne botulism has decreased from 50 to 70% in the 1800s and 1900s to 5 to 10% in the past 20 to 30 years (19, 50, 91). Convalescence and recovery from botulism are usually prolonged, requiring weeks to months depending on the serotype of toxin and the quantity consumed. Recovery is usually complete, and patients regain full normal function (5, 21).

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

448 Diagnosis of botulism includes clinical assessment of the initial visual disturbances and other cranial effects, skeletal muscle weakness, and detection of botulinum neurotoxin from appropriate specimens by the mouse bioassay (7, 50, 121). The definitive diagnosis is detection of botulinum neurotoxin in the patient’s serum, stool, wound, vomitus, and/or suspect contaminated food (7, 50, 121). However, detection of botulinum neurotoxin is often negative in serum and feces after 48 h of exposure, and foods may be unavailable for analysis. Electrodiagnostic testing can provide presumptive diagnosis of botulism and is particularly useful for patients with clinical signs but whose samples test negative in the mouse bioassay (7, 21). Electrophysiologic testing is also useful for differential diagnosis of other causes of acute flaccid paralysis syndromes such as Guillain-Barré syndrome, myasthenia gravis, tick paralysis, and LambertEaton syndrome (7, 21). Guidelines for electrodiagnostic testing for botulism have been described (21). Currently, there is no treatment for botulism except for passive administration of antibodies (available from the CDC and certain State Health Departments) at early stages in the disease before botulinum neurotoxin has begun internalization into nerves (7, 8, 19, 50). Antitoxin can effectively neutralize unbound toxin in the circulation, but it will not prevent the disease once receptor binding and the internalization process of botulinum neurotoxin is under way. Therefore, antitoxin must be administered as quickly as possible. The major treatment of botulism is supportive nursing care, with specific attention given to respiratory ability and the need for mechanical ventilation (7, 21). Nasogastric or parenteral nutritional support may also be required. A human botulism immunoglobulin intravenous treatment (BabyBIG) has been prepared from immunized human donors for treatment of infants with botulism (8). These antibody preparations reduce the severity of botulism in infants and reduce the duration of hospital stay (8). Infants treated with BabyBIG-IV had a decreased duration of hospital stay, from 6.4 to 2.0 weeks and 2.2 to 1.5 weeks for type A and B botulism, respectively (8). Information on the drug can be obtained from www.infantbotulism.org, supported by the California Department of Health Services.

EPIDEMIOLOGY OF FOODBORNE BOTULISM Foodborne botulism occurs mainly in clustered geographic regions of the world (54). In many countries, it is very rare, although the actual incidence of botulism is undoubtedly greater than is reported. It is likely

SMP_Food Microbiology_CH17.indd

448

that mild cases are not diagnosed and patients are not admitted for treatment. Certain countries currently do not have adequate public health facilities for botulism diagnosis (see reference 140 for examples). Since it is a rare disease, botulism may be misdiagnosed as another neurologic disorder (7, 21). Nonetheless, in the United States and certain other countries with capable public health facilities, the characteristic paralytic symptoms and records of release of antitoxin for treatment probably make hospitalized botulism cases one of the most accurately reported foodborne diseases (19). In pioneering ecological surveys and studies of C. bot­ ulinum, Meyer and colleagues concluded that the risk of botulism is greater in geographic regions that have a high prevalence of type A, B, and E spores (90, 91). This has been supported by other studies of C. botuli­ num spore prevalence, generally following a botulism outbreak. Current epidemiologic and microbiological evidence indicates that types A, B, and E (and rarely F and C) toxins are responsible for human botulism. Since their discovery in the mid-1980s (41, 50, 51), C. baratii and C. butyricum strains that produce botulinum neurotoxins have been recognized as causes of infant botulism (41, 50). Botulinogenic C. butyricum has also been associated with foodborne outbreaks (137). The primary regions of the world with reports of human foodborne botulism are East Asia (China and Japan), North America, certain countries in Europe (Russia, Poland, Germany, France, Italy, Spain, Portugal, Denmark, and Norway), the Middle East (Iran), Latin America, and South Africa (53, 64). Cases of hospitalized human botulism are rare in the United Kingdom, although notable outbreaks have occurred including the Loch Maree tragedy, the Birmingham incident, and the hazelnut yogurt outbreak (64). Human botulism is also rarely reported in Africa, Australia, Israel, Taiwan, Greece, New Zealand, India, Mexico, and several South American countries (53). However, these conclusions regarding the epidemiology of botulism require careful consideration with an increasing global food supply and regional and worldwide trade and travel. In the United States, botulism has been primarily associated with consumption of in-home-prepared foods (19, 121, 129), although restaurant-associated foods have been associated with several outbreaks during the past 2 decades (64, 121, 129). Botulism outbreaks in the United States reached a peak during the 1930s, when procedures for diagnosis were developed, and then declined due to vigorous preventive measures in the commercial canned food industry and the development of guidelines for home canning (19, 90, 91). From 1899 to 1949, there were in the United States 1,281 reported

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

449

cases and 830 deaths; from 1950 to 1977, 678 cases and 169 deaths; from 1978 to 1993, 423 cases and 31 deaths; and from 1990 to 2000, 263 cases and 11 deaths (19, 90, 91). Currently, approximately 25 to 50 foodborne botulism cases are diagnosed annually (121, 130). In the United States from 1899 to the early 1990s, approximately 60% of foodborne botulism cases were caused by type A toxin, 18% by type B, and 22% by type E (19). Botulism is endemic to the Pacific Coast states and in Alaska. During the period from 1990 to 2000, 39% of the cases occurred in Alaska and were caused by type E toxin in traditional Alaska Native home-prepared fermented fish and sea mammals (121, 129). In the lower 49 states, the most common vehicles were home-canned vegetables (129). Nine percent of the cases were due to two restaurant-associated outbreaks, each affecting 25 individuals (129). During the period from 1981 to 2002 in the United States, of 1,269 reported cases, 13 adult type F botulism cases caused by toxigenic C. baratii were diagnosed (129). In only one case was a food vehicle identified, and C. baratii was isolated from leftovers of spaghetti and tuna. Outbreaks associated with consumption of commercial canned chili and related products, carrot juice, and various other products have occurred during the past 10 years (18, 58, 59, 114, 122, 147). Toxigenic C. butyricum has been reported to have caused foodborne botulism in India (137). Foods involved in mainland U.S. botulism are usually home-canned fruits and vegetables, followed in incidence by home-prepared meats, fish, and other miscellaneous foods (Table 17.3) (121, 129). Of 182 outbreaks associated with foods from 1971 to 1989, 137 were caused by fruits and vegetables, 15 by meats, 13 by fish, and 17 by other foods, including mixed vehicles. As expected, vegetables harvested from soils have been the cause of botulism, particularly onions, garlic, and potatoes (121, 129, 130). These outbreaks resulted mainly from improper preservation procedures of the implicated food such as storing at ambient conditions garlic pieces in oil and baked potatoes in foil (2, 130). Such handling proce-

Table 17.3  Typical foods associated with foodborne botulism Fermented fish products Home-canned vegetables (green beans, mushrooms, olives) Foods preserved in oil (garlic, onions) Cooked foods stored at room temperature (potatoes, onions) Foods packaged in air-tight or modified-atmosphere containers with inappropriate refrigeration

SMP_Food Microbiology_CH17.indd

449

dures created an anaerobic environment for growth and toxin production by C. botulinum. Although C. botu­ linum spores are found on certain vegetables, surveys have revealed that the numbers of spores on these foods are usually quite low (29). Botulinum spores are rarely found in most commercially produced meats, with the exception of certain varieties of fish. Botulism from consumption of commercial and restaurant­-prepared foods has unexpectedly occurred with changes in food handling and preservation practices. Contributing factors to several of these outbreaks have been described (2, 12, 42, 47, 54, 121, 129, 130). An excellent example is the poor process control and maintenance of chilled distribution of smoked whitefish from the Great Lakes that resulted in a resurgence of botulism in the 1960s (40, 54). In the 1970s, changes in packaging procedures and underprocessing led to botulinum toxin production by C. botulinum in canned mushrooms prepared by seven U.S. commercial producers (85). Production of foods with minimal processing or no preservatives and relying primarily on refrigeration for controlling C. botulinum has led to outbreaks of botulism (47). Increased consumption of foods in restaurants and from food service establishments in the United States and certain other countries have impacted the occurrence of foodborne botulism (47, 64, 121, 129). In 2001, an outbreak of type A botulism affecting 15 individuals occurred at a church supper from consumption of chili. In 1993, an outbreak of foodborne botulism affecting 17 persons occurred in Texas from skordalia dip prepared using baked potatoes cooked in foil and stored at ambient temperature following cooking (2). A botulism outbreak affecting eight persons involved a commercial cheese sauce that was properly processed commercially but then was left opened in the food service establishment without proper refrigeration. Type A spores were introduced in handling, grew, and produced botulinum neurotoxin. In July 1998, a large botulism outbreak reportedly affecting 1,400 persons with 19 deaths was attributed to consumption of spoiled meat and poultry products, including paté and a processed meat called “casher” (47). Poor hygiene in the processing plant and inadequate refrigeration in distribution contributed to the outbreak. In 1997, an outbreak of foodborne botulism affecting 27 patients and leading to one death was caused by locally made cheese in Iran (47). A type A botulism outbreak that affected nine bus drivers in Argentina was caused by a restaurant­-prepared Argentine meat roll (matambre) (140). A summary of selected botulism outbreaks involving commercially processed or restaurant-prepared foods that have occurred in various regions of the world has been reviewed (47).

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

450 These outbreaks highlight the need for implementation of safe food handling procedures, a rigorous inspection program, and education to prevent future outbreaks of botulism. The only food that has been definitively associated with infant botulism is honey (5, 41). In California, approximately 30% of infant botulism cases were associated with honey consumption following its discovery in 1976 (5, 106), but this has decreased to less than 10% with enhanced public awareness. In Europe, honey consumption was associated with nearly 50% of the cases of infant botulism (41). Analyses of botulism­implicated honey have revealed that the quantities of C. botulinum spores consumed that cause illness can be quite low, as few as 5 to 70 spores per gram. As with many other foods produced today, commercial honeys are often blends from several worldwide geographic regions. There have also been several reported incidences in which honey containing botulinum spores was fed to infants, who did not contract botulism (5). Type B infant botulism was diagnosed in a 5-month-old female baby and was considered to be associated with infant formula, but epidemiologic analysis using physiological tests and pulsed-field gel electrophoresis did not support the conclusion that the outbreak vehicle was unopened formula (66). A recent study revealed that environmental sources such as dust are vehicles for C. botulinum spores in infant botulism. Botulism from consumption of commercially prepared foods has been extremely rare in the United States and many other countries (12, 19, 42, 85). The current good safety record for commercial foods is due, in large part, to the diligence of food manufacturers in formulating, processing, and controlling temperature during distribution of foods (47, 64). Other contributing factors include the low prevalence of C. botulinum spores in prepared foods (53, 54, 127), competition of C. botu­ linum with spoilage organisms, and consumption of foods before toxin production can occur.

PREVENTION OF GROWTH AND TOXIN FORMATION IN FOODS Foodborne botulism is the class of botulism that can most readily be prevented through proper food processing, preservation, and temperature control. Prevention of infant, adult intestinal, and wound botulism relies currently on minimizing exposure to spores, but because up to 90% of spore sources are currently unknown in infant botulism (honey being the exception) (5, 29, 42, 127) and dust is a known vehicle (28), preventing exposure to spores is not practical in many instances.

SMP_Food Microbiology_CH17.indd

450

In contrast, the primary factors that affect survival of spores during processing and subsequent growth of C. botulinum in foods have been well studied, and models have been developed for control, thereby resulting in the ability to design and implement appropriate food formulation and processing principles for prevention of botulism (9, 10, 47, 54, 88, 125, 126). Empirical testing results of the potential for C. botulinum to grow and produce botulinal neurotoxin in a variety of foods have also been extensively reported to assist the food industry in designing and producing botulism-safe foods. In summary, prevention of botulinum neurotoxin formation in foods can be achieved by (i) avoiding contamination of foods by spores; (ii) inactivating spores that are present in foods; (iii) preventing spores from germination and vegetative cell growth resulting in botulinal neurotoxin formation; and (iv) inactivation of botulinal neurotoxins in a food. Practically, it is difficult to prevent contamination of many foods by spores due to their widespread distribution and contamination of foods during harvesting, formulation, and processing. Hence, considerable research and development has been devoted to spore inactivation and prevention of growth in foods (see above; for reviews, see references 74, 127, and 133).

Botulinal Spore Inactivation

Thermal processing methods for inactivating C. botu­ linum spores have been described in chapter 3) and are only briefly summarized in this section. The excellent safety record of botulism prevention in commercial, low-acid canned foods attests to the efficacy of industry practices (19, 47, 53, 85, 91, 95). Most foodborne botulism occurs from home-canned/prepared and homefermented foods (19, 127, 129). Guidelines to prevent botulism in home-prepared foods can be obtained from the CDC, FDA, U.S. Department of Agriculture (USDA), local health departments, and many university outreach programs. Thermal processing is of considerable importance in the food and sterilization industries for inactivation of C. botulinum (23, 105; see also chapter 3). Heat processes used for inactivation of C. botulinum spores in low-acid canned foods will also inactivate other foodborne organisms of public health concern. Group I C. botulinum spores (serotype A and proteolytic strains of serotypes B and F) have a much higher heat resistance than group II spores (type E and nonproteolytic strains of serotypes B and F) (Table 17.2). Spores from nonproteolytic type B and F strains generally have slightly higher heat resistances than serotype E (chapter 3). The basis for differences among the six groups of botulinogenic

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

451

clostridia has not been elucidated, but with the availability of additional completed genome sequences from all six groups of the botulinum neurotoxin-producing strains (14, 101), it is anticipated that genes controlling sporulation and resistance properties will be identified (98). The greatest heat resistance occurs in certain strains of C. botulinum serotype A, in which a D121°C (250°F) of 0.21 minutes is often used by canning and packaging industries as a cardinal value. Interestingly, in C. sporogenes, which has been proposed as a surrogate for C. botulinum, spores can have a D121°C of 0.5 to 6 minutes (see chapter 3). This large difference in thermal resistance and certain other physiological properties brings into question the use of C. sporogenes as a surrogate in the development of processing procedures and for testing of formulations. Nontoxigenic mutant strains with deletions of the entire botulinum toxin gene cluster and flanking sequences, as well as mutants with insertionally inactivated toxin gene, have been isolated in our laboratory from C. botulinum strain 62A (15, 63). However, an ideal nontoxigenic surrogate would be derived from C. botulinum strains by complete deletion of the toxin gene. Inactivation of C. botulinum spores in buffers and in foods by ionizing irradiation has been investigated (54). In general, bacterial spores are 5 to 15 times more resistant than their corresponding vegetative cells. The kinetics of inactivation initially reveal a soft shoulder, followed by a logarithmic rate of inactivation (54). Spores from proteolytic strains have D values in the range of 0.2 to 0.45 megarads (2.0 to 4.5 kGy) at −50°C to −10°C. In the range of −20°C to +50°C, D values decreased by approximately 1 kilorad per °C (54). In contrast to thermal inactivation, type E spores were not significantly more sensitive than type A spores. The presence of oxygen and other factors such as pH appear to affect inactivation and recovery, but the mechanisms are not known (54, 74). From a practical food processing perspective, it appears unlikely that ionizing irradiation at relatively high concentrations would satisfactorily destroy C. botulinum spores and retain a food’s acceptable quality characteristics. Potential food preservation methods under development including high pressure, pulsed electric fields, ohmic heating, and high-intensity light and sound treatments are in preliminary stages of evaluation for inactivating spores, particularly C. botulinum spores (54, 74; see also chapter 3). Conducting research with C. botulinum requires a CDC and/or USDA-registered select agent facility (19), and very few laboratories with capabilities of applying the aforementioned processes to botulinum spores are available. The limited studies

SMP_Food Microbiology_CH17.indd

451

evaluating high-pressure treatments have revealed that spore inactivation does occur, particularly when combined with heat (110). There is a need for critical and careful evaluation of advanced physical preservation techniques for Clostridium and Bacillus endospores. It should be kept in mind that when a physical process is introduced into a food production process, inadvertent events that may enable growth and toxin formation by C. botulinum can occur (see “Growth Requirements of C. botulinum in Foods” below). Chemicals and gases have been evaluated for spore inactivation. Chlorine and related compounds are among the most effective chemicals for destruction of spores (54, 60). Pretreatment of spores with chlorine may also sensitize C. botulinum spores to other inac­ tivation mechanisms such as heat (54). The effectiveness of chlorine is dependent on its concentration and form of use, the presence of organic material, pH, exposure time, temperature, and other factors (54). Evidence suggests that group I C. botulinum, with the highest heat resistance, is also more resistant to chlorine than group II C. botulinum. Since chlorine dioxide is less susceptible to sequestration by organic materials, its use can have advantages over that of hypochlorite (54). For decontamination of clean surfaces, solutions of 100 to 200 mg of hypochlorite per liter for 2 minutes is sufficient for C. botulinum spore inactivation. In water used for cooling of heat-treated cans, 1 to 2 mg of hypochlorite per ml has been recommended (54, 95). The absence of hypochlorite in cooling water probably contributed to one of the largest and most economically devastating episodes of botulism in recent history, which involved canned salmon (85, 95). C. botulinum spores can be inactivated by hydrogen peroxide. For sporicidal activity, relatively high concentrations (ca. 35%) are required (54). Hydrogen peroxide has the greatest stability at pH 3.5 to 4.0 and loses activity as the pH increases. Higher temperatures also increase the activity. Hydrogen peroxide has utility for sterilization of aseptic processing materials and certain packaging systems (54). Ozone at 5 to 6 mg/liter with exposure times of 32 min can also inactivate C. botuli­ num spores (54).

Botulinal Neurotoxin Inactivation

The instability of botulinum neurotoxins in foods and clinical samples is an important aspect contributing to the safety of foods from botulism, but neurotoxin inactivation has not been well studied. Inactivation of botulinal neurotoxin toxicity is dependent on serotype and the type of protein complex in which the neurotoxin is present. In general, botulinum neurotoxins are labile

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

452 to heat, alkali, chlorine, and certain other physical and chemical treatments (54, 118, 123). Heat has long been recognized to be highly effective for inactivation of botulinum neurotoxins (123, 139). Botulinum neurotoxins are most stable to heat under acidic conditions such as pH 3.5 to 5 and in the presence of organic acids, proteins, and certain ions such as Ca2+ and Mg2+ (54, 103). Heating at 70°C for 1 h, 80°C for 30 min, or boiling for 5 minutes inactivated toxicity in buffers and foods (123). However, the rate of thermal inactivation does not follow a linear curve, and considerable tailing has been observed during heat inactivation of botulinum neurotoxin inactivation in various buffers and foods (54, 123). The biphasic curves observed preclude the use of D values to model thermal inactivation of botulinal neurotoxin, and it has been proposed that heat resistance be expressed as the time required for reduction of activity to a minimal LD50 (F value) (54). The basis for heat resistance of a small percentage of botulinal neurotoxin is unknown, and it has been suggested that it could be due to the formation of protein complexes, aggregation, changes in physical shape, or renaturation on cooling. The tailing observed could affect the degree of inactivation of high levels of botulinum neurotoxin during pasteurization processes used for various foods such as milk, and the possibility for retention of toxicity has opened a debate regarding the adequacy of current timetemperature conditions used in commercial pasteurization procedures (1, 108, 142, 143; Larson and Johnson, unpublished). The stability of botulinum neurotoxins in drinking water, lake water, and other liquids depends on the presence of organic matter, higher acidity, and hypochlorite (146). Botulinal neurotoxin is stable in distilled water for 7 days, and stability is enhanced at lower pH values such as 4 to 5 and in the presence of proteins. Botulinum neurotoxins are sensitive to chlorine, and the concentrations of chlorine generally used in drinking water are adequate for inactivation. Contaminated objects or surfaces can be decontaminated with 0.1 to 0.5% hypochlorite solution. For routine sanitation and cleansing of surfaces, exposure to 0.1% hypochlorite (undiluted commercial bleach is about 5.25% hypochlorite) is adequate, and for higher levels of contamination 0.5% should be used with 20 to 30 min of exposure, followed by rinsing with distilled water. For decontamination of spills, the toxin solution on a surface should be covered with absorbent material to prevent aerosolization. Botulinum neurotoxin/A toxin complex is inactivated by alkali at pH values of >9.5 to 10 (ca. 0.1 M NaOH) (54). Limited studies have revealed that botulinum neurotoxin is inactivated by 250 ppm ozone (54).

SMP_Food Microbiology_CH17.indd

452

In general, botulinal neurotoxins are not affected by freezing, particularly in the presence of proteins and organic acids at pH values of 5 to 6.5. The notable exception to stabilization by organic acids during freezing is acetate buffer, in which all toxic activity is lost for botulinal type A neurotoxin (119). Botulinum neurotoxins are not significantly inactivated by gamma irradiation from a 60Co source (123). Drying by vacuum drying or lyophilization can stabilize botulinum toxins, especially in the presence of proteins such as gelatin or albumin and excipients such as trehalose (54, 118). Overall, the effects of physical and chemical treatments on botulinum neurotoxin and its protein complexes have not been thoroughly studied, although such information would be useful to enhance the safety of foods.

Growth Requirements of C. botulinum in Foods

The primary factors that influence the growth of C. bot­ ulinum in foods are temperature, pH and acidity, water activity (aw), redox potential, nutrient sufficiency, the presence of antimicrobials, and competitive microflora (Table 17.4) (47, 54, 64). Compilations of conditions for growth and neurotoxin formation by C. botulinum in various foods and media are available (47). Cardinal values required for growth of botulinogenic clostridia in foods are presented in Table 17.2. In practice, these limit values apply to relatively few strains, and most strains require slightly less stringent values for growth and toxin formation to occur. In foods formulated to be botulism safe, inhibition relies on the inhibitory activities of combinations of factors (47, 54, 127). Among the most important inhibitory factors are acidic pH and aw. The critical pH for botulism safety of foods is pH 4.6,

Table 17.4  Primary physical treatments and antimicrobials

used in formulation of botulism-safe foodsa Thermal treatment pH and acidity aw and solute composition

Presence of antimicrobials and competitive microflora Organic acids Nitrites, sulfites, phenolic compounds Polyphosphates Fatty acids and esters Gas composition Naturally occurring antimicrobials Competitive microflora Indirect antimicrobials a

Modified from reference 47.

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

453

below which C. botulinum spores will not germinate and outgrowth of vegetative cells will not take place. Food products with pH of £ 4.6 are termed high-acid foods and have an excellent record of botulism safety (96). Low-acid foods are those having an equilibrium pH of >4.6 and an aw of >0.85. These foods require suitable processing or preservation procedures to prevent survival and growth of C. botulinum (47, 54). In refrigerated, minimally processed foods in which growth of group II C. botulinum is of concern, a pH of 5.0 is sufficient to prevent germination and growth. Short-chain organic acids (e.g., lactic, acetate, and malic) with pKa values in the range of 3.0 to 5.0 are generally used as acidulants, as these are more inhibitory than mineral acids at equivalent pH values. When acids are used to adjust the pH of foods, adequate time and mixing are required for diffusion of the acids throughout the foods, occasionally several hours to days. If the foods will not be subsequently processed, then these acidulated foods should be refrigerated until equilibrium is reached. The classification of foods into groups of high and low acidity has been a useful criterion for ensuring the botulism safety of most foods. However, botulism outbreaks have been attributed to high-acid foods, particularly tomato products (19, 54, 64, 91, 133). Certain conditions can enable toxin formation in high-acid foods, including inadequate penetration of acids into larger pieces of foods, hence creating microenvironments of higher pH. A second important factor has been termed metabiosis, whereby yeasts, molds, or bacteria can metabolize acids and other components such as proteins and raise the pH to a level at which C. botulinum can grow (54, 133). For example, fungal mats can form at the surface of certain foods, thereby raising the pH under the mat until permissive conditions for C. botulinum growth are achieved. Growth of C. botulinum has been reported in acid media in which precipitated protein or meat particles are present, probably forming a permissive microenvironment pocket for C. botulinum proliferation (54). Upper pH limits for growth of C. botulinum are in the range of pH 8 to 9, but inhibition by alkalinization is not practical in foods. Control of aw, particularly by addition of NaCl, is of considerable utility for C. botulinum control. Food preservation by brining is mainly due to the reduction of aw, whereby sufficient free water is not available to the organism for growth. aw is defined as the vapor pressure of the food divided by the vapor pressure of pure water or the equivalent of relative humidity/100. Brining is the most common practice for reducing aw in food preservation. The percentage of brine in a food is defined as follows: % NaCl ´ 100/%H2O + %NaCl). The limit-

SMP_Food Microbiology_CH17.indd

453

ing values for inhibition by brine alone are 10% and 5% for group I and group II C. botulinum, respectively (54). Most group I and II strains are inhibited at slightly higher aw values. High concentrations of sucrose can also inhibit C. botulinum by reducing the aw, but approximate concentrations of 30% and 15% sucrose are required to inhibit group I and group II strains, respectively (47, 62, 99). Similarly, glycerol, organic polymers, ions such as potassium, and other food components can bind free water and reduce the aw, but their efficacy in inhibiting C. botulinum growth is generally relatively poor on a weight % basis compared to that of NaCl (47, 62, 125). Therefore, when these compounds are used as preservatives, it is essential to measure the aw of the food during its formulation and to conduct challenge studies of the food (33, 47) during development and before commercialization. Temperature is commonly used to prevent C. botu­ linum growth in foods. The minimum temperature for growth of group I strains is commonly accepted to be 10°C (50°F) (54, 64), whereas group II strains can grow at temperatures as low as 3.3°C (38°F) (47, 54). In practice, these minimum temperatures do not apply to many C. botulinum strains. Growth at lower temperatures can require weeks to months due to the slowing of metabolism and growth rate. The upper temperature limits for growth for group I and group II strains are 45 to 50°C (113 to 122°F) and 40 to 45°C (104 to 113°F), respectively. Refrigeration alone has been considered for inhibition of C. botulinum under otherwise permissive conditions in foods, as many consumers have expressed preferences for “healthy” foods that have minimal processing and contain low levels of salt or other preservatives. Although certain minimally processed foods depend on refrigeration alone for botulism safety, this practice can result in C. botulinum growth and formation of botulinal toxin because of poor temperature control or temperature abuse in food stores, in food service establishments, or by consumers. Temperature abuse is one of the most common mishandling practices that result in botulinum neurotoxin production and botulism outbreaks. Therefore, it is not recommended to rely solely on low temperature for botulism safety, and adequate processing and/or formulation by inclusion of secondary barriers such as antimicrobials is recommended (Table 17.4). Since C. botulinum is a strict anaerobe, a permissive redox potential (Eh) is required for germination of spores and growth of vegetative cells. However, relatively high Eh levels of approximately +200 mV can also allow germination and growth initiation of spores under certain conditions (54). Whiting and Naftulin (144) reported

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

454 that the critical level of oxygen that prevents germination and growth by C. botulinum is approximately 1 to 2%. The presence of carbon dioxide has been reported to enhance germination of botulinum spores (54). Although reduced oxygen and increased carbon dioxide levels may be permissive for C. botulinum growth and toxin production, increased safety risk depends not only on the gas environment but also on the product, storage temperature, packaging film used, and indigenous competitive flora. Modified-atmosphere packaging and high-barrier films can be used to effectively extend the shelf life of foods under refrigerated conditions. Reduced oxygen levels and increased concentrations of carbon dioxide in refrigerated modified-atmosphere-packaged foods can decrease oxidative and chemical deterioration, as well as inhibiting common aerobic spoilage microbes such as fungi and many bacteria (47). Concerns have been raised that these conditions may also select for anaerobic psychrotrophic microbes such as group II C. ­botulinum and increase the botulism risk (54, 133). The combination of high-barrier films and respiring foods such as vegetables may also lead to decreased oxygen levels for packaged foods in an ambient atmosphere. The reduced oxygen content, in turn, may enhance the growth of C. botu­ linum. In packaged mushrooms, a high-barrier plastic film reduced oxygen concentration through respiration of the mushrooms and permitted botulinum toxin production (133). Toxin was detected in prepackaged mushrooms after 3 days of storage, although mushrooms were still considered organoleptically acceptable. A simple solution to prevent botulinum toxin formation was to increase gas exchange by introducing 3 to 4 holes in the wrap covering the mushrooms. Similarly, packaging of garlic in oil and relying solely on refrigeration for botulism safety resulted in outbreaks and changes in regulatory formulation requirements. In summary, the presence of oxygen at low concentrations alone cannot be relied upon to inhibit botulinum growth and toxin production. The effect of food components, such as sulfhydryls or competitive microflora, may also reduce oxidative-reduction potential to a level at which C. botulinum can grow. Therefore, similar to relying solely on temperature to control C. botulinum growth, neither redox-potential nor enhanced oxygen transfer is recommended on its own to prevent the growth of C. botulinum; hence, suitable processing and inclusion of secondary barriers are recommended. An important factor for prevention of C. botulinum growth and toxin formation in many foods is the presence of a competitive microflora. Most foods have low levels of C. botulinum spores, and the pathogen is often

SMP_Food Microbiology_CH17.indd

454

inhibited by competitive microflora such as lactic acid bacteria or yeasts that ferment sugars and other substrates in the food with production of inhibitory levels of organic acids, alcohols, and bacteriocins. Competitive microflora or fermentates derived from cultures have been included in food formulations to control growth of C. botulinum (47, 49, 72, 112). For most foods, the use of a high level of a single inhibitor of C. botulinum growth reduces the organoleptic acceptability of the food. Therefore, many foods are formulated to be organoleptically desirable by using combinations of inhibitory factors at sublethal levels. The individual actions of these inhibitors often have cumulative or occasionally synergistic effects for inhibiting C. botulinum growth. An example of a low-acid food prepared in a botulism-safe way by using a combination of treatments and antimicrobials is cured meats in which minimal heat processing is combined with the addition of nitrite (54, 133). Another example is pasteurized process cheese, in which sublethal heat processing is combined with subinhibitory levels of aw (moisture and NaCl content), acid, and phosphate salts (47, 134). Process cheese and related products have an excellent safety record with regard to botulism (47). Extensive studies on process cheese at the Food Research Institute (Madison, WI) led to one of the first empirical models for botulism-safe product formulation (134). Strategies for controlling C. botulinum by using combinations of physical treatments and antimicrobials in various classes of foods have been described (10, 47, 54, 88, 126). The nutrient composition of a food is also important in the control of C. botulinum. The minimal nutrient requirements have been determined for growth of C. botulinum types A, B, and E in chemically defined media (64, 118, 145). Certain nutrients are required in high quantities for growth of C. botulinum, such as arginine for group I C. botulinum and tryptophan for type E (63, 64, 118, 145). Increasing the arginine content also promotes growth in otherwise inhibitory foods, probably through active metabolism with release of ammonia and an increase in pH (64). The presence of glutamic acid in growth medium increases the salt tolerance in group I C. botulinum possibly by serving as a compatible solute or as a precursor to a protective osmolyte (64). Substrate utilization as well as protein and lipid degradation varies among the different botulinogenic clostridial groups (Table 17.5) (16, 42, 50, 64, 70, 145). All of the groups except C. argentinense (group IV) are able to utilize glucose as a primary carbon source. Depending on the group, certain strains can ferment fructose, maltose, sucrose, and sorbitol to varying degrees. While C. baratii and C. butyricum utilize lactose, insignificant acid pro-

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

455

Table 17.5  Nutritional substrates metabolized by botulinogenic clostridiaa Clostridium species and type(s) (group)

Glucose

Fructose

Sucrose

Maltose

Lactose

Gelatin

Milk

Meat

Lipase

Lecithinase

Types A, B, F (I)

+

+/-

+/-

+/-

-

+

+

+

+

-

Types B, E, F (II)

+

+

+/-

+/-

-

+

-

-

+

-

Types C, D (III)

+

+/-

+/-

+/-

-

+

+/-

-

+

-

C. argentinense (IV)

-

-

-

-

-

+

+

+/-

-

-

C. baratii toxigenic (V)

+

+

+

+/-

+/-

-

-

-

-

+

C. butyricum toxigenic (VI)

+

+

+

+

+

-

-

-

-

-

C. botulinum

a

Symbols: +, positive; -, negative; +/-, weak or variable.

duction and growth from lactose are observed with C. botulinum groups I to IV. Similarly, the ability of the botulinogenic clostridia to digest proteins differs significantly among the groups (16, 42, 50, 64, 70). Strong proteolysis is observed by digestion of meat particles or milk (casein as substrates), whereas gelatin is much more easily digested. Botulinogenic clostridia within groups I, III, and IV are proteolytic and can digest milk, meat particles, and gelatin. Group II C. botulinum is nonproteolytic and does not digest casein or meat, but it does degrade gelatin. C. butyricum and C. baratii do not digest the complex proteins, nor do they utilize gelatin. Lipase and lecithinase reactions on plates are commonly used for characterization of clostridia (16, 50, 51, 70). Lipase hydrolyzes the breakdown of triglycerides into glycerol and fatty acids and is observed as a pearly, lustrous film surrounding the colonies. Lecithinase mediates the breakdown of lecithin to diglyceride and phosphorylcholine and is seen as an opaque whitish halo of precipitation around the colony. Groups I to III of C. botulinum have a positive lipase reaction, whereas C. argentinense, C. butyricum type E, and C. baratii are negative in this activity. A summary of substrate characteristics is presented in Table 17.5.

USE OF PREDICTIVE MODELING AND CHALLENGE STUDIES FOR EVALUATION OF C. BOTULINUM NEUROTOXIN FORMATION Predictive modeling can be useful in assessing the processing and formulation parameters for a botulism-safe food. Among the earliest applications of predictive modeling in food microbiology was the development of processing models to determine commercial sterility in

SMP_Food Microbiology_CH17.indd

455

canned foods (see chapter 3). During the past 2 decades, advances in computing and software development have led to significant advances in the field of predictive microbiology. These advances have led to the development and availability of predictive modeling tools (10, 38, 47, 54, 60, 64, 88, 111, 126, 141). These modeling approaches have provided valuable insights into combined effects of biological and environmental factors on the growth, survival, and death of C. botulinum. Predictive models provide valuable information for the development of botulism-safe foods; however, they should be validated in specific foods by challenge studies in a qualified laboratory (33, 47). The National Advisory Committee on the Microbiological Criteria for Foods developed guidelines for botulinum challenge studies on minimally processed refrigerated foods with extended shelf life (33). Similar approaches can be applied to shelf-stable, low-acid foods with the additional recommendation that botulinum neurotoxin production should be determined for twice the expected shelf life rather than 1.5 times as recommended for refrigerated foods. The challenge study should mimic as closely as possible the commercial process for production of the food and should utilize suitable strains of C. botulinum rather than related clostridia that have been proposed as “surrogate” microbes. It is anticipated that as modeling procedures continue to advance, they will provide more in-depth information and guidelines to assist the food industry in the formulation, processing, production, and marketing of safe foods.

Laboratory Procedures

Since botulism is quite rare, it is recommended that toxin and organism diagnostic tests be performed in a suitable reference laboratory that has the necessary experience

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

456 and reference cultures, toxins, and antitoxins for the procedures (19, 50, 119). Relatively few industry, government (local, state, and national), and academic laboratories have these capabilities. The primary laboratory in the United States is the National Botulism Surveillance and Reference Laboratory, Centers for Disease Control and Prevention, Atlanta, GA. Information regarding botulism can be found at http://www.cdc.gov and http:// www.bt.gov/agent/botulism/index.asp, and the emergency telephone number to report botulism cases and request antitoxin is 770-488-7100. Emergency 24-hour telephone numbers or Internet connections of qualified local and state public health laboratories (http://www. cdc.gov/other.htm#states or http://www.astho.org/state. html) can also be contacted (7, 19). Detection of botulinum neurotoxin from representative specimens is the cornerstone of botulism diagnosis (19, 21, 50, 118, 119, 132, 133). The accepted procedure for botulinum neurotoxin detection is the mouse bioassay consisting of two essential steps: (i) determining whether a food substrate or an extract in gel-­phosphate (0.1 M sodium phosphate–0.2% gelatin, pH 6.2) is lethal on intraperitoneal or intravenous injection into mice and (ii) confirming the lethal agent as botulinum neurotoxin by neutralization with specific botulinum antitoxin (19, 50, 132, 133). Detailed procedures, required controls, and difficulties and pitfalls of the mouse bioassay have been described (19, 50, 131, 133). When botulinum neurotoxin of a nonproteolytic C. botulinum strain is tested, it is necessary to activate the toxin to the light and heavy chains by limited proteolysis using trypsin (19, 50, 132, 133). One problem often encountered when testing clinical and food samples is that these specimens may contain nonbotulinum substances that are lethal to mice and cause nonspecific deaths. These can be detected by evaluation of characteristic botulism symptoms and onset time of symptoms in mice (generally >3 to 4 hours in mice injected intraperitoneally), by “diluting out” these substances, which generally have a lower toxicity than botulinal neurotoxin, and most definitely by use of serotype-specific antitoxins in a subset of the samples used for injection (46, 50, 69). Another difficulty encountered is that samples may contain more than one serotype of toxin due to contamination by C. botulinum producing different serotypes of toxins or by single strains that produce more than one serotype of toxin. These difficulties in toxin analysis have been described (19, 46, 50, 131). When delays in specimen collection and analysis occur or if mouse-lethal substances other than botulinum neurotoxin are present in samples, laboratory detection of botulinum neurotoxin may not be conclusive. In clinical cases of botulism, a positive test

SMP_Food Microbiology_CH17.indd

456

was obtained in only about 30% of samples obtained more than 2 days after human exposure to neurotoxin. Therefore, it is important to process samples rapidly and to culture neurotoxigenic clostridia from the foods as described below. C. botulinum may produce reduced amounts of toxin after repeated laboratory culturing, and clinical or wild isolates should be preserved in liquid nitrogen or at −80°C in oxygen-impermeable containers (64, 118). Toxin titers produced by strains and the identity of botulinal neurotoxin should be periodically confirmed using pure stock isolates and by mouse bioassay and neutralization by specific antitoxins (64, 118). For many decades and to the present, the mouse bioassay is the most important and accepted laboratory method for detection and identification of botulinum neurotoxin (19, 50, 118, 119, 133). However, due to certain drawbacks in the mouse assay as well as increased regulatory and ethical concerns in using animals for toxin determinations, there is considerable interest in assays not employing animals. The primary alternative assays employed have been based on immunological detection, particularly enzyme-linked-immunosorbent assays and related immunological methods (19, 52, 131). Since botulinum neurotoxins are zinc metallopeptidases with high specificity for their neuronal substrates, methods based on catalytic activity combined with sensitive detection methods, including high-throughput fluorogenic reporters and fluorescence resonance energy transfer sensing systems, have been developed (3, 32, 48). A promising methodology for detection of botulinum neurotoxins and their protein complexes is mass spectroscopy, including high-resolution platforms to characterize the toxins and to detect reaction products from proteolytic cleavage of the neuronal substrates or detection of signature amino acid sequences of the neurotoxins (11). Mass spectrometry can detect toxin equivalents of as little as 0.01 MLD50 and concentrations as low as 0.62 MLD50 per ml (71). Biosensor devices, including microfluidic and nanofluidic platforms, synaptic chips, and other platforms, are also in development (62). Currently, most of these in vitro assays have the drawbacks that they are less sensitive than the mouse ­ bioassay, have not been adequately validated for detecting botulinum neurotoxin in foods and clinical samples, and do not measure each of the molecular steps required for intoxication and botulism (76, 93, 118). Modifications and alternatives to animal assays have also been considered for detection of antibodies to botulinal neurotoxin in sera of patients who have been treated medicinally with the toxin (32, 102–104). Nonanimal assays that depend on all the steps in the intoxication mechanism, including receptor binding,

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

457

internalization, and catalytic cleavage of neuronal substrates, such as cell-based assays (32, 102–104), would be ideal for determination of toxin titers and antibodies in patients’ sera. Of particular promise is the use of neuronal cell cultures because toxicity to these cells depends on all domains of botulinal neurotoxin and the steps needed for intoxication (32, 102–104). However, many neuronal cell lines are fairly insensitive to botulinum neurotoxins, particularly cell lines derived from neuroblastomas. Recently, our and other laboratories have developed sensitive cell lines, including primary spinal cord neurons and human stem cell pluripotent neurons (93, 102–104), and these have outstanding potential for assay of and determination of the mechanisms of botulinum neurotoxins. Recently, the FDA approved a cell-based assay for determination of botulinum type A toxin for use in medicine (www.fda.gov).

Culturing Botulinogenic Clostridia

A food or clinical sample can be definitively determined to be the cause of botulism only if botulinum neurotoxin activity is demonstrated. However, an important aspect of investigations of botulism outbreaks is the culturing of the responsible botulinogenic bacteria (19, 41, 50, 52, 57, 67, 70, 131). The isolation, identification, and maintenance of pure cultures of botulinogenic bacteria present certain practical difficulties. Clostridia tend to grow as consortia, and pure cultures are sometimes difficult to achieve and maintain (41, 64, 67, 70). The purity of botulinogenic clostridia must be ascertained by microscopy and by plating on nonselective media. Cultures should be routinely tested for botulinum neurotoxin formation using the mouse bioassay and for spore formation. Since botulinogenic clostridia have complex nutrient requirements, rich media are commonly used for cultivation (19, 41, 50, 52, 67, 70, 131). Enrichment of neurotoxic clostridia is often carried out by heating the samples (e.g., 60 to 80°C for 10 min) or by treating the samples with ethanol to kill vegetative bacteria and enrich for sporeformers (19, 50, 52, 67, 70, 131, 133). Non-heat-treated samples should also be cultured because spores may not be present in the samples. From a practical perspective, usually a small quantity of food and clinical specimens, ca. 1-g samples, are added to 10 ml of cooked meat medium +0.5% glucose, or trypticase-peptone-glucose-yeast extract broth (19, 50, 133). For isolation of C. botulinum type E, the addition of trypsin to the medium may enhance recovery by the inactivation of bacteriocins that may be present (50, 133). When pure cultures are obtained, they are characterized by various tests, particularly by demonstration of botulinal neurotoxin by the mouse bioassay and specific

SMP_Food Microbiology_CH17.indd

457

neutralization by antibodies. Further characterization often includes lipase reaction, microscopic observation, and Gram stain, proteolysis, and substrate utilization patterns (50, 57, 67, 70). Formerly, the determination of volatile fatty acid products was routinely determined by gas-liquid chromatography, but this is becoming less common, while molecular techniques such as the determination of the nucleotide sequence of genes encoding rRNA or interspatial regions and PCR analyses for toxin genes have become integral for identification (35, 41, 42, 67, 70). Reference texts describe preparation of media, methods for anaerobic culture, and phenotypic and metabolic tests for identification (19, 81, 138).

Safety Precautions in Working with C. botulinum and Botulinal Neurotoxins

Botulinum neurotoxins are extremely toxic molecules and are considered to be the most potent poisons known (7, 64, 118). They have an estimated lethal human intravenous dose of 0.1 to 1 ng per kg of body weight and an oral lethal dose of 0.1 to 1 mg per kg (7, 116, 118). Because the consequences of an accidental intoxication are so severe, safety must be a primary concern of scientists studying these toxins (19, 50, 67, 118). The CDC recommends biosafety level 3 primary containment and personnel precautions for facilities producing large quantities of the botulinum neurotoxins and in working with high-toxin-producing C. botulinum strains (19, 67). The CDC has defined an “exempt” quantity of botulinum neurotoxin as £0.5 mg, which can be used in biosafety level 2 laboratories. All personnel who work in the laboratory should be thoroughly educated on the hazards of working with C. botulinum and its toxins. They must have knowledge of spill control and toxin inactivation (0.1 to 0.5% hypochlorite or 0.1 M NaOH). Operations should be performed to prevent the formation of aerosols (107, 117), e.g., use of closed containers during centrifugation, avoiding pressurized containers containing active toxin, and application of absorbent materials on spills prior to decontamination. Proper personal care protection must be used, including eye protection and use of gloves and lab coats. Personnel who work with large concentrations of toxin (³0.5 mg and/or high-toxin-producing cultures) should be immunized. Formerly, pentavalent (A through E) toxoid was available from the CDC, but it has been phased out and a heptavalent toxoid is under evaluation. A biosafety manual should be posted in the laboratory and should contain the proper emergency phone numbers and procedures for emergency response, spill control, and decontamination. When possible, culture and toxin handling and manipulation should be performed in a class II

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

458 or III biological safety cabinet with appropriate respiratory protection. The use of needles and syringes for bioassays requires extreme caution. Beginning in 1997, C. botulinum cultures and toxins were included in a group of select agents whose transfer has been controlled by the CDC and for which restricted and secure working areas are mandatory (19). To transfer select agents, both the laboratory and personnel sending and receiving cultures and toxins must be registered with the USDA and/ or CDC and exchange the appropriate approval forms through their Biosafety Office. The laboratory and Principal Investigator should maintain excellent and fre­ quent communication with the Biosafety Office and Responsible Official within the organization.

of the disease, but improved toxoids can be used for immunization of researchers and for military personnel who may be exposed to botulinum neurotoxin in warfare (7). Since immunization provides only partial protection against intoxication with large quantities of toxin, immunized researchers must still follow scrupulous laboratory practices in working with botulinum neurotoxin, including avoidance of aerosols, handling toxin in the biological safety cabinets, and use of closed containers during centrifugation and other procedures (19). Considerable efforts in the United States and in certain other countries are being devoted to the development of heptavalent vaccines, small-molecule inhibitors, and other countermeasures.

GENOMICS OF C. BOTULINUM

CONCLUSIONS

The genomic sequences of several Clostridium species such as C. tetani, C. perfringens, and C. acetobutyli­ cum (35) have recently been determined, and others are being completed, including those of C. difficile and C. botulinum (14). The genomic sequences of relatively few strains of each species have been determined, and it is not clear if the sequences will be representative of most strains, particularly for C. botulinum and C. perfrin­ gens, in which several distinct groups of bacteria occur within the species. Additional genome ­sequencing is being performed for more botulinum neurotoxin-producin­g strains. Analyses of the genomes of the ­clostridia to date have revealed interesting features regarding pathogenicity, spore formation, and metabolism (35, 98). The availability of the genomic sequences of C. botulinum and other botulinogenic clostridial species will enable rational approaches for identification of genes/proteins for development of inhibitors and novel antimicro­ bials that target the pathogen for enhancement of food safety.

Remarkable advances have been achieved during the past 2 decades in elucidating the biochemistry, structure, and pharmacological mechanisms of botulinum neurotoxins. Structural and biochemical studies of these potent neurotoxins have provided much insight into the mechanisms of substrate catalysis, neurospecific binding, and trafficking of botulinal neurotoxin to their neuronal targets. These advances have contributed to the remarkable success in using botulinum neurotoxin as a pharmacological agent for the treatment of various neuronal diseases and may lead to improved vaccines and countermeasures. The availability of genomic sequences and comparative genomic analyses, together with the development of genetic tools such as efficient gene replacement techniques and shuttle vectors for controlled gene expression, will be invaluable in elucidating pathogenic mechanisms of botulinogenic clostridia. Botulism is a rare disease, but its occurrence from consumption of foods can have great economic impact on the food industry, as well as tremendous negative exposure by the media. Hence, inactivation of C. botuli­ num spores in foods that support production of botulinum toxin or the prevention of growth by preservation methods and food formulation is an important goal of the food industry. Considerable information is available regarding the microbiological features of botulinogenic clostridia and preservation and formulation strategies for their control in foods. Nonetheless, botulism continues to occur through consumption of foods, and new technologies and research are needed to enhance control. Newer processing procedures such as pulsed electric fields, ohmic heating, high pressure, and high-­intensity light and sound require careful evaluation before they are widely applied in the food industry. Research for control of botulinogenic clostridia in food systems has

BOTULINUM NEUROTOXIN AND FOOD BIOTERRORISM Botulinal neurotoxin has been considered as a potential biological warfare agent that could be administered in aerosols, foods, or water (7), and some history supports this consideration (7, 17). Botulinum neurotoxin is absorbed through mucous membranes, and three cases of botulism were documented in laboratory workers who apparently inhaled the toxin (19). Botulinum neurotoxin is labile to many environmental conditions and chemicals, and preparation of an aerosol weapon would be difficult. Immunization is not feasible for protection of human populations from botulism owing to the ­rarity

SMP_Food Microbiology_CH17.indd

458

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

459

contributed to fundamental and applied knowledge in the food industry, and it is anticipated that more discoveries will arise from the study of botulinogenic clostridia and their neurotoxins. Research in my laboratory has been supported by the NIH, USDA, University of Wisconsin, and industry sponsors of the Food Research Institute, University of Wisconsin–Madison. I thank Marite Bradshaw for excellent assistance in preparing the manuscript. I am grateful to my laboratory personnel over the years and collaborators and mentors on various projects involving neurotoxigenic clostridia.

References 1. Alberts, B. 2005. Modeling attacks on the food supply. Proc. Natl. Acad. Sci. USA 102:9737–9738. 2. Angulo, F. J., J. Getz, J. P. Taylor, K. A. Hendricks, C. L. Hatheway, S. S. Barth, H. M. Solomon, A. E. Larson, E. A. Johnson, L. N. Nickey, and A. A. Ries. 1998. A large outbreak of botulism: the hazardous baked potato. J. Infect. Dis. 178:172–177. 3. Anne, C., F. Cornille, and C. Lenoir. 2001. Highthroughput fluorogenic assay for determination of botulinum type B protease activity. Anal. Biochem. 291:253–261. 4. Arndt, J. W., M. J. Jacobson, E. E. Abole, C. M. Forsyth, W. H. Tepp, J. D. Marks, E. A. Johnson, and R. C. Stevens. 2006. A structural perspective of the sequence variability within botulinum neurotoxin subtypes A1–A4. J. Mol. Biol. 362:733–742. 5. Arnon, S. S. 2004. Infant botulism, p. 1758–1766. In R. D. Feigen and J. D. Cherry (ed.), Textbook of Pediatric Infectious Diseases, 5th ed. W. B. Saunders, Philadelphia, PA. 6. Arnon, S. S., K. Damus, and J. Chin. 1981. Infant botulism: epidemiology and relation to sudden infant death syndrome. Epidemiol. Rev. 3:45–66. 7. Arnon, S. S., R. Schechter, T. V. Inglesby, D. A. Henderson, J. G. Bartlett, M. Ascher, E. Eitzen, A. D. Fine, J. Hauer, M. Layton, S. Lillibridge, M. T. Osterholm, E. O’Toole, G. Parker, T. M. Perl, P. K. Russell, D. L. Swerdlow, K. Tonat, and the Working Group on Civilian Biodefense. 2001. Botulinum toxin as a biological weapon. JAMA 285:1059–1070. 8. Arnon, S. S., R. Schechter, S. E. Maslanka, N. P. Jewell, and C. L. Hatheway. 2006. Human botulism immune globulin for the treatment of infant botulism. N. Engl. J. Med. 354:462–471. 9. Baird-Parker, A. C., and B. Freame. 1967. Combined effect of water activity, spores, and temperature on the growth of Clostridium botulinum from spores and vegetative cell inocula. J. Appl. Bacteriol. 30:420–429. 10. Baranyi, J., and T. A. Roberts. 2000. Principles and applications of predictive modeling of the effects of preservative factors on microorganisms, p. 342–358. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Foods, vol. 1. Aspen Publishers, Gaithersburg, MD.

SMP_Food Microbiology_CH17.indd

459

11. Barr, J., H. Moura, A. E. Boyer, A. R. Wollfitt, S. R. Kalb, A. Pavlopoulos, L. G. McWilliams, J. G. Schmidt, R. A. Martinez, and D. L. Ashley. 2005. Botulinum neurotoxin detection and differentiation by mass spectroscopy. Emerg. Infect. Dis. 10:1578–1583. 12. Bell, C., and A. Kyriades. 2000. Clostridium botulinum. A Practical Approach to Its Control in Foods. Blackwell Science Ltd., Oxford, United Kingdom. 13. Bradshaw, M., S. S. Dineen, N. D. Maks, and E. A. Johnson. 2004. Regulation of neurotoxin complex expression in Clostridium botulinum strains 62A, Hall Ahyper, and NCTC 2916. Anaerobe 10:321–333. 14. Bradshaw, M., and E. A. Johnson. 2010. Genetic manipulation of Clostridium, p. 238–261. In R. H. Baltz, A. L. Demain, and J. E. Davies (ed.), Manual of Industrial Microbiology and Biotechnology, 3rd ed. ASM Press, Washington, DC. 15. Bradshaw, M., K. M. Marshall, J. T. Heap, W. H. Tepp, N. P. Minton, and E. A. Johnson. 2010. Construction of a nontoxigenic Clostridium botulinum strain for food challenge studies. Appl. Environ. Microbiol. 76:387–393. 16. Cato, E. P., W. L. George, and S. M. Finegold. 1986. Genus Clostridium, p. 1141–1200. In P. H. A. Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt (ed.), Bergey’s Manual of Systematic Bacteriology, vol. 2. The Williams & Wilkins Co., Baltimore, MD. 17. Caya, J. G. 2001. Clostridium botulinum and the ophthalmologist: a review of botulism, including biological warfare ramifications of botulinum toxin. Surv. Ophthamol. 56:25–34. 18. Cengiz, M., M. Yilmaz, L. Dosemici, and A. Ramaza­ noglu. 2006. A botulism outbreak from roasted canned mushrooms. Hum. Exp. Toxicol. 25:273–278. 19. Centers for Disease Control and Prevention. 1998. Botulism in the United States, 1899–1996. In Hand­ book for Epidemiologists, Clinicians, and Laboratory Workers. Centers for Disease Control and Prevention, Atlanta, GA. 20. Cheng, L. W., B. Onisko, E. A. Johnson, J. R. Reader, S. M. Griffey, A. E. Larson, W. H. Tepp, L. H. Stanker, D. L. Brandon, and J. M. Carter. 2008. Effects of purification on the bioavailability of botulinum neurotoxin type A. Toxicology 249:123–129. 21. Cherington, M. 2004. Botulism: update and review. Semin. Neurol. 24:155–163. 22. Chertow, D. S., E. T. Tan, S. E. Maslanka, J. Schulte, et al. 2006. Botulism in 4 adults following cosmetic injections with an unlicensed, highly concentrated botulinum preparation. JAMA 296:2476–2479. 23. Chotyakul, N., G. Velasquez, and J. A. Torres. 2011. Assessment of the uncertainty in thermal food processing decisions based on microbial safety objectives. J. Food Eng. 102:247–256. 24. Coban, A., Z. Matur, H. A. Hanagasi, and U. Parman. 2010. Iatrogenic botulism after botulinum toxin type A injections. Clin. Neuropharmacol. 33:158–160. 25. Cooksley, C. M., I. J. Davis, K. Winzer, W. C. Chan, M. W. Peck, and N. P. Minton. 2010. Regulation of

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

460

26. 27. 28.

29.

30. 31. 32.

33. 34. 35. 36.

37. 38. 39. 40. 41.

42.

­ eurotoxin production and sporulation by a putative n agrBD signaling system in proteolytic Clostridium botu­ linum. Appl. Environ. Microbiol. 76:4448–4460. Devers, K. G., and J. S. Nine. 2010. Autopsy findings in botulinum toxin poisoning. J. Forensic Sci. 55:1649–1651. Dickson, E. C. 1918. Botulism. A clinical and experimental study. Rockefeller Inst. Med. Res. Monog. 8:1–117. Dodds, K. L. 1993. Clostridium botulinum in the environment, p. 21–51. In A. H. W. Hauschild and K. L. Dodds (ed.), Clostridium botulinum: Ecology and Control in Foods. Marcel Dekker, New York, NY. Dodds, K. L. 1993. Clostridium botulinum in foods, p. 53–68. In A. H. W. Hauschild and K. L. Dodds (ed.), Clostridium botulinum: Ecology and Control in Foods. Marcel Dekker, New York, NY. Dolman, E. C. 1964. Botulism as a world health problem. In K. H. Lewis and K. Cassel (ed.), Botulism. U. S. Public Health Services, Washington, DC. Domingo, R. M., J. S. Haller, and M. Gruenthal. 2008. Infant botulism: two recent cases and literature review. J. Child Neurol. 23:1336–1346. Dong, M., W. H. Tepp, E. A. Johnson, and E. R. Chapman. 2004. Using fluorescent sensors to detect botulinum neurotoxin activity in vitro and in living cells. Proc. Natl. Acad. Sci. USA 101:14701–14706. Doyle, M. P. 1991. Evaluating the potential risk from ­extended-shelf-life refrigerated foods by Clostridium bot­ ulinum inoculation studies. Food Technol. 45:154–156. Dunn, E. C. 2008. Postsocialist spores: disease, bodies, and the state in the Republic of Georgia. Am. Ethnol. 35:243–258. Dürre, P. (ed.). 2005. Handbook of Clostridia. CRC Press, Boca Raton, FL. Eklund, M. W., D. I. Wieler, and F. T. Poysky. 1967. Outgrowth and toxin production of non-proteolytic type B Clostridium botulinum at 3.3 to 5.6oC. J. Bacteriol. 93:1461–1462. Eklund, M. W., and V. R. Dowell, Jr. (ed.). 1987. Avian Botulism: An International Perspective. Charles C. Thomas, Springfield, IL. Esty, J. R., and K. F. Meyer. 1922. The heat resistance of the spores of B. botulinus and allied anaerobes. XI. J. Infect. Dis. 31:650–653. Fagan, R. P., J. B. McLaughlin, and J. P. Middaugh. 2009. Persistence of botulinum toxin patients’ sera, 1959–2009. J. Infect. Dis. 199:1029–1031. Foster, E. M. 1997. Historical overview of key issues in food safety. Emerg. Infect. Dis. 3:481–482. Franciosa, G., J. L. Ferreira, and C. L. Hatheway. 1994. Detection of type-A, type-B, and type-E botulism neurotoxin genes in Clostridium botulinum and other Clostridium species by PCR—evidence of unexpressed type-B toxin genes in type-A toxigenic organisms. J. Clin. Microbiol. 32:1911–1917. Franciosa, G., P. Aureli, P., and R. Schechter. 2003. Clostridium botulinum, p. 61–89. In M. D. Miliotis and J. W. Bier (ed.), International Handbook of Foodborne Pathogens. Marcel Dekker, New York, NY.

SMP_Food Microbiology_CH17.indd

460

43. Franciosa, G., A. Maugliani, C. Scalfaro, and P. Aureli. 2009. Evidence that plasmid-borne botulinum neurotoxin type B genes are widespread among Clostridium botulinum serotype B strains. PLoS ONE 4:e4829. 44. Franciosa, G., C. Scalfaro, P. di Bonito, M. Vitale, and P. Aureli. 2011. Identification of novel linear megaplasmids carrying a beta-lactamase gene in neurotoxigenic Clostridium botulinum type E strains. PLoS ONE 6: 21706. 45. Ghoddusi, H., and R. Sherburn. 2010. Preliminary study on the isolation of Clostridium butyricum strains from natural sources in the UK and screening of the isolates for the presence of the type E botulinal toxin gene. Int. J. Food Microbiol. 142:202–206. 46. Giménez, D. F., and J. A. Giménez. 1993. Serological subtypes of botulinal neurotoxins, p. 421–431. In B. R. Dasgupta (ed.), Botulism and Tetanus Neurotoxins: Neurotransmission and Biomedical Aspects. Plenum Press, New York, NY. 47. Glass, K. A., and E. A. Johnson. 2002. Formulating low-acid foods for botulinal safety, p. 323–350. In V. K. Juneja and J. N. Sofos (ed.), Control of Foodborne Microorganisms. Marcel Dekker, New York, NY. 48. Hallis, B., B. A. F. James, and C. C. Shone. 1996. Development of novel assays for botulinum type A and B neurotoxins based on their endopeptidase activity. J. Clin. Microbiol. 34:1934–1938. 49. Hammes, W. P., and P. S. Tichaczek. 1994. The potential of lactic acid bacteria for the production of safe and wholesome food. Z. Lebensm. Unters. Forsch. 198:193–201. 50. Hatheway, C. L. 1988. Botulism, p. 111–133. In A. Balows et al. (ed.), Laboratory Diagnosis of Infectious Diseases: Principles and Practice. Springer-Verlag, New York, NY. 51. Hatheway, C. L. 1993. Clostridium botulinum and other organisms that produce botulinum neurotoxin, p. 3–20. In A. H. W. Hauschild and K. L. Dodds (ed.), Clostridium botulinum: Ecology and Control in Foods. Marcel Dekker, New York, NY. 52. Hatheway, C. L., and E. A. Johnson. 1998. Clostridium: the spore-bearing anaerobes, p. 731–782. In L. Collier, A. Balows, and M. Sussman (ed.), Topley & Wilson’s Microbiology and Microbial Infections, 9th ed., vol. 2. Systematic Bacteriology. Arnold, London, United Kingdom. 53. Hauschild, A. H. W. 1993. Epidemiology of foodborne botulism, p. 69–104. In A. H. W. Hauschild and K. L. Dodds (ed.), Clostridium botulinum: Ecology and Control in Foods. Marcel Dekker, New York, NY. 54. Hauschild, A. H. W. 1989. Clostridium botulinum, p. 111–189. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY. 55. Heap, J. T., S. A. Kuehne, M. Ehsaan, S. T. Cartman, C. M. Cooksley, J. C. Scott, and N. P. Minton. 2010. The ClosTron: Mutagenesis in Clostridium refined and streamlined. J. Microbiol. Methods 80:49–55. 56. Hill, K. K., T. J. Smith, C. H. Helma, L. O. Ticknor, B. T. Foley, R. T. Svensson, J. L. Brown, E. A. Johnson, L. A. Smith, R. T. Okinaka, P. J. Jackson, and J. D. Marks.

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

57. 58. 59. 60.

61.

62.

63. 64.

65. 66.

67.

68.

69. 70.

71.

461

2007. Genetic diversity among botulinum neurotoxinproducing clostridial strains. J. Bacteriol. 189:818–832. Holdeman, L. V., E. P. Cato, and W. E. C. Moore. 1979. Anaerobe Laboratory Manual, 4th ed. Virginia Polytechnic Institute and State University, Blacksburg, VA. Horowitz, B. Z. 2010. Type E botulism. Clin. Toxicol. 48:880–895. Horowitz, B. Z. 2011. The ripe olive scare and hotel Loch Maree tragedy: botulism under glass in the 1920’s. Clin. Toxicol. 49:345–347. Ito, K. A., D. J. Seslar, W. A. Mercern, and K. F. Meyer. 1967. The thermal and chlorine resistance of C. botu­ linum types A, B, and E spores. In M. Ingram and T. A. Roberts (ed.), Botulism 1966. Chapman and Hall, London, United Kingdom. Jacobson, M. J., G. Lin, W. Tepp, J. Dupuy, P. Stenmark, R. C. Stevens, and E. A. Johnson. 2011. Purification, modeling, and analysis of botulinum neurotoxin subtype A5 (BoNT/A5) from Clostridium botulinum strain A661222. Appl. Environ. Microbiol. 77:4217–4222. Johnson, E. A. 2005. Clostridium botulinum and Clostridium tetani, p. 1035–1088. In S. P. Borriello, P. R. Murray, and G. Funke (ed.), Topley and Wilson’s Microbiology and Microbial Infections, 8th ed. Hodder Arnold, London, United Kingdom. Johnson, E. A., and M. Bradshaw. 2001. Clostridium botulinum. A metabolic and cellular perspective. Toxicon 39:1703–1722. Johnson, E. A., and M. C. Goodnough. 1998. Botulism, p. 723–741. In L. Collier, A. Balows, and M. Sussman (ed.), Topley & Wilson’s Microbiology and Microbial Infections, 9th ed., vol. 2. Systematic Bacteriology. Arnold, London, United Kingdom. Johnson, E. A., and C. Montecucco. 2008. Botulism, p. 333–368. In A. Engel (ed.), Handbook of Clinical Neurology, vol. 91. Elsevier, New York, NY. Johnson, E. A., W. H. Tepp, M. Bradshaw, R. J. Gilbert, P. E. Cooke, and E. D. G. McIntosh. 2006. Characterization of Clostridium botulinum strains associated with an infant botulism case in the United Kingdom. J. Clin. Microbiol. 43:2602–2607. Johnson, E. A., P. Summanen, and S. M. Finegold. 2007. Clostridium, p. 889–910. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry, and M. A. Pfaller (ed.), Manual of Clinical Microbiology, 9th ed., vol. 2. ASM Press, Washington, DC. Jones, R. G., M. Ochiai, Y. Liu, T. Ekong, and D. Sesardic. 2008. Development of improved SNAP25 endopeptidase immuno-assays for botulinum type A and E toxins. J. Immunol. Methods 329:92–101. Jones, R. G. A., M. J. Corbel, and D. Sesardic. 2006. A review of WHO international standards for botulinum antitoxins. Biologicals 34:223–226. Jousimies-Somer, H. R., P. Summanen, D. M. Citron, E. J. Baron, H. M. Wexler, and S. M. Finegold. 2002. Anaerobic Laboratory Manual, 6th ed. Star Publishing Company, Belmont, CA. Kalb, S. R., H. Mooura, A. E. Boyer, L. G. McWilliams, J. L. Pirkle, and J. R. Barr. 2006. The use of Endopep-

SMP_Food Microbiology_CH17.indd

461

72. 73. 74. 75. 76.

77. 78.

79. 80. 81. 82.

83.

84.

85. 86. 87.

88. 89.

MS for the detection of botulinum neurotoxins in serum and stool samples. Anal. Biochem. 351:84–92. Kautter, D. A., S. M. Harmon, R. Y. Lynt, and T. Lilly. 1966. Antagonistic effect on Clostridium botulinum by organisms resembling it. Appl. Microbiol. 14:616–622. Keller, J. E, E. A. Neale, G. Oyler, and M. Adler. 1999. Persistence of botulinum neurotoxin action in cultured spinal cord cells. FEBS Lett. 456:137–142. Kempner, W. 1897. Further contributions to the knowledge of meat poisoning. The antitoxin to botulism. Z. Hyg. Infekt. Krankh. 26:481–500. (In German.) Koepke, R., J. Sobel, and S. S. Arnon. 2008. Global ­occurrence of infant botulism, 1976–2006. Pediatrics 122: e73–e82. Lacy, D. B., W. Tepp, A. C. Cohen, B. R. DasGupta, and R. C. Stevens. 1998. Crystal structure of botulinum neurotoxin type A and implications for toxicity. Nat. Struct. Biol. 5:898–902. Lambowitz, A. M., and S. Zimmerly. 2004. Mobile group II introns. Annu. Rev. Genet. 38:1–35. Larson, A. E., and E. A. Johnson. 1999. Evaluation of botulinal toxin production in packaged freshcut cantaloupe and honeydew melons. J. Food Prot. 62:948–952. Reference deleted. Reference deleted. Lindstrom, M., and H. Korkeala. 2006. Laboratory diagnostics of botulism. Clin. Microbiol. Rev. 46:86–94. Lindstrom, M., M. Nevas, K. Kurki, R. Sauna-Aho, A. Latvala-Kiesila, I. Polonen, and H. Korkeala. 2004. Type C botulism due to toxic feed affecting 52,000 farmed foxes and minks in Finland. J. Clin. Microbiol. 42:4718–4725. Ludwig, W., K.-H. Schleifer, and W. Whitman. 2009. Revised road map to the phylum Firmicutes, p. 1–13. In P. De Vos, G. M. Garity, D. Jones, N. R. Krieg, W. Ludwig, F. A. Rainey, K.-H. Schleifer, and W. B. Whitman (ed.), Bergey’s Manual of Systemic Bacteriology, 2nd ed., vol. 3. The Firmicutes. Springer, New York, NY. Lúquez, C., J. K. Dykes, R. A. Yu, B. H. Raphael, and S. E. Maslanka. 2010. First report of an infant botulism case due to Clostridium botulinum type E. J. Clin. Microbiol. 48:326–328. Lynt, R. K., D. A. Kautter, and R. B. Read, Jr. 1975. Botulism in commercially canned foods. J. Milk Food Technol. 38:546–550. Marshall, K. M., M. Bradshaw, and E. A. Johnson. 2010. Conjugative botulinum neurotoxin-encoding plasmids. PLoS ONE 5:e11087. Marshall, K. M., M. Bradshaw, S. Pellett, and E. A. Johnson. 2007. Plasmid encoded neurotoxin genes in Clostridium botulinum serotype A subtypes. Biochem. Biophys. Res. Commun. 361:49–54. McKellar, R. C., and X. Lu (ed.). 2004. Modeling Micro­ bial Responses in Foods. CRC Press, Boca Raton, FL. McLauchlin, K., A. Grant, and C. L. Little. 2006. Foodborne botulism in the United Kingdom. J. Public Health 28:337–342.

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

462 90. Meyer, K. F. 1956. The status of botulism as a world health problem. Bull. W. H. O. 15:281–298. 91. Meyer, K. F., and B. Eddie. 1950. Fifty years of botulism in the United States and Canada. George Williams Hooper Foundation, San Francisco, CA. 92. Montal, M. 2010. Botulinum neurotoxin: a marvel of protein design. Annu. Rev. Biochem. 79:591–617. 93. Montecucco, C., and G. Schiavo. 1995. Structure and function of tetanus and botulinum neurotoxins. Q. Rev. Biophys. 28:423–472. 94. Myllykoski, M. Lindström, E. Bekema, I. Pölönen, and H. Korkeala. 2011. Fur animal botulism due to feed. Res. Vet. Sci. 90:412–418. 95. NFPA/CMI Container Integrity Task Force, Microbio­ logical Assessment Group Report. 1984. Botulism risk from post-processing contamination of commercially canned foods in metal containers. J. Food Prot. 47:801–816. 96. Odlaug, T. E., and I. J. Pflug. 1978. Clostridium botuli­ num and acid foods. J. Food Prot. 41:566–573. 97. Onyenoke, R. U., J. A. Brill, K. Farahi, and J. Wiegel. 2004. Sporulation genes in members of the low G+C gram-type-positive phylogenetic branch (Firmicutes). Arch. Microbiol. 182:182–192. 98. Paredes, C. J., K. V. Alsaker, and E. T. Papoutsakis. 2005. A comparative genomic view of clostridial sporulation and physiology. Nat. Rev. Microbiol. 3:969–978. 99. Peck, M. W. 2009. Biology and genomic analysis of Clos­ tridium botulinum. Adv. Microb. Physiol. 55:183–265. 100. Peck, M. W., K. E. Goodburn, R. P. Betts, and S. C. Stringer. 2008. Assessment of the potential for growth and neurotoxin formation by non-proteolytic Clostridium botulinum in short shelf-life commercial foods designed to be chilled. Trends Food Sci. Technol. 19:201–216. 101. Peck, M. W., S. C. Stringer, and A. T. Carter. 2011. Clostridium botulinum in the post-genomic era. Food Microbiol. 28:183–191. 102. Pellett, S., W. H. Tepp, C. M. Clancy, G. E. Borodic, and E. A. Johnson. 2007. A neuronal cell-based botulinum neurotoxin assay for highly sensitive detection of neutralizing serum antibodies. FEBS Lett. 581:4803–4808. 103. Pellett, S., W. H. Tepp, S. I. Toth, and E. A. Johnson. 2010. Comparison of the primary rat spinal cord assay cell (RSC) assay and the mouse bioassay for botulinum neurotoxin type A potency determination. J. Pharmacol. Toxicol. Methods 61:304–310. 104. Pellett, S., Z. W. Zu, C. L. Pier, W. H. Tepp, S. C. Zhang, and E. A. Johnson. 2011. Sensitive and quantitative detection of botulinum neurotoxin in neurons derived from mouse embryonic stem cells. Biochem. Biophys. Res. Commun. 404:388–392. 105. Pflug, I. J. 2010. Science, practice, and human errors in controlling Clostridium botulinum in heat-preserved food in hermetic containers. J. Food Prot. 73:993–1002. 106. Pickett, J., B. Berg, E. Chaplin, and M. A. Brunstershafer. 1976. Syndrome of botulism in infancy—clinical and electrophysiological study. N. Engl. J. Med. 295:770–772.

SMP_Food Microbiology_CH17.indd

462

107. Pitt, M. L. M., and R. D. LeClaire. 2005. Pathogenesis by aerosol, p. 65–78. In L. E. Lindler, F. J. Lebeda, and G. W. Korch, (ed.), Biological Weapons Defense. Infectious Diseases and Counterbioterrorism. Humana Press, Totowa, NJ. 108. Rasooly, R., and P. M. Do. 2010. Clostridium botuli­ num neurotoxin type B is heat-stable in milk and not inactivated by pasteurization. J. Agric. Food Chem. 58:12557–12561. 109. Rebagliati, V., S. Chianelli, M. Tornese, L. Rossi, and A. Troncoso. 2008. Documented outbreaks of botulism: the impact of food-borne transmission. Asian Pac. J. Trop. Med. 1:71–75. 110. Reddy, N. R., H. M. Solomon, R. C. Tetzloff, and E. J. Rhodehamel. 2003. Inactivation of Clostridium botulinum type A spores by high-pressure processing at ­elevated temperatures. J. Food Prot. 66:1402–1407. 111. Roberts, T. A. 1997. Maximizing the usefulness of food microbiology research. Emerg. Infect. Dis. 3:523–528. 112. Rodgers, S. 2004. Novel approaches in controlling safety of cook-chill meals. Trends Food Sci. Technol. 15:366–372. 113. Rodolico, C., E. Barca, L. Fenicia, F. Anniballi, A. U. Sinardi, and P. Girlanda. 2010. Wound botulism in drug users: a still underestimated diagnosis. Neurol. Sci. 31:825–827. 114. Rowlands, R. E. G., C. A. Ristori, G. I. S. Lopez, A. M. R. de Paula, H. Sakuma, R. Grigaliunas, R. L. Filho, D. S. Gelli, M. B. Eduardo, and M. Jakabi. 2010. Botulism in Brazil, 2000–2008: epidemiology, clinical findings and laboratory diagnosis. Rev. Inst. Trop. Sao Paulo 54:183–186. 115. Sachdeva, A., S. L. H. Defibaugh-Chávez, J. B. Day, D. Zink, and S. K. Sharma. 2010. Detection and confirmation of Clostridium botulinum in water used for cooling at a plant producing low-acid canned foods. Appl. Environ. Microbiol. 76:7653–7657. 116. Sakaguchi, G. 1983. Clostridium botulinum toxins. Pharmacol. Ther. 19:165–194. 117. Sanford, D. C., R. Barnewall, M. L. Vassar, N. Niemuth, K. Metcalfe, R. V. House, I. Henderson, and J. D. Shearer. 2010. Inhalational botulism in rhesus macaques exposed to botulinum neurotoxin complex serotypes A1 and B1. Clin. Vaccine Immunol. 17:1293–1304. 118. Schantz, E. J., and E. A. Johnson. 1992. Properties and use of botulinum toxin and other microbial neurotoxins in medicine. Microbiol. Rev. 56:80–99. 119. Schantz, E. J., and D. A. Kautter. 1978. Microbiological methods: standardized assay for Clostridium botulinum toxins. J. Assoc. Off. Anal. Chem. 61:96–99. 120. Schiavo, G., M. Matteoli, and C. Montecucco. 2000. Neurotoxins affecting neuroexocytosis. Physiol. Rev. 80:717–766. 121. Shapiro, R. L., C. L. Hatheway, and D. L. Swerdlow. 1998. Botulism in the United States: a clinical and epidemiologic review. Ann. Intern. Med. 129:221–228. 122. Sheth, A. N., P. Wiersma, D. Atrubin, V. Dubey, D. Zink, G. Skinner, F. Doer, P. Juliao, G. Gonzalez, C. Burnett, C. Drenzek, C. Shuler, J. Austin, A. Ellis, S. Maslanka, and J. Sobel. 2008. International outbreak of

Manila Typesetting Company

11/08/2012 07:00AM

17.  Clostridium botulinum

123.

124.

125.

126.

127. 128.

129. 130. 131.

132. 133. 134.

135.

463

severe botulism caused by commercial carrot juice. Clin. Infect. Dis. 47:1245–1251. Siegel, L. S. 1993. Destruction of botulinum toxins in food and water, p. 323–341. In A. H. W. Hauschild and K. L. Dodds (ed.), Clostridium botulinum. Ecology and Control in Foods. Marcel Dekker, New York, NY. Skarin, H., T. Hafstrom, J. Westerberg, and B. Segerman. 2011. Clostridium botulinum group III: A group with dual identity shaped by plasmids, phages and mobile elements. BMC Genomics 12:185. Sleator, R. D., and C. Hill. 2008. Molecular analysis of the microbial food safety implications of food reformulations for improved health. Foodborne Pathog. Dis. 5:499–504. Smeldt, J. P. P. M., P. C. Hellemons, P. C. Wouters, and S. J. C. van Gerwen. 2002. Physiological and mathematical aspects in setting criteria for decontamination of foods by physical means. Int. J. Food Microbiol. 78:57–77. Smith, L. D. S., and H. Sugiyama. 1988. Botulism. The Organism, Its Toxins, the Disease. Charles C. Thomas, Springfield, IL. Smith, T. J., J. Lou, I. N. Geren, C. M. Forsyth, R. Tsai, S. L. LaPorte, W. H. Tepp, M. Bradshaw, E. A. Johnson, L. A. Smith, and J. D. Marks. 2005. Sequence variation with botulinum neurotoxin serotypes impacts antibody binding and neutralization. Infect. Immun. 73:5450–5457. Sobel, J., N. Tucker, A. Sulka, J. McLaughlin, and S. Maslanka. 2004. Foodborne botulism in the United States, 1990–2004. Emerg. Infect. Dis. 10:1606–1611. Solomon, H. A., and D. A. Kautter. 1988. Outgrowth and toxin production by Clostridium botulinum in bottled, chopped garlic. J. Food Prot. 51:862–865. Solomon, H. M., E. A. Johnson, D. T. Bernard, S. S. Arnon, and J. L. Ferreira. 2001. Clostridium botulinum and its toxins, p. 317–324. In F. P. Downes and K. Ito (ed.), Compendium for the Microbiological Examination of Foods, 4th ed. American Public Health Association, Washington, DC. Sugiyama, H. 1980. Clostridium botulinum neurotoxin. Microbiol. Rev. 44:419–448. Sugiyama, H. 1990. Botulism, p. 107–125. In D. O. Cliver (ed.), Foodborne Diseases. Academic Press, San Diego, CA. Tanaka, N., E. Traisman, P. Plantinga, L. Finn, W. Flom, L. Meske, and J. Guggisberg. 1986. Evaluation of factors involved in antibotulinal properties of pasteurized process cheese spreads. J. Food Prot. 49:526–53l. Therre, H. 1999. Botulism in the European Union. Euro Surveill. 4:2–7.

SMP_Food Microbiology_CH17.indd

463

136. Topakaian, R., C. Heibl, K. Stieglbauer, B. Dreer, M. Nagl, P. Knoflaxh, and F. T. Aichner. 2009. Quantitative autonomic testing in the management of botulism. J. Neurol. 256:803–809. 137. Tsukamoto, K., M. Mikamoto, T. Kohda, H. Ihara, S. Wang, T. Maegawa, S. Nakamura, T. Karasawa, and S. Kozaki. 2002. Characterization of Clostridium butyr­ icum neurotoxin associated with food-borne botulism. Microb. Pathog. 33:177–184. 138. U.S. Food and Drug Administration. 2001. Bacterio­ logical Analytical Manual (Online). http://www.cfsan. fda.gov/~ebam/bam-toc.html. 139. van Ermengem, E. 1979. Classics in infectious disease. A new anaerobic bacillus and its relation to botulism. Rev. Infect. Dis. 1:701–719. Originally published in 1897 as: Ueber einen neuen anaeroben Bacillus und seine Beziehungen zum Botulismus. Z. Hyg. Infektionskr. 26:1–56. 140. Villar, R. G., R. L. Shapiro, S. Busto, C. Riva-Posse, G. Verdejo, M. I. Farace, R. Rossetti, J. A. San Juan, C. M. Julia, J. Becher, S. E. Maslanka, and D. L. Swerdlow. 1999. Outbreak of type A botulism and development of a botulism surveillance and antitoxin r­elease system in Argentina. JAMA 281:1334–1338. 141. Webster, K., C. Jardine, S. B. Cash, and L. M. McMullen. 2010. Risk ranking: investigating expert and public differences in evaluating food safety hazards. J. Food Prot. 73:1875–1885. 142. Wein, L. M., and Y. Liu.. 2005. Analyzing a bioterror attack on the food supply: the case of botulinum toxin in milk. Proc. Natl. Acad. Sci USA 102:9984–9989. 143. Weingart, O. G., T. Schreiber, C. Mascher, D. Pauly, M. B. Dorner, T. F. H. Berger, C. Effer, F. Gessler, M. J. Loessner, M.-A. Avondet, and B. G. Dorner. 2010. The case of botulinum toxin in milk: experimental data. Appl. Environ. Microbiol. 76:3293–3300. 144. Whiting, R. C., and K. A. Naftulin. 1992. Effect of headspace oxygen concentration on growth and toxin production by proteolytic strains of Clostridium botuli­ num. J. Food Prot. 55:23–27. 145. Whitmer, M. E., and E. A. Johnson. 1988. Development of improved defined media for Clostridium botulinum serotypes A, B, and E. Appl. Environ. Microbiol. 54:753–759. 146. Woodburn, M. J., E. Somers, J. Rodriguez, and E. J. Schantz. 1979. Heat inactivation rates of botulinum toxins A, B, E, and F in some foods and buffers. J. Food Sci. 44:1658–1661. 147. Zhang, J.-C., L. Sun, and Q.-H. Nie. 2010. Botulism, where are we now. Clin. Toxicol. 48:867–879.

Manila Typesetting Company

11/08/2012 07:00AM

SMP_Food Microbiology_CH17.indd

464

Manila Typesetting Company

11/08/2012 07:00AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch18

Bruce A. McClane Susan L. Robertson Jihong Li

18

Clostridium perfringens

Clostridium perfringens was first recognized as an important cause of foodborne disease in the 1940s and 1950s (34). It later became apparent that C. perfringens causes two quite different human foodborne diseases, i.e., C. perfringens type A food poisoning and enteritis necroticans (also known as Darmbrand or Pig-Bel). Since foodborne enteritis necroticans is rare in industrialized societies, this chapter focuses mainly on C. perfringens type A food poisoning; details regarding enteritis necroticans are available elsewhere (36).

CHARACTERISTICS OF THE BACTERIUM Clostridium perfringens is a gram-positive, rod-shaped, encapsulated, nonmotile anaerobe that causes a spectrum of human and veterinary diseases (34, 47). The virulence of this bacterium largely results from its prolific toxin-producing ability, including several toxins (e.g., C. perfringens enterotoxin [CPE] and b-toxin) with activity on the human gastrointestinal (GI) tract (47). However, individual C. perfringens cells vary in their toxin gene carriage, providing the basis for a toxin typing system that classifies C. perfringens isolates (47) into one of five types (A through E), depending upon their

ability to express the four “typing” toxins (alpha, beta, epsilon, and iota) (Table 18.1). Whereas C. perfringens toxin typing once involved laborious toxin-antiserum neutralization tests in mice, multiplex-PCR-based toxin genotyping of C. perfringens isolates is now typically performed instead. Distinct toxin types are associated with each of the two foodborne diseases caused by C. perfringens. Enteritis necroticans, a life-threatening illness, is usually caused by type C isolates, with b-toxin considered the primary virulence factor responsible for this illness (66). C. perfringens type A food poisoning, typically self-­limiting, is (as implied by its name) associated with type A isolates producing CPE. This enterotoxin is encoded by a gene (cpe) located on either the chromosome or large plasmids (11, 12, 74; see also “Genetics of CPE” below). About 75% of food poisoning cases in the United States and Europe involve type A isolates carrying a chromosomal cpe gene. In addition to producing toxins active in the GI tract, C. perfringens possesses several other characteristics favoring its ability to cause foodborne disease. First, this bacterium has an exceptionally short doubling time, which allows rapid multiplication in foods to reach a

Bruce A. McClane, Susan L. Robertson, and Jihong Li, Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, 420 Bridgeside Point II Bldg., 450 Technology Dr., Pittsburgh, PA 15219.

465

SMP_Food Microbiology_CH18.indd

465

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

466 Table 18.1  Toxinotyping of C. perfringens strains Toxin production Type

Alpha

Beta

Epsilon

Iota

A

+

-

-

-

B

+

+

+

-

C

+

+

-

-

D

+

-

+

-

E

+

-

-

+

pathogenic burden. Type A chromosomal cpe isolates are especially fast growing, with a mean doubling time of <10 min at the optimum growth temperature of 43°C (40). Second, C. perfringens forms spores tolerant of food environment stresses such as radiation, desiccation, freezing, refrigeration, and heat (34), which favors survival in incompletely cooked or inadequately held foods. As discussed in detail in the next section, vegetative cells and spores of chromosomal cpe food poisoning strains possess particularly strong stress resistance characteristics (38, 40, 65). C. perfringens is considered an “anaerobe” because it does not produce colonies on agar plates continuously exposed to air (34). However, this bacterium tolerates moderate exposure to oxygen. Compared to most other anaerobes, C. perfringens requires only relatively modest reductions in oxidation-reduction potential (Eh) for growth (34).

Influence of Intrinsic and Extrinsic Factors of Foods on C. perfringens

C. perfringens growth in food is affected by a variety of environmental factors, including temperature, Eh, pH, and water activity (aw).

Temperature

Spore resistance against heat and cold contributes to the ability of C. perfringens to cause food poisoning by facilitating survival in undercooked foods or during low-temperature storage. The heat resistance of C. perfringens spores is influenced by both environmental and genetic factors. An example of an environmental influence is the medium in which C. perfringens spores are heated (34, 65). The variations in spore heat resistance among different C. perfringens isolates reflect genetic differences that influence the bacterium’s thermal resistance. Of particular note, spores of type A chromosomal cpe isolates average 60-fold-greater heat resistance at 100°C than spores of other C. perfringens strains (34, 65); the molecular basis for the exceptional spore heat resistance of type A chromosomal cpe isolates is dis-

SMP_Food Microbiology_CH18.indd

466

cussed below (see “Virulence Factors Contributing to C. perfringens Type A Food Poisoning”). Importantly, incomplete cooking of foods not only fails to kill C. perfringens spores in foods but can favor development of C. perfringens type A food poisoning by inducing spore germination. The vegetative cells of type A chromosomal cpe isolates are also more heat resistant than other C. perfringens vegetative cells. For example, at 55°C, vegetative cells of chromosomal cpe isolates have approximately twofold-greater survival than other C. perfringens vegetative cells (65). In addition, the vegetative cells of type A chromosomal cpe isolates have higher maximum growth temperatures (~53°C) than other C. perfringens vegetative cells (40). While C. perfringens vegetative cells are not notably tolerant of cold temperatures, the cells of type A chromosomal cpe strains survive refrigeration and freezing much better than other C. perfringens vegetative cells (34, 40). Growth rates for all C. perfringens strains rapidly decrease markedly at lower temperatures (40), but type A chromosomal cpe isolates possess lower minimum growth temperatures (~12°C) than other C. perfringens strains. However, no known C. perfringens strains can grow at 6°C (34). C. perfringens spores are considerably more cold resistant than vegetative cells (34). This resistance is particularly notable for spores of type A chromosomal cpe food poisoning strains (40), which survive refrigeration and freezing much better than spores of other C. perfringens strains (Table 18.2). Cold resistance properties Table 18.2  Spore resistance properties against food environ­

ment stressesa

Avg survival value for type A: Treatment, time (unit for survival value)

Chromosomal cpe isolates

Plasmid cpe isolates

61.1

0.9

4°C, 6 mo

  0.3

1.2

−20°C, 3 mo

  0.8

1.5

NaCl, 6 mo

  0.2

1.1

NaNO2, 3 mo

  0.9

3.5

Temp Heat at 100°C (D value in min) Cold (log reduction)

Chemical treatment (log reduction)

Spores of type A cpe-negative isolates have resistance properties similar to those of spores of type A plasmid cpe isolates. Data compiled from references 38, 40, and 64. a

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

467

of vegetative cells and spores of type A cpe-positive isolates, particularly chromosomal cpe isolates, may foster food poisoning if refrigerated or frozen foods are later inadequately warmed for serving.

Other Environmental Factors

C. perfringens growth in food is also affected by aw, Eh, pH, and probably the presence of curing agents (34). The lowest aw supporting growth of C. perfringens is 0.93 when other growth conditions are near optimal (34). For an anaerobe, C. perfringens does not require a highly reduced environment for growth. Provided the environmental Eh is suitably low for initiating growth (the exact Eh value needed to initiate C. perfringens growth depends on environmental factors such as pH), C. perfringens can produce reducing molecules such as ferredoxin to modify the Eh of its environment and create favorable growth conditions (34). As a practical guide for food microbiologists, the Eh of many common foods such as raw meats and gravies is often sufficient to support the growth of C. perfringens (34). It has not yet been determined whether type A chromosomal cpe isolates are more tolerant of Eh conditions than other C. perfringens isolates. Growth of C. perfringens is also pH sensitive, with optimal growth occurring at pH 6 to 7. C. perfringens grows poorly, if at all, at pH values of £5 and ³8.3 (34). Vegetative cells and spores of type A chromosomal cpe isolates have pH sensitivity similar to those of other C. perfringens isolates (38). At commercially applicable concentrations, the effectiveness of curing agents in limiting C. perfringens growth in foods is an unsettled subject (34). Early studies revealed that curing salt concentrations necessary for significantly affecting C. perfringens growth exceed commercially relevant levels, e.g., growth inhibition of C. perfringens reportedly requires at least 6 to 8% NaCl and 10,000 ppm NaNO3 or 400 ppm NaNO2. In contrast, later studies determined that curing salts can be at least partially effective at preventing C. perfringens growth in food, even when used at commercially acceptable levels. For example, (i) coapplication with other preservation factors, such as heating and acidic pH, can increase C. perfringens sensitivity to curing salts; (ii) simultaneous use of curing agents with other antimicrobials can exert a synergistic effect to inhibit C. perfringens growth; and (iii) foods often contain lower initial populations of C. perfringens cells and spores than those used in laboratory studies evaluating the effectiveness of curing agents for inhibiting C. perfringens growth. One practical argument supporting the ability of curing agents to influence C. perfringens growth in

SMP_Food Microbiology_CH18.indd

467

foods is the relatively rare association of commercially cured meat products with C. perfringens type A food poisoning outbreaks (34). Recent studies revealed that the vegetative cells and spores of type A chromosomal cpe isolates survive better than those of other C. perfringens strains under osmo­ tic stress from NaCl or killing by nitrites (Table 18.2). For spores, resistance against nitrite-induced killing involves the same molecular mechanism that mediates the exceptional temperature resistance properties of spores of type A chromosomal cpe isolates. Preservation factors such as pH, aw, and (likely) curing agents also control C. perfringens populations in foods by inhibiting the outgrowth of germinating C. perfringens spores (34). However, ungerminated spores can remain viable in foods containing preservation factors that prevent cell growth; those spores may later undergo germination/outgrowth if the growth-inhibiting factor(s) is removed during food preparation.

Genetic Basis for the Resistance and Growth Phenotypes of Type A Chromosomal cpe Isolates

The broad nature of their exceptional growth and resistance properties suggested that the type A chromosomal cpe isolates commonly causing food poisoning might be a sublineage of C. perfringens, even distinct from type A plasmid cpe isolates (which usually have growth and resistance phenotypes similar to those of cpe-negative C. perfringens type A isolates). Multilocus sequence typing analyses involving sequence comparisons of eight housekeeping genes recently confirmed that type A chromosomal cpe isolates differ from other C. perfringens strains (16). Based upon those multilocus sequence typing findings, chromosomal cpe isolates may have arisen from integration of a mobile genetic element carrying the cpe gene onto the chromosome of a C. perfringens isolate already possessing survival traits favorable for the food environment. If so, this acquisition event has produced an exceptionally fit foodborne pathogen.

RESERVOIRS FOR C. PERFRINGENS TYPE A FOOD POISONING STRAINS C. perfringens is widely distributed throughout the natural environment (34), including soil (at levels of 103 to 104 CFU/g), foods (e.g., approximately 50% of raw or frozen meat contains C. perfringens), dust, and the intestinal tract of humans and domestic animals (e.g., human feces can contain 104 to 106 C. perfringens/g). The widespread natural occurrence of C. perfringens is a contributing factor to the frequent occurrence of C. perfringens

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

468 type A food poisoning outbreaks. However, <5% of global C. perfringens isolates carry the enterotoxin gene (cpe) necessary for causing C. perfringens type A food poisoning. Therefore, simply determining where any C. perfringens isolates reside in nature clearly has limited significance for understanding the specific reservoirs for C. perfringens type A food poisoning isolates. Reservoirs have been identified for type A isolates carrying a plasmid-borne cpe gene, which apparently cause approximately 25% of food poisoning cases. These strains are present in soil and the intestines of healthy human carriers (5, 43). In contrast, the ecology of the type A chromosomal cpe isolates causing most food poisoning cases remains poorly understood. Those chromosomal cpe isolates have not been identified in soil (5, 43) and are only rarely carried by healthy human carriers (5, 24). Interestingly, type A chromosomal cpe isolates have been identified in ~1.4% of raw meats sold at retail stores in the United States (79), including turkey, chicken, pork, and fish. Those chromosomal cpe food isolates were highly heat resistant and capable of producing CPE, indicating that C. perfringens isolates with full food poisoning potential are present in some U.S. meats at the time of retail purchase. The presence of these cpe-positive, highly stress-resistant isolates in raw meats also indicates that the heat resistance properties described in the preceding section of this chapter are intrinsic traits of type A chromosomal cpe isolates, rather than resulting from selection of survivors during cooking or warming of foods. Clearly, current knowledge of the reservoir(s) for C. perfringens type A food poisoning isolates remains deficient, which is unfortunate because it impairs efforts to rationally control/reduce outbreaks of C. perfringens type A food poisoning.

C. PERFRINGENS TYPE A FOOD POISONING OUTBREAKS

Incidence

The most recent statistics from the Centers for Disease Control and Prevention (7) rank C. perfringens type A food poisoning as the second most commonly identified bacterial cause of foodborne disease outbreaks in the United States. In 2007, 45 confirmed or suspected C. perfringens type A food poisoning outbreaks (representing 4% of total foodborne disease outbreaks) occurred in the United States, involving 2,062 cases (7% of total cases of foodborne diseases). Since most C. perfringens type A food poisoning cases remain unidentified, these official CDC statistics understate the true incidence and

SMP_Food Microbiology_CH18.indd

468

impact of this foodborne disease. Conservative estimates are that 250,000 cases of C. perfringens type A foodborne illness occur annually in the United States, causing an average of 7 deaths per year (49). Economic costs associated with C. perfringens type A food poisoning exceed $120 million/year in 1989 dollars (76). Identified C. perfringens type A food poisoning outbreaks are usually large (with an average outbreak size of ~50 to 100 cases) and typically occur in institutionalized settings. The large size of most recognized C. perfringens type A foodborne disease outbreaks is attributable to two factors: (i) the fact that foods at large institutions are often prepared in advance and then held for later serving, thereby allowing growth of C. perfringens if those foods are temperature abused; and (ii) the relatively mild and nondistinguishing symptoms of most C. perfringens type A food poisoning cases, which usually results in public health officials becoming involved in investigating and reporting this foodborne illness only when large numbers of people become ill. C. perfringens type A food poisoning occurs throughout the year but is slightly more common during the summer, perhaps because higher ambient temperatures facilitate temperature abuse of foods during cooling and holding.

Food Vehicles for C. perfringens Type A Food Poisoning

The leading food vehicles for C. perfringens type A poisoning in the United States are meats and poultry (34). Meat-containing products, e.g., gravies and stews, and Mexican foods are other important vehicles for C. perfringens type A foodborne illness.

Contributing Factors Leading to C. perfringens Type A Food Poisoning

C. perfringens type A food poisoning usually results from temperature abuse during the cooking, cooling, or holding of foods. Improper storage or holding temperatures contributes to nearly all C. perfringens type A food poisoning outbreaks for which contributing factors were identified. Improper cooking also contributes to about one-third of these outbreaks, whereas use of contaminated equipment is a factor in a few outbreaks. The importance of temperature abuse in C. perfringens type A food poisoning is not surprising, considering the aforementioned exceptional heat resistance of C. perfringens vegetative cells and spores, particularly those of type A chromosomal cpe isolates. Undercooking or improperly holding foods can increase the likelihood of illness by inducing germination of, yet not killing, C. perfringens spores present in foods (34). Vegetative

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

469

cells resulting from these germinated spores can then rapidly multiply to the populations necessary to cause foodborne illness if the foods are improperly cooled or stored.

Prevention and Control of C. perfringens Type A Food Poisoning

Thorough cooking of food is important to prevent/ control C. perfringens type A food poisoning. This is particularly true for large roasts and turkeys because, due to their size, it is difficult to obtain the high internal temperatures needed to kill C. perfringens spores. The difficulty of cooking large roasts and whole poultry carcasses to sufficiently high internal temperatures explains, in part, why those foods are such common food vehicles for C. perfringens type A food poisoning. A second, perhaps even more important, step in preventing C. perfringens type A foodborne illness is to rapidly cool cooked foods and then store/serve those foods at nonpermissive conditions for C. perfringens growth (e.g., storage at either refrigeration temperature or temperatures >60°C).

Example of a Recent C. perfringens Type A Food Poisoning Outbreak

Beginning the morning of 7 May 2010, more than 40 patients and staff at a Louisiana mental health care facility developed GI symptoms (59). During this outbreak, three fatalities occurred, including a 43-year-old woman, a 41-year-old man, and a 52-year-old man. Epidemiologic questionnaires implicated chicken salad as the food vehicle, with people ingesting that item being 23 times more likely to show symptoms. It was also determined the chicken salad had been improperly stored prior to serving. Testing confirmed the outbreak as C. perfringens type A food poisoning. When this chapter was being written, a complete report of this outbreak had not been published. However, this outbreak is described here because it illustrates several typical features of a C. perfringens type A food poisoning outbreak, as well as two important atypical features. With respect to typical features, it is notable that this outbreak occurred in an institutional setting and was large, involving many cases. The food vehicle, a poultry product, is also typical of C. perfringens food poisoning outbreaks. This Louisiana outbreak also involved a high fatality rate, which is unusual for C. perfringens type A food poisoning outbreaks. Particularly atypical is that these fatalities occurred in middle-aged people, while most fatalities from C. perfringens type A food poisoning involve the elderly. The high fatality rate in middle-aged

SMP_Food Microbiology_CH18.indd

469

people of this outbreak is reminiscent of an unusual C. perfringens type A food poisoning outbreak that occurred in 2001 in an Oklahoma residential care facility for the mentally challenged (1). Three of the 7 victims in that outbreak developed necrotizing enteritis, resulting in two fatalities in middle-aged persons. The severity of the 2001 Oklahoma outbreak was attributed, at least in part, to psychiatric drug therapy, which produced constipation and fecal impaction as side effects. Fecal impaction would prevent the expulsion of CPE from the intestines through the usual protective effects of diarrhea, allowing CPE more contact time to damage the GI tract of the patients. Consistent with this explanation, severe colonic necrosis was observed at autopsy of the deceased patients. Whether similar circumstances contributed to the 2010 Louisiana C. perfringens food poisoning outbreak has not yet been reported. Regardless, the severity of these two outbreaks highlights the critical importance of proper food preparation and storage at mental health facilities.

Identification of C. perfringens Type A Food Poisoning Outbreaks

Public health agencies use descriptive criteria, such as incubation time, symptoms, food vehicle history (e.g., involvement of a temperature-abused meat/poultry food vehicle), and food consumption based on case-control comparisons to help identify C. perfringens type A food poisoning outbreaks. However, clinical features alone are not sufficient for identifying these outbreaks, given the similarities between the onset times and symptomology of C. perfringens type A food poisoning and certain other foodborne illnesses (particularly the diarrheal form of Bacillus cereus food poisoning). In response, public health agencies usually include laboratory analyses for more reliable identification of C. perfringens type A food poisoning outbreaks. The most specific approach for diagnosing this food poisoning is fecal CPE detection. Several commercially available serologic assays are available for fecal CPE detection, including a reverse-passive latex agglutination assay (Oxoid) and rapid enzyme-linked immunosorbent assays (Tech Lab or R-Biopharm). CPE is also present in the feces of some people suffering from nonfoodborne GI tract diseases (such as antibiotic-associated diarrhea), so detecting CPE in feces from a single ill individual is insufficient to establish a C. perfringens type A food poisoning outbreak. However, demonstrating the presence of CPE in feces from several epidemiologically associated ill individuals provides strong evidence for a C. perfringens type A food poisoning outbreak, particularly when those individuals consumed a common food and

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

470 developed illness within the normal incubation period for this food poisoning. Since CPE can be labile in feces, a limitation to the use of fecal CPE detection approaches for identifying C. perfringens food poisoning outbreaks is the need for rapid collection of fecal samples after the onset of food poisoning symptoms. Bacteriologic criteria can also be used to support identification of C. perfringens type A food poisoning outbreaks. Such approaches can involve demonstrating the presence of large numbers of C. perfringens in stool from two or more ill persons or in epidemiologically implicated food. However, there are limitations to simply demonstrating the presence of C. perfringens in suspect food or feces for identifying an outbreak of C. perfringens type A food poisoning. CPE-­negative C. perfringens isolates are widely distributed in the environment, including a presence, sometimes at high levels, in foods or in feces from healthy people. In theory, demonstrating the presence of CPE-producing C. perfringens food poisoning isolates in foods or in feces represents a more specific microbiology-based ­ approach for diagnosing C. perfringens type A food poisoning outbreaks. However, while CPE serologic detection assays can be used to evaluate the enterotoxigenic potential of a food/fecal isolate in vitro, this approach can be challenging in practice because an isolate must sporulate in laboratory medium to demonstrate CPE expression (CPE expression is sporulation associated) and in vitro C. perfringens sporulation is often difficult to achieve (31). Alternative microbiology-based approaches include PCR assays, which can not only identify cpe-­positive isolates but also distinguish whether a cpe-positive isolate carries a plasmid-borne cpe gene or the chromosomal cpe gene more commonly associated with food poisoning. However, PCR assays have the obvious limitation that they may detect silent cpe genes, leading to false-positive conclusions. Despite their limitations, microbiologic approaches can provide useful supplemental information for epidemiologic investigations, particularly when several approaches are employed together. For example, investigators of the 2001 Oklahoma outbreak (discussed above) conclusively identified that outbreak as C. perfringens type A food poisoning by (i) using a multiplex toxin genotyping PCR to determine that fecal isolates obtained from the outbreak victims were cpe positive and belonged to genotype A, (ii) using a cpe genotyping PCR assay to establish that those type A disease isolates carried the chromosomal cpe gene typical of food poisoning isolates, and (iii) using Western blotting to confirm that those disease isolates expressed CPE when induced to sporulate in vitro.

SMP_Food Microbiology_CH18.indd

470

In summary, the laboratory plays an increasingly important role in identifying C. perfringens type A food poisoning outbreaks. Within the proper epidemiologic/ clinical context, demonstrating the presence of CPE in fecal specimens obtained from several ill individuals provides compelling evidence for a C. perfringens type A food poisoning outbreak. Newer approaches, particularly PCR-based assays, are also increasingly useful as epidemiologic tools for investigating these outbreaks.

CHARACTERISTICS OF C. PERFRINGENS TYPE A FOOD POISONING Symptoms of C. perfringens type A food poisoning generally develop about 8 to 18 hours after ingestion of contaminated food and then resolve spontaneously within the next 12 to 24 hours (47). Victims of C. perfringens type A food poisoning almost always suffer diarrhea and abdominal cramps, with vomiting and fever being more variable symptoms. While death rates from C. perfringens type A food poisoning are usually low, fatalities can occur in people who are debilitated, elderly, or medicated. The typical pathogenesis of C. perfringens type A food poisoning is illustrated in Fig. 18.1. Initially, a food becomes contaminated with a cpe-positive C. perfringens type A isolate. If the food is temperature abused, the bacteria rapidly multiply until they are consumed with the contaminated food. Many of the ingested C. perfringens vegetative cells die when exposed to stomach acidity, but if the food vehicle was contaminated with >106 C. perfringens cells/g, some ingested bacteria survive passage through the stomach and remain viable when entering the small intestine, where they multiply and sporulate. CPE is expressed by these sporulating C. perfringens cells and is eventually released into the intestinal lumen, where the sporulating cells lyse to release their endospores. Once released, CPE quickly binds to intestinal epithelial cells and exerts its action, which induces intestinal tissue damage. This CPE-induced intestinal tissue damage initiates intestinal fluid loss, which clinically manifests as diarrhea. Two factors probably help to explain why most C. perfringens type A food poisoning cases are relatively mild and self-limited: (i) CPE preferentially affects villus tip cells (70), which are the oldest intestinal cells and can be rapidly replaced in young, healthy individuals by normal turnover of intestinal cells; and (ii) the diarrhea associated with C. perfringens type A food poisoning probably helps to mitigate the severity of illness by flushing unbound CPE and many sporulating

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

471

Figure 18.1  Pathogenesis of C. perfringens type A food poisoning. Vegetative cells of an enterotoxin (CPE)-producing C. perfringens strain multiply rapidly in contaminated food (usually a meat or poultry product) and, after ingestion, sporulate in the small intestine. Sporulating C. perfringens cells then produce CPE, which is released at the completion of sporulation, when the mother cell lyses to release its endospore. CPE then causes morphologic damage to the small intestine, resulting in diarrhea and abdominal cramps. Modified and reproduced from Journal of Food Safety, 1992, 12:237–252, by permission of John Wiley and Sons. doi:10.1128/9781555818463.ch18f1

C. perfringens cells containing intracellular CPE from the small intestine. As mentioned earlier, in those situations where food poisoning victims do not develop diarrhea, e.g., due to severe constipation from drug therapy, the intestines endure prolonged exposure to CPE (and possibly other toxins), which can result in severe disease (including necrotizing enteritis).

INFECTIOUS DOSE AND SUSCEPTIBLE POPULATIONS FOR C. PERFRINGENS TYPE A FOODBORNE ILLNESS Since C. perfringens cells are susceptible to killing by stomach acidity, cases of C. perfringens type A food poisoning usually develop only after consumption of a heavily contaminated food, i.e., >106 to 107 C. perfringens vegetative cells per gram of food (34). C. perfringens type A food poisoning almost always occurs when

SMP_Food Microbiology_CH18.indd

471

CPE, the toxin responsible for the disease symptoms, is produced in vivo as C. perfringens isolates sporulate in the intestines; therefore, this illness is typically an infection rather than an intoxication. Rare reports describe the onset of early symptoms that could be consistent with preformed CPE in foods occasionally contributing to food poisoning. Nevertheless, the typically long incubation period of this food poisoning, despite the quick action of CPE, indicates that preformed CPE involvement in C. perfringens type A food poisoning symptoms must be uncommon (34). Everyone is susceptible to C. perfringens type A food poisoning; however, this illness tends to be more serious in elderly, debilitated, or medicated individuals. Many individuals develop a serum antibody response to CPE following illness, but there is no evidence to indicate that previous exposure to this type of food poisoning provides significant future protection.

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

472

VIRULENCE FACTORS CONTRIBUTING TO C. PERFRINGENS TYPE A FOOD POISONING

Factors Involved in Spore Resistance Properties The enhanced food environment stress resistance properties of vegetative cells of most type A chromosomal cpe isolates remain poorly understood. However, considerable progress has now been achieved towards understanding why the spores of those isolates exhibit particularly strong food stress-associated resistance characteristics. During sporulation, Bacillus and Clostridium spp. produce several different small acid-soluble proteins (SASPs) that bind to spore DNA and thereby offer protection from such stresses as heat and radiation. It had long been thought that C. perfringens produces three SASPs, named SASP-1, -2, and -3, and Sarker’s group had shown that production of those three SASPs is important for C. perfringens spore resistance (62). However, SASPs 1 through 3 could not explain the enhanced resistance phenotype of spores made by type A chromosomal cpe isolates because those same studies also determined that similar levels of SASP-1, -2, and -3 are produced by type A chromosomal cpe isolates and other C. perfringens strains. Therefore, Li and McClane performed a bioinformatic search (37) on the genome of SM101, a type A food poisoning derivative that forms highly resistant spores and has a chromosomal cpe gene. That analysis identified an open reading frame (ORF) encoding a protein (named SASP-4) possessing sequence characteristics of a SASP. When the gene encoding this new SASP was sequenced in numerous C. perfringens strains, SASP-4 sequence variations were identified,

which helped to explain the enhanced spore resistance properties of type A chromosomal cpe strains versus other C. perfringens strains. Specifically, type A chromosomal cpe strains with exceptionally resistant spores consistently encoded a SASP-4 variant with an Asp at residue 36, whereas C. perfringens strains with sensitive spores consistently encoded a SASP-4 variant with a Gly at residue 36. Supporting the relevance of this observation for explaining spore resistance differences, inactivation of the gene encoding SASP-4 in SM101 resulted in spores with sharply reduced resistance against heat, cold, and nitrites (Table 18.3). Complementation of that mutant to express the SASP-4 Asp variant produced spores with substantially greater stress resistance than that of the spores made after that mutant was complemented to express the Gly SASP-4 variant (Table 18.3). The greater spore protection offered by the SASP-4 Asp variant had tighter DNA binding than the SASP-4 Gly variant. Interestingly, SASP-4 was bound more tightly to AT-rich DNA, whereas C. perfringens SASP-2 bound more tightly to GC-rich DNA (37, 42). This result is significant since it supports the importance of SASP-4 in protecting the C. perfringens genome, which has an ~72% AT content. Furthermore, these findings suggest that different SASPs may act cooperatively to protect spore DNA by binding to different C. perfringens DNA sequences varying in their AT content. Inactivating the SASP-4-encoding gene in SM101 sharply reduced spore resistance properties, but those SASP-4 null mutant spores still exhibited greater resistance against food environment stresses than spores made by other C. perfringens strains. Therefore, while clearly very important, SASP-4 is not the only factor

Table 18.3  Resistance properties of C. perfringens strains showing importance

of the SASP-4 variant with an Asp at residue 36a

Strain

D value (min) after heat treatment at 100°C

Log reduction of spore viability after treatment with: Nitrous acid after 60 min

Cold after 6 mo at: 4°C

−20°C

F4969WT

  0.5

4.0

0.88

1.23

SM101WT

59.1

1.1

0.35

0.58

SM101::ssp4

  8.7

4.0

0.82

1.91

SM101::ssp4(pCS)

44.7

1.1

0.40

0.65

SM101::ssp4(pCF)

16.4

3.2

0.50

1.22

Strains include F4969, a type A plasmid cpe strain; SM101, a type A chromosomal cpe strain; SM101:: ssp4, which is an SASP-4 null mutant of SM101; SM101::ssp4(pCS), which is the SM101 ssp4 null mutant complemented to express the SASP-4 made by SM101; and SM101::ssp4(pCF), which is the SM101 ssp4 null mutant complemented to express the SASP-4 made by F4969. WT, wild type. Data compiled from references 37 and 42. a

SMP_Food Microbiology_CH18.indd

472

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

473

contributing to the exceptional resistance properties of spores of most type A chromosomal cpe isolates. Two groups have suggested that spore ultrastructural differences may also contribute to the stress resistance phenotype of spores made by type A chromosomal cpe isolates (57, 58). In both studies, the spore core was smaller in spores of type A chromosomal cpe isolates than in spores of other C. perfringens isolates, possibly because the smaller spores are more dehydrated.

C. perfringens Enterotoxin Evidence that CPE Is Involved in C. perfringens Type A Food Poisoning

Epidemiologic studies provided strong initial evidence that CPE plays a pivotal role in C. perfringens type A foodborne illness. Those studies (45) revealed the following: (i) a strong positive correlation between illness and the presence of CPE in feces, since, depending on assay sensitivity and how quickly a fecal sample was collected after the onset of symptoms, 80 to 100% of feces from individuals ill with C. perfringens type A food poisoning test CPE positive, whereas virtually no feces from healthy individuals test CPE positive; (ii) the presence of CPE in the feces of food poisoning victims at levels known to cause significant GI tract effects in ­experimental animals; (iii) development of the characteristic GI symptoms of C. perfringens type A food poisoning in human volunteers who were fed purified CPE; (iv) CPE-positive C. perfringens food poisoning isolates are considerably more effective than CPE-negative C. perfringens isolates at inducing fluid accumulation in rabbit ileal loops or diarrhea in human volunteers; and (v) rabbit ileal loop effects induced by CPE-positive isolates can be neutralized with CPE-specific antisera. Later studies fulfilling molecular Koch’s postulates experimentally supported the importance of CPE in the GI tract pathogenesis of C. perfringens food poisoning (64). Those experiments revealed (Fig. 18.2) that sporulating (but not vegetative) culture lysates of SM101, a transformable derivative of a wild-type CPE-positive food poisoning isolate, can induce both fluid accumulation and histopathologic damage in rabbit ileal loops. This observation is consistent with CPE (whose expression is sporulation associated) being necessary for the GI tract activity of SM101. That important conclusion was confirmed when (i) neither vegetative nor sporulating culture lysates of an isogenic SM101 cpe knockout mutant were able to induce intestinal fluid accumulation or histopathologic damage and (ii) full GI virulence was restored to sporulating cultures by complementing the SM101 cpe knockout mutant with a shuttle plasmid carrying the wild-type cpe gene.

SMP_Food Microbiology_CH18.indd

473

C. perfringens type A food poisoning is not the only GI disease involving CPE. C. perfringens type A isolates producing CPE also cause several nonfoodborne human GI illnesses, including antibiotic-associated diarrhea and sporadic diarrhea, as well as some veterinary diarrheas (47). However, the cpe-positive C. perfringens type A isolates causing nonfoodborne GI diseases typically carry a plasmid-borne cpe gene and are genetically distinct from the chromosomal cpe isolates causing most food poisoning cases (11). Molecular Koch’s postulate studies using a cpe knockout mutant of F4969, a nonfoodborne human GI disease isolate, revealed that CPE plays an important role in the pathogenesis of nonfoodborne human GI disease caused by type A isolates carrying a plasmid-borne cpe gene (64). The pathogenesis of nonfoodborne GI diseases may sometimes also involve additional toxins besides CPE, since F4969 is a relatively atypical type A nonfoodborne human GI isolate that lacks the cpb2 gene encoding beta2 toxin, i.e., beta2 toxin could sometimes also play a role in CPE-­associated nonfoodborne human GI diseases (19).

Genetics of CPE

The presence of the cpe gene in only ~1 to 5% of all C. perfringens isolates, most of which are type A (15, 31), suggested that this toxin gene might be associated with mobile genetic elements. The first conclusive evidence supporting this association was provided by Southern blot analyses of pulsed-field gel electrophoresis gels, which localized the cpe gene in nonfoodborne human GI disease isolates and animal isolates to large plasmids (11, 12, 74). Subsequent studies (3) revealed that the cpe plasmid of isolate F4969 (and probably most other nonfoodborne human GI disease isolates) is transferable, at high efficiency, by conjugation to cpe-negative C. ­perfringens isolates. Two major cpe-carrying plasmid families among type A nonfoodborne human GI disease isolates, with one of those cpe plasmids also encoding beta2 toxin, were subsequently identified (19). Representatives of those two major cpe plasmid families in type A isolates have been sequenced (55), which revealed they share about 50% homology, including a common Tn916-like region named tcp (transfer of clostridial plasmids) that likely mediates their conjugative transfer. The chromosomal cpe gene, which is in most type A food poisoning isolates, is also apparently associated with a mobile genetic element. Specifically, the discovery of IS1470 sequences upstream and downstream of the chromosomal cpe ORF in NCTC8239 and other food poisoning isolates (Fig. 18.3A), led to the proposal (4) that the chromosomal cpe gene is present on a 6.3-kb

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

474

Figure 18.2  Fulfilling molecular Koch’s postulates demonstrates that CPE is important for the gastrointestinal virulence of C. perfringens type A food poisoning isolates. Tissue specimens shown were collected from rabbit ileal loops that had been treated with either concentrated vegetative (FTG) or sporulating (DS) culture lysates of C. perfringens strain SM101, an electroporatable derivative of food poisoning strain NCTC 8798; MRS101, which is a cpe knockout mutant of SM101; or MRS101(pJRC 200), which is the MRS101 mutant complemented with a shuttle plasmid carrying the cloned, wild-type cpe gene. Note that (i) tissue specimens from loops treated with a concentrated FTG lysate of SM101, which does not contain CPE, were indistinguishable from control ileal loop tissue specimens, while (ii) tissue damage (and fluid accumulation) was observed only in loops treated with DS culture lysates of SM101 or MRS101(pJRC200), both of which contain CPE. Reprinted with permission of John Wiley and Sons from reference 64. doi:10.1128/9781555818463.ch18f2

transposon (named Tn5565) with terminal IS1470 ­elements. PCR evidence (2) suggests this putative cpe­carrying transposon has a circular intermediate form, but actual excision and reintegration of Tn5565 has not yet been experimentally demonstrated. Interestingly, recent studies (41) revealed that the plasmid-borne cpe locus in some type C isolates resembles the chromosomal cpe locus of type A isolates. Presuming the type C cpe plasmid is conjugative, this finding suggests a possible origin for type A chromosomal cpe isolates. The type C cpe plasmid may have transferred into a type A isolate, followed by excision from that plasmid of a mo-

SMP_Food Microbiology_CH18.indd

474

bile genetic element carrying the cpe gene. That excised cpe-carrying element may then have integrated onto the chromosome of the host type A isolate, creating a progenitor for most current food poisoning isolates. Interestingly, no upstream IS1470 element is associated with the plasmid cpe gene of type A isolates (Fig. 18.3A). Furthermore, instead of the downstream IS1470 sequence present in the chromosomal cpe locus, there is typically either an IS1151 sequence or an IS1470-like sequence present downstream of the plasmid-borne cpe gene in type A isolates. These cpe locus variations have enabled development of a mul-

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

475

(A)

(B) Figure 18.3  Variations in cpe locus arrangements. (A) Comparison of cpe locus arrangement in different C. perfringens isolates. The top and middle region maps show the arrangement of the plasmid cpe locus in type A human nonfoodborne GI disease isolates F4969 and F4013, respectively. Note that the cpe plasmid of F4013 and other type A isolates with a similar cpe locus also encode b2 toxin. The map on the bottom shows the arrangement of the chromosomal cpe locus in food poisoning isolate NCTC8239. The chromosomal cpe locus is similarly arranged in most other food poisoning isolates. (B) Multiplex PCR cpe subtyping assay. Representative results obtained with this assay are shown for culture lysates from type A isolates known to carry a chromosomal cpe gene (lanes 2 to 6 from the left), a plasmid cpe gene with an associated IS1470-like sequence (lanes 7 to 11 from the left), or a plasmid cpe gene with an associated IS1151 sequence (lanes 12 to 15 from the left). The migration positions of molecular size markers are shown on the left. The sizes of expected PCR products are shown on the right. Compiled from references 4, 19, 54, and 56. doi:10.1128/9781555818463.ch18f3

tiplex PCR assay capable of discriminating between most type A isolates carrying a plasmid-borne versus a chromosomal cpe gene (Fig. 18.3B). This PCR cpe subtyping assay is now proving valuable for epidemiologic investigations.

SMP_Food Microbiology_CH18.indd

475

Expression and Release of CPE

The expression and release of CPE from C. perfringens isolates have at least three interesting features: (i) CPE production is tightly regulated, i.e., this toxin is expressed during sporulation, but not vegetative growth;

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

476 (ii) during sporulation, many CPE-positive isolates produce extremely large amounts of this toxin; and (iii) CPE is not actually secreted by sporulating C. perfringens cells but is instead released into the intestines when the mother cell lyses upon the completion of sporulation.

Regulation of CPE Synthesis

Duncan’s classic studies first established a relationship between CPE expression and sporulation, e.g., C. perfringens mutants blocked at stage 0 of sporulation completely lost their ability to produce CPE (50). Western blot studies (14) later confirmed a relationship between sporulation and CPE expression by revealing 1,500fold-greater CPE expression by sporulating cells versus vegetative cells of cpe-positive type A food poisoning isolate NCTC8239. Therefore, it is not surprising that the master initiator of sporulation, i.e., Spo0A, is necessary for CPE production (25). More-recent studies have revealed that three of the four alternative sigma factors, i.e., SigF, SigE, and SigK, are also necessary for CPE

production, with SigF controlling production of SigE and SigK (23, 39), as shown in Fig. 18.4.

Synthesis of CPE CPE synthesis begins shortly after sporulation is induced and then progressively increases for the next 6 to 8 hours (50). Late in sporulation, CPE can represent up to 15% of the total cell protein present inside a C. perfringens sporulating cell (14). In general, the better a C. perfringens isolate sporulates, the more CPE it produces; however, this correlation is not absolute (10). Why do some CPE-positive C. perfringens type A strains produce so much enterotoxin during sporulation? This high-level CPE expression is not related to cpe gene location, as type A isolates carrying a chromosomal or a plasmid cpe gene produce similar levels of CPE (10–12). Nor is this strong enterotoxin expression due to a gene dosage effect, since most, if not all, cpe-positive isolates carry a single copy of the cpe gene. RNA slot blot and Northern blot analyses (13, 51, 82) demonstrated regulation of CPE production at the

Figure 18.4  Regulation of C. perfringens sporulation and CPE expression by alternative sigma factors. When induced by stress, a regulatory cascade is induced and results in production of SigF, an alternative sigma factor. Production of SigF then controls expression of three other alternative sigma factors (SigE, SigK, and SigG) that are necessary for sporulation. In addition, SigE and SigK mediate CPE expression from SigE- and SigK-dependent promoters for the cpe gene. Based upon data presented in references 23 and 39. doi:10.1128/9781555818463.ch18f4

SMP_Food Microbiology_CH18.indd

476

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

477

transcriptional level, with significant amounts of cpe mRNA expressed during sporulation but little or no cpe mRNA produced during vegetative growth of C. perfringens. Northern blot studies (13) further revealed that cpe mRNA is transcribed as a monocistronic message of ~1.2 kb, which is consistent with initial primer extension analysis studies (51) indicating that cpe mRNA transcription starts ~200 bp upstream of the cpe ORF translation start site. Subsequent primer extension, RNase T2 protection, and deletion mutagenesis studies (82) identified at least three start sites (i.e., P1, P2, and P3) for the initiation of cpe mRNA transcription. Therefore, the presence of multiple promoters is probably a major contributor to strong CPE expression. P1 shares homology with SigK-dependent promoters, whereas P2 and P3 share homology with SigEdependent promoters. Those promoter homologies likely explain (Fig. 18.4) why SigE and SigK are important for regulating CPE production, as reported previously (23). Posttranscriptional effects may also help regulate CPE expression levels. For example, the functional half-life of cpe mRNA in sporulating C. perfringens cells is reportedly 58 min (35), which is unusually long for a bacterial message. Such exceptional message stability could contribute to the abundant CPE expression noted for sporulating cells of some C. perfringens strains. Given that stem-loop structures can contribute to message stability, the putative stability of cpe mRNA could result from a stem-loop structure present 36 bp downstream of the 3¢ end of the cpe ORF (14). This stem-loop structure is followed by an oligo(dT) tract, suggesting it also functions as a rho-independent transcriptional terminator (14), which would be consistent with the previously described transcriptional start sites identified 200 bp upstream of the cpe initiation codon and the 1.2-kb size of cpe mRNA observed in Northern blot studies (13).

Release of CPE from C. perfringens Unlike most C. perfringens toxins, CPE is not secreted via a classic bacterial transport system (14). Consistent with this, cpe does not encode the 5¢ signal peptide associated with many secreted exotoxins (14). Instead, newly synthesized CPE accumulates in the cytoplasm of the mother cell, sometimes reaching sufficiently high concentrations to induce formation of cytoplasmic CPE-containing paracrystalline inclusion bodies. This intracellular CPE is eventually released into the intestines upon the completion of sporulation, i.e., when the mother cell lyses to free its mature spore. The dependency upon mother cell lysis for CPE release helps explain why, despite CPE’s quick intestinal action, C. perfringens type A food poisoning symptoms develop 8 to 24 hours after ingestion of contaminated foods, i.e., sporulating C. perfringens

SMP_Food Microbiology_CH18.indd

477

cells need at least 8 to 12 hours to complete sporulation and then release CPE into the intestine.

CPE Biochemistry

Early studies (48) characterized CPE as a single polypeptide of ~35,000 Da, with a pI of 4.3. Later, sequencing of the cloned cpe ORF (14) determined that the CPE protein is comprised of 319 amino acids, with a precise Mr of 35,317. Additional cpe ORF sequencing studies (10) revealed that the CPE amino acid sequence is highly conserved by most CPE-positive C. perfringens type A isolates. The CPE sequence lacks homology with other proteins, except for some limited similarity, of unknown significance, with a nonneurotoxic protein produced by C. botulinum (50). CPE is a heat-labile protein; its biologic activity can be inactivated by heating for 5 minutes at 60°C. The enterotoxin is also quite sensitive to pH extremes (pH of <6 or >8) but is relatively resistant to some proteolytic treatments. In fact, limited trypsin or chymotrypsin treatment actually increases CPE activity about two- to threefold (46), suggesting that intestinal proteases could activate CPE during food poisoning.

CPE Action

Studies of the in vivo and in vitro effects of CPE have indicated that CPE has a novel mechanism of action, as described below.

CPE effects on the GI tract CPE is classified as an enterotoxin because it induces fluid and electrolyte losses from the GI tract of many mammalian species (48). In rabbits, the principal target organ for CPE is the small intestine, with the ileum being particularly sensitive (48). Interestingly, the rabbit colon is relatively insensitive to CPE, despite the strong binding of this enterotoxin to rabbit colonic cells (48). Results from ex vivo studies (18) revealed that the human ileum is also very sensitive to CPE, whereas human colonic tissue had only a mild histopathologic response to CPE treatment, at least under the conditions used in those ex vivo studies. The biologic activity of CPE is readily distinguishable from such classical enterotoxins as cholera and Escherichia coli heat-labile enterotoxins (48): (i) CPE does not increase intestinal cyclic AMP levels; (ii) CPE inhibits glucose absorption; and (iii) CPE quickly induces histopathologic damage, including epithelial desquamation and severe villus shortening, in the small intestine. While some other bacterial enterotoxins, e.g., Shiga toxin and C. difficile toxins, also cause intestinal tissue damage, CPE is unique with respect to its ability to induce intestinal damage within as little as 15 to 30 minutes.

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

478

Figure 18.5  Model for the cellular action of CPE. (Left to right) CPE binds to claudin receptors, forming a small complex. At 37°C, several small complexes interact to form an ~450kDa CH-1 prepore. The CH-1 prepore then inserts into membranes to form an active pore. The CH-1 active pore allows a Ca2+ influx. With high CPE doses, a massive Ca2+ influx occurs and triggers oncosis; with low CPE doses, there is a more moderate Ca2+ influx that triggers apoptosis. Morphologic damage caused by membrane permeability alterations allows unbound CPE access to the basolateral surface, resulting in formation of a second CPE complex, named CH-2, containing claudins and occludin. doi:10.1128/9781555818463.ch18f5

Two observations strongly suggest that CPE-induced tissue damage (Fig. 18.2) plays a major role in initiating CPE-induced fluid/electrolyte intestinal transport alterations. First, fluid transport alterations develop concurrently with tissue damage in CPE-treated rabbit ileum (67). Second, only those CPE doses capable of inducing tissue damage can induce intestinal fluid and electrolyte transport alterations in the rabbit ileum (48). Based upon these two observations, CPE apparently initiates its GI tract effects during C. perfringens type A food poisoning by causing tissue damage that disrupts villus integrity and induces a breakdown of the normal intestinal secretion-absorption equilibrium. CPE fragments can affect the paracellular permeability properties of the intestinal epithelium (73), suggesting that CPE-induced paracellular permeability could also contribute to intestinal fluid/electrolyte secretion in CPE-treated intestinal tissue (however, see the discussion arguing against this possibility later in this ­chapter). Additionally, the ability of CPE to induce significant release of some

SMP_Food Microbiology_CH18.indd

478

proinflammatory cytokines (33, 78) and kill cells by the proinflammatory process of oncosis raises the possibility that inflammation could be another contributor to CPEinduced intestinal effects, particularly late in illness.

Overview of the cellular action of CPE The cytotoxic action of CPE is responsible for the tissue damage that initiates CPE-induced intestinal fluid/ electrolyte alterations. A model describing the cytotoxic action of CPE is presented in Fig. 18.5. This model indicates that CPE acts on the intestine via a multistep cytotoxic process involving several early events. CPE first binds to a protein receptor(s). This binding localizes CPE in an ~90-kDa small complex. Several small complexes then likely interact with one another to form an ~450kDa CH-1 large complex on the surface of mammalian plasma membranes. The initial CH-1 complex prepore then inserts in plasma membranes to form an active membrane pore that alters normal plasma membrane permeability characteristics. Those plasma membrane

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

479

permeability alterations lead to calcium influx into the CPE-treated cell, causing cell death. The development of cellular morphologic damage due to calcium influx then promotes the formation of a second, ~600-kDa large complex, named CH-2, that could induce tight junction (TJ) rearrangements and thereby possibly contribute to paracellular permeability changes in the CPE-treated intestinal epithelium. This model emphasizes the uniqueness of CPE cellular/ molecular activity by predicting that CPE closely interacts with eukaryotic proteins at every step in its action. No other known membrane-active toxin interacts so closely with eukaryotic proteins throughout its action.

Cellular action of CPE: early events Step no. 1: receptor binding.  While very high con­ centrations of CPE can induce cation channel formation in protein-free artificial membranes, ample evidence indicates that, at pathophysiologic CPE concentrations, the cytotoxic effect of this enterotoxin requires receptor binding. For example, CPE interactions with sensitive mammalian cells exhibit the specificity and saturability expected of a receptor-mediated process. CPE-specific binding is also rapid and temperature sensitive, with less CPE binding occurring at 4°C than 37°C. The small intestines of all tested mammalian species can specifically bind CPE, helping to explain why CPE induces intestinal fluid and electrolyte losses in many mammalian species (48). Additional evidence using mouse fibroblast provides persuasive evidence that, at pathophysiologic concentrations, the cytotoxic action of CPE requires a receptor. Fibroblasts naturally do not bind toxin or respond to challenge with pathophysiologic CPE concentrations (26, 27, 69), yet become very CPE sensitive when transfected to express certain members of the claudin family of TJ proteins (26, 27). Claudins are ~22-kDa proteins that play a critical role in maintaining normal TJ structure and function. More than 20 different claudins, which vary mainly in their C-terminal cytoplasmic tail sequences, have been identified to date. It has thus far been determined that claudins -3, -4, -6, -7, -8, and -14, but not claudins -1, -2, -5, or -10, can bind CPE at pathophysiologic concentrations. The structure of a claudin has not yet been fully elucidated, but these proteins are thought to contain two extracellular loops. It is now well established that CPE binds to the second putative extracellular loop of receptor claudins. In addition, recent fibroblast transfectants studies identified an Asn residue present in the middle of this second loop as being particularly critical for CPE binding (Fig. 18.6).

SMP_Food Microbiology_CH18.indd

479

It has already been mentioned that, in vivo, CPE intestinal damage starts at the villus tip. Recent studies revealed this pattern results from increased CPE binding to this villus region due to the abundant presence of claudin-4 receptors at this location (71).

Step no. 2: small complex formation.  Upon binding to sensitive mammalian cells, CPE becomes localized in an ~90-kDa, CPE-containing small complex (81). Coimmunoprecipitation experiments using CPEtreated Caco-2 cells revealed that, when sequestered in this small complex, CPE physically interacts with claudin receptors, such as claudin-4 (63). Interestingly, those experiments also revealed that the small complex may also contain claudin-1, which is not an efficient CPE receptor. This observation may reflect the known claudin-claudin binding interactions that occur between receptor claudins and claudin-1. The small CPE complex is important for CPE action, since it forms in all CPE-sensitive cells examined to date (81). When sequestered in a small complex, CPE largely remains exposed on the membrane surface. For example, CPE localized in a small complex remains fully accessible to externally applied antibodies and proteases (30, 80). Step no. 3: formation of a prepore large CPE complex on the membrane surface.  Under physio­logic conditions, i.e., at 37°C, the ~90-kDa small complex rapidly transitions into a larger complex, named CH1 (Fig. 18.5). The small CPE complex appears to be a direct precursor for this ~450-kDa CPE-containing large complex. For example, when Vero cells treated with CPE at 4°C, a temperature at which all bound toxin is present in small complexes, are shifted to 37°C, formation of CH-1 occurs almost immediately (81). At 37°C, CH1 is the predominant CPE-containing complex present when Vero cells or Caco-2 cells are briefly treated with CPE (69). However, if Caco-2 or Vero cell monolayers are CPE treated at 37°C for longer periods, a second CPE-containing large complex named CH-2 becomes discernible. The small ~90-kDa CPE complex is easily distinguished from the two large CPE complexes on the basis of its size, sensitivity to sodium dodecyl sulfate (SDS) (both large complexes are stable in SDS, whereas the small complex dissociates in the presence of SDS), and temperature sensitivity (formation of the large CPE complexes is blocked at 4°C, whereas the small complex forms at low temperatures). Electroelution and coimmunoprecipitation studies (63) have revealed that, in addition to CPE, CH-1 contains receptor claudins and claudin-1, which supports CH-1 formation resulting from small complex ­oligomerization.

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

480

(A)

(B)

SMP_Food Microbiology_CH18.indd

480

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

481

Heteromer gel shift analyses (Fig. 18.7) revealed that CH-1 and CH-2 each contain 6 CPE molecules; hence, the designations CH-1 and CH-2 are abbreviations for “CPE hexamer-1” and “CH-2 hexamer-2.” Consistent with its content of 6 CPE molecules and several claudins, CH-1 is very large, estimated to be ~450 kDa using gel filtration and native electrophoresis techniques (63). A CPE mutant lacking amino acids 81 to 106 was able to form CH-1, but it did not induce membrane permeability alterations (72). Relative to the CH-1 formed by native CPE, the CH-1 formed by this CPE mutant is much more sensitive to degradation by externally applied proteases. Those results suggest that CPE initially oligomerizes as a prepore on the membrane surface prior to inserting into a lipid bilayer to form an active pore. Oligomerization of many pore-forming toxins depends upon lipid raft-induced clustering of toxin-receptor complexes. However, CPE oligomerization during CH-1 formation is lipid raft independent (6). This unusual aspect of CPE action may reflect the fact that claudins, including claudin receptors, naturally cluster together in cells in the form of long TJ strands.

Step no. 4: plasma membrane permeability alterations.  Insertion of CH-1 into membranes apparently results in pore formation, which is responsible for cytotoxicity. For example, a close kinetic correlation (68) exists between CH-1 formation and the onset of membrane permeability alterations that are responsible for the death of CPE-treated Caco-2 cell or Vero cell monolayers. Furthermore, CPE point and deletion mutants unable to form CH-1 are nontoxic (28, 70), whereas CPE deletion fragments exhibiting enhanced cytotoxicity, relative to native CPE, are more efficient than the native enterotoxin at CH-1 formation (29). Initially, CPE-induced plasma membrane permeability alterations are restricted to small molecules (<200 Da), suggesting that the CPE-induced membrane pore is ~0.5 to 1 nm2. CPE induces a rapid increase in both influx and efflux of many small molecules, including ions (preferably cations) and amino acids. These small-molecule permeability alterations severely perturb the levels of ions

and other small molecules present in a CPE-treated cell, leading to cell death. The kinetic development of these membrane permeability alterations is CPE dose dependent, but initial effects can develop within 5 minutes.

Cellular action of CPE: consequences of CPEinduced membrane permeability alterations Cell death.  There is now considerable understanding of how CPE kills sensitive mammalian cells (8, 9). Caco-2 cells treated with low CPE doses (1 μg/ml) exhibit morphologic alterations, DNA cleavage into a distinct fragment ladder, mitochondrial membrane depolarization, cytochrome c release from mitochondria, and caspase 3/7 activation. Very late events, e.g., DNA cleavage and morphologic damage, resulting from this low CPE dose treatment could be blocked by a caspase-3 inhibitor, but not by a caspase-1 inhibitor or by glycine, an oncosis inhibitor. In contrast, higher CPE doses (10 μg/ml) induce morphological damage and random DNA shearing without mitochondrial membrane depolarization, cytochrome c release, or caspase-3/7 activation. The morphological damage and random DNA shearing induced by this higher CPE dose could be inhibited by glycine, but not by caspase-1 or caspase-3/7 inhibitors. Collectively, these observations indicate that lower CPE doses kill mammalian cells via a classical caspase 3/7 apoptosis, whereas mammalian cells treated with higher CPE doses die from oncosis. This identification of CPE dose-dependent variations in cell death pathway activation may have direct in vivo relevance. Evidence of inflammation has been observed in some, but not all, CPE-treated rabbit ileal loops (8, 9). Generally, inflammatory cell infiltration is more prominent in CPE-treated ileal loops treated with higher CPE doses. The results with Caco-2 cells described above suggest that a putative association between high CPE doses and inflammation may reflect the ability of high CPE doses to induce oncosis, which (unlike classical apoptosis) is proinflammatory. Since (i) inflammation contributes to the pathogenesis of several other foodborne GI illnesses and (ii) high CPE doses similar to those caus-

Figure 18.6  CPE interactions with claudin receptors. (A) Alignment of strong CPE-binding claudin receptors versus weak or non-CPE binding receptors. Note the presence of an Asn residue in the middle of the extracellular loop 2 (ECL-2) domain of all strong CPE-binding claudins. The equivalent ECL-2 residue is never an Asn in the weak- or non-CPE-binding claudins. TM domains 3 and 4 refer to putative transmembrane domains of claudins. (B) Evidence that the Asn residue in the middle of ECL-2 of CPE receptor claudins is important for CPE binding and cytotoxicity. The equivalent ECL-2 residues in claudin-1 and -4 (residues 150 in claudin-1 and 149 in claudin-4) were changed from D to N and from N to D, respectively. Photos show CPE sensitivity of parent fibroblasts or fibroblast transfectants expressing claudin-1, claudin-1D150N, claudin-4, or claudin-4N149D. The graph shows the cytotoxic effects of CPE on these cells. Note that as shown in both the photos and the graph, changing this key ECL-2 Asn residue profoundly affects CPE sensitivity. Reproduced with permission from Infection and Immunity, 2010, 78:505–517. doi:10.1128/9781555818463.ch18f6

SMP_Food Microbiology_CH18.indd

481

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

482

Figure 18.7  Heteromer gel shift analyses indicate that CPE is present as a hexamer in large complexes. In this experiment, two different-sized CPE variants mixed together at varying ratios are applied to Caco-2 cells. After cell lysis, the CPE large complexes are separated by electrophoresis on an SDS-containing polyacrylamide gel and visualized by Western blotting using CPE antibodies. The stoichiometry of CPE in each large complex is calculated by counting the number of CPE complex bands (which is 7) on these gels and then subtracting one (because there are two homomer complexes). Therefore, it is concluded that CPE is present as a hexamer in CH-2 (shown here). CH-1 was also determined to contain six copies of CPE (not shown). Reproduced with permission from reference 63. doi:10.1128/9781555818463.ch18f7

ing oncosis in vitro have been detected in the feces of some C. perfringens type A food poisoning victims, the proinflammatory effects of oncosis could contribute to intestinal pathology in some cases of CPE-induced GI disease. When lower CPE concentrations are present in the intestinal lumen, human intestinal epithelial cells probably die from apoptosis and inflammation plays a lesser (or no) role in pathology. How do membrane permeability alterations trigger death of CPE-treated cells, and why is cell death pathway activation CPE dose dependent? CPE-induced membrane permeability alterations (8) cause a Ca2+ influx (probably through the CPE pore), which then serves as the direct trigger for cell death. Specifically, high CPE

SMP_Food Microbiology_CH18.indd

482

doses cause a rapid and massive change in cytoplasmic Ca2+ levels, which results in oncosis, whereas lower CPE doses cause a slower-developing and more modest elevation in cytoplasmic Ca2+ levels, which produce a classical caspase 3/7-dependent apoptosis. Interestingly, both CPE-induced apoptosis and oncosis involve the cytoplasmic proteins calmodulin and calpain. However, high CPE doses induce a more rapid and stronger activation of calpain proteolytic activity than do low CPE doses. Further studies of CPE-induced cell death pathways are under way.

Formation of CH-2.  Upon the onset of morpho­ logic damage, CPE-treated cells form a second large

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

483

Figure 18.8  Linear map of CPE functional regions. CPE regions involved in CH-1 oligomerization are located between residues 45 and 54 of CPE, with residues D48 and I51 being required for both those events to occur. A putative transmembrane stem domain that may mediate CPE insertion into membranes during pore formation has been localized to residues 80 to 110 of CPE. The C-terminal region of CPE, which also reacts with MAb 3C9, is depicted as located between residues 290 to 319 of CPE. The extreme N-terminal sequences of CPE (residues 1 to 44) are unnecessary for cytotoxicity, and some of these sequences may be removed during disease by intestinal proteases (see the text). Compiled from references 23, 28, 29, 44, 45, and 70. doi:10.1128/9781555818463.ch18f8

complex, named CH-2. This second CPE complex, which also contains six CPE molecules (Fig. 18.7), is larger than CH-1 with a molecular mass of ~600 kDa. In addition to CPE, CH-2 also contains receptor claudins and nonreceptor claudins. However, unlike CH-1, the CH-2 complex also contains another TJ protein, named occludin.

CPE-induced changes in TJs.  Native CPE can induce TJ rearrangements, with a brief (1-h) native CPE treatment causing TJ rearrangements in rat liver (61). Since treatment with CPE can induce internalization of claudins and occludin, CPE may also trigger TJ structural rearrangements in the intestines. In fact, a noncytotoxic, but binding-capable, CPE fragment increases intestinal drug absorption, apparently by promoting increased paracellular permeability. This effect might suggest that CPE-induced intestinal paracellular permeability changes contribute to diarrhea. However, arguing against that possibility are observations that a noncytotoxic, but binding-capable, CPE fragment was unable to cause intestinal fluid transport changes in rabbit ileal loops (71). Whether native CPE effects on TJ rearrangements contribute to CPE’s intestinal activity remains unclear. Nevertheless, available data overwhelmingly reveal that the primary pathologic effect of CPE involves, via its cytotoxic action, the induction of histopathologic damage in the ileum. In support of this view, the onsets of histopathologic damage fluid and electrolyte losses closely correlate in CPE-treated rabbit ileal loops (67). Furthermore, CPE-induced TJ rearrangements in rat liver were observed only when enterotoxin was applied to the basal surface (60, 73). Since CPE initially interacts with the apical side of enterocytes, the ability of CPE to cause TJ rearrangements only when applied to the basolateral surface suggests that TJ rearrangements would develop

SMP_Food Microbiology_CH18.indd

483

in the intestinal epithelium after CPE’s cytotoxic effects had already produced sufficient histopathologic damage to provide the enterotoxin with access to the basolateral surface of epithelial cells.

CPE Structure/Function Relationships

Structure/function studies (28, 29, 46, 70) have provided considerable insights into how the CPE protein mediates the pathophysiologic effects described above. In particular, studies using CPE fragments and point mutants have localized several important functional regions on the CPE protein. For the CPE receptor-binding region, those results have now been coupled with structural analyses provided by X-ray diffraction analyses (77). An overview of CPE functional regions, as currently understood, is depicted in Fig. 18.8.

Receptor-binding domain Receptor-binding activity has been localized to the C terminus of the CPE protein. For example, C-terminal CPE fragments (or even a synthetic peptide corresponding to amino acids 290 to 319 of native CPE, i.e., the 30 C-terminal CPE amino acids) efficiently compete against 125 I-CPE binding to cells or isolated membranes (46). More recently, site-directed mutagenesis has implicated CPE residues Tyr306, Tyr310, Tyr312, and Leu 315 in receptor binding (17). In addition, the structure of the CPE receptor-binding domain (residues 194 to 319) was recently solved to 1.75 Å (Fig. 18.9). This domain was shown to consist of a nine-strand b-sandwich with structural similarity to the receptor-binding domain of the large Cry toxin family of Bacillus thuringiensis. Furthermore, coupling this new structure with site­directed mutagenesis data strongly suggests that a large surface loop between strands b8 and b9 of the receptor-

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

484

Figure 18.9  Structure of the C-terminal half of the CPE protein. (A) Structure of the C-terminal half of the CPE protein, showing a structure consisting of a 9-strand b-sandwich and a large loop containing Tyr residues 306, 310, and 312, which are important for CPE binding to claudins. (B) The C-terminal­ half of CPE has structural, but not sequence, similarity to the receptor binding domain of Cry4Ba, a toxin produced by Bacillus thuringiensis. Reproduced with permission from reference 77. doi:10.1128/9781555818463.ch18f9

binding domain mediates the binding of CPE to claudin receptors.

Cytotoxicity domain Despite their receptor-binding ability, C-terminal CPE fragments, such as CPE194-319, are not themselves cytotoxic, i.e., they do not induce small-molecule permeability alterations in mammalian cells (46). This finding indicated that mere occupancy of the CPE receptor is insufficient to trigger CPE-induced cytotoxicity. It also supports the importance of postbinding steps for CPE activity and implicates sequences in the N-terminal half of the CPE molecule as being necessary for cytotoxicity, i.e., like most bacterial toxins, CPE segregates its receptor binding and activity regions. However, the entire N-terminal half of CPE is not needed for cytotoxicity. For example, deletion mutagenesis studies (29) revealed that removing the first 44 Nterminal amino acids of CPE increases cytotoxic activity two- to threefold. This effect may have some in vivo relevance. Intestinal proteases such as trypsin and chy-

SMP_Food Microbiology_CH18.indd

484

motrypsin can remove limited numbers of N-terminal amino acids from native CPE to produce an activated enterotoxin (20–22). This observation suggests CPE may be proteolytically activated in the intestines during C. perfringens type A foodborne illness. Why does limited removal of N-terminal sequences activate CPE cytotoxicity? Deletion mutagenesis studies (29) revealed that activated CPE fragments, i.e., CPE fragments lacking the extreme N-terminal sequences of native CPE, bind to receptors and form small complexes in a fashion similar to that of native enterotoxin. However, relative to native enterotoxin, these activated CPE fragments exhibit an enhanced ability to form CH1. Besides explaining why activated CPE fragments are more cytotoxic, this observation provides additional support for the important role played by CH-1 in CPEinduced cytotoxicity. It was also shown that removing eight additional Nterminal amino acids from a CPE45-319 fragment eliminates all cytotoxic activity (29). This elimination of cytotoxic activity is specifically due to the CPE53-319 fragment being blocked for CH-1 formation (29). Collectively, CPE deletion fragment studies (29) mapped a region responsible for CH-1 formation to amino acids 45 to 53 of the native enterotoxin. This conclusion was supported by later studies (28) using random CPE point mutants; for example, a CPE Gly49Asp random mutant, which binds and forms the small complex, was defective for forming CH-1 and also was noncytotoxic. Alanine scanning mutagenesis studies (70) have identified the core CPE sequence responsible for CH-1 formation as extending from residues Gly47 to Ile51, with residues 48 and 51 playing particularly important roles in CPE-induced cytotoxicity and CH-1 formation. Saturation mutagenesis studies (70) then revealed that the charge and size of the CPE Asp48 residue and the length of the aliphatic side chain on the CPE Ile51 residue are critical for CH-1 complex formation and cytotoxicity. CPE contains a stretch of amino acids (residues 81 to 106) alternating in their side chain polarity. That pattern is common to the transmembrane domain of bbarrel pore-forming toxins, suggesting this CPE region might be involved in pore formation. This possibility is supported by studies (72) using a CPE variant in which amino acids 81 to 106 were deleted. This CPE variant was noncytotoxic and unable to form a pore, although it bound to receptors and formed CH-1. The loss of cytotoxic activity for this CPE variant appears to be due to the CH-1 complex it forms being blocked for insertion into lipid bilayers. Instead, the CH-1 made by this variant remains on the plasma membrane surface, as evidenced by its greater susceptibility to externally applied

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

485

proteases compared to the CH-1 formed by native CPE. As mentioned earlier, these findings reveal a prepore step in CPE action. Which CPE region is involved in inducing TJ rearrangements and (possibly) paracellular permeability changes? The noncytotoxic C-terminal half of CPE can slowly induce TJ rearrangements in MDCK cells (73), likely because C-terminal CPE fragments can interact with claudin receptors, which are also major TJ structural proteins. TJ rearrangements are induced more rapidly by native CPE than by C-terminal CPE fragments, suggesting N-terminal CPE sequences are important for native CPE-induced TJ rearrangements and, possibly, for causing rapid paracellular permeability changes. The ability of native CPE to induce TJ rearrangements more rapidly than C-terminal CPE fragments may involve the same N-terminal region of amino acids 45 to 53 of native CPE that mediates cytotoxicity. Consistent with this hypothesis, CPE mutants containing alanine substitutions at either residue 48 or 51 were deficient for formation of CH-2, which contains occludin, as well as for CH-1 formation (73). The ability of native CPE, but not C-terminal CPE fragments capable of binding claudins, to associate with occludin suggests a role for the CPE protein in localizing occludin within CH-2, i.e., formation of CH-2 does not simply result from interactions between claudin and occludin. Collectively, these findings indicate that native CPE disrupts TJs by simultaneously interacting with claudins, via its C-terminal region, and either directly or indirectly with occludin via its N-terminal region of amino acids 45 to 53.

CPE Epitopes: Is a CPE Vaccine Possible?

CPE fragments prepared during structure/function mapping studies were subjected to reactions (22) with a series of CPE-specific monoclonal antibodies (MAbs). These epitope-mapping studies revealed at least four or five disparate regions scattered throughout the enterotoxin primary sequence that are involved in the presentation of CPE epitopes. Of most significance, the CPE epitope recognized by MAb 3C9 maps to the extreme C terminus of the enterotoxin protein. Since MAb 3C9 is a neutralizing monoclonal antibody that blocks CPE binding to cells (22), the presence of the MAb 3C9 epitope in the C terminus of CPE provides additional evidence for this CPE region having receptor binding activity. Moreover, the presence of a neutralizing linear epitope in C-terminal CPE fragments, which are not by themselves cytotoxic, suggests those fragments could be potential CPE vaccine candidates. The possible use of C-terminal CPE fragments for developing a CPE vaccine was explored by

SMP_Food Microbiology_CH18.indd

485

preparing a 30-mer synthetic peptide corresponding to the extreme C-terminal CPE sequence and then chemically conjugating that peptide to a thyroglobulin carrier (53). When the resultant conjugate was administered intravenously to mice, the conjugate-immunized mice developed very high titers of serum antibodies capable of neutralizing the cytotoxicity of native CPE (53). Effective immunity against C. perfringens type A food poisoning may require a secretory immunoglobulin A response in the intestinal lumen. Therefore, if a CPE vaccine becomes desirable, it may be necessary to pursue approaches that can specifically stimulate the development of intestinal immunoglobulin A immunity against CPE.

Possible Medical Applications of CPE

Pancreatic, breast, and prostate cancer cells often exhibit upregulated expression of claudins capable of serving as CPE receptors (75). This observation has opened the intriguing possibility of using CPE as an anticancer agent, and initial results to this end appear promising. For example, when CPE was injected into a Panc-1 (human pancreatic cancer cell) tumor xenograft growing on the back of a mouse, tumor necrosis and shrinkage occurred without harming the mouse (52). In addition, CPE reduced tumor growth in a mouse brain model of breast cancer metastasis (32). However, many questions must be addressed before CPE can be employed therapeutically in people. For example, do cancer cells readily develop resistance to CPE? Will toxicity limit use of CPE to treating solid tumors by direct injection? Will immune responses against CPE limit its effectiveness? Nevertheless, the possibility of harnessing this toxic protein for medical applications is exciting. In addition, there is current interest in exploiting, for enhanced drug delivery, the ability of noncytotoxic C-terminal CPE fragments to increase paracellular permeability.

CONCLUDING REMARKS Since the publication of the first edition of this book, remarkable progress has been achieved towards understanding the unique pathogenesis of C. perfringens type A food poisoning. Despite those many advances, numerous challenges remain for researchers: identifying the reservoirs of the chromosomal cpe C. perfringens isolates responsible for foodborne illness; improving our understanding of the composition and stoichiometry of mammalian proteins present in the small and large CPE complexes; unraveling the three-dimensional structure of the intact CPE molecule and CPE-containing ­structures;

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

486 better mapping of CPE functional regions mediating pore formation; and evaluating whether inflammation contributes to CPE-induced GI disease. Answering these and other questions through basic research should lead to practical and effective approaches for controlling CPE-associated diseases, including C. perfringens type A food poisoning. For example, identifying the reservoirs for all isolates causing C. perfringens food poisoning would enable public health agencies to design specific hygienic measures to reduce food contamination with these bacteria. Additional studies may also lead to the development of agents capable of blocking CPE expression or activity. Finally, continued research on the mechanism of CPE activity may lead to the development of potent new anticancer agents or to better delivery of therapeutic agents. Preparation of this chapter was supported by Public Health Service grant R37 AI19844-30 from the National Institute of Allergy and Infectious Disease. NOTE ADDED IN PROOF Since this chapter was written, there have been several important developments regarding C. perfringens type A food poisoning. First, the Centers for Disease Control and Prevention have recently increased their estimate of the occurrence of this disease to 1 million cases/year in the United States (CDC, http:// www.cdc.gov/foodborneburden/clostridium-perfringens.html, 2011), making this the second most common bacterial foodborne illness. Second, a report of a severe C. perfringens type A food poisoning outbreak in a Louisiana mental health care facility has now been published (CDC, Morb. Mortal. Weekly Rep., 61:605–608, 2012). Finally, the complete structure of CPE has recently been solved by two independent research groups (D. C. Briggs, C. E. Naylor, J. G. Smedley, 3rd, N. Lukoyanova, S. Robertson, D. S. Moss, B. A. McClane, and A. K. Basak, J. Mol. Biol. 413:138–149, 2011; K. Kitadokoro, K. Nishimura, S. Kamitani, A. Fukui-Miyazaki, H. Toshima, H. Abe, Y. Kamata, Y. Sugita-Konishi, S. Yamamoto, H. Karatani, and Y. Horiguchi, J. Biol. Chem. 286:19549–19555, 2011).

References 1. Bos, J., L. Smithee, B. A. McClane, R. F. Distefano, F. Uzal, J. G. Songer, S. Mallonee, and J. M. Crutcher. 2005. Fatal necrotizing enteritis following a foodborne outbreak of enterotoxigenic Clostridium perfringens type A infection. Clin. Infect. Dis. 15:78–83. 2. Brynestad, S., and P. E. Granum. 1999. Evidence that Tn5565, which includes the enterotoxin gene in Clostridium perfringens, can have a circular form which may be a transposition intermediate. FEMS Microbiol. Lett. 170:281–286. 3. Brynestad, S., M. R. Sarker, B. A. McClane, P. E. Granum, and J. I. Rood. 2001. The enterotoxin (CPE) plasmid from

SMP_Food Microbiology_CH18.indd

486

Clostridium perfringens is conjugative. Infect. Immun. 69:3483–3487. 4. Brynestad, S., B. Synstad, and P. E. Granum. 1997. The Clostridium perfringens enterotoxin gene is on a transposable element in type A human food poisoning strains. Microbiology 143:2109–2115. 5. Carman, R. J., S. Sayeed, J. Li, C. W. Genheimer, M. F. Hiltonsmith, T. D. Wilkins, and B. A. McClane. 2008. Clostridium perfringens toxin genotypes in the feces of healthy North Americans. Anaerobe 14:102–108. 6. Caserta, J. A., M. L. Hale, M. R. Popoff, B. G. Stiles, and B. A. McClane. 2008. Evidence that membrane rafts are not required for the action of Clostridium perfringens enterotoxin. Infect. Immun. 76:5677–5685. 7. Centers for Disease Control and Prevention. 2007. Sur­ veillance for foodborne disease outbreak—United States, 2007. MMWR Morb. Mortal. Wkly. Rep. 59:973–979. 8. Chakrabarti, G., and B. A. McClane. 2005. The importance of calcium influx, calpain, and calmodulin for the activation of Caco-2 cell death pathways by Clostridium perfringens enterotoxin. Cell. Microbiol. 7:129–146. 9. Chakrabarti, G., X. Zhou, and B. A. McClane. 2003. Death pathways activated in Caco-2 cells by Clostridium perfringens enterotoxin. Infect. Immun. 71:4260–4270. 10. Collie, R. E., J. F. Kokai-Kun, and B. A. McClane. 1998. Phenotypic characterization of enterotoxigenic Clostridium perfringens isolates from non-foodborne human gastrointestinal diseases. Anaerobe 4:69–79. 11. Collie, R. E., and B. A. McClane. 1998. Evidence that the enterotoxin gene can be episomal in Clostridium perfringens isolates associated with nonfoodborne human gastrointestinal diseases. J. Clin. Microbiol. 36:30–36. 12. Cornillot, E., B. Saint-Joanis, G. Daube, S. Katayama, P. E. Granum, B. Carnard, and S. T. Cole. 1995. The enterotoxin gene (cpe) of Clostridium perfringens can be chromosomal or plasmid-borne. Mol. Microbiol. 15:639–647. 13. Czeczulin, J. R., R. E. Collie, and B. A. McClane. 1996. Regulated expression of Clostridium perfringens enterotoxin in naturally cpe-negative type A, B, and C isolates of C. perfringens. Infect. Immun. 64:3301–3309. 14. Czeczulin, J. R., P. C. Hanna, and B. A. McClane. 1993. Cloning, nucleotide sequencing, and expression of the Clostridium perfringens enterotoxin gene in Escherichia coli. Infect. Immun. 61:3429–3439. 15. Daube, G., P. Simon, B. Limbourg, C. Manteca, J. Mainil, and A. Kaeckenbeeck. 1996. Hybridization of 2,659 Clostridium perfringens isolates with gene probes for seven toxins (a, b, e, i, q, m, and enterotoxin) and for sialidase. Am. J. Vet. Res. 57:496–501. 16. Deguchi, A., K. Miyamoto, T. Kuwahara, Y. Miki, I. Kaneko, J. Li, B. A. McClane, and S. Akimoto. 2009. Genetic characterization of type A enterotoxigenic Clostridium perfringens strains. PLoS ONE 4:e5598. 17. Ebihara, C., M. Kondoh, M. Harada, M. Fujii, H. Mizuguchi, S. Tsunoda, Y. Horiguchi, K. Yagi, and Y. Watanabe. 2007. Role of Tyr306 in the C-terminal fragment of Clostridium perfringens enterotoxin for modulation of tight junction. Biochem. Pharmacol. 73:824–830.

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

487

18. Fernandez-Miyakawa, M. E., V. Pistone-Creydt, F. Uzal, B. A. McClane, and C. Ibarra. 2005. Clostridium perfringens enterotoxin damages the human intestine in vitro. Infect. Immun. 73:8407–8410. 19. Fisher, D. J., K. Miyamoto, B. Harrison, S. Akimoto, M. R. Sarker, and B. A. McClane. 2005. Association of beta2 toxin production with Clostridium perfringens type A human gastrointestinal disease isolates carrying a plasmid enterotoxin gene. Mol. Microbiol. 56:747–762. 20. Granum, P. E., and M. Richardson. 1991. Chymotrypsin treatment increases the activity of Clostridium perfringens enterotoxin. Toxicon 29:445–453. 21. Granum, P. E., J. R. Whitaker, and R. Skjelkvale. 1981. Trypsin activation of enterotoxin from Clostridium perfringens type A. Biochim. Biophys. Acta 668:325–332. 22. Hanna, P. C., E. U. Wieckowski, T. A. Mietzner, and B. A. McClane. 1992. Mapping functional regions of Clostridium perfringens type A enterotoxin. Infect. Immun. 60:2110–2114. 23. Harry, K. H., R. Zhou, L. Kroos, and S. B. Melville. 2009. Sporulation and enterotoxin (CPE) synthesis are controlled by the sporulation-specific sigma factors SigE and SigK in Clostridium perfringens. J Bacteriol. 191:2728–2742. 24. Heikinheimo, A., M. Lindstrom, P. E. Granum, and H. Korkeala. 2006. Humans as reservoir for enterotoxin gene-carrying Clostridium perfringens type A. Emerg. Infect. Dis. 12:1724–1729. 25. Huang, I. H., M. Waters, R. R. Grau, and M. R. Sarker. 2004. Disruption of the gene (spo0A) encoding sporulation transcription factor blocks endospore formation and enterotoxin production in enterotoxigenic Clostridium perfringens type A. FEMS Microbiol. Lett. 233:233–240. 26. Katahira, J., N. Inoue, Y. Horiguchi, M. Matsuda, and N. Sugimoto. 1997. Molecular cloning and functional characterization of the receptor for Clostridium perfringens enterotoxin. J. Cell Biol. 136:1239–1247. 27. Katahira, J., H. Sugiyama, N. Inoue, Y. Horiguchi, M. Matsuda, and N. Sugimoto. 1997. Clostridium perfringens enterotoxin utilizes two structurally related membrane proteins as functional receptors in vivo. J. Biol. Chem. 272:26652–26658. 28. Kokai-Kun, J. F., K. Benton, E. U. Wieckowski, and B. A. McClane. 1999. Identification of a Clostridium perfringens enterotoxin region required for large complex formation and cytotoxicity by random mutagenesis. Infect. Immun. 67:6534–6541. 29. Kokai-Kun, J. F., and B. A. McClane. 1997. Deletion analysis of the Clostridium perfringens enterotoxin. Infect. Immun. 65:1014–1022. 30. Kokai-Kun, J. F., and B. A. McClane. 1996. Evidence that region(s) of the Clostridium perfringens enterotoxin molecule remain exposed on the external surface of the mammalian plasma membrane when the toxin is sequestered in small or large complex. Infect. Immun. 64:1020–1025. 31. Kokai-Kun, J. F., J. G. Songer, J. R. Czeczulin, F. Chen, and B. A. McClane. 1994. Comparison of Western immunoblots and gene detection assays for identification of

SMP_Food Microbiology_CH18.indd

487

potentially enterotoxigenic isolates of Clostridium perfringens. J. Clin. Microbiol. 32:2533–2539. 32. Kominsky, S. L., M. Vali, D. Korz, T. G. Gabig, S. A. Weitzman, P. Argani, and S. Sukumar. 2004. Clostridium perfringens enterotoxin elicits rapid and specific cytolysis of breast carcinoma cells mediated through tight junction proteins claudin 3 and 4. Am. J. Pathol. 164:1627–1634. 33. Krakauer, T., B. Fleischer, D. L. Stevens, B. A. McClane, and B. G. Stiles. 1997. Clostridium perfringens enterotoxin lacks superantigenic activity but induces an interleukin-6 response from human peripheral blood mononuclear cells. Infect. Immun. 65:3485–3488. 34. Labbe, R. G. 1989. Clostridium perfringens, p. 192–234. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY. 35. Labbe, R. G., and C. L. Duncan. 1977. Evidence for stable messenger ribonucleic acid during sporulation and enterotoxin synthesis by Clostridium perfringens type A. J. Bacteriol. 129:843–849. 36. Lawrence, G. W. 1997. The pathogenesis of enteritis necroticans, p. 198–207. In J. I. Rood, B. A. McClane, J. G. Songer, and R. W. Titball (ed.), The Clostridia: Molecular Genetics and Pathogenesis. Academic Press, London, England. 37. Li, J., and B. A. McClane. 2008. A novel small acid soluble protein variant is important for spore resistance of most Clostridium perfringens food poisoning isolates. PLoS Pathog. 4:e1000056. 38. Li, J., and B. A. McClane. 2006. Comparative effects of osmotic, sodium nitrite-induced, and pH-induced stress on growth and survival of Clostridium perfringens type A isolates carrying chromosomal or plasmid-borne enterotoxin genes. Appl. Environ. Microbiol. 72:7620–7625. 39. Li, J., and B. A. McClane. 2010. Evaluating the involvement of alternative sigma factors SigF and SigG in Clostridium perfringens sporulation and enterotoxin synthesis. Infect. Immun. 78:4286–4293. 40. Li, J., and B. A. McClane. 2006. Further comparison of temperature effects on growth and survival of Clostridium perfringens type A isolates carrying a chromosomal or ­ plasmid-borne enterotoxin gene. Appl. Environ. Microbiol. 72:4561–4568. 41. Li, J., K. Miyamoto, S. Sayeed, and B. A. McClane. 2010. Organization of the cpe locus in CPE-positive Clostridium perfringens type C and D isolates. PLoS ONE 5:e10932. 42. Li, J., D. Paredes-Sabja, M. R. Sarker, and B. A. McClane. 2009. Further characterization of Clostridium perfringens small acid soluble protein-4 (Ssp4) properties and expression. PLoS ONE 4:e6249. 43. Li, J., S. Sayeed, and B. A. McClane. 2007. Prevalence of enterotoxigenic Clostridium perfringens isolates in Pittsburgh (Pennsylvania) area soils and home kitchens. Appl. Environ. Microbiol. 73:7218–7224. 44. McClane, B. A. 1994. Clostridium perfringens enterotoxin acts by producing small molecule permeability alterations in plasma membranes. Toxicology 87:43–67. 45. McClane, B. A., D. M. Lyerly, and T. D. Wilkins. 2006. Enterotoxic clostridia: Clostridium perfringens type A

Manila Typesetting Company

11/08/2012 07:00AM

Foodborne Pathogenic Bacteria

488 and Clostridium difficile, p. 703–714. In V. A Fischetti, R. P. Novick, J. J. Ferretti, D. A. Portnoy, and J. I. Rood (ed.), Gram-Positive Pathogens, 2nd ed. ASM Press, Washington, DC. 46. McClane, B. A., and J. I. Rood. 2001. Clostridial toxins involved in human enteric and histotoxic infections, p. 169–209. In H. Bahl and P. Duerre (ed.), Clostridia: Biotechnology and Medical Applications. Wiley-VCH, Weinheim, Germany. 47. McClane, B. A., F. A. Uzal, M. F. Miyakawa, D. Lyerly, and T. Wilkins. 2006. The enterotoxic clostridia, p. 688– 752. In S. Falkow, M. Dworkin, E. Rosenburg, H. Schleifer, and E. Stackebrandt (ed.), The Prokaryotes, 3rd ed. Springer, New York, NY. 48. McDonel, J. L. 1986. Toxins of Clostridium perfringens types A, B, C, D, and E, p. 477–517. In F. Dorner and H. Drews (ed.), Pharmacology of Bacterial Toxins. Pergamon Press, Oxford, England. 49. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffen, and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5:607–625. 50. Melville, S. B., R. E. Collie, and B. A. McClane. 1997. Regulation of enterotoxin production in Clostridium perfringens, p. 471–487. In J. I. Rood, B. A. McClane, J. G. Songer, and R. Titball (ed.), The Molecular Genetics and Pathogenesis of the Clostridia. Academic Press, London, England. 51. Melville, S. B., R. Labbe, and A. L. Sonenshein. 1994. Expression from the Clostridium perfringens cpe promoter in C. perfringens and Bacillus subtilis. Infect. Immun. 62:5550–5558. 52. Michl, P., M. Buchholz, M. Rolke, S. Kunsch, M. Lohr, B. McClane, S. Tsukita, G. Leder, G. Adler, and T. M. Gress. 2001. Claudin-4: a new target for pancreatic cancer treatment using Clostridium perfringens enterotoxin. Gastroenterology 121:678–684. 53. Mietzner, T. A., J. F. Kokai-Kun, P. C. Hanna, and B. A. McClane. 1992. A conjugated synthetic peptide corresponding to the C-terminal region of Clostridium perfringens type A enterotoxin elicits an enterotoxinneutralizing antibody response in mice. Infect. Immun. 60:3947–3951. 54. Miyamoto, K., G. Chakrabarti, Y. Morino, and B. A. McClane. 2002. Organization of the plasmid cpe locus of Clostridium perfringens type A isolates. Infect. Immun. 70:4261–4272. 55. Miyamoto, K., D. J. Fisher, J. Li, S. Sayeed, S. Akimoto, and B. A. McClane. 2006. Complete sequencing and diversity analysis of the enterotoxin-encoding plasmids in Clostridium perfringens type A non-food-borne human gastrointestinal disease isolates. J. Bacteriol. 188:1585–1598. 56. Miyamoto, K., Q. Wen, and B. A. McClane. 2004. Multiplex PCR genotyping assay that distinguishes between isolates of Clostridium perfringens type A carrying a chromosomal enterotoxin gene (cpe) locus, a plasmid cpe locus with an IS1470-like sequence or a plasmid cpe locus with an IS1151 sequence. J. Clin. Microbiol. 41:1552–1558.

SMP_Food Microbiology_CH18.indd

488

57. Novak, J. S., V. K. Juneja, and B. A. McClane. 2003. An ultrastructural comparison of spores from various strains of Clostridium perfringens and correlations with heat resistance parameters. Int. J. Food Microbiol. 86:239–247. 58. Orsburn, B., S. B. Melville, and D. L. Popham. 2008. Factors contributing to heat resistance of Clostridium perfringens endospores. Appl. Environ. Microbiol. 74: 3328–3335. 59. ProMED-mail. 2010. C. perfringens food intoxication, fatal—USA: (LA) nosocomial. ProMED-mail 2010, 30 May 2010. 60. Rahner, C., L. Mitic, and J. Anderson. 2001. Hetero­ geneity in expression and subcellular localization of claudins 2, 3, 4, and 5 in the rat liver, pancreas, and gut. Gastroenterology 120:411–422. 61. Rahner, C., L. L. Mitic, B. A. McClane, and J. M. Anderson. 1999. Clostridium perfringens enterotoxin impairs bile flow in the isolated perfused rat liver and induces fragmentation of tight junction fibrils. Hepatology 30:326A. 62. Raju, D., P. Setlow, and M. R. Sarker. 2007. AntisenseRNA-mediated decreased synthesis of small, acid-soluble spore proteins leads to decreased resistance of Clostridium perfringens spores to moist heat and UV radiation. Appl. Environ. Microbiol. 73:2048–2053. 63. Robertson, S. L., J. G. Smedley III, U. Singh, G. Chakrabarti, C. M. Van Itallie, J. M. Anderson, and B. A. McClane. 2007. Compositional and stoichiometric analysis of Clostridium perfringens enterotoxin complexes in Caco-2 cells and claudin 4 fibroblast transfectants. Cell. Microbiol. 9:2734–2755. 64. Sarker, M. R., R. J. Carman, and B. A. McClane. 1999. Inactivation of the gene (cpe) encoding Clostridium perfringens enterotoxin eliminates the ability of two cpe-positive C. perfringens type A human gastrointestinal disease isolates to affect rabbit ileal loops. Mol. Microbiol. 33:946–958. 65. Sarker, M. R., R. P. Shivers, S. G. Sparks, V. K. Juneja, and B. A. McClane. 2000. Comparative experiments to examine the effects of heating on vegetative cells and spores of Clostridium perfringens isolates carrying plasmid versus chromosomal enterotoxin genes. Appl. Environ. Microbiol. 66:3234–3240. 66. Sayeed, S., F. A. Uzal, D. J. Fisher, J. Saputo, J. E. Vidal, Y. Chen, P. Gupta, J. I. Rood, and B. A. McClane. 2008. Beta toxin is essential for the intestinal virulence of Clostridium perfringens type C disease isolate CN3685 in a rabbit ileal loop model. Mol. Microbiol. 67:15–30. 67. Sherman, S., E. Klein, and B. A. McClane. 1994. Clostridium perfringens type A enterotoxin induces concurrent development of tissue damage and fluid accumulation in the rabbit ileum. J. Diarrhoeal Dis. Res. 12:200–207. 68. Singh, U., L. L. Mitic, E. Wieckowski, J. M. Anderson, and B. A. McClane. 2001. Comparative biochemical and immunochemical studies reveal differences in the effects of Clostridium perfringens enterotoxin on polarized CaCo-2 cells versus Vero cells. J. Biol. Chem. 276: 33402–33412.

Manila Typesetting Company

11/08/2012 07:00AM

18.  Clostridium perfringens

489

69. Singh, U., C. M. Van Itallie, L. L. Mitic, J. M. Anderson, and B. A. McClane. 2000. CaCo-2 cells treated with Clostridium perfringens enterotoxin form multiple large complex species, one of which contains the tight junction protein occludin. J. Biol. Chem. 275:18407–18417. 70. Smedley, J. G., III, and B. A. McClane. 2004. Fine-mapping of the N-terminal cytotoxicity region of Clostridium perfringens enterotoxin by site-directed mutagenesis. Infect. Immun. 72:6914–6923. 71. Smedley, J. G., III, J. Saputo, J. C. Parker, M. E. Fernandez-Miyakawa, S. L. Robertson, B. A. McClane, and F. A. Uzal. 2008. Noncytotoxic Clostridium perfringens enterotoxin (CPE) variants localize CPE intestinal binding and demonstrate a relationship between CPEinduced cytotoxicity and enterotoxicity. Infect. Immun. 76:3793–3800. 72. Smedley, J. G., III, F. A. Uzal, and B. A. McClane. 2007. Identification of a prepore large-complex stage in the mechanism of action of Clostridium perfringens enterotoxin. Infect. Immun. 75:2381–2390. 73. Sonoda, N., M. Furuse, H. Sasaki, S. Yonemura, J. Katahira, Y. Horiguchi, and S. Tsukita. 1999. Clostridium perfringens enterotoxin fragments remove specific claudins from tight junction strands: evidence for direct ­involvement of claudins in tight junction barrier. J. Cell Biol. 147:195–204. 74. Sparks, S. G., R. J. Carman, M. R. Sarker, and B. A. McClane. 2001. Genotyping of enterotoxigenic Clostridium perfringens isolates associated with gastrointestinal disease in North America. J. Clin. Microbiol. 39:883–888.

SMP_Food Microbiology_CH18.indd

489

75. Swisshelm, K., R. Macek, and M. Kubbies. 2005. Role of claudins in tumorigenesis. Adv. Drug Deliv. Rev. 57:919–928. 76. Todd, E. C. D. 1989. Preliminary estimates of costs of foodborne disease in the United States. J. Food Prot. 52:595–601. 77. Van Itallie, C. M., L. Betts, J. G. Smedley III, B. A. McClane, and J. M. Anderson. 2008. Structure of the claudin-binding domain of Clostridium perfringens enterotoxin. J. Biol. Chem. 283:268–274. 78. Wallace, F. M., A. S. Mach, A. M. Keller, and J. A. Lindsay. 1999. Evidence for Clostridium perfringens enterotoxin inducing a mitogenic and cytokine response in vitro and a cytokine response in vivo. Curr. Microbiol. 38:96–100. 79. Wen, Q., and B. A. McClane. 2004. Detection of enterotoxigenic Clostridium perfringens type A isolates in American retail foods. Appl. Environ. Microbiol. 70:2685–2691. 80. Wieckowski, E., J. F. Kokai-Kun, and B. A. McClane. 1998. Characterization of membrane-associated Clostri­ dium perfringens enterotoxin following Pronase treatment. Infect. Immun. 66:5897–5905. 81. Wieckowski, E. U., A. P. Wnek, and B. A. McClane. 1994. Evidence that an ~50kDa mammalian plasma membrane protein with receptor-like properties mediates the amphiphilicity of specifically-bound Clostridium perfringens ­enterotoxin. J. Biol. Chem. 269:10838–10848. 82. Zhao, Y., and S. B. Melville. 1998. Identification and characterization of sporulation-dependent promoters upstream of the enterotoxin gene (cpe) of Clostridium ­perfringens. J. Bacteriol. 180:136–142.

Manila Typesetting Company

11/08/2012 07:00AM

SMP_Food Microbiology_CH18.indd

490

Manila Typesetting Company

11/08/2012 07:00AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch19

Per Einar Granum Toril Lindbäck

Bacillus cereus

The Bacillus cereus group presently consists of seven Bacillus species, i.e., B. anthracis, B. cereus, B. mycoides, B. pseudomycoides, B. thuringiensis, B. weihenstephanensis, and the most recently recognized member of the group, B. cytotoxicus, which is thermotolerant (80). These species are so closely related that they could be within one species but differentiated from B. anthracis (not treated in this text), which has specific large virulence plasmids. The broad spectrum of B. cereus toxicity ranges from avirulent strains used as probiotics for humans to highly toxic strains responsible for food-related fatalities (81). B. cereus can cause two different types of foodborne illness: the diarrheal type, which was first recognized after a hospital outbreak associated with vanilla sauce in Oslo, Norway, in 1948 (44), and the emetic type, which was described about 20 years later after several outbreaks associated with fried rice in London, England (65). The diarrheal type of foodborne illness is caused by an enterotoxin(s) produced during vegetative growth of B. cereus in the small intestine (39), whereas the emetic illness is caused by a toxin that is preformed by B. cereus while growing in the food (53). For both types of foodborne illness, the food involved usually has been heat treated and surviving spores are the source of the food poisoning, following germination. B. cereus is not a very competitive microorganism but grows well in food after cooking and cooling (<48°C). The heat treat-

19 ment will cause spores to germinate, and in the absence of competing flora, B. cereus grows well, with a generation time (for some strains) as short as 12 minutes under optimal conditions (14). The members of the B. cereus group are common soil saprophytes and are easily spread to many types of foods, especially of plant origin (rice and pasta), but are also frequently isolated from meat, eggs, and dairy products (53). Increased numbers of psychrotolerant strains (mainly B. weihenstephanensis), largely in the dairy industry, have led to increased surveillance of B. cereus in recent years (18, 38, 41, 54, 85). B. cereus foodborne illness is likely highly underreported, as both types of illness are usually relatively mild and typically last for less than 24 hours (53). However, occasional reports of more severe forms of the diarrheal type of B. cereus foodborne illness have been described (39), including three deaths caused by a necrotic enterotoxin (58). Deaths due to ingestion of large amounts of the emetic toxin also have also been reported (22, 61, 77).

CHARACTERISTICS OF THE ORGANISM The aerobic endospore-forming bacteria have traditionally all been placed in the genus Bacillus. Over the past 3 decades, this genus has expanded to accommodate more than 100 species (6), and the family Bacillaceae has been divided into

Per Einar Granum and Toril Lindbäck, Department of Food Safety and Infection Biology, Norwegian School of Veterinary Science, Oslo, Norway.

491

Foodborne Pathogenic Bacteria

492 nine different genera (see the taxonomic outline in Bergey’s Manual 2001 [http://www.cme.msu.edu/Bergeys/]). The first genus is Bacillus, which includes members of the B. cereus group B. anthracis, B. cereus, B. mycoides,­ B. thuringiensis, and more recently, B. pseudo­mycoides (66), B. weihenstephanensis (54), and B. cytotoxicus (80). These bacteria have highly similar 16S and 23S rRNA sequences, indicating that they have diverged from a common evolutionary line relatively recently (6, 7). Although B. anthracis is related to the other species within the B. cereus group based on rRNA sequences, it is the most distinctive member of this group, both in its highly virulent pathogenicity and taxonomically. But even strains of B. cereus have caused anthrax-like symptoms (46); hence, it is even more difficult to identify definite criteria to differentiate the two species. In addition, extensive genomic studies of DNA (including full-genome sequencing) from strains of B. cereus and B. thuringiensis have revealed there is no taxonomic basis for separate species status (17). Nevertheless, the name B. thuringiensis is retained for those strains that synthesize a crystalline inclusion (Cry protein) or δ-endotoxin that is highly toxic to specific insects. The cry genes are usually located on plasmids, and without the relevant plasmid(s), the bacterium is indistinguishable from B. cereus. Cells of the seven species in the B. cereus group are large (cell width, >0.9 µm) and produce central-to-terminal ellipsoid or cylindrical spores that do not distend the sporangia (19, 54, 66). Bacteria of these Bacillus species sporulate easily on most media after 1 to 3 days. Both B. cereus and B. thuringiensis lose their motility during the early stages of sporulation. The B. cereus group can be differentiated using the criteria indicated in Table 19.1.

RESERVOIRS B. cereus is widespread in nature and frequently isolated from soil and growing plants (53). From this natural environment, it is easily spread to foods, especially those of plant origin. It is frequently present in raw materials and ingredients used in the food industry such as vegetables,

starch, and spices (30% of samples contain 102 to 105 CFU/g). Through cross-contamination, it can be spread to other foods, such as meat products (51). Large cell populations of B. cereus have also been detected in feces of cows, hence the possibility of it being spread directly to meat. B. cereus is largely spread to milk and dairy products through soil and grass in contact with the udder of the cow and then into raw milk. B. cereus spores survive milk pasteurization, and after germination, the cells are free from competition from other vegetative cells (5). According to the classic literature (19), B. cereus is unable to grow at temperatures below 10°C and cannot grow in milk and milk products stored at temperatures between 4 and 8°C. However, the psychrotolerant strains (mainly B. weihenstephanensis) can grow well at temperatures as low as 4 to 6°C (18, 38, 85). In addition to rice, pasta, and spices, dairy products are among the most common food vehicles for B. cereus. The majority (if not all) of B. weihenstephanensis isolates are unable to cause food poisoning (82) and will outcompete mesophilic B. cereus in foods stored at temperatures below 8 to 10°C. The closely related B. thuringiensis can produce enterotoxins (20, 68, 69) and can cause foodborne illness when administered to human volunteers (68). This may develop into a problem, as spraying of this organism to protect crops against insect infestations has become a common practice in several countries. It was recently determined that 31 of the 40 randomly selected B. cereus-like strains isolated from foods were classified as B. thuringiensis due to crystal production and/or carriage of cry genes (70). B. thuringiensis has reportedly caused an outbreak of foodborne illness (51). However, because the procedures normally used for identification of B. cereus would not differentiate between the two species of Bacillus (Table 19.1), outbreaks caused by B. thuringiensis may have been unrecognized. To ensure safe spraying of B. thuringiensis, the organism should not produce enterotoxins. Although all members of the B. cereus group harbor genes for at least one of the enterotoxins (Nhe), some strains do not produce detectable amounts of toxin(s).

Table 19.1  Criteria to differentiate members of the B. cereus groupa Species B. cereus B. anthracis B. thuringiensis B. mycoides

Colony morphology White White White/gray Rhizoid

Hemolysis

Motility

+ – + (+)

+ – + –

Susceptible Parasporal to penicillin ­crystal inclusion – + – –

– – + –

a   Symbols: +, positive; –, negative; (+), weakly positive. B. weihenstephanensis can be differentiated from B. cereus based on growth at <7°C and not at 43°C; it can be identified rapidly using rRNA gene- or cspA (cold shock protein A gene)-targeted PCR (54); B. pseudomycoides is not distinguishable from B. mycoides by physiological and morphological characteristics but can be clearly differentiated based on fatty acid composition and 16S RNA sequences (66).

19.  Bacillus cereus

493

FOODBORNE OUTBREAKS The number of outbreaks of B. cereus foodborne illness is highly underestimated. This is mainly due to the relatively short duration of both types of illness (usually <24 hours) and to the fact that complete recovery is, with a few exceptions, rapid after the symptoms subside. The dominating type of disease caused by B. cereus differs among countries. In Japan, the emetic type is reported about 10 times more frequently than the diarrheal type (75), whereas in Europe and North America the diarrheal type is most frequently reported (53, 72). This is likely attributed to eating habits, although contaminated milk was reported to cause at least one large outbreak of the emetic type in Japan (74). Some patients experience both types of B. cereus foodborne illness concurrently (51), with about 5% of B. cereus strains producing both types of toxins (26, 53). Surveillance of foodborne illnesses differs greatly among countries. Hence, it is not possible to directly compare the incidence of outbreaks reported by different countries. The percentage of outbreaks and cases attributed to B. cereus in Japan, North America, and Europe varies from approximately 1 to 47% for outbreaks and from approximately 0.5 to 33% for cases (reports are from different periods between 1960 and 2005) (53, 72). The greatest number of reported B. cereus outbreaks and cases are from Iceland, The Netherlands, and Norway.

There are relatively few outbreaks of salmonellosis and campylobacter enteritis reported in Norway and Iceland; these are the two most frequently reported causes of foodborne illness in most of Europe and the United States. B. cereus in The Netherlands, between 1993 and 1998, was responsible for 12% of outbreaks for which the causative agent was identified. However, the actual incidence of B. cereus illnesses was only 2.0% of the total number of reported cases because most cases of foodborne illness were of unknown etiology (72). Examples of foods involved in foodborne outbreaks are shown in Table 19.2.

CHARACTERISTICS OF DISEASE There are two types of B. cereus foodborne illness. The first type, which is caused by an emetic toxin, results in vomiting, whereas the second type, which is caused by enterotoxin(s), results in diarrhea (53). In a small number of cases, both types of symptoms occur (53), likely due to production of both types of toxin. There has been some debate about whether enterotoxin(s) can be preformed in foods. In reviewing the literature, it appears that the incubation time (>6 hours; average, 12 hours) is too long for the diarrheal illness to be caused by preformed enterotoxin (53), and experiments in animal models revealed that the enterotoxin(s) is degraded as it proceeds to the ileum (35).

Table 19.2  Examples of foods involved in B. cereus food poisoning events Type of food Barbecued chicken Cooked noodles Cream cake Eclair (pastry) Fish soup Hibachi steak Lobster pâté Meat loaf Meat with rice Milk Milkshake Pea soup Sausages School lunch Scrambled egg Several rice dishes Stew Turkey Vanilla sauce Vegetable sprouts Wheat flour dessert

Country

No. of people involved

Type(s) of syndromea

Many countries Spain Norway Thailand Norway United States United Kingdom United States Denmark Many countries United States The Netherlands Ireland, China Japan Norway Many countries Norway United Kingdom, United States Norway (many countries) United States Bulgaria

—b 13 5 >400 20 11 — — >200 — 36 — — 1,877 12 — 152 — >200 3 —

E, D D D E (D) D E, D D D D E, D ? D D E D E, D D D D E, D D

E, emetic syndrome; D, diarrheal syndrome. —, not available.

a  

b  

Foodborne Pathogenic Bacteria

494 However, there is no doubt that B. cereus enterotoxins can be preformed in food, but the number of B. cereus cells necessary to produce enterotoxin in such food is at least 2 orders of magnitude higher than the number of cells needed to cause emetic toxin food poisoning (18, 35). Usually, foods with such large populations of B. cereus will no longer be acceptable to the consumer, although food containing >107 B. cereus cells/ml may not always appear spoiled. The characteristics of the two types of B. cereus foodborne illness are described in Table 19.3. Some recent reports of B. cereus foodborne illness have been of increased concern, including three outbreaks with fatal outcomes. An outbreak in Norway associated with eating stew containing approximately 104 to 105 B. cereus cells per serving affected 17 people, of which 3 were hospitalized, one for 3 weeks. The onset for these three patients was late, greater than 24 h (see “The B. cereus Spore” below). In another case of B. cereus foodborne illness, the emetic toxin cereulide was responsible for the death of a 17-year-old Swiss boy, due to fulminant liver failure (61). A large amount of B. cereus emetic toxin was detected in residue in the pan used to reheat the implicated food (pasta) and in the boy’s liver and bile. Similarly, a 7-year-old girl died in Belgium only 13 hours after ingesting a pasta salad that had been stored over several days in a refrigerator where the temperature was 14°C. Her 2-year-old brother, who had eaten only a small amount of the salad, was hospitalized for 8 days (22). The latest report of a serious outbreak of B. cereus emetic type is from Japan (77), where three family members began vomiting 30 minutes after consuming reheated fried rice. After 6 hours, a 1-year-old boy died of acute encephalopathy. A 2-yearold girl recovered rapidly only after plasma exchange and subsequent hemodialysis. Their mother recovered after fluid therapy.

The most recently discovered B. cereus enterotoxin,­ c­ ytotoxin K (CytK), is similar to the b-toxin of Clostridium perfringens (and other related toxins) and was the causative agent in a severe outbreak of B. cereus foodborne illness in France in 1998 (58). Several people in this outbreak developed bloody diarrhea, and three died. This was the first recorded outbreak of B. cereus necrotic enteritis, although it is not nearly as severe as C. perfringens type C foodborne illness (16). The B. cereus strain isolated from this outbreak is different from the other members of the B. cereus group and grows at higher temperatures (thermotolerant). This strain has recently been recognized as the seventh member of the B. cereus group (B. cytotoxicus) (80).

DOSE AND SUSCEPTIBLE POPULATIONS After the first recognized diarrheal outbreak of B. cereus foodborne illness in Oslo (vanilla sauce), Professor Hauge isolated the causative agent, grew it to 4 × 106 CFU/ml, and drank 200 ml of the culture (44). Approximately 13 hours later, he developed abdominal pain and watery diarrhea that lasted for approximately 8 hours. The cell population of B. cereus ingested was approximately 8 × 108. Counts of B. cereus ranging from 200 to 109 CFU/g (or ml) (35, 37, 44, 53) have been reported in foods incriminated in outbreaks, indicating the total dose ranged from approximately 5 × 104 to 1011. The total number of B. cereus cells required to be ingested to produce illness is likely in the range of 105 to 108 viable cells or spores. For the lower dose, it is likely that only the spores, which will survive the stomach acid barrier, can cause illness. Still, the wide range of infective dose is also in part due to differences in the amount of enterotoxin produced by different strains (35). Hence, food containing more than 103 B. cereus spores/g cannot be considered completely safe for consumption. Little is known

Table 19.3  Characteristics of the two types of illness caused by B. cereusa Characteristic Dose causing illness Toxin produced Type of toxin Incubation period Duration of illness

Diarrheal syndrome

10 –10 cells (total) In the small intestine of the host Protein; enterotoxin(s) 8–16 h (occasionally >24 h) 12–24 h (occasionally several days) Symptoms Abdominal pain, watery diarrhea occasionally with nausea Foods most frequently Meat products, soups, vegetables, implicated puddings/sauces, and milk/milk products 5

7

Based on references 35, 53, and 74.

a 

Emetic syndrome 10 –10 cells (per g in foods) Preformed in foods Cyclic peptide; emetic toxin 0–5 h 6–24 h 5

8

Nausea, vomiting, and malaise (sometimes followed by diarrhea, due to production of enterotoxin) Fried and cooked rice, pasta, pastry, and noodles

19.  Bacillus cereus

495

about susceptible populations, but the more severe types of illness have occasionally involved young athletes (<19 years) or the elderly (>60 years) (36, 37, 58). More studies are needed to address this important question.

VIRULENCE FACTORS/MECHANISMS OF PATHOGENICITY The two types of B. cereus foodborne illness are caused by very different types of toxins. The emetic toxin, cereulide, is a 1.2-kDa dodecadepsipeptide synthesized by a nonribosomal peptide synthetase (2, 3, 27), whereas the diarrheal disease is caused by enterotoxins of protein structures (9, 12, 35, 36, 53, 58).

The Emetic Toxin

The emetic toxin (Table 19.4) causes emesis (vomiting) only, and its structure was for many years a mystery because the only detection system involved living primates (49, 53). However, the discovery that the toxin could be detected by HEp-2 cells (vacuolation activity) (49) has led to its isolation and structure determination (2, 3). Although there have been some questions as to whether the emetic toxin and the vacuolating factor are the same component (74, 76), there is now no doubt that they are identical (2, 3, 75). The emetic toxin has been named cereulide and consists of a ring structure of three repeats of four amino- and/or oxy-acids: (dTable 19.4  Properties of the B. cereus emetic toxin cereulidea Trait Molecular mass Structure Isoelectric point Antigenic Biological activity in living primates and Asian house shrew Receptor Ileal loop tests (rabbit, mouse) Cytotoxic HEp-2 cells Stability to heat Stability to pH Effect of proteolysis (trypsin, pepsin) Conditions under which toxin is produced Mechanisms of production

Property/activity 1.2 kDa Ring-shaped peptide Uncharged No Vomiting

5-HT3 (stimulation of the vagus afferent) None No Vacuolation activity 90 min at 121°C Stable at pH 2–11 None In food (rice and milk at 12–32°C) Produced by a nonribosomal peptide synthetase

Based on references 2, 3, 27, 53, 74, 75, and 82.

a 

O-Leu-d-Ala-l-O-Val-l-Val)3. This ring structure (dode­ cadepsi­peptide) has a molecular mass of 1.2 kDa, and acts as a K+ ionophore, like valinomycin (2). Cereulide is produced by cereulide synthetase (Ces) via a nonribosomal peptide synthetase (NRPS) mechanism (27). The complete 24-kb ces gene cluster is comprised of seven genes encoding the machinery required for the synthesis of cereulide and is located on a large plasmid (25, 47). Cereulide is resistant to heat, pH, and proteolysis and is not antigenic (2, 3, 53) (Table 19.4). Cereulide causes inhibition of mitochondrial activity (inhibition of fatty acid oxidation) and stimulates the vagus afferent by binding to the 5-HT3 receptor (3). Mice injected intraperitoneally with synthetic cereulide developed histopathological changes in the liver, and massive degeneration of hepatocytes was observed (85). The serum values of hepatic enzymes were highest on days 2 and 3 after the inoculation of cereulide and rapidly decreased thereafter. General recovery from pathological changes and regeneration of hepatocytes were observed after 4 weeks. Until recently, neither of the characterized emetic strains grew below 10°C, and neither of them produces the Hbl enterotoxin (26). However, in 2006 a few strains of B. weihenstephanensis were also reported to produce cereulide (84). Although they grow at lower temperatures than other members of the B. cereus group, the emetic toxin is not produced below 8°C (83). For most emetic strains, the toxin is produced best at temperatures in the range of 16 to 25°C.

Enterotoxins

B. cereus produces different proteins (or protein complexes), referred to as enterotoxins (8, 39, 45, 58), that can cause foodborne illness (Table 19.5). Two of the enterotoxins have three components and are structurally related, whereas the third (CytK) is a single 34-kDa protein. The three-component hemolysin (Hbl, consisting of three proteins, B, L1, and L2) with enterotoxin activity was the first to be fully characterized (12, 13). This toxin also has dermonecrotic and vascular permeability activities and causes fluid accumulation in ligated rabbit ileal loops. Hbl has been suggested to be a primary virulence factor in B. cereus diarrhea (9), but several strains without genes for this toxin have also caused food poisoning. Evidence has revealed that all three components of Hbl are necessary for maximal enterotoxin activity (9). It was first suggested that the B protein is the component that binds Hbl to the target cells and that L1 and L2 have lytic functions (9). However, more recently another model for the action of Hbl has been proposed, suggesting that the components of Hbl

Foodborne Pathogenic Bacteria

496 Table 19.5  Toxins produced by B. cereus Toxin Hemolysin BL (Hbl) Nonhemolytic enterotoxin (Nhe) Cytotoxin K (CytK) Emetic toxin (cereulide)

Type/size

Food poisoning

Reference(s)

Protein, 3 components Protein, 3 components

Probably Yes

9, 10, 45, 71 40, 56, 59, 60

Protein, 1 component, 34 kDa Cyclic peptide, 1.2 kDa

Yes, 3 deaths Yes, several deaths

58 2, 3, 22, 61, 77

bind to target cells independently and then constitute a membrane-attacking complex resulting in a colloid osmotic lysis mechanism (10). More research is needed to elucidate the mode of action for Hbl. A 1:1:1 ratio of the three components yields the greatest biological activity (9). Substantial heterogeneity has been observed in the Hbl components, and individual strains can produce different combinations of single or multiple bands of each component (73). This is due to sequence variation of multiple genes of hbl (52). About 50% of B. cereus strains harbor hbl genes. A second three-component enterotoxin, Nhe, was first described by Lund and Granum (59, 60). The Nhe enterotoxin was purified from a B. cereus strain ­isolated after a large food-associated outbreak in Norway in 1995. Nhe is a pore-forming toxin with structural and functional properties similar to those of the ClyA/SheA family of hemolysins in Escherichia coli (30). Initially, the hemolytic activity of Nhe was not detected, but more recently Nhe was found to possess hemolytic ­activity, although not as high as that of Hbl (30). The three components of Nhe are different from those of Hbl, although they are very similar in three-­dimensional structure, possessing an a-helix bundle and a unique subdomain containing a hydrophobic b-hairpin (30). The hydrophobic b-hairpin of NheC is essential for binding of NheC to the cell membrane (57). The enterotoxin is most active when the ratio between NheA, NheB, and NheC is 10:10:1 (30, 56); however, NheC is expressed in lower ratios in vivo (23) than the optimal ratio. This is likely why NheC is not detected on twodimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (33). If higher-than-optimal ratios for NheC activity are used in experiments, the biological activity is rapidly reduced, and at a 1:1:1 ratio, very little biological activity is left (30, 56). Recent studies have revealed that full activity­ of Nhe requires a specific binding order of the three components, whereby the first step is associated with the binding of NheC and NheB to the cell surface followed by binding of NheA, leading to cell lysis (57). More than 99% of B. cereus strains are able to produce Nhe, although not all at 32 to 37°C.

The last characterized B. cereus toxin associated with food poisoning is CytK. This toxin belongs to a family of b-barrel toxins and is similar to the b-toxin of C. perfringens and was the cause of symptoms in a severe outbreak of B. cereus foodborne illness in a nursing home in France in 1998 (58). In this outbreak (B. cereus necrotic enteritis), several people developed bloody diarrhea and three patients died. Similar to the other members of this b-barrel toxin family, CytK is inserted into membranes as a heptamer and makes a pore of 7Å in diameter in planar lipid bilayers (43). The B. cereus strain causing the fatal outbreak in France (58) is highly cytotoxic and will be the type strain for B. cytotoxicus. The CytK produced by this strain differs from CytK produced by most other B. cereus strains (cytK is present in about 35 to 40% of strains), and the toxin is named CytK-1 (31, 58). CytK-1 is 89% identical in amino acid sequence to the more commonly occurring CytK (28). The high virulence of this strain is thought to be due to the greater cytotoxic activity of CytK-1 compared to CytK and to a high level of cytK expression. CytK-1 is about 5 times more toxic on epithelial cells than the other CytK, although the activities of these two toxins on erythrocytes are very similar (31). CytK-1 is produced by a few strains that are genetically remote from other strains of the B. cereus group (28). These strains harbor a novel gene variant encoding Nhe, which is not detected using ordinary nhe PCR primers or monoclonal antibodies against the Nhe proteins (28). Although studies have revealed that the total toxicity of B. cereus supernatant fluids for most strains is largely due to Nhe (64), Hbl and CytK cannot be excluded as contributing to diarrhea, and it is documented that CytK can be fatal when produced in large amounts by some strains.

Regulation of Enterotoxin Production

All three proteins of the Hbl complex (8) are transcribed from one operon (hbl) in a 5-kb polycistronic mRNA, transcribing hblC (component L2), hblD (component L1), and hblA (component B) (71). The three genes encoding the Nhe components, nheA, nheB, and nheC, constitute the nhe operon (Fig. 19.1) and are transcribed in a 3.7-kb

19.  Bacillus cereus

497

Figure 19.1  The nhe operon with promoter and regulatory sites. The indicated regulatory sequences are from B. cereus type strain (ATCC 14579). The consensus PlcR box (1) and the putative PlcR box with one mismatch in strain ATCC 14579 (underlined) and six bases between the palindromic flanks of the recognition sequence instead of four as in the established consensus (40) and the two predicted cre sites (with mismatches towards the B. subtilis consensus underlined) are shown as boxes. The inverted repeat between nheB and nheC (40) is indicated as a stem-loop structure. The bent arrows indicate the positions of transcriptional start sites, preceded by putative −10 and −35 regions. The transcriptional start site closest to the nheA gene was identified using RNA isolated from strains NVH 0075/95 and NVH 1230/88 (56), whereas the one further upstream was identified using a plasmid carrying the nhe promoter from B. thuringiensis strain 407 (1). The scale (in bp) is shown in the lower part of the figure. doi:10.1128/9781555818463.ch19f1

polycistronic mRNA (56). The properties of the three Nhe proteins are described in Table 19.6. There is a gap of 109 bp between nheB and nheC, and this gap, containing an inverted repeat of 13 bp, is likely important for regulation of translation of nheC mRNA (40, 56). The low concentration of the NheC component necessary for full activity of the Nhe complex may be due to translational repression when the ribosome-binding site is hidden in a secondary structure. Twenty members of the B. cereus group have been completely sequenced, and database searches reveal that the organization of both the hbl and the nhe operons is very well conserved between the strains. Sec-type signal peptides have been identified in CytK and all components of Hbl and Nhe, and the three toxins appear to be secreted via the Sec pathway (29). Expression of the B. cereus toxins Hbl, Nhe, and CytK is regulated by the PlcR quorum-sensing system (1, 78). Bacteria use quorum sensing to coordinate bacterial processes in response to environmental situations such as cell density. These communication systems Table 19.6  Properties of the Nhe proteinsa Protein NheA NheB NheC

Signal peptide Active Mol wt size (aa) protein size (aa) (active protein) 26 30 30

Based on reference 40.

a  

360 372 329

41,019 39,820 36,481

pI 5.13 5.61 5.28

are based on the secretion and recognition of cell-tocell signaling molecules. The transcriptional activator PlcR was first identified as an activator required for transcription of plcA at the onset of stationary phase in B. thuringiensis (55). PlcR controls expression of 45 genes: 22 of the proteins encoded by the PlcR regulon are secreted, mainly toxins, phospholipases, and proteases; 18 are cell wall proteins; and 5 are cytoplasmic proteins (32). The activity of PlcR depends on peptide-activating PlcR (PapR). The papR gene encodes a 48-amino-acid polypeptide with an N-terminal signal peptide sequence. PapR is secreted via the SecA pathway, processed, and then reimported into the bacterial cell via the oligopeptide permease (Opp) system (34). It has recently been determined that the in vivo active form of PapR is the C-terminal heptapeptide of the precursor (15). Crystal structure studies have revealed that PlcR is composed of an N-terminal helix-turn-helix DNAbinding domain and a C-terminal regulatory domain composed of five degenerated tetratricopeptide repeats (TPRs) (21). TPRs are structurally conserved helical domains involved in protein-protein or protein-peptide interactions (13), and when PapR binds to the TPR domain, the helix-turn-helix domain is rearranged, allowing DNA association (21). PapR-PlcR then binds to a specific DNA sequence called the “PlcR box” (TATGNAN4TNCATA) located upstream from the

498 controlled genes, resulting in ­ upregulation of expression of plcR and papR as well as the other genes in the PlcR regulon (1, 78). The PlcR activation is strain specific. Four groups of PlcR-PapR pairs defining four different pherotypes have been identified: PlcRI, PlcRII, PlcRIII, and PlcRIV (78, 79). The strain specificity depends on PapR originating from the same strain or from a member of the same pherogroup. Some level of specificity is determined by the first and the fifth residues of the PapR heptapeptide (78, 79). Crystal structure studies of PlcR revealed a striking similarity of PlcR to the Enterococcus faecalis sex pheromone receptor PrgX (21). They revealed that gram-positive bacterial quorum sensors that bind autoinducer peptides are derived from a common ancestor and comprise the RNPP (Rap/NprR/PlcR/PgX) family, whereas NprR from B. thuringiensis is a transcriptional regulator and Rap proteins from bacilli control competence and sporulation by binding to cellular ligands (21). In B. anthracis, PlcR is inactivated by a nonsense mutation in the plcR gene (62), and consequently, genes regulated by PlcR are expressed at very low levels compared to those of B. cereus. Interestingly, most B. weihenstephanensis strains also produce enterotoxins, but usually at temperatures below 30°C (82). This temperature regulation is not under the control of the above-described (PlcR-PapR) system and has yet to be elucidated. Hbl and Nhe have also been suggested to be subject to catabolite repression, at least during anaerobiosis, because transcription of hbl was repressed by increasing concentrations of glucose (24) and growth on sucrose produced higher levels of Hbl and Nhe than did growth on glucose (67). Genes regulated by catabolite repression harbor catabolite responsive element(s) (cre sites), for which the consensus sequence in B. subtilis has been determined to be TGWNANCGNT NWCA (48) or WWTGNAARCGNWW WCAWW (63). A search of the hbl, nhe, and cytK regulatory regions for the presence of this sequence revealed two potential cre sites in the nhe regulatory region (Fig. 19.1). The concept of B. cereus toxin regulation by catabolite repression is perhaps not unexpected, as from a bacterial point of view, deploying virulence factors to liberate required nutrients does not appear necessary when easily metabolized carbohydrates are available.

The B. cereus Spore

The spore of B. cereus is an important factor in contributing to foodborne illness. First, the B. cereus spore is more hydrophobic than spores from any other Bacillus spp., which enables it to adhere to several types of sur-

Foodborne Pathogenic Bacteria faces (50). Hence, it is difficult to remove during cleaning and is a challenging target for disinfectants. B. cereus spores also contain appendages and/or pili (4, 50) that are, at least partly, involved in adhesion (50), although these structures also contribute to spores’ ability to stick together in large aggregates. Not only can these properties of B. cereus spores enable them to withstand cleaning and disinfection and hence remain present on equipment surfaces to contaminate different foods, but also they aid in adherence to epithelial cells. Studies have revealed that spores of many strains associated with foodborne outbreaks (4) can adhere to Caco-2 cells in culture and that these properties are associated with hydrophobicity and possibly to the spore’s appendages (5).

Food Spoilage

As described above, B. cereus contamination is very difficult to avoid in many food products. Spices and other dried components (starch) typically contain from a few to more than 103 spores per gram; hence, many food products will be contaminated. If the final product is not sufficiently heat treated (>100°C), the spores will survive, and outgrowth will occur if the food is held at >10°C over time. This is the reason for outbreaks (mainly emetic) involving heat-treated rice and pasta stored at ambient temperatures overnight. Other lightly heat-treated products may also be at risk, but most such products (e.g., those packaged sous vide) are usually stored cold and heat treated a second time at home before consumption. A special problem is pasteurized milk, in which the mesophilic strains will not grow if it is stored properly; however, B. weihenstephanensis will grow in milk at approximately 4°C and will reach large enough cell numbers in the milk to cause what is referred to as “sweet curdling” and “bitty cream.” Long before reaching this point, the number of B. cereus cells is high enough to cause food poisoning. Yet milk is rarely associated with outbreaks of B. cereus food poisoning because B. weihenstephanensis dominates in milk, and most of these strains are unable to produce enterotoxin in the human gut, although they can produce the toxins at lower temperatures (81).

Commercial Methods for Detection of the B. cereus Toxins

No commercial kit for detection of the emetic toxin (cereulide) is yet available. With the known structure of cereulide (2), it is likely that a kit will be available in the near future. However, because the genes encoding the NRPS responsible for cereulide production are known (25, 27), PCR can be used for detection of these genes. Screening

19.  Bacillus cereus for cereulide production by B. cereus strains can also easily be done by using a sperm motility assay (42). Neither of the two available commercial immunoassays can quantify the toxicity of the B. cereus enterotoxins. The assay from Oxoid measures the presence of the HblC (L2) component (11), whereas the Tecra kit detects mainly the NheA (45-kDa) component (38, 39, 60). However, if one or both of the commercial kits react positively with proteins from B. cereus supernatant fluids, it is likely that the strain is enterotoxin positive. If culture supernatant fluids are also cytotoxic, the strains can be regarded as enterotoxin positive. At present, there is no commercial method available for detecting CytK.

Concluding remarks B. cereus is a normal inhabitant of soil but is also well adapted to grow in the gut of warm-blooded animals. B. cereus is frequently isolated from a variety of foods, including vegetables, dairy products, and meat. This bacterium causes an emetic or a diarrheal type of foodassociated illness that largely occurs in developed countries. The diarrheal type of illness is most prevalent in the western hemisphere, whereas the emetic type is most prevalent in Japan and other Asian countries. Desserts, meat dishes, and dairy products are most frequently associated with diarrheal illness, whereas rice and pasta are the most common vehicles of emetic illness. A B. cereus emetic toxin has been isolated and characterized and also synthesized. Three types of B. cereus enterotoxins involved in outbreaks of foodborne illness have been identified. Two of these enterotoxins possess three components and are related, whereas the third is a one-component protein (CytK). Deaths have been caused by the emetic toxin and by a strain producing only CytK, whereas factors yet to be identified appear to be involved in the diarrheal disease. Some strains of the B. cereus group are able to grow at refrigeration temperature. These variants raise concerns regarding the safety of cooked, refrigerated foods with extended shelf lives. B. cereus spores are adhesive to many surfaces and can survive normal cleaning and disinfection (but not hypochlorite, some peroxides, or UVC) procedures. B. cereus foodborne illness is likely to be highly underreported because of its relatively mild symptoms with short duration. However, increased consumer interest for precooked, chilled food products with extended shelf lives may be well suited for B. cereus survival and growth. Such foods could increase the prominence of B. cereus as a foodborne pathogen.

499

References   1. Agaisse, H., M. Gominet, O. A. Okstad, A. B. Kolsto, and D. Lereclus. 1999. PlcR is a pleiotropic regulator of extracellular virulence factor gene expression in Bacillus thuringiensis. Mol. Microbiol. 32:1043–1053.   2. Agata, N., M. Mori, M. Ohta, S. Suwan, I. Ohtani, and M. Isobe. 1994. A novel dodecadepsipeptide, cereulide, isolated from Bacillus cereus causes vacuole formation in HEp-2 cells. FEMS Microbiol. Lett. 121:31–34.   3. Agata, N., M. Ohta, M. Mori, and M. Isobe. 1995. A novel dodecadepsi­peptide, cereulide, is an emetic toxin of Bacillus cereus. FEMS Microbiol. Lett. 129:17–20.   4. Andersson, A., P. E Granum, and U. Rönner. 1998. The adhesion of Bacillus cereus spores to epithelial cells might be an additional virulence mechanism. Int. J. Food Microbiol. 39:93–99.   5. Andersson, A., U. Rönner, and P. E. Granum. 1995. What problems does the food industry have with the sporeforming pathogens Bacillus cereus and Clostridium perfringens? Int. J. Food Microbiol. 28:145–156.   6. Ash, C., and M. D. Collins. 1992. Comparative analysis of 23S ribosomal RNA gene sequence of Bacillus anthracis and emetic Bacillus cereus determined by PCRdirect sequencing. FEMS Microbiol. Lett. 73:75–80.   7. Ash, C., J. A. Farrow, M. Dorsch, E. Steckebrandt, and M. D. Collins. 1991. Comparative analysis of Bacillus anthracis, Bacillus cereus, and related species on the basis of reverse transcriptase sequencing of 16S rRNA. Int. J. Syst. Bacteriol. 41:343–346.   8. Beecher, D. J., and J. D. Macmillan. 1991. Characterization of the components of hemolysin BL from Bacillus cereus. Infect. Immun. 59:1778–1784.   9. Beecher, D. J., J. L. Schoeni, and A. C. L. Wong. 1995. Enterotoxin activity of hemolysin BL from Bacillus cereus. Infect. Immun. 63:4423–4428.   10. Beecher, D. J., and A. C. L Wong. 1997. Tripartite hemo­ lysin BL from Bacillus cereus. Hemolytic analysis of component interaction and model for its characteristic para­ doxical zone phenomenon. J. Biol. Chem. 272:233–239.   11. Beecher, D. J., and A. C. L. Wong. 1994. Identification and analysis of the antigens detected by two commercial Bacillus cereus diarrheal enterotoxin immunoassay kits. Appl. Environ. Microbiol. 60:4614–4616.   12. Beecher, D. J., and A. C. L. Wong. 1994. Improved purification and characterization of hemolysin BL, a hemolytic dermonecrotic vascular permeability factor from Bacillus cereus. Infect. Immun. 62:980–986.   13. Blatch, G. L., and M. Lassle. 1999. The tetratrico­peptide repeat: a structural motif mediating protein-protein interactions. Bioessays 21:932–939.   14. Borge, G. A., M. Skeie, T. Langsrud, and P. E. Granum. 2001. Growth and toxin profiles of Bacillus cereus isolated from different food sources. Int. J. Food Microbiol 69:237–246.   15. Bouillaut, L., S. Perchat, S. Arold, S. Zorrilla, L. Slamti, C. Henry, M. Gohar, N. Declerck, and D. Lereclus. 2008. Molecular basis for group-specific activation of the virulence regulator PlcR by PapR heptapeptides. Nucleic Acids Res. 36:3791–3801.

500   16. Brynestad, S., and P. E. Granum. 2002. Clostridium perfringens and foodborne infections. Int. J. Food Microbiol. 74:195–202.   17. Carlson, C. R., D. A. Caugant, and A.-B. Kolstø. 1994. Genotypic diversity among Bacillus cereus and Bacillus thuringiensis strains. Appl. Environ. Microbiol. 60:1719–1725.   18. Christiansson, A., A. S. Naidu, I. Nilsson, T. Wadström, and H.-E. Pettersson. 1989. Toxin production by Bacillus cereus dairy isolates in milk at low temperatures. Appl. Environ. Microbiol. 55:2595–2600.   19. Claus, D., and R. C. W. Berkeley. 1986. Genus Bacillus, p. 1105–1139, vol . 2. In P. H. A. Seneath (ed.), Bergey’s Manual of Systematic Bacteriology. The Williams and Wilkins Co., Baltimore, MD.   20. Damgaaerd, P. H., H. D. Larsen, B. M. Hansen, J. Bresciani, and K. Jørgensen. 1996. Enterotoxin-producing strains of Bacillus thuringiensis isolated from food. Lett. Appl. Microbiol. 23:146–150.   21. Declerck, N., L. Bouillaut, D. Chaix, N. Rugani, L. Slamti, F. Hoh, D. Lereclus, and S. T. Arold. 2007. Structure of PlcR: insights into virulence regulation and evolution of quorum sensing in Gram-positive bacteria. Proc. Natl. Acad. Sci. USA 104:18490–18495.   22. Dierick, K., E. Van Coillie, I. Swiecicka, G. Meyfroidt, H. Devlieger, A. Meulemans, G. Hoedemaekers, L. Fourie, M. Heyndrickx, and J. Mahillon. 2005. Fatal family outbreak of Bacillus cereus-associated food poisoning. J. Clin. Microbiol. 43:4277–4279.   23. Dietrich, R., M. Moravek, C. Bürk, P. E. Granum, and E. Märtlbauer. 2005. Production and characterization of antibodies against each of the three subunits of the Bacillus cereus non-hemolytic enterotoxin complex. Appl. Environ. Microbiol. 257:293–298.   24. Duport, C., S. Thomassin, G. Bourel, and P. Schmitt. 2004. Anaerobiosis and low specific growth rates enhance hemolysin BL production by Bacillus cereus F4430/73. Arch. Microbiol. 182:90–95.   25. Ehling-Schulz, M., M. Fricker, H. Grallert, P. Rieck, M. Wagner, and S. Scherer. 2006. Cereulide synthetase gene cluster from emetic Bacillus cereus: structure and location on a mega virulence plasmid related to Bacillus anthracis toxin plasmid pXO1. BMC Microbiol. 6:20.   26. Ehling-Schulz, M., B. Svensson, M.-H. Guinebritiere, T. Lindbäck, M. Andersson, A. Schulz, A. Christiansson, P. E. Granum, E. Märtelbauer, C. Nyguyen-The, M. Salkinoja-Salonen, and S. Scherer. 2005. Emetic toxin formation of Bacillus cereus is restricted to a single evolutionary lineage of closely related strains. Microbiology 151:183–197.   27. Ehling-Schulz, M., N. Vukov, A. Schulz, R. Shaheen, M. Andersson, E. Martlbauer, and S. Scherer. 2005. Identification and partial characterization of the nonribosomal peptide synthetase gene responsible for cereulide production in emetic Bacillus cereus. Appl. Environ. Microbiol. 71:105–113.   28. Fagerlund, A., J. Brillard, R. Furst, M. H. Guinebretiere, and P. E. Granum. 2007. Toxin production in a rare and genetically remote cluster of strains of the Bacillus cereus group. BMC Microbiol. 7:43.

Foodborne Pathogenic Bacteria   29. Fagerlund, A., T. Lindbäck, and P. E. Granum. 2010. Bacillus cereus cytotoxins Hbl, Nhe and CytK are secreted via the Sec translocation pathway. BMC Microbiol. 10:304.   30. Fagerlund, A., T. Lindbäck, A. K. Storset, P. E. Granum, and S. P. Hardy. 2008. Bacillus cereus Nhe is a poreforming toxin with structural and functional properties similar to the ClyA (HlyE, SheA) family of haemolysins, able to induce osmotic lysis in epithelia. Microbiology 154:693–704.   31. Fagerlund, A., O. Ween, T. Lund, S. P. Hardy, and P. E. Granum. 2004. Different cyto-toxicity of CytK and CytK-like proteins from Bacillus cereus. Microbiology 150:2689–2697.   32. Gohar, M., K. Faegri, S. Perchat, S. Ravnum, O. A. Okstad, M. Gominet, A. B. Kolsto, and D. Lereclus. 2008. The PlcR virulence regulon of Bacillus cereus. PLoS ONE 3:e2793.   33. Gohar, M., O. A. Okstad, N. Gilois, V. Sanchis, A. B. Kolsto, and D. Lereclus. 2002. Two-dimensional electrophoresis analysis of the extracellular proteome of Bacillus cereus reveals the importance of the PlcR regulon. Proteomics 2:784–791.   34. Gominet, M., L. Slamti, N. Gilois, M. Rose, and D. Lereclus. 2001. Oligopeptide permease is required for expression of the Bacillus thuringiensis plcR regulon and for virulence. Mol. Microbiol. 40:963–975.   35. Granum, P. E. 1994. Bacillus cereus and its toxins. J. Appl. Bacteriol. 76:61S–66S.   36. Granum, P. E., A. Andersson, C. Gayther, M. C. te Giffel, H. D. Larsen, T. Lund, and K. O’Sullivan. 1996. Evidence for a further enterotoxin complex produced by Bacillus cereus. FEMS Microbiol. Lett. 141:145–149.   37. Granum, P. E., and T. C. Baird-Parker. 2000. Bacillus spp., p. 1029–1039. In B. Lund, T. Baird-Parker, and G. Gould (ed.), The Microbiological Safety and Quality of Food. Aspen Publishers, Aspen Hill, MD.   38. Granum, P. E., S. Brynestad, and J. M. Kramer. 1993. Analysis of enterotoxin production by Bacillus cereus from dairy products, food poisoning incidents and non-gastrointestinal infections. Int. J. Food Microbiol. 17:269–279.   39. Granum, P. E., and T. Lund. 1997. Bacillus cereus enterotoxins. FEMS Microbiol. Lett. 157:223–228.   40. Granum, P. E., K. O’Sullivan, and T. Lund. 1999. The sequence of the non-haemolytic enterotoxin operon from Bacillus cereus. FEMS Microbiol. Lett. 177:225–229.   41. Griffiths, M. W. 1990. Toxin production by psychrotrophic Bacillus spp. present in milk. J. Food Prot. 53:790–792.   42. Haggblom, M. M., C. Apetroaie, M. A. Andersson, and M. S. Salkinoja-Salonen. 2002. Quantitative analysis of cereulide, the emetic toxin of Bacillus cereus, produced under various conditions. Appl. Environ. Microbiol. 68:2479–2483.   43. Hardy, S. P., T. Lund and P. E. Granum. 2001. CytK toxin of Bacillus cereus forms pores in planar lipid bilayers and is cytotoxic to intestinal epithelia. FEMS Microbiol. Lett. 197:47–51.

19.  Bacillus cereus   44. Hauge, S. 1955. Food poisoning caused by aerobic spore forming bacilli. J. Appl. Bacteriol. 18:591–595.   45. Heinrichs, J. H., D. J. Beecher, J. M. MacMillan, and B. A. Zilinskas. 1993. Molecular cloning and characterization of the hblA gene encoding the B component of hemolysin BL from Bacillus cereus. J. Bacteriol. 175:6760–6766.   46. Hoffmaster, A. R., J. Ravel, D. A. Rasko, G. D. Chapman, M. D. Chute, C. K. Marston, B. K. De, C. T. Sacchi, C. Fitzgerald, L. W. Mayer, M. C. Maiden, F. G. Priest, M. Barker, L. Jiang, R. Z. Cer, J. Rilstone, S. N. Peterson, R. S. Weyant, D. R. Galloway, T. D. Read, T. Popovic, and C. M. Fraser. 2004. Identification of anthrax toxin genes in a Bacillus cereus associated with an illness resembling inhalation anthrax. Proc. Natl. Acad. Sci. USA 101:8449–8454.   47. Hoton, F. M., L. Andrup, I. Swiecicka, and J. Mahillon. 2005. The cereulide genetic determinants of emetic Bacillus cereus are plasmid-borne. Microbiology 151: 2121–2124.   48. Hueck, C. J., and W. Hillen. 1995. Catabolite repression in Bacillus subtilis: a global regulatory mechanism for the gram-positive bacteria? Mol. Microbiol. 15:395–401.   49. Hughes, S., B. Bartholomew, J. C. Hardy, and J. M. Kramer. 1988. Potential application of a HEp-2 cell assay in the investigation of Bacillus cereus emeticsyndrome food poisoning. FEMS Microbiol. Lett. 52:7–12.   50. Husmark, U. 1993. Adhesion mechanisms of bacterial spores to solid surfaces. Ph.D. thesis. Department of Food Science, Chalmers University of Technology and SIK, The Swedish Institute for Food Research, Göteborg, Sweden.   51. Jackson, S. G., R. B. Goodbrand, R. Ahmed, and S. Kasatiya. 1995. Bacillus cereus and Bacillus thuringiensis isolated in a gastroenteritis outbreak investigation. Lett. Appl. Microbiol. 21:103–105.   52. Klimowicz, A. K., T. A. Benson, and J. Handelsman. 2010. A quadruple enterotoxin-deficient mutant of Bacillus thuringiensis remains insecticidal. Microbiology 156:3575–3583.   53. Kramer, J. M., and R. J. Gilbert. 1989. Bacillus cereus and other Bacillus species, p. 21–70. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, New York, NY.   54. Lechner, S., R. Mayr, K. P. Francic, B. M. Prub, T. Kaplan, E. Wieber-Gunkel, G. A. S. B. Stewart, and S. Scherer. 1998. Bacillus weihenstephanensis sp. nov. is a new psychrotolerant species of the Bacillus cereus group. Int. J. Syst. Bacteriol. 48:1373–1382.   55. Lereclus, D., H. Agaisse, M. Gominet, S. Salamitou, and V. Sanchis. 1996. Identification of a Bacillus thuringiensis gene that positively regulates transcription of the phosphatidyl­inositol-specific phospholipase C gene at the onset of the stationary phase. J. Bacteriol. 178:2749–2756.   56. Lindbäck, T., A. Fagerlund, M. S. Rødland, and P. E. Granum. 2004. Characterization of the Bacillus cereus Nhe enterotoxin. Microbiology 150:3959–3967.

501   57. Lindbäck, T., S. P. Hardy, R. Dietrich, M. Sodring, A. Didier, M. Moravek, A. Fagerlund, S. Bock, C. Nielsen, M. Casteel, P. E. Granum, and E. Martlbauer. 2010. Cytotoxicity of the Bacillus cereus Nhe enterotoxin requires specific binding order of its three exoprotein components. Infect. Immun. 78:3813–3821.   58. Lund, T., M. L. De Buyser, and P. E. Granum. 2000. A new cytotoxin from Bacillus cereus that may cause necrotic enteritis. Mol. Microbiol. 38:254–261.   59. Lund, T., and P. E. Granum. 1996. Characterisation of a non-haemolytic enterotoxin complex from Bacillus cereus isolated after a foodborne outbreak. FEMS Microbiol. Lett. 141:151–156.   60. Lund, T., and P. E. Granum. 1997. Comparison of biological effect of the two different enterotoxin complexes isolated from three different strains of Bacillus cereus. Microbiology 143:3329–3336.   61. Mahler, H., A. Pasi, J. M. Kramer, P. Schulte, A. C. Scoging, W. Bar, and S. Krahenbuhl. 1997. Fulminant liver failure in association with the emetic toxin of Bacillus cereus. N. Engl. J. Med. 336:1142–1148.   62. Mignot, T., M. Mock, D. Robichon, A. Landier, D. Lereclus, and A. Fouet. 2001. The incompatibility between the PlcR- and AtxA-controlled regulons may have selected a nonsense mutation in Bacillus anthracis. Mol. Microbiol. 42:1189–1198.   63. Miwa, Y., A. Nakata, A. Ogiwara, M. Yamamoto, and Y. Fujita. 2000. Evaluation and characterization of ­catabolite-responsive elements (cre) of Bacillus subtilis. Nucleic Acids Res. 28:1206–1210.   64. Moravek, M., R. Dietrich, C. Buerk, V. Broussolle, M. H. Guinebretiere, P. E. Granum, C. Nguyen-The, and E. Martlbauer. 2006. Determination of the toxic potential of Bacillus cereus isolates by quantitative enterotoxin analyses. FEMS Microbiol. Lett. 257:293–298.   65. Mortimer, P. R., and G. McCann. 1974. Food poisoning episodes associated with Bacillus cereus in fried rice. Lancet i:1043–1045.   66. Nakamura, L. K. 1998. Bacillus pseudomycoides sp. nov. Int. J. Syst. Bacteriol. 48:1031–1035.   67. Ouhib, O., T. Clavel, and P. Schmitt. 2006. The production of Bacillus cereus enterotoxins is influenced by carbohydrate and growth rate. Curr. Microbiol. 53:222–226.   68. Ray, D. E. 1991. Pesticides derived from plants and other organisms, p. 585–636. In W. J. Hayes and E. R. Laws, Jr. (ed.), Handbook of Pesticide Toxicology. Academic Press, Inc., New York, NY.   69. Rivera, A. M. G., P. E. Granum, and F. G Priest. 2000. Common occurrence of enterotoxin genes and enterotoxicity in Bacillus thuringiensis. FEMS Microbiol. Lett. 190:151–155.   70. Rosenquist, H., L. Smidt, S. R. Andersen, G. B. Jensen, and A. Wilcks. 2005. Occurrence and significance of Bacillus cereus and Bacillus thuringiensis in ready-to-eat food. FEMS Microbiol Lett. 250:129–136.   71. Ryan, P. A., J. M. Macmillan, and B. A. Zilinskas. 1997. Molecular cloning and characterization of the genes encoding the L1 and L2 components of hemolysin BL from Bacillus cereus. J. Bacteriol. 179:2551–2556.

502   72. Schmidt, K. (ed). 2001. WHO Surveillance Programme for Control of Foodborne Infections and Intoxications in Europe. Seventh Report. FAO/WHO Collaborating Centre for Research and Training in Food Hygiene and Zoonoses, Berlin, Germany.   73. Schoeni, J. L., and A. C. L Wong. 1999. Heterogeneity observed in the components of hemolysin BL, an enterotoxin produced by Bacillus cereus. Int. J. Food Microbiol. 53:159–167.   74. Shinagawa, K. 1993. Serology and characterization of Bacillus cereus in relation to toxin production. Bull. Int. Dairy Fed. 287:42–49.   75. Shinagawa, K., H. Konuma, H. Sekita, and S. Sugii. 1995. Emesis of rhesus monkeys induced by intragastric administration with the HEp-2 vacuolation factor (cereulide) produced by Bacillus cereus. FEMS Microbiol. Lett. 130:87–90.   76. Shinagawa, K., S. Otake, N. Matsusaka, and S. Sugii. 1992. Production of the vacuolation factor of Bacillus cereus isolated from vomiting-type food poisoning. J. Vet. Med. Sci. 54:443–446.   77. Shiota, M., K. Saitou, H. Mizumoto, M. Matsusaka, N. Agata, M. Nakayama, M. Kage, S. Tatsumi, A. Okamoto, S. Yamaguchi, M. Ohta, and D. Hata. 2010. Rapid detoxification of cereulide in Bacillus cereus food poisoning. Pediatrics 125:e951–e955.   78. Slamti, L., and D. Lereclus. 2002. A cell-cell signaling peptide activates the PlcR virulence regulon in bacteria of the Bacillus cereus group. EMBO J. 21:4550–4559.   79. Slamti, L., and D. Lereclus. 2005. Specificity and polymorphism of the PlcR-PapR quorum-sensing system in the Bacillus cereus group. J. Bacteriol. 187:1182–1187.

Foodborne Pathogenic Bacteria   80. Guinebretiere, M. H., S. Auger, N. Galleron, M. Contzen, B. De Sarrau, M. L. De Buyser, G. Lamberet, A. Fagerlund, P. E. Granum, D. Lereclus, P. De Vos, C. Nguyen-The, and A. Sorokin. 2012. Bacillus cytotoxicus sp. nov. is a new thermotolerant species of the Bacillus cereus group occasionally associated with food poisoning. Int. J. Syst. Evol. Microbiol. 17 Feb 2012. [Epub ahead of print] PubMed PMID: 22328607.   81. Stenfors Arnesen, L. P., A. Fagerlund, and P. E. Granum. 2008. From soil to gut: Bacillus cereus and its food poisoning toxins. FEMS Microbiol. Rev. 32:579–606.   82. Stenfors Arnesen, L. P., K. O’Sullivan, and P. E. Granum. 2007. Food poisoning potential of Bacillus cereus strains from Norwegian dairies. Int. J. Food Microbiol. 116:292–296.   83. Thorsen, L., B. B. Budde, L. Henrichsen, T. Martinussen, and M. Jakobsen. 2009. Cereulide formation by Bacillus weihenstephanensis and mesophilic emetic Bacillus cereus at temperature abuse depends on pre-incubation conditions. Int. J. Food Microbiol. 134:133–139.   84. Thorsen, L., B. M. Hansen, K. F. Nielsen, N. B. Hendriksen, R. K. Phipps, and B. B. Budde. 2006. Characterization of emetic Bacillus weihenstephanensis, a new cereulide-producing bacterium. Appl. Environ. Microbiol. 72:5118–5121.   85. van Netten, P., A. van de Moosdijk, P. van Hoensel, D. A. A. Mossel, and I. Perales. 1990. Psychrotrophic strains of Bacillus cereus producing enterotoxin. J. Appl. Bacteriol. 69:73–79.   86. Yokoyama, K., M. Ito, N. Agata, M. Isobe, K. Shibayama, T. Horii, and M. Ohta. 1999. Pathological effect of synthetic cereulide, an emetic toxin of Bacillus cereus, is reversible in mice. FEMS Immunol. Med. Microbiol. 24:115–120.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch20

Elliot T. Ryser Robert L. Buchanan

Listeria monocytogenes†

Listeriosis has emerged as a major foodborne disease during the past 30 years, after a 1981 outbreak of listeriosis in Nova Scotia, Canada, was traced to contaminated cole slaw (324). However, the discovery of the causative agent of listeriosis, Listeria monocytogenes, dates back to the mid 1920s, when E. G. D. Murray and James Pirie independently reported detailed descriptions of listeriosis in small animals (302). A fascinating account of the discovery of L. monocytogenes has been compiled by Jim McLauchlin (247). The first documented retrospective case of human listeriosis involved a soldier who suffered from meningitis at the end of World War I. There is a suggestion in the literature that listeriosis may have been the cause of Queen Anne’s 17 unsuccessful pregnancies (320). Between 1930 and 1950, a few human listeriosis cases were reported. Today, the estimated incidence of listeriosis in most developed countries is 2 to 5 cases per 1,000,000 population, with nearly 1,600 cases occurring annually in the United States (321). The emergence of listeriosis is the result of complex interactions between various factors reflecting changes in social patterns. These factors include

•

•

•

•

•

•

This chapter represents a substantive update of chapter 21, “Listeria monocytogenes,” in the 3rd edition of Food Microbiology: Fundamentals and Frontiers. We fully acknowledge the contributions of the prior authors (B. Swaminathan, D. Cabanes, W. Zhang, and P. Cossart) to the current chapter. †

20

improvements during the past 50 years in medicine, public health, sanitation, and nutrition that have resulted in increased life expectancy, particularly in developed countries the greatly increased population of immunocompromised individuals at increased risk of listeriosis and other diseases as a result of the increased use of immunosuppressive medications for the treatment of malignancies, management of organ transplantations, and autoimmune diseases, as well as the ongoing epidemic of AIDS changes toward centralized and more consolidated food production and processing, the ever-expanding national and international distribution of foods, and increased use of refrigeration as a primary means of food preservation changes in food habits including increased consumer demand for preservative-free, fresh-tasting convenience foods that can be purchased ready-to-eat (RTE), refrigerated, and rapidly prepared with minimal or no cooking before consumption; and changes in handling and preparation practices expanded use of antacids and gastric-acid-suppressive medications improved diagnostic methods and enhanced public health surveillance.

Listeriosis is an atypical foodborne illness of major public health concern because of the severity of the ­ disease

Elliot T. Ryser, Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824-1225. Robert L. Buchanan, Center for Food Safety and Security Systems, College of Agriculture and Natural Resources, University of Maryland, College Park, MD 20742.

503

504 (meningitis, septicemia, and spontaneous abortion), the high case-fatality rate (approximately 20 to 30%), the long incubation time (up to 70 days), and a ­predilection for individuals who have an underlying condition that leads to impairment of T-cell-mediated immunity. Control of L. monocytogenes in foods represents a significantly greater challenge than most foodborne pathogens in that it is widely distributed in the environment, is resistant to diverse environmental conditions, including low pH and high NaCl concentrations, and is facultatively anaerobic and psychrotrophic. The various ways L. monocytogenes can enter food processing plants, its ability for prolonged survival in the environment (soil, plants, and water), on foods, and in food processing plants, and its ability to grow at low temperatures (2 to 4°C) and to survive in biofilms or on/in foods and food contact surfaces for prolonged periods under adverse conditions have made this bacterium a major concern of the agrifood industry for more than 25 years. The significance of L. monocytogenes as a foodborne pathogen is complex. The severity and case-fatality rate of the disease require appropriate preventive measures, but the characteristics of the microorganism are such that it is unrealistic to expect all food to be Listeria-free. This dilemma has generated an ongoing debate concerning both the various strategies for prevention of listeriosis and the regulation of L. monocytogenes in foods. Since 1985, considerable research has been conducted to identify the routes of contamination, the behavior of this pathogen in a wide range of foods, and various microbial intervention strategies including the use of Listeria growth inhibitors, thermal/nonthermal processing, and postpackaging pasteurization. Epidemiologic investigations of outbreaks have helped identify the vehicles of transmission and have led to an expanding list of RTE foods that have been associated with outbreaks. Basic research on the genetics, molecular biology, and immunologic response of animals and humans to L. monocytogenes has provided detailed insights into the virulence characteristics of this fascinating pathogen.

CHARACTERISTICS OF THE ORGANISM

Classification—The Genus Listeria

Listeria belongs to the Clostridium subbranch together with Staphylococcus, Streptococcus, Lactobacillus, and Brochothrix. This phylogenetic position of Listeria is consistent with its low G+C% DNA content (36 to 42%) (26, 303). The genus Listeria contains eight species. L. monocytogenes is one of six closely related species forming a group that also includes L. ivanovii, L. innocua, L. seeligeri, L. welshimeri, and L. marthii nom. prov. (159,

Foodborne Pathogenic Bacteria 302). Within this group, two subspecies of L. ivanovii have been described: L. ivanovii subsp. ivanovii and L. ivanovii subsp. londoniensis (30). The two remaining species, L. grayi and L. rocourtiae, are more phylogenetically distant (85, 86, 159). Based on results of DNA-DNA hybridization, multilocus enzyme analysis, and 16S rRNA sequencing, six of the eight best-known species in the genus Listeria follow two different lines of descent: (i) L. monocytogenes and its closely related species, namely L. innocua, L. ivanovii (subsp. ivanovii and subsp. londoniensis), L. welshimeri, and L. seeligeri; and (ii) L. grayi. Within the genus Listeria, only L. monocytogenes and L. ivanovii are considered to be pathogenic, as evidenced by their 50% lethal dose (LD50) in mice and their ability to grow in mouse spleen and liver. L. monocytogenes is the only human pathogen of major public health concern, and L. ivanovii infections are primarily confined to animals. The methods used for isolating L. monocytogenes from food and clinical samples have been reviewed and described elsewhere (40, 96, 130). Identification of Listeria to species level is based on a limited number of biochemical markers, among which hemolysin production has been traditionally used to differentiate between L. monocytogenes and the most frequently encountered nonpathogenic Listeria species, L. innocua (336). Other biochemical tests used to discriminate between species include acid production from d-xylose, l-rhamnose, alpha-methyl-d-mannoside, and d-mannitol (Fig. 20.1). A wide range of miniaturized biochemical test kits are commercially available for species confirmation in addition to more rapid methods that utilize enzyme-linked immunosorbent assays and DNA probes for specific genes (312).

Further Characterization and Subtyping of L. monocytogenes

L. monocytogenes isolates are often characterized below the species level for the purposes of public health surveillance and to assist in outbreak investigations. Serotyping has proven its value over many years. There are 13 sero­ types of L. monocytogenes that can cause disease, but more than 90% of human isolates belong to three serotypes: 1/2a, 1/2b, and 4b (229, 314, 337). Because of the low discriminatory power of this method, phage-typing systems were developed and were a primary means of distinguishing between strains of the same serotype prior to the introduction of molecular typing methods. Since 1989, various molecular typing methods have been applied to L. monocytogenes, including multilocus enzyme electrophoresis, ribotyping, DNA microrestriction (DNA fragments generated with high-frequency cutting enzymes

20.  Listeria monocytogenes

505

Figure 20.1  Phenotypic identification of Listeria species. doi:10.1128/9781555818463.ch20f1

and separated by conventional agarose gel electrophoresis) and macrorestriction (large DNA fragments generated with infrequently cutting enzymes and separated by pulsed-field gel electrophoresis [PFGE]), random amplification of polymorphic DNA (124, 158). Due to their ability to type all strains and the high discriminatory power of some of them, these methods have become standard tools for epidemiologic investigations. In addition, unlike serotyping and phage typing, which require specialized reagents (typing sera and bacteriophages) available to only a few reference laboratories, these molecular methods can be performed in any reasonably equipped laboratory. This includes the routine subtyping and reporting of Listeria isolates by public health and regulatory laboratories in the United States and other countries via the Centers for Disease Control and Prevention’s (CDC) PulseNet. This international network of laboratories use highly standardized subtyping protocols (i.e., PFGE) to detect foodborne disease clusters that may have a common source. L. monocytogenes was added to PulseNet in 1999, and a standardized PFGE protocol is available (157). Serotyping is still useful for first-level discrimination between isolates before application of more-sensitive subtyping methods. This is often needed because subtyping based on virulence markers alone can be complex due to the sharing of virulence determinants across L. monocytogenes lineage groups (see below). Recently, a PCR-based method has been developed and validated for grouping L. monocytogenes isolates on the basis of their serotype (98). The past decade has seen a plethora of new molecular methods proposed for subtyping L. monocytogenes. These include multilocus sequence typing (253), multivirulence-locus sequence typing (66, 392), multiple-locus variable-number tandem-repeat analysis (375, 394), DNA ­sequencing-based subtyping (52), and macroarray analyses (33, 34, 97). Whether any of these approaches offer significant advantages over current molecular subtyping

methods and whether they can be used routinely in public health laboratories for subtyping clinical isolates of L. monocytogenes remains to be determined. A number of these systems are being automated so that they can be used for routine analyses outside research laboratory environments (316). L. monocytogenes serotype 4b strains are responsible for 33 to 50% of sporadic human cases worldwide and almost all of the major foodborne outbreaks in Europe and North America. The remainder of the sporadic cases are largely associated with serotypes 1/2a and 1/2b. However, most food isolates belong to the serotypes 1/2a and 1/2c. Although the reasons for this have not been fully elucidated, some recent observations provide intriguing clues. DNA from several epidemic-associated strains of L. monocytogenes 4b were resistant to restriction by Sau3A and other restriction enzymes known to be sensitive to cytosine methylation at 5¢ GATC 3¢ sites. This modification of Sau3A restriction appears to be host mediated (393). A putative restriction-modification system (85M, 85R, and 85S) has been identified in these strains (390). Also, they contain certain unique genetically unlinked DNA sequences. Epidemic-associated strains exhibiting these unique characteristics have been designated epidemic clone I (ECI). In addition, a novel gene (gtcA) involved in the incorporation of galactose and glucose to the cell wall teichoic acid appears to be present only in serotype 4b and other serotype 4 isolates (292). A multiplex single nucleotide polymorphism (SNP)-based method has been developed for identifying and differentiating epidemic clones of L. monocytogenes (230). Evans et al. (110) examined and classified several outbreak strains of L. monocytogenes 4b as ECI and ECII. The ECI isolates were involved in outbreaks prior to 1998, whereas ECII isolates were involved in multistate outbreaks in the United States in 1998 and 2002. ECII isolates have diverged in the serotype-specific region of

506 the genome compared with other serotype 4b strains. Since that initial work, two additional EC groups have been identified: ECIII, which is associated with specific serotype 1/2a strains, and ECIV, a group of serotype 4b strains with a different set of SNPs from those observed with ECI and ECII (197, 230). Wiedmann et al. (384) classified L. monocytogenes isolates into three distinct lineages based on the combination of their ribotype patterns and allelic analysis of two virulence genes (hlyA and actA). Lineage I was primarily composed of human clinical isolates associated with sporadic and outbreak-associated disease; lineage 2 contained human and animal isolates but none from epidemic outbreaks; and lineage III contained only animal isolates. Lineage I isolates were of serotypes 1/2b and 4b, whereas lineage II isolates were of serotypes 1/2a and 1/2c. The predominance of human clinical isolates in lineage I was confirmed in subsequent studies from the Wiedmann laboratory (160, 186, 271). Zhang et al. (391) used DNA microarray analysis to identify lineage-specific and serotype-specific differences in the genome content of lineage I and II strains and reported that 47 genes in 16 different contiguous segments of the genome that were present in serotype 1/2a strains of lineage II were absent in lineage I strains and an additional 9 genes were altered exclusively in 4b strains. Ward et al. (379) targeted the prfA virulence gene cluster to probe the intraspecific phylogeny of L. monocytogenes and subsequently classified outbreak-associated 4b isolates in their lineages I and III. They suggested that the low frequency of association between lineage III strains and human disease was likely to be due to the rarity of exposure and not reduced virulence of lineage III strains. Liu et al. (229) observed that the serotype 4b strains in lineage III appeared to lack some virulence determinants found in serotype 4b strains clustered in lineage I. While this lineage was originally subdivided into sublineages IIIA and IIIB, recent phylogenetic analysis of these strains has led to the reassignment of lineage IIIB to a new lineage group, lineage IV (84, 380).

Susceptibility to Physical and Chemical Agents

L. monocytogenes can grow at 0 to 45°C, with growth occurring more slowly at lower temperatures. The average generation times for 39 L. monocytogenes strains were 43, 6.6, and 1.1 h at 4, 10, and 37°C, respectively, and the respective lag times were 151, 48, and 7.3 h (18). Temperatures below 0°C preserve or moderately inactivate the bacterium. Survival and injury during frozen storage depend on the substrate and the rate of freezing. Among foodborne pathogens, it is considered one of the most resistant to freezing injury. L. mono-

Foodborne Pathogenic Bacteria cytogenes is inactivated by exposure to temperatures above 50°C. Zheng and Kathariou (393) cloned three genes (ltrA, ltrB, and ltrC) of L. monocytogenes that are essential for low-temperature growth. When a 1.2-kb internal fragment of ltrB was used as a probe in Southern hybridizations of HindIII-digested L. monocytogenes DNA, a 9.5-kb DNA fragment that hybridized with the probe was found to be unique to outbreak-associated ECI serotype 4b strains of L. monocytogenes. The pH range for the growth of L. monocytogenes was thought to be 5.6 to 9.6, although the bacterium can initiate growth in laboratory media at pH values as low as 4.4. Growth at low pH values is influenced by the incubation temperature and the type of acid. At pH values below 4.3, listeriae may survive but do not multiply. Experimentally, the presence of up to 0.1% acetic, citric, and lactic acid in tryptose broth inhibits the growth of L. monocytogenes, with inhibition increasing as the incubation temperature decreases. The antilisterial activity of these acids is related to their degree of dissociation, with citric and lactic acids being less detrimental for the pathogen at an equivalent pH than acetic acid. L. monocytogenes grows optimally at water activity (aw) of ³0.97. For most strains, the minimum aw for growth is 0.93, but some strains may grow at aw values as low as 0.90. Further, the bacterium may survive for long periods at aw values as low as 0.83 (339). An inverse relationship exists between the thermal resistance of L. monocytogenes and the aw of the medium in which it is suspended (357), which must be addressed by manufacturers who rely on low aw and thermal treatment for food preservation. L. monocytogenes is able to grow in the presence of 10 to 12% sodium chloride and can grow to high populations at moderate salt concentrations (6.5%). The b­acterium survives for long periods at high salt concentrations, with survival in such environments significantly increased at lower temperatures. The inoculum level of L. monocytogenes affects the ability of listeriae to grow under adverse environmental conditions (temperature, pH, and aw). For example, at 25°C and aw of 0.997, the minimum pH at which growth can be initiated is 4.45 at a cell concentration of 7.3 CFU/ ml, whereas growth can be observed at a pH as low as 3.9 when the cell concentration exceeds 6 × 106 CFU/ ml (207). Multiple predictive microbiology models based on data obtained from studies using microbiological media and various foods describe the effects and interactions of temperature, pH, aw, sodium chloride content, organic acid concentration, and sodium nitrite on both growth and survival of L. monocytogenes.

20.  Listeria monocytogenes Listeriosis and RTE Foods Certain RTE processed foods that support the growth of L. monocytogenes are high-risk vehicles for transmitting listeriosis to susceptible populations as determined by active surveillance for sporadic listeriosis and epidemiologic investigation of listeriosis outbreaks. These foods are usually preserved by refrigeration and offer an appropriate environment for the multiplication of L. monocytogenes during manufacture, aging, transportation, and storage. The risk is likely to be increased if the RTE foods are likely to be exposed to marginal abuse temperatures (8 to 15°C) or have extended refrigerated shelf lives. Foods in this category include unpasteurized milk and dairy products prepared from unpasteurized milk such as soft unripened and surface-ripened cheeses, frankfurters that were not reheated before consumption, delicatessen meats, smoked fish, and some seafood. In Canada, the regulatory policy directs inspection and compliance action on RTE foods that can support the growth of L. monocytogenes. The highest priority is given to those foods that have caused listeriosis and those that have greater than 10 days of shelf life (112). The need to establish effective, risk-based food safety controls has resulted in L. monocytogenes being the focus of a large number of qualitative and quantitative microbiological risk assessments in various RTE foods (4, 6, 7, 9, 43, 64, 108, 172, 227, 250, 287, 288, 289, 290, 306, 363). The most comprehensive of these assessments was conducted by the U.S. Food and Drug Administration (FDA), in collaboration with the U.S. Department of Agriculture (USDA) Food Safety and Inspection Service and the CDC. The assessment was undertaken to predict the potential relative risk of listeriosi­s from eating certain RTE foods among three age-based groups of people: perinatal (16 weeks after fertilization to 30 days after birth), elderly (60 years of age and older), and intermediate age (general population, less than 60 years of age). This assessment evaluated foods within 23 categories considered to be principal p­otential sources of Listeria (Table 20.1) and found that RTE meats, including deli meats and frankfurters, posed the greatest risk. Subsequent risk assessments have examined in detail a variety of RTE foods and various sectors of the food chain. From the exposure models and “what-if” scenarios used in the risk assessment, it was determined that the following five factors affected consumer exposure to L. monocytogenes at the time of food consumption: (i) amount and frequency of the food consumed, (ii) prevalence and levels of L. monocytogenes in the food, (iii) likelihood of L. monocytogenes growth in the

507 food during refrigerated storage, (iv) refrigerated storage temperature, and (v) duration of refrigerated storage of the food before consumption. The risk assessment model was used to estimate the likely impact of control strategies by changing one or two input parameters and measuring the change in the model outputs. For example, one “what-if” scenario determined that the predicted number of listeriosis cases would be reduced by 69% if all home refrigerators were consistently operating at or below 7.2°C. Another scenario determined that reducing the maximum storage time of deli meats from 28 to 14 days would reduce the median number of cases in the elderly population by 13.6%.

Fluid Milk Products

Raw milk is a well-documented source of L. monocytogenes. The first major outbreak of foodborne listeriosis was traced to California-produced Mexican-style cheese in 1985, and a 2000 outbreak in North Carolina was likely caused by the use or partial use of unpasteurized milk for cheesemaking (228, 237). The prevalence of L. monocytogenes in bulk tank raw milk varies from 1 to 13%, while its prevalence in milk processing plants ranges from 7 to 28% (275). Therefore, raw milk and food products made from raw milk could be potential sources of L. monocytogenes. Based on numerous surveys, about 2.5, 3.6, and 5.2% of the raw milk samples tested in North America, Europe, and elsewhere, respectively, were positive for L. monocytogenes (309). However, the ability of L. monocytogenes to survive and proliferate in raw dairy products stored at refrigeration temperatures makes this bacterium a particular concern for the dairy industry. Contamination levels that are initially low (e.g., <1 CFU/25 g) may increase to high cell numbers that could pose a human health hazard if milk and certain cheeses are subjected to long-term refrigerated storage (309). When suspended in milk, most thermal inactivation studies have shown that cells of L. monocytogenes were effectively inactivated by high temperature-short time (HTST) pasteurization (71°C for 15 s or equivalent) (100, 309). One multistate survey showed a very low frequency (1 of 5,519 samples) of isolation for L. monocytogenes in commercial pasteurized fluid milk products sold in the United States (122), with the same being true for dairy products prepared from pasteurized milk (310). When thermal inactivation of freely suspended L. monocytogenes cells was compared with that of cells that were internalized within phagocytic leukocytes, differences were observed (99). The physiological state (actively growing cells versus cells in stationary phase) and growth of Listeria at elevated temperatures (e.g., in infected cows that may have developed fever) before pasteurization are some of the potential factors that

Foodborne Pathogenic Bacteria

508

Table 20.1  Relative risk ranking and predicted number of cases of listeriosis for the total U.S. population on a per serving and

per annum basisa

Predicted median no. of cases of listeriosis for 23 food categories Relative risk ranking 1 2 3

Per serving basisb Food High risk

Per annum basisc Cases

Food

Deli meats Frankfurters, not reheated Pâté and meat spreads

7.7 × 10 6.5 × 10−8 3.2 × 10−8 7.1 × 10−9 6.2 × 10−9 5.1 × 10−9

8

Unpasteurized fluid milk Smoked seafood Cooked ready-to-eat crustaceans High fat and other dairy products Soft unripened cheese

9 10 11 12 13

Pasteurized fluid milk Fresh soft cheese Frankfurters, reheated Preserved fish Raw seafood

1.0 × 10−9 1.7 × 10−10 6.3 × 10−11 2.3 × 10−11 2.0 × 10−11

Fruits Dry/semidry fermented sausages Semisoft cheese Soft ripened cheese Vegetables Deli-type salads Ice cream and other frozen dairy products Processed cheese Cultured milk products Hard cheese

1.9 × 10−11 1.7 × 10−11

4 5 6 7

14 15 16 17 18 19 20 21 22 23

Moderate risk

Low risk

−8

Very high risk High risk

Moderate risk

Deli meats Pasteurized fluid milk High fat and other dairy products Frankfurters, not reheated Soft unripened cheese Pâté and meat spreads

Cases 1,598.7 90.8 56.4 30.5 7.7 3.8

2.7 × 10−9

Unpasteurized fluid milk

3.1

1.8 × 10−9

Cooked ready-to-eat crustaceans Smoked seafood Fruits Frankfurters, reheated Vegetables Dry/semidry fermented sausages Fresh soft cheese Semisoft cheese

2.8

6.5 × 10−12 5.1 × 10−12 2.8 × 10−12 5.6 × 10−13 4.9 × 10−14 4.2 × 10−14 3.2 × 10−14 4.5 × 10−15

Low risk

Soft ripened cheese Deli-type salads Raw seafood Preserved fish Ice cream and other frozen dairy products Processed cheese Cultured milk products Hard cheese

1.3 0.9 0.4 0.2 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1

Table adapted from reference 6. Food categories were classified as high risk (>5 cases per billion servings), moderate risk (£5 but ³1 case per billion servings), and low risk (<1 case per billion servings). c Food categories were classified as very high risk (>100 cases per annum), high risk (>10 to 100 cases per annum), moderate risk (>1 to 10 cases per annum), and low risk (<1 case per annum). a

b

may have caused these cells to become more heat resistant. Lado and Yousef (209) state that the following factors should be considered in evaluating the effect of pasteurization on L. monocytogenes: (i) the safety margin of pasteurization for inactivation of the bacterium may be lower than previously thought; (ii) the level of contamination in pooled milk is lower than that used in most inoculation studies; (iii) homogenization of milk destroys the integrity of phagocytic cells in milk, thus removing any protection offered to bacterial cells; and (iv) thermoduric spoilage microorganisms surviving HTST are likely to outcompete L. monocytogenes. Hence, there is general agreement with

the observations of the WHO informal study group, which concluded that “pasteurization is a safe process which reduces the number of L. monocytogenes in raw milk to levels that do not pose an appreciable risk to human health” (5). Recognizing the small margin of safety offered by HTST pasteurization, most raw milk processors have adopted processes that employ temperatures well above the minimum legal requirements for pasteurized milk. L. monocytogenes grows in pasteurized milk, with the numbers i­ncreasing 10-fold in 7 days at 4°C. Further, listeriae grow more rapidly in pasteurized milk than in raw milk when incubated at 7°C. Therefore, fluid milk that is contaminated

20.  Listeria monocytogenes after pasteurization and stored under refrigeration may attain very high populations of L. monocytogenes after 1 week. Temperature abuse may further enhance growth rates, as was evidenced in pasteurized-milk-associate­d outbreaks in Illinois (78) and Massachusetts (75). Additional details may be obtained from a review on the prevalence, growth, and survival of L. monocytogenes in fluid milk and unfermented dairy products (122, 309).

Cheeses

L. monocytogenes can survive the cheese manufacturing and ripening process because of its relative hardiness to temperature and pH fluctuations, ability to multiply at refrigeration temperature, and salt tolerance. Its growth in cheese milk is retarded but not completely inhibited by lactic starter cultures. During cheesemaking, L. monocytogenes is concentrated about 10-fold in the curd, with the remaining cells appearing in the whey. The behavior of listeriae in the curd is influenced by the type of cheese, ranging from growth in feta cheese to significant inactivation during cottage cheese manufacture. During cheese ripening, the L. monocytogenes population may increase (Camembert), decrease gradually (Cheddar or Colby), or decrease rapidly during early ripening and then stabilize (blue cheese). Consumption of soft Hispanic-style (e.g., queso fresco or queso blanco) and surface-ripened cheeses (e.g., Brie or Camembert) by susceptible persons is an important risk factor for sporadic and epidemic listeriosis in North America and Europe. Over 2 decades ago, 49% of the sporadic listeriosis cases in France were attributed to consumption of soft cheese. Prevention efforts have reduced this incidence by 68% between 1987 and 1997 (82, 151). The first documented foodborne listeriosis outbreak in Japan resulted in 86 cases of infection (mainly gastro­enteritis) and was traced to L. monocytogenes serotype 1/2a contamination in washedtype cheese (238). In the United States, the Centers for Disease Control and Prevention recommends that persons at high risk for listeriosis avoid consumption of cheeses made from unpasteurized milk (59).

Meat and Poultry Products

The growth potential for L. monocytogenes in meat and poultry products depends on the meat variety, pH, and type and number of competing microorganisms. A recent survey in the United States revealed that 3.5% of ­commercial ground beef samples were positive for L. monocytogenes (315). Poultry supports the growth of L. monocytogenes better than other meats, whereas properly fermented sausage does not permit Listeria growth following fermentation. Contamination typically occurs via symptomatic or asymptomatic carriage of L. monocytogenes by the animal

509 before slaughter, followed by contamination of the carcass during or after slaughter. Because L. monocytogenes tends to concentrate and multiply in the kidney, mesenteric and mammary lymph nodes, liver, and spleen of infected animals, organ meats may be more hazardous to consume than muscle tissue (114). Regardless of the contamination route, L. monocytogenes strongly attaches to the surface of raw meats and is difficult to remove or inactivate. L. monocytogenes multiplies readily in meat products, including vacuum-packaged beef, at pH values near 6.0, whereas little or no multiplication occurs at pH 5.0 (114, 141). RTE meat products that have received a heat treatment followed by cooling in brine before packaging may provide a particularly conducive environment for multiplication of L. monocytogenes because of the reduction in competitive flora and the high salt tolerance of the bacterium. Based on the USDA monitoring program for L. monocytogenes, the prevalence of this pathogen in cooked RTE meat and poultry products decreased from 4.61% in 1990 to 0.32% in 2010 (365), with much of this decrease due to the adoption of mandatory hazard analysis and critical control point programs for the industry beginning in 1998. Since 2004, producers of such products in the United States must also use one of the three USDA alternatives for control of Listeria, namely postpackaging pasteurization with or without the addition of Listeria growth inhibitors (e.g., lactate or diacetate) to the product formulation or enhanced environmental sampling for Listeria (364). Surveys of both meat and poultry processing facilities have identified both transient and persistent strains of L. monocytogenes, with this pathogen in some cases persisting at the same location for months or years (234, 235, 338, 387). Cooked, RTE meat and poultry products have been frequent sources of sporadic and outbreak-associate­d listeriosis in both North America and Europe. Consumption of unreheated frankfurters and undercooked chicken was first identified as a risk factor for sporadic cases of listeriosis in the United Sates (335) after a contaminated turkey frankfurter product was linked to a sporadic L. monocytogenes infection in a cancer patient in 1989 (57). Nationwide outbreaks of listeriosis were traced to contaminated frankfurters or deli meat in the United States in 1998, 2000, and 2002 (147, 252, 276) and in Canada in 2008 (8). Cross-contamination of RTE meat and poultry products with L. monocytogenes can occur during both production and retail sale (225). In a large U.S. survey of RTE foods obtained from different retail markets, in-storepackaged deli meats were about 7 times more likely to harbor L. monocytogenes (2.7% positive) than manufacture-package­d products (0.4% positive) (245). Using more

510 recent retail contamination rates (289) and consumer purchasing habits (56), approximately 80% of all listeriosis cases and deaths involving deli meat can be attributed to products sliced at retail with mechanical delicatessen slicers now recognized as the major source of Listeria crosscontamination (199, 200, 226, 289, 376). One survey in New York State showed that many cases of human listeriosis were in fact associated with these persistent L. monocytogenes strains in retail environments (290).

Seafoods

The role of seafoods in human listeriosis was reviewed in 2000 (304). The same year, Huss et al. (177) classified the following seafoods as potential high-risk foods for listeriosis: (i) mollusks, including fresh and frozen mussels, clams, and oysters in shell or shucked; (ii) raw fish; (iii) lightly preserved fish products including salted, marinated, fermented, cold-smoked, and gravad fish; and (iv) mildly heat-processed fish products and crustaceans, with smoked seafood and cooked RTE crustaceans identified as moderate- to high-risk foods in the FDA 2003 risk assessment (6). Shrimp, smoked mussels, and imitation crabmeat have been implicated as sources of human listeriosis. Gravad rainbow trout and coldsmoked rainbow trout were responsible for two outbreaks in Sweden and Finland, respectively (109, 259). In the United States, L. monocytogenes has been isolated from both domestic and imported fresh, frozen, and processed seafood products, including crustaceans, molluscan shellfish, and finfish (190, 258). An FDA survey of domestic and imported refrigerated or frozen cooked crab meat in 1987-1988 determined contamination levels of 4.1% for domestic products and 8.3% for imported products (107). Crab and smoked fish samples analyzed by the FDA between 1991 and 1996 yielded L. monocytogenes contamination rates of 7.5% and 13.6%, respectively. From 1999 to 2003, 2.9% and 14.9% of the hot- and cold-smoked, respectively, seafood samples (primarily fish) tested positive for L. monocytogenes (190) with 3.6% of RTE crustaceans (primarily crab and shrimp) also positive during the same period. However, consumption of smoked seafood in the United States is still much less than that of meats and cheeses, and the much smaller scale of production of these products helps to explain the general lack of large-scale outbreaks in the United States. Cold-smoked fish continues to be the focus of efforts to develop new control measures and systems for L. monocytogenes (94). As was true for meat processing plants, L. monocytogenes is also frequently found in facilities producing raw and RTE seafood. When Hoffman et al. (175) compared the ribotypes of L. monocytogenes strains isolated

Foodborne Pathogenic Bacteria from smoked fish processing environments and raw fish, various unique ribotypes were recovered from raw fish samples, indicating that the raw fish was possibly contaminated from a source other than the processing environment. These findings are also supported by results from two other studies (175, 176).

EFFECTS OF NEWER METHODS OF FOOD PRESERVATION New trends in food preservation have recently been developed, including the use of bacteriocins as biopreservatives, as well as vacuum and modified atmosphere packaging, all of which have been reviewed by Lado and Yousef (209). Nisin, produced by certain strains of Lactococcus lactis subsp. lactis, is by far the most widely recognized bacteriocin. The antagonistic effect of nisin on L. monocytogenes, which is well documented, is strongly dependent on the pH and chemical composition of the food to which it is added. Pediocins (from Pediococcus pentosaceus and Pediococcus acidilactici) and bavaracin A (211) from Lactobacillus bavaricus transiently affect L. monocytogenes growth and survival in various beef systems, especially at low temperature. Meatborne lactic acid bacteria can effectively inhibit the growth of L. monocytogenes by rapid production of lactic acid and bacteriocins and can be used as potential biopreservatives in cooked meat products (373). A similar effect was observed with a bacteriocin carnocin from Carnobacterium piscicola in broth and skimmed milk (246). However, a drawback to the use of bacteriocins is the emergence of resistant mutants, as has been observed with bavaracin A and nisin (80, 211). According to Modi et al. (263) one nisin-resistant strain of L. monocytogenes that grew in the presence of nisin was more heat sensitive than the wild-type strain. In vitro studies have also demonstrated antibacterial activity of essential oils against L. monocytogenes, which functions by partitioning the lipids of the cell membrane and mitochondria, resulting in increased membrane permeability and cell leakage. Examples of essential oils include carvacrol, thymol, eugenol, perillaldehyde, cinnamaldehyde, and cinnamic acid, which have MICs of 0.05 to 5 ml/ml in vitro (48). A survey of vacuum-packaged processed meat in retail stores in the early 1990s revealed that 53% of the samples tested were contaminated with L. monocytogenes and that 4% contained more than 1,000 CFU/g (155). This observation corroborates experimental evidence that the growth of L. monocytogenes (a facultative anaerobe) is not significantly affected by vacuum packaging. There has been considerable interest in modified atmosphere packaging

20.  Listeria monocytogenes of meat products (low oxygen and high carbon dioxide concentrations) over recent years because of the increasing demand for refrigerated convenience foods with extended shelf lives. Studies with meat juice, raw chicken, and precooked chicken nuggets revealed that such atmospheres do not significantly affect the growth of L. monocytogenes. L. monocytogenes is sensitive to low doses of irradiation. Sublethal heating, freezing, sanitizer exposure, or exposure to various nonthermal processing methods including high-intensity pulsed light, high pressure, irradiation, and pulsed electric fields can injure a substantial portion of surviving cells (383). Heat-stressed L. monocytogenes may be considerably less pathogenic than nonstressed cells. While injured cells can repair the damage induced by stress and grow on nonselective agar, the additional stress encountered when grown on selective agar leads to variable recovery. Hence, the presence of sublethally damaged cells in foods may lead to differences in cell recovery rates.

RESERVOIRS The role of improperly fermented silage in the transmission of animal listeriosis, first referred to as “circling

511 disease” in 1929, to ruminants is now well known. This pathogen is geographically widespread in both rural and urban environments (120, 319) and can survive and/or grow in improperly fermented silage, pasture grass, decaying vegetation, soil, sewage, and various aquatic environments, including surface water of canals, lakes, ditches, and freshwater tributaries (67, 89, 164, 178, 381). Alfalfa plants and other crops grown on soil treated with sewage sludge can be contaminated with Listeria (2). In one study, one-half of the radish samples grown in L. monocytogenes-inoculated soil were confirmed positive 3 months later (369). The widespread presence of L. monocytogenes in soil is likely due to contamination by decaying plant and fecal material with the soil providing a cool, moist environment and the decaying material providing the nutrients (317). L. monocytogenes has been isolated from the feces of many healthy animals and birds; listeriosis in many animal species has been recorded. Humans exhibiting symptoms of listeriosis and asymptomatic carriers shed the organism in their feces. Figure 20.2 illustrates the many ways in which L. monocytogenes can be spread from the environment to animals and humans and back to the environment.

Figure 20.2  Potential routes of transmission of L. monocytogenes. Adapted from reference 93. Circles or ovals indicate areas of greatest risk of L. monocytogenes multiplication. Boxes indicate where direct consumption of minimally processed products (e.g., whole fresh vegetables, cooked carcass cuts of meat and fish, and effectively pasteurized milk) presents a low risk. Double arrows indicate consumer at risk. doi:10.1128/9781555818463.ch20f2

512

Food Processing Plants Entry of L. monocytogenes into food processing plants occurs through soil on workers’ shoes and clothing, transport equipment, animals that excrete the bacterium or have contaminated hides or surfaces, raw plant tissue, raw food of animal origin, and possibly healthy human carriers. Growth of listeriae is favored by high humidity and the presence of nutrients. L. monocytogenes is most often recovered from moist areas such as floor drains, condensate, stagnant water, floors, and residues on processing equipment (74). L. monocytogenes can attach to various types of surfaces including stainless steel, glass, plastic, and rubber and form biofilms as has been described in meat and dairy processing environments (189). Listeria spp. can survive on fingers after hand washing and in aerosols. The presence of L. monocytogenes in the food chain is evidenced by the widespread distribution of the listeriae in processed products. Contaminated effluents from food processing plants increase the spread of L. monocytogenes in the environment. Sources of L. monocytogenes in dairy processing plants include the environment (floors and floor drains, especially in areas in and around coolers or places subject to outside contamination) and raw milk. Efforts to ensure that milk is safe from L. monocytogenes contamination should focus on promoting appropriate methods of pasteurization and on identifying and eliminating sources of postpasteurization contamination (378). Strict adherence to the guidance in the Pasteurized Milk Ordinance helps minimize Listeria contamination in dairy processing environments and better ensure the safety and quality of the final product. The presence of L. monocytogenes on carcasses is usually attributed to contamination by fecal matter during slaughter, evisceration, and fabrication. A high percentage (11 to 52%) of healthy animals are fecal carriers. Up to 45% of pigs can harbor L. monocytogenes in their tonsils, and 24% of cattle have contaminated internal retropharyngeal nodes (47, 347). L. monocytogenes has been recovered from both unclean and clean zones (especially on workers’ hands) in slaughterhouses, with the most heavily contaminated working areas being cow dehiding and pig stunning/hoisting. Surveys conducted at both turkey and broiler slaughterhouses failed to recover L. monocytogenes from feathers, scald tank water overflow, neck skin, livers, hearts, ceca, or large intestines. In contrast, L. monocytogenes was isolated from feather plucker drip water, chill water overflow, recycling water for cleaning gutters, and mechanically deboned meat. These findings demonstrate the importance of the defeathering machine, chillers, and recycled water in product cross-contamination (137).

Foodborne Pathogenic Bacteria Contamination of processed RTE foods with L. monocytogenes primarily occurs after processing. To date, there is no evidence to indicate that L. monocytogenes can survive validated heat processing protocols used to render foods safe. Having the propensity to adhere to food contact surfaces and form biofilms, L. monocytogenes is particularly difficult to eliminate from the food processing environment even when well-designed sanitation programs are used (206). For example, L. monocytogenes can persistently survive on stainless steel food contact surfaces such as those found on dicing machines and repeatedly contaminate RTE meats (226, 234). Because L. monocytogenes is a frequent contaminant of raw foods and ingredients, ample opportunity exists for reintroduction of listeriae into food processing facilities (99). Extensive information on the problem of L. monocytogenes in various food processing environments and approaches to control has been provided by Kornacki and Gurtler (206). Tompkin (361) outlined a six-step Listeria control program for food processing environments that includes (i) prevention of the establishment and growth of Listeria species in niches or other sites that can lead to contamination of RTE foods; (ii) implementation of a sampling program to assess how well the control program is working; (iii) rapid and effective response when the sampling program yields positive results for Listeria species; (iv) verification by follow-up sampling to ensure that the source of contamination has been identified and corrected; (v) short-term assessment of the last 4 to 8 samplings to facilitate early detection of problems and trends; and (vi) long-term assessment at appropriate intervals (quarterly, annually, etc.) to identify widely scattered contamination events and to measure overall progress towards continuous improvement. Longitudinal studies based on various molecular subtyping analyses assist in identifying contamination patterns of L. monocytogenes in food processing plants and critical validation of intervention strategies (175, 176, 210, 234, 271, 387). The design of the environmental sampling program and the response to positive findings by the sampling program determine the overall effectiveness of the Listeria control program in food processing environments (361).

Surveys of Foods for Prevalence of L. monocytogenes and Regulatory Status in Different Countries

Numerous surveys that have been reviewed and summarized elsewhere confirm that L. monocytogenes contamination of fermented (310) and unfermented (309) dairy products, meat (114), poultry and eggs (311), fish and seafood (190), and fresh produce (36) is widespread in many parts of the world. One large-scale qualitative/quantitative

20.  Listeria monocytogenes survey of retail RTE foods was conducted by investigators at the National Food Processors Association in the United States (145). Product categories examined included luncheon meats, deli salads, Mexican-style cheeses, packaged salads, blue-veined and soft-ripened cheeses, smoked seafood, and seafood salads. Of 31,705 samples examined, 1.82% (the range by sample category varied from 0.17% to 4.7%) were positive for L. monocytogenes. Pathogen levels in positive samples ranged from <0.3 CFU/g to 1.5 × 105 CFU/g. A dose-response model was developed by combining the food survey data with concurrent data on illness in the population that consumed the foods surveyed. Based on this model, control strategies focused on foods that yielded higher cell numbers of L. monocytogenes are likely to have greater public health impact than the current zero-tolerance policy (7, 43, 65). In a survey conducted in Denmark during 1994– 1995, L. monocytogenes was isolated from 14.2% and 30.9% of raw fish and raw meats, respectively (190). Preserved, non-heat-treated fish and meat products were more frequently contaminated (10.8% and 23.5%, respectively) than heat-treated meat products (5%). L. monocytogenes was present at levels exceeding 100 CFU/g in 1.3% of the preserved (not heattreated and heat-treated) fish and meat products that were packed under vacuum or modified atmosphere for extended shelf life. In contrast, another survey covering the same product types that were not packed under vacuum or modified atmospheres revealed significantly lower contamination levels (0.3 to 0.6%) (270). In a Japanese survey of retail foods, L. monocytogenes was isolated from 12, 20, 37, and 25% of minced beef, pork, chicken, and pork-beef mixture, respectively, with only 5 chicken samples having populations greater than 100 CFU/g. L. monocytogenes was isolated from 5.4% of smoked salmon samples and 3.3% of RTE uncooked seafoods (180). A survey conducted in Barcelona, Spain, revealed L. monocytogenes contamination in 9.3% of RTE foods and 2.9% of foods intended to be cooked before consumption (87). In another investigation, L. monocytogenes was isolated from 22% of retail smoked fish samples (95). L. monocytogenes has been isolated from many vegetables, including bean sprouts, cabbage, cucumber, leafy vegetables, potatoes, prepackaged salads, radishes, salad vegetables, and tomatoes in North America, Europe, and Asia (12), with coleslaw, chopped celery, and most recently, cantaloupe having been linked to major outbreaks of listeriosis. Because of the frequent occurrence of L. monocytogenes in foods and the persistence of this pathogen in food processing environments, food regulatory agen-

513 cies in many countries have established tolerance levels other than “zero” (i.e., absence in the sample tested) for L. monocytogenes. Following a nationwide outbreak of listeriosis in Canada in 2008 that was traced to RTE deli meats, the Canadian government revised their policy on Listeria monocytogenes in RTE foods to include new end product compliance criteria in addition to environmental monitoring and the use of postlethality treatments and/or Listeria growth inhibitors for RTE meat products. Foods are divided into two risk categories, with category 1 foods (those that support L. monocytogenes growth to ³100 CFU/g) regulated more stringently than foods in categories 2a (those that support L. monocytogenes growth to £100 CFU/g throughout shelf life) and 2b (those that do not support L. monocytogenes during the expected product shelf life) (169). The French position directs that foods (25-g samples) should be L. monocytogenes-free, but when this is not possible, one must try to obtain the lowest level possible. In France, foods are divided into three groups as follows: (i) food for populations “at risk,” (ii) foods heated in their wrapping or aseptically conditioned after treatment, and (iii) raw foods or foods susceptible to recontamination after treatment. In contrast, the United Kingdom and the United States, while acknowledging the widespread distribution of L. monocytogenes in the food supply and the difficulties in producing L. monocytogenes-free foods, have decided not to adopt tolerance levels for L. monocytogenes in RTE foods. Both countries argue that any “acceptable” levels for L. monocytogenes would require knowledge of the number of listeriae unlikely to cause human infection. While progress is continually being made to establish such tolerance levels through various risk assessments, these two countries have thus far maintained their “zero-tolerance” policy for L. monocytogenes in RTE foods (340).

Human Carriage

Asymptomatic fecal carriage of L. monocytogenes has been studied in various human populations, including healthy people, pregnant women, outpatients with gastroenteritis, slaughterhouse workers, laboratory workers handling Listeria, food handlers, and patients undergoing renal transplantation or hemo­ dialysis (265). L. monocytogenes can be found in 2 to 6% of fecal samples from healthy people, whereas listeriosis patients often excrete higher numbers. In one study, stool specimens from 21% of patients contained ³104 L. monocytogenes CFU/g, and 18% of household contacts of patients with listeriosis fecally shed the same serotype and multilocus enzyme type of L. monocytogenes as the index case (187, 331). Among

514 ­ ousehold contacts of 18 pregnant women with listerih osis, 8.3% asymptomatically shed L. monocytogenes, whereas no listeriae were isolated from 30 household contacts of age-, sex-, and hospital-matched controls (245). Results from a 1985 outbreak investigation in California showed that community-acquired outbreaks might be amplified through secondary transmission by fecal carriers. The very low prevalence of L. monocytogenes in human stools and the short duration of fecal shedding argue against routine stool screening of persons with possible work-related exposure to the pathogen (e.g., dairy workers) as a tool for prevention of listeriosis (162, 318). L. monocytogenes isolated from asymptomatic carriers may carry the full complement of virulence genes or may be attenuated by truncated virulence-associated genes such as inlA and actA (184, 273, 274). L. monocytogenes has not been isolated from oropharyngeal samples of healthy people, and the presence of listeriae in cervicovaginal specimens may be associated with pregnancy-related listeriosis. The role of healthy carriers in the epidemiology of listeriosis is unclear.

FOODBORNE OUTBREAKS Foodborne transmission of listeriosis, although suggested in the early medical literature, was not definitively documented until 1981, when a case-control study was simultaneously used with strain typing investigation in an outbreak in Canada (324). Since 1981, epidemiologic investigations have repeatedly revealed that consumption of contaminated food is the primary mode for transmission of listeriosis. Since 1981, more than 10 major (>30 cases) outbreaks of listeriosis have been traced to various foods including soft cheeses, delicatessen meats, and most recently, cantaloupe. An outbreak of listeriosis in Massachusetts in 1979 may have been caused by raw produce, but the food source was not positively identified. Twenty patients with L. monocytogenes serotype 4b infection were hospitalized during a 2-month period; only nine cases had been detected in the previous 26 months. Ten of the patients were immunosuppressed adults and five died. Fifteen patients are thought to have acquired the infection in the hospital. Consumption of tuna fish, chicken salad, or cheese was associated with illness, but no specific brand was implicated. It was postulated that the raw celery and lettuce, served as a garnish with the three foods, may have been contaminated with L. monocytogenes (173). Although the source of infection was not definitively identified in this outbreak, consumption of cimetidine or antacids was implicated as a risk factor

Foodborne Pathogenic Bacteria for listeriosis, a factor that has been identified in several case-control studies. Decreased gastric acidity might have increased the survival of L. monocytogenes cells as they passed through the stomach. The first confirmed foodborne outbreak of listeriosis occurred in 1981 in Nova Scotia, Canada. Thirty-four pregnancy-associated cases and seven cases in nonpregnant adults were reported during a 6-month period. A case-control study implicated locally prepared coleslaw as the vehicle, and the epidemic strain was subsequently isolated from an unopened package of the product. Cabbage fertilized with manure from sheep suspected of having Listeria meningitis was the probable source. Harvested cabbage was stored over the winter and spring in an unheated shed, which provided a definite growth advantage for this psychrotrophic pathogen (324). Pasteurized milk was identified as the most likely source of infection in another large outbreak of listeriosis, in Boston, MA, in 1983 (116). Forty-nine cases occurred during a 2-month period, 42 in immunosuppressed adults and 7 in pregnant women; the overall case-fatality rate was 29% (95); a case-control study implicated pasteurized milk with 2% fat as the vehicle. Multiple serotypes of L. monocytogenes were isolated from raw milk at the implicated dairy, but none belonged to the outbreak strain. No deviations from approved pasteurization processes were noted at the dairy, suggesting that the contamination occurred after pasteurization. In 1985, the first outbreak of listeriosis to attract widespread attention occurred in Los Angeles County, CA, when contaminated soft Mexican-style cheese was linked to 142 cases over an 8-month period (228). Pregnant women accounted for 93 cases. The remaining 49 cases were nonpregnant adults; 48 of 49 nonpregnant adults had a predisposing condition for listeriosis. Among pregnancy-associated cases, 87% occurred in Hispanic women. The case-fatality rates were 32% for perinatal cases and 32% for nonpregnant adults. Inadequate pasteurization and mixing of raw milk with pasteurized milk likely resulted in the contaminated cheese. L. monocytogenes-contaminated Vacherin Mont d’ Or cheese (a soft surface-ripened variety) was responsible for a 4-year-long (1983 to 1987) outbreak in Switzerland involving 122 cases (J. Bille, presented at the Foodborne Listeriosis Symposium, Wiesbaden, Germany, 7 September 1988). Contaminated pâté was epidemiologically linked to 300 cases of listeriosis in the United Kingdom during 1989–1990 (248). Contaminated pork tongue in aspic marketed in France was the principal v­ehicle for 279 cases of listeriosis ­ reported during a

20.  Listeria monocytogenes 10-month period in 1992 (182). Potted pork (“rillettes”) was associated with 39 cases in 1993, and soft cheese was the vehicle for 33 cases in 1995 (152). Recalling the implicated food, advising the general population through the mass media to avoid consuming the implicated product, and taking appropriate action to prevent further L. monocytogenes contamination during manufacture and handling terminated these large outbreaks. A large multistate outbreak of listeriosis occurred in the United States between August 1998 and March 1999. A total of 101 outbreak-associated cases (including 15 perinatal cases) were identified in 22 states. Fifteen adult deaths and six miscarriages or stillbirths were associated with this outbreak. A case-control study implicated turkey franks from a major manufacturer produced at one facility. L. monocytogenes serotype 4b of the epidemic PFGE subtype was isolated from opened and unopened packages of frankfurters from the implicated factory (156, 252). Between December 1998 and February 1999, an increase in listeriosis cases due to L. monocytogenes serotype 3a was recognized in Finland. A total of 25 cases, most of which were hematological or organ transplant patients, were identified as part of the outbreak; 6 patients died from Listeria infection. Butter served at the tertiary care hospital was implicated as the source of infection. The epidemic strain was isolated from all 13 butter samples obtained from the hospital kitchen and from several lots from the dairy and wholesale store. One sample contained 11,000 L. monocytogenes CFU/g, with the others having lower counts (5 to 60 CFU/g) (236). Another outbreak of listeriosis associated with butter occurred in 2003 in the United Kingdom. Between May and November of 2000, a listeriosis outbreak in the United States impacted 11 states. When subtyped, the L. monocytogenes isolates from these cases were all serotype 1/2a and were indistinguishable from each other by PFGE. Eight perinatal and 21 nonperinatal cases were reported. Among the 21 nonperinatal case patients, the median age was 65 years (range, 29 to 92 years); 13 (62%) were female. This outbreak resulted in four deaths and three miscarriages/stillbirths. A casecontrol study implicated sliced processed turkey meat from a delicatessen. A traceback investigation identified a single food processing plant as the source of this outbreak and led to the recall of 16 million pounds of processed turkey meat. The same plant had been identified in a Listeria contamination event that had occurred a decade previously (276). An outbreak of listeriosis, identified in North Carolina in 2000, affected 13 people, of whom 12 were 18- to 34-year-old Hispanic females. This outbreak, which

515 resulted in five stillbirths, was traced to Mexican-style cheese produced from contaminated raw milk at a local dairy (237). Another listeriosis outbreak in Texas was associated with legally imported but illegally distributed cheese prepared in Mexico. Five of six case patients reported eating the implicated cheese. These outbreaks underscore the need for educating Hispanic women about food safety considerations during pregnancy and enforcing existing laws that regulate the illegal sale and importation of certain cheeses that have been prepared from raw milk. During October 2002, another multistate outbreak of listeriosis was recognized in Pennsylvania and eight other states in the United States (147). This outbreak of 54 cases included eight deaths and three fetal deaths. Case-control studies implicated delicatessen turkey meat as the source of the outbreak. In traceback studies, the outbreak strain, which was related to the 1998–1999 outbreak strain by PFGE, was found in the environment of one turkey processing plant and in turkey products produced by another plant. Together, the two plants recalled more than 30 million pounds of turkey meat products (Table 20.2). This outbreak, together with the outbreak in 2000 that was traced to delicatessen turkey, led to the three previously discussed alternatives for control of Listeria in RTE meats, with no additional large-scale RTE meat-relate­d outbreaks having been identified in the United States since 2002. However, in 2008 a nationwide outbreak in Canada that included 57 cases of illness and 23 fatalities was traced to various delicatessen meats that became contaminated during packaging at a larger manufacturing facility in the Province of Ontario. The L. monocytogenes isolates from this outbreak belonged to serotype 1/2a but were of two distinct PFGE patterns (139). This outbreak led to the adoption of additional preventative measures in Canada as described earlier. Since the initial link between consumption of coleslaw and listeriosis in 1981 that confirmed L. monocytogenes as a foodborne pathogen, fresh produce was thought to play a minor role in invasive illness. However, in 2010 consumption of commercially diced celery was linked to 10 cases of listeriosis, including 5 fatalities in the state of Texas (179). In 2011, a much larger outbreak of listeriosis in the United States was traced to cantaloupes that were grown and packed in Colorado, with a total of 146 cases of illness, including 30 deaths, reported across 28 states from July 31 to October 27. The victims ranged in age from <1 to 96 years with a median age of 77 years (60). An unusual feature of this outbreak was the involvement of four different L. monocytogenes strains, belonging to serotypes 1/2a or 1/2b, that exhibited widely divergent PFGE patterns (61). Three of these

Foodborne Pathogenic Bacteria

516 Table 20.2  Selected invasive listeriosis outbreaks, 1990 to 2011 Yr(s)

Geographic location

No. of persons affected

No. of deaths/ fetal deaths

Vehicle

Serotype

Reference(s)

Blue mold cheese or hard cheese Pork tongue in jelly Smoked mussels Pork rillettes Chocolate milk Gravad rainbow trout and coldsmoked rainbow trout Raw-milk soft ripened cheese Imitation crab (?) Pont-l’Evêque soft cheese Frankfurters

4b 4b 1/2 4b 1/2b 4b

188 152 17 154 78, 291 109

4b 1/2b 4b 4b

154 113 185 252

Butter Commercially prepared fruit salad Pâté

3a NAa

236 305

4b

272

4b 4b 1/2a

88 88 276

1989–90 1992 1992 1993 1994 1994–95

Denmark France New Zealand France Illinois Sweden

26 279 4 38 3 6

6 85 2 11 0 1

1995 1996 1997 1998–99

France Ontario, Canada France United States (multiple states) Finland Australia

33 2 14 108

4 0 0 14/4

11 6

4 5

Connecticut, Maryland, New York France France United States (multiple states) North Carolina

11

0

10 32 30

3 10 4/3

Rillettes Pork tongue in jelly Turkey deli meat

13

0/5

4b

237

United States (multiple states) Quebec, Canada British Columbia, Canada Texas United Kingdom Switzerland United States

54

8/3

Homemade Mexican-style cheese Delicatessen turkey meat

4b

147

17 47

0 0

Cheese made from raw milk Cheese

NA 4b

131 278

13 5 10 5

2 0 3 3

4b 1/2 1/2a NA

272 81 27 75

Canada Canada Austria, Germany, Czech Republic Louisiana United States United States

57 40 34

23 2 8

Mexican-style cheese Prepacked sandwiches Soft cheese Pasteurized flavored and nonflavored milk RTE deli meats Cheese Acid curd cheese “Quargel”

1/2a 1/2a

293 132 121

8 10 146

2 5 30

Hog head cheese Celery Cantaloupe

1/2a NA 1/2a and 1/2b

62 10 60, 61

1998–99 1998–99 1999 1999–2000 1999–2000 2000 2000 2002 2002 2002 2003 2003 2005 2007 2008 2008 2009–10 2010 2010 2011 a

NA, not available.

four strains were recovered from the processing line or packing area, suggesting multiple incoming sources of contamination (366). The L. monocytogenes strains responsible for the major outbreaks from 1981 to 1992 (Canada, 1981; California, 1985; Switzerland, 1983–1987; France, 1992) were all serotype 4b and belonged to a small number of closely related clones as evidenced by ribotyping, multilocus enzyme

electrophoresis, and DNA macrorestriction pattern analysis (45). This group of closely related, epidemic-associated strains was designated ECI. The 1998–1999 and 2002 multistate outbreaks in the United States were caused by strains of a novel epidemic-associated lineage, ECII (110). During the past decade, several outbreaks of febrile gastroenteritis caused by L. monocytogenes have been documented (Table 20.3). These gastroenteritis

Yr(s)

Geographic location

Incubation time (h)

1986–87

Pennsylvania

Unknown

Unknown

1989

United States

NA

9

1991 1993 1996

Tasmania, Australia Italy Illinois

NA 18–43 20

4 18 80

1998?

Finland

<27

5

1997

Northern Italy

24

1,566

2000

Winnipeg, Canada

NA

>20

2000

Australia

2001

Los Angeles, CA

NA

16

2001

Sweden

31

120

2001

Japan

24–144

38

2002

British Columbia, Canada

NA

86

a

No. of persons affected

31

Contamination level (CFU/g or ml)

Reference

Multiple serotypes

a

NA

299

4b

NA

299

Smoked mussels Rice salad Chocolate milk, temperature abused Cold-smoked rainbow trout Cold salad of corn and tuna Defective cans of aerosol-dispensed whipping cream RTE corned beef and ham Precooked, sliced turkey

1/2a 1/2b 1/2b

NA NA 109

262 313 78

1/2a

1.9 × 105

259

4b

106

14

NA

NA

240

1/2

1.8 × 107

344

1/2a

1.6 × 109

123

Fresh, raw milk cheese

1

6.3 × 107

79

Locally made cheese

1/2b

238

Cheese made from pasteurized milk/water

NA

<0.3 to 4.6 × 107 2 10 to 109

Symptoms Fever, vomiting, and diarrhea in the week before positive culture Fever, muscle pain, gastroenteritis Gastrointestinal Gastrointestinal, flu-like Fever, diarrhea Nausea, abdominal cramps, diarrhea, fever Headache, abdominal pain, diarrhea, fever Gastrointestinal

Fever, muscle pain, headache, diarrhea Body aches, fever, headache, diarrhea, vomiting Diarrhea, fever, stomach cramps, vomiting Gastrointestinal, flu-like symptoms Febrile gastroenteritis

Vehicle Unknown

Shrimp

Serotype

20.  Listeria monocytogenes

Table 20.3  Outbreaks of gastrointestinal manifestation of listeriosis in humans with no known predisposing condition

278

NA, not available.

517

518 outbreaks differ in several respects from the invasive outbreaks just described. They affect persons with no known predisposing risk factors for listeriosis. The infectious dose appears to be much higher (1.9 × 105 to 1.6 × 109 CFU/g or ml) than that for typical invasive listeriosis in the susceptible population. Finally, the symptoms appear within a day or less (typically in 18 to 27 h) of exposure (similar to other bacterial enteric infections) in gastrointestinal listeriosis in contrast to the several weeks or more of incubation observed for invasive listeriosis.

CHARACTERISTICS OF DISEASE Human disease caused by L. monocytogenes predominantly occurs in certain well-defined high-risk groups that include pregnant women, neonates, immunocompromised adults, and the elderly but has also been documented in individuals with no predisposing underlying conditions. In nonpregnant adults, L. monocytogenes primarily causes septicemia, meningitis, and meningoencephalitis, with a mortality rate of 20 to 30%. Other infrequent manifestations of listeriosis in this population include endocarditis in persons with underlying cardiac lesions (including prosthetic or porcine valves) and various types of focal infections, including endo-ophthalmiti­s, septic arthritis, osteomyelitis, pleural infection, and peritonitis (279). Clinical conditions known to predispose persons to the serious manifestations of listeriosis include malignancy, organ transplants, immunosuppressive therapy, HIV infection, and advanced age. Although pregnant women, particularly in the third trimester of pregnancy, may experience only mild flu-like symptoms (fever and myalgias with or without diarrhea) as a result of L. monocytogenes infection, the infection has serious consequences for the fetus, leading to stillbirth or abortion. In neonates who are less than 7 days old, sepsis and pneumonia are the predominant syndromes, whereas in older neonates, the infection manifests as meningitis and sepsis. Several recent investigations of listeriosis outbreaks show that L. monocytogenes also causes febrile gastro­ enteritis in otherwise healthy hosts (Table 20.3). Interestingly, Murray and Pirie had clearly stated in their original descriptions of cases from the 1920s that diarrhea was a common feature of listeriosis in small ­animals (247). The most compelling evidence for the gastrointestinal manifestation of L. monocytogenes infections comes from outbreak investigations by Dalton et al. (78) and Aureli et al. (14). Fever and diarrhea are the most consistent symptoms in gastrointestinal listeriosis. Also, the incubation time for the enteric form

Foodborne Pathogenic Bacteria is rather short, usually in the range of 18 to 27 hours. However, the incubation time varied between 24 and 144 hours in the 2001 Japan outbreak, possibly because of wide-ranging contamination levels in the implicated cheese (Table 20.3). In the five investigations in which attempts were made to quantify the numbers of L. monocytogenes cells in the implicated foods, the contamination levels were very high (range, 1.9 × 105 to 1.6 × 109 CFU). From the data available thus far for gastrointestinal listeriosis, it appears that infection requires a high dose of L. monocytogenes. It is not known whether the strains involved in the gastrointestinal forms of listeriosis possess additional virulence factors similar to those of common enteric pathogens. While listeriosis outbreaks attract the most attention, most cases of human listeriosis occur sporadically. However, some of these sporadic cases may be unrecognized common-source clusters. The source and route of infection of most of these cases remain unknown, although foodborne transmission has been demonstrated in some instances. Many cases have not been associated with food because of the difficulties of prospectively investigating sporadic cases of the disease. Long incubation times (up to 10 weeks) make accurate food histories difficult if not impossible to obtain, and incriminated foodstuffs almost always have been consumed or discarded. Understanding the epidemiology of sporadic cases is critical to the development of effective control strategies. An active surveillance program for listeriosis in the United States from 1989 to 1991 based on a population of 19 million people yielded an estimated incidence of 7.9 cases per million. A casecontrol study of dietary risk factors conducted during the same period indicated that foodborne transmission was responsible for about one-third of cases and that patients with listeriosis were more likely than controls to have eaten soft cheeses or foods purchased from deli­ catessen counters (284, 330). By 1993, the incidence of sporadic cases of listeriosis in the active surveillance sites had declined to 4.4 cases per million (358), with this decrease attributed to enhanced efforts by the food industry and regulatory agencies to prevent contamination of processed foods. The most recent 2010 data from CDC’s foodborne diseases active surveillance program (FoodNet) indicate that the incidence of listeriosis in the United States had declined to 2.4 cases per million population (58) with an estimated 1,600 cases, including 255 fatalities, occurring annually (321). The reported incidence of listeriosis in eight western or northern European countries varied between 3.9 and 10.3 cases per million in 2006 with a mean of 6.5 per million persons (153). There has been an

20.  Listeria monocytogenes upward trend in listeriosis cases in several European countries during the past several years, and there has been a concerted effort to identify the factors contributing to this increased incidence. The mean rate of listeriosis in Australia was 3.0 cases per million from 2004 to 2008 (277). According to a report from 2000, the estimated incidence of listeriosis in Japan was far lower, 0.65 cases per million people (180). The most reliable surveillance data for listeriosis come from industrialized countries; prevalence in other areas is either unknown or low. Whether this reflects differences in consumption patterns, dietary habits, host susceptibility, food processing and storage technologies, or lack of awareness or laboratory facilities is not known. Although exposure to L. monocytogenes is common, invasive listeriosis is rare. It is unclear whether this is due to early acquired protection, to intrinsic resistance of the average healthy host, or to most strains being only weakly virulent. Despite improvements in the emerging picture concerning the key events affecting the relative susceptibility of humans to this pathogen (44), there remain gaps in our understanding of the onset of disease (see below). Many (2 to 6%) healthy individuals are asymptomatic fecal carriers of L. monocytogenes. The risk of clinical disease in intestinal carriers of L. monocytogenes is unknown but must be very low given the rarity of diagnosis. Endogenous infections by L. monocytogenes in the gut are plausible, especially in patients receiving immunosuppressive therapy, which can impair resistance to infection and alter the intestinal defense mechanisms to favor listerial invasion. Nevertheless, asymptomatic fecal carriage has been observed in pregnant women who proceed to normal birth at term, and women who have given birth to infected infants do not necessarily suffer the same problem in later pregnancies. Similarly, recent transplant recipients may harbor L. monocytogenes in the gut without developing the disease. Epidemiologic studies since 1981 have focused on the role of contaminated food in transmission of listeriosis. However, two unusual transmission routes have been described. Hospital-acquired listeriosis is sporadically described, mainly in nurseries, with equipment serving as the vehicle. During intrauterine infections, amniotic fluid can contain up to 108 L. monocytogenes CFU/ml (249). Mineral oil was also implicated in one outbreak of neonatal listeriosis (332). Primary cutaneous infections without systemic involvement have been observed as an occupational disease in veterinarians and farmers with most cases caused by manipulation of presumably infected bovine fetuses or cows.

519

INFECTIOUS DOSE AND SUSCEPTIBLE POPULATIONS

Infectious Dose

The infectious dose of L. monocytogenes depends on many factors, including the immunological status of the host, the type of food consumed, the degree of virulence of the particular strain, and the numbers ingested, as seen from the previously discussed outbreaks (279). The ­severity of the disease is such that tests with human volunteers are impossible. Studies involving monkeys and various rodents suggest that reducing levels of exposure will reduce clinical disease (111, 385, 386). However, these experiments do not help to determine the minimal infective dose for humans. Further, the current assumption is that for infectious and toxicoinfectious agents, a single cell has a small but finite probability of causing disease. In the case of L. monocytogenes, the FAO/WHO risk assessment estimated that the median probability for a single cell causing listeriosis is in the range of 10–12 to 10–14 (7) based on exposure data and annual disease statistics. When exposed to L. monocytogenes at the beginning of the third trimester, pregnant rhesus monkeys showed an increased risk of delivering a stillborn infant with pathology similar to that of humans, including acute inflammation, placentitis, fetal liver necrosis, and isolation of Listeria from the placental and fetal tissues (352). Published data indicate that the populations of L. monocytogenes in contaminated food responsible for epidemic and sporadic foodborne cases usually contained more than 100 CFU/g. The frankfurters implicated in the 1998 listeriosis outbreak in the United States contained less than 0.3 CFU/g (252). However, these results may not accurately reflect the numbers consumed, since Listeria populations in the food would be expected to change between the time of consumption and analysis. It is clear that the risk of foodborne listeriosis is overwhelmingly associated with RTE foods that support the growth of L. monocytogenes at refrigeration or marginal abuse temperatures. Hence, additional epidemiologic information is clearly needed to more accurately assess the infectious dose.

Susceptible Populations

Most human cases of listeriosis occur in individuals who have a predisposing disease that leads to impairment of their T-cell-mediated immunity (279). The percentage of patients suffering from a known underlying condition varies from 70 to 85% in some surveys to nearly 100% in others (46, 346). The most commonly affected populations include those at the extreme ages (neonates and the elderly), pregnant women, and individuals who are immunosuppressed by medication

520 (corticosteroids or cytotoxic drugs), especially after organ transplantation or illness (hematologic malignancies, such as leukemia, lymphoma, and myeloma as well as solid malignancies). Listeriosis is 300 times more frequent in people with AIDS than in the general population (333). In addition to impaired T-cell immunity, a small percentage of listeriosis patients suffer from chronic diseases not usually associated with immunosuppression, such as congestive heart failure, diabetes, cirrhosis, alcoholism, and systemic lupus erythematosus, alone or in association with known predisposing diseases. Mortality in listeriosis cases is almost exclusively associated with predisposing diseases and conditions (138), with recent transplant patients being the subpopulation at greatest risk, as they are ~2,500 times more susceptible than healthy adults under 65 years of age (7). Concurrent infection can also influence susceptibility to listeriosis. This was exemplified by a cluster of cases in 1987 in Philadelphia that were caused by a number of different strains. Clinical and epidemiologic investigations suggested that individuals who were previously asymptomatic for listerial infection but whose gastrointestinal tract harbored L. monocytogenes became symptomatic, possibly because of a coinfecting agent (333). However, a single food vehicle could not be identified due to the diversity of strains.

VIRULENCE FACTORS/MECHANISMS OF PATHOGENICITY In the early 1980s, L. monocytogenes was appearing as an attractive model system to study both intracellular parasitism and the immunological response of humans and animals to infection. This was due to five main reasons: (i) L. monocytogenes grows well in culture media, (ii) it can be genetically manipulated, (iii) it belongs to a genus that contains pathogenic and nonpathogenic species, (iv) it infects mammalian cells in tissue culture, and (v) several laboratory animals are susceptible to Listeria (215). The L. monocytogenes genome sequence was determined in 2001, paving the way for the discovery of a series of new virulence factors (106, 201, 372). Since that initial study, over 25 L. monocytogenes genome sequences have been published, including those of representative strains from all four lineages (86). In addition, genome sequences for L. innocua, L. welshimeri, L. seeligeri, L. marthii, and L. grayii are available (86). Listeria is one of the most intensively studied pathogens (71). For example, one of the authors makes available a database of Listeria research articles, which

Foodborne Pathogenic Bacteria is approaching 5,500 entities (http://www.foodrisk. org/exclusives/cfs3/).

Pathogenicity of L. monocytogenes

Many tests for addressing L. monocytogenes pathogenicity have been developed, including tissue culture assays and tests using laboratory animals, in particular immunocompetent and immunocompromised mice (216, 285). Animal models routinely involve intraperitoneal (i.p.) or intragastric (i.g.) infection, and virulence is evaluated either by comparing the LD50s or by enumerating bacteria in infected target organs, in particular the spleen and liver. Heterogeneity in the virulence of L. monocytogenes has been observed in several in vitro and in vivo studies (LD50 values of L. monocytogenes strains range from 103 to 109 CFU) (7, 41, 368, 386), depending on the animal or in vitro system being used and whether the immune system of the animal was suppressed prior to exposure. No clear correlation between the level of virulence and the origin (human, animal, category of food, or environment) or the strain characteristics (serotype, phage type, multilocus enzyme type, or DNA micro- or macrorestriction patterns) could be established. However, only 3 of the 13 known serovars of L. monocytogenes, 1/2a, 1/2b, and 4b, account for more than 90% of human cases of listeriosis (232, 333). Among the listeriosis-associated serotypes, 4b strains cause a large proportion of listeriosis outbreak cases worldwide, serotypes 1/2a and 1/2b account for a significant portion of sporadic cases, and serotypes 1/2a, 1/2b, and 1/2c predominate in food isolates (28, 29, 304, 384). As discussed earlier, specific epidemic clones of serotype 4b and 1/2a have been linked to specific SNPs. The relatively low percentage of cases from serotypes 1/2a, 1/2b, and 1/2c from foods appears to be strongly influenced by mutations leading to premature stop codons in inlA, which encodes internalin A (183, 192, 370). Although rare nonpathogenic or weakly pathogenic L. monocytogenes isolates have been reported, all strains of L. monocytogenes are considered to be potentially capable of causing human disease. There has been substantial debate over the appropriateness of various routes of administration and ­animal models for studying L. monocytogenes pathogenicity, particularly in regard to fetal infections. Since most listeriosis cases are associated with foodborne transmission, it is generally accepted that an oral route of administration is most suited, particularly since there can be substantial differences in the ability of L. monocytogenes strains to cross the intestinal barrier. Crossing both the intestinal and placental barriers involves two proteins, internalin A (InlA) and internalin

20.  Listeria monocytogenes B (InlB), which mediate high-specificity attachment to specific epithelial cells and low-specificity attachment to a wide variety of cell types, respectively (29). The former attaches to the E-cadherin receptor, while the latter attaches to the hepatocyte growth factor (Met) receptor. E-cadherin receptors differ among mammalian species, with InlA from L. monocytogenes binding to the receptors on the intestinal epithelial cells of humans, guinea pigs, rabbits, and gerbils, but not mice or rats, bringing into question the appropriateness of using mice to study orally transmitted ­ listeriosis. Similarly, there are differences in host specificity related to the ability of L. monocytogenes to cross the placental barrier, with the pathogen invading the barrier in humans, rats, rabbits, and gerbils, but not mice or guinea pigs. High-efficiency invasion of the placental barrier requires receptors for both InlA and InlB (31, 90). There have been an increasing number of dose-response trials and related research that have focused on the use of Rhesus monkeys, guinea pigs, gerbils, or “humanized” mice (91, 129, 220, 295, 352, 385).

Experimental Infection and Cell Biology of the Infectious Process

Much of the original work on the disease course of L. monocytogenes was performed with mice injected intravenously (i.v.), with subsequent monitoring of bacterial growth kinetics in the spleen and liver. Within 10 minutes after i.v. injection, 90% of the inoculum is taken up by the liver and 5% to 10% by the spleen. During the first 6 hours, the number of viable listeriae in the liver decreases 10-fold, indicating rapid destruction of most of the bacteria. Surviving listeriae then multiply within susceptible macrophages and grow exponentially in the spleen and liver for the next 48 hours, peaking at day 2 or 3 postinfection (12). Rapid inactivation ensues during the next 3 to 4 days, indicating recovery of the host. Convalescent mice are resistant to challenge and have a delayed-type hypersensitivity characterized by swelling of the footpads injected with crude cell preparations of L. monocytogenes. However, as described above, the natural route of foodborne infections with L. monocytogenes in humans is via the gastrointestinal tract. L. monocytogenes infects intestinal epithelial cells in a process that requires the interaction of InlA, expressed at the bacterial surface, with epithelial E-cadherin, expressed at the surface of epithelial cells (257). Although L. monocytogenes has a relatively broad host range, the relative susceptibility of mammalian species to ingested L. monocytogenes varies based in large part on the

521 a­ttachment/invasion of the intestinal epithelium. As indicated above, this reflects differences in the epithelial receptors. For example, mice are relatively resistant to intestinal infection with L. monocytogenes because of a single-amino-acid difference between human and mouse E-cadherin (219). Recent studies using different L. monocytogenes strains report that, in contrast to strains from other serovars, serotype 4b epidemic strains appear to be able to cause systemic infection in mice infected orally, suggesting that there might be serovar-specific virulence factors playing a role in mouse susceptibility to orally acquired listeriosis (76). The probability that an oral infection will lead to perinatal infections has been related to the probability of specific “key events”: (i) survival of the pathogen in the upper gastrointestinal tract, (ii) attachments and uptake into the intestinal epithelial cells, (iii) survival and escape from phagosomes and transfer to phagocytes, (iv) transfer across placenta, and (v) pathogen growth leading to fetal morbidity and mortality (44). From the intestinal lumen, bacteria traverse the ­epithelial-cell layer and disseminate via the bloodstream to other organs, such as the spleen and liver, where they are internalized by splenic and hepatic macrophages in which they can survive and replicate (Fig. 20.3). When the bacteria reach the liver and the spleen, most Listeria cells are rapidly killed. In the initial phase of infection, infected hepatocytes are the targets for neutrophils and later for mononuclear phagocytes that are responsible for control and resolution of infection. The bacterial cells are subsequently transported via the blood to regional lymph nodes. Depending on the level of T-cell response induced in the first days following initial infection, further dissemination via the blood to the brain or, in the pregnant animal, the placenta, may subsequently occur. L. monocytogenes possesses the ability to cross the maternofetal barrier and leads to placental abscesses, chorioamnionitis, and finally infection of the fetus. As the highest concentrations of L. monocytogenes are encountered in the gut and in the lung, it is thought that infection might be amplified through ingestion of contaminated amniotic fluid rather than solely as a consequence of the hematogenous transplacental route (231, 325). Several animal models have been developed to study pregnancy-associated (1, 16, 218, 264) and pulmonary (264) L. monocytogenes infections. L. monocytogenes has, in addition to its ability to cross initially the intestinal barrier and the maternofetal barrier, the capacity to cross the blood-brain barrier and reach the central nervous system (CNS) and cause meningitis, encephalitis, and brain abscesses.

Foodborne Pathogenic Bacteria

522

Figure 20.3  Schematic representation of the pathophysiology of Listeria infection. doi:10.1128/9781555818463.ch20f3

Invasion of brain cell endothelial cells seems a prerequisite for meningeal pathogens that penetrate the CNS. It has been proposed that L. monocytogenes utilizes penetration of human brain microvascular endothelial cells as a means of crossing the bloodbrain barrier. L. monocytogenes adheres to human brain microvascular endothelial cells through the microvilli and then infects them by an InlB-dependent m­echanism (161). Once inside a susceptible host, L. monocytogenes has the potential to disseminate and multiply in a wide variety of cell types and tissues. Listeria is primarily found intracellularly due to its ability to induce its own phagocytosis into many cell lines (e.g., macrophages, fibroblasts, heptocytes, and epithelial cells) that are normally nonphagocytic (71). Detailed analysis of infected-tissuecultured cells reveals a complex series of interactions between the bacteria and the cell (Fig. 20.4). Host cell infection begins with the adhesion and internalization of the bacteria either by phagocytosis in the case of macrophages or induced phagocytosis (invasion) in the case of normally nonphagocytic cells. It has been suggested that at the intestinal epithelium, InlA provides high-specificity adhesion while InlB activates the c-MET receptor, which in turn accelerates junctional endocytosis into the enterocytes (282), with goblet cells being the specific location of attachment (269). Bacterial invasion starts by a close contact with the plasma membrane that progressively enwraps the bacterium. This process is u­sually referred to as the “zipper” mechanism in con-

trast to the “trigger” mechanism used by Salmonella or Shigella (71). Following internalization, bacteria reside within membrane-bound vacuoles (phagosomes) for ~30 min. L. monocytogenes escapes the vacuoles by perforating the membrane by the combined action of listeriolysin O (LLO) and two phospholipases. Neumann et al. (267) concluded that L. monocytogenes could escape ­phagosomes independent of listeriolysin in interleukin4-deactivated human macrophages. Once free in the cytosol, L. monocytogenes replicates and becomes covered with actin filaments. These filaments rearrange within 2 hours into long comet tails (up to 40 µm in length) left behind in the cytosol while the bacteria move ahead at a speed of ~0.3 mm/s (77, 359). When moving bacteria contact the plasma membrane, they induce the formation of bacterium-containing protrusions. Contact between these protrusions and neighboring cells results in the internalization of the protrusion (301). In the newly infected cell, the bacterium is surrounded by two plasma membranes that must be lysed to initiate a new cycle of multiplication and movement. Lysis of the two membrane vacuoles requires at least four virulence factors: LLO, the metalloprotease Mpl, and two phospholipases (see below). The entire cycle is completed in about 5 hours. This process allows L. monocytogenes to disseminate by direct cell-to-cell transfer, thereby avoiding humoral defenses such as antibodies. This accounts for the early observations that antibodies are not protective and that immunity to Listeria is T cell mediated (215).

20.  Listeria monocytogenes

523

Figure 20.4  Schematic representation of the successive steps of the cell infectious process. Factors implicated in the different steps are indicated. doi:10.1128/9781555818463.ch20f4

Virulence Factors

The overall pathogenesis of L. monocytogenes and identification of individual virulence determinants have been studied extensively using a variety of genetic techniques. Conjugative transposons from the Tn1545-Tn916 family or Tn917 transposons were initially exploited to generate mutant libraries allowing the identification of several L. monocytogenes virulence factors (72, 127, 196, 286). More recently, Tn1545- or Tn917-derived tagged transposons were also used for signature-tagged mutagenesis in L. monocytogenes (15, 102, 239). As previously discussed, genome sequencing of multiple strains has allowed the direct comparison of pathogenic and nonpathogenic isolates for the occurrence of specific virulence determinants. Plasmid vectors originating from Bacillus subtilis or Escherichia coli are used for genetic studies in L. monocytogenes, including allelic exchange of chromosomal DNA, cloning, gene expression, or reporter gene fusion (119). For allelic exchange, in-frame deletion, or site-specific mutagenesis, thermosensitive vectors such as pKSV7 (351) or pAUL-A (322) are used. A novel thermosensitive plasmid, pMAD, harboring the b-galactosidase gene, facilitates the screening and the generation of allelic exchanges (11). For complementation and gene expression, shuttle plasmids replicating in E. coli and

L. monocytogenes, i.e., pAT18, pMK4, and pAM401, are mostly used (356, 362, 388). Complementation of deletion mutants can now be achieved using a new integrative phage-derived plasmid, pPL2, that allows a single-copy integration at a specific location in the L. monocytogenes chromosome (214). Listeriae are not naturally competent; however, transformation with plasmid DNA can be obtained on protoplasts or by electroporation (233, 254, 374). Transduction has relatively recently become a tool for the identification of bacteriophages that grow at 30°C (174). SNP analysis has been used extensively to study the evolution of Listeria virulence and the emergence of epidemic clones. These tools have led to the successive identification of multiple virulence determinants that are necessary to modulate the ability of L. monocytogenes to cause disease and determine the relative virulence ability of individual isolates. Gene expression analysis is providing an appreciation of the large number of genes (>70) that may be involved in in vivo infections and the complexity associated with the disease course and the ability of the microorganism to evade cellular defenses (53). Some of the well-defined virulence determinants are described in more detail below in relation to their role in L. monocytogenes pathogenesis.

524

Adhesion to Mammalian Cells

The first step in the disease course of foodborne listeriosis is the attachment of L. monocytogenes to the intestinal epithelium. In addition to its role as the first step in the internalization and passage through the intestinal epithelium, adhesion may promote colonization of the gastrointestinal tract, direct the bacteria to appropriate target cells or tissues such as the CNS or the placenta, and may even facilitate cell invasion by activating cell signal transduction pathways or triggering the synthesis of a target cell receptor required for invasion. Adherence involves a number of surface proteins, including InlA, ActA, Ami, p104, and FbpA. InlA is the first member of the internalin multigene family, a family characterized by the presence of a leucine-rich repeat (LRR) region (127). InlA promotes adherence and entry into cells expressing its receptor, the adhesion protein E-cadherin (257) (see above and below). ActA, the surface protein required for actin-based motility, may also promote attachment via host cell proteoglycans (3, 354). An ActA-deficient mutant is significantly impaired in attachment/entry into IC-21 murine macrophages and Chinese hamster ovary epithelial cells. Ami is a bacterial surface amidase that presents an N-terminal domain similar to the amidase domain of the Atl autolysin of Staphylococcus aureus and a C-terminal, cell wall-anchoring domain made up of eight GW dipeptide modules (37). Besides its lytic activity on Listeria cell walls (251), Ami is implicated in bacterial adhesion to target cells (261). Ami mutants are attenuated, indicating that Ami plays a role in the virulence of L. monocytogenes. The involvement of a cell surface protein of 104 kDa (p104) in the adhesion of L. monocytogenes to the human intestinal cell line Caco-2 has also been proposed (280). Finally, FbpA is a fibronectinbinding protein shown to be required for intestinal and liver colonization after oral infection of transgenic mice expressing human E-cadherin. FbpA binds to immobilized human fibronectin and increases adherence of wildtype L. monocytogenes to HEp-2 cells. Despite the lack of conventional secretion/anchoring signals, FbpA is detected on the bacterial surface and was shown to be a substrate of the SecA2 pathway (102). In addition, FbpA behaves as a chaperone for two important virulence factors, LLO and InlB, probably preventing their degradation. The roles of surface proteins, including adhesion, have been reviewed by Bierne and Cossart (23).

Entry into Mammalian Cells

InlA and InlB were the first listerial factors identified as mediating bacterial invasion into different target cell types. As discussed earlier, their cellular receptors have also been identified. The molecular signaling cascades

Foodborne Pathogenic Bacteria triggered during Listeria entry into host cells are being characterized in detail. InlA and InlB are just two of a family of internalins (e.g., InlC, InlD, InlH, and InlJ) that are characterized by the presence of LRRs (49). Some internalins are present in most lineages, whereas others are associated with specific lineages or serotypes (86). Some are associated with adhesion to enterocytes (InlA), whereas others may play more of a role in stimulating junction endocytosis (see above). Recent findings suggest that other molecules are also necessary for inter­ nalization (P60, ActA, Auto, and Vip), revealing a complex dialogue between Listeria and eukaryotic cells d­uring the early phases of the infection cycle.

Internalins InlA (800 amino acids) displays a classical signal ­sequence followed by an LRR region comprising 15 repeats of 22 amino acids. An interrepeat (IR) region separates the LRR region from a second repeat region (called the B repeat region). The carboxy terminus displays a cell wall-sorting motif, LPTTG, which is the substrate for the enzyme sortase A and allows covalent linkage of the protein to the peptidoglycan. The LRR and IR regions are necessary and sufficient to promote Listeria entry into human epithelial cells (219). E-cadherin, the InlA receptor (257), belongs to the cadherin superfamily of cell adhesion molecules, which are transmembrane glycoproteins located at adherens junctions and allow cell-cell adhesion. Listeria exploits the whole molecular junctional complex involved in adherens junctions for inducing its entry into target cells (353). The speciesspecific nature of the interaction between L. monocytogenes and animal hosts (see above) involves recognition of a proline at position 16 (Pro16) of the E-cadherin molecule (217). Also, as previously mentioned, the truncation of InlA that occurs naturally in some strains a­ssociated with foods helps to explain the decreased incidence of listeriosis cases associated with 1/2a, 1/2b, and 1/2c from food sources, and a recent epidemiologic survey indicates that clinical strains express a full-length InlA far more frequently than do strains recovered from food products (183, 193), revealing that InlA is critical for the pathogenesis of human listeriosis. InlB (630 amino acids) is another member of the internalin family involved in entry of Listeria into a broad range of cell lines including hepatocytes and nonepithelial cells (101) (see above). It contains a signal peptide, seven LRR repeats, one IR region, and one B repeat. The C-terminal domain presents three tandem repeats (GW modules) that mediate attachment of InlB to the bacterial cell wall through noncovalent interaction with lipoteichoic acids (192). The LRR region of InlB is suf-

20.  Listeria monocytogenes ficient to confer invasiveness to noninvasive L. innocua or to latex beads (39). The crystal structures of the LRR domain (242), of the LRR/IR domains (329), and of the full-length InlB (241) indicate that the protein can accommodate several ligands. Indeed, InlB interacts with several cellular receptors, which likely cooperate to promote bacterial uptake. gC1qR was the first InlB receptor identified (38). gC1qR was initially identified as the receptor for C1q, the first component of the complement cascade (281). InlB-dependent entry is blocked by antigC1qR antibodies and by C1q, and transfection with human gC1qR enhances cell invasion (38). However, gC1qR has no transmembrane domain or glycosylphosphatidylinositol-anchored domain, suggesting that it likely behaves as a coreceptor for a signaling protein. The signaling receptor, Met, is the receptor for the hepatocyte growth factor. Met is a transmembrane protein that mediates several signaling pathways triggered by InlB. The interaction with the receptor Met occurs through the LRR domain of InlB (342). In addition, InlB directly binds to cellular glycosaminoglycans (GAGs) through its GW modules. InlB-dependent entry into epithelial cells is strongly affected by depletion of the cellular plasma membrane GAGs. It was proposed that interaction of InlB with GAGs through its GW modules leads to its detachment from the bacterial surface, allowing its clustering at the cellular surface through binding to Met by its LRR domain and favoring the local activation of the signaling pathway downstream of Met (194).

P60 For a long time, Diap (encoding invasion-associated protein) mutants could not be obtained, suggesting that the protein was essential for bacterial viability. Therefore, the role of P60 (encoded by iap) was first evaluated in spontaneous rough mutants expressing lower levels of P60 and forming long filamentous structures composed of bacterial chains (208). The rough mutants are less virulent and enter less efficiently in certain eukaryotic cells, suggesting a role for P60 in bacterial invasion (165, 171, 208). Ultimately, a viable Diap mutant was obtained, allowing more precise studies of the role of P60. As for rough mutants, it also had a defect in septum formation and in virulence after i.v. infection of mice. In addition, the Diap mutant is impaired in bacterial movement and spreading from cell to cell due to an improper localization of ActA at the surface of L. monocytogenes (223, 283). However, in contrast to studies using rough mutants, the invasiveness of this Diap mutant in mouse ­fibroblasts and human epithelial cells is only slightly diminished compared with that of wild-type bacteria (223, 283). P60 is an autolysin both secreted by and associated

525 with the bacterial cell wall (208, 308, 389). Secretion is mediated as for FbpA by the recently identified auxiliary secretion system SecA2, which mediates the secretion of at least 17 secreted and surface proteins of L. monocytogenes (223, 283). In addition, P60 plays an important role in the immune response against L. monocytogenes infection. Antibodies specific for P60 can act as opsonins and may play a role in preventing systemic infections in immunocompetent individuals (205). Moreover, P60 is a major protective antigen that induces both T-CD8 and Th1 protective immune responses (135, 136, 168).

ActA ActA is a surface protein implicated in the attachment to cells (see above) and responsible for the actin-based motility of Listeria (see below). Moreover, expression of ActA in L. innocua is sufficient to promote bacterial entry in some epithelial cell lines (354). Auto A novel autolysin-encoding gene, aut, was identified recently (50). It is the only autolysin gene that is absent from the nonpathogenic species L. innocua. The aut gene encodes a surface protein, Auto, with an autolytic activity, as expected from the presence of a domain harboring homologies with autolysin-encoding genes, especially Nacetylglucosaminidases. The protein Auto possesses a Cterminal cell wall-anchoring domain made up of four GW modules, similar to those observed in the other autolysin, Ami, and in InlB (49). The morphology of an aut deletion mutant is similar to that of the wild type, with no defect in septation and cell division, suggesting no role for Auto in these functions (50). Auto is required for entry of L. monocytogenes into different nonphagocytic eukaryotic cell lines. An aut deletion mutant has a reduced virulence following i.v. inoculation of mice and oral infection of guinea pigs, which correlates with its low invasiveness. However, the autolytic activity of Auto by itself, rather than an invasive ability, might be critical for virulence. Indeed, Auto may control the general surface architecture exposed to the host by L. monocytogenes and/or the composition of the surface products released by the bacteria, hence affecting the host response to infection (50). Vip The vip gene encodes an LPXTG surface protein absent from nonpathogenic Listeria species (51). vip is positively regulated by PrfA, the transcriptional activator of the major Listeria virulence factors. Vip is anchored to the Listeria cell wall by sortase A and is required for entry into some mammalian cells. Using a ligand overlay approach, a Vip cellular receptor was identified. It is the endoplasmic

526 reticulum resident protein Gp96, reported to also interact with toll-like receptors (224). The Vip-Gp96 interaction is critical for bacterial entry into some cells. Comparative infection studies using oral and i.v. inoculation of nontransgenic and transgenic mice expressing human E-cadherin demonstrated a role for Vip in Listeria virulence not only at the intestine level but also in late stages of the infectious process. Vip thus appears to be a new virulence factor exploiting Gp96 as a receptor for cell invasion and/or signaling events that may interfere with the host immune response in the course of the infection (51).

Sortases The genome of L. monocytogenes (strain EGD-e) contains 41 genes that encode LPXTG proteins including InlA and Vip (49). LPXTG proteins are anchored to the peptidoglycan by transpeptidases named sortases (326). srtA encodes a sortase responsible for the anchoring of InlA, Vip, and at least 13 other proteins to the peptidoglycan (25, 51, 294). L. monocytogenes mutants lacking srtA are defective in internalizing into human enterocytic-like Caco-2 cells and hepatocytic HepG2 cells, as well as in colonizing the liver and spleen of mice infected orally. The Listeria genome encodes another sortase, SrtB, which has a limited number of substrates containing a C-terminal NXXTN sorting motif. SrtB is not required for the infectious process (24).

Escape from the Phagosome and Intracellular Growth

Once internalized, the next key step in the disease course of L. monocytogenes is to escape the vacuole/phagosome via the concerted action of LLO and two phospholipases C (PLCs), each of which is considered a virulence determinant. Once in the target cell cytoplasm, Listeria then multiplies, acquiring nutrients from the cytosol.

Listeriolysin O Nonhemolytic L. monocytogenes strains can be found in the environment but are avirulent in experimentally infected mice. Nonhemolytic mutants generated by conjugative transposons were found to be similarly avirulent in mice (126, 196, 286). Both wild-type and nonhemolytic strains entered mammalian cells but were unable to grow intracellularly due to an inability to escape from the phagocytic vacuole (126). Subsequent analysis indicated that the mutations resulted in disruption of hly, which encodes LLO. LLO is a pore-forming toxin that enables the bacterium to escape into the cytosol, where it replicates. LLO accomplishes this by binding to cholesterol and then oligomerizing into large complex pores (181). LLO is a 60-kDa secreted protein involved in Listeria’s escape from both

Foodborne Pathogenic Bacteria primary and secondary vacuoles (70, 103, 196, 286). LLO is a member of the pore-forming, cholesterol-dependent cytolysin family. These thiol-activated toxins are produced by several gram-positive bacteria, including Streptococcus, Pneumococcus, Clostridium, and Bacillus species (255). They are active only on cholesterol-containing membranes, with cholesterol likely acting as the receptor in the membrane. After an initial step of binding, LLO monomers diffuse laterally to form ring-shaped ion-permeable pores of 30 nm of diameter, without disrupting the plasma membrane (298). Unlike all other thiol-activated toxins, LLO has a maximal activity at pH 5 and is inactive at pH 7, and thus its deleterious effect on cellular membranes is impaired when the bacterium is free in the cytosol (143, 328). In addition, optimal pore formation by LLO occurs between pH 5.5 and 6.0 (the pH of the early phagosome) (19). A PEST-like sequence (P, Pro; E, Glu; S, Ser; T, Thr) has been detected in the LLO molecule (83). This motif was first thought to target LLO for degradation when present in the cytosol, restricting its function to the vacuolar environment and inhibiting the lysis of host cells. LLO is denaturized in the cytosol (334), and the PEST sequence does not mediate proteasomal degradation by the host (327). A second hemolysin, listeriolysin S, has been isolated from specific strains of lineage I L. monocytogenes, leading to consideration that this second hemolysin may account for the increased pathogenicity of this lineage (73).

Phospholipases C Two phospholipases are produced and secreted by L. monocytogenes: a phosphoinositide-phospholipase (PIPLC), and a broad-range phospholipase, phosphatidylinositoyl-specific phospholipase C (PC-PLC) (143). The two phospholipases have a membrane-damaging activity and are involved in bacterial escape from primary and/or secondary phagosomes (55, 146, 163, 244, 371). Each of the two L. monocytogenes phospholipases is ­important for virulence, since mutants deficient in either PI-PLC or PC-PLC are attenuated (54, 350). More importantly, double mutants deficient in both phospholipases are 500 times less virulent than single mutants, highlighting the importance and the complementarity of these factors in listeriosis (350). The PC-PLC is synthesized as a proenzyme maturated by the metalloprotease encoded by mpl (92, 218, 244). It hydrolyzes a wide variety of phospholipids including sphingomyelin. Both phospholipases act synergistically with LLO in lysing primary and secondary vacuoles (54,133). In the absence of LLO, the PC-PLC can also promote lysis of primary vacuoles in human epithelial cell lines (163, 244). In synergy with LLO, the PI-PLC induces hydrolysis of PI and production of diacylglycerol in macrophages (144, 343), leading to

20.  Listeria monocytogenes mobilization of protein kinase Cd and subsequent elevation of intracellular calcium levels (377). pH regulates the activity of both Mpl on itself and PC-PLC (117). Once free in the cytosol, L. monocytogenes starts multiplying, with an approximate doubling time of 1 hour (360). A number of genes are upregulated as the microorganism grows within the host tissue (53). This includes several genes encoding virulence or metabolic determinants induced during L. monocytogenes’s intracellular life, including those involved in phagosomal lysis, actin-based motility, and cell-to-cell spreading (42, 120, 202). In contrast to most bacteria, L. monocytogenes replicates into the cytosol when it is directly m­icroinjected into cells (142). The cytosol permissiveness for L. monocytogenes growth is likely due to the bacterium’s ability to use a variety of cytosolic nutrients, as suggested by the fact that intracellular multiplication of several auxotrophic mutants is not affected (243). The precise nutritional requirements and the mechanisms used by Listeria to obtain nutrients from the host cell cytosol are only starting to be elucidated. Intracytosolic growth of L. monocytogenes is highly dependent on the hpt gene, which encodes a sugar uptake system (142), and is tightly regulated by PrfA (260). The ability to accomplish both uptake and biosynthesis of

527 thiamine precursors appears to be needed for intracellular replication of L. monocytogenes (323). Branchedchain fatty acids synthesis through lipoate dependency facilitates the intracellular growth of L. monocytogenes (198). Access to adequate zinc nutrition via the two zinc uptake systems in L. monocytogenes appears to be a prerequisite for intracellular growth (68).

Intracytoplasmic Movement and Cell-to-Cell Spreading

Induced phagocytosis, escape from the phagosome, and intracellular multiplication are essential steps for infection of individual cells but are not sufficient to achieve infection. L. monocytogenes must also efficiently infect tissues by direct cell-to-cell spreading. As described earlier, direct cell-to-cell spreading is the result of intracellular movement, protrusion formation, phagocytosis of the protrusion by a neighboring cell, formation of a ­doublemembrane-bound vacuole, and lysis of the two-membrane vacuole. Intracytoplasmic movement is strictly coupled to continuous actin assembly that provides the force for bacterial propulsion (Fig. 20.5). Actin polymerization requires expression of the actA gene. actA mutants are invasive, escape from the phagosome, replicate, and form intracellular microcolonies but are not covered with actin,

Figure 20.5  Directional actin polymerization by L. monocytogenes. L. monocytogenes isolates were processed for triple-labeling fluorescence microscopy, 5 h after starting the infection of Vero cells. Bacteria (red) were visualized with a polyclonal anti-Listeria antibody, actin (green) with phalloidin, and cell nuclei (blue) with DAPI (4¢,6¢-diamidino-2-phenylindole). Magnification, ×100. doi:10.1128/9781555818463.ch20f5

Foodborne Pathogenic Bacteria

528 do not move intracellularly, and do not spread from cell to cell. ActA is a 639-amino-acid protein that presents a signal sequence at its amino-terminal domain and a transmembrane motif at its carboxyl-terminal domain that anchors the molecule to the bacterial surface (93, 204). A central repeat domain containing four proline-rich repeats stimulates the Listeria actin-based motility (212). This domain binds members of the enabled/vasodilatorstimulated phosphoprotein (Ena/VASP) family of proteins (63, 230, 268), which modulate bacterial speed and directionality (13, 134). Immunolocalization of the ActA protein in infected cells has revealed that ActA is asymmetrically distributed on the bacterial surface, with an increasing concentration towards the opposite pole of the bacterial cell, which is the site of comet tail formation. The amino-terminal region of ActA alone can induce bacterial movement (213). It binds and activates Arp2/3, a sevenprotein host complex that induces actin polymerization and the generation of a dendritic array of actin filaments (69). By doing so, ActA mimics WASP (Wiskott-Aldrich protein) family proteins (35, 345). Actin filaments in Listeria tails, as in Shigella tails, are branched, in contrast to those in Rickettsia conorii tails, which display longer and unbranched actin filaments (148–150).

Coregulation of Virulence Factors

Most Listeria virulence genes involved in the steps of the infection cycle (i.e., prfA, plcA, hly, mpl, actA, plcB,

inlA, inlB, inlC, hpt, bsh, and vip) are regulated by the transcriptional activator PrfA (Fig. 20.6) (51, 104, 222, 256, 260). PrfA is a protein comprising 233 amino acids that belongs to the Crp/Fnr family. Genes regulated by PrfA contain a binding site in the −41 region, the PrfA box, consisting of a 14-bp palindromic sequence to which the putative helix-turn-helix motif of PrfA binds in vitro (341). Expression of prfA is a complex phenomenon. prfA is the second gene of the plcA-prfA operon, whose primary promoter is regulated by prfA. Hence, prfA autoregulates (activates) its own synthesis. prfA can also be transcribed from its own promoter region. It has been determined that in a prfA mutant transcription at the prfA-specific promoter is upregulated (118). PrfA-regulated virulence factors such as LLO, PlcA, PlcB, and Vip are produced at higher levels as temperature increases to 37°C, the temperature of the host (51, 221). At a temperature lower than 30°C, the low levels of expression are correlated to undetectable levels of PrfA, although its transcript is synthesized (297). At low temperatures, the untranslated leader region of the prfA mRNA forms a stable secondary structure masking the Shine-Dalgarno sequence (191). Consequently, PrfA is not translated. At high temperatures, this structure, named “thermosensor,” melts. Mutations destabilizing the thermosensor at 30°C mimic the effect of an increase in temperature, i.e., they unmask the ribosome-binding site and lead to virulence gene expression. Knowledge of

Figure 20.6  Schematic representation of the L. monocytogenes virulence factors. The localization of factors implicated in adhesion (green), entry (blue), escape from the phagosome and intracellular growth (red), and intracytoplasmic movement and cell-to-cell spreading (purple) is indicated. The names of factors whose expression is regulated by PrfA are in orange. doi:10.1128/9781555818463.ch20f6

20.  Listeria monocytogenes the L. monocytogenes genome sequence (140) allowed analysis of the complete PrfA regulon. A whole-genome array based on the sequence of the L. monocytogenes EGDe strain was constructed to study the PrfA regulon (260). This transcriptomic analysis revealed that PrfA could act both as an activator and as a repressor. It led to the identification of many new PrfA-regulated genes in addition to genes previously known to be part of the PrfA regulon. However, in many cases, the effect of PrfA is probably indirect. This analysis has been expanded upon in a second transcriptional analysis to identify the genes that are positively controlled by PrfA and upregulated in the host (53).

Stress Proteins of L. monocytogenes

L. monocytogenes has evolved a series of adaptive responses in order to cope with a large variety of stresses to survive and/or multiply under harsh environmental conditions, outside as well as inside the host. The LisRK two-component signal transduction system is implicated in virulence, acid and ethanol tolerance, and oxidative stress (195). virR, a gene encoding the response regulator of another two-component system, was shown to regulate a key regulon in L. monocytogenes, controlling virulence by a global regulation of surface component modifications (25). The DnaK heat shock chaperone protein is required for survival of L. monocytogenes under high temperatures and acidic conditions as well as for efficient phagocytosis with macrophages (166, 167). The major heat shock chaperones, GroES and GroEL, are induced at high temperature, at low pH, and during cell infection (125, 167). The ClpC stress protein is required for survival under iron deprivation or high temperature or osmolarity in bone marrow macrophages and in organs of mice (307). Two other proteases have been reported to be involved in stress response and virulence, namely ClpE and ClpP (128, 266). The fri gene, which encodes ferritin, is necessary for optimal growth in minimal medium in both the presence and absence of iron, as well as after cold and heat shock. Ferritin also provides protection against reactive oxygen species and is essential for full virulence of L. monocytogenes. A comparative proteomic analysis revealed an effect of the fri deletion on the levels of LLO and several stress proteins (105, 170). Adaptation to osmotic stress depends on the intracellular accumulation of osmolytes and thus on uptake systems that include BetL, GbuABC, OpuC, OpuB, and OppA (32, 203, 348, 349, 367, 382). Bile tolerance of L. monocytogenes involves the bile salt hydrolase Bsh, an enzyme that deconjugates bile salts and that is required for both intestinal and hepatic phases of listeriosis (104). The transporter BtlA (21) and other

529 systems such as the putative transporter of the glutamate decarboxylase GadE and the zinc uptake regulator ZurR contribute to tolerance to bile or various other stresses, e.g., low pH, salt, ethanol, detergent, and antibiotic (20). The alternative sigma factor B contributes to the ability of L. monocytogenes not only to survive and/ or multiply under stressful conditions outside the host (acid, osmotic, or oxidative stresses, low temperature, or carbon starvation) (115) but also to persist within the host during the infectious process. It has been demonstrated that sigma B contributes to survival following exposure to bile salts (22) and also to transcription of the virulence gene activator PrfA (265). Characterization of the sigma B-dependent general stress regulon confirmed the broad role of this sigma factor (355). The lessons learned in relation to L. monocytogenes responses to stressors within the host are beginning to be used to study the impact of the pathogen’s expression of stress responses and virulence determinants in foods and food processing environments (296, 300).

CONCLUDING REMARKS Public health surveillance, outbreak investigations, and applied and basic research conducted during the past 30 years have helped to characterize the disease listeriosis, define the magnitude of its public health problem and its impact on the food industry, identify the risk factors associated with disease, and develop appropriate and targeted control strategies. The food processing industry has made progress towards reducing the prevalence of L. monocytogenes in processing plant environments and in high-risk foods, and preventive measures have been developed and implemented for persons at increased risk of infection. These gains were reflected in a substantial decline in the incidence of listeriosis during the late 1990s and early 2000s. For example, in the United States the incidence of listeriosis declined, in 2005, to 0.3 cases per 100,000 population. However, since that time the number of listeriosis cases has plateaued, with no substantial declines in the United States, and several European countries have experienced increased incidences of listeriosis during the past 5 years. Furthermore, we have seen the association of listeriosis outbreaks with new foods such as cut produce, something that was predictable based on the multiple-risk assessments related to L. monocytogenes. Thus, despite ever-increasing knowledge concerning this pathogen, translation of that knowledge into effective control programs has lagged. It also implies that “doing the same old thing” is not likely to further lower the public health burden of L. monocytogenes. Hopefully, the same p­rogress in

Foodborne Pathogenic Bacteria

530 u­nderstanding the basic biology of this important pathogen during the past decade will translate in the next decade into effective risk-based food safety systems for preventing this rare but potentially life-threatening foodborne disease.

  12.   13.

References   1. Abram, M., D. Schluter, D. Vuckovic, B. Wraber, M. Doric, and M. Deckert. 2003. Murine model of pregnancy-associated Listeria monocytogenes infection. FEMS Immunol. Med. Microbiol. 35:177–182.   2. Al-Ghazali, M., and S. Al-Azawi. 1990. Listeria monocytogenes contamination of crops grown on soil treated with sewage sludge cake. J. Appl. Bacteriol. 69:642–674.   3. Alvarez-Dominguez, C., J. A. Vazquez-Boland, E. Carrasco-Marin, P. Lopez-Mato, and F. Leyva-Cobian. 1997. Host cell heparan sulfate proteoglycans mediate attachment and entry of Listeria monocytogenes, and the listerial surface protein ActA is involved in heparan sulfate receptor recognition. Infect. Immun. 65:78–88.   4. Anonymous. 2002. Final Assessment Report: Proposal 239 - Listeria Risk Assessment & Risk Management Strategy. Food Standards Australia New Zealand. www. foodstandards.gov.au/_srefiles/P239_DAR.021002.pdf. Accessed 24 May 2006.   5. Anonymous. 1988. Food Listeriosis—Report of the WHO Informal Working Group. WHO/EHE/FOS/88.5. World Health Organization, Geneva, Switzerland.   6. Anonymous. 2003. Interpretive Summary: Quantitative Assessment of the Relative Risk to Public Health from Foodborne Listeria monocytogenes among Selected Categories of Ready-To-Eat Meats. Center for Food Safety and Applied Nutrition, Food and Drug Admin­ istration, U.S. Department of Health and Human Services; Food Safety and Inspection Service, U.S. Department of Agriculture. www.foodsafety.gov/~dms/ lmr2toc.html. Accessed 24 May 2006.   7. Anonymous. 2004. Risk Assessment of Listeria monocytogenes in Ready-To-Eat Foods: Technical Report. Microbiological Risk Assessment Series 5. World Health Organization and Food and Agriculture Organization, Geneva, Switzerland.   8. Anonymous. 2008. 2008 Listeriosis Outbreak in Ontario—Chronology of Events. http://www.health. gov.on.ca/english/public/pub/disease/listeria/listeriosis_ outbreak_chronology.pdf. Accessed 7 February 2012.   9. Anonymous. 2010. FSIS comparative risk assessment for Listeria monocytogenes in ready-to-eat meat and poultry deli meats. U.S. Department of Agriculture—Food Safey and Inspection Service. http://www.fsis.usda.gov/ PDF/Comparative_RA_Lm_Report_May2010.pdf.   10. Anonymous. 2010. DSHS orders Sanger Produce to close, recall products. http://www.dshs.state.tx.us/news/ releases/20101020.shtm. Accessed 31 March 2012.   11. Arnaud, M., A. Chastanet, and M. Debarbouille. 2004. New vector for efficient allelic replacement in naturally

  14.

  15.

  16.

  17.   18.

  19.

  20.

  21.

  22.

  23.   24.

  25.

nontransformable, low-GC-content, gram-positive bacteria. Appl. Environ. Microbiol. 70:6887–6891. Audurier, A., P. Pardon, J. Marly, and F. Lantier. 1980. Experimental infection of mice with Listeria monocytogenes and L. innocua. Ann. Microbiol. 131B:47–57. Auerbuch, V., J. J. Loureiro, F. B. Gertler, J. A. Theriot, and D. A. Portnoy. 2003. Ena/VASP proteins contribute to Listeria monocytogenes pathogenesis by controlling temporal and spatial persistence of bacterial actin-based motility. Mol. Microbiol. 49:1361–1375. Aureli, P., G. C. Fiorucci, D. Caroli, G. Marchiaro, O. Novara, L. Leone, and S. Salmaso. 2000. An outbreak of febrile gastroenteritis associated with corn contaminated by Listeria monocytogenes. N. Engl. J. Med. 342:1236–1241. Autret, N., I. Dubail, P. Trieu-Cuot, P. Berche, and A. Charbit. 2001. Identification of new genes involved in the virulence of Listeria monocytogenes by signature-tagged transposon mutagenesis. Infect. Immun. 69:2054–2065. Bakardjiev, A. I., B. A. Stacy, S. J. Fisher, and D. A. Portnoy. 2004. Listeriosis in the pregnant guinea pig: a model of vertical transmission. Infect. Immun. 72:489–497. Baker, M., M. Brett, P. Short, L. Calder, and R. Thornton. 1993. Listeriosis and mussels. CDNZ 93: 12–15. Barbosa, W. B., L. Cabedo, H. J. Wederquist, J. N. Sofos, and G. R. Schmidt. 1994. Growth variation among species and strains of Listeria monocytogenes. J. Food Prot. 57:765–769. Beauregard, K. E., K. D. Lee, R. J. Collier, and J. A. Swanson. 1997. pH-dependent perforation of macrophage phagosomes by listeriolysin O from Listeria monocytogenes. J. Exp. Med. 186:1159–1163. Begley, M., C. G. Gahan, and C. Hill. 2002. Bile stress response in Listeria monocytogenes LO28: adaptation, cross-protection, and identification of genetic loci involved in bile resistance. Appl. Environ. Microbiol. 68:6005–6012. Begley, M., C. Hill, and C. G. Gahan. 2003. Identification and disruption of btlA, a locus involved in bile tolerance and general stress resistance in Listeria monocytogenes. FEMS Microbiol. Lett. 218:31–38. Begley, M., R. D. Sleator, C. G. Gahan, and C. Hill. 2005. Contribution of three bile-associated loci, bsh, pva, and btlB, to gastrointestinal persistence and bile tolerance of Listeria monocytogenes. Infect. Immun. 73:894–904. Bierne, H., and P. Cossart. 2007. Listeria monocytogenes surface proteins: from genome predictions to function. Microbiol. Mol. Biol. Rev. 71:377–397. Bierne, H., C. Garandeau, M. G. Pucciarelli, C. Sabet, S. Newton, F. Garcia-del Portillo, P. Cossart, and A. Charbit. 2004. Sortase B, a new class of sortase in Listeria monocytogenes. J. Bacteriol. 186:1972–1982. Bierne, H., S. K. Mazmanian, M. Trost, M. G. Pucciarelli, G. Liu, P. Dehoux, L. Jansch, F. Garcia-del Portillo, O. Schneewind, P. Cossart, and T. E. L. G. Consortium.

20.  Listeria monocytogenes

  26.

  27.

  28.

  29.

  30.   31.

  32.

  33.

  34.

  35.

  36.

  37.

2002. Inactivation of the srtA gene in Listeria monocytogenes inhibits anchoring of surface proteins and affects virulence. Mol. Microbiol. 43:464–489. Bille, J., J. Rocourt, and B. Swaminathan. 2003. Listeria and Erysipelothrix, p. 461–471. In P. R. Murray, E. Jo Baron, J. H. Jorgensen, M. A. Pfaller, and R. H. Yolken (ed.), Manual of Clinical Microbiology, 8th ed., vol. 1. ASM Press, Washington, DC. Bille, J., D. S. Blanc, H. Schmid, K. Boubaker, A. Baumgartner, H. H. Siegrist, M. L. Tritten, R. Lienhard, D. Berner, R. Anderau, M. Treboux, J. M. Ducommun, R. Malinverni, D. Genné, P. H. Erard, and U. Waespi. 2006.  Outbreak of human listeriosis associated with Tomme cheese in Northwest Switzerland, 2005. Euro. Surveill. 11:91–93. Boerlin, P., F. Boerlin-Petzold, E. Bannerman, J. Bille, and T. Jemmi. 1997. Typing Listeria monocytogenes isolates from fish products and human listeriosis cases. Appl. Environ. Microbiol. 63:1338–1343. Boerlin, P., and J. Piffaretti. 1991. Typing of human, animal, food, and environmental isolates of Listeria monocytogenes by multilocus enzyme electrophoresis. Appl. Environ. Microbiol. 57:1624–1629. Boerlin, P., J. Rocourt, F. Grimont, P. A. D. Grimont, C. Jacquet, and J. C. Piffaretti. 1992. Listeria ivanovii subspecies londoniensis. Int. J. Syst. Bacteriol. 15:42–46. Bonazzi, M., M. Lecuit, and P. Cossart. 2009. Listeria monocytogenes internalin and E-cadherin: from bench to bedside. Cold Spring Perspect. Biol. 2009:1a003087. http://www.ncbi.nlm.nih.gov/pmc/­ articles/PMC2773623/. Accessed 21 April 2012. Borezee, E., E. Pellegrini, and P. Berche. 2000. OppA of Listeria monocytogenes, an oligopeptide-binding protein required for bacterial growth at low temperature and involved in intracellular survival. Infect. Immun. 68:7069–7077. Borucki, M. K., S. H. Kim, D. R. Call, S. C. Smole, and F. Pagotto. 2004. Selective discrimination of Listeria monocytogenes epidemic strains by a mixed-genome DNA microarray compared to discrimination by pulsedfield gel electrophoresis, ribotyping, and multilocus sequence typing. J. Clin. Microbiol. 42:5270–5276. Borucki, M. K., J. Reynolds, D. R. Call, T. J. Ward, B. Page, and J. Kadushin. 2005. Suspension microarray with dendrimer signal amplification allows direct and high-throughput subtyping of Listeria monocytogenes from genomic DNA. J. Clin. Microbiol. 43:3255–3259. Boujemaa-Paterski, R., E. Gouin, G. Hansen, S. Samarin, C. Le Clainche, D. Didry, P. Dehoux, P. Cossart, C. Kocks, M. F. Carlier, and D. Pantaloni. 2001. Listeria protein ActA mimics WASp family proteins: it activates filament barbed end branching by Arp2/3 complex. Biochemistry 40:11390–11404. Brackett, R. E. 2007. Incidence and behavior of Listeria monocytogenes in products of plant origin, p. 655–680. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. Braun, L., S. Dramsi, P. Dehoux, H. Bierne, G. Lindahl, and P. Cossart. 1997. InlB: an invasion protein of

531

  38.

  39.

  40.

  41.

  42.

  43.

  44.

  45.

  46.

  47.   48.   49.

  50.

Listeria monocytogenes with a novel type of surface association. Mol. Microbiol. 25:285–294. Braun, L., B. Ghebrehiwet, and P. Cossart. 2000. gC1qR/p32, a C1q-binding protein, is a receptor for the InlB invasion protein of Listeria monocytogenes. EMBO J. 19:1458–1466. Braun, L., F. Nato, B. Payrastre, J. C. Mazie, and P. Cossart. 1999. The 213-amino-acid leucine-rich repeat region of the Listeria monocytogenes InlB protein is sufficient for entry into mammalian cells, stimulation of PI 3-kinase and membrane ruffling. Mol. Microbiol. 34:10–23. Brehm-Stecher, F., and E. A. Johnson. 2007. Rapid methods for detection of Listeria, p. 257–282. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. Brosch, R., B. Catimel, G. Milon, C. Burchrieser, E. Vindel, and J. Rocourt. 1993. Virulence heterogeneity of Listeria monocytogenes strains from various sources (food, human, animal) in immunocompetent mice and its association with typing characteristics. J. Food Prot. 56:296–301. Bubert, A., Z. Sokolovic, S. K. Chun, L. Papatheodorou, A. Simm, and W. Goebel. 1999. Differential expression of Listeria monocytogenes virulence genes in mammalian host cells. Mol. Gen. Genet. 261:323–336. Buchanan, R. L., W. G. Damert, R. C. Whiting, and M. van Schothorst. 1997. An approach for using epidemiologic and microbial food survey data to develop a “purposefully conservative” estimate of the dose­response relationship between Listeria monocytogenes levels and the incidence of foodborne listeriosis. J. Food Prot. 60:918–922. Buchanan, R. L., A. H. Havelaar, M. A. Smith, R. C. Whiting, and E. Julien. 2009. The key events doserespons­e framework: its potential for application to foodborne pathogenic microorganisms. Crit. Rev. Food Sci. Nutr. 49:718–728. Buchrieser, C., R. Brosch, B. Catimel, and J. Rocourt. 1993. Pulsed-field electrophoresis applied for comparing Listeria monocytogenes strains involved in outbreaks. Can. J. Microbiol. 39:395–401. Bula, C., J. Bille, and M. Glauser. 1995. An epidemic of food-borne listeriosis in Western Switzerland: description of 57 cases involving adults. Clin. Infect. Dis. 20:66–72. Buncie, S. 1991. The incidence of Listeria monocytogenes in slaughtered animals, in meat, and in meat products in Yugoslavia. Int. J. Food Microbiol. 12:173–180. Burt, S. 2004. Essential oils: their antibacterial properties and potential applications in foods—a review. Int. J. Food Microbiol. 94:223–253. Cabanes, D., P. Dehoux, O. Dussurget, L. Frangeul, and P. Cossart. 2002. Surface proteins and the pathogenic potential of Listeria monocytogenes. Trends Microbiol. 10:238–245. Cabanes, D., O. Dussurget, P. Dehoux, and P. Cossart. 2004. Auto, a surface associated autolysin of Listeria monocytogenes required for entry into eukaryotic cells and virulence. Mol. Microbiol. 51:1601–1614.

Foodborne Pathogenic Bacteria

532   51. Cabanes, D., S. Sousa, A. Cebria, M. Lecuit, F. Garciadel Portillo, and P. Cossart. 2005. Gp96 is a receptor for a novel Listeria monocytogenes virulence factor, Vip, a surface protein. EMBO J. 24:2827–2838.   52. Cai, S., D. Y. Kabuki, A. Y. Kuaye, T. G. Cargioli, M. S. Chung, R. Nielsen, and M. Wiedmann. 2002. Rational design of DNA sequence-based strategies for subtyping Listeria monocytogenes. J. Clin. Microbiol. 40:3319–3325.   53. Camejo, A., C. Buchrieserr, E. Couve, F. Carvalho, O. Reis, P. Ferreira, S. Sousa, P. Cossart, and D. Cabanes. 2009. In vivo transcriptional profiling of Listeria monocytogenes and mutagenesis identify new virulence factors involved in infection. PLoS Pathog. 5(5):e1000449. www.plospathogens.org. Accessed 21 March 2012.   54. Camilli, A., H. Goldfine, and D. A. Portnoy. 1991. Listeria monocytogenes mutants lacking phosphatidylinositol-specific phospholipase C are avirulent. J. Exp. Med. 173:751–754.   55. Camilli, A., L. G. Tilney, and D. A. Portnoy. 1993. Dual roles of PlcA in Listeria monocytogenes pathogenesis. Mol. Microbiol. 8:143–157.   56. Cates, S. C., R. A. Morales, S. A. Karns, L.-A. Jaykus, K. M. Kosa, T. Teneyck, C. M. Moore, and P. Cowen. 2006 Consumer knowledge, storage, and handling practices regarding Listeria in frankfurters and deli meats: results of a web-based survey. J. Food Prot. 69:1630–1639.   57. Centers for Disease Control and Prevention. 1989. Listeriosis associated with consumption of turkey franks. MMWR Morb. Mortal. Wkly. Rep. 38:267–268.   58. Centers for Disease Control and Prevention. 2010. Vital signs: incidence and trends of infection with pathogens transmitted commonly through food—Foodborne Diseases Active Surveillance Network, 10 U.S. sites, 1996– 2010. MMWR Morb. Mortal. Wkly. Rep. 60:749–755.   59. Centers for Disease Control and Prevention. 1996. Preventing Foodborne Illness: Listeriosis. U.S. Depart­ ment of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Infectious Diseases, Atlanta, GA.   60. Centers for Disease Control and Prevention. 2011. Multistate outbreak of listeriosis linked to whole cantaloupes from Jensen Farms, Colorado. http:// www.cdc.gov/listeria/outbreaks/cantaloupes-jensenfarms/120811/index.html. Accessed 12 February 2011.   61. Centers for Disease Control and Prevention. 2011. Multistate outbreak of listeriosis associated with Jensen Farms cantaloupe—United States, August-September 2011. http://www.cdc.gov/mmwr/preview/mmwrhtml/ mm6039a5.htm?s_cid=mm6039a5_w. Accessed 12 February 2011.   62. Centers for Disease Control and Prevention. 2010. Outbreak of invasive listeriosis associated with consumption of hog head cheese—Louisiana, 2010. MMWR Morb. Mortal. Wkly. Rep. 60:401–405.   63. Chakraborty, T., F. Ebel, E. Domann, K. Niebuhr, B. Gerstel, S. Pistor, C. J. Temm-Grove, B. M. Jockusch, M. Reinhard, and U. Walter. 1995. A focal adhesion ­factor

  64.

  65.   66.

  67.   68.

  69.   70.   71.   72.

  73.

  74.

  75.

  76.

  77.

directly linking intracellularly motile Listeria monocytogenes and Listeria ivanovii to the actin-based cytoskeleton of mammalian cells. EMBO J. 14:1314–1321. Chen, Y., W. H. Ross, M. J. Gray, M. Wiedmann, R. C. Whiting, and V. N. Scott. 2006. Attributing risk to Listeria monocytogenes subgroups: dose response in relation to genetic lineages. J. Food Prot. 69:335–344. Chen, Y., W. H. Ross, V. N. Scott, and D. E. Gombas. 2003. Listeria monocytogenes: low levels equal low risk. J. Food Prot. 66:570–577. Chen, Y., W. Zhang, and S. J. Knabel. 2005. Multi­virulence-locus sequence typing clarifies epidemiology of recent listeriosis outbreaks in the United States. J. Clin. Microbiol. 43:5291–5294. Colburn, K., C. Kaysner, C. Abeyta, Jr., and M. Wekell. 1990. Listeria species in a California estuarine environment. Appl. Environ. Microbiol. 56:2007–2011. Corbett, D., J. Wang, S. Schuler, G. Lopez-Castejon, S. Glenn, D. Brough, P. W. Andres, J. S. Cavet, and I. S. Roberts. 2012. Two zinc uptake systems contribute to the full virulence of Listeria monocytogenes during growth in vitro and in vivo. Infect. Immun. 80: 14–21. Cossart, P. 2000. Actin-based motility of pathogens: the Arp2/3 complex is a central player. Cell. Microbiol. 2:195–205. Cossart, P., and J. Mengaud. 1989. Listeria monocytogenes: a model system for the molecular study of intracellular parasitism. Mol. Biol. Med. 6:463–474. Cossart, P., and P. J. Sansonetti. 2004. Bacterial invasion: the paradigms of enteroinvasive pathogens. Science 304:242–248. Cossart, P., M. F. Vicente, J. Mengaud, F. Baquero, J. C. Perez-Diaz, and P. Berche. 1989. Listeriolysin O is essential for virulence of Listeria monocytogenes: direct evidence obtained by gene complementation. Infect. Immun. 57:3629–3636. Cotter, P. D., L. A. Draper, E. M. Lawton, K. M. Daly, D. S. Groeger, P. G. Casey, R. P. Ross, and C. Hill. 2008. Listeriolysin S, a novel peptide haemolysin associated with a subset of lineage I Listeria monocytogenes. PLoS Pathog. 4(9):e1000144. www.plospathogens.org. Accessed 21 March 2012. Cox, L., T. Kleiss, J. Cordier, C. Cordellana, P. Konkel, C. Pedrazzini, R. Beumer, and A. Siebenga. 1989. Listeria spp. in food processing, non-food and domestic environments. Food Microbiol. 6:49–61. Cummings, M. P. Kudt, B. Matyas, A. DeMaria, T. Stiles, L. Han, M. Gilchrist, P. Neves, E. Fitzgibbons, and S. Condon. 2008. Outbreak of Listeria monocytogenes infections associated with pasteurized milk from a local dairy—Massachusetts, 2007. MMWR Morb. Mortal. Wkly. Rep. 57:1097–1100. Czuprynski, C. J., N. G. Faith, and H. Steinberg. 2003. A/J mice are susceptible and C57BL/6 mice are resistant to Listeria monocytogenes infection by intragastric inoculation. Infect. Immun. 71:682–689. Dabiri, G. A., J. M. Sanger, D. A. Portnoy, and F. S. Southwick. 1990. Listeria monocytogenes moves r­apidly

20.  Listeria monocytogenes

  78.

  79.

  80.   81.

  82.

  83.   84.

  85.

  86.

  87.

  88.

  89.

through the host-cell cytoplasm by inducing directional actin assembly. Proc. Natl. Acad. Sci. USA 87:6068–6072. Dalton, C. B., C. C. Austin, J. Sobel, P. S. Hayes, W. F. Bibb, L. M. Graves, B. Swaminathan, M. E. Proctor, and P. M. Griffin. 1997. An outbreak of gastroenteritis and fever due to Listeria monocytogenes in milk. N. Engl. J. Med. 336:100–105. Danielsson-Tham, M. L., E. Eriksson, S. Helmersson, M. Leffler, L. Ludtke, M. Steen, S. Sorgjerd, and W. Tham. 2004. Causes behind a human cheese-borne outbreak of gastrointestinal listeriosis. Foodborne Pathog. Dis. 1:153–159. Davies, E., and M. Adams. 1994. Resistance of Listeria monocytogenes to the bacteriocin nisin. Int. J. Food Microbiol. 21:341–347. Dawson, S. J., M. R. W. Evans, D. Willby, J. Bardwell, N. Chamberlain, and D. A. Lewis. 2006.  Listeria outbreak associated with sandwich consumption from a hospital retail shop, United Kingdom. Euro Surveill. 11:89–91. De Buyser, M. L., B. Dufour, M. Maire, and V. Lafarge. 2001. Implication of milk and milk products in foodborne diseases in France and in different industrialised countries. Int. J. Food Microbiol. 67:1–17. Decatur, A. L., and D. A. Portnoy. 2000. A PEST-like sequence in listeriolysin O essential for Listeria monocytogenes pathogenicity. Science 290:992–995. den Bakker, H. C., B. M. Bowen, L. D. RodriguezRivera, and M. Wiedmann. 2012. FSL J1-208, a virulent uncommon phylogenetic linage IV Listeria monocytogenes strains with a small chromosome size and a putative virulence plasmid carrying internalin-like genes. Appl. Environ. Microbiol. 78:1876–1889. den Bakker, H. C., B. N. Bundrant, E. D. Fortes, R. H. Orsi, and M. Wiedmann. 2010. A population and phylogenetic approach to understand the evolution of virulence in the genus Listeria. Appl. Environ. Microbiol. 76:6085–6100. den Bakker, H. C., C. A. Cummings, V. Ferreira, P. Vatta, R. H. Orsi, L. Degoricija, M. Baker, O. Petrauskene, M. R. Furtado, and M. Wiedmann. 2010. Comparative genomics of the bacterial genus Listeria: genome evolution is characterized by limited gene acquisition and limited gene loss. BMC Genomics 11:688. http://www. biomedcentral.com/1471-2164/11/688. Accessed 30 March 2012. de Simon, M., and M. D. Ferrer. 1998. Initial numbers, serovars and phagevars of Listeria monocytogenes isolated in prepared foods in the city of Barcelona (Spain). Int. J. Food Microbiol. 44:141–144. de Valk, H., V. Vaillant, C. Jacquet, J. Rocourt, F. Le Querrec, F. Stainer, N. Quelquejeu, O. Pierre, V. Pierre, J. C. Desenclos, and V. Goulet. 2001. Two consecutive nationwide outbreaks of listeriosis in France, October 1999-February 2000. Am. J. Epidemiol. 154:944–950. Dijkstra, R. 1982. The occurrence of Listeria monocytogenes in surface water of canals and lakes, in ditches of one big polder and in the effluents and canals of a

533 sewage treatment plant. Zentralbl. Bakteriol. Hyg. Abt. 1 Orig. B 176:202–205.   90. Disson, O., S. Grayo, E. Huillet, G. Nikitas, F. LangaVives, O. Dussurget, M. Ragon, A. Le Monnier, C. Babinet, P. Cossart, and M. Lecuit. 2008. Conjugated action of two species-specific invasion proteins for fetoplacental listeriosis. Nature 455:1114–1118.   91. Disson, O., G. Nikitas, S. Grayo, O. Dussurget, P. Cossart, and M. Lecuit. 2009. Modeling human listeriosis in natural and genetically engineered animals. Nat. Protocols 4:799–810.   92. Domann, E., M. Leimeister-Wachter, W. Goebel, and T. Chakraborty. 1991. Molecular cloning, sequencing, and identification of a metalloprotease gene from Listeria monocytogenes that is species specific and physically linked to the listeriolysin gene. Infect. Immun. 59:65–72.   93. Domann, E., J. Wehland, M. Rohde, S. Pistor, M. Hartl, W. Goebel, M. Leimeister-Wachter, M. Wuenscher, and T. Chakraborty. 1992. A novel bacterial virulence gene in Listeria monocytogenes required for host cell microfilament interaction with homology to the proline-rich region of vinculin. EMBO J. 11:1981–1990.   94. Domenech, E., J. A. Amoros, S. Martorell, and I. Escriche. 2012. Safety assessment of smoked fish related to Listeria monocytogenes prevalence using risk management metrics. Food Control 25:233–238.   95. Dominguez, C., I. Gomez, and J. Zumalacarregui. 2001. Prevalence and contamination levels of Listeria monocytogenes in smoked fish and pate sold in Spain. J. Food Prot. 64:2075–2077.   96. Donnelly, C. W., and D. G. Nyachuba. 2007. Conventional methods to detect and isolate Listeria monocytogenes, p. 215–256. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL.   97. Doumith, M., C. Cazalet, N. Simoes, L. Frangeul, C. Jacquet, F. Kunst, P. Martin, P. Cossart, P. Glaser, and C. Buchrieser. 2004. New aspects regarding evolution and virulence of Listeria monocytogenes revealed by comparative genomics and DNA arrays. Infect. Immun. 72:1072–1083.   98. Doumith, M., C. Jacquet, P. Gerner-Smidt, L. M. Graves, S. Loncarevic, T. Mathisen, A. Morvan, C. Salcedo, M. Torpdahl, J. A. Vazquez, and P. Martin. 2005. Multicenter validation of a multiplex PCR assay for differentiating the major Listeria monocytogenes serovars 1/2a, 1/2b, 1/2c, and 4b: toward an international standard. J. Food Prot. 68:2648–2650.   99. Doyle, M. 1988. Effect of environmental and processing conditions on Listeria monocytogenes. Food Technol. 42:169–171. 100. Doyle, M. P., K. A. Glass, J. T. Berry, G. A. Garcia, D. J. Pollard, and R. D. Schultz. 1987. Survival of Listeria monocytogenes in milk during high-temperature, short-time pasteurization. Appl. Environ. Microbiol. 53:1433–1438. 101. Dramsi, S., I. Biswas, L. Braun, E. Maguin, P. Mastroenni, and P. Cossart. 1995. Entry into hepatocytes requires expression of the inlB gene product. Mol. Microbiol. 16:251–261.

534 102. Dramsi, S., F. Bourdichon, D. Cabanes, M. Lecuit, H. Fsihi, and P. Cossart. 2004. FbpA, a novel multifunctional Listeria monocytogenes virulence factor. Mol. Microbiol. 53:639–649. 103. Dramsi, S., and P. Cossart. 2002. Listeriolysin O: a genuine cytolysin optimized for an intracellular parasite. J. Cell Biol. 156:943–946. 104. Dussurget, O., D. Cabanes, P. Dehoux, M. Lecuit, C. Buchrieser, P. Glaser, and P. Cossart. 2002. Listeria monocytogenes bile salt hydrolase is a PrfA-regulated virulence factor involved in the intestinal and hepatic phases of listeriosis. Mol. Microbiol. 45:1095–1106. 105. Dussurget, O., E. Dumas, C. Archambaud, I. Chafsey, C. Chambon, M. Hebraud, and P. Cossart. 2005. Listeria monocytogenes ferritin protects against multiple stresses and is required for virulence. FEMS Microbiol. Lett. 250:253–261. 106. Dussurget, O., J. Pizarro-Cerda, and P. Cossart. 2004. Molecular determinants of Listeria monocytogenes virulence. Annu. Rev. Microbiol. 58:587–610. 107. Elliott, E. L., and J. E. Kvenberg. 2000. Risk assessment used to evaluate the U.S. position on Listeria monocytogenes in seafood. Int. J. Food Microbiol. 62:253–260. 108. Endrikat, S., D. Gallagher, R. Pouillet, H. H. Quesenberry, et al. 2010. A comparative risk assessment for Listeria monocytogenes in prepackaged versus retail sliced deli meat. J. Food Prot. 73:612–619. 109. Ericsson, H., A. Eklow, M. L. Danielsson-Tham, S. Loncarevic, L. O. Mentzing, I. Persson, H. Unnerstad, and W. Tham. 1997. An outbreak of listeriosis suspected to have been caused by rainbow trout. J. Clin. Microbiol. 35:2904–2907. 110. Evans, M. R., B. Swaminathan, L. M. Graves, E. Altermann, T. R. Klaenhammer, R. C. Fink, S. Kernodle, and S. Kathariou. 2004. Genetic markers unique to Listeria monocytogenes serotype 4b differentiate epidemic clone II (hot dog outbreak strains) from other lineages. Appl. Environ. Microbiol. 70:2383–2390. 111. Farber, J., E. Coates, N. Beausoleil, and J. Fournier. 1991. Feeding trials of Listeria monocytogenes with a nonhuman primate model. J. Clin. Microbiol. 29:2606–2608. 112. Farber, J. M. 2000. Present situation in Canada regarding Listeria monocytogenes and ready-to-eat seafood products. Int. J. Food Microbiol. 62:247–251. 113. Farber, J. M., E. M. Daley, M. T. MacKie, and B. Limerick. 2000. A small outbreak of listeriosis potentially linked to the consumption of imitation crab meat. Lett. Appl. Microbiol. 31:100–104. 114. Farber, J. M., and P. I. Peterkin. 2007. Incidence and behavior of Listeria monocytogenes in meat products, p. 503–570. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 115. Ferreira, A., D. Sue, C. P. O’Byrne, and K. J. Boor. 2003. Role of Listeria monocytogenes sigma(B) in survival of lethal acidic conditions and in the acquired acid tolerance response. Appl. Environ. Microbiol. 69:2692–2698.

Foodborne Pathogenic Bacteria 116. Fleming, D., S. Cochi, K. MacDonald, J. Brondum, P. Hayes, B. Plikaytis, M. Holmes, A. Audurier, C. Broome, and A. Reingold. 1985. Pasteurized milk as a vehicle of infection in an outbreak of listeriosis. N. Engl. J. Med. 312:404–407. 117. Forster, B. M., A. P. Bitar, E. R. Slepkov, K. J. Kota, H. Sondermann, and H. Marquis. 2011. The metalloprotease of Listeria monocytogenes is regulated by pH. J. Bacteriol. 193:5090–5097. 118. Freitag, N., L. Rong, and D. Portnoy. 1993. Regulation of the prfA transcriptional activator of Listeria monocytogenes: multiple promoter elements contribute to intracellular growth and cell-to-cell spread. Infect. Immun. 61:2537–2544. 119. Freitag, N. E. 2000. Genetic tools for use with Listeria monocytogenes, p. 488–498. In V. A. Fischetti, R. P. Novick, J. J. Ferretti, D. A. Portnoy, and J. I. Rood (ed.), Gram-Positive Pathogens. ASM Press, Washington, DC. 120. Freitag, N. E., and K. E. Jacobs. 1999. Examination of Listeria monocytogenes intracellular gene expression by using the green fluorescent protein of Aequorea victoria. Infect. Immun. 67:1844–1852. 121. Fretz, R., J. Pichler, U. Sagel, P. Much, W. Ruppitsch, A. T. Pietzka, A. Stöger, S. Huhulescu, S. Heuberger, G. Appl, D. Werber, K. Stark, R. Prager, A. Flieger, R. Karpíšková, G. Pfaff, and F. Allerberger. 2010.  Update: Multinational listeriosis outbreak due to “Quargel”, a sour milk curd cheese, caused by two different L. monocytogenes serotype 1/2a strains, 2009-2010. Euro. Surveill. 15:19543. 122. Frye, C., and C. W. Donnelly. 2005. Comprehensive survey of pasteurized fluid milk produced in the United States reveals a low prevalence of Listeria monocytogenes. J. Food Prot. 68:973–979. 123. Frye, D. M., R. Zweig, J. Sturgeon, M. Tormey, M. LeCavalier, I. Lee, L. Lawani, and L. Mascola. 2002. An outbreak of febrile gastroenteritis associated with delicatessen meat contaminated with Listeria monocytogenes. Clin. Infect. Dis. 35:943–949. 124. Fugett, E. B., D. Schoonmaker-Bopp, N. B. Dumas, J. Corby, and M. Wiedmann. 2007. Pulsed-field gel electrophoresis (PFGE) analysis of temporally matched Listeria monocytogenes isolated from human clinical cases, foods, ruminant farms, and urban and natural environments reveals source-associated as well as widely distributed PFGE types. J. Clin. Microbiol. 45:865–873. 125. Gahan, C. G., J. O’Mahony, and C. Hill. 2001. Characterization of the groESL operon in Listeria monocytogenes: utilization of two reporter systems (gfp and hly) for evaluating in vivo expression. Infect. Immun. 69:3924–3932. 126. Gaillard, J., P. Berche, C. Frehel, E. Gouin, and P. Cossart. 1991. Entry of L. monocytogenes into cells is mediated by internalin, a repeat protein reminiscent of surface antigens from Gram-positive cocci. Cell 65:1127–1141. 127. Gaillard, J. L., P. Berche, and P. Sansonetti. 1986. Transposon mutagenesis as a tool to study the role of hemolysin in the virulence of Listeria monocytogenes. Infect. Immun. 52:50–55.

20.  Listeria monocytogenes 128. Gaillot, O., E. Pellegrini, S. Bregenholt, S. Nair, and P. Berche. 2000. The ClpP serine protease is essential for the intracellular parasitism and virulence of Listeria monocytogenes. Mol. Microbiol. 35:1286–1294. 129. Garner, M. R., B. L. Njaa, M. Wiedmann, and K. J. Boor. 2006. Sigma B contributes to Listeria monocytogenes gastrointestinal infection but not to systemic spread in the guinea pig infection model. Infect. Immun. 74:876–886. 130. Gasanov, U., D. Hughes, and P. M. Hansbro. 2005. Methods for the isolation and identification of Listeria spp. and Listeria monocytogenes: a review. FEMS Microbiol. Rev. 29:851–875. 131. Gaulin, C., D. Ramsay, L. Ringuette, and J. Ismail. 2003. First documented outbreak of Listeria monocytogenes in Quebec, 2002. Can. Commun. Dis. Rep. 29:181–186. 132. Gaulin, C., and D. Ramsay. 2010. Rapport d’investigation et d’intervention à la suite de l’éclosion d’infections à Listeria monocytogenes pulsovar 93 liée à la consommation de fromages québecois, 2008. Cited from Health Canada (2011) Policy on Listeria monocytogenes in Ready-to-Eat Foods. http://www.hc-sc.gc. ca/fn-an/legislation/pol/policy_listeria_monocytogenes_ 2011-eng.php. Accessed 31 March 2012. 133. Gedde, M., D. Higgins, L. Tilney, and D. Portnoy. 2000. Role of listeriolysin O in cell-to-cell spread of Listeria monocytogenes. Infect. Immun. 68:999–1003. 134. Geese, M., J. J. Loureiro, J. E. Bear, J. Wehland, F. B. Gertler, and A. S. Sechi. 2002. Contribution of Ena/ VASP proteins to intracellular motility of Listeria requires phosphorylation and proline-rich core but not F-actin binding or multimerization. Mol. Biol. Cell 13:2383–2396. 135. Geginat, G., M. Lalic, M. Kretschmar, W. Goebel, H. Hof, D. Palm, and A. Bubert. 1998. Th1 cells specific for a secreted protein of Listeria monocytogenes are protective in vivo. J. Immunol. 160:6046–6055. 136. Geginat, G., T. Nichterlein, M. Kretschmar, S. Schenk, H. Hof, M. Lalic-Multhaler, W. Goebel, and A. Bubert. 1999. Enhancement of the Listeria monocytogenes p60specific CD4 and CD8 T cell memory by nonpathogenic Listeria innocua. J. Immunol. 162:4781–4789. 137. Genigeorgis, C., D. Dutulescu, and J. Fernandez Garayzabal. 1989. Prevalence of Listeria spp. in poultry meat at the supermarket and slaughterhouse level. J. Food Prot. 52:618–624. 138. Gerner-Smidt, P., S. Ethelberg, P. Schiellerup, J. J. Christensen, J. Engberg, V. Fussing, A. Jensen, C. Jensen, A. M. Petersen, and B. G. Bruun. 2005. Invasive listeriosis in Denmark 1994-2003: a review of 299 cases with special emphasis on risk factors for mortality. Clin. Microbiol. Infect. 11:618–624. 139. Gilmour, M. W., M. Graham, G. Van Domselaar, S. Tyler, H. Kent, K. M. Trout-Yakel, O. Larios, V. Allen, B. Lee, and C. Nadon. 2010. High-throughput genome sequencing of two Listeria monocytogenes clinical isolates during a large foodborne outbreak. BMC Genomics 11:120–134. 140. Glaser, P., L. Frangeul, C. Buchrieser, C. Rusniok, A. Amend, F. Baquero, P. Berche, H. Bloecker, P. Brandt,

535

141.

142.

143. 144.

145. 146.

147.

148.

149.

150.

T. Chakraborty, A. Charbit, F. Chetouani, E. Couve, A. de Daruvar, P. Dehoux, E. Domann, G. DominguezBernal, E. Duchaud, L. Durant, O. Dussurget, K. D. Entian, H. Fsihi, F. G. Portillo, P. Garrido, L. Gautier, W. Goebel, N. Gomez-Lopez, T. Hain, J. Hauf, D. Jackson, L. M. Jones, U. Kaerst, J. Kreft, M. Kuhn, F. Kunst, G. Kurapkat, E. Madueno, A. Maitournam, J. M. Vicente, E. Ng, H. Nedjari, G. Nordsiek, S. Novella, B. de Pablos, J. C. Perez-Diaz, R. Purcell, B. Remmel, M. Rose, T. Schlueter, N. Simoes, A. Tierrez, J. A. Vazquez-Boland, H. Voss, J. Wehland, and P. Cossart. 2001. Comparative genomics of Listeria species. Science 294:849–852. Glass, K. A., and M. P. Doyle. 1989. Fate of Listeria monocytogenes in processed meat products during refrigerated storage. Appl. Environ. Microbiol. 55: 1565–1569. Goetz, M., A. Bubert, G. Wang, I. Chico-Calero, J. A. Vazquez-Boland, M. Beck, J. Slaghuis, A. A. Szalay, and W. Goebel. 2001. Microinjection and growth of bacteria in the cytosol of mammalian host cells. Proc. Natl. Acad. Sci. USA 98:12221–12226. Goldfine, H., and S. J. Wadsworth. 2002. Macrophage intracellular signaling induced by Listeria monocytogenes. Microbes Infect. 4:1335–1343. Goldfine, H., S. J. Wadsworth, and N. C. Johnston. 2000. Activation of host phospholipases C and D in macrophages after infection with Listeria monocytogenes. Infect. Immun. 68:5735–5741. Gombas, D. E., Y. Chen, R. S. Clavero, and V. N. Scott. 2003. Survey of Listeria monocytogenes in ready-to-eat foods. J. Food Prot. 66:559–569. Gonzalez-Zorn, B., G. Dominguez-Bernal, M. Suarez, M. T. Ripio, Y. Vega, S. Novella, and J. A. VazquezBoland. 1999. The smcL gene of Listeria ivanovii encodes a sphingomyelinase C that mediates bacterial escape from the phagocytic vacuole. Mol. Microbiol. 33:510–523. Gottlieb, S. L., E. C. Newbern, P. M. Griffin, L. M. Graves, R. M. Hoekstra, N. L. Baker, S. B. Hunter, K. G. Holt, F. Ramsey, M. Head, P. Levine, G. Johnson, D. Schoonmaker-Bopp, V. Reddy, L. Kornstein, M. Gerwel, J. Nsubuga, L. Edwards, S. Stonecipher, S. Hurd, D. Austin, M. A. Jefferson, S. D. Young, K. Hise, E. D. Chernak, and J. Sobel. 2006. Multistate outbreak of listeriosis linked to turkey deli meat and subsequent changes in US regulatory policy. Clin. Infect. Dis. 42:29–36. Gouin, E., C. Egile, P. Dehoux, V. Villiers, J. Adams, F. Gertler, R. Li, and P. Cossart. 2004. The RickA protein of Rickettsia conorii activates the Arp2/3 complex. Nature 427:457–461. Gouin, E., H. Gantelet, C. Egile, I. Lasa, H. Ohayon, V. Villiers, P. Gounon, P. J. Sansonetti, and P. Cossart. 1999. A comparative study of the actin-based motility of the pathogenic bacteria Listeria monocytogenes, Shigella flexneri and Rickettsia conorii. J. Cell Sci. 112:1697–1708. Gouin, E., M. D. Welch, and P. Cossart. 2005. Actinbased motility of intracellular pathogens. Curr. Opin. Microbiol. 8:35–45.

536 151. Goulet, V., H. de Valk, O. Pierre, F. Stainer, J. Rocourt, V. Vaillant, C. Jacquet, and J. C. Desenclos. 2001. Effect of prevention measures on incidence of human listeriosis, France, 1987–1997. Emerg. Infect. Dis. 7:983–989. 152. Goulet, V., C. Jacquet, V. Vaillant, I. Rebiere, E. Mouret, E. Lorente, F. Steiner, and J. Rocourt. 1995. Listeriosis from consumption of raw milk cheese. Lancet 345:1581–1582. 153. Goulet, V., C. Hedberg, A. Le Monnier, and H. de Valk. 2008. Increasing incidence of listeriosis in France and other European countries. Emerg. Infect. Dis. 14:734–740. 154. Goulet, V. 1995. Investigation en cas d’épidémie de listériose. Méd. Mal. Infect. 25:184–190. 155. Grau, F., and P. Vanderlinde. 1992. Occurrence, numbers and growth of Listeria monocytogenes on some vacuum-packaged processed meats. J. Food Prot. 55: 4–7. 156. Graves, L. M., S. B. Hunter, A. R. Ong, D. SchoonmakerBopp, K. Hise, L. Kornstein, W. E. DeWitt, P. S. Hayes, E. Dunne, P. Mead, and B. Swaminathan. 2005. Microbiological aspects of the investigation that traced the 1998 outbreak. J. Clin. Microbiol. 43:2350–2355. 157. Graves, L. M., and B. Swaminathan. 2005. PulseNet’s step-by-step laboratory protocol for molecular subtyping of Listeria monocytogenes by macrorestriction and pulsed-field gel electrophoresis, p. 57–72. In C. Adley (ed.), Food-Borne Pathogens, vol. 21. Humana Press, Totowa, NJ. 158. Graves, L. M., B. Swaminathan, and S. Hunter. 2007. Subtyping Listeria monocytogenes, p. 283–304. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 159. Graves, L. M., L. O. Helsel, A. G. Steigerwalt, R. E. Morey, M. I. Daneshvar, S. E.Roof, R. H. Orsi, E. D. Fortes, S. R. Milillo, H. C. den Bakker, M. Weidmann, B. Swaminathan, and B. D. Saunders. 2010. Listeria marthii sp. nov., isolated from the natural environment, Finger Lakes Natural Forest. Int. J. Syst. Evol. Microbiol. 60:1280–1288. 160. Gray, M. J., R. N. Zadoks, E. D. Fortes, B. Dogan, S. Cai, Y. Chen, V. N. Scott, D. E. Gombas, K. J. Boor, and M. Wiedmann. 2004. Listeria monocytogenes isolates from foods and humans form distinct but overlapping populations. Appl. Environ. Microbiol. 70:5833–5841. 161. Greiffenberg, L., W. Goebel, K. S. Kim, J. Daniels, and M. Kuhn. 2000. Interaction of Listeria monocytogenes with human brain microvascular endothelial cells: an electron microscopic study. Infect. Immun. 68:3275–3279. 162. Grif, K., G. Patscheider, M. P. Dierich, and F. Allerberger. 2003. Incidence of fecal carriage of Listeria monocytogenes in three healthy volunteers: a one-year prospective stool survey. Eur. J. Clin. Microbiol. Infect. Dis. 22:16–20. 163. Grundling, A., M. D. Gonzalez, and D. E. Higgins. 2003. Requirement of the Listeria monocytogenes broad-range phospholipase PC-PLC during Infection of human epithelial cells. J. Bacteriol. 185:6295–6307.

Foodborne Pathogenic Bacteria 164. Guenich, H., H. Muller, A. Schrettenbrunner, and H. Seeliger. 1985. The occurrence of different Listeria species in municipal waste water. Zentralbl. Bakteriol. Hyg. Abt. 1 Orig. Reihe B 181:563–565. 165. Gutekunst, K. A., L. Pine, E. White, S. Kathariou, and G. M. Carlone. 1992. A filamentous-like mutant of Listeria monocytogenes with reduced expression of a 60-kilodalton extracellular protein invades and grows in 3T6 and Caco-2 cells. Can. J. Microbiol. 38:843–851. 166. Hanawa, T., M. Fukuda, H. Kawakami, H. Hirano, S. Kamiya, and T. Yamamoto. 1999. The Listeria monocytogenes DnaK chaperone is required for stress tolerance and efficient phagocytosis with macrophages. Cell Stress Chaperones 4:118–128. 167. Hanawa, T., T. Yamamoto, and S. Kamiya. 1995. Listeria monocytogenes can grow in macrophages without the aid of proteins induced by environmental stresses. Infect. Immun. 63:4595–4599. 168. Harty, J. T., and E. G. Pamer. 1995. CD8 T lymphocytes specific for the secreted p60 antigen protect against Listeria monocytogenes infection. J. Immunol. 154:4642–4650. 169. Health Canada. 2011. Policy on Listeria monocytogenes on ready-to-eat foods. http://hc-sc.gc.ca/fn-an/­legislation/ pol/policy_listeria_monocytogenes_2011-eng.php. Accessed 11 February 2012. 170. Hebraud, M., and J. Guzzo. 2000. The main cold shock protein of Listeria monocytogenes belongs to the family of ferritin-like proteins. FEMS Microbiol. Lett. 190:29–34. 171. Hess, J., I. Gentschev, G. Szalay, C. Ladel, A. Bubert, W. Goebel, and S. H. Kaufmann. 1995. Listeria monocytogenes p60 supports host cell invasion by and in vivo survival of attenuated Salmonella typhimurium. Infect. Immun. 63:2047–2053. 172. Hitchens, A. D., and R. C. Whiting. 2001. Food-borne Listeria monocytogenes risk assessment. Food Additives Contam.18:1108–1117. 173. Ho, J. L., K. N. Shands, G. Friedland, P. Eckind, and D. W. Fraser. 1986. An outbreak of type 4b Listeria monocytogenes infection involving patients from eight Boston hospitals. Arch. Intern. Med. 146:520–524. 174. Hodgson, D. A. 2000. Generalized transduction of serotype 1/2 and serotype 4b strains of Listeria monocytogenes. Mol. Microbiol. 35:312–323. 175. Hoffman, A. D., K. L. Gall, D. M. Norton, and M. Wiedmann. 2003. Listeria monocytogenes contamination patterns for the smoked fish processing environment and for raw fish. J. Food Prot. 66:52–60. 176. Hu, Y., K. Gall, H. Alphina, R. Ivanek, Y.T. Grohn, and M. Wiedmann. 2006. Daily variability of Listeria contamination patterns in a cold-smoked salmon processing operation. J. Food Prot. 69:2123–2133. 177. Huss, H. H., A. Reilly, and P. K. Ben Embarek. 2000. Prevention and control of safety hazards in cold smoked salmon production. Food Control 11:149–156. 178. Husu, J., S. Sivela, and A. Rauramaa. 1990. Prevalence of Listeria species as related to chemical quality of farmensiled grass. Grass Forage Sci. 45:309–314.

20.  Listeria monocytogenes 179. Iannelli, V. 2010. Listeriosis—Listeria and the Texas celery recall. http://pediatrics.about.com/b/2010/10/21/ listeriosis-listeria-and-the-texas-celery-recall.htm. Accessed 12 February 2011. 180. Inoue, S., A. Nakama, Y. Arai, Y. Kokubo, T. Maruyama, A. Saito, T. Yoshida, M. Terao, S. Yamamoto, and S. Kumagai. 2000. Prevalence and contamination levels of Listeria monocytogenes in retail foods in Japan. Int. J. Food Microbiol. 59:73–77. 181. Jacobs, T., A. Darji, N. Frahm, M. Rohde, J. Wehland, T. Chakraborty, and S. Weiss. 1998. Listeriolysin O: cholesterol inhibits cytolysis but not binding to cellular membranes. Mol. Microbiol. 28:1081–1089. 182. Jacquet, C., B. Catimel, R. Brosch, C. Buchrieser, P. Dehaumont, V. Goulet, V. Lepoutre, P. Veit, and J. Rocourt. 1995. Investigations related to the epidemic strain involved in the French listeriosis outbreak in 1992. Appl. Environ. Microbiol. 61:2242–2246. 183. Jacquet, C., M. Doumith, J. I. Gordon, P. M. Martin, P. Cossart, and M. Lecuit. 2004. A molecular marker for evaluating the pathogenic potential of foodborne Listeria monocytogenes. J. Infect. Dis. 189:2094–2100. 184. Jacquet, C., E. Gouin, D. Jeannel, P. Cossart, and J. Rocourt. 2002. Expression of ActA, Ami, InlB, and listeriolysin O in Listeria monocytogenes of human and food origin. Appl. Environ. Microbiol. 68:616–622. 185. Jacquet, C., C. Saint-Cloment, F. Brouille, B. Catimel, and J. Rocourt. 1998. La listériose humaine en France en 1997. Données du Centre National de Référence des Listeria. Bull. Epidémiol. Hebdomadaire 33: 142–143. 186. Jeffers, G. T., J. L. Bruce, P. L. McDonough, J. Scarlett, K. J. Boor, and M. Wiedmann. 2001. Comparative genetic characterization of Listeria monocytogenes isolates from human and animal listeriosis cases. Microbiology 147:1095–1104. 187. Jensen, A. 1993. Excretion of Listeria monocytogenes in faeces after listeriosis: rate, quantity and duration. Med. Microbiol. Lett. 2:176–182. 188. Jensen, A., W. Frederiksen, and P. Gerner-Smidt. 1994. Risk factors for listeriosis in Denmark, 1989–1990. Scand. J. Infect. Dis. 26:171–178. 189. Jeong, D., and J. Frank. 1994. Growth of Listeria monocytogenes at 10°C in biofilms with microorganisms isolated from meat and dairy processing environments. J. Food Prot. 57:576–586. 190. Jinneman, K. C., M. M. Wekell, and M. W. Eklund. 2007. Incidence and behavior of Listeria monocytogenes in fish and seafood, p. 617–653. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 191. Johansson, J., P. Mandin, A. Renzoni, C. Chiaruttini, M. Springer, and P. Cossart. 2002. An RNA thermosensor controls expression of virulence genes in Listeria monocytogenes. Cell 110:551–561. 192. Jonquieres, R., H. Bierne, F. Fiedler, P. Gounon, and P. Cossart. 1999. Interaction between the protein InlB of Listeria monocytogenes and lipoteichoic acid: a novel

537

193.

194.

195.

196.

197. 198.

199.

200.

201.

202.

203.

204.

205.

mechanism of protein association at the surface of Gram-positive bacteria. Mol. Microbiol. 34:902–914. Jonquieres, R., H. Bierne, J. Mengaud, and P. Cossart. 1998. The inlA gene of Listeria monocytogenes LO28 harbors a nonsense mutation resulting in release of internalin. Infect. Immun. 66:3420–3422. Jonquieres, R., J. Pizarro-Cerda, and P. Cossart. 2001. Synergy between the N- and C-terminal domains of InlB for efficient invasion of non-phagocytic cells by Listeria monocytogenes. Mol. Microbiol. 42:955–965. Kallipolitis, B. H., and H. Ingmer. 2001. Listeria monocytogenes response regulators important for stress tolerance and pathogenesis. FEMS Microbiol. Lett. 204:111–115. Kathariou, S., P. Metz, H. Hof, and W. Goebel. 1987. Tn916-induced mutations in the hemolysin determinant affecting virulence of Listeria monocytogenes. J. Bacteriol. 169:1291–1297. Kathariou, S. 2002. Listeria monocytogenes virulence and pathogenicity, a food safety persepctive. J. Food Prot. 65:1811–1829. Keeney, K., L. Colosi, W. Weber, and M. O’Riordan. 2009. Generation of branched-chain fatty acids through lipoate-dependent metabolism facilitates intracellular growth of Listeria monocytogenes. J. Bacteriol. 191:2187–2196. Keskinen, L. A., E. C. D. Todd, and E. T. Ryser. 2008. Transfer of surface-dried Listeria monocytogenes from stainless steel knife blades to roast turkey breast. J. Food Prot. 71:176–181. Keskinen, L. A., E. C. D. Todd, and E. T. Ryser. 2008. Impact of bacterial stress and biofilm-forming ability on transfer of surface-dried Listeria monocytogenes during slicing of delicatessen meats. Int. J. Food Microbiol. 127:298–304. Khelef, N., M. Lecuit, C. Buchrieser, D. Cabanes, O. Dussurget, and P. Cossart. 2004. Listeria monocytogenes and the Genus Listeria. http://141.150.157.117:8 080/prokPUB.index.htm. In M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt (ed.), The Prokaryotes: An Evolving Electronic Resource for the Microbiological Community, 3rd ed. Springer, New York, NY. Klarsfeld, A., P. Goossens, and P. Cossart. 1994. Five Listeria monocytogenes preferentially expressed in mammalian cells: plcA, purH, purD, pyrE, and an arginine ABC transporter gene, arpJ. Mol. Microbiol. 13:585–597. Ko, R., and L. T. Smith. 1999. Identification of an ATPdriven, osmoregulated glycine betaine transport system in Listeria monocytogenes. Appl. Environ. Microbiol. 65:4040–4048. Kocks, C., E. Gouin, M. Tabouret, P. Berche, H. Ohayon, and P. Cossart. 1992. Listeria monocytogenes induced actin assembly requires the ActA gene product, a surface protein. Cell 68:521–531. Kolb-Maurer, A., S. Pilgrim, E. Kampgen, A. D. McLellan, E. B. Brocker, W. Goebel, and I. Gentschev. 2001. Antibodies against listerial protein 60 act as an opsonin for phagocytosis of Listeria monocytogenes by human dendritic cells. Infect. Immun. 69:3100–3109.

538 206. Kornacki, J. L., and J. B. Gurtler. 2007. Listeria in food processing facilities, p. 683–766. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 207. Koutsoumanis, K. P., and J. N. Sofos. 2005. Effect of inoculum size on the combined temperature, pH and aw limits for growth of Listeria monocytogenes. Int. J. Food Microbiol. 104:83–91. 208. Kuhn, M., and W. Goebel. 1989. Identification of an extracellular protein of Listeria monocytogenes possibly involved in intracellular uptake by mammalian cells. Infect. Immun. 57:55–61. 209. Lado, B. H., and A. E. Yousef. 2007. Characteristics of Listeria monocytogenes important to food processors, p. 157–213. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 210. Lappi, V. R., J. Thimothe, K. K. Nightingale, K. Gall, V. N. Scott, and M. Wiedmann. 2004. Longitudinal studies on Listeria in smoked fish plants: impact of intervention strategies on contamination patterns. J. Food Prot. 67:2500–2514. 211. Larsen, A., and B. Norrung. 1993. Inhibition of Listeria monocytogenes by bavaricin A, a bacteriocin produced by Lactobacillus bavaricus Ml401. Lett. Appl. Microbiol. 17:132–134. 212. Lasa, I., V. David, E. Gouin, J. B. Marchand, and P. Cossart. 1995. The amino-terminal part of ActA is ­critical for the actin-based motility of Listeria monocytogenes; the central proline-rich region acts as a stimulator. Mol. Microbiol. 18:425–436. 213. Lasa, I., E. Gouin, M. Goethals, K. Vancompernolle, V. David, J. Vandekerckhove, and P. Cossart. 1997. Identification of two regions in the N-terminal domain of ActA involved in the actin comet tail formation by Listeria monocytogenes. EMBO J. 16:1531–1540. 214. Lauer, P., M. Y. Chow, M. J. Loessner, D. A. Portnoy, and R. Calendar. 2002. Construction, characterization, and use of two Listeria monocytogenes sitespecific phage integration vectors. J. Bacteriol. 184: 4177–4186. 215. Lecuit, M., and C. Cossart. 2001. Listeria monocytogenes, p. 1437–1462. In S. Sussman (ed.), Molecular Medical Microbiology, vol. 2. Academic Press, London, United Kingdom. 216. Lecuit, M., and P. Cossart. 2002. Genetically-modifiedanimal models for human infections: the Listeria paradigm. Trends Mol. Med. 8:537–542. 217. Lecuit, M., S. Dramsi, C. Gottardi, M. Fedor-Chaiken, B. Gumbiner, and P. Cossart. 1999. A single amino acid in E-cadherin responsible for host specificity towards the human pathogen Listeria monocytogenes. EMBO J. 18:3956–3963. 218. Lecuit, M., D. M. Nelson, S. D. Smith, H. Khun, M. Huerre, M. C. Vacher-Lavenu, J. I. Gordon, and P. Cossart. 2004. Targeting and crossing of the human maternofetal barrier by Listeria monocytogenes: role of internalin interaction with trophoblast E-cadherin. Proc. Natl. Acad. Sci. USA 101:6152–6157.

Foodborne Pathogenic Bacteria 219. Lecuit, M., H. Ohayon, L. Braun, J. Mengaud, and P. Cossart. 1997. Internalin of Listeria monocytogenes with an intact leucine-rich repeat region is sufficient to promote internalization. Infect. Immun. 65:5309–5319. 220. Lecuit, M., S. Vandormael-Pournin, J. Lefort, M. Huerre, P. Gounon, C. Dupuy, C. Babinet, and P. Cossart. 2001. A transgenic model for listeriosis: role of internalin in crossing the intestinal barrier. Science 292:1722–1725. 221. Leimeister-Wachter, M., E. Domann, and T. Chakraborty. 1992. The expression of virulence genes in Listeria monocytogenes is thermoregulated. J. Bacteriol. 174:947–952. 222. Leimeister-Wachter, M., C. Haffner, E. Domann, W. Goebel, and T. Chakraborty. 1990. Identification of a gene that positively regulates expression of listeriolysin, the major virulence factor of Listeria monocytogenes. Proc. Natl. Acad. Sci. USA 87:8336–8340. 223. Lenz, L. L., S. Mohammadi, A. Geissler, and D. A. Portnoy. 2003. SecA2-dependent secretion of autolytic enzymes promotes Listeria monocytogenes pathogenesis. Proc. Natl. Acad. Sci. USA 100:12432–12437. 224. Li, Z., J. Dai, H. Zheng, B. Liu, and M. Caudill. 2002. An integrated view of the roles and mechanisms of heat shock protein gp96-peptide complex in eliciting immune response. Front. Biosci. 7:731–751. 225. Lianou, A., and J. N. Sofos. 2007. A review of the incidence and transmission of Listeria monocytogenes in ready-to-eat products in retail and food service environments. J. Food Prot. 70:2172–2198. 226. Lin, C. M., K. Takeuchi, L. Zhang, C. B. Dohm, J. D. Meyer, P. A. Hall, and M. P. Doyle. 2006. Cross-­contamination between processing equipment and deli meats by Listeria monocytogenes. J. Food Prot. 69:71–79. 227. Lindqvist, R., and A. Westeroo. 1999. Quantitative risk assessment for Listeria monocytogenes in smoked or gravad salmon or rainbow trout in Sweden. Int. J. Food Microbiol. 58:181–196. 228. Linnan, M., L. Mascola, X. Lou, V. Goulet, S. May, C. Salminen, D. Hird, M. Yonekura, P. Hayes, R. Weaver, A. Audurier, B. Plikaytis, S. Fannin, A. Kleks, and C. Broome. 1988. Epidemic listeriosis associated with Mexican-style cheese. N. Engl. J. Med. 319:823–828. 229. Liu, D., M. L. Lawrence, L. Gorski, R. E. Mandrell, A. J. Ainsworth, and F. W. Austin. 2006. Listeria monocytogenes serotype 4b strains belonging to lineages I and III possess distinct molecular features. J. Clin. Microbiol. 44:214–217. 230. Lomonaco, S., S. J. Knabel, A. Dalmasso, T. Civera, and M. T. Bottero. 2011. Novel multiplex single nucleotide polymorphism-based method for identifying epidemic clones of Listeria monocytogenes. Appl. Environ. Microbiol. 77:6290–6294. 231. Lorber, B. 1997. Listeriosis. Clin. Infect. Dis. 24:1–9. 232. Low, J. C., F. Wright, J. McLauchlin, and W. Donachie. 1993. Serotyping and distribution of Listeria isolates from cases of ovine listeriosis. Vet. Rec. 133:165–166. 233. Luchansky, J. B., P. M. Muriana, and T. R. Klaenhammer. 1988. Application of electroporation for transfer of plasmid DNA to Lactobacillus, Lactococcus, Leuconostoc, Listeria,

20.  Listeria monocytogenes

234.

235.

236.

237.

238.

239.

240.

241.

242.

243.

244.

245.

246.

Pediococcus, Bacillus, Staphylococcus, Enterococcus and Propionibacterium. Mol. Microbiol. 2:637–646. Lunden, J. M., T. J. Autio, and H. J. Korkeala. 2002. Transfer of persistent Listeria monocytogenes contamination between food-processing plants associated with a dicing machine. J. Food Prot. 65:1129–1133. Lunden, J. M., T. J. Autio, A. M. Sjoberg, and H. J. Korkeala. 2003. Persistent and nonpersistent Listeria monocytogenes contamination in meat and poultry processing plants. J. Food Prot. 66:2062–2069. Lyytikainen, O., T. Autio, R. Maijala, P. Ruutu, T. Honkanen-Buzalski, M. Miettinen, M. Hatakka, J. Mikkola, V. J. Anttila, T. Johansson, L. Rantala, T. Aalto, H. Korkeala, and A. Siitonen. 2000. An outbreak of Listeria monocytogenes serotype 3a infections from butter in Finland. J. Infect. Dis. 181:1838–1841. MacDonald, P. D., R. E. Whitwam, J. D. Boggs, J. N. MacCormack, K. L. Anderson, J. W. Reardon, J. R. Saah, L. M. Graves, S. B. Hunter, and J. Sobel. 2005. Outbreak of listeriosis among Mexican immigrants as a result of consumption of illicitly produced Mexicanstyle cheese. Clin. Infect. Dis. 40:677–682. Makino, S. I., K. Kawamoto, K. Takeshi, Y. Okada, M. Yamasaki, S. Yamamoto, and S. Igimi. 2005. An outbreak of food-borne listeriosis due to cheese in Japan, during 2001. Int. J. Food Microbiol. 104:189–196. Mandin, P., H. Fsihi, O. Dussurget, M. Vergassola, E. Milohanic, A. Toledo-Arana, I. Lasa, J. Johansson, and P. Cossart. 2005. VirR, a response regulator critical for Listeria monocytogenes virulence. Mol. Microbiol. 57:1367–1380. Manitoba Health. 2000. Faulty food product identified as source of listeria foodborne outbreak. http://www. gov.mb.ca/health/publichealth/cmoh/docs/listeria.pdf. Accessed 31 March 2012. Marino, M., M. Banerjee, R. Jonquieres, P. Cossart, and P. Ghosh. 2002. GW domains of the Listeria monocytogenes invasion protein InlB are SH3-like and mediate binding to host ligands. EMBO J. 21:5623–5634. Marino, M., L. Braun, P. Cossart, and P. Ghosh. 1999. Structure of the InlB leucine-rich repeats, a domain that triggers host cell invasion by the bacterial pathogen L. monocytogenes. Mol. Cell 4:1063–1072. Marquis, H., H. Bouwer, D. Hinrichs, and D. Portnoy. 1993. Intracytoplasmic growth and virulence of Listeria monocytogenes auxotrophic mutants. Infect. Immun. 61:3756–3760. Marquis, H., V. Doshi, and D. A. Portnoy. 1995. The broad-range phospholipase C and a metalloprotease mediate listeriolysin O-independent escape of Listeria monocytogenes from a primary vacuole in human epithelial cells. Infect. Immun. 63:4531–4534. Mascola, L., F. Sorvillo, V. Goulet, B. Hall, R. Weaver, and M. Linnan. 1992. Fecal carriage of Listeria monocytogenes—observations during a community wide, ­commonsource outbreak. Clin. Infect. Dis. 15:557–558. Matthieu, F., M. Michel, A. Lebrihi, and G. Lefebvre. 1994. Effect of the bacteriocin carnocin CP5 and of the producing strain Carnobacterium piscicola CP5 on the

539

247. 248. 249. 250.

251. 252.

253.

254.

255.

256.

257.

258. 259.

260.

viability of Listeria monocytogenes ATCC 15313 in salt solution, broth and skimmed milk, at various incubation temperatures. Int. J. Food Microbiol. 22:155–172. McLauchlin, J. 1997. The discovery of Listeria. PHLS Microbiol. Dig. 14:76–78. McLauchlin, J., S. Hall, S. Velani, and R. Gilbert. 1991. Human listeriosis and pate—a possible association. Br. Med. J. 303:773–775. McLauchlin, J., and P. Hoffman. 1989. Neonatal crossinfection from Listeria monocytogenes. Commun. Dis. Rep. 6:3–4. McLauchlin, J., R. T. Mitchell, W. J. Smerdon, and K. Jewell. 2004. Listeria monocytogenes and listeriosis: a review of hazard characterisation for use in microbiological risk assessment of foods. Int. J. Food Microbiol. 92:15–33. McLaughlan, A. M., and S. J. Foster. 1998. Molecular characterization of an autolytic amidase of Listeria monocytogenes EGD. Microbiology 144:1359–1367. Mead, P. S., E. F. Dunne, L. Graves, M. Wiedmann, M. Patrick, S. Hunter, E. Salehi, F. Mostashari, A. Craig, P. Mshar, T. Bannerman, B. D. Sauders, P. Hayes, W. Dewitt, P. Sparling, P. Griffin, D. Morse, L. Slutsker, and B. Swaminathan. 2005. Nationwide outbreak of listeriosis due to contaminated meat. Epidemiol. Infect. 134:744–751. Meinersmann, R. J., R. W. Phillips, M. Wiedmann, and M. E. Berrang. 2004. Multilocus sequence typing of Listeria monocytogenes by use of hypervariable genes reveals clonal and recombination histories of three lineages. Appl. Environ. Microbiol. 70:2193–2203. Mengaud, J., C. Braun-Breton, and P. Cossart. 1991. Identification of phosphatidylinositol-specific phospholipase C activity in Listeria monocytogenes: a novel type of virulence factor? Mol. Microbiol. 5:367–372. Mengaud, J., J. Chenevert, C. Geoffroy, J. L. Gaillard, and P. Cossart. 1987. Identification of the structural gene encoding the SH-activated hemolysin of Listeria monocytogenes: listeriolysin O is homologous to streptolysin O and pneumolysin. Infect. Immun. 55:3225–3227. Mengaud, J., S. Dramsi, E. Gouin, J. Vasquez-Boland, G. Milon, and P. Cossart. 1991. Pleiotropic control of Listeria monocytogenes virulence factors by a gene which is autoregulated. Mol. Microbiol. 5:2273–2283. Mengaud, J., H. Ohayon, P. Gounon, R. M. Mege, and P. Cossart. 1996. E-cadherin is the receptor for internalin, a surface protein required for entry of L. monocytogenes into epithelial cells. Cell 84:923–932. Miettinen, H., and G. Wirtanen. 2005. Prevalence and location of Listeria monocytogenes in farmed rainbow trout. Int. J. Food Microbiol. 104:135–143. Miettinen, M. K., A. Siitonen, P. Heiskanen, H. Haajanen, K. J. Bjorkroth, and H. J. Korkeala. 1999. Molecular epidemiology of an outbreak of febrile gastroenteritis caused by Listeria monocytogenes in cold-smoked rainbow trout. J. Clin. Microbiol. 37:2358–2360. Milohanic, E., P. Glaser, J. Y. Coppee, L. Frangeul, Y. Vega, J. A. Vazquez-Boland, F. Kunst, P. Cossart, and C. Buchrieser. 2003. Transcriptome analysis of

Foodborne Pathogenic Bacteria

540 Listeria monocytogenes identifies three groups of genes differently regulated by PrfA. Mol. Microbiol. 47:1613–1625. 261. Milohanic, E., R. Jonquieres, P. Cossart, P. Berche, and J. L. Gaillard. 2001. The autolysin Ami contributes to the adhesion of Listeria monocytogenes to eukaryotic cells via its cell wall anchor. Mol. Microbiol. 39:1212–1224. 262. Mitchell, D. L. 1991. A case cluster of listeriosis in Tasmania. Commun. Dis. Intell. 15:427. 263. Modi, K., M. Chikindas, and T. Montville. 2000. Sensitivity of nisin-resistant Listeria monocytogenes to heat and the synergistic action of heat and nisin. Lett. Appl. Microbiol. 30:249–253. 264. Munder, A., A. Zelmer, A. Schmiedl, K. E. Dittmar, M. Rohde, M. Dorsch, K. Otto, H. J. Hedrich, B. Tummler, S. Weiss, and T. Tschernig. 2005. Murine pulmonary infection with Listeria monocytogenes: differential susceptibility of BALB/c, C57BL/6 and DBA/2 mice. Microbes Infect. 7:600–611. 265. Nadon, C. A., B. M. Bowen, M. Wiedmann, and K. J. Boor. 2002. Sigma B contributes to PrfA-mediated virulence in Listeria monocytogenes. Infect. Immun. 70:3948–3952. 266. Nair, S., C. Frehel, L. Nguyen, V. Escuyer, and P. Berche. 1999. ClpE, a novel member of the HSP100 family, is involved in cell division and virulence of Listeria monocytogenes. Mol. Microbiol. 31:185–196. 267. Neumann, K., E. Eppler, L. Filgueira, P. Groscurth, E. Gasal, A. Schaffner, G. Schoedon, and M. Schneemann. 2003. Listeria species escape from the phagosomes of interleukin-4-deactivated human macrophages independent of listeriolysin. Immunol. Cell. Biol. 81:431–439. 268. Niebuhr, K., F. Ebel, R. Frank, M. Reinhard, E. Domann, U. D. Carl, U. Walter, F. B. Gertler, J. Wehland, and T. Chakraborty. 1997. A novel proline-rich motif present in ActA of Listeria monocytogenes and cytoskeletal proteins is the ligand for the EVH1 domain, a protein module present in the Ena/VASP family. EMBO J. 16:5433–5444. 269. Nikitas, G., C. Deschamps, O. Disson, T. Niault, P. Cossart, and M. Lecuit. 2011. Transcytosis of Listeria monocytogenes across the intestinal barrier upon specific targeting of goblet cell accessible E-cadherin. J. Exp. Med. 208:2263–2277. 270. Norrung, B., J. K. Andersen, and J. Schlundt. 1999. Incidence and control of Listeria monocytogenes in foods in Denmark. Int. J. Food Microbiol. 53:195–203. 271. Norton, D. M., J. M. Scarlett, K. Horton, D. Sue, J. Thimothe, K. J. Boor, and M. Wiedmann. 2001. Characterization and pathogenic potential of Listeria monocytogenes isolates from the smoked fish industry. Appl. Environ. Microbiol. 67:646–653. 272. Norton, D. M., and C. R. Braden. 2007. Foodborne listeriosis, p. 305–356. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 273. Olier, M., D. Garmyn, S. Rousseaux, J. P. Lemaitre, P. Piveteau, and J. Guzzo. 2005. Truncated internalin A

274.

275.

276.

277.

278.

279.

280.

281. 282.

283.

284.

and asymptomatic Listeria monocytogenes carriage: in vivo investigation by allelic exchange. Infect. Immun. 73:644–648. Olier, M., F. Pierre, S. Rousseaux, J. P. Lemaitre, A. Rousset, P. Piveteau, and J. Guzzo. 2003. Expression of truncated Internalin A is involved in impaired internalization of some Listeria monocytogenes isolates carried asymptomatically by humans. Infect. Immun. 71:1217–1224. Oliver, S. P., B. M. Jayarao, and R. A. Almeida. 2005. Foodborne pathogens in milk and the dairy farm environment: food safety and public health implications. Foodborne Pathog. Dis. 2:115–129. Olsen, S. J., M. Patrick, S. B. Hunter, V. Reddy, L. Kornstein, W. R. MacKenzie, K. Lane, S. Bidol, G. A. Stoltman, D. M. Frye, I. Lee, S. Hurd, T. F. Jones, T. N. LaPorte, W. Dewitt, L. Graves, M. Wiedmann, D. J. Schoonmaker-Bopp, A. J. Huang, C. Vincent, A. Bugenhagen, J. Corby, E. R. Carloni, M. E. Holcomb, R. F. Woron, S. M. Zansky, G. Dowdle, F. Smith, S. Ahrabi-Fard, A. R. Ong, N. Tucker, N. A. Hynes, and P. Mead. 2005. Multistate outbreak of Listeria monocytogenes infection linked to delicatessen turkey meat. Clin. Infect. Dis. 40:962–967. OzFoodNet Working Group. 2010. Monitoring the incidence and causes of diseases potentially transmitted by food in Australia: annual report of the OzFoodNet Network, 2009. Commun. Dis. Intell. 34: 396–426. Pagotto, F., L.-K. Ng, C. Clark, J. Farber, and the Canadian Public Health Laboratory Network. 2006. Canadian Listeriosis Reference Service. Foodborne Pathog. Dis. 3:132–137. Painter, J., and L. Slutsker. 2007. Listeriosis in humans, p. 85–109. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. Pandiripally, V. K., D. G. Westbrook, G. R. Sunki, and A. K. Bhunia. 1999. Surface protein p104 is involved in adhesion of Listeria monocytogenes to human intestinal cell line, Caco-2. J. Med. Microbiol. 48:117–124. Peerschke, E. I., K. B. Reid, and B. Ghebrehiwet. 1994. Identification of a novel 33-kDa C1q-binding site on human blood platelets. J. Immunol. 152:5896–5901. Pentecost, M., J. Kumaran, P. Ghosh, and M. R. Amieva. 2010. Listeria monocytogenes internalin B activates junctional endocytosis to accelerate intestinal invasion. PLoS Pathog. 6(5):e1000900. Pilgrim, S., A. Kolb-Maurer, I. Gentschev, W. Goebel, and M. Kuhn. 2003. Deletion of the gene encoding p60 in Listeria monocytogenes leads to abnormal cell division and loss of actin-based motility. Infect. Immun. 71:3473–3484. Pinner, R., A. Schuchat, B. Swaminathan, P. Hayes, K. Deaver, R. Weaver, B. Plikaytis, M. Reeves, C. Broome, and J. Wenger. 1992. Role of foods in sporadic listeriosis. 2. Microbiologic and epidemiologic investigation. JAMA 267:2046–2050.

20.  Listeria monocytogenes 285. Pizarro-Cerda, J., M. Lecuit, and P. Cossart. 2002. Measuring and analysing invasion of mammalian cells by bacterial pathogens: the Listeria monocytogenes system. Methods Microbiol. 31:161–177. 286. Portnoy, D. A., P. S. Jacks, and D. J. Hinrichs. 1988. Role of hemolysin for the intracellular growth of Listeria monocytogenes. J. Exp. Med. 167:1459–1471. 287. Pouillet, R., N. Miconnet, A. L. Afchain, M. L. DelignetteMuller, A. Beafort, L. Rosso, J.-B. Denis, and M. Cornu. 2007. Quantitative risk assessment of Listeria monocytogenes in French cold-smoked salmon. I. Quantitative exposure assessment. Risk Anal. 27:683–700. 288. Pouillet, R., V. Goulet, M. L. Delignette-Muller, and M. Cornu. 2009. Quantitative risk assessment of Listeria monocytogenes in French cold-smoked salmon. II. Risk characterization. Risk Anal. 29:806–819. 289. Pradhan, A. K., R. Ivanek, Y. T. Gröhn, I. Geornaras, J. N. Sofos, and M. Wiedmann. 2009. Quantitative risk assessment for Listeria monocytogenes in selected categories of deli meats: impact of lactate and diacetate on listeriosis cases and deaths. J. Food Prot. 72:978–989. 290. Pradhan, A. K., R. Ivanek, Y. T. Gröhn, R. Bukowski, I. Geornaras, J. N. Sofos, and M. Wiedmann. 2010. Quantitative risk assessment of listeriosis-associated deaths due to Listeria monocytogenes contamination of deli meats originating from manufacture and retail. J. Food Prot. 73:620–630. 291. Proctor, M. E., R. Brosch, J. W. Mellen, L. A. Garrett, C. W. Kaspar, and J. B. Luchansky. 1995. Use of pulsedfield gel electrophoresis to link sporadic cases of invasive listeriosis with recalled chocolate milk. Appl. Environ. Microbiol. 61:3177–3179. 292. Promadej, N., F. Fiedler, P. Cossart, S. Dramsi, and S. Kathariou. 1999. Cell wall teichoic acid glycosylation in Listeria monocytogenes serotype 4b requires gtcA, a novel, serogroup-specific gene. J. Bacteriol. 181:418–425. 293. Public Health Agency of Canada. 2010. Update to 2008 Listeria monocytogenes case numbers. Public Health Agency of Canada, Ottawa, Ontario, Canada. 294. Pucciarelli, M. G., E. Calvo, C. Sabet, H. Bierne, P. Cossart, and F. Garcia-Del Portillo. 2005. Identification of substrates of the Listeria monocytogenes sortases A and B by a non-gel proteomic analysis. Proteomics 5:4808–4817. 295. Racz, P., K. Tenner, and E. Mero. 1972. Experimental Listeria enteritis. I. An electron microscopic study of the epithelial phase in experimental listeria infection. Lab. Investig. 26:694–700. 296. Rantsiou, K., A. Greppi, M. Garosi, A. Acquadro, M. Mataragas, and L. Cocolin. 2012. Strain dependent expression of stress response and virulence genes of Listeria monocytogenes in meat juices as determined by microarray. Int. J. Food Microbiol. 152:116–122. 297. Renzoni, A., A. Klarsfeld, S. Dramsi, and P. Cossart. 1997. Evidence that PrfA, the pleiotropic activator of virulence genes in Listeria monocytogenes, can be present but inactive. Infect. Immun. 65:1515–1518.

541 298. Repp, H., Z. Pamukci, A. Koschinski, E. Domann, A. Darji, J. Birringer, D. Brockmeier, T. Chakraborty, and F. Dreyer. 2002. Listeriolysin of Listeria monocytogenes forms Ca2+-permeable pores leading to intracellular Ca2+ oscillations. Cell. Microbiol. 4:483–491. 299. Riedo, F., R. Pinner, M. Tosca, M. Carter, L. Graves, M. Reeves, R. Weaver, B. Plikaytis, and C. Broome. 1994. A point-source foodborne listeriosis outbreak: documented incubation period and possible mild illness. J. Infect. Dis. 170:693–696. 300. Ringus, D. L., R. A. Ivy, M. Wiedmann, and K. J. Boor. 2012. Salt stress-induced transcription of sB-and CtsRregulated genes in persistent and non-persistent Listeria monocytogenes strains from food processing plants. Foodborne Pathog. Dis. 9:198–206. 301. Robbins, J., A. Barth, H. Marquis, E. de Hostos, W. Nelson, and J. Theriot. 1999. Listeria monocytogenes exploits normal host cell processes to spread from cell to cell. J. Cell Biol. 146:1333–1349. 302. Rocourt, J., and C. Buchreiser. 2007. The genus Listeria and Listeria monocytogenes: phylogenetic position, taxonomy, and identification, p. 1–20. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis, and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 303. Rocourt, J., and C. Jacquet. 1994. Epidémiologie des infections humaines à Listeria monocytogenes en 1994: certitudes et interrogations. Ann. Inst. Pasteur 5:168–174. 304. Rocourt, J., C. Jacquet, and A. Reilly. 2000. Epidemiology of human listeriosis and seafoods. Int. J. Food Microbiol. 62:197–209. 305. Rooney, J., and P. Sutherland. 2001. Independent review of food safety in the Hunter area health service. HAHS Review of Food Safety. Personal Communication. Cited from Health Canada (2011) Policy on Listeria monocytogenes in Ready-to-Eat Foods. http://www.hc-sc.gc. ca/fn-an/legislation/pol/policy_listeria_monocytogenes_ 2011-eng.php. Accessed 31 March 2012. 306. Ross, T., S. Rasmussen, A. Fazil, G. Paoli, and J. Sumner. 2009. Quantitative risk assessment of Listeria monocytogenes in ready-to-eat meats in Australia. Int. J. Food Microbiol. 131:128–137. 307. Rouquette, C., C. de Chastellier, S. Nair, and P. Berche. 1998. The ClpC ATPase of Listeria monocytogenes is a general stress protein required for virulence and promoting early bacterial escape from the phagosome of macrophages. Mol. Microbiol. 27:1235–1245. 308. Ruhland, G. J., M. Hellwig, G. Wanner, and F. Fiedler. 1993. Cell-surface location of Listeria-specific protein p60—detection of Listeria cells by indirect immunofluorescence. J. Gen. Microbiol. 139:609–616. 309. Ryser, E. T. 2007. Incidence and behavior of Listeria monocytogenes in unfermented dairy products, p. 357– 403. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 310. Ryser, E. T. 2007. Incidence and behavior of Listeria monocytogenes in fermented dairy products, p. 405–501.

Foodborne Pathogenic Bacteria

542 In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 311. Ryser, E. T. 2007. Incidence and behavior of Listeria monocytogenes in poultry and egg products, p. 571–615. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 312. Ryser, E. T., and C. W. Donnelly. 2001. Listeria, p. 343– 356. In Compendium of Methods for the Microbiological Examination of Foods, 4th ed. American Public Health Association, Washington, DC. 313. Salamina, G., E. Dalle Donne, A. Niccolini, G. Poda, D. Cesaroni, M. Bucci, R. Fini, M. Maldini, A. Schuchat, B. Swaminathan, W. Bibb, J. Rocourt, N. Binkin, and S. Salmaso. 1996. A foodborne outbreak of gastroenteritis involving Listeria monocytogenes. Epidemiol. Infect. 117:429–436. 314. Salova, N., N. N. Filatov, E. V. Sizykh, A. N. Gerasimov, and L. A. Riapis. 2005. Pulse-electrotypes of Listeria monocytogenes strains, isolated in Moscow. Zh. Mikrobiol. Epidemiol. Immunobiol. July-August:19-22. (In Russian.) 315. Samadpour, M., M. W. Barbour, T. Nguyen, T. M. Cao, F. Buck, G. A. Depavia, E. Mazengia, P. Yang, D. Alfi, M. Lopes, and J. D. Stopforth. 2006. Incidence of enterohemorrhagic Escherichia coli, Escherichia coli O157, Salmonella, and Listeria monocytogenes in retail fresh ground beef, sprouts, and mushrooms. J. Food Prot. 69:441–443. 316. Sammarco, M. L., M. Vitullow, M. Tamburro, M. Pontello, and G. Ripabelli. 2011. Amplified fragment length polymorphism analysis of Listeria monocytogenes by ExperionTM automated microfluidic electrophoresis. J. Microbiol. Methods 87:119–124. 317. Sauders, B. D., and M. Wiedmann. 2007. Ecology of Listeria species and L. monocytogenes in the natural environment, p. 21–53. In E. T. Ryser and E. H. Marth (ed.), Listeria, Listeriosis and Food Safety, 3rd ed. CRC Press, Boca Raton, FL. 318. Sauders, B. D., D. Pettit, B. Currie, P. Suits, A. Evans, K. Stellrecht, D. M. Dryja, D. Slate, and M. Wiedmann. 2005. Low prevalence of Listeria monocytogenes in human stool. J. Food Prot. 68:178–181. 319. Sauders, B. D., M. Z. Durzak, E. Fortes, K. Windham, Y. Schukken, A. J. Lembo, Jr., B. Akey, K. K. Nightengale, and M. Wiedmann. 2006. Molecular characterization of Listeria monocytogenes from natural and urban environments. J. Food Prot. 69:93–105. 320. Saxbe, W. B., Jr. 1972. Listeria monocytogenes and Queen Anne. Pediatrics 49:97–101. 321. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M.-C. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne pathogens acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. 322. Schaferkordt, S., and T. Chakraborty. 1995. Vector plasmid for insertional mutagenesis and directional cloning in Listeria spp. BioTechniques 19:720–722. 323. Schauer, K., J. Stolz, S. Scherer, and T. M. Fuchs. 2009. Both thiamine uptake and biosynthesis of thiamine

324.

325. 326. 327.

328. 329.

330.

331.

332.

333. 334. 335.

336.

337. 338.

precursors are required for intracellular replication of Listeria monocytogenes. J. Bacteriol. 191:2218–2227. Schlech, W. F., III, P. Lavigne, R. Bortolussi, A. Allen, E. Haldane, A. Wort, A. Hightower, S. Johnson, S. King, E. Nicholls, and C. V. Broome. 1983. Epidemic listeriosis—evidence for transmission by food. N. Engl. J. Med. 308:203–206. Schlech, W. F., III. 2000. Foodborne listeriosis. Clin. Infect. Dis. 31:770–775. Schneewind, O., P. Model, and V. A. Fischetti. 1992. Sorting of protein A to the staphylococcal cell wall. Cell 70:267–281. Schnupf, P., D. A. Portnoy, and A. L. Decatur. 2006. Phosphorylation, ubiquitination and degradation of listeriolysin O in mammalian cells: role of the PEST-like sequence. Cell. Microbiol. 8:353–364. Schnupf, P., and D. A. Portnoy. 2007. Listeriolysin O: a phagosome-specific lysin. Microbes Infect. 9:1176–1187. Schubert, W. D., G. Gobel, M. Diepholz, A. Darji, D. Kloer, T. Hain, T. Chakraborty, J. Wehland, E. Domann, and D. W. Heinz. 2001. Internalins from the human pathogen Listeria monocytogenes combine three distinct folds into a contiguous internalin domain. J. Mol. Biol. 312:783–794. Schuchat, A., K. Deaver, P. Hayes, L. Graves, L. Mascola, and J. Wenger. 1993. Gastrointestinal carriage of Listeria monocytogenes in household contacts of patients with listeriosis. J. Infect. Dis. 167:1261–1262. Schuchat, A., K. Deaver, J. Wenger, B. Plikaytis, L. Mascola, R. Pinner, A. Reingold, and C. Broome. 1992. Role of foods in sporadic listeriosis. 1. Case-control study of dietary risk factors. JAMA 267:2041–2045. Schuchat, A., C. Lizano, C. Broome, B. Swaminathan, C. Kim, and K. Win. 1991. Outbreak of neonatal listeriosis associated with mineral oil. Pediatr. Infect. Dis. J. 10:183–189. Schuchat, A., B. Swaminathan, and C. V. Broome. 1991. Epidemiology of human listeriosis. Clin. Microbiol. Rev. 4:169–183. Schuerch, D. W., E. M. Wilson-Kubalek, and R. K. Tweten. 2005. Molecular basis of listeriolysin O pH dependence. Proc. Natl. Acad. Sci. USA 102:12537–12542. Schwartz, B., D. Hexter, C. Broome, A. Hightower, R. Hischorn, J. Porter, P. Hayes, W. Bibb, B. Lorber, and D. Faris. 1989. Investigation of an outbreak of listeriosis: new hypotheses for the etiology of epidemic Listeria monocytogenes infections. J. Infect. Dis. 159: 680–685. Seeliger, H., and D. Jones. 1986. Listeria, p. 1235–1245. In J. G. Holt et al. (ed.), Bergey’s Manual of Systematic Bacteriology, vol. 2. Williams & Wilkins, Baltimore, MD. Seeliger, H. P. R., and K. Hohne. 1979. Serotyping of Listeria monocytogenes and related species. Methods Microbiol. 13:31–49. Senczek, D., R. Stephan, and F. Untermann. 2000. Pulsed-field gel electrophoresis (PFGE) typing of Listeria

20.  Listeria monocytogenes

339.

340. 341.

342.

343.

344.

345.

346.

347. 348.

349. 350.

351.

strains isolated from a meat processing plant over a 2year period. Int. J. Food Microbiol. 62:155–159. Shahamat, M., A. Seaman, and M. Woodbine. 1980. Survival of Listeria monocytogenes in high salt concentrations. Zentralbl. Bakteriol. Hyg. I. Abt. Orig. A 246:506–511. Shank, F., E. L. Elliott, I. K. Wachsmuth, and M. E. Losikoff. 1996. U.S. position on Listeria monocytogenes in foods. Food Control 7:229–234. Sheehan, B., A. Klarsfeld, R. Ebright, and P. Cossart. 1996. A single substitution in the putative helix-turnhelix motif of the pleiotropic activator PrfA attenuates Listeria monocytogenes virulence. Mol. Microbiol. 20:785–797. Shen, Y., M. Naujokas, M. Park, and K. Ireton. 2000. InIB-dependent internalization of Listeria is mediated by the Met receptor tyrosine kinase. Cell 103: 501–510. Sibelius, U., T. Chakraborty, B. Krogel, J. Wolf, F. Rose, R. Schmidt, J. Wehland, W. Seeger, and F. Grimminger. 1996. The listerial exotoxins listeriolysin and phosphatidylinositol-specific phospholipase C synergize to elicit endothelial cell phosphoinositide metabolism. J. Immunol. 157:4055–4060. Sim, J., D. Hood, L. Finnie, M. Wilson, C. Graham, M. Brett, and J. A. Hudson. 2002. Series of incidents of Listeria monocytogenes non-invasive febrile gastroenteritis involving ready-to-eat meats. Lett. Appl. Microbiol. 35:409–413. Skoble, J., D. A. Portnoy, and M. D. Welch. 2000. Three regions within ActA promote Arp2/3 complex-mediated actin nucleation and Listeria monocytogenes motility. J. Cell Biol. 150:527–538. Skogberg, K., J. Syrjanen, M. Jahkola, O. Renkonen, J. Paavonen, J. Ahonen, S. Kontiainen, P. Ruutu, and V. Valtonen. 1992. Clinical presentation and outcome of listeriosis in patients with and without immunosuppressive therapy. Clin. Infect. Dis. 14:815–821. Skovgaard, N., and B. Norrung. 1989. The incidence of Listeria spp. in faeces of Danish pigs and in minced pork meat. Int. J. Food Microbiol. 8:59–63. Sleator, R. D., C. G. Gahan, T. Abee, and C. Hill. 1999. Identification and disruption of BetL, a secondary glycine betaine transport system linked to the salt tolerance of Listeria monocytogenes LO28. Appl. Environ. Microbiol. 65:2078–2083. Sleator, R. D., C. G. Gahan, and C. Hill. 2003. A postgenomic appraisal of osmotolerance in Listeria monocytogenes. Appl. Environ. Microbiol. 69:1–9. Smith, G. A., H. Marquis, S. Jones, N. C. Johnston, D. A. Portnoy, and H. Goldfine. 1995. The two distinct phospholipases C of Listeria monocytogenes have overlapping roles in escape from a vacuole and cell-to-cell spread. Infect. Immun. 63:4231–4237. Smith, K., and P. Youngman. 1992. Use of a new integrational vector to investigate compartment-specific expression of the Bacillus subtilis spoIIM gene. Biochimie 74:705–711.

543 352. Smith, M. A., K. Takeuchi, R. E. Brackett, H. M. McClure, R. B. Raybourne, K. M. Williams, U. S. Babu, G. O. Ware, J. R. Broderson, and M. P. Doyle. 2003. Nonhuman primate model for Listeria monocytogenesinduced stillbirths. Infect. Immun. 71:1574–1579. 353. Sousa, S., D. Cabanes, C. Archambaud, F. Colland, E. Lemichez, M. Popoff, S. Boisson-Dupuis, E. Gouin, M. Lecuit, P. Legrain, and P. Cossart. 2005. ARHGAP10 is necessary for alpha-catenin recruitment at adherens junctions and for Listeria invasion. Nat. Cell. Biol. 7:954–960. 354. Suarez, M., B. Gonzalez-Zorn, Y. Vega, I. Chico-Calero, and J. A. Vazquez-Boland. 2001. A role for ActA in epithelial cell invasion by Listeria monocytogenes. Cell. Microbiol. 3:853–864. 355. Sue, D., D. Fink, M. Wiedmann, and K. J. Boor. 2004. sigmaB-dependent gene induction and expression in Listeria monocytogenes during osmotic and acid stress conditions simulating the intestinal environment. Microbiology 150:3843–3855. 356. Sullivan, M. A., R. E. Yasbin, and F. E. Young. 1984. New shuttle vectors for Bacillus subtilis and Escherichia coli which allow rapid detection of inserted fragments. Gene 29:21–26. 357. Sumner, S. S., T. M. Sandros, M. Harmon, V. N. Scott, and D. T. Bernard. 1991. Heat resistance of Salmonella typhimurium and Listeria monocytogenes in sucrose solutions of various water activities. J. Food Sci. 56:1741–1743. 358. Tappero, J., A. Schuchat, K. Deaver, L. Mascola, and J. Wenger. 1995. Reduction in the incidence of human listeriosis in the United States. Effectiveness of prevention efforts. JAMA 273:1118–1122. 359. Theriot, J. A., T. J. Mitchison, L. G. Tilney, and D. A. Portnoy. 1992. The rate of actin-based motility of intracellular Listeria monocytogenes equals the rate of actin polymerization. Nature 357:257–260. 360. Tilney, L. G., and D. A. Portnoy. 1989. Actin filaments and the growth, movement, and spread of the intracellular bacterial parasite, Listeria monocytogenes. J. Cell Biol. 109:1597–1608. 361. Tompkin, R. B. 2002. Control of Listeria monocytogenes in the food-processing environment. J. Food Prot. 65:709–725. 362. Trieu-Cuot, P., C. Carlier, C. Poyart-Salmeron, and P. Courvalin. 1991. Shuttle vectors containing a multiple cloning site and a lacZ alpha gene for conjugal transfer of DNA from Escherichia coli to gram-positive bacteria. Gene 102:99–104. 363. Tromp, S. O., H. Rugersberg, and E. Franz. 2010. Quantitative microbial risk assessment for Escherichia coli, Salmonella enterica, and Listeria monocytogenes in leafy green vegetables consumed at salad bars, based on supply chain logistics. J. Food Prot. 73:1830–1840. 364. U.S. Department of Agriculture, Food Safety and Inspection Service. 2004. Control of Listeria monocytogenes in ready-to-eat poultry and meat products. Fed. Regist. 69:70051–70053.

544 365. U.S. Department of Agriculture, Food Safety and Inspection Service. 2010. The FSIS microbiological testing program for ready-to-eat (RTE) meat and poultry products, 1990-2010. http://www.fsis.usda.gov/Science/ Micro_Testing_RTE/. Accessed 2 February 2012. 366. U.S. Food and Drug Administration. 2012. Information of the recalled Jensen Farms cantaloupes. http://www. fda.gov/Food/FoodSafety/CORENetwork/ucm272372. htm. Accessed 12 February 2012. 367. Utratna, M., I. Shaw, E. Starr, C. P. O’Bryne. 2011. Rapid, transient, and proportional activation of sB in response to osmotic stress in Listeria monocytogenes. Appl. Environ. Microbiol. 77:7841–7845. 368. Van Langendonck, N., E. Bottreau, S. Bailly, M. Tabouret, J. Marly, P. Pardon, and P. Velge. 1998. Tissue culture assays using Caco-2 cell line differentiate virulent from non-virulent Listeria monocytogenes strains. J. Appl. Microbiol. 85:337–346. 369. Van Renterghem, B., F. Huysman, R. Rygole, and W. Verstraete. 1991. Detection and prevalence of Listeria monocytogenes in the agricultural ecosystem. J. Appl. Bacteriol. 71:211–217. 370. Van Stelten, A. J. M. Simpson, T. J. Ward, and K. K. Nightingale. 2010. Revelation by single-nucleotide polymorphism genotyping that mutations leading to a premature stop codon in inlA are common among Listeria monocytogenes isolates from ready-to-eat foods but not human listeriosis cases. Appl. Environ. Microbiol. 76:2783–2790. 371. Vazquez-Boland, J. A., C. Kocks, S. Dramsi, H. Ohayon, C. Geoffroy, J. Mengaud, and P. Cossart. 1992. Nucleotide sequence of the lecithinase operon of Listeria monocytogenes and possible role of lecithinase in cell-to-cell spread. Infect. Immun. 60:219–230. 372. Vazquez-Boland, J. A., M. Kuhn, P. Berche, T. Chakraborty, G. Dominguez-Bernal, W. Goebel, B. Gonzalez-Zorn, J. Wehland, and J. Kreft. 2001. Listeria pathogenesis and molecular virulence determinants. Clin. Microbiol. Rev. 14:584–640. 373. Vermeiren, L., F. Devlieghere, and J. Debevere. 2006. Co-culture experiments demonstrate the usefulness of Lactobacillus sakei 10A to prolong the shelf-life of a model cooked ham. Int. J. Food Microbiol. 108: 68–77. 374. Vicente, M. F., F. Baquero, and J. C. Perez-Diaz. 1987. A protoplast transformation system for Listeria sp. Plasmid 18:89–92. 375. Volpe Sperry, K. E., S. Kathariou, J. S. Edwards, and L. A. Wolf. 2008. Multiple-locus variable-number tandemrepea­t analysis as a tool for subtyping Listeria monocytogenes strains. J. Clin. Microbiol. 46:1435–1450. 376. Vorst, K. L., E. C. D. Todd, and E. T. Ryser. 2006. Transfer of Listeria monocytogenes during mechanical slicing of turkey breast, bologna, and salami. J. Food Prot. 69:619–626. 377. Wadsworth, S. J., and H. Goldfine. 2002. Mobilization of protein kinase C in macrophages induced by Listeria monocytogenes affects its internalization and escape from the phagosome. Infect. Immun. 70:4650–4660.

Foodborne Pathogenic Bacteria 378. Walker, R., L. Jensen, H. Kinde, A. Alexander, and L. Owen. 1991. Environment survey for Listeria species in frozen milk plants in California. J. Food Prot. 54:178–182. 379. Ward, T. J., L. Gorski, M. K. Borucki, R. E. Mandrell, J. Hutchins, and K. Pupedis. 2004. Intraspecific phylogeny and lineage group identification based on the prfA virulence gene cluster of Listeria monocytogenes. J. Bacteriol. 186:4994–5002. 380. Ward, T. J., T. F. Ducey, T. Usgaard, K. A. Dunn, and J. P. Bielawski. 2008. Multilocus genotyping assays for single nucleotide polymorphism-based subtyping of Listeria monocytogenes isolates. Appl. Environ. Microbiol. 74:7629–7642. 381. Weis, J., and H. Seeliger. 1975. Incidence of Listeria monocytogenes in nature. Appl. Microbiol. 30:29–32. 382. Wemekamp-Kamphuis, H. H., J. A. Wouters, R. D. Sleator, C. G. Gahan, C. Hill, and T. Abee. 2002. Multiple deletions of the osmolyte transporters BetL, Gbu, and OpuC of Listeria monocytogenes affect virulence and growth at high osmolarity. Appl. Environ. Microbiol. 68:4710–4716. 383. Wesche, A. M., J. Gurtler, B. P. Marks, and E. T. Ryser. 2009. Inactivation of sublethally injured bacterial pathogens in foods. J. Food Prot. 72:1121–1138. 384. Wiedmann, M., J. L. Bruce, C. Keating, A. E. Johnson, P. L. McDonough, and C. A. Batt. 1997. Ribotypes and virulence gene polymorphisms suggest three distinct Listeria monocytogenes lineages with differences in pathogenic potential. Infect. Immun. 65:2707–2716. 385. Williams, D., E. A. Irvin, R. A. Chmielewski, J. F. Frank, and M. A. Smith. 2007. Dose response of Listeria monocytogenes after oral exposure in pregnant guinea pigs. J. Food Prot. 70:1122–1128. 386. Williams, D., J. Castleman, C.-C. Lee, B. Mote, and M. A. Smith. 2009. Risk of fetal mortality after exposure to Listeria monocytogenes based on dose-response data from pregnant guinea pigs and primates. Risk Anal. 29:1495–1505. 387. Williams, S. K., S. Roof, E.A. Boyle, D. Burson, H. Thippareddi, I. Geornaras, J.N. Sofos, M. Wiedmann, and K. Nightingdale. 2011. Molecular ecology of Listeria monocytogenes and other Listeria species in small and very small ready-to-eat meat processing plants. J. Food Prot. 74:63–77. 388. Wirth, R., F. Y. An, and D. B. Clewell. 1986. Highly efficient protoplast transformation system for Streptococcus faecalis and a new Escherichia coli-S. faecalis shuttle vector. J. Bacteriol. 165:831–836. 389. Wuenscher, M., S. Kohler, A. Bubert, U. Gerike, and W. Goebel. 1993. The iap gene of Listeria monocytogenes is essential for cell viability and its gene product, p60, has bacteriolytic activity. J. Bacteriol. 175:3491–3501. 390. Yildirim, S., W. Lin, A. D. Hitchins, L. A. Jaykus, E. Altermann, T. R. Klaenhammer, and S. Kathariou. 2004. Epidemic clone I-specific genetic markers in strains of Listeria monocytogenes serotype 4b from foods. Appl. Environ. Microbiol. 70:4158–4164.

20.  Listeria monocytogenes 391. Zhang, C., M. Zhang, J. Ju, J. Nietfeldt, J. Wise, P. M. Terry, M. Olson, S. D. Kachman, M. Wiedmann, M. Samadpour, and A. K. Benson. 2003. Genome diversification in phylogenetic lineages I and II of Listeria monocytogenes: identification of segments unique to lineage II populations. J. Bacteriol. 185:5573–5584. 392. Zhang, W., B. M. Jayarao, and S. J. Knabel. 2004. Multi-virulence-locus sequence typing of Listeria monocytogenes. Appl. Environ. Microbiol. 70:913–920.

545 393. Zheng, W., and S. Kathariou. 1997. Host-mediated modification of Sau3AI restriction in Listeria monocytogenes: prevalence in epidemic-associated strains. Appl. Environ. Microbiol. 63:3085–3089. 394. Zunabovic, M., K. J. Domig, I. Pichler, and W. Kneifel. 2012. Monitoring transmission routes of Listeria spp. in smoked salmon production with repetitive element sequenced-based PCR techniques. J. Food Prot. 75:504–511.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch21

Keun Seok Seo Gregory A. Bohach

Staphylococcus aureus

Staphylococcal food poisoning (SFP) is among the most prevalent causes of gastroenteritis worldwide. It results from ingestion of one or more preformed staphylococcal enterotoxins (SEs) in staphylococcus-contaminated food. The etiological agents of SFP are members of the genus Staphylococcus, predominantly Staphylococcus aureus. This form of food poisoning is considered an intoxication; it does not involve infection by, and growth of, the bacteria in the host. The association of staphylococci with foodborne illness was made more than a century ago. Barber, in 1914, was the first to implicate a toxin in SFP (9). He reported that repeated ingestion of contaminated milk produced symptoms typical of the illness. Barber cultured the milk, demonstrated the presence of a putative causative staphylococcal agent, and provided the first evidence that a soluble toxin was responsible for the disease. The next major advance in understanding SFP etiology was reported in 1930 by Dack et al. (33), who voluntarily consumed supernatants from cultures of “a yellow hemolytic staphylococcus” grown from contaminated sponge cake. Upon ingestion of the filtrates, Dack became ill with vomiting, abdominal cramps, and diarrhea. At that time, the only other foodborne toxin that had been recognized was botulinum toxin. However, staphylococcal toxin, which exerted an effect on the gastrointestinal tract, was the first true enterotoxin described. It was particularly unique in

21

comparison to botulinum toxin because its activity was “not entirely destroyed by heating even for 30 minutes at 100°C.” S. aureus has been extensively characterized. This bacterium produces a variety of extracellular products. Many of these, including the SEs, are virulence factors that have been implicated in diseases of humans and animals. As a group, the SEs elaborate a set of biological properties that enable staphylococci to cause at least two common human diseases, toxic shock syndrome (TSS) and SFP. This chapter will primarily address SFP; however, in regard to the SEs, there is significant overlap in the natural histories of both diseases. Hence, TSS also will be discussed in some sections in which this overlap is most relevant.

CHARACTERISTICS OF THE ORGANISM

Nomenclature, Characteristics, and Distribution of SE-Producing Staphylococci

The term “staphylococci” describes a group of small, spherical, gram-positive bacteria. Depending on the species and culture conditions, their cells have diameters ranging from approximately 0.5 to 1.5 mm. They are catalase-positive chemoorganotrophs with a DNA composition of 30 to 40 mol% guanine + cytosine content. Staphylococci have a typical gram-positive cell wall containing peptidoglycan and teichoic acids. Except

Keun Seok Seo and Gregory A. Bohach, Department of Basic Sciences, Mississippi State University, Mississippi State, MS 39762.

547

Foodborne Pathogenic Bacteria

548 for clinical isolates such as some community-acquired methicillin-resistant S. aureus strains (127) and strains exposed to antimicrobial therapy, most staphylococci are sensitive to β-lactams, tetracyclines, macrolides, lincosamides, novobiocin, and chloramphenicol but are resistant to polymyxin and polyene. Some differential characteristics of S. aureus and several other selected species of staphylococci are summarized in Table 21.1. There have been many useful schemes for classification of the staphylococci. According to Bergey’s Manual of Determinative Bacteriology (120), staphylococci are classified in the family Micrococcaceae. This family includes the genera Micrococcus, Staphy­ lococcus, Stomatococcus, and Planococcus. The genus Staphylococcus is further subdivided into 32 species and subspecies. Many of these are present in food as a result of human, animal, or environmental contamination. Several species of Staphylococcus, including both coagulase-negative and coagulase-positive isolates, can produce SEs. Although several species, including some coagulase-negative staphylococci, have the potential to cause gastroenteritis (17), nearly all cases of SFP are attributed to S. aureus. This is a reflection of the relatively high incidence of SE production by S. aureus in comparison to that of other staphylococcal species. Although the reason for this is unknown, the SEs are superantigens (SAgs) and therefore are potential immunomodulating agents (see below). Hence, SE production is proposed to provide a selective advantage to S. aureus, a species that is common to both humans and animals, the two most common sources of food contamination. Enterotoxigenic strains of staphylococci have been well characterized on the basis of a number of genotypic

and phenotypic characteristics. An extensive phage-typing system is available for S. aureus. Most SE-producing isolates belong to phage group I or III or are nontypeable. Although SE production by other phage groups is less common, it has been documented. Hajek and Marsalek (47) developed a classification scheme based largely on the animal host of origin. They were able to differentiate S. aureus into at least six biotypes. By far, SE production was most prevalent among human isolates within biotype A. SE production by other biotypes is rare except for biotype C bovine and ovine mastitis isolates. SEs may also be produced by Staphylococcus intermedius and S. hyicus (formerly S. aureus biotypes E and F, respectively) and coagulase-negative staphylococci, albeit considerably less frequently than by S. aureus. Currently known toxins in the SE family, with their physical and functional properties, are summarized in Table 21.2. Structure-function and mechanisms of pathogenicity of the SEs are discussed later in this chapter. This section will introduce the nomenclature and evolution of the SE family of toxins.

Classification Scheme Based on Antigenicity

Major advances in characterization of SEs were made approximately 2 decades after Dack and colleagues (33) associated SFP with an exotoxin. Bergdoll and coworkers were the first investigators to produce purified SE preparations and develop specific antisera. They and others, using purified or partially purified toxins, induced protective antibodies in several species of animals. Immunity was strain specific and did not provide protection against strains other than the one that was used to induce the initial immune response (13). It soon became apparent that S. aureus could

Table 21.1  General characteristics of selected Staphylococcus species Characteristic Coagulase Thermostable nuclease Clumping factor Yellow pigment Hemolytic activity Phosphatase Lysostaphin Hyaluronidase Mannitol fermentation Novobiocin resistance a

ND, not determined.

S. aureus

S. chromogenes

S. hyicus

S. intermedius

S. epidermidis

S. saprophyticus

+ +

− −

+ +

+ +

− +/−

− −

+ + + + Sensitive + +

− + − + Sensitive − +/−

− − − + Sensitive + −

+ − + + Sensitive − +/−

− − +/− +/− Slightly sensitive +/− −

− +/− − − NDa ND +/−

−

−

−

−

−

+

21.  Staphylococcus aureus

549

Table 21.2  Biochemical and functional properties of SEs SE or SEla

Molecular mass (kDa)

Zinc site(s)

Binding to MHC α/β chain

SEA

27.1

Cleft, domain 2

α/β

SED SEE SElJ SElP SElN SElO SEH SES SEB SEC1 SEC2 SEC3 SElU SEG SER SEI SElK SElL SElM SElQ SElX SET TSST-1

26.9 26.8 31.2 27.1 26.1 26.8 25.1 26.2 28.4 27.5 26.6 26.6 27.2 27.0 27.0 24.9 30.0 26.8 24.8 25.0 19.2 22.6 21.9

Cleft, domain 2 Domain 2 NDb ND ND ND Domain 2 ND − Cleft Cleft Cleft ND − ND ND ND ND ND ND ND ND −

α/β α/β ?/β ND ND ND ND/β ND α/− α/− α/− α/− ND α/− ND ND/β ND/β ND/β ND/β ND ND ND α/−

a b c

Human TCR interactionsc 1.1, 5.3, 6.3, 6.4, 6.9, 7.3, 7.4, 9, 16, 18, 21.3, 22.1, 23 1, 5.3, 6.9, 7.4, 8, 12 5.1, 6.3, 6.4, 8.1, 8.2, 13.1, 18, 21.6 ND ND 9 5, 7, 22 Vα10e 9, 16 1, 3.2, 6.4, 12, 14, 15, 17, 20 3.2, 5, 6.4, 6.9, 12, 13.2,15, 17, 20 5, 12, 13.1, 13.2, 14, 15, 17, 20 2, 3, 5, 12, 13.1, 13.2, 14, 17, 20 ND 3, 12, 13.6, 14 3, 11, 12, 13.2, 14 1, 5, 5.3, 6, 23 5.1, 5.2, 6.7 5, 7, 22 6, 8, 9, 18, 21.3 2, 5.1, 21.3 1, 6, 18, 21 ND 2, 8

Reference(s) 90 6, 69 90 147 76 63 63 106 101 34 34 34 34 78 63 100 63 103 63 63 63 145 101 30

The nomenclature used in this table follows the new nomenclature rules of the International Nomenclature Committee for Staphylococcal SAgs (INCSS) (80). ND, not determined. Numbers indicate Vb elements unless otherwise indicated for SEl.

produce multiple toxins with similar molecular weights and biological and physicochemical properties. Initially, differentiation between antigenic forms of SE was based on the observation that many food isolates produce one common antigenic type of toxin, tentatively designated the “F” type toxin. Most other enterotoxigenic strains, such as those from enteritis patients, also produced a second antigenic form, which was classified as “E” toxin. The discovery of additional isolates that did not conform to this pattern prompted adoption of an improved nomenclature system. A committee assembled in 1962 established an alphabetical nomenclature (25), which, with some modification (80), is still currently recommended. Accordingly, SEs are sequentially assigned a letter of the alphabet in the order of their discovery. The “F” and “E” type toxins were designated SEA and SEB, respectively. Between 1962 and 1972, three additional “classical” SE serotypes (SEC, SED, and SEE) were re-

ported (12). Initially, SEF was used in reference to an exotoxin commonly produced by isolates of S. aureus associated with TSS. This designation was dropped when the toxin was confirmed to lack emetic activity. SEF was retired from use in the SE nomenclature system and is now referred to as toxic shock syndrome toxin 1 (TSST-1) (16). Compared to the rate of identification of the classical SEs (SEA through SEE), modern genomic analysis has greatly increased the rate at which previously unrecognized SEs and related proteins are discovered. As a result, classification based solely on antigenic properties is not practical. Instead, nomenclature is now based predominantly on molecular relatedness of the primary sequences. Furthermore, it is not common to confirm the enterotoxigenic activities of putative SEs prior to publication. As a result, some toxins originally reported to be SEs were later found to lack emetic activity. To ensure an orderly

Foodborne Pathogenic Bacteria

550

Figure 21.1  Alignment of primary sequences of mature SEs and SEls in the current literature. Also shown are the consensus sequence (at the bottom) and dashes (–) to indicate gaps in the sequences made by alignment. Sequence alignment and output were conducted using the CLUSTAL W Program (136). doi:10.1128/9781555818463.ch21f1

assignment for newly identified confirmed or putative toxins, investigators should contact the International Nomenclature Committee for Staphylococcal SAgs (INCSS) for guidance prior to publication (80). The INCSS also recommends that proteins related to the

SEs but not confirmed to exhibit emetic activity in the monkey feeding assay (see below) be designated SE-like (SEl) until their enterotoxic activity can be confirmed. Primary sequences and other properties of currently reported SEs and SEls are summarized in

21.  Staphylococcus aureus Fig. 21.1 and Table 21.2 (6, 14, 32, 56, 68, 83, 101, 115, 122, 129, 135). The incidence of SE involvement in SFP appears to change with time. Prior to 1971, SEA was the predominant toxin identified in cases of SFP, followed in frequency by SED and SEC. SEB was only rarely associated with SFP. In some cases, SEA was also identified as the causative agent in combination with SEC or SED; SEB was only rarely associated with SFP (87). In a study examining more recent outbreaks of SFP from 1977 through 1981, SEA remained the most common toxin implicated (52). However, in contrast to previous surveys, SEB was the only other SE identified. Holmberg and Blake (52) suggested that the observed decrease in cases attributed to other SEs was due to improved conditions for the processing and storage of milk, which had been commonly contaminated by SECand SED-producing strains of S. aureus in the past. In a more recent report, SEE was described as a classical SE, least commonly associated with SFP (64). Of the newly described staphylococcal toxins, SEG, SEI, and other toxins in the egc operon (described below) (63) appear to be very common and widely distributed among staphylococcal human and animal isolates (125).

SE Antigenic Subtypes and Molecular Variants

Designation of SEs based on serological typing has been useful. However, sequence analysis and detailed immunological studies have produced some examples in which the antigenic characteristics of the proteins do not reflect their molecular or biological uniqueness. The best-documented examples are with SEC. It had been noted for some time that the SEC serological variant can be further divided into at least three subtypes (SEC1, SEC2, and SEC3) based on minor differences in immunological reactivity. However, within each subtype, significant sequence variability may occur. For example, the SECs produced by strains FRI-909 and FRI-913 were both designated SEC3 according to their immunological reactivity. However, it was later shown that the sequences of both toxins differ by nine residues (83). The SEC variants produced by bovine and ovine isolates of S. aureus have very similar sequences and are apparently indistinguishable from SEC1 in immunological assays. In contrast, they behave differently from SEC1 in biological assays. For example, although SEC-bovine differs from SEC1 by only three residues, the potencies of the two toxins differ by several orders of magnitude in lymphocyte proliferation assays (83). Examples of heterogeneity among both the classical SEs and newly described toxins are becoming more common with the rapid rate of sequence determinations.

551

Staphylococcal Genetics and Evolutionary Aspects of SE Production SEs Are Superantigens and Belong to a Large Pyrogenic Toxin Family

In discussing the genetics and evolution of the SEs, one must also consider other staphylococcal toxins, plus some toxins produced by other bacteria, especially group A streptococci. SEs are part of a large family of related toxins produced by S. aureus and Streptococcus pyogenes (18). This family of toxins has been termed the pyrogenic toxin (PT) family. Members of this family are grouped together based on shared biological and biochemical properties. The one feature common to all PTs, including SEs, is their unique ability to act as SAgs (84). SAgs are molecules that have the ability to stimulate an exceptionally high percentage of T cells. The mechanism by which this occurs distinguishes them from mitogens and conventional antigens (Ags). In regard to T-cell stimulation, SAgs are bifunctional molecules that interact with major histocompatibility complex class II (MHCII) molecules on Ag-presenting cells (APCs). Unlike the situation with conventional Ags, this interaction does not require processing and occurs outside the MHCII peptide-binding groove (Fig. 21.2). The MHCII-SAg complex

Figure 21.2  Interactions between APC and T cells facilitated by conventional antigens (Ags) and superantigens (SAgs). Following processing by the APC, conventional Ags are presented to highly specific T-cell receptors (TCR) in association with the Ag-binding groove of the MHCII molecule. SAgs interact with MHCII molecules (without processing) outside the Ag binding groove. The SAg/MHCII bimolecular complex binds to the TCR through specificity determined only by the V region of the receptor α or β chain. doi:10.1128/9781555818463.ch21f2

552 interacts with the T-cell receptor (TCR). The interaction with the TCR is also nonconventional and relatively nonspecific; for most SAgs, binding occurs at a variable (V) location on the TCR b-chain (the Vb region). Since SAgs bind outside the area on the TCR used for Ag recognition, they activate a much higher percentage of T cells than can be activated by conventional Ags. However, compared to mitogens, which stimulate T cells in an indiscriminate manner, there is some degree of specificity in SAg action because only certain Vb (or rarely Va) sequences are recognized. Hence, not all T cells are stimulated. The staphylococcal and streptococcal PTs are prototype microbial SAgs that exert a variety of immunomodulatory effects leading to shock, immunosuppression, and other systemic abnormalities associated with TSS. While SEs are included with the PTs, they have the unique distinction of possessing an additional ability to induce an emetic response upon oral ingestion and are thus solely responsible for SFP. It is generally agreed that many of the toxins in this family, including members of the SE family, arose from a common ancestral gene that crossed the genus barrier and became stably established in both Staphylococcus and Streptococcus genera. Evidence for this idea is strongly supported by the observation that the structural genes for some SEs and related streptococcal PTs are carried on discrete genetic elements (see below).

Role of Genetic Elements in Generation and Dissemination of the SEs and SEls

Some staphylococcal and streptococcal PTs are encoded by structural genes located on bacteriophage genomes. Although the streptococcal toxins have been best studied, the genetic mobility of SEA and SEE appears to be similar to what is observed among group A streptococci. Betley and Mekalanos confirmed that the SEA structural gene (sea) is carried by a lysogenic phage. sea was cloned directly from the induced bacteriophage genome (15). Both the SEA and the SEE genes map near the phage attachment site (att site) on their respective phage genomes. In a high percentage of cases, the toxin-encoding phages cannot be induced to a lytic cycle and appear to be defective. The most likely explanation for these observations is that the toxin genes were located originally on a bacterial genome but were subsequently obtained by the phage upon abnormal excision from the chromosome. This phenomenon is documented for some bacterial toxins that are transferred by lysogenic conversion in other genera of bacteria such as Streptococcus and Corynebacterium (67). The role of plasmids has received considerable attention in relation to transmission of PT genes. Of the

Foodborne Pathogenic Bacteria PTs, SED was the first to be demonstrated as plasmid encoded. Bayles and Iandolo reported that in more than 20 characterized sed+ isolates, the sed structural gene is localized to a stable 27.6-kb plasmid (pIB485), which also encodes penicillin and cadmium resistance (6). Subsequent studies also demonstrated that pIB485-related plasmids also harbor the genes for SElJ, SER, SES, and SET (99, 147). While the literature includes reports associating SEC and SEB genes with transmissible penicillin resistance plasmids (96), it is now generally agreed that both of these toxins are chromosomally encoded and harbored on pathogenicity islands in most strains. The best-characterized staphylococcal pathogenicity island (SaPI) is SaPI1, which harbors the genes for SElK and SElL in addition to TSST-1 (96). This SaPI and another TSST-1encoding island, SaPI2, can be mobilized to different extents by superinfection. Early reports suggested that seb is located on a DNA element (66) that has been recently sequenced and designated SaPI3. The sec gene has been identified on SaPI4, which is highly related to SaPI1 (but lacks tst) and SaPIbov, an island common to bovine isolates (expressing SEC, TSST-1, and SElK) (39). In another molecular variation, the genes for SEG and SEI are clustered on an operon within the staphylococcal chromosome that also contains genes for SElM, SElN, SElO, and several pseudogenes (39, 63). This enterotoxin gene cluster (designated egc) has had several variations identified and is proposed to be a nursery of new toxin genes. Some staphylococcal toxin genes are insertion sites for genetic elements carrying other virulence determinants (14). The expression of several SEs is affected by this feature. For example, SEB and TSST-1 synthesis are mutually exclusive in S. aureus. At least in some isolates, this is due to the fact that SaPI3 and SaPI1 share a common insertion site (96). Likewise, the phage that harbors sea utilizes the b-toxin locus hlb as its insertion site so that sea+ isolates do not produce b-toxin.

Mechanisms and Rationale for Generation of SE Diversity

Based on amino acid sequences, currently known members of the SE family are divided into four groups (Fig. 21.3). Group 1 contains the highly related SEA, SElP, SEE, SEIJ, and SED, with the last being more distantly related. Group 2 contains SEB, the SEC subtypes and molecular variants, SEG, SER, and SElU. Toxins in this group are highly related to each other and to several streptococcal PTs. Group 3 contains SEI, SEIL, SEIK, SEIM, and SEIQ. Group 4 contains SET and nonemetic TSST-1, which are smaller than other SEs.

21.  Staphylococcus aureus

553

Figure 21.3  Tree representation demonstrating molecular relatedness of the currently known SE family and compared to TSST-1. This tree was created with the clustering feature of the PHYLIP program (37). doi:10.1128/9781555818463.ch21f3

Most staphylococcal and streptococcal PTs, even those with no significant overall homology, contain four highly conserved stretches of primary sequence (51). This suggests that there is a selective advantage in host-parasite interactions for these bacteria to maintain certain toxin characteristics. At the same time, modification of selected regions of the proteins could allow the bacteria to broaden their host range. This molecular diversity may explain how a group of toxins with the same function but different host specificities could have arisen for the purpose of exploiting a broader repertoire of receptors. Sequence comparisons of members within the PT family have provided several examples where diversity among the toxins appears to have arisen through gene duplication and/or homologous recombination. For example, SEC1 is most similar to SEC2 and SEC3. However, residues 14 through 26 of SEC1 (Fig. 21.1) are identical to the analogous region of SEB but significantly different from the

other SEC subtypes (14). Genetic recombination between seb and sec in a strain producing both SEB and SEC2 (or SEC3) could explain the generation of SEC1. Additional minor variations have resulted from point mutations; even closely related SEs display some sequence differences. This may reflect fine-tuning of the toxin sequences for interacting with cells from a variety of hosts, as demonstrated with the SEC subtype variants. The sequences of SEC toxins produced by strains of S. aureus isolated from humans differ slightly (>95% identity) from sequences of SEC variants produced by bovine and ovine isolates (83).

Staphylococcal Regulation of SE Expression General Considerations

S. aureus is a widely distributed bacterium. It can exist in harsh environments as well as in various animals and

554 humans. This versatile survival ability relies on a complex network of virulence factors and adaptability. More than 50 genes involved in pathogenesis produce proteins that are either cell free or expressed on the bacterial surface. These enable the bacterium to evade host defenses, adhere to cells and the tissue matrix, propagate within the host, and destroy cells and tissues. Expression of virulence factors is temporarily controlled in response to cell density, nutrient availability, and environmental signals so that these factors are produced when required. During growth, expression of surface proteins is upregulated early, whereas that of cell-free secreted proteins is upregulated postexponentially. These changes in expression are regulated by a complex network of regulatory genes, described below. In addition, several environmental signals also affect the production of extracellular proteins such as high-salt conditions, pH, and subinhibitory concentrations of antibiotics or glycerol monolaurate (50). SEs are produced in extremely low quantities throughout most of the exponential growth phase (10, 93). There is generally a large increase in expression during the late exponential or early stationary phases of growth, with SEA and SED accumulating somewhat earlier than other SEs (14). Production of SE is dependent on de novo synthesis within the cell. The quantity of toxin produced is strain dependent, with SEB and SEC generally being produced in the highest quantities, up to 350 mg/ml. Even though there is a lower production of SEA, SED, and SEE, these SEs are generally detectable by gel diffusion assays, which detect as little as 100 ng of SE per ml of culture. Some strains produce very low levels of toxins and require more-sensitive analytical methods for detection. Some studies have revealed that SED and SEIJ are the toxins most likely to be undetected in S. aureus cultures, followed by SEC, SEA, and SEB in decreasing order.

Molecular and Environmental Regulation of SE Production

Several loci in S. aureus are involved in regulation of virulence factor expression. S. aureus genomic analyses reveal at least 16 two-component regulatory systems, of which agrCA (accessory gene regulator), saeRS (S. aureus exoprotein expression), lytRS (two-component regulatory system involved in autolysis), arlRS (autolysis­related locus), srrAB (staphylococcal respiratory response), and yccFG (two-component regulatory system involved in cell wall synthesis) have been characterized (22, 40, 43, 108). In addition to SarA (Staphylococcal accessory regulator A), a family of proteins homologous to SarA including Rot (repressor of toxin) and Xpr (exoprotein regulator) has been characterized (28, 85, 124).

Foodborne Pathogenic Bacteria The ability of S. aureus to respond to environmental changes involves the density-sensing agr system (95), with some properties typical of bacterial two-component sensors/regulators (128). agr expression coincides temporally with expression of most SEs during the bacterial growth cycle. All are maximally expressed during late exponential and postexponential growth. SEA, which is not regulated by agr, is produced earlier (14). The agr locus maps at approximately 4 o’clock on the staphylococcal genome, near bla (b-lactamase) and sea, on the standard map of S. aureus. It contains two divergent operons separated by approximately 120 bp (Fig. 21.4). Transcription can be initiated from three promoters (P1, P2, and P3). P1 is weakly constitutive and transcribes agrA. P2 and P3 are induced strongly during late exponential and early stationary phases but are only weakly expressed earlier. The P2 transcript, RNAII, encodes four proteins designated AgrA, AgrB, AgrC, and AgrD. A model for signal transduction through the agr system is shown in Fig. 21.4. Together these four proteins form a quorum-sensing system enabling the bacterium to respond to its environment. A unique cyclic thiolactone pheromone peptide (autoinducing peptide [AIP]) derived from AgrD residues 46 to 53 is the activating molecule for the agr system. AIP is processed from AgrD by AgrB (146). Increased levels of AIP are recognized by the membrane-bound receptor AgrC, which in turn initiates a signal transduction pathway (146). Its homology to conserved domains of histidine protein kinases, in particular a conserved histidine autophosphorylation site (128), further suggests that the 423-residue agrC gene product is likely to be the sensor. Autophosphorylation of the agrC product at the end of the exponential phase is thought to allow AgrC‑P to phosphorylate AgrA, the response regulator, by phosphorylation of the aspartic acid residue, thereby activating expression of both RNAII and RNAIII (97). AgrA has homology to activator proteins in other systems such as OmpR. AgrA and other typical response regulators exhibit a highly conserved N terminus of approximately 120 residues in length. Three amino acids, two aspartic acids and a lysine, are conserved among all response regulators. Mutations in the agr locus result in decreased expression of several SEs and other exoproteins. Regulation of gene expression by agr can be transcriptional or translational. It regulates a-hemolysin at both the transcriptional and translational levels, while SEB and SEC are regulated at the transcriptional level. Not all SEs are regulated by agr. SEA expression is not affected by agr mutations (139). At least 24 genes are under agr control; hence, it is a global regulator (95).

21.  Staphylococcus aureus

555

Figure 21.4  General characteristics of the agr locus in S. aureus. This physical map shows the relative location of genes within the locus and other interacting regulatory genes and gene products (not drawn to scale). doi:10.1128/9781555818463.ch21f4

Strains with mutations in AgrA, AgrB, AgrC, or AgrD have an Agr− phenotype and do not initiate transcription from either P2 or P3. The 514-nucleotide transcript from P3, designated RNAIII, encodes the 26-residue staphylococcal d-hemolysin and contains a significant amount of untranslated sequence. Interestingly, Agr − phenotypes can be complemented with plasmids encoding RNAIII under control of an inducible promoter, even when the d-hemolysin gene, hld, has been inactivated (74). Therefore, it appears that RNAIII is a diffusible element that plays a key role in the agr regulation of exoprotein structural genes, including those for the SEs (61, 74, 95). This RNA molecule can replace the regulatory function of the entire agr locus and acts reciprocally, upregulating transcription of most of the extracellular protein genes and downregulating many surface protein genes at the transcriptional level (95). The exact mechanism by which transcriptional regulation occurs has not been determined. The sae locus is a positive effector of cell-free band a-hemolysins, coagulase, nuclease, and protein A but apparently does not affect SEA, protease, lipase,

s­ taphylokinase, or cell-bound protein A. Mutations in sae had no effect on the production of RNAIII (95). These results suggest that the sae locus is distinct from other previously identified regulatory loci and may not act completely at the level of transcription. The SarA protein family is a family of proteins homologous to SarA. Included in this family are SarA homologs (SarA, R, S, T, U, and V), MgrA (a global regulator of autolysis and virulence), and Rot (the repressor of toxin) (28, 60, 116). Sequence alignment and crystal structure studies demonstrated that SarA, R, and S are typical “winged-helix DNA binding proteins” with helix-turn-helix motifs (79, 81, 82). SarA binds to several target promoters, including agr, hla, spa, and fnbA. SarA binds to an intergenic region between agr promoters P2 and P3 that upregulates expression of RNAIII, hence influencing the expression of agr-regulated genes (29). Transcriptional gene fusion analysis revealed that sarA upregulates both tst and seb (26). Mutation in the sar locus causes increased expression of a-toxin and decreased expression of d- and b-toxins (27).

556 Rot is a member of the SarA family of transcriptional factors. Rot acts as both a negative and a positive regulator of gene expression (116). Inactivation of rot was able to partially restore a-toxin and protease-positive phenotypes in an agr-null mutant (85). It also regulates expression of SEB (140). Deletion of the upstream promoter element to position −59 resulted in a dramatic reduction in the seb mRNA level. The sequence between −93 and −53 contains a Rot-binding site that negatively regulates transcription of seb. Even though the transcription of rot was not affected by agr, its activity is negatively regulated by agr. With the induction of RNAIII during the postexponential growth phase, the activity of Rot as a repressor of exotoxin expression is inactivated (85). However, the exact mechanism of Rot regulation by the Agr system remains to be elucidated. Sigma (s) factors may also affect temporal expression of SEs. RNA polymerase purified from exponential-phase cultures of S. aureus contains a s-70‑related factor, similar to that of Escherichia coli (111). The holoenzyme containing this s-factor transcribed sea more efficiently than sec or agr P2, suggesting additional complexity in the differential expression of certain SEs. While agr and most exoprotein genes, including sec, are expressed as the cells enter stationary phase, sea is expressed earlier during the exponential phase, simultaneously with the s-70-like factor. Hence, differential SE expression at specific points of the bacterial growth phase may coincide with availability of compatible s-factors. The alternative sigma factor (sB) is the major response to environmental stimuli. The sigB operon is constitutively transcribed under control of s70 and encodes SigB (sB) and the anti-SigB factor RsbW, the antirepressor RsbV, and RsbU. RsbW usually binds to sB, which phosphorylates RsbV (75). Under environmental stresses such as high temperature, alkaline pH, high levels of NaCl, and catabolites (glucose, galactose, sucrose, glycerol, and maltose), phosphorylated RsbV is dephosphorylated by RsbU or RsbP and then binds RsbW, subsequently releasing and activating sB (14, 89). sB recognizes a unique promoter, GTTT(N14-17)GGGTAT, which has been identified for 23 different S. aureus genes including the sar locus (44). This suggests that environmental signals activate sB, which in turn activates global virulence regulation. SEB expression is negatively regulated by sB. When sB is activated by stress-induced conditions, expression of SEB is reduced (123). In contrast, expression of SEB is increased in a sigB mutant; however, the promoter of seb differs from the unique promoter for sB. This would suggest that expression of SEB is not directly regulated by sB.

Foodborne Pathogenic Bacteria SEC expression is affected by glucose through at least two different mechanisms. First, metabolism of glucose indirectly influences SEC production through agr by reducing pH (114). Since agr is maximally expressed at neutral pH, growth in a nonbuffered environment containing glucose lowers pH levels, which directly reduces agr expression. Consequently, expression of sec and other agr target genes is affected correspondingly (114). Glucose also reduces sec expression in strains lacking agr (Dagr strains). This suggests the existence of a second glucose-dependent mechanism for reduction of SE expression, independent of agr and apparently not involving pH (see below).

Other Relevant Molecular Aspects of SE Expression

Although many chemical and physical factors selectively inhibit SE expression (42, 58, 70), their effects on regulation and signal transduction are only beginning to be elucidated. agr is not the only signal transduction mechanism for S. aureus, as revealed by investigations into the inhibitory effect of glucose. The negative regulatory effect of glucose described above cannot be attributed entirely to lower pH levels in cultures grown on glucose because cultures containing the carbohydrate produce less SE, even when the pH is stably maintained. Although this effect has some attributes of catabolite repression, there are major differences between the glucose inhibitory effect in S. aureus and catabolite repression in E. coli. For example, inhibition by glucose cannot be reversed by adding cyclic AMP to staphylococcal cultures. The significance of these differences is still unclear. Even the catabolite-repressible staphylococcal lac operon has features that are significantly different from those of the analogous operon in E. coli, and it is apparently unresponsive to cyclic AMP. S. aureus is an osmotolerant bacterium (see below). While it is able to survive and grow in environments of low water activity (aw), production of some SEs, especially SEB and SEC, is reduced when the bacteria are grown under osmotic stress. In experiments performed with SEC-producing strains, levels of sec mRNA and SEC protein are both reduced in response to high NaCl concentrations. However, addition of osmoprotectants reverses the effect. This reduced expression is also seen in Dagr strains, indicating that the signal transduction pathway used in this mechanism involves an alternative pathway. Low concentrations of the commonly used emulsifier glycerol monolaurate (GML) inhibit transcription of many exoprotein genes, including sea, without inhibiting S. aureus growth (109). Inhibition of SE production

21.  Staphylococcus aureus is not associated with a simultaneous effect on agr transcription and occurs in Dagr mutants as well as wild-type strains. These results, plus the finding that constitutive expression of some genes is not affected, suggest that GML interferes with non-agr-mediated signal transduction. It has been proposed that GML and a variety of related food additives exert this effect by insertion into the staphylococcal membrane, altering the membrane protein conformation and thereby interfering with signal transduction.

RESERVOIRS

Sources of Staphylococcal Food Contamination

Humans are the main reservoir for staphylococci involved in human disease, including S. aureus. Although most species are considered to be normal inhabitants of the external regions of the body, S. aureus is also a leading human pathogen. Colonized individuals may be carriers and are an important source for dissemination of staphylococci to others and to food. In humans, the anterior nares are the predominant site of colonization, although S. aureus can be present on other sites such as the skin or perineum. Dissemination of S. aureus among humans and from humans to food can occur by direct contact, indirectly by skin fragments, or through respiratory tract droplet nuclei. Today, most sources of SFP are traced to humans who contaminate food during preparation. In addition to contamination by food preparers who are carriers, S. aureus may also be introduced into food by contaminated equipment used in food processing such as meat grinders, knives, storage utensils, cutting blocks, and saw blades. A survey of over 700 foodborne disease outbreaks revealed the following conditions most often associated with food poisoning: (i) inadequate refrigeration; (ii) preparation of foods far in advance; (iii) poor personal hygiene, e.g., not washing either hands or instruments properly; (iv) inadequate cooking or heating of food; and (v) prolonged use of warming plates when serving foods, a practice that promotes staphylococcal growth and SE production (23). Animals, also an important source of S. aureus, are often heavily colonized with staphylococci. Predisposing factors that facilitate the survival of the bacterium are major concerns in the maintenance and processing of domestic animals and their products. For example, one very serious problem for the dairy industry is mastitis, an infectious disease of mammary tissue often caused by S. aureus. The combined losses and expenses associated with bovine mastitis make it the single most

557 costly disease of animal agriculture in the United States. Colonization of animals by S. aureus is also a public health concern because it may result in contamination of milk and dairy products with S. aureus prior to or during processing. It is not always possible to trace the source of staphylococcal food contamination to human or animal origin. Regardless of its source, many studies have demonstrated the common presence of S. aureus in many types of food products (Table 21.3). The cell numbers of staphylococci are often low initially. However, their widespread presence provides a potential source of bacteria capable of causing SFP if conditions appropriate for SE expression are provided.

Resistance to Adverse Environmental Conditions

Some unique resistance properties of S. aureus facilitate its contamination and growth in food. Outside the body, S. aureus is one of the most resistant non-spore-forming human pathogens and can survive for extended periods in a dry state. Its survival is facilitated by organic material, which is likely to be associated with the staphylococci from an inflammatory lesion. Isolation of staphylococci from air, dust, sewage, and water is relatively easy, and environmental sources of food contamination have been documented in several outbreaks of SFP. S. aureus is known for acquiring genetic resistance to heavy metals and antimicrobial agents used in clinical medicine. However, generally the resistance of this bacterium to common food preservative methods is unremarkable. One noteworthy exception is its ­osmotolerance, which permits growth in media containing the equivalent of 3.5 M NaCl and survival at aw of less than 0.86. This is especially problematic because other bacteria with which S. aureus does not compete efficiently are likely to be inhibited under these conditions. The molecular basis for staphylococcal osmotolerance has been investigated in recent years, although systems for responding to osmotic stress have been more intensively studied in less tolerant bacteria. Considering the unique resistance of staphylococci, it would not be surprising to find they have developed a highly efficient osmoprotectant system. As in other bacteria, several compounds accumulate in the cell or enhance staphylococcal growth under osmotic stress. Glycine betaine appears to be the most important osmoprotectant for S. aureus. To varying degrees, other compounds, including l-proline, proline betaine, choline, and taurine, can also act as compatible solutes for this bacterium. In S. aureus, intracellular levels

Foodborne Pathogenic Bacteria

558 Table 21.3  Prevalence of S. aureus in several foods Product Ground beef

No. of samples tested

% Positive for S. aureus No. of S. aureus CFU/ga

Reference

74

57

³100

1,830

8

³1,000

1,090

9

³100

107

Big-game meat

112

46

67 50

25 6

³10 100

126

Pork sausage Ground turkey Salmon steaks Oysters Blue crabmeat

75 86 59 896

80 2 10 52

Peeled shrimp

1,468

27

Lobster tail

1,315 465

Assorted cream pies Tuna pot pie Delicatessen salads a

³10 >3.4 >3.6 >3.6

135 24

135 47

³3

48 36 36 148

³3

137

24

³3

137

1

³25

141

1,290

2

³10

149

517

12

³3

104

Determined by either direct plate count or most-probable-number techniques.

of proline and glycine betaine accumulate to very high levels in response to increased concentrations of NaCl in the environment. Although the signal transduction pathway is not known for staphylococci, in other bacteria it involves a loss in turgor pressure in the cell and activation of required transport systems. High-affinity and low-affinity transport systems operate in S. aureus for both proline and glycine betaine (138, 143). The low-affinity systems are primarily stimulated by osmotic stress, have broad substrate specificity, and may be the same transporter shared by both osmoprotectants. By itself, the demonstration of a stress response system in S. aureus does not explain the unusual staphylococcal osmotolerance. Other less tolerant microorganisms possess mechanisms for counteracting osmotic stress. The efficiency of the staphylococcal system may reflect an unusually high endogenous level of intracellular K+ and lack of need for de novo transporter synthesis. For example, in other well-studied systems such as that of E. coli, changes in osmotic stress activate K+ transport systems. The elevated intracellular K+ levels that result are required for induction of proU and eventually lead to synthesis of transporters for glycine betaine and other osmoprotectants. In S. aureus, K+ levels are high in unstressed cells. Therefore, the transport system is constitutively present in this bacterium and is preformed when high-salt conditions are encountered. The net result is a very rapid and efficient response. S. aureus cells accumulate a 21-fold increase in proline

after less than 3 minutes of exposure to high salt concentrations (138).

FOODBORNE OUTBREAKS

Incidence of SFP

SFP occurs as either isolated cases or outbreaks affecting a large number of people. Since SFP is usually self-limiting, the incentive to report cases has not been as great as for other foodborne illnesses. Although there is a national surveillance system for SFP, it is not an officially reportable disease. It has been estimated that only 1 to 5% of all SFP cases are reported in the United States, usually at the state health department level. Most of these occur within highly publicized outbreaks. Isolated cases occurring in the home are not usually reported. Staphylococci account for an estimated 14% of the total of foodborne disease outbreaks within the United States (52). In a study conducted by the U.S. Department of Agriculture Economic Research Service, more than an estimated 1.5 million cases of SFP occurred in a single year (1993), resulting in 1,210 deaths and costing $1.2 billion (24). There has been an average of approximately 25 major outbreaks of SFP reported annually within the United States. Occurrence of SFP is cyclical, with the highest incidence typically occurring in the late summer, when temperatures are warm and food is more likely to be stored improperly (54). A second peak occurs in November and December. Approximately

21.  Staphylococcus aureus one-third of these are associated with leftover holiday food. SFP may be a leading cause of foodborne illness worldwide, although reporting in other countries is even less complete than in the United States. In one study, 40% of outbreaks of foodborne gastroenteritis in Hungary were due to SFP (9). The percentage is slightly lower in Japan (approximately 20 to 25%), where contamination of rice balls during preparation is a potential problem (9). Outbreaks due to improper manufacturing of canned corned beef have been reported in England, Brazil, Argentina, Malta, northern Europe, and Australia. Cases in Great Britain have been attributed to contaminated milk and cheese produced from milk from mastitic sheep (64). In some countries, ice cream has been a major vehicle of SFP. Despite increased knowledge of the epidemiology of SFP and attempts to educate the public, large outbreaks continue to represent a significant health hazard. For example, in 1998, approximately 4,000 individuals acquired SFP at a church event in Brazil (35). More recently, a massive outbreak in Osaka, Japan, in 2000 resulted in more than 10,000 cases from SEA and SEH in reconstituted milk (59).

Characteristics of a Recent Large Typical SFP Outbreak

The following is a summary of an outbreak of SFP reported by the U.S. Food and Drug Administration (4). Many of the aspects of this outbreak such as the type of food involved, mechanism of contamination, inadequate safety measures in food handling, and clinical manifestations of illness are typical. This particular outbreak originated from one meal that was fed to 5,824 elementary schoolchildren at 16 sites in Texas. Of all the children exposed, a total of 1,364 developed clinical signs of SFP. Investigation into the source of the illness revealed that 95% of the children who became ill had eaten chicken salad, from which large populations of S. aureus were cultured. The series of events leading up to the outbreak were as follows. Preparation of the meals was performed in a centralized kitchen facility and began on the preceding day. Frozen chickens used for the salad were boiled for 3 hours. After cooking, the chickens were deboned, cooled to room temperature with a fan, ground into small pieces, placed into 30.5-cm-deep aluminum pans, and stored overnight in a walk-in refrigerator at 5.6 to 7.2°C. On the following morning, the remaining ingredients of the salad were added and the mixture was blended with an electric mixer. The food was placed in

559 thermal containers and transported by truck to the various schools between 9:30 a.m. and 10:30 a.m. It was held at room temperature until served between 11:30 a.m. and noon. It is believed that the chicken became contaminated after cooking, when it was deboned. Most likely, the storage of the warm chicken in the deep aluminum pans did not permit rapid cooling and provided an environment favorable for staphylococcal growth and SE production. Further growth of the bacteria probably occurred during the period when the food was kept in the warm classrooms. Prevention of the incident would have entailed more-rapid cooling of the chicken and refrigeration of the salad after preparation.

CHARACTERISTICS OF DISEASE

Symptoms of SFP in Humans

SFP is usually a self-limiting illness presenting with emesis following a short incubation period. However, vomiting is not the only symptom that is commonly observed. Likewise, a significant number of patients with SFP do not vomit. Other common symptoms include nausea, abdominal cramps, diarrhea, headaches, muscular cramping, and/or prostration. In a summary of clinical symptoms involving 2,992 patients diagnosed with SFP, 82% complained of vomiting, 74% felt nauseated, 68% had diarrhea, and 64% exhibited abdominal pain (52). In all cases of diarrhea, vomiting was always present. Diarrhea is usually watery but may contain blood as well. The absence of high fever is consistent with a lack of infection or significant toxemia in this type of food poisoning, although some patients present with low-grade fever. Other symptoms can include general weakness, dizziness, chills, and perspiration. Symptoms typically develop within 6 hours after ingestion of contaminated food. In one report, 75% of the exposed individuals exhibited symptoms of SFP within 6 to 10 hours postingestion (52). The mean incubation rate is 4.4 hours, although incubation periods as short as 1 hour have been reported. In another outbreak, symptoms lasted for 1 to 88 hours, with a mean of 26.3 hours. Death, usually from severe dehydration or electrolyte imbalance, occurs in a small percentage of patients, with a fatality rate ranging from 0.03% for the general public to 4.4% for more susceptible populations such as children and the elderly (52). Approximately 10% of patients with confirmed SFP seek medical attention. Treatment is minimal in most cases, although administration of fluids is indicated when diarrhea and vomiting are severe.

560

TOXIC DOSE AND SUSCEPTIBLE POPULATIONS

Numbers of Staphylococci Required

Since many variables affect the amount of SE produced, one cannot predict with certainty the number of S. aureus cells in food required to cause SFP. Factors contributing to toxin concentrations have been extensively studied and include environmental conditions such as food composition, temperature, other physical and chemical parameters, and the presence of antimicrobial growth inhibitors. Also, bacterial factors to be considered include potential differences in the types, amounts, and numbers of different SEs that the strain involved has the physiological ability to produce. It is likely that these combined conditions are unique for each isolated case or outbreak of SFP. Despite this variability, there are several general guidelines that are useful for assessing general risk. According to the U.S. Food and Drug Administration, doses of SE causing illness result when populations of greater than 105 S. aureus cells per gram of contaminated food are present (4). In other studies, 105 to 108 S. aureus cells were observed to represent the typical range, despite the fact that lower levels were sometimes implicated (52).

Toxin Dose Required

Many investigations have been conducted to assess SE potency and the amount of toxin in food required to initiate SFP symptoms. Some of the most useful information in this regard has come from analysis of food recovered from outbreaks of the illness. Although SEs are quite potent, the amount of SE required to induce symptoms is relatively large compared to many other exotoxins that are acquired through contaminated food. A basal level of approximately 1 ng of SE per gram of contaminated food is sufficient to cause symptoms associated with SFP. Although 1 to 5 μg of ingested toxin is typically associated with many outbreaks, the actual levels of detectable SE were considerably less (<0.01 mg) in 16 SFP outbreaks (45). One of the most useful studies for predicting the minimal oral dose of SE required to induce SFP in humans was a well-documented investigation of an outbreak associated with ingestion of contaminated chocolate milk (36). In that study, the minimum dose of SEA required to cause SFP in school children was found to be 144 ± 50 ng. Another massive SFP outbreak in Japan, caused by low-fat milk contaminated with SEA, showed that the total intake of SEA per capita was found to be 20 to 100 ng (5). The minimal number of S. aureus cells required to produce a toxic level of SE in food has been investigated. Noleto et al. (94) showed that an S. aureus strain

Foodborne Pathogenic Bacteria harboring sea, seb, and sed produced detectable levels of SEB and SED (1 ng/ml) at a density of 6 × 106 CFU/ ml and SEA (4 ng/ml) at a density of 3 × 107 CFU/ml under nutrient-enriched conditions. However, when grown in milk, this strain produced detectable levels of SEA and SED as low as 104 and 107 CFU/ml, respectively. Considering the variability in strains, culture conditions, and genetic regulation of SE production, this information needs to be interpreted with caution. Many factors contribute to the likelihood of developing symptoms and their severity. The most important include susceptibility of the individual to the toxin, the total amount of food ingested, and the overall health of the affected person. The toxin type may also influence the likelihood and severity of disease. Though SFP outbreaks attributed to ingestion of SEA are much more common, individuals exposed to SEB exhibit more-severe symptoms. Forty-six percent of 2,291 individuals exposed to SEB exhibited SFP symptoms severe enough to be admitted to the hospital (52). Only 5% of 1,813 individuals exposed to SEA required such treatment. It is possible that these observations reflect differences in the levels of toxin present because SEB is generally produced at higher concentrations than SEA. Human volunteers and several species of macaque monkeys have been used to determine the minimal amount of purified SEs required to induce emesis when administered orally. Generally, monkeys are less susceptible to SE-­induced enterotoxicity than humans. The study of purified SEs has provided useful comparative information and has important research applications, but its direct relevance to SFP is uncertain because potential stabilization of unpurified SEs by food is an important consideration. Based on a study in which human volunteers ingested partially purified toxin, Raj and Bergdoll estimated that 20 to 25 μg of SEB (0.4 μg/kg) is sufficient to cause vomiting in humans (110). In the Rhesus monkey model, the 50% emetic dose is between 5 and 20 μg per animal (or approximately 1 μg/kg of body weight) when administered intragastrically. In our investigations, the minimal emetic dose of SEC1 for pigtail monkeys (Macaca nemestrina) is consistently between 0.1 and 1.0 μg/kg.

VIRULENCE FACTORS/MECHANISMS OF PATHOGENICITY

SE Structure-Function Associations Basic Structural and Biophysical Features

SEs and SEls are globular proteins of approximately 25 kilodaltons. SE sequence analysis, relatedness, and

21.  Staphylococcus aureus diversity have been discussed above (Fig. 21.1 and 21.3 and Table 21.2). Although detailed studies have not been performed with every toxin, as a group they are stable molecules in many respects. Their recognition as heat-stable toxins arose from early studies in which enterotoxicity and antigenicity were not completely destroyed upon boiling crude preparations. Furthermore, less extreme temperature treatments such as those used for pasteurization of milk had little or no effect on SE toxicity. The heat resistance of the SEs has been extensively studied. The general conclusion from this combined work is that SEs are difficult to inactivate by heating and have increased stability when present in high concentrations or in crude states such as in the environment of food. Since temperatures required to inactivate SEs are much higher than those needed to kill S. aureus under the same environmental conditions, toxic food involved in many cases of SFP is devoid of viable staphylococci at the time of serving. One additional SE property that has potential significance toward development of SFP is their resistance to inactivation by proteases present in the gastrointestinal tract. Resistance to pepsin, especially in a relatively low pH environment, is a key requirement for SE activity in vivo. Each of the SEs tested has some resistance to pepsin, a property not shared by at least one nonemetic staphylococcal PT, TSST-1. SEB is susceptible to degradation by pepsin at very low pH values, but partial neutralization of gastric acidity by ingested food is presumed to temporarily provide a protective environment for the toxin (9). SEs may be cleaved by other common proteases; however, unless the fragments generated are separated in the presence of denaturing agents, proteolysis alone may not be sufficient to cause a loss of biological activity. This is apparently representative of inherent SE molecular stability, which can be demonstrated by renaturing studies. For example, denaturation occurs only under strong denaturing conditions employing high concentrations of urea or guanidine hydrochloride. If the denaturing conditions are removed, SEs may spontaneously renature and regain biological activity (126). Differences in stability exist among the toxins. For example, SEB and SEC1 are approximately 50-fold more stable under denaturing conditions than SEA (141). It has been suggested that the SE disulfide bond is responsible for its inherent molecular stability. Several investigators have determined that the closed disulfide bond does contribute at least some degree of conformational stabilization but its disruption has only minimal effects on the overall stability and activity of the molecule (53) (see below).

561

Three-Dimensional Structure

Despite their sequence diversity with only partial sequence conservation, structural studies revealed that the three-dimensional topology of SEs is remarkably uniform (133). The molecule has an overall ellipsoidal shape and is folded into two domains of aminoand carboxy-terminal domains that are connected by α-helices in the center (Fig. 21.5). The amino-terminal domain, domain 1, contains residues near the N terminus but not the N-terminal residues themselves. The folding conformation of this domain may have potential significance for the function of the toxin. Its topology, in which a β-barrel structure is capped at one end by an a-helix, is known as the oligonucleotide/oligosaccharide binding (OB) fold. The internal portion of its β-barrel is rich in hydrophobic residues, and the potential oligomer binding surface is covered with mainly hydrophilic residues. This OB-folding pattern is found in staphylococcal nuclease and the B subunit of AB5 heat-labile enterotoxins such as cholera toxin, pertussis toxin, and Shiga toxin (86), which share the feature of exerting their activity of interacting with either oligosaccharides or oligonucleotides.

Figure 21.5  Schematic diagrams of the SEC3 crystal structure illustrating major structural features. Numerical designations defining the location of select residues and each a-helix and β-strand are shown within the two major domains. Also indicated are the N and C termini. The intramolecular disulfide linkage between Cys residues 93 and 110 (stick) connects the disulfide loop to the β5 strand containing the conserved residues (Fig. 21.7) potentially important for emesis. The zinc atom bound by SEC3 faces the back of the SEC3 molecule between domains 1 and 2 and is coordinated by D83, H118, and H122. In contrast, the high-affinity zinc-binding site in SEA is positioned on the opposite edge of domain 2. The conformational topology of domain 1 is the same as the OB binding domains of several other proteins described in the text. doi:10.1128/9781555818463.ch21f5

562

Foodborne Pathogenic Bacteria

The other prominent feature of domain 1 is that in most toxins of this family it contains two cysteine residues responsible for forming the disulfide linkage characteristic of most SEs. This bond and the cysteine loop are located at the end of the domain opposite its a-helix cap. Crystallographic data of SEA, SEB, and SEC3 indicate that the loop region of all three toxins is quite flexible (21, 118, 133). The larger carboxy-terminal domain (domain 2) can be described as a five-strand antiparallel β-sheet wall overlaid with a group of a-helices forming a β-grasp motif. It has structural similarity to the immunoglobulin-binding motifs of streptococcal proteins G and L, ubiquitin, 2F2-2S ferredoxin, and the Ras-binding domains of the Ser/Thr-specific protein kinase Raf-1 (46, 144). The structural similarities among SEs and other bacterial proteins suggest that SEs evolved through the recombination of these proteins (88).

Binding of Zinc by SEs

Biochemical and structural studies of SEs have revealed that some SEs are dependent on zinc ions to be functional and to be able to properly bind MHCII. Fraser et al. (41) determined that SEA and SEE bind zinc via a single site with a dissociation constant of 1 to 2 μM. The binding site was subsequently predicted by mutagenesis of SEA to be comprised of a nonlinear stretch of residues (H187, H225, and D227) in the concave β-sheet within the C-terminal domain and N-terminal serine in the metal coordination (Fig. 21.6) (118). Crystallographic analysis of SEA crystals soaked in 10 mM ZnCl2 revealed a second zinc-binding site, similar to that of the SEC2 site (see below). It is coordinated by D86 and H114 (corresponding to D83 and H118 in SEC2), a water molecule replaces the second histidine residue in SEC2 (H122), and E39 is analogous to D9 in SEC2 (Fig. 21.5) (117). SEC2 and SEC3 bind zinc through a low-affinity mechanism (20, 21). The zinc ion is tetrahedrally coordinated by D83, H118, and H122 from one molecule and D9 from a neighboring molecule. The zinc-binding site is located in a classical motif (H-E-X-X-H) at the cleft in the a5 groove of SEC3 between the two domains. This zinc-binding motif is typically found in the catalytic site of metalloenzymes such as thermolysin (49) and in certain other bacterial protease-dependent toxins such as botulinum and tetanus neurotoxins (72). However, none of the SEs is known to possess protease activity. Although the function of the zinc atom in SEC is not certain, it has been suggested that it plays a minor role in stabilizing the toxin structure.

Figure 21.6  (A) Schematic diagram of the SEA crystal structure. SEA has two MHCII binding sites. A relatively low-affinity MHCII binding occurs at a generic binding site that is conserved in most of SEs. A high-affinity MHCII binding site is located on the external surface of domain 2. This includes the high-affinity zinc-binding site, formed by His187, His225, and Asp227. The zinc ion mediates cross-linking of SEA with MHCII molecules and is crucial for maximal B- and T-cell activation. (B) A hypothetical model of MHCII-SEC-TCR complex based on the modeling predicted from the crystal structures of the SEC3-HLA-DR1 (low-affinity binding site) and SEB-Vb complexes. doi:10.1128/9781555818463.ch21f6

Crystallographic studies revealed that SED binds two zinc ions and forms a homodimer in a zinc-dependent manner (132). One zinc ion is tetrahedrally coordinated by D182, H220, and D222 from one molecule and H218 from the other molecule, which allows it to bind both MHCII a and b chains (see below). The second zinc-binding site is similar to that of SEC2.

Binding to MHCII Molecules

Despite their structural similarities, the mechanisms by which various SEs bind MHCII are diverse. SEs interact

21.  Staphylococcus aureus with MHCII in three ways: (i) binding a single α chain of MHCII, (ii) binding a single β chain of MHCII, or (iii) cross-linking two MHCII molecules. The first characterized type of SE binding is to the a chain of MHCII. SEB, SEC1, SEC2, SEC3, and TSST-1 bind in this manner and share a common MHCII binding site at domain 1 (referred to as the generic site) with a comparatively low affinity (0.4 to 0.7 µM). The crystal structure of the SEB-HLA-DR1 complex showed that one major interaction is mediated by a salt bridge provided by Glu67 of SEB and Lys39 on HLA-DR1 (62). Although SEC2 and SEC3 are able to bind zinc, crystal structure data and site-directed mutagenesis studies revealed that the zinc ion is not involved in the MHCII interaction (131). The second type of interaction is with the b-chain of MHCII. A crystal structure of SEH complexed with HLA-DR1 revealed that toxin interacts with the b-chain of the receptor, and binding requires its bound zinc ion (105). The zinc ion cross-linking plays an important role in the interaction of SEH and HLA-DR1 by providing cross-linking through His81 on HLA-DR1 and His206 and Asp208 on SEH. The third type of interaction involved SE cross-linking of two MHCII molecules. SEA has a low-affinity MHCII a chain binding site that overlaps that of SEB and TSST-1, plus a high-affinity site on the outside of domain 2 near the N terminus (57). The SEA high-affinity site involves coordination of zinc through three toxin residues (His187, His225, and Asp227) and His 81 of the MHCII b-chain. Two MHCII binding sites per toxin allow two MHCII molecules to be crosslinked (137) (Fig. 21.6). Similar modes of action, including formation of homodimers by some toxins, have been proposed for other SEs with high-affinity zinc-binding sites on the outer face of domain 2 (SED and SEE) (3, 132).

Binding to TCR

SEs and other SAgs interact with a characteristic repertoire of TCR sequences (Table 21.1). The TCR specificity of each SE is determined by toxin residues in the shallow cavity at the top of the molecule (34). Crystallographic analysis of a molecular complex between SEC3 and a portion of the murine TCR β chain (38) indicated that the toxin binds to the CDR1, CDR2, and HV4 loops of Vβ. Modeling permitted the deduction of a trimolecular complex model containing the SAg bridging the T cell and APC. Using SEC3 and SEB, this model shows that these toxins form a “wedge” between the TCR and the MHCII, orient-

563 ing the peptide-binding cleft away from the TCR (Fig. 21.6B). Although different SAgs are predicted to bind with variations of this theme, in general, SAg binding results in receptor interactions significantly different from that in typical Ag presentation. The overall affinity of the entire complex determines the effectiveness of the stimulation (77), and binding by toxins with low affinity for the TCR can be compensated for by stronger binding to MHCII, and vice versa. SEH appears to have the most divergent mechanism of binding and T-cell stimulation. This toxin binds to MHCII in a mechanism that does not allow efficient interactions between SEH and the TCR Vβ or between MHCII and Vβ. In one study (106), SEH stimulated T cells by interacting with TCR Va (Va10), and Vβ-specific expansion could not be demonstrated.

Molecular Regions of SEs Responsible for Enterotoxicity

The structural aspects of SEs that enable them to survive degradation by pepsin and other enzymes in the gastrointestinal tract are required for the toxins to induce SFP. However, stability alone is not sufficient. SEs must also be able to interact with the appropriate target, leading to emesis, diarrhea, and other gastrointestinal tract symptoms. Initial attempts to define molecular regions responsible for enterotoxicity involved testing the biological activity of protease-generated fragments derived from SEA, SEB, or SEC1 (126). Three main conclusions were drawn from this work. First, only large toxin fragments containing central and C-terminal portions of the SEs retained enough of the native structure to cause emesis. Second, N-terminal residues of the SEs were not required for emesis. SEC1, modified by removal of the 59 N-terminal residues, retains the ability to cause emesis in monkeys. Smaller toxin fragments from SEC1 and other SEs were inactive. The third conclusion was that emesis seemed to require preservation of structure in the area of the conserved SE disulfide loop. The disulfide bond is a structural feature that is characteristic of most, but not all, SEs. The disulfide bond has long been associated with emesis and is not found uniformly in nonemetic staphylococcal and streptococcal exotoxins. With a few known exceptions, SEs contain exactly two cysteine residues, which could potentially form an intramolecular linkage and a spacer disulfide loop. SEG and the SEC molecular variant produced by isolates from sheep (SEC-ovine) deviate from this pattern and possess an additional third cysteine (Fig. 21.7). There is speculation regarding the importance in emesis of the disulfide bond, located near the center of

564 every classical and most other SEs. Based on data acquired from analysis of recently identified SEs and SEls in the monkey feeding assay, evidence suggests that a stable disulfide bond facilitates an emetic response in vivo. Specifically, toxins lacking one or both cysteine residues are generally nonemetic or weakly emetic (8, 92, 102–104, 134). Potential structural contributions from the Cys-Cys bond that could contribute to the SE conformation necessary for emetic activity could include one or more of the following features: (i) proper positioning of cysteine residues upon formation of the disulfide bond, (ii) exposure and/or orientation of crucial residues in the loop formed between the two linked cysteines, (iii) exposure and/or orientation of residues immediately adjacent to the disulfide linkage but not contained within the cysteine loop, and (iv) contributions to the overall SE conformation by the linkage of the two cysteine residues. Each of these possibilities has been considered. The cysteine residues and the loop probably do not play a direct role in the emetic response. It has been possible to make substitutions at the cysteine residues in several SEs by site-directed mutagenesis and show that neither of these two residues is absolutely critical. Although most of the SEs have a cysteine loop or analogous feature, the lengths and composition of residues within the loops of different SE types vary greatly. This lack of consistency among SE loop properties suggests that they are unlikely to have a shared enterotoxigenic function (Fig. 21.7). Furthermore, proteolytic nicking of toxins in their loops has no effect on their ability to cause emesis (126). In regard to overall protein conformation, Warren et al. (142) determined that the disulfide bond contributes only minimally. Presumably then, if the disulfide linkage is important in emesis, the effect is likely to provide a particular orientation of residues near the disulfide linkage, but not in the loop. The most convincing evidence in support of this possibility has been provided by mutagenesis of SEC1 in which its cysteine residues were replaced by either serine or alanine (53). It was found that mutants with serine substitutions were emetic, whereas the analogous mutants with alanine substitutions were nonemetic. Although serine and alanine are both considered to be conservative substitutions for cysteine, one difference between these two amino acids is that serine has the ability to hydrogen bond. Thus, hydrogen bonding by serine may be able to replace the disulfide linkage stabilization of local structure. A previous study showed that a very high dose (150 µg/kg) of SEI induced emesis in one of four animals in the monkey feeding assay (92, 101). A recent publication in which SET was reported to be

Foodborne Pathogenic Bacteria emetic should be interpreted with caution because the emetic response observed in the animals was not typical of SE-induced intoxication (101). Which critical local residues require proper orientation by the disulfide bond (or hydrogen bonding at the same positions) in order for the SEs to induce emesis is unknown. Possible candidates are those within a highly conserved stretch of residues directly adjacent to, and downstream from, the disulfide loop (Fig. 21.7). In SEC3, these residues are located on the b5-strand. An attractive hypothesis is that in addition to stability in the gut, two other structural requirements need to be met for enterotoxicity. First, the appropriate conserved residues must be present in the toxin. Many PTs, emetic and nonemetic, have a similar set of highly conserved residues in a location analogous to the b5-strand of SEC3. Second, they must be positioned properly for interacting with their target in the gut. The unique SE disulfide bond may serve this function. Of the entire PT family, two non-SE toxins produced by S. pyogenes (streptococcal pyrogenic exotoxin and streptococcal SAg) could potentially form a disulfide linkage (113). However, the presence of more than two cysteines in these toxins suggests that the structure in this area and the degree of local stabilization by their putative disulfide linkage may not be identical to those of the SEs. Additional studies on SEA revealed that single-site substitutions of residues close to the N terminus also influence emesis (48). This was especially the case for mutants constructed by substitution with glycine. For example, mutagenesis of residues 25, 47, and 48 causes significant reduction in the emetic potency of SEA. Although these residues are far from the disulfide bond in the SEA primary sequence, they are located near or within domain 1 of SEA and could potentially influence the area near the disulfide bond.

SE Antigenic Epitopes

The need for reagents that detect SEs in food and clinical samples, as well as the desire to differentiate antigenically different toxins in the toxin family, has been the impetus for considerable effort directed toward epitope characterization and mapping. Although the increased number of identified SEs and putative SEs is beginning to make immunological detection of SEs obsolete, several commercial reagents relying on this technique are still widely used. However, detection based on antigenicity is gradually being replaced by molecular techniques, especially multiplex PCR (91, 98). Individually, each classical SE type and subtype has a sufficient degree of antigenic distinctness to allow its differentiation from other PTs by using highly specific poly-

21.  Staphylococcus aureus

565

Figure 21.7  Comparison of cysteine loop and adjacent sequences for SEs and the analogous regions of the SEls. Evidence suggests that proper positioning of the critical downstream residues by a stable disulfide bond is required for emesis. Toxins designated as emetic are those reported as inducing emesis in the monkey feeding assay (8, 92, 101–104, 134). SEI was reported to be weakly emetic (92). doi:10.1128/9781555818463.ch21f7

clonal antisera and monoclonal antibodies. However, some degree of cross-reactivity between several SEs can often be demonstrated. The level of cross-reactivity generally correlates with shared primary sequences. The Type C SE subtypes and their molecular variants exhibit a substantial amount of cross-reactivity, as do SEA and SEE, the two major serological types with the greatest sequence relatedness. For these toxins, cross-reactivity can even be demonstrated in relatively insensitive assays such as immunodiffusion assays, in which these two toxins produce lines of partial identity with the heterologous antiserum. SEB and the SEC subtypes and molecular variants (see above) are also highly related at the amino acid level. Cross-reactivity between SEB and SEC may be demonstrated occasionally by immunodiffusion, but more consistently using sensitive methods such as radioimmunoassays or immunoblotting. Generally, it has not been possible to produce useful antibodies that

cross-react among less related SEs. Although one investigator has produced a monoclonal antibody that crossreacts with all five major SE antigenic types A through E, this antibody has low affinity and cross-reacts with other staphylococcal proteins (11). The two most distantly related SEs determined to be recognized by a common epitope are SEA and SED (12). The mapping of conserved and specific antigenic epitopes on SEs and their differentiation from potentially toxic regions have potential applications toward rational development of nontoxic vaccines. Considering the array of SE antigenic types, the most efficient toxoid would presumably contain one or more epitopes that are shared by multiple toxins. Several approaches have been used to partially localize antigenic epitopes on the SEs and differentiate them from toxic regions. One of the earlier methods used for this purpose was to identify protease-generated toxin fragments from several SEs

566 that bind to cross-reactive antibodies (19). These studies revealed that both N- and C-terminal toxin fragments contain cross-reactive epitopes, but they were unable to define shorter stretches of residues. There is some evidence that immunization with short highly conserved peptides could have merit. For example, the use of synthetic peptides from highly conserved stretches of primary sequence has resulted in the production of neutralizing antibody for several of the SEs. Immunization with synthetic peptides corresponding to residues 130 to 160 of SEB or the same region of SEC1 (residues 148 to 162) induced antibodies that neutralized both native toxins (51, 65). The highly conserved SE sequence K-K-X-V-T-X-Q-E-L-D (Fig. 21.1), encompassed by both peptides, may represent part of an epitope that could be useful for protective immunity. It is yet to be determined if major epitopes identified on other toxins such as SEA also show promise (107).

SE Mode of Action in Induction of Emesis and Other Symptoms Related in SFP SE-Induced Emesis Requires Nerve Stimulation

Except for rare SFP cases in which massive doses of SEs are consumed, systemic dissemination of the toxins does not contribute significantly to illness. When fed to rodents, SEs do enter circulation but are rapidly removed by the kidneys (7). Most studies using the simian model indicate that the SE site of action following ingestion is the abdominal viscera. Early studies into the mechanism of action of SEs tested the emetic responsiveness of animals to the toxins after disrupting well-defined neural systems or after visceral deafferation. The characteristic emetic response was determined to result from stimulation of local neural receptors in the abdomen (130), which transmit impulses through the vagus and sympathetic nerves, ultimately stimulating the medullary emetic center.

Cellular Histopathology in the Gastrointestinal Tract

Information on the histological effects of oral doses of SEs in humans is extremely limited. Most of what is known has been derived from information obtained from experiments in Rhesus monkeys. Upon ingestion of the toxin, pathological changes compatible with a definition of gastroenteritis are observed in several parts of the gastrointestinal tract (71). The primate stomach becomes hyperemic and is marked by lesions that begin with the influx of neutrophils into the lamina propria and epithelium. A mucopurulent exudate in the gastric lumen is also typically observed. Also

Foodborne Pathogenic Bacteria characteristic are mucus-filled surface cells that eventually release their contents. Later in the illness, neutrophils become replaced by macrophages. Eventually, upon resolution of the symptoms, the cellular infiltrate clears. A similar cellular infiltrate and lumen exudate occur in the small intestine, although they decrease in severity in sections taken from lower, compared to upper, portions of the intestine. Clearly evident in the jejunum are extension of crypts, disruption or loss of the brush border, and an extensive infiltrate of neutrophils and macrophages into the lamina propria. Changes in the colon are minimal in the monkey model. Only a mild cellular exudate and mucus depletion are evident. The only other significant effect in monkeys is acute lymphadenitis in the mesenteric lymph nodes.

Search for the Gastrointestinal Tract Target

The specific cells that SEs interact with in the abdomen have not been clearly identified, nor has their receptor. Evidence suggests that interaction of SEs with their target directly or indirectly causes production of inflammatory mediators that induce SFP symptoms. Jett et al. (65) determined that oral administration of SEB produced elevated levels of arachidonic acid cascade products. Specifically, they observed significant increases in prostaglandin E2, leukotriene B4, and 5-hydroxyeicosatetraenoic acid. These three compounds are potent vasoactive inflammatory mediators that can also act as chemoattractants for neutrophils. Both of these activities are consistent with the histopathology described above in the SFP monkey model. Scheuber et al. (119) could not demonstrate an effect of prostanoid inhibitors on SEB-induced emesis but were able to correlate gastrointestinal symptoms with cysteinyl leukotriene generation. Intoxication of animals with SEB resulted in a 10-fold increase in levels of leukotriene E4 in bile, plus an unidentified leukotriene in the urine. This group of investigators suggested a role for mast cells in the pathogenesis of SFP. Although induction of histamine production by SEB was responsible for some secondary nonenteric immediate hypersensitivity skin reactions, it did not correlate with emesis. Evidence suggests that ingestion of SEs causes a stimulation of mast cells and possibly other inflammatory cells in the abdomen. Thus far, the abdominal receptor has not been identified. Experiments using anti-idiotype antibodies in binding assays and protection assays have provided circumstantial evidence for its existence on monkey mast cells (112). Komisar et al. showed that SEB stimulates rodent peritoneal mast cells as well and provided evidence for a protein receptor (73). It is possible that SEs act directly on their receptor and circum-

21.  Staphylococcus aureus vent the typical two-stage mast cell immunoglobulin E antibody-Ag interaction (119). Despite these observations, Alber et al. (2) were unable to directly stimulate monkey or human skin and intestinal mast cells with SEB to release inflammatory mediators. It was suggested that stimulation of mast cells in vivo occurs through a nonimmunological mechanism requiring the generation of neuropeptides released from peripheral terminals of primary sensory nerves. At least one putative mast cell-stimulating peptide, substance P, was implicated in SEB-induced toxicity by use of antibodies and a variety of inhibitors. The attractiveness of this explanation is that it is consistent with earlier predictions of a neural involvement in the pathogenesis of SFP. A recent study using the house musk shrew demonstrated that SEA induces 5-hydroxytryptamine (5-HT) release in intestine (55). Pretreatment with a 5-HT synthesis inhibitor, or 5-HT(3) receptor, or surgical vagatomy was able to inhibit SEA-induced emesis. Furthermore, cannabinoid receptor agonists significantly decreased the release of 5-HT and SEA-induced emesis. These results suggested that, in this model, SEA-induced emesis is mediated by the release of 5-HT and 5-HT(3) receptors on vagal afferent neurons and is downregulated by the cannabinoid receptor system by decreasing the release of 5-HT. However, results with other animal models including kittens, piglets, house musk shrews, and ferrets, particularly those that require systemic administration rather than oral feeding, must be interpreted with caution, since they may mimic TSS rather than SFP.

Is There a Relationship between Superantigenicity, TSS, and SFP?

The discovery of the mechanism of SAg action and the unique properties of this class of proteins provided an explanation for the multiple systemic effects seen in TSS patients. The massive cellular stimulation induced by superantigenic PTs explained the long-recognized fact that TSS patients had elevated serum cytokine and other mediator levels that mediate many symptoms of the disease. The realization that at least some of the pathogenesis of TSS could be attributed to immune cell stimulation led to the prediction by some investigators that SFP could also be a reflection of SAg function. Consistent with this prediction was the fact that TSS patients often have a gastrointestinal tract illness component characterized by vomiting and diarrhea. Also, patients with endotoxin shock have elevated cytokine levels and similarly display vomiting and diarrhea. If superantigenicity is responsible for SFP, the SEs presumably act directly on T cells and APCs in the gut. As mentioned above, while some

567 SEs enter the host’s circulatory system, they appear to be rapidly cleared by the kidneys so that significant systemic concentrations are unlikely to be achieved (7). This, added to the fact that TSS-associated symptoms (i.e., shock and fever) are not observed in SFP patients, suggests that SEs do not mediate SFP through systemic cytokines. Despite their similarities and the evidence cited above, several lines of evidence suggest that the partial overlap between SFP and TSS symptoms is probably coincidental and that superantigenicity is not directly responsible for SFP. First, as discussed above, nonimmunological mast cell stimulation has been linked to the release of inflammatory mediator affecting nerve interactions. The second line of evidence has come through mutagenesis of several SEs. Studies have revealed that T-cell stimulation and induction of emesis are separable functions and are determined by distinct portions of the SE molecules (1, 48, 53). It has been possible to construct SEA, SEB, or SEC1 mutants that are deficient in T-cell stimulatory activity but retain the ability to induce emesis and vice versa. Finally, although all PTs have been reported to have SAg function, only the SEs are emetic when ingested. The lack of emesis-inducing ability in some nonenterotoxic PTs has been attributed to instability in the gastrointestinal tract. However, at least one nonemetic PT, streptococcal pyrogenic exotoxin A, is very stable in gastric fluid (121). If superantigenicity does not explain SFP, how do PTs cause vomiting and diarrhea in TSS? Several possibilities could explain how SEs and other PTs act on the gastrointestinal tract if they are not consumed through the oral route. First, the toxins’ ability to induce TSS symptoms may be limited to the systemic circulation. If so, cytokines would need to enter the abdomen or gain access to the central nervous system from the circulation and mediate gastroenteritis pathogenesis. Alternatively, the PTs could enter the gut from the circulation and act directly at the local level as SAgs or through other mechanisms as proposed for SFP. The latter possibility is less likely since even nonenterotoxic PTs, including those that are susceptible to degradation in the gut, are known to cause gastrointestinal symptoms when they are associated with TSS. These complex and unresolved issues are relevant to interpreting models for studying the enterotoxic activity of SEs. Oral administration of SEs, culture filtrates, or suspected food to monkeys is the preferred method for detecting enterotoxic properties of the SEs. However, intravenous administration of SEs to primates and other animals such as cats has been used as an alternative to feeding (31, 103). Considering that systemic exposure

Foodborne Pathogenic Bacteria

568 to SEs mimics the situation leading to TSS symptoms, intravenous administration may not accurately reflect the pathogenesis and toxin properties required to induce emesis in SFP. Recently, other models that demonstrate emesis following administration of toxins to shrews and piglets have been proposed as alternatives to the gold standard, the monkey feeding assay (54, 134). Additional studies will be required to determine whether these alternatives can replace the primate feeding assay.

CONCLUDING REMARKS Progress has been made toward understanding the molecular aspects relevant to SFP. S. aureus has some unique properties that promote its ability to produce foodborne illness. However, further understanding of the molecular aspects of these unique properties should facilitate the implementation of more efficient ways to selectively inhibit staphylococcal survival, growth, and SE production in food. For example, exploitation of properties of the global regulation of virulence factors could be used to block exotoxin production and could have potential applications in the food industry. One issue that continues to puzzle the scientific community is the questionable rationale for staphylococci to produce an emetic toxin. Unlike enteric pathogens that inhabit the gut, the ability to induce emesis and diarrhea as a mechanism to promote exit from the host and dissemination does not appear to be important for staphylococci. One may ask a similar question regarding the ability of SEs to induce lethal shock in TSS, since it is generally agreed that lethality is not advantageous to microorganisms. Instead, long-term host-pathogen coexistence usually relies upon adaptations that allow the microorganism to survive for extended periods without harming the host or being cleared by the immune response. Based on what is now known regarding the super­ antigenic properties and proposed host-specific molecular adaptation of the SEs, one may propose that SEs are produced for the purpose of immunomodulating the host. Exposure of animals and peripheral blood mononuclear cell cultures to SEs and other SAgs has repeatedly been shown to induce at least a transient immunosuppression. Similarly, TSS patients often fail to produce a significant immune response to causative superantigenic PTs and remain susceptible to subsequent toxigenic illnesses. Hence, one may speculate that the harmful effects on the host (SFP and TSS) are merely secondary effects of the staphylococci attempting to affect immune cell function to allow the bacteria to survive and persistently colonize its many animal hosts.

Our efforts in the preparation of this manuscript were sup­ ported by grants from the U.S. Department of Agriculture (2008-00892).

References   1. Alber, G., D. K. Hammer, and B. Fleischer. 1990. Relationship between enterotoxic- and T lymphocytestimulating activity of staphylococcal enterotoxin B. J. Immunol. 144:4501–4506.   2. Alber, G., P. H. Scheuber, B. Reck, B. Sailer-Kramer, A. Hartmann, and D. K. Hammer. 1989. Role of substance P in immediate-type skin reactions induced by staphylococcal enterotoxin B in unsensitized monkeys. J. Allergy Clin. Immunol. 84:880–885.   3. Al-Daccak, R., K. Mehindate, F. Damdoumi, P. Etongue-Mayer, H. Nilsson, P. Antonsson, M. Sundstrom, M. Dohlsten, R. P. Sekaly, and W. Mourad. 1998. Staphylococcal enterotoxin D is a promiscuous superantigen offering multiple modes of interactions with the MHC class II receptors. J. Immunol. 160:225–232.   4. Anonymous. 1992. Foodborne Pathogenic Microorga­n­ isms and Natural Toxins. Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, Rockville, MD.   5. Asao, T., Y. Kumeda, T. Kawai, T. Shibata, H. Oda, K. Haruki, H. Nakazawa, and S. Kozaki. 2003. An extensive outbreak of staphylococcal food poisoning due to low-fat milk in Japan: estimation of enterotoxin A in the incriminated milk and powdered skim milk. Epidemiol. Infect. 130:33–40.   6. Bayles, K. W., and J. J. Iandolo. 1989. Genetic and molecular analyses of the gene encoding staphylococcal enterotoxin D. J. Bacteriol. 171:4799–4806.   7. Beery, J. T., S. L. Taylor, L. R. Schlunz, R. C. Freed, and M. S. Bergdoll. 1984. Effects of staphylococcal enterotoxin A on the rat gastrointestinal tract. Infect. Immun. 44:234–240.   8. Bergdoll, M. S. 1988. Monkey feeding test for staphylococcal enterotoxin. Methods Enzymol. 165:324–333.   9. Bergdoll, M. S. 1985. The staphylococcal enterotoxin— an update, p. 247–254. In J. Jeljaszewicz (ed.), The Staphylococci. Gustav Fischer Verlag, Stuttgart, Germany.   10. Bergdoll, M. S. 1979. Staphylococcal intoxications, p.  443–494. In H. Riemann and F. L. Bryan (ed.), Food-Borne Infections and Intoxications. Academic Press, Inc., New York, NY.   11. Bergdoll, M. S. 1989. Staphylococcus aureus, p.  463– 523. In M. P. Doyle (ed.), Foodborne Bacterial Pathogens. Marcel Dekker, Inc., New York, NY.   12. Bergdoll, M. S., C. R. Borja, R. N. Robbins, and K. F. Weiss. 1971. Identification of enterotoxin E. Infect. Immun. 4:593–595.   13. Bergdoll, M. S., M. J. Surgalla, and G. M. Dack. 1959. Staphylococcal enterotoxin. Identification of a specific precipitating antibody with enterotoxin-neutralizing property. J. Immunol. 83:334–338.

21.  Staphylococcus aureus   14. Betley, M. J., D. W. Borst, and L. B. Regassa. 1992. Staphylococcal enterotoxins, toxic shock syndrome toxin and streptococcal pyrogenic exotoxins: a comparative study of their molecular biology. Chem. Immunol. 55:1–35.   15. Betley, M. J., and J. J. Mekalanos. 1985. Staphylococcal enterotoxin A is encoded by phage. Science 229:185–187.   16. Betley, M. J., P. M. Schlievert, M. S. Bergdoll, G. A. Bohach, J. J. Iandolo, S. A. Khan, P. A. Pattee, and R. R. Reiser. 1990. Staphylococcal gene nomenclature. ASM News 56:182.   17. Blaiotta, G., D. Ercolini, C. Pennacchia, V. Fusco, A. Casaburi, O. Pepe, and F. Villani. 2004. PCR detection of staphylococcal enterotoxin genes in Staphylococcus spp. strains isolated from meat and dairy products. Evidence for new variants of seG and seI in S. aureus AB-8802. J. Appl. Microbiol. 97:719–730.   18. Bohach, G. A., D. J. Fast, R. D. Nelson, and P. M. Schlievert. 1990. Staphylococcal and streptococcal pyrogenic toxins involved in toxic shock syndrome and related illnesses. Crit. Rev. Microbiol. 17:251–272.   19. Bohach, G. A., C. J. Hovde, J. P. Handley, and P. M. Schlievert. 1988. Cross-neutralization of staphylococcal and streptococcal pyrogenic toxins by monoclonal and polyclonal antibodies. Infect. Immun. 56:400–404.   20. Bohach, G. A., L. M. Jablonski, C. F. Deobald, Y. I. Chi, and C. V. Stauffacher. 1995. Functional domains of staphylococcal enterotoxin, p. 339–356. In M. Ecklund, J. L. Richard, and M. Mise (ed.), Molecular Approaches to Food Safety; Issues Involving Toxic Microorganisms. Alaken Inc., Fort Collins, CO.   21. Bohach, G. A., C. V. Stauffacher, D. H. Ohlendorf, Y. I. Chi, G. M. Vath, and P. M. Schlievert. 1996. The staphylococcal and streptococcal pyrogenic toxin family, p. 131–154. In B. R. Singh and A. T. Tu (ed.), Natural Toxins II. Plenum Publishing Corporation, New York, NY.   22. Brunskill, E. W., and K. W. Bayles. 1996. Identification of LytSR-regulated genes from Staphylococcus aureus. J. Bacteriol. 178:5810–5812.   23. Bryan, F. L. 1976. Staphylococcus aureus, p. 12–128. In M. P. deFigueiredo and D. F. Splittstoesser (ed.), Food Microbiology: Public Health and Spoilage Aspects. AVI, Westport, CT.   24. Buzby, J. C., T. Roberts, C. T. Lin, and J. M. McDonald. 1996. Bacterial Foodborne Disease Medical Costs and Productivity Losses. Agricultural Economic Report 147. U.S. Department of Agriculture, Washington, DC. http://www.ers.usda.gov/publications/aer-agriculturaleconomic-report/aer741.aspx.   25. Casman, E. P., M. S. Bergdoll, and J. Robinson. 1963. Designation of staphylococcal exterotoxins. J. Bacteriol. 85:715–716.   26. Chan, P. F., and S. J. Foster. 1998. Role of SarA in virulence determinant production and environmental signal transduction in Staphylococcus aureus. J. Bacteriol. 180:6232–6241.   27. Cheung, A. L., Y. T. Chien, and A. S. Bayer. 1999. Hyperproduction of alpha-hemolysin in a sigB mu-

569

  28.   29.   30.

  31.   32.   33.

  34.

  35.

  36.

  37.   38.

  39.

  40.

  41.

tant is associated with elevated SarA expression in Staphylococcus aureus. Infect. Immun. 67:1331–1337. Cheung, A. L., and G. Zhang. 2002. Global regulation of virulence determinants in Staphylococcus aureus by the SarA protein family. Front. Biosci. 7:d1825–d1842. Chien, Y., and A. L. Cheung. 1998. Molecular interactions between two global regulators, sar and agr, in Staphylococcus aureus. J. Biol. Chem. 273:2645–2652. Choi, Y., J. A. Lafferty, J. R. Clements, J. K. Todd, E.  W. Gelfand, J. Kappler, P. Marrack, and B. L. Kotzin. 1990. Selective expansion of T cells expressing V beta 2 in toxic shock syndrome. J. Exp. Med. 172:981–984. Clark, W. G., and J. S. Page. 1968. Pyrogenic responses to staphylococcal enterotoxins A and B in cats. J. Bacteriol. 96:1940–1946. Couch, J. L., M. T. Soltis, and M. J. Betley. 1988. Cloning and nucleotide sequence of the type E staphylococcal enterotoxin gene. J. Bacteriol. 170:2954–2960. Dack, G. M., W. E. Cary, O. Woolper, and H. Wiggers. 1930. An outbreak of food poisoning proved to be due to a yellow hemolytic Staphylococcus. Can. J. Microbiol. 4:167–175. Deringer, J. R., R. J. Ely, S. R. Monday, C. V. Stauffacher, and G. A. Bohach. 1997. Vbeta-dependent stimulation of bovine and human T cells by host-specific staphylococcal enterotoxins. Infect. Immun. 65:4048–4054. Do Carmo, L. S., C. Cummings, V. R. Linardi, R. S. Dias, J. M. De Souza, M. J. De Sena, D. A. Dos Santos, J. W. Shupp, R. K. Pereira, and M. Jett. 2004. A case study of a massive staphylococcal food poisoning incident. Foodborne Pathog. Dis. 1:241–246. Everson, M. L., M. W. Hinds, R. S. Bernstein, and M.  S. Bergdoll. 1988. Estimation of human dose of staphylococcal enterotoxin A from a large outbreak of staphylococcal food poisoning involving chocolate milk. Int. J. Food Microbiol. 7:311–316. Felsenstein, J. 1989. PHYLIP-Phylogeny Inference Package (Version 3.2). Cladistics 5:164–166. Fields, B. A., E. L. Malchiodi, H. Li, X. Ysern, C. V. Stauffacher, P. M. Schlievert, K. Karjalainen, and R.  A. Mariuzza. 1996. Crystal structure of a T-cell receptor beta-chain complexed with a superantigen. Nature 384:188–192. Fitzgerald, J. R., S. R. Monday, T. J. Foster, G. A. Bohach, P. J. Hartigan, W. J. Meaney, and C. J. Smyth. 2001. Characterization of a putative pathogenicity island from bovine Staphylococcus aureus encoding multiple superantigens. J. Bacteriol. 183:63–70. Fournier, B., and A. Klier. 2004. Protein A gene expression is regulated by DNA supercoiling which is modified by the ArlS-ArlR two-component system of Staphylococcus aureus. Microbiology 150:3807–3819. Fraser, J. D., S. Lowe, M. J. Irwin, N. R. Gascoigne, and K. R. Hudson. 1993. Structural model of staphylococcal enterotoxin A interaction with MHC class II antigens, p. 7–30. In B. T. Huber and E. Palmer (ed.), Superantigens: a Pathogens’s View of the Immune System. Cold Spring Harbor Laboratory Press, Plainview, NY.

Foodborne Pathogenic Bacteria

570   42. Friedman, M. E. 1966. Inhibition of staphylococcal enterotoxin B formation in broth cultures. J. Bacteriol. 92:277–278.   43. Fukuchi, K., Y. Kasahara, K. Asai, K. Kobayashi, S. Moriya, and N. Ogasawara. 2000. The essential twocomponent regulatory system encoded by yycF and yycG modulates expression of the ftsAZ operon in Bacillus subtilis. Microbiology 146(Pt. 7):1573–1583.   44. Gertz, S., S. Engelmann, R. Schmid, A. K. Ziebandt, K. Tischer, C. Scharf, J. Hacker, and M. Hecker. 2000. Characterization of the sigma(B) regulon in Staphylococcus aureus. J. Bacteriol. 182:6983–6991.   45. Gilbert, R. J., and A. A. Wieneke. 1973. Staphylococcal food poisoning with special reference to the detection of enterotoxin in food, p. 273–285. In B. C. Hobbs and J. H. Christian (ed.), The Microbiological Safety of Food. Academic Press, Inc., New York, NY.   46. Gronenborn, A. M., D. R. Filpula, N. Z. Essig, A. Achari, M. Whitlow, P. T. Wingfield, and G. M. Clore. 1991. A novel, highly stable fold of the immunoglobulin binding domain of streptococcal protein G. Science 253:657–661.   47. Hajek, V., and E. Marsalek. 1973. The occurrence of enterotoxigenic Staphylococcus aureus strains in hosts of different animal species. Zentralbl. Bakteriol. Orig. A 223:63–68.   48. Harris, T. O., and M. J. Betley. 1995. Biological activities of staphylococcal enterotoxin type A mutants with Nterminal substitutions. Infect. Immun. 63:2133–2140.   49. Hase, C. C., and R. A. Finkelstein. 1993. Bacterial extra­ cellular zinc-containing metalloproteases. Microbiol. Rev. 57:823–837.   50. Herbert, S., P. Barry, and R. P. Novick. 2001. Subinhibitory clindamycin differentially inhibits transcription of exoprotein genes in Staphylococcus aureus. Infect. Immun. 69:2996–3003.   51. Hoffmann, M. L., L. M. Jablonski, K. K. Crum, S. P. Hackett, Y. I. Chi, C. V. Stauffacher, D. L. Stevens, and G. A. Bohach. 1994. Predictions of T-cell ­receptorand major histocompatibility complex-binding sites on staphylococcal enterotoxin C1. Infect. Immun. 62:3396–3407.   52. Holmberg, S. D., and P. A. Blake. 1984. Staphylococcal food poisoning in the United States. New facts and old misconceptions. JAMA 251:487–489.   53. Hovde, C. J., J. C. Marr, M. L. Hoffmann, S. P. Hackett, Y. I. Chi, K. K. Crum, D. L. Stevens, C. V. Stauffacher, and G. A. Bohach. 1994. Investigation of the role of the disulphide bond in the activity and structure of staphylococcal enterotoxin C1. Mol. Microbiol. 13:897–909.   54. Hu, D. L., K. Omoe, Y. Shimoda, A. Nakane, and K. Shinagawa. 2003. Induction of emetic response to staphylococcal enterotoxins in the house musk shrew (Suncus murinus). Infect. Immun. 71:567–570.   55. Hu, D. L., G. Zhu, F. Mori, K. Omoe, M. Okada, K. Wakabayashi, S. Kaneko, K. Shinagawa, and A. Nakane. 2007. Staphylococcal enterotoxin induces emesis through increasing serotonin release in intestine

  56.

  57.

  58.   59.

  60.

  61.

  62.

  63.

  64.   65.

  66.   67.

  68.   69.

and it is downregulated by cannabinoid receptor 1. Cell Microbiol 9:2267–2277. Huang, I. Y., and M. S. Bergdoll. 1970. The primary structure of staphylococcal enterotoxin B. 3. The cyanogen bromide peptides of reduced and aminoethylated enterotoxin B, and the complete amino acid sequence. J. Biol. Chem. 245:3518–3525. Hudson, K. R., R. E. Tiedemann, R. G. Urban, S.  C. Lowe, J. L. Strominger, and J. D. Fraser. 1995. Staphylococcal enterotoxin A has two cooperative binding sites on major histocompatibility complex class II. J. Exp. Med. 182:711–720. Iandolo, J. J., and W. M. Shafer. 1977. Regulation of staphylococcal enterotoxin B. Infect. Immun. 16:610–616. Ikeda, T., N. Tamate, K. Yamaguchi, and S. Makino. 2005. Mass outbreak of food poisoning disease caused by small amounts of staphylococcal enterotoxins A and H. Appl. Environ. Microbiol. 71:2793–2795. Ingavale, S., W. van Wamel, T. T. Luong, C. Y. Lee, and A. L. Cheung. 2005. Rat/MgrA, a regulator of autolysis, is a regulator of virulence genes in Staphylococcus au­ reus. Infect. Immun. 73:1423–1431. Janzon, L., and S. Arvidson. 1990. The role of the deltalysin gene (hld) in the regulation of virulence genes by the accessory gene regulator (agr) in Staphylococcus au­ reus. EMBO J. 9:1391–1399. Jardetzky, T. S., J. H. Brown, J. C. Gorga, L. J. Stern, R. G. Urban, Y. I. Chi, C. Stauffacher, J. L. Strominger, and D. C. Wiley. 1994. Three-dimensional structure of a human class II histocompatibility molecule complexed with superantigen. Nature 368:711–718. Jarraud, S., M. A. Peyrat, A. Lim, A. Tristan, M. Bes, C. Mougel, J. Etienne, F. Vandenesch, M. Bonneville, and G. Lina. 2001. egc, a highly prevalent operon of enterotoxin gene, forms a putative nursery of superantigens in Staphylococcus aureus. J. Immunol. 166:669–677. Jay, J. M. 1986. Modern Food Microbiology, 3rd ed., p. 437–458. Van Nostrand Reinhold Company, New York, NY. Jett, M., R. Neill, C. Welch, T. Boyle, E. Bernton, D. Hoover, G. Lowell, R. E. Hunt, S. Chatterjee, and P. Gemski. 1994. Identification of staphylococcal enterotoxin B sequences important for induction of lymphocyte proliferation by using synthetic peptide fragments of the toxin. Infect. Immun. 62:3408–3415. Johns, M. B., Jr., and S. A. Khan. 1988. Staphylococcal enterotoxin B gene is associated with a discrete genetic element. J. Bacteriol. 170:4033–4039. Johnson, L. P., and P. M. Schlievert. 1983. A physical map of the group A streptococcal pyrogenic exotoxin bacteriophage T12 genome. Mol. Gen. Genet. 189:251–255. Jones, C. L., and S. A. Khan. 1986. Nucleotide sequence of the enterotoxin B gene from Staphylococcus aureus. J. Bacteriol. 166:29–33. Kappler, J., B. Kotzin, L. Herron, E. W. Gelfand, R.  D. Bigler, A. Boylston, S. Carrel, D. N. Posnett, Y. Choi, and P. Marrack. 1989. V beta-specific stimulation of human T cells by staphylococcal toxins. Science 244:811–813.

21.  Staphylococcus aureus   70. Katsuno, S., and M. Kondo. 1973. Regulation of staphylococcal enterotoxin B synthesis and its relation to other extracellular proteins. Jpn. J. Med. Sci. Biol. 26:26–29.   71. Kent, T. H. 1966. Staphylococcal enterotoxin gastro­ enteritis in rhesus monkeys. Am. J. Pathol. 48:387–407.   72. Klimpel, K. R., N. Arora, and S. H. Leppla. 1994. Anthrax toxin lethal factor contains a zinc metallo­ protease consensus sequence which is required for lethal toxin activity. Mol. Microbiol. 13:1093–1100.   73. Komisar, J., J. Rivera, A. Vega, and J. Tseng. 1992. Effects of staphylococcal enterotoxin B on rodent mast cells. Infect. Immun. 60:2969–2975.   74. Kornblum, J., B. Kreiswirth, S. J. Projan, H. Ross, and R. P. Novick. 1990. Agr: a polycistronic locus regulating exoprotein synthesis in Staphylococcus aureus, p. 373– 402. In R. P. Novick and R. Skurray (ed.), Molecular Biology of the Staphylococci. VCH Publishers, New York, NY.   75. Kullik, I. I., and P. Giachino. 1997. The alternative sigma factor sigmaB in Staphylococcus aureus: regulation of the sigB operon in response to growth phase and heat shock. Arch. Microbiol. 167:151–159.   76. Kuroda, M., T. Ohta, I. Uchiyama, T. Baba, H. Yuzawa, I. Kobayashi, L. Cui, A. Oguchi, K. Aoki, Y. Nagai, J. Lian, T. Ito, M. Kanamori, H. Matsumaru, A. Maruyama, H. Murakami, A. Hosoyama, Y. MizutaniUi, N. K. Takahashi, T. Sawano, R. Inoue, C. Kaito, K. Sekimizu, H. Hirakawa, S. Kuhara, S. Goto, J. Yabuzaki, M. Kanehisa, A. Yamashita, K. Oshima, K. Furuya, C. Yoshino, T. Shiba, M. Hattori, N. Ogasawara, H. Hayashi, and K. Hiramatsu. 2001. Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet 357:1225–1240.   77. Leder, L., A. Llera, P. M. Lavoie, M. I. Lebedeva, H. Li, R. P. Sekaly, G. A. Bohach, P. J. Gahr, P. M. Schlievert, K. Karjalainen, and R. A. Mariuzza. 1998. A mutational analysis of the binding of staphylococcal enterotoxins B and C3 to the T cell receptor beta chain and major histocompatibility complex class II. J. Exp. Med. 187:823–833.   78. Letertre, C., S. Perelle, F. Dilasser, and P. Fach. 2003. Identification of a new putative enterotoxin SEU encoded by the egc cluster of Staphylococcus aureus. J. Appl. Microbiol. 95:38–43.   79. Li, R., A. C. Manna, S. Dai, A. L. Cheung, and G. Zhang. 2003. Crystal structure of the SarS protein from Staphylococcus aureus. J. Bacteriol. 185:4219–4225.   80. Lina, G., G. A. Bohach, S. P. Nair, K. Hiramatsu, E. Jouvin-Marche, and R. Mariuzza. 2004. Standard nomenclature for the superantigens expressed by Staphylococcus. J. Infect. Dis. 189:2334–2336.   81. Liu, Y., A. Manna, R. Li, W. E. Martin, R. C. Murphy, A. L. Cheung, and G. Zhang. 2001. Crystal structure of the SarR protein from Staphylococcus aureus. Proc. Natl. Acad. Sci. USA 98:6877–6882.   82. Liu, Y., A. C. Manna, C. H. Pan, I. A. Kriksunov, D.  J. Thiel, A. L. Cheung, and G. Zhang. 2006. Structural and function analyses of the global regulatory protein

571

  83.

  84.   85.

  86.   87.

  88.   89.

  90.

  91.

  92.

  93.   94.   95.   96.

  97.

SarA from Staphylococcus aureus. Proc. Natl. Acad. Sci. USA 103:2392–2397. Marr, J. C., J. D. Lyon, J. R. Roberson, M. Lupher, W.  C. Davis, and G. A. Bohach. 1993. Characterization of novel type C staphylococcal enterotoxins: biological and evolutionary implications. Infect. Immun. 61:4254–4262. Marrack, P., and J. Kappler. 1990. The staphylococcal enterotoxins and their relatives. Science 248:1066. McNamara, P. J., K. C. Milligan-Monroe, S. Khalili, and R. A. Proctor. 2000. Identification, cloning, and initial characterization of rot, a locus encoding a regulator of virulence factor expression in Staphylococcus aureus. J. Bacteriol. 182:3197–3203. Merritt, E. A., and W. G. Hol. 1995. AB5 toxins. Curr. Opin. Struct. Biol. 5:165–171. Merson, M. H. 1973. The epidemiology of staphylo­ coccal foodborne disease, p. 20–37. In Proceedings of the Staphylococci in Foods Conference. Pennsylvania State University Press, University Park, PA. Mitchell, D. T., D. G. Levitt, P. M. Schlievert, and D. H. Ohlendorf. 2000. Structural evidence for the evolution of pyrogenic toxin superantigens. J. Mol. Evol. 51:520–531. Miyazaki, E., J. M. Chen, C. Ko, and W. R. Bishai. 1999. The Staphylococcus aureus rsbW (orf159) gene encodes an anti-sigma factor of SigB. J. Bacteriol. 181:2846–2851. Monday, S. R., and G. A. Bohach. 2001. Genes encoding staphylococcal enterotoxins G and I are linked and separated by DNA related to other staphylococcal enterotoxins. J. Nat. Toxins 10:1–8. Monday, S. R., and G. A. Bohach. 1999. Use of multiplex PCR to detect classical and newly described pyrogenic toxin genes in staphylococcal isolates. J. Clin. Microbiol. 37:3411–3414. Munson, S. H., M. T. Tremaine, M. J. Betley, and R. A. Welch. 1998. Identification and characterization of staphylococcal enterotoxin types G and I from Staphylococcus aureus. Infect. Immun. 66:3337–3348. Noleto, A. L., L. M. Malburg, Jr., and M. S. Bergdoll. 1987. Production of staphylococcal enterotoxin in mixed cultures. Appl. Environ. Microbiol. 53:2271–2274. Noleto, A. L., A. Tibana, C. A. Silva, L. Rubin, and S. Schueller. 1980. Bacteriological analysis of milk. An. Microbiol. (Rio J.) 25:41–51. (In Portuguese.) Novick, R. P. 2003. Autoinduction and signal transduction in the regulation of staphylococcal virulence. Mol. Microbiol. 48:1429–1449. Novick, R. P. 2003. Mobile genetic elements and bacterial toxinoses: the superantigen-encoding pathogenicity islands of Staphylococcus aureus. Plasmid 49:93–105. Novick, R. P., S. J. Projan, J. Kornblum, H. F. Ross, G. Ji, B. Kreiswirth, F. Vandenesch, and S. Moghazeh. 1995. The agr P2 operon: an autocatalytic sensory transduction system in Staphylococcus aureus. Mol. Gen. Genet. 248:446–458.

572   98. Omoe, K., D. L. Hu, H. Takahashi-Omoe, A. Nakane, and K. Shinagawa. 2005. Comprehensive analysis of classical and newly described staphylococcal super­ antigenic toxin genes in Staphylococcus aureus isolates. FEMS Microbiol. Lett. 246:191–198.   99. Omoe, K., D. L. Hu, H. Takahashi-Omoe, A. Nakane, and K. Shinagawa. 2003. Identification and characterization of a new staphylococcal enterotoxin-related putative toxin encoded by two kinds of plasmids. Infect. Immun. 71:6088–6094. 100. Omoe, K., K. Imanishi, D. L. Hu, H. Kato, H. TakahashiOmoe, A. Nakane, T. Uchiyama, and K. Shinagawa. 2004. Biological properties of staphylococcal enterotoxin-like toxin type R. Infect. Immun. 72:3664–3667. 101. Ono, H. K., K. Omoe, K. Imanishi, Y. Iwakabe, D. L. Hu, H. Kato, N. Saito, A. Nakane, T. Uchiyama, and K. Shinagawa. 2008. Identification and characterization of two novel staphylococcal enterotoxins, types S and T. Infect. Immun. 76:4999–5005. 102. Orwin, P. M., J. R. Fitzgerald, D. Y. Leung, J. A. Gutierrez, G. A. Bohach, and P. M. Schlievert. 2003. Characterization of Staphylococcus aureus enterotoxin L. Infect. Immun. 71:2916–2919. 103. Orwin, P. M., D. Y. Leung, H. L. Donahue, R. P. Novick, and P. M. Schlievert. 2001. Biochemical and biological properties of staphylococcal enterotoxin K. Infect. Immun. 69:360–366. 104. Orwin, P. M., D. Y. Leung, T. J. Tripp, G. A. Bohach, C. A. Earhart, D. H. Ohlendorf, and P. M. Schlievert. 2002. Characterization of a novel staphylococcal enterotoxinlike superantigen, a member of the group V subfamily of pyrogenic toxins. Biochemistry 41:14033–14040. 105. Petersson, K., M. Hakansson, H. Nilsson, G. Forsberg, L. A. Svensson, A. Liljas, and B. Walse. 2001. Crystal structure of a superantigen bound to MHC class II displays zinc and peptide dependence. EMBO J. 20:3306–3312. 106. Petersson, K., H. Pettersson, N. J. Skartved, B. Walse, and G. Forsberg. 2003. Staphylococcal enterotoxin H induces V alpha-specific expansion of T cells. J. Immunol. 170:4148–4154. 107. Pontzer, C. H., J. K. Russell, and H. M. Johnson. 1989. Localization of an immune functional site on staphylococcal enterotoxin A using the synthetic peptide approach. J. Immunol. 143:280–284. 108. Pragman, A. A., J. M. Yarwood, T. J. Tripp, and P. M. Schlievert. 2004. Characterization of virulence factor regulation by SrrAB, a two-component system in Staphylococcus aureus. J. Bacteriol. 186:2430–2438. 109. Projan, S. J., S. Brown-Skrobot, P. M. Schlievert, F. Vandenesch, and R. P. Novick. 1994. Glycerol monolaurate inhibits the production of beta-lactamase, toxic shock toxin-1, and other staphylococcal exoproteins by interfering with signal transduction. J. Bacteriol. 176:4204–4209. 110. Raj, H. D., and M. S. Bergdoll. 1969. Effect of enterotoxin B on human volunteers. J. Bacteriol. 98:833–834. 111. Rao, L., R. K. Karls, and M. J. Betley. 1995. In vitro transcription of pathogenesis-related genes by puri-

Foodborne Pathogenic Bacteria fied RNA polymerase from Staphylococcus aureus. J. Bacteriol. 177:2609–2614. 112. Reck, B., P. H. Scheuber, W. Londong, B. Sailer-Kramer, K. Bartsch, and D. K. Hammer. 1988. Protection against the staphylococcal enterotoxin-induced intestinal disorder in the monkey by anti-idiotypic antibodies. Proc. Natl. Acad. Sci. USA 85:3170–3174. 113. Reda, K. B., V. Kapur, J. A. Mollick, J. G. Lamphear, J. M. Musser, and R. R. Rich. 1994. Molecular characterization and phylogenetic distribution of the streptococcal superantigen gene (ssa) from Streptococcus pyogenes. Infect. Immun. 62:1867–1874. 114. Regassa, L. B., J. L. Couch, and M. J. Betley. 1991. Steady-state staphylococcal enterotoxin type C mRNA is affected by a product of the accessory gene regulator (agr) and by glucose. Infect. Immun. 59:955–962. 115. Ren, K., J. D. Bannan, V. Pancholi, A. L. Cheung, J. C. Robbins, V. A. Fischetti, and J. B. Zabriskie. 1994. Characterization and biological properties of a new staphylococcal exotoxin. J. Exp. Med. 180:1675–1683. 116. Said-Salim, B., P. M. Dunman, F. M. McAleese, D. Macapagal, E. Murphy, P. J. McNamara, S. Arvidson, T. J. Foster, S. J. Projan, and B. N. Kreiswirth. 2003. Global regulation of Staphylococcus aureus genes by Rot. J. Bacteriol. 185:610–619. 117. Schad, E. M., A. C. Papageorgiou, L. A. Svensson, and K. R. Acharya. 1997. A structural and functional comparison of staphylococcal enterotoxins A and C2 reveals remarkable similarity and dissimilarity. J. Mol. Biol. 269:270–280. 118. Schad, E. M., I. Zaitseva, V. N. Zaitsev, M. Dohlsten, T. Kalland, P. M. Schlievert, D. H. Ohlendorf, and L. A. Svensson. 1995. Crystal structure of the superantigen staphylococcal enterotoxin type A. EMBO J. 14:3292–3301. 119. Scheuber, P. H., C. Denzlinger, D. Wilker, G. Beck, D. Keppler, and D. K. Hammer. 1987. Staphylococcal enterotoxin B as a nonimmunological mast cell stimulus in primates: the role of endogenous cysteinyl leukotrienes. Int. Arch. Allergy Appl. Immunol. 82:289–291. 120. Schleifer, K. H. 1986. Gram positive cocci, p. 999–1100. In P. A. Sneath (ed.), Bergey’s Manual of Systematic Bacteriology, 1st ed., vol. 2. Williams & Wilkins Co., Baltimore, MD. 121. Schlievert, P. M., L. M. Jablonski, M. Roggiani, I. Sadler, S. Callantine, D. T. Mitchell, D. H. Ohlendorf, and G. A. Bohach. 2000. Pyrogenic toxin superantigen site specificity in toxic shock syndrome and food poisoning in animals. Infect. Immun. 68:3630–3634. 122. Schmidt, J. J., and L. Spero. 1983. The complete amino acid sequence of staphylococcal enterotoxin C1. J. Biol. Chem. 258:6300–6306. 123. Schmidt, K. A., N. P. Donegan, W. A. Kwan, Jr., and A. Cheung. 2004. Influences of sigmaB and agr on expression of staphylococcal enterotoxin B (seb) in Staphylococcus aureus. Can. J. Microbiol. 50:351–360. 124. Smeltzer, M. S., M. E. Hart, and J. J. Iandolo. 1993. Phenotypic characterization of xpr, a global regulator of

21.  Staphylococcus aureus

125.

126.

127. 128. 129. 130. 131.

132.

133. 134.

135.

136.

extracellular virulence factors in Staphylococcus aureus. Infect. Immun. 61:919–925. Smyth, D. S., P. J. Hartigan, W. J. Meaney, J. R. Fitzgerald, C. F. Deobald, G. A. Bohach, and C. J. Smyth. 2005. Superantigen genes encoded by the egc cluster and SaPIbov are predominant among Staphylococcus aureus isolates from cows, goats, sheep, rabbits and poultry. J. Med. Microbiol. 54:401–411. Spero, L., B. Y. Griffin, J. L. Middlebrook, and J. F. Metzger. 1976. Effect of single and double peptide bond scission by trypsin on the structure and activity of staphylococcal enterotoxin C. J. Biol. Chem. 251:5580–5588. Stefani, S., and P. E. Varaldo. 2003. Epidemiology of methicillin-resistant staphylococci in Europe. Clin. Microbiol. Infect. 9:1179–1186. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol. Rev. 53:450–490. Su, Y. C., and A. C. Wong. 1995. Identification and purification of a new staphylococcal enterotoxin, H. Appl. Environ. Microbiol. 61:1438–1443. Sugiyama, H., and T. Hayama. 1965. Abdominal viscera as site of emetic action for staphylococcal enterotoxin in the monkey. J. Infect. Dis. 115:330–336. Sundberg, E. J., M. W. Sawicki, S. Southwood, P. S. Andersen, A. Sette, and R. A. Mariuzza. 2002. Minor structural changes in a mutated human melanoma antigen correspond to dramatically enhanced stimulation of a CD4+ tumor-infiltrating lymphocyte line. J. Mol. Biol. 319:449–461. Sundstrom, M., L. Abrahmsen, P. Antonsson, K. Mehindate, W. Mourad, and M. Dohlsten. 1996. The crystal structure of staphylococcal enterotoxin type D reveals Zn2+-mediated homodimerization. EMBO J. 15:6832–6840. Swaminathan, S., W. Furey, J. Pletcher, and M. Sax. 1992. Crystal structure of staphylococcal enterotoxin B, a superantigen. Nature 359:801–806. Taylor, S. L., L. R. Schlunz, J. T. Beery, D. O. Cliver, and M. S. Bergdoll. 1982. Emetic action of staphylococcal enterotoxin A on weanling pigs. Infect. Immun. 36:1263–1266. Thomas, D. Y., S. Jarraud, B. Lemercier, G. Cozon, K. Echasserieau, J. Etienne, M. L. Gougeon, G. Lina, and F. Vandenesch. 2006. Staphylococcal enterotoxin-like toxins U2 and V, two new staphylococcal superantigens arising from recombination within the enterotoxin gene cluster. Infect. Immun. 74:4724–4734. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive

573

137.

138.

139.

140. 141. 142.

143.

144.

145.

146.

147.

multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680. Tiedemann, R. E., and J. D. Fraser. 1996. Cross-linking of MHC class II molecules by staphylococcal enterotoxin A is essential for antigen-presenting cell and T cell activation. J. Immunol. 157:3958–3966. Townsend, D. E., and B. J. Wilkinson. 1992. Proline transport in Staphylococcus aureus: a high-affinity system and a low-affinity system involved in osmoregulation. J. Bacteriol. 174:2702–2710. Tremaine, M. T., D. K. Brockman, and M. J. Betley. 1993. Staphylococcal enterotoxin A gene (sea) expression is not affected by the accessory gene regulator (agr). Infect. Immun. 61:356–359. seng, C. W., and G. C. Stewart. 2005. Rot repression of enterotoxin B expression in Staphylococcus aureus. J. Bacteriol. 187:5301–5309. Warren, J. R. 1977. Comparative kinetic stabilities of staphylococcal enterotoxin types A, B, and C1. J. Biol. Chem. 252:6831–6834. Warren, J. R., L. Spero, and J. F. Metzger. 1974. Stabilization of native structure by the closed disulfide loop of staphylococcal enterotoxin B. Biochim. Biophys. Acta 359:351–363. Wengender, P. A., and K. J. Miller. 1995. Identification of a PutP proline permease gene homolog from Staphylococcus aureus by expression cloning of the high-affinity proline transport system in Escherichia coli. Appl. Environ. Microbiol. 61:252–259. Wikstrom, M., T. Drakenberg, S. Forsen, U. Sjobring, and L. Bjorck. 1994. Three-dimensional solution structure of an immunoglobulin light chain-binding domain of protein L. Comparison with the IgG-binding domains of protein G. Biochemistry 33:14011–14017. Wilson, G. J., K. S. Seo, T. Connelley, O. N. ChuangSmith, J. A. Merriman, C. M. Guinane, R. A. Cartwright, J. Y. Park, G. A. Bohach, P. M. Schlievert, W. I. Morrison, and J. R. Fitzgerald. 2011. A novel core genome-encoded superantigen contributes to lethality of community-associated MRSA necrotizing pneumonia. PLoS Pathog. 7(10):e1002271. Zhang, L., L. Gray, R. P. Novick, and G. Ji. 2002. Transmembrane topology of AgrB, the protein involved­ in the post-translational modification of AgrD in Staphylococcus aureus. J. Biol. Chem. 277:34736–34742. Zhang, S., J. J. Iandolo, and G. C. Stewart. 1998. The enterotoxin D plasmid of Staphylococcus aureus encodes a second enterotoxin determinant (sej). FEMS Microbiol. Lett. 168:227–233.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch22

22

Craig W. Hedberg

Epidemiology of Foodborne Diseases

The epidemiology of foodborne diseases is constantly changing (42). As new foods and new sources of foods become available, new opportunities for foodborne disease transmission often follow. For example, the development of export production of raspberries in Guatemala was closely followed by widespread outbreaks of diarrheal illnesses caused by the coccidian parasite Cyclospora cayetanensis in the United States and Canada (48). Diets change in response to tastes and concerns about health. Americans are being challenged to eat five servings of fresh fruits and vegetables every day to help prevent cancer, heart disease, obesity, and diabetes (77, 95). As a nation, we increased consumption of fresh fruits and vegetables during the early 1990s, but that trend plateaued and we continue to fall short of this goal (26, 61, 85, 92). However, the number and variety of outbreaks of foodborne illness associated with fresh fruits and vegetables have increased (86) (Fig. 22.1). Other unanticipated consequences of health-related behaviors may be developing. In recent years, there has been a substantial increase in the use of statins to reduce cholesterol levels broadly across the population (76). In a recent outbreak, statin use was associated with an increased risk of norovirus infection

(78), and laboratory studies suggest that inhibition of cholesterol biosynthesis using statins may increase viral replication, a potential mechanism that would enhance susceptibility to infection (27). Moreover, as the population ages, the number and proportion of people at increased risk for serious complications of foodborne illnesses are increasing. Hence, even if the incidence of foodborne illnesses remains constant, the public health burden of these illnesses will increase. Not all change is bad, however. During the early 1990s, eating hamburgers at fast-food restaurants was identified as an important risk factor for Escherichia coli O157:H7 infections (87). Following a large foodborne outbreak of E. coli O157:H7 in the Pacific Northwest in 1993, public health recommendations and industry actions addressed this problem (3). Subsequently, a Foodborne Diseases Active Surveillance Network (FoodNet) case-control study of sporadic E. coli O157:H7 infections conducted during 1996–1997 revealed that eating hamburgers at fast-food restaurants was not associated with illness (55). Further regulatory and industry activities have led to reductions in the incidence of E. coli O157:H7 infections to levels meeting the National Health Objectives for 2010 (35)

Craig W. Hedberg, Division of Environmental Health Sciences, School of Public Health, University of Minnesota, Minneapolis, MN 55455.

575

SMP_Food Microbiology_CH22.indd

575

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

576

50

M e a n N o . O u tb re a k s / Y e a r

45 40

Salmonella

35

E. coli

30

Norovirus

25 20 15 10 5 0 1973 - 1982

1983-1987

1988-1992

1993-1997

1998-2002

2003-2008

Figure 22.1  Confirmed foodborne outbreaks associated with fresh fruits and vegetables, United States, 1973 to 2008. Source, CDC, Foodborne Outbreak Surveillance Summaries (26a). doi:10.1128/9781555818463.ch22f1

(Fig. 22.2). Similarly, following the 2002 to 2004 implementation of FDA rules requiring development of hazard analysis and critical control point (HACCP) plans for production of fruit juices, the number of juice­associated outbreaks reported to the Centers for Disease Control and Prevention (CDC) declined (96). Does the constantly changing epidemiology of foodborne diseases mean food is getting safer or less safe? When we get beyond looking at specific pathogens or foods, it is difficult to say. Food safety is the product of complex interactions between environmental, cultural, and socioeconomic factors (74). Before FoodNet was established as part of the CDC’s Emerging Infection Program, there was no systematic basis for even estimating the magnitude of foodborne illnesses in the United States (83). In 1999, Mead and colleagues at CDC (68) estimated that 76 million foodborne illnesses occur each year in the United States. This estimate was larger than previously reported estimates, and it has recently been revised downward to 48 million by Scallan and colleagues (82). Changes in the methodology and data used to estimate key parameters of this burden of illness (discussed below) make it difficult to determine whether the incidence of foodborne illness in the United States has actually decreased in the past decade. Similar methods are being used internationally to estimate the global burden of foodborne illness (30, 58). Given the increasing globalization of our food supply, having consistent and reliable estimates of foodborne illness disease

SMP_Food Microbiology_CH22.indd

576

burdens will be critical for evaluating and improving the performance of our food safety systems. This chapter will provide an introduction to epidemiology and epidemiologic methods as they are applied to problems of foodborne diseases. An understanding of epidemiology is important, because despite all of our best efforts to prevent foodborne diseases, humans remain the ultimate bioassay for low-level or sporadic contamination of our food supply (42). Epidemiologic methods of foodborne disease surveillance are needed to detect outbreaks, identify their causes, and assess the effectiveness of control measures. Epidemiologic data are also important in establishing food safety priorities, allocating food safety resources, stimulating public interest in food safety issues, establishing risk reduction strategies and public education campaigns, and evaluating the effectiveness of food safety programs (93). The examples provided in this chapter were drawn from the relevant literature and from the author’s own experiences as a foodborne disease epidemiologist. They were chosen to be illustrative, not necessarily representative. Epidemiologic data relevant to individual foodborne diseases are presented in their respective chapters.

EPIDEMIOLOGY Epidemiology is the study of events in populations. Events are usually thought of in terms of diseases. For example, FoodNet was established, in part, to determine actively

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases

577

O157 Incidence O157 At Retail O157 At Plant 1.4 1.2

2.5

1

2

0.8 1.5

1

National Health Objective for 2010 = 1.0 case / 100,000 population

0.6 0.4

0.5

0.2

0

% Positive Ground Beef Samples

O157 Cases/100,000 Population

3

0 1996

1997 1998

1999

2000 2001

2002

2003

2004 2005

2006

2007 2008

2009

Figure 22.2  Prevalence of E. coli O157:H7 in ground beef and incidence of O157 infections­ in FoodNet, 1996 to 2009. Adapted from reference 93a. doi:10.1128/9781555818463.ch22f2

how many laboratory-confirmed infections caused by Campylobacter, E. coli O157:H7, and Salmonella occur in the populations under surveillance. A second major FoodNet activity has been to conduct periodic surveys to determine how many diarrheal illnesses occur in the same populations that are under active surveillance for specific foodborne pathogens (8). By comparing the incidence and proportion of specific diseases attributable to foodborne transmission with the incidence of occurrence and medical evaluation for clinical syndromes, such as diarrhea, it becomes possible to determine what proportion of the diarrheal illnesses may be foodborne. This was the basis of the approach taken by Scallan and colleagues in determining that 48 million foodborne illnesses occur each year in the United States (82). Epidemiology also includes the study of factors associated with the occurrence of illnesses. The same population survey that FoodNet conducts to assess the frequency of diarrheal illness has also surveyed the population to find out how often people have potential exposures, such as eating undercooked hamburger, eating runny eggs, failing to wash their hands after handling raw chicken, or making a salad on the same cutting board they used to cut up raw chicken. Many of these factors are likely to

SMP_Food Microbiology_CH22.indd

577

be assessed in the context of outbreak investigations to enable identification of the factors actually contributing to the occurrence of an outbreak. The careful description of events in populations is generally called descriptive epidemiology. This process of determining characteristics of person, place, and time associated with illness occurrence forms the basis of all surveillance systems. The defining example of this process was the work of John Snow, during the London cholera epidemic of 1860 (75). He located cases on a map of London, saw a cluster around the Broad Street pump, and pulled the pump handle. The contaminated water source was shut down, and the outbreak ended. Such dramatic examples of consequential epidemiology are not common. However, surveillance of foodborne disease, based on the principles of descriptive epidemiology, remains a cornerstone on which other food safety activities are built. Surveillance is a function of public health agencies that can be highlighted by a classic investigation. In mid-September 1994, a microbiologist at the Minnesota Department of Health’s (MDH) Public Health Laboratory noted an increase in the number of Salmonella enterica serovar Enteritidis isolates being

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

578 identified. MDH epidemiologists began tracking the age, gender, and county of residence of the cases. After tracking new cases for a couple of weeks, it was apparent that cases were clustered in southeastern Minnesota. The conclusion was that a regional outbreak was occurring (46). In contrast to the mere description of events, the comparison of different rates at which they occur between groups is known as analytical epidemiology. This involves determining a measure of association and a measure of the variability, or uncertainty, in the measurement of the point estimate of that association. For etiologic studies to determine the cause of an outbreak, the critical measure is the difference between the risk of illness among those exposed to the “contaminated” food and the risk of illness among those who were not exposed. This is customarily presented as an odds ratio or risk ratio, depending on the form of the epidemiologic study design. The classic tool of analytical epidemiology for foodborne diseases is the use of a case-control study to identify the source of an outbreak. Case-control studies are frequently used in foodborne outbreak investigations, because in most investigations it is easier to identify ill persons (cases) than it is to identify the entire group at risk. Outbreak investigations initiated from pathogenspecific surveillance will typically include case-control studies. After concluding that a regional outbreak was occurring in southeastern Minnesota, epidemiologists constructed a questionnaire to ascertain what foods individual case patients had eaten in the 5 days before they became ill. The same questionnaire was used to interview healthy individuals of about the same age from the same communities. By comparing the responses of the cases to the responses of the healthy individuals, who served as controls, it was determined that the outbreak was due to consumption of commercially manufactured ice cream (46). Ten (67%) of 15 cases, but only 2 (13%) of 15 controls, had eaten brand A ice cream (matched odds ratio [OR] = 10.0; 95% confidence interval [CI] = 1.4 to 434). The magnitude of the OR is a measure of the strength of the association between illness and exposure (Table 22.1). The strength of this association left little doubt that the ice cream was the source of the outbreak. A public health intervention was initiated days before Salmonella was actually isolated from the suspected product. In addition, results of the case-control study suggested the problem likely involved multiple ice cream flavors and days of production. In response, a nationwide product recall was initiated. Further investigation led to the determination that the ice cream premix had been contaminated by raw egg

SMP_Food Microbiology_CH22.indd

578

Table 22.1  Evaluating the strength of association between

illness and exposure in epidemiologic studiesa

The strength of an association increases with the size of the odds ratio or relative risk: 1.  1 = no association; 2.  <5 = relatively weak association; 3.  5–10 = relatively strong association; 4.  >10 = very strong association Other key factors to consider when evaluating the association: 1.  Was time between exposure and illness onset consistent with a reasonable incubation period? 2.  Was a dose-response behavior observed? 3.  Do results of traceback suggest a common source? a

Adapted from reference 28.

residues in tanker trailers used to haul the premix from two other suppliers. Ultimately, what initially appeared to be a regional outbreak in southeastern Minnesota involved an estimated 224,000 illnesses, with cases reported from 41 states (46). Recent developments in molecular subtyping (discussed below) have made it possible to compare cases caused by an outbreak-associated strain with cases caused by ­unrelated strains (57, 67, 91). Because the occurrence of an outbreak implies a common source, the cases caused by unrelated strains with many different likely sources of exposure make a very efficient control group. This type of case-case comparison study has been critical for investigating Listeria outbreaks, where the risk for illness is dependent on underlying health conditions (38). The same methods are also being applied to outbreaks of E. coli O157:H7 and Salmonella infections (20, 41). The ability of molecular subtyping to discriminate outbreak­associated from unrelated cases may become increasingly important in regions where public health resources may not be adequate to recruit and interview healthy control subjects. Another closely related tool of analytical epidemiology is the cohort study. In outbreak investigations, retro­ spective cohort studies are conducted when an entire group (or population) with a common exposure can be identified. Individuals can be interviewed without prior knowledge of whether they were ill or not. Identifying groups based on their exposure status rather than their illness status allows for directly calculating a risk ratio for specific exposures. The risk ratio is the percentage of exposed persons who were ill divided by the percentage of unexposed persons who were ill. For example, in an outbreak of Campylobacter infections that was identified in a school in Madrid, Spain, 81 of 253 people interviewed developed a diarrheal illness consistent with

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases campylobacteriosis, for an overall attack rate of 32% (53). The epidemic curve suggested a common source exposure on a particular day. Information on foods eaten at the school on that day was available for 199 persons. Of these, 171 people ate a custard dessert and 77 became ill, for an attack rate of 45%. However, only 4 of 28 people who did not eat the custard became ill, for an attack rate of 14% (53). Thus, the risk ratio for eating custard was 3.2, meaning that exposed individuals (i.e., persons who ate the custard) were 3 times more likely to become ill than were individuals who were not exposed. Typical settings for cohort studies involve banquets, wedding receptions, and events where a list of all persons attending can be obtained. For specific exposures, such as a particular food item, the percentage of persons who became ill after eating the food can be determined. Hence, in addition to implicating the food item, it is possible to determine the attack rate among persons who ate it. Information on attack rates may also help in evaluating the factors contributing to the occurrence of the outbreak. For example, for many bacterial foodborne agents the attack rate may be related to the exposure dose. A high attack rate may imply high levels of contamination, such as may occur after prolonged temperature abuse. In both case-control and cohort studies, much attention is paid to the precision of the calculated estimate, as measured by the P value or 95% CI. Most conventions treat a P value of <0.05 as being significant. Similarly, 95% CIs are expected to exclude 1.0 when a significant exposure occurs. While these criteria have served as useful guides to data analysis over the years, it must always be noted that the main contributor to the P value and CI is the size of the sample. The larger the sample size, the more likely a statistically significant result may be obtained. Investigators should not discard potentially important associations just because they fail to achieve a P value of <0.05. Conversely, “highly significant” results may be totally spurious. The concern is that bias in sampling may have introduced reasons other than contamination of food for the observed difference in risk of illness. In particular, if food items are usually consumed together, one food item may be identified while the other is overlooked. Mistakes in the identification of the source of two large foodborne outbreaks appear to have resulted from this type of error (2, 48). In 1996, outbreaks of illness caused by Cyclospora were initially associated with strawberries from California, while in 2008, a nationwide outbreak of S. enterica serovar Saintpaul infections was associated with tomatoes. In both outbreaks, subsequent epidemiologic analyses

SMP_Food Microbiology_CH22.indd

579

579 identified the actual vehicles: raspberries and jalapeno peppers, respectively. The harm done by this type of error to the affected producers and to confidence in the public health system may be substantial. However, when potential sources of bias are properly controlled and accounted for, epidemiologic methods have demonstrated an impressive ability to accurately identify the source and contributing causes of foodborne outbreaks. As powerful as epidemiologic methods are, they do have limitations. In an outbreak setting where everyone ate a contaminated food item, there is no opportunity to demonstrate a difference in risk of illness between those who were exposed and those who were not. Alternative approaches, such as looking at the amount eaten or when it was eaten, may allow some epidemiologic discrimination, but results are not likely to be “significant.” For example, in the outbreaks of shigellosis associated with parsley, parsley was associated with illness at one restaurant in Minnesota, but not at a second restaurant. Chopped parsley was used on so many dishes at the second restaurant that most of the cases and the controls ate it (69). In epidemiologic studies, the absence of an association does not necessarily mean that the food item or exposure in question is safe. For example, in the FoodNet case-control study of sporadic E. coli O157:H7 infections, there was no association between sporadic infections and consumption of unpasteurized apple cider, even though concern about outbreaks associated with unpasteurized fruit juices led the FDA to develop the juice HACCP regulations (55, 96). Similarly, the same case-control study failed to reveal an association with attendance at day care, which has been associated with outbreaks of disease and remains an important potential source of infection. Neither unpasteurized apple cider nor attending child day care was identified as a primary source of sporadic E. coli O157:H7, but it would be misleading to interpret the results to imply that there is no risk. Similarly, in a FoodNet case-control study of S. Enteritidis infections, eating chicken prepared outside the home was the only significant risk factor associated with illness in a multivariate analysis (56). However, a U.S. Department of Agriculture (USDA) risk assessment estimated that 182,000 egg-associated S. Enteritidis infections occurred in the United States during 2000 (84). Additionally, from 1 May to 30 November 2010, over 1,900 illnesses were attributed to an outbreak of S. Enteritidis infections associated with related egg producers in Iowa (23). The concepts and methods of epidemiology can be used to examine the relationships between disease and all levels of food safety, from production and

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

580 ­ istribution to preparation and consumption. However, d as noted above, epidemiologic methods are more useful for some approaches than others. Although similar analytic techniques are used in outbreak investigations and case-control studies of sporadic infections, they are much more robust when applied to outbreak investigations. This difference can be illustrated by looking at two hypothetical investigations into the role of peanut butter as a source for Salmonella infections (19, 21). First, if 10% of sporadic Salmonella infections were caused by consumption of peanut butter, an unmatched case-control study with 600 cases and 1,200 controls would be required to produce a statistically significant result. Approximately one-half of the population consumes peanut butter during a given week (17). Hence, we would expect approximately 600 controls to have a history of eating peanut butter. The 10% of cases caused by consumption of peanut butter would account for 60 cases. However, the remaining cases would be expected to have peanut butter consumption patterns that reflected general consumption patterns in the population, as represented by controls. Hence, approximately 330 of the 600 cases would have reported eating peanut butter (Table 22.2). This distribution of exposure histories across cases and controls produces an odds ratio of 1.2 with a 95% CI of 1.004 to 1.5 and a P value of 0.045. Because peanut butter is a commonly eaten food item, a sporadic case-control study would have little power to attribute illnesses to it. In contrast, if an outbreak of salmonellosis was caused by a specific brand of peanut butter with a 40% market share and 60% of cases due to eating that specific brand of peanut butter, only 25 cases and 50 controls would be needed to implicate the product (Table 22.3). The distribution of exposure histories across cases and controls would produce an OR of 8.5 with a 95% CI of 4.9 to 14.6 and a P value of <0.001. In contrast to the large sample size required for the sporadic case-control study, with relatively few cases, a specific exposure can be reliably linked to a specific disease outbreak. Table 22.2  Hypothetical case-control study investigating

the role of peanut butter as a source for sporadic Salmonella infectionsa Exposure history

No. of cases

No. of controls

Total no.

Ate peanut butterb

330

  600

  930

Did not eat peanut butter

270

  600

  870

Total

600

1,200

1,800

a b

OR = 1.2; 95% CI = 1.004 to 1.5; P = 0.045. Control exposure frequency approximated from reference 17.

SMP_Food Microbiology_CH22.indd

580

Table 22.3  Hypothetical case-control study investigating the

role of a specific brand of peanut butter as the source for a Salmonella outbreaka No. of cases

No. of controls

Total no.

Ate specific brand of peanut butterb

17

10

42

Did not eat specific brand or did not eat peanut butter

 8

40

33

Total

25

50

75

Exposure history

OR = 8.5; 95% CI = 4.9 to 14.6; P < 0.001. b Control exposure frequency approximated from reference 17, with assumption that the specific brand had a 40% market share. For cases, 60% assumed due to consumption of the specific brand. Cases not due to the specific brand would have reported exposure to the specific brand similar to that of controls. a

AGENTS AND TRANSMISSION Information about specific agents and their transmission is contained in their respective chapters. However, in discussing the epidemiology of foodborne diseases, it is important to keep in mind the chain of infection. This includes the agent, the reservoir that contains the agent, a means of escape from the reservoir, a mode of transmission to a susceptible host, and a means of entry to the host. These elements define the agent-host environment that results in the occurrence of illness. Specific agents vary with respect to the types of illness they cause, from self-limited gastroenteritis associated with bacterial enterotoxins to invasive and life-threatening diseases associated with Listeria monocytogenes. Some agents, such as norovirus, are cleared from the gastrointestinal tract in a matter of days, whereas others, such as Giardia, may cause prolonged carriage, over several weeks. Infectious doses may range from ingestion of a few E. coli O157:H7 cells to hundreds of thousands of Clostridium perfringens cells. Incubation periods range from a few hours for staphylococcal intoxication to weeks for viral hepatitis type A. Some agents, such as Shigella, infect only humans, whereas Salmonella, Campylobacter, and E. coli O157:H7 primarily originate from animal sources. Many agents are associated with typical food vehicles associated with their animal reservoirs, such as S. Enteritidis with eggs and E. coli O157:H7 with ground beef, or with food vehicles susceptible to environmental contamination from the animal reservoir, such as E. coli O157:H7 with leafy green vegetables (16). The characteristics of specific agents affect the potential sources of contamination during food production and preparation and form the basis for hazard analysis for specific foods. They also help shape the conduct of

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases foodborne disease investigations and allow the investigation of source and transmission to be developed around the particular features of the disease.

FOODBORNE DISEASE SURVEILLANCE Surveillance involves the systematic collection of data with analysis and dissemination of results (12). Surveillance systems may be passive or active, national or regional in scope, or based on a sentinel system of individual sites designed to provide estimates that can be generalized. Types of systems, their use, and their limitations are discussed below. There are four primary purposes for conducting foodborne disease surveillance: 1. Identify, control, and prevent outbreaks of foodborne illness 2. Determine the causes of foodborne illness 3. Monitor trends in occurrence of foodborne illness 4. Quantify the magnitude of foodborne illness. “Pulling the pump handle” is the goal of all public health epidemiologists conducting foodborne disease surveillance. Hence, it is necessary to identify, control, and prevent outbreaks of foodborne illness. For products with a long shelf life, such as cured meat products, frozen foods, or cereals, even a relatively long investigation can identify a contaminated product quickly enough to remove it from the marketplace in time to prevent many illnesses (38, 46). For highly perishable products, such as fresh fruits and vegetables, often the outbreak will have run its course before the food item can be implicated (44, 86). Even if it is impossible to directly intervene to control the outbreak, there is still value in determining the causes of foodborne illness. HACCP systems for protecting the safety of foods rely on accurate knowledge of potential hazards (70). Many of these hazards, either in terms of specific agents, specific food ingredients, or various agent-food interactions were originally determined as a result of foodborne illness surveillance. Because food sources and foodborne disease agents are constantly changing, hazard analysis is an ongoing process that requires continuous support from public health surveillance of foodborne diseases. Additionally, Scallan and colleagues’ estimates of the burden of foodborne illness suggest that most foodborne illnesses are caused by unspecified agents (82). As the experience with Cyclospora illness outbreaks from Guatemalan raspberries reveals, foodborne disease surveillance may be the only way to identify both new causes of foodborne disease and potential foodborne hazards.

SMP_Food Microbiology_CH22.indd

581

581 A third traditional purpose for conducting foodborne disease surveillance is to monitor trends in occurrence of foodborne illness. This is important to identify priorities for new food safety activities and to monitor the effectiveness of existing programs. For example, during the 1980s, the increased occurrence of sporadic S. Enteritidis infections and outbreaks in New England led to the identification of a new problem with S. Enteritidis contamination of grade A shell eggs (89). In the United States, the USDA and FDA worked with the egg industry to develop and implement a number of control strategies (49) culminating with FDA’s implementation of the Egg Safety Final Rule in 2010 (31). The incidence of S. Enteritidis infections in FoodNet sites declined 46% from 1996 to 1999, suggesting that these control strategies have had a positive impact (66). Importantly, the incidence decreased by 71% in Connecticut, the FoodNet state with the highest incidence of S. Enteritidis infections (66). Unfortunately, not all producers complied with the voluntary strategies, and a large multistate outbreak occurred shortly before implementation of the final rule (23). FoodNet surveys of the population, physicians, and clinical laboratories linked to active surveillance of laboratory-confirmed infections provide the most comprehensive data available from which to estimate the magnitude of foodborne diseases. The burden of illness estimates produced by Mead and revised by Scallan have been used as a basis for setting public health priorities (68, 82). FoodNet trend analysis for specific agents has been used for evaluating the effectiveness of prevention measures. However, significant gaps remain in our knowledge base. Both Mead and Scallan concluded that approximately 80% of estimated foodborne illnesses cannot be attributed to known agents (68, 82). Moreover, timely estimates for the burden of illness associated with all major foodborne pathogens, not just those under FoodNet surveillance, are needed to improve our evaluation of prevention measure effectiveness.

Methods of Foodborne Disease Surveillance

Foodborne disease surveillance requires close collaboration between acute-disease epidemiologists, public health laboratories, and environmental health specialists to determine the likely sources of exposure and routes of food contamination. There are four primary components of foodborne disease surveillance (Table 22.4). The first component is identification and reporting of outbreaks associated with events and establishments, usually by ill persons who were part of the outbreak. There is a bias towards detecting outbreaks at events such as a wedding, where a large group gathers with a

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

582 Table 22.4  Primary components of foodborne disease

surveillance

Investigation of outbreaks associated with events and establishments. Pathogen-specific surveillance to identify clusters of cases caused by the same microorganism. Determining risk factors for sporadic cases of infection with common foodborne pathogens. Population surveillance to determine the frequency of gastrointestinal illnesses, health care-seeking behavior, food consumption, and personal prevention measures.

common experience and has some reason to discuss it later (60). People eating at a restaurant may never know that their illness was also experienced by other restaurant patrons, who are unknown to them. Outbreaks associated with events and establishments require prompt and thorough investigation to identify both the agent and the source. These outbreaks are usually recognized because of the occurrence of common symptoms, such as diarrhea and vomiting, and are reported to public health officials before a specific diagnosis has been established (60). Unfortunately, in many outbreak investigations, laboratory testing is not sought, not available, or not adequate to identify the causative agent. Of 2,751 outbreaks of foodborne disease reported to the CDC from 1993 to 1997, 1,873 (68%) were classified as of unknown etiology (73).

From 1998 to 2002, 67% of reported outbreaks were classified as of unknown etiology (15). However, many of these likely were outbreaks of norovirus gastroenteritis, which is not detected by clinical laboratory testing. With the development and use of norovirus testing by many state public health laboratories, the proportion of confirmed norovirus outbreaks among foodborne outbreaks in the United States increased from 1% in 1991 to 12% in 2000 (97). The overall proportion of outbreaks with an unknown etiology decreased to 29% in 2006, when suspected and confirmed norovirus outbreaks accounted for 40% of all confirmed foodborne outbreaks reported to CDC (25). This shift in attribution of unknown etiologies to norovirus is highlighted in Fig. 22.3. However, the shift in attribution to norovirus was not solely due to improved access to laboratory testing. As seen in Fig 22.4, approximately one-third to one-half of outbreaks classified as norovirus were not confirmed but were suspected based on clinical and epidemiologic features consistent with noroviruses and distinct from outbreaks caused by known bacterial infections or toxins (54). These include median periods of 24 to 48 hours for incubation and of 12 to 60 hours for the duration of symptoms and a relatively high proportion of cases experiencing vomiting (54). Such epidemiologic profiling of outbreaks is useful to guide laboratory testing as well. Clinical laboratories and many public health laboratories do not test for the presence of enterotoxigenic E. coli. However, if a high

900

Unknown Etiology

No. Outbreaks / Year

800

Norovirus

700 600 500 400 300 200 100 0

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Figure 22.3  Changing distribution of outbreaks attributed to norovirus or of unknown e­ tiology, reported to Centers for Disease Control and Prevention, Foodborne Disease Outbreak Surveillance System, 1998–2008. Adapted from reference 26a. doi:10.1128/9781555818463.ch22f3

SMP_Food Microbiology_CH22.indd

582

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases

583

400 Norovirus- confirmed

No. Outbreaks / Year

350

Norovirus- suspected

300 250 200 150 100 50 0 1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Figure 22.4  Numbers of confirmed and suspected outbreaks of norovirus reported to Centers for Disease Control and Prevention, Foodborne Disease Outbreak Surveillance System, 1998 to 2008. Adapted from reference 26a. doi:10.1128/9781555818463.ch22f4

proportion of cases became ill several days after a common exposure and experienced diarrhea and cramps, with little vomiting or fever, the investigator should seek specialized laboratory testing for enterotoxigenic E. coli (29). Just such a clinical and epidemiologic profile led to the identification of an atypical enteropathogenic strain of E. coli serogroup O39 as the cause of a foodborne outbreak at a restaurant (43). Similarly, investigation of a series of event-associated outbreaks with unusually long median incubation periods of a week led to the recognition of Cyclospora as a foodborne pathogen (48). The investigation of outbreaks associated with events is the oldest form of foodborne disease surveillance and is frequently overlooked in discussions of how to improve foodborne disease surveillance, even though approximately three-quarters of foodborne outbreaks are detected through this route (60). Moreover, it remains the primary way to identify “new” foodborne pathogens, such as Cyclospora and diarrheagenic E. coli, which are not routinely identified by clinical laboratories. Across much of the developing world, recognition of outbreaks associated with events or establishments, usually involving large groups and short incubation periods, is the extent of foodborne disease surveillance available (Table 22.5). This basic level of foodborne disease surveillance requires very little public health infrastructure. A major focus of public health surveillance for foodborne illnesses has been to improve pathogen-specific surveillance (91). Serotype-based surveillance for

SMP_Food Microbiology_CH22.indd

583

Salmonella infections reported electronically to CDC from state public health laboratories through the Public Health Laboratory Information System enabled the detection of community outbreaks associated with large clusters, uncommon serotypes, and geographic and temporal clustering; many were associated with previously unrecognized vehicles, most notably fresh fruits and vegetables (44, 86). Outbreaks of salmonellosis caused by cantaloupes, tomatoes, and alfalfa sprouts were all recognized because of unusual temporal clusters of cases with an uncommon serotype. Epidemiologic investigations of these cases identified the sources.

Table 22.5  Levels of foodborne disease surveillance system

acuity based on ability to detect types of outbreaks Level of surveillance

Type of outbreaks detected

Basic

Outbreaks associated with events or establishments: large groups, short incubation periods.

Intermediate

Community outbreaks associated with large clusters, uncommon serotypes, geographic and temporal clustering.

Advanced

Widely dispersed outbreaks, caused by common serotypes. Sensitive, nonculture methods available to identify a wide range of pathogens for which no clinical laboratory testing is available.

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

584 An automated Salmonella outbreak detection algorithm (SODA) was developed to look for unusual case clusters based on the 5-year mean number of cases from the same geographic area and week of the year (51). Detection of clusters by SODA has helped confirm the existence of multistate outbreaks of S. Stanley infections associated with alfalfa sprouts in 1995, and S. Agona infections associated with toasted oats cereal in 1998 (10, 63). However, SODA was not especially sensitive in detecting outbreaks caused by common serotypes, such as S. Typhimurium or S. Enteritidis. For these serovars, additional subtyping, such as phage typing or pulsed-field gel electrophoresis (PFGE), has been used to increase strain discrimination and facilitate outbreak detection. For example, outbreaks of S. Enteritidis infection associated with raw almonds in Canada and the United States and raw-milk cheese in France were detected because they were caused by S. Enteritidis strains having uncommon phage types (40, 52). Molecular subtyping schemes, such as PFGE, have greatly improved the investigation of outbreaks by distinguishing unrelated sporadic cases from the main outbreak-associated strain. The ability to distinguish specific subtypes among relatively common pathogens, such as E. coli O157:H7 and S. Typhimurium, is the basis of the National Molecular Subtyping Network (PulseNet), which has replaced the Public Health Laboratory Information System as the primary tool for pathogenspecific surveillance of E. coli O157:H7 and Salmonella in the United States (91). PulseNet takes advantage of the combined revolutions in molecular biology and information technology. Highly reproducible PFGE patterns are generated under standardized conditions, and the PFGE patterns can be transmitted electronically between participating laboratories. PFGE subtyping improves the specificity of the case definition and makes it more likely that the source of the outbreak can be identified. PFGE has been used in outbreaks caused by E. coli O157:H7, Salmonella, and Shigella (69, 79). For example, an outbreak of E. coli O157:H7 infections in Colorado was associated with consumption of a nationally distributed ground beef product. Within days, it was possible to compare the PFGE profile of the outbreak strain to PFGE patterns of E. coli O157:H7 isolates throughout the United States (9). This gives PulseNet the potential to be the “backbone” of a public health surveillance system that can provide truly national surveillance for a variety of foodborne pathogens in a manner timely enough to be an early warning system for outbreaks of foodborne illness. Furthermore, the usefulness and reliability of PulseNet has led to the development of PulseNet International.

SMP_Food Microbiology_CH22.indd

584

The adoption of standard laboratory protocols and information systems will facilitate detection of the global spread of emerging foodborne pathogens. The public health utility of incorporating molecular subtyping into routine surveillance of E. coli O157:H7 and Salmonella has been repeatedly demonstrated (4). However, many subtype-cluster investigations initiated on a national basis fail to identify a common source exposure (37). This may reflect the difficulties in conducting timely multistate outbreak investigations. For example, between 1 January 2007 and 29 October 2007, at least 272 isolates of Salmonella I 4,[5],12:i:– with an indistinguishable genetic fingerprint were reported to PulseNet from ill persons in 35 states (18). The increased occurrence of this particular subtype led to an attempt to collate responses from individual state investigations by the end of June. By mid-August, the failure to detect any potential common source of exposure led CDC to request that individual states use a common, standardized questionnaire. In September, CDC investigators changed direction again and conducted a series of broad-based hypothesis-generating interviews with a handful of recent cases. This led to the development of a second, standardized questionnaire that was sent to all of the states participating in the multistate investigation. Ultimately, the source of the outbreak was detected by investigators at the MDH. Foodborne disease surveillance in Minnesota is centralized at MDH, and MDH uses a group of public health graduate students, collectively known as “Team Diarrhea,” to conduct case interviews using a detailed exposure source questionnaire. During the course of this outbreak investigation, the first two Minnesota cases revealed no remarkable histories of exposure. However, the third case reported consuming a particular brand of frozen pot pie. This exposure was viewed as being significant because the demographics of this outbreak were similar to those of earlier outbreaks associated with frozen food items (88). This new exposure was added to the basic questionnaire, and the next case interviewed also reported eating the same brand of frozen pot pies. That same day, the first two cases were reinterviewed, and both had eaten the same brand of frozen pot pies. This information was relayed to CDC and the other states, the pot pies were confirmed as the source of the outbreak, and the manufacturer initiated a nationwide recall. After months of nonproductive investigations across multiple states, the source of the outbreak was identified within a few days following the best practices developed by MDH (28). Molecular subtyping also can be used to enhance the investigation of outbreaks associated with events or es-

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases tablishments, such as when multiple, apparently independent outbreaks of Shigella sonnei infections across the United States and Canada were determined to have been caused by a common strain (69). Although the independent investigations did not identify the cause, the linkage of the outbreaks allowed investigators to identify parsley that had been imported from Mexico as the common source. All outbreak investigations require close collaboration with environmental health specialists and field investigators to identify and trace the source of food items that may have been a vehicle for the causative agent. Although product tracebacks are frequently started only after a food item has been implicated, detailed product information is actually critical to epidemiologic analysis. In the outbreak of S. Enteritidis infection associated with brand A ice cream, for example, if cases and controls had initially been asked only if they had eaten ice cream, investigators would never have identified the source of the outbreak (46). Thirteen (87%) cases and 10 (67%) controls ate some type of ice cream (OR = 2.5; 95% CI = 0.4 to 26.3). These data would not have raised suspicions about ice cream being the source of the outbreak. Hence, collecting product source information should begin as early in the investigation as possible. The importance of this point is further emphasized by the increasing occurrence of outbreaks associated with foods such as salads and salsas that routinely contain multiple ingredients. Had source tracing of tomatoes been pursued before tomatoes were implicated as the source of the S. Saintpaul outbreak in 1998, alternative conclusions might have been drawn earlier. However, it is likely that investigators were anticipating tomatoes to be identified as the source, given the frequency of recent tomato-associated outbreaks and the apparent strength of the association based on the results of the initial casecontrol study (2). A Global Foodborne Infections Network has been established by the World Health Organization (WHO) to promote pathogen-specific surveillance for Salmonella and E. coli O157:H7 and provide training to international partners. The Global Salmonella Surveillance platform has accumulated information on over 1.5 million human isolates from more than 80 countries (65). However, most of the participants represent developed countries, primarily in Europe and the Americas. In addition, rates of submission of serotype information appear to have declined from 1995 to 2009 for several countries (65). Declining public health resources in developed and developing countries may severely limit the usefulness of this system and the international distribution of PulseNet. Building the capacity to conduct mo-

SMP_Food Microbiology_CH22.indd

585

585 lecular subtype surveillance is necessary to detect widely dispersed outbreaks caused by common serotypes. Failure to invest in these advanced surveillance systems will consign developing countries to a basic level of surveillance (Table 22.5). Determining risk factors for sporadic cases of infection with common foodborne pathogens is a third major component of foodborne disease surveillance. This can help identify targets for intervention and provide a basis to evaluate its effectiveness. Case-control studies of sporadic S. Enteritidis infections in the United States and Europe helped establish the role of Grade A shell eggs in the epidemiology of these infections. Case-control studies of Campylobacter infections helped establish chicken as a primary source of Campylobacter (90). Similarly, case-control studies of sporadic E. coli O157:H7 infections confirmed the importance of ground beef, particularly undercooked ground beef, as a source of the pathogen (59, 87). However, as discussed above, this method is inherently limited in its ability to evaluate exposure to commonly eaten food items. FoodNet has conducted population-based, case-control studies for E. coli O157:H7, Salmonella serogroups B and D, Campylobacter, Cryptosporidium, and L. monocytogenes. These have provided updated and original estimates of the proportion of these infections that are attributable to specific food items. Results of the E. coli O157:H7 case-control study revealed that eating hamburgers at a fast-food restaurant was not associated with illness, as it had been in previous case-control studies (55). This appears to be due to the fact that fast-food restaurants are thoroughly cooking the hamburgers they serve. At home, or at table service restaurants, where consumers have a choice to order an undercooked hamburger, eating undercooked hamburgers was associated with illness (55). Case-control studies of Campylobacter, S. Enteritidis, and S. enterica serovar Heidelberg infections also revealed associations with eating out, not at home (47, 56). Other interesting findings included associations between S. Enteritidis infection and chicken consumption, S. Heidelberg infection and egg consumption, and both Campylobacter and Cryptosporidium infections with foreign travel (80). The final important component of foodborne disease surveillance is population surveillance to determine the frequency of gastrointestinal illnesses, health care-seeking behavior, food consumption, and personal prevention measures. This type of syndrome-specific surveillance is not usually conducted by state or local health departments. However, these measures formed the basis of CDC’s recent estimates that 48 million foodborne diseases occur each year in the United States. These data

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

586 are also useful for understanding the results of active surveillance. By describing when and why people seek medical care when they are ill, these data can help to estimate how many illnesses occur for each one that gets a specific diagnosis. Finally, the population-based food consumption data are being used to estimate the proportion of specific diseases attributable to specific exposures in the FoodNet case-control studies. Following models developed to build on FoodNet findings, international collaborations are being coordinated through WHO to strengthen disease surveillance and to determine the burden of acute gastroenteritis (30). Results of these studies are intended to provide national estimates that can be aggregated to provide global estimates of (i) the burden associated with acute gastroenteritis of foodborne origin, (ii) the burden caused by specific pathogens commonly transmitted by food, and (iii) the burden caused by specific foods or food groups. Although there are no reliable estimates for the global burden of foodborne illness, other national estimates in developed countries are consistent with the illness rates estimated for the United States (30). The global burden of Salmonella gastroenteritis has been estimated at approximately 94 million cases and 155,000 deaths annually (65).

Surveillance for Food Hazards

Risk assessments for specific pathogens such as E. coli O157:H7 in ground beef or S. Enteritidis in shell eggs require making measurements or assumptions about many parameters. At every step from farm to table, potential sources of contamination can be identified, measured, and modeled to determine the relative contribution to the overall risk of foodborne illness. For many foodborne pathogens of public health importance, contamination of finished food products is a sporadic and rare occurrence. While end product pathogen testing is difficult to justify as a primary prevention measure, end product testing may provide important information as part of a comprehensive quality assurance testing program. Given the scale of modern agriculture and food production systems, contamination levels below the statistical and microbiological threshold can still be large public health problems. For example, internal contamination of eggs with S. Enteritidis is rare, occurring in an estimated 1 in 20,000 eggs on a national basis in the United States, with a range from 2.5 to 62.5 per 10,000 eggs from environmentally positive flocks (50). Despite this sporadic occurrence, consumption of contaminated shell eggs is estimated to cause more than 100,000 illnesses per year (84). For this reason, surrogate measures of contamination, such as identify-

SMP_Food Microbiology_CH22.indd

586

ing infected hens or environmental contamination in egg laying houses, have been widely accepted as being more efficient and effective than testing individual eggs. Alternatively, measuring a relatively common bacterium such as commensal strains of E. coli may provide surveillance for control of operations that can be independently associated with the degree of risk for disease transmission. For example, in a series of baseline surveys of beef slaughter plants and ground beef conducted by USDA in preparation for their introduction of HACCP regulation in the meat industry, 8.2% of steer and heifer carcasses and 15.8% of cow and bull carcasses were contaminated by commensal E. coli (32, 33). However, only 4 (0.2%) of 2,081 steer and heifer carcasses and none of 2,112 cow and bull carcasses were contaminated with E. coli O157:H7. Of 563 finished ground beef samples, 78.6% were contaminated by commensal E. coli, but none by E. coli O157:H7 (34). Hence, microbiological testing for commensal E. coli may provide useful information for monitoring HACCP, even though preventing contamination by E. coli O157:H7 is the desired outcome. In addition to microbiological surveys of food products conducted by USDA and FDA, the USDA implemented the National Animal Health Monitoring System in 1983 to conduct national studies and compile data from industry sources to address emerging issues such as the association between calf management practices and the presence of E. coli O157:H7 and Salmonella in cattle herds (36, 62). The veterinary Diagnostic Laboratory Reporting System compiles and analyzes reports from state veterinary diagnostic laboratories to assess trends in infectious diseases among food animals (81). Behavioral risk factors for foodborne disease range from choosing to eat alfalfa sprouts or undercooked ground beef to failing to wash hands between using a toilet and making a salad. These factors are evaluated and can be identified during the course of outbreak investigations. The frequency of these behaviors in the general population can be assessed through surveys, such as those being conducted as part of FoodNet. For example, 7.7% of respondents to the FoodNet population survey conducted from 1998 to 1999 reported eating alfalfa sprouts during the 7 days before the interview (11). However, 10.5% of California and Oregon residents reported eating sprouts. Hence, it was not entirely surprising that sprout-associated outbreaks of salmonellosis were more common on the west coast than in the rest of the United States (94). Among other potentially important food exposures, 25% of persons who ate eggs consumed runny eggs; 11% of persons who ate hamburgers ate hamburgers that were pink; 4.4% of respon-

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases dents drank unpasteurized apple juice or apple cider; 3.4% drank unpasteurized milk; and 2.5% ate fresh oysters (11). Among persons who handled raw chicken, 88.6% reported washing their hands with soap and water after touching the raw chicken, but only 48.3% of respondents had a meat thermometer in the home (11). Information of this type can provide important perspectives for understanding the occurrence of foodborne disease outbreaks and sporadic infections. It also serves as the basis for developing and modifying public education campaigns to improve food safety, such as the Partnership for Food Safety Education’s national public education campaign, Fight Bac! (6). For example, the 2002-2003 FoodNet population survey revealed a decline in alfalfa sprout consumption of 51% overall and 39% in California and Oregon (13). However, while rates of alfalfa sprout consumption declined a further 17% in Oregon in the 2006 FoodNet population survey, they stabilized in California and increased 16% overall (17). As evidenced by a large, multistate outbreak of S. Newport infections associated with alfalfa sprouts in 2009, FDA guidelines for alfalfa sprout production have not been adequate to prevent the distribution of sprouts grown from contaminated seed stock (22). Contributing factors to the occurrence of foodborne disease outbreaks are compiled by CDC from national surveillance data and by individual states. National data are typically published in summaries covering 5-year periods, but the last time that contributing factor data were summarized was for 1998 to 2002 (15). These summaries include detailed tables, by year, but very little in the way of synthesis or even trend analysis. For example, in outbreaks for which contributing factors were reported between 1988 and 1997, the two most commonly reported contributing factors were improper holding temperatures (60%) and poor personal hygiene (34%) (1, 73). During the first 5 years of this period, poor personal hygiene declined as a contributing factor, from 44% of outbreaks in 1988 to 29% of outbreaks in 1992. During the second 5-year period, the role of poor personal hygiene increased from 27% of outbreaks in 1993 to 38% of outbreaks in 1997. It would be useful to know whether these data represented actual trends in the occurrence of foodborne disease outbreaks, or trends in the way outbreaks are investigated and reported, or mere variations in passively collected data that do not correlate with the actual occurrence of foodborne diseases. Unfortunately, contributing factors were reported from only 58% of all outbreaks reported to CDC, and not all outbreaks were reported (1, 73). Bias in what is reported is the largest systematic problem for interpreting national foodborne disease outbreak surveillance data.

SMP_Food Microbiology_CH22.indd

587

587 Beginning in 1998, CDC began reporting an expanded list of 14 factors that could contribute to contamination of foods, 5 factors associated with survival or lack of inactivation of foodborne pathogens, and 12 factors associated with their proliferation and amplification (5). While these refined summaries were intended to provide better information on conditions found to contribute to the occurrence of individual outbreaks, contributing factors were reported from only 46% of outbreaks reported from 1998 to 2003. During this period, bare-hand contact with food was identified as a contributing factor in 25% of outbreaks, and handling of food by an infected food worker was identified as a contributing factor in 20%. In contrast to these results, the Environmental Health Specialists Network systematically evaluated outbreaks in restaurants occurring within selected FoodNet sites (45). Norovirus accounted for more than 40% of outbreaks, with handling of food by an infected food worker and bare-hand contact with food identified as contributing factors in 65% and 35% of outbreaks, respectively (45). However, to put this in perspective, it would be useful to know how often food workers in restaurants work while ill or have bare-hand contact with ready-to-eat foods. For comparison, 5% of food workers responding to a FoodNet population survey reported that they had worked while ill with vomiting or diarrhea (39). In a 2003 survey on the occurrence of foodborne illness risk factors in various food service settings, FDA reported that 57% of observations on the prevention of hand contact with ready-to-eat foods in full-service restaurants were out of compliance (7). Hence, the proportion of norovirus outbreaks attributable to these contributing factors is not simply the sum of all outbreaks in which they were identified. These indirect comparisons suggest that the presence of an infected food worker is by far the most important risk factor for outbreaks of norovirus in restaurants. This example highlights a critical need for addressing likely food hazards; first to determine the proportion of illnesses and outbreaks attributable to the hazard, and then to estimate the predictive value of the hazard for causing disease. This is part of the process of risk assessment and is an important application of foodborne disease epidemiology (71). In a restaurant setting, we may know how many times bare-hand contact with foods or failure to wash hands is identified as a contributing factor in an outbreak of foodborne illness, but what is the frequency of this failure in restaurants? As noted above, the FDA began to address these data needs in 1997 with the creation of the FDA Retail Food Program Database of Foodborne Illness Risk Factors. This was followed by a second survey in 2003, with a third in 2008 to assess

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

588 trends (7). Results of these surveys revealed better compliance with hand washing in fast-food restaurants but not in full-service restaurants (7). Both fast-food and full-service restaurants increased compliance with provisions regarding bare-hand contact with ready-to-eat foods. These studies provide data to address important environmental questions using epidemiologic methods applied to surveillance for the hazards.

Evaluating Food Safety Systems

Public health surveillance of foodborne disease is critical­ to the performance of food safety systems that are based on HACCP plans. Surveillance is required to identify new hazards. It also provides the ultimate feedback on the efficacy of HACCP plans. Recalling the outbreak of salmonellosis due to commercial ice cream, cross­contamination during transportation was identified as a hazard that had not been addressed in the manufacturer’s HACCP plan. Epidemiologic data can also measure the success of food safety interventions. After implementation of the nationwide recall of brand A ice cream, a survey of brand A customers in Georgia revealed that 50% had not heard about the recall until 5 days after it was issued and 9% did not hear about it at all (64). Furthermore, 26% of persons who heard the recall did not initially believe the products were unsafe, and many subsequently ate the ice cream. Of an estimated 11,000 cases of illness in Georgia associated with this outbreak, 16% were exposed after the recall (64). As noted earlier, case-control studies of sporadic E. coli O157:H7 conducted before and after the Jack-inBox E. coli O157:H7 outbreak of 1993 revealed that there was a change in the risk of illness associated with eating hamburgers at fast-food restaurants. The recommendations that followed the outbreak appeared to reduce the risk of eating hamburgers in fast-food restaurants. This appears to be due to thorough cooking of hamburger in these restaurants, since overall incidence rates for E. coli O157:H7 did not decline during this time period and the improved sensitivity of testing for E. coli O157:H7 provided a more complete picture of its prevalence (Fig. 22.2). E. coli O157:H7 was isolated from fewer than 1 in 1,000 samples of ground beef sampled at the retail level by USDA from 1995 to 1998 (35). Following the introduction of more-sensitive culture methods late in 1999, the level of E. coli O157:H7-positive samples increased to 3.2 per 1,000. In 2000, more than 1% of samples tested positive. This increase was likely due primarily to the use of moresensitive testing methods rather than more contamination of the finished product (72). However, following the 2002 USDA Food Safety Inspection Service notice

SMP_Food Microbiology_CH22.indd

588

to manufacturers of raw ground beef products that they must consider E. coli O157:H7 a potential hazard in their HACCP plans, the percentage of positive samples at retail decreased markedly, and the incidence of E. coli O157:H7 infections decreased as well (35, 72) (Fig. 22.2). However, available microbiological and epidemiologic data continue to support the need for terminal pathogen inactivation treatments, such as cooking or irradiation, to further reduce the risks of E. coli O157:H7 infections. The FoodNet surveillance network was established, in part, to provide a tool to evaluate the public health impact of the USDA’s implementation of meat processing plant HACCP systems. FoodNet sites provide population-based measures of the incidence of several important foodborne pathogens. For purposes of trend analysis, the first 3 years of FoodNet surveillance (1996 to 1998) were established as a baseline. From 1998 to 2004, declines in the reported incidence of infections caused by Campylobacter, Listeria, Salmonella, Shiga toxin-producing E. coli (STEC) O157, Shigella, and Yersinia were observed, but the incidence of Vibrio infection increased. However, since 2004, the rates of these infections have been stable, indicating a need for further efforts from industry, regulatory agencies, and the public to improve prevention of foodborne infections (14, 24). It has been noted that declining incidence rates of foodborne illnesses from 1998 to 2004 occurred at the same time as implementation of changes in meat and poultry processing plants to enhance the microbiological safety of their products, revisions to the FDA Food Code for restaurants, and increased attention to good agricultural practices for fresh fruits and vegetables and eggs on farms (14). These are the types of effects FoodNet was designed to measure. However, determining cause and effect is not as simple as measuring the trends. To begin with, there is considerable regional variability in the incidence of specific infections among the FoodNet sites. For example, in 1996, the reported incidence of Campylobacter infections ranged from 58 per 100,000 in California to 14 per 100,000 in Georgia (8). Similarly, the incidence of E. coli O157:H7 infections ranged from 5 per 100,000 in Minnesota to 0.6 per 100,000 in Georgia (8). Although the rates of Salmonella infections did not vary much between sites, the distribution of serotypes did; S. Enteritidis was the most common serotype in Connecticut but was rare in Georgia. Overall, the incidence of Campylobacter infections in FoodNet declined from 25 per 100,000 in 1996 to 12.9

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases per 100,000 in 2004. However, a large portion of that overall trend was due to marked declines in the incidence of Campylobacter infections in California, from 58 per 100,000 in 1996 to 28.6 per 100,000 in 2004 (14). Campylobacter rates in California and overall have remained stable through 2009 (24). Data for both incidence and trends continue to indicate considerable regional variability that cannot be readily explained. Hence, it is important to remember that the underlying factors that contribute to the regional variability of these infections may also lead to regional differences in how they respond to prevention and control measures.

SUMMARY Humans remain the ultimate bioassay for low-level or sporadic contamination of our food supply (42). Epidemiologic methods of foodborne disease surveillance are powerful tools because they take advantage of events that are occurring throughout the population. This population-based lens, focused by advances in molecular subtyping and information technology available to public health laboratories, is particularly well suited to dealing with foodborne diseases associated with massproduced and widely distributed food products. Epidemiologic methods of foodborne disease surveillance are needed to detect outbreaks. In outbreak settings, rapid ascertainment of detailed information on food source and production is needed to confirm the cause of the outbreak and identify contributing factors. Epidemiologic data are also very important to establish food safety priorities, allocate food safety resources, stimulate public interest in food safety issues, establish risk reduction strategies and public education campaigns, and evaluate the effectiveness of food safety programs. The principles and methods of epidemiologic surveillance can be applied to surveillance for food hazards as well as to surveillance for foodborne diseases. The same methods of observation and analysis that form the basis of foodborne disease surveillance in the United States are also being practiced across the world in developed countries and are being adapted for use in many developing countries. Since national food supplies are rapidly becoming global in origin, the need for an international system for foodborne disease surveillance exists as well. Models such as PulseNet provide opportunities to conduct multinational surveillance for at least the major bacterial foodborne disease agents. Because foodborne disease problems imported into one country may represent disease problems endemic to the food-producing country, growing awareness of these problems could stimulate investment in interventions

SMP_Food Microbiology_CH22.indd

589

589 that improve the health of both countries. In particular, the global trade in fresh produce, spices, and seed stock requires safe water distribution and sanitary sewage disposal systems wherever these products are handled.

References   1. Bean, N. H., J. S. Goulding, C. Lao, and F. J. Angulo. 1996. Surveillance for foodborne-disease outbreaks—United States, 1988-1992. MMWR Surveill. Summ. 45(SS-5):1–55.   2. Behravesh, C. B, R. K. Mody, J. Jungk, L. Gaul, J. T. Redd, S. Chen, S. Cosgrove, E. Hedican, D. Sweat, L. ChávezHauser, S. L. Snow, H. Hanson, T. A. Nguyen, S. V. Sodha, A. L. Boore, E. Russo, M. Mikoleit, L. Theobald, P. Gerner-Smidt, R. M. Hoekstra, F. J. Angulo, D. L. Swerdlow, R. V. Tauxe, P. M. Griffin, and I. T. Williams, for the Salmonella Saintpaul Outbreak Investigation Team. 2011. 2008 outbreak of Salmonella Saintpaul infections associated with raw produce. N. Engl. J. Med. 364:918–927. doi:10.1056/NEJMoa1005741.   3. Bell, B. P., M. Goldoft, P. M. Griffin, M. A. Davis, D. C. Gordon, P. I. Tarr, C. A. Bertelson, J. H. Lewis, T. J. Barrett, J. G. Wells, R. Baron, and J. Kobayashi. 1994. A multistate outbreak of Escherichia coli O157: H7-associated bloody diarrhea and hemolytic uremic syndrome from hamburgers: the Washington experience. JAMA 272:1349–1353.   4. Bender, J. B., C. W. Hedberg, J. M. Besser, D. J. Boxrud, K. L. MacDonald, and M. T. Osterholm. 1997. Surveillance for Escherichia coli O157:H7 infections in Minnesota by molecular subtyping. N. Engl. J. Med. 337:388–394.   5. Bryan, F. L., J. J. Guzewich, and E. C. D. Todd. 1997. Surveillance of foodborne disease III. Summary and presentation of data on vehicles and contributory factors; their value and limitations. J. Food Prot. 60:701–714.   6. Carr, C. J., and F. C. Lu. 1998. Partnership for food safety education—“Fight BAC!.” Regul. Toxicol. Pharmacol. 27:281–282.   7. Center for Food Safety and Applied Nutrition, Office of Compliance, Food and Drug Administration. 2004. FDA Report on the Occurrence of Foodborne Illness Risk Factors in Selected Institutional Foodservice, Restaurant, and Retail Food Store Facility Types. Food and Drug Administration, Washington, DC. http://www.cfsan.fda. gov/~dms/retrsk2.html.   8. Centers for Disease Control and Prevention. 1997. Foodborne diseases active surveillance network, 1996. MMWR Morb. Mortal. Wkly. Rep. 46:258–261.   9. Centers for Disease Control and Prevention. 1997. Escherichia coli O157:H7 infections associated with eating a nationally distributed commercial brand of frozen ground beef patties and burgers—Colorado, 1997. MMWR Morb. Mortal. Wkly. Rep. 46:777–778. 10. Centers for Disease Control and Prevention. 1998. Multistate outbreak of Salmonella serotype Agona infections linked to toasted oats cereal—United States, April-May, 1998. MMWR Morb. Mortal. Wkly. Rep. 47:462–464.

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

590 11. Centers for Disease Control and Prevention. 1999. Food­ borne Diseases Active Surveillance Network (FoodNet): Population Survey Atlas of Exposures: 1998–1999. Centers for Disease Control and Prevention, Atlanta, GA. 12. Centers for Disease Control and Prevention. 2001. Updated guidelines for evaluating public health surveillance systems. MMWR Recommend. Rep. 50(RR13): 1–35. 13. Centers for Disease Control and Prevention. 2004. Food­ borne Diseases Active Surveillance Network (FoodNet): Popu­lation Survey Atlas of Exposures: 2002–2003. Cen­ ters for Disease Control and Prevention, Atlanta, GA. http:// www.cdc.gov/foodnet/surveys/pop/2002/2002Atlas.pdf. 14. Centers for Disease Control and Prevention. 2005. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food—10 sites, United States, 2004. MMWR Morb. Mortal. Wkly. Rep. 54:352–356. 15. Centers for Disease Control and Prevention. 2006. Sur­ veillance for foodborne-disease outbreaks—United States, 1998-2002. MMWR Surveill. Summ. 55(SS10):1–34. 16. Centers for Disease Control and Prevention. 2006. Ongoing multistate outbreak of Escherichia coli serotype O157:H7 infections associated with consumption of fresh spinach—United States, September 2006. MMWR Morb. Mortal. Wkly. Rep. 55:1045–1046. 17. Centers for Disease Control and Prevention. 2007. Foodborne Diseases Active Surveillance Network (FoodNet): Population Survey Atlas of Exposures: 2006–2007. Centers for Disease Control and Prevention, Atlanta,  GA.  http://www.cdc.gov/foodnet/surveys/ FoodNetExposureAtlas0607_508.pdf. 18. Centers for Disease Control and Prevention. 2007. Inves­ tigation of Outbreak of Human Infections Caused by Salmonella I 4,[5],12:i:-. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/salmonella/ 4512eyeminus.html. 19. Centers for Disease Control and Prevention. 2007. Multistate outbreak of Salmonella serotype Tennessee infections associated with peanut butter—United States, 2006– 2007. MMWR Morb. Mortal. Wkly. Rep. 56:521–524. 20. Centers for Disease Control and Prevention. 2009. Multistate Outbreak of E. coli O157:H7 Infections Linked to Eating Raw Refrigerated, Prepackaged Cookie Dough. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/ecoli/2009/0630.html. 21. Centers for Disease Control and Prevention. 2009. Multistate outbreak of Salmonella infections associated with peanut butter and peanut butter-containing products—United States, 2008–2009. MMWR Morb. Mortal. Wkly. Rep. 58:85–90. 22. Centers for Disease Control and Prevention. 2009. Outbreak of Salmonella serotype Saintpaul infections associated with eating alfalfa sprouts—United States, 2009. MMWR Morb. Mortal. Wkly. Rep. 58:500–503. 23. Centers for Disease Control and Prevention. 2010. Investigation update: multistate outbreak of human Salmonella Enteritidis infections associated with shell eggs.

SMP_Food Microbiology_CH22.indd

590

Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/salmonella/enteritidis/index.html. 24. Centers for Disease Control and Prevention. 2010. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food— 10 States, 2009. MMWR Morb. Mortal. Wkly. Rep. 59:418–422. 25. Centers for Disease Control and Prevention. 2010. Sur­ veillance for foodborne disease outbreaks—United States, 2007. MMWR Morb. Mortal. Wkly. Rep. 59:973–979. 26. Centers for Disease Control and Prevention. 2010. Statespecific trends in fruit and vegetable consumption among adults—United States, 2000–2009. MMWR Morb. Mortal. Wkly. Rep. 59:1125–1130. 26a. Centers for Disease Control and Prevention. Foodborne Outbreak Online Database. http://wwwn.cdc.gov/food borneoutbreaks/. Accessed 13 September 2012. 27. Chang, K. O. 2009. Role of cholesterol pathways in norovirus replication. J. Virol. 83:8587–8595. 28. Council to Improve Foodborne Outbreak Response. 2009. Guidelines for Foodborne Disease Outbreak Response. Council to Improve Foodborne Outbreak Response, Atlanta, GA. http://www.cifor.us/documents/CIFORGui delinesforFoodborneDiseaseOutbreakResponse-updated. pdf. 29. Dalton, C. B., E. D. Mintz, J. G. Wells, C. A. Bopp, and R. V. Tauxe. 1999. Outbreaks of enterotoxigenic Escherichia coli infection in American adults: a clinical and epidemiologic profile. Epidemiol. Infect. 123:9–16. 30. Flint, J. A., Y. T. Van Duynhoven, F. J. Angulo, S. M. DeLong, P. Braun, M. Kirk, E. Scallan, M. Fitzgerald, G. K. Adak , P. Sockett, A. Ellis, G. Hall, N. Gargouri, H. Walke, and P. Braam. 2005. Estimating the burden of acute gastroenteritis, foodborne disease, and pathogens commonly transmitted by food: an international review. Clin. Infect. Dis. 41:698–704. 31. Food and Drug Administration. 2009. Egg Safety Final Rule. Food and Drug Administration, Washington, DC. www.fda.gov/Food/FoodSafety/Product-Specific Informa tion/EggSafety/EggSafetyActionPlan/ucm170615.htm. 32. Food Safety and Inspection Service, U.S. Department of Agriculture. 1994. Nationwide Beef Microbiological Data Collection Program: Steers and Heifers. U.S. Department of Agriculture, Washington, DC. 33. Food Safety and Inspection Service, U.S. Department of Agriculture. 1996. Nationwide Beef Microbiological Data Collection Program: Cows and Bulls. U.S. Department of Agriculture, Washington, DC. 34. Food Safety and Inspection Service, U.S. Department of Agriculture. 1996. Nationwide Federal Plant Ground Beef Microbiological Survey. U.S. Department of Agriculture, Washington, DC. 35. Food Safety and Inspection Service, U. S. Department of Agriculture. 2005. Microbiological Testing Program for Escherichia coli O157:H7 in Raw Ground Beef, 1995 through 2004 Data. U.S. Department of Agriculture, Washington, DC. http://www.fsis.usda.gov/Frame/Frame Redirect.asp?main=http://www.fsis.usda.gov/OPHS/ecoltest/tables1.htm.

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases 36. Garber, L. P., S. J. Wells, D. D. Hancock, M. P. Doyle, J. Tuttle, J. A. Shere, and T. Zhao. 1995. Risk factors for fecal shedding of Escherichia coli O157:H7 in dairy calves. J. Am. Vet. Med. Assoc. 207:46–49. 37. Gerner-Smidt, P., J. Kincaid, K. Kubota, K. Hise, S. B. Hunter, M. A. Fair, D. Norton, A. Woo-Ming, T. Kurzynski, M. J. Sotir, M. Head, K. Holt, and B. Swaminathan. 2005. Molecular surveillance of Shigatoxigenic Escherichia coli O157 by PulseNet USA. J. Food Prot. 68:1926–1931. 38. Graves, L. M., S. B. Hunter, A. R. Ong, D. SchoonmakerBopp, K. Hise, L. Kornstein, W. E. DeWitt, P. S. Hayes, E. Dunne, P. Mead, and B. Swaminathan. 2005. Microbiological aspects of the investigation that traced the 1998 outbreak of listeriosis in the United States to contaminated hot dogs and establishment of molecular subtyping-based surveillance for Listeria monocytogenes in the PulseNet network. J. Clin. Microbiol. 43:2350–2355. 39. Green, L., C. Selman, A. Banerjee, R. Marcus, C. Medus, F. J. Angulo, V. Radke, S. Buchanan, and the EHSNet Working Group. 2005. Food service workers’ self­reported food preparation practices: an EHS-Net study. Int. J. Hyg. Environ. Health 208:27–35. 40. Haeghebaert, S., P. Sulem, L. Deroudille, E. VannerovAdenot, O. Bagnis, P. Bouvet, F. Grimont, A. Brisabois, F. Le Querrec, C. Hervy, E. Espie, H. de Valk, and V. Vaillant. 2003. Two outbreaks of Salmonella enter­ itidis phage type 8 linked to the consumption of Cantal cheese made with raw milk, France, 2001. Euro Surveill. 8:151–156. 41. Harker, K. S., C. Lane, E. De Pinna, and G. K. Adak. 2010. An outbreak of Salmonella Typhimurium DT191a associated with reptile feeder mice. Epidemiol. Infect. 14:1–8. 42. Hedberg, C. W., K. L. MacDonald, and M. T. Osterholm. 1994. Changing epidemiology of foodborne disease: a Minnesota perspective. Clin. Infect. Dis. 18:671–682. 43. Hedberg, C. W., S. J. Savarino, J. M. Besser, C. J. Paulus, V. M. Thelen, L. J. Meyers, D. N. Cameron, T. J. Barrett, J. B. Kaper, M. T. Osterholm, and the Investigation Team. 1997. An outbreak of foodborne illness caused by Escherichia coli O39:NM, an agent not fitting into the existing scheme for classifying diarrheogenic E. coli. J. Infect. Dis. 176:1625–1628. 44. Hedberg, C. W., F. J. Angulo, K. E. White, C. W. Langkop, W. L. Schell, M. G. Stobierski, A. Schuchat, J. M. Besser, S. Dietrich, L. Helsel, P. M. Griffin, J. W. McFarland, M. T. Osterholm, and the Investigation Team. 1999. Outbreaks of salmonellosis associated with eating uncooked tomatoes: implications for public health. Epidemiol. Infect. 122:385–393. 45. Hedberg, C. W., S. J. Smith, E. Kirkland, V. Radke, T. F. Jones, C. A. Selman, and the EHS-Net Working Group. 2006. Systematic environmental evaluations to identify differences between outbreak and nonoutbreak restaurants. J. Food Prot. 69:2697–2702. 46. Hennessy, T. W., C. W. Hedberg, L. Slutsker, K. E. White, J. M. Besser-Wiek, M. E. Moen, J. Feldman, W. W. Coleman, L. M. Edmonson, K. L. MacDonald, M. T. Osterholm, and the Investigation Team. 1996. A national

SMP_Food Microbiology_CH22.indd

591

591 outbreak of Salmonella enteritidis infections from ice cream. N. Engl. J. Med. 334:1281–1286. 47. Hennessy, T. W., L. R. Cheng, H. Kassenborg, S. D. Ahuja, J. Mohle-Boetani, R. Marcus, B. Shiferaw, F. Angulo, and the FoodNet Working Group. 2004. Egg consumption is the principal risk factor for sporadic Salmonella serotype Heidelberg infections: a case-control study in FoodNet sites. Clin. Infect. Dis. 38:S237–S243. 48. Herwaldt, B. L., M. L. Ackers, and the Cyclospora Working Group. 1999. An outbreak in 1996 of cyclosporiasis associated with imported raspberries. N. Engl. J. Med. 336:1548–1556. 49. Hogue, A., P. White, J. Guard-Petter, W. Schlosser, R. Gast, E. Ebel, J. Farrar, T. Gomez, J. Madden, M. Madison, A. M. McNamara, R. Morales, D. Parham, P. Sparling, W. Sutherlin, and D. Swerdlow. 1997. Epidemiology and control of egg-associated Salmonella enteritidis in the United States of America. Rev. Sci. Tech. 16:542–553. 50. Hope, B. K., R. Baker, E. D. Edel, A. T. Hogue, W. D. Schlosser, R. Whiting, R. M. McDowell, and R. A. Morales. 2002. An overview of the Salmonella enteritidis risk assessment for shell eggs and egg products. Risk Anal. 22:203–218. 51. Hutwagner, L. C., E. K. Maloney, N. H. Bean, L. Slutsker, and S. M. Martin. 1997. Using laboratory-based surveillance data for prevention: an algorithm for detecting Salmonella outbreaks. Emerg. Infect. Dis. 3:395–400. 52. Isaacs, S., J. Aramini, B. Ciebin, J. A. Farrar, R. Ahmed, D. Middleton, A. U. Chandran, L. J. Harris, M. Howes, E. Chan, A. S. Pichette, K. Campbell, A. Gupta, L. Y. Lior, M. Pearce, C. Clark, F. Rogers, F. Jamieson, I. Brophy, A. Ellis, and the Salmonella Enteritidis PT30 Outbreak Investigation Working Group. 2005. An international outbreak of salmonellosis associated with raw almonds contaminated with a rare phage type of Salmonella enter­ itidis. J. Food Prot. 68:191–198. 53. Jimenez, M., P. Soler, J. D. Venanzi, C. Varela, and F. MartinezNavarro. 2005. An outbreak of Campylobacter jejuni enteritis in a school of Madrid, Spain. Euro Surveill. 10:118–121. 54. Kaplan, J. E., G. W. Gary, R. C. Baron, N. Singh, L. B. Schonberger, R. Feldman, and H. B. Greenberg. 1982. The epidemiology of Norwalk gastroenteritis and the role of Norwalk virus in outbreaks of acute nonbacterial gastroenteritis. Ann. Intern. Med. 96:756–761. 55. Kassenborg, H. D., C. W. Hedberg, M. Hoekstra, M. C. Evans, A. E. Chin, R. Marcus, D. J. Vugia, K. Smith, S. D. Ahuja, L. Slutsker, and P. M. Griffin. 2004. Farm visits and undercooked hamburgers as major risk factors for sporadic Escherichia coli O157:H7 infection: data from a case-control study in 5 FoodNet sites. Clin. Infect. Dis. 38(S3):S271–S278. 56. Kimura, A. C., V. Reddy, R. Marcus, P. R. Cieslak, J. C. Mohle-Boetani, H. D. Kassenborg, S. D. Segler, F. P. Hardnett, T. Barrett, D. L. Swerdlow, and the Emerging Infections Program FoodNet Working Group. 2004. Chicken consumption is a newly identified risk factor for sporadic Salmonella enterica serotype Enteritidis infections in the United States: a case-control study in Foodnet sites. Clin. Infect. Dis. 38(S3):S285–S296.

Manila Typesetting Company

11/08/2012 07:04AM

Foodborne Pathogenic Bacteria

592 57. Krumkamp, R., R. Reintjes, and M. Dirksen-Fischer. 2008. Case-case study of a Salmonella outbreak: an ­epidemiologic method to analyse surveillance data. Int. J. Hyg. Environ. Health 211:163–167. 58. Kuchenmüller, T., S. Hird, C. Stein, P. Kramarz, A. Nanda, and A. H. Havelaar. 2009. Estimating the global burden of foodborne diseases—a collaborative effort. Euro Surveill. 14:19195. 59. Le Saux, N. J. S. Spika, B. Friesen, I. Johnson, D. Mlnychuck, C. Anderson, R. Dion, M. Rahman, and W. Tostowarky. 1993. Ground beef consumption in non-commercial settings is a risk factor for sporadic Escherichia coli O157:H7 infection in Canada. J. Infect. Dis. 167:500–502. 60. Li, J., K. Smith, D. Kaehler, K. Everstine, J. Rounds, and C. Hedberg. 2010. Evaluation of a statewide foodborne illness complaint surveillance system in Minnesota, 2000 through 2006. J. Food Prot. 73:2059–2064. 61. Li, R., M. Serdula, S. Bland, A. Mokdad, B. Bowman, and D. Nelson. 2000. Trends in fruit and vegetables consumption among adults in 16 U.S. states: behavioral risk factor surveillance system. Am. J. Public Health 90:777–781. 62. Losinger, W. C., S. J. Wells, L. P. Garber, H. S. Hurd, and L. A. Thomas. 1995. Management factors related to Salmonella shedding by dairy heifers. J. Dairy Sci. 78:2464–2472. 63. Mahon, B. E., A. Ponka, W. N. Hall, K. Komatsu, S. E. Dietrich, A. Siitonen, G. Cage, P. S. Hayes, M. A. LambertFair, N. H. Bean, P. M. Griffin, and L. Slutsker. 1997. An international outbreak of Salmonella infections caused by alfalfa sprouts grown from contaminated seeds. J. Infect. Dis. 175:876–882. 64. Mahon, B. E., L. Slutsker, L. Hutwagner, C. Drenzek, K. Maloney, K. Toomey, and P. M. Griffin. 1999. Consequences in Georgia of a nationwide outbreak of Salmonella infections: what you don’t know might hurt you. Am. J. Public Health 89:31–35. 65. Majowicz, S. E., J. Musto, E. Scallan, F. J. Angulo, M. Kirk, S. J. O’Brien, T. F. Jones, A. Fazil, R. M. Hoekstra, and the International Collaboration on Enteric Disease ‘Burden of Illness’ Studies. 2010. The global burden of nontyphoidal Salmonella gastroenteritis. Clin. Infect. Dis. 50:882–889. 66. Marcus, R., T. Rabatsky-Ehr, J. C. Mohle-Boetani, M. Farley, C. Medus, B. Shiferaw, M. Carter, S. Zansky, M. Kennedy, T. J. Van Gilder, and J. L. Hadler. 2004. Dramatic decrease in the incidence of Salmonella serotype Enteritidis infections in 5 FoodNet sites: 1996–1999. Clin. Infect. Dis. 38(S3):S135–S141. 67. McCarthy, N., and J. Giesecke. 1999. Case-case comparisons to study causation of common infectious diseases. Int. J. Epidemiol. 28:764–768. 68. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5:607–625. 69. Naimi, T. S., J. H. Wicklund, S. J. Olsen, G. Krause, J. G. Wells, J. M. Bartkus, D. J. Boxrud, M. Sullivan, H. Kassenborg, J. M. Besser, E. D. Mintz, M. T. Osterholm, and C. W. Hedberg. 2003. Concurrent outbreaks of

SMP_Food Microbiology_CH22.indd

592

Shigella sonnei and enterotoxigenic Escherichia coli infections associated with parsley: implications for surveillance and control of foodborne illness. J. Food Prot. 66:535–541. 70. National Advisory Committee on Microbiological Criteria for Foods. 1998. Hazard analysis and critical control point principles and application guidelines. J. Food Prot. 61:762–765. 71. National Advisory Committee on Microbiological Criteria for Foods. 1998. Principles of risk assessment for illness caused by foodborne biological agents. J. Food Prot. 61:1071–1074. 72. Naugle, A. L., K. G. Holt, P. Levine, and R. Eckel. 2005. Food Safety and Inspection Service regulatory testing program for Escherichia coli O157:H7 in raw ground beef. J. Food Prot. 68:462–468. 73. Olsen, S. J., L. C. MacKinon, J. S. Goulding, N. H. Bean, and L. Slutsker. 2000. Surveillance for foodborne-disease outbreaks—United States, 1993-1997. MMWR Surveill. Summ. 49(SS-1):1–51. 74. Potter, M. E., S. Gonzalez Ayala, and N. Silarug. 1997. Epidemiology of foodborne diseases, p. 376–390. In M. P. Doyle, L. R. Beuchat, and T. J. Montville (ed.), Food Microbiology: Fundamentals and Frontiers. ASM Press, Washington, DC. 75. Richardson, B. W., and W. H. Frost. 1936. Snow on cholera. The Commonwealth Fund, New York, NY. 76. Robinson, J. G., and B. Booth. 2010. Statin use and lipid levels in older adults: National Health and Nutrition Examination Survey, 2001 to 2006. J. Clin. Lipidol. 4:483–490. 77. Rolls, B. J., and E. A. Bell. 2000. Dietary approaches to the treatment of obesity. Med. Clin. N. Am. 84:401–418. 78. Rondy, M., M. Koopmans, C. Rotsaert, T. Van Loon, B. Beljaars, G. Van Dijk, J. Siebenga, S. Svraka, J. W. Rossen, P. Teunis, W. Van Pelt, and L. Verhoef. 2011. Norovirus disease associated with excess mortality and use of statins: a retrospective cohort study of an outbreak following a pilgrimage to Lourdes. Epidemiol. Infect. 139:453–463. 79. Rounds, J. M., C. W. Hedberg, S. Meyer, D. J. Boxrud, and K. E. Smith. 2010. Investigation of Salmonella enterica pulsed-field gel electrophoresis clusters in Minnesota, 2001–2007. Emerg. Infect. Dis. 16:1678–1685. 80. Roy, S. L., S. M. DeLong, S. A. Stenzel, B. Shiferaw, J. M. Roberts, A. Khalakdina, R. Marcus, S. D. Segler, D. D. Shaw, S. Thomas, D. J. Vugia, S. M. Zansky, V. Dietz, M. J. Beach, and the Emerging Infections Program FoodNet Working Group. 2004. Risk factors for sporadic cryptosporidiosis among immunocompetent persons in the United States from 1999 to 2001. J. Clin. Microbiol. 42:2944–2955. 81. Salman, M. D., G. R. Frank, D. W. MacVean, J. S. Reif, J. K. Collins, and R. Jones. 1988. Validation of disease diagnoses reported to the National Animal Health Monitoring System from a large Colorado beef feedlot. J. Am. Vet. Med. Assoc. 192:1069–1073. 82. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Widdowson, S. L. Roy, J. L. Jones, and P. M.

Manila Typesetting Company

11/08/2012 07:04AM

22.  Epidemiology of Foodborne Diseases Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. 83. Scallan, E. 2007. Activities, achievements, and lessons learned during the first 10 years of the Foodborne Diseases Active Surveillance Network: 1996-2005. Clin. Infect. Dis. 44:718–725. 84. Schroeder, C. M., A. L. Naugle, W. D. Schlosser, A. T. Hogue, F. J. Angulo, J. S. Rose, E. D. Ebel, W. T. Disney, K. G. Holt, and G. P. Goldman. 2005. Estimate of illnesses from Salmonella enteritidis in eggs, United States, 2000. Emerg. Infect. Dis. 11:113–115. 85. Serdula, M. K., C. Gillespie, L. Kettel-Khan, R. Farris, J. Seymour, and C. Denny. 2004. Trends in fruit and vegetable consumption among adults in the United States: behavioral risk factor surveillance system, 1994-2000. Am. J. Public Health 94:1014–1018. 86. Sivapalasingam, S., C. R. Friedman, L. Cohen, and R. V. Tauxe. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J. Food Prot. 67:2342–2353. 87. Slutsker L., A. A. Ries, K. Maloney, J. G. Wells, K. D. Greene, and P. M. Griffin. 1998. A nationwide case­control study of Escherichia coli O157:H7 infection in the United States. J. Infect. Dis. 177:962–966. 88. Smith, K. E., C. Medus, S. D. Meyer, D. J. Boxrud, F. Leano, C. W. Hedberg, K. Elfering, C. Braymen, J. B. Bender, and R. N. Danila. 2008. Outbreaks of salmonellosis associated with frozen, microwaveable, breaded, stuffed chicken products, Minnesota, 1998-2006. J. Food Prot. 71:2153–2160. 89. St. Louis, M. E., D. L. Morse, M. E. Potter, T. M. Demelfig, J. J. Guzewich, R. V. Tauxe, P. A. Blake, and the Salmonella enteritidis Working Group. 1988. The emergence of Grade A eggs as a major source of Salmonella enteritidis infections: new implications for the control of salmonellosis. JAMA 259:2103–2107. 90. Tauxe, R. V. 1992. Epidemiology of Campylobacter jejuni infections in the United States and other industri-

SMP_Food Microbiology_CH22.indd

593

593 alized nations, p. 9–19. In I. Nachamkin, M. J. Blaser, and L. S. Tompkins (ed.), Campylobacter jejuni: Current Status and Future Trends. ASM Press, Washington, DC. 91. Tauxe, R. V. 1998. New approaches to surveillance and control of emerging foodborne diseases. Emerg. Infect. Dis. 4:455–456. 92. Thompson, B., W. Demark-Wahnefried, G. Taylor, J. W. McClelland, G. Stables, S. Havas, Z. Feng, M. Topor, J. Heimendinger, K. D. Reynolds, and N. Cohen. 1999. Baseline fruit and vegetable intake among adults in seven 5 a day study centers located in diverse geographic areas. J. Am. Diet. Assoc. 99:1241–1248. 93. Todd, E. C. D. 1996. Worldwide surveillance of foodborne disease: the need to improve. J. Food Prot. 59:82–92. 93a. United States Department of Agriculture Food Safety and Inspection Service. 2011. Microbiological results of raw ground beef products analyzed for Escherichia coli O157: H7, summarized by calendar year. http://www.fsis.usda.gov/ Science/Ecoli_O157_Summary_Tables/index.asp. Last modified 11 July 2011. 94. Van Beneden, C. A., W. E. Keene, R. A. Strang, D. H. Werker, A. S. King, B. Mahon, K. Hedberg, A. Bell, M. T. Kelly, V. K. Balan, W. R. MacKenzie, and D. Fleming. 1999. Multinational outbreak of Salmonella enterica Newport infections due to contaminated alfalfa sprouts. JAMA 281:158–162. 95. van’t Veer, P., M. C. Jansen, M. Klerk, and F. J. Kok. 2000. Fruits and vegetables in the prevention of cancer and cardiovascular disease. Public Health Nutr. 3:103–107. 96. Vojdani, J. D., L. R. Beuchat, and R. V. Tauxe. 2008. Juice-associated outbreaks of human illness in the United States, 1995 through. 2005. J. Food Prot. 71:356–364. 97. Widdowson, M. A., A. Sulka, S. N. Bulens, R. S. Beard, S. S. Chaves, R. Hammond, E. D. Salehi, E. Swanson, J. Totaro, R. Woron, P. S. Mead, J. S. Bresee, S. S. Monroe, and R. I. Glass. 2005. Norovirus and foodborne disease, United States, 1991-2000. Emerg. Infect. Dis. 11:95–102.

Manila Typesetting Company

11/08/2012 07:04AM

SMP_Food Microbiology_CH22.indd

594

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

SMP_Food Microbiology_CH23.indd

595

IV Manila Typesetting Company

11/08/2012 07:04AM

SMP_Food Microbiology_CH23.indd

596

Manila Typesetting Company

11/08/2012 07:04AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch23

23

Marta H. Taniwaki John I. Pitt

Mycotoxins

The term “mycotoxin” is derived from the Greek word mykes, which means “fungi,” and from the Latin toxi­ can, meaning “toxins.” Mycotoxins have been defined as “fungal metabolites which when ingested, inhaled or absorbed through the skin, can cause disease or death in man and domestic animals, including birds” (121). By general agreement this definition excludes the toxins produced by macrofungi (the mushrooms), and compounds that cause disease only in lower animals or plants. Fungi produce a vast number of secondary metabolites. Only a small number of such compounds are classified as mycotoxins; i.e., they have been demonstrated to cause illness in humans or domestic animals. Specific mycotoxins are produced only by specific fungi, usually by only a few species. A particular species of fungus may produce more than one mycotoxin, though never more than one of the major compounds described here. Mycotoxin production occurs only as a result of fungal growth: the presence of spores of a particular fungus on a foodstuff is not a good guide to mycotoxin production. However, if environmental conditions, particularly of temperature and water activity (aw), are conducive to fungal growth, toxin production may occur at any period during growing, harvesting, drying, or storage of food

commodities. Mycotoxins may occur in processed foods, but are much less important than when they occur in commodities such as grains or nuts. Mycotoxins are typically chemically stable once formed, and persist in food even after the destruction of the fungi that produced them. Molecular structures of mycotoxins vary widely (Fig. 23.1), so their effects on human and animal health also vary widely: they may be neurotoxins, teratogens, nephrotoxins, hepatotoxins, immunosuppressive agents, or carcinogens. The International Agency for Research on Cancer (IARC) (63, 64) has evaluated important mycotoxins according to the risk of carcinogenicity to humans (Table 23.1). Perhaps the most important point is that toxicity due to mycotoxins is almost always insidious, with no overt indication of effects on health in the short term. For this reason, the health effects of mycotoxins are among the more neglected areas of medical science. Many fungal secondary metabolites are toxic, but only a few are significant in relation to food safety. Most are of little concern, either because their toxicity is limited or because they are produced by species that are uncommon in foods. The most important mycotoxins are aflatoxins, ochratoxin A, fumonisins, deoxynivalenol, and zearalenone (98). Each of these is dealt with below.

Marta H. Taniwaki, Instituto de Tecnologia de Alimentos, Av. Brasil, 2880, Campinas, São Paolo CEP 13070-178, Brazil. John I. Pitt, CSIRO Animal, Food and Health Sciences, P.O. Box 52, North Ryde, New South Wales 1670, Australia.

597

SMP_Food Microbiology_CH23.indd

597

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

598

Figure 23.1  Structures of aflatoxin B1, aflatoxin G1, aflatoxin M1, and ochratoxin A. doi:10.1128/9781555818463.ch23f1

Aflatoxins

Chemical Characterization

Aflatoxins are highly substituted coumarins, with a fused dihydrofurofuran moiety. Four naturally occurring aflatoxins are important: aflatoxins B1 and B2, named because of their blue fluorescence under UV light; and Table 23.1  Evaluation of mycotoxins in humansa Mycotoxin(s)

IARC evaluation

Aflatoxins B and G

Group 1, carcinogenic to humans

Aflatoxin M1

Group 2B, possibly carcinogenic to humans

Ochratoxin A

Group 2B, possibly carcinogenic to humans

Fumonisins

Group 2B, possibly carcinogenic to humans

Deoxynivalenol

Group 3, not classifiable as to carcinogenicity

Zearalenone

Group 3, not classifiable as to carcinogenicity

a

Data from references 63 and 64.

SMP_Food Microbiology_CH23.indd

598

aflatoxins G1 and G2, which fluoresce greenish yellow (Fig. 23.1). When aflatoxins B1 and G1 are ingested by lactating animals, small proportions (1 to 2%) are excreted in milk as aflatoxins M1 and M2, which are hydroxylated derivatives of the parent compounds (29).

Fungal Sources

Aflatoxins are produced in foods primarily by Aspergillus flavus and the closely related species Aspergillus para­ siticus. On standard identification media, Czapek yeast extract agar (CYA) (127) and malt extract agar (MEA) (127), both species produce rapidly growing colonies characterized by green conidia (asexual spores) borne on typical Aspergillus fruiting structures (Fig. 23.2). The two species differ in small details: A. flavus usually bears conidia on two sets of supporting cells (metulae and phialides), while A. parasiticus spores are borne on phialides alone. Conidia of A. parasiticus exhibit rougher walls than those of A. flavus. A. flavus isolates produce B aflatoxins and some also produce cyclopiazonic acid. On a worldwide basis, about 40% of A. flavus isolates produce aflatoxins (61),

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

599

Figure 23.2  (a) A. flavus, CYA, 7 days at 25°C; (b) A. flavus fruiting structure (bar = 20  mm); (c) A. flavus conidia (bar = 5 mm). (d) A. parasiticus, CYA, 7 days at 25°C; (b) A. parasiticus fruiting structure (bar = 10 mm); (c) A. parasiticus conidia (bar = 5 mm) doi:10.1128/9781555818463.ch23f2

though percentages of toxin-producing isolates may vary with land use (61). Isolates of A. parasiticus produce both B and G aflatoxins but not cyclopiazonic acid, and almost all isolates are toxigenic. At least eight other Aspergillus species make aflatoxins. Only two of these are of possible importance in foods: A. nomius and A. minisclerotigenes. Both resemble A. flavus in culture, but A. nomius produces bullet-shaped sclerotia, as distinct from the large spherical sclerotia produced by many A. flavus isolates, while A. minisclerotigenes produces small spherical sclerotia. Both of these species produce both B and G aflatoxins. Detection of these species from foods or soils is facilitated by the use of Aspergillus flavus and parasiticus agar (128). After 42 to 48 h of incubation on A. flavus and parasiticus agar at 30°C, colonies of A. flavus and A. parasiticus exhibit a brilliant orange-yellow reverse coloration.

SMP_Food Microbiology_CH23.indd

599

A. flavus will grow at from a minimum of 10 to 12°C to a maximum of 43 to 48°C, and optimally at about 33°C. It has been shown to grow at a minimum aw of 0.82 at 25°C, 0.81 at 30°C, and 0.80 at 37°C (162). Growth of A. flavus occurred over the pH range 2.1 to 11.2 at 25, 30, and 37°C, with optimal growth over a broad range from pH 3.4 to 10 (127). Atmospheres of 20% CO2 and <0.5% O2 and 80% CO2 with 20% O2 did not inhibit A. flavus growth on CYA and potato dextrose agar. However, CO2 >40% and <0.5% O2 inhibited A. flavus (151, 152). The most reliable figures for heat resistance of A. fla­ vus conidia indicate a D45 value of >160 h, a D50 of 16 h, a D52 of 40 to 45 min, and a D60 of 1 min, at neutral pH and high aw, with z values from 3.3 to 4.1 C° (127). These values indicate ready destruction at pasteurization temperatures. The physiology of A. parasiticus is similar.

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

600

Genetics

The aflatoxin biosynthetic pathway in A. flavus has been studied extensively and is now quite well understood. The pathway is a single unit on a chromosome within linkage group VII, and includes more than 20 enzymes and gene products. Many of these have now been isolated and characterized. Two regulatory genes, AflR and AflJ, are known. The gene products of AflR regulate the biosynthetic pathway at the transcription level. The function of AflJ is not known, but disruption prevents aflatoxin accumulation (21).

Ecology

A. flavus is of universal occurrence in food crops in the tropical and warm temperate zones of the world. A. fla­ vus is associated with peanuts, maize, and cotton­seed and frequently produces aflatoxins in these commodities. It also occurs in tree nuts, especially pistachios and Brazil nuts (109), and less commonly in hazelnuts, walnuts, coconuts, copra, pecans, and kola nuts (13, 57, 111, 127, 141). Evidence suggests that A. para­ siticus has a more limited geographical range than A. flavus. It occurs where peanuts are grown and appears to be uncommon in most other natural environments. The occurrence of A. flavus and related species in foods has been reviewed extensively elsewhere (127).

Toxicity of Aflatoxins

Aflatoxins have a likely involvement in five toxic effects: acute toxicity, liver carcinogenicity, liver cirrhosis, immunosuppression, and growth retardation in children.

Acute Toxicity—Aflatoxicosis

An outbreak of hepatitis due to aflatoxin ingestion from maize occurred in India in 1974: almost 400 cases were identified, and 106 deaths were reported. Clinical features were jaundice preceded by fever, vomiting, and anorexia, with ascites and edema in the lower limbs in extreme cases. Levels in analyzed foods were often extremely high, with estimated daily ingestion of up to 6 mg of aflatoxins (8). Two outbreaks of aflatoxicosis have been documented from Kenya, in 1981 and 2004. In the latter outbreak, 317 cases and 125 deaths were reported (83). A smaller outbreak occurred in Malaysia in 1988, where 13 children died from acute hepatic encephalopathy from consumption of commercially pre­pared noodles contaminated with aflatoxin. With the exception of the Malaysian incident, aflatoxicosis only occurs when drought or famine causes exceptionally high levels of aflatoxins in the diet or forces the eating of substandard food.

SMP_Food Microbiology_CH23.indd

600

Liver Carcinogenicity

More insidious, and more frequent, is the development of liver cancer from consumption of much lower levels of aflatoxin over longer time periods. The International Agency for Research on Cancer (IARC) recognizes aflatoxin B1 and naturally occurring mixtures of aflatoxins as Group 1 carcinogens; i.e., they are recognized as carcinogenic to humans. Studies in the past 20 years have determined that hepatitis B virus also causes human liver cancer; both agents are causal and synergistic. In its risk assessment of aflatoxins, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) has derived two potency factors for cancer formation by aflatoxins: for aflatoxin alone, 0.01 cases per 100,000 people per annum per ng kg–1 body weight per day; and for individuals carrying hepatitis B infection, 0.30 cases. Thus, the two agents together are about 30 times as potent as aflatoxin alone (64). In the liver, aflatoxin B1 is converted by cytochrome P450 enzymes to the 8,9-epoxide. This is capable of binding to liver proteins, leading to liver failure and potentially to aflatoxicosis. This epoxide is also able to bind to DNA, a precursor step to the development of liver cancer.

Stunting in Children

A number of studies have now provided evidence that aflatoxin exposure before birth and in early childhood is associated with stunted growth—defined by the WHO as height for age being more than 2 standard deviations below average height for age in a given population. It is apparent, from the high numbers of people believed to be consuming uncontrolled levels of aflatoxin, that stunting is an important disease burden, only recently recognized (55). Growth suppression has also been observed in a number of mammals and avian species.

Immunosuppression

Aflatoxins have been shown to suppress the cell-­mediated immune response in both cell lines and domestic animals. Effects include the impairment of delayed-type hypersensitivity, decreases in the phagocytic activity of macrophages, increased susceptibility to infection, and reduced response to vaccines (64). Few studies have been reported in humans, but it is apparent that if the effects in humans mirror those in animals even approximately, then the immunosuppressive effects of aflatoxins probably have very wide implications for human health.

Liver Cirrhosis

Direct evidence that aflatoxins are involved in liver cirrhosis is limited, but as death from cirrhosis of the liver is an important cause of mortality, any enhancement by

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

601

aflatoxin of other factors in causing cirrhosis will have important consequences.

Analysis of Aflatoxins

Aflatoxins are usually very unevenly distributed in commodities, especially peanuts, where few nuts are contaminated but contamination levels in individual nuts are often high, up to 1 mg kg–1. For that reason, sampling is often the largest source of error in aflatoxin assays. Sampling plans have been developed for continuous lines, for 10-tonne lots, and for bag stacks (28, 163, 164). All of the recommended methods use sample sizes of 8 kg or more. Entire samples should be comminuted in a vertical chopper or similar mill. Subsamples, ideally of 500 g or more, may be extracted by a variety of mixed polar and nonpolar solvents, depending on the food matrix being analyzed. Methanol and water (80:20) is recommended for commodities such as maize, peanuts, and cottonseed (6). The traditional methodology for aflatoxin assays is thin-layer chromatography (TLC), and as this is inexpensive and reliable, it is recommended for less developed economies. For acceptable/not acceptable testing, immunochemical methods are most frequently used. For advanced users such as high-volume analytical laboratories or regulatory authorities, liquid chromatography, sometimes coupled with mass spectroscopy, has become normal practice. Limits of detection are now well below 1 mg kg–1. Techniques for mycotoxin analysis have been updated elsewhere (73, 140).

Occurrence of Aflatoxins in Foods, and Regulation

Levels of aflatoxins in foods are highly variable. Only a few commodities are at serious risk if good agricultural and manufacturing practice is observed. Peanuts, maize, and cottonseed frequently contain unacceptable levels of aflatoxins, and assays are usually carried out at wholesale intakes, and frequently in finished products as well. In developed countries, regulations are generally stringently applied to human foods and animal feeds that contain appreciable amounts of these commodities. Other closely watched commodities are tree nuts, especially pistachios and brazil nuts, figs, and spices. Although levels in these commodities may exceed regulatory limits from time to time, such levels seldom pose a long-term risk to human health. Under inadequate storage conditions, other grains including sorghum and rice may also permit growth of A. flavus and aflatoxin production. These commodities have very high consumption levels in many communities, so careful storage is of great importance.

SMP_Food Microbiology_CH23.indd

601

Regulation of aflatoxin levels in foods commenced in about 1970, using what was then the limit of detection, 5 mg kg–1, as the permitted limit. However, as peanut producers could not reach that limit, higher limits were set up in peanut-exporting countries, including the United States and Australia. As analytical techniques improved with the introduction of liquid chromatography, lower limits were frequently set by importing countries, especially in Europe. Recent clearer understanding of aflatoxin toxicology has produced a compromise safe limit for human consumption of 15 mg kg–1 of peanuts. Levels in most foodstuffs, where aflatoxin is less likely to occur, remain lower, usually 5 mg kg–1 or less. Although safe levels of aflatoxins have now been established and aflatoxin levels are closely controlled in developed countries, it has been estimated that up to 5 billion people worldwide are at risk from exposure to uncontrolled levels of aflatoxins in their diets (145).

Control of Aflatoxins in Foods

Control of aflatoxins has proved very difficult, as invasion by A. flavus (and A. parasiticus in peanuts) occurs before harvest in all of the major crops affected (127). In peanuts, it appears likely that infection by A. flavus while nuts are still in the ground is a prerequisite for high levels of aflatoxins to be formed after harvest (120). In the absence of high preharvest infection levels, and with rapid and effective drying, peanuts can be produced free of any appreciable level of aflatoxin. The major causes of preharvest infection are high spore numbers in soil and plant stress induced by drought and/or high soil temperatures (123). A. parasiticus and A. fla­ vus are commensals in peanuts, as both fungi are able to grow in peanut plants and developing peanuts (122). Partial control of spore numbers can be achieved by crop rotation, as numbers of A. flavus in soil decrease under cultivation of small grains or pasture. Irrigation, which eliminates drought stress, is regarded as the most effective method for reducing aflatoxin formation in peanuts (30). However, peanuts throughout the world are recognized as a drought-resistant crop and are mostly grown under dry culture, with irrigation reserved for more moisture-sensitive crops such as rice or vegetables. In many areas where peanuts are grown, irrigation is not an option. Under these circumstances, reduction in drought stress by good agricultural practices can be a beneficial approach, for example, by weed control and wider spacing between rows. Harvesting early when drought stress occurs can help reduce contamination. Rapid drying of peanuts using mechanical dryers as soon as possible after pulling has a major effect in reducing­

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

602 the levels of aflatoxins in peanuts. Maintaining dryness during storage is also essential (59). A. flavus is also a commensal in maize and cottonseed (72, 84), so preharvest infection is common in these crops. In maize, the infection route is usually through insect damage, and in cotton, through the nectaries. For maize, irrigation and improved farm management practices have a beneficial effect on aflatoxin formation (117). However, as maize is usually dried in the field, rapid drying techniques are not commonly practiced. Control of insect damage in maize (51) and a number of practices including good storage conditions (134) and planting genetically modified crops (90) are employed to control aflatoxin levels. Cottonseed is a by-product of cotton production, so field drying is normal. Breeding of cotton without nectaries has been proposed as one means of limiting A. flavus access to cotton bolls (72). In tree nuts, type, timing, and amount of irrigation are very important methods to reduce aflatoxin formation. Insect control is also of great importance (133). Entry of A. flavus into pistachio nuts depends on the time of splitting of hulls. Nuts in which hull splitting occurs early are much more susceptible to A. flavus invasion on the tree (37). It is known that some cultivars are more prone to early splitting than others, and this is especially important where nuts are harvested from the ground, after contact with the soil. Figs are sometimes infected by A. flavus, both because of their unique structure developed for insect fertilization and because figs are harvested from the ground in some countries. Immature figs are not colonized by A. flavus, but once they are ripe, infection occurs readily and fungal growth continues during drying (76). The proportion of figs infected is only about 1% (142). Good agricultural practice has been the standard method in developed countries for attempting to control aflatoxin levels in commodities, particularly maize and peanuts. Good agricultural practice involves good farm management, including weed control, optimal row and plant spacing, insect control, adequate water supplies, rapid and complete drying, removal of defects, and effective control of storage conditions. By themselves, these approaches are often inadequate; too often drought stress and/or insect damage results in aflatoxin formation before harvest, out of farmer control, even in developed countries. The only effective HACCP for aflatoxins remains end product testing.

Reducing Aflatoxin Levels in Foods

In peanuts, reduction in aflatoxin levels is accomplished by color sorting of individual kernels after shelling. The process was developed originally to reject commercially unac-

SMP_Food Microbiology_CH23.indd

602

ceptable discolored nuts, regardless of cause, but as fungal growth is a prime cause of discoloration, the process is also an effective nondestructive means of removing most nuts containing aflatoxins. In crops under severe drought stress, peanuts begin to dry in the ground, and under these conditions luxuriant growth of A. flavus can occur, with high aflatoxin production. In this case, blanching to remove skins and roasting to increase discoloration permit effective color sorting to be carried out (130). Maize and fig samples are screened for the presence of aflatoxin by the examination of cracked kernels or fruit by UV light. No effective nonchemical testing techniques exist for cottonseed or pistachios, and as with other commodities, nondestructive chemical assays are not available. It is normal practice to assay aflatoxin levels in all consignments of peanuts and maize in major developed producing countries, often repeatedly, from intake to shellers to final product. Such controls rarely exist in less developed countries. Other commodities are similarly screened according to needs and markets. The extent to which aflatoxins are destroyed during heating is largely dependent on the process used. Less than 25% of the aflatoxin content of a commodity is destroyed by boiling water (107), extrusion (26), and autoclaving (144). However, dry roasting of peanuts can reduce aflatoxin levels by up to 80% (107). Nixtamalization, the process of making tortilla, reduced more than 90% of aflatoxin in contaminated maize (95). The alkali process usually practiced to produce refined table oil completely removes aflatoxin (65).

Interventions to Reduce Aflatoxins

In recent years, several approaches have been put forward to reduce aflatoxin levels in foods and feeds. The most advanced preharvest approach is biocontrol by competitive exclusion for both peanut and cotton crops. The technique, developed independently in the United States and Australia, relies on the fact that only about 40% of A. flavus strains produce aflatoxins. Selected nontoxigenic strains that are both competitive in the field and unlikely to revert to toxicity are introduced, in high numbers, into soils in fields where peanut or cotton crops are being grown (36, 122, 126). The nontoxigenic spores compete with the existing toxin-producing spores in the soil for infection sites on developing nuts. In sufficiently high numbers, control can be very effective. This process is used commercially in the United States for peanuts and cotton (33, 67). Pilot-scale work is currently in progress on maize crops and under development for some tree nuts. Other reported experimental studies have aimed to improve farm efficiency by providing advice and im-

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

603

proved management for farmers in less developed countries (155). Some success has been achieved. A different approach has resulted from the discovery that certain clays, among a variety of natural materials including activated charcoal and bentonite, adsorb aflatoxins very strongly. Animal studies have been very effective, and some human trials have yielded promising results. It is too early to say whether this approach will be effective in practice (165).

Ochratoxin A

Chemical Characterization

Ochratoxins are mycotoxins containing a dihydroisocoumarin ring linked to phenylalanine through a 7carboxy group (Fig. 23.1). Several ochratoxins are known: ochratoxin A (OTA); its ethyl ester, ochratoxin C; and dechlorinated analogues, i.e., ochratoxin B and its methyl and ethyl esters. Ochratoxins A and B are the only ones so far detected in naturally contaminated foods or feeds. OTA is the most toxic compound, and the only one treated here.

Fungal Sources

OTA is produced by three well-defined groups of fungi: (i) the ocher-colored aspergilli—Aspergillus ochraceus, A. westerdijkiae, A. steynii, and a few other much less important, closely related species; (ii) the black aspergilli—A. carbonarius and A. niger (which produces OTA only infrequently); and (iii) Penicillium species—P. verrucosum plus the closely related species P. nordicum (48, 75, 127).

Aspergillus ochraceus and Related Species

After 1 week at 25°C on the standard growth medium CYA, A. ochraceus and closely related species produce colonies 40 to 55 mm in diameter with ocher (light ­yellow brown) conidia, characteristics distinctive for this group of species. A. westerdijkiae and A. steynii are distinguished from A. ochraceus by their inability to grow on CYA at 37°C. A. westerdijkiae produces finely roughened, spherical conidia and A. steynii smoothwalled, ellipsoidal conidia, while those of A. ochraceus are smooth-walled and spherical (127) (Fig. 23.3a to c). These three species have similar growth requirements so far as is known, apart from the difference in maximum growth temperature, ca. 40°C for A. ochraceus and about 5°C lower for the other species. Although A. ochraceus was the species first recognized as producing OTA, recent studies indicate that A. west­ erdijkiae is much more common, and much more likely to produce OTA. Studies reporting A. ochraceus­ before 2005 were likely to include all three species, but conclude

SMP_Food Microbiology_CH23.indd

603

that the occurrence of OTA in coffee was associated with A. ochraceus (93, 146, 150). More recently, advances in molecular and fungal metabolite techniques resulted in the description of A. westerdijkiae and A. steynii (49), with A. westerdijkiae recognized as the main source of OTA in coffee (103). These three species are xerophilic, growing down to water activities of 0.8 or below, and as such can be isolated from stored dry foods.

Aspergillus carbonarius and Aspergillus niger

These two species produce rapidly growing black or deep reddish brown colonies on CYA, usually 60 mm or more after 7 days at 25°C. A. carbonarius produces much larger conidia (6 to 7 µm in diameter) than A. niger (4 to 5 mm). A. carbonarius grows at from 10 to 40°C, and down to 0.85 aw (127) (Fig. 23.3d to f). A. carbonarius has a major habitat in vineyards and grape drying yards, where its ability to tolerate high temperatures and strong sunlight (as a result of its dark hyphae and spores) provides a competitive advantage. It has been recognized as the primary source of OTA contamination in grapes and grape products, including wines, throughout the world (1, 12, 77). Grapes are usually sun-dried without preservatives, so raisins and sultanas sometimes contain unacceptable levels of OTA (62, 85, 96). OTA can also result from growth of A. car­ bonarius in coffee beans (68, 81, 150) and cocoa beans (32, 104). Other sources, including figs (38), peanuts and maize (88, 89), and paprika (5), are less important. A. niger has usually been regarded as a benign fungus and has been widely used in enzyme production and ingredients for food processing. It holds GRAS (generally regarded as safe) status from the U.S. Food and Drug Administration. Usually only a low percentage of A. niger isolates are able to produce OTA. Although A. niger is very frequently isolated, it is not a significant source of OTA as a rule.

Penicillium verrucosum and Penicillium nordicum

P. verrucosum and P. nordicum are very similar, slowly growing species, reaching only 15 to 25 mm in diameter on CYA after 7 days at 25°C, and colored dull green to dark green. Both grow from about 0 to 30°C and down to 0.80 aw (127) (Fig. 23.3g to i). P. verrucosum is the major OTA producer in cereals grown in cool temperate climates, ranging across northern and central Europe, Canada, and northern Asia (86). In these regions, OTA is frequently present in cereal products, especially bread and flour-based foods, and in the meat of animals that eat cereals as a major dietary component (87). P. verrucosum is not found in

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

604

Figure 23.3  (a) A. ochraceus, CYA, 7 days at 25°C; (b) A. ochraceus fruiting structure (bar = 20 mm); (c) A. ochraceus conidia (bar = 5 mm). (d) A. carbonarius, CYA, 7 days at 25°C; (e) A. carbonarius fruiting structure (bar = 40 mm); (f) A. carbonarius conidia (bar = 5 mm). (g) P. verrucosum, CYA, 7 days at 25°C; (h) P. verrucosum fruiting structure (bar = 10 mm); (i) P. verrucosum conidia (bar = 5 mm). doi:10.1128/9781555818463.ch23f3

SMP_Food Microbiology_CH23.indd

604

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

605

warmer climates, so small grains from the tropics and warm temperate zones do not contain OTA (127). P. nordicum is closely related to P. verrucosum (75) but does not occur in cereals. It has been found in manufactured meat products such as salami and ham and can produce OTA there. This does not appear to be a major problem.

Toxicity of OTA

OTA is a chronic nephrotoxin, affecting kidney function in all animal species tested. OTA is readily absorbed through the intestines and, once it enters the bloodstream, has a long half-life, up to 3 weeks in monkeys. As a result, the blood of healthy humans regularly contains detectable amounts of OTA in areas where it is frequently part of the diet (168). Because P. verruco­ sum has a restricted distribution, this phenomenon is confined almost entirely to northern Europe, northern North America, and northern Asia. Recent indications of OTA in human blood from other parts of the world are almost certainly due to the presence of ochratoxigenic Aspergillus species in foods. OTA also has carcinogenic properties, but the mechanism of carcinogenicity remains unclear. The carcinogenic effects in animals are considered to be of less importance than the nephrotoxicity (74, 167). Although OTA is demonstrably toxic to animals of all kinds, its effects in humans remain unclear and the subject of debate. Both genotoxic and nongenotoxic modes of action have been proposed (74, 168). IARC has classified OTA as a possible human carcinogen (Group 2B), based on sufficient evidence of carcinogenicity in experimental animal studies and inadequate evidence in humans. The target organ of toxicity in all mammalian species tested is the kidneys, in which lesions can be produced by both acute and chronic exposure (167). Although the mechanism of action of OTA is unclear, its structural similarity to phenylalanine and the fact that it inhibits many enzymes and processes that are dependent on phenylalanine strongly suggest that it acts by disrupting phenylalanine metabolism (24). OTA toxicity in animals can be reduced by treatment with phenylalanine, as it is a phenylalanine analogue (92).

Exposure Assessment

Assessment of OTA exposure has proved to be very difficult. People in Europe and northern North America are exposed to ochratoxin A in barley and wheat and their products, especially bread, and also from meat, especially pork, from animals fed contaminated feed. Low levels also occur in beer, wine, coffee, cocoa, chocolate, and dried vine fruits. However, as wheat and barley

SMP_Food Microbiology_CH23.indd

605

crops from warmer climates are not infected by P. ver­ rucosum, OTA intake is much lower in tropical and subtropical regions (125).

Risk Characterization

In a recent evaluation, JECFA has maintained the provisional tolerable weekly intake (PTWI) of OTA at 100 ng kg–1 body weight. This is based on the lowest observed level of chronic toxicity in pigs, as risk assessments have indicated that the chronic toxicity occurs at lower intake levels than any carcinogenic effect (168). However, a more recent reevaluation of the risk associated with OTA has reduced the PTWI to 21 ng kg–1 body weight, based on applying a larger uncertainty factor to available toxicity data (74). JECFA (168) has assessed the dietary exposure to OTA from processed cereals as 8 to 17 ng kg–1 body weight per week, well below the PTWI. However, in a previous evaluation, JECFA had determined that the 95th percentile for OTA consumption was about 90 ng kg–1 body weight, approaching the PTWI (167). Even with some additional intake from wine and coffee, the hazard from intake of OTA in Europe appears low. Intake in other parts of the world is likely to be much lower, in the absence of appreciable levels in cereals. Dietary exposure is based primarily on data from Europe, where processed cereal foods sometimes contain high OTA levels (69, 143, 168). Cereals in tropical and warm temperate climates seldom contain appreciable OTA, as P. verrucosum is a cool-climate fungus. Tropical foods, such as coffee and cocoa, are sometimes contaminated with OTA as the result of growth by Aspergillus species, but levels are well within the PTWI (168).

Control of OTA Formation in Crops

None of the fungi producing OTA are known to be systemic invaders or pathogens of food crops. In consequence, control of the formation of OTA relies on good practice. For cereal grains, coffee cherries (dry process), and grapes, rapid drying will ensure low production of OTA. However, climatic conditions often make that difficult to achieve. A. carbonarius occurs in vineyards, and can infect grapes before harvest, as the result of damage by plantpathogenic fungi including powdery mildews, Rhizopus spp., or Botrytis spp., or from skin splitting due to rain or storm damage. The resulting Aspergillus bunch rot generally occurs near harvest when the sugar content of berries is highest (77, 80) and causes production of OTA in grape juice and wine. OTA occurs in wine from the warmer growing areas throughout the world, but levels are usually low, as the fermentation process stops

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

606 growth of the fungus. When grapes are dried, however, mechanical damage during harvesting and time in the drying yard increase the probability of OTA formation. The population of A. carbonarius in vineyards can be reduced by good management practices such as irrigation, pruning to improve airflow through vines, use of cover crops between rows, and appropriate fungicide applications (14, 80, 129). Fungi capable of producing OTA are rarely present on coffee cherries at harvest. After harvesting from the tree, coffee cherries are processed under two basic systems: (i) the dry processing system, which produces what is called natural coffee or dried coffee cherry (the seed is enclosed in the whole fruit); and (ii) the wet processing system, which generates what is called parchment coffee, where the seed is enclosed in the inner integument or endocarp (27). Slow drying in both systems can potentially lead to OTA formation. Effective sun drying or a combination of sun drying and mechanical dehydration provide effective control (150). If grapes for wine production are substantially sound, i.e., free from rotting fruit, and are crushed quickly after harvest, leading rapidly to anaerobic conditions, OTA can be held to low limits (80). If grains are stored under safe conditions, OTA formation will also be minimal (110, 123).

The Effect of Processing on OTA

OTA is largely removed during the winemaking process, as it is bound to solid fractions and sediment. Use of some fining agents can also reduce OTA levels in wine. Red wines retain slightly more OTA than white wines, but overall, the carryover from grapes into finished wine is between 1 and 8% (43, 78, 79). The kinetics of OTA destruction during coffee roasting has been studied at four temperatures (180, 200, 220, and 240°C) over three time periods (44). A reduction of 8 to 98% of OTA was observed, with reduction depending directly on the degree of roasting. Although thorough roasting of coffee beans may destroy OTA effectively, the use of highly contaminated beans in coffee blends is not recommended, as they may affect beverage quality and reduce the level of chlorogenic acid, believed to be beneficial for human health. The effect of processes such as roasting, drink preparation, and instant coffee production on OTA destruction has been reviewed elsewhere (54, 149). Bran and germ removal can affect OTA levels in flour and subsequent products (112). A toxigenic strain of P. verrucosum was grown on two samples of hard and soft wheat, and then wholemeal flour, white flour, and bread were produced from them. Different processes greatly influenced the final levels of OTA in the bread. Milling

SMP_Food Microbiology_CH23.indd

606

hard wheat to produce white flour produced an ~65% reduction in OTA, and a further 10% decrease occurred during baking. Wholemeal flour and bread showed much less reduction in OTA during processing, as might be expected, because less of the grain is discarded (112). The stability of OTA in cereals under heating was studied by growing a toxigenic strain of A. ochraceus on wheat (19). Dry and moistened samples were heated at several temperatures and times, and the times to 50% destruction of OTA under various heating conditions were calculated. A 50% inactivation of OTA at 100, 153, 200, and 250°C required 707, 201, 12, and 6 min, respectively, for dry wheat. For moist wheat, the corresponding inactivation times at 100, 150, and 200°C were 145, 64, and 19 min, respectively. Complete destruction did not occur even at 250°C (19).

Fusarium mycotoxins The most important general observation to be made about Fusarium mycotoxins is that all Fusarium species grow only at high (>0.9) water activities (127), so that toxin production in crops occurs only before harvest or during early stages of drying. Production of mycotoxins ceases long before the crops are fully dried, and only occurs during storage under catastrophic conditions, such as flooding. Production of Fusarium mycotoxins occurs as the result of growth of the causal fungus in the living plant and seed.

Fumonisins

Chemical Characterization

Fumonisins are a family of compounds comprising longchain aliphatic acids with esterified carboxylic acid side chains. Fumonisins are large molecules: the most important, fumonisin B1 (CAS [Chemical Abstracts Service] 116355-30-0), has the molecular formula C34H59NO15, with a molecular weight of 721. Fumonisins are structurally similar to the sphingoid base backbone of sphingolipids, important membrane constituents (7) (Fig. 23.4).

Fungal Sources

Fumonisins are produced by Fusarium verticillioides (known in older literature as Fusarium moniliforme) and some closely related species, in particular Fusarium proliferatum. These species are systemic in maize world­ wide, being always present in the plants and even in healthy kernels (97). It has been shown recently that some fumonisins are also produced by A. niger, an entirely unexpected source. F. verticillioides grows rapidly at 25°C on any standard mycological medium including CYA, MEA, and

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

607

Figure 23.4  Structures of fumonisin B1, deoxynivalenol, and zearalenone. doi:10.1128/9781555818463.ch23f4

PDA (127). Colonies are white to pale salmon-colored, with low and often ropy mycelia and a powdery texture due to production of chains of microconidia. The reverse color on PDA is variable, pale salmon, grayish violet, brownish violet to deep violet. Slow growth occurs at 37°C on these media. Characteristic Fusarium spores, macroconidia, are produced on dichloran chloramphenicol peptone agar (127) or other specialist media. Macroconidia are long (40 to 50 mm) and slender, almost straight, thin-walled, with foot-shaped basal cells. Microconidia are pointed at both ends or club-shaped, 7 to 10 mm long, and are produced in chains from long single phialides (fruiting cells) (127) (Fig. 23.5a and b). F. proliferatum is very similar, but colonies on PDA lack purple colors in the reverse, and microconidia are produced from phialides with more than one fertile neck (82). Fumonisin production in maize is favored by relatively high temperatures (105). Fumonisins are found

SMP_Food Microbiology_CH23.indd

607

primarily in maize and sorghum, as F. verticillioides and F. proliferatum rarely infect other crops. Levels are mainly influenced by local climatic conditions near harvest. Under plant stress, the symptomless endophytic relationship may alter to a disease- and/or mycotoxinproducing interaction (2, 9, 10). It is likely that water stress and insect predation, factors related to the onset of this change in the fungus-plant interaction, may be involved in this conversion (39). Biological interactions between the crop plant, maize, and the fungus are complex (169). Endophytic growth of F. verticillioides within the maize plant may be of benefit to plant growth, as observed in other members of the Gramineae (170). F. ver­ ticillioides has been reported to suppress the growth of Fusarium species that cause Gibberella ear rot (131). Frisvad et al. (50) reported for the first time the production of fumonisin B2 in A. niger, including the ex-type culture that had been fully sequenced (118). Until that time, production of fumonisin had been reported only in

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

608

Figure 23.5  (a) F. verticillioides, PDA, 7 days at 25°C; (b) F. verticillioides macroconidia (bar = 10 mm). (c) F. graminearum, PDA, 7 days at 25°C; (d) F. graminearum macroconidia (bar = 10 mm). doi:10.1128/9781555818463.ch23f5

species of Fusarium. A. niger has been shown to produce fumonisin B2 and fumonisin B4 on grapes and raisins (102), while fumonisin B2 was produced on coffee­ (108). A. niger is among the fungi most commonly ­ reported from foods, and the possibility of co-­occurrence of OTA and fumonisin B2 in foods is of concern.

Toxicology

Fumonisins are remarkable for the wide range of effects caused in animals and man. Fumonisins act by inhibiting ceramide synthase, causing accumulation of intermediates of sphingolipid metabolism, and also causing depletion of complex sphingolipids. These effects interfere with the function of some membrane proteins, including folate binding. The most dramatic effect occurs in horses, where the disease called equine leukoencephalomalacia occurs. It is a rapidly progressing disease that turns equine brains to mush. For horses, consumption of feed containing >10 mg/kg fumonisin B1 in the diet (equivalent to 0.2 mg/kg body weight per

SMP_Food Microbiology_CH23.indd

608

day) was associated with increased risk of developing this disease (167). In pigs, fumonisins cause pulmonary edema due to left ventricle heart failure (167). In rats, the primary effect is liver cancer (52), but fumonisins also cause programmed cell death (apoptosis) (153). It is unusual for a single toxin to have such diverse effects in different animal species, and the reasons for that remain unknown (35). In humans, fumonisins and F. verticillioides are associated with esophageal cancer. Extensive studies in areas of low and high maize consumption in South Africa have established this connection (132, 167). This disease is also prevalent in areas of China (148) and occurs at significantly higher levels than background also in parts of Iran, northern Italy, Kenya, and a small area of the southern United States (167). In all of those areas consumption of maize and maize products is very high. Some evidence exists that high intake of fumonisins from maize may be associated with neural tube defects

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

609

such as spina bifida in areas of Guatemala, South Africa, China, and a population along the Texas-Mexico border (91). In animal models folate supplementation has been shown to reduce the incidence of neural tube defects (53). Numerous studies have shown that antioxidants can prevent some of the cytotoxic effects of fumonisins and other mycotoxins (58, 167).

The United States has established guidelines for fumonisin levels: dry milled grain products should contain no more than 2 mg/kg of total fumonisins (46). In the European Union, the maximum limit is 1 mg/kg for maize flour, semolina, germ, and oil and 0.4 mg/kg for products ready for consumption, except that for babies and children the limit is 0.2 mg/kg (40).

Chemical Analysis

Control

Screening methods for fumonisin determination in foods have been based either on TLC separation after cleanup or more commonly on enzyme-linked immunosorbent assays. Most analytical techniques have concentrated on the estimation of fumonisin B1.

Extraction

The most effective method for extraction of fumonisins is shaking with methanol and water (75:25) (139).

Purification

Of the methods considered by various authors, the most satisfactory is the use of a SAX solid-phase extraction cartridge (Varian Bond, Harbor City, CA) (139).

Assays

The traditional methodology of TLC is still in use, and is recommended for less developed economies as it is inexpensive and reliable. For fumonisins a reversedphase system is preferable. It has been shown that derivatization with fluorescamine before spotting on the TLC plate was more effective than after running. The preferred developing medium is methanol/4% aqueous potassium chloride (70:30) (139). A suitable developing solvent is benzene/methanol/acetic acid (18:1:1). Fumonisin B1 is visualized under long-wave UV light as a greenish yellow spot with Rf of ~0.35. Levels down to at least 1.0 mg/kg can be estimated. Enzyme-linked immunosorbent assay techniques are also effective for fumonisin analysis. Although antibodies are raised against fumonisin B1, cross-reactivity­ usually occurs with fumonisins B2 and B3 as well. Commercial kits are available from several manufacturers, and a range of methods have been described (17).

Occurrence and Regulation in Foods

The major source of fumonisins in foods is maize, though other small grains may have low levels at times. With the discovery that A. niger may produce fumonisins,­ the range of foodstuffs where fumonisins may be found has become much wider. The toxicological implications of this discovery have not yet been assessed, but are unlikely to have a major impact on food safety.

SMP_Food Microbiology_CH23.indd

609

Fumonisins occur in maize preharvest. It is known that drought stress and insect damage are major factors inducing production, so good agricultural practice, rain, irrigation, and the cultivation of Bt cultivars are all important in limiting fumonisin formation. Some progress has been made in breeding cultivars resistant to ear rot (97, 105). Freshly harvested maize should be rapidly dried to a suitable moisture level immediately (41, 97, 106, 161), but this is important only in the initial stages of drying because Fusarium species do not grow below about 0.9 aw, so once the kernel moisture content has been reduced below that figure, fumonisin accumulation ceases. This frequently occurs in field drying before harvest of the cobs. For the same reason, fumonisins will not be produced in storage. Even if very high moisture occurs due to water ingress, competition with other micro­organisms at such high water activities will prevent any significant increase in fumonisin levels. In most geographical areas, the main methods for controlling fumonisin levels are visual inspection of lots for fungal damage, followed by fumonisin analyses and rejection of lots that do not meet specifications. Visible sorting of maize grain as a technique to reduce fumonisin levels by subsistence farmers has been proposed (3) and optimized (157, 158). In Central America, the process of nixtamalization removes almost all fumonisins, resulting in tortillas and other maize-based foods being substantially free of these mycotoxins (34).

Effect of Processing on Fumonisins

Maize is wet milled to obtain maize starch, germ, and fiber, while dry milling produces bran, germ, and fractions of decreasing particle size—grits, corn meal, and flour (4). Fumonisins are not destroyed during these processes and are found in all fractions, with higher concentrations in bran and germ (20, 70, 119). Fumonisin levels are reduced by processing at temperatures above 150°C, including maize meal production (138), frying (66), baking, roasting, and alkaline cooking (25, 66, 71). Extrusion processing is used extensively in the production of breakfast cereal, snack,

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

610 and textured foods. The highest reduction in fumonisin levels occurs at extrusion temperatures of 160°C or higher and in the presence of glucose (22, 23, 34).

Deoxynivalenol

Chemical Characterization

Deoxynivalenol (DON) belongs to the family of chemicals known as the trichothecenes, sesquiterpenoid compounds that are characterized by a 12,13-epoxy ring. At least 100 trichothecene molecules are known, differentiated by hydroxyl or acetyl groups and side chains. DON, still sometimes known as vomitoxin in the United States, is 12,13-epoxy-3,7,15-trihydroxytrichothec-9-en-8-one, CAS number 51481-10-8 (31) (Fig. 23.4). It is the most commonly produced trichothecene. Nivalenol differs from DON by the substitution of a hydroxyl group for the hydrogen atom at the 4 position. The most toxic of these molecules is known as T-2 toxin, but it is not commonly produced in foodstuffs.

Fungal Sources

DON is produced by Fusarium graminearum (often listed as Gibberella zeae, its sexual stage), Fusarium culmorum, and less commonly some related species. F. graminearum grows rapidly on any standard mycological medium including CYA, MEA, and PDA (127). Colonies on CYA and MEA are colored grayish rose, grayish yellow, or paler, with reverses orange red to yellowish brown. On PDA, colonies are colored yellowish brown to reddish brown, sometimes with a central mass of red brown to orange areas bearing macroconidia, with the reverse dark red. Characteristic Fusarium conidia, macroconidia, are produced on dichloran chloramphenicol peptone agar or other specialist medium, usually with five septa, thick-walled, straight to moderately curved, with the basal cell distinctly foot-shaped; microconidia are not produced (127) (Fig. 23.5c and d). Colonies of F. culmorum are similar to those of F. graminearum, but colored pale red to pastel red on CYA, MEA, and PDA; the reverse on CYA and PDA is pastel red to deep red and on MEA brown to reddish brown. Macroconidia are relatively short, wide, and only slightly curved, with four or five septa, 30 to 45 mm long, with basal cells with a slight to definite notch. Microconidia are not produced. F. graminearum occurs in maize, and both F. gra­ minearum and F. culmorum in small grains, especially wheat and barley. These species are frank pathogens,

SMP_Food Microbiology_CH23.indd

610

invading plants and grains by causing diseases known as Gibberella ear rot in maize and and Fusarium head blight in wheat, barley, and triticale. Epidemics of Gibberella ear rot require the congruence of three factors: airborne or insect-borne spores, inoculation at the susceptible time, and appropriate moisture and temperature (97, 105). This disease is prevalent in north temperate climates especially in wet years, and much less common in the tropics. Fusarium head blight affects all commercial cultivars of wheat and barley. The formation of DON or nivalenol by F. graminearum depends on the geographical origin of the fungal strain (99, 160), while F. culmorum always produces DON (99, 154).

Toxicology

Like all trichothecenes, DON and nivalenol are inhibitors of protein synthesis (42). Nivalenol is much more toxic than DON, but is produced in much lower quantities in grains, and is not considered to be a significant mycotoxin (100). Trichothecene toxicoses in humans appear to be rare (147), but DON can cause gastro­ intestinal problems and immunotoxicity in humans (18). Feed refusal by pigs or more serious problems can cause economic losses in domestic animals too (114). Apart from a general effect on protein synthesis, the mechanism of action of DON remains unclear (167).

Chemical Analysis

Analysis for DON usually requires gas chromatography and mass spectroscopy (6, 101). Masked DON, in the form of the 3-glucoside, is a problem because it is not assayed by the usual techniques (135). Extraction may use chloroform-ethyl acetate or 70% methanol, and cleanup is accomplished by filtering through a C18 or silica gel column (6).

Occurrence and Regulations

DON is found in maize and small grains in all areas where these crops are grown. It is especially prevalent in cooler areas where rainfall is higher, such as Canada (137) and Argentina (115), and occurs less commonly in drier, hotter areas such as Australia (124). A number of studies have shown that fungal infection rates are higher in crops planted in fields previously planted with maize, particularly when residues from those crops were left in the field. Once grains are dried, increases in levels of Fusarium mycotoxins rarely occur. Favorable weather conditions are critical for infection to occur in wheat heads. Research-based field ob-

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

611

servations have confirmed that temperature and moist conditions during heading and anthesis are the major factors of importance (60). The U.S. Food and Drug Administration has issued guidelines for DON in foods and feeds as follows: for finished wheat products that may be consumed by humans, 1 mg kg–1; for grains and by-products for feedlot and dairy cattle, 10 mg kg–1, except that for dairy cattle the total DON content in feed should not exceed 5 mg kg–1; in feed for pigs, 5 mg kg–1, but not exceeding 20% of the total diet; and for all other animals, 5 mg kg–1, not exceeding 40% of the total diet (47).

Control

Some success has been achieved in controlling DON formation in wheat by the use of azole fungicides at anthesis (116, 171). Forecasting systems to advise farmers of the likelihood of DON formation have been developed in Canada (136) and Europe (56, 156). Otherwise, control relies on reducing levels of Fusarium species in the field by good management and crop rotation (167).

Zearalenone

Chemical Characterization

Zearalenone (ZEA) is an estrogenic mycotoxin, described as belonging to the resorcylic acid lactone group. Its molecular formula is C18H22O5 and CAS registry number is 17924-92-4 (29) (Fig. 23.4).

Fungal Sources

ZEA is produced by the same Fusarium species that produce DON and nivalenol, and generally speaking under the same conditions, the main sources being maize and small grains (172).

Toxicology

ZEA has a low acute toxicity, but it and its metabolites possess estrogenic activity in pigs, cattle, and sheep. The most obvious problems are seen in pigs: doses of ZEA as low as 1 to 5 mg/kg can induce vulvovaginitis and vaginal and rectal prolapse in young female pigs (113). ZEA is also considered to be hepatotoxic, hematotoxic, immunotoxic, and genotoxic, but no studies of human carcinogenicity have been reported. The nature of its toxicity is not completely understood (172). Metabolism of ZEA involves reduction of the 6-keto group, resulting in the formation of a- and b-zearalenol. a-Zearalenol has greater estrogenic activity than ZEA (45). ZEA has been implicated in early puberty and ad-

SMP_Food Microbiology_CH23.indd

611

vanced growth of girls in Hungary and Italy (94). Its importance in human health is still poorly understood.

Analysis

ZEA is usually analyzed by TLC (6).

Occurrence and Regulation

ZEA occurs in the same situations as DON, i.e., maize and small grains, and it occurs in all regions of the world (see reference 172 for a detailed review). No internationally recognized limits for ZEA exist, with specific regulations varying from 20 to 1,000 mg/ kg or more (172). JECFA has established a provisional maximum tolerable daily intake of ZEA and its metabolites of 0.5 mg/kg body weight per day (166).

Control

Control of ZEA in crops is similar to that related to DON.

Genomics Genomics, the study of entire genomes, provides basic information to build the knowledge base of gene function that will assist in understanding mycotoxin formation and reduction in crops. Understanding the molecular basis of plant resistance, where it exists, will hopefully lead to identification of biochemical factors for use in plant breeding or genetic engineering (133). Studies on economically important fungi at the genomic level will assist in understanding mycotoxin biosynthesis and also help to understand the biology, evolution, biochemical function, and genetic regulation of the genes in these fungal systems. Genomic studies have been carried out on several toxigenic fungi, including A. flavus, A. niger, F. graminearum, and F. verticillioides (15, 16). Genome sequences have been determined for some mycotoxigenic fungi, notably F. graminearum, F. verticillioides, and Fusarium solani. The availability of large amounts of genomic and expressed sequence tag data has provided tools for large-scale functional analysis of gene expression in F. graminearum and F. verticillioides. The sequencing of the A. niger genome led to a most unexpected result. Searches for biochemically active genes showed the presence of the genes for ­ fumonisin (11), which was independently verified (118). It was soon confirmed that this gene cluster (consisting of at least 15 genes) was complete and active, and the ability of at least some strains of A. niger to produce ­fumonisins was confirmed (50). Indeed, recent information indicates that fumonisin production by A. niger is

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

612 c­ ommon. In one study on A. niger strains from a sample of Californian raisins, 50 of 66 strains (77%) produced fumonisins (102). In a second study, where isolates were taken from 13 samples of dried vine fruits from several countries, 20 of 30 (67%) were producers (159). Research on Fusarium head blight has used fungal genomics approaches to determine the role of DON in pathogenicity and the effect of this toxin accumulation on crop production. The information may be useful in developing resistant germplasm (133).

References 1. Abarca, M. L., F. Accensi, M. R. Bragulat, G. Castellá, and F. J. Cabanes. 2003. Aspergillus carbonarius as the main source of ochratoxin A contamination in dried vine fruits from Spanish market. J. Food Prot. 66:504–506. 2. Abbas, H. K., R. D. Cartwright, and W. T. Shier. 2006. Aflatoxin and fumonisin contamination of corn (maize, Zea mays) hybrids in Arkansas. Crop Prot. 25:1–9. 3. Afolabi, C. G., R. Bandyopadhyay, J. F. Leslie, and E. J. Ekpo. 2006. Effect of sorting on incidence and occurrence of fumonisins and Fusarium verticillioides on maize from Nigeria. J. Food Prot. 69:2019–2023. 4. Alexander, R. J. 1988. Maize dry milling: processes, products and applications, p. 351–376. In S. A. Watson and P. E. Ramstad (ed.), Maize: Chemistry and Technology. American Association of Cereal Chemists, St. Paul, MN. 5. Almeida, L., V. Rabe, B. Sanchez, F. Torrella, J. P. López-Pérez, J. A. Gabaldon, and L. Guardiola. 2007. Ochratoxin A in red paprika: relationship with the origin of the raw material. Food Microbiol. 24:319–327. 6. AOAC International. 2005. Official Methods of Analysis of AOAC International, 18th ed. AOAC International, Gaithersburg, MD. 7. ApSimon, J. W. 2001. Structure, synthesis, and biosynthesis of fumonisin B1 and related compounds. Environ. Health Perspect. 109(Suppl. 2):S245–S249. 8. Azziz-Baumgartner, E., K. Lindblade, K. Gieseker, H. S. Rogers, S. Kieszak, H. Njapau, R. Schleicher, L. F. McCoy, A. Misore, K. DeCock, C. Rubin, and L. Slutsker. 2005. Case-control study of an acute aflatoxicosis outbreak, Kenya, 2004. Environ. Health Perspect. 113:1779–1783. 9. Bacon, C. W., and P. E. Nelson. 1994. Fumonisin production in corn by toxigenic strains of Fusarium mo­ niliforme and Fusarium proliferatum. J. Food Prot. 57:514–521. 10. Bacon, C. W., and D. M. Hinton. 1996. Symptomless endophytic colonization of corn by Fusarium monili­ forme. Can. J. Bot. 74:1195–1202. 11. Baker, S. E. 2006. Aspergillus niger genomics: past, present and into the future. Med. Mycol. 44:S17–S21. 12. Battilani, P., and A. Pietri. 2002. Ochratoxin A in grapes and wine. Eur. J. Plant Pathol. 108:639–643.

SMP_Food Microbiology_CH23.indd

612

13. Bayman, P., J. L. Baker, and N. E. Mahoney. 2002. Aspergillus on tree nuts: incidence and associations. Mycopathologia 155:161–169. 14. Bellí, N., S. Marín, V. Sanchis, and A. J. Ramos. 2006. Impact of fungicides on Aspergillus carbonarius growth and ochratoxin A production on synthetic grape-like medium and on grapes. Food Addit. Contam. 23:1021–1029. 15. Bhatnagar, D., J. Yu, and T. E. Cleveland. 2002. Applying the genomic wrench—new tool for an old problem. Mycopathologia 155:159. 16. Bhatnagar, D., R. Proctor, G. A. Payne, J. Wilkinson, J. Yu, T. E. Cleveland, and W. C. Nierman. 2006. Genomics of mycotoxigenic fungi, p. 157– 177. In D. Barug, D. Bhatnagar, H. P. van Egmond, J. W. van der Kamp, W. A. Van Osenbruggen, and A. Visconti (ed.), The Mycotoxin Factbook. Wageningen Academic Publishers, Wageningen, The Netherlands. 17. Bird, C. B., B. Malone, L. G. Rice, P. F. Ross, R. Eppley, and M. M. Abouzied. 2002. Determination of total fumonisins in corn by competitive direct enzyme-linked immunosorbent assay: collaborative study. J. AOAC Int. 85:404–410. 18. Bondy, G. S., and J. J. Pestka. 2000. Immunomodulation by fungal toxins. J. Toxicol. Environ. Health B 3:109–143. 19. Boudra, H., P. Le Bars, and L. Le Bars. 1995. Thermostability of ochratoxin A in wheat under two moisture conditions. Appl. Environ. Microbiol. 61:1146–1158. 20. Brera, C., F. Debegnach, S. Grossi, and M. Miraglia. 2003. Effect of industrial processing on the distribution of fumonisin B1 in dry milling corn fractions. J. Food Prot. 67:1261–1266. 21. Brown, M. P., C. S. Brown-Jenco, and G. A. Payne. 1999. Genetic and molecular analysis of aflatoxin biosynthesis. Fungal Genet. Biol. 26:81–98. 22. Bullerman, L. B., and A. Bianchini. 2007. Stability of mycotoxins during food processing. Int. J. Food Microbiol. 119:140–146. 23. Bullerman, L. B., A. Bianchini, M. A. Hanna, L. S. Jackson, J. Jablonski, and D. Ryu. 2008. Reduction of fumonisin B1 in corn grits by single-screw extrusion J. Agric. Food Chem. 56:2400–2405. 24. CAST. 2003. Mycotoxins: Risks in Plant and Animal Systems. Task Force Report 138. Council for Agricultural Science and Technology, Ames, IA. 25. Castelo, M. M., S. S. Sumner, and L. B. Bullerman. 1998. Occurrence of fumonisins in maize-based food products. J. Food Prot. 61:704–707. 26. Cazzaniga, D., J. C. Basílico, R. J. González, R. L. Torres, and D. M. de Greef. 2001. Mycotoxins inactivation by extrusion cooking of corn flour. Lett. Appl. Microbiol. 33:144–147. 27. Codex Alimentarius. 2009. Code of practice for the prevention and reduction of ochratoxin A contamination in coffee (CAC/RCP 69-2009). Joint FAO/WHO Food Standards Programme, FAO, Rome, Italy. 28. Coker, R. D., M. J. Nagler, P. R. Defize, G. B. Derksen, H. Buchholz, H. A. Putzka, H. P. Hoogland, A. H. Roos, and A. Boenke. 2000. Sampling plans for the

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

613

determination of aflatoxin B1 in large shipments of animal feedstuffs. J. AOAC Int. 83:1252–1258. 29. Cole, R. J., and M. A. Schweikert. 2003. Handbook of Secondary Fungal Metabolites, vol. I. Academic Press, San Diego, CA. 30. Cole, R. J., R. A. Hill, P. D. Blankenship, T. H. Sanders, and K. H. Garren. 1982. Influence of irrigation and drought stress on invasion by Aspergillus flavus of corn kernels and peanut pods. Dev. Ind. Microbiol. 23:229–236. 31. Cole, R. J., B. B. Jarvis, and M. A. Schweikert. 2003. Handbook of Secondary Fungal Metabolites, vol. III. Academic Press, San Diego, CA. 32. Copetti, M. V., J. L. Pereira, B. T. Iamanaka, J. I. Pitt, and M. H. Taniwaki. 2010. Ochratoxigenic fungi and ochratoxin A in cocoa during farm processing. Int. J. Food Microbiol. 143:67–70. 33. Cotty, P. 1994. Influence of field application of an atoxigenic strain of Aspergillus flavus on the populations of Aspergillus flavus infecting cotton bolls and on the aflatoxin content of cottonseed. Phytopathology 84:1270–1277. 34. De la Campa, R., J. D. Miller, and K. Hendricks. 2004. Fumonisin in tortillas produced in small-scale facilities and effect of traditional masa production methods on this mycotoxin. J. Agric. Food Chem. 52:4432–4437. 35. Desjardins, A. E. 2006. Fusarium Mycotoxins: Chemistry, Genetics and Biology. APS Press, St. Paul, MN. 36. Dorner, J. W., R. J. Cole, and P. D. Blankenship. 1998. Effect of inoculum rate of biological control agents on preharvest aflatoxin contamination of peanuts. Biol. Control 12:171–176. 37. Doster, M. A., and T. J. Michailides. 1995. The relationship between date of hull splitting and decay of pistachio nuts by Aspergillus species. Plant Dis. 79:766–769. 38. Doster, M. A., T. J. Michailides, and D. P. Morgan. 1996. Aspergillus species and mycotoxins in figs from California orchards. Plant Dis. 80:484–489. 39. Dowd, P. F. 2003. Insect management to facilitate preharvest mycotoxin management. J. Toxicol. Toxin Rev. 22:327–350. 40. European Commission. 2005. Commission Recom­men­ dation on the prevention and reduction of Fusarium toxins in cereals and cereal products. Draft SANCO/1719/2005. European Commission, Brussels, Belgium. 41. Fandohan, P., B. Gnonlonfin, K. Hell, W. F. O. Marasas, and M. J. Wingfield. 2005. Natural occurrence of Fusarium and subsequent fumonisin contamination in preharvest and stored maize in Benin, West Africa. Int. J. Food Microbiol. 99:173–183. 42. Feinberg, B., and C. S. MacLaughlin. 1989. Biochemical mechanism of action of trichothecene mycotoxins, p. 27– 36. In V.R. Beasley (ed.), Trichothecene Mycotoxins: Pathophysiological Effects, vol. 1. CRC Press, Boca Raton, FL. 43. Fernandes, A., N. Ratola, A. Cerdeira, A. Alves, and A. Venancio. 2007. Changes in ochratoxin A concentration during winemaking. Am. J. Enol. Viticul. 58:92–96.

SMP_Food Microbiology_CH23.indd

613

44. Ferraz, M. B. M., A. Farrah, B. T. Iamanaka, D. Perrone, M. V. Copetti, V. X. Marques, A. A. Vitali, and M. H. Taniwaki. 2010. Kinetics of ochratoxin A destruction during coffee roasting. Food Control 21:872–877. 45. Fink-Gremmels, J., and H. Malekinejad. 2007. Clinical effects and biochemical mechanisms associated with exposure to the mycoestogen zearalenone. Animal Feed Sci. Technol. 137:326–341. 46. Food and Drug Administration. 2001. Fumonisin Lev­ els in Human Foods and Animal Feeds. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, College Park, MD. 47. Food and Drug Administration. 2010. Guidance for Industry and FDA: Advisory Levels for Deoxynivalenol (DON) in Finished Wheat Products for Human Con­ sumption and Grains and Grain By-Products Used for Animal Feed. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, College Park, MD. 48. Frisvad, J. C., U. Thrane, R. A. Samson, and J. I. Pitt. 2006. Important mycotoxins and the fungi which produce them, p. 3–31. In A. D. Hocking, J. I. Pitt, R. A. Samson, and U. Thrane (ed.), Advances in Food Mycology. Springer, New York, NY. 49. Frisvad, J. C., J. M. Frank, J. A. Houbraken, A. F. Kuijpers, and R. A. Samson. 2004. New ochratoxin A producing species of Aspergillus section Circumdati. Stud. Mycol. 50:23–43. 50. Frisvad, J. C., J. Smedsgaard, R. A. Samson, T. O. Larsen, and U. Thrane. 2007. Fumonisin B2 production by Aspergillus niger. J. Agric. Food Chem. 55:9727–9732. 51. Garcia, S., and N. Heredia. 2006. Mycotoxins in Mexico: epidemiology, management, and control strategies. Mycopathologia 162:255–264. 52. Gelderblom, W. C., N. P. Kriek, W. F. Marasas, and P. G. Theil. 1991. Toxicity and carcinogenicity of the Fusarium moniliforme metabolite fumonisin B1 in rats. Carcinogenesis 12:1247–1251. 53. Gelineau-van Waes, J., K. A. Voss, V. L. Stevens, M. C. Speer, and R. T. Riley. 2009. Maternal fumonisin exposure as a risk factor for neural tube defects. Adv. Food Nutr. Res. 56:145–181. 54. Gollucke, A. P. B., D. Q. Tavares, and M. H. Taniwaki. 2004. Efeito do processamento sobre a ocratoxina A, em café. Higiene Alimentar. 18:38–48. 55. Gong, Y. Y., K. Cardwell, A. Hounsa, S. Egal, P. C. Turner, A. J. Hall, and C. P. Wild. 2002. Dietary aflatoxin exposure and impaired growth in young children from Benin and Togo: cross sectional study. Br. Med. J. 325:20–21. 56. Gourdain, E., F. Piraux, and B. Barrier-Guillot. 2011. A model combining agronomic and weather factors to predict occurrence of deoxynivalenol in durum wheat kernels. World Mycotoxin J. 4:129–139. 57. Gurses, M. 2006. Mycoflora and aflatoxin content of hazelnuts, walnuts, peanuts, almonds and roasted chickpeas (leblebi) sold in Turkey. Int. J. Food Prop. 9:395–399. 58. He, Q., J. Kim, and R. P. Sharma. 2004. Silymarin protects against liver damage in BALB/c mice exposed to

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

614

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69. 70.

71.

fumonisin B1 despite increasing accumulation of free sphingoid bases. Toxicol. Sci. 80:335–342. Highley, E., E. Wright, H. J. Banks, and B. R. Champ (ed.). 1994. Stored Product Protection. Proceedings of the 6th International Working Conference on Stored Product Protection. CAB International, Wallingford, United Kingdom. Hooker, D. C., A. W. Schaafsma, and L. TamburicIlincic. 2002. Using weather variables pre- and postheading to predict deoxynivalenol in winter wheat. Plant Dis. 88:611–619. Horn, B. W., and J. W. Dorner. 1999. Regional differences in production of aflatoxin B1 and cyclopiazonic acid by soil isolates of Aspergillus flavus along a transect within the United States. Appl. Environ. Microbiol. 65:1444–1449. Iamanaka, B. T., M. H. Taniwaki, H. C. Menezes, E. Vicente, and M. H. Fungaro. 2005. Incidence of toxigenic fungi and ochratoxin A in dried fruits sold in Brazil. Food Addit. Contam. 22:1258–1263. International Agency for Research on Cancer. 1993. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, vol. 56. Some Naturally Occurring Substances: Food Items and Constituents, Heterocyclic Aromatic Amines and Mycotoxins. International Agency for Research on Cancer, Lyon, France. International Agency for Research on Cancer. 2002. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, vol. 82. Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene. International Agency for Research on Cancer, Lyon, France. International Commission on Microbiological Speci­ fications for Foods. 1996. Toxigenic fungi: Aspergillus, p. 347–381. In Microorganisms in Foods. 5. Micro­ biological Specifications of Food Pathogens. Blackie Academic and Professional, London, United Kingdom. Jackson, L. S., S. K. Katta, D. D. Fingerhut, J. W. DeVries, and L. B. Bullerman. 1997. Effects of baking and frying on the fumonisin B1 content of corn-based foods. J. Agric. Food Chem. 45:4800–4805. Jaime-Garcia, R., and P. J. Cotty. 2007. Influence of application timing on sporulation of the biocontrol product Aspergillus flavus AF36 in cotton fields of Arizona. Phytopathology 97:S26. Joosten, H. M. L. J., J. Goetz, A. Pittet, M. Schellenberg, and P. Bucheli. 2001. Production of ochratoxin A by Aspergillus carbonarius on coffee cherries. Int. J. Food Microbiol. 65:39–44. Jorgensen, K. 2005. Occurrence of ochratoxin A in commodities and processed food—a review of EU occurrence data. Food Addit. Contam. 25(Suppl. 1):S26–S30. Katta, S. K., A. E. Cagampang, L. S. Jackson, and L. B. Bullerman. 1997. Distribution of Fusarium molds and fumonisins in dry-milled corn fractions. Cereal Chem. 74:858–883. Katta, S. K., L. S. Jackson, S. S. Sumner, M. A. Hanna, and L. B. Bullerman. 1999. Effect of temperature and

SMP_Food Microbiology_CH23.indd

614

screw speed on stability of fumonisin B1 in extrusioncooked maize grits. Cereal Chem. 76:16–20. 72. Klich, M. A., and M. A. Chmielewski. 1985. Nectaries as entry sites for Aspergillus flavus in developing cotton bolls. Appl. Environ. Microbiol. 50:602–604. 73. Krska, R., P. Schubert-Ullrich, A. Molinelli, M. Sulyok, S. MacDonald, and C. Crews. 2008. Mycotoxin analysis: an update. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 25:152–163. 74. Kuiper-Goodman, T., C. Hilts, S. M. Billiard, Y. Kiparissis, I. D. K. Richard, and S. Hayward. 2010. Health risk assessment of ochratoxin A for all age-sex strata in a market economy. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 27:212–240. 75. Larsen, T. O., A. Svendsen, and J. Smedsgaard. 2001. Biochemical characterization of ochratoxin A-producing strains of the genus Penicillium. Appl. Environ. Microbiol. 67:3630–3635. 76. Le Bars, J. 1990. Contribution to a practical strategy for preventing aflatoxin contamination of dried figs. Microbiol. Aliment. Nutr. 8:265–270. 77. Leong, S. L., A. D. Hocking, and J. I. Pitt. 2004. Occurrence of fruit rot fungi (Aspergillus section Nigri) on some drying varieties of irrigated grapes. Aust. J. Grape Wine Res. 10:83–88. 78. Leong, S. L., A. D. Hocking, and E. S. Scott. 2006. Effect of temperature and water activity on growth and ochratoxin A production by Australian Aspergillus car­ bonarius and A. niger isolates on a simulated grape juice medium. Int. J. Food Microbiol. 110:209–216. 79. Leong, S. L., A. D. Hocking, P. Varelis, G. Giannikopoulos, and E. S. Scott. 2006. Fate of ochratoxin A during vinification of Semillon and Shiraz grapes. J. Agric. Food Chem. 54:6460–6464. 80. Leong, S. L., A. D. Hocking, J. I. Pitt, B. A. Kazi, R. W. Emmett, and E. S. Scott. 2006. Australian research on ochratoxigenic fungi and ochratoxin A. Int. J. Food Microbiol. 111(Suppl. 1):S10–S17. 81. Leong, S. L., L. T. Hien, T. V. An, N. T. Trang, A. D. Hocking, and E. S. Scott. 2007. Ochratoxin A-producing Aspergilli in Vietnamese green coffee. Lett. Appl. Microbiol. 45:301–306. 82. Leslie, J. F., and B. A. Summerell. 2006. The Fusarium Laboratory Manual. Blackwell Publishing Professional, Ames, IA. 83. Lewis, L., M. Onsongo, H. Njapau, H. Schurz-Rogers, G. Luber, S. Kieszak, J. Nyamongo, J. Backer, A. M. Dahiye, A. Misore, K. DeCock, and C. Rubin. 2005. Aflatoxin contamination of commercial maize products during an outbreak of acute aflatoxicosis in eastern and central Kenya, 2004. Environ. Health Perspect. 113:1763–1767. 84. Lillehoj, E. B., W. W. McMillian, W. D. Guthrie, and D. Barry. 1980. Aflatoxin-producing fungi in preharvest corn: inoculum source in insects and soils. J. Environ. Qual. 9:691–694. 85. Lombaert, G. A., P. Pellaers, G. Neumann, D. Kitchen, V. Huzel, R. Trelka, S. Kotello, and P. M. Scott. 2004.

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

615

Ochratoxin A in dried vine fruits on the Canadian retail market. Food Addit. Contam. 21:578–585. 86. Lund, F., and J. C. Frisvad. 2003. Penicillium verrucosum in wheat and barley indicates presence of ochratoxin A. J. Appl. Microbiol. 95:1117–1123. 87. Ministry of Agriculture, Fisheries and Food. 1997. Survey of Aflatoxins and Ochratoxin A in Cereals and Retail Products. MAFF Food Surveillance Information Sheet 130. U.K. Ministry of Agriculture, Fisheries and Food, London, United Kingdom. 88. Magnoli, C., A. Astoreca, L. Ponsone, M. Fernandez-Juri, S. Chiacchiera, and A. M. Dalcero. 2006. Ochratoxin A and the occurrence of ochratoxin A-producing black aspergilli in stored peanut seeds from Córdoba, Argentina. J. Sci. Food Agric. 86:2369–2373. 89. Magnoli, C., C. Hallak, S. Chiacchiera, and A. M. Dalcero. 2006. Occurrence of ochratoxin A-producing fungi in commercial corn kernels in Argentina. Mycopathologia 161:53–58. 90. Marasas, W. F., and H. F. Vismer. 2003. Food and thought about mycotoxins, organic and genetically modified foods, p. 423–427. In P. F. Credfield, D. M. Armitage, C. H. Bell, P. M. Cogan, and E. Highley (ed.), Advances in Stored Product Protection. CAB International, Wallingford, United Kingdom. 91. Marasas, W. F., R. T. Riley, K. A. Hendricks, V. L. Stevens, T. W. Sadler, J. Gelineau-van Waes, S. A. Missmer, J. Cabrera, O. Torres, W. C. Gelderblom, J. Allegood, C. Martínez, J. Maddox, J. D. Miller, L. Starr, M. C. Sullards, A. V. Roman, K. A. Voss, E. Wang, and A. H. Merrill, Jr. 2004. Fumonisins disrupt sphingolipid metabolism, folate transport, and neural tube development in embryo culture and in vivo: a potential risk factor for human neural tube defects among populations consuming fumonisin-contaminated maize. J. Nutr. 134:711–716. 92. Marquardt, R. R., and A. A. Frohlich. 1992. A review of recent advances in understanding ochratoxicosis. J. Anim. Sci. 70:3968–3988. 93. Martins, M. L., H. M. Martins, and A. Gimeno. 2003. Incidence of microflora and of ochratoxin A in green coffee beans (Coffea arabica). Food Addit. Contam. 20:1127–1131. 94. Massart, F., V. Meuca, G. Saggese, and G. Soldani. 2008. High growth rate of girls with precocious puberty exposed to estrogenic mycotoxins. J. Pediatrics 152:690–695. 95. Méndes-Albores, J. A., G. Arámbula-Villa, M. G. LoarcaPiña, J. Gonzalez-Hernández, E. Castaño-Tostado, and E. Moreno-Martínez. 2004. Aflatoxins’ fate during the nixtamalization of contaminated maize by two tortillamaking processes. J. Stored Prod. Res. 40:87–94. 96. Meyvaci, K. B., A. Altindisli, U. Aksoy, R. Eltem, H. Turgut, Z. Arasiler, and N. Kartal. 2005. Ochratoxin A in sultanas from Turkey 1. Survey of unprocessed sultanas from vineyards and packing-houses. Food Addit. Contam. 22:1138–1143. 97. Miller, J. D. 1994. Epidemiology of Fusarium ear diseases of cereals. p. 19–36. In J. D. Miller and H. L.

SMP_Food Microbiology_CH23.indd

615

Trenholm (ed.), Mycotoxins in Grain—Compounds other than Aflatoxin. Eagan Press, St. Paul, MN. 98. Miller, J. D. 1995. Fungi and mycotoxins in grains: implications for stored products research. J. Stored Prod. Res. 31:1–16. 99. Miller, J. D., R. Greenhalgh, Y. Z. Wang, and M. Lu. 1991. Mycotoxin chemotypes of three Fusarium species. Mycologia 83:121–130. 100. Miller, J. D., J. W. ApSimon, B. A. Blackwell, R. Greenhalgh, and A. Taylor. 2001. Deoxynivalenol: a 25 year perspective on a trichothecene of agricultural importance, p. 310–320. In B.A. Summerell, J. F. Leslie, D. Backhouse, W. L. Bryden, and L. W. Burgess (ed.), Fusarium: Paul E. Nelson Memorial Symposium. APS Press, St. Paul, MN. 101. Mirocha, C. J., E. Kolaczkowski, W. P. Xie, H. Yu, and H. Jelen. 1999. Analysis of deoxynivalenol and its derivatives (batch and single kernel) using gas chromatography mass specrometry. J. Agric. Food Chem. 46:1414–1418. 102. Mogensen, J. M., J. C. Frisvad, U. Thrane, and K. F. Nielsen. 2010. Production of fumonisin B2 and B4 by Aspergillus niger on grapes and raisins. J. Agric. Food Chem. 58:954–958. 103. Morello, L. G., D. Sartori, A. L. Martinez, M. L. Vieira, M. H. Taniwaki, and M. H. Fungaro. 2007. Detection and quantification of Aspergillus westerdijkiae in coffee beans based on selective amplification of b-tubulin gene by using real-time PCR. Int. J. Food Microbiol. 119:270–276. 104. Mounjouenpou, P., D. Gueule, A. Fontana-Tachon, B. Guyot, P. R. Tondje, and J. P. Guiraud. 2008. Filamentous fungi producing ochratoxin A during cocoa processing in Cameroon. Int. J. Food Microbiol. 130:234–241. 105. Munkvold, G. P. 2003. Epidemiology of Fusarium diseases and their mycotoxins in maize ears. Eur. J. Plant Pathol. 109:705–713. 106. Munkvold, G. P., and A. E. Desjardins. 1997. Fumonisins in maize—can we reduce their occurrence? Plant Dis. 81:556–565. 107. Njapau, H., E. M. Muzungaile, and R. C. Changa. 1998. The effect of village processing techniques on the content of aflatoxins in corn and peanuts in Zambia. J. Sci. Food Agric. 76:450–456. 108. Noonim, P., W. Mahakarnchanakul, K. F. Nielsen, J. C. Frisvad, and R. A. Samson. 2009. Fumonisin B2 production by Aspergillus niger in Thai coffee beans. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 26:94–100. 109. Olsen, M., P. Johnsson, T. Moller, R. Paladino, and M. Lindblad. 2008. Aspergillus nomius, an important aflatoxin producer in Brazil nuts? World Mycotoxin J. 1:123–126. 110. Olsen, M., N. Jonsson, N. Magan, J. Banks, C. Fanelli, A. Rizzo, A. Haikara, A. Dobson, J. Frisvad, S. Holmes, J. Olkku, S. J. Perrsson, and T. Borjesson. 2006. Prevention of ochratoxin A in cereals in Europe, p. 317–342. In A. D.

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

616 Hocking, J. I. Pitt, R. A. Samson, and U. Thrane (ed.), Advances in Food Mycology. Springer, New York, NY. 111. Onifade, A. K., and Y. A. Jeff-Agboola. 2003. Effect of fungal infection on proximate nutrient composition of coconut (Cocos nucifera Linn) fruit. J. Food Agric. Environ. 1:141–142. 112. Osborne, B. G., F. Ibe, G. L. Brown, F. Petagine, K. A. Scudamore, J. N. Banks, M. T. Hetmanski, and C. T. Leonard. 1996. The effects of milling and processing on wheat contaminated with ochratoxin A. Food Addit. Contam. 13:141–153. 113. Osweiler, G. D. 1988. Occurrence and clinical manifestations of trichothecene toxicoses and zearalenone toxicoses, p. 31–42. In J. L. Richard and J. R. Thurston (ed.), Diagnosis of Mycotoxicoses. Martinus Nijhoff, Dordrecht, The Netherlands. 114. Osweiler, G. D. 2000. Mycotoxins—contemporary issues of food animal health and productivity. Vet. Clin. North Am. Food Anim. Pract. 16:511–530. 115. Pacin, A. M., S. L. Resnik, and E. J. Martinez. 2011. Concentrations and exposure estimates of deoxynivalenol in wheat products from Argentina. Food Addit. Contam. B 4:125–131. 116. Paul, P. A., P. E. Lipps, D. E. Hershman, M. P. McMullen, M. A. Draper, and L. V. Madden. 2008. Efficacy of triazole-based fungicides for Fusarium head blight and deoxynivalenol control in wheat: a multi­ variate meta-analysis. Phytopathology 98:999–1011. 117. Payne, G. A., D. K. Cassel, and C. R. Adkins. 1988. Reduction in aflatoxin contamination in corn by irrigation and tillage. Phytopathology 76:679–684. 118. Pei, H. J., J. H. Winde, D. B. Archer, P. S. Dyer, G. Hofmann, P. J. Schaap, G. Turner, R. P. de Vries, R. Albang, K. Albermann, M. R. Andersen, J. D. Bendtsen, J. A. Benen, M. van den Berg, S. Breestraat, M. X. Caddick, R. Contreras, M. Cornell, P. M. Coutinho, E. G. Danchin, A. J. Debets, P. Dekker, P. W. van Dijck, A. van Dijk, L. Dijkhuizen, A. J. Driessen, C. d‘Enfert, S. Geysens, C. Goosen, G. S. Groot, P. W. de Groot, T. Guillemette, B. Henrissat, M. Herweijer, J. P. van den Hombergh, C. A. van den Hondel, R. T. van der Heijden, R. M. van der Kaaij, F. M. Klis, H. J. Kools, C. P. Kubicek, P. A. van Kuyk, J. Lauber, X. Lu, M. J. van der Maarel, R. Meulenberg, H. Menke, M. A. Mortimer, J. Nielsen, S. G. Oliver, M. Olsthoorn, K. Pal, N. N. van Peij, A. F. Ram, U. Rinas, J. A. Roubos, C. M. Sagt, M. Schmoll, J. Sun, D. Ussery, J. Varga, W. Vervecken, P. J. van de Vondervoort, H. Wedler, H. A. Wösten, A.-P. Zeng, A. J. van Ooyen, J. Visser, and H. Stam. 2007. Genome sequencing and analysis of the versatile cell factory Aspergillus niger CBS 513.88. Nat. Biotechnol. 25:221–231. 119. Pietri, A., M. Zanetti, and T. Bertuzzi. 2009. Distri­ bution of aflatoxins and fumonisins in dry-milled maize fractions. Food Addit. Contam. 26:372–380. 120. Pitt, J. I. 1989. Field studies on Aspergillus flavus and aflatoxins in Australian groundnuts, p. 223–235. In Aflatoxin Contamination of Groundnut: Proceedings of the International Workshop, 6–9 October 1987,

SMP_Food Microbiology_CH23.indd

616

ICRISAT Center, India. International Crops Research Institute for the Semi-Arid Tropics, Patancheru, India. 121. Pitt, J. I. 1996. What are mycotoxins? Aust. Mycotoxin Newsl. 7(4):1. 122. Pitt, J. I. 2004. Biocontrol of aflatoxins in peanuts, p. 141–152. In D. Barug, H. van Egmond, R. LopezGarcia, T. van Osenbruggen, and A. Visconti (ed.), Meeting the Mycotoxin Menace. Wageningen Academic Publishers, Wageningen, The Netherlands. 123. Pitt, J. I. 2006. Fungal ecology and the occurrence of mycotoxins, p. 33–41. In H. Njapau, S. Trujillo, H. P. van Egmond, and D. L. Park (ed.), Mycotoxins and Phycotoxins: Advances in Determination, Toxicology and Exposure Management. Wageningen Academic Publishers, Wageningen, The Netherlands. 124. Pitt, J. I., and L. Tomaska. 2001. Are mycotoxins a health hazard in Australia? 1. Aflatoxins and Fusarium toxins. Food Aust. 53:535–539. 125. Pitt, J. I., and L. Tomaska. 2002. Are mycotoxins a health hazard in Australia? 2. Ochratoxin A. Food Aust. 54:39–43. 126. Pitt, J. I., and A. D. Hocking. 2006. Mycotoxins in Australia: biocontrol of aflatoxins in peanuts. Mycopathologia 162:233–243. 127. Pitt, J. I., and A. D. Hocking. 2009. Fungi and Food Spoilage, 3rd ed. Springer, New York, NY. 128. Pitt, J. I., A. D. Hocking, and D. R. Glenn. 1983. An improved medium for the detection of Aspergillus flavus and A. parasiticus. J. Appl. Bacteriol. 54:109–114. 129. Pollastro, S., C. Dongiovanni, R. M. De Miccolis Angelini, A. Abbatecola, P. Natale, M. A. De Guido, and F. Faretra. 2005. Grape rot and contamination of wine by ochratoxin A. Inf. Fitopatol. 55:15–21. 130. Read, M. 1989. Removal of aflatoxin contamination from the Australian groundnut crop. p. 133–140. In Aflatoxin Contamination of Groundnut: Proceedings of the International Workshop, 6–9 October 1987, ICRISAT Center, India. International Crops Research Institute for the Semi-Arid Tropics, Patancheru, India. 131. Reid, L. M., R. W. Nicol, T. Ouellet, M. Savard, J.  D. Miller, J. C. Young, D. W. Stewart, and A. W. Schaafsma. 1999. Interaction of Fusarium gramine­ arum and F. moniliforme in maize ears: disease progress, fungal biomass,­ and mycotoxin accumulation. Phytopathology 89:1028–1037. 132. Rheeder, J. P., W. F. Marasas, P. G. Thiel, G. S. Shephard, and D. J. van Schalkwyk. 1992. Fusarium moniliforme and fumonisins in corn in relation to human esophageal cancer in Transkei. Phytopathology 82:353–357. 133. Robens, J. 2006. Research and regulatory priorities in the USA, p. 95–104. In D. Barug, D. Bhatnagar, H. P. van Egmond, J. W. van der Kamp, W. A. van Osenbruggen, and A. Visconti (ed.), The Mycotoxin Factbook. Wageningen Academic Publishers, Wageningen, The Netherlands. 134. Saleemullah, A. I., I. A. Khalil, and H. Shah. 2006. Aflatoxin contents of stored and artificially inoculated cereals and nuts. Food Chem. 98:699–703.

Manila Typesetting Company

11/08/2012 07:04AM

23.  Mycotoxins

617

135. Sasanya, J. J., C. Hall, and C. Wolf-Hall. 2008. Analysis of deoxynivalenol, masked deoxynivalenol, and Fusarium graminearum pigment in wheat samples, using liquid chromatography-UV-mass spectrometry. J. Food Prot. 71:1205–1213. 136. Schaafsma, A. W., and D. C. Hooker. 2007. Climatic models to predict occurrence of Fusarium toxins in wheat and maize. Int. J. Food Microbiol. 119:116–125. 137. Schaafsma, A. W., L. Tanburic-Ilinic, J. D. Miller, and D. C. Hooker. 2001. Agronomic considerations for reducing deoxynivalenol in wheat grain. Can. J. Plant Pathol. 23:279–285. 138. Scott, P. M., and G. A. Lawrence. 1995. Analysis of beer for fumonisins. J. Food Prot. 58:1379–1382. 139. Shephard, G. S., and V. Sewram. 2004. Determination of the mycotoxin fumonisin B1 by reversed-phase thin layer chromatography: a collaborative study. Food Addit. Contam. 21:498–505. 140. Shephard, G. S., F. Berthiller, J. Dorner, R. Krska, G. A. Lombaert, B. Malone, C. Maragos, M. Sabino, M. Solfrizzo, M. W. Trucksess, H. P. van Egmond, and T. B. Whitaker. 2010. Developments in mycotoxin analysis: an update for 2008–2009. World Mycotoxin J. 3:3–23. 141. Srinivasulu, B., K. Aruna, K. V. Kumar, and D. V. Rao. 2003. Investigations on postharvest aflatoxin contamination of copra. Indian Coconut J. 34:8–9. 142. Steiner, W. E., R. H. Rieker, and R. Battaglia. 1988. Aflatoxin contamination in dried figs: distribution and association with fluorescence. J. Agric. Food Chem. 36:88–91. 143. Steyn, P. S., W. C. A. Gelderblom, G. S. Shephard, and F. R. van Heerden. 2008. Mycotoxins with a special focus on aflatoxins, ochratoxins and fumonisins, p. 3467–3527. In B. Ballantyne, T. Marrs, and T. Syversen (ed.), General and Applied Toxicology, 3rd ed. John Wiley, Chichester, United Kingdom. 144. Stoloff, L., M. Trucksess, P. W. Anderson, and E. F. Glabe. 1978. Determination of the potential for myco­ toxin contamination of pasta products. J. Food Sci. 3:228–230. 145. Strosnider, H., E. Azziz-Baumgartner, M. Banziger, R. V. Bhat, R. Breiman, M. N. Brune, K. DeCock, A. Dilley, J. Groopman, K. Hell, S. H. Henry, D. Jeffers, C. Jolly, P. Jolly, G. N. Kibata, L. Lewis, X.-M. Liu, G. Luber, L. McCoy, P. Mensah, M. Miraglia, A. Misore, H. Njapau, C.-N. Ong, M. T. K. Onsongo, S. W. Page, D. Park, M. Patel. T. Phillips, M. Pineiro, J. Pronczuk, H. S. Rogers, C. Rubin, M. Sabino, A. Schaafma, G. Shephard, J. Stroka, C. Wild, J. T. Williams, and D. Wilson. 2006. Workgroup report: public health strategies for reducing aflatoxin exposure in developing countries. Environ. Health Perspect. 114:1898–1903. 146. Suárez-Quiroz, M., O. González-Rios, M. Barel, B. Guyot, S. Schorr-Galindo, and J. P. Guiraud. 2004. Study of ochratoxin A producing strains in coffee processing. Int. J. Food Sci. Technol. 39:501–507. 147. Sudakin, D. L. 2003. Trichothecenes in the environment: relevance to human health. Toxicol. Lett. 143:97–107.

SMP_Food Microbiology_CH23.indd

617

148. Sun, G., S. Wang, X. Hu, J. Su, T. Huang, J. Yu, L. Tang, W. Gao, and J.-S. Wang. 2007. Fumonisin B1 contamination of home-grown corn in high-risk areas for esophageal and liver cancer in China. Food Addit. Contam. 24:181–185. 149. Taniwaki, M. H. 2006. An update on ochratoxigenic fungi and ochratoxin A in coffee, p. 189–202. In A. D. Hocking, J. I. Pitt, R. A. Samson, and U. Thrane (ed.), Advances in Food Mycology. Springer, New York, NY. 150. Taniwaki, M. H., J. I. Pitt, A. A. Teixeira, and B. T. Iamanaka. 2003. The source of ochratoxin A in Brazilian coffee and its formation in relation to processing methods. Int. J. Food Microbiol. 82:173–179. 151. Taniwaki, M. H., A. D. Hocking, J. I. Pitt, and G. H. Fleet. 2009. Growth and mycotoxin production by food spoilage fungi under high carbon dioxide and low oxygen atmospheres. Int. J. Food Microbiol. 132:100–108. 152. Taniwaki, M. H., A. D. Hocking, J. I. Pitt, and G. H. Fleet. 2010. Growth and mycotoxin production by fungi in atmospheres containing 80% carbon dioxide and 20% oxygen. Int. J. Food Microbiol. 143:218–225. 153. Tolleson, W. H., K. L. Dooley, W. G. Sheldon, J. D. Thurman, T. J. Bucci, and P. C. Howard. 1996. The mycotoxin fumonisin induces apoptosis in cultured human cells and livers and kidneys of rats. Adv. Exp. Med. Biol. 392:237–250. 154. Toth, B., A. Mesterházy, P. Nicholson, P. Téren, and J. Varga. 2004. Mycotoxin production and molecular variability of European and American isolates of Fusarium culmorum. Eur. J. Plant Pathol. 110:587–599. 155. Turner, P. C., A. Sylla, Y. Y. Gong, M. S. Diallo, A. E. Sutcliffe, A. J. Hall, and C. P. Wild. 2005. Reduction in exposure to carcinogenic aflatoxin by post-harvest intervention measures in West Africa: a community-based intervention study. Lancet 365:1950–1956. 156. Van der Fels-Klerx, H. J., M. C. Kandhai, S. Brynestad, M. Dreyer, T. Börgesson, H. M. Martins, M. Uiterwijk, E. Morrison, and C. H. Booij. 2009. Development of a European system for identification of emerging mycotoxins in wheat supply chains. World Mycotoxin J. 2:119–127. 157. van der Westhuizen, L., G. S. Shephard, J. P. Rheeder, H.-M. Burger, W. C. A. Gelderblom, C. P. Wild, and Y. Y. Gong. 2010. Simple intervention method to reduce fumonisin exposure in a subsistence maize-­farming community in South Africa. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 27:1582–1588. 158. van der Westhuizen, L., G. S. Shephard, J. P. Rheeder, H.-M. Burger, W. C. A. Gelderblom, C. P. Wild, and Y. Y. Gong. 2011. Optimising sorting and washing of home-grown maize to reduce fumonisin contamination under laboratory-controlled conditions. Food Control 22:396–400. 159. Varga, J., S. Kocsubé, K. Suri, G. Szigeti, A. Szekeres, M. Varga, B. Tóth, and T. Bartók. 2010. Fumonisin contamination and fumonisin producing black Aspergilli in dried vine fruits of different origin. Int. J. Food Microbiol. 143:143–149.

Manila Typesetting Company

11/08/2012 07:04AM

Nonbacterial Pathogens

618 160. Wallwijk, C., P. Kastelein, I. de Vries, Z. Kerenyi, T. van der Lee, T. Hesselink, J. Koehl, and G. Kema. 2003. Major changes in Fusarium spp. in wheat in the Netherlands. Eur. J. Plant Pathol. 109:743–754. 161. Warfield, C. Y., and D. G. Gilchrist. 1999. The influence of kernel age on fumonisin B1 production in maize by Fusarium moniliforme. Appl. Environ. Microbiol. 65:2853–2856. 162. Wheeler, K. A., A. D. Hocking, and J. I. Pitt. 1988. Water relations of some Aspergillus species isolated from dried fish. Trans. Br. Mycol. Soc. 91:631–638. 163. Whitaker, T. B., J. Springer, P. R. Defize, W. J. DeKoe, and R. D. Coker. 1995. Evaluation of sampling plans used the United States, United Kingdom and the Netherlands to test raw shelled peanuts for aflatoxin. J. AOAC Int. 78:1010–1018. 164. Whitaker, T. B., M. W. Truckess, A. S. Johansson, F. G. Giesbrecht, W. M. Hagler, Jr., and D. T. Bowman. 1998. Variability associated with testing shelled corn for fumonisin. J. AOAC Int. 81:1162–1168. 165. Williams, J. H., T. D. Phillips, P. E. Jolly, J. K. Stiles, C. M. Jolly, and D. Aggarwal. 2004. Human aflatoxicosis in developing countries: a review of toxicology, exposure, potential health consequences, and interventions. Am. J. Clin. Nutr. 80:1106–1122. 166. World Health Organization. 2000. Safety Evaluation of Certain Food Additives and Contaminants. WHO

SMP_Food Microbiology_CH23.indd

618

Food Additive Series 44. World Health Organization, Geneva, Switzerland. 167. World Health Organization. 2001. Safety Evaluation of Certain Food Additives and Contaminants. WHO Food Additive Series 47. World Health Organization, Geneva, Switzerland. 168. World Health Organization. 2008. Safety Evaluation of Certain Food Additives and Contaminants. WHO Food Additives Series 59. World Health Organization, Geneva, Switzerland. 169. Yates, I. E., and D. Sparks. 2008. Fusarium verticilliodes dissemination among maize ears of field-grown plants. Crop Prot. J. 27:606–613. 170. Yates, I. E., N. W. Widstrom, C. W. Bacon, A. Glenn, D. M. Hinton, D. Sparks, and A. J. Jaworski. 2005. Field performance of maize grown from Fusarium verti­ cillioides-inoculated seed. Mycopathologia 159:65–73. 171. Zhang, W. J., P. S. Fan, X. Zhang, C. J. Chen, and M. G. Zhou. 2009. Quantification of Fusarium gramine­ arum in harvested grain by real-time polymerase chain reaction to assess efficacies of fungicides on Fusarium head blight, deoxynivalenol contamination and yield of winter wheat. Phytopathology 99:95–100. 172. Zinedine, A., J. M. Soriano, J. C. Moltó, and J. Manˇes. 2007. Review on the toxicity, occurrence, metabolism, detoxification, regulations and intake of zearalenone: an oestrogenic mycotoxin. Food Chem. Toxicol. 45:1–18.

Manila Typesetting Company

11/08/2012 07:04AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch24

Lee-Ann Jaykus Doris H. D’Souza Christine L. Moe

24

Foodborne Viral Pathogens

Human enteric viruses are responsible for substantial morbidity worldwide. Transmitted predominantly by the fecal-oral route and exclusively in association with human sewage and/or vomitus, these viruses come into contact with humans by a variety of routes, including the consumption of contaminated foods. A functional rather than taxonomic group, the human enteric viruses are represented by many different virus families and genera (Table 24.1) From an epidemiologic perspective, the most significant of these are human noroviruses (NoVs), which are the most common cause of acute gastroenteritis in industrialized countries (81, 177) and are now recognized as one of the leading causes of foodborne disease. For example, in the United States, NoVs are esti­mated to be responsible for about 58% of all domestically acquired foodborne disease of known etiology, constituting more than 5 million estimated cases annually (192). This is confirmed by recent epidemiologic surveillance studies revealing that they were the cause of 54% of all foodborne disease outbreaks of confirmed etiology reported via the U.S. Centers for Disease Control and Prevention (CDC) Electronic Foodborne Outbreak Reporting System (eFORS) in 2006 (34). Furthermore,

NoVs were estimated to be responsible for ~15,000 (range, 8,097 to 23,323) hospitalizations every year, second only to Salmonella enterica as the leading cause of foodborne disease hospitalizations. They also caused an estimated 150 U.S. deaths annually (192). Enteric viruses in general may also be responsible for a large proportion of foodborne disease of unknown etiology, the burden of which is substantial (193, 242). Enteric viruses can be transmitted directly by personto-person contact or indirectly by consumption of contaminated food or water or contact with fomites. The usual source of enteric virus contamination is human fecal matter, which can easily harbor 106 to 1012 genomic RNA copies per gram when shed by infected, symptomatic individuals (123). However, the role of vomitus cannot be overlooked, particularly for NoVs, as virus particles can be liberated during a vomiting episode and this is likely a significant factor contributing to transmissibility (243). Aerosolization of vomitus can result in infection of exposed subjects who inhale and subsequently swallow the aerosolized virus (145); it can also provide a source of virus to contaminate nearby surfaces.

Lee-Ann Jaykus, Department of Food, Bioprocessing and Nutrition Sciences, North Carolina State University, Raleigh, NC 27695-7624. Doris H. D’Souza, Department of Food Science and Technology, University of Tennessee, Knoxville, TN 37996. Christine L. Moe, Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, GA 30322.

619

SMP_Food Microbiology_CH24.indd

619

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

620 Table 24.1  Human enteric virusesa Virus family

Genus (type species)

Disease syndrome(s)

Role of foodborne transmission

Enterovirus (poliovirus)

Human strains frequently cause asymptomatic or mild forms of gastroenteritis, meningitis, encephalitis, myelitis, myocarditis, and conjunctivitis

Credible but not well documented

Hepatovirus (hepatitis A virus)

Causes relatively mild form of acute hepatitis, usually self-limiting; symptoms and severity increase with age

Well documented

Parechovirus (human parechovirus)

Respiratory and gastrointestinal symptoms in young children, with occasional infection of the central nervous system

Credible but not well documented

Kabuvirus (Aichi virus)

Causes gastroenteritis in humans

Recently documented for Aichi virus via consumption of molluscan shellfish

Norovirus (Norwalk virus)

Causes a common form of acute viral gastroenteritis that is usually self-limiting

Well documented (genus forms a phylogenetic clade of five genogroups; genotypes I and II cause disease in humans, genotype II being more prevalent)

Sapovirus (Sapporo virus)

Causes sporadic outbreaks and cases of gastroenteritis; also self-limiting

Documented; epidemiologic importance of foodborne transmission unknown

Astroviridae

Mamastrovirus (human astrovirus)

Human astroviruses cause acute gastroenteritis in children and the immunocompromised

Credible but not well documented

Hepeviridae

Hepevirus (hepatitis E virus)

Causes outbreaks and cases of enterically transmitted acute hepatitis that are usually self-limiting, but pregnant women are at increased risk for mortality

Primarily transmitted by waterborne routes; foodborne transmission credible but not yet documented

Reoviridae

Rotavirus (rotavirus A)

Five species (types A to E); type A causes gastroenteritis and dehydration in infants <24 mos, milder disease in older children, life-threatening diarrhea in the malnourished; type B most often associated with large epidemics in human children and adults in China and causes severe gastroenteritis

Foodborne outbreaks documented but rare

Adenoviridae

Mastadenovirus (human adenovirus C)

Causes acute infantile gastroenteritis; second in prevalence to that caused by rotaviruses

Credible but not documented

Picornaviridae

Caliciviridae

a

Data from reference 64a.

SMP_Food Microbiology_CH24.indd

620

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens Human enteric viruses have properties that are unique from those of bacterial foodborne pathogens. Viruses are usually species specific and tissue tropic, meaning that the human enteric viruses are believed to infect only humans. Since they must resist the enzymatic conditions and extremes of pH encountered in the gastrointestinal tract, enteric viruses are also resistant to a wide range of commonly used food-processing, preservation, and storage treatments. They are also notably persistent in foods and the environment, frequently surviving for days to weeks without substantial loss in infectivity. Although frequently present in low numbers in contaminated foods, their infectious doses are also low (219), meaning that very low levels of contamination may pose a human health risk.

Human Enteric Viruses of Epidemiologic Significance and their Diseases Virus groups that can be transmitted by foodborne routes include the human enteroviruses (poliovirus, coxsackievirus, and echovirus), enteric adenoviruses, hepatitis A and E viruses, parvovirus, rotavirus, NoVs, and sapoviruses, among others. From an epidemiologic perspective, human NoV and hepatitis A virus (HAV) are the two most important. The former is most significant by virtue of the sheer numbers of cases; the latter because it causes a relatively more severe disease. The Norwalk virus was first reported in 1972 by Kapikian and colleagues, who identified a small, roundstructured virus of 27 nm in diameter by electron microscopy when viewing fecal material obtained from a 1968 gastroenteritis outbreak (82). Since that time, many similar so-called small, round-structured viruses have been reported. Virtually all of these viruses are now considered to be members of the Caliciviridae (which means “cuplike” shape) family. Research over the last 2 decades has resulted in significant advances in Caliciviridae taxonomy, and current classifications include five genera: (i) Vesivirus; (ii) Lagovirus; (iii) Norovirus, with Norwalk virus as the prototype strain; (iv) Sapovirus, represented by the Sapporo virus; and (v) a fifth, unclassified genus (13). Members of the Vesivirus and Lagovirus genera infect animals, posing no known human disease risk, whereas the Sapovirus and Norovirus genera are responsible for epidemic gastroenteritis in humans, the latter of which is the most significant (81). The genomes of viruses within the Caliciviridae family consist of a single strand of positive-sense RNA ranging in size from 7.4 to 8.3 kb. For example, the Norwalk

SMP_Food Microbiology_CH24.indd

621

621 virus genome sequence is 7,642 nucleotides (nt) in length, excluding the 3¢ poly(A) tail, and has a base composition of 48% G+C. The genome encodes three open reading frames: ORF1 (nt 146 to 5359) is the largest (~1,700 amino acids) and encodes a nonstructural polyprotein that contains the genes for p48, NTPase, p22, VPg, protease, and RNA polymerase; ORF2 (nt 5346 to 6935) encodes the viral capsid protein (550 amino acids, molecular weight ~56,600); and ORF3 (nt 6938 to 7573) encodes a small basic structural protein of unknown function (57) (Fig. 24.1). The Norovirus genus consists of more than 40 strains further subdivided into five genogroups (GI to GV) based on RNA genome sequence analysis. There is only 51 to 56% nucleic acid sequence similarity when comparing across genogroups. Each genogroup can be further subdivided into several genotypes. Strains within a genotype share 69 to 97% nucleic acid sequence homology; there are 8 distinct genotypes of GI and 17 for GII NoVs (57). Genogroups I and II are most important to human disease, with GII strains being responsible for the vast majority of human disease (27, 157). Over the last decade, the epidemic GII.4 strain and its variants have been of particular interest, constituting the pandemic strain and being responsible for most human NoV outbreaks worldwide (27, 133, 250). Interestingly, GII.4 strains tend to be transmitted more commonly by person-to-person spread rather than by food or waterborne mechanisms (121). First discovered in 1972, HAV is classified as a member of the genus Hepatovirus within the family­ Picornaviridae (99). The virus is a naked, round particle with a diameter of 27 to 32 nm. Its genome is a linear, single-stranded, 7.5-kb positive-sense RNA molecule that is enclosed in an icosahedral capsid that consists of three major proteins designated VP1, VP2, and VP3. The genome can be subdivided into three regions: a long 5¢ terminal untranslated region (5¢UTR) of about 735 nt (nt 1 to 735), a large ORF encoding a polyprotein of 2,227 amino acids (nt 741 to 7415), and a short 3¢UTR (nt 7416 to 7478) with a poly(A) tail (Fig. 24.2). The single ORF is divided into three regions designated P1, P2, and P3. P1 codes for the capsid proteins and is co- and posttranslationally cleaved into four smaller structural proteins (VP1, VP2, VP3, and VP4) and nonstructural proteins (2A, 2B, 2C, 3A, 3B, 3C, and 3D) by a virus-encoded protease (99). The approximate genome locations of these various proteins have been defined (54). Human isolates of HAV comprise a single serotype, and monoclonal antibodies raised to different isolates fail to distinguish them from one another. Nonetheless,

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

622

Figure 24.1  Norwalk virus genome organization (adapted from Kapikian et al. [114]). VPg, genome-linked virion protein. ORF1 encodes polyprotein posttranslationally cleaved into nonstructural proteins helicase (with a predicted nucleotide triphosphate-binding domain [NTPase]), proteinase (Pro), and RNA-dependent RNA polymerase (Pol). ORF2 encodes capsid protein, which is translated into major and minor structural proteins which consist of shell and protruding domains. N, NH2-terminal arm. ORF3 encodes a basic protein of unknown function. doi:10.1128/9781555818463.ch24f1

there is substantial sequence heterogeneity within select genome regions that encode the putative VP1-2A junction of HAV. Using this region as the basis for typing, HAV isolates can be differentiated into seven unique genotypes such that a genotype is defined as a group of viruses that differ from each other in sequence homology by no more than 15%. Of these seven genotypes (designated I to VII), only genotypes I, II, III, and VII are associated with human disease, whereas genotypes IV, V, and VI have been isolated from simians. Genotypes I and III are the predominant human genotypes and are further divided into subtypes A and B, whereas genotypes II and VII have only one human isolate each (6). There are other human enteric viruses that can be transmitted by contaminated foods, although their epidemiologic significance is not well understood. The human enteroviruses, for example, are smooth, round, nonenveloped particles 27 nm in diameter with singlestranded, positive-sense RNA. These viruses are members of the Picornaviridae family (poliovirus is the prototype) and historically have been transmitted through the consumption of contaminated water and unpasteurized milk (211). Although the disease can occasionally be problema­ tic in developing countries, outbreaks of foodborne poliomyelitis usually do not occur in developed countries due to effective vaccination. Other human enteroviruses

SMP_Food Microbiology_CH24.indd

622

such as the coxsackie- and echoviruses have also been associated occasionally with foodborne disease (42). The symptoms of enteroviral infection are diverse and virus specific but may range from gastrointestinal, respiratory, neurological, and skin manifestations. Rotaviruses usually are transmitted by either waterborne or person-to-person routes, but occasionally foodborne outbreaks occur (26). This virus group is the leading cause of infantile diarrhea worldwide (139) and is responsible for significant morbidity and more than 500,000 deaths per year; the vast majority of these deaths occur in developing countries (34). Rotaviruses are 70 to 75 nm in diameter and appear roughly spherical in electron micrographs. They consist of 11 segmented, double-stranded RNA molecules encased in a double-layered protein coat. The 5¢ and 3¢ ends of the double-stranded RNA are highly conserved, with sizes ranging from 667 bp (segment 11) to 3,302 bp (segment 1) and totaling 6,120 kDa, or 18,555 bp (52). All rotavirus genes are monocistronic except for segments 9 and 11. There are six structural viral proteins (termed VP1, VP2, VP3, VP4, VP6, and VP7) and five nonstructural proteins (termed NSP1 through NSP5). Based on VP6 reactivity with monoclonal antibodies, there are at least seven different rotavirus groups (A through G). Second-generation rotavirus vaccines are

Manila Typesetting Company

11/08/2012 07:06AM

Figure 24.2  HAV genome organization (adapted from Hollinger and Emerson [99]). 5´UTR, 624 to 1,199 nt long; VPg, genome-linked virion protein, 22 to 24 amino acids. The single ORF encodes polyprotein posttranslationally cleaved into VP1, VP2, and VP3, which form the capsid; VP4 is the inner surface capsid protein. 2A, unknown function; 2B, RNA synthesis and cell membrane permeability; 2C, RNA replication; 3A and 3B, RNA replication proteins, cofactor for 3D; 3C, viral proteinases; 3D, RNA-dependent RNA polymerase. Poly(A) tail, about 35 to 100 nt long; 3´UTR, 47 to 125 nt long. doi:10.1128/9781555818463.ch24f2

24.  Foodborne Viral Pathogens

SMP_Food Microbiology_CH24.indd

623

623 now recommended for global use by the World Health Organization, although the risk of intussusception is still elevated, at around 1 in every 60,000 vaccinated infants (176). Other significant human enteric viruses include the parvoviruses, astroviruses, and hepatitis E virus. Parvoviruses are perhaps the smallest of the enteric viruses, with diameters of 20 to 26 nm (42). They are ­single-stranded DNA viruses with a smooth protein coat. Astroviruses are small, round-structured viruses that are star shaped and have single-stranded, positivesense RNA that is 6.8 to 7.2 kb in length. These viruses cause a diarrheal disease with an incubation period of 1 to 3 days and symptoms lasting for 1 to 4 days. Hepatitis E virus is a small (30 nm) virus with singlestranded, positive-sense RNA and three ORFs. Recently classified in the genus Hepevirus, hepatitis E virus appears to have a single serotype and at least four major genotypes (109). Hepatitis E virus causes a disease similar to that caused by HAV but with severe manifestations in pregnant women. It is much more prevalent in developing countries, although there is recent evidence of its emergence in the developed world (182). Hepatitis E virus is transmitted predominantly through sewagecontaminated water and person-to-person contact (42), although concern for foodborne transmission has recently been expressed (218).

Foodborne Transmission of Human Enteric Viruses From a foodborne disease standpoint, three types of commodities are commonly associated with viral disease outbreaks, namely, (i) molluscan shellfish contaminated by feces-impacted growing waters; (ii) fresh produce items contaminated by human feces during production or packing, usually through workers’ hands or contact with contaminated water; and (iii) ready-to-eat (RTE) and prepared foods contaminated by infected food handlers as a result of poor personal hygiene (Fig. 24.3). Contamination of marine waters with human sewage is the critical factor with respect to virus contamination of molluscan shellfish (e.g., mussels, clams, cockles, and oysters). Sources of contamination in shellfish harvest waters include illegal dumping of human waste, failing septic systems along shorelines, sewage treatment plants overloaded with storm water, and discharges of treated and untreated municipal wastewater and sludge. Molluscan shellfish are of particular concern because they are filter feeders, meaning that they filter and even concentrate microorganisms in their gut during the process of feeding. This is further complicated by the fact

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

624

Figure 24.3  Transmission routes of foodborne viruses. doi:10.1128/9781555818463.ch24f3

that viruses tend to be environmentally persistent and the animals are usually consumed either raw or only lightly cooked. Furthermore, there is no significant relationship between the presence of virus contamination and the levels of coliform and fecal coliform indicator bacteria, which are used in most countries to classify molluscan shellfish harvesting waters. This means that there is no reliable means of screening by which to prevent virally contaminated shellfish from reaching the marketplace (79, 120).

SMP_Food Microbiology_CH24.indd

624

At the time of this writing, there are many unanswered questions with respect to how fresh produce items become contaminated with viruses. Certainly, produce may become contaminated when grown in fields irrigated with wastewater or fertilized and conditioned with improperly decontaminated sewage effluent. In fact, common sewage sludge treatments such as drying, pasteurization, anaerobic digestion, and composting reduce but do not necessarily eliminate enteric viruses. Viruses can survive in contaminated soil for long pe-

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens riods of time depending upon factors such as growing season, soil composition, temperature, sunlight, moisture, rainfall, resident microflora, and virus type (238). Despite the credibility of preharvest contamination, the role of infected farm workers who pick, wash, and pack produce cannot be overlooked, especially since simple produce items often associated with outbreaks (e.g., strawberries, raspberries, and green onions) are most often harvested by hand. While washing of produce may reduce contamination loads, it will not eliminate enteric viruses if they are present (120). Without question, poor personal hygienic practices of infected food handlers are the most important contributor to the spread of viral foodborne disease. The U.S. CDC estimates that about 50% of all NoV foodborne outbreaks are linked to ill food service workers (242). There is strong evidence suggesting that contaminated hands frequently play a role in enteric virus spread, acting as either virus donors or recipients. Hands usually become contaminated by direct contact with any viruscontaining fluid from oneself or others; they may also become contaminated by contact with virus-contaminated surfaces or objects. The extent of such contamination will vary depending on a variety of factors, including the virus load, the degree of discharge from the host, the hand-washing habits of the infected person, and the efficiency with which the virus is transferred and persists. Studies of Norwalk virus persistence have revealed that viral RNA can be detected on the hands of volunteers for at least 2 hours after deliberate contamination (138). Food handlers may transmit viruses to foods from contaminated hands, a contaminated surface, or between food items. Technically, any food handled by a virus carrier, symptomatic or asymptomatic, and then consumed without a subsequent heating step can be a potential vehicle for transmission of enteric viruses. Common food items falling in this category include hand-sliced deli meats and cheeses; sandwiches; meat, vegetable, and fruit salads; and various desserts.

Epidemiology of Foodborne Viral Disease Although many different types of enteric viruses can contaminate bivalve mollusks, only a few (HAV, NoVs, and astroviruses) have been epidemiologically linked to shellfish-associated viral disease, particularly in oysters (183). The first documented epidemiologic linkage between human NoVs and shellfish-associated gastroenteritis occurred in the mid-1970s, and many more outbreaks have been reported since then. For example,­

SMP_Food Microbiology_CH24.indd

625

625 three large outbreaks associated with improper discharge of untreated human waste material occurred in Louisiana in the 1990s alone (16, 119). Between 2003 and 2004, three distinct human NoV outbreaks in Australia were associated with the consumption of imported oyster meat whose harvest was traced back to a single Japanese estuary system (237). In 2010, 334 NoV cases associated with shellfish consumption were reported in five European countries (239). Interestingly, a 1988 outbreak of HAV in Shanghai, China, which was linked to the consumption of shellfish harvested from a feces-impacted site, ranks among the largest foodborne disease outbreaks ever reported, with a total of about 300,000 cases (85). Enteric virus outbreaks have been associated with fresh produce as well. When considering “simple” food products (i.e., those consisting of a single ingredient), berries and green onions have been the most common culprits. For example, between 2005 and 2009, berryassociated NoV outbreaks were reported in France (44), Denmark (64), Sweden (97), and Finland (146). In most cases, these products were imported. A high-profile multistate HAV outbreak occurred in the United States in 1997. More than 200 cases of illness were reported, of which >90% occurred among school employees and students. Case-control studies revealed food items containing frozen strawberries as the likely vehicles of infection. The implicated strawberries were grown in Mexico, processed and frozen at a California facility, and distributed through the U.S. Department of Agriculture school lunch programs. Although regulatory inspection was unable to identify a definitive source of contamination, three of the Mexican growing sites had evidence of inadequate hygiene and toilet facilities (102). An interesting group of HAV outbreaks occurred in 2003 in the United States. The largest of these occurred in patrons of a single Pennsylvania restaurant over a 3-day period, with a total of 601 confirmed illnesses, 3 deaths, and at least 124 hospitalizations. A case-­control study implicated foods containing green onions, particularly mild salsa. Although some of the restaurant workers became symptomatic, all became ill at the same time as the patrons, ruling them out as the source of the outbreak (240). A series of smaller outbreaks occurred among restaurant patrons in three other states in the months preceding this one, also implicating green onions (5). In both cases, the origin of the green onions could be traced back to northern Mexico. Comparative analysis of the HAV genome sequences associated with these 2003 outbreaks revealed substantial sequence similarity to endemic Mexican HAV strains, suggesting that

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

626 the green onions were likely contaminated before arrival at the restaurants (38, 240). Although the ultimate source of viral contamination (e.g., water or harvester) in the case of the green onion and berry outbreaks was never identified, it should be noted that these two products are almost always harvested by hand and hence subject to substantial human handling. The potential importance of human hands is also supported in a recent study by León-Félix et al. (129), who determined that between one-quarter and one-third of northern Mexican field workers’ hands had evidence of human NoV contamination, with 30 to 45% of the green peppers they harvested having contamination as well. There is a clear need to better characterize the source of viral contamination in fresh produce items. The contamination of RTE and prepared foods most frequently results from poor hand-washing practices of infected food handlers after toilet use, as fecal material can be left on hands or even under nails and then can come in contact with food products. Consequently, handling cooked products with bare hands has been identified as a major factor for pathogen transfer to RTE foods. Many hepatitis A outbreaks occur as the result of a single infected food handler at a single food establishment, and have been reported in association with lettuce, salads, sandwiches, hamburgers, spaghetti, and bakery products (41). For example, in 1992, up to 5,000 people may have been exposed to HAV following the consumption of a variety of gourmet foods prepared by an infected food handler in Denver, Colorado (50). Human NoVs have been associated with many outbreaks caused by poor hygiene by food handlers, including outbreaks involving the consumption of contaminated salads and sandwiches, delicatessen meats, and bakery products (53). Perhaps not unexpectedly, these viruses are the leading cause of foodborne disease outbreaks in school settings (231). In recent outbreak investigations, scientists have combined molecular and immunological methods to support epidemiologic evidence indicating a common food source for NoV transmission. These methods have enabled investigators to link multiple clusters of illnesses as well as identify the contaminated food items. Several of these recent outbreaks are particularly interesting. For example, Friedman et al. (73) described a NoV outbreak among 2,700 guests attending 46 different weddings on a single weekend. All of the wedding cakes were baked at the same bakery, where there were at least two ill employees. The same NoV strain was detected in stool specimens from two wedding guests, one wedding hall employee, and one ill bakery employee.

SMP_Food Microbiology_CH24.indd

626

Using molecular epidemiologic evidence, Malek et al. (143) were able to link 137 cases of NoV occurring on 13 Colorado River rafting trips to one batch of delicatessen meat purchased from a single processing plant. The meat had been sliced by an employee who did not wear gloves and had just returned to work 1 day after a gastroenteritis episode. This latter outbreak was a wake-up call to U.S. Department of Agriculture Food Safety and Inspection Service-inspected facilities, which frequently do not consider enteric viruses as hazards in their hazard analysis and critical control point plans. Secondary transmission after a primary foodborne outbreak of NoV must also be considered. In a particularly interesting case, a GI NoV that initially infected athletes on a single college football team who consumed sandwiches made by an infected food handler was later passed on to the rival team after contact during a football game in which the first team’s members were clearly ill (14). Patterson et al. (178) reported on the role of a kitchen assistant who vomited into a sink used to prepare a potato salad that was subsequently identified as the vehicle in a NoV outbreak. Isakbaeva et al. (104) investigated a NoV gastroenteritis outbreak that affected six consecutive cruises despite extensive sanitation measures. Using genetic sequence analysis, the investigators documented strain persistence with likely foodborne, environmental, and person-to-person transmission (37). In general, foodborne NoV outbreaks have higher primary attack rates than outbreaks in other settings (156), but as illustrated above, a single contamination event can affect many more people that those simply consuming the contaminated food.

Disease, Pathogenesis, and Treatment

NoVs

The disease caused by NoVs is self-limiting and characterized by nausea, vomiting, diarrhea, and abdominal pain, with occasional headache and low-grade fever (81). Data from outbreak investigations and human challenge studies indicate that the incubation period ranges from 12 to 51 hours, with a mean of 24 hours, and illness lasts no more than 48 to 72 hours (8). In a community-based cohort study of NoV infection in The Netherlands, Rockx et al. (184) reported a median duration of symptoms of 5 days and found that diarrhea was more prevalent in NoV-infected children of <1 year of age, whereas infected children of ³5 years of age were more likely to vomit. Some outbreak investigations have revealed less frequent vomiting in older individuals (33). Severe illness or hospitalization is uncommon except

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens in children, the elderly, and the immunocompromised, for whom rehydration therapy may be necessary. Some deaths in elderly patients occur (230). Outbreaks associated with NoV are characterized by vomiting in >50% of cases, an average duration of symptoms of 12 to 72 hours, a high attack rate, and stool specimens that are negative for bacterial pathogens. Average attack rates are usually 30 to 45%, but in some outbreaks up to 80% (33, 120, 157, 224). Asymptomatic NoV infections have been documented in outbreaks (74), human challenge studies (8, 135), and community- and hospital-based studies of sporadic NoV infection (4, 249). Virus can be shed in the stool both before symptoms occur and for up to 56 days after the onset of infection (8, 76, 184). Atmar et al. (8) reported that median peak virus concentration in stool specimens from volunteers infected with Norwalk virus was approximately 1011 genomic copies per gram of feces. As stated above, virus can also be shed in high numbers in vomitus (243). Host susceptibility to infection appears to be due both to genetic determinants and acquired immunity. Human challenge studies with Norwalk virus, a GI human NoV, have revealed that some volunteers remained uninfected even after exposure to high doses (108, 135). Further investigation revealed that these subjects lacked the H type 1 antigen, a histo-blood group antigen present in saliva and also on the surface of epithelial cells that likely serves as a receptor or coreceptor for Norwalk virus binding. In vitro assays provide additional evidence that recombinant Norwalk virus capsid protein binds to H type 1 antigen (89); specifically, the P domain of the capsid protein is now recognized as the exact binding domain (56, 57, 217). Synthesis of the H type 1 antigen depends on the a -1,2-fucosyltransferase enzyme that is encoded by the FUT2 gene and determines the “secretor” status of an individual. Individuals who are homozygous recessive for the FUT2 gene are secretor negative and appear to have innate genetic resistance to Norwalk virus infection. About 20% of Europeans are secretor negative, and this proportion is higher in Asian and African ethnic groups. In addition to secretor status, blood type also has a role in susceptibility to Norwalk virus infection, as blood group O individuals in Norwalk virus human challenge studies were more susceptible to Norwalk virus infection than group A or B individuals (103, 135). In contrast to Norwalk virus, a human challenge study with Snow Mountain virus, a GII.2 NoV, demonstrated no relationship between infection with the virus and blood group secretor status (134).

SMP_Food Microbiology_CH24.indd

627

627 There have been mixed reports about whether there is a relationship between infections with other NoV strains and histo-blood group antigens in outbreaks. Hennessy et al. (93), in their investigation of an outbreak in a military field hospital in Afghanistan, reported that individuals with blood group B were less susceptible to symptomatic NoV infection than people of other blood types. Tan et al. (216) reported that secretor status and histo-blood group antigen were associated with risk of infection in two GII outbreaks in China. Positive secretor status was strongly associated with risk of symptomatic infection in both outbreaks (P = 0.0007), and in the GII.4 outbreak, individuals with blood type A had a significantly increased risk of infection, whereas type O individuals had a decreased infection risk. In contrast, Halperin et al. (86) did not find a relationship between infection with GII NoV and ABO histo-blood group in their investigation of two outbreaks in military units in Israel. The GII.4 NoVs appear to bind to all histo-blood group antigens (not just type 1) (101) and, hence, are likely to have a greater pool of susceptible individuals in the population. This characteristic may explain the greater frequency of GII.4 outbreaks compared with other NoV strains. The relationship between NoV infection and histo-blood group antigens has been recently reviewed (198, 215). Both reviews conclude that recognition and binding to carbohydrate receptors is an important feature of NoV infection and may also play a role in virus evolution. Immunity to NoV has been recently reviewed by Leon et al. (128). There is some evidence of acquired immunity, although the mechanisms for this have not been elucidated. Some human challenge studies have revealed that volunteers who became ill after a primary NoV challenge did not become ill when rechallenged with the same virus 6 to 14 weeks later, but did become ill when rechallenged with same virus 24 to 42 months after the primary challenge (174, 245). This suggests the development of short-term protective immunity but perhaps not long-term immunity. The high attack rates observed in NoV outbreaks also suggest that most of the population do not have protective immunity against NoV infection, either because the protective immunity is so short-lived or because there are many NoV strains and protective immunity is not cross-reactive. The role of serum antibodies in NoV infection is also not well understood. Sera collected in human challenge studies and NoV outbreak investigations indicate that NoV-specific humoral antibody titers (immunoglobulin G [IgG], IgM, and IgA) increase in response to infection. Serum IgG produced in response to infection with one NoV strain often cross-reacts with antigens from other

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

628 NoV strains, and common epitopes within and between genogroups have been identified (173). IgA and IgM responses appear to be more type specific than that of IgG. However, it is not clear that serum antibodies provide any protection against subsequent NoV infection, and these antibodies may only be markers of susceptibility to NoV infection. In vitro experiments suggest that serum IgG produced in response to Norwalk virus infection is capable of blocking Norwalk virus attachment to synthetic H type 1 antigen (89). However, volunteers with anti-Norwalk virus IgG in their prechallenge sera were more likely to become infected after Norwalk virus challenge than those who did not have anti-Norwalk virus IgG (135). There are few studies of cellular immune response to NoV infection (128, 132, 134, 247). Overall, these studies have revealed that peripheral blood mononuclear cells from subjects in NoV challenge studies respond to NoV antigen in vitro and secrete anti-inflammatory and inflammatory cytokines, but gamma interferon secretion responses differ by individual and are probably confounded by the effect of individual prechallenge exposure history. One recent Norwalk virus human challenge study revealed that 62% of secretor-positive volunteers became infected, suggesting that some mechanism exists to protect genetically susceptible individuals from Norwalk virus infection. Examination of Norwalk virus-specific salivary IgA levels revealed that the secretor-positive volunteers who did not become infected mounted an early (before 5 days postchallenge) Norwalk virus-specific salivary IgA response to the virus. In contrast, infected secretor-positive volunteers did not mount a Norwalk virus-specific salivary IgA response until after 5 days postchallenge (135). The role of the mucosal response in protective immunity to Norwalk virus infection is not understood. It is possible that this early, specific salivary IgA response may represent a memory response that enables rapid production of neutralizing antibodies that may block virus binding to cell receptors. Why some secretor-positive volunteers developed this salivary IgA response and whether it provides long-term or short-term immunity are not known. However, this observation of a protective response does suggest that induction of a mucosal immune response should be considered as an important part of a vaccine strategy for NoV infection. Vaccine development efforts are under way and have been reviewed by Harrington et al. (88), Vinjé (234), and Herbst-Kralovetz et al. (92). Researchers are exploring various vaccination strategies, including the use of recombinant NoV capsid protein viruslike particles (VLPs) made in insect cells and plants to stimulate immunity in oral and nasal vaccination strategies (92, 188);

SMP_Food Microbiology_CH24.indd

628

Venezuelan equine encephalitis virus replicon-based vaccines (88); and NoV P particles (214). Phase 1 clinical trials in humans have found that Norwalk virus VLPs are safe and immunogenic, stimulating the production of systemic and mucosal anti-Norwalk virus antibodies and antibody-secreting cells (12, 63, 212). Because NoV infection outbreaks continue to occur in countries with high standards of sanitation and hygiene, vaccination of high-risk subgroups (such as food handlers) may be the only effective way to control epidemic NoV infection. However, at this stage there are many questions that need to be addressed about whether the current vaccine approaches stimulate a protective immune response, whether it is possible to develop long-term immunity, and how to measure immune correlates of protection. Furthermore, these viruses seem to be evolving rapidly, and there are many strains. The number of antigens that would be needed for an effective vaccine and the degree of cross-protection provided against other strains are not known. The development of a chimpanzee model for NoV infection was recently reported and used to test GI (Norwalk virus) and GII (MD145 ) VLP vaccines (20). This landmark study revealed that the Norwalk virus VLP vaccine induced protective homologous immunity in chimpanzees that were challenged with Norwalk virus 18 months later, but the chimpanzees that received the MD145 VLP vaccine or placebo were not protected from Norwalk virus infection. Patients who suffer from NoV illness have broadened and blunted villi of the proximal small intestine with infiltration of mononuclear cells and cytoplasmic vacuolization (114). Biopsy specimens collected during human challenge studies indicate that histopathologic changes appear within 24 hours of virus challenge and usually persist for up to 2 weeks (2). Histopathologic testing revealed that the intestinal mucosa appear intact and viruses are not detected in the epithelial cells of the mucosa by electron microscopy. Also, transient maladsorption of fat, d-xylose, and lactose was observed, along with decreased levels of small intestine brush border enzymes (trehalase and alkaline phosphatase). However, adenylate cyclase activities of the jejunum were not elevated. Reduced gastric motility is likely responsible for the nausea and vomiting caused by NoV. Detection of NoV RNA in the sera of children with gastroenteritis has been reported (213). NoV infection is self-limiting, and if necessary, oral fluid and electrolyte replacement therapy is usually adequate for replenishing fluid loss. The symptoms of the disease can also be reduced by the oral administration of bismuth subsalicylate (114). In cases of severe vomiting and diarrhea, parenteral administration of fluids

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens may be necessary. Infrequently, hospitalization for severe dehydration is necessary; and death in elderly patients has been documented (90). Repeated diarrheal episodes may promote intestinal mucosal damage, resulting in eventual malnutrition (114). The development of antiviral drugs that block NoV binding to host cells may be valuable as a prophylactic approach in settings where NoV infection outbreaks are common. For instance, Chang and George (40) have reported that interferon and ribavirin inhibit Norwalk virus replication in replicon-bearing cells and may have therapeutic value.

HAV

The incubation period for HAV infection is 15 to 50 days, with an average of 28 days. The virus is more readily transmitted during the latter half of the incubation period, meaning that food workers in retail settings with acute HAV infection can readily contaminate RTE products if they do not practice adequate personal hygiene. Symptomatic hepatitis A infection generally presents in four phases. The first phase is characterized by viral replication but a lack of symptoms. In the second, so-called preicteric, or prodromal, phase, which usually lasts 5 to 7 days, patients may experience anorexia, nausea, vomiting, alterations in taste, arthralgias, malaise, fatigue, urticaria, and pruritus. When seen by a health care provider during this phase, patients are often diagnosed as having gastroenteritis or a generalized viral syndrome. The third, or icteric, phase presents as darkening of the urine, pale-colored stools, jaundice, and right-upper-quadrant pain with hepatomegaly (99). Fecal shedding and viremia are maximal at the onset of the icteric phase, and subside during its 7- to 28-day duration (206). During the last, or convalescent, phase, symptoms resolve and liver enzymes return to normal. Although icteric disease occurs in fewer than 10% of children younger than 6 years of age, it occurs in 40 to 50% of older children and in 70 to 80% of adults. In most cases infection is mild and self-limiting; however, in older patients more severe disease is possible. Complications such as acute liver failure, cholestatic hepatitis, and relapsing hepatitis occur but are rare. A variety of extrahepatic manifestations occasionally occur in patients with acute hepatitis A, including hemolysis, acalculous cholecystitis, acute renal failure, pleural or pericardial effusion, acute reactive arthritis, and pancreatitis; on occasion, neurological syndromes have also been reported (49). The overall mortality rate for HAV infection is approximately 0.01%. Immunity after infection is complete and considered to be lifelong in duration (206). Hepatitis A infection begins with ingestion of the virus, which reaches the gastrointestinal tract in the bile and replicates in the hepatocyte. Approximately

SMP_Food Microbiology_CH24.indd

629

629 104 virions/ml of blood can be detected in the viremic phase of the disease (100). Studies are under way to elucidate the nature of the host cell receptor that determines the tissue tropism of HAV to the hepatocyte. The virus may also circulate in the blood enclosed in lipid-associated membrane fragments, enabling it to be protected from neutralizing antibodies. Possible “enterohepatic” cycling of HAV may occur via ingestion of infected material, absorption from the stomach or small intestine, replication in the liver, secretion into bile, and excretion in stool or reabsorption (49). Enlargement of Kupffer cells is the first evidence of infection by HAV. For example, in experimental studies with owl monkeys, viral antigens were first detected in Kupffer cells of the liver at 14 days and in hepatocytes at 21 days after oral administration of the HM-175 strain of HAV (7). Cardinal pathologic features of acute hepatitis include the presence of hepatocellular degeneration, characterized as either ballooning or acidophilic (apoptotic) change, together with varying degrees of portal and lobular inflammation and hepatocyte regeneration (49). Interestingly, although HAV targets liver cells, it usually does not kill those cells (48). Rather, immunemediated lysis, particularly involving natural killer and cytotoxic T cells, is recognized as the most probable cause of hepatic inflammation (100). In addition, human gamma interferon produced by HAV-specific T cells may participate in pathogenesis and probably promotes clearance of HAV-infected hepatocytes, as increased levels of interferon have been detected in the serum of infected patients (142, 229). Biochemical changes associated with active HAV infection include bilirubin excretion in the urine and elevated alkaline phosphatase, alanine aminotransferase, and aspartate aminotransferase/serum glutamic-oxaloacetic transaminase levels, all of which persist until recovery from illness (99). The immune response against HAV infection occurs on two levels, humoral and cellular. In the humoral response, IgM, IgG, and IgA antibodies are directed against conformational epitopes, mostly the surface proteins VP1 and VP3 and precursor protein VP0 of the HAV particle. The IgA and IgM responses precede the IgG response, but the latter is long-lived and provides immunity (49). Because immune-mediated injury causes hepatic inflammation, cytotoxic T cells also play a significant role in the immune response. Since detection of virus particles, viral antigen, or RNA is complicated, the method of choice for diagnosis of infection is immunoassay, which detects the presence of IgM anti-HAV antibody in the blood (206).

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

630 Historically, administration of Igs at doses as low as 0.01 to 0.04 ml/kg of body weight were used for HAV postexposure prophylaxis. These can be effective in controlling both the incidence and severity of disease as long as they are administered within 2 weeks of exposure (179). The prophylactic administration of Ig is recommended for patrons and food handlers provided all of the following criteria exist: (i) the infected worker was responsible for handling RTE foods and was not wearing gloves; (ii) the infected worker demonstrated poor hygienic practices or already had diarrheal symptoms; and (iii) the patrons can be identified and treated within 2 weeks of exposure (43). As Ig provides only short-term protection against HAV infection, its administration is not recommended for preexposure prophylaxis (68). For that sort of protection, two FDA-licensed formalin-killed whole-virus vaccines are available, generally administered as a single primary immunization followed by a booster dose after 6 to 12 months (126). The efficacy of the vaccine is 94 to 100%, and protection should last for about 20 years (68). Routine vaccination of all food handlers and restaurant employees is usually cost prohibitive, even during a hepatitis A epidemic (152). However, HAV vaccination can and perhaps should be offered to high-risk populations such as health care workers in infectious diseases and pediatrics, medical staff in laboratories handling stool samples, and staff in sewage treatment plants (98). Recently, HAV vaccination has been recommended for children up to 2 years of age (36).

Barriers to the Study of Human Enteric Viruses Most naturally occurring human enteric viruses cannot be cultivated in vitro, nor are there relevant animal models to facilitate their study. This has been the single most important barrier to our ability to understand these important agents of foodborne disease. Human NoVs are the best example of this phenomenon. For instance, the only way to work with these viruses in their natural form is to obtain fecal material from infected individuals. Needless to say, there are only limited supplies of this material, usually obtained from federal, state, or local health departments or from the few investigators engaged in human challenge studies. In the early days following their identification, NoVs could be detected only by electron microscopic methods, which lack sensitivity. Antibodies were later raised against some of these viruses, but it quickly became apparent that NoVs are antigenically diverse, with antibodies raised to one

SMP_Food Microbiology_CH24.indd

630

strain having minimal cross-reactivity with other strains. Likewise, these viruses are genetically diverse. With the advent of the molecular biology revolution, the first human NoV genome was cloned and sequenced (246) and investigators began making VLPs, or intact virus capsids lacking only the viral RNA (83). These have become important reagents for binding studies and anti­ body production. Molecular techniques also provided an opportunity to detect NoVs and other enteric viruses using nucleic acid amplification methods, most notably reverse transcription-PCR (RT-PCR). While a vast improvement over electron microscopy, RT-PCR is not perfect: it relies on the availability of broadly reactive primers and probes; it is prone to matrix-associated inhibition; and interpretation of positive results must be approached cautiously as detection of viral RNA does not necessarily correlate with the presence of infectious virus. In the absence of a cultivable human NoV strain, surrogate viruses are often used as proxies. In the early years, bacteriophages were commonly used surrogates, but these viruses have significant structural and functional differences compared with mammalian viruses. Over the last 15 years, two mammalian cultivable viruses have emerged as the most widely used surrogates for human NoVs, i.e., feline calicivirus (FCV), which can be readily propagated in feline kidney cells (59); and murine norovirus (MNV), which is slightly more difficult to propagate but shares more likeness to human NoVs with respect to biochemical and genetic features (244). The availability of these surrogates has enabled investigators to extrapolate the behavior of human NoV strains with respect to persistence, resistance, and inactivation kinetics. However, there is growing evidence that the behavior of the cultivable surrogates does not always mimic that of human strains. For example, FCV is more sensitive to extremes of pH than is MNV or human NoVs; MNV is more sensitive to ethanol and to desiccation (32, 190). This means that surrogate data must be interpreted with caution. Ideally, comparative studies using both human NoVs and cultivable surrogates would provide a more comprehensive picture of the behavior of these viruses, but only a few such studies have been reported to date (172).

Persistence and Resistance of Enteric Viruses It is generally recognized that human enteric viruses are environmentally persistent, much more so than vegetative bacteria that cause foodborne illness. Most enteric virus persistence studies were conducted before the year 2000 using viruses for which mammalian cell culture hosts were

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens available. For example, the vaccine strain of poliovirus is readily cultivatable, and cell culture-adapted strains exist for HAV and rotavirus. Many factors influence the stability of the enteric viruses in the environment, the most important of which are relative humidity, temperature, degree of inoculum drying, type of suspending medium (fecal or otherwise), virus type, and the type of surface contaminated. These early studies are reviewed elsewhere (21, 170). There is compelling evidence that virtually all enteric viruses remain infectious for hours to days after they have been deposited on environmental surfaces. Food and fecal matrices provide additional stability, prolonging persistence from days to weeks (46). Of more recent interest has been the environmental stability of human NoVs (21). Epidemiologic evidence from prolonged outbreaks occurring on cruise ships and in hotels and hospitals has long indicated that these viruses are very stable in the environment (37, 104). In fact, it has become relatively routine to process environmental swabs for detection of viral RNA by nucleic acid amplification when evaluating infection control strategies undertaken after NoV outbreaks in institutional settings (136). In these cases, viral RNA can be detected for weeks after removal of the source of contamination. Recent laboratory-based studies have sought to further characterize NoV persistence, in most cases using the surrogates as models. For example, Doultree et al. (59) reported that FCV persists on glass surfaces for more than 60 days at 4°C, with complete inactivation after 14 to 28 days at 25°C and after 1 to 10 days at 37°C. Liu et al. (138) determined that two human NoV strains persisted for 42 days when inoculated onto stainless steel, ceramic, and Formica coupons; in general, there was a <1-log10 decrease in viral genome copy number over that 42-day experimental period. Bae and Schwab (10) determined long-term persistence (generally <0.2-log10 PFU/day) of several NoV surrogates in inoculated surface water and groundwater. Kingsley and Richards (116) determined that in oysters allowed to accumulate HAV by exposure to contaminated water, infectious virus could be detected for 3 weeks. Persistence on fingerpads has also been documented (138). Enteric viruses are also readily transferred from one surface to another, with the efficiency of transfer dependent upon factors such as pressure, friction, time, temperature, and the presence of moisture (105, 170). At its greatest, investigators have observed the transfer of as much as 10 to 50% of the initial virus inoculum in a single tactile action, although transfer is likely to be somewhat less efficient in natural situations (18, 105). Further, the degree of virus transfer is quite variable, even when evaluated under highly controlled condi-

SMP_Food Microbiology_CH24.indd

631

631 tions. The combined effect of environmental persistence and ease of transfer makes for a potentially dangerous situation if viruses are present in food production, processing, or preparation environments in which humans are handling food. Human enteric viruses tend to be quite resistant to most of the intrinsic and extrinsic parameters commonly used by food processors, including manipulation of temperature (freezing and refrigeration), water activity, pH, gaseous environment, natural and intentionally added inhibitors, and the presence of competitive microflora. Enteric viruses cannot grow in foods; hence, manipulation of a gaseous environment or competitive microflora is of little value. In general, there is a direct correlation between maintenance of virus infectivity and reduced temperature, with freezing temperatures being particularly protective (105). For example, HAV survived freezing for up to 2 years in a multistate outbreak associated with consumption of frozen strawberries (163). Enteric viruses are also resistant to extremes of pH, which makes sense because they must survive gastric acidity in order to reach target sites in the small intestine. It has been determined that cell culture-adapted HAV can remain infectious after being held at pH 1 for 5 hours and that human enteroviruses retain stability at pH 3 (168). Hewitt and Greening (95) determined that HAV remained infectious in acidic marinade (pH ~3.75) when held at 4°C for 4 weeks. Epidemiologic evidence has revealed that NoVs are similarly resistant to low pH. For example, Norwalk virus caused infection in human volunteers after incubation for 3 hours at pH 2.7 (55). Another example of the resistance of human NoVs to prolonged acidic pH was revealed by a large gastroenteritis outbreak in which orange juice (pH ~3.5) was implicated (69). By contrast, the commonly used NoV surrogate FCV appears to be quite unstable at lower pH values (32, 62), raising questions regarding its universal suitability as a NoV surrogate. Heat is the most commonly used food processing and preparation technique, and the mechanism of virus inactivation by heat is denaturation of the virus capsid (62). Cannon et al. (32), using a classic thermal inactivation experimental design (capillary tubes), determined the inactivation kinetics of MNV and FCV. At 56, 63, and 72°C, they observed D values of 3.5, 0.4, and 0.2 minutes, respectively, for MNV and 6.7, 0.4, and 0.1 minutes, respectively, for FCV. These results revealed that standard pasteurization conditions (for milk) should result in inactivation of these viruses. Similar data have been obtained by others (11, 94). Because of the absence of an in vitro cultivation technique, it is difficult to

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

632 conduct­ similar experiments for human NoVs. However, recent studies by Topping et al. (223) have revealed that human NoVs are more resistant to heat than are their cultivable surrogate counterparts. The suspension assay design provides the best means to characterize the degree of viral inactivation attributable to commonly used disinfectants under ideal conditions. The efficacy of standard disinfectants depends on concentration, contact time, product formulation, and surrogate virus. There are not many suspension assay results for HAV or poliovirus. Results are inconclusive for the human NoV surrogates. For example, some studies have shown efficacy for ethanol (77, 141), whereas others have not (59, 164, 172, 241). The same is true for chlorine; however, relatively high concentrations (exceeding 500 ppm) are needed to obtain complete inactivation of the surrogate viruses (59, 62, 180). Quaternary ammonium compounds are not effective (59, 164, 241). Likewise, triclosan (a phenolic) and chlorhexidine are relatively ineffective (172).

Control of Viruses in Foods and the Food Production, Processing, and Preparation Environment

Molluscan Shellfish

The role of molluscan shellfish in the transmission of human enteric viral disease has been recognized for decades, with accompanying efforts to develop effective pre- and postharvest controls for this food commodity. Preharvest controls continue to include prevention of human sewage pollution of shellfish harvesting waters and the establishment of alternative microbiological indicators. Prevention of sewage pollution can be difficult because of the diversity of sources and unpredictable nature of a contamination event; however, preventing illegal dumping of human waste in marine waters is likely to be a particularly important control (16). Oyster beds close to sewage plants may need special consideration because of concerns about the potential for sewage overflow, particularly during times of excessive rainfall (39). Preventing human sewage contamination of molluscan shellfish overlay waters is complicated by the lack of a reliable microbiological indicator whose presence and/or levels are correlated with the presence of enteric virus contamination (106). Bacteriophages, especially F-specific coliphages (FRNA phages or male-specific coliphages), somatic coliphages, and phages of Bacteroides fragilis, have long been proposed as alternative indicators of viral contamination of the environment. Of these, the distribution and survival of FRNA phages are considered to be more similar to those of human enteric viruses (58, 199), although

SMP_Food Microbiology_CH24.indd

632

shellfish may contain pathogenic viruses even though they lack FRNA phages (3, 71). Similar difficulties in establishing relationships between the presence and levels of somatic coliphages and those infecting B. fragilis have been reported (162). The feasibility of using human adenoviruses as indicators of human enteric viruses in environmental and shellfish samples has likewise also been proposed (72), but recent evidence shows poor correlations (17). Hence, the identification of a reliable microbiological indicator for enteric virus contamination in shellfish and their harvesting waters has yet to be established. Controlled purification, widely used in Europe, is an early postharvest control strategy. Depuration and relaying are based on the principle that shellfish can purge at least some of their microbial contaminants by extended feeding in clean water, provided that the feeding conditions (e.g., temperature, salinity, and dissolved oxygen) are favorable (183). In depuration, freshly harvested shellfish are placed for several days in a controlled environment, usually in tanks provided with a supply of clean, disinfected seawater under specific operating conditions. In relaying, shellfish are transferred from contaminated growing areas to approved or naturally unpolluted (pristine) areas (201). Unfortunately, the scientific consensus is that enteric viruses are not eliminated during controlled purification. For example, Schwab et al. (197) found only a 7% reduction in human NoV concentration after 48 hours of depuration treatment. Other postharvest controls for virus contamination of shellfish include heating and alternative food-processing technologies. Early reports documented outbreaks of both HAV and viral gastroenteritis linked to the consumption of cooked shellfish (151, 160). Subsequent thermal inactivation studies revealed that when artificially contaminated cockles were immersed for 1 minute in water at 85, 90 or 95°C or were steamed for 1 minute, only partial reduction in HAV titers was achieved (154). Similarly, Croci et al. (47) determined that heat treatments at 100°C for 2 minutes were necessary to completely inactivate HAV in contaminated mussels. Sow et al. (202), who studied the inactivation of HAV and MNV in spiked soft-shell clam meat, determined that heating at 90°C for 90 seconds reduced MNV titers by 3.3 log10 and HAV titers by 2.7 log10, whereas at 90°C for 180 seconds both MNV-1 and HAV were completely inactivated (defined as a reduction of ³5.5 log10). Data such as these have resulted in European heat processing recommendations (internal temperature of 90°C for 1.5 min) to control HAV in shellfish; this should also eliminate the human NoV surrogate FCV (59, 200). Unfortunately, such high temperatures often result in unacceptable sensory changes to the product.

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens In general, enteric viruses are quite resistant to ionizing radiation, and studies have determined this to be an ineffective control for molluscan shellfish, at least at the relatively low doses necessary to maintain organoleptic quality of the product (87, 144). There has, however, been much interest in the efficacy of high-pressure processing (HPP) as a postharvest treatment to inactivate human enteric viruses in shellfish, particularly because it works well for inactivating pathogenic Vibrio species under conditions that result in a product that maintains its “freshlike” properties. In early work, Kingsley et al. (117) determined that HAV and FCV suspended in tissue culture medium were inactivated after exposure to 450 MPa for 5 minutes and 275 MPa for 5 minutes, respectively. In later studies, researchers further characterized the degree of HAV inactivation in artificially contaminated shellfish, finding infectivity reductions of >1, >2, and >3 log10 after treatment at 350, 375, and 400 MPa, respectively, for 1 minute at temperatures ranging from ca. 8.7 to 10°C (30). Kingsley et al. (115) determined that MNV suspended in tissue culture medium could be inactivated by 6.9 log10 after a 5-minute treatment at 450 MPa at room temperature. In oyster tissue, a 4.1-log10 inactivation of MNV was observed after 5 minutes of treatment at 400 MPa at 5°C. Despite these promising preliminary results, Leon et al. (127) determined that when human subjects consumed Norwalk virus-contaminated oysters with and without prior HPP treatment, infection occurred among one or more volunteers consuming HPP-treated products, except those products receiving the most severe treatment (600 MPa). In addition, this treatment resulted in a product having a mildly cooked, whitish appearance. Hence, more research is needed before it can be determined whether HPP is a viable postprocess technology for inactivation of enteric viruses in molluscan shellfish.

Fresh Produce

Preharvest measures directed at prevention of viral contamination of fresh produce are addressed by the good agricultural practices recommendations provided by federal agencies and industry trade groups (227, 228). One important consideration in these documents is ensuring the absence of human fecal contamination in irrigation water and waters used to mix pesticides and fungicides, or in manure, fertilizer, and/or compost. Unfortunately, sewage spills, storm-related contamination of surface waters, illicit discharge of waste, and residential septic system failures are widely recognized as the leading sources of surface water and groundwater contamination, and these are relatively common and frequently unavoidable occurrences (210). Since many produce

SMP_Food Microbiology_CH24.indd

633

633 items are subject to extensive human handling during harvesting, preharvest food safety strategies must also focus on individuals harvesting the food, particularly if the product is hand-picked. This might include issues such as ready access to clean, on-site toilet and handwashing facilities; restriction of children and ill workers in production fields; and adequate employee education (227, 228). From a postharvest perspective, it is generally recognized that current preservation and storage processes such cooling, freezing, acidification (pH ³4.5), and moderate heat treatments (pasteurization) are usually inadequate to completely inactivate human NoVs if they are present on or in a food matrix (159). Most of the decontamination efforts applied to fresh produce have focused on washing. Certainly, the water used for washing and ice used for transport and cooling must be free of human fecal contamination. For produce washing, waters are frequently supplemented with sanitizers, the most common of which is chlorine, although others such as chlorine dioxide and organic acids have been used. Generally, washing fresh produce can reduce but not eliminate foodborne pathogens. In some instances, simple home-use remedies such as water immersion, hand rubbing, or use of detergents or vinegar (10%) can be moderately effective, providing approximately 1-log10 reductions in virus concentration on the surface of berries (140). Alternatives to chlorine include hydrogen peroxide (140), peroxyacetic acid-hydrogen peroxide formulations (84), and trisodium phosphate (140, 208). Use of ozonated wash water also has promise (96, 131). Particularly promising, Predmore and Li (181) recently determined that supplementation of chlorinated (200 ppm) wash water with surfactants (sodium dodecyl sulfate, polysorbates) enhanced the efficiency of removal of MNV from fresh produce by approximately 100-fold. Since some of these compounds are approved as food additives and/or GRAS (generally recognized as safe), this may be a promising control strategy for the future. Although novel surface sanitizers continue to be evaluated, the data on chemical disinfection for the inactivation of viruses from food surfaces remain less than desired. Not only is efficacy relatively poor, but some of the novel agents (such as hydrogen peroxide) cause unacceptable organoleptic changes to the product. Product-specific surface morphology and physiological characteristics make disinfection complicated; leafy vegetables can be more difficult to decontaminate because of their rough or wrinkled surfaces, and small fruits like raspberries and blackberries have more porous and complex surfaces that can entrap virus particles (183). Furthermore,

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

634 disinfectants incorporated in wash water may not be effective in removing or inactivating viruses that have penetrated through the skin of the product or those that might have entered tissues through cuts and abrasions. There is substantial interest in several emerging food­processing technologies for inactivating viruses on the surface of fresh produce. For example, FCV and poliovirus are highly susceptible to inactivation by UV radiation, with 3-log10 reductions obtained when treating contaminated water with doses of 23 and 40 mJ/cm2, respectively (78, 220). Fino and Kniel (67) determined that when green onions, lettuce, and strawberries were artificially contaminated with HAV, Aichi virus, or FCV and subsequently exposed to UV light at doses of £240 mW∙s/cm2, they obtained about a 4.6-log10 reduction for lettuce; a reduction of 2.5 to 5.6 log10 for green onions; and a reduction of 1.9 to 2.6 log10 for strawberries. Su et al. (209) studied the effect of high-intensity ultrasound on the survival of MNV and FCV in buffer and orange juice. Using a treatment of 20 kHz for 2, 5, 10, 15, 20, and 30 minutes with pulses of 30 seconds on/30 seconds off, a 15-minute treatment was required for complete inactivation (reduction of 4 log10) of FCV, but only a 1.6-log10 reduction in MNV was achieved, even after a 30-minute treatment. Based on preliminary data, novel processing technologies have promise, but much more research is needed to validate their efficacy and major investment is needed to commercialize the most promising ones. As is the case for molluscan shellfish, ionizing radiation is not a promising technology for inactivating viruses in produce, as high doses are relatively ineffective and likely to adversely affect product quality (65). A combination of two or more controls, so-called hurdle approaches, may be necessary to completely eliminate viral contamination of fresh produce. A few studies have investigated such combinations. For example, Li et al. (130) determined that a minimal reduction in the titer of MNV (<1 log10) inoculated on shredded iceberg lettuce occurred with a treatment of liquid (2.1%) or vaporized (2.5%) hydrogen peroxide; however, the combination of UV light and vaporized hydrogen peroxide substantially increased the efficacy of inactivation. Butot et al. (28) determined the effect of freeze-drying, freezedrying combined with heating, and steam blanching on inactivating HAV and FCV on the surface of berries and herbs. While freeze-drying itself was only minimally effective, the addition of a terminal dry heat treatment at 120°C improved the degree of inactivation.

RTE and Prepared Foods

Adherence to strict hygienic practices when handling and preparing foods is critical to control viral contamination in RTE food products. These issues are reviewed

SMP_Food Microbiology_CH24.indd

634

extensively by others (153, 248). The first line of defense in this regard is employee education and regular supervision of employees. The importance of controlling viral contamination is further illustrated by several modifications to the 2005 Food Code (which have been retained in the 2009 revision), including (i)  exclusion and restriction of ill employees, (ii) emphasis on hand-washing procedures, and (iii) prohibition of bare-hand contact with RTE foods (225, 226). Because ill employees shed viruses in large numbers, managers of food manufacturing, catering, and food service industries should restrict these individuals from working directly with food or food equipment and should provide a sick leave policy that allows workers to stay home while ill. Even with such controls, virus shedding by individuals is still a concern because it can occur pre- or postsymptomatically as well as asymptomatically (8, 165, 166). Glove use (and changing of gloves if damaged or soiled), frequent hand washing, and prevention of cross-contamination during handling and preparation of foods are essential. One of the most important considerations in hand hygiene is the role of fingernails, as virus-containing fecal material can be harbored here and long, artificial fingernails are a risk factor for NoV outbreaks (155). Hand disinfection remains an important, although somewhat frustrating, issue. Reliable hand decontamination relies on three important factors, i.e., (i) an effective disinfecting agent, (ii) adequate use instructions, and (iii) regular compliance (191). Compliance with recommended hand-washing regimens is notoriously poor (153). Additionally, although hand washing helps prevent and control the transfer of viruses, hand-washing agents differ in their ability to inactivate viruses (15, 169). Consistent with the results of suspension assays described previously, fingerpad assays confirm that commonly used hand-washing agents have minimal efficacy against human enteric viruses, rarely exceeding 1- to 2-log10 inactivation when used at manufacturer-recommended concentrations (18, 113, 122, 137, 149). Some have suggested that many of the commercial hand sanitizers are not even as effective as soap and water alone (18, 19). There is increasing consensus for the need to establish new standards for the selection of effective formulations for hand-washing agents with regard to antiviral activities, particularly with respect to labeling (191). Recently, Macinga et al. (141) reported on a new ethanol-based hand sanitizer containing a synergistic blend of polyquaternium polymer and organic acid that reduced the infectivity of FCV and MNV by >3 log10 after a 30-second exposure, as evaluated by both suspension and fingerpad assays.

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens Frequently handled objects, such as taps, doorknobs, and even food preparation surfaces, can also serve as sources of virus contamination in the food preparation environment, and standard surface disinfection is used as a control measure (21). Not unexpectedly, the efficacy against enteric viruses of most chemical disinfectants commonly used in both institutional and domestic environments is questionable (169). Early research revealed that sodium hypochlorite was most effective for HAV and rotavirus inactivation, and its efficacy improved when used at high concentrations and for a relatively long contact time (1, 107, 150). Virtually the same results have been observed in most of the surface decontamination studies involving human NoV surrogates (80, 107, 241). In particular, bleach used at concentrations ranging from 1,000 to as high as 5,000 ppm appears to be the most effective surface-sanitizing regimen relative to other disinfectants tested (59, 171, 189). However, the specific log10 inactivation achieved using sodium hypochlorite varies with virus type, surface type, and study design (59, 189). In addition, the presence of fecal material not only increases virus survival but is likely to protect viruses from the antimicrobial activity of bleach (171). Based on such data, the CDC recommends a minimum of 1,000 ppm for disinfecting surfaces contaminated with human NoVs (35).

Detection methods Despite extensive efforts to develop robust methods to detect viral contamination in foods, this remains a challenging area of research (60, 61). Because the numbers of virus particles present in contaminated foods are usually low and there is no universal or rapid culturebased method for growing NoVs or HAV, cultural enrichment is not an option. Hence, the viruses must be concentrated and purified from the food matrix before applying detection methodology, with molecular amplification being the detection method of choice. Sequencebased determination of amplicon identity is used for further confirmation. The major steps for the detection of viruses in foods can be designated as follows: (i) virus concentration and purification, (ii) nucleic acid extraction, (iii) detection of amplicons, and (iv) confirmation of amplicon identity.

Virus Concentration and Purification

Virus concentration and purification schemes are designed to reduce sample volume and remove at least some of the matrix while simultaneously recovering most of the contaminating viruses. In order to achieve these goals, sample manipulations are undertaken that

SMP_Food Microbiology_CH24.indd

635

635 capitalize on the behavior of enteric viruses to act as proteins in solutions, to cosediment by simple centrifugation when adsorbed to larger particles, and to remain infectious at extremes of pH or in the presence of organic solvents. Many virus concentration and purification methods use manipulation of pH and/or ionic conditions to favor virus adsorption to, or elution from, the food matrix. This is then followed by relatively lowspeed centrifugation, after which the virus-containing phase (either precipitate or eluate) is recovered for further purification. Other steps in the virus concentration and purification process may include various forms of filtration (crude filtration and ultrafiltration), ultracentrifugation, precipitation (achieved through the addition of polyethylene glycol or organic flocculants or by manipulation of pH), organic solvent extraction (to remove matrix-associated lipids), ligand-bound magnetic separation (using immunobeads or cationic particles), and/or enzyme pretreatment (to break down matrix-associated organic matter, particularly complex carbohydrates). In almost all instances, virus extraction is accomplished by combining two or more of these steps in series (105). It is generally recognized that elution of viruses from the surface of foods is relatively inefficient, mostly because of underlying hydrophobic, ionic, and van der Waals interactions that tend to favor adherence of viruses to organic matter. In addition, “other” forces may influence virus binding to solids, including virus association with sugar or carbohydrate moieties on food tissues, as has been observed for shellfish and fresh produce (75, 222). Recent efforts to capitalize on these interactive forces have brought about some new methods to facilitate virus concentration and purification. For example, Papafragkou et al. (169) used cationically charged beads to facilitate HAV capture from a variety of food matrices, achieving recoveries of about 50% of input virus and detection limits of ca. 10 PFU per 25-g sample. Cannon and Vinjé (31) used the specificity of histo-blood group antigens as potential binding ligands for human NoVs, revealing that a magnetic bead-based histo-blood group antigen assay could be developed for the recovery of low numbers (30 to 200 genomic units) of human NoVs from seeded environmental waters. This approach has since been applied and expanded upon by others who have used recirculating magnetic capture equipment to increase the sample size/volume (161, 221). A good virus concentration and purification method (Fig. 24.4) should be relatively simple and result in sample volume reductions of 10- to 1,000-fold and recovery of >90% of the target virus. However, recovery efficiency is both virus and matrix specific (66). For example, HAV recovery tends to be low in comparison to the

Manila Typesetting Company

11/08/2012 07:06AM

636

SMP_Food Microbiology_CH24.indd 636 Manila Typesetting Company

Nonbacterial Pathogens

11/08/2012 07:06AM

Figure 24.4  Candidate methods for the concentration and detection of viruses in foods. doi:10.1128/9781555818463.ch24f4

24.  Foodborne Viral Pathogens relatively high recovery of human enteroviruses (e.g., poliovirus); simple sample matrices such as lettuce tend to be easier to work with than more complex matrices such as sandwiches and RTE salads. All methods have their own advantages and disadvantages. For example, immunocapture methods tend to be simple, requiring fewer sample manipulations with a lower likelihood of virus loss and favoring the recovery of infectious virus; however, they may be hindered by high specificity since broadly reactive antisera are not available for human NoVs. Presently, there is no universal extraction method that can be applied to all foods, particularly when contamination levels are low (105).

Nucleic Acid Extraction

Although viral RNA can be released from capsids and made available for amplification using a simple heat treatment (99°C for 5 minutes), this is rarely done because, even with the best virus concentration method, residual matrix­associated components persist and oftentimes interfere with nucleic acid amplification. Therefore, an efficient nucleic acid extraction step is critical. The science of RNA extraction has developed rapidly over the last decade, and what was once a complex procedure is now much simpler and more reliable (105). Over the years, there have been many comparative studies to identify the “best” RNA extraction method(s) for use with food systems. Many investigators are now using combination guanidinium–silica-based methods, which have recently been automated by a number of diagnostics manufacturers (203).

Detection of Amplicons

RT-PCR it the most commonly used method for the detection of viral RNA. The sensitivity and specificity of RT-PCR depends upon the efficiency of the virus concentration and purification steps, as well as the nucleic acid extraction. It also depends on primer choice and amplification conditions. The use of primers with low degeneracy and high melting temperature is recommended because this reduces the likelihood of nonspecific amplification, which is common when nucleic acids are extracted directly from a complex sample matrix. For HAV detection, broadly reactive primers with high annealing temperatures are available, usually corresponding to the VP1-2A junction or 5¢UTR of the viral genome (110). The high degree of genetic diversity for the NoVs has made the development of broadly reactive primers difficult. Four “regions” of the NoV genome have been used for primer design (designated regions A, B, C, and D) (148, 235), but the ORF1ORF2 junction (just downstream of region B) is the most conserved and is frequently used for genogroup-

SMP_Food Microbiology_CH24.indd

637

637 specific detection (111, 112). For strain comparison (as might be appropriate in outbreak investigations), primers corresponding to the NoV capsid region (region D) can be used, although these sometimes have a higher degree of degeneracy, which may affect efficiency of amplification (148). It is recommended that multiple primer pairs be used when screening food samples for NoV contamination (25), as one pair may perform better than another, depending on virus load and the matrix from which the sample was derived. Historically, nested reactions have been used to increase the sensitivity of detection, particularly for naturally contaminated food products (25). During the last decade, real-time or quantitative RTPCR (qRT-PCR) methods have replaced the traditional method. The term “real time” refers to the simultaneous detection and confirmation of amplicon identity as the amplification progresses. It is said that real-time assays can be made quantitative, but residual matrix­­associated inhibitors frequently affect the reliable quantification of viral load in naturally contaminated samples. Therefore, simple presence-absence testing, rather than quantification, is generally used for virus detection in foods and environmental samples. Further, there is some debate as to whether an RNA or DNA standard is best when designing the standard curve for real-time quantification. While using a DNA standard provides a defined assessment of the efficiency of DNA amplification, it does not take into account an inefficient reverse transcription step, which may result in an underestimate of the number of viral genome copies in the sample (105).

Confirmation of Amplicon Identity

When RT-PCR products are obtained from naturally contaminated foods, the amplicons must be sequenced for further confirmation and strain typing. qRT-PCR is not very amenable to the sequencing that needs to be done for confirmation, so presumptively positive samples are almost always rescreened with traditional RT-PCR prior to cloning and sequencing of amplicons. Sequences derived from foods can be compared to those obtained from clinical specimens in an effort to make causal associations. This is a cumbersome process, and although the approach has been used successfully to link contaminated foods to outbreaks of viral gastroenteritis, more often than not it is impossible to achieve amplification of viral RNA from implicated foods (186). Several ways to facilitate confirmation are under development. For example, a recent study has reported the development of an RT-PCR assay that allows the user to detect and genotype human NoV strains in a single

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

638 step (195). Efforts are under way to develop DNA micro­ arrays for the simultaneous detection and genotyping of NoVs (167) and even for the identification of a broad range of foodborne viruses, including HAV, NoVs, rotavirus, and select human enteroviruses (9). These methods will undoubtedly be improved upon over time.

Recent Advances in Detection and Diagnostics

Initial efforts in the development of methods to detect viral contamination in foods were limited almost exclusively to molluscan shellfish. By the mid-2000s, relatively standard methods were developed for detecting HAV and human NoVs in shellfish (111), with many ensuing studies done to characterize the prevalence of contamination, particularly in European shellfish (45). Since then, there has been an explosion of interest in developing methods for other foods. For example, there are now well-accepted methods for the detection of viruses in berries (29, 203, 204), and the European Committee for Standardization is in the process of establishing standard methods for the detection of NoVs and HAV in selected foods and bottled water (124, 194). There has also been increased scrutiny relative to the need for inclusion of appropriate controls when attempting to detect viruses in naturally contaminated products. Not only are the standard negative and positive amplification controls needed, but there is now a strong recommendation from the international community for the inclusion of sample process (extraction) controls and internal amplification controls (IACs). Extraction controls are intended to evaluate the efficiency of the virus concentration and purification steps; as such, the analyst adds a relatively high concentration (106 to 108 infectious units) of a harmless virus with physical characteristics similar to enteric viruses. Viruses used as process controls have included Mengo virus, MNV, FCV, and various bacteriophages. The concept is that if the process control is recovered and detected by qRT-PCR, this indicates that the sample preparation method was successful; if not, it calls into question the efficiency of that method (124, 203). IACs are sequences of nontarget RNA or DNA, usually flanked by the target-specific primer binding sites. These are added to the amplification reactions to account for reaction failure due to residual matrix­associated amplification inhibition (91, 124). Another important area of research has focused on differentiating infectious and noninfectious viruses. Several candidate methods have been identified, to be used in conjunction with RT-PCR, to aid in this discrimination. These methods fall into two general categories, i.e., those estimat-

SMP_Food Microbiology_CH24.indd

638

ing capsid integrity and those targeting genome integrity. Candidates in the former category are more popular and include the use of nucleic acid intercalating agents (175), RNase pretreatment (164, 223), ligand/antibody capture (31, 196), and direct detection of damaged protein (187). The most widely used method to date is RNase pretreatment. While these methods have been reviewed (185), a comprehensive, side-by-side comparison of techniques has not yet been undertaken.

Applications of Virus Detection Methods to Naturally Contaminated Samples

Regardless of their drawbacks, there has been an increase in the use of virus detection methods as applied to naturally contaminated samples (118). For example, in their 2007 U.S. market survey, DePaola et al. (51) determined (as evaluated by qRT-PCR) a prevalence of virus contamination (NoVs or HAV) in U.S. oysters ranging from 5.2 to 11.3%, depending upon the region from which they were harvested. Mattison et al. (147) confirmed the presence of human NoV RNA in 16/275 (6%) marketpackaged leafy greens collected in 2009 from Canadian commercial sources. Detection of NoVs in environmental swabs and hand samples has been used for NoV outbreak investigations of restaurants and cruise ships (23, 24). Boxman et al. (25) successfully detected NoV RNA in foods associated with two outbreaks, although only with the aid of nested RT-PCR. Recently, there has been detection of environmental contamination by NoVs even in the absence of outbreaks (22). Despite the tremendous strides that have been made in virus detection, these protocols are still underutilized and usually applied only in response to known or suspected foodborne disease outbreaks. The primary reasons for their limited use include (i) the inability of molecular amplification methods to discriminate between infectious and inactivated virus; (ii) the lack of widely accepted, collaboratively tested methods, as available methods have been used only to a limited extent (125, 236); (iii) the requirement that most methods be product specific, meaning that universal approaches for detecting a variety of viruses do not exist; and (iv) the cost and need for highly trained personnel. In addition, care should be taken when interpreting both negative and positive test results. For example, while a sequenceconfirmed positive test assures the presence of the viral genome in the product, it does not necessarily confirm that the contaminant was infectious. Negative test results are fraught with interpretive challenges related to sampling, the potential for virus inactivation (particularly if an extended time occurs between sampling and testing), low levels of virus contamination, and the impact of

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens the matrix on assay detection limits. Hence, a negative test result cannot be relied upon to ensure the absence of virus contamination (186). An additional emerging issue is how to interpret marginally positive qRT-PCR results, especially when amplicon concentrations are too low to be used in sequencing reactions (205).

Risk Assessment Interest in applying the principles of risk analysis to foodborne viruses has increased in the past 5 years. The Food and Agriculture Organization and the World Health Organization conducted an expert consultation in 2007 to address this topic (70). Overall, the panel believed that there were so many unanswered questions that embarking on a formal risk assessment would be premature. However, the document produced from their meeting served as the basis for a new work charge for the Codex Alimentarius Committee on Food Hygiene. In the meantime, the academic sector has sought to apply, on a preliminary basis, mathematical modeling to the development of criteria for enteric pathogens in irrigation waters (207) and the transmission of human NoVs in the food-handling environment (158). A particularly interesting study seeks to use human NoV genotype profiles for selection and differentiation of foodborne outbreaks. The first phase of this work focused on developing a predictive model that could serve prospectively in the selection of outbreaks most likely to be associated with foodborne transmission. Using prospective epidemiologic surveillance data from the European Union, the investigators developed a logistic regression model that was adapted to a Web-based tool. The odds of an outbreak being foodborne were highly influenced by venue (home versus institution versus restaurant) and genotype (most foodborne outbreaks are associated with GI strains) (233). A follow-up project was initiated to determine whether NoV genotype frequency distributions could be used to enhance detection of the sources of foodborne outbreaks (232). Both studies revealed how epidemiologic data might eventually be used to support better source attribution of foodborne virus illnesses and perhaps a move toward eventual risk-based decision making for management of foodborne viral diseases.

Conclusions Epidemiologic evidence reveals that foods play an important role in the transmission of human enteric viruses. This can in large part be attributed to improvements in the ability to detect viral pathogens in clinical specimens, mostly related to recent advances in molecular detection

SMP_Food Microbiology_CH24.indd

639

639 techniques. However, because both HAV and NoVs can be transmitted by a variety of nonfoodborne routes, the relative importance of foods as a vehicle of transmission is poorly understood. Unfortunately, widespread epidemiologic surveillance for viral gastroenteritis is complicated by the fact that the disease, while miserable, is usually short-lived and recovery is complete, so many cases are not reported to or detected by health care professionals. In addition, tests for the detection of NoVs in stool specimens are not routinely applied. At the current time, sporadic cases of viral gastroenteritis usually are unrecognized and only widespread outbreaks are investigated. When outbreaks are investigated, rarely is the implicated food tested for viral contamination, and when it is, detection limits minimize the usefulness of the results. Improved and more widespread reporting and investigation of foodborne viral disease outbreaks, as well as targeted epidemiologic studies to identify the risk factors for acquiring viral gastroenteritis, would improve our understanding of source attribution. The implementation of programs such as CaliciNet may help to provide this information. Even if better detection of viral disease were implemented, the availability of effective controls remains a stumbling block. For molluscan shellfish, the absence of a reliable indicator of viral contamination of harvesting waters, along with the weaknesses in efficacy of conventional control strategies (depuration, relaying, and mild heat), suggests that occasional viral contamination of shellfish will continue to occur. Fresh produce can become contaminated with viruses through a variety of routes, but our understanding of how, why, and how often such contamination events occur is quite limited. Water contamination and subsequent produce contamination is likely to occur sporadically, and implementing effective measures to improve personal hygiene at the farm level is challenging. The lack of truly effective antiviral disinfectant agents for both hands and surfaces is a further complication. Multiple control strategies will likely be needed, requiring significant resources. Perhaps the best place to start is with education. As more people understand why and how enteric viruses are spread, then day-to-day changes such as routine hand washing and hand and surface sanitation will become more of a way of life and lead to better control of foodborne virus transmission. Our understanding of the importance of enteric viruses to the overall burden of foodborne disease has improved dramatically over the past 20 years. However, many unanswered questions remain and much research is still needed. The field has been limited by the small number of scientists working with foodborne viruses,

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

640 as well as the absence of effective in vitro culture methods, which limits studies on infectious virus strains. It does appear that this is changing as more food microbiologists begin to tackle this fascinating and challenging subject. It is likely that the next decade will bring exciting new developments as more is done to understand and control these important causes of human foodborne disease.

References 1. Abad, F. X., R. M. Pintó, and A. Bosch. 1997. Disinfection of human enteric viruses on fomites. FEMS Microbiol. Lett. 156:107–111. 2. Agus, S. G., R. Dolin, R. G. Wyatt, A. J. Tousimis, and R. S. Northup. 1973. Acute infectious nonbacterial gastroenteritis: intestinal histopathology. Histologic and enzymatic alterations during illness produced by the Norwalk agent in man. Ann. Intern. Med. 79:18–25. 3. Allwood, P. B., Y. S. Malik, C. W. Hedberg, and S. M. Goyal. 2003. Survival of F-specific RNA coliphage, feline calicivirus, and Escherichia coli in water: a comparative study. Appl. Environ. Microbiol. 69:5707–5710. 4. Amar, C. F., J. Gray. M. Iturriza-Gomar, E. A. Maclure, and J. McLauchlin. 2007. Detection by PCR of eight groups of enteric pathogens in 4,627 faecal samples: re-examination of the English case-control Infectious Intestinal Disease Study (1993–1996). Eur. J. Clin. Microbiol. Infect. Dis. 26:311–323. 5. Amon, J. J., R. Devasia, G. Xia, O. V. Nainan, S. Hall, B. Lawson, J. S. Wolthuis, P. D. M. MacDonald, C. W. Shepard, I. T. Williams, G. L. Armstrong, J. A. Gabel, P. Erwin, L. Sheeler, W. Kuhnert, P. Patel, G. Vaughan, A. Weltman, A. S. Craig, B. P. Bell, and A. Fiore. 2005. Molecular epidemiology of foodborne hepatitis A outbreaks in the United States, 2003. J. Infect. Dis. 192:1323–1330. 6. Arauz-Ruiz, P., L. Sundqvist, Z. García, L. Taylor, K. Visoná, H. Norder, and L. O. Magnius. 2001. Presumed common source outbreaks of hepatitis A in an endemic area confirmed by limited sequencing within the VP1 region. J. Med. Virol. 65:449–456. 7. Asher, L. V., L. N. Binn, T. L. Mensing, R. H. Marchwicki, R. A. Vassell, and G. D. Young. 1995. Pathogenesis of hepatitis A in orally inoculated owl monkeys (Aotus trivirgatus). J. Med. Virol. 47:260–268. 8. Atmar, R. L., A. R. Opekun, M. A. Gilger, M. K. Estes, S. E. Crawford, F. H. Neill, and D. Y. Graham. 2008. Norwalk virus shedding after experimental human infection. Emerg. Infect. Dis. 14:1553–1557. 9. Ayodeji, M., M. Kulka, S. A. Jackson, I. Patel, M. Mammel, T. A. Cebula, and B. B. Goswami. 2009. A microarray based approach for the identification of common foodborne viruses. Open Virol. J. 3:7–20. 10. Bae, J., and K. J. Schwab. 2008. Evaluation of murine norovirus, feline calicivirus, poliovirus, and MS2 as surrogates for human norovirus in a model of viral persis-

SMP_Food Microbiology_CH24.indd

640

tence in surface water and groundwater. Appl. Environ. Microbiol. 74:477–484. 11. Baert, L., C. E. Wobus, E. V. Coillie, L. B. Thackray, J. Debevere, and M. Uyttendaele. 2008. Detection of murine norovirus 1 by using plaque assay, transfection assay, and real-time reverse transcription-PCR before and after heat exposure. Appl. Environ. Microbiol. 74:543–546. 12. Ball, J. M., D. Y. Graham, A. R. Opekun, M. A. Gilger, R. A. Guerrero, and M. K. Estes. 1999. Recombinant Norwalk virus-like particles given orally to volunteers: phase I study. Gastroenterology 117:40–48. 13. Bank-Wolf, B. R., M. König, and H. J. Thiel. 2010. Zoonotic aspects of infections with noroviruses and sapoviruses. Vet. Microbiol. 140:204–212. 14. Becker, K. M., C. L. Moe, K. L. Southwick, and J. N. MacCormack. 2000. Transmission of Norwalk virus during a football game. N. Engl. J. Med. 343:1223–1227. 15. Belliot, G., A. Lavaux, D. Souihel, D. Agnello, and P. Pothier. 2008. Use of murine norovirus as a surrogate to evaluate resistance of human norovirus to disinfectants. Appl. Environ. Microbiol. 74:3315–3318. 16. Berg, D. E., M. A. Kohn, T. A. Farley, and L. M. McFarland. 2000. Multi-state outbreaks of acute gastroenteritis traced to fecal-contaminated oysters harvested in Louisiana. J. Infect. Dis. 181(Suppl. 2): S381–S386. 17. Beuret, C., A. Baumgartner, and J. Schluep. 2003. Virus-contaminated oysters: a three-month monitoring of oysters imported to Switzerland. Appl. Environ. Microbiol. 69:2292–2297. 18. Bidawid, S., N. Malik, O. Adegbunrin, S. A. Sattar, and J. M. Farber. 2004. Norovirus cross-contamination during food handling and interruption of virus transfer by hand antisepsis: experiments with feline calicivirus as a surrogate. J. Food Prot. 67:103–109. 19. Bidawid, S., J. M. Farber, and S. A. Sattar. 2000. Contamination of foods by food handlers: experiments on hepatitis A virus transfer to food and its interruption. Appl. Environ. Microbiol. 66:2759–2763. 20. Bok, K., G. I. Parra, T. Mitra, E. Abente, C. K. Shaver, D. Boon, R. Engle, C. Yu, A. Z. Kapikian, S. V. Sosnovtsev, R. H. Purcell, and K. Y. Green. 2011. Chimpanzees as an animal model for human norovirus infection and vaccine development. Proc. Natl. Acad. Sci. USA 108:325–330. 21. Boone, S. A., and C. P. Gerba. 2007. Significance of fomites in the spread of respiratory and enteric viral disease. Appl. Environ. Microbiol. 73:1687–1696. 22. Boxman, I. L., L. Verhoef, R. Dijkman, G. Hägele, N. A. te Loeke, and M. Koopmans. 2011. Yearround prevalence of norovirus in the environment of catering companies without a recently reported outbreak of gastroenteritis. Appl. Environ. Microbiol. 77:2968–2974. 23. Boxman, I. L., R. Dijkman, N. A. te Loeke, G. Hägele, J. J. Tilburg, H. Vennema, and M. Koopmans. 2009. Environmental swabs as a tool in norovirus outbreak investigation, including outbreaks on cruise ships. J. Food Prot. 72:111–119.

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens 24. Boxman, I., R. Dijkman, L. Verhoee, A. Maat, G. van Dijk, H. Vennema, and M. Koopmans. 2009. Norovirus on swabs taken from hands illustrate route of transmission: a case study. J. Food Prot. 72:1753–1755. 25. Boxman, I. L., J. J. Tilburg, N. A. te Loeke, H. Vennema, E. de Boer, and M. Koopmans. 2007. An efficient and rapid method of recovery for norovirus from food associated with outbreaks of gastroenteritis. J. Food Prot. 70:504–508. 26. Bresee, J. S., M. A. Widdowson, S. S. Monroe, and R. I. Glass. 2002. Foodborne viral gastroenteritis: challenges and opportunities. Clin. Infect. Dis. 35:748–753. 27. Bull, R. A., J. S. Eden, W. D. Rawlinson, and P. A. White. 2010. Rapid evolution of pandemic noro­ viruses of the GII.4 lineage. PLoS Pathog. 6: e1000831. 28. Butot, S., T. Putallaz, R. Amoroso, and G. Sánchez. 2009. Inactivation of enteric viruses in minimally processed berries and herbs. Appl. Environ. Microbiol. 75:4155–4161. 29. Butot, S., T. Putallaz, and G. Sánchez. 2007. Procedure for rapid concentration and detection of enteric viruses from berries and vegetables. Appl. Environ. Microbiol. 73:186–192. 30. Calci, K. R., G. K. Meade, R. C. Tezloff, and D. H. Kingsley. 2005. High-pressure inactivation of hepatitis A virus within oysters. Appl. Environ. Microbiol. 71:339–343. 31. Cannon, J. L., and J. Vinjé. 2008. Histo-blood group antigen assay for detecting noroviruses in water. Appl. Environ. Microbiol. 74:6818–6819. 32. Cannon, J. L., E. Papafragkou, G. W. Park, J. Osborne, L. Jaykus, and J. Vinjé. 2006. Surrogates for the study of norovirus stability and inactivation in the environment: a comparison of murine norovirus and feline calicivirus. J. Food Prot. 69:2761–2765. 33. Carrique-Mas, J., Y. Andersson, B. Petersén, K.-O. Hedlund, N. Sjögren, and J. Giesecke. 2003. A Norwalklike virus waterborne community outbreak in a Swedish village during peak holiday season. Epidemiol. Infect. 131:737–744. 34. Centers for Disease Control and Prevention. 2009. Surveillance of foodborne disease outbreaks—United States, 2006. MMWR Morb. Mortal. Wkly. Rep. 58:609–615. 35. Centers for Disease Control and Prevention. 2011. Updated norovirus outbreak management and disease prevention guidelines. MMWR Recomm. Rep. 60(RR-3):1–18. 36. Centers for Disease Control and Prevention. 2005. Hepatitis A vaccination coverage among children aged 24–35 months—United States, 2003. MMWR Morb. Mortal. Wkly. Rep. 54:141–144. 37. Centers for Disease Control and Prevention. 2003. Outbreaks of gastroenteritis associated with noroviruses on cruise ships—United States, 2002. JAMA 289:167–169. 38. Centers for Disease Control and Prevention. 2003. Hepatitis A outbreak associated with green onions at

SMP_Food Microbiology_CH24.indd

641

641

39. 40. 41. 42.

43.

44.

45.

46.

47.

48.

49. 50. 51.

52. 53.

a restaurant—Monaca, Pennsylvania. MMWR Morb. Mortal. Wkly. Rep. 52:1155–1157. Chalmers, J. W. T., and J. H. McMillan. 1995. An outbreak of viral gastroenteritis associated with adequately prepared oysters. Epidemiol. Infect. 115:163–167. Chang, K. O., and D. W. George. 2007. Interferons and ribavirin effectively inhibit Norwalk virus replication in replicon-bearing cells. J. Virol. 81:12111–12118. Cliver, D. O. 1997. Virus transmission via food. Food Technol. 51:71–78. Cliver, D. O., and S. M. Matsui. 2002. Viruses, p.161–175. In D. O. Cliver and H. P. Reimann (ed.), Foodborne Diseases, 2nd ed. Academic Press, Boston, MA. Committee on Infectious Diseases, American Academy of Pediatrics. 1991. Hepatitis A, p. 234–237. In Report of the Committee on Infectious Diseases, 22nd ed. American Academy of Pediatrics, Elk Grove, IL. Cotterelle, B., C. Drougard, J. Rolland, M. Becamel, M. Boudon, S. Pinede, O. Traoré, K. Balay, P. Pothier, and E. Espié. 2005. Outbreak of norovirus infection associated with the consumption of frozen raspberries, France, March 2005. Euro Surveill. 10:E050428.1. Croci, L. M. N. Losio, E. Suffredini, E. Pavoni, S. Di Pasquale, F. Fallacara, and G. Ancangeli. 2007. Assessment of human enteric viruses in shellfish from the northern Adriatic sea. Int. J. Food Microbiol. 114:252–257. Croci, L., D. De Medici, G. Morace, A. Fiore, C. Scalfaro, F. Beneduce, and L. Toti. 2002. The survival of hepatitis A virus in fresh produce. Int. J. Food Microbiol. 73:29–34. Croci, L., M. Ciccozzi, D. De Medici, S. Di Pasquale, A. Fiore, A. Mele, and L. Toti. 1999. Inactivation of hepatitis A virus in heat-treated mussels. J. Appl. Microbiol. 87:884–888. Cromeans, T., O. V. Nainan, H. A. Fields, M. O. Favorov, and H. S. Margolis. 1994. Hepatitis A and E viruses, p. 1–56. In Y. H. Hui, J. R. Gorham, K. D. Murrell, and D. O. Cliver (ed.), Foodborne Disease Handbook, vol. 2. Diseases Caused by Viruses, Parasites, and Fungi. Marcel Dekker, New York, NY. Cuthbert, J. A. 2002. Hepatitis A: old and new. Clin. Microbiol. Rev. 14:38–58. Dalton, C. B., A. Haddix, R. E. Hoffman, and E. E. Mast. 1996. The cost of a food-borne outbreak of hepatitis A in Denver, Colo. Arch. Intern. Med. 156:1013–1016. DePaola, A., J. L. Jones, J. Woods, W. Burkhardt III, K. R. Calci, J. A. Krantz, J. C. Bowers, K. Kasturi, R. H. Byars, E. Jacobs, D. Williams-Hill, and K. Nabe. 2010. Bacterial and viral pathogens in live oysters: 2007 United States market survey. Appl. Environ. Microbiol. 76:2754–2768. Desselberger, U. 2000. Rotaviruses: methods and protocols, p. 1–8. In J. Gray and U. Desselberger (ed.), Methods in Molecular Medicine, vol. 34. Humana Press, Totowa, NJ. de Wit, M. A., M. A. Widdowson, H. Vennema, E. de Bruin, T. Fernandes, and M. Koopmans. 2007. Large outbreak of norovirus: the baker who should have known better. J. Infect. 55:188–193.

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

642 54. Divizia, M., R. Gabrielli, A. Macaluso, B. Bagnato, L. Palombi, E. Buonomo, F. Cenko, L. Leno, S. Bino, A. Basha, and A. Panà. 2005. Nucleotide correlation between HAV isolates from human patients and environmental samples. J. Med. Virol. 75:8–12. 55. Dolin, R., N. R. Blacklow, H. DuPont, R. F. Buscho, R. G. Wyatt, J. A. Kasel, R. Hornick, and R. M. Chanock. 1972. Biological properties of Norwalk agent of acute infectious nonbacterial gastroenteritis. Proc. Soc. Exp. Biol. Med. 140:578–583. 56. Donaldson, E. F., L. C. Lindesmith, A. D. LoBue, and R. S. Baric. 2010. Viral shape-shifting: norovirus evasion of the human immune system. Nat. Rev. Microbiol. 8:231–241. 57. Donaldson, E. F., L. C. Lindesmith, A. D. Lobue, and R. S. Baric. 2008. Norovirus pathogenesis: mechanisms of persistence and immune evasion in human populations. Immunol. Rev. 225:190–211. 58. Doré, W. J., M. Mackie, and D. N. Lees. 2003. Levels of male-specific RNA bacteriophage and Escherichia coli in molluscan bivalve shellfish from commercial harvesting areas. Lett. Appl. Microbiol. 36:92–96. 59. Doultree, J. C., J. D. Druce, C. J. Birch, D. S. Bowden, and J. A. Marshall. 1999. Inactivation of feline calicivirus, a Norwalk virus surrogate. J. Hosp. Infect. 41:51–57. 60. D’Souza, D. H., and L. Jaykus. 2006. Molecular approaches for the detection of foodborne viral pathogens, p. 91–118. In J. Maurer (ed.), PCR Methods in Foods, Springer, New York, NY. 61. D’Souza, D. H., J. Jean, and L. Jaykus. 2006. Methods for detection of viral and parasitic pathogens, p. 188-1– 188-23. In Y. H. Hui (ed.), Handbook of Food Technology and Engineering, vol. 4. CRC/Taylor and Francis, Boca Raton, FL. 62. Duizer, E., P. Bijkerk, B. Rockx, A. de Groot, F. Twisk, and M. Koopmans. 2004. Inactivation of caliciviruses. Appl. Environ. Microbiol. 70:4538–4543. 63. El-Kamary, S. S., M. F. Pasetti, P. M. Mendelman, S. E. Frey, D. I. Bernstein, J. J. Treanor, J. Ferreira, W. H. Chen, R. Sublett, C. Richardson, R. F. Bargatze, M. B. Sztein, and C. O. Tacket. 2010. Adjuvanted intranasal Norwalk virus-like particle vaccine elicits antibodies and antibody-secreting cells that express homing receptors for mucosal and peripheral lymphoid tissues. J. Infect. Dis. 202:1649–1658. 64. Falkenhorst, G., L. Krusell, M. Lisby, S. B. Madsen, B. Böttiger, and K. Mølbak. 2005. Imported frozen raspberries cause a series of norovirus outbreaks in Denmark, 2005. Euro Surveill. 10:E050922.2. 64a. Fauquet, C. M., M. A. Mayo, J. Maniloff, U. Desselberger, and L. A. Ball. 2005. Virus Taxonomy, Eighth Report of the International Committee on Taxonomy of Viruses. Elsevier Academic Press, New York, NY. 65. Feng, K., E. Diver, Y. Ma, and J. Li. 2011. Inactivation of a human norovirus surrogate, human norovirus viruslike particles, and vesicular stomatitis virus by gamma irradiation. Appl. Environ. Microbiol. 77:3507–3517. 66. Fino, V. R., and K. E. Kniel. 2008. Comparative recovery of foodborne viruses from fresh produce. Foodborne Pathog. Dis. 5:819–825.

SMP_Food Microbiology_CH24.indd

642

67. Fino, V. R., and K. E. Kniel. 2008. UV light inactivation of hepatitis A virus, Aichi virus, and feline calicivirus on strawberries, green onions, and lettuce. J. Food Prot. 71:908–913. 68. Fiore, A. E. 2004. Hepatitis A transmitted by food. Clin. Infect. Dis. 38:705–715. 69. Fleet, G. H., P. Heiskanen, I. Reid, and K. A. Buckle. 2000. Foodborne viral illness—status in Australia. Int. J. Food Microbiol. 59:127–136. 70. Food and Agriculture Organization/World Health Organization. 2008. Viruses in Food: Scientific Advice to Support Risk Management Activities, Meeting Report. Microbiological Risk Assessment Series No. 13. Food and Agriculture Organization/World Health Organization, Rome, Italy and Geneva, Switzerland. 71. Formiga-Cruz, M., A. K. Allard, A. C. Conden-Hansson, K. Henshilwood, B. E. Hernroth, J. Jofre, D. N. Lees, F. Lucena, M. Papapetropoulou, R. E. Rangdale, A. Tsibouxi, A. Vantarakis, and R. Girones. 2003. Evaluation of potential indicators of viral contamination in shellfish and their applicability to diverse geographical areas. Appl. Environ. Microbiol. 69:1556–1563. 72. Formiga-Cruz, M., G. Tofiño-Quesada, S. BofillMas, D. N. Lees, K. Henshilwood, A. K. Allard, A. C. Conden-Hansson, B. E. Hernroth, A. Vantarakis, A. Tsibouxi, M. Papapetropoulou, M. D. Furones, and R. Girones. 2002. Distribution of human virus contamination in shellfish from different growing areas in Greece, Spain, Sweden, and the United Kingdom. Appl. Environ. Microbiol. 68:5990–5998. 73. Friedman, D. S., D. Heisey-Grove, F. Argyros, E. Berl, J. Nsubuga, T. Stiles, J. Fontana, R. S. Beard, A. Moneow, M. E. McGrath, H. Sutherby, R. C. Dicker, A. DeMaria, and B. T. Matyas. 2005. An outbreak of norovirus gastroenteritis associated with wedding cakes. Epidemiol. Infect. 133:1057–1063. 74. Gallimore, C. I., D. Cubitt, N. du Plessis, and J. J. Gray. 2004. Asymptomatic and symptomatic excretion of noroviruses during a hospital outbreak of gastroenteritis. J. Clin. Microbiol. 42:2271–2274. 75. Gandhi, K. M., R. E. Mandrell, and P. Tian. 2010. Binding of virus-like particles of Norwalk virus to romaine lettuce veins. Appl. Environ. Microbiol. 76:7997–8003. 76. Gaulin, C., M. Frigon, D. Poirier, and C. Fournier. 1999. Transmission of calicivirus by a foodhandler in the pre-symptomatic phase of illness. Epidemiol. Infect. 123:475–478. 77. Gehrke, C., J. Steinmann, and P. Goroncy-Bermes. 2004. Inactivation of feline calicivirus, a surrogate of norovirus (formerly Norwalk-like viruses), by different types of alcohol in vitro and in vivo. J. Hosp. Infect. 56:49–55. 78. Gerba, C. P., D. M. Gramos, and N. Nwachuku. 2002. Comparative inactivation of enteroviruses and adenovirus 2 by UV light. Appl. Environ. Microbiol. 68:5167–5169. 79. Gersberg, R. M., M. A. Rose, R. Robles-Sikisaka, and A. K. Dhar. 2006. Quantitative detection of hepatitis A

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens virus and enteroviruses near the United States-Mexico border and correlation with levels of fecal indicator bacteria. Appl. Environ. Microbiol. 72:7438–7444. 80. Girard, M., S. Ngazoa, K. Mattison, and J. Jean. 2010. Attachment of noroviruses to stainless steel and their inactivation, using household disinfectants. J. Food Prot. 73:400–404. 81. Glass, R. I., U. D. Parashar, and M. K. Estes. 2009. Norovirus gastroenteritis: an update. N. Engl. J. Med. 361:1776–1785. 82. Green, K. Y., R. M. Chanock, and A. Z. Kapikian. 2001. Human caliciviruses, p. 841–874. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields Virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA. 83. Green, K. Y., J. F. Lew, X. Jiang, A. Z. Kapikian, and M. K. Estes. 1993. Comparison of the reactivities of baculovirus-expressed recombinant Norwalk virus capsid antigen with those of the native Norwalk virus antigen in serologic assays and some epidemiologic observations. J. Clin. Microbiol. 31:2185–2191. 84. Gulati, B. R., P. B. Allwood, C. W. Hedberg, and S. M. Goyal. 2001. Efficacy of commonly used disinfectants for the inactivation of calicivirus on strawberry, lettuce, and a food-contact surface. J. Food Prot. 64:1430–1434. 85. Halliday, M. L., L. Y. Kang, T. K. Zhou, M. D. Hu, Q. C. Pan, T. Y. Fu, Y. S. Huang, and S. L. Hu. 1991. An epidemic of hepatitis A attributable to the ingestion of raw clams in Shanghai, China. J. Infect. Dis. 164:852–859. 86. Halperin, T., H. Vennema, M. Koopmans, G. K. BarGal, R. Kayouf, T. Sela, R. Ambar, and E. Klement. 2008. No association between histo-blood group antigens and susceptibility to clinical infections with genogroup II norovirus. J. Infect. Dis. 197:63–65. 87. Harewood, P., S. Rippey, and M. Montesalvo. Effect of gamma irradiation on shelf life and bacterial and viral loads in hard-shelled clams (Mercenaria mercenaria). Appl. Environ. Microbiol. 60:2666–2670. 88. Harrington, P. R., L. Lindesmith, B. Yount, C. L. Moe, J. LePendu, and R. S. Baric. 2003. Norovirus attachment, susceptibility and vaccine design. Recent Res. Dev. Virol. 5:19–44. 89. Harrington, P. R., L. Lindesmith, B. Yount, C. L. Moe, and R. S. Baric. 2002. Binding of Norwalk viruslike particles to ABH histo-blood group antigens is blocked by antisera from infected human volunteers or experimentally vaccinated mice. J. Virol. 76:12335–12343. 90. Harris, J. P., W. J. Edmunds, R. Pebody, D. W. Brown, and B. A. Lopman. 2008. Deaths from norovirus among the elderly, England and Wales. Emerg. Infect. Dis. 14:1546–1552. 91. Hata, A., H. Katayama, M. Kitajima, C. Visvanathan, C. Nol, and H. Furumai. 2011. Validation of internal controls for extraction and amplification of nucleic acids from enteric viruses in water samples. Appl. Environ. Microbiol. 77:4336–4343.

SMP_Food Microbiology_CH24.indd

643

643 92. Herbst-Kralovetz, M., H. S. Mason, and Q. Chen. 2010. Norwalk virus-like particles as vaccines. Expert Rev. Vaccines 9:299–307. 93. Hennessy, E. P., A. D. Green, M. P. Connor, R. Darby, and P. MacDonald. 2003. Norwalk virus infection and disease is associated with ABO histo-blood group type. J. Infect. Dis. 188:176–177. 94. Hewitt, J., M. Rivera-Aban, and G. E. Greening. 2009. Evaluation of murine norovirus as a surrogate for human norovirus and hepatitis A virus in heat activation studies. J. Appl. Microbiol. 107:65–71. 95. Hewitt, J., and G. E. Greening. 2004. Survival and persistence of norovirus, hepatitis A virus, and feline calicivirus in marinated mussels. J. Food Prot. 67:1743–1750. 96. Hirneisen, K. A., S. M. Markland, and K. E. Kniel. 2011. Ozone inactivation of norovirus surrogates on fresh produce. J. Food Prot. 74:836–839. 97. Hjertqvist, M., A. Johansson, N. Svensson, P. E. Abom, C. Magnusson, M. Olsson, K. O. Hedlund, and Y. Andersson. 2006. Four outbreaks of norovirus gastroenteritis after consuming raspberries, Sweden, June– August 2006. Euro Surveill. 11:E060907.1. 98. Hofmann, F., G. Wehrle, H. Berthold, and D. Koster. 1992. Hepatitis A as an occupational hazard. Vaccine 10:82–84. 99. Hollinger, F. B., and S. U. Emerson. 2001. Hepatitis A virus, p. 799–840. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields Virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA. 100. Hollinger, F. B., and J. R. Ticehurst. 1996. Hepatitis A virus, p. 735–774. In B. N. Fields, D. M. Knipe, P. M. Howley, M. Chanock, J. L. Melnick, and T. P. Monath (ed.), Fields Virology, 3rd ed. Lippincott-Raven Publishers, Philadelphia, PA. 101. Huang, P., T. Farkas, S. Marionneau, W. Zhong, N. Ruvoën-Clouet, A. L. Morrow, M. Altaye, L. K. Pickering, D. S. Newburg, J. LePendu, and X. Jiang. 2003. Noroviruses bind to human ABO, Lewis and secretor histo-blood group antigens: identification of 4 distinct strain-specific patterns. J. Infect. Dis. 188:19–31. 102. Hutin, Y. J., V. Pool, E. H. Cramer, O. V. Nainan, J. Weth, I. T. Williams, S. T. Goldstein, K. F. Gensheimer, B. P. Bell, C. N. Shapiro, M. J. Alter, and H. S. Margolis for the National Hepatitis A Investigation Team. 1999. A multistate, foodborne outbreak of hepatitis A. N. Engl. J. Med. 340:595–601. 103. Hutson, A. M., R. L. Atmar, D. Y. Graham, and M. K. Estes. 2002. Norwalk virus infection and disease is associated with ABO histo-blood group type. J. Infect. Dis. 185:1335–1337. 104. Isakbaeva, E. T., M. A. Widdowson, R. S. Beard, S. N. Bulens, J. Mullins, S. S. Monroe, J. Bresee, P. Sassano, E. H. Cramer, and R. I. Glass. 2005. Norovirus transmission on cruise ship. Emerg. Infect. Dis. 11:154–158. 105. Jaykus, L., and B. Escudero-Abarca. 2010. Human pathogenic viruses in food, p. 195–232. In V. J. Juneja and J. N. Sofos (ed.), Pathogens and Toxins in Foods: Challenges and Interventions. ASM Press, Washington, DC.

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

644 106. Jaykus, L., M. T. Hemard, and M. D. Sobsey. 1994. Human enteric pathogenic viruses, p. 92–153. In C. R. Hackney and M. D. Pierson (ed.), Environmental Indicators and Shellfish Safety. Chapman and Hall, New York, NY. 107. Jean, J., J.-F. Vachon, O. Moroni, A. Darveau, I. Kukavica-Ibrulj, and I. Fliss. 2003. Effectiveness of commercial disinfectants for inactivating hepatitis A virus on agri-food surfaces. J. Food Prot. 66:115–119. 108. Johnson, P. C., J. J. Mathewson, H. L. DuPont, and H. B. Greenberg. 1990. Multiple-challenge study of host susceptibility to Norwalk gastroenteritis in US adults. J. Infect. Dis. 161:18–21. 109. Jothikumar, N., T. L. Cromeans, B. H. Robertson, X. J. Meng, and V. R. Hill. 2006. A broadly reactive one-step real-time RT-PCR assay for rapid and sensitive detection of hepatitis E virus. J. Virol. Methods 131:65–71. 110. Jothikumar, N., T. L. Cromeans, M. D. Sobsey, and B. H. Robertson. 2005. Development and evaluation of a broadly reactive TaqMan assay for rapid detection of hepatitis A virus. Appl. Environ. Microbiol. 71:3359–3363. 111. Jothikumar, N., J. A. Lowther, K. Henshilwood, D. N. Lees, V. R. Hill, and J. Vinjé. 2005. Rapid and sensitive detection of noroviruses by using TaqMan-based one-step reverse transcription-PCR assays and application to naturally contaminated shellfish samples. Appl. Environ. Microbiol. 71:1870–1875. 112. Kageyama, T., S. Kojima, M. Shinohara, K. Uchida, S. Fukushi, F. B. Hoshino, N. Takeda, and K. Katayama. 2003. Broadly reactive and highly sensitive assay for Norwalk-like viruses on real-time quantitative reverse transcription-PCR. J. Clin. Microbiol. 41:1548–1557. 113. Kampf, G., D. Grotheer, and J. Steinmann. 2005. Efficacy of three ethanol-based hand rubs against feline calicivirus, a surrogate virus for norovirus. J. Hosp. Infect. 60:144–149. 114. Kapikian, A. Z., M. K. Estes, and R. M. Chanock. 1996. Norwalk group of viruses, p. 783–810. In B. N. Fields, D. M. Knipe, P. M. Howley, M. Chanock, J. L. Melnick, and T. P. Monath (ed.), Fields Virology, 3rd ed. Lippincott-Raven Publishers, Philadelphia, PA. 115. Kingsley, D. H., D. R. Holliman, K. R. Calci, H. Chen, and G. J. Flick. 2007. Inactivation of a norovirus by high-pressure processing. Appl. Environ. Microbiol. 73:581–585. 116. Kingsley, D. H., and G. P. Richards. 2003. Persistence of hepatitis A virus in oysters. J. Food Prot. 66:331–334. 117. Kingsley, D. H., D. G. Hoover, E. Papafragkou, and G. P. Richards. 2002. Inactivation of hepatitis A virus and a calicivirus by high hydrostatic pressure. J. Food Prot. 65:1605–1609. 118. Kobayashi, S., K. Natori, N. Takeda, and K. Sakae. 2004. Immunomagnetic capture RT-PCR for detection of norovirus from foods implicated in a foodborne outbreak. Microbiol. Immunol. 48:201–204. 119. Kohn, M. A., T. A. Farley, T. Ando, M. Curtis, S. A. Wilson, Q. Jin, S. S. Monroe, R. C. Baron, L. M.

SMP_Food Microbiology_CH24.indd

644

McFarland, and R. I. Glass. 1995. An outbreak of Norwalk virus gastroenteritis associated with eating raw oysters. JAMA 273:466–471. 120. Koopmans, M., and E. Duizer. 2004. Foodborne viruses: an emerging problem. Int. J. Food Microbiol. 90:23–41. 121. Kroneman, A., L. Verhoef, J. Harris, H. Vennema, E. Duizer, Y. van Duynhoven, J. Gray, M. Iturriza, B. Böttiger, G. Falkenhorst, C. Johnsen, C. H. von Bonsdorff, L. Maunula, M. Kuusi, P. Pothier, A. Gallay, E. Schreier, M. Höhne, J. Koch, G. Szücs, G. Reuter, K. Krisztalovics, M. Lynch, P. McKeown, B. Foley, S. Coughlan, F. M. Ruggeri, I. Di Bartolo, K. Vainio, E. Isakbaeva, M. Poljsak-Prijatelj, A. H. Grom, J. Z. Mijovski, A. Bosch, J. Buesa, A. S. Fauquier, G. Hernandéz-Pezzi, K. O. Hedlund, and M. Koopmans. 2008. Analysis of integrated virological and epidemiological reports of norovirus outbreaks collected within the Foodborne Viruses in Europe network from 1 July 2001 to 30 June 2006. J. Clin. Microbiol. 46:2959–2965. 122. Lages, S. L., M. A. Ramakrishnan, and S. M. Goyal. 2008. In-vivo efficacy of hand sanitisers against feline calicivirus: a surrogate for norovirus. J. Hosp. Infect. 68:159–163. 123. Lee, N., M. C. Chan, B. Wong, K. W. Choi, W. Sin, G. Lui, P. K. Chan, R. W. Lai, C. S. Crockram, J. J. Sung, and W. K. Leung. 2007. Fecal viral concentration and diarrhea in norovirus gastroenteritis. Emerg. Infect. Dis. 13:1399–1401. 124. Le Guyader, F. S., S. Parnaudeau, J. Schaeffer, A. Bosch, F. Loisy, M. Pommepuy, and R. L. Atmar. 2009. Detection and quantification of noroviruses in shellfish. Appl. Environ. Microbiol. 75:618–624. 125. Le Guyader, F., L. Haugarreau, L. Miossec, E. Dubois, and M. Pommepuy. 2000. Three-year study to assess human enteric viruses in shellfish. Appl. Environ. Microbiol. 66:3241–3248. 126. Lemon, S. M. 1997. Type A hepatitis: epidemiology, diagnosis, and prevention. Clin. Chem. 43:1494–1499. 127. Leon, J. S., D. H. Kingsley, J. S. Montes, G. P. Richards, G. M. Lyon, G. M. Abdulhafid, S. R. Seitz, M. L. Fernandez, P. F. Teunis, G. J. Flick, and C. L. Moe. 2011. Randomized, double-blinded clinical trial for human norovirus inactivation in oysters by high hydrostatic pressure processing. Appl. Environ. Microbiol. 77:5476–5482. 128. Leon, J. S., S. Menira, Q. Wang, E. R. Smith, L. J. Saif, and C. L. Moe. 2008. Immunology of norovirus infection, p. 219–262. In M. Vajdy (ed.), Immunity against Mucosal Pathogens. Springer, Boston, MA. 129. León-Félix, J., R. A. Martínez-Bustillos, M. BáezSañudo, F. Peraza-Garay, and C. Chaidez. 2010. Norovirus contamination of bell pepper from handling during harvesting and packing. Food Environ. Virol. 2:211–217. 130. Li, D., L. Baert, M. De Jonghe, E. Van Coillie, J. Ryckeboer, F. Devlieghere, and M. Uyttendaele. 2011. Inactivation of murine norovirus 1, coliphage jX174, and Bacteroides [corrected] fragilis phage B40-8 on

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens

131. 132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.

surfaces and fresh-cut iceberg lettuce by hydrogen peroxide and UV light. Appl. Environ. Microbiol. 77:1399–1404. Lim, M. Y., J. M. Kim, J. E. Lee, and G. P. Ko. 2010. Characterization of ozone disinfection of murine norovirus. Appl. Environ. Microbiol. 76:1120–1124. Lindesmith, L., E. Donaldson, J. Leon, C. L. Moe, J. A. Frelinger, R. E. Johnston, D. J. Weber, and R. S. Baric. 2010. Heterotypic humoral and cellular immune response following Norwalk virus infection. J. Virol. 84:1800–1815. Lindesmith, L. C., E. F. Donaldson, A. D. LoBue, J. L. Cannon, D. P. Zheng, J. Vinje, and R. S. Baric. 2008. Mechanisms of GII.4 norovirus persistence in human populations. PLoS Med. 5:e31. Lindesmith, L., C. Moe, J. LePendu, J. A. Frelinger, J. Treanor, and R. S. Baric. 2005. Cellular and humoral immunity following Snow Mountain virus challenge. J. Virol. 79:2900–2909. Lindesmith, L., C. Moe, S. Marionneau, N. Ruvoen, X. Jiang, J. Lindblad, P. Stewart, J. LePendu, and R. Baric. 2003. Human susceptibility and resistance to Norwalk virus infection. Nat. Med. 9:548–553. Liu, B., P. Maywood, L. Gupta, and B. Campbell. 2003. An outbreak of Norwalk-like virus gastroenteritis in an aged-care residential hostel. N. S. W. Public Health Bull. 14:105–109. Liu, P., Y. Yuen, H.-M. Hsiao, L. Jaykus, and C. Moe. 2010. Effectiveness of liquid soap and hand sanitizer against Norwalk virus on contaminated hands. Appl. Environ. Microbiol. 76:394–399. Liu, P., Y.-W. Chien, E. Papafragkou, H.-M. Hsiao, L. Jaykus, and C. L. Moe. 2009. Persistence of human noroviruses on food preparation surfaces and human hands. Food Environ. Virol. 1:141–147. Lopman, B., M. Reacher, Y. van Duijnhoven, F. X. Hanon, D. Brown, and M. Koopmans. 2003. Viral gastroenteritis outbreaks in Europe, 1995–2000. Emerg. Infect. Dis. 9:90–96. Lukasik, J., M. L. Bradley, T. M. Scott, M. Dea, A. Koo, W.-Y. Hsu, J. A. Bartz, and S. R. Farrah. 2003. Reduction of poliovirus 1, bacteriophages, Salmonella montevideo, and Escherichia coli 0157:H7 on strawberries by physical and disinfectant washes. J. Food Prot. 66:188–193. Macinga, D. R., S. A. Sattar, L. Jaykus, and J. W. Arbogast. 2008. Improved inactivation of non-enveloped enteric viruses and their surrogates by a novel alcohol-based hand sanitizer. Appl. Environ. Microbiol. 74:5047–5052. Maier, K., P. Gabriel, E. Koscielniak, Y.-D. Stierhof, K. H. Wiedmann, K. B. Flehmig, and A. Vallbracht. 1998. Human gamma interferon production by cytotoxic T lymphocytes sensitized during hepatitis A virus infection. J. Virol. 62:3756–3763. Malek, M., E. Barzilay, A. Kramer, B. Camp, L. Jaykus, B. Escudero-Abarca, G. Derrick, P. White, C. Gerba, C. Higgins, J. Vinje, R. Glass, M. Lynch, and M. A. Widdowson. 2009. Outbreak of norovirus infection

SMP_Food Microbiology_CH24.indd

645

645

144. 145.

146.

147.

148.

149.

150.

151.

152. 153.

154. 155. 156. 157.

among river rafters associated with packaged delicatessen meat, Grand Canyon, 2005. Clin. Infect. Dis. 48:31–37. Mallet, J. C., L. E. Beghian, T. G. Metcalf, and J. D. Kaylor. 1991. Potential of irradiation technology for improving shellfish sanitation. J. Food Safety 11:231–245. Marks, P. J., I. B. Vipond, D. Carlisle, D. Deakin, R. E. Fey, and E. O. Caul. 2000. Evidence for airborne transmission of Norwalk-like virus (NLV) in a hotel restaurant. Epidemiol. Infect. 124:481–487. Maunula, L., M. Roivainen, M. Keränen, S. Mäkela, K. Söderberg, M. Summa, C. H. von Bonsdorff, M. Lappalainen, T. Korhonen, M. Kuusi, and T. Niskanen. 2009. Detection of human noroviruses from frozen raspberries in a cluster of gastroenteritis outbreaks. Euro Surveill. 14:pii=19435. Mattison, K., J. Harlow, V. Morton, A. Cook, F. Pollari, S. Bidawid, and J. M. Farber. 2010. Enteric viruses in ready-to-eat packaged leafy greens. Emerg. Infect. Dis. 11:1815–1817. Mattison, K., E. Grudeski, B. Auk, H. Charest, S. J. Drews, A. Fritzinger, N. Gregoricus, S. Hayward, A. Houde, B. E. Lee, X. L. Pang, J. Wong, T. F. Booth, and J. Vinjé. 2009. Multicenter comparison of two norovirus ORF2-based genotyping protocols. J. Clin. Microbiol. 47:3927–3932. Mbithi, J. N., V. S. Springthorpe, and S. A. Sattar. 1993. Comparative in vivo efficiencies of hand-washing agents against hepatitis A virus (HM-175) and poliovirus type 1 (Sabin). Appl. Environ. Microbiol. 59:3463–3469. Mbithi, J. N., V. S. Springthorpe, and S. A. Sattar. 1990. Chemical disinfection of hepatitis A virus on environmental surfaces. Appl. Environ. Microbiol. 56:3601–3604. McDonnell, S., K. B. Kirkland, W. G. Hlady, C. Aristeguieta, R. S. Hopkins, S. S. Monroe, and R. I. Glass. 1997. Failure of cooking to prevent shellfishassociated viral gastroenteritis. Arch. Intern. Med. 157:111–116. Meltzer, M. I., C. N. Shapiro, E. E. Mast, and C. Arcari. 2001. The economics of vaccinating restaurant workers against hepatitis A. Vaccine 19:2138–2145. Michaels, B., C. Keller, M. Blevins, G. Paoli, T. Ruthman, E. Todd, and C. J. Griffith. 2004. Prevention of food worker transmission of foodborne pathogens: risk assessment and evaluation of effective hygiene intervention strategies. Food Service Technol. 4:31–49. Millard, J., H. Appleton, and J. V. Parry. 1987. Studies on heat inactivation of hepatitis A virus with special reference to shellfish. Epidemiol. Infect. 98:397–414. Moe, C. L. 2003. Next time you have vomiting patients in your office—consider noroviruses. Atlanta Med. 77:13–19. Moe, C. L. 2009. Preventing norovirus transmission: how should we handle food handlers? Clin. Infect. Dis. 48:38–40. Moe, C. L., E. Honorat, J. Leon, and J. Eisenberg. 2007. Epidemiologic patterns in published norovirus outbreak reports, 1981­–2006, abstract P3-18. Third

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

646 International Calicivirus Conference, Cancun, Mexico, November 2007. 158. Mokhtari, A., and L. Jaykus. 2009. Quantitative exposure model for the transmission of norovirus in retail food preparation. Int. J. Food Microbiol. 133:38–47. 159. Mormann, S., M. Dabisch, and B. Becker. 2010. Effects of technological processes on the tenacity and inactivation of norovirus genogroup II in experimentally contaminated foods. Appl. Environ. Microbiol. 76:536–545. 160. Morse, D. L., J. J. Guzewich, J. P. Hanrahan, R. Stricof, M. Shayegani, R. Deibel, J. C. Grabau, N. A. Nowak, J. E. Herrmann, G. Cukor, and N. R. Blacklow. 1986. Widespread outbreaks of clam- and oyster-associated gastroenteritis. Role of Norwalk virus. N. Engl. J. Med. 314:678–681. 161. Morton, V., J. Jean, J. Farber, and K. Mattison. 2009. Detection of noroviruses in ready-to-eat foods by using carbohydrate-coated magnetic beads. Appl. Environ. Microbiol. 75:4641–4643. 162. Muniain-Mujika, I., M. Calvo, F. Lucena, and R. Girones. 2003. Comparative analysis of viral pathogens and potential indicators in shellfish. Int. J. Food Microbiol. 83:75–85. 163. Niu, M. T., L. B. Polish, B. H. Robertson, B. K. Khanna, B. A. Woodruff, C. N. Shapiro, M. A. Miller, J. D. Smith, J. K. Gedrose, M. J. Alterand, and H. S. Margolis. 1992. Multistate outbreak of hepatitis A associated with frozen strawberries. J. Infect. Dis. 166:518–524. 164. Nowak, P., J. R. Topping, V. Fotheringham, C. I. Gallimore, J. J. Gray, M. Iturriza-Gomara, and A. I. Knight. 2011. Measurement of the virolysis of human GII.4 norovirus in response to disinfectants and sanitisers. J. Virol. Methods 174:7–11. 165. Okabayashi, T., S. Yokota, Y. Ohkoshi, H. Ohuchi, Y. Yoshida, M. Kikuchi, K. Yano, and N. Fujii. 2008. Occurrence of norovirus unrelated to norovirus outbreaks in an asymptomatic food handler population. J. Clin. Microbiol. 46:1985–1988. 166. Ozawa, K., T. Oka, N. Tekeda, and G. S. Hansman. 2007. Norovirus infections in symptomatic and asymp­ tomatic food handlers in Japan. J. Clin. Microbiol. 45:3996–4005. 167. Pagotto, F., N. Corneau, K. Mattison, and S. Bidawid. 2008. Development of a DNA microarray for the simultaneous detection and genotyping of noroviruses. J. Food Prot. 71:1434–1441. 168. Pallansch, M. A., and R. P. Roos. 2001. Enteroviruses: polioviruses, coxsackieviruses, echoviruses, and newer enteroviruses, p. 723–775. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields Virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA. 169. Papafragkou, E., M. Plante, K. Mattison, S. Bidawid, K. Karthikeyan, J. M. Farber, and L. Jaykus. 2007. Rapid and sensitive detection of hepatitis A virus in representative food matrices. J. Virol. Methods 147:177–187. 170. Papafragkou, E., D. H. D’Souza, and L. Jaykus. 2006. Foodborne viruses: prevention and control, p. 289–330. In S.M. Goyal (ed.), Food Virology. Kluwer Academic/ Plenum Publishers, New York, NY.

SMP_Food Microbiology_CH24.indd

646

171. Park, G. W., and M. D. Sobsey. 2011. Simultaneous comparison of murine norovirus, feline calicivirus, coliphage MS2, and GII.4 norovirus to evaluate the efficacy of sodium hypochlorite against human norovirus on a fecally soiled stainless steel surface. Foodborne Pathog. Dis. 8:1005–1010. 172. Park, G. W., L. Barclay, D. Macinga, D. Charbonneau, C. A. Pettigrew, and J. Vinjé. 2010. Comparative efficacy of seven hand sanitizers against murine norovirus, feline calicivirus, and GII.4 norovirus. J. Food Prot. 73:2232–2238. 173. Parker, T. D., N. Kitamoto, T. Tanaka, A. M. Hutson, and M. K. Estes. 2005. Identification of genogroup I and genogroup II broadly reactive epitopes on the norovirus capsid. J. Virol. 79:7402–7409. 174. Parrino, T. A., D. S. Schreiber, J. S. Trier, A. Z. Kapikian, and N. R. Blacklow. 1977. Clinical immunity in acute gastroenteritis caused by Norwalk agent. N. Engl. J. Med. 297:86–89. 175. Parshionikar, S., I. Laseke, and G. S. Fout. 2010. Use of propidium monoazide in reverse transcriptase PCR to distinguish between infectious and noninfectious enteric viruses in water samples. Appl. Environ. Microbiol. 76:4318–4326. 176. Patel, M. M., V. R. López-Collada, M. M. Bulhões, L. H. De Oliveira, A. Bautista Márquez, B. Flannery, M. Esparza-Aguilar, E. I. Montenegro Renoiner, M. E. Luna-Cruz, H. K. Sato, L. del Carmen HernándezHernández, G. Toledo-Cortina, M. Cerón-Rodríguez, N. Osnaya-Romero, M. Martínez-Alcazar, R. G. AguinagaVillasenor, A. Plascencia-Hernández, F. FojacoGonzález, G. Hernández-Peredo Rezk, S. F. GutierrezRamírez, R. Dorame-Castillo, R. Tinajero-Pizano, B. Mercado-Villegas, M. R. Barbosa, E. M. Maluf, L. B. Ferreira, F. M. de Carvalho, A. R. dos Santos, E. D. Cesar, M. E. de Oliveira, C. L. Silva, M. de Los Angeles Cortes, C. Ruiz Matus, J. Tate, P. Gargiullo, and U. D. Parashar. 2011. Intussusception risk and health benefits of rotavirus vaccination in Mexico and Brazil. N. Engl. J. Med. 364:2283–2292. 177. Patel, M. M., M. A. Widdowson, R. I. Glass, K. Akazawa, J. Vinjé, and U. D. Parashar. 2008. Systematic literature review of role of norovirus in sporadic gastroenteritis. Emerg. Infect. Dis. 14:1224–1231. 178. Patterson, W., P. Haswell, P. T. Fryers, and J. Green. 1997. Outbreak of small round structured virus gastroenteritis arose after kitchen assistant vomited. Commun. Dis. Rep. CDR Rev. 7:R101–R103. 179. Pavia, A. T., L. Nielson, L. Armington, D. J. Thurman, E. Tierney, and C. R. Nichols. 1990. A communitywide outbreak of hepatitis A in a religious community: impact of mass administration of immune globulin. Am. J. Epidemiol. 131:1085–1093. 180. Poschetto, L. F., A. Ike, T. Papp, U. Mohn, R. Böhm, and R. E. Marschang. 2007. Comparison of the sensitivities of noroviruses and feline calicivirus to chemical disinfection under field-like conditions. Appl. Environ. Microbiol. 73:5494–5500. 181. Predmore, A., and J. Li. 2011. Enhanced removal of a human norovirus surrogate from fresh vegetables and

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens fruits by a combination of surfactants and sanitizers. Appl. Environ. Microbiol. 77:4829–4838. 182. Purdy, M. A., and Y. E. Khudyakov. 2011. The molecular epidemiology of hepatitis E virus infection. Virus Res. 161:31–39. 183. Richards, G. P. 2001. Enteric virus contamination of foods through industrial practices: a primer on intervention strategies. J. Ind. Microbiol. Biotechnol. 27:117–125. 184. Rockx, B., M. de Wit, H. Vennema, J. Vinjé, E. de Bruin, Y. van Duynhoven, and M. Koopmans. 2002. Natural history of human calicivirus infection: a prospective cohort study. Clin. Infect. Dis. 35:246–253. 185. Rodriguez, R. A., I. L. Pepper, and C. P. Gerba. 2009. Application of PCR-based methods to assess the infectivity of enteric viruses in environmental samples. Appl. Environ. Microbiol. 75:297–307. 186. Rutjes, S. A., F. Lodder-Verschoor, W. H. van der Poel, Y. T. van Duijnhoven, and A. M. de Roda Husman. 2006. Detection of noroviruses in foods: a study on virus extraction procedures in foods implicated in outbreaks of human gastroenteritis. J. Food Prot. 69:1949–1956. 187. Sano, D., R. M. Pinto, T. Omura, and A. Bosch. 2010. Detection of oxidative damages on viral capsid protein for evaluating structural integrity and infectivity of human norovirus. Environ. Sci. Technol. 44:808–812. 188. Santi, L., L. Batchelor, Z. Huang, B. Hjelm, J. Kilbourne, C. J. Arntzen, Q. Chen, and H. S. Mason. 2008. An efficient plant viral expression system generating orally immunogenic Norwalk virus-like particles. Vaccine 26:1846–1854. 189. Sattar, S. A. 2004. Microbicides and the environmental control of nosocomial viral infections. J. Hosp. Infect. 56:564–569. 190. Sattar, S. A., M. Ali, and J. A. Tetro. 2011. In vivo comparison of two human norovirus surrogates for testing ethanol-based handrubs: the mouse chasing the cat! PLoS One 6:e17340. 191. Sattar, S. A., V. S. Springthorpe, J. Tetro, R. Vashon, and B. Keswick. 2002. Hygienic hand antiseptics: should they not have activity and label claims against viruses? Am. J. Infect. Control 30:355–372. 192. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. 193. Scallan, E., P. M. Griffin, F. J. Angulo, R. V. Tauxe, and R. M. Hoekstra. 2011. Foodborne illness acquired in the United States—unspecified agents. Emerg. Infect. Dis. 17:16–22. 194. Schultz, A. C., E. Vega, A. Dalsgaard, L. S. Christensen, B. Nørrung, J. Hoorfar, and J. Vinjé. 2011. Development and evaluation of novel one-step TaqMan realtime RT-PCR assays for the detection and direct genotyping of genogroup I and II noroviruses. J. Clin. Virol. 50:230–234. 195. Schultz, A. C., S. Perelle, S. Di Pasquale, K. Kovac, D. De Medici, P. Fach, H. M. Sommer, and J. Hoorfar. 2011. Collaborative validation of a rapid method for ef-

SMP_Food Microbiology_CH24.indd

647

647

196.

197.

198. 199.

200.

201.

202.

203.

204.

205.

206. 207.

208. 209.

210.

ficient virus concentration in bottled water. Int. J. Food Microbiol. 145(Suppl. 1):S158–S166. Schwab, K. J., F. H. Neill, F. Le Guyader, M. K. Estes, and R. L. Atmar. 2001. Development of a reverse ­transcriptionPCR-DNA enzyme immunoassay for detection of “Norwalk-like” viruses and hepatitis A virus in stool and shellfish. Appl. Environ. Microbiol. 67:742–749. Schwab, K. J., F. H. Neill, M. K. Estes, T. G. Metcalf, and R. L. Atmar. 1998. Distribution of Norwalk virus within shellfish following bioaccumulation and subsequent depuration by detection using RT-PCR. J. Food Prot. 61:1674–1680. Shirato, H. 2011. Norovirus and histo-blood group antigens. Jpn. J. Infect. Dis. 64:95–103. Sinton, L. W., C. H. Hall, P. A. Lynch, and R. J. DaviesColley. 2002. Sunlight inactivation of fecal indicator bacteria and bacteriophages from waste stabilization pond effluent in fresh and saline waters. Appl. Environ. Microbiol. 68:1122–1131. Slomka, M. J., and H. Appleton. 1998. Feline calicivirus as a model system for heat inactivation studies of small round structured viruses in shellfish. Epidemiol. Infect. 121:401–407. Sobsey, M. D., and L. Jaykus. 1991. Human enteric viruses and depuration of bivalve molluscs, p. 71–114. In W. S. Otwell, G. E. Rodrick, and R. E. Martin (ed.), Molluscan Shellfish Depuration. CRC Press, Boca Raton, FL. Sow, H., M. Desbiens, R. Morales-Rayas, S. E. Ngazoa, and J. Jean. 2011. Heat inactivation of hepatitis A virus and a norovirus surrogate in soft-shell clams (Mya arenaria). Foodborne Pathog. Dis. 8:387–393. Stals, A., L. Baert, A. De Keuckelaere, E. Van Coillie, and M. Uyttendaele. 2011. Evaluation of norovirus detection methodology for ready-to-eat foods. Int. J. Food Microbiol. 145:420–425. Stals, A., L. Baert, A. De Keuckelaere, E. Van Coillie, and M. Uyttendaele. 2011. Evaluation of norovirus detection methodology for soft red fruits. Food Microbiol. 28:52–58. Stals, A., L. Baert, V. Jasson, E. Van Coillie, and M. Uyttendaele. 2011. Screening of fruit products for norovirus and the difficulty of interpreting positive PCR results. J. Food Prot. 74:425–431. Stapleton, J. T. 1995. Host immune response to hepatitis A virus. J. Infect. Dis. 171(Suppl. 1):S9–S14. Stine, S. W., I. Song, C. Y. Choi, and C. P. Gerba. 2005. Application of microbial risk assessment to the development of standards for enteric pathogens in water used to irrigate fresh produce. J. Food Prot. 68:913–918. Su, X., and D. H. D’Souza. 2011. Trisodium phosphate for foodborne virus reduction on produce. Foodborne Pathog. Dis. 8:713–717. Su, X., S. Zivanovic, and D. H. D’Souza. 2010. Inactivation of human enteric virus surrogates by high-intensity ultrasound. Foodborne Pathog. Dis. 7:1055–1061. Suslow, T. V., M. P. Oria, L. R. Beuchat, E. H. Garrett, M. E. Parish, L. J. Harris, J. N. Farber, and F. F. Busta.

Manila Typesetting Company

11/08/2012 07:06AM

Nonbacterial Pathogens

648

211. 212.

213.

214.

215. 216.

217.

218. 219.

220.

221.

222.

223.

2003. Production practices as risk factors in microbial food safety of fresh and fresh-cut produce. Comp. Rev. Food Sci. Food Safety 2(Suppl.):1–40. Svensson, L. 2000. Diagnosis of foodborne viral infections in patients. Int. J. Food Microbiol. 59:117–126. Tacket, C. O., M. B. Sztein, G. A. Losonsky, S. S. Wasserman, and M. K. Estes. 2003. Humoral, mucosal and cellular immune responses to oral Norwalk virus-like particles in volunteers. Clin. Immunol. 108:241–247. Takanashi, S., S. Hashira, T. Matsunaga, A. Yoshida, T. Shiota, P. G. Tung, P. Khamrin, S. Okitsu, M. Mizuguchi, T. Igarashi, and H. Ushijima. 2009. Detection, genetic characterization, and quantification of norovirus RNA from sera of children with gastroenteritis. J. Clin. Virol. 44:161–163. Tan, M., H. Pengwei, M. Xia, P.-A. Fang, W. Zhong, M. McNeal, C. Wei, W. Jiang, and X. Jiang. 2011. Norovirus P particle, a novel platform for vaccine development and antibody production. J. Virol. 85:753–764. Tan, M., and X. Jiang. 2010. Norovirus gastroenteritis, carbohydrate receptors, and animal models. PLoS Pathog. 6:e1000983. Tan, M., M. Jin, H. Xie, Z. Duan, X. Jiang, and Z. Fang. 2008. Outbreak studies of a GII-3 and a GII-4 norovirus revealed an association between HBGA phenotype and viral infection. J. Med. Virol. 80:1296–1301. Tan, M., R. S. Hegde, and X. Jiang. 2004. The P domain of norovirus capsid protein forms dimer and binds to histo-blood group antigen receptors. J. Virol. 78:6233–6242. Teo, C. G. 2010. Much meat, much malady: changing perceptions of the epidemiology of hepatitis E. Clin. Microbiol. Infect. 16:24–32. Teunis, P. F., C. L. Moe, P. Liu, S. Miller, L. Lindesmith, R. S. Baric, J. LePendu, and R. L. Calderon. 2008. Norwalk virus: how infectious is it? J. Med. Virol. 80:1468–1476. Thurston-Enriquez, J. A., C. N. Haas, J. Jacangelo, K. Riley, and C. P. Gerba. 2003. Inactivation of feline calicivirus and adenovirus type 40 by UV radiation. Appl. Environ. Microbiol. 69:577–582. Tian, P., D. Yang, and R. Mandrell. 2011. A simple method to recover norovirus from fresh produce with large sample size by using histo-blood group antigenconjugated to magnetic beads in a recirculating affinity magnetic separation system (RCAMS). Int. J. Food Microbiol. 147:223–227. Tian, P., A. L. Engelbrektson, X. Jiang, W. Zhong, and R. E. Mandrell. 2007. Norovirus recognizes histo-blood group antigens on gastrointestinal cells of clams, mussels, and oysters: a possible mechanism of bioaccumulation. J. Food Prot. 70:2140–2147. Topping, J. R., H. Schnerr, J. Haines, M. Scott, M. J. Carter, M. M. Willcocks, K. Bellamy, D. W. Brown, J. J. Gray, C. I. Gallimore, and A. I. Knight. 2009. Temperature inactivation of Feline calicivirus vaccine strain FCV F-9 in comparison with human noroviruses using an RNA exposure assay and reverse transcribed quantitative real-time polymerase chain reaction—a

SMP_Food Microbiology_CH24.indd

648

novel method for predicting virus infectivity. J. Virol. Methods 156:89–95. 224. Tseng, F., J. S. Leon, J. N. MacCormack, J. M. Maillard, and C. L. Moe. 2007. Molecular epidemiology of gastroenteritis outbreaks in North Carolina, United States, 1995–2000. J. Med. Virol. 79:84–91. 225. U.S. Food and Drug Administration. 2005. 2005 Food Code. U.S. Food and Drug Administration, College Park, MD. 226. U.S. Food and Drug Administration. 2009. 2009 Food Code. U.S. Food and Drug Administration, College Park, MD. 227. U.S. Food and Drug Administration. 2006. Guide to minimize microbial food safety hazards of fresh-cut fruits and vegetables. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, College Park, MD. 228. U.S. Food and Drug Administration. 1998. Guidance to industry: guide to minimize microbial food safety hazards for fruits and vegetables. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, College Park, MD. 229. Vallbracht, A., B. Fleischer, and F. W. Busch. 1993. Hepatitis A: hepatotropism and influence on myelopoiesis. Intervirology 35:133–139. 230. van Asten, L., J. Siebenga, C. van den Wijngaard, R. Verheij, H. van Vliet, M. Kretzschmar, H. Boshuizen, W. van Pelt, and M. Koopmans. 2011. Unspecified gastroenteritis illness and deaths in the elderly associated with norovirus epidemics. Epidemiology 22:336–343. 231. Venuto, M., B. Halbrook, M. Hinners, A. Lange, and S. Mickelson. 2010. Analyses of the eFORS (Electronic Foodborne Outbreak Reporting System) surveillance data (2000–2004) in school settings. J. Environ. Health 72:8–13. 232. Verhoef, L., H. Vennema, W. van Pelt, D. Lees, H. Boshuizen, K. Henshilwood, and M. Koopmans for the Foodborne Viruses in Europe Network. 2010. Use of norovirus genotype profiles to differentiate origins of foodborne outbreaks. Emerg. Infect. Dis. 16:617–624. 233. Verhoef, L., A. Kroneman, Y. van Duynhoven, H. Boshuizen, W. van Pelt, and M. Koopmans for the Foodborne Viruses in Europe Network. 2009. Selection tool for foodborne norovirus outbreaks. Emerg. Infect. Dis. 15:31–38. 234. Vinjé, J. 2010. A norovirus vaccine on the horizon? J. Infect. Dis. 202:1623–1625. 235. Vinjé, J., R. A. Hamidjaja, and M. D. Sobsey. 2004. Development and application of a capsid VP1 (region D) based reverse transcription PCR assay for genotyping of genogroup I and II noroviruses. J. Virol. Methods 116:109–117. 236. Vinjé, J., H. Vennema, L. Maunula, C. H. von Bonsdorff, M. Hoehne, E. Schreier, A. Richards, J. Green, D. Brown, S. S. Beard, S. S. Monroe, E. de Bruin, L. Svensson, and M. P. Koopmans. 2003. International collaborative study to compare reverse transcriptase PCR assays for detection and genotyping of noroviruses. J. Clin. Microbiol. 41:1423–1433.

Manila Typesetting Company

11/08/2012 07:06AM

24.  Foodborne Viral Pathogens 237. Webby, R. J., K. S. Carville, M. D. Kirk, G. Greening, R. M. Ratcliff, S. K. Crerar, K. Dempsey, M. Sarna, R. Stafford, M. Patel, and G. Hall. 2007. Internationally distributed frozen oyster meat causing multiple outbreaks of norovirus infection in Australia. Clin. Infect. Dis. 44:1026–1031. 238. Wei, J., Y. Jin, J. T. Sims, and K. E. Kniel. 2010. Fate of human enteric viruses during dairy manure-based composting. J. Food Prot. 73:1543–1547. 239. Westrell, T., V. Dusch, S. Ethelberg, J. Harris, M. Hjertqvist, N. Jourdan-da Silva, A. Koller, A. Lenglet, M. Lisby, and L. Vold. 2010. Norovirus outbreaks linked to oyster consumption in the United Kingdom, Norway, France, Sweden and Denmark, 2010. Euro Surveill. 15:pii=19524. 240. Wheeler, C., T. M. Vogt, G. L. Armstrong, G. Vaughan, A. Weltman, O. V. Nainan, V. Dato, G. Xia, K. Waller, J. Amon, T. M. Lee, A. Highbaugh-Battle, C. Hembree, S. Evenson, M. A. Ruta, I. T. Williams, A. E. Fiore, and B. P. Bell. 2005. An outbreak of hepatitis A associated with green onions. N. Engl. J. Med. 353:890–897. 241. Whitehead, K., and K. A. McCue. 2010. Virucidal efficacy of disinfectant actives against feline calicivirus, a surrogate for norovirus, in a short contact time. Am. J. Infect. Control 38:26–30. 242. Widdowson, M. A., A. Sulka, S. N. Bulens, R. S. Beard, S. S. Chaves, R. Hammond, E. D. Salehi, E. Swanson, J. Totaro, R. Woron, P. S. Mead, J. S. Bresee, S. S. Monroe, and R. I. Glass. 2005. Norovirus and foodborne disease, United States, 1991–2000. Emerg. Infect. Dis. 11:95–102. 243. Wikswo, M. E., J. Cortes, A. J. Hall, G. Vaughan, C. Howard, N. Gregoricus, and E. H. Cramer. 2011. Disease transmission and passenger behaviors during

SMP_Food Microbiology_CH24.indd

649

649

244.

245.

246. 247.

248.

249.

250.

a high morbidity norovirus outbreak on a cruise ship, January 2009. Clin. Infect. Dis. 52:1116–1122. Wobus, C. E., S. M. Karst, L. B. Thackray, K. O. Chang, S. V. Sosnovtsev, G. Belliot, A. Krug, J. M. Mackenzie, K. Y. Green, and H. W. Virgin IV. 2004. Replication of Norovirus in cell culture reveals a tropism for dendritic cells and macrophages. PLoS Biol. 2:e432. Wyatt, R. G., R. Dolin, N. R. Blacklow, H. L. DuPont, R. F. Buscho, T. S. Thornhill, A. Z. Kapikian, and R. M. Chanock. 1974. Comparison of three agents of acute infectious nonbacterial gastroenteritis by cross-challenge in volunteers. J. Infect. Dis. 129:709–714. Xi, J. N., D. Y. Graham, K. N. Wang, and M. K. Estes. 1990. Norwalk virus genome cloning and characterization. Science 14:1580–1583. Xia, M., T. Farkas, and X. Jiang. 2007. Norovirus capsid protein expressed in yeast forms virus-like particles and stimulates systematic and mucosal immunity in mice following an oral administration of raw yeast extracts. J. Med. Virol. 79:74–83. Zainazor, C. T., M. S. Hidayah, L. C. Chai, R. Tunung, F. M. Ghazali, and R. Son. 2010. The scenario of norovirus contamination in food and food handlers. J. Microbiol. Biotechnol. 20:229–237. Zhang, S., T. H. Chen, J. Wang, C. Dong, J. Pan, C. Moe, W. Chen, L. Yang, X. Wang, H. Tang, X. Li, and P. Liu. 2011. Symptomatic and asymptomatic infections of rotavirus, norovirus, and adenovirus among hospitalized children in Xi’an, China. J. Med. Virol. 83:1476–1484. Zheng, D. P., M. A. Widdowson, R. I. Glass, and J. Vinjé. 2010. Molecular epidemiology of genogroup IIgenotype 4 noroviruses in the United States between 1994 and 2006. J. Clin. Microbiol. 48:168–177.

Manila Typesetting Company

11/08/2012 07:06AM

SMP_Food Microbiology_CH24.indd

650

Manila Typesetting Company

11/08/2012 07:06AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch25

25

Paul W. Brown Linda A. Detwiler

Bovine Spongiform Encephalopathy

Bovine spongiform encephalopathy (BSE), widely known as “mad cow disease,” is a subacute degenerative disease affecting the central nervous system (CNS) of cattle. The first cases were recognized in the United Kingdom in 1986, and because of recycling of slaughtered carcass remains into animal feed, the disease rapidly became epidemic in the United Kingdom and spread to other countries both inside and outside the European Union. Worldwide, the number of cases at the end of 2011 was 190,000, all but 6,000 of which were within the United Kingdom. In addition to the officially reported and confirmed cases, it is estimated that as many as 3.5 million animals were infected and may have entered the food and feed chains without being detected (36), and it should also be noted that the absence of reported cases in a country may not indicate so much the absence of disease as a lack of adequate surveillance. BSE is a member of the family of diseases known as the transmissible spongiform encephalopathies (TSEs), which affect both animals and humans (Fig. 25.1). The animal diseases also include scrapie of sheep and goats, chronic wasting disease of cervids, and transmissible

mink encephalopathy. Human TSEs include the prototypic disease kuru, which is limited to Papua New Guinea and is now almost extinct; Creutzfeldt-Jakob disease (CJD) together with its variant form (vCJD) due to BSE infection; fatal familial insomnia; and the Gerstmann-Sträussler-Scheinker syndrome. All TSEs share many common characteristics, including: • •

•

• •

•

incubation periods ranging from years to decades illnesses of weeks to months, with invariable progression to death accumulation in the brain and other tissues of fibrillar amyloid protein aggregates (PrPTSE) pathological changes confined to the CNS the absence of any detectable agent-specific immune response transmissibility by either natural or experimental means

Although these features define the fundamental biological identity of TSEs, some important differences occur in pathogenesis, routes of natural transmission, and distribution of infectivity in tissues that must be taken into account for diagnosis, prevention, and control.

Paul W. Brown, Commissariat à l’Énergie Atomique (CEA), Service d’Étude des Prions et des Infections Atypiques (SEPIA), 92265 Fontenayaux-Roses, France. Linda A. Detwiler, Department of Pathobiology and Population Medicine, College of Veterinary Medicine, Mississippi State University, Mississippi State, MS 39762.

651

Nonbacterial Pathogens

652

CWD

1980

? TME 1965

?

?

1936

Scrapie

? BSE 1988 Kuru

1966

vCJD 1997

? ? ? ? 1900 10

CJD

1968 GSS

1981 FFI

20

30

40

50

60

70

80

1995

90 2000 10

Figure 25.1  Chronology of the TSEs. Bars represent known or presumed (striped) time periods of disease occurrence. Disease names are placed at time of first recognition (except for scrapie, first described in the 18th century). Dates within or under bars correspond to year of first experimental transmission in the laboratory; vertical arrows are placed at times of known or possible interspecies transmissions in nature. Horizontal arrows indicate probable extension of occurrence along time lines. CWD, chronic wasting disease; TME, transmissible mink encephalopathy; CJD, sporadic and familial forms of CreutzfeldtJakob disease; GSS, Gerstmann-Sträussler-Scheinker disease; FFI, fatal familial insomnia. doi:10.1128/9781555818463.ch25f1

CHARACTERISTICS OF THE BSE AGENT

Typical (Classic) BSE

The unusual features of TSE have been recognized ever since it became apparent in the 1930s that scrapie could be experimentally transmitted in sheep, with an incubation period exceeding 2 years, and that no pathogenic agent was detectable by light microscopy (30), leading to the description of the agent as a “slow” or “unconventional” virus. The mystery deepened in the 1960s when the results of radiation experiments suggested that infectivity did not depend on nucleic acids (1). Building on this heretical observation, further research in the 1980s led to the discovery that a hostencoded (rather than foreign) glycoprotein consisting of 253 amino acids and having a molecular weight of 35 kDa was closely associated with infectivity (12, 65). A quarter century of investigation following this discovery has focused on defining a mechanism by which a protein could “replicate” and so mimic the behavior of a nucleic acid-containing pathogen, in particular, how it could be compatible with the existence of different pathogenic agent strains.

The explanation is still imperfect, but a wealth of accumulating evidence has led to the conclusions that (i) a misfolded form of the protein (PrPTSE), known as a prion, acts as a template to induce normal protein molecules to cascade into the same misfolded configuration; (ii) maximal infectivity is associated with an aggregate (or polymer) of 14 to 28 misfolded protein molecules; (iii) an as yet unidentified host molecule (chaperone, ganglioside, noncoding RNA) is probably necessary as a cofactor in replication; (iv) the degree of similarity in the primary structure of the protein in different species influences the ease with which the protein can induce interspecies disease; and (v) the entire process appears to occur spontaneously in the sporadic form of disease, but can be initiated (i.e., transmitted) by the introduction of tissue from a diseased host into a healthy host, as is likely to have happened when humans consumed BSE-contaminated meat products. The agent of BSE has some further interesting features, first among which is the fact that most cases of the disease appear to be caused by a single strain that, even when experimentally transmitted to nonbovine species such as sheep and laboratory rodents, “breeds true”

25.  Bovine Spongiform Encephalopathy and is recognizable by a unique biological fingerprint. When infected tissue is subjected to electrophoresis and Western blotting (using a labeled antibody to visualize the protein), the resulting three-band pattern displays a combination of differing band intensities and migration speeds not seen in any other form of TSE (Fig. 25.2) (45). More importantly, the biological behavior of the BSE agent can be recognized by a distinctive combination of incubation period length and brain lesion distribution when inoculated into a panel of inbred mouse strains (20).

Atypical BSE

In 2004, cases of a bovine prion disease with different neuropathology and Western blot patterns than those already documented as classic BSE were described by scientists in Italy (26) and France (10) (Fig. 25.3). In both countries the cattle were over 8 years of age and had none of the clinical features of the classic disease, often simply dying as aged “downer” cattle. Immunohistochemical study of the brains revealed a large number of PrPTSE amyloid deposits throughout the cerebral cortex and thalamus, quite unlike the brainstem neuropathology of classic cases, hence the name “bovine amyloidotic spongiform encephalopathy” (BASE). The Italian cases were characterized by a Western blot protein triplet with a slightly lower molecular weight of the lowest band (giving rise to the designation of these cases as “L” atypicals). Western blot studies of the French cases revealed a slightly heavier molecular mass of the lowest band than seen in classic BSE, and these were therefore designated as “H” atypical cases. Since these two publications, more than 60 cases have been identified in other countries throughout Europe (43, 54, 90), Canada (37), and the United States (75).

CJD

S

S

V

Codon 129 MM VV MM

21kDa

19kDa

Figure 25.2  Western blot patterns of PrPTSE in sporadic (S) and variant (V) CJD brain tissue homogenates treated with proteinase K. MM, methionine homozygote; VV, valine homozygote. The vCJD pattern is distinct from both codon 129 genotypes of sCJD. Courtesy of Mark Head, CJD Surveil­ lance Unit, Western General Hospital, Edinburgh, Scotland. doi:10.1128/9781555818463.ch25f2

653

BSE

H

C

L

21kDa PNGase: -

19kDa

-

-

+

+

+

Figure 25.3  Western blot patterns of PrPTSE in typical classic (C) and in H and L types of atypical BSE brain tissue homogenates treated with proteinase K. In this blot the slight differences in the migration of the lowest band are more easily appreciated after treatment with N-glycosidase (PNGase). Courtesy of Gianluigi Zanusso, Department of Neurosciences, University of Verona, Verona, Italy. doi:10.1128/9781555818463.ch25f3

One of the most important and still unanswered questions about atypical BSE concerns its significance with respect to human and animal health. Atypical BSE has been experimentally transmitted to wild-type and transgenic mice (21, 55), nonhuman primates (28), and cattle (7). Results of inoculation studies using humanized transgenic mice and nonhuman primates indicate that L-type BSE may have a higher zoonotic potential than classic BSE (9, 66), and very recent bioassays of muscle tissue in transgenic mice that overexpress bovine PrP have revealed transmission rates of 10 to 90% (82). An unexpected observation that atypical BSE transforms into classic BSE during passage in transgenic mice has led to speculation about whether atypical BSE was the original strain of bovine disease that upon passage in cattle changed into the typical form (29). The origin and natural routes of transmission, if any, have yet to be determined. Almost all cases have been in older cattle (usually more than 8 years of age) and have shown little resemblance to the ­ clinicalpathological picture seen in classic disease. It has been suggested that the disease may be sporadic or be caused by a genetic mutation (74), but there is no convincing evidence to support either of these ideas. The correct answer will probably only come by study of the future annual incidence curves of both types of disease (19). Regardless of the origin of atypical BSE, the combination of ambiguous clinical diagnosis and experimental transmissibility­ mandates a continuation of feed and specific risk materials (SRMs) bans, together with diagnostic testing programs, for some time to come.

Nonbacterial Pathogens

654

STABILITY OF THE INFECTIOUS AGENT The agent of BSE shares with other TSE agents the property of unusual resistance to destruction (Table 25.1). None of the standard disinfection methods is effective, including irradiation or exposure to various chemical disinfectants. Even harsher conditions that are capable of inactivating all other known pathogens (including bacterial spores), such as heating under pressure at 121°C, exposure to dry heat at 600°C, or immersion in 0.1 N NaOH or 0.5% bleach, cannot ensure complete inactivation (15–17, 84, 87). Better procedures are exposure to dry heat at 1,000°C (15), immersion in either 1 N NaOH or fresh undiluted bleach (16), and steam heating under pressure at 132°C (17), and the best method is sequential exposure to both NaOH and steam-autoclaving inactivation treatments (83, 85, 109). It is evident that foodstuffs cannot be subjected to these conditions and remain organoleptically acceptable; hence, the problem of processing food, including beef that might contain the BSE agent, would appear to be insoluble. However, exposure of beef or beef products to high temperature (132°C) under ultrahigh pressure (≥690 mPa = 1,000 atm) retains beef quality and reduces the very low level of infectivity that could realistically be present in commercial products to undetectable or nontransmissible levels (Tables 25.2 and 25.3) (18, 24).

In addition to the contamination of food with BSE, it is likely that infectious particles enter the environment, where because of their extraordinary durability they may persist for many years (14, 62, 79). Wastewater and tissue remnants from commercial slaughterhouses, custom processors, necropsy laboratories, etc., may be sent to municipal sewage treatment plants for disposal. One study has revealed that if prions were to enter these municipal treatment systems, most of the agent would partition to activated sludge solids, survive mesophilic anaerobic digestion, and be present in treated biosolids (47). Additional research is needed to determine the real risk from this potential route of exposure.

BODILY DISTRIBUTION OF INFECTIVITY The distribution of infectivity (and PrPTSE) in bovine tissues must be known to determine the level of risk from human exposure to BSE. As of 2010, knowledge about the bodily distribution of infectivity and PrPTSE in BSE and other members of the TSE family of diseases is summarized in a set of tables available online at the World Health Organization website, including an extensive bibliography of primary sources (110). Much research has gone into defining the pathogenesis of BSE, the most important of which is studies in

Table 25.1  Effectiveness of various chemical and physical inactivation treatments Ineffective Chemical treatmentsa Alcohol Ammonia b-Propiolactone (0.2%) Detergents Ethylene oxide Formaldehyde Hydrochloric acid Liquid H2O2 Peracetic acid Permanganate Phenol (5–10%)e Physical treatmentsa Boiling (100°C) Microwave radiation UV radiation

Partially effective

Most effective

NaOCl (0.5%) NaOH (0.1 N) KOH (0.6 N) + detergent Chlorine dioxide Glutaraldehyde Iodophores Guanidine thiocyanate (4 M) Sodium dichloroisocyanurate Sodium metaperiodate Urea (6–8 M) Environ LpHf

NaOCl (2%)b NaOH (1 N) NaOH (0.75 N) + SDSc Formic acid (100%) Gaseous H2O2 (2 mg/liter)d Phenol (90%)

Steam heat (121°C) Dry heat (300°C) Ionizing radiation (³50 kG) UV radiation/TiO2 (2–12 h)d Dry heat (121°C) at 1,000 MPa (5 min)

Steam heat (³133°C) Dry heat (>600°C)

Exposure times from 15 to 60 min except as noted. Commercial bleach is a 5.25% NaOCl solution (2% = 20,000 ppm). c SDS, sodium dodecyl sulfate. d Results based on single studies. e The concentration range of most commercial phenol-based disinfectants. f A mixture of o-benzyl-p-chlorophenol (6.4%), p-tertiary-amylphenol (3%), and o-phenylphenol (0.5%). a

b

25.  Bovine Spongiform Encephalopathy Table 25.2  Proteinase-resistant protein (PrPTSE) and infecti­

vity reductions under various pressure, temperature, and exposure time conditionsa Pressure (MPa)b

Temperature Cooking time Log10 PrPTSE Log LD50 (°C) (min) reduction reduction

690 690 1,000 1,000 1,200 1,200

125 120 135 135 135 135

3 10 3 10 3 10

1.5 1.5 ≥3.0 ≥3.0 ≥3.5 ≥3.5

2.8 3.0 3.8 5.7 5.8 5.6

a A scrapie agent (263K)-infected hamster brain was homogenized in hot dog. The pretreatment (input) infectivity was 8 log LD50/g. b 690 MPa = 100,000 lb/in2.

which cattle were infected orally (98, 100, 101) and their tissues were examined at various times thereafter, up to and including the fully developed stage of illness. More recent bioassay studies conducted in bovinized transgenic mice and tissue PrPTSE detection studies in various experimental hosts, including primates (44), have completed the picture, but are subject to a number of caveats. Many of the experimental studies have used intracerebral (i.c.) inoculation, which does not necessarily represent the situation in orally acquired bovine disease. The fact that pigs can be infected by small i.c. doses but not by massive oral doses illustrates this point (99). Even those studies that have used oral dosing of nonbo­ vine species cannot be assumed to produce­ results comparable to those involving oral dosing of cattle under natural conditions. Hence, it is necessary to use caution in the interpretation of bioassay results from all nonbovine species, particularly if the results are

655 based on ultrasensitive prion amplification techniques or bioassays in transgenic mice overexpressing the PrP gene. Similar caution is needed for the interpretation of immunohistological detection of PrPTSE in tissues, because it may be present in tissues in the absence of transmissibility (73) or, conversely, it may be absent in tissues that are infectious (2, 6). The issue is especially relevant to recent studies in which PrPTSE has been detected in muscle tissue of various animal species infected either naturally or experimentally with various strains of TSE (3, 4, 91, 95). Only a few such studies have included parallel bioassays, and these have usually been performed by i.c. inoculation of transgenic mice. Notwithstanding these caveats, an overall picture of BSE pathogenesis has evolved that is consistent with pathogenesis in other members of the TSE family, and preliminary work with atypical BSE has yielded results similar to the far more extensively studied classic form of disease (7) except for the recent finding of a much higher frequency of infectivity in muscle tissue of cows with atypical BSE (82). In oral infections, the pathogenic agent first enters the gut, where it begins to multiply in Peyer’s patches, and then moves to the CNS following the parasympathetic and sympathetic nerve fibers of the autonomic nervous system (48). The distribution of infectivity in cattle during the preclinical stage of BSE is primarily restricted to the intestines, tonsils, lymphoid tissue of the third eyelid, and CNS. As the disease progresses, the agent replicates to high levels in the CNS and spreads centrifugally to the peripheral nervous system, invading the vagus,

Table 25.3  Misfolded-protein (PrPTSE) reductions in various meat products containing 263K scrapie

agent-infected hamster braina Food product Hot dog Hamburger Corned beef Beef paté Baby food Cat food

Time relative to treatment Before After Before After Before After Before After Before After Before After

Weight equivalent of brain (µg) in loaded sampleb 330

100

33

10

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

– + – + – + – + –

3.3

1

0.3

0.1

+ – + – + – + – + – + –

+

±

–

+

+

–

+

+

–

±

–

±

–

+

±

–

Cooked for 5 min at 130°C under 690 MPa of pressure. The weight of the loaded sample is inversely related to the concentration of PrPTSE. No PrPTSE was detected in any of the treated samples, and hence all samples showed at least a 2.5- to 3-log reduction of PrPTSE after cooking under high pressure. a

b

Nonbacterial Pathogens

656 facial and sciatic nerves, nasal mucosa, tongue, adrenal gland, and muscle (7, 22, 39, 48, 53, 60). In the fully developed disease, the CNS and its associated tissues (e.g., retinal and ganglionic tissues) invariably contain high levels of infectivity, usually in the range of 104 to 106 mean lethal doses per gram of tissue (LD50/g). The lymphoreticular system (including tonsils and lymphoid tissue of the third eyelid and distal ileum) contains much lower levels of infectivity, in the range of 101 to 102 LD50/g (100, 101; G. A. H. Wells, personal communication). Peripheral infectivity in extraintestinal tissues is only detectable at the clinical stage of disease (22). A very recent study has revealed infectivity not only in the distal ileum, but also in the ileocecal junction and jejunum (6). This is significant because (unlike Europe) Canada and the United States limit the definition of SRMs to the distal ileum of the small intestine, hence raising the possibility of infectivity entering the food and animal feed chains via other intestinal locations. Bovine tissues can enter the human food chain in many ways, both obvious and subtle (Table 25.4). Examples of the obvious are two staples of the human diet, muscle (meat) and milk. PrPTSE and/or infectivity has been detected in muscle tissue of humans with vCJD (95) and of animals with either natural or experimental infections with scrapie (3), chronic wasting disease of deer (4), or even rodent-adapted BSE (25, 91). The original extensive cattle bioassay study of many different tissues failed to find infectivity in muscle at any of several time points after experimental oral exposure (100, 101); a more recent study of tissues from symptomatic cattle revealed infectivity in semitendinosus muscle (1 of 10 animals), but not in latissimus dorsi muscle (0 of 13 animals) (22). Evidence that infectivity is not associated with milk includes temporospatial epidemiologic observations failing to detect evidence of significant maternal transmission; clinical observations of calves nursed by infected cows, where the calves have not developed BSE (35, 103); and experimental observations that milk from infected cows does not transmit disease when administered i.c. or orally to mice (61, 86). However, similarly negative early observations made in scrapie-infected sheep have been contradicted by recent research in which a proportion of nursing lambs were found to have contracted scrapie (56, 57, 59). Examples of less obvious bovine tissues consumed by humans are tallow and its derivatives that are used in many different food and drug products such as emulsifiers and binders. Like muscle and milk, tallow does not appear to pose a risk to human health.

Table 25.4  Consumable bovine materials used by humans Used directly (or after minimal processing) Meat on the bone (T-bone, ox tail) Deboned meat Offalsa (e.g., liver, lung, heart, kidney, thymus, and brain) Fat (suet) Bone (soup and broth) Brain and endocrine powders in some unregulated “health food” supplements Used after processing Milk and milk products (e.g., butter, lactose, and casein) Rennet (chymosin and pepsin derived from abomasum) used in the production of cheese, whey, and whey products such as lactose Bone and skin to make gelatin, gelatin derivatives, and collagen Meat, including tongue, meat extracts used as flavoring, and mechanically recovered meatb Tripe (forestomachs) Tallowc (used for frying, in food as shortening, and to make tallow derivatives) Fat (suet), beef drippings Blood and blood products (e.g., hemoglobin to clarify wine) Intestines (duodenum to rectum for natural sausage casings) a SRMs including offals as well as skull, ganglia, eyes, tonsils, and spinal cord in most countries are compulsorily removed for destruction. b Production of mechanically recovered meat from ruminant animals is banned in many countries. Where used, its inclusion in meat products usually does not exceed 5 to 10% by weight. c Tallow (bovine rendered fat) is one of the two major end products from rendering bovine carcass waste. Tallow derivatives include glycerol, fatty acids and their esters, stearates, polysorbates, and sorbitan esters. They are produced from tallow by hydrolysis at temperatures and pressures that inactivate TSE infectivity.

Experimental studies of rendered bovine carcasses spiked with BSE-infected brain tissue have not detected any infectivity in the tallow fraction (88, 89). Should a small amount of infectivity ever be found in tallow, its processing into derivatives includes exposure to NaOH and steam heat under conditions that should inactivate the BSE agent.

FEED-BORNE AND FOODBORNE OUTBREAKS

Cattle

BSE probably arose sometime in the early 1980s, and although the species of origin of BSE is unknown, there is ample epidemiologic evidence to indicate that once cattle were infected, the epidemic was perpetuated by the feeding of BSE-contaminated meat and bone meal (MBM) to cattle (105, 106). It was initially thought that BSE was the result of an increase in the amount of scrapie infectivity entering the

25.  Bovine Spongiform Encephalopathy

657

animal feed system coincident with a change in the rendering process in the United Kingdom. This hypothesis has the merit of explaining both the timing and geographic location of the epidemic. However, two observations have challenged this assumption; specifically, neither the classic nor either type of atypical BSE strain is identical to any known scrapie strain (20), and the disease produced in cattle that are experimentally infected with scrapie does not resemble naturally occurring BSE (31, 32). Hence, although the origin of BSE may never be identified with certainty, the evidence is at least partially consistent with any of the three following sources (27, 51, 72):

700 new cases reported each week at the peak of the epidemic in the United Kingdom during 1992–1993, whereas there were fewer than 200 cases during the entire year of 2005 and only 11 cases during 2010. The first cases outside the United Kingdom were identified only 2 years after those in the United Kingdom, and it will never be known if the outbreaks in other countries were due to infective tissue (dead or alive) imported from the United Kingdom or from simultaneous endogenous miniepidemics of BSE due to widespread similar changes in the rendering of slaughtered cattle. However, the peak of the epidemic in the EU occurred nearly 10 years after its peak in the United Kingdom, which is consistent with the time required to generate “virgin soil” outbreaks from imported feed, as well as from recycled infected carcasses of imported cattle. Cases of BSE that continued to emerge after the 1988 feed ban in the United Kingdom (and after the initial feed bans in the European Union) were likely due to cross-feeding and cross-contamination on farms (34, 49, 50, 81). Rendering may reduce but does not eliminate infectivity (78, 88, 89). Given that BSE can be orally transmitted to cattle with as little as 1 mg of infected tissue (98), a very low degree of contamination would be sufficient to recycle the disease. This was especially true in countries that do not have dedicated lines and

•

•

•

sheep or goats with an uncharacterized scrapie strain or a strain that was modified in the course of its adaptation to cattle cattle that developed TSE spontaneously or from a somatic or germ line mutation other species such as wild ungulates infected with TSE that entered into the feed system

As a consequence of recycled ruminant tissue’s having caused the BSE epidemic, changes in feeding practices have led to a precipitous decline in BSE, which has now all but disappeared from even the most heavily burdened countries, as revealed by the annual mortality curves for the United Kingdom and European Union (EU), shown in Fig. 25.4 (67). There were nearly

BSE in the UK and EC

4000 3500

Number of cases

3000

UK (X 10) EC

2500 2000 1500 1000 500 0

86

89

92

95

98

01

04

07

10

Year of onset

Figure 25.4  The chronology of BSE in the United Kingdom (black) and EU (gray). Numbers of United Kingdom cases are 10× greater than scale. doi:10.1128/9781555818463.ch25f4

Nonbacterial Pathogens

658 e­ quipment to manufacture and process feed for ruminants and nonruminants. Cross-feeding is the practice of feeding meal for poultry or pigs or pet food (which had legally contained ruminant MBM) to cattle on the same farm, usually due to simple human error or negligence. The continuance of BSE cases due to cross-contamination and cross-feeding of ruminant MBM has resulted in extended feed bans that have prohibited feeding of all mammalian or animal MBM back to any animals used for human food. All evidence indicates that BSE does not spread horizontally among cattle. The question of maternal transmission between an infected cow and her calf remains unanswered, and although it appears that this risk is small to nonexistent, the possibility has not been entirely eliminated. Nevertheless, international recommendations requiring the removal of offspring of an infected dam from the implicated herd have recently been withdrawn (50, 68, 104).

France, the largest importer of calves and MBM from the United Kingdom (64). A few individuals living in other countries (Canada, the United States, Saudi Arabia, Japan, and Taiwan) had, with one exception (Saudi Arabia), all visited or lived in the United Kingdom for periods of months to years in the decade before their illness began. Humans most likely became infected with the agent that causes BSE through the consumption of beef products contaminated by CNS tissue, such as mechanically recovered meat that was pressure-extracted from carcasses and often contained spinal cord and paraspinal ganglia in addition to residual muscle shards (19, 96). One puzzling question about vCJD is the limited number of actual cases in a United Kingdom population heavily exposed to BSE: based on statistical analyses of tissue surveys conducted on postoperative appendices, it is estimated that many thousands of silent infections may have occurred (46, 93). A related question concerns the surprising fact that there is only a single documented instance of more than one case within a household, where the same foods are shared (76). The most plausible explanation is that small “packets” of low-level infectivity were heterogeneously scattered through batches of mechanically recovered meat used in the production of meat products, and in consequence caused very few illnesses that were correspondingly rare and randomized.

Humans

Human infection with BSE results in a variant form of CJD (vCJD) that was first recognized in the United Kingdom in 1996, approximately 10 years after the BSE epidemic began (107) (Fig. 25.5). By 2012, the number of primary cases had increased to 173, and 43 additional cases had been identified in other European Union countries, nearly two-thirds of which were in

BSE and vCJD in the UK

40 35

Number of cases

30

BSE (X 1000) vCJD

25 20 15 10 5 0

86

89

92

95

98

01

04

07

10

Year of onset

Figure 25.5  The chronology of BSE (black) and vCJD (gray) in the United Kingdom. Numbers of BSE cases are 1,000× greater than scale. doi:10.1128/9781555818463.ch25f5

25.  Bovine Spongiform Encephalopathy

659

The period of maximum exposure to BSE in the United Kingdom occurred in the mid- to late 1980s before the ban on high-risk tissues was instituted, and the delay in the peak occurrence of vCJD clearly delineates an average incubation period of about 12 to 15 years. The delay in the appearance of cases outside the United Kingdom reflects the delay in exposure to BSE in the EU (Fig. 25.6). Also, because all foodborne cases have so far been homozygous for methionine at codon 129 of the prion gene, it is still possible that individuals with alternative genotypes may have a significantly longer incubation period that will generate a “second wave” of cases that has not yet begun. Provisional support for this possibility comes from an ongoing United Kingdom immunohistological study of archived postoperative appendices that has so far yielded a total of 16 PrPTSE-positive samples in a survey of 32,000 tests, of which at least 2 samples had codon 129 genotypes other than homozygous methionine (11, 52, 93). In four instances, patients with orally acquired disease have transmitted infection via blood transfusions, causing grave concern in the blood donor and recipient communities (58, 70, 92). Most countries have programs of blood donor deferrals of individuals whose histories include specified lengths of residence in the United Kingdom

(and in some cases other European countries) or who have received blood or blood products from individuals in the United Kingdom. Plasma products are considered to pose a lower risk, because experimental evidence indicates large reductions of infectivity by one or more of the processing steps used in their manufacture (40); however, one possible instance of infection in a hemophilia patient treated with factor VIII has been reported (71).

CHARACTERISTICS OF DISEASE

Cattle

Affected animals develop a progressive degeneration of the nervous system. They may display changes in temperament, abnormalities of posture and movement, and changes in sensation, including signs of apprehension; nervousness or aggression; incoordination, especially hind-limb ataxia; tremor and difficulty in rising; and hyperesthesia to sound and touch. In addition, many animals have decreased milk production and loss of body condition despite continued appetite. Only a small proportion of affected cattle exhibit what would be considered typical “mad cow” signs. Many suspects have several, but not all, of the signs described above if they are closely observed.

vCJD in the UK and EC

40 35

Number of cases

30

UK EC

25 20 15 10 5 0

86

89

92

95

98

01

04

07

10

Year of onset Figure 25.6  The chronology of vCJD in the United Kingdom (black) and EU (gray). doi:10.1128/9781555818463.ch25f6

Nonbacterial Pathogens

660 BSE can be mistaken for other conditions or go unnoticed due to the subtlety of the signs. Neurologic, metabolic, or other kinds of disease that affect coordination and gait often predispose an animal to injuries such as broken limbs or soft tissue damage. If the animal then becomes recumbent because of a broken leg or torn ligament, the injury may be the prominent or sole presenting sign, and without a complete diagnostic workup and history of disease progression, the true underlying BSE cause of the nonambulatory condition may be overlooked. Even more troubling is the occurrence of a significant number of atypical BSE cases in older, recumbent cows in the absence of any preceding abnormalities. It is thought that an animal usually becomes infected within the first year of life. The average incubation period of natural BSE is estimated to range from 2 to 8 years. Following the onset of clinical signs, the animal’s condition gradually deteriorates until the animal ­becomes recumbent, dies, or is destroyed. The clinical progression of BSE may last from 2 weeks to 6 months. Most cases in the United Kingdom have occurred in dairy cows between 3 and 6 years of age, with the youngest confirmed victim being 20 months of age and the oldest over 22 years of age (102).

Humans

Variant CJD (vCJD) has certain clinical and neuropathological features that set it apart from other forms of CJD (19, 108). The most distinctive feature is the young age at onset of illness (Fig. 25.7) with many adolescents afflicted, but only occasional cases in adults

older than 50 years of age. This pattern contrasts with the peak occurrence of sporadic CJD in patients between 50 and 70 years of age. The clinical presentation is usually characterized by some form of psychiatric disturbance, such as depression or anxiety, and complaints of sensory symptoms, particularly limb pain. As the illness progresses, however, the clinical distinction between the sporadic and variant forms of illness becomes progressively blurred. The combination of psychiatric and sensory symptoms in an adolescent or young adult is sufficient to raise a suspicion of vCJD in patients who reside or have resided in countries in which BSE has occurred. The average duration of illness is 14 months, or about twice as long as that of sporadic CJD (Fig. 25.8). Neuropathological examination, without which a definitive diagnosis cannot be made, reveals diffuse spongiform changes that are especially severe in the basal ganglia, posterior thalamus, and cerebellum, together with a feature unique to vCJD: myriad amyloid plaques surrounded by halos of vacuolation, the so-called florid or daisy plaques (Fig. 25.9).

GOVERNMENT REGULATORY MEASURES Bovine products and by-products are widely used for both food and pharmaceuticals and hence require the highest level of safety. Because of the hardy nature of the BSE agent and its high potential for cross-contamination, the most effective approach to protect bovine products and bovine-derived materials from contamination by BSE is to ensure that infected animals or carcasses never enter

250

Number of cases

200

sCJD (n = 607) vCJD (n = 159)

150 100 50 0

10-19

20-29

30-39

40-49 50-59 60-69 Age at onset of illness

70-79

80-89

Figure 25.7  Age at onset of cases of variant (black) and sporadic (gray) CJD in the United Kingdom, 1994 to 2005. Courtesy of Robert Will, CJD Surveillance Unit, Western General Hospital, Edinburgh, Scotland. doi:10.1128/9781555818463.ch25f7

25.  Bovine Spongiform Encephalopathy

661

120

Number of cases

100 80

Median sCJD = 4 mos sCJD (n = 607) vCJD (n = 159)

60 40

Median vCJD = 14 mos

20 0

1 2 3 4 5 6 7 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 Duration of illness (months)

Figure 25.8  Duration of illness of cases of variant (black) and sporadic (gray) CJD in the United Kingdom, 1994 to 2005. Courtesy of Robert Will, CJD Surveillance Unit, Western General Hospital, Edinburgh, Scotland. doi:10.1128/9781555818463.ch25f8

­ rocessing plants. Because there are presently no diagnostic p tools sensitive enough for detection­of the disease during its long preclinical incubation, governments must rely on measures to prevent exposure through feed (Table 25.5). These measures may be applied to the national herd in a BSE-free country or to the next generation of cattle in a country that has already identified the disease. In a country in which BSE has been identified in cattle or in which there has been substantial exposure, specific measures must also be taken to protect human health.

Protection of Animal Health

Countries should conduct a risk assessment to evaluate possible exposures from both external and internal sources of BSE and for potential recycling of the agent within the cattle production system. A country with no known exposure to BSE can best protect its national herd by prohibiting the importation of ruminant MBM, including MBM from other species if there is any possibility of cross-contamination from BSE, and by prohibiting the importation

Figure 25.9  The pathognomonic vCJD “daisy plaque” consisting of a core of amyloid protein surrounded by vacuolar “petals.” Courtesy of James Ironside, CJD Surveillance Unit, Western General Hospital, Edinburgh, Scotland. doi:10.1128/9781555818463.ch25f9

Nonbacterial Pathogens

662 Table 25.5  Principal governmental measures taken to protect human and animal health Precautionsa Ban on ruminant protein in ruminant feed Ban on importation of live ruminants and most ruminant products from any country reporting BSE Ban on export of UK cattle born before July 1988 feed ban Ban on export of UK cattle >6 months of age BSE passive surveillance initiated Ban on SBOc in animal and human nutrition Ban on export of SBO and feed containing SBO to EU countries High-risk waste to be rendered at 133°C/3 bar/20 min Ban on export from UK of SBO and feed containing SBO to non-EU countries FDA initiates series of guidelines to insure freedom from BSE contamination of source materials for numerous products used by humans, including gelatin, cosmetics, vaccines, and blood and tissue donations Ban on mammalian MBMd in ruminant feed Ban on mammalian protein in ruminant feede Rendering methods must sterilize BSE SBO ban broadened to include the boney skull MRMf from bovine vertebral column banned and export prohibited Removal of lymph nodes and visible nervous tissue from UK bovine meat >30 months exported to EU Ban on export of all UK cattle and cattle products except milk Slaughtered/dead cattle >30 months (except certain beef cattle >42 months) ruled unfit for any use (hides for leather excluded) Mammalian MBM prohibited from all animal feed/fertilizer Mammalian MBM and MBM-containing feed recalled Mammalian waste (except fat) to be rendered at 133°C/3 bar/20 min BSE cohort cattle in UK ordered slaughtered and destroyed Ban on most mammalian protein to ruminants Ban on import of live ruminants and most ruminant products from all European countries Replace human plasma and plasma products for use in the UK with imported sources Slaughter and destruction of offspring born to BSE-affected UK cattle after July 1996 Leukodepletion of whole blood donations from UK residents Ban on cattle and sheep SRMg throughout the EU Ban on MRM production from any part of cattle, sheep, and goats Ban on mammalian protein in all livestock feed Ban on slaughter techniques that could contaminate cattle carcasses with brain emboli (e.g., pithing or pneumatic stun guns) Immunologic brain examination on all slaughtered cattle >30 months of age All “downers” destroyed; SRMs banned from human food Ban on some SRMs from all animal feed; brains and spinal cords from all cattle aged ³30 months Mandatory testing of cattle at slaughter: age raised to 72 months

Great Britainb

European Unionb

United States

July 1988 July 1989 July 1989 Mar 1990 May 1990 Sept 1989/90 Sept 1990 Nov 1990 July 1991 Dec 1993 through Nov 1999 June 1994 Nov 1994 Jan 1995 Aug 1995 Dec 1995 Jan 1996 Mar 1996 Mar 1996 Mar/Apr 1996 June 1996 July 1996 Jan 1997 Aug 1997 Dec 1997 Aug 1998 Jan 1999 July/Nov 1999 Jul 2000 Jan 2001 Jan 2001 Jan 2001 Jan 2001 Jan 2004 April 2009 July 2011

July 2011

The most important measures are shown in boldface type.

a 

In Northern Ireland and Scotland, dates of implementation were sometimes different from those shown for England and Wales; also, individual EU countries often

b 

adopted different measures on different dates. SBO, specified bovine offals (brain, spinal cord, thymus, tonsil, spleen, and intestines from cattle >6 months).

c 

MBM, meat-and-bone meal (high-protein residue produced by rendering).

d 

Some exemptions, e.g., milk, blood, and gelatin.

e  f

MRM, mechanically recovered meat (residual meat derived from bones, including vertebral column with dorsal root ganglia and possibly spinal cord in situ). SRM, specified risk materials (all tissues shown to be infectious in cattle, sheep, or goats; where infectivity is limited to animals over a certain age, the ban applies

g 

to animals over that age. The definition of SRM changes as new information is acquired).

25.  Bovine Spongiform Encephalopathy

663

of cattle from countries with BSE or with high risk ­factors for BSE. Imported live cattle pose a risk if they are eventually slaughtered, rendered, and incorporated into MBM. Given that the primary, if not sole, route of BSE transmission is through the feeding of contaminated MBM to cattle, countries with any risk factors need to implement feed controls. The level of restriction is usually dependent upon the amount of contamination thought to be in the system. There are three main factors that can increase the stability of a national feed production system:

also included additional tissues that are known to be infectious in scrapie, even though they have not been shown to be infectious in BSE-affected cattle. Processing can increase the BSE risk in edible products via cross-contamination, especially considering that TSE agents tend to adhere to surfaces and are unusually durable. Standard measures that are used in slaughter plants to reduce the level of microbial contamination, e.g., dipping in 82.2°C water, do not inactivate the BSE agent. The time, temperature, and chemical treatments that would reduce levels of the TSE agent are extremely caustic and could corrode foodprocessing equipment and adversely affect food quality. However, certain less onerous practices can at least reduce the risk of cross-contamination, including:

1. Feed bans. These regulations can range from the basic prohibition of feeding ruminant MBM back to ruminants to prohibiting most animal proteins from being fed to all animals used for food production, including fish. 2. SRMs ban. This ban requires that high-infectivitylevel tissues such as bovine brain and spinal cord be removed from both the food and feed chains and be destroyed. The intent of this control is to remove the primary source of infectivity from the entire system to prevent the possibility of cross-contamination. Canada and, to a lesser extent, the United States have each implemented a ban requiring the removal of certain SRMs from animal feed. This was done in response to the finding that most Canadian BSE victims were born after the 1997 feed ban (23, 94). 3. Regulation of rendering. Although no rendering process can completely remove all detectable infectivity, some are more effective than others. The best procedure identified to date requires a 20-minute autoclave exposure at 133°C under 3 bars of pressure (78). Experience in countries that have spent considerable effort to eliminate BSE has underlined the need for an extremely high level of compliance with feed controls in order to remove the agent from the system and prevent new infections in cattle. There can be no complacency.

Protection of Human Health

Standard cooking temperatures do not inactivate the BSE agent, and there is no screening test to guarantee that an infected animal would not enter a processing plant. Therefore, the primary public health protection measure is to remove SRMs from the food supply and to mandate procedures to prevent the possibility of crosscontamination between SRMs and edible tissue. The basic list of SRMs includes brain, spinal cord, trigeminal ganglia, dorsal root ganglia, eye, skull, vertebral column, intestine, and tonsil. Some countries have

•

•

•

The use of dedicated equipment (e.g., knives, hooks, steels, etc.) for SRM removal. Such equipment should never be used for edible tissues. Color coding of such equipment increases awareness and reduces the opportunity for misuse (Fig. 25.10). Removal of SRMs from the slaughter floor. This prevents contamination of the area where fabrication occurs and the edible product that could be contaminated, and also provides an opportunity for quality and safety checks to ensure that the SRMs have been removed prior to further processing. Employee training on proper removal of SRMs and the significance of the procedure.

Although testing will not guarantee identification of all infected animals, many countries have used testing at slaughter to reduce the amount of infectivity in the system by eliminating the carcasses of animals that test positive. In addition, carcasses that may have been contaminated by close association with the positive carcass can also be eliminated. Of particular importance is the need to include testing of animals that have passed beyond the usual age range for classic BSE, because atypical forms of BSE that cannot be clinically suspected occur predominantly in older cattle. In 1996, the sale of beef for human consumption from most cattle over the age of 30 months was prohibited in the United Kingdom. This was done because studies revealed that the levels of infectivity in SRM tissue were very high after this age. The decline of BSE cases in the United Kingdom epidemic in combination with greater practicality of BSE testing has led to a recent regulation change to allow the use of such animals if they test negative for BSE. The tenacious nature of the agent makes the complete elimination of risk in countries with BSE extremely difficult. Hence, it is imperative for the government and

Nonbacterial Pathogens

664

Figure 25.10  Red “RM” identification labels are used to differentiate between slaughterhouse tools used to handle SRMs and those used for edible products. Courtesy of Ana Carolina Alonso Simplicio de Oliveira, Frigoalta, Brazil. doi:10.1128/9781555818463.ch25f10

the industry to reduce the avenues for contamination of food or pharmaceutical products by having multiple effectively implemented safeguards in place. Given the state of knowledge about atypical BSE, the current feed regulations should limit spread to cattle, sheep, goats, and other ruminants. Likewise, SRM bans should provide protection to the public. If evidence indicates that BSE is a spontaneously occurring and/or a genetic disease, no country in the world with cattle would be able to claim freedom. It would then be prudent for all countries to prohibit the practice of recycling ruminant by-products as feed for other ruminants.

METHODS OF DETECTION

Postmortem

Historically, the diagnosis of BSE relied on the occurrence of clinical signs of the disease confirmed by postmortem histopathological examination of brain tissue (97). A diagnosis could also be made by electron microscopy detection of fibrils in denatured brain extracts, called scrapie-associated fibrils because they were first observed in the brains of scrapie-infected sheep. In the 1990s, the development of tests to detect the pathognomonic PrPTSE misfolded protein greatly enhanced the diagnostic capabilities for BSE and other

TSEs, both because of the tests’ improved sensitivity and the fact that they could be used on frozen or partially autolyzed tissue. Two types of tests for the detection of PrPTSE have been internationally approved for the confirmatory diagnosis of BSE: immunohistochemistry and Western blotting. In addition, a number of rapid immunoassays have been developed and approved by governments for use as screening tests, with positive results subjected to confirmatory Western blotting.

Premortem

There is at present no test that has been rigorously and reproducibly proven capable of detecting BSE (or any other TSE) during the preclinical phase of disease. By the turn of the century, a widespread effort to develop such a test led to a number of different strategies (13), of which the most interesting were (i) the use of antibodies that are specific for the misfolded form of the protein and thus eliminate background “noise” from the presence of normal protein; (ii) polypeptide or synthetic ligands that take advantage of exposed epitopes unique to the misfolded form of the protein; and (iii) in vitro amplification of the protein to vastly increase test sensitivity. Although most of the tests ran into specificity problems in the transition from brain extract to serum samples that defeated a timely resolution, the amplification technique has very recently been refined

25.  Bovine Spongiform Encephalopathy

665

into a high-sample-throughput format (quaking-induced conversion, or QuIC, test), with a sensitivity in the femtogram range of PrPTSE, and is likely to outstrip all other approaches to a practical screening test (5, 69). In fact, its sensitivity may have surpassed the threshold for disease transmission, an issue that will require parallel bioassays to resolve. Several laboratories using the method have reported promising results in a comparatively limited number of tests to detect scrapie in sheep, BSE in cattle, and vCJD in humans, and data are accumulating for the detection of protein in the blood of presymptomatic animals, i.e., a truly practical screening test for infection. However, none of the tests has yet been validated in a large, standardized, coded panel of blood samples.

lion (95% confidence interval, 1 per 100,000 to 3 per million) if only the two indigenously infected positives were counted, and a prevalence of 4 per million (95% confidence interval, 1 per 100,000 to 0.8 per million) if the imported cow was included (which from the standpoint of both public health and U.S. trading partners would be the more important analysis). The Enhanced BSE Surveillance Program was terminated in 2006, since which time the U.S. Department of Agriculture’s Animal and Plant Health Inspection Service testing has been limited to approximately 40,000 animals each year.

BSE PREVALENCE IN THE UNITED STATES Four cows with BSE have been identified in the United States through a testing program initiated in 1990 that targeted cattle with neurologic symptoms, were recumbent (downers), or died for unexplained reasons. The first cow, detected in December 2003 with classic BSE, had been imported several years earlier from Canada. The second cow was detected in November 2004, the third in March 2006, and the fourth in April 2012: the second and third were H-type atypical and the last was an L-type atypical. All three were born and raised in the United States, and hence indigenously infected (or sporadic) (75). A question thus arises as to the actual prevalence of BSE in the United States, which is made exceedingly difficult to answer by virtue of a widely varying rate of testing during the life of the program. The USDA currently tests approximately 40,000 bovines each year, focusing on groups in which BSE is most likely to be found. The targeted population for ongoing surveillance includes cattle exhibiting signs of central nervous disorders or signs associated with BSE, nonambulatory animals, and dead cattle. Discovery of the 2012 atypical cow resulted from a random sample selected from deadstock collected at a rendering facility in California.  Only after the test came back positive was it learned that the 10-year-old dairy cow was euthanized after lameness resulted in recumbency. Using the number of tested animals as the statistical population, analyses were performed for two time periods: (i) from the inception of the program in 1990 through March 2006 and (ii) from January 2002 through March 2006, a period in which the test rate was much greater than that during the 1990s and that still included all of the positive tests. Both analyses yielded very similar results: a prevalence of approximately 2.7 per mil-

EVALUATION OF POSSIBLE SPORADIC BSE If in addition to orally acquired disease BSE were to occur in a sporadic form (like CJD in humans), it would be necessary to continue indefinitely the prohibition of recycling contaminated tissues via supplemental feeding practices. Testing for sporadic disease would need to be done in countries with large national herds that are still free from orally acquired BSE, such as Argentina and Australia. The criteria for selection of cattle to answer the question of spontaneous BSE are different from those for orally acquired BSE. Most importantly, we do not know at what age spontaneous cases of BSE may occur. It is unlikely to be the 3- to 5-year-old age group in which orally acquired BSE is most prevalent, and if the age distribution for spontaneous disease in cattle were to mimic that of sporadic (spontaneous) CJD in humans, which peaks in the 50- to 75-year-old age group, it would not peak until 14 to 20 years of age (the last third of the 20odd-year natural life span of a cow). Significant numbers of such older cattle do not exist, and hence it may never be possible to state with assurance that spontaneous BSE does not occur. However, approximately 10% of sporadic CJD cases occur in patients 25 to 50 years of age, in whom the calculated incidence of CJD is 1 case per 10 million per year. This age bracket approximates the middle third of a cow’s normal life span, or 7 to 13 years of age. For similar age distributions of sporadic BSE and CJD, negative tests in a total of approximately 3 million animals would allow us to be 95% confident that sporadic BSE is not present at a prevalence higher than 1 per million, and 30 million negative tests would lower the maximum prevalence to 1 per 10 million cattle.

BSE IN NONBOVINE SPECIES A final point concerns the potential risk to humans of BSE in nonbovine species. Of the three other principal

Nonbacterial Pathogens

666 livestock sources of food for human consumption, sheep and goats are susceptible by i.c., parenteral, and oral infection (38, 41, 42, 80); pigs are susceptible to i.c. but not oral infection (99); and chickens are resistant to both i.c. and oral infection (63). Current U.S. regulations prohibit the feeding of most mammalian tissues to sheep and goats, but permit feeding to pigs and chickens. Natural cases of BSE have been confirmed in at least two goats in Europe (38,80), and if BSE were somehow to find its way into sheep, it could spread naturally within an experimental flock (8). European red deer are susceptible to i.c. infection (oral infection has not been studied) (33). The susceptibility of fish remains inconclusive (77). In the broader context of the risk to humans from consumption of other TSE-contaminated foodstuffs, chronic wasting disease (CWD) of deer and elk, originally limited to comparatively small areas east of the Rocky Mountains in the US and Canada, continues its spread into previously unaffected areas of both countries. Although mainly of concern to hunters who consume the meat (and other tissues), a wider population is at risk from distributions among family and friends or from restaurant offerings. However, laboratory studies indicate a limited ability of CWD to infect noncervid species, including humans, and to date there is no evidence of any human CWD infections. There are at present no other recognized sources of TSE contamination from the animal world. The most obvious source would be scrapie-infected sheep, but in the nearly three centuries since its recognition in Europe and global spread, during which time there have occurred uncountable instances of the unwitting consumption of contaminated tissues, not a single case of infection has been recognized in humans.

•

•

CONCLUSIONS Modifications of feed production and practices have resulted in a steady decline in cases of BSE around the world. Likewise, precautionary measures to exclude high-risk cattle and SRMs from all cattle from entering the human food chain have now, 15 years after the original reports, virtually eliminated new cases of vCJD. Despite this extremely positive trend, governments and industry cannot become complacent about measures to minimize animal and human risk from this family of diseases. There are several important issues about BSE, vCJD, and other TSEs that need to be monitored during the next several years: •

Secondary cases of human-to-human transmission of vCJD through blood transfusion. The costs associated

•

with a loss of blood donation sources, leukodepletion of the blood supply, and the extreme restrictions on the use of surgical equipment are significant. The extent to which latent BSE infection may be present in the population of the United Kingdom and other countries with vCJD is unknown. Recent evidence indicates that susceptibility may not be strictly limited to individuals who are homozygous for methionine at codon 129 of the prion protein gene, and statistical analysis of tissue surveys in the United Kingdom suggests that a much larger number of humans may have been infected than the current tally of 173 cases of vCJD. It is not known whether such individuals pose a risk to others via the blood supply or the use of surgical instruments, and there is no validated blood test to screen the population for inapparent infections. Given the broader public health implications from humanto-human transmissions, it is imperative for countries to prevent primary foodborne transmission. Countries where BSE may be present but not detected. Many countries have imported vast amounts of MBM from countries with BSE-infected cattle, some of which do not have adequate surveillance programs and have not implemented policies to prevent contamination of animal feed and human food chains. These countries may still serve as a source of the disease. Distribution of infectivity in bovine tissues. BSE infectivity may be more widely distributed than has been ­realized (e.g., muscle infectivity in atypical BSE). The correlation of infectivity and detectable PrPTSE in tissues is not perfect, as illustrated by bioassay-proven infectivity in the absence of PrPTSE in the tongue and nasal mucosa of cattle with end-stage BSE. Similar disparate results have been reported from studies of experimental infections of human CJD and atypical scrapie in sheep. As sensitive testing methodologies come to be more widely employed, the definition of risk material and requirements for feed controls may change. The emergence of new strains or species adaptation of existing strains. The origin of BSE has not been identified with certainty, and its emergence should alert us to the possibility that TSEs may occur in other animal species and have unpredictable characteristics that will provide new challenges.

Finally, it is important that regulatory policies be modified in accord with advances in experimental and epidemiologic knowledge to minimize adverse consequences to both public and animal health. In particular, the development of preclinical diagnostic tests may vastly improve the precision of proactive measures to minimize risks to human and animal health.

25.  Bovine Spongiform Encephalopathy References 1. Alper, T., W. A. Cramp, D. A. Haig, and M. C. Clarke. 1967. Does the agent of scrapie replicate without nucleic acid? Nature 214:764–766. 2. Andréoletti, O., L. Orge, S. L. Benestad, V. Beringue, C. Litaise, S. Simon, A. Le Dur, H. Laude, H. Simmons, S. Lugan, F. Corbière, P. Costes, N. Morei, F. Scheicher, and C. Lacroux. 2011. Atypical/Nor98 scrapie infectivity in sheep peripheral tissues. PLoS Pathog. 7:e1001285. 3. Andréoletti, O., S. Simon, C. Lacroux, N. Morel, G. Tabouret, A. Chabert, S. Lugan, F. Corbière, P. Ferré, G. Foucras, H. Laude, F. Eychenne, J. Grassi, and F. Schelcher. 2004. PrPSc accumulation in myocytes from sheep incubating natural scrapie. Nat. Med. 10:591–593. 4. Angers, R. C., S. R. Browning, T. S. Seward, C. J. Sigurdson, M. W. Miller, E. A. Hoover, and G. C. Telling. 2006. Prions in skeletal muscles of deer with chronic wasting disease. Science 311:1117. 5. Atarashi, R., J. M. Wilham, L. Christensen, A. G. Hughson, R. A. Moore, L. M. Johnson, H. A. Onwubiko, S. A. Priola, and B. Caughey. 2008. Simplified ultrasensitive prion detection by recombinant PrP conversion with shaking. Nat. Methods 5:211–212. 6. Balkema-Buschmann, A., M. Eiden, C. Hoffman, M. Kaatz, U. Ziegler, M. Keller, and M. H. Groschup. 2011. BSE infectivity in the absence of detectable PrPSc accumulation in the tongue and nasal mucosa of terminally diseased cattle. J. Gen. Virol. 92:467–476. 7. Balkema-Buschmann, A., C. Fast, M. Kaatz, M. Eiden, U. Ziegler, L. McIntyre, M. Keller, B. Hills, and M. H. Groschup. 2011. Pathogenesis of classic and atypical BSE in cattle. Prev. Vet. Med. 102:112–117. 8. Bellworthy, S. J., G. Dexter, M. Stack, M. Chaplin, S. A. C. Hawkins, M. M. Simmons, M. Jeffrey, S. Martin, L. Gonzalez, and P. Hill. 2005. Natural transmission of BSE between sheep within an experimental flock. Vet. Rec. 157:206. 9. Béringue, V., L. Herzog, F. Reine, A. Le Dur, C. Casalone, J.-L. Vilotte, and H. Laude. 2008. Transmission of atypical bovine prions to mice transgenic for human prion protein. Emerg. Infect. Dis. 14:1898–1901. http://www. cdc.gov/EID/content/14/12/1898.htm 10. Biacabe, A.-G., J.-L. Laplanche, S. Ryder, and T. Baron. 2004. Distinct molecular phenotypes in bovine prion diseases. EMBO Rep. 5:110–114. 11. Bishop, M. T., P. Hart, L. Aitchison, H. N. Baybutt, C. Plinston, V. Thomson, N. L. Tuzi, M. W. Head, J. W. Ironside, R. G. Will, and J. C. Manson. 2006. Predicting susceptibility and incubation time of human-to-human transmission of vCJD. Lancet Neurol. 5:393–398. 12. Bolton, D. C., M. P. McKinley, and S. P. Prusiner. 1982. Identification of a protein that purifies with the scrapie prion. Science 218:1309–1311. 13. Brown, P. 2005. Blood infectivity, processing and screening tests in transmissible spongiform encephalopathies. Vox Sanguinis 89:63–70. 14. Brown, P., and D. C. Gajdusek. 1991. Survival of scrapie virus after 3 years’ internment. Lancet 337:269–270.

667 15. Brown, P., E. H. Rau, P. Lemieux, B. K. Johnson, A. Bacote, and D. C. Gajdusek. 2004. Infectivity studies of both ash and air emissions from simulated incineration of scrapie-contaminated tissues. Environ. Sci. Technol. 38:6155–6160. 16. Brown, P., R. G. Rohwer, and D. C. Gajdusek. 1986. Newer data on the inactivation of scrapie virus or Creutzfeldt-Jakob disease virus in brain tissue. J. Infect. Dis. 153:1145–1148. 17. Brown, P., R. G. Rohwer, E. M. Green, and D. C. Gajdusek. 1982. Effect of chemicals, heat, and histopathologic processing on high infectivity hamsteradapted scrapie virus. J. Infect. Dis. 145:683–687. 18. Brown, P., R. Meyer, F. Cardone, and M. Pocchiari. 2003. Ultra-high-pressure inactivation of prion infectivity in processed meat: a practical method to prevent human infection. Proc. Natl. Acad. Sci. USA 100:6093–6097. 19. Brown, P., R. G. Will, R. Bradley, D. M. Asher, and L. Detwiler. 2001. Bovine spongiform encephalopathy and variant Creutzfeldt-Jakob disease: background, evolution, and current concerns. Emerg. Infect. Dis. 7:6–16. 20. Bruce, M. E., R. G. Will, J. W. Ironside, I. McConnell, D. Dummond, A. Suttie, L. McCardle, A. Chree, J. Hope, C. Birkett, S. Cousens, H. Fraser, and C. J. Bostock. 1997. Transmissions to mice indicate that “new variant” CJD is caused by the BSE agent. Nature 389:498–501. 21. Buschmann, A., A. Gretzshel, A.-G. Biacabe, K. Schiebel, C. Corona, C. Hoffmann, M. Eiden, T. Baron, C. Casalone, and M. Groschup. 2006. Atypical BSE in Germany—proof of transmissibility and biochemical characterization. Vet. Microbiol. 117:103–116. 22. Buschmann, A., and M. Groschup. 2005. Highly bovine spongiform encephalopathy-sensitive transgenic mice confirm the essential restriction of infectivity to the nervous system in clinically diseased cattle. J. Infect. Dis. 192:934–942. 23. Canadian Food Inspection Agency. 2007. Enhanced animal health protection from BSE. http://www.inspection. gc.ca/animals/terrestrial-animals/diseases/enhancedfeed-ban/eng/1299870250278/1334278201780. Modified March 11, 2011. 24. Cardone, F., P. Brown, R. Meyer, and M. Pocchiari. 2006. Inactivation of transmissible spongiform encephalopathy agents by ultra high pressure-temperature treatment. Biochim. Biophys. Acta 1764:558–562. 25. Cardone, F., A. Thomzig, W. Schulz-Schaeffer, A. Valanzano, M. Sbriccoli, H. Abdel-Haq, S. Graziano, M. Puopolo, P. Brown, M. Beekes, and M. Pocchiari. 2009. PrPTSE in muscle-associated lymphatic tissue during the preclinical stage of mice orally-infected with BSE. J. Gen. Virol. 90:2563–2568. 26. Casalone, C., G. Zanusso, P. Acutis, S. Ferrari, L. Capucci, F. Tagliavini, S. Monaco, and M. Caramelli. 2004. Identification of a second bovine amyloidotic spongiform encephalopathy: molecular similarities with sporadic Creutzfeldt-Jakob disease. Proc. Natl. Acad. Sci. USA 101:3065–3070.

668 27. Colchester, A. C. F., and N. T. H. Colchester. 2005. The origin of bovine spongiform encephalopathy: the human prion disease hypothesis. Lancet 366:856-861. 28. Comoy, E. E., C. Casalone, N. Lescoutra-Etchegaray, G. Zanusso, S. Freire, D. Marcé, F. Auvré, M.-M. Ruchoux, S. Ferrari, S. Monaco, N. Salès, M. Caramelli, P. Leboulch, P. Brown, C. Lasmézas, and J.-P. Deslys. 2008. Atypical BSE (BASE) transmitted from asymptomatic aging cattle to a primate. PLoS ONE 3:e3017. 29. Copobianco, R., C. Casalone, S. Suardi, M. Mangieri, C. Miccolo, L. Limido, M. Catania, G. Rossi, G. Di Fede, G. Giaccone, M. G. Bruzzone, L. Minati, C. Corona, P. Acutis, D. Gelmetti, G. Lombardi, M. H. Groschup, A. Buschmann, G. Zanusso, S. Monaco, M. Caramelli, and F. Tagliavini. 2007. Conversion of the BASE prion strain into the BSE strain: the origin of BSE? PLoS Pathog. 3:e31. 30. Cuillé, J., and P. I. Chelle. 1936. La maladie dite «tremblante» de mouton; est-elle inoculable? Compte Rendu Acad. Sci. 203:1552. 31. Cutlip, R. C., J. M. Miller, A. N. Hamir, J. Peters, M. M. Robinson, A. L. Jenny, H. D. Lehmkuhl, W. D. Taylor, and F. D. Bisplinghoff. 2001. Resistance of cattle to scrapie by the oral route. Can. J. Vet. Res. 65:131–132. 32. Cutlip, R. C., J. M. Miller, R. E. Race, A. L. Jenny, J. B. Katz, H. D. Lehmkuhl, B. M. DeBey, and M. M. Robinson. 1994. Intracerebral transmission of scrapie to cattle. J. Infect. Dis. 169:814–820. 33. Dagleish, M. P., S. Martin, P. Steele, J. Finlayson, S. Sisó, S. Hamilton, F. Chianini, H. W. Reid, L. González, and M. Jeffrey. 2008. Experimental transmission of bovine spongiform encephalopathy to European red deer (Cervus elaphus elaphus). BMC Vet. Res. 4:17. 34. Doherr, M. G., A. R. Hett, J. Rufenacht, A. Zurbriggen, and D. Heim. 2002. Geographical clustering of cases of bovine spongiform encephalopathy (BSE) born in Switzerland after the feed ban. Vet. Rec. 151:467–472. 35. Donnelly, C. A. 1998. Maternal transmission of BSE: interpretation of the data on the offspring of BSE-­affected pedigree suckler cows. Vet. Rec. 142:579–580. 36. Donnelly, C. A., N. M. Ferguson, A. C. Ghani, and R. M. Anderson. 2002. Implications of bovine spongiform encephalopathy (BSE) screening data for the scale of the British BSE epidemic and current European infection levels. Proc. R. Soc. Lond. Ser. B 269:2179–2190. 37. Dudas, S., J. Yang, C. Graham, M. Czub, T. A. McAllister, M. B. Coulthart, and S. Czub. 2010. Molecular, biochemical and genetic characteristics of BSE in Canada. PLoS ONE 5:e10638. 38. Eloit, M., K. Adjou, M. Coulpier, J. J. Fontaine, R. Hamel, T. Lilin, S. Messiaen, O. Andréoletti, T. Baron, A. Bencsik, A.-G. Biacabe, V. Beringue, H. Laude, A. Le Dur, J.-L. Vilotte, E. Comoy, J.-P. Deslys, J. Grassi, S. Simon, F. Lantier, and P. Sarradin. 2005. BSE agent signatures in a goat. Vet. Rec. 156:523–524. 39. Espinosa, J. C., M. Morales, J. Castilla, M. Rogers, and J. M. Torres. 2007. Progression of prion infectivity in asymptomatic cattle after oral bovine spongiform encephalopathy challenge. J. Gen. Virol. 88:1379–1383.

Nonbacterial Pathogens 40. Flan, B., and J.-T. Aubin. 2005. Evaluation de l’efficacité des procédés de purification des protéines plasmatiques à éliminer les agents transmissibles non conventionnels. Virologie 9:S45–S56. 41. Foster, J. D., M. Bruce, I. McConnell, A. Chree, and H. Fraser. 1996. Detection of BSE infectivity in brain and spleen of experimentally infected sheep. Vet. Rec. 138:546–548. 42. Foster, J. D., D. W. Parnham, N. Hunter, and M. Bruce. 2001. Distribution of the prion protein in sheep terminally affected with BSE following experimental oral transmission. J. Gen. Virol. 82:2319–2326. 43. Gavier-Widén, D., M. Nöremark, J. P. M. Langeveld, M. Stack, A.-G. Biacabe, J. Vulin, M. Chaplin, J. A. Richt, J. Jacobs, C. Acín, E. Monleón, L. Renström, B. Klingeborn, and T. G. M. Baron. 2008. Bovine spongiform encephalopathy in Sweden: an H-type variant. J. Vet. Diagn. Invest. 20:2–10. 44. Herzog, C., N. Salès, N. Etchegaray, A. Charbonnier, S. Freire, D. Dormont, J.-P. Deslys, and C. I. Lasmézas. 2004. Tissue distribution of bovine spongiform enceph­ alopathy agent in primates after intravenous or oral infection. Lancet 363:422–428. 45. Hill, A. F., M. Desbruslais, S. Joiner, K. C. L. Sidle, I. Gowland, J. Collinge, L. J. Doey, and P. Lantos. 1997. The same prion strain causes vCJD and BSE. Nature 389:448–450. 46. Hilton, D. A., A. C. Ghani, L. Conyers, P. Edwards, L. McCardle, D. Ritchie, M. Penney, D. Hegazy, and J. W. Ironside. 2004. Prevalence of lymphoreticular prion protein accumulation in UK tissue samples. J. Pathol. 203:733–739. 47. Hinckley, G. T., C. J. Johnson, K. H. Jacobson, C. Bartholomay, K. D. McMahon, D. McKenzie, J. M. Aiken, and J. A. Pedersen. 2008. Persistence of pathogenic prion protein during simulated wastewater treatment processes. Environ. Sci. Technol. 42:5254–5259. 48. Hoffmann, C., U. Ziegler, A. Buschmann, A. Weber, L. Kupfer, A. Oelschlegel, B. Hammerschmidt, and M. H. Groschup. 2007. Prions spread via the autonomic nervous system from the gut to the central nervous system in cattle incubating bovine spongiform encephalopathy. J. Gen. Virol. 88:1048–1055. 49. Hoinville, L. J. 1994. Decline in the incidence of BSE in cattle born after the introduction of the “feed ban.” Vet. Rec. 134:274–275. 50. Hoinville, L. J., J. W. Wilesmith, and M. S. Richards. 1995. An investigation of risk factors for cases of bovine spongiform encephalopathy born after the introduction of the “feed ban.” Vet. Rec. 136:312–318. 51. Horn, G., M. Bobrow, M. E. Bruce, M. Goedert, A. McLean, and J. Webster. 2001. Review of the Origin of BSE. Stationery Office, London, United Kingdom. 52. Ironside, J. W., M. T. Bishop, K. Connelly, D. Hegazy, S. Lowrie, M. LeGrice, D. L. Ritchie, L. M. McCardle, and D. A. Hilton. 2006. Variant Creutzfeldt-Jakob disease: prion protein genotype analysis of positive appen-

25.  Bovine Spongiform Encephalopathy

669

dix tissue samples from a retrospective prevalence study. BMJ 332:1186–1188. Iwata, N., Y. Sato, Y. Higuchi, K. Nohtomi, N. Nagata, H. Hasegawa, M. Tobiume, Y. Nakamura, K. Hagiwara, H. Furuoka, M. Horiuchi, Y. Yamakawa, and T. Sata. 2006. Distribution of PrPSc in cattle with bovine spongiform encephalopathy slaughtered at abattoirs in Japan. Jpn. J. Infect. Dis. 59:100–107. Jacobs, J. G., J. P. M. Langeveld, A.-G. Biacabe, P.-L. Acutis, M. P. Polak, D. Gavier-Widén, A. Buschmann, M. Caramelli, C. Casalone, M. Mazza, M. Groschup, J. H. F. Erkens, A. Davidse, F. G. van Zijderveld, and T. Baron. 2007. Molecular discrimination of atypical bovine spongiform encephalopathy strains from a geographical region spanning a wide area in Europe. J. Clin. Microbiol. 45:1821–1829. Kong, Q., M. Zheng, C. Canalone, L. Qing, S. Huang, B. Chakrabort, P. Wang, F. Chen, I. Cali, C. Corona, F. Martucci, B. Iulini, P. Acutis, L. Wang, J. Liang, M. Wang, X. Li, S. Monaco, G. Zanusso, W. Q. Zou, M. Caramelli, and P. Gambetti. 2008. Evaluation of the human transmission risk of an atypical bovine spongiform encephalopathy prion strain. J. Virol. 82:3697–3701. Konold, T., S. J. Moore, S. J. Bellworthy, and H. A. Simmons. 2008. Evidence of scrapie transmission via milk. BMC Vet. Res. 4:14. Lacroux, C., S. Simon, S. L. Benestad, S. Maillet, J. Mathey, S. Lugan, F. Corbière, H. Cassar, P. Costes, D. Bergonier, J.-L. Weisbecker, T. Moldal, H. Simmons, F. Lantier, C. Feraudet-Tarisse, N. Morel, F. Schelcher, J. Grassi, and O. Andréoletti. 2008. Prions in milk from ewes incubating natural scrapie. PLoS Pathog. 4: e1000238. Llewelyn, C. A., P. E. Hewitt, R. S. G. Knight, K. Amar, S. Cousens, J. Mackenzie, and R. G. Will. 2004. Possible transmission of variant Creutzfeldt-Jakob disease by blood transfusion. Lancet 363:417–421. Maddison, B. C., C. A. Baker, H. C. Rees, L. A. Terry, L. Thorne, S. J. Bellworthy, G. C. Whitelam, and K. C. Gough. 2009. Prions are secreted in milk from clinically normal scrapie-exposed sheep. J. Virol. 83:8293–8296. Masujin, K., D. Matthews, G. A. H. Wells, S. Mohri, and T. Yokoyama. 2007. Prions in the peripheral nerves of bovine spongiform encephalopathy-affected cattle. J. Gen. Virol. 88:1850–1858. Middleton, D. J., and R. M. Barlow. 1993. Failure to transmit bovine spongiform encephalopathy to mice by feeding them with extraneural tissues of affected cattle. Vet. Rec. 132:545–547. Miller, W. M., E. S. Williams, T. Hobbs, and L. L. Wolfe. 2004. Environmental sources of prion transmission in mule deer. Emerg. Infect. Dis. 10:1003–1006. Moore, J., S. A. Hawkins, A. R. Austin, T. Konold, R. B. Green, I. W. Blamire, I. Dexter, M. J. Stack, M. J. Chaplin, J. P. Langeveld, M. M. Simmons, Y. I. Spencer, P. R. Webb, M. Dawson, and G. A. Wells. 2011. Studies of the transmissibility of the agent of bovine spongiform encephalopathy to the domestic chicken. BMC Res. Notes 4:501.

64. National Creutzfeldt-Jakob Disease Research & Surveil­ lance Unit (NCJDRSU). 2012. Variant CreutzfeldtJakob disease current data (July 2012). http://www.cjd. ed.ac.uk/vcjdworld.htm. 65. Oesch, B., D. Westaway, M. Walchli, M. P. McKinley, S. B. H. Kent, R. Aebersold, R. A. Barry, P. Tempst, D. B. Teplow, L. E. Hood, S. B. Prusiner, and C. Weismann. 1985. A cellular gene encodes scrapie PrP 27-30 protein. Cell 40:735–746. 66. OIE (World Organisation for Animal Health). 2011. Bovine spongiform encephalopathy, chapter 11.5. In Terrestrial Animal Health Code, 20th ed. OIE, Paris, France. http://www.oie.int/index.php?id= 169&L=0&htmfile=chapitre_1.11.5.htm  67. OIE (World Organisation for Animal Health). 2011. Geographical distribution of countries that reported BSE confirmed cases since 1989. http://www.oie. int/animal-health-in-the-world/bse-specific-data. 68. Ono, F., Y. Yamakawa, M. Tobiume, Y. Sato, H. Katano, K. Hagiwara, I. Itagaki, A. Hiyaoka, K. Komatuzaki, Y. Emoto, H. Shibata, Y. Murayama, K. Terao, Y. Yasutomi, and T. Sata. 2010. Transmission of classic and atypical (L-type) bovine spongiform encephalopathy (BSE) prions to cynomolgus macaques, poster Ppo2-26. Prion 2010, Salzburg, Austria, 8 to 11 September 2010. 69. Orrú, C. D., and B. Caughey. 2011. Prion seeded conversion and amplification assays. Top. Curr. Chem. 305:121–133. 70. Peden, A., M. W. Head, D. L. Ritchie, J. E. Bell, and J. W. Ironside. Preclinical vCJD after blood transfusion in a PRNP codon 129 heterozygous patient. Lancet 364:527–529. 71. Peden, A., L. McCardle, M. W. Head, S. Love, H. J. Ward, S. N. Cousens, D. M. Keeling, C. M. Millar, F. G. Hill, and J. W. Ironside. 2010. Variant CJD infection in the spleen of a neurologically asymptomatic UK adult patient with haemophilia. Haemophilia 16:296–304. 72. Phillips, Lord, of Worth Matravers, J. Bridgeman, and M. Ferguson-Smith. 2000. The BSE Inquiry. HM Stationery Office, London, United Kingdom. http:// col l ections.europarchive.org/tna/20090505194948/ http://bseinquiry.gov.uk/report/index.htm. 73. Piccardo, P., J. C. Manson, D. King, B. Ghetti, and R. M. Barron. 2007. Accumulation of prion protein in the brain that is not associated with transmissible disease. Proc. Natl. Acad. Sci. USA 104:4712–4717. 74. Richt, J. A., and S. M. Hall. 2008. BSE case associated with prion protein gene mutation. PLoS Pathog. 4: e1000156. 75. Richt, J. A., R. A. Kunkle, D. Alt, E. M. Nicholson, A. N. Hamir, S. Czub, J. P. Kluge, A. J. Davis, and S. M. Hall. 2007. Identification and characterisation of two bovine spongiform encephalopathy cases diagnosed in the United States. J. Vet. Diagn. Invest. 19:142–154. 76. Riverol, M., J-A. Palma, M. Alañá, C. GuerreroMárquez, M. R. Luquin, and A. Rábano. 2012. Variant Creutzfeldt-Jakob disease occurring in mother and son. J. Neurol. Neurosurg. Psychiatry 83:235–236.

53.

54.

55.

56. 57.

58.

59.

60.

61.

62. 63.

670 77. Salta, E., C. Panagiotidis, K. Teliousis, S. Petrakis, E. Eleftheriadis, F. Arapoglou, N. Grigoriadis, A. Nicolaou, E. Kaldrymidou, G. Krey, and T. Sklaviadis. 2009. Evaluation of the possible transmission of BSE and scrapie to gilthead sea bream (Sparus aurata). PLoS ONE 4:e6715. 78. Schreuder, B. E. C., R. E. Geertsma, L. J. M. van Keulen, J. A. A. M. van Asten, P. Enthoven, R. C. Oberthür, A. A. de Koeijer, and A. D. M. E. Osterhaus. 1998. Studies on the efficacy of hyperbaric rendering procedures in inactivating bovine spongiform encephalopathy (BSE) and scrapie agents. Vet. Rec. 142:474–480. 79. Seidel, B., A. Thomzig, A. Buschmann, M. H. Groschup, R. Peters, M. Beekes, and K. Terytze. 2007. Scrapie agent (strain 263K) can transmit disease via the oral route after persistence in soil over years. PLoS ONE 2:e435. 80. Spiropoulos, J., R. Lockey, R. E. Sallis, L. A. Terry, L. Thorne, T. M. Holder, K. E. Beck, and M. M. Simmons. 2011. Isolation of prion with BSE properties from farmed goat. Emerg. Infect. Dis. 17:2253–2261. 81. Stevenson, M. A., J. W. Wilesmith, J. B. M. Ryan, R. S. Morris, A. B. Lawson, D. U. Pfeiffer, and D. Lin. 2000. Descriptive spatial analysis of the epidemic of bovine spongiform encephalopathy in Great Britain to June 1997. Vet. Rec. 147:379–384. 82. Suardi, S., C. Vimercati, C. Casalone, D. Gelmetti, C. Corona, B. Iulini, M. Mazza, G. Lombardi, F. Moda, M. Ruggerone, I. Campagnani, E. Piccoli, M. Catania, M.H. Groschup, A. Balkema-Buschmann, M. Caramelli, S. Monaco, G. Zanusso, and F. Tagliavini. 2012. Infectivity in skeletal muscle of cattle with atypical bovine spongiform encephalopathy. PLoS ONE 7:e31449. 83. Taguchi, F., Y. Tamai, K. Uchida, R. Kitajima, H. Kojima, T. Kawaguchi, Y. Ohtani, and S. Miura. 1991. Proposal for a procedure for complete inactivation of the Creutzfeldt-Jakob disease agent. Arch. Virol. 119:297–301. 84. Taylor, D. M. 2000. Inactivation of transmissible spongiform encephalopathy agents: a review. Vet. J. 159:10–17. 85. Taylor, D. M. 2004. Resistance of transmissible spongiform encephalopathy agents to decontamination. Contrib. Microbiol. 11:136–145. 86. Taylor, D. M., C. E. Ferguson, C. J. Bostock, and M. Dawson. 1995. Absence of disease in mice receiving milk from cows with bovine spongiform encephalopathy. Vet. Rec. 136:592. 87. Taylor, D. M., H. Fraser, I. McConnell, D. A. Brown, K. L. Brown, K. A. Lamza, and G. R. A. Smith. 1999. Decontamination studies with the agents of bovine spongiform encephalopathy and scrapie. Arch. Virol. 139:313–326. 88. Taylor, D. M., S. L. Woodgate, and M. J. Atkinson. 1995. Inactivation of the bovine spongiform encephalopathy agent by rendering procedures. Vet. Rec. 137:605–610. 89. Taylor, D. M., S. L. Woodgate, A. J. Fleetwood, and R. J. G. Cawthorne. 1997. Effect of rendering procedures on the scrapie agent. Vet. Rec. 141:643–649.

Nonbacterial Pathogens 90. Tester, S., V. Juillerat, M. G. Doherr, B. Haase, M. Polak, F. Ehrensperger, T. Leeb, A. Zurbriggen, and T. Seuberlich. 2009. Biochemical typing of pathological prion protein in aging cattle with BSE. Virol. J. 6:64. 91. Thomzig, A., F. Cardone, D. Krugr, M. Pocchiari, P. Brown, and M. Beekes. 2006. Pathological prion protein in muscles of hamsters and mice infected with rodent-adapted BSE or vCJD. J. Gen. Virol. 87:251–254. 92. UK Health Protection Agency. 2006. Variant CJD and blood. http://www.hpa.org.uk/Topics/InfectiousDiseases/ InfectionsAZ/CreutzfeldtJakobDisease/Variant CJDAndBlood/ 93. UK Health Protection Agency. 2012. Summary results of the second national survey of abnormal prion prevalence in archived appendix specimens. http://www.hpa. org.uk/hpr/archives/2012 /news3212.htm#bnrmlprn. 94. U.S. Food and Drug Administration. 2009. Substances prohibited from use in animal food or feed; final rule 21 CFR part 589. http://edocket.access.gpo.gov/2008/081180.htm 95. Wadsworth, J. D. F., S. Joiner, A. F. Hill, T. A. Campbell, M. Desbruslais, P. J. Luthert, and J. Collinge. 2001. Tissue distribution of protease resistant prion protein in variant Creutzfeldt-Jakob disease using a highly sensitive immunoblotting assay. Lancet 358:171–180. 96. Ward, H. J. T., D. Everington, S. Cousens, B. SmithBathgate, M. Leitch, S. Cooper, C. Heath, R. G. Knight, P. Smith, and R. G. Will. 2006. Risk factors for variant Creutzfeldt-Jakob disease: a case control study. Ann. Neurol. 59:111–120. 97. Wells, G. A. H., A. C. Scott, C. T. Johnson, R. F. Gunning, R. D. Hancock, M. Jeffrey, M. Dawson, and R. Bradley. 1987. A novel progressive spongiform encephalopathy in cattle. Vet. Rec. 121:419–420. 98. Wells, G. A. H., T. Konold, M. E. Arnold, A. R. Austin, S. A. C. Hawkins, M. Stack, M. M. Simmons, Y. H. Lee, D. Gavier-Widén, M. Dawson, and J. W. Wilesmith. 2007. Bovine spongiform encephalopathy: the effect of oral exposure dose on attack rate and incubation period in cattle. J. Gen. Virol. 88:1363–1373. 99. Wells, G. A. H., S. A. C. Hawkins, A. R. Austin, S. J. Ryder, S. H. Done, R. B. Green, I. Dexter, M. Dawson, and R. H. Kimberlin. 2003. Studies of the transmissibility of the agent of bovine spongiform encephalopathy to pigs. J. Gen. Virol. 84:1021–1031. 100. Wells, G. A. H., S. A. Hawkins, R. B. Green, A. R. Austin, I. Dexter, Y. I. Spencer, M. J. Chaplin, M. J. Stack, and M. Dawson. 1998. Preliminary observations on the pathogenesis of experimental bovine spongiform encephalopathy (BSE): an update. Vet. Rec. 142:103–106. 101. Wells, G. A. H., J. Spiropoulos, S. A. C. Hawkins, and S. J. Ryder. 2005. Pathogenesis of experimental bovine spongiform encephalopathy: preclinical infectivity in tonsil and observations on the ­ distribution of lingual tonsil in slaughtered cattle. Vet. Rec. 156:401–407.

25.  Bovine Spongiform Encephalopathy

671

102. Wilesmith, J. W., and J. B. Ryan. 1992. Bovine spongiform encephalopathy: recent observations on the agespecific incidences. Vet. Rec. 130:491–492. 103. Wilesmith, J. W., and J. B. M. Ryan. 1997. Absence of BSE in the offspring of pedigree suckler cows affected by BSE in Great Britain. Vet. Rec. 141:250–251. 104. Wilesmith, J. W., G. A. H. Wells, J. B. M. Ryan, D. GavierWiden, and M. M. Simmons. 1997. A cohort study to examine maternally associated risk factors for bovine spongiform encephalopathy. Vet. Rec. 141:239–243. 105. Wilesmith, J. W., G. A. H. Wells, M. P. Cranwell, and J. B. M. Ryan. 1988. Bovine spongiform encephalopathy: epidemiological studies. Vet. Rec. 123:638–644. 106. Wilesmith, J. W., J. B. Ryan, and W. D. Hueston. 1992. Bovine spongiform encephalopathy: case-control studies of calf feeding practices and meat and bonemeal inclusion in proprietary concentrates. Res. Vet. Sci. 52:325–331. 107. Will, R. G., J. W. Ironside, M. Zeidler, S. N. Cousens, K. Estibeiro, A. Alperovitch, S. Poser, M. Pocchiari,

A. Hofman, and P. G. Smith. 1996. A new variant of Creutzfeldt-Jakob disease in the UK. Lancet 347:921–925. 108. Will, R. G., M. Zeidler, G. E. Stewart, J. A. Macleod, J. W. Ironside, S. N. Cousens, J. Mackenzie, K. Estibeiro, A. J. E. Green, and R. S. G. Knight. 2000. Diagnosis of new variant Creutzfeldt-Jakob disease. Ann. Neurol. 47:575–582. 109. World Health Organization (WHO). 1999. Infection Control Guidelines for Transmissible Spongiform Encephalopathies: Report of a WHO Consultation. WHO/CDS/CSR/APH/2000.3 World Health Organiza­ tion, Geneva, Switzerland. www.who.int/csr/resources/ publications/bse/whocdscsraph2003.pdf. 110. World Health Organization (WHO). 2010. WHO Tables on Tissue Infectivity Distribution in Transmissible Spongiform Encephalopathies. WHO/EMP/QSM/2010.1. World Health Organization, Geneva, Switzerland. www. who.int/entity/bloodproducts/tablestissueinfectivity.pdf.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch26

26

H. Ray Gamble Dante S. Zarlenga

Helminths in Meat

Foodborne parasites pose a risk to human health in virtually all regions of the world. In addition to the direct effect that these parasites have on human health, zoonotic parasites found in food animals often serve as trade barriers for countries where these parasites occur. A considerable body of legislation has been developed for the purpose of preventing and controlling zoonotic parasites in food animals, including very costly meat inspection programs. There are four meat-borne helminths of medical significance: Trichinella spp., Taenia solium, and Taenia asiatica, which occur primarily in pork; and Taenia saginata, which is found in beef. Despite the availability of sensitive, specific diagnostic tests, veterinary public health programs (meat inspection), and effective chemotherapeutic agents for human tapeworm carriers, these parasites continue to be a threat to public health throughout the world. There are a variety of reasons for this, including animal management systems that perpetuate infection, inadequate or poorly enforced inspection requirements for slaughtered animals, new sources of infection, and demographic changes in human populations that introduce new culinary practices of preparing meats. Thus, current control and preventive procedures are often inadequate, and

more effective control measures are needed to ensure safe meat for human consumption.

TRICHINELLOSIS The history of trichinellosis is fascinating, going back to a period when infection by Trichinella species was presumed more than proven. Whether the commandment in the Bible (Leviticus 11 and Deuteronomy 14) to not eat the flesh of cloven-footed animals (swine) was due in part to the potential danger of contracting trichinellosis is only speculative. In the seventh century a.d., Mohammed prohibited the eating of pork. Hence, to this day, trichinellosis is rare among Jews and Muslims. The earliest known case of trichinellosis may be evidenced by the mummy of a person who probably lived near the Nile River ca. 1200 b.c., an observation made only in 1980 when Trichinella larvae were presumptively identified in the mummy’s intercostal muscle (102). Trichinella spiralis was first observed by Paget in 1835 in the muscles of a man during postmortem dissection (18). A landmark in the history of clinical trichinellosis was Zenker’s demonstration in 1860 that encapsulated larvae in the arm muscle caused the illness and death of

H. Ray Gamble, National Research Council, 500 Fifth Street NW, Washington, DC 20001. Dante S. Zarlenga, U.S. Department of Agriculture, Agricultural Research Service, 10300 Baltimore Avenue, Beltsville, MD 20705.

673

SMP_Food Microbiology_CH26.indd

673

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

674 a young woman. Even after 1860, many cases of trichinellosis were undoubtedly not diagnosed because of the difficulty in recognizing the infection clinically.

Species of Trichinella

Nematodes of the genus Trichinella are ubiquitous in animals, both domestic and wild. Historically, it was assumed that there was only one species of Trichinella, T. spiralis. However, our understanding of the composition of this genus has changed remarkably in recent years. A large volume of information is now available on species within the genus and the molecular and biological characteristics that distinguish them (76, 110, 128, 129). Initial knowledge of variation within the genus Trichinella was based primarily on observed biological differences. In 1972, new species were proposed for an isolate of Trichinella from arctic carnivores, designated T. nativa (13); an isolate first obtained from a hyena in Kenya, designated T. nelsoni (13); and a nonencapsulating isolate from a raccoon dog in Russia, designated T. pseudospiralis (57). These new species demonstrated clear differences from T. spiralis in both distribution and biological characteristics. T. nativa is found only in the Holarctic region, and its distinguishing biological characteristics include resistance to freezing and low infectivity for pigs (76). T. nelsoni occurs in equatorial Africa and is found in a wide range of wild animals, the primary hosts being the bush pig and the warthog (111). Like T. nativa, T. nelsoni has low infectivity for the domestic pig, and it develops a capsule slowly compared with other encapsulating Trichinella species (76). T. pseudospiralis, believed initially to be primarily a parasite of carnivorous birds, is smaller than the other species and, more importantly, does not form a capsule in host musculature (57). Since its description, this species has been found in various birds and mammals throughout the world (76, 129) and has been implicated in human disease (2, 75). The systematics of Trichinella have evolved rapidly over the past 30 years, with the use of biochemical, and then molecular, techniques to further delineate the phylogenetic relationships of an increasing number of isolates from various animals, including humans. Molecular differentiation has supported observed biological (76) and biochemical (87) differences, and as a result, new species and genotypes of Trichinella have emerged. Currently, the genus Trichinella includes eight named species and three related genotypes, broadly grouped into those species that form capsules in the host and those that do not form a capsule (129). Those species forming a capsule include T. spiralis, T. nativa and the related genotype T-6, T. britovi and the related

SMP_Food Microbiology_CH26.indd

674

genotypes T-8 and T-9, T. murrelli, and T. nelsoni. The nonencapsulating species include T. pseudospiralis, T. papuae, and T. zimbabwensis. Encapsulated species of Trichinella have been reported only in mammalian hosts. Of these, T. spiralis is distributed in temperate regions worldwide and remains the species most closely associated with infections in domestic swine and synanthropic rodents and in human infections resulting from the ingestion of pork. The freeze-resistant species T. nativa occurs in the Holoarctic region worldwide and has a broad host distribution (128). Due to the ability of T. nativa to persist in frozen muscle tissue, it poses a public health threat to hunters and consumers of game meats in regions where trichinellosis is endemic. Human trichinellosis resulting from walrus meat infected with T. nativa has been well documented (47, 88), leading public health officials in Canada to develop an inspection program for walruses killed for human consumption (137). The other freezeresistant genotype of Trichinella, designated T-6, is found in North America. It has low infectivity for pigs, is found in a variety of wild mammals, and has been implicated in human disease (31). T. britovi and related genotypes occur in temperate regions of Europe, Asia (T-9 from Japan), and Africa (T-8 from South Africa and Namibia). The distribution of T. britovi has recently been extended to West Africa (134). T. britovi has some intermediate characteristics of other species, including some tolerance to low temperatures, moderate infectivity for swine, and slow capsule formation (larvae have been confused for nonencapsulating species in some cases) (128). T. britovi has been the most common species implicated in human infections resulting from the ingestion of infected horsemeat in Europe. T. murrelli (130) is a North American species found in wildlife and occasionally horses and humans. It has low infectivity for domestic pigs but poses a risk to humans who eat game meats. T. nelsoni appears to have a distribution limited to the eastern part of the Afrotropical region from Kenya to South Africa (134). The three nonencapsulating species of Trichinella are found in a variety of hosts, including mammals, birds, and reptiles (135). T. pseudospiralis has been recovered from raptorial birds, wild carnivores including wild boar, rats, and marsupials in Europe, Asia, North America, and the Australian subcontinent. Several human outbreaks due to T. pseudospiralis have been reported. T. papuae (133) has been detected in several regions of Papua New Guinea, while T. zimbabwensis has been reported in Zimbabwe and Mozambique (132). T. papuae has been found in domestic and wild pigs as well as saltwater crocodiles (fed pig meat). Due to its abil-

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

675

ity to infect reptiles, it is the suspected agent of human trichinellosis resulting from the ingestion of turtle and lizard meat. T. zimbabwensis is similar to T. papuae in its ability to infect reptiles and wild carnivores (132), but it has not been implicated in human disease. It is noteworthy that these two Trichinella species develop at host body temperatures of 26 to 32°C; encapsulating species inoculated into the same hosts were unable to develop at these temperatures. In contrast to T. papuae and T. zimbabwensis, T. pseudospiralis does not develop in reptiles but does develop in chickens. As molecular techniques are refined and comparative studies are performed on Trichinella isolates, further taxonomic resolution should be applied to this genus. Several excellent reviews of the current status of Trichinella species are available (76, 110, 128, 135).

Life Cycles of Trichinella Species

All species of Trichinella complete their life cycle within one host; no intermediate hosts or extrinsic development is required. When larvae encysted in raw or inadequately cooked meat are ingested, the muscle fibers and capsules that enclose the parasites are digested in the stomach. In the intestine, the liberated larvae burrow into the lamina propria of the villi in the jejunum and ileum. Four molts occur within 48 h, and by the third day, the worms are sexually mature. The small tapered head of the adult worm has a round, unarmed mouth that opens into a tubular esophagus. Approximately one-third of the anterior portion of the body is composed of the stichosome, consisting of stacks of discoid stichocyte cells. The secreted contents of these cells are important for serological detection of infection and also contain antigens that confer protective resistance to the host. Immediately below the esophagus and the stichosome lies a thin-walled intestine, the hind portion of which terminates in the rectum, a muscular tube lined with chitin. The female worm measures about 3.5 mm in length and possesses a vulval opening about one-fifth the body length from the anterior end (Fig. 26.1). The male measures 1.3 to 1.6 mm in length and possesses a single testis that originates in the posterior portion of the body and extends anteriorly to near the posterior end of the esophagus, where it turns posteriorly to form the vas deferens, which becomes the enlarged vesicula seminalis. The vesicula seminalis becomes the ejaculatory duct to join the copulatory tube in the cloaca. The copulatory tube forms the copulatory bell that is extruded during copulation (Fig. 26.2). There are two ventrally located copulatory appendages on each side of the cloacal opening, which possibly serve to clasp the female in copula. Between these appendages lie four tubercles, or papillae.

SMP_Food Microbiology_CH26.indd

675

Figure 26.1  Scanning electron micrograph of female adult worm of T. spiralis with its prominent vulval opening (×2,450). doi:10.1128/9781555818463.ch26f1

Variations in size and differences in the cuticle have been noted for different species and genotypes. For example, T. pseudospiralis is as much as one-fourth to one-third smaller than other isolates. In the host, sexually mature adult worms reenter the lumen of the small intestine, where copulation takes place. The adult males die shortly after copulation. The female worms reburrow into the mucosa and begin to larviposit, usually into the central lacteals of the villi, about 7 days after infection and may continue to do so for a period up to a few weeks. Each female worm bears approximately 1,500 newborn larvae, but this number varies based on Trichinella species and host. The tiny newborn larvae (100 by 6 mm) are carried from the intestinal lymphatic vessels to the regional lymph nodes and into the thoracic duct and the venous blood, passing through the right side of the heart, through the pulmonary capillaries back to the left side of the heart, and into peripheral circulation. During migration, the larvae are known to enter many tissues, including those of the myocardium, brain, and other sites, but here they either are destroyed or reenter the bloodstream. Generally, only larvae that reach striated muscles are able to continue development. They penetrate the sarcolemma of the fibers, where they mature, reaching approximately 700 to 1,100 mm in length. They become coiled within the fibers and, in the case of most species, are encapsulated as a result of the host’s cellular response. This host-parasite complex, called the nurse cell, is capable of supporting the infective larvae for months or even years. An increased vascular supply to the nurse cell provides nutrients and oxygen vital to the parasite’s survival. The encapsulated cyst eventually becomes calcified, and as a result the larva dies.

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

676

Figure 26.2  Scanning electron micrograph of male adult of T. spiralis with its copulatory bell (×1,400). doi:10.1128/9781555818463.ch26f2

Infection in humans typically represents a dead end in the parasite’s life cycle. However, in animal hosts, the carcass serves as a source of infection. Typically, infected animals are refractory to subsequent infection. However, molecular evidence has documented dual infection in animals (128), and it is possible that these dual infections result from multiple exposures.

Epidemiology

Trichinellosis, human disease resulting from infection with species of the genus Trichinella, is considered a zoonosis because infection occurs as a result of ingestion of raw or poorly cooked meat from infected animals. Sources of human infection include domestic livestock, primarily pigs and horses, and wild mammals including, most notably, bear and feral pigs. The transmission patterns of Trichinella spp. can be divided, based on hosts, into synanthropic (domestic-animal) and wildlife (sylvatic) cycles (128). Humans can become involved in both of these cycles. The domestic cycle is defined by habitat and, by definition, must involve domesticated (farm) animals. There are various patterns of transmission within the domestic cycle. Pigs may become infected through the deliberate feeding of uncooked or undercooked animal flesh containing infective larvae. Infected pigs may then serve as a source of infection to other pigs through cannibalism. Within the domestic cycle, rats may serve as a reservoir host, acquiring infection from wildlife, carrion, or pigs. Rats may transmit infection directly to pigs that ingest them. Pigs may acquire infection directly from wildlife, ingesting meat from dead animals when allowed to wander under unrestricted conditions or through the deliberate feeding of wildlife carcasses. The species of Trichinella most frequently associated with the domestic cycle in pigs is T. spiralis;

SMP_Food Microbiology_CH26.indd

676

however, there are an increasing number of reports of T. britovi as a cause of human infection caused by meat from domestic pigs (131). Lower infectivity of other Trichinella species for the domestic pig diminishes their importance in the domestic cycle. Because cooking, freezing, and other processing methods kill Trichinella larvae in meat, most human infections have resulted from instances where meat preparation was not adequate. Pork products such as fresh sausage, summer sausage, and dried or smoked sausage have all been implicated as sources of human trichinellosis. While countries where pigs are reared in confinement management systems have essentially no Trichinella infection in commercial pork, pork and pork products continue to serve as a major source of infection in many parts of the world (27, 119, 153). Another domesticated animal that has emerged as a major source of human trichinellosis is the horse. Outbreaks in 1975 and 1985 in the southern suburbs of Paris were attributed to the consumption of raw horsemeat in the form of steak tartare (11). It was initially difficult to believe that human infections resulted from ingestion of meat of an herbivorous animal, but a series of large outbreaks over the past 3 decades have documented the role of horsemeat as a cause of human trichinellosis. More than 3,200 cases of human trichinellosis have been reported in France and Italy as a result of ingesting raw or undercooked horsemeat. Other countries where horsemeat is consumed have not had human cases, as meat is cooked thoroughly before consumption. We know that three species have contributed to human trichinellosis from horsemeat, T. spiralis, T. britovi, and T. murrelli (9). Speculation following the first outbreaks was that the horsemeat was contaminated with meat from other species (6). However, the recovery of larvae from suspected meat (62) and identification of naturally infected horses (4) have solidified this source of human exposure. The route by which horses become infected has been the subject of considerable speculation. The majority of horses linked to human outbreaks in Italy and France originated from countries or regions where a high incidence of Trichinella infection is known to occur in pigs, rats, and other animal species. One epidemiological study reported that the intentional feeding of animal products and kitchen waste is a common occurrence among horse owners in parts of Eastern Europe, and that horses are more willing to consume meat than previously realized (108). With better awareness of the risk of acquiring trichinellosis from horsemeat, inspection programs have been implemented that have reduced the incidence of infection from this source.

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

677

Other domestic sources of trichinellosis have been reported sporadically. An outbreak in China was reported to be due to the consumption of mutton (20). An unusual outbreak in northern Germany was attributed to the consumption of air-dried meat known as pastyrma that had been purchased as camel meat in Egypt (10). It was not certain whether this delicacy was indeed camel meat, since pastyrma is usually made from beef. In either case, there may have been adulteration with pork. Dog meat has been implicated as a source of human trichinellosis in Thailand, China, and Russia (21, 23, 68). Dog meat is prepared as a special dish and is favored by the Vietnamese, Chinese, and hill tribe Thais. Farmed crocodile flesh has also been reported to harbor Trichinella larvae (106), but no human cases of trichinellosis from this source have been reported. The sylvatic cycle of Trichinella involves more than 100 species of mammals, and many of these species pose a risk to humans who eat raw or undercooked game meats. In the sylvatic cycle, wild carnivorous or omnivorous animals scavenge the carrion of dead animals or eat meat from prey animals. Higher infection rates typically occur in animals near the top of the food chain. For example, in the arctic environment, Trichinella is quite prevalent in polar bears, grizzly bears, foxes, and wolves. Transmission in the arctic environment is further facilitated by the survival of larvae in frozen meat. Trichinous polar bear meat may conceivably have been responsible for the deaths of three Swedish explorers on an expedition to the North Pole in 1897. More than 50 years later, laboratory examination verified the presence of larvae in minute particles of bear meat still remaining on the explorers’ equipment (142). In Alaska, all human cases of trichinellosis have been traced to the consumption of bear or walrus meat, and the etiological agent of these infections is T. nativa (47). Polar bear meat is considered a delicacy among the Eskimos. Among the Inuit Eskimos in northeastern Canada, there have been outbreaks resulting from the consumption of undercooked walrus meat (158) and probably other mammals (96). The repeated outbreaks of trichinellosis caused by the consumption of infected walrus meat have led to an inspection program for harvested walruses prior to distribution of the meat to Inuit communities (137). In the former Soviet Union, up to 96% of human trichinellosis cases have been traced to the consumption of wild animals, particularly bears (7). Human infection resulting from badger meat has been reported in Korea (150) as well as Russia (121) and other countries of the former Soviet Union. In the United States, bear meat is the most commonly incriminated game meat, and feral pig is second (143).

SMP_Food Microbiology_CH26.indd

677

Prevalence of Human Disease

Human trichinellosis is essentially nonexistent in some countries of the world but remains a significant problem in many others. While many thousands of clinical cases are reported each year throughout the world, it is impossible to measure true prevalence in humans. In the United States, a National Institutes of Health report published in 1943 found 16.2% of the population to be infected with Trichinella (170). This information led to considerable publicity on the dangers of eating pork and was responsible for strict federal control of methods used to prepare ready-to-eat pork products. The number of cases of clinical trichinellosis reported to the Centers for Disease Control and Prevention declined from about 500/year in the 1940s to fewer than 50/year in the 1980s and fewer than 10/year on average over the past decade. Further, the majority of these cases resulted from nonpork sources such as bear and other game meats. Trends in human infection patterns change with demographics, as evidenced in certain immigrant population in the United States (79). Historically, trichinellosis was more common in individuals of German, Italian, and Polish descent because of culinary preferences for raw or undercooked pork. While human trichinellosis was generally on the decline in the United States, between 1975 and 1984 infection among Southeast Asians, who prepare some dishes containing essentially raw pork, was 25 times more frequent than in the general population (152). Of the 1,260 cases of trichinellosis reported to the Centers for Disease Control and Prevention during this period, 60 (4.8%) were among refugees from Southeast Asia. One of the largest outbreaks ever reported in the United States occurred in 1990 when Southeast Asian refugees from six states and Canada developed trichinellosis after eating pork sausage at a wedding held in Iowa (100). In several countries of the European Union, where slaughter inspection of pigs for Trichinella infection has been required for more than 100 years, the parasite has been essentially eradicated from domestic swine and human trichinellosis resulting from domestic pork has not been reported for several decades. These countries do report sporadic outbreaks due to game meats or imported meats. In contrast, trichinellosis remains a serious public health problem in many parts of the world and may be considered an emerging or reemerging disease in some countries. An example of the emergence of trichinellosis comes from Romania (8). Between 1980 and 1989, a total of 4,345 cases of human trichinellosis were reported. Between 1990 and 2004, a total of 23,948 cases were reported. In the year with the greatest number of human cases, 1993, more than 250 human

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

678 cases were reported per 1 million inhabitants. This spike in human infection is attributed to changes in the pork production systems due to turmoil in the social and political systems. Reemergence of trichinellosis as a serious health risk has likewise been reported in Serbia (174), where infection rates in pigs between the years 1995 and 2006 were nearly 1.0%. These regions, which are highly endemic for Trichinella infection in pigs, have also been the source of many of the horses implicated in outbreaks of human trichinellosis in France and Italy (9). In addition to countries of Eastern Europe, Trichinella continues to be a significant public health risk in parts of Asia and in Central and South America. Wang and Cui (164, 165) reported 23,004 cases of trichinellosis in China between 1964 and 1999. Almost all cases (95.8%) were linked to pork. These same authors found up to 5.6% of pork samples from local markets to be infected. In addition to outbreaks associated with pork, nine outbreaks occurring in China between 1974 and 1998 were associated with ingestion of dog meat (21). In Argentina, the number of human cases increased from 908 between 1971 and 1981 to 6,919 between 1990 and 2002. The most affected areas were in the central region of Argentina, including Buenos Aires, Santa Fe, and Cordoba. The consumption of locally produced pigs is linked to human outbreaks, and these outbreaks occur more frequently during the winter season when pork is used to make chacinados, the principal meat dish involved in human outbreaks (140). In most cases, poor hygienic conditions for raising pigs and inadequate veterinary control at slaughter are the reasons for high rates of human infection. Pigs raised outdoors, fed raw garbage, or raised in contact with large rodent populations are at risk of acquiring infection. Inadequate veterinary inspection may result from insensitive or improperly performed testing methods or from pigs that are butchered outside of normal inspection channels (backyard pigs). All of these factors are involved in the persistence or resurgence of swine and human infections as discussed above (27). In addition to gaps in inspection programs, culinary habits (improper cooking) impact the risk of human exposure, and this is exemplified in some Asian countries where epidemics have been reported, including Thailand, Laos, Japan, China, and Hong Kong. A dish known as nahm is a common source among the inhabitants of northern Thailand (77). There are also other Thai dishes, such as lu (lahb [raw spiced meat] mixed with fresh blood) and satay (small pieces of spiced pork grilled rare to medium on a bamboo skewer), that are sources of infection. Sometimes satay is eaten raw. In both dishes, the larvae may remain alive. Outbreaks of trichinellosis

SMP_Food Microbiology_CH26.indd

678

in Laos have been due to the consumption of pork dishes known as som-mou, lap mou, and lap leuat. Laotian immigrants in the United States still prepare these traditional dishes, resulting in outbreaks (19).

Pathogenesis and Pathology

Damage caused by Trichinella infection varies with the intensity of the infection and the tissues invaded. The in-and-out movements of the adult worms, especially the females in intestinal tissue, cause an acute inflammatory response and petechial hemorrhages. The cellular response consists primarily of neutrophils with eosinophils. This is followed by an infiltration of lymphocytes, plasma cells, and macrophages that peaks at about 12 days after infection, gradually declining thereafter. Thus, lesions in the intestine are due to the host’s response to the adult worms or their protein products. The newborn larvae cause an acute inflammatory response as they pass through or become lodged in various tissues and organs. The infiltration consists of lymphocytes, neutrophils, and especially eosinophils. Although there is myocarditis, viable larvae are more numerous in the pericardial fluid than in the myocardium. Pulmonary hemorrhage and bronchopneumonia may be observed during this stage of larval migration through the capillaries. Rarely, encephalitis may result if the larvae migrate through cerebral capillaries. When larvae encyst in striated muscle fibers, there is an immediate tissue response consisting of inflammation of the sarcolemma of the involved muscle fibers. The disturbance of ultrastructure and metabolic processes in muscle fibers results in basophilic transformation. This is followed by destruction of the muscle fibers and the eventual formation of a capsule. The larvae gradually die, provoking an intense granulomatous reaction or foreign body cellular response that culminates in calcification. The most heavily parasitized striated muscles include the diaphragm, intercostals, and ocular and masseter muscles. Those larvae that enter tissues other than striated muscles disintegrate and are eventually absorbed.

Clinical Manifestations

The vast majority of individuals infected with Trichinella spp. are asymptomatic, probably because low numbers of larvae are ingested. Classical trichinellosis is ­usually described as a febrile disease with gastrointestinal symptoms,­ periorbital edema, myalgia, petechial hemorrhage,­ and eosinophilia (82). During the intestinal stage of infection, gastrointestinal symptoms such as nausea, ­vomiting, and “toxic” diarrhea or dysentery, as well as fever (over 38°C) and sweating, may be observed.

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

679

Malaise can be severe and lasts longer than fever. The onset of intestinal symptoms occurs usually within 72 h after infection and may last for 2 weeks or longer. Generally, from the second week, when the newborn larvae are migrating and can reach almost any tissue within the body, a characteristic edema is noted around the eyes and in some cases around the sides of the nose, at the temples, and even on the hands. Periorbital edema is present in about 85% of patients and is often complicated by subconjunctival and subungual splinter hemorrhages that disappear within 2 weeks. Eosinophilia greater than 50% is the most consistent manifestation. In severe cases, however, the number of eosinophils may be low. As the larvae enter striated muscle fibers, a striking myositis and muscular pain are evident. In addition to inflammation and pain, eosinophilia is observed during this phase. Serum creatine phosphokinase and transaminase levels can be elevated as a result of leakage of these enzymes from muscle fibers into the serum. Higher levels of enzymes in serum may not correlate well with the severity of the clinical condition. Hypoalbuminemia, which is believed to be due to a demand for protein by the parasite, is usually accompanied by hypopotassemia. Muscle atrophy and contractures may also occur. Respiratory symptoms, including dyspnea, cough, and hoarseness, may either be due to myositis of respiratory muscles or be secondary to pulmonary congestion. Neurologic manifestations are rare but may result from invasion of the brain by migrating larvae. In severe cases, death may result following cardiac decompensation or respiratory failure, cyanosis, and coma. Intimate contact between the parasite and host tissues stimulates the production of antibodies that can be demonstrated in the serum. The exact role of antibodies in acquired immunity is not entirely clear, but antibodies appear to be involved in destruction of newborn larvae. On the basis of experimental animal data, it would be safe to assume that in most cases in humans, the severity of infection is reduced considerably by the development of immunity from previous subclinical infections. Local gut immunity in mice has been shown to be T-cell dependent (61). It has been reported that different species of Trichinella present somewhat different clinical pictures in humans. For example, Serhir et al. (148) described a diarrheic syndrome in outbreaks caused by T. nativa in arctic regions of North America. This syndrome, characterized by persistent diarrhea and little edema or fatigue, differs from the traditional myopathic syndrome characterized by edema, fatigue, fever, and rash. These distinct syndromes occur concurrently in outbreaks; following the accumulation of serological and

SMP_Food Microbiology_CH26.indd

679

epidemiological evidence, it was concluded that the diarrheic syndrome is seen in patients who have preexisting immunity to T. nativa from prior exposure (148). Similarly, infection with T. murrelli results in a greater number of patients developing a rash (44%) and fewer patients developing facial edema (58%) (30). Only a few outbreaks of human trichinellosis have been attributed to nonencapsulating species of Trichinella (14, 138). The only observed clinical difference in these cases was a prolonged period of symptoms (fever and myalgias), which lasted twice as long as those reported for encapsulating species.

Diagnosis

It is difficult to clinically diagnose trichinellosis because it mimics so many other infections as a result of its dissemination throughout the body. Thus, individual or small groups of cases may frequently go undiagnosed, whereas common-source outbreaks make diagnosis easier (14). Diagnosis is based on a history of eating infected meat, symptoms, laboratory findings (including serology), and recovery of larvae from muscles (52). Classical symptoms of trichinellosis include fever, myalgia, and periorbital edema, although, as discussed previously, these symptoms may vary depending on the species of Trichinella. The most characteristic feature of laboratory tests is marked eosinophilia. Although eosinophilia is not restricted to trichinellosis, its presence in 30 to 85% of infected individuals is a constant and important diagnostic aid. Eosinophilia is initiated in the second week of infection, reaching its peak by about day 20. It may be absent not only in patients with very severe and fatal cases but also in individuals with secondary infection. Muscle enzymes, including creatine phosphokinase and lactate dehydrogenase, are frequently elevated in trichinellosis patients, as are overall leukocyte counts (14, 29). The most definitive diagnostic method is muscle biopsy to detect encapsulated or nonencapsulated larvae. The diagnostic success of biopsy depends on chance distribution of larvae in the particular striated muscle that is sampled. Gastrocnemius, pectoralis major, deltoid, or biceps muscles are commonly used because of easy accessibility. The muscle strip is compressed tightly between two microscope slides and examined for the presence of larvae (Fig. 26.3). Part of the biopsy sample can be digested or fixed and then sectioned, stained, and examined. The presence of active larvae following digestion with artificial gastric juice indicates a recent infection. There are numerous immunological tests available for the diagnosis of trichinellosis (78, 80, 107). Recommendations for the use of these tests have been

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

680

Treatment

Figure 26.3  Pressed muscle containing T. spiralis larvae. doi:10.1128/9781555818463.ch26f3

summarized by the International Commission on Trichinellosis (54). The enzyme-linked immunosorbent assay (ELISA), using an excretory-secretory (ES) antigen collected from muscle larvae or a synthetic tyvelose antigen containing a dominant epitope, is the test of choice for human trichinellosis based on a high degree of sensitivity and specificity. Both ES and tyvelose antigens are preferable to crude extracts of T. spiralis muscle larvae due to a risk of cross-reactions with other helminth infections. Sensitivity of nearly 100% has been obtained for humans infected with T. spiralis when immunoglobulin G (IgG) is measured by ELISA (15). Serological tests measuring other classes of antibodies result in lower sensitivity (107); however, detection of specific IgM or IgA is indicative of a recent infection. The use of antigens from other stages (adults, newborn larvae) has not resulted in comparable serodiagnosis. Western blotting to determine the presence of antibodies to ES antigens in the range of 40 to 70 kDa has been used as a confirmatory test. The sandwich ELISA has also been used to detect circulating antigens in sera (70). Trichinellosis patients typically become seropositive between the second and fifth week of infection, although this time will vary with the species of Trichinella. The time required for seroconversion is inversely correlated with the infective dose. Thus, it is advisable to take multiple serum samples at intervals of several weeks in order to demonstrate seroconversion in patients whose sera were initially negative or to detect rising titers. Once patients become seropositive, antibody levels do not correlate with the severity of the disease (107). Patients typically remain seropositive for several years, but sero­ positivity has been reported to last for up to 35 years following infection (30, 83).

SMP_Food Microbiology_CH26.indd

680

The efficacy of treatment of trichinellosis depends on the intensity of infection, the species of Trichinella, the stage of infection, and the character and intensity of the host response. The purpose of treatment during the intestinal phase is to destroy adult worms and to interfere with the production of newborn larvae. The drug of choice is mebendazole (Vermox) at dosages of 200 to 400 mg three times a day for 3 days and then 400 to 500 mg three times a day for 10 days (52, 82). Mebendazole is also believed to be active against developing and encysted larvae but at dosages higher than those used against adult worms. Mebendazole is administered during the first week of infection before newborn larvae migrate, in moderate or severe infections in combination with corticosteroids, and in infections with T. nativa, which responds poorly to treatment with nonbenzimidazole compounds. Mebendazole is not recommended for women in the first trimester of pregnancy. Thiabendazole (Mintezol) is no longer used because of adverse reactions. Albendazole (Valbazen and Zentel) is better absorbed and is probably as active as or even more active than mebendazole (82). Pyrantel (Antiminth) at 10 mg/kg of body weight per day for 4 days and levamisole at 2.5 mg/kg/day (maximum of 1 to 50 mg) are active only against adult worms in the intestine. To minimize hypersensitivity, it is recommended that corticosteroids be given in combination with anthelmintic drugs. Corticosteriods are recommended for acute severe trichinellosis not only for antiallergic action but also for anti-inflammatory and antishock actions. Supportive therapy, such as bed rest, is very important.

Prevention and Control

Prevention of human exposure to Trichinella from domestic pigs can be accomplished in a variety of ways: (i) by prevention of infection on the farm (preharvest), (ii) by testing animals at slaughter, (iii) by postslaughter processing using methods that have been proven effective for killing Trichinella larvae in meat, and (iv) by proper cooking by the consumer. The recommended application of these methods for the control of trichinellosis in pork has been summarized by the International Commission on Trichinellosis (53) and in guidelines published by the World Organisation for Animal Health (OIE) (28).

Prevention of Infection on the Farm

Knowledge of the routes of transmission of Trichinella to domestic pigs has allowed for the design of swine

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

681

management systems that essentially eliminate the risk of exposure of pigs to this parasite. By following a series of good management practices, including biosecure housing, effective rodent control, and good manufacturing and storage practices for feed, it is possible to certify the safety of pork without subsequent slaughter inspection or further processing. The requirements for producing pigs under conditions that greatly reduce or preclude exposure to Trichinella infection are outlined by several published sources. The International Commission on Trichinellosis provides guidance for Trichinella-free pig farming in a document entitled “Recommendations on Methods for the Control of Trichinella in Domestic and Wild Animals Intended for Human Consumption” (53). The essential elements of these recommendations relative to Trichinella-free pig farming are reiterated in the U.S. Trichinae Herd Certification Program Standards (155) and in Commission Regulation (EC) No 2075/2005 of the European Union (33). According to Regulation (EC) No 2075/2005, carcasses and meat of domestic pigs kept solely for fattening and slaughter are exempt from Trichinella examination at slaughter when the animals come from a holding (farm) or category of holdings that has been officially recognized by a competent authority as free from Trichinella.

Slaughter Testing

In virtually all developed countries, animals slaughtered for human consumption are examined by a federal or state inspector to determine if the animal is healthy and suitable for use as human food. In many countries, part of the inspection process for slaughtered pigs includes testing each pig carcass for Trichinella infection by using one of several approved direct methods of inspection. Inspection methods for Trichinella infection have historically been performed by a trichinoscope examination, in which a piece of muscle is compressed between two glass slides and scanned under a microscope. This method is both tedious and ineffective, since it has a sensitivity of only about 5 larvae per gram of tissue and nonencapsulating species of Trichinella are very difficult to visualize (49); consequently, it is no longer recommended for routine use in testing pigs. The development of a high-volume slaughter test, the pooled sample digestion method, takes advantage of the ability of Trichinella larvae to survive treatment with acidified pepsin (as they do in the stomach of a host). Samples of tissue collected from sites of parasite predilection (diaphragm, cheek muscle, tongue) are subjected to digestion in acidified pepsin. Larvae, freed from their muscle cell capsules, are recovered by a series of sedimen-

SMP_Food Microbiology_CH26.indd

681

tation steps and then visualized and enumerated under a microscope. Requirements for performing the digestion test are found in the Directives of the European Union (33), in the U.S. Code of Federal Regulations (154), and in the OIE Manual of Diagnostic Tests and Vaccines for Terrestrial Animals (117). Using methods of inspection testing as practiced on pig carcasses, the sensitivity of the digestion method is approximately 3 larvae per gram of tissue (48, 51). This level of detection is considered effective for identifying pork that poses a significant public health risk. Although there is insufficient information to determine the exact number of larvae that are necessary to cause clinical human disease (and these figures will be affected by the type of Trichinella, the amount of meat eaten, and the health of the individual), it is generally considered that infections of >1 larvae per gram of tissue are a public health risk. Thus, most infections that could cause clinical human disease would be detected by currently employed direct testing methods. Through the application of rigorous slaughter testing using artificial digestion methods, Trichinella infection has been virtually eliminated from domestic pigs in some countries. However, this effort is not without cost, with an estimated $570 million spent annually by the European Union for Trichinella inspection (127). In some countries, testing programs are in place, but not all pigs are processed through official slaughter facilities. These gaps in veterinary control are one reason that trichinellosis remains a significant public health problem.

Postslaughter Meat Processing

Pork products that are considered ready to eat, and have not otherwise been demonstrated to be free from Trichinella infection, must be processed by heating, freezing, or curing to kill any potential Trichinella larvae. Commercial processing methods that have been proven experimentally to render pork free from infective Trichinella larvae are described in the U.S. Code of Federal Regulations (154). Pork meat must be heated to 58°C or frozen at one of several time-temperature combinations. With flash freezing, Trichinella larvae are killed instantaneously at –35°C. The effectiveness of curing depends on a combination of salt concentration, temperature, and time. Each method should be tested experimentally to determine effectiveness as no model for curing conditions has been devised.

Proper Cooking

The most effective measure to prevent trichinellosis caused by fresh pork from areas where the disease is endemic is education of the public. The responsibility lies

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

682 with consumers to ensure that larvae are killed before the pork is eaten. The U.S. Department of Agriculture recommends that fresh pork be cooked to an internal temperature of 145°F (63°C) before consumption in the home (107). Rapid cooking methods, such as the use of microwave ovens, may not heat the pork uniformly or for sufficient time to destroy the larvae (84). Commercial cooking of pork products requires lower temperatures to ensure inactivation of Trichinella, presuming that temperature is more closely controlled by commercial meat processors. The U.S. Code of Federal Regulations (154) requires that pork be frozen for 20 days at 5°F (–15°C), 10 days at –10°F (–23°C), or 6 days at –20°F (–29°C) to kill the larvae, provided that the meat is less than about 6 in. (15 cm) thick. These freezing temperature and time requirements are not effective for killing the arctic isolate T. nativa or other freeze-resistant types of Trichinella.

TAENIASIS Until the mid-1980s, it was believed that taeniasis was the result of ingesting tissue cysts of either T. saginata from cows or T. solium from pigs; however, shortly thereafter, Fan (34) identified and characterized a third human taeniid with adult morphological characteristics similar to those of T. saginata and larval characteristics more in line with those of T. solium. Pigs were putatively identified as the intermediate host for this unique taeniid that had a predilection for liver tissue in a multitude of experimental hosts. Since that time, controversy has arisen over the classification and therefore the naming of this third human taeniid. For this reason, both Taenia asiatica (32) and Taenia saginata asiatica (41) have appeared and continue to appear in the literature as the scientific names identifying this organism. However, sufficient morphological, biological, and genetic differences exist to warrant use of the name T. asiatica, which better defines this organism as a unique lineage and eliminates confusion in comparisons of the disparate epidemiological information and clinical manifestations associated with this organism and T. saginata. Recent sequencing of the entire mitochondrial genome of T. asiatica (74) offers further genetic support for its unique classification. In addition, a DNA sequence-based study using formalinfixed samples dating back to 1929 clearly showed a partitioning of the taeniids into T. saginata, T. solium, and a distinct third species, T. asiatica, from the East (73). In Korea and China, these were found to be sympatric (72). For these reasons, the term T. asiatica will be used to define this species here (67).

SMP_Food Microbiology_CH26.indd

682

T. saginata Taeniasis

The history of taeniids has been described in the ancient literature dating back to 1500 b.c., with intriguing theories as to the parasites’ origin and nature. However, it was not until 1782 that Goeze (58) identified a difference between the bovine and swine parasites, which was later confirmed in 1857 by Küchenmeister based on the morphologies of the scolices (85). In 1861, Leuckart was the first to identify a relationship between the adult worm in humans and the larval bladder worm in cattle.

Life Cycle of T. saginata

The life cycle of T. saginata begins with cattle ingesting T. saginata eggs during grazing. The eggs hatch in the duodenum, liberating a six-hooked embryo that penetrates mesenteric venules or the lymphatics and reaches skeletal muscles or the heart, where it develops into the cysticercus larva. The cysticercus is essentially a miniature scolex and neck invaginated into a fluid-filled bladder that measures about 10 by 6 mm. The bladder larva becomes infective within 8 to 10 weeks following ingestion and can remain infective for more than a year. A person eating raw or poorly prepared beef harboring viable larval cysts (often referred to as cysticercus bovis) is subject to infection whereby the larvae are released from their surrounding muscle tissues by digestion in the small intestine. The scolex of the cysticercus evaginates from the vesicle or bladder and attaches to the mucosa of the jejunum, where the larva develops into a mature adult worm in 8 to 10 weeks. A mature adult worm is characterized by a scolex with four hemispherical suckers of 0.7 to 0.8 mm in diameter situated at the four angles of the head. The entire strobila can grow to 17 m in length and possess 1,000 to 2,000 proglottids or segments immediately following the neck in a symmetric series of immature, mature, and gravid proglottids as they proceed posteriorly. The mature and gravid proglottids contain both male and female reproductive organs, resulting in self-fertilization when distal proglottids come together by folding of the worm’s body. The presence of only a single worm per definitive host suggests that the mechanism of genetic variation results from other than crossbreeding among adult worms. The scolex attaches to the mucosal surface of the upper jejunum by means of the four suckers, but because of its length, the entire worm might extend down to the terminal ileum. The developing proglottids extend down the small intestine, where the most distal gravid proglottids (20 mm long and 5 to 7 mm wide) detach singly from the rest of the strobila and independently migrate through the rectum to the outside. Each gravid

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

683

proglottid contains about 80,000 eggs, which are expressed from the proglottids and deposited on the perianal skin. When gravid proglottids come to rest on the ground, eggs are extruded. The eggs may also be present as a result of indiscriminate, promiscuous defecation. Cattle ingest T. saginata eggs during grazing to complete the cycle.

Epidemiology

Taeniasis results when raw or inadequately cooked beef containing tissue cysts is consumed. Raw or rare beef is popular in many countries, particularly in the form of beef tartare (basterma) in the Middle East; the equivalent of steak tartare (kitfo) in parts of Africa, especially Ethiopia; shish kebab in India; lahb in Thailand; and yuk hoe in Korea. Although the meat of reindeer in northern Siberia (81), as well as meat of other herbivorous animals, has been reported to contain cysticerci of T. saginata and may have been responsible for taeniasis in humans, the only evidence of wild intermediate hosts in the Americas has been a single report at the turn of century involving llama and pronghorn antelope (126). Thus, beef is the primary source of T. saginata taeniasis in humans. With respect to the incidence of taeniasis in humans, geographic areas can be classified into three groups: countries or regions where infection is highly endemic, with a prevalence in the human population exceeding 10%; countries with moderate infection rates; and countries with a very low prevalence (below 0.1%) or even no infections caused by the parasite. The areas of high endemicity are Central and East African countries, such as Ethiopia, Kenya, and Zaire; Caucasian South Central Asian republics of the former Soviet Union; Middle Eastern countries, such as Syria and Lebanon; and parts of the former Yugoslavia. A high incidence has been identified in Bali, Indonesia (163), where prevalence among three disparate villages was decidedly variable (1.1 to 27.5%). This is likely the result of a cultural shift that is replacing minced pork products with uncooked minced beef, even in a strongly Hindu nation. In older studies in the Asian republics of the former Soviet Union, the prevalence rate was shown to reach 45% (125). The predominant cestode found in Europe is T. saginata, and T. solium is much less represented (17), especially in areas such as Poland (160–162). The incidence is low in the United States, where only 443 patients were reportedly treated for taeniasis due to T. saginata in 1981 (109). More recently, a populationbased survey was performed in the southwestern United States along the border with Mexico (5) among neighboring cities. Results indicated a combined prevalence

SMP_Food Microbiology_CH26.indd

683

of 3%; however, compared with the residents of Juárez, Mexico, residents of El Paso, Texas, were 8.6-fold more likely to be tapeworm carriers. This difference between the United States and Mexico was attributed to more widespread use of anthelmintics in Mexican border towns. It should be noted that this study did not delineate between the pig and bovine tapeworms. Chronic persistence of and periodic increases in human T. saginata infections correlate with factors that govern infections in cattle. Increased risk of infection to cattle is directly associated with environmental factors and access to pastures contaminated with human feces or sewage effluent (17), as well as feed, irrigation ditches, or cattle pens contaminated by farm workers. In 2002, an outbreak on a feedlot in Alberta, Canada, resulted in the killing of nearly 3,000 animals, of which 67 were infected with T. saginata (89). In this instance, the water supply was implicated as the likely source of infection (146). Reports of cysticercosis, also known as “beef measles” because of the cystlike larvae in the muscle, continue to surface in U.S. cattle (101) as well. Federal regulations prohibit carcasses harboring as few as one dead larva from being approved for human consumption unless the carcass is first treated to destroy any remaining parasites. Generally, only about 5% of carcasses are found to be heavily infected and totally condemned. Eggs of T. saginata are capable of surviving for long periods in the environment and are resistant to moderate desiccation, disinfectants, and low temperatures (4 to 5°C). Longevity of 71 days in liquid manure, 33 days in river water, and 154 days on pasture has been reported (149). The eggs can also survive many sewage treatment processes and can remain viable in fluid effluent or dried sludge. Reviews (17, 26) summarize both biological and nonbiological treatments of sludge to reduce egg viability; however, a 2005 study in France (105) demonstrated that cows raised on pastures fertilized with liquid sludge did not acquire live cysticerci provided a 6-week delay occurred between application and grazing. Despite an apparently low rate of human taeniasis in many developed countries, the incidence of bovine cysticercosis seems not to have decreased in recent years. Among affluent Americans, infection has been attributed to greater freedom of international travel in addition to a general preference for exotic diets. The penchant for raw beef and steak tartare is not limited to affluent Americans and Europeans but is also found, for example, among Ethiopians who can afford beef. A combination of poor sanitation and culinary preference for raw and rare beef in many regions of high endemicity­

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

684 continues to contribute to a high risk of infection and transmission.

Pathogenesis and Pathology

The scolex of the adult worm generally is lodged in the upper part of the jejunum. Usually, only a single worm is present. Although multiple infections have been reported, these have come under scrutiny with the identification of the Asian species. It has been suggested that these multiple infections with T. saginata were actually misdiagnosed infections with T. asiatica, which is well documented as being capable of synchronically infecting human hosts (94). The adult tapeworm of T. saginata does not cause pathologic changes; however, mucosal biopsies have shown some inflammation, suggesting that the presence of the worm can trigger irritation that in turn results in bowel distention or spasm. Migration to unusual sites is rare, but complications such as appendicitis, invasion of pancreatic or bile ducts, intestinal obstruction, or perforation and vomiting of proglottids with aspiration have been reported.

Clinical Manifestations

Although most cases of taeniasis are asymptomatic, up to one-third of patients complain of nausea or abdominal “hunger” pain that is often relieved by eating. Epigastric pain may be accompanied by weakness, weight loss, increased appetite, headache, constipation, dizziness, and diarrhea. The patient usually becomes aware of the infection when a proglottid is passed in the stool or is found on the perianal area or even on underclothing. The adult worm is weakly immunogenic, as manifested by moderate eosinophilia and increased levels of IgE in serum. Allergic reactions, such as urticaria and pruritus, may be due to the worm and its metabolites. The adult worms induce the production of antibodies. The persistence for years of a large, actively growing worm reflects lack of protective immunity to resident worms but is consistent with concomitant immunity observed in other helminth infections.

Diagnosis

Definitive diagnosis is based on the proglottid, since the eggs of T. saginata cannot be distinguished from those of other species of Taenia or those of Multiceps and Echinococcus species. The gravid proglottid of T. saginata has 15 to 20 lateral branches of the uterus on each side of the main uterine stem, a characteristic feature (Fig. 26.4); however, a similar range of uterine branches has been identified in T. asiatica as well, suggesting that the geographic source of the infection may play a role in diagnosis if uterine branches are used in the analysis. If

SMP_Food Microbiology_CH26.indd

684

Figure 26.4  Gravid proglottid of T. saginata. Note at least 16 lateral uterine branches. doi:10.1128/9781555818463.ch26f4

the gravid proglottid is treated with 10% formaldehyde and injected with India ink, the uterine branches are very prominent. Uterine branches also can be seen by gently pressing the proglottid between two microscope slides and holding them in front of a bright light. If the scolex is present, the four characteristic hookless suckers can be used as a distinguishing feature for identification. The egg is nearly spherical in shape, measuring 30 to 40 mm in diameter; has characteristic radial striations on its thick shell; and contains a hexacanth embryo with delicate lancet-shaped hooklets (Fig. 26.5). Rather than looking for the eggs in the stool, it is better to use the commercial Scotch tape method to obtain the eggs or proglottids from the perianal region. Since detection of eggs is nonspecific in terms of species, DNA probes that allow differentiation of tapeworm species by hybridization techniques were among the first DNA techniques to be developed (46, 63). However, more recent advancements use PCR to discriminate among various Taenia species. The amplification of a 0.55-kb DNA fragment provides a unique T. saginata genomic DNA template (60). T. saginata can also be easily delineated from T. asiatica by differential amplification of external transcribed rRNA gene repeat fragments (173) and multiplex PCR (59). Restriction fragment length polymor-

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

685

Prevention and Control

Figure 26.5  T. saginata egg (×590). doi:10.1128/9781555818463.ch26f5

phism (RFLP), PCR-linked RFLP, random amplified ­polymorphic DNA, single-strand conformation polymorphism, and microsatellite analysis are other techniques that are available for differentiating species (for reviews, see Yamasaki et al. [172] and Ito et al. [69]). Current technologies will also allow the amplification of specific regions of DNA from eggs or proglottids followed by sequencing and comparison to other common taeniids for confirmatory diagnosis. Although an antibody response is induced by the adult worm, specific serological tests such as those used for the diagnosis of trichinellosis have not been routinely used for diagnosis of taeniasis. This is predicated upon the fact that taeniasis induced by T. saginata is not life-threatening and proglottids shed in human feces remain the easiest and most definitive test. However, preliminary data have shown that a <12-kDa antigen prepared from the cyst fluid of Taenia hydatigena may be capable of differentiating between the types of human taeniasis (64).

Treatment

The drug of choice in treating taeniasis is a single oral dose of 5 to 10 mg of praziquantel (Biltricide) per kg of body weight for adults or children. This has been shown to achieve very high success rates (145). A good alternative is niclosamide (Niclocide and Yomesan) at 2 g orally once for adults and 50 mg/ kg orally once for children. This is a taeniacide that is effective in damaging the worm to such an extent that a purge following therapy will often produce the scolex. Other drugs such as albendazole and mebendazole can also be used but only under specialized circumstances.

SMP_Food Microbiology_CH26.indd

685

In addition to effective drug therapy against the adult Taenia worms, there are measures that can be taken to prevent cysticercosis in cattle. The most important of these measures is adequate sewage disposal so that tapeworm eggs are unavailable to cattle. Since humans are the only definitive hosts of T. saginata and thus the only disseminators of eggs, which can be shed at a rate of 480,000 to 720,000 daily, the success of this measure depends on educating livestock producers and their employees about modes of transmission. Furthermore, examining workers’ stools for Taenia species and providing adequate toilet facilities in cattle feeding establishments will also help to reduce the incidence of infection. Where toilet facilities are available, overloading of systems should be avoided. In addition to education and monitoring, prevention should include sanitary protection of cattle feed as well. Bovine cysticercosis resulting in extensive carcass condemnation has resulted from feeding with cottonseed harvested from an area where an overtaxed municipal water treatment facility and hard rains combined to contaminate the feed with T. saginata eggs. Currently, meat inspection is used to detect cysticercosis in slaughtered cattle and thereby reduce taeniasis. Although practiced extensively, this methodology has limitations. Even routine examination of the heart and masseter muscles can miss a significant percentage of infected cattle. It is estimated that as few as 25% of infected cattle are detected by methods currently employed for meat inspection (139). This is due to assumptions regarding predilection sites in host tissues, the ability to use the heart as a representative site for whole-body infections, and the generally low number of cysts that occur in any animal subject to low-level, environmentally derived infections. In the United States, the percentage of cattle inspected by state and municipal authorities varies. Cattle from small farms where sanitary conditions can be poor are frequently slaughtered in local abattoirs without inspection requirements. Clearly, all infected carcasses should be condemned; however, approximately 70% of infected carcasses are trimmed of visible cysts and then passed unrestricted for consumption (147). Currently, the U.S. Code of Federal Regulations (9CFR, §318.10) (154) requires infected carcasses to be frozen for a minimum of 10 days at 15°F (–9.4°C) or 5 days at 0°F (–18°C) to kill the organism (124). Nonetheless, the combination of the practice of trimming the cysts with the less than 100% animal inspection rate and the poor rate of identification at the abattoir is consistent with a low level of positive identification.

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

686 An alternative approach for assessing cysticercosis in cattle is a sensitive serological test (144). ELISA techniques have been evaluated over the years (1, 24, 25, 116, 157), but the efficacy of these methods has not been proven to be superior to that of physical inspection. Antibody tests have been based upon recombinant proteins, peptides, and secreted parasite antigens (ES) (1, 116, 157). Antigens secreted from metacestodes have generated reasonable results; however, using ­parasite-derived natural products severely limits the use of such a test. Also, the association between direct and serological methods is poor primarily because of the disparate immune responses among the hosts (166) and the lack of correlation between the level of parasitism and the level of the immune response. Serological testing might be useful for epidemiological purposes, especially if cysticercosis is suspected in a cattle herd. However, no large-scale study has been performed to validate the diagnostic efficacy of native or recombinant antigens with respect to sensitivity and specificity for detection of infection. The development of an effective cattle vaccine for use in areas of high endemicity would not only prevent initial infection and reduce human infections but also protect cattle against challenge infections. Recombinant vaccines for bovine cysticercosis and their orthologous antigens for swine cysticercosis have been tested with remarkable success (90, 92, 93); however, economic and logistical issues have thus far deterred application of vaccines as a prophylactic measure for bovine cysticercosis. On the part of the consumer, public health education concerning the risks of eating raw or inadequately cooked beef is important. The cysticerci in meat are inactivated by freezing meat at –10°C for 10 days or –18°C for 5 days, heating to an internal temperature of 56°C, or salt curing under appropriate conditions. Major obstacles to this approach have been reluctance by consumers to modify preferred culinary habits and to break with long-established cultural traditions. Thus, attempts to eradicate T. saginata taeniasis have been unsuccessful.

Life Cycle of T. solium

T. solium Taeniasis

Epidemiology

Taeniasis caused by T. solium was known at the time of Hippocrates. The Greeks described the larval stage in the tongue of swine as resembling a hailstone. In contrast to T. saginata, T. solium, the pork tapeworm, possesses an armed rostellum, has a smaller number of lateral uterine branches, and lacks a vaginal sphincter (136). Its phylogenetic classification as a unique species in 1758 is attributed to Linnaeus.

SMP_Food Microbiology_CH26.indd

686

When pork containing a viable cyst of the larval stage of T. solium (often referred to as cysticercus cellulosae) is ingested, the head of the larva evaginates from the fluid-filled milky white bladder. The scolex bears four suckers and an apical crown of hooklets. The cysticerci are referred to as pork measles and are larger (5 to 20 mm in diameter) than cysticercus bovis. They attach to the wall of the small intestine and mature into adult worms in 5 to 12 weeks. The adult worm measures up to 8 m (usually 1.5 to 5 m) and has a scolex that is roughly quadrate, possessing a conspicuous rounded rostellum armed with a double row of large and small hooklets, numbering 22 to 32. A short cervical region is anterior to a series of proglottids or segments. Immature proglottids are broader than they are long, mature ones being nearly square, while gravid ones are longer than they are broad. The total number of distinct proglottids is less than 1,000. The terminal proglottids become separated from the rest of the strobila and migrate out of the anus or are passively expelled in the stool. A single gravid proglottid contains fewer than 50,000 eggs. Upon escape from the ruptured uterus of the gravid proglottid and after deposition on the soil, the eggs may remain viable for many weeks. The eggs of T. solium are more apt than those of T. saginata to appear in the stool. In the normal cycle, the eggs are ingested by pigs, the usual intermediate host. The hexacanth embryo hatches in the duodenum, migrates through the intestinal wall to reach the blood and lymphatic channels, and is carried to the skeletal muscles and the myocardium. The embryo develops in 8 to 10 weeks into a cysticercus. In the abnormal cycle, humans can serve as intermediate, but terminal, hosts and harbor cysticerci acquired by accidental ingestion of eggs or by the autochthonous cycle (an infection that results from the movement of eggs or gravid proglottids from the intestine back into the stomach by reverse peristalsis). The cysticercus develops most commonly in striated muscles and subcutaneous tissues but also in the brain, eye, heart, lung, and peritoneum. T. solium taeniasis has a worldwide distribution and is important in all countries where pork is consumed, especially those with suboptimal practices of sanitation and pig husbandry. Humans are the only known natural definitive host for this parasite. It is very difficult to evaluate the prevalence of human infection because coproscopical methods cannot differentiate between infections caused by T. saginata and those caused by

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

687

T.  solium. Serological methods have been developed for this purpose but have not yet been tested in large field studies. Infections resulting from T. solium have much more serious consequences than those caused by T. saginata because the larval or cysticercus stage of T. solium, which infects pigs, can also infect humans. Cysticercosis in humans is an important public health problem in developing countries where hygiene and sanitary conditions are deficient or nonexistent and where swinerearing practices are primitive (122). The relatively high prevalence of T. solium taeniasis in the Bantu population of South Africa has been attributed to a widespread practice of using tapeworm proglottids as a component of muti, a medication used by native herbalists for the treatment of intestinal worms (65). Infection of humans with larvae results from inadvertent ingestion of eggs in contaminated water or foods, e.g., vegetables fertilized with night soil (human excrement), from indirect contact with hands of individuals already infected with the adult worm (external autoinfection), or from direct contact with a tapeworm carrier. Another important mode of transmission involves patients infected with adult tapeworms, in which the eggs are carried by reverse peristalsis back to the duodenum or stomach, where they are stimulated to hatch and subsequently invade extra­intestinal tissues to become cysticerci (internal auto­ infection). This is a key cause of neurocysticercosis. The prevalence of human neurocysticercosis is ­increasing in some parts of the world, and a considerable body of literature is devoted to the diagnosis, treatment, and prevention of this disease (55, 151, 167). The relevance of neurocysticercosis as a public health problem in the United States has been highlighted by reports of autoch­ thonous cases (159). Furthermore, with the increase in illegal immigration and the number of migrant workers from Central America, cysticercosis has garnered attention as a neglected infection in the United States that is linked to poverty (141). In some parts of the world such as in Muslim countries (125), human infection is practically nonexistent; however, it is quite common in Mexico, Central and South America, Central and South Africa, Asia, non-Islamic Southeast Asia, Slavic countries, and Southern and Eastern Europe. Prevalence rates of cysticercosis have been reported to be 6.1% in Pondicherry, South India (123), 7 to 21% among different provinces of Madagascar (3), 6 to 9% based upon viable cysts or 4 to 36% based upon serology in three rural Venezuelan communities (42), 0.2 to 7.2% in central and northern provinces of Vietnam (169), 21% in rural Peruvian communities (103), and 0.1 to 8% in Western and Central Africa (175). Although it is present

SMP_Food Microbiology_CH26.indd

687

in Mexico, current evidence suggests that the incidence and severity of infection did not increase between 1994 and 2009 (45), offering encouragement for advancing efforts to control this disease. It should be noted that most of these studies, as well as other similar studies, have used indirect serology-based assays, which are prone to false-positive and false-negative results. In general, specific habits of eating raw or undercooked pork are linked to foci of taeniasis; however, in some regions there is no link between porcine and human cysticercosis where non-pork eaters have as great a chance of contracting the disease as pork eaters (97). An interesting development is that reports worldwide are showing an association between the presence of neurocysticercosis and the incidence of epilepsy, suggesting that many infected patients are being misdiagnosed as epileptics (16, 112, 114, 141).

Pathogenesis and Pathology

As in T. saginata infection, there is usually only a single adult worm, and pathologic changes are similar. There is mild local inflammation of the intestinal mucosa as a result of attachment by the suckers and especially the hooklets. However, because of its smaller size, T. solium is less likely to cause intestinal obstruction. Rare instances of intestinal perforation with secondary peritonitis and gallbladder infection have been reported. Pathologic changes due to cysticerci can be serious, depending on the tissue invaded and the number of cysticerci that become established. Damage results from pressure caused by encapsulated larvae on surrounding tissue, since the cysticerci produce occupying lesions. There may be no prodromal symptoms or only slight muscular pain and a mild fever during invasion of the muscles and subcutaneous tissues. In ocular cysticercosis, which accounts for about 20% of neurocysticercosis cases, there may be loss of vision. Invasion of the meninges, cortex, cerebral substance, and ventricles evokes tissue reactions leading to focal epileptic attacks or other motor or sensory involvement. The reasons for the predilection of cysticerci for the central nervous system are still obscure.

Clinical Manifestations

Since only a single adult worm is usually present in T. solium taeniasis, there are no symptoms of epigastric fullness. Patients are asymptomatic and become aware of the infection only when they find proglottids in their stools or on perianal skin. However, there may be vague abdominal discomfort, hunger pains, anorexia, and nervous disorders. Rare instances of intestinal perforation with secondary peritonitis and gallbladder infection

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

688 have been reported. Eosinophilia, as high as 28%, and leukopenia can occur. The persistence for years of actively growing tapeworms does not appear to be consistent with the development of protective immunity. As mentioned earlier, seizures that often accompany forms of neurocysticercosis are to a high degree misdiagnosed as indicative of epilepsy primarily because of the inability to adequately diagnose the infection.

Diagnosis

Diagnosis is based on examination of stool specimens and perianal scrapings. The sensitivity of both methods is increased by three examinations daily. Since T. solium eggs cannot be distinguished from those of T. saginata, specific diagnosis is based on the identification of the gravid proglottid, which has fewer lateral branches (7 to 12) on each side of the main uterine stem (Fig. 26.6) than that of T. saginata. If the scolex is obtained, it possesses hooklets in addition to four suckers. The development of DNA probes (63), and more recently the use of PCR (66, 98, 99, 115), multiplex PCR (171), and loop-mediated isothermal amplification (113), have made it possible to distinguish T. solium from T. saginata. Neurocysticercosis, on the other hand, is becoming a significant public health issue in the United States. Clinical presentation is unpredictable but often accompanied by neurologic symptoms, intracranial hypertension, mental changes, or sudden unexplained death among individuals at highest risk for disease (for a review, see reference 104). Histologic studies are always the most definitive; however, whole-body imaging (86), neuro­imaging (56, 120), and other lab results can assist in final diagnosis and treatment. Clearly, image scanning is the most sensitive; however, a recent study using cerebrospinal fluid to compare antibody and PCR-based diagnostic tests showed that PCR was the most sensitive of the tests (95.9%), but with variable specificity (80 or 100%), and that most antibody/antigen-based tests were of similar sensitivity (95). The sensitivity of serological tests for taeniasis and cysticercosis varies, depending on the particular method and the clinical form of infection. Efforts have been made to improve the sensitivity of serology-based modern methods (43, 44, 156, 168), although emphasis on the use of serology has shifted primarily to the diagnosis of cysticercosis rather than taeniasis. A highly specific immunoassay for diagnosis of cysticercosis, using the <12-kDa T. hydatigena antigen, was reported (64) to be so specific as to distinguish between human clinical cases of cysticercosis and taeniasis; however, this has never been pursued. Furthermore, as with T. saginata, obtaining sufficient parasite antigen poses logistical problems in the absence of a recombinant product.

SMP_Food Microbiology_CH26.indd

688

Figure 26.6  Gravid proglottid of T. solium. Note fewer lateral uterine branches than in T. saginata in Fig. 26.4. doi:10.1128/9781555818463.ch26f6

Treatment

Niclosamide is the drug of choice because of its effectiveness against the scolex and proliferative zone of the strobila. Single doses of praziquantel and mebendazole have also been reported to be effective. It is imperative to treat patients harboring the adult worm, since cysticercosis can occur from internal autoinfection. The major concerns in treating patients with the adult worm are to prevent vomiting and to ensure rapid expulsion of disintegrated proglottids from the intestine.

Prevention and Control

Prevention and eradication of T. solium adults in the intestine and of cysticerci in various tissues in humans and in pigs is a concern in areas of endemicity where economic, social, and sanitary conditions are substandard. Needed changes are monumental in scope. Without fundamental changes, the most scrupulous personal hygiene and eating habits will not prevent or eradicate infection in underdeveloped or developing countries. In developed countries, infection can be avoided by adherence to modern animal husbandry practices. The best preventive measure in interrupting transmission from humans to animals is to introduce and maintain proper sanitary facilities to dispose of contaminated feces. Even with proper toilet facilities, measures must be taken to make sure that sewage treatment is adequate to kill the eggs. Since one infected individual can infect literally thousands of pigs via contaminated feedlots, management personnel and employees must be educated regarding the parasite and means of avoiding transmission. Prospective employees who will come in contact with animals should be examined for tapeworm infection before employment and semiannually during employment, and chemical toilets should be installed and properly maintained at convenient locations on slaughtering premises. The most

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

689

important practice is to keep pigs in enclosures or indoors to prevent access to human fecal matter. Meat inspection programs are only partially effective for identifying and condemning infected carcasses. Meat inspection for measly pork is performed in the United States for animals intended for interstate commerce, but inspection can miss lightly infected carcasses. Thus, the consumer should make sure that meat is cooked to an internal temperature of 56°C or higher or is frozen at –10°C for at least 14 days to kill cysticerci (109). Other means of inactivating cysticerci in pork, such as irradiation, have not been commercially applied. The best preventive measure is to avoid eating raw or uninspected pork. With the idea that eliminating swine cysticercosis will greatly assist in eradicating human taeniasis and neurocysticercosis, efforts to immunize pigs with recombinant antigens have met with remarkable success in both laboratory studies and field trials (90–93). Unfortunately, vaccination as a means of prevention and control is subject to problems other than a lack of efficacy. Economic and regulatory issues as well as dissemination and stability of the vaccine also come into play. As a result, a swine vaccine has yet to be made commercially available.

Asian Taeniasis

Asian taeniasis was first described in the aboriginal population in the mountainous regions of Taiwan (34). Initially, the etiologic agent was considered to be T. saginata because of morphological similarities among the adult worms, although notable differences, including shorter length, fewer proglottids, wider diameter of the scolex, and fewer testes in the mature proglottid of the agent, were found (34). Using cloned rRNA fragments and sequence amplification by PCR, Zarlenga et al. (173) were the first to show that T. asiatica is similar but genetically distinct from T. saginata. Later, sequence variation in the 28S rRNA and mitochondrial cytochrome c oxidase I genes and RFLP pattern variation in the cytochrome c oxidase I and rRNA internal transcribed spacer confirmed T. asiatica as genetically distinct from other taeniid cestodes (12). More recently, PCR (59, 72) and multiplex PCR (71) have been used to unequivocally differentiate T. saginata from T. asiatica. Although further studies are needed, the close relationship between T. asiatica and T. saginata in the absence of neurocysticercosis among those populations exhibiting the highest prevalence of infection suggests that T. asiatica is not likely a cause of human cysticercosis (50). Claims have been made suggesting that T. asiatica and T. saginata are capable of hybridization (118); however, given that both these parasites show substantial genetic variation among geographic isolates, hybridization can

SMP_Food Microbiology_CH26.indd

689

be difficult to demonstrate in other than an experimental setting.

Epidemiology

T. asiatica infection has been found in several Asian countries, notably the mountainous regions of Taiwan, Cheju Island of Korea, and Samosir Island of Indonesia. It is estimated that public health costs in these areas alone exceed $35 million annually (37). In Taiwan, of 1,661 aboriginal cases of Asian taeniasis, the overall clinical infection rate was 76% among nine aboriginal tribes in 10 counties in mountainous areas (39). Multiple studies since that time suggest that the infection rate is quite variable among the aboriginal populations of Taiwan; nonetheless, the rate remains comparatively high. Pigs, cattle, goats, wild boars, and monkeys can serve as intermediate hosts, and this has been validated in experimentally infected animals (35). Of these, the wild boar appears to be the probable natural intermediate host in Taiwan. This may be true also in Indonesia, where people have become infected, presumably from consuming pork (22). The cysticercus of T. asiatica is armed with tiny rostellar hooklets like that of T. solium and develops in a period shorter than that required for either T. saginata or T. solium. Interestingly, it is found mainly in the parenchyma of the liver (34), whereas cysticerci of T. saginata and T. solium are found primarily in the muscles of cattle and pigs, respectively. Thus, the custom of eating the viscera, especially the liver, of freshkilled animals appears to be a major contributing factor to transmission.

Clinical Manifestations

Infected individuals pass proglottids in their feces, even for 30 years or more, suggesting that the life span of this form of Taenia is very long (39). Common clinical manifestations include pruritis ani, nausea, abdominal pain, and dizziness. Abdominal pain is usually localized along the midline of the epigastrium or in the umbilical region and varies in intensity from a dull, aching, gnawing, or burning feeling to an intense, colic-like, sharp pain. Infection is often accompanied by a change in appetite (positive or negative), headache, diarrhea, and/or constipation. In contrast to human infection with T. solium and T. saginata, humans have been reported to be capable of harboring multitudes of adult worms of T. asiatica (94).

Treatment

Clinical trials performed some time ago in Taiwan have shown that a single dose (150 mg) of praziquantel was highly effective against T. asiatica infection (36, 38).

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

690 This treatment regimen remains current today. As with T. saginata, niclosamide is also effective.

Prevention and Control

The best preventive measure is to avoid eating raw, uninspected pork or other meat that contains the cysticerci. Since many cases of Asian taeniasis have been reported to result from the consumption of infected pork (40), the meat should be cooked thoroughly or should be frozen for at least 4 days at –10°C to kill the cysticerci, the same procedures used to prevent T. solium taeniasis.

References

















1. Abuseir, S., M. Kühne, T. Schnieder, G. Klein, and C. Epe. 2007. Evaluation of a serological method for the detection of Taenia saginata cysticercosis using serum and meat juice samples. Parasitol. Res. 101:131–137. 2. Andrews, J. R. H., R. Ainsworth, and D. Abernethy. 1994. Trichinella pseudospiralis in humans: description of a case and its treatment. Trans. R. Soc. Trop. Med. Hyg. 88:200–203. 3. Andriantsimahavandy, A., V. E. Ravaoalimalala, P. Rajaonarison, P. Ravoniarimbinina, M. Rakotondrazaka, N. Raharilaza, D. Rakotoarivelo, M. Ratsitorahina, L. P. Rabarijaona, C. E. Ramarokoto, P. Leutscher, and R. Migliani. 2003. The current epidemiological situation of cysticercosis in Madagascar. Arch. Inst. Pasteur Madagascar 69:46–51. 4. Arriaga, C., L. Yépez-Mulia, N. Viveros, L. A. Adame, D. S. Zarlenga, J. R. Lichtenfels, E. Benitez, and M. G. Ortega-Pierres. 1995. Detection of Trichinella spiralis muscle larvae in naturally infected horses. J. Parasitol. 81:781–783. 5. Barton Behravesh, C., L. F. Mayberry, J. R. Bristol, V. M. Cardenas, K. D. Mena, J. Martínez-Ocaña, A. Flisser, and K. F. Snowden. 2008. Population-based survey of taeniasis along the United States–Mexico border. Ann. Trop. Med. Parasitol. 102:325–333. 6. Bellani, L., A. Mantovani, S. Pampiglione, and I. Fillippini. 1978. Observations on an outbreak of human trichinellosis in Northern Italy, p. 535–539. In C. W. Kim and Z. S. Pawlowski (ed.), Trichinellosis. University Press of New England, Hanover, NH. 7. Bessonov, A. S. 1981. Changes in the epizootic and epidemic situation of trichinellosis in the USSR, p. 365–368. In C. W. Kim, E. J. Ruitenberg, and J. S. Tepperna (ed.), Trichinellosis. Reedbooks, Chertsey, United Kingdom. 8. Blaga, R., B. Durand, S. Antoniu, C. Gherman, C. M. Cretu, V. Cozma, and P. Boireau. 2007. A dramatic increase in the incidence of human trichinellosis in Romania over the past 25 years: impact of political changes and regional food habits. Am. J. Trop. Med. Hyg. 76:983–986. 9. Boireau, P., I. Vallee, T. Roman, C. Perret, L. Mingyuan, H. R. Gamble, and A. A. Gajadhar. 2000. Trichinella in horses: a low frequency infection with high human risk. Vet. Parasitol. 93:309–320.

SMP_Food Microbiology_CH26.indd

690

10. Bommer, W., H. Kaiser, W. Mannweiler, H. Mergerian, and G. Pottkamper. 1985. An outbreak of trichinellosis in northern Germany caused by imported air-dried meat from Egypt, p. 314–317. In C. W. Kim (ed.), Trichinellosis. State University of New York Press, Albany, NY. 11. Bouree, P., J. L. Leymarie, and C. Aube. 1989. Epidemiological study of two outbreaks of trichinosis in France, due to horse meat, p. 382–386. In C. E. Tanner, A. R. Martinez-Fernandez, and F. Bolas-Fernandez (ed.), Trichinellosis. CSIC Press, Madrid, Spain. 12. Bowles, I., and D. P. McManus. 1994. Genetic characterization of the Asian Taenia, a newly described taeniid cestode of humans. Am. J. Trop. Med. Hyg. 50:33–34. 13. Britov, V. A., and S. N. Boev. 1972. Taxonomic rank of various strains of Trichinella and their circulation in nature. Vestn. Akad. Nauk. SSSR 28:27–32. 14. Bruschi, F., and K. D. Murrell. 2002. New aspects of human trichinellosis: the impact of new Trichinella species. Postgrad. Med. J. 78:15–22. 15. Bruschi, F. A. Moretti, D. Wassom, and D. PiergiliFioretti. 2001. The use of a synthetic antigen for the serological diagnosis of human trichinellosis. Parasite 8:141–143. 16. Bucardo, F., A. Meza-Lucas, F. Espinoza, R.C. GarcíaJerónimo, R. García-Rodea, and D. Correa. 2005. The seroprevalence of Taenia solium cysticercosis among epileptic patients in León, Nicaragua, as evaluated by ELISA and western blotting. Ann. Trop. Med. Parasitol. 99:41–45. 17. Cabaret, J., S. Geerts, M. Madeline, C. Ballandonne, and D. Barbier. 2002. The use of urban sewage sludge on pastures: the cysticercosis threat. Vet. Res. 33:575–597. 18. Campbell, W. C. 1983. Historical introduction, p. 1–30. In W. C. Campbell (ed.), Trichinella and Trichinosis. Plenum, New York, NY. 19. Centers for Disease Control. 1982. Common-source outbreaks of trichinosis—New York City, Rhode Island. MMWR Morb. Mortal. Wkly. Rep. 31:161–164. 20. Coordinating Group for Prevention and Treatment of Trichinosis, Harbin City. 1981. A Survey of Trichinosis Due to Eating Scalded Mutton. WHO/ HELM/82.5. World Health Organization, Geneva, Switzerland. 21. Cui, J., and Z. Q. Wang. 2001. Outbreaks of human trichinellosis caused by consumption of dog meat in China. Parasite 8:74–77. 22. Depary, A. A., and M. L. Kosman. 1990. Taeniasis in Indonesia with special reference to Samosir Island, North Sumatra. Southeast Asian J. Trop. Med. Public Health 22:239–241. 23. Dissamarn, R., and P. Indrakamhang. 1985. Trichinosis in Thailand during 1962–1983. Int. J. Zoonoses 12:257–266. 24. Dorny, P., I. Phiri, S. Gabriel, N. Speybroeck, and J. Vercruysse. 2002. A sero-epidemiological study of bovine cysticercosis in Zambia. Vet. Parasitol. 104:211–215.

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

691

25. Dorny, P., F. Vercammen, J. Brandt, W. Vansteenkiste, D. Berkvens, and S. Geerts. 2000. Sero-epidemiological study of Taenia saginata cysticercosis in Belgian cattle. Vet. Parasitol. 88:43–49. 26. Dumontet, S., H. Dinel, and B. Baloda. 1999. Pathogen reduction in biosolids by composting and other biological treatments: a literature review. Biol. Agric. Hortic. 16:409–430. 27. Dupouy-Camet, J. 2000. Trichinellosis: a worldwide zoonosis. Vet. Parasitol. 93:191–200. 28. Dupouy-Camet, J., and K. D. Murrell (ed.). 2007. FAO/WHO/OIE Guidelines for the Surveillance, Management, Prevention and Control of Trichinellosis. World Orga­nisation for Animal Health, Paris, France. 29. Dupouy-Camet, J., W. Kociecka, F. Bruschi, F. BolasFernandez, and E. Pozio. 2002. Opinion on the diagnosis and treatment of human trichinellosis. Expert Opin. Pharmacother. 3:1117–1129. 30. Dupouy-Camet, J., A. Paugam, G. De Pinieux, V. Lavarde, and A. Vieillefond. 2001. Trichinella murrelli: pathological features in human muscles at different delays after infection. Parasite 8:176–179. 31. Dworkin, M. S., H. R. Gamble, D. S. Zarlenga, and P. O. Tennican. 1996. Outbreak of trichinellosis associated with eating cougar jerky. J. Infect. Dis. 174:663–666. 32. Eom, K. S., and H. J. Rim. 1993. Morphologic descriptions of Taenia asiatica sp. Korean J. Parasitol. 31:1–6. 33. European Commission. 2005. Commission Regulation (EC) No 2075/2005 of 5 December 2005 laying down specific rules on official controls for Trichinella in meat. Off. J. Eur. Union L338:60–82. 34. Fan, P. C. 1988. Taiwan Taenia and taeniasis. Parasitol. Today 4:86–88. 35. Fan, P. C. 1990. Asian Taenia saginata: species or strain? Southeast Asian J. Trop. Med. Public Health 22:245–250. 36. Fan, P. C. 1995. Review of taeniasis in Asia. Chung Hua Min Kuo Wei Sheng Wu Chi Mien I Hsueh Tsa Chih 28:79–94. 37. Fan, P. C., and W. C. Chung. 1997. Sociocultural factors and local customs related to taeniasis in East Asia. Kao Hsiung I Hsueh Ko Hsueh Tsa Chih 13:647–652. 38. Fan, P. C., W. C. Chung, C. H. Chan, Y. A. Chen, F. Y. Cheng, and M. C. Hsu. 1986. Studies on taeniasis in Taiwan. V. Field trial on evaluation of therapeutic efficacy of mebendazole and praziquantel against taeniasis. Southeast Asian J. Trop. Med. Public Health 17:82–90. 39. Fan, P. C., W. C. Chung, C. Y. Lin, and C. H. Chan. 1992. Clinical manifestations of taeniasis in Taiwan aborigines. J. Helminthol. 66:118–123. 40. Fan, P. C., W. C. Chung, C. T. Soh, and M. L. Kosman. 1992. Eating habits of East Asian people and transmission of taeniasis. Acta Trop. 50:305–315. 41. Fan, P. C., C. Y. Lin, C. C. Chen, and W. C. Chung. 1995. Morphological description of Taenia saginata

SMP_Food Microbiology_CH26.indd

691

42.

43.

44.

45.

46. 47. 48.

49.

50. 51. 52.

53.

54.

asiatica (Cyclophyllidea: Taeniidae) from man in Asia. J. Helminthol. 69:299–303. Ferrer, E., Z. Cabrera, G. Rojas, M. Lares, A. Vera, B. A. de Noya, I. Fernandez, H. U. Romero, L. J. Harrison, R. M. Parkhouse, and M. M. Cortez. 2003. Evidence for high seroprevalence of Taenia solium cysticercosis in individuals from three rural communities in Venezuela. Trans. R. Soc. Trop. Med. Hyg. 97:522–526. Ferrer, E., M. M. Cortez, Z. Cabrera, G. Rojas, I. Davila, B. Alarcon de Noya, H. A. Perez, I. Fernandez, H. U. Romero, L. J. Harrison, R. M. Parkhouse, and T. Garate. 2005. Oncospheral peptide-based ELISAs as potential seroepidemiological tools for Taenia solium cysticercosis/neurocysticercosis in Venezuela. Trans. R. Soc. Trop. Med. Hyg. 99:568–576. Fleury, A., C. Beltran, E. Ferrer, T. Garate, L. J. Harrison, R. M. Parkhouse, E. Garcia, G. Fragoso, J. Costa-Cruz, G. Biondi, S. Agapejev, and E. Sciutto. 2003. Application of synthetic peptides to the diagnosis of neurocysticercosis. Trop. Med. Int. Health 12:1124–1130. Fleury, A., J. Moreno García, P. Valdez Aguerrebere, M. de Sayve Durán, P. Becerril Rodríguez, C. Larralde, and E. Sciutto. 2010. Neurocysticercosis, a persisting health problem in Mexico. PLoS Negl. Trop. Dis. 4:e805. Flisser, A., A. Reid, E. Garcia-Zepeda, and D. P. McManus. 1988. Specific detection of Taenia saginata eggs by DNA hybridisation. Lancet 2:1429–1430. Forbes, L. B. 2000. The occurrence and ecology of Trichinella in marine mammals. Vet. Parasitol. 93:321–334. Forbes, L.B., and A. A. Gajadhar. 1999. A validated Trichinella digestion assay and an associated sampling and quality assurance system for use in testing pork and horse meat. J. Food Prot. 62:1308–1313. Forbes, L. B., S. Parker, and W. B. Scandrett. 2003. Comparison of a modified digestion assay with trichinoscopy for the detection of Trichinella larvae in pork. J. Food Prot. 66:1043–1046. Galan-Puchades, M. T., and M. V. Fuentes. 2000. The Asian Taenia and the possibility of cysticercosis. Korean J. Parasitol. 38:1–7. Gamble, H. R. 1998. Sensitivity of artificial digestion and enzyme immunoassay methods for inspection of trichinellosis in pigs. J. Food Prot. 61:339–343. Gamble, H. R., and K. D. Murrell. 1988. Trichinellosis, p. 1018–1024. In W. Balows (ed.), Laboratory Diagnosis of Infectious Disease: Principles and Practice. SpringerVerlag, New York, NY. Gamble, H. R., A. S. Bessonov, K. Cuperlovic, A. A. Gajadhar, F. van Knapen, K. Noeckler, H. Schenone, and X. Zhu. 2000. International Commission on Trichinellosis: recommendations on methods for the control of Trichinella in domestic and wild animals intended for human consumption. Vet. Parasitol. 93:393–408. Gamble, H. R., E. Pozio, F. Bruschi, K. Nöckler, C. M. O. Kapel, and A. A. Gajadhar. 2004. International

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

692

55. 56.

57. 58. 59.

60.

61.

62.

63.

64.

65. 66.

67.

Commission on Trichinellosis: recommendations on the use of serological tests for the detection of Trichinella infection in animals and man. Parasite 11:3–13. Garcia, H. H., and O. H. Del Brutto. 2000. Taenia solium cysticercosis. Infect. Dis. Clin. North. Am. 14:97–119. Garcia, H. H., O. H. Del Brutto, T. E. Nash, A. C. White, Jr., V. C. Tsang, and R. H. Gilman. 2005. New concepts in the diagnosis and management of neurocysticercosis (Taenia solium). Am. J. Trop. Med. Hyg. 72:3–9. Garkavi, B. L. 1972. The species of Trichinella isolated from wild carnivores. Veterinariia 10:90–91. (In Russian.) Goeze, J. A. E. 1782. Versuch einer Naturgeschichte der Eingeweidewürmer thierischer Körper. P. Pape, Blankenberg, Germany. Gonzalez, L. M., E. Montero, N. Morakote, S. Puente, J. L. Diaz De Tuesta, T. Serra, R. Lopez-Velez, D. P. McManus, L. J. Harrison, R. M. Parkhouse, and T. Garate. 2004. Differential diagnosis of Taenia saginata and Taenia saginata asiatica taeniasis through PCR. Diagn. Microbiol. Infect. Dis. 49: 183–188. Gottstein, B., P. Deplazes, I. Tanner, and J. S. Skaggs. 1991. Diagnostic identification of Taenia saginata with the polymerase chain reaction. Trans. R. Soc. Trop. Med. Hyg. 85:248–249. Grencis, R. K., and D. Wakelin. 1985. Analysis of lymphocyte subsets involved in mediation of intestinal immunity to Trichinella spiralis in the mouse, p. 26–30. In C. W. Kim (ed.), Trichinellosis. State University of New York Press, Albany, NY. Haeghebaert, S., and E. Maillot. 1999. CommunityWide Outbreak of Trichinellosis: Tarn et Garonne, Haute Garonne, Tarn Districts, January–March 1998. Réseau National de Santé Publique, Rennes, France. Harrison, L. J. S., J. Delgado, and R. M. E. Parkhouse. 1990. Differential diagnosis of Taenia saginata and Taenia solium with DNA probes. Parasitology 100:459–461. Hayunga, E. G., M. P. Sumner, M. L. Rhoads, K. D. Murrell, and R. S. Isenstein. 1991. Development of a serologic assay for cysticercosis, using an antigen isolated from Taenia spp. cyst fluid. Am J. Vet. Res. 52:462–470. Heinz, H., and G. Macnab. 1965. Cysticercosis in the Bantu of southern Africa. S. Afr. J. Med. Sci. 30:19–31. Hernández, M., L. M. Gonzalez, A. Fleury, B. Saenz, R. M. Parkhouse, L. J. Harrison, T. Garate, and E. Sciutto. 2008. Neurocysticercosis: detection of Taenia solium DNA in human cerebrospinal fluid using a semi-nested PCR based on HDP2. Ann. Trop. Med. Parasitol. 102:317–323. Hoberg, E. P. 2006. Phylogeny of Taenia: species definitions and origins of human parasites. Parasitol. Int. 55:23–30.

SMP_Food Microbiology_CH26.indd

692

68. Hou, H. W. 1983. A Survey of an Outbreak of Trichinosis Caused by Eating Roasted Dog Meat. WHO/HELM/84.15. World Health Organization, Geneva, Switzerland. 69. Ito, A., M. Nakao, Y. Sako, K. Nakaya, T. Yanagida, and M. Okamoto. 2009. Taenia, p. 839–850. In D. Liu (ed.), Molecular Detection of Foodborne Pathogens. CRC Press, Boca Raton, FL. 70. Ivanoska, D., K. Cuperlovic, H. R. Gamble, and K. D. Murrell. 1989. Comparative efficacy of antigen and antibody detection tests for human trichinellosis. J. Parasitol. 75:38–41. 71. Jeon, H. K., J. Y. Chai, Y. Kong, J. Waikagul, B. Insisiengmay, H. J. Rim, and K. S. Eom. 2009. Differential diagnosis of Taenia asiatica using multiplex PCR. Exp. Parasitol. 121:151–156. 72. Jeon, H. K., K. H. Kim, J. Y. Chai, H. J. Yang, H. J. Rim, and K. S. Eom. 2008. Sympatric distribution of three human Taenia tapeworms collected between 1935 and 2005 in Korea. Korean J. Parasitol. 46:235–241. 73. Jeon, H. K., K. H. Kim, and K. S. Eom. 2011. Molecular identification of Taenia specimens after long-term preservation in formalin. Parasitol Int. 60:203–205. 74. Jeon, H. K., K. H. Lee, K. H. Kim, U. W. Hwang, and K. S. Eom. 2005. Complete sequence and structure of the mitochondrial genome of the human tapeworm, Taenia asiatica (Platyhelminthes; Cestoda). Parasitology 130:717–726. 75. Jongwutiwes, S., N. Chantachum, P. Kraivichian, P. Siriyasatien, C. Putaporntip, A. Tamburrini G. La Rosa, C. Sreesunpasirikul, P. Yingyourd, and E. Pozio. 1998. First outbreak of human trichinellosis caused by Trichinella pseudospiralis. Clin. Infect. Dis. 26:111–115. 76. Kapel, C. M. O. 2000. Host diversity and biological characteristics of the Trichinella genotypes and their effect on their transmission. Vet. Parasitol. 93:263–278. 77. Khamboonruang, C. 1990. The present status of trichinellosis in Thailand. Southeast Asian J. Trop. Med. Public Health 22:312–315. 78. Kim, C. W. 1975. The diagnosis of parasitic diseases. Prog. Clin. Pathol. 6:267–288. 79. Kim, C. W. 1991. The significance of changing trends in trichinellosis. Southeast Asian J. Trop. Med. Public Health 22:316–320. 80. Kim, C. W. 1994. A decade of progress in trichinellosis, p. 35–47. In W. C. Campbell (ed.), Trichinellosis. Istituto Superiore di Sanita Press, Rome, Italy. 81. Kirichek, V. S., M. N. Belousov, and A. S. Nikitin. 1984. New data on the epidemiology of taeniarhynchiasis in regions of the Far North in the USSR (based on observational data in the Yamalo-Nenets Autonomous Okrug). Med. Parazitol. (Mosk.) 6:27–33. (In Russian.) 82. Kociecka, W. 2000. Human disease, pathology, diagnosis and treatment. Vet. Parasitol. 93:365–383. 83. Kociecka, W., F. Bruschi, C. Marini, B. Mrozewicz, and L. Pielok. 2001. Clinical appraisal of patients and detection of serum antibodies by ELISA and CIA tests

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

84. 85. 86.

87.

88.

89.

90. 91. 92. 93.

94.

95.

96.

97.

98.

693

in late periods of Trichinella sp. invasion. Parasite 8:147–151. Kotula, A. W., K. D. Murrell, L. Acosta-Stein, L. Lamb, and L. Douglass. 1983. Destruction of Trichinella spiralis during cooking. J. Food Sci. 48:765–768. Küchenmeister, F. 1857. Animal and Vegetable Parasites. The Sydenham Society, London, United Kingdom. Kumar, A., A. H. Goenka, A. Choudhary, J. K. Sahu, and S. Gulati. 2010. Disseminated cysticercosis in a child: whole-body MR diagnosis with the use of parallel imaging. Pediatr. Radiol. 40:223–227. La Rosa, G., G. Marucci, and E. Pozio. 2003. Biochemical analysis of encapsulated and non-encapsulated species of Trichinella (Nematoda, Trichinellidae) from cold- and warm-blooded animals reveals a high genetic divergence in the genus. Parasitol. Res. 91:462–466. Leclair, D., L. B. Forbes, S. Suppa, J. F. Proulx, and A. A. Gajadhar. 2004. A preliminary investigation on the infectivity of Trichinella larvae in traditional preparations of walrus meat. Parasitol. Res. 93:507–509. Lees, W., J. Nightingale, D. Brown, B. Scandrett, and A. Gajadhar. 2002. Outbreak of Cysticercus bovis (Taenia saginata) in feedlot cattle in Alberta. Can. Vet. J. 43:227–228. Lightowlers, M. W. 2006. Cestode vaccines: origins, current status and future prospects. Parasitology 133(Suppl.):S27–S42. Lightowlers, M. W. 2010. Eradication of Taenia solium cysticercosis: a role for vaccination of pigs. Int. J. Parasitol. 40:1183–1192. Lightowlers, M. W., and C. G. Gauci. 2001. Vaccines against cysticercosis and hydatidosis. Vet. Parasitol. 101:337–352. Lightowlers, M. W., R. Rolfe, and C. G. Gauci. 1996. Taenia saginata: vaccination against cysticercosis in cattle with recombinant oncosphere antigens. Exp. Parasitol. 84:330–338. Liu, H.-Y., D. Chao, and P.-C. Fan. 1981. Prevalence and chemotherapy of taeniasis among the aborigines in Nan-ao district, I-lan County, Northeastern Taiwan. Proc. Nat. Sci. Counc. A 5:188–195. Lorraine, M., A. Fleury, E. Sciutto, E. Kendjo, G. Fragoso, L. Paris, and B. Bouteille. 2011. Human neurocysticercosis: comparison of different diagnostic tests using cerebrospinal fluid. J. Clin. Microbiol. 49:195–200. MacLean, J. D., J. Viallet, C. Law, and M. Staudt. 1989. Trichinosis in the Canadian arctic: report of five outbreaks and a new clinical syndrome. J. Inject. Dis. 160:513–520. Mafojane, N. A., C. C. Appleton, R. C. Krecek, L. M. Michael, and A. L. Willingham III. 2003. The current status of neurocysticercosis in Eastern and Southern Africa. Acta Trop. 87:25–33. Mayta, H., R. H. Gilman, E. Prendergast, J. P. Castillo, Y. O. Tinoco, H. H. Garcia, A. E. Gonzalez, and C. R. Sterling for the Cysticercosis Working Group in Peru.

SMP_Food Microbiology_CH26.indd

693

99.

100.

101.

102.

103.

104. 105.

106.

107. 108.

109. 110.

111. 112.

2008. Nested PCR for specific diagnosis of Taenia solium taeniasis. J. Clin. Microbiol. 46:286–289. Mayta, H., A. Talley, R. H. Gilman, J. Jimenez, M. Verastegui, M. Ruiz, H. H. Garcia, and A. E. Gonzalez. 2000. Differentiating Taenia solium and Taenia saginata infections by simple hematoxylin-eosin staining and PCR-restriction enzyme analysis. J. Clin. Microbiol. 38:133–137. McAuley, J. B., M. K. Michelson, A. W. Hightower, S. Engeran, L. A. Wintermeyer, and P. M. Schantz. 1992. A trichinosis outbreak among Southeast Asian refugees. Am. J. Epidemiol. 135:1404–1410. McFadden, A. M., D. D. Heath, C. M. Morley, and P. Dorny. 2011. Investigation of an outbreak of Taenia saginata cysts (cysticercus bovis) in dairy cattle from two farms. Vet. Parasitol. 176:177–184. Millet, N. B., G. D. Hart, T. A. Reyman, M. R. Zimmerman, and P. K. Lewin. 1980. ROM I: mummification for the common people, p. 71 –84. In A. Cockburn and E. Cockburn (ed.), Mummies, Disease and Ancient Cultures. Cambridge University Press, Cambridge, United Kingdom. Moro, P. L., L. Lopera, N. Bonifacio, R. H. Gilman, B. Silva, M. Verastegui, A. Gonzales, H. H. Garcia, and L. Cabrera. 2003. Taenia solium infection in a rural community in the Peruvian Andes. Ann. Trop. Med. Parasitol. 97:373–379. Moskowitz, J., and G. Mendelsohn. 2010. Neuro­ cysticercosis. Arch. Pathol. Lab. Med. 134:1560–1563. Moussavou-Boussougou, M. N., S. Geerts, M. Madeline, C. Ballandonne, D. Barbier, and J. Cabaret. 2005. Sewage sludge or cattle slurry as pasture fertilizers: comparative cysticercosis and trichostrongylosis risk for grazing cattle. Parasitol. Res. 97:27–32. Mukaratirwa, S., and C. M. Foggin. 1999. Infectivity of Trichinella sp. isolated from Crocodylus niloticus to the indigenous Zimbabwean pig. Int. J. Parasitol. 29:1129–1131. Murrell, K. D., and F. Bruschi. 1994. Clinical trichinellosis. Prog. Clin. Parasitol. 4:117–150. Murrell, K. D., M. Djordjevic, K. Cuperlovic, L. Sofronic, M. Savic, M. Djordjevic, and S. Damjanovic. 2004. Epidemiology of Trichinella infection in the horse: the risk from animal product feeding practices. Vet. Parasitol. 123:223–233. Murrell, K. D., R. Fayer, and J. P. Dubey. 1986. Parasitic organisms. Adv. Meat Res. 2:311–377. Murrell, K. D., J. R. Lichtenfels, D. S. Zarlenga, and E. Pozio. 2000. The systematics of the genus Trichinella with a key to species. Vet. Parasitol. 93:293–307. Nelson, G. S., and J. Mikundi. 1963. A strain of Trichinella spiralis from Kenya of low infectivity to rats and domestic pigs. J. Helminthol. 37:329–338. Nicoletti, A., A. Bartoloni, V. Sofia, F. Bartalesi, J. R. Chavez, R. Osinaga, F. Paradisi, J. L. Dumas, V. C. Tsang, A. Reggio, and A. J. Hall. 2005. Epilepsy and

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

694

113.

114.

115.

116.

117.

118.

119. 120. 121.

122. 123. 124.

125.

126.

neurocysticercosis in rural Bolivia: a population‑based survey. Epilepsia 46:1127–1132. Nkouawa, A., Y. Sako, T. Li, X. Chen, T. Wandra, I. K. Swastika, M. Nakao, T. Yanagida, K. Nakaya, D. Qiu, and A. Ito. 2010. Evaluation of a loop-mediated isothermal amplification method using fecal specimens for differential detection of Taenia species from humans. J. Clin. Microbiol. 48:3350–3352. Nsengiyumva, G., M. Druet-Cabanac, B. Ramanan­ kandrasana, B. Bouteille, L. Nsizabira, and P. M. Preux. 2003. Cysticercosis as a major risk factor for epilepsy in Burundi, east Africa. Epilepsia 44:950–955. Nunes, C. M., A. K. Dias, F. E. Dias, S. M. Aoki, H. B. de Paula, L. G. Lima, and J. F. Garcia. 2005. Taenia saginata: differential diagnosis of human taeniasis by polymerase chain reaction-restriction fragment length polymorphism assay. Exp. Parasitol. 110:412–415. Ogunremi, O., and J. Benjamin. 2010. Development and field evaluation of a new serological test for Taenia saginata cysticercosis. Vet. Parasitol. 169:93–101. OIE (World Organisation for Animal Health). 2008. Trichinellosis, ch. 2.1.6. In Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, 6th Ed. OIE, Paris, France. Okamoto, M., M. Nakao, D. Blair, M. T. Anantaphruti, J. Waikagul, and A. Ito. 2010. Evidence of hybridization between Taenia saginata and Taenia asiatica. Parasitol. Int. 59:70–74. Ortega Pierres, M. G., C. Arriaga, and L. Yepez-Mulia. 2000. Epidemiology of trichinellosis in Central and South America. Vet. Parasitol. 93:201–225. Osborn, A. G., and M. T. Preece. 2006. Intracranial cysts: radiologic-pathologic correlation and imaging approach. Radiology 239: 650–664. Ozeretskovskaya, N. N., L. G. Mikhailova, T. P. Sabgaida, and A. S. Dovgalev. 2005. New trends and clinical patterns of human trichinellosis in Russia at the beginning of the XXI century. Vet. Parasitol. 132:167–171. Pal, D. K., A. Carpio, and J. W. Sander. 2000. Neurocysticercosis and epilepsy in developing countries. J. Neurol. Neurosurg. Psychiatry 68:137–143. Parija, S. C., and P. S. Sahu. 2003. A serological study of human cysticercosis in Pondicherry, South India. J. Commun. Dis. 35:283–289. Pawlowski, Z. S. 1982. Taeniasis and cysticercosis, p. 313–348. In J. H. Steele (ed.), Parasitic Zoonosis, vol. 1. CRC Handbook Series in Zoonoses. CRC Press, Boca Raton, FL. Pawlowski, Z. S. 1990. Cestodiasis, p. 490–504. In K. S. Warren and A. A. F. Mahmoud (ed.), Tropical and Geographic Medicine, 2nd ed. McGraw-Hill, New York, NY. Pawlowski, Z., and M. G. Schultz. 1972. Taeniasis and cysticercosis (Taenia saginata). Adv. Parasitol. 10:269–343.

SMP_Food Microbiology_CH26.indd

694

127. Pozio, E. 1998. Trichinellosis in the European Union: epidemiology, ecology and economic impact. Parasitol. Today 14:35–38. 128. Pozio, E. 2000. Factors affecting the flow among domestic, synanthropic and sylvatic cycles of Trichinella. Vet. Parasitol. 93:241–262. 129. Pozio, E. 2005. The broad spectrum of Trichinella hosts: from cold- to warm-blooded animals. Vet. Parasitol. 132:3–11. 130. Pozio, E., and G. La Rosa. 2000. Trichinella murrelli n. sp: etiological agent of sylvatic trichinellosis in temperate areas of North America. J. Parasitol. 86:134–139. 131. Pozio, E., G. La Rosa, and M. A. Gomez Morales. 2001. Epidemiology of human and animal trichinellosis in Italy since its discovery in 1887. Parasite 8:106–108. 132. Pozio, E., G. Marucci, A. Casulli, L. Sacchi, S. Mukaratirwa, C. M. Foggin, and G. La Rosa. 2004. Trichinella papuae and Trichinella zimbabwensis induce infection in experimentally infected varans, caimans, pythons and turtles. Parasitology 128:333–342. 133. Pozio, E., I. L. Owen, G. La Rosa, L. Sacchi, P. Rossi, and S. Corona. 1999. Trichinella papuae n. sp. (Nematoda), a new non-encapsulated species from domestic and sylvatic swine of Papua New Guinea. Int. J. Parasitol. 29:1825–1839. 134. Pozio, E., P. Pagani, G. Marucci, D. S. Zarlenga, E. P. Hoberg, D. De Meneghi, G. La Rosa, and L. Rossi. 2005. Trichinella britovi etiological agent of sylvatic trichinellosis in the Republic of Guinea (West Africa) and a re-evaluation of geographical distribution for encapsulated species in Africa. Int. J. Parasitol. 35:955–960. 135. Pozio, E., and D. S. Zarlenga. 2005. Recent advances on the taxonomy, systematics and epidemiology of Trichinella. Int. J. Parasitol. 35:1191–1204. 136. Proctor, E. M. 1972. Identification of tapeworms. S. Afr. Med. J. 46:234–238. 137. Proulx, J. F., J. D. MacLean, T. W. Gyorkos, D. Leclair, A. K. Richter, B. Serhir, L. Forbes, and A. A. Gajadhar. 2002. Novel prevention program for trichinellosis in Inuit communities. Clin. Infect. Dis. 34:1508–1514. 138. Ranque, S., B. Faugere, E. Pozio, G. La Rosa, A. Tamburrini, J.-F. Pellissier and P. Brouqui. 2000. Trichinella pseudospiralis outbreak in France. Emerg. Infect. Dis. 6:543–547. 139. Rhoads, M. L., K. D. Murrell, G. W. Dilling, M. M. Wong, and N. F. Baker. 1985. A potential diagnostic reagent for bovine cysticercosis. J. Parasitol. 71:779–787. 140. Ribicich, M., H. R. Gamble, A. Rosa, J. Bolpe, and A. Franco. 2005. Trichinellosis in Argentina: an historical review. Vet. Parasitol. 132:137–141. 141. Richards, F. O., P. M. Schantz, E. Ruiz-Tiben, and F. J. Sorvillo. 1985. Cysticercosis in Los Angeles County. JAMA 254:3444–3448.

Manila Typesetting Company

11/08/2012 07:07AM

26.  Helminths in Meat

695

142. Roberts, D. 1986. The last trace. Am. Photogr. 16:64–68. 143. Roy, S. L., A. S. Lopez, and P. M. Schantz. 2003. Trichinellosis surveillance—United States, 1997–2001. MMWR Surveill. Summ. 52:1–8. 144. Ruitenberg, E. J., F. Van Knapen, and J. W. Weiss. 1979. Foodborne parasitic infections—a review. Vet. Parasitol. 5:1–10. 145. Ruiz-Perez, A., M. Santana-Ane, B. Villaverde-Ane, F. Bandera-Tirado, and N. Santana-Santos. 1995. The minimum dosage of praziquantel in the treatment of Taenia saginata, 1986–1993. Rev. Cubana Med. Trop. 47:219–220. 146. Scandrett, W. B., and A. A. Gajadhar. 2004. Recovery of putative taeniid eggs from silt in water associated with an outbreak of bovine cysticercosis. Can. Vet. J. 45:758–760. 147. Schultz, M. G., J. A. Hermos, and J. H. Steele. 1970. Epidemiology of beef tapeworm infection in the United States. Public Health Rep. 85:169–176. 148. Serhir, B., J. D. MacLean, S. Healey, B. Segal, and L. Forbes. 2001. Outbreak of trichinellosis associated with arctic walruses in northern Canada, 1999. Can. Commun. Dis. Rep. 27:31–36. 149. Snyder, G. R., and K. D. Murrell. 1983. Bovine cysticercosis, p. 161–170. In G. Woods (ed.), Practices in Veterinary Health and Preventive Medicine. Iowa State University Press, Ames, IA. 150. Sohn, W.-M., H.-M. Kim, D.-I. Chung, and S. T. Yee. 2000. The first human case of Trichinella spiralis infection in Korea. Korean J. Parasitol. 38:111–115. 151. Sorvillo, F., P. Wilkins, S. Shafir, and M. Eberhard. 2011. Public health implications of cysticercosis acquired in the United States. Emerg. Infect. Dis. 17:1–6. 152. Stehr-Green, J. K., and P. M. Schantz. 1986. Trichinosis in Southeast Asian refugees in the United States. Am. J. Public Health 76:1238–1239. 153. Takahashi, Y., L. Mingyuan, and J. Waikagul. 2000. Epidemiology of trichinellosis in Asia and the Pacific Rim. Vet. Parasitol. 93:227–239. 154. U.S. Code of Federal Regulations. 1990. Title 9: Animals and Animal Products, vol. 1, p. 212–220. Office of the Federal Register, Government Printing Office, Washington, DC. 155. U.S. Code of Federal Regulations. 2010. Title 9: Animals and Animal Products, Part 149, Voluntary Trichinae Certification Program. Office of the Federal Register, Government Printing Office, Washington, DC. 156. Verastegui, M., R. H. Gilman, H. H. Garcia, A. E. Gonzalez, Y. Arana, C. Jeri, I. Tuero, C. M. Gavidia, M. Levine, V. C. Tsang, and the Cysticercosis Working Group in Peru. 2003. Prevalence of antibodies to unique Taenia solium oncosphere antigens in taeniasis and human and porcine cysticercosis. Am. J. Trop. Med. Hyg. 69:438–444. 157. Vicentini-Oliveira, J. C., M. A. Golim, S. de Cássia Paulan, G. F. Biondi, R. Rossi-Ferreira, E. Deffune, and C. M. Nunes. 2010. Taenia saginata: production and characterization of monoclonal antibodies against

SMP_Food Microbiology_CH26.indd

695

158.

159. 160. 161. 162. 163.

164. 165. 166.

167. 168.

169.

170.

171.

172.

Taenia saginata metacestode antigens. Exp. Parasitol. 126:621–625. Viens, P., and P. Auger. 1981. Clinical and epidemiological aspects of trichinosis in Montreal, p. 275–277. In C. W. Kim, E. J. Ruitenberg, and J. S. Teppema (ed.), Trichinellosis. Redbooks, Chertsey, United Kingdom. Wallin, M. T., and J. F. Kurtzke. 2004. Neurocysticercosis in the United States: review of an important emerging infection. Neurology 63:1559–1564. Waloch, M. 2004. Cestode infections in Poland in 2002. Przegl. Epidemiol. 58:165–169. Waloch, M. 2005. Cestode infections in Poland in 2003. Przegl. Epidemiol. 59:331–335. Waloch, M. 2010. Cestode infections in Poland in 2008. Przegl. Epidemiol. 64:261–264. Wandra, T., P. Sutisna, N. S. Dharmawan, S. S. Margono, R. Sudewi, T. Suroso, P. S. Craig, and A. Ito. 2006. High prevalence of Taenia saginata taeniasis and status of Taenia solium cysticercosis in Bali, Indonesia, 2002–2004. Trans. R. Soc. Trop. Med. Hyg. 100:346–353. Wang, Z. Q., and J. Cui. 2001. Epidemiology of swine trichinellosis in China. Parasite 8:67–70. Wang, Z. Q., and J. Cui. 2001. The epidemiology of human trichinellosis in China during 1964–1999. Parasite 8:63–66. Wanzala, W., J. A. Onyango-Abuje, E. K. Kang’ethe, H. Ochanda, and L. J. Harrison. 2002. Serodiagnosis of bovine cysticercosis by detecting live Taenia saginata cysts using a monoclonal antibody-based antigenELISA. J. S. Afr. Vet. Assoc. 73:201–206. White, A. C., Jr. 2000. Neurocysticercosis: updates on epidemiology, pathogenesis, diagnosis, and management. Annu. Rev. Med. 51:187–206. Wilkins, P. P., J. C. Allan, M. Verastegui, M. Acosta, A. G. Eason, H. H. Garcia, A. E. Gonzalez, R. H. Gilman, and V. C. Tsang. 1999. Development of a serologic assay to detect Taenia solium taeniasis. Am. J. Trop. Med. Hyg. 60:199–204. Willingham, A. L., III, N. V. De, N. Q. Doanh, D. Cong le, T. V. Dung, P. Dorny, P. D. Cam, and A. Dalsgaard. 2003. Current status of cysticercosis in Vietnam. Southeast Asian J. Trop. Med. Public Health 34:35–50. Wright, W. H., K. B. Kerr, and L. Jacobs. 1943. Studies on trichinosis. XV. Summary of the findings of Trichinella spiralis in a random sampling and other samplings of the population of the United States. Public Health Rep. 58:1293–1313. Yamasaki, H., J. C. Allan, M. O. Sato, M. Nakao, Y. Sako, K. Nakaya, D. Qiu, W. Mamuti, P. S. Craig, and A. Ito. 2004. DNA differential diagnosis of taeniasis and cysticercosis by multiplex PCR. J. Clin. Microbiol. 42:548–553. Yamasaki, H., M. Nakao, Y. Sako, K. Nakaya, M. O. Sato, and A. Ito. 2006. Mitochondrial DNA diagnosis for taeniasis and cysticercosis. Parasitol. Int. 55(Suppl): S81–S85.

Manila Typesetting Company

11/08/2012 07:07AM

Nonbacterial Pathogens

696 173. Zarlenga, D. S., D. P. McManus, P. C. Fan, and J. H. Cross. 1991. Characterization and detection of a newly described Asian taeniid using cloned ribosomal DNA fragments and sequence amplification by the polymerase chain reaction. Exp. Parasitol. 72:174–183. 174. Zivojinovic, M., G. Dimitrijevic, M. Lazic, M. Petrovic, and L. Sofronic-Milosavljevic. 2009.

SMP_Food Microbiology_CH26.indd

696

Trichinella prevalence in swine in an endemic district in Serbia: epidemiology and control. Vet. Parasitol. 159:358–360. 175. Zoli, A., O. Shey-Njila, E. Assana, J. P. Nguekam, P. Dorny, J. Brandt, and S. Geerts. 2003. Regional status, epidemiology and impact of Taenia solium cysti­ cercosis in Western and Central Africa. Acta Trop. 87:35–42.

Manila Typesetting Company

11/08/2012 07:07AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch27

27

Eugene G. Hayunga

Helminths Acquired from Finfish, Shellfish, and Other Food Sources

A variety of human helminthic infections can be acquired through the consumption of food products from infected animals and plants, through the accidental ingestion of infected invertebrates in foodstuffs or drinking water, or through inadvertent fecal contamination by humans or animals (Table 27.1). Effective prevention involves exploiting parasite vulnerabilities (35) and requires a sound understanding of parasite life cycles and modes of transmission. Unlike bacteria, the infective stages of helminths generally do not propagate. As a result, the critical control point for foodborne helminthiases is initial food preparation, not subsequent storage, reheating, or processing. Although these helminthic infections can readily be prevented, the reality is that safe water supplies, adequate sanitation, and reliable food handling simply do not exist for much of the world’s population, a fact generally not appreciated by tourists “who explore tropical countries with a zeal undamped by any knowledge of preventive medicine” (45). Foodborne helminths, although taxonomically diverse, share the common characteristic of requiring more than one host to complete their life cycles. Termed biohelminths by Kisielewska (42), their transmission requires close behavioral contact between hosts. Typically,

the definitive host of a biohelminth occupies the highest trophic level of the food chain. Prevention of biohelminth infections can be accomplished by avoiding the intermediate hosts or by adequately cooking foods. In contrast, helminths with eggs or free-living stages that can survive a certain length of time in the external environment, termed geohelminths (42), are typically transmitted via contaminated water or foods and are best controlled by improved sanitation. In addition, any parasite capable of penetrating the skin can also penetrate the buccal epithelium.

HELMINTHS ACQUIRED FROM FINFISH AND SHELLFISH

Anisakis and Related Roundworms

Several related nematodes of the genera Anisakis, Pseudoterranova, and Contracaecum may be acquired by eating raw fish or squid in seafood dishes such as sushi, sashimi, ceviche, and lomi-lomi (58). The noninvasive form of anisakiasis is generally asymptomatic, resulting in “tingling throat syndrome” when worms are released from seafood following digestion and migrate up the esophagus into the pharynx, where they

Eugene G. Hayunga, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892.

697

SMP_Food Microbiology_CH27.indd

697

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

698 Table 27.1  Sources of infection with some foodborne helminths of humans Beef or porka

Helminth

Finfish

Shellfish

Other invertebrates

Fruits or vegetables

Other food sources

Fecal Contamination

Xb

Alaria americana Angiostrongylus sp.

X

Anisakis sp.

X

X

X

X

c

Ascaris lumbricoides

X

Baylisascaris procyonis

X

Capillaria philippinensis

X

Clonorchis sinensis

X

X X

Dicrocoelium dendriticum Diphyllobothrium sp.

Xd

X

Dipylidium caninum

X

Dracunculus medinensis

X X

Echinococccus granulosus

X

Echinococcus multiocularis Echinostomum sp.

X

Eustrongylides sp.

X

X

Xe

Fasciola hepatica

X

Fasciolopsis buski

X

Gnathostoma spinigerum

X

X

X

Xf Xg

X

Heterophyes heterophyes

X

Hymenolepis diminuta

X

Hymenolepis nana X

Ligula intestinalis

X

Macracanthorhynchus hirudinaceus X

Metagonimus yokogawai

X

Moniliformis moniliformis

X

Multiceps multiceps X

Nanophyetus salmincola

Xc

Nybelinia surmenicola Opisthorchis sp.

X

Xh X

Phaeneropsolos bonnei Philometra sp.

X X

Prosthodendrium molenkampi Spirometra sp.

X X

Paragonimus westermani

X

X

X

Strongyloides sp.

Xi Xj

Taenia saginata

X

Taenia solium

X

X X

Toxocara canis Trichinella spiralis

X

Xk

Trichostrongylus sp.

X

SMP_Food Microbiology_CH27.indd

698

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods

Figure 27.1  Anisakis embedded in the human gastric mucosa as visualized by gastroscopy. (Photograph contributed by Tomoo Oshima; illustration courtesy of the Armed Forces Institute of Pathology, Washington, D.C., AFIP 76-2118.) doi:10.1128/9781555818463.ch27f1

subsequently may be expectorated (59). In the invasive form, worms typically penetrate the mucosa or submucosa of the stomach or small intestine (Fig. 27.1), resulting in epigastric pain, nausea, vomiting, and diarrhea, usually 12 h after consumption of the infected seafood. Chronic anisakiasis may mimic peptic ulcer, appendicitis, enteritis, Crohn’s disease, or gastric carcinoma. The only effective treatment is surgical removal of the worms. The life cycles of these helminths are not completely known. The adult worms are intestinal parasites of dolphins, whales, seals, and sea lions. Eggs passed in the feces of marine mammals embryonate in seawater and develop into larvae that are eaten by krill. The infected crustaceans are next eaten by fish or squid, and the larvae develop further. The life cycle is completed in marine mammals, but when fish or squid are eaten by the unsuitable human host, the parasites do not develop further or reproduce. Anisakidae larvae have been found in rockfish, herring, cod, halibut, mackerel, and salmon.

699

The prominent reddish brown larvae of Pseudoter­ ranova are readily visible in contrast to the whitish fish tissue, but the smaller, lighter-colored Anisakis larvae are more difficult to detect (58). Fish raised in commercial pens are less likely to be infected than wild-caught fish (15). Comparison of salmon steaks with salmon fillets indicates predilection sites for the larvae and suggests that certain cuts may pose a lower risk for infection (19). In some fishes, most of the juvenile larvae are found in the viscera (16), suggesting that immediate evisceration after catching would prevent postmortem migration of larvae into the musculature. However, only thorough cooking or prolonged freezing will kill the parasites and completely eliminate the risk of infection. The U.S. Food and Drug Administration recommends that all finfish and shellfish intended to be eaten raw, partly cooked, or marinated be blast frozen to –35°C or below for 15 h or regularly frozen to –20°C or below for 7 days (70). Cold smoking and most methods of brining fish are not reliable preventive measures (62). Most human infections with the so-called sushi parasite have been reported in Japan and The Netherlands. In Japan, hypochlorhydria or achlorhydria has been found in more than half of anisakiasis patients and may predispose them to infection (22). In the United States, fewer than 10 cases of anisakiasis are diagnosed annually (70). Considering the increasing popularity of raw seafood dishes and the proliferation of sushi restaurants, the low rate of reported infection is remarkable. Although we do not know the infective dose for this parasite, the risk of acquiring anisakiasis is most likely related to the intensity of infection and the possible predilection of the worm for some organs or muscles over others. The majority of human infections acquired in the United States have been associated with dishes prepared at home, which suggests that the skill and care of the food handler and the selection of fish prepared for consumption may be important risk factors. The source of commercial fish harvesting with regard to the distribution of reservoir hosts may also contribute to the epidemiology of this disease, as geographic variation has

Described in chapter 26. Frog, raccoon, and opossum. Squid. d Acquired by ingestion of ants on unwashed herbs and vegetables. e Frog and tadpole. f The condition halzoun occurs when adult worms in uncooked sheep liver attach to the pharynx. g Pork, chicken, duck, frog, eel, snake, and rat. h Wild boar. i Acquired by ingestion of procercoids in copepods and plerocercoids in frogs, tadpoles, lizards, snakes, birds, and mammals; infection has also been reported to occur from eating undercooked pork. j Transmammary infection has been reported. k Reported to be acquired from a variety of animals including bear, walrus, and horse. a

b c

SMP_Food Microbiology_CH27.indd

699

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

700 been reported for rates of infection of fish from different localities (17).

Diphyllobothrium latum

The broad tapeworm Diphyllobothrium latum occurs in northern temperate regions of the world where raw or undercooked freshwater fishes are eaten. Prevalent since neolithic times (43), infection is closely related to dietary and cultural practices in food preparation (21). Infective larvae may be found in whitefish, trout, pike, and salmon. Cases of infection have been reported throughout Europe, particularly in the Baltic countries; and in the Great Lakes region of the United States; Canada; Japan; South America; and Australia. D. latum was introduced into the United States by European immigrants in the middle of the 19th century (17). Although it is the largest of the human tapeworms, measuring up to 9 m in length, infections may be mild or asymptomatic. Nausea, abdominal pain, diarrhea, and weakness are common manifestations. D. latum may also cause pernicious anemia and vitamin B12 deficiency because the worm is highly efficient in competing with its host for available vitamins. D. latum eggs shed in feces require approximately 12 days to embryonate, at which time the ciliated coracidium exits the egg through the operculum. When eaten by a Cyclops or Diaptomus copepod, the coracidium penetrates the body cavity of the freshwater crustacean and develops into the procercoid stage of the parasite. When small fish such as minnows eat infected crustaceans, the procercoid penetrates into the viscera or muscles of the fish and develops into another unsegmented stage called the plerocercoid. The carnivo­ rous fish that eat these fish are termed paratenic hosts because although the plerocercoid penetrates into their viscera or muscles, it does not develop beyond this stage. Further development from the plerocercoid into the adult worm requires ingestion by the human definitive host. Diagnosis of infection is based on the presence of eggs in the feces. The distinctive D. latum egg is ovoid, approximately 45 to 70 µm, with a prominent operculum at one end and a characteristic knob at the abopercular end; proglottids may occasionally be found in the feces. Infection is best prevented by adequate cooking or freezing of fish before consumption. Improved sanitation measures can also help reduce prevalence by interrupting the parasite’s life cycle. Praziquantel is the recommended anthelminthic (1). Diphyllobothrium ­ dendriticum and Ligula intestinalis, tapeworms of piscivorous birds, and Diphyllobothrium pacificum, a tapeworm of seals, have also been reported in humans.

SMP_Food Microbiology_CH27.indd

700

Clonorchis sinensis

Although the correct scientific name for the Chinese liver fluke is Opisthorchis sinensis, its original name, Clonorchis sinensis, remains more commonly accepted. Widespread throughout Asia, the parasite is typically acquired by eating infected freshwater fishes; the infective stage of C. sinensis has also been found in crayfish (23). Reservoir hosts include dogs, cats, foxes, pigs, rats, mink, badgers, and tigers. Adult worms, measuring approximately 1.2 to 2.4 cm in length and 0.3 to 0.5 cm in width, reside in the bile duct. When eggs passed in the feces are eaten by Parafossarulus manchouricus or other hydrobiid snails such as Bulimus, Semisulcospira, Alocinma, or Melanoides species, the miracidium is released into the digestive tract and penetrates into the hemocoel, where it develops into first the sporocyst and then the redia stage. The redia gives rise to free-swimming cercariae that leave the snail and penetrate the second intermediate host, a cyprinid fish, where they encyst as metacercariae in the gills, fins, or muscles or under the skin. When the definitive host eats an infected fish, metacercariae excyst in the duodenum, migrate into the bile duct, and develop into adult worms. C. sinensis may live in the human host for as long as 25 to 30 years, and massive infections with as many as 500 to 1,000 parasites have been reported. The severity of symptoms is related to the intensity and duration of infection. Diarrhea, epigastric pain, and anorexia are typical manifestations of acute clonorchiasis. The adult worm produces localized tissue damage that may result in hyperplasia or metaplasia of the bile duct epithelium, duct thickening, fibrosis, biliary stasis, and secondary bacterial infection. Pancreatitis may occur when worms enter the pancreatic duct. An association between cholangiocarcinoma and C. sinensis infection has also been reported (37). Diagnosis is made by identifying eggs in the feces. The operculate C. sinensis eggs measure approximately 27 to 16 µm and are characterized by distinctive opercular shoulders and a small spinelike process at the abopercular end. Infection is best prevented by avoiding raw, undercooked, or pickled finfish and shellfish. Snail control and improved sanitation can also help reduce transmission. Praziquantel and albendazole are effective anthelminthics (1). Two closely related bile duct flukes of dogs and cats may also be acquired by eating uncooked cyprinid fishes. Opisthorchis felineus occurs throughout Eastern Europe and portions of former Soviet Union, while Opisthorchis viverrini has been reported to occur in Southeast Asia. The clinical picture for human infection with these species is very similar to that for clonorchiasis.

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods

The Human Lung Fluke Paragonimus westermani

Paragonimus westermani is the best-known and most widely distributed lung fluke, although other species of this genus may parasitize humans (76). P. westermani infection is common throughout the Far East and occurs to a lesser extent in parts of Africa and the Indian subcontinent. P. westermani is a common parasite of mink in Canada and the eastern United States. Other reservoir hosts include a variety of animals that may eat crustaceans. Infection has been reported in dogs, cats, tigers, and cattle. The reddish brown, thick-bodied flatworms are found encapsulated in cystic structures adjacent to the bronchi or bronchioles. They measure approximately 0.8 to 1.6 cm in length by 0.4 to 0.8 cm in width and are 0.3 to 0.5 cm thick. Eggs are released by the parasite into the bronchioles, where they may be expectorated or swallowed and passed in the feces. Eggs require several weeks in an aqueous environment for development. Upon hatching, the free-swimming miracidium penetrates a snail of the genus Brotia, Semisulcospira, Tarebia, or Thiara, where it undergoes further development into a sporocyst, and then two generations of rediae. Approximately 11 weeks after the snail is infected, cercariae are shed and then penetrate any of a variety of freshwater crabs and crayfish, where they become encysted in the muscles, gills, and other organs as metacercariae. Important shellfish hosts include the freshwater and brackish water crabs Eriocheir, Potamon, and Sundathelphusa and the crayfish Procambarus. Humans become infected by eating raw or inadequately cooked freshwater crabs or crayfish. Dishes such as raw crayfish salad, jumping salad (live shrimp), drunken crab (live crabs in wine), and crayfish curd are popular throughout the Orient. Pickling does not kill the parasite, but the infective metacercariae can be killed by boiling the crabs for several minutes until the meat has congealed and turned opaque (40). Infection may also be acquired from shellfish juices used in food dishes or folk remedies, from food prepared by using contaminated utensils or chopping blocks, or by drinking water contaminated with metacercariae released from dead or injured crustaceans. Migration of parasites through host tissues produces localized hemorrhage and infiltration of lymphocytes. Pulmonary symptoms include dyspnea (labored breathing), chronic cough, chest pain, night sweats, hemoptysis, and persistent rales. The severity of symptoms appears to be related to the number of parasites present. Pleural effusion and fibrosis may occur in long-standing infections, although there is also evidence that lesions

SMP_Food Microbiology_CH27.indd

701

701

may resolve without treatment. Neurological complications result from migration of the parasite into the spinal cord or brain. P. westermani has also been found in the intestinal wall, peritoneum, pleural cavity, and testes. Paragonimus szechuanensis and Paragonimus hueitun­ gensis cause cutaneous larva migrans characterized by eosinophilia, anemia, and low-grade fever. Diagnosis is based on the presence of eggs in expectorant or feces. The characteristic golden-brown eggs are 80 to 120 µm long by 48 to 60 µm wide. Chest X rays may show nodular shadows or calcified spots. Praziquantel and bithionol are effective anthelminthics.

Capillaria philippinensis

Capillaria philippinensis, a nematode found in freshwater fishes, causes a severe and potentially fatal infection in humans. In 1967 and 1968, the disease reached epidemic proportions in The Philippines with over 1,000 confirmed cases and more than 100 deaths (11). The infection appears to be acquired as a result of eating raw fish that contain infective larvae. The freshwater fishes Ambassis miops, Eleotris melanosoma, and Hypseleotris bipartita have been implicated in experimental infections, but the life cycle of C. philippinensis has not been fully elucidated nor have natural reservoir hosts been identified. Both larvae and adult worms may be found embedded in the intestine (Fig. 27.2). Diagnosis of infection is made by identifying eggs in the feces. Symptoms of capillariasis include borborygmus (intestinal noises associated with the movement of gas), abdominal pain,

Figure 27.2  Transverse section of larval Capillaria philippi­ nensis embedded in human intestinal glands. Magnification, ×280. (Illustration courtesy of the Armed Forces Institute of Pathology, AFIP 69-1066.) doi:10.1128/9781555818463.ch27f2

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

702 nausea, vomiting, diarrhea, and anorexia during the acute phases. If the infection is untreated, intestinal malabsorption and intractable diarrhea lead to cachexia and possibly death. Mebendazole and albendazole are effective anthelminthics (1).

Gnathostoma Roundworms

Roundworms of the genus Gnathostoma reside in the stomach walls of a variety of carnivorous mammals. The life cycle of this parasite typically involves two intermediate hosts, a copepod and a freshwater fish; however, a variety of animals may serve as paratenic or transport hosts. Unable to mature in human hosts, the parasite wanders aimlessly through the tissues, causing a severe larva migrans that may persist for years. Gnathostomiasis has been reported to occur throughout the world, particularly in Thailand, and is considered to be an emerging infection in several parts of Latin America. Gnathostoma spinigerum has been found in tigers, leopards, lions, domestic cats, mink, and dogs. The stout, reddish female worms range in length from 25 to 54 mm and are characterized by spines covering the anterior half of the body and by a prominent cephalic bulb covered by rows of sharp hooklets. Male worms are about half as long as females. Human infection typically results from eating raw, marinated, or poorly cooked freshwater fish (62). In addition, infection can be acquired by eating pork, chicken, duck, frog, eel, snake, or rat or by accidentally ingesting infected copepods in drinking water. It is also possible that larvae may pene­ trate the skin during food handling (13). Larvae can be killed by cooking or by immersion in strong vinegar for 5 h or longer; immersion in lime juice and chilling at 4°C for 1 month are not effective (5). Raw foods, particularly fish and chicken, should be avoided in areas where infection is endemic, and drinking water should be filtered before consumption. Nausea, abdominal pain, and vomiting usually develop between 24 and 48 h following the ingestion of infected meat or fish. Symptoms of larva migrans or creeping eruption include pruritus (itching), urticaria (hives), tenderness, and painful subcutaneous swelling. Invasion of the central nervous system may result in meningitis and neuropathy. Diagnosis is difficult, and chemotherapy is of questionable value. Albendazole and ivermectin have been reported as alternatives to surgical removal of the larvae (1).

Other Helminths Associated with Seafood

Heterophyes heterophyes, an intestinal fluke acquired from the mullet, has been reported in Egypt, Israel, and

SMP_Food Microbiology_CH27.indd

702

throughout Asia. The parasite may cause nausea, diarrhea, and abdominal pain, but light infections are often asymptomatic. There have been reports of fatal myocarditis and neurological complications when helminth eggs penetrate the intestine and enter the circulatory system. The closely related trematode Metagonimus yokagawai, acquired from salmonid fishes, has been reported in Asia, the Balkans, Israel, Spain, and portions of the former Soviet Union. Fisheating mammals and birds, such as dogs, cats, and pelicans, serve as reservoir hosts. Infection with species of another trematode, Echinostomum, has been reported in Japan and has been attributed to eating raw freshwater fish, in particular sashimi (33). Diagnosis of infection is made by identifying eggs passed in the feces. Intestinal symptoms are mild and depend upon the number of worms present. Praziquantel is an effective anthelminthic. Salmon poisoning, a severe and frequently fatal disease of dogs in the Pacific Northwest, is associated with the intestinal trematode Nanophyetus (Troglotrema) salmincola. Disease in dogs occurs because N. sal­ mincola serves as a vector for the rickettsial pathogen Neorickettsia helmintheca. Human N. salmincola infection, characterized by nausea, diarrhea, and intestinal discomfort, may be acquired by eating uncooked salmonid fishes and has also been attributed to handling freshly killed coho salmon (34). The nematode Eustrongylides is a common parasite of piscivorous birds. In addition to fish, reptiles and amphibians may also serve as intermediate hosts and thus contain infective larvae. An infection in New York City was attributed to eating raw fish prepared at home (75). Cases in Maryland and New Jersey were reported in fishermen who had eaten live bait (9, 26). As with Anisakis infections, surgical removal of worms is the only effective treatment. Philometra, a nematode closely related to Dracun­ culus, was acquired by a fisherman in Hawaii as a result of filleting a carangid fish (18). Infection with the tapeworm Nybelinia surmenicola has been attributed to eating raw squid (41). Sparganosis may be transmitted by a variety of animals, including fish. Although typically associated with snails, Angiostrongylus may also be acquired from freshwater prawns and land crabs.

HELMINTHS ACQUIRED FROM VEGETATION

The Sheep Liver Fluke Fasciola hepatica

Sheep liver rot, caused by the digenetic trematode Fasciola hepatica, was recognized as early as the 14th

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods century (20), and F. hepatica was the first trematode for which a complete life cycle was elucidated (44). Sheep, cattle, and other herbivores acquire the infection by eating metacercariae encysted on aquatic plants. F. hepatica has been found in goats, horses, deer, rabbits, camels, vicuña, swine, dogs, and squirrels. Human infection is prevalent in sheep-rearing areas throughout the world. In the United States, human fascioliasis has been reported in California (51) and Puerto Rico (36). F. hepatica is a large fluke, measuring approximately 3 cm in length by 1.5 cm in width, and is readily identified by its characteristic “cephalic zone,” a distinct conical projection at the anterior end. Adult worms reside in the biliary passages and gall bladder. Eggs passed with feces into the water require 9 to 15 days to mature, but may remain viable for several months in soil if they remain moist. Upon hatching, the free-swimming miracidium penetrates a lymnaeid snail of the genus Lymnaea, Succinea, Fossaria, or Practicolella, where it develops further. Free-swimming cercariae are shed from the snail and attach to aquatic vegetation, where they encyst as metacercariae. The metacercariae are susceptible to drying but can survive over winter (54). Humans become infected by ingesting infested freshwater plants or free metacercariae in drinking water. Human cases are frequently traced to watercress, Nasturtium officinalis (39). Transmission can be controlled by using molluscicides, by draining ponds, and by protecting crops and water supplies from contact with livestock. Some inflammation is associated with migration of the parasite, but mild infections are often asymptomatic. Tissue destruction occurs when worms penetrate the liver. In sheep, liver rot causes massive damage. There may be mechanical obstruction of bile ducts, hyperplasia of the biliary epithelium, and proliferation of the connective tissue. Worms may erode the walls of the bile ducts and invade the liver parenchyma. Secondary bacterial infection and portal cirrhosis have been reported, but liver calcification appears to be rare. A pharyngeal form of disease, called halzoun, may occur following the ingestion of raw liver from infected animals when worms present in liver attach to the pharynx. Pain, bleeding, and edema of the face and neck are associated with halzoun. Diagnosis is based on the presence of eggs in the feces. The relatively large, operculate F. hepatica eggs are approximately 130 to 150 µm long by 63 to 90 µm wide. Ingestion of liver from infected sheep or cattle may result in spurious infection when eggs present in the food are passed in feces. Such false-positive findings can be ruled out by subsequent stool examinations. Although other

SMP_Food Microbiology_CH27.indd

703

703

anthelminthics have been reported in the literature, triclabendazole is presently the treatment of choice.

The Giant Intestinal Fluke Fasciolopsis buski

Fasciolopsis buski, the largest trematode parasite of humans, is endemic throughout Asia. The pig is an important reservoir host, but infection has also been reported in dogs and rabbits. Although only 0.8 to 3.0 mm thick, F. buski can grow to 7.5 cm in length by 2 cm in width. Adult worms attach to the bowel walls, primarily along the duodenum and jejunum. Eggs passed in the feces are unembryonated and require 3 to 7 weeks in freshwater to develop. Upon hatching, the free-swimming miracidium penetrates a snail of the genus Segmentina or Hippeutis, where it develops further. Approximately 4 to 7 weeks after the snail is infected, cercariae are shed into the water, and then attach to vegetation, where they encyst as metacercariae. Humans become infected by eating infested water chestnuts, bamboo, caltrop, or lotus. Individuals may also acquire the parasite by peeling the hulls of plants with their teeth. The metacercariae excyst in the small intestine, attach to the mucosa, and develop into adult worms in about 3 months. Drying or cooking the plants before eating kills the metacercariae (6). Immersing of vegetables in boiling water for a few seconds, or even peeling and washing them in clear water, is sufficient to preclude infection (5). Poor sanitation and the use of human and swine feces as fertilizer are major factors in disease transmission (57). Adult worms feed not only on intestinal contents but also on the intestinal epithelium, leading to local ulceration and hemorrhage. Nausea, abdominal pain, diarrhea, and hunger pangs are common. In heavy infections, stools are profuse and light yellow in color, suggestive of malabsorption; intestinal obstruction and ascites have been reported. Diagnosis is based on the presence of eggs in the feces. The yellow-brown eggs have a small operculum and are approximately 130 to 140 µm long by 80 to 85 µm wide. Adult worms are seen in feces only following chemotherapy or purgation. Praziquantel is the anthelminthic of choice.

Other Helminths Associated with Vegetation

Fresh vegetables grown in areas where night soil (human waste) is used as fertilizer are frequently contaminated and, thus, may facilitate transmission of any of a number of geohelminths. Human infection by Dicrocoelium is explained by the accidental ingestion of ants on vegetation, and Angiostrongylus costaricensis is thought to be acquired by eating raw fruits and vegetables on

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

704 which snails have left larvae in mucus deposits or by accidentally ingesting infected snails on unwashed vegetation. Trichostrongylus and Echinococcus granulosus infections have also been attributed to contaminated vegetation (5).

HELMINTHS ACQUIRED FROM INVERTEBRATES IN DRINKING WATER

The Guinea Worm Dracunculus medinensis

The long, threadlike roundworm Dracunculus has plagued humans since antiquity. It is believed to be the “fiery serpent” described in the Bible (Numbers 21:4–9) and is symbolically depicted in the caduceus, the insignia of the medical profession. Although infection is rarely fatal, this parasite causes considerable discomfort and disability. The female nematode is almost a meter in length but less than 2 mm thick; males are inconspicuous and only 2 cm long. Worms develop to maturity in the body cavity or deep connective tissues. Females then migrate to the subcutaneous tissues, become gravid, and stimulate the formation of a blister that eventually ruptures to expose part of the worm. Upon contact with freshwater, the worm bursts to release larvae, which are then ingested by copepods of the genera Cyclops, Mesocyclops, or Thermocyclops. Humans become infected by swallowing infected copepods present in drinking water (Fig. 27.3). The larvae penetrate the digestive tract and take about a year to reach maturity. Transmission is clearly related to poverty, the quality of drinking water, and water contact by infected individuals. A measure as simple as sieving drinking water through a piece of cloth will remove

Figure 27.3  Arrows show Dracunculus larvae within the body cavity of the intermediate host, Cyclops. Magnification, ×85. (Specimen contributed by E. L. Schiller, Johns Hopkins University School of Public Health, Baltimore, MD; illustration courtesy of the Armed Forces Institute of Pathology, AFIP 68-4629.) doi:10.1128/9781555818463.ch27f3

SMP_Food Microbiology_CH27.indd

704

copepods and prevent infection. The provision of safe drinking water supplies in a village is usually followed by disappearance of the disease (49). As a result of an aggressive and highly effective eradication program, the worldwide annual incidence of dracunculiasis has been reduced by 99% from that estimated in 1986 and the disease has been completely eliminated from many countries; however, reemergence can occur when safe water systems fail (4, 73). Dracunculus infections have been reported in dogs in China and the former Soviet Union, but it is not clear whether canines play any significant role as reservoir hosts. Surgical removal of Dracunculus has been practiced in India and Pakistan, but the traditional treatment of winding the worm around a small stick and slowly extruding it can still be effective provided asepsis is maintained. Attempts to develop an effective anthelminthic against this parasite have had only limited success.

Other Helminths Acquired from Copepods

There is evidence that the tissue-dwelling nematode Gnathostoma spinigerum may be acquired directly from copepods (12). Sparganosis, although typically acquired from a variety of vertebrate hosts, may also result from ingesting infected copepods.

HELMINTHS ACQUIRED FROM OTHER INVERTEBRATES

Snails

The rodent lungworm Angiostrongylus cantonensis is a slender roundworm about 25 mm in length that typically resides in the pulmonary arteries of rodents. The adult nematodes lay eggs that hatch in the lungs. Firststage larvae migrate to the trachea, are swallowed, and pass in the feces, where they are ingested by any of a variety of slugs, land snails, or planarians. When rodents eat infected molluscs, the larvae migrate to the brain, developing into adults in about 4 weeks, and then to the pulmonary arteries, where they begin to lay eggs after an additional 2 weeks. In humans, migration of the parasite through the brain typically presents as eosinophilic meningitis but may also result in ocular involvement and even fatal consequences (56, 67). Cases of infection have been reported in Taiwan, Thailand, India, The Philippines, Japan, Hawaii, several Pacific islands, Cuba, and the Ivory Coast. Foodborne infections have been attributed to the consumption of raw or undercooked molluscs such as the freshwater snail Pila or the giant African land snail Achatina fulica. The parasite has been found in other potential food sources such as the land crab

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods Cardisoma hirtipes and the coconut crab Birgus latro (2). Freshwater prawns, which are frequently eaten raw in Thailand, Vietnam, and Tahiti, have been implicated in human infection (56), while shrimp, crabs, and frogs serve as paratenic hosts. In addition, the parasite can be acquired by the accidental ingestion of slugs on lettuce, helminth larvae left by snails in mucus deposits on fruits and vegetables, and contaminated drinking water. A. cantonensis infection has also been reported in the United States in a child who had deliberately consumed a raw snail as a result of peer pressure (50). Transmission can be prevented by thorough cooking and appropriate attention to food-handling practices, particularly the careful washing of fruits and vegetables and washing of hands after exposure to molluscs while gardening. A. costaricensis is a closely related roundworm found in the mesenteric arteries of the cotton rat Sigmodon hispidus. Human abdominal angiostrongyliasis has been reported in Costa Rica, Honduras, Panama, Mexico, Brazil, and Venezuela. The veronicellid slug Vaginulus plebius has been implicated in the transmission of A. costaricensis. Human infection is acquired from accidental ingestion of slugs or by contact with contaminated fruits, vegetables, or grass. Several species of the intestinal trematode Echinos­ tomum may be acquired by eating metacercariae encysted in snails. Human echinostomiasis occurs primarily throughout Asia (33), although there is archeological evidence of human infection in the New World (63). Infections are generally mild and often asymptomatic. Adult worms are found in a variety of domestic animals and birds, and the Norway rat is believed to be an important reservoir host. Diagnosis is based on identification of unembryonated, operculate eggs measuring approximately 83 to 116 mm in length by 58 to 69 mm in width in stool specimens.

705

Fleas

Dipylidium caninum is a common tapeworm of dogs and cats throughout the world. Tapeworm eggs are ingested by the flea Ctenocephalides, where they develop into the cysticercoid stage. Human infection appears limited to young children and results from accidental ingestion of infected fleas. Infection can best be prevented by controlling fleas on pets and by periodic worming of animals when necessary.

Beetles, Cockroaches, and Other Insects

Hymenolepis diminuta is a tapeworm of rats, mice, and other rodents. Eggs passed in rodent feces are ingested by flour beetles (Tribolium, Tenebrio), cockroaches (Blattella, Periplaneta), or fleas, where they develop into the cysticercoid stage. Human infection, acquired by the accidental ingestion of infected insects, is typically asymptomatic, although nausea, abdominal pain, and diarrhea may occur. Preventive measures include protection of grains and foodstuffs from insects and rodent control. The acanthocephalan Moniliformis moniliformis is an intestinal parasite of rats that also utilizes cockroaches and beetles as intermediate hosts. As with H. diminuta, human infection usually occurs in young children. The giant leechlike acanthocephalan Mac­ racanthorhynchus hirudenaceus is a cosmopolitan parasite of swine. The spiny proboscis embeds itself in the intestinal mucosa; female worms measure up to 65 cm in length. Human infections are rare but have been directly attributed to eating raw beetles. The spirurid nematode Gongylonema is typically acquired by ingestion of cockroaches; an infection was reported in an adult resident of New York City (25). The trematodes Prosthodendrium molenkampi and Phaeneropsolos bonnei may be acquired from ingesting dragonfly and damselfly aquatic larvae.

Ants

Dicrocoelium dendriticum is a trematode found in the biliary passages of sheep, deer, and other herbivores. The intermediate hosts are land snails and ants. Parasites are shed from snails in slime balls that are deposited on the grass and eaten by ants; the mammalian host becomes infected by ingestion of infected ants. Reports of human infection are frequently spurious, as ingestion of liver from infected sheep can result in a false positive diagnosis when eggs present in the food are passed in feces. However, genuine human cases have been reported in Europe, Asia, and Africa. Transmission of Dicrocoelium can best be prevented by careful washing of herbs and vegetables to remove ants.

SMP_Food Microbiology_CH27.indd

705

HELMINTHS ACQUIRED FROM OTHER FOOD SOURCES

Tapeworms Causing Sparganosis

Sparganosis is infection with the plerocercoid stage of the diphyllobothrid tapeworm Sparganum or Spirometra. Human infection may be acquired by the accidental ingestion of copepods infected with the procercoid stage of the tapeworm. Alternatively, humans may serve as paratenic hosts by eating any of a variety of animals infected with the plerocercoid stage, such as frogs, tadpoles, lizards, snakes, birds, and mammals. Foodborne transmission has also been attributed to eating raw pork

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

706 (10). Infection may also result from the folk medicine practice of applying poultices of frog or snake to the eyes, skin, or vagina, as sparagana are capable of migrating out of the infected animal flesh and penetrating the lesion. The sparganum is a wrinkled, ivory-white, ribbonlike flatworm that can grow up to 30 cm in length but is only about 3 mm wide (Fig. 27.4). Histological sections reveal parenchymal tissue typical of undifferentiated cestode plerocercoids. Although usually found in subcutaneous tissue, spargana are highly motile and may migrate into muscle, viscera, brain, and eye tissue. Sparganum pro­ liferum is an especially hyperplastic species, exhibiting extensive branching and the capability of asexual reproduction by budding into separate organisms. Migration of spargana is characterized by a painful, localized inflammatory reaction. Excessive lacrimation, periorbital edema, and swelling of the eyelids are associated with ocular sparganosis. Subcutaneous lesions may develop into abscesses; if the lesion ulcerates, the sparganum may be mistaken for a guinea worm. Diagnosis of infection is presumptive and usually not confirmed until the worm is removed and identified through histological sectioning. Although some success with anthelminthics has been reported for laboratory animals, surgical removal of spargana remains the treatment of choice.

Figure 27.4  A sparganum removed from a subcutaneous nodule in the inguinal region. Magnification, ×3.8. (Illustration courtesy of the Armed Forces Institute of Pathology, AFIP 707392.) doi:10.1128/9781555818463.ch27f4

SMP_Food Microbiology_CH27.indd

706

Other Biohelminths

Alaria americana is a strigeid trematode typically found in the intestines of dogs and foxes. The normal life cycle involves sporocysts in snails and mesocercariae (a nonencysted stage) in tadpoles and frogs. Human infection has been attributed to consumption of inadequately cooked frog legs, raccoon, and opossum. Parasites have been recovered from the eye and from intradermal lesions, and at least one massive, fatal infection has been reported with mesocercariae disseminated into the lungs and other internal organs (29). Human infection by the trematode Echinostomum has also been attributed to consumption of crustaceans, tadpoles, and frogs (33). Pentastomids, or “tongue worms,” are larval arthropods that have been recovered from liver, spleen, lungs, and eye. Human infection has been attributed to the eating of raw snake, lizard, goat, and sheep. Ocular involvement probably results from direct contact with pentastomid eggs in water. Most infections are asymptomatic, although respiratory discomfort and intestinal obstruction have been reported (46). Hoarseness and coughing due to young worms attached to the pharynx, a condition known as halzoun in the Near East, has been attributed to pentastomid infection (6). Halzoun may also be caused by Fasciola hepatica in sheep liver. Ingestion of liver from sheep infected with adult F. he­ patica or Dicrocoelium dentriticum worms may result in false-positive diagnoses of infections when parasite eggs are released into the alimentary canal. Although typically associated with fish and copepods, Gnathostoma may also be acquired from pork, chicken, duck, frog, eel, snake, or rat. Trichinella spiralis, described in chapter 26, is typically associated with ingestion of raw or undercooked pork. However, human infection has also been acquired from sources as diverse as bear and walrus. An outbreak of trichinellosis in France was traced to horsemeat imported from the United States (3). Wild boar, a paratenic host for Paragonimus westermani, has been implicated as another potential source of human infection by that species (48). Prenatal and transmammary transmission of infective larval stages of helminths is more common than generally realized and in some instances may be the major route of infection. It has been observed for cestodes, trematodes, and most frequently, nematodes (47). Larvae of Strongyloides fuelleborni have been recovered from human milk (7), and Strongyloides stercoralis may also be transmitted in this manner (5). Similar reports from dairy animals suggest a risk to humans from drinking raw milk. Strongyloidiasis has been reported as a common cause of infant death in Papua New Guinea (72).

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods HELMINTHS ACQUIRED FROM FECAL CONTAMINATION Inadequate washing of produce or poor hygiene among food handlers can result in a variety of helminthic infections (60). Personal hygiene is critical because helminth eggs are often adherent and contamination may be found not only on hands but also under fingernails, on clothing, and in wash water (53). Parasites can be acquired from both animal and human waste, and the use of night soil to fertilize crops represents a major source of infection. In addition to the direct transmission of zoonoses from animal to man, there is evidence that birds and dogs may play an important ancillary role in disseminating helminth eggs found in untreated human waste (64, 68). Climate change may also contribute to the emergence of soil-transmitted helminths (74). A variety of geohelminth species utilize the fecaloral route for person-to-person and animal-to-person transmission. Ascaris lumbricoides, the largest intestinal roundworm of humans (Fig. 27.5), has been known since antiquity. Ascaris eggs have been found in mummified remains in Egypt and in archeological artifacts from the sites of Roman legion encampments. In parts of Central and South America, infection rates have been reported to approach 45% (5). It has been estimated that, at one time, in China alone 18,000 tons of Ascaris eggs were produced each year (66). Unembryonated eggs shed in the feces are resistant to desiccation, moderate freezing temperatures, and chemical treatment of sewage and

Figure 27.5  Numerous adult Ascaris lumbricoides worms obstructing the jejunum of a 13-year-old Zairian. Magnification, ×2.1. (Illustration courtesy of the Armed Forces Institute of Pathology, AFIP 72-13204.) doi:10.1128/9781555818463.ch27f5

SMP_Food Microbiology_CH27.indd

707

707

may remain dormant in the soil for years (8). The swine ascarid, Ascaris suum, can also infect humans but does not develop to maturity (14). The whipworm Trichuris trichiura (Fig. 27.6) occurs throughout the world but is most common in the tropics and in regions where sanitation is poor. In the United States, infection now occurs primarily in the southeast, but archeological evidence reveals that distribution was once far more widespread (27). Trichuris eggs can survive in the soil for several years. There is no reservoir host for T. trichiura, but Trichuris suis and Trichuris vulpis, parasites of swine and dogs, respectively, have been found in humans. Trichostrongylus is a common intestinal nematode of herbivores that resembles the hookworm. Human trichostrongyliasis, caused by the accidental ingestion of larvae on contaminated vegetation or in drinking water, occurs primarily in Asia but has been reported throughout the world. Oesophagostomum is a common intestinal nematode of primates, swine, and domestic animals throughout Asia, Africa, and South America. In humans, accidentally ingested larvae produce nodular lesions and abscesses, approximately 1 to 2 cm in diameter, in the intestinal wall and sometimes other organs (32). As in anisakiasis, there is no effective anthelminthic, and surgical removal of the parasite is the only effective treatment. Definitive diagnosis is based upon identifying worms in surgical or biopsy specimens. In livestock, severe diarrhea may kill the animal and parasitic nodules render the intestine unsuitable for making sausage casings, thus representing a serious economic loss for farmers (24). The dwarf tapeworm Hymenolepis nana is the only human tapeworm that does not require an intermediate host; its life cycle is maintained by person-to-person contact and by autoinfection, although rodents may serve as reservoir hosts. Distributed throughout the world, H. nana is common in southern Europe, the former Soviet Union, India, and Latin America. Young children are most frequently infected. Eggs shed in feces are fully embryonated and immediately infective but have poor resistance outside the host. Internal autoinfection may occur when eggs hatch in the small intestine and penetrate the villi to repeat the developmental cycle without leaving the host. Eggs ingested by fleas and flour beetles develop into cysticercoids, but insects do not appear to play a major role in transmission. Mild infections are typically asymptomatic, and even large numbers of H. nana are well tolerated. Toxocara canis, an intestinal roundworm of dogs and foxes, appears to be the principal cause of visceral larva migrans in the United States (55). Toxocara cati is a related species in domestic felines. Following the

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

708

Figure 27.6  Trichuris trichiura adult worms. The photograph shows the slender anterior ends threaded beneath the colonic epithelium. Magnification, ×3.3. (Illustration courtesy of the Armed Forces Institute of Pathology, AFIP 69-3583.) doi:10.1128/9781555818463.ch27f6

accidental ingestion of soil contaminated with embryonated Toxocara eggs, the larvae migrate throughout the body and may cause serious complications when they invade the central nervous system or the eye. Eggs may survive for years in the environment, and the typically high rates of contamination in municipal parks, playgrounds, and schoolyards pose a serious potential risk for young children (69). Appropriate preventive measures include animal control, particularly leash laws and waste disposal regulations; the periodic worming of pets; and scrupulous encouragement of hand washing before meals. Ocular involvement is a serious complication, and tragically, there are reports that eyes with benign Toxocara inflammatory lesions have been unnecessarily enucleated because of suspected retinoblastoma (30). Ocular lesions may be more likely to occur in mild infections (31). Baylisascaris procyonis, an ascarid of raccoons, can cause severe neurological complications and even death when this aggressive species migrates through the tissues of an unsuitable host. Fatal human infections have been reported in children believed to have been exposed to Baylisascaris eggs present on pieces of firewood or in raccoon feces deposited by animals nesting in unused

SMP_Food Microbiology_CH27.indd

708

hearths and chimneys (28, 38); fatal infections have also been reported for animals (61). Echinococcus granulosus is a tapeworm of dogs and other canids. Human infection following the accidental ingestion of eggs in canine feces is common in sheepraising areas throughout the world and among indigenous populations such as Eskimos, where there is a close ecological relationship between humans and dogs. Echinococcus multilocularis is a closely related species producing an alveolar cyst that grossly resembles an invading neoplasm. The course of the disease resembles that of a growing carcinoma, and the disease is among the most lethal of all helminthic infections (5). Risk factors for Echinococcus infection include exposure to infected dogs; consumption of contaminated water, ice, or snow; contact with infected foxes or their skins; and ingestion or handling of contaminated berries or other vegetation (65). Increases in grassland rodent and red fox populations in some parts of Europe, attributed to changes in land use as a result of agricultural policy, may further contribute to the risk of human infection (71). Multiceps multiceps develops into a similar proliferative stage called a coenurus. In sheep, coenuriasis of the brain or spinal cord is relatively common and is known as gid or staggers. Human infections may occur in any organ but frequently involve subcutaneous tissue, brain, and eye. Cysticercosis, described in chapter 26, results from the accidental ingestion of Taenia solium eggs. Racemose cysticercosis has been described as an aberrant cysticercus form of T. solium but may represent coenurus infection. I wish to thank Judy A. Sakanari, University of California, San Francisco, for sharing research findings about Anisakis and other fish-borne parasites, and John H. Cross, Uniformed Services University of the Health Sciences, Bethesda, Maryland, for providing background information about capillariasis and other helminthic diseases in the Far East. I am especially in­ debted to John S. Mackiewicz, State University of New York at Albany, for encouraging attention to the importance of eco­ logical factors when considering parasitic diseases. Illustrations for this chapter were obtained from the extensive slide collec­ tion of the Armed Forces Institute of Pathology, Washington, D.C.

References 1. Abramowicz, M. (ed.). 2010. Drugs for Parasitic Infections, 2nd ed. The Medical Letter, Inc., New Rochelle, NY. 2. Alicata, J. E. 1965. Notes and observations on murine angiostrongylosis and eosinophilic meningoencephalitis in Micronesia. Can. J. Zool. 43:667–672. 3. Ancelle, T., J. Dupouy-Camet, M. E. Bougnoux, V. Fourestie, H. Petit, G. Mougeot, J. P. Nozais, and J. LaPierre. 1988. Two outbreaks of trichinosis caused

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods by horsemeat in France in 1985. Am. J. Epidemiol. 127:1302–1311. 4. Barry, M. 2007. The tail end of guinea worm—global eradication without a drug or a vaccine. N. Engl. J. Med. 356:2561–2564. 5. Beaver, P. C., R. C. Jung, and E. W. Cupp. 1984. Clinical Parasitology, 9th ed. Lea & Febiger, Philadelphia, PA. 6. Bogitsh, B. J., and T. C. Cheng. 1990. Human Parasitology. Saunders College Publications, Holt, Rinehart & Winston, New York, NY. 7. Brown, R. C., and M. H. F. Girardeau. 1977. Transmammary passage of Strongyloides sp. larvae in the human host. Am. J. Trop. Med. Hyg. 26:215–219. 8. Bryan, F. D. 1977. Diseases transmitted by foods contaminated by waste water. J. Food Prot. 40:45–56. 9. Centers for Disease Control. 1982. Intestinal perforation caused by larval Eustrongylides—Maryland. MMWR Morb. Mortal. Wkly. Rep. 31:383–389. 10. Corkum, K. C. 1966. Sparganosis in some vertebrates of Louisiana and observations on a human infection. J. Parasitol. 52:444–448. 11. Cross, J. H. 1992. Intestinal capillariasis. Clin. Microbiol. Rev. 5:120–129. 12. Daengsvang, S. 1971. Infectivity of Gnathostoma spinige­ rum larvae in primates. J. Parasitol. 57:476–478. 13. Daengsvang, S., B. Sermswatsri, P. Youngyi, and D. Guname. 1970. Development of adult Gnathostoma spin­ igerum in the definitive host (cat and dog) by skin penetration of the advanced third-stage larvae. Southeast Asia J. Trop. Med. Public Health 1:187–192. 14. Davies, N. J., and J. M. Goldsmid. 1978. Intestinal obstruction due to Ascaris suum infection. Trans. R. Soc. Trop. Med. Hyg. 72:107. 15. Deardorff, T. L., and M. L. Kent. 1989. Prevalence of larval Anisakis simplex in pen-reared and wild-caught salmon (Salmonidae) from Puget Sound, Washington. J. Wildl. Dis. 25:416–419. 16. Deardorff, T. L., and R. M. Overstreet. 1981. Larval Hysterothylacium (=Thynnascaris) (Nematoda: Anisakidae) from fishes and invertebrates in the Gulf of Mexico. Proc. Helminthol. Soc. Wash. 48:113–126. 17. Deardorff, T. L., and R. M. Overstreet. 1990. Seafoodtransmitted zoonoses in the United States: the fishes, the dishes, and the worms, p. 211–265. In D. R. Ward and D. R. Hackney (ed.), Microbiology of Marine Food Products. Van Nostrand Reinhold, New York, NY. 18. Deardorff, T. L., R. M. Overstreet, M. Okihiro, and R. Tam. 1986. Piscine adult nematode invading an open lesion in a human hand. Am. J. Trop. Med. Hyg. 35:827–830. 19. Deardorff, T. L., and R. Throm. 1988. Commercial blast freezing of third-stage Anisakis simplex larvae encapsulated in salmon and rockfish. J. Parasitol. 74:600–603. 20. de Brie, J. 1879. Le Bon Berger, ou le Vray Régime et Gouvernement des Bergers et Bergères: Compose par le Rustique Jehan de Brie le Bon Berger. Isidore Liseux, Paris, France.

SMP_Food Microbiology_CH27.indd

709

709

21. Desowitz, R. S. 1981. New Guinea Tapeworms and Jewish Grandmothers. Norton and Company, New York, NY. 22. Dooley, J. R., and R. C. Neafie. 1976. Anisakiasis, p. 475– 481. In C. H. Binford and D. H. Connor (ed.), Pathology of Tropical and Extraordinary Diseases, vol. 2. Armed Forces Institute of Pathology, Washington, DC. 23. Dooley, J. R., and R. C. Neafie. 1976. Clonorchiasis and opisthorchiasis, p. 509–516. In C. H. Binford and D. H. Connor (ed.), Pathology of Tropical and Extraordinary Disease, vol. 2. Armed Forces Institute of Pathology, Washington, DC. 24. Dooley, J. R., and R. C. Neafie. 1976. Oesophagostomiasis, p. 440–445. In C. H. Binford and D. H. Connor (ed.), Pathology of Tropical and Extraordinary Diseases, vol. 2. Armed Forces Institute of Pathology, Washington, DC. 25. Eberhard, M. L., and C. Busillo. 1999. Human Gongylonema infection in a resident of New York City. Am. J. Trop. Med. Hyg. 61:51–52. 26. Eberhard, M. L., H. Hurwitz, A. M. Sun, and D. Coletta. 1989. Intestinal perforation caused by larval Eustrongylides (Nematode: Dioctophymatoidae) in New Jersey. Am. J. Trop. Med. Hyg. 40:648–650. 27. Faulkner, C. T., S. E. Cowie, P. E. Martin. S. R. Martin, C. S. Mayes, and S. Patton. 2000. Archeological evidence of parasitic infection from the 19th century company town of Fayette, Michigan. J. Parasitol. 86:846–849. 28. Fox, A. S., K. R. Kazacos, N. S. Gould, P. T. Heydemann, C. Thomas, and K. M. Boyer. 1985. Fatal eosinophilic meningoencephalitis and visceral larva migrans caused by the raccoon ascarid Baylisascaris procyonis. N. Engl. J. Med. 312:1619–1623. 29. Freeman, R. S., P. F. Stuart, J. B. Cullen, A. C. Ritchie, A. Mildon, B. J. Fernandes, and R. Bonin. 1976. Fatal human infection with mesocercariae of the trematode Alaria americana. Am. J. Trop. Med. Hyg. 25:803–807. 30. Glickman, L. T. 1984. Toxocariasis. In K. S. Warren and A. A. F. Mahmoud (ed.), Tropical and Geographical Medicine. McGraw-Hill, New York, NY. 31. Glickman, L. T., P. M. Schantz, and R. H. Cypress. 1979. Canine and human toxocariasis: review of transmission, pathogenesis and clinical disease. J. Am. Vet. Med. Assoc. 175:1265–1269. 32. Gordon, J. A., C. M. D. Ross, and H. Affleck. 1969. Abdominal emergency due to an oesophagostome. Ann. Trop. Med. Parasitol. 63:161–164. 33. Graczyk, T. K., and B. Fried. 1998. Echinostomiasis: a common but forgotten food-borne disease. Am. J. Trop. Med. Hyg. 58:501–504. 34. Harrell, L. W., and T. L. Deardorff. 1990. Human nanophyetiasis: transmission by handling naturally infected coho salmon (Oncorhynchus kisutch). J. Infect. Dis. 161:146–148. 35. Hayunga, E. G. 1989. Parasites and immunity: tactical considerations in the war against disease—or, how did the worms learn about Clausewitz? Perspect. Biol. Med. 32:349–370. 36. Hillyer, G. V. 1981. Fascioliasis in Puerto Rico: a review. Bol. Assoc. Med. P. R. 73:94–101.

Manila Typesetting Company

11/08/2012 07:10AM

Nonbacterial Pathogens

710 37. Hou, P. C. 1965. Hepatic clonorchiasis and carcinoma of the bile duct in a dog. J. Pathol. Bacteriol. 89:365. 38. Huff, D. S., R. C. Neafie, M. J. Binder, G. A. DeLeon, L. W. Brown, and K. R. Kazacos. 1984. Case 4. The first fatal Baylisascaris infection in humans: an infant with eosinophilic meningoencephalitis. Pediatr. Pathol. 2:345–352. 39. Jones, E. A., J. M. Key, H. P. Milligan, and D. Owens. 1977. Massive infection with Fasciola hepatica in man. JAMA 63:836–842. 40. Katz, M., D. D. Despommier, and R. W. Gwadz. 1982. Parasitic Diseases. Springer-Verlag, New York, NY. 41. Kikuchi, Y., T. Takenouchi, M. Kamiya, and H. Ozake. 1981. Trypanorhynchiid cestode larva found on the human palatine tonsil. Jpn. J. Parasitol. 30:3497–3499. 42. Kisielewska, K. 1970. Ecological organization of intestinal helminth groupings in Clethrionomys glareolus (Schreb.) (Rodentia). I. Structure and seasonal dynamics of helminth groupings in a host population in the Bialowieza National Park. Acta Parasitol. Polonica 18: 121–147. 43. Le Bailly, M., U. Leuzinger, H. Schlichtherle, and F. Bouchet. 2005. Diphyllobothrium: neolithic parasite? J. Parasitol. 91:957–959. 44. Leuckart, K. G. 1882. Zur Entwickelungsgeschichte des Leberegels. Zweite Mittheilung. Zool. Anz. 5:524–528. 45. MacArthur, W. P. 1933. Cysticercosis as seen in the British Army, with special reference to the production of epilepsy. Trans. R. Soc. Trop. Med. Hyg. 27:343–363. 46. Meyers, W. M., R. C. Neafie, and D. H. Connor. 1976. Pentastomiasis, p. 546–550. In C. H. Binford and D. H. Connor (ed.), Pathology of Tropical and Extraordinary Diseases, vol. 2. Armed Forces Institute of Pathology, Washington, DC. 47. Miller, G. C. 1981. Helminths and the transmammary route of infection. Parasitology 82:335–342. 48. Miyazaki, I., and S. Habe. 1976. A newly recognized mode of human infection with the lung fluke, Paragonimus wes­ termani (Kerbert 1978). J. Parasitol. 62:646–648. 49. Muller, R. 1971. Dracunculus and dracunculiasis. Adv. Parasitol. 9:73–151. 50. New, D., M. D. Little, and J. Cross. 1995. Angiostrongylus cantonensis infection from eating raw snails. N. Engl. J. Med. 332:1105–1106. 51. Norton, R. A., and L. Monroe. 1961. Infection by Fasciola he­ patica acquired in California. Gastroenterology 41:46–48. 52. Reference deleted. 53. Ockert, G., and J. Obst. 1973. Ausstreuung umhüllter Onkosphären durch Bandwurmträger. Monatsh. Veterinarmed. 28:97–98. 54. Ollerenshaw, C. B. 1980. Forecasting liver fluke disease, p. 33–52. In A. E. R. Taylor and R. Muller (ed.), 12th Annual Symposium of the British Society for Parasitology. Blackwell, London, United Kingdom. 55. Paul, A. J., K. S. Todd, Jr., and J. A. Dipietro. 1988. Environmental contamination by eggs of Toxocara species. Vet. Parasitol. 26:339–342. 56. Rosen, L., G. Loison, J. Laigret, and G. D. Wallace. 1967. Studies on eosinophilic meningitis. 3. Epidemiologic and

SMP_Food Microbiology_CH27.indd

710

clinical observations on Pacific islands and the possible etiologic role of Angiostrongylus cantonensis. Am. J. Epidemiol. 85:17–44. 57. Sadun, E. H., and C. Maiphoom. 1953. Studies in the epidemiology of the human intestinal fluke, Fasciolopsis buski (Lankester) in Central Thailand. Am. J. Trop. Med. Hyg. 2:1070–1084. 58. Sakanari, J. A., H. M. Loinaz, T. L. Deardorff, R. B. Raybourne, H. H. McKerrow, and J. G. Frierson. Intestinal anisakiasis: a case diagnosed by morphologic and immunologic methods. Am. J. Clin. Pathol. 90:107–113. 59. Sakanari, J. A., and J. H. McKerrow. 1989. Anisakiasis. Clin. Microbiol. Rev. 2:278–284. 60. Sanchez, J. L., I. Hernandez-Fragoso, C. Rios, and C. K. Ho. 1990. Parasitological evaluation of a foodhandler population cohort in Panama: risk factors for intestinal parasitism. Mil. Med. 156:250–255. 61. Sato, H., Y. Une, S. Kawakami, E. Saito, H. Kamiya, N. Akao, and H. Furuoka. 2005. Fatal Baylisascaris larva migrans in a colony of Japanese macaques kept by a safari-style zoo in Japan. J. Parasitol. 91: 716–719. 62. Schantz, P. M. 1989. The dangers of eating raw fish. N. Engl. J. Med. 320:1143–1145. 63. Sianto, L., K. J. Reinhard, M. Chame, S. Mendonca, M. L. C. Concalves, A. Fernandes, L. F. Ferreira, and A. Araujo. 2005. The finding of Echinostoma (Trematoda: Digenea) and hookworm eggs in coprolites collected from a Brazilian mummified body dated 600–1,200 years before present. J. Parasitol. 91:972–975. 64. Silverman, P. H., and R. B. Griffiths. 1955. A review of methods of sewage disposal in Great Britain with special reference to the epizootiology of Cysticercus bovis. Ann. Trop. Med. Parasitol. 49:436–450. 65. Stehr-Green, J. K., P. A. Stehr-Green, P. M. Schantz, J. F. Wilson, and A. Lanier. 1988. Risk factors for infection with Echinococcus multilocularis in Alaska. Am. J. Trop. Med. Hyg. 38:380–385. 66. Stoll, N. R. 1947. This wormy world. J. Parasitol. 33:1–18. 67. Toma, H., S. Matsumura, C. Oshiro, and Y. Sato. 2002. Ocular angiostrongyliasis without meningitis symptoms in Okinawa, Japan. J. Parasitol. 88:211–213. 68. Traub, R. J., I. D. Robertson, P. Irwin, N. Mencke, and R. C. A. Thompson. 2002. The role of dogs in transmission of gastrointestinal parasites in a remote tea-growing community in northeastern India. Am. J. Trop. Med. Hyg. 67:539–545. 69. Uga, S., and N. Kataoka. 1995. Measures to control Toxocara egg contamination in sandpits of public parks. Am. J. Trop. Med. Hyg. 52:21–24. 70. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. 2009. Anisakis simplex and related worms. In Foodborne Pathogenic Microorganisms and Natural Toxins Handbook (Bad Bug Book). U.S. Food and Drug Administration, Silver Spring, MD. http://www. fda.gov/Food/FoodSafety/FoodborneIllness/FoodborneI llnessFoodbornePathogensNaturalToxins/BadBugBook/ ucm070768.htm. 71. Viel, J. F., P. Giraudoux, V. Abrial, and S. Bresson-Hadni. 1999. Water vole (Arvicola terrestris scherman) density

Manila Typesetting Company

11/08/2012 07:10AM

27.  Helminths from Finfish, Shellfish, and Other Foods as risk factor for human alveolar echinococcosis. Am. J. Trop. Med. Hyg. 61:559–565. 72. Vince, J. D., R. W. Ashford, M. J. Gratten, and J. Bana-Koiri. 1979. Strongyloides species infestation in young infants of Papua New Guinea: association with generalized oedema. Papua New Guinea Med. J. 22:120–127. 73. Voelker, R. 2007. Persistence pays off in guinea worm fight. JAMA 298:1856–1857.

SMP_Food Microbiology_CH27.indd

711

711

74. Weaver, H. J., J. M. Hawdon, and E. P. Hoberg. 2010. Soiltransmitted helminthiases: implications of climate change and human behavior. Trends Parasitol. 26:574–581. 75. Wittner, M., J. W. Turner, G. Jacquotte, L. R. Ash, M. P. Salgo, and H. B. Tanowitz. 1989. Eustrongylidiasis—a parasitic infection acquired by eating sushi. N. Engl. J. Med. 320:1124–1126. 76. Yokogawa, M. 1969. Paragonimus and paragonimiasis. Adv. Parasitol. 7:375–387.

Manila Typesetting Company

11/08/2012 07:10AM

SMP_Food Microbiology_CH27.indd

712

Manila Typesetting Company

11/08/2012 07:10AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch28

28

Ynes R. Ortega

Protozoan Parasites

Protozoan parasites have long been associated with foodborne and waterborne outbreaks of disease in humans. Difficulties arise with the inactivation of these organisms because of their resistance to environmental stresses. Groups of protozoan parasites transmitted via foods include Apicomplexa, flagellates, ciliates, and amoebae (Table 28.1). A major characteristic of apicomplexan parasites is that a vertebrate host is required to complete the complex life cycle and produce infectious cysts. Of this group, Cryptosporidium species, Cyclospora cayetanen­ sis, Sarcocystis hominis and Sarcocystis suihominis, and Isospora belli frequently inhabit the intestinal mucosa and produce diarrheal illnesses in humans. These coccidian parasites affect immunocompetent as well as immunocompromised individuals, causing more severe and prolonged symptoms in the latter. Cryptosporidium can infect animals, and some genotypes have been described as having host preferences or specificities. C. cayetanen­ sis has been isolated exclusively from humans. Another apicomplexan, Toxoplasma gondii, infects human tissues other than the intestinal mucosa and can cause birth defects, blindness, and chorioretinitis. Sarcocystis species in humans can also infect muscles and other organs. The life cycle stages of apicomplexan parasites are

produced intracellularly in the host. For Cyclospora, Toxoplasma, and Isospora, sporogony typically occurs outside the host, requiring the passage of time before oocysts are infective to a new host, while Cryptosporidium oocysts are excreted already sporulated and are infectious when shed. Microsporidia, once considered to be protozoan parasites, have been reclassified as fungi based on recent phylogenetic analyses. Microsporidia have been implicated in human foodborne and waterborne diseases and are able to propagate and complete their life cycle intracellularly. Of the microsporidia, members of five genera have been implicated in human diseases: Encephalitozoon, Enterocytozoon, Septata, Pleistophora, and Vittaforma. None of these species are host, tissue, or organ specific, with the exception of Enterocytozoon bieneusi, which appears to infect only the human intestinal tract. Flagellates, ciliates, and amoebae can propagate in more than one host. All three types of parasites can cause disease in humans and animals. The infectious stage of this group is the cyst. Once ingested by the susceptible host, the trophozoites or motile forms are released from the cyst and colonize the host’s intestinal cells. In contrast with Apicomplexa, microsporidia, and trypanosomes, flagellates, ciliates, and amoebae do

Ynes R. Ortega, Center for Food Safety, University of Georgia, 1109 Experiment Street, Griffin, GA 30223-1797.

713

SMP_Food Microbiology_CH28.indd

713

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

714 Table 28.1  Protozoa of medical importance acquired from

food and water

Phylum or group Apicomplexa

Protozoa(n) Cryptosporidium parvum

Oocyst

Cyclospora cayetanensis Isospora belli Toxoplasma gondii

Oocyst Oocyst Oocyst/tissue cyst Oocyst Oocyst Sarcocyst Cyst Spore Spore

Sarcocystis hominis Sarcocystis suihominis Sarcocystis spp. Ciliophora Balantidium coli Microsporaa Enterocytozoon bieneusi Encephalitozoon intestinalis Sarcomastigophora Acanthamoeba spp.b

Euglenozoa

Infective stage

Trophozoite

Dientamoeba fragilis Entamoeba disparc

Trophozoite Cyst

Entamoeba histolytica

Cyst

Giardia intestinalis

Cyst

Naegleria fowlerib

Trophozoite

Trypanosoma cruzi

Metacyclic trypomastigote

a Microsporidia were originally considered parasites but have been reclassified as fungi. b Species of normally free-living amoebae that invade through the mucosa and can in some instances infect the central nervous system of humans. Amoebic meningitis is usually fatal. c Commensal, morphologically similar to E. histolytica, and not normally capable of inducing disease.

not present intracellular stages but instead attach with specialized structures to the luminal surface of the intestinal cells and feed off the nutrients and cellular debris of the intestinal epithelium. Giardia lamblia, a flagellate previously considered to be a commensal organism, is now well recognized as an etiological agent of acute or chronic diarrhea in humans. Balantidium coli, a ciliate, also causes diarrhea in humans, and if the infection is not treated, it can cause ulcerative colitis. Various species of amoebae can infect humans. Most of them are commensal, not producing disease; however, others can cause diarrhea, dysentery, or amebomas if the infection is not treated. Trypanosoma cruzi is a vector-borne flagellate para­ site that in the past has been responsible for various foodborne outbreaks. If the infection is not treated effectively during the acute phase, infection is usually lifelong, as there is no treatment for the chronic stage of the infection. The outcome is heart disease and may be complicated with megasyndromes, particularly megaesopha-

SMP_Food Microbiology_CH28.indd

714

gus and megacolon (121). In Venezuela, an outbreak of American trypanosomiasis affected 103 individuals at a college community and was associated with consumption of guava juice. Of those, 75% were symptomatic, 44% had documented parasitemia, and one child died (5). In Brazil, 178 cases of acute Chagas’ disease were associated with consumption of acai juice (127). The emergence of Chagas’ disease in the United States is possible, as opossums (sylvatic reservoir) and insect vectors infected with T. cruzi have been identified in the southern part of the country (54).

CRYPTOSPORIDIUM Cryptosporidium was first isolated in 1910 from the intestines of mice. Various species of Cryptosporidium capable of infecting animals have since been described. It was not until 1976 that Cryptosporidium parvum infection in humans was first described. Since 1982, C. parvum has been one of the most frequently identified opportunistic pathogens associated with diarrhea and wasting syndrome in patients with AIDS. Cryptosporidiosis in immunocompromised patients produces a life-threatening, prolonged, cholera-like illness (166). In immunocompetent patients, C. parvum causes self‑resolving, acute diarrheal disease with variability in the severity of symptoms. Isolate variations and associations with the severity of disease have been subjects of intensive investigation for some time. All Cryptosporidium infections were initially thought to be caused by C. parvum, even though there was some evidence that more than one Cryptosporidium species might be implicated. As molecular tools were developed, speciation of Cryptosporidium allowed for the description of C. hominis, previously recognized as C. parvum genotype I, which is isolated exclusively from humans. Genotype II, which belongs to the C. parvum group, was also described. This genotype is isolated from cattle and humans and is readily infective to laboratory animals. Subsequently, other Cryptosporidium species, originally called C. parvum, were described. PCR targeting the small-subunit rRNA gene is usually employed to genotype Cryptosporidium isolated from humans, animals, and water samples. Genotype differentiation based on the sequencing of the thrombospondinrelated adhesive protein (TRAP-C2) has also been used (124, 138). Currently, the GP60 gene is being used to subtype Cryptosporidium. Six subtypes have been described for C. hominis and 11 subtypes for C. parvum. The denomination of each subtype includes the group type (I or II) and the type and number of the TCA, TCG, or TCT repeats. In some cases, other repeats (of 6 bp or 13 to 15 bp) located after the trinucleotide repeats are also in-

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

715

cluded in the C. parvum or C. hominis subtype name, respectively. Human infections with Cryptosporidium are most frequently caused by five Cryptosporidium species (C. hominis, C. parvum, C. felis, C. canis, and C. meleagridis) and occur most frequently in HIV-positive individuals (33, 55, 67, 78, 120, 137, 142). Other species occasionally identified in humans include C. muris, C. andersoni, C. suis, and Cryptosporidium cervine, horse, rabbit, skunk, and chipmunk I genotypes (174). Cryptosporidiosis is acquired after ingesting food or water contaminated with infective Cryptosporidium oocysts. Once ingested, the oocysts excyst (Fig. 28.1). The released sporozoites proceed to invade the enterocytes. A parasitophorous vacuole is formed as the parasite enters

the cell. This structure contains the intracellular parasites and communicates with the host cell via a feeder organelle. The parasite goes through two asexual multiplication stages called merogony: type I meronts contain eight merozoites, which are released and proceed to invade other enterocytes to form type II meronts, each containing four merozoites. Once merozoites from type II meronts are released, they infect enterocytes but differentiate into sexual stages identified as macrogametocytes (female) and microgametocytes (male). The union of a microgametocyte and a macrogametocyte produces a zygote (immature oocyst), which matures within its host into a fully sporulated oocyst (Fig. 28.2). Two types of oocysts are produced and excreted. Thin-walled

Figure 28.1  Scanning electron micrographs of oocysts and excysting sporozoites of Cryptosporidium parvum. (A) Intact oocyst prior to excystation (×11,200). (B) Three sporozoites (Sp) excysting from oocyst simultaneously via the cleaved suture (Su) (×11,200). (C) Empty oocyst (×11,200). (D) Excysted sporozoite (×9,800). Ae, apical end. From Reduker et al. (147). doi:10.1128/9781555818463.ch28f1

SMP_Food Microbiology_CH28.indd

715

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

716

Figure 28.2  Life cycle of Cryptosporidium. From Dubey et al. (58). doi:10.1128/9781555818463.ch28f2

oocysts excyst endogenously, resulting in autoinfection. Thick-walled oocysts, which are environmentally resistant, are shed in the feces and are immediately infectious to other hosts (46, 178). An additional life stage that may represent an extracellular trophozoite/gamont stage has been described in C. andersoni, a parasite that infects the gut of cattle. This stage has also been reported in C. parvum. This finding is still under debate, as other groups have not been able to reproduce it. Phylogenetic analysis indicates that Cryptosporidium has a closer phylogenetic affinity with the gregarines than with the coccidia (36, 93, 153). To date, 15 species of Cryptosporidium have been identified in various animals, of which C. canis (dogs), C. felis (cats), C. muris (rodents), C. parvum (cattle, sheep, and goats), C. meleagridis (turkeys), and C. hom­ inis (humans) are known to infect humans. Generally, C. parvum infects the brush border of the intestinal epithelium and causes villous atrophy, though the exact mechanism for this pathological alteration is not yet

SMP_Food Microbiology_CH28.indd

716

fully understood. Immunocompetent patients develop a profuse watery diarrhea accompanied by epigastric cramping, nausea, and anorexia, which is usually selflimiting and lasts for about 15 days. Extraintestinal dissemination has been observed in immunocompromised patients. Cryptosporidium may be found in epithelial cells from other organs such as those of the respiratory tract and the biliary tree (113). Immunocompromised patients (e.g., AIDS patients or those receiving immunosuppressive drugs) develop severe diarrhea (3 to 6 liters/day), which persists for several weeks to months or even years. These cases are the most severe and life threatening, with continuous shedding of oocysts. In patients with HIV infections, the CD4+ cell count is the best marker for the ability of the immune system to self-resolve the infection. In the developed world, patients with CD4+ counts of 180 cells/mm3 or higher usually develop self-limiting cryptosporidiosis, while those with counts below 180 cells/mm3 usually develop chronic and profuse diarrhea, which is exacer-

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

717

bated by the lack of effective therapy (72). The pathogenesis of the diarrhea is not clear, and the presence of a toxin has been suggested but not yet demonstrated. The mean prevalence of C. parvum infection in Europe and the United States is 1 to 3%, and the prevalence is considerably higher in developing countries. Outbreaks of disease associated with Cryptosporidium have been reported in the United States as well as in other countries (178). Cryptosporidium infection is highly associated with travel abroad, exposure to farm animals, and person-to-person transmission in settings such as day care centers and medical institutions. A large number of waterborne outbreaks of cryptosporidiosis have been documented, and some of the most notable were those in Milwaukee, Wisconsin (116), and in Georgia (89), with a total of more than 400,000 cases reported. In Minnesota in 1995, Cryptosporidium was associated with cases of acute gastroenteritis experienced by attendees of a social event. This outbreak was epidemiologically associated with the consumption of contaminated chicken salad (9). In 1993 in central Maine and in 1996 in New York, apple cider was associated with outbreaks of cryptosporidiosis (11, 122). In 1998 in Spokane, Washington, another foodborne outbreak affected about 50 people, but it could not be traced to a specific type of food. In 2000, an infected food handler was responsible for a foodborne outbreak at a cafeteria of a university in Washington (143). Cryptosporidium has been identified in produce intended for human consumption. In 2008, a foodborne outbreak of cryptosporidiosis affecting 72 individuals in Helsinki was associated with a salad mixture (140). In Norway, 26% of lettuce samples and 74% of mung bean sprout samples examined were Cryptosporidium positive. Giardia was identified in dill (20%), lettuce (20%), mung bean sprouts (30%), radish (10%), and strawberry (20%) samples (142). The risk of acquiring cryptosporidiosis and giardiasis based on frequency and level of mung bean sprout consumption was assessed to be 20 cases per 100,000 (151, 152). Analysis of produce obtained from a regional market in Peru revealed that 14.5% of samples contained Cryptosporidium oocysts (133). The immunofluorescence assay specific to Cryptosporidium has been used as a gold standard assay. An alternative method for identifying Cryptosporidium from fresh produce has been developed by using polyclonal sera raised to a recombinant viral capsid protein of an RNA virus that is a symbiont of Cryptosporidium (103). In a recent report, it was estimated that 8% of cases of cryptosporidiosis are foodborne and the annual number of episodes of domestically acquired foodborne

SMP_Food Microbiology_CH28.indd

717

Cryptosporidium spp. was placed at 57,616 (156). This is much smaller than the number estimated in 1999. Although there is no effective and definite therapeutic agent for the treatment of cryptosporidiosis, spiramycin has been reported to decrease diarrhea in the early stages of infections (125, 141, 155); however, it is not effective in advanced infections. Azithromycin, nitazoxanide, and paromomycin have been reported to be effective in AIDS patients with cryptosporidiosis (27, 48, 74, 79). Alternative experimental therapies that have been evaluated in humans include the use of passive immunotherapies such as hyperimmune bovine colostrum. Monoclonal antibodies and polyclonal hyperimmune hen yolk against Cryptosporidium have also been tested in animal models with some success (35, 65, 139). To date, nitazoxanide is the only drug approved by the Food and Drug Administration for the treatment of cryptosporidiosis. This treatment seems most effective in malnourished children. Nitazoxanide, paromomycin, and rifamycin derivatives have been reported to be partially effective in severely immunocompromised patients with cryptosporidiosis (32). Cryptosporidium oocysts are 4 to 6 mm in diameter. Because of their size, they can be overlooked in examinations of feces or confused with yeast cells. Cryptosporidium can be identified by light microscopy using acid-fast staining. Immunoassays (fluorescent-antibody immunoassay or enzyme immunoassay) are available and are very sensitive and specific diagnostic procedures (15, 16, 37, 169, 170). Cryptosporidium can be mistaken for Cyclospora, which is also positive in acid-fast staining but larger (Fig. 28.3). It is important that laboratories carefully measure the diameters of cells, particularly if the cells appear to be larger than those of Cryptosporidium. New molecular diagnostic tests such as PCR are being developed and could improve the sensitivity in detecting Cryptosporidium oocysts in produce (20, 45, 51, 81, 102, 103, 106, 150). It has been demonstrated that shellfish can concentrate large volumes of Cryptosporidium oocysts along with other particulate matter (66, 80, 83, 85, 87, 117). Cryptosporidium has also been identified in shellfish organs by using conventional diagnostic assays, and molecular tools (multiplex nested PCR) have been examined to optimize parasite identification (80–82, 84). The use of natural grassland buffers may contribute to the control of the parasite load that may potentially be deposited in water banks and oceans as a consequence of rangeland runoff, thus contaminating shellfish (18). To date, shellfish have not been associated with outbreaks of Cryptosporidium infections. Although Cryptosporidium oocysts are highly resistant to chlorine and chlorine dioxide, treatment with 1,000 mg/ml for 1 min resulted in 0.5- and 2-log reductions,

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

718

respectively, according to the viability assay (in vitro excystation or tissue culture infectivity) (38). More than 90% inactivation was achieved when Cryptosporidium oocysts were treated with ozone at 1 mg/ml for 5 min, chlorine dioxide at 1.3 mg/ml for 1 h, and chlorine and monochloramine at 80 mg/ml for 90 min (105).

CYCLOSPORA

Figure 28.3  Acid-fast staining. (A) Cryptosporidium par­ vum. (B) Cyclospora cayetanensis. (C) Isospora belli. (Bars = 20 mm.) doi:10.1128/9781555818463.ch28f3

SMP_Food Microbiology_CH28.indd

718

Cyclospora was probably first reported in humans in 1979 by Ashford, who described it as an Isospora-like coccidian affecting humans in Papua, New Guinea (17). Thereafter, other investigators found similar structures in fecal samples of patients with diarrhea, but because of the morphologies of the unsporulated oocysts and their autofluorescence, they were considered to be CLBs (coccidian-like bodies or cyanobacterium-like bodies) (108, 109, 161). In 1993, conclusive identification was made and the CLBs were fully characterized as coccidian parasites and placed in the genus Cyclospora with the proposed species name cayetanensis (131, 134). Cyclospora belongs to the family Eimeriidae, subphylum Apicomplexa. Cyclospora species infect moles, rodents, insectivores, snakes, and humans. In 1998, Cyclospora species that morphologically resemble Cyclospora isolated from humans were isolated from nonhuman primates (60, 112, 165). However, phylogenetic analysis has demonstrated that these isolates are different species (60, 112). Even before their true identity was established, epidemiological information on the intriguing CLBs was being collected worldwide. In Nepal, the prevalence of Cyclospora infection is highest in adult expatriates (162), whereas in areas of Peru where infection is endemic, children under 10 years of age are the most susceptible to infection, though most are asymptomatic (118). Adults from areas of endemicity do not develop symptoms of the infection, but adults of medium to high socioeconomic status living outside of the areas of endemicity, as well as foreign travelers, present with clinical disease. These observations suggest that prior exposure to the parasite may result in protective immunity. Newly shed oocysts (Fig. 28.4A) of C. cayetanensis require 2 weeks to sporulate and become infectious (Fig. 28.4B) under optimal laboratory conditions (134). The requirement of this period for the oocyst to become infectious suggests that contamination of produce occurs with oocysts that are fully sporulated or almost fully sporulated. Otherwise, unsporulated oocysts would require optimal time and environmental conditions to induce oocyst sporulation, while produce would still remain edible. In studies performed in areas of endemicity, Cyclospora oocysts have been isolated from produce by washing thoroughly with distilled water. The washes are concentrated

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

719

Figure 28.4  Cyclospora cayetanensis oocysts. (A) Phase-contrast microscopy of unsporulated oocysts. (B) Oocysts (OO) in the process of excystation. Note the two sporozoites (SP) free of the sporocyst (SC). (C) Transmission electron microscopy of human small intestine showing Cyclospora intracellular stages. ME, merozoite. doi:10.1128/9781555818463.ch28f4

by centrifugation, and pellets are fixed and preserved in 10% formalin. Samples are examined directly by using epifluorescence microscopy and phase-contrast microscopy (134). Nested PCR targeted to amplify the 18S rRNA gene is also used with these preparations; however, it should be noted that further testing using restriction fragment length polymorphism is required, since the described PCR cross-reacts with Eimeria species. Eimeria parasites are infectious to animals but not to humans and can be readily found in the environment (100, 160). Cyclosporiasis is characterized by mild to severe nausea, anorexia, abdominal cramping, mild fever, and watery diarrhea. Diarrhea alternating with constipation has been commonly reported. Some patients present with flatulent dyspepsia and, less frequently, joint pain and night sweats. The onset of illness is usually sudden, and symptoms persist for an average of 7 weeks (42, 132). C. cayetanensis infects epithelial cells of the duodenum and jejunum of humans. Merogony and gametog-

SMP_Food Microbiology_CH28.indd

719

ony occur intracytoplasmically within the parasitophorous vacuoles in intestinal cells (Fig. 28.4C) (132, 168). Duodenal and jejunal biopsy specimens from patients with cyclosporiasis show various degrees of jejunal villous blunting, atrophy, and crypt hyperplasia (42, 132). Extensive lymphocytic infiltration into the surface epithelium is present, especially at the tips of the shortened villi. The reactive inflammatory response of the host does not correlate with the number of intracellular parasites present in the tissues (132). Routes of transmission for Cyclospora are undocumented, although the fecal-oral route, either directly or via water and food, is probably the major one. In the United States, epidemiological evidence suggests that water has been responsible for sporadic cases of cyclosporiasis. In Utah, a man became infected after cleaning his basement that was flooded with runoff from a nearby farm following heavy rains. In 1990, an outbreak involving residents in a physicians’ dormitory in a Chicago hospital was

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

720 epidemiologically associated with tap water from unprotected reservoir tanks that served the building and which had a broken water pump (13, 97). In Pokhara, Nepal, British Gurkha soldiers were confirmed to have cyclosporiasis; oocysts were isolated from drinking water. The water, consisting of a mixture of river and municipal water, was routinely chlorinated and served the houses in the camp where the soldiers were stationed (144). A few reports have described the isolation of C. cay­ etanensis oocysts from animals (chickens, ducks, and dogs) (39, 77, 175, 176). To date, after several experimental studies attempting to infect these and additional animal species, it seems that Cyclospora species are host specific (61). Since the early 1990s, Cyclospora has been linked to waterborne and foodborne disease outbreaks. Foodborne outbreaks have been epidemiologically associated with the consumption of contaminated fresh produce such as raspberries, lettuce, basil, and snow peas (10, 12, 90, 91). In 1996, a large foodborne outbreak occurred following a wedding reception at a restaurant in Boston, resulting in 57 guests having cyclosporiasis. Berries were implicated as the vehicle of transmission for Cyclospora oocysts (71). In 1999, another foodborne outbreak among attendees of two events was reported. Sixty-two cases of infection were documented, and the illness was associated with consumption of chicken pasta salad and tomato basil salad. The most likely vehicle of illness was fresh basil, grown either in Mexico or in the United States, which was included in both salads (110). However, Cyclospora oocysts were not recovered from or detected in the produce associated with these outbreaks. In the United States, most cases of cyclosporiasis have been reported in the months of April to August, suggesting some possible seasonality. In the United States in 1996 and 1997, Cyclospora infections were associated with imported raspberries (91, 92). It was speculated that contamination of produce could have occurred when raspberries were sprayed with insecticide possibly diluted with contaminated surface water. Analysis of irrigation water demonstrated the presence of Cyclospora oocysts (25). In Peru, C. cayetanensis oocysts have been isolated from vegetables obtained from markets in areas of endemicity. In studies where vegetables were experimentally inoculated with C. cayetanensis oocysts, it was demonstrated that washing with water does not remove all the oocysts (133). Cyclospora has also been isolated from produce from Nepal (159). In Germany in 2000, 34 individuals acquired Cyclospora infections, which were associated with

SMP_Food Microbiology_CH28.indd

720

consumption of lettuce spiced with green leafy herbs all imported from southern Europe (56). That same year, another outbreak occurred at a wedding reception in Philadelphia. The wedding cake tested positive, and the cream filling containing raspberries was the food item strongly associated with illness (95). In 2001, in British Columbia, Canada, 17 cases of cyclosporiasis were reported. Eleven of these individuals had consumed Thai basil that was imported from the United States (96). In 2004, approximately 50 cases of cyclosporiasis were linked to consumption of raw Guatemalan snow peas (14). In 2009, a Cyclospora outbreak was reported in Sweden, involving 12 confirmed and 6 probable cases, and was associated with imported sugar snap peas from Guatemala (99). In the most recent report from the Centers for Disease Control and Prevention, it is estimated that 99% of cases of Cyclospora infection in the United States are foodborne and that there are an estimated 11,407 cases annually (156). Neither the minimum infectious dose of oocysts nor the sporulation and survival rates under different environmental conditions are known, although human volunteer studies have been attempted without success (6a). Because of the potential low number of oocysts present in foods, molecular assays and methodologies to improve parasite recovery have been studied (100, 101, 130). Analysis of the intervening transcribed spacer-1 (ITS1) region sequence of Cyclospora isolates from the 1996 outbreak demonstrated that all were identical, suggesting that a single source of contamination caused the infection (2). Another study examined C. cayetanensis and Cyclospora papionis isolates. Although high sequence variability is present, conserved species-specific ITS-1 sequences were identified. This consistent and remarkable diversity among Cyclospora species ITS-1 sequences argues for polyparasitism and simultaneous transmission of multiple strains (128). There is strong evidence to suggest seasonality of Cyclospora infections. In Peru, more than 6 years of prospective epidemiological studies investigating endemic Cyclospora infections found that nearly all infections occurred between December and July (118). Rarely were infections documented at any other time of the year. In the United States, major cases have occurred during the period from May to July. In Nepal, infection and illness occur most frequently from May to August. The specific reasons for this marked seasonality have not been elucidated. The only successful antimicrobial treatment for Cy­ clospora is trimethoprim-sulfamethoxazole (TMP-SMX)

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

721

(118, 135, 172). AIDS patients appear to have higher parasite infestation levels than immunocompetent individuals infected with Cyclospora. However, the prevalence of Cyclospora infection among HIV patients is not higher than among immunocompetent populations. This is probably due to the frequent use of TMP-SMX for Pneumocystis (carinii) jiroveci prophylaxis among HIV patients (135). Ciprofloxacin has also been examined as an alternative treatment for sulfa-sensitive patients (172).

ISOSPORA Isospora is a coccidian parasite that infects humans. It is frequently identified in AIDS patients (21, 40). Isospora can be acquired by ingestion of contaminated food or water. Oocysts excyst in the intestine, and sporocysts are released. The sporocysts in turn release sporozoites that infect intestinal epithelial cells. Asexual and sexual life cycle stages occur in the cytoplasm of enterocytes. Unsporulated Isospora oocysts require 12 to 48 h to mature and become infectious outside the host. Isospora can be detected in stools from infected patients by observing unsporulated oocysts shed in feces. I. belli oocysts (10 to 19 by 20 to 30 mm) can be detected in direct wet mount preparations in heavy infections during ovum and parasite examination (Fig. 28.5). However,

Figure 28.5  Bright-field photomicrograph of Isospora belli oocysts unsporulated (A) and sporulated (B). doi:10.1128/9781555818463.ch28f5

SMP_Food Microbiology_CH28.indd

721

most infections are not heavy, and rates of shedding of oocysts may be variable; therefore, it is necessary to examine a series of samples and perform modified acid-fast staining (Fig. 28.3C). I. belli infects the entire intestine and produces severe intestinal disease (41). Deaths from overwhelming infections have been reported, especially in immunocompromised patients. Symptoms include diarrhea, nausea, steatorrhea, headache, and weight loss. The disease may persist for months and even years (49). Isospora infection is rare in immunocompetent people but occurs in 0.2 to 0.3% of immunocompromised AIDS patients in the United States and 8 to 20% of AIDS patients in Africa and Haiti. Isospora infection is endemic in many parts of Africa, Asia, and South America. The treatment of choice is TMP-SMX. In HIV patients, recurrence is common after discontinuation of therapy (135).

TOXOPLASMA T. gondii is a coccidian parasite that infects a variety of warm-blooded hosts. Cats are the definitive hosts, and other warm-blooded animals can serve as intermediate hosts. Cats excrete oocysts in their feces. Oocysts are environmentally resistant and can survive several years in moist, shaded conditions. Infections are acquired principally by ingestion of food and water containing oocysts, ingestion of animal tissues containing cystic forms (bradyzoites), or transplacental transmission (57). The unsporulated oocysts require 24 h outside the host to differentiate and become infectious. When oocysts are ingested by the intermediate host, the oocyst walls are ruptured and the sporozoites are released (Fig. 28.6). They invade epithelial cells and rapidly multiply asexually, producing tachyzoites. Tachyzoites multiply by endodyogeny, a process in which the mother tachyzoite is consumed by the formation of two daughter zoites (Fig. 28.7). Eventually bradyzoites, which are slowly multiplying forms, develop, forming tissue cysts. Cysts persist for the duration of the life of the host. By encysting, the parasite evades the host’s immune response and ensures its viability (52, 58). When animal tissues containing cysts are ingested by cats, proteolytic enzymes digest the cyst walls and the bradyzoites are released. They infect and multiply in the intestinal epithelial cells, transforming into tachyzoites. These in turn disperse via blood and lymph. When tissues of other animal species are ingested by felines, tachyzoites or bradyzoites begin the enteroepithelial cycle and sexual multiplication also occurs (Fig. 28.6). Macro- and microgametocytes are produced. After fertilization, the zygote differentiates into oocysts, which are passed in the feces (58).

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

722

Figure 28.6  Life cycle of Toxoplasma gondii. From Dubey and Beattie (57). doi:10.1128/9781555818463.ch28f6

Although most Toxoplasma infections occur by ingestion of contaminated meat, other foods, or water, infections can also be acquired by organ transplantation or by blood transfusion (22, 31, 43, 146). Disseminated toxoplasmosis may occur in patients who have received organ transplants and are receiving immunosuppressive therapy. Chorioretinitis is frequently observed in adults who acquire the infection. In immunosuppressed patients, toxoplasmosis can reactivate from latent infections (53, 148). In the United States, it is estimated that 50% of cases of toxoplasmosis are foodborne, representing 76,840 cases per year (156). Toxoplasmosis can also be acquired vertically by transplacental transmission when a pregnant woman becomes infected. After multiplying in the placenta, tachyzoites spread into the fetal tissues. Infection can occur at any stage of the pregnancy, but the fetus is affected the most when infection occurs during the first

SMP_Food Microbiology_CH28.indd

722

months of pregnancy. Most infected children do not show any signs of the disease until later in life, when they may present with chorioretinitis and mental retardation (3, 7, 30, 50, 98). The overall prevalence in humans and animals varies according to eating habits and lifestyle. The prevalence of T. gondii is highest in swine with outdoor access. Confined housing reduces the exposure to cat feces and infected rodents (3, 24). Toxoplasmosis can be acquired by ingestion of lamb, poultry, horse, and wild game animals. Cooking, freezing, or gamma irradiation will kill the Toxoplasma cysts and oocysts. Temperatures of 61°C or higher for 3.6 min will inactivate the parasites, and freezing at –13°C will result in death of cysts (57, 115). Pyrimethamine in combination with folinic acid or trisulfapyrimidine is the treatment of choice for acute infections. TMP-SMX is effective and frequently used to prevent recurrence of acute infections in AIDS patients (73).

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

723

Figure 28.7  Transmission electron micrographs of Toxoplasma gondii. (A) Sporozoite in parasitophorous vacuole (Pv) of host cell (Hc) at 24 h after inoculation. Am, amylopectin granule; Co, conoid; Mn, microneme; Nu, nucleus of sporozoite; Rh, rhoptry. (B) Final stage of endodyogeny to form two daughter tachyzoites that are still attached to the posterior ends (arrowheads). DG, dense granules; HC, host cell; IT, intravacuolar tubules; M, mitochondrion; MC, microneme; MI, micropore; N, nucleus; RO, rhoptry. From Dubey and Beattie (57). doi:10.1128/9781555818463.ch28f7

SARCOCYSTIS Another coccidian that causes significant foodborne infection is Sarcocystis. Muscular sarcocystosis is caused by several species of Sarcocystis. The oocysts, which are excreted in the feces of the infected animal or individual, contain two sporocysts, each with four sporozoites. The oocyst cell wall is thin and usually breaks, and the sporocysts are more commonly identified in the feces. The intermediate hosts are pigs, cattle, sheep, etc., and humans acquire the infection when they ingest meats containing muscle cysts (sarcocysts) full of bradyzoites (63). Humans can also be definitive hosts for S. hominis and S. suihominis. These sarcocysts are released from the infected muscle and infect the intestinal lamina propria. The bradyzoites differentiate into sexual stages, and after fertilization, oocysts are produced. Oocysts measure 12.3 to 14.6 by 18.5 to 20 mm (63). Patients with muscular sarcocystosis present with musculoskeletal pain, fever, rash, cardiomyopathy, bronchospasm, and subcutaneous swelling. The intestinal infection is characterized by nausea, vomiting, diarrhea, dyspnea, tachycardia, and stomachache (63). The infection can be acquired by ingestion of raw or undercooked meats, other foods in-

SMP_Food Microbiology_CH28.indd

723

cluding vegetables, and water contaminated with feces of a variety of animals containing oocysts/sporocysts.

MICROSPORIDIA Most microsporidium infections have been reported to occur in AIDS patients (39). Five genera of microsporidia have been associated with human infections: Enterocytozoon, Septata, Pleistophora, Encephalitozoon, and Vittaforma. E. bieneusi, the only species that seems to be tissue specific, is associated only with human enteric infections. Microsporidium spores, which are highly resistant environmental forms, vary in size among the different species. They are ovoid or piriform and 1 to 2 mm in diameter. Encephalitozoon intestinalis spores are bigger than those of E. bieneusi. These spores contain a polar filament that is used to eject the sporoplasm and penetrate the host cell cytoplasm. After entering the cell, the parasite multiplies asexually and eventually forms spores, lysing the host cell and invading neighboring cells. Although the mechanisms of transmission are not clear, it is believed that the infection can be acquired by ingesting spores in contaminated water and produce (119).

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

724 Infection with microsporidia is characterized by watery and large-volume stools. Microsporidia are occasionally associated with biliary tract disease and may potentially cause cholangiopathy, as observed in AIDS patients (23). Microsporidium spores can be histologically identified in tissue, but identification in fecal samples requires special staining processes. Microsporidia can be identified in tissues by using various conventional stains such as hematoxylin-eosin, Gram, and Giemsa. Detection in fecal samples can be achieved by using nonspecific staining techniques such as the use of calcofluor white and a modified trichrome with Chromotrope 2R; however, small structures and yeast spores are also stained by these procedures (24, 114, 157). Identification of microsporidia in the environment presents an additional challenge. Most species of the animal kingdom can be parasitized by species of microsporidia, and noninfectious species can be confused with those infectious to humans. Definitive identification is limited to transmission electron microscopy (19, 28, 29, 34, 47). New tests using specific monoclonal antibodies and molecular tools such as PCR are being developed to aid in the detection of microsporidia in tissues and in fecal and environmental samples (68, 102, 126, 164). Infections caused by Septata can be treated with albendazole or with metronidazole and atovaquone. To date, there is no effective treatment for infections caused by E. bieneusi (44) However, nitazoxanide was reported to resolve the infection in an AIDS patient (26).

GIARDIA Giardia is a protozoan flagellate that belongs to the phylum Sarcomastigophora. Initially thought to be a commensal organism in humans, Giardia is now clearly recognized as a common cause of diarrhea and malabsorption. Giardia infects millions of people throughout the world in both epidemic and sporadic forms. Most human infections result from ingestion of contaminated water or food or direct fecal-oral transmission such as would occur in person-to-person contact in child care centers and in male homosexual activity (1). Three species of Giardia have been described based on differences discernible in cysts and trophozoites by light microscopy: G. agilis from amphibians; G. muris from rodents, birds, and reptiles; and G. lamblia (also called G. intestinalis or G. duodenalis) from various mammals, including humans. Two additional species that are indistinguishable from G. lamblia by light microscopy, G. ardeae from herons and G. psittaci from psittacine birds, have been identified based on ultrastructural morphologic differences (62).

SMP_Food Microbiology_CH28.indd

724

G. lamblia does not appear to be host restricted, and wild animals such as beavers and muskrats have been implicated in waterborne outbreaks of giardiasis in humans. More recently, molecular classification using smallsubunit rRNA has placed Giardia as one of the most primitive eukaryotic organisms (171). Two major molecular groups or assemblages (A and B) of G. lamblia have been recognized as infecting humans. Other assemblages infect animals exclusively: dogs (C and D); cats (F); rats (G); and cattle, goats, sheep, and pigs (E) (123). Giardia can be observed in two forms: the trophozoite and the cyst. The cyst is the infectious form and is relatively inert and environmentally resistant. After cysts are ingested, excystation occurs in the duodenum after exposure to the acidic gastric pH and the pancreatic enzymes chemotrypsin and trypsin. Each cyst releases two vegetative trophozoites. The trophozoites replicate in the crypts of the duodenum and upper jejunum and reproduce asexually by binary fission. Some of the trophozoites then encyst in the ileum, possibly as a result of exposure to bile salts or cholesterol starvation (1). The trophozoites and cysts are excreted in the feces. Cysts are round or oval. Each cyst measures 4 to 11 by 7 to 10 mm, has four nuclei, and contains axonemes and median bodies. Trophozoites have the shape of a teardrop (viewed dorsally or ventrally) and measure 10 to 20 mm in length by 5 to 15 mm in width. The trophozoite has a concave sucking disk with four pairs of flagella, two axonemes, two median bodies, and two nuclei. The ventral disks act as suction cups, allowing mechanical attachment to the surface of the intestine (Fig. 28.8) (107). Infections may result from the ingestion of 10 or fewer Giardia cysts (154). Boiling is very effective to inactivate Giardia cysts, but cysts can survive after freezing for a few days. Ozone is an excellent disinfectant and can even be used to inactivate microorganisms such as protozoa, which are very resistant to conventional disinfectants. The apparent activation energy for the inactivation of protozoa is 80 kJ/mol. By-products of ozonation include bromate, iodate, and chlorate, which may be of concern depending on the chemical composition of the water to be disinfected. A 2-log reduction in active G. lamblia and G. muris cysts with ozone required 2.4 times longer than the contact time recommended by the Surface Water Treatment Rule. A surrogate organism used as an indicator for protozoan parasite inactivation is Bacillus subtilis because it has reported values similar to those for Cryptosporidium. Chlorine dioxide at the dose of 2.5 mg/ml reduces the number of spores by ca. 2.0 log and 0.5 log at water

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

725

Figure 28.8  Scanning electron micrograph of Giardia lam­ blia trophozoites. One trophozoite shows the dorsal surface and the other shows the ventral surface with sucking disk and flagella. doi:10.1128/9781555818463.ch28f8

temperatures of 23.2 and 5.2°C, respectively. This chlorine concentration is below the maximum treatment concentration set by the U.S. Environmental Protection Agency (69, 145, 149, 173). G. lamblia is prevalent worldwide and is especially common in areas where poor sanitary conditions and insufficient water treatment facilities prevail. Seasonality of giardiasis has been reported during late summer in the United Kingdom, the United States, and Mexico. The majority of Giardia infections are asymptomatic, but they can present as chronic diarrhea. Travelers to areas of endemicity are at high risk for developing symptomatic giardiasis. A study in Leningrad (now St. Petersburg), Russia, reported that 95% of travelers developed symptomatic giardiasis. Hikers and campers are also at increased risk, since Giardia cysts, often of animal origin, can be found in freshwater lakes and streams. The prevalence of Giardia infection can be as high as 35% in children attending child care centers. Although these children are frequently asymptomatic, they may infect other family members, who may develop symptomatic giardiasis (8, 70, 111). In waterborne outbreaks of diarrhea in which the etiologic agent was identified, Giardia has been the most common agent. Waterborne transmission is usually a result of inadequate water treatment or sewage contamination of drinking, well, or surface water. Giardiasis has also been associated with exposure to contaminated recreational water such as that in swimming pools. Giardia cysts are susceptible to inactivation by ozone and halogens; however, the concentration of chlorine used for drinking water may not

SMP_Food Microbiology_CH28.indd

725

cause inactivation. Inactivation by chlorine requires prolonged contact time, and filtration is the recommended means for purifying water (4, 104, 136). It is estimated that 76,840 Giardia episodes associated with foodborne illness occur in the U.S. (156). Symptomatic patients present with loose, foul-smelling stools and increased levels of fat and mucus in fecal samples. Flatulence, abdominal cramps, bloating, and nausea are common, as are anorexia, malaise, and weight loss. Blood is not present in stools. Fever is occasionally present at the beginning of the infection. In contrast with most other forms of acute infectious diarrhea, G. lamblia infection results in prolonged symptoms. Although giardiasis may resolve spontaneously, the illness frequently lasts for several weeks and sometimes for months if left untreated. Those with chronic giardiasis have profound malaise and diffuse epigastric and abdominal discomfort. Although diarrhea may persist, it may be replaced by constipation or even by normal bowel habits (1). Malabsorption associated with giardiasis may be responsible for substantial weight loss. Even in asymptomatic infections, malabsorption of fats, carbohydrates, and vitamins may occur. Reduced intestinal disaccharidase activity may persist even after Giardia is eradicated. Lactase deficiency is the most common residual deficiency and occurs in 20 to 40% of cases (163). Villous blunting, lymphocytic infiltration, and malabsorption are observed in biopsy samples in symptomatic cases. No tissue invasion is observed, and high numbers of trophozoites are sometimes present in the crypts without obvious pathology. To date, the presence of a toxin has not been demonstrated and no other potential mechanisms by which Giardia causes diarrhea have been identified. Giardia infection can be diagnosed by finding cysts or, less commonly, trophozoites, in fecal specimens (Fig. 28.9A). Giardia in feces can be detected by enzyme immunoassays, indirect and direct immunofluorescence assays using monoclonal antibodies, or PCR. All of these procedures are highly sensitive and specific for environmental and stool samples. In some patients with chronic diarrhea and malabsorption, results of stool examinations are repeatedly negative despite ongoing suspicion of giardiasis (6, 88, 177). Giardia cysts were found in 41.5% of nondepurated mussels from the Galician coast that were destined for human consumption (86). Effective treatment for patients with symptomatic giardiasis is mainly a single treatment course with metro­ nidazole. In refractory cases, multiple or combination courses have occasionally been required. Tinidazole is widely used throughout the world, and a single dose is effective for the treatment of giardiasis (94, 167).

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

726

Figure 28.9  Hematoxylin and eosin staining. (A) Giardia lamblia cyst: (a) with two nuclei and trophozoite; (b) showing two nuclei and median body visible at one pole. (B) Balantidium coli cyst: (a) showing large macronucleus and cilia beneath cyst wall and trophozoite; (b) showing oval macronucleus. (C) Entamoeba histolytica trophozoite: (a) showing a nucleus and few red blood cells in the cytoplasm and cyst; (b) showing two of four nuclei and rodshaped inclusion bodies with rounded ends. (Courtesy of Lynne S. Garcia.) doi:10.1128/9781555818463.ch28f9

BALANTIDIUM Balantidium coli is a ciliate parasite that, although found worldwide, is not highly prevalent. It is a commensal parasite of pigs. The trophozoites reside in the large intestine and multiply by binary fission. In humans, B.

SMP_Food Microbiology_CH28.indd

726

coli can cause ulcerative colitis and diarrhea. The ulcers differ from those caused by Entamoeba histolytica in that the epithelial surface is damaged but with lesions more superficial than those caused by amoebae. The parasite encysts and is excreted in the feces. The cyst,

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

727

which is the environmentally resistant form, is large and oblong, 45 to 65 μm in diameter (Fig. 28.9B). Both cysts and trophozoites contain two nuclei. Trophozoites move via their cilia and rotate on their longitudinal axis (75). Treatment of infection is preferentially done with tetracycline and, alternatively, with iodoquinol and metronidazole.

AMOEBIASIS Various amoebae can infect humans as commensals; however, E. histolytica is pathogenic to humans. Entamoeba dispar, which is morphologically similar to E. histolytica, is not pathogenic. Cysts and trophozoites of E. histolytica are excreted in the feces of infected individuals (Fig. 28.9C), with cysts being environmentally resistant. Once cysts are ingested, the trophozoite excysts, colonizes the large intestine, and multiplies by binary fission followed by encystation. Most patients are asymptomatic even when shedding cysts in their feces. In other instances, parasites can invade the mucosa and cause an ulceration that goes from the luminal surface of the intestine through the lamina propria and to the muscularis mucosa. The parasite then spreads laterally, forming a flask-shaped ulcer. The trophozoites feed on cell debris and red blood cells (76). Infection can progress, producing ulcerative amoebic colitis and causing perforation of the intestinal wall. Patients complain of diarrhea with stools containing blood and mucus, back pain, tenesmus, dehydration, and abdominal tenderness. Fulminant colitis is characterized by severe bloody diarrhea, fever, and abdominal tenderness due to transmural necrosis of the bowel. Another presentation of amoebiasis is formation of amebomas, which resemble carcinomas and are not necessarily associated with pain. Amoebae can also disseminate to the liver. Complications with amoebiasis are observed when the liver parenchyma is gradually replaced with necrotic debris, inflammatory cells, and trophozoites. Patients present with hepatomegaly, weight loss, and anemia (76, 158). Asymptomatic amoebiasis can be treated with iodoquinol, paromomycin, or diloxanide. If mild to severe infection and hepatic abscesses are present, amoebiasis can be treated with metronidazole or tinidazole followed with iodoquinol to treat asymptomatic amoebiasis.

SUMMARY Inactivation of the protozoan parasites has been a challenging task. Enrichment media are not available for cultivating parasites; therefore, isolation and identifica-

SMP_Food Microbiology_CH28.indd

727

tion methodologies need to be extremely sensitive and specific to detect small numbers of various forms of parasites that may be present in foods. Molecular tools such as PCR, restriction fragment length polymorphism, and variations of these techniques are being developed to improve the sensitivity and specificity of detection and identification processes. Treatment of drinking water by chlorination at permissible concentrations is relatively ineffective at inactivating spores, cysts, and oocysts of parasites and cannot be recommended as a sole method for water treatment. Boiling can inactivate cysts, oocysts, and spores. Irradiation of produce to inactivate pathogens that may be present has also been examined. This methodology effectively inactivates Toxoplasma cyst forms and oocysts. There has been an increase in consumers’ acceptance of purchasing irradiated produce (59), thereby providing additional incentive to use irradiation as a treatment to reduce or eliminate at least some genera of parasites in food and water. Finally, we face new challenges to ensure the safety of produce and other foods. Foodborne disease outbreaks caused by parasites are being documented with increased frequency. The consumption of fresh fruits and vegetables in developed countries has increased, and local production cannot meet this demand. Thus, importation of produce and other foods from countries where sanitary standards are minimal or lacking makes it even more difficult to control foodborne diseases (129).

References   1. Adam, R. D. 2001. Biology of Giardia lamblia. Clin. Microbiol. Rev. 14:447–475.   2. Adam, R. D., Y. R. Ortega, R. H. Gilman, and C. R. Sterling. 2000. Intervening transcribed spacer region 1 variability in Cyclospora cayetanensis. J. Clin. Microbiol. 38:2339–2343.   3. Ades, A. E. 1991. Evaluating the sensitivity and predictive value of tests of recent infection: toxoplasmosis in pregnancy. Epidemiol. Infect. 107:527–535.   4. Adin, A. 2004. Particle count and size alteration for membrane fouling reduction in non-conventional water filtration. Water Sci. Technol. 50:273–278.   5. Alarcón de Noya, B., Z. Díaz-Bello, C. Colmenares, R. Ruiz-Guevara, L. Mauriello, R. Zavala-Jaspe, J. A. Suarez, T. Abate, L. Naranjo, M. Paiva, L. Rivas, J. Castro, J. Márques, I. Mendoza, H. Acquatella, J. Torres, and O. Noya. 2010. Large urban outbreak of orally acquired acute Chagas disease at a school in Caracas, Venezuela. J. Infect. Dis. 201:1308–1315.   6. Aldeen, W. E., D. Hale, A. J. Robison, and K. Carroll. 1995. Evaluation of a commercially available ELISA assay for detection of Giardia lamblia in fecal specimens. Diagn. Microbiol. Infect. Dis. 21:77–79.

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

728   6a. Alfano-Sobsey, E. M., M. L. Eberhard, J. R. Seed, D. J. Weber, K. Y. Won, E. K. Nace, and C. L. Moe. 2004. Human challenge pilot study with Cyclospora cayeta­ nensis. Emerg. Infect. Dis. 10:726–728.   7. Amos, C. S. 1977. The ocular manifestations of congenital infections produced by Toxoplasma and cytomegalovirus. J. Am. Optom. Assoc. 48:532–538.   8. Ang, L. H. 2000. Outbreak of giardiasis in a daycare nursery. Commun. Dis. Public Health 3:212–213.   9. Anonymous. 1996. Foodborne outbreak of diarrheal illness associated with Cryptosporidium parvum—Minnesota, 1995. MMWR Morb. Mortal. Wkly. Rep. 45:783–784.   10. Anonymous. 1997. Outbreak of cyclosporiasis—northern Virginia-Washington, D.C.-Baltimore, Maryland, Metropolitan area, 1997. Can. Commun. Dis. Rep. 23:166–168.   11. Anonymous. 1997. Outbreaks of Escherichia coli O157:H7 infection and cryptosporidiosis associated with drinking unpasteurized apple cider—Connecticut and New York, October 1996. MMWR Morb. Mortal. Wkly. Rep. 46:4–8.   12. Anonymous. 1997. Outbreaks of cyclosporiasis— United States, 1997. MMWR Morb. Mortal. Wkly. Rep. 46:451–452.   13. Anonymous. 2000. Outbreaks of diarrhea associated with cyanobacteria (blue-green algae)-like-bodies— Chicago and Nepal, 1989 and 1990. MMWR Morb. Mortal. Wkly. Rep. 40:325–327.   14. Anonymous. 2004. Outbreak of cyclosporiasis associated with snow peas—Pennsylvania, 2004. MMWR Morb. Mortal. Wkly. Rep. 53:876–878.   15. Anusz, K. Z., P. H. Mason, M. W. Riggs, and L. E. Perryman. 1990. Detection of Cryptosporidium parvum oocysts in bovine feces by monoclonal antibody capture enzyme-linked immunosorbent assay. J. Clin. Microbiol. 28:2770–2774.   16. Arrowood, M. J., and C. R. Sterling. 1989. Comparison of conventional staining methods and monoclonal antibody-based methods for Cryptosporidium oocyst detection. J. Clin. Microbiol. 27:1490–1495.   17. Ashford, R. W. 1979. Occurrence of an undescribed coccidian in man in Papua New Guinea. Ann. Trop. Med. Parasitol. 73:497–500.   18. Atwill, E. R., K. W. Tate, M. D. Pereira, J. Bartolome, and G. Nader. 2006. Efficacy of natural grassland buffers for removal of Cryptosporidium parvum in rangeland runoff. J. Food Prot. 69:177–184.   19. Awadalla, H. N., I. F. el Naga, M. M. el-Temsahi, and A. Y. Negm. 1998. Detection of Microsporidia by different staining techniques. J. Egypt. Soc. Parasitol. 28:729–738.   20. Balatbat, A. B., G. W. Jordan, Y. J. Tang, and J. Silva, Jr. 1996. Detection of Cryptosporidium parvum DNA in human feces by nested PCR. J. Clin. Microbiol. 34:1769–1772.   21. Ballal, M., T. Prabhu, A. Chandran, and P. G. Shivananda. 1999. Cryptosporidium and Isospora belli diarrhoea in immunocompromised hosts. Indian J. Cancer 36:38–42.

SMP_Food Microbiology_CH28.indd

728

  22. Barcan, L. A., M. L. Dallurzo, L. O. Clara, A. Valledor, S. Macias, E. Zorkin, S. Gerona, and B. Livellara. 2002. Toxoplasma gondii pneumonia in liver transplantation: survival after a severe case of reactivation. Transpl. Infect. Dis. 4:93–96.   23. Beaugerie, L., M. F. Teilhac, A. M. Deluol, J. Fritsch, P. M. Girard, W. Rozenbaum, Y. Le Quintrec, and F. P. Chatelet. 1992. Cholangiopathy associated with Microsporidia infection of the common bile duct mucosa in a patient with HIV infection. Ann. Intern. Med. 117:401–402.   24. Berlin, O. G. W., L. R. Ash, C. N. Conteas, and J. B. Peter. 1999. Rapid, hot chromotrope stain for detecting microsporidia. Clin. Infect. Dis. 29:209.   25. Bern, C., B. Hernandez, M. B. Lopez, M. J. Arrowood, M. A. de Mejia, A. M. de Merida, A. W. Hightower, L. Venczel, B. L. Herwaldt, and R. E. Klein. 1999. Epidemiologic studies of Cyclospora cayetanensis in Guatemala. Emerg. Infect. Dis. 5:766–774.   26. Bicart-See, A., P. Massip, M. D. Linas, and A. Datry. 2000. Successful treatment with nitazoxanide of Enterocytozoon bieneusi microsporidiosis in a patient with AIDS. Antimicrob. Agents Chemother. 44:167–168.   27. Bissuel, F., L. Cotte, M. Rabodonirina, P. Rougier, M. A. Piens, and C. Trepo. 1994. Paromomycin: an effective treatment for cryptosporidial diarrhea in patients with AIDS. Clin. Infect. Dis. 18:447–449.   28. Blanshard, C., W. S. Hollister, C. S. Peacock, D. G. Tovey, D. S. Ellis, E. U. Canning, and B. G. Gazzard. 1992. Simultaneous infection with two types of intestinal microsporidia in a patient with AIDS. Gut 33:418–420.   29. Boldorini, R., A. Tosoni, G. Mazzucco, M. Cernuschi, P. Caramello, E. Maran, G. Costanzi, and G. Monga. 1996. Intracellular protozoan infection in small intestinal biopsies of patients with AIDS. Light and electron microscopic evaluation. Pathol. Res. Pract. 192:249–259.   30. Brady-McCreery, K. M., M. A. Hussein, and E. A. Paysse. 2003. Congenital toxoplasmosis with unusual retinal findings. Arch. Ophthalmol. 121:1200–1201.   31. Brinkman, K., S. Debast, R. Sauerwein, F. Ooyman, J. Hiel, and J. Raemaekers. 1998. Toxoplasma retinitis/encephalitis 9 months after allogeneic bone marrow transplantation. Bone Marrow Transplant. 21:635–636.   32. Cabada, M. M., and A. C. White, Jr. 2010. Treatment of cryptosporidiosis: do we know what we think we know? Curr. Opin. Infect. Dis. 23:494–499.   33. Caccio, S., E. Pinter, R. Fantini, I. Mezzaroma, and E. Pozio. 2002. Human infection with Cryptosporidium felis: case report and literature review. Emerg. Infect. Dis. 8:85–86.   34. Cali, A., L. Weiss, P. Takvorian, H. Tanowitz, and M. Wittner. 1994. Ultrastructural identification of AIDS associated microsporidiosis. J. Eukaryot. Microbiol. 41:24S.   35. Cama, V. A., and C. R. Sterling. 1991. Hyperimmune hens as a novel source of anti-Cryptosporidium antibodies suitable for passive immune transfer. J. Protozool. 38:42S–43S.

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

729

  36. Carreno, R. A., D. S. Martin, and J. R. Barta. 1999. Cryptosporidium is more closely related to the gregarines than to coccidia as shown by phylogenetic analysis of apicomplexan parasites inferred using small-subunit ribosomal RNA gene sequences. Parasitol. Res. 85:899–904.   37. Casemore, D. P., M. Armstrong, and R. L. Sands. 1985. Laboratory diagnosis of cryptosporidiosis. J. Clin. Pathol. 38:1337–1341.   38. Chauret, C. P., C. Z. Radziminski, M. Lepuil, R. Creason, and R. C. Andrews. 2001. Chlorine dioxide inactivation of Cryptosporidium parvum oocysts and bacterial spore indicators. Appl. Environ. Microbiol. 67:2993–3001.   39. Chu, D. M., J. B. Sherchand, J. H. Cross, and P. A. Orlandi. 2004. Detection of Cyclospora cayetanensis in animal fecal isolates from Nepal using an FTA filter-base polymerase chain reaction method. Am. J. Trop. Med. Hyg. 71:373–379.   40. Chui, D. W., and R. L. Owen. 1994. AIDS and the gut. J. Gastroenterol. Hepatol. 9:291–303.   41. Comin, C. E., and M. Santucci. 1994. Submicroscopic profile of Isospora belli enteritis in a patient with acquired immune deficiency syndrome. Ultrastruct. Pathol. 18:473–482.   42. Connor, B. A., J. Reidy, and R. Soave. 1999. Cyclosporiasis: clinical and histopathologic correlates. Clin. Infect. Dis. 28:1216–1222.   43. Conrath, J., A. Mouly-Bandini, F. Collart, and B. Ridings. 2003. Toxoplasma gondii retinochoroiditis after cardiac transplantation. Graefes Arch. Clin. Exp. Ophthalmol. 241:334–338.   44. Costa, S. F., and L. M. Weiss. 2000. Drug treatment of microsporidiosis. Drug Resist. Updat. 3:384–399.   45. Coupe, S., C. Sarfati, S. Hamane, and F. Derouin. 2005. Detection of Cryptosporidium and identification to the species level by nested PCR and restriction fragment length polymorphism. J. Clin. Microbiol. 43:1017–1023.   46. Current, W. L., and P. H. Bick. 1989. Immunobiology of Cryptosporidium spp. Pathol. Immunopathol. Res. 8:141–160.   47. Curry, A. 2003. Electron microscopy and the investigation of new infectious diseases. Int. J. Infect. Dis. 7:251–257.   48. Danziger, L. H., T. P. Kanyok, and R. M. Novak. 1993. Treatment of cryptosporidial diarrhea in an AIDS patient with paromomycin. Ann. Pharmacother. 27:1460–1462.   49. DeHovitz, J. A., J. W. Pape, M. Boncy, and W. D. Johnson, Jr. 1986. Clinical manifestations and therapy of Isospora belli infection in patients with the acquired immunodeficiency syndrome. N. Engl. J. Med. 315:87–90.   50. De Marco, R., R. Ceccarelli, R. Frulio, C. Palmero, and P. Vittone. 2003. Retinochoroiditis associated with congenital toxoplasmosis in children: IgG antibody profiles demonstrating the synthesis of local antibodies. Eur. J. Ophthalmol. 13:74–79.   51. Deng, M. Q., and D. O. Cliver. 2000. Comparative detection of Cryptosporidium parvum oocysts from apple juice. Int. J. Food Microbiol. 54:155–162.

SMP_Food Microbiology_CH28.indd

729

  52. Derouin, F. 1992. Pathogeny and immunological control of toxoplasmosis. Braz. J. Med. Biol. Res. 25:1163–1169.   53. Derouin, F., A. Debure, E. Godeaut, M. Lariviere, and H. Kreis. 1987. Toxoplasma antibody titers in renal transplant recipients. Pretransplant evaluation and posttransplant follow-up of 73 patients. Transplantation 44:515–518.   54. Diaz, J. H. 2007. Chagas disease in the United States: a cause for concern in Louisiana? J. La. State Med. Soc. 159:21–29.   55. Ditrich, O., L. Palkovic, J. Steˇrba, J. Prokopic, J. Loudová, and M. Giboda. 1991. The first finding of Cryptosporidium baileyi in man. Parasitol. Res. 77:44–47.   56. Doller, P. C., K. Dietrich, N. Filipp, S. Brockmann, C. Dreweck, R. Vonthein, C. Wagner-Wiening, and A. Wiedenmann. 2002. Cyclosporiasis outbreak in Germany associated with the consumption of salad. Emerg. Infect. Dis. 8:992–994.   57. Dubey, J. P., and C. P. Beattie. 1988. Toxoplasmosis of Animals and Man. CRC Press, Inc., Boca Raton, FL.   58. Dubey, J. P., N. L. Miller, and J. K. Frenkel. 1970. Toxoplasma gondii life cycle in cats. J. Am. Vet. Med. Assoc. 157:1767–1770.   59. Dubey, J. P., D. W. Thayer, C. A. Speer, and S. K. Shen. 1998. Effect of gamma irradiation on unsporulated and sporulated Toxoplasma gondii oocysts. Int. J. Parasitol. 28:369–375.   60. Eberhard, M. L., A. J. da Silva, B. G. Lilley, and N. J. Pieniazek. 1999. Morphologic and molecular characterization of new Cyclospora species from Ethiopian monkeys: C. cercopitheci sp.n., C. colobi sp.n., and C. papionis sp.n. Emerg. Infect. Dis. 5:651–658.   61. Eberhard, M. L., Y. R. Ortega, D. E. Hanes, E. K. Nace, R. Q. Do, M. G. Robl, K. Y. Won, C. Gavidia, N. L. Sass, K. Mansfield, A. Gozalo, J. Griffiths, R. Gilman, C. R. Sterling, and M. J. Arrowood. 2000. Attempts to establish experimental Cyclospora cay­ etanensis infection in laboratory animals. J. Parasitol. 86:577–582.   62. Erlandsen, S. L., and W. J. Bemrick. 1987. SEM evidence for a new species, Giardia psittaci. J. Parasitol. 73:623–629.   63. Fayer, R. 2004. Sarcocystis spp. in human infections. Clin. Microbiol. Rev. 17:894–902.   64. Fayer, R., C. A. Spear, and J. P. Dubey. 1990. The general biology of Cryptosporidium, p. 1–41. In R. Fayer (ed.), Cryptosporidium and Cryptosporidiosis. CRC Press, Boca Raton, FL.   65. Fayer, R., M. Tilley, S. J. Upton, A. J. Guidry, D. W. Thayer, M. Hildreth, and J. Thomson. 1991. Production and preparation of hyperimmune bovine colostrum for passive immunotherapy of cryptosporidiosis. J. Protozool. 38:38S–39S.   66. Fayer, R., J. M. Trout, E. J. Lewis, M. Santin, L. Zhou, A. A. Lal, and L. Xiao. 2003. Contamination of Atlantic coast commercial shellfish with Cryptosporidium. Parasitol. Res. 89:141–145.

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

730   67. Fayer, R., J. M. Trout, L. Xiao, U. M. Morgan, A. A. Lai, and J. P. Dubey. 2001. Cryptosporidium canis n. sp. from domestic dogs. J. Parasitol. 87:1415–1422.   68. Fedorko, D. P., N. A. Nelson, and C. P. Cartwright. 1995. Identification of microsporidia in stool specimens by using PCR and restriction endonucleases. J. Clin. Microbiol. 33:1739–1741.   69. Finch, G. R., E. K. Black, C. W. Labatiuk, L. Gyürék, and M. Belosevic. 1993. Comparison of Giardia lam­ blia and Giardia muris cyst inactivation by ozone. Appl. Environ. Microbiol. 59:3674–3680.   70. Flanagan, P. A. 1992. Giardia—diagnosis, clinical course and epidemiology. A review. Epidemiol. Infect. 109:1–22.   71. Fleming, C. A., D. Caron, J. E. Gunn, and M. A. Barry. 1998. A foodborne outbreak of Cyclospora cayetanensis at a wedding: clinical features and risk factors for illness. Arch. Intern. Med. 158:1121–1125.   72. Furio, M. M., and C. J. Wordell. 1985. Treatment of infectious complications of acquired immunodeficiency syndrome. Clin. Pharm. 4:539–554.   73. Gallant, J. E., R. D. Moore, and R. E. Chaisson. 1994. Prophylaxis for opportunistic infections in patients with HIV infection. Ann. Intern. Med. 120:932–944.   74. Galvagno, G., G. Cattaneo, and E. Reverso-Giovantin. 1993. Chronic diarrhea due to Cryptosporidium: the efficacy of spiramycin treatment. Pediatr. Med. Chir. 15:297–298. (In Italian.)   75. Garcia, L. S. 2007. Intestinal protozoa: flagellates and ciliates, p. 31–48. In Diagnostic Medical Parasitology, 5th ed. American Society for Microbiology, Washington, DC.   76. Garcia, L. S. 2007. Intestinal protozoa: amebae, p. 6–30. In Diagnostic Medical Parasitology, 5th ed. American Society for Microbiology, Washington, DC.   77. Garcia-López, H. L., L. E. Rodríguez-Tovar, and C. E. Medina-De la Garza. 1996. Identification of Cyclospora in poultry. Emerg. Infect. Dis. 2:356–357.   78. Gatei, W., R. W. Ashford, N. J. Beeching, S. K. Kamwati, J. Greensill, and C. A. Hart. 2002. Cryptosporidium muris infection in an HIV-infected adult, Kenya. Emerg. Infect. Dis. 8:204–206.   79. Giacometti, A., F. Burzacchini, O. Cirioni, F. Barchiesi, M. Dini, and G. Scalise. 1999. Efficacy of treatment with paromomycin, azithromycin, and nitazoxanide in a patient with disseminated cryptosporidiosis. Eur. J. Clin. Microbiol. Infect. Dis. 18:885–889.   80. Gomez-Bautista, M., L. M. Ortega-Mora, E. Tabares, V. Lopez-Rodas, and E. Costas. 2000. Detection of infectious Cryptosporidium parvum oocysts in mussels (Mytilus galloprovincialis) and cockles (Cerastoderma edule). Appl. Environ. Microbiol. 66:1866–1870.   81. Gómez-Couso, H., F. Freire-Santos, C. F. Amar, K. A. Grant, K. Williamson, M. E. Ares-Mazás, and J. McLauchlin. 2004. Detection of Cryptosporidium and Giardia in molluscan shellfish by multiplexed nestedPCR. Int. J. Food Microbiol. 91:279–288.   82. Gómez-Couso, H., F. Freire-Santos, G. A. HernándezCórdova, and M. E. Ares-Mazás. 2005. A histological

SMP_Food Microbiology_CH28.indd

730

  83.

  84.

  85.

  86.

  87.   88.

  89.

  90.   91.

  92.

  93.

  94.   95.

study of the transit of Cryptosporidium parvum oocysts through clams (Tapes decussatus). Int. J. Food Microbiol. 102:57–62. Gómez-Couso, H., F. Freire-Santos, J. Martínez-Urtaza, O. García-Martín, and M. E. Ares-Mazás. 2003. Contamination of bivalve molluscs by Cryptosporidium oocysts: the need for new quality control standards. Int. J. Food Microbiol. 87:97–105. Gómez-Couso, H., F. Freire-Santos, M. R. OrtegaIñarrea, J. A. Castro-Hermida, and M. E. Ares-Mazás. 2003. Environmental dispersal of Cryptosporidium par­ vum oocysts and cross transmission in cultured bivalve molluscs. Parasitol. Res. 90:140–142. Gómez-Couso, H., F. Méndez-Hermida, J. A. CastroHermida, and E. Ares-Mazás. 2006. Cryptosporidium contamination in harvesting areas of bivalve molluscs. J. Food Prot. 69:185–190. Gómez-Couso, H., F. Méndez-Hermida, J. A. CastroHermida, and E. Ares-Mazás. 2005. Giardia in shellfish-farming areas: detection in mussels, river water and waste waters. Vet. Parasitol. 133:13–18. Graczyk, T. K., and K. J. Schwab. 2000. Foodborne infections vectored by molluscan shellfish. Curr. Gastroenterol. Rep. 2:305–309. Guerreiro, N. M., P. M. Herrera, L. de Escalona, C. E. de Kolster, V. G. de Yanes, O. de Febres, O. Naveda, and M. de Naveda. 1991. Giardia lamblia: comparison of two diagnostic methods and evaluation of response to treatment with metronidazole. G. E. N. 45:105–110. (In Spanish.) Hayes, E. B., T. D. Matte, T. R. O’Brien, T. W. McKinley, G. S. Logsdon, J. B. Rose, B. L. Ungar, D. M. Word, M. A. Wilson, E. G. Long, E. S. Hurwitz, and D. D. Juranek. 1989. Large community outbreak of cryptosporidiosis due to contamination of a filtered public water supply. N. Engl. J. Med. 320:1372–1376. Herwaldt, B. L. 2000. Cyclospora cayetanensis: a review, focusing on the outbreaks of cyclosporiasis in the 1990s. Clin. Infect. Dis. 31:1040–1057. Herwaldt, B. L., M. L. Ackers, and the Cyclospora Working Group. 1997. An outbreak in 1996 of cyclosporiasis associated with imported raspberries. N. Engl. J. Med. 336:1548–1556. Herwaldt, B. L., M. J. Beach, and the Cyclospora Working Group. 1999. The return of Cyclospora in 1997: another outbreak of cyclosporiasis in North America associated with imported raspberries. Ann. Intern. Med. 130:210–220. Hijjawi, N. S., B. P. Meloni, U. M. Ryan, M. E. Olson, and R. C. Thompson. 2002. Successful in vitro cultivation of Cryptosporidium andersoni: evidence for the existence of novel extracellular stages in the life cycle and implications for the classification of Cryptosporidium. Int. J. Parasitol. 32:1719–1726. Hill, D. R. 1993. Giardiasis. Issues in diagnosis and management. Infect. Dis. Clin. North Am. 7:503–525. Ho, A. Y., A. S. Lopez, M. G. Eberhart, R. Levenson, B. S. Finkel, A. J. da Silva, J. M. Roberts, P. A. Orlandi, C. C. Johnson, and B. L. Herwaldt. 2002. Outbreak

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

  96.

  97.

  98.   99.

100.

101.

102.

103.

104.

105.

106. 107.

108.

731

of cyclosporiasis associated with imported raspberries, Philadelphia, Pennsylvania, 2000. Emerg. Infect. Dis. 8:783–788. Hoang, L. M., M. Fyfe, C. Ong, J. Harb, S. Champagne, B. Dixon, and J. Isaac-Renton. 2005. Outbreak of cyclosporiasis in British Columbia associated with imported Thai basil. Epidemiol. Infect. 133:23–27. Huang, P., J. T. Weber, D. M. Sosin, P. M. Griffin, E. G. Long, J. J. Murphy, F. Kocka, C. Peters, and C. Kallick. 1995. The first reported outbreak of diarrheal illness associated with Cyclospora in the United States. Ann. Intern. Med. 123:409–414. Hume, O. S. 1972. Toxoplasmosis and pregnancy. Am. J. Obstet. Gynecol. 114:703–715. Insulander, M., B. Svenungsson, M. Lebbad, L. Karlsson, and B. de Jong. 2010. A foodborne outbreak of Cyclospora infection in Stockholm, Sweden. Foodborne Pathog. Dis. 7:1585–1587. Jinneman, K. C., J. H. Wetherington, W. E. Hill, A. M. Adams, J. M. Johnson, B. J. Tenge, N. L. Dang, R. L. Manger, and M. M. Wekell. 1998. Template preparation for PCR and RFLP of amplification products for the detection and identification of Cyclospora sp. and Eimeria spp. oocysts directly from raspberries. J. Food Prot. 61:1497–1503. Jinneman, K. C., J. H. Wetherington, W. E. Hill, C. J. Omiescinski, A. M. Adams, J. M. Johnson, B. J. Tenge, N. L. Dang, and M. M. Wekell. 1999. An oligonucleotide-ligation assay for the differentiation between Cyclospora and Eimeria spp. polymerase chain reaction amplification products. J. Food Prot. 62:682–685. Karasudani, T., S. Aoki, J. Takeuchi, M. Okuyama, M. Oseto, S. Matsuura, T. Asai, and H. Inouye. 2001. Sensitive detection of Cryptosporidium oocysts in environmental water samples by reverse transcription-PCR. Jpn. J. Infect. Dis. 54:122–124. Kniel, K. E., and M. C. Jenkins. 2005. Detection of Cryptosporidium parvum oocysts on fresh vegetables and herbs using antibodies specific for a Cryptosporidium parvum viral antigen. J. Food Prot. 68:1093–1096. Kong, L. I., L. J. Swango, B. L. Blagburn, C. M. Hendrix, D. E. Williams, and S. D. Worley. 1988. Inactivation of Giardia lamblia and Giardia canis cysts by combined and free chlorine. Appl. Environ. Microbiol. 54:2580–2582. Korich, D. G., J. R. Mead, M. S. Madore, N. A. Sinclair, and C. R. Sterling. 1990. Effects of ozone, chlorine dioxide, chlorine, and monochloramine on Cryptosporidium parvum oocyst viability. Appl. Environ. Microbiol. 56:1423–1428. Laberge, I., M. W. Griffiths, and M. W. Griffiths. 1996. Prevalence, detection and control of Cryptosporidium parvum in food. Int. J. Food Microbiol. 32:1–26. Lanfredi-Rangel, A., J. A. Diniz, Jr., and W. de Souza. 1999. Presence of a protrusion on the ventral disk of adhered trophozoites of Giardia lamblia. Parasitol. Res. 85:951–955. Long, E. G., A. Ebrahimzadeh, E. H. White, B. Swisher, and C. S. Callaway. 1990. Alga associated with diarrhea

SMP_Food Microbiology_CH28.indd

731

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119. 120. 121.

in patients with acquired immunodeficiency syndrome and in travelers. J. Clin. Microbiol. 28:1101–1104. Long, E. G., E. H. White, W. W. Carmichael, P. M. Quinlisk, R. Raja, B. L. Swisher, H. Daugharty, and M. T. Cohen. 1991. Morphologic and staining characteristics of a cyanobacterium-like organism associated with diarrhea. J. Infect. Dis. 164:199–202. Lopez, A. S., D. R. Dodson, M. J. Arrowood, P. A. Orlandi, Jr., A. J. da Silva, J. W. Bier, S. D. Hanauer, R. L. Kuster, S. Oltman, M. S. Baldwin, K. Y. Won, E. M. Nace, M. L. Eberhard, and B. L. Herwaldt. 2001. Outbreak of cyclosporiasis associated with basil in Missouri in 1999. Clin. Infect. Dis. 32:1010–1017. Lopez, C. E., A. C. Dykes, D. D. Juranek, S. P. Sinclair, J. M. Conn, R. W. Christie, E. C. Lippy, M. G. Schultz, and M. H. Mires. 1980. Waterborne giardiasis: a communitywide outbreak of disease and a high rate of asymptomatic infection. Am. J. Epidemiol. 112:495–507. Lopez, F. A., J. Manglicmot, T. M. Schmidt, C. Yeh, H. V. Smith, and D. A. Relman. 1999. Molecular characterization of Cyclospora-like organisms from baboons. J. Infect. Dis. 179:670–676. López-Vélez, R., R. Tarazona, C. A. Garcia, E. GomezMampaso, A. Guerrero, V. Moreira, and R. Villanueva. 1995. Intestinal and extraintestinal cryptosporidiosis in AIDS patients. Eur. J. Clin. Microbiol. Infect. Dis. 14:677–681. Luna, V. A., B. K. Stewart, D. L. Bergeron, C. R. Clausen, J. J. Plorde, and T. R. Fritsche. 1995. Use of the fluorochrome calcofluor white in the screening of stool specimens for spores of microsporidia. Am. J. Clin. Pathol. 103:656–659. Lunden, A., and A. Uggla. 1992. Infectivity of Toxoplasma gondii in mutton following curing, smoking, freezing or microwave cooking. Int. J. Food Microbiol. 15:357–363. Mac Kenzie, W. R., N. J. Hoxie, M. E. Proctor, M. S. Gradus, K. A. Blair, D. E. Peterson, J. J. Kazmierczak, D. G. Addiss, K. R. Fox, J. B. Rose, and J. P. Davis. 1994. A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public water supply. N. Engl. J. Med. 331:161–167. MacRae, M., C. Hamilton, N. J. Strachan, S. Wright, and I. D. Ogden. 2005. The detection of Cryptosporidium parvum and Escherichia coli O157 in UK bivalve shellfish. J. Microbiol. Methods 60:395–401. Madico, G., J. McDonald, R. H. Gilman, L. Cabrera, and C. R. Sterling. 1997. Epidemiology and treatment of Cyclospora cayetanensis infection in Peruvian children. Clin. Infect. Dis. 24:977–981. Marshall, M. M., D. Naumovitz, Y. Ortega, and C. R. Sterling. 1997. Waterborne protozoan pathogens. Clin. Microbiol. Rev. 10:67–85. Matos, O., M. Alves, L. Xiao, V. Cama, and F. Antunes. 2004. Cryptosporidium felis and C. meleagridis in persons with HIV, Portugal. Emerg. Infect. Dis. 10:2256–2257. Miles, M. A. 2010. Orally acquired Chagas disease: lessons from an urban school outbreak. J. Infect. Dis. 201:1282–1284.

Manila Typesetting Company

11/08/2012 08:57PM

Nonbacterial Pathogens

732 122. Millard, P. S., K. F. Gensheimer, D. G. Addiss, D. M. Sosin, G. A. Beckett, A. Houck-Jankoski, and A. Hudson. 1994. An outbreak of cryptosporidiosis from fresh-pressed apple cider. JAMA 272:1592–1596. 123. Monis, P. T., R. H. Andrews, G. Mayrhofer, and P. L. Ey. 2003. Genetic diversity within the morphological species Giardia intestinalis and its relationship to host origin. Infect. Genet. Evol. 3:29–38. 124. Morgan-Ryan, U. M., A. Fall, L. A. Ward, N. Hijjawi, I. Sulaiman, R. Fayer, R. C. Thompson, M. Olson, A. Lal, and L. Xiao. 2002. Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporidiidae) from Homo sapiens. J. Eukaryot. Microbiol. 49:433–440. 125. Moskovitz, B. L., T. L. Stanton, and J. J. Kusmierek. 1988. Spiramycin therapy for cryptosporidial diarrhoea in immunocompromised patients. J. Antimicrob. Chemother. 22(Suppl. B):189–191. 126. Müller, A., K. Stellermann, P. Hartmann, M. Schrappe, G. Fätkenheuer, B. Salzberger, V. Diehl, and C. Franzen. 1999. A powerful DNA extraction method and PCR for detection of microsporidia in clinical stool specimens. Clin. Diagn. Lab. Immunol. 6:243–246. 127. Nóbrega, A. A., M. H. Garcia, E. Tatto, M. T. Obara, E. Costa, J. Sobel, and W. N. Araujo. 2009. Oral transmission of Chagas disease by consumption of açaí palm fruit, Brazil. Emerg. Infect. Dis. 15:653–655. 128. Olivier, C., S. van de Pas, P. W. Lepp, K. Yoder, and D. A. Relman. 2001. Sequence variability in the first internal transcribed spacer region within and among Cyclospora species is consistent with polyparasitism. Int. J. Parasitol. 31:1475–1487. 129. Orlandi, P. A., D. M. Chu, J. W. Bier, and G. J. Jackson. 2002. Parasites in the food supply. Food Technol. 56:72. 130. Orlandi, P. A., and K. A. Lampel. 2000. Extraction-free, filter-based template preparation for rapid and sensitive PCR detection of pathogenic parasitic protozoa. J. Clin. Microbiol. 38:2271–2277. 131. Ortega, Y. R., R. H. Gilman, and C. R. Sterling. 1994. A new coccidian parasite (Apicomplexa: Eimeriidae) from humans. J. Parasitol. 80:625–629. 132. Ortega, Y. R., R. Nagle, R. H. Gilman, J. Watanabe, J. Miyagui, H. Quispe, P. Kanagusuku, C. Roxas, and C. R. Sterling. 1997. Pathologic and clinical findings in patients with cyclosporiasis and a description of intracellular parasite life-cycle stages. J. Infect. Dis. 176:1584–1589. 133. Ortega, Y. R., C. R. Roxas, R. H. Gilman, N. J. Miller, L. Cabrera, C. Taquiri, and C. R. Sterling. 1997. Isolation of Cryptosporidium parvum and Cyclospora cayetanen­ sis from vegetables collected in markets of an endemic region in Peru. Am. J. Trop. Med. Hyg. 57:683–686. 134. Ortega, Y. R., C. R. Sterling, R. H. Gilman, V. A. Cama, and F. Diaz. 1993. Cyclospora species—a new protozoan pathogen of humans. N. Engl. J. Med. 328:1308–1312. 135. Pape, J. W., R. I. Verdier, M. Boncy, J. Boncy, and W. D. Johnson, Jr. 1994. Cyclospora infection in adults infected with HIV. Clinical manifestations, treatment, and prophylaxis. Ann. Intern. Med. 121:654–657.

SMP_Food Microbiology_CH28.indd

732

136. Payment, P. 1999. Poor efficacy of residual chlorine disinfectant in drinking water to inactivate waterborne pathogens in distribution systems. Can. J. Microbiol. 45:709–715. 137. Pedraza-Díaz, S., C. F. Amar, J. McLauchlin, G. L. Nichols, K. M. Cotton, P. Godwin, A. M. Iversen, L. Milne, J. R. Mulla, K. Nye, H. Panigrahl, S. R. Venn, R. Wiggins, M. Williams, and E. R. Youngs. 2001. Cryptosporidium me­ leagridis from humans: molecular analysis and description of affected patients. J. Infect. 42:243–250. 138. Peng, M. M., L. Xiao, A. R. Freeman, M. J. Arrowood, A. A. Escalante, A. C. Weltman, C. S. Ong, W. R. Mac Kenzie, A. A. Lal, and C. B. Beard. 1997. Genetic polymorphism among Cryptosporidium parvum isolates: evidence of two distinct human transmission cycles. Emerg. Infect. Dis. 3:567–573. 139. Perryman, L. E., and J. M. Bjorneby. 1991. Immunotherapy of cryptosporidiosis in immunodeficient animal models. J. Protozool. 38:98S–100S. 140. Pönka, A., H. Kotilainen, R. Rimhanen-Finne, P. Hokkanen, M. L. Hänninen, A. Kaarna, T. Meri, and M. Kuusi. 2009. A foodborne outbreak due to Cryptosporidium parvum in Helsinki, November 2008. Euro Surveill. 14:pii=19269. 141. Portnoy, D., M. E. Whiteside, E. Buckley III, and C. L. MacLeod. 1984. Treatment of intestinal cryptosporidiosis with spiramycin. Ann. Intern. Med. 101:202–204. 142. Posada, G., D. Pizarro, and E. Mohs. 1987. Oral rehydration in children with Cryptosporidium muris diarrhea. Bol. Med. Hosp. Infant. Mex. 44:740–744. (In Spanish.) 143. Quiroz, E. S., C. Bern, J. R. MacArthur, L. Xiao, M. Fletcher, M. J. Arrowood, D. K. Shay, M. E. Levy, R. I. Glass, and A. Lal. 2000. An outbreak of cryptosporidiosis linked to a foodhandler. J. Infect. Dis. 181:695–700. 144. Rabold, J. G., C. W. Hoge, D. R. Shlim, C. Kefford, R. Rajah, and P. Echeverria. 1994. Cyclospora outbreak associated with chlorinated drinking water. Lancet 344:1360–1361. 145. Radziminski, C., L. Ballantyne, J. Hodson, R. Creason, R. C. Andrews, and C. Chauret. 2002. Disinfection of Bacillus subtilis spores with chlorine dioxide: a benchscale and pilot-scale study. Water Res. 36:1629–1639. 146. Raisanen, S. 1978. Toxoplasmosis transmitted by blood transfusions. Transfusion 18:329–332. 147. Reduker, D. W., C. A. Speer, and J. A. Blixt. 1985. Ultrastructure of Cryptosporidium parvum oocysts and excysting sporozoites as revealed by high resolution scanning electron microscopy. J. Protozool. 32:708–711. 148. Reiter-Owona, I., H. Seitz, U. Gross, M. Sahm, J. K. Rockstroh, and H. M. Seitz. 2000. Is stage conversion the initiating event for reactivation of Toxoplasma gondii in brain tissue of AIDS patients? J. Parasitol. 86:531–536. 149. Rice, E. W., and J. C. Hoff. 1981. Inactivation of Giardia lamblia cysts by ultraviolet irradiation. Appl. Environ. Microbiol. 42:546–547. 150. Ripabelli, G., A. Leone, M. L. Sammarco, I. Fanelli, G. M. Grasso, and J. McLauchlin. 2004. Detection of

Manila Typesetting Company

11/08/2012 08:57PM

28.  Protozoan Parasites

151. 152.

153. 154. 155.

156.

157.

158. 159.

160.

161.

162.

163.

164.

733

Cryptosporidium parvum oocysts in experimentally contaminated lettuce using filtration, immunomagnetic separation, light microscopy, and PCR. Foodborne Pathog. Dis. 1:216–222. Robertson, L. J., and B. Gjerde. 2001. Occurrence of parasites on fruits and vegetables in Norway. J. Food Prot. 64:1793–1798. Robertson, L. J., J. D. Greig, B. Gjerde, and A. Fazil. 2005. The potential for acquiring cryptosporidiosis or giardiosis from consumption of mung bean sprouts in Norway: a preliminary step-wise risk assessment. Int. J. Food Microbiol. 98:291–300. Rosales, M. J., G. P. Cordón, M. S. Moreno, and C. M. Sánchez. 2005. Extracellular like-gregarine stages of Cryptosporidium parvum. Acta Trop. 95:74–78. Rose, J. B., C. N. Haas, and S. Regli. 1991. Risk assessment and control of waterborne giardiasis. Am. J. Public Health 81:709–713. Sáez-Llorens, X., C. M. Odio, M. A. Umaña, and M. V. Morales. 1989. Spiramycin vs. placebo for treatment of acute diarrhea caused by Cryptosporidium. Pediatr. Infect. Dis. J. 8:136–140. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M. A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg. Infect. Dis. 17:7–15. Schottelius, J., E. M. Kuhn, and R. Enriquez. 2000. Microsporidia and Candida spores: their discrimination by Calcofluor, trichrome-blue and methylene-blue combination staining. Trop. Med. Int. Health 5:453–458. Sepulveda, B. 1982. Amebiasis: host-pathogen biology. Rev Infect. Dis. 4:1247–1253. Sherchand, J. B., J. H. Cross, M. Jimba, S. Sherchand, and M. P. Shrestha. 1999. Study of Cyclospora cayeta­ nensis in health care facilities, sewage water and green leafy vegetables in Nepal. Southeast Asian J. Trop. Med. Public Health 30:58–63. Shields, J. M., and B. H. Olson. 2003. PCR-restriction fragment length polymorphism method for detection of Cyclospora cayetanensis in environmental waters without microscopic confirmation. Appl. Environ. Microbiol. 69:4662–4669. Shlim, D. R., M. T. Cohen, M. Eaton, R. Rajah, E. G. Long, and B. L. Ungar. 1991. An alga-like organism associated with an outbreak of prolonged diarrhea among foreigners in Nepal. Am. J. Trop. Med. Hyg. 45:383–389. Shlim, D. R., C. W. Hoge, R. Rajah, R. M. Scott, P. Pandy, and P. Echeverria. 1999. Persistent high risk of diarrhea among foreigners in Nepal during the first 2 years of residence. Clin. Infect. Dis. 29:613–616. Singh, K. D., D. K. Bhasin, S. V. Rana, K. Vaiphei, R. Katyal, V. K. Vinayak, and K. Singh. 2000. Effect of Giardia lamblia on duodenal disaccharidase levels in humans. Trop. Gastroenterol. 21:174–176. Sironi, M., C. Bandi, S. Novati, and M. Scaglia. 1997. A PCR-RFLP method for the detection and species

SMP_Food Microbiology_CH28.indd

733

identification of human microsporidia. Parassitologia 39:437–439. 165. Smith, H. V., C. A. Paton, R. W. Girdwood, and M. M. Mtambo. 1996. Cyclospora in non-human primates in Gombe, Tanzania. Vet. Rec. 138:528. 166. Soave, R. 1988. Cryptosporidiosis and isosporiasis in patients with AIDS. Infect. Dis. Clin. North Am. 2:485–493. 167. Speelman, P. 1985. Single-dose tinidazole for the treatment of giardiasis. Antimicrob. Agents Chemother. 27:227–229. 168. Sun, T., C. F. Ilardi, D. Asnis, A. R. Bresciani, S. Goldenberg, B. Roberts, and S. Teichberg. 1996. Light and electron microscopic identification of Cyclospora species in the small intestine. Evidence of the presence of asexual life cycle in human host. Am. J. Clin. Pathol. 105:216–220. 169. Tee, G. H., A. H. Moody, A. H. Cooke, and P. L. Chiodini. 1993. Comparison of techniques for detecting antigens of Giardia lamblia and Cryptosporidium par­ vum in faeces. J. Clin. Pathol. 46:555–558. 170. Ungar, B. L. 1990. Enzyme-linked immunoassay for detection of Cryptosporidium antigens in fecal specimens. J. Clin. Microbiol. 28:2491–2495. 171. van Keulen, H., S. R. Campbell, S. L. Erlandsen, and E. L. Jarroll. 1991. Cloning and restriction enzyme mapping of ribosomal DNA of Giardia duodenalis, Giardia ardeae and Giardia muris. Mol. Biochem. Parasitol. 46:275–284. 172. Verdier, R. I., D. W. Fitzgerald, W. D. Johnson, Jr., and J. W. Pape. 2000. Trimethoprim-sulfamethoxazole compared with ciprofloxacin for treatment and prophylaxis of Isospora belli and Cyclospora cayetanensis infection in HIV-infected patients. A randomized, controlled trial. Ann. Intern. Med. 132:885–888. 173. von Gunten, U. 2003. Ozonation of drinking water: part II. Disinfection and by-product formation in presence of bromide, iodide or chlorine. Water Res. 37:1469–1487. 174. Xiao, L. 2010. Molecular epidemiology of cryptosporidiosis: an update. Exp. Parasitol. 124:80–89. 175. Yai, L. E., A. R. Bauab, M. P. Hirschfeld, M. L. de Oliveira, and J. T. Damaceno. 1997. The first two cases of Cyclospora in dogs, Sao Paulo, Brazil. Rev. Inst. Med. Trop. Sao Paulo 39:177–179. 176. Zerpa, R., N. Uchima, and L. Huicho. 1995. Cyclospora cayetanensis associated with watery diarrhoea in Peruvian patients. J. Trop. Med. Hyg. 98:325–329. 177. Zimmerman, S. K., and C. A. Needham. 1995. Comparison of conventional stool concentration and preserved-smear methods with Merifluor Cryptosporidium/ Giardia Direct Immunofluorescence Assay and ProSpecT Giardia EZ Microplate Assay for detection of Giardia lamblia. J. Clin. Microbiol. 33:1942–1943. 178. Zu, S. X., and R. L. Guerrant. 1993. Cryptosporidiosis. J. Trop. Pediatr. 39:132–136.

Manila Typesetting Company

11/08/2012 08:57PM

SMP_Food Microbiology_CH28.indd

734

Manila Typesetting Company

11/08/2012 08:57PM

Preservatives and Preservation Methods

SMP_Food Microbiology_CH29.indd

735

Manila Typesetting Company

V 11/08/2012 07:11AM

SMP_Food Microbiology_CH29.indd

736

Manila Typesetting Company

11/08/2012 07:11AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch29

29

Ahmed E. Yousef V. M. Balasubramaniam

Physical Methods of Food Preservation

Preservation methods were originally developed to extend the shelf life of food by protecting the product from microbiological, chemical, and physical changes that could lead to spoilage. The microbiological changes are prevented by eliminating spoilage microorganisms or simply suppressing their metabolic activity. Modern preservation methods are designed not only to extend the shelf life of food, but also to ensure its safety by inactivating pathogenic microorganisms and viruses of concern, or in some cases just preventing their growth in the product. The word “inactivation,” as used in this chapter, refers to the destruction of microorganisms as judged by their inability to recover on microbiological media. The most commonly used preservation methods are physical in nature. Treatment of food with heat (i.e., thermal processing) inactivates spoilage-initiating microorganisms and enzymes, as well as disease-causing microorganisms. The process also may destroy heat-labile toxins that some pathogens may secrete. Removal of heat to refrigerate or freeze food suppresses microbial metabolism and multiplication, and the process also may inactivate a fraction of the food microbiota. Decreasing water availability is effectively used in preserving many foods through concentration or drying or by addition of water activity

(aw) modifiers. Alternatives to thermal processing have been introduced with the hope of preserving food effectively while minimizing deterioration of product quality during processing. Most of these alternative technologies also are considered physical preservation methods. These include gamma radiation, which is gradually gaining acceptance as an effective preservation method. Use of ultrahigh pressure to preserve prepackaged value-added food is increasing. Emerging preservation approaches also include using pulsed electric fields, UV light, and ultrasound, with the aim of ensuring food safety while minimizing adverse impacts of processing on product quality. Many of these physical treatments are addressed in this chapter, with emphasis on engineering background, microbiological considerations, and applications in food processing.

PRESERVATION BY HIGH TEMPERATURES When the temperature of a medium exceeds the optimum for the growth of a microorganism, cell multiplication slows and eventually ceases (52). An additional increase in temperature has a detrimental effect, and microbial cells experience injury or death. Mildly high temperatures (usually in the range of 55 to 90°C)

Ahmed E. Yousef and V. M. Balasubramaniam, Department of Food Science and Technology, The Ohio State University, 2015 Fyffe Court, Columbus, OH 43210.

737

SMP_Food Microbiology_CH29.indd

737

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

738

Figure 29.1  Microbial cell components that are altered or damaged during heat treatment. doi:10.1128/9781555818463.ch29f1

i­nactivate psychrotrophic and mesophilic microorganisms by damaging various cell components (Fig. 29.1). Membranes, proteins, and ribosomes are damaged by this heat treatment, leading to cell death (74, 82, 137). Thermal pasteurization is an example of a preservation process in which mildly high temperatures are applied. Higher dry heat can inactivate bacterial spores by altering their DNA (118). Inactivation of key components of the spore germination system plays a role in thermal inactivation of spores by heat (45). Temperatures that are high enough to inactivate bacterial spores are used in thermal sterilization of food.

Thermal Process Development, Validation, and Monitoring

Developing a thermal process for pasteurizing or sterilizing food requires extensive experimentation to ensure product safety and to confirm that the end product has acceptable quality. Considering the safety aspects, the thermal process should eliminate viable microorganisms of known public health significance that are capable of growing in the food at the conditions (e.g., temperature) at which the product is likely to be held during distribution and storage. These microorganisms will be referred to as “pathogens of concern.” Designing a process that totally eradicates a large population of these pathogens (when they are artificially added to food) is often unnecessary. When present in food, these pathogens are usually in small populations. Therefore, some processes are designed to be just sufficient to minimize the risk of disease transmission by food. Temperature and time are the most significant variables that processors optimize to achieve the aforementioned goal.

SMP_Food Microbiology_CH29.indd

738

Once a thermal process is designed, it should be validated (20, 71). The U.S. Food and Drug Administration (FDA) defines validation as establishing, by objective evidence, that a process consistently produces a result or product meeting its predetermined specification (141). In thermal processing, validation involves treating food, inoculated with the pathogen of concern or alternative biological indicator, at the predetermined process temperature and time and determining if the treatment achieves the desired pasteurization or sterilization. Once a process is successfully validated and allowed by regulatory agencies, processors can initiate commercial production. However, the process should always be monitored to ensure that desired pathogen lethality is delivered. Monitoring often includes observing and recording the temperature and time during processing, but the use of biological indicators in selected processing runs may be practiced. If these biological indicators meet certain criteria, they may be described as pathogen surrogates.

Pathogens of Concern and Biological Indicators

Food preservation has been reliably achieved by thermal processing because high temperatures can be lethal to all forms of life in food. Heat treatment is so effective that it can be customized to eliminate almost any microbial target, including spoilage and pathogenic microorganisms. However, commercial thermal processes are made just severe enough to protect product from spoilage and consumers from pathogens of concern; excessive heating is avoided to minimize product quality damage. A pathogen of concern is a microorganism with a known history of causing a disease transmitted by the product. Therefore, process severity is matched with process efficacy against this pathogen of concern. Examples of current pathogens of concern include enterohemorrhagic Escherichia coli in fruit juices, Salmonella enterica serovar Enteritidis in liquid eggs, and Listeria monocytogenes in milk and ready-to-eat meat. These products are thermally pasteurized with the goal of eliminating the risk of transmission of the pathogens indicated. Thermal commercial sterilization is a heat-intensive treatment that usually targets Clostridium botulinum spores. Therefore, spores of C. botulinum represent the pathogen of concern in most sterilization processes. Adequate processing should eliminate the most resistant strain of the pathogen of concern in the food. These pathogenic targets are investigated under controlled laboratory settings, as experiments involving these pathogens cannot be replicated in the foodprocessing environment for safety reasons. Therefore,

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation biological indicators are used in lieu of the pathogens to validate pilot- and industrial-scale processes in food-manufacturing facilities. Biological indicators are microorganisms with relatively high and stable thermal resistance that can be quantifiably measured and with characteristics that make them suitable for thermal process validation (143). Spores of Geobacillus stearothermophilus, Bacillus subtilis, Bacillus coagulans, and Clostridium sporogenes are biological indicators used in validating moist heat sterilization processes. Spores of certain strains of C. sporogenes have frequently replaced those of C. botulinum for the validation of commercial sterilization of food (138). Biological indicators used in thermal food processing, particularly pasteurization, also are described as pathogen surrogates. The surrogate, as applied in food processing, is a nonpathogenic species and strain responding to a particular processing treatment in a manner equivalent to that of the pathogenic counterpart (140). Surrogate microbes are often selected based on close genetic relationships to the target pathogen, assuming that genetic similarities equate to similar resistance patterns. It is tempting to simply seek surrogates that once were pathogenic strains but have had their virulence impaired naturally or through genetic manipulation. Despite the merits of using these mutants as surrogates, there are undesirable implications to this approach. Accidental release in the processing facility of a surrogate that is closely related to the targeted pathogen may constitute a burden on processors. The surrogate may be mistakenly isolated during rapid screening tests as a putative pathogen, warranting a regulatory action (145). Alternatively, researchers seek pathogen surrogates among GRAS (Generally Recognized as Safe) bacterial species. Lactic acid bacteria are often considered for use as suitable surrogates.

Engineering Principles

Thermal processing for preservation purposes involves heating the food material to a predefined temperature (process temperature), maintaining this temperature for a prespecified time (holding time), and subsequently cooling the product. Pasteurization and sterilization are typical examples of preservation approaches that utilize thermal processing principles. Rapid heating (initially) and cooling (after processing) is essential for minimizing product damage while achieving the goal of preservation. Transfer of heat from the heating medium to food occurs by one of three modes, or combinations thereof: conduction, convection, and radiation. The basic principles of heat transfer are briefly summarized; these can

SMP_Food Microbiology_CH29.indd

739

739 be reviewed in detail in the food engineering literature (see, e.g., references 31 and 120). Conductive heat transfer is primarily associated with foods having a solid or semisolid matrix (e.g., meat or cheese), in which transfer of energy occurs from one molecule to another. The heat transfer does not involve product movement; hence, these products often heat or cool slowly and temperature distribution within the product is not uniform. Since the heat transfer by conduction occurs from outside to inside the food, the geometric center of the product is usually the last portion of the product to receive the heat treatment. Considering a one-dimensional heat transfer when heating a rectangular-shaped slab of food material with a given surface area (A) and thickness (X), the rate of heat flow (q) through the material can be expressed as: q = -k A dT / dX where k is the thermal conductivity (W/[m·°C]) of the food material and dT is the temperature difference (°C). This simply means such food will heat faster when it has a larger surface area, smaller thickness, and greater temperature differential between the heating medium and the product’s coldest spot. Heat is transferred by convection in fluid foods, e.g., juice and milk. The heat exchange takes place between the fluid food and a solid interface or between the food and the heating (or cooling) medium. The rate of heat transfer (q) is a function of the solid surface area (A) and the temperature difference between the solid and fluid (Ts – Tf): q = h A (Ts - Tf ) where h is the surface convective heat transfer coefficient [W/(m2·°C)]. The thermal gradient results in a density difference in heated product, thus causing fluid movement and heat transfer by natural convection. Heat transfer also is facilitated by mechanical agitation, which cases forced convection. Movement of liquids in a can in a still retort is an example of natural convection. Forced convection occurs when a mechanical stirrer is used for fluid agitation, e.g., agitation of milk in a heating or cooling tank. The bulk product movement during convection causes reasonable uniformity in temperature distribution within the product, especially when mechanical agitation is applied. Transfer of heat by radiation occurs when one surface emits radiant heat (e.g., infrared waves) and another absorbs these waves. Heat transfer by radiation requires no physical medium and can occur in a vacuum.

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

740 Opacity of food to radiant heat depends on the form of this energy. The rate of heat transfer (q) from an object of surface area A can be expressed as q = σεA T 4A where s is the Stefan-Boltzmann constant [5.669 × 10–8 W/ (m2·K4)], T is absolute temperature, A is area (m2), and e is the emissivity.

Factors Influencing Heat Transfer

Considering the practical aspect of thermal processing, rate of heat transfer and process uniformity are influenced by many factors (37, 47). These include: •

•

•

•

•

•

Product type. Liquid foods are often heated by convection; consequently, heating rate is rapid. However, solid foods are heated slowly by conduction. A combination of conduction and convection heating occurs in some foods. Starch-containing liquid foods, for example, undergo convection heating initially, but thermally induced gelatinization of the starch shifts the process to conductive heating. Container material. Glass and metal are commonly used as packaging materials, but they differ considerably in thermal diffusivity. Glass containers heat more slowly than metal containers under similar processing and product conditions. Container shape. Containers that are tall and narrow promote heat transfer by convection; hence, the transfer of heat to the product is more rapid than with containers of other shapes. Container size. The center of a small container reaches the target temperature more rapidly than the center of a larger container. Agitation. Agitation and mixing assist convection current, and thereby increase heat transfer efficiency, especially in viscous or semiviscous products. Temperature of heating medium. The greater the difference in temperature between the heat transfer medium and the product, the more rapid the rate of heat transfer to the product.

lation survivors during heating time. These observations are best described in survivor plots that depict the logarithmic nature of population inactivation over time (Fig. 29.2A). Linearity of the survivor plots makes it possible to measure the death rate. Instead of death rate, decimal reduction time (D value) is a more practical value that can be deduced from survivor plots. D value (DT) is defined as the time for a 10-fold (or 1-log cycle) reduction in the number of survivors of a given microorganism at a specified temperature, T (Fig. 29.2A). The greater the D value at a given temperature, the more resistant is the microorganism to heat. Mathematically, DT can be expressed as DT = −

where N0 and N are the initial and final populations of a microorganism. DT serves as a reliable measure of heat resistance provided that the test strain is prepared and grown under defined conditions and heat treated using specified conditions. However, DT (of a given microbial species) changes with the strain and its physiological status, characteristics of the suspending medium (e.g., food composition, aw, and pH), and experimental conditions. Thermal inactivation D values of selected bacterial spores are provided in Table 29.1. The D value of a microorganism decreases as the processing temperature increases. A plot of log10 DT against temperature is generally linear. This “thermal resistance” plot is useful in comparing heat resistance of a microorganism at different temperatures. The plot can be used to calculate the z value (the thermal resistance constant), which mathematically is equal to the negative reciprocal of the slope (Fig. 29.2B). Hence, a z value represents the change in temperature required to change the DT values of a microorganism by an order of magnitude (i.e., 10-fold). Mathematically, z values are related to DT as described in the following equation: z=-

Kinetics of Microbial Death by Heat

Despite the multiplicity of factors affecting thermal treatments, temperature and time are the major parameters governing any thermal process operation. Processors often need to determine the heating time, at a specified temperature, that is sufficient to process the food effectively. Knowledge of the kinetics of microbial death helps in determining these critical processing parameters. Kinetics of microbial inactivation, at a given temperature, are determined by observing the decline in popu-

SMP_Food Microbiology_CH29.indd

740

∆t log N − log N0

DT log DT2 - log DT1

A larger z value indicates that the microorganism has greater resistance to variations in the heating temperature. Once the z value of a microorganism is determined, it is possible to determine its DT at any temperature, T, from the knowledge of decimal reduction time (D0) estimated at a specific reference temperature, T0: T0 −T z

DT = D0 10

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation

741

Figure 29.2  Graphic approach for determination of the decimal reduction time (D value) and the thermal resistance constant (z value). doi:10.1128/9781555818463.ch29f2

Thermal Lethality Measurements

To determine heat processes required to sterilize food, the thermal death time (F value) was introduced. F value (FT) is defined as the time in minutes, at a specified temperature, required to achieve a targeted reduction in a homogeneous population having a specific z value. Population reduction targets vary with the intended thermal process, i.e., pasteurization versus sterilization. Unlike D value, which defines the time it takes to reduce the population (at a given temperature) by 1 log, FT represents the time required to accomplish a targeted reduction in microbial population. Therefore, the relationship between F and D values is expressed as follows:

Because process lethality varies with the microorganism targeted and the processing temperature applied, it is important to designate a reference F value. A reference F value (F0) for canning represents the thermal death time when a microorganism with a z value of 10°C is treated at 121°C (250°F). The F0 value can be used as a measure of the lethality of a given heat treatment. Similar to the D value, F0 is affected by food composition (Table 29.2). The lethality concept is useful in comparing different temperature-time combinations that can be used to achieve the desired log reduction of a targeted microorganism. After determining the z value for the targeted microorganism and the F0 of the process, FT at any other temperature can be calculated using this relationship:

FT = DT (log N Initial - log NFinal ) where NInitial and NFinal are the initial and final populations of the microorganism, respectively, during the heat treatment. For example, if a thermal process (i.e., with defined temperature and time) is designed to reduce a microbial population from 102 to 10–3 CFU/g, the corresponding F value would be the holding time required to achieve 5-log reduction at the temperature tested. It is customary in the canning industry to associate F value with the time (at a given temperature) that is sufficient to achieve commercial sterility, which is assumed when the process can reduce a C. botulinum spore population by 12 log. C. botulinum spores, the target pathogen of concern in low-acid, shelf-stable canned foods, have a D121°C (D250°F) of 0.21 min and a z value of 10°C (18°F). Therefore, F121°C for a process necessary to commercially sterilize a low-acid food is the product of multiplication of 12 by 0.21, which equals 2.52 min (129).

SMP_Food Microbiology_CH29.indd

741

FTz = F0 10

T0 −T z

If a thermal treatment has an F0 of 1 min, this means the process accomplishes a predetermined microbial inactivation through heating the product at 121°C for 1 min. Processes equivalent to the one just described (i.e., timetemperature combinations that produce similar microbial inactivation) can be determined using the equation immediately above. For example, processing the same product at 111 and 131°C would require 10 min and 0.1 min, respectively. Therefore, processes having an F0 of 1 min, F111°C of 10 min, and F131°C of 0.1 min are all equivalent in severity. Hence, increasing the process temperature allows processors to reduce the processing time. On the contrary, a product needs to be held for a longer time at lower process temperatures to receive a lethal treatment similar to that achieved at the reference temperature. Plotting the changes in microbial population during thermal treatment may produce an asymptotic curve instead of a linear relationship. Even though conventional

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

742

Table 29.1  Comparison of heat resistance of selected bacterial spores in thermally processed foodsa Microorganism(s)

Significance

Geobacillus stearothermophilus Clostridium botulinum types A and B Clostridium sporogenes

Bacillus coagulans

Paenibacillus polymyxa and Paenibacillus macerans Clostridium pasteurianum Bacillus licheniformis

Thermophile. Flat-sour spoilage of low-acid foodb. Mesophilic anaerobe. Toxin producer in low-acid food. Mesophilic anaerobe. Spoilage of low-acid food. Thermophile. Spoilage of acid and acidified foodc. Mesophiles. Spoilage of acid and acidified food. Mesophilic anaerobe. Spoilage of acid and acidified food. Mesophile. Spoilage of acid and acidified food.

Reference temperature (°C)

D value (min)

z value (°C)

121.1

4.0–5.0

7.8–12.2

121.1

0.10–0.20

7.8–10.0

121.1

0.10–1.5

7.8-10.0

121.1

0.01–0.07

7.8–10.0

100

0.10–0.50

6.7–8.9

100

0.10–0.50

6.7–8.9

93.3

4.5

15.0

 Adapted from reference 141.  Low-acid food, pH >4.6.  Acid or acidified food, pH 4.0 to 4.6.

a

b c

thermal process calculations assume linearity (Fig. 29.2), the survivor plot often deviates from linearity. The nonlinearity (expressed in the form of shoulders and tailing) could be attributed to many factors, including the presence of heat-resistant subpopulations, deficiencies in enumeration protocols, or cell clumping. Despite this nonlinearity, conventional thermal processing relies on D and z values calculated from the linear portion of the survivor plot, but processors add margins of safety during these calculations to ensure product safety. Additional complications in thermal process calculations arise from the fact that the heating and cooling of food do not occur instantaneously. The inadvertent gradual increase in product temperature to the assigned process temperature contributes to process lethality. To avoid overprocessing, lethality during process come-up time should be taken into account. Thermal lethality measurements are described in greater detail in several publications (see, e.g., references 101 and 130).

Resistance of Microorganisms to Heat

High temperatures injure or kill living cells by denaturing vital proteins, disrupting membrane integrity, and

SMP_Food Microbiology_CH29.indd

742

damaging DNA. Some prokaryotic microorganisms (i.e., bacteria and archaea) have greater innate protective mechanisms against heat than others (80). These mechanisms, which include protein cross-linking and DNA structural changes, protect bacteria against temperatures higher than their upper growth range. For example, thermophiles express genes involved in disulfide bonding, and this response contributes to the heat resistance of these bacteria (12). Regardless of the optimum growth range, bacteria have a number of stress responses to heat shock. These responses include expression of heat shock proteins, often called chaperones. Some of the functions of chaperones include preventing misfolding or undesired aggregation of newly synthesized proteins and correcting improper protein folding (119). Lack of expression of the genes encoding these proteins decreases the heat tolerance of bacteria (150). Stress response genes are often involved in more than one stress response and are considered to be components of a general bacterial stress response (49). Stage of growth and state of dormancy affect the heat resistance of microorganisms. Bacterial cells in the stationary phase of growth are more heat resistant than those actively multiplying during the exponential phase

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation

743

The pH of a food is a critical factor in determining the extent of heating needed for pasteurization or commercial sterilization. Low pH sensitizes microorganisms to heat; hence, a milder treatment is needed to process an acidic than a low-acid food. Acidity is so important that regulatory agencies classify foods into low-acid and acid categories (142), and greater regulatory scrutiny is given to the former than the latter. Low-acid foods (pH ³ 4.6) are presumed to support growth and toxin production by C. botulinum. Products such as milk, meat, vegetables, and soups are examples of low-acid foods. Commercial sterilization of low-acid foods is achieved by heating the product at a high temperature (e.g., 121°C) for a duration that is sufficient to inactivate C. botulinum spores. Food products with a natural pH of < 4.6 are considered acid foods. Citrus fruit juices (e.g., orange juice) are good examples of acid foods. These foods may be commercially sterilized at £100°C to destroy both vegetative cells and heat-resistant spores of spoilage organisms, particularly fungal spores.

and enzymes and either decreases considerably or eliminates the population of the most heat-resistant pathogens of public health significance. At the advent of pasteurization, it was intended for raw milk and the main pathogens targeted by the process were Mycobacterium spp. (54). Subsequently, other pathogens were identified as being transmitted by milk (64); hence the significance of pasteurization increased. Currently, known milk pathogens include Bacillus cereus, Brucella spp., Campylobacter jejuni, Coxiella burnetii, enterohemorrhagic E. coli, L. monocytogenes, several Mycobacterium spp. (e.g., M. paratuberculosis), Salmonella spp., and Yersinia enterocolitica (110). Some heat-resistant microbes, including bacterial spores, survive pasteurization processes; these are often described as thermoduric microorganisms. Therefore, refrigeration should follow pasteurization to keep these microorganisms from growing. Additionally, pasteurized products should be properly packaged to prevent recontamination. In addition to milk, other foods that are thermally processed and labeled as “pasteurized” include juices, liquid eggs, wholeshell eggs, almonds, and others. Thermal pasteurization involves heating the food to a predetermined temperature and holding the product at this temperature for a specified duration. Different combinations of temperature and time can be used to pasteurize a food (37). For example, milk can be pasteurized at 63°C for 30 min or at 71.8°C for 15 s. Acid foods are pasteurized at lower treatment intensity than that needed for low-acid foods. Acidity sensitizes microorganisms to heat. As a moderate heat treatment, pasteurization often causes minimal changes in the sensory properties of food. Pasteurized products often have a shelf life of 2 to 3 weeks under refrigerated storage. The goal of pasteurization is to extend the shelf life of food and to make it safe for human consumption. Safety cannot be quantified experimentally, but regulatory agencies set guidelines for processors so that product safety is reasonably assured. According to regulatory guidelines in the United States, processors can label some products (e.g., citrus juices and shell eggs) as pasteurized if they have been processed to produce at least 5-log reduction of the most processing-resistant pathogen of the greatest public health concern for that particular food (90). Although heat is often used to pasteurize food, there is a trend to use alter­ native lethal agents, provided these alternative technologies accomplish the goals stated earlier for pasteurization.

Pasteurization

Sterilization

Table 29.2  Approximate F0 values needed to commercially

sterilize selected canned foodsa Product

F0 value (min)

Green beans in brine Carrots in brine Cream-style corn Dog food Meat loaf a

3.5 3–4 5–6 12 6

 Adapted from references 37 and 91.

(147). Spore formation dramatically increases the heat resistance of bacteria. Although vegetative cells of sporeforming bacteria are inactivated readily at 55 to 72°C, spores of these same bacteria require heating at 99 to 115°C for a similar degree of inactivation (146). The heat sensitivity of fungal mycelia and conidia generally is similar to that of vegetative bacteria, but ascospores and sclerotia of some fungi are more resistant to heat (16, 127). Physical and chemical properties of food may affect the heat resistance of contaminating microorganisms. The aw, pH, gross composition, and antimicrobial constituents of a food have a profound influence on the heat resistance of its microbiota (27, 128).

Food pH and Resistance of Microorganisms to Heat

Thermal pasteurization is a relatively mild heat treatment intended to prolong the shelf life of food and protect consumers from diseases likely to be transmitted by that particular food. The process inactivates spoilage microorganisms

SMP_Food Microbiology_CH29.indd

743

Sterilization is the process of rendering a product free of any living organisms. Proper autoclaving, for example, renders microbiological media sterile. Application of this strict definition of sterilization to food would require

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

744 exposing the product to excessive heating and result in products with unacceptable quality. Alternatively, food may be heat treated just sufficiently to produce a shelfstable and safe product. Such a process is known as commercial sterilization. Commercial sterility of thermally processed food, as defined by the U.S. FDA, is the condition achieved by the application of heat that renders the food free of (i) microorganisms capable of reproducing in the food under normal nonrefrigerated storage and distribution conditions and (ii) viable microbial cells or spores of public health significance (26). Consequently, commercially sterile food may contain a small number of viable, but dormant, nonpathogenic bacterial spores. Commercial sterilization involves heating the food to a predetermined temperature and holding the product at this temperature for a specified duration of time, with the goal of inactivating spores of bacteria of greatest public health significance. The food is often prepackaged in hermetically sealed containers; therefore, anaerobic sporeformers capable of causing disease are the pathogens of concern in commercially sterile foods. C. botulinum is an anaerobic bacterium that produces heat-resistant spores, and the corresponding vegetative cells are capable of producing a highly toxic neurotoxin (123). Low-acid foods (with

pH ³4.6) are a favorable environment for C. botulinum spores to germinate, cells to multiply, and toxins to be produced. Hence, strict measures are taken during sterilization of low-acid foods to ensure inactivation of C. botulinum spores. Historically, food processors applied thermal treatments sufficient to reduce a population of C. botulinum spores by 12 log (i.e., 12D processes) to commercially sterilize low-acid canned foods. This treatment may not totally eliminate spores of some spoilage microorganisms, such as G. stearothermophilus (72). This bacterium, however, is a thermophile that is unlikely to grow at ambient temperatures of most storage conditions.

Advances in Thermal Processing Aseptic Processing

Canning involves commercially sterilizing prepackaged (i.e., canned) food, whereas aseptic processing is applied to foods that are thermally treated prior to packaging. Commercial sterilization of prepackaged food is inherently a batch process. Commercial sterilization of canned food in a retort, which is a batch process, causes prolonged exposure to heat during temperature come-up, holding, and cooling. Excessive heat-

Figure 29.3  Schematic diagram of an aseptic processing system. doi:10.1128/9781555818463.ch29f3

SMP_Food Microbiology_CH29.indd

744

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation ing, which is unavoidable in batch processes, degrades products’ sensory and nutritional qualities. These drawbacks can be avoided by aseptic packaging, where the food and packaging material are commercially sterilized separately and then combined under aseptic conditions (109). Aseptic processing is a continuous thermal process in which the pumpable food material is heated at elevated temperatures (>130°C) by passing it through a set of heat exchangers (Fig. 29.3). The hot product is held in a tube until the process’s holding time is achieved and then passes through a set of cooling heat exchangers. The cooled product is filled in presterilized packages and hermetically sealed in an atmosphere that is free of microorganisms. Aseptic processing is widely used for preservation of fruit juices, dairy products, and sauces. A typical example of aseptically processed food is ultrahigh-temperaturetreated milk. It is significant that this technology is not being applied to low-acid, particulate-containing foods (109). Several publications, e.g., Sastry and Cornelius (114), provide detailed information on the aseptic processing of foods. Aseptically processed food may be heated by conventional means. Alternatively, advanced thermal treatments have been introduced, including ohmic and microwave heating technologies. These alternative heating technologies are described here briefly.

Ohmic Heating

In ohmic heating, electricity is passed through food, which serves as an electrical resistor. When the electric current passes through the food, the energy is dissipated as heat, which results in rapid and uniform volumetric heating of the product (115). Electricity for ohmic heaters can be obtained from a normal electrical source (60 Hz in the United States and 50 Hz in Europe). Electrical conductivity of food materials determines how efficiently the electric current passes through the food, and hence, knowledge of the electrical conductivity of various products is critical. Electrical conductivity of food is influenced by ionic (salt) content; therefore, it may be possible to adjust this food property to achieve desired ohmic heating. The advantages of ohmic heating include its capacity for rapid and volumetric heating of food. In contrast, heat must travel from the outside surface to the inside of food during conventional heating. It is generally assumed that the microbial inactivation that occurs during ohmic heating is primarily due to thermal effects (98). Recent research has revealed that there may be nonthermal effects within the range of electric field strengths commonly used with ohmic heating (113).

SMP_Food Microbiology_CH29.indd

745

745

Microwave Heating

Microwaves are a part of the electromagnetic spectrum, covering the frequency range of 300 to 300,000 MHz (120). Application of microwave energy causes rotation of dipolar molecules in food, i.e., water molecules. These rotations create intermolecular friction, which produces heat (87). Penetration of microwaves in food produces a volumetrically distributed heat. Compared with conventional heating, where heat is usually transferred from the surface to the interior, microwaves generate heat throughout the material, leading to faster heating rates and shorter processing times. However, distribution of energy and associated heat in a microwave field is inherently uneven (94); this leads to thermal nonuniformity and uneven lethality of microorganisms in the treated product (53). This limitation has hindered application of microwave heating in commercial-scale food processing. Inactivation of microorganisms is attributed to the thermal effect of microwaves, and any nonthermal effects are assumed to be negligible (59).

LOW-TEMPERATURE PRESERVATION Both the addition and removal of heat are governed by the same principles of heat transfer. Conduction, convection, and radiation are the methods of heat transfer whether the food gains or loses heat. Addition of heat to food causes microbial destruction, whereas heat removal (for chilling or freezing) merely suppresses microbial growth. As food gains energy and the temperature increases to ambient, most food microbiota can grow. Freezing, chilling, and thermal treatment, nevertheless, share a common theme; all are considered temperaturebased preservation methods. Freezing stops the metabolic activity of most foodborne microorganisms, and hence product deterioration by microorganisms is prevented. Freezing and thawing cycles may cause physical damage and kill living cells, including food microbiota. Toxins in food are generally not affected by the freezing process. Chilling suppresses the growth and metabolic activity of most food microbiota; however, psychrophilic and psychrotrophic food microbiota grow at these temperatures (108). Low temperatures preserve food not only by suppressing microbial growth and metabolism but also by retarding chemical and biochemical deterioration. It is generally accepted that lowering the temperature by 10°C decreases reaction rates by 50%. Reactions that cause product deterioration include lipid oxidation, Maillard reactions, enzymatic browning, proteolysis, and lipolysis; these are suppressed during chilling and freezing. Low-temperature preservation is addressed

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

746 briefly in this chapter, and the topic is covered in greater depth in the published literature (30, 37, 47).

Chilling

Chill storage refers to holding food below ambient temperature and above freezing, generally in the range of –2 to ~16°C. Most foods remain unfrozen at –2°C even though pure water freezes at 0°C (Fig. 29.4). Chilling is accomplished by removing sensible heat energy from the product using mechanical refrigeration or cryogenic systems. Fresh fruits and vegetables may be chilled by evaporative or vacuum cooling to remove field heat from the product. Many foods, particularly sensitive products such as raw milk and poultry, are rapidly chilled after harvest. If chilled slowly, these products may spoil or support growth of pathogenic microbes. Chilling poultry carcasses, for example, is so important that rate of cooling is regulated in the United States (25). Slow cooling can lead to multiplication of spoilage and pathogenic microorganisms on poultry carcasses (63). Some fresh fruits and vegetables, however, may undergo chilling injury if held below critical temperatures within the chilling range described earlier (155). After processing or cooking, many foods are chilled and kept refrigerated during storage and retailing. Foods receiving minimal preservation treatment (i.e., pasteurization, acidification,

and fermentation) are promptly refrigerated to prevent growth of the microorganisms that survive processing; hence chilling is a secondary barrier to microbial growth. Cook-chill products such as roast meat and precooked meals are chilled after the cooking treatment. It is important that chilled foods are protected against psychrotrophic pathogens, which are capable of growing at low temperatures (108). Aeromonas hydrophila, B. cereus, some nonproteolytic C. botulinum strains, L. monocytogenes, Vibrio parahaemolyticus, and Y. enterocolitica are disease-causing psychrotrophic bacteria (78). The presence of these pathogens in minimally processed foods can compromise the safety of these products. Activity of psychrotrophic spoilage microorganisms often determines the shelf life of chilled products. Common psychrotrophic spoilage microbes include Pseudomonas spp. and some lactic acid bacteria. Numerous fungi are psychrotrophic in nature and cause deterioration of many food products (104). Low temperature generally decreases metabolic reactions that contribute to microbial growth and metabolism; however, response to these low temperatures varies considerably with the type of microorganism. The temperature range for growth of a microorganism depends on how well the microbe can regulate cellular lipid fluidity (50). For psychrotrophic bacteria, membrane phospholipids must remain in a liquid-crystalline state to

Figure 29.4  Schematic diagram showing changes in the temperature of food and water during freezing. doi:10.1128/9781555818463.ch29f4

SMP_Food Microbiology_CH29.indd

746

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation maintain membrane fluidity, which enables their growth at low temperature. The composition of the lipid bilayer determines whether membrane phospholipids are in the liquid-crystalline state. For example, membranes of L. monocytogenes contain >95% branched-chain fatty acids (4). When grown at 37°C, the major fatty acids are anteiso-C15:0 (41 to 52%), anteiso-C17:0 (24 to 51%), and iso-C15:0 (2 to 18%). When listeriae are grown at 5°C, the anteiso-C15:0 becomes predominant, representing 65 to 85% of total membrane fatty acids. Reduction in the proportion of long aliphatic chains (C17:0) and the increase in asymmetric branching reduce van der Waals bonds among membrane constituents. This, in turn, reduces the tight packing of membrane phospholipids at low temperature, which helps maintain membrane fluidity. Once it is chilled, the cold temperature of the product should be maintained during refrigerated storage. In addition to temperature, other factors could be controlled during refrigerated storage, depending on the food. These factors include relative humidity and the gaseous environment in the package. Temperature fluctuation during refrigerated storage should be avoided. Increasing the temperature by a few degrees could increase the microbial growth rate, leading to product spoilage or increased risk of foodborne illness. Depending upon the type of food and temperature of storage, shelf life can vary from a few days to several weeks.

Modified-Atmosphere Packaging

Modified atmosphere refers to an altered gaseous environment, which is applied to food enclosed in a gas barrier material to extend the shelf life of the product. Modified-atmosphere packaging (MAP) involves replacing air in the food package with a single gas or mixture of gases (86). Many combinations of carbon dioxide, nitrogen, and oxygen can be used in MAP (42). Maintaining refrigeration during storage of the packaged food is critical for successful MAP. Vacuum-packaging is considered MAP in which a vacuum replaces any air in the package. Modifying the atmosphere, in conjunction with refrigerated storage, maintains the quality of fresh meat, vegetables and fruits, bakery and dairy products, and seafood. “Sous vide” processed foods are vacuumpackaged, then cooked, chilled, and subsequently stored at refrigeration temperature (24). Applying MAP to meat helps to control aerobic spoilage bacteria, including those belonging to the genera Pseudomonas, Acinetobacter, and Moraxella (35). However, psychrotrophic anaerobes or microaerophiles can grow in vacuum-packaged or MAP products, of which some are pathogenic bacteria. Lactic acid bacte-

SMP_Food Microbiology_CH29.indd

747

747 ria are one of the most significant groups of spoilage microbes for these products (42).

Controlled-Atmosphere Storage

Controlled-atmosphere storage is similar to MAP, as both technologies involve replacing air around the food with gases of different composition; however, these two processes are different in other aspects. Controlledatmosphere storage implies precise control of gas composition throughout storage, whereas in MAP, modification of gas composition occurs at the beginning of storage and no further control is applied (86). Controlled-atmosphere storage of fruits and vegetables is used widely (1). This involves reducing the oxygen concentration to 2 to 5% and increasing the carbon dioxide concentration to 8 to 10%. These changes suppress respiration and adverse quality changes of stored fruits and vegetables while inhibiting the growth of some spoilage microorganisms.

Food Freezing Engineering Aspects

Freezing of food, which usually contains an ample amount of water, is normally accomplished by decreasing its temperature to at least –18°C and maintaining the food at this low temperature. Food may be frozen by direct contact with a chilling agent. The freezing medium can be a heat exchanger plate (maintained at –30 to –40°C), dry ice, or liquid refrigerant (e.g., liquid N2 at –196°C). Initially, sensible heat is removed from the product through conduction and convection heat transfer. After removal of sensible heat, the food temperature decreases below the freezing point of water, and the liquid component reaches the “supercooling” stage (Fig. 29.4). Freezing (i.e., phase conversion of water to ice) normally begins at –1 to –3°C and continues over a broad temperature range. The latent heat is removed from the product and its temperature continues to decrease. Gradual formation of ice crystals increases the concentration of food solutes into the remainder of the unfrozen water. This increasingly concentrated solution requires lower temperatures to freeze. The concentration process reaches a level at which the solutes precipitate and the residual water freezes. At this point (the eutectic point), the food reaches a totally frozen state. This state is reached at –15 to –20°C for fruits and vegetables, –25°C for ice cream mix, and below –40°C for meats. Figure 29.4 depicts changes in temperature during freezing of food (37). During thawing, energy is supplied to the frozen food to melt the ice crystals. Thawing creates a temperature

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

748 gradient, with the product’s outer surface rising above freezing temperature first. Since water has lower thermal conductivity values than ice, the outer layer of water decreases the rate of heat transfer to the product center. Therefore, it is generally faster to freeze than to thaw food.

Microbiological Considerations

While food temperature decreases during freezing, aw decreases; therefore, only cold-tolerant and xerotolerant microorganisms can grow. Some fungi have these characteristics and thus may grow on frozen products such as meat and butter (104). Other microorganisms experience cold shock, osmotic shock, or both, which may injure their cells. The concentration of intracellular liquids changes the pH and ionic strength, thereby affecting enzymes’ activity, denaturating proteins, and hampering many biological functions. Mechanical injury can occur when relatively large crystals form inside microbial cells. Mechanical damage occurs more often when the product freezes at a slow rather than a fast rate (85). Cell recovery or death from injury during freezing depends on the microorganism, the medium (i.e., food matrix), and how the freezing process was applied. Gram-positive bacteria are more resistant to freezing than gram-negative bacteria (58). Parasites can be destroyed relatively easily by subzero freezing temperatures. For example, the viability of Trichinella spiralis in pork can be effectively eliminated if the raw product is held briefly at –23.3°C (39). The composition of the medium profoundly affects the survival of freezing by suspended or embedded microorganisms. The presence of sodium chloride, for example, reduces the freezing point of solutions, thereby prolonging cells’ exposure to high solute concentrations before freezing occurs. This condition is deleterious to some microorganisms. To the contrary, the presence of glycerol, propylene glycol, and some proteins generally has a cryoprotective effect (55). Compared with rapid or instantaneous freezing (e.g., by dipping in liquid nitrogen), slow freezing (e.g., using a household freezer) tends to produce larger ice crystals and to expose microbial cells to osmotic shock for a longer period of time. Under these conditions, microbial cells are more susceptible to injury and death during freezing (85). Recovery of freeze-injured microorganisms, or multiplication of their healthy cells, may begin during the thawing process. Thawing often is accompanied by the release of nutrient-rich liquid from damaged food cells. The microbiota that survives freezing may multiply in this liquid or after the microbes gain entry into damaged

SMP_Food Microbiology_CH29.indd

748

food tissues. Therefore, thawed products may be vulnerable to rapid microbial spoilage or an increased risk of growth of pathogens.

Quality Considerations

Proper freezing maintains product safety, but quality may gradually deteriorate during storage. Freezer burn and unfavorable color change may occur during freezing. The cells and the tissue structure of foods also may be damaged by freezing, frozen storage, and thawing. Hence, the storage life of frozen food is determined by quality changes. The storage life of frozen food ranges from months to more than a year (38).

PRESERVATION BY DECREASING WATER AVAILABILITY Water is one of the major constituents of food and exists in bound or unbound form. The bound portion of water adheres to food components by physical or chemical binding mechanisms. Within the food matrix, there are many sites that bind water; these include hydroxyl groups of polysaccharides, the carbonyl and amino groups of proteins, and salt ions. Unbound (or free) water is present within the pores or the interstitial tissue of the food. The amount and form of water influence microbial behavior, as well as the physicochemical and biological reactions in food. The presence of unbound water is essential for microbial growth. Removing or restricting the availability of water can preserve the food by suppressing microbial metabolism and multiplication, thus extending the product’s shelf life and enhancing its safety. There are a variety of approaches to modify water availability in food. Drying is the most commonly used method of removing water from food. Adding humectants (i.e., additives used mainly to help retain moisture in food) such as glycerol or other chemical binding agents (e.g., salt) is used to limit water availability for microbial growth in food. The concept of water availability is addressed briefly in this chapter, but many published articles provide more detailed coverage of this topic (10, 47, 65).

aw

If a moist food (e.g., cheese sauce) is contained in a package with sufficient headspace, water molecules gradually leave the food and enter the surrounding atmosphere (Fig. 29.5). The movement continues until a state of equilibrium is achieved between water in the food (i.e., its moisture content) and water vapor in the air surrounding the food. At the equilibrium state, food neither gains water from nor loses water to its environment; therefore, the food’s water is described

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation

749

Figure 29.5  Illustration of the water availability concept. aw is the ratio between the water vapor pressure on food (P) and that on pure water (P0), measured at the same temperature and when the system is at equilibrium. doi:10.1128/9781555818463.ch29f5

as equilibrium moisture content and water in the air around the food is expressed as water vapor pressure (P). The greater the moisture content of the food, the higher its water vapor pressure. In addition to moisture content, which is often expressed in weight percentage, other food components determine the availability of water for migration to the environment surrounding the food. For example, the presence of water-binding constituents could diminish this water movement. If pure water, instead of food, was in the package (Fig. 29.5), a similar equilibrium is achieved, but water vapor pressure (P0) would be at the highest state. If these were side-by-side experiments conducted at the same temperature, the ratio of P and P0 would be indicative of the availability of water in the food, i.e., the aw. Therefore, aw is defined as the ratio between the water vapor pressure resulting from the food (P) when in complete equilibrium with the surrounding air media and the saturation vapor pressure of pure water (P0) under identical conditions (see the following equation). Since changes in temperature lead to differences in the energy of water molecules and their movement from food to air, it is important that vapor pressure of food and water are measured at the same temperature.

or biological activities are better described by a product’s aw than by the amount of water present in the food. Foods can be divided into approximately three categories depending on their aws (88). High-moisture or highly hydrated foods, such as milk, meat, and fresh fruits and vegetables, are those having aws of >0.90. If other compositional and environmental factors are favorable, most foodborne microorganisms can grow in foods of this category. Intermediate-moisture foods are products having aws in the range of 0.65 to 0.90, depending on product composition and physical characteristics. Because of the moderately low aw, intermediatemoisture foods are relatively resistant to microbial spoilage. Examples of intermediate-moisture foods are raisins and jams. Dry foods such as milk powder and crackers have aws of <0.65. These products do not support the growth of foodborne microorganisms and have a characteristically long shelf life. Although high aw is conducive to microbial spoilage of food, some chemical deterioration is favored at intermediate aw. The rate of nonenzymatic browning caused by Maillard reactions in a model system decreases with increasing water content (33). Lipid oxidation also is favored at aws in the range of 0.6 to 0.7 (68).

aw = P / P0

Water is the largest constituent of a microbial cell. The functionality of these cells depends greatly on water availability. For example, nutrients are transferred into cells and waste materials are removed from cells in soluble forms. An aqueous intracellular environment is needed for metabolic reactions. It is not surprising that removal of water from food and from contaminating microorganisms halts microbial proliferation. Growth of most foodborne microorganisms is favored in products with the highest aw. The presence of microbial contaminants in a food having favorable aw, in combination with other factors such as low acidity, can lead to quality deterioration or compromise the safety of foods. Conversely, microorganisms have limiting

For pure water, aw equals 1. For foods, the aw value is in the range of 0 to 1, depending on the moisture content and the composition of the product. The ratio between P and P0 is directly related to the equilibrium relative humidity (ERH) in the air surrounding the food, as indicated by the equation ERH(%) = aw ´ 100 The relationship between aw (or ERH) and the moisture content of food is not linear (67). There is a clear distinction between aw and moisture content. Chemical

SMP_Food Microbiology_CH29.indd

749

Microbiological Considerations

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

750 aw levels below which they will not grow (116, 117). These limits vary for different microorganisms. Bacteria in food have the highest aw requirements for growth. Many spoilage (e.g., Pseudomonas spp.) and pathogenic (e.g., C. botulinum type E) bacteria grow only at aws of 0.95 or higher (Table 29.3). Spore-forming bacteria, L. monocytogenes, and Vibrio spp. are examples of bacteria having growth-limiting aws in the range of 0.91 to 0.95. Staphylococcus aureus is the foodborne bacterial pathogen with the lowest aw requirement. The bacterium grows in media with an aw of ³0.86; however, it needs media with a much higher aw for production of enterotoxins (60, 105). Yeasts, in general, have aw requirements that are lower than those of bacteria (Table 29.3). Debaryomyces hansenii, Candida pseudotropicalis, Zygosaccharomyces bailii, and Zygosaccharomyces rouxii are spoilage yeasts that grow in foods with aw values that are low enough to suppress bacterial growth. Intermediate-moisture foods

(aw from 0.65 to 0.90) are often subject to spoilage by these yeasts. In addition to their tolerance to low aw, yeasts can grow in acid or acidified foods and cause major economic losses (104). Many molds are considered xerotolerant and grow at an aw lower than that required for growth of other foodborne microorganisms. Minimum aw growth requirements of Eurotium spp., for example, are in the range of 0.7 to 0.8 (Table 29.3). Xerophilic fungi, such as Xeromyces bisporus, have very low aw requirements and also do not grow at high-aw (>0.97) conditions (102). Fungi are not only spoilage microbes, as some also produce potent toxic and carcinogenic agents. Fortunately, low-aw conditions that support growth of these molds do not support toxin production by them (103). There are limitations for using aw to establish the microbial safety and stability of food. Growth at reduced water availability varies not only with the microorganism but also with the humectant used to lower the aw

Table 29.3  Approximate aw minima for growth of foodborne microorganisms and examples of relevant foodsa aw

Bacteria

0.95–0.97

0.91–0.94

0.87–0.90

Fungib

Clostridium botulinum type E, Clostridium perfringens, Pseudomonas spp., Enterobacteriaceae (e.g., Escherichia coli and Salmonella spp.), Vibrio cholerae Clostridium botulimum types A and B, Bacillus spp., Listeria monocytogenes, Vibrio parahaemolyticus Staphylococcus aureus

Fresh vegetables and fruits, milk, fresh meat, hot dogs, fresh sausages, soft cheeses, canned food (these are sterile unless cans are opened)

0.81–0.85

0.71–0.80

0.61–0.70

0.20–0.60

a

Product (example)

Rhizopus nigricans

Cured meats, some cheeses, bakery goods (e.g., bread), evaporated milk

Trichothecium roseum (Candida spp.), (Saccharomyces cerevisiae) Aspergillus fumigatus, Aspergillus parasiticus, Penicillium expansum, Penicillium patulum, Penicillium viridicatum, Rhizopus spp., (Debaryomyces hansenii) Aspergillus flavus, Aspergillus niger, Aspergillus ochraceus, Eurotium spp., (Zygosaccharomyces bailii) Eurotium halophilicum, Xeromyces bisporus, (Zygosaccharomyces rouxii)

Aged cheese, fermented sausages (e.g., salami), dried meat (jerky), bacon, fruit juice concentrates, chocolate syrup, fruit cakes, sweetened condensed milk

No microbial growth

Jams, salted fish, marmalade, molasses, dried figs, salted fish Corn syrup, marshmallows, chewing gums, dried fruits, dry grains, dry pet food Caramel, honey, noodles, crackers, cakes mixes, infant formula (powder), milk powder, flour

 Adapted from references 6 and 104.  Yeast species are in parentheses.

b

SMP_Food Microbiology_CH29.indd

750

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation Table 29.4  Comparison between NaCl and glycerol, when

used as humectants to decrease water availability, on the minimum aw that supports the growth of pathogenic bacteriaa aw attained by the humectant

Bacterium

Clostridium botulinum type E Escherichia coli Clostridium perfringens Clostridium botulinum types A and B Vibrio parahaemolyticus Bacillus cereus Listeria monocytogenes Staphylococcus aureus

NaCl

Glycerol

0.966 0.949 0.945 0.940

0.943 0.940 0.930 0.930

0.932 0.930 0.920 0.860

0.911 0.920 0.900 0.890

a

 Adapted from reference 10.

(Table 29.4). Glycerol can permeate the bacterial membrane and may have a different inhibitory effect than that of sodium chloride or sucrose. It should also be obvious that lowering a food’s aw inhibits microbial growth but may not kill the microorganisms in the food. These microorganisms may remain dormant until the food is rehydrated, and then resume their metabolic activity and multiply. Therefore, it is important to decontaminate dried foods that may become rehydrated (e.g., milk powders).

Microbial Osmotolerance

In response to relatively high salt contents (e.g., 6 to 8% NaCl) in agar media, bacteria form elongated cells or filamentous structures (62, 79). This phenotypic change may be accompanied by colonies growing with a rough surface and irregular borders. Changes in the cell and colony morphology of L. monocytogenes in response to the medium salinity alter the adhesion properties of the pathogen (15). It is not known how these morphologic changes help bacteria cope with the reduced aw in the media. Microorganisms remain viable under some osmotic stress by maintaining intracellular osmoregulation. When exposed to osmotic stress, microorganisms accumulate compatible solutes (osmoprotectants) in cell cytoplasm. These solutes bind water but do not interfere with the metabolic activities of the cell. In response to low aw, some bacteria accumulate potassium ions and amino acids (21). Xerotolerant fungi may accumulate glycerol, erythritol, and arabitol (32). Variability among microorganisms in osmoregulatory capacities may explain the differences in aw limits for growth (44).

SMP_Food Microbiology_CH29.indd

751

751

Methods To Remove Water from Food Drying

Drying is a process of mobilizing water present in the internal food matrix to its surface and then removing the surface water by evaporation. Drying often involves simultaneous heat and mass transfer. Most drying operations involve converting liquid water present within the food material to vapor form and removing it by passing hot air over the product. During drying, the heat is transferred from the external heating medium to the food. The moisture within the food moves toward its surface and evaporates into the heat transfer medium (usually air). Transfer of heat to the food can be accomplished through conduction, convection, and radiation. Although convection is the dominant heat transfer mechanism at the surface, conduction transfers the heat within the food. The moisture moves within the food by diffusion. Dehydration helps food processors to achieve the following goals: (i) preserving food by preventing the proliferation of spoilage and pathogenic microorganisms; (ii) reducing the weight and bulkiness of the food by removing water; and (iii) developing novel food products, such as snacks, by controlling the textural properties of food. Drying technologies are covered in depth in many books (see, e.g., reference 112).

Freeze-Drying

During freeze-drying, the product is frozen first and the moisture is subsequently removed by sublimation (47, 56). Sublimation converts frozen moisture (solid state) to vapor (gaseous phase) without going through the intermediate liquid state. Typically, freeze-drying includes these steps: prefreezing, freezing, and primary and secondary drying. The food is pretreated and then frozen by decreasing its temperature below the eutectic point (»–40°C); this converts all the moisture in the product into ice. It is favorable that the freezing technique promotes the formation of small ice crystals in the food. This may be accomplished by immersing the product in liquid nitrogen. For liquid foods, a drum freeze-dryer may be used. During the primary drying stage, moisture is removed from food by sublimation. This is accomplished by controlling the vacuum in the freeze-dryer and through heat input. Vacuum control is typically accomplished by maintaining pressure of 1 to 2 torr during freeze-drying. The sublimation of ice to water vapor is facilitated by providing the necessary heat of sublimation (»2.6 to 2.8 MJ/kg at 100°C). This heat may be provided through conduction or radiation, but without raising the temperature of

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

752 the food above its melting point. Primary drying decreases product moisture to approximately 15 to 20%. Secondary drying removes any remaining unfrozen water, while heat is added continuously but at a slower rate. Loss of moisture occurs through diffusion. Freeze-drying has minimal impact on the structure and flavor of the food; hence the product has superior quality compared with that produced by other dehydration techniques. The relatively high cost of freezedrying limits its application to value-added products. Heat-sensitive products such as instant coffee, readyto-eat meals, vegetables, and powders from liquid extracts are prepared by freeze-drying. Freeze-drying may be assisted by the application of microwaves, which may decrease freeze-drying times and energy costs.

Osmotic Dehydration

Osmotic dehydration is a simple procedure that involves removing water from fresh food using hypertonic solutions (56). Sugars (e.g., sucrose), salts (e.g., sodium or calcium chloride), or other humectants (e.g., glycerol) can be used to prepare the hypertonic solutions. The process is called osmotic because it relies on the ability of membranes (that constitute the walls of food cells) to selectively permit the passage of water out of the food without allowing the surrounding solutes to enter these cells. As a result, the food is dehydrated by losing moisture. During most osmotic dehydration, however, a two-way mass exchange takes place: (i) water, with some water-soluble compounds inside the food, diffuses into the surrounding solution; and (ii) the osmotic substance is diffused into the dehydrated tissue from the surrounding solution. The concentration of the hypertonic solution and the solution temperature are important factors that determine the duration of osmotic dehydration and the quality of the dehydrated product. When done correctly, osmotic dehydration retains the food’s natural volatile compounds. Osmotic dehydration may not completely remove all the moisture from food, so it may be used as a pretreatment to other methods of dehydration. Osmotic dehydration is primarily used for dehydrating fruits and vegetables such as apples, blueberries, pineapples, mangoes, apricots, plums, cherries, carrots, and onions.

PRESERVATION BY IONIZING IRRADIATION The electromagnetic spectrum includes many frequencies that can be used in food processing. Gamma radiation, X rays, UV light, and microwave energy are the

SMP_Food Microbiology_CH29.indd

752

most usable components of that spectrum. Recently introduced radiation technologies include radiofrequency and pulsed light. The electromagnetic radiations with shorter wavelengths (e.g., gamma radiation) often provide the most lethal effect on microorganisms. When food is treated with the longer-wavelength radiations (e.g., microwaves), these are converted into heat, which is the direct cause of microbial lethality. In this chapter, microwave energy is covered as a thermal treatment. The ionizing radiation technologies are described briefly, but gamma radiation is emphasized because of its current applicability to food preservation.

X Rays

Ionizing radiation includes X rays and gamma radiation. X rays are produced when an electron beam strikes a metal such as tantalum, tungsten, or stainless steel. Xray sources used to irradiate food are restricted to maximum energies of 5 MeV or less (5, 84). Several food applications of X rays have been developed. Treatment of fresh vegetables with X rays produces substantial microbial lethality (75, 76). Drawbacks of X-ray systems include the high power requirement due to low efficiency and the high cooling requirement. About 80 to 90% of the energy of the electron beam is converted into heat in the metal target. Conversion efficiency depends on the angle of electron impingement, the target metal and its thickness, and the acceleration energy of the electrons (131).

Gamma Radiation

Gamma radiation is emitted by an excited nucleus of a radioisotope, which permits the nucleus to go to its lowest energy or ground state. The radiation sources that can be used in food preservation are cobalt-60 (60Co) and cesium-137 (137Cs). 60Co has a half-life of 5.3 years, whereas 137Cs has a half-life of 30 years. Most commercial irradiators use 60Co as the radiation source. The source is stored underwater, but during food treatment it is contained in concrete-and-steel structures to protect operators. When used on food, the upper energy limit of the gamma-radiation source is restricted to 1.3 MeV (84). There are several advantages of using gamma radiation in food processing. The technology is well developed, the radiation has high penetration power, and the emitted energy is constant over the duration of the food treatment. Unlike other electromagnetic sources, a gamma irradiator is a continuous source that cannot be turned off. At the food irradiator, the radiation source is isolated and secured by the use of thick concrete walls and lead shields. The typical process involves exposing the

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation prepackaged or bulk food material to an ionizing radiation dose for a certain time. The food travels in pallets on a conveyor into and through a chamber where it is exposed to the radiation source. The irradiation dosage can be controlled by adjusting the conveyor speed and the thickness of the packaged product. The radiation dose delivered for killing microorganisms in food is not sufficient to destroy natural enzymes in the product. Thus, to minimize undesirable changes caused by enzymes during extended storage, blanching of vegetables and mild heat treatment for meats may be desired before these products are gamma irradiated (37, 43).

Irradiation Dose and Treatment Efficacy

The radiation energy absorbed by a material is measured in grays (Gy), where 1 Gy represents the absorption of 1 J/kg. The dosage unit often used for food irradiation is the kGy. When a microorganism is exposed to an increasing radiation dose, a logarithmic decrease in its population is often observed (34). This relationship is similar to that described earlier for survivor plots in response to thermal treatments. Similar to heat resistance, the response of a microbial population to an irradiation dose can be represented by the decimal reduction dose (DkGy value), expressed in kGy units. When the dosesurvival curve is a straight line, the DkGy is the negative reciprocal of its slope, DkGy =

Radiation Dose log N0 - log N

where N0 is the initial microorganism population and N is the population surviving the radiation dose. Irradiation can be used to treat a variety of foods, including meat, fish, chicken, nuts, grains, vegetables, spices, poultry, and seafood. Depending upon the intensity of the lethal dose, irradiation can be used to pasteurize or sterilize the food. The following dosage levels are typically used in food preservation depending upon the end objective of the preservation. •

•

•

Radappertization. This involves applying high-dose irradiation (10 to 75 kGy) to produce commercial sterility, an outcome similar to that produced by heat to inactivate bacterial spores in low-acid foods. Radicidation. Medium-dose irradiation (1 to 10 kGy) to produce an effect equivalent to thermal pasteurization. This treatment inactivates vegetative bacteria, but spores are not inactivated. Radurization. This involves application of low irradiation doses (0.05 to 1.00 kGy) to control insects in

SMP_Food Microbiology_CH29.indd

753

753 grains, inhibit sprouting of potatoes, delay fruit ripening, and similar applications. Since radiation is absorbed while a food material passes through an irradiation chamber, packages closer to the source may receive a higher irradiation dose than those that are farther away from the source. The uniformity of dose distribution is expressed as Dmax/Dmin. For radiation-sensitive foods such as chicken, the maximumto-minimum dose ratio should be kept as small as possible (e.g., 1.5), whereas other foods such as onions can tolerate a higher dosage variation. The irradiation dosage received by the food is monitored by means of dosimeters. These are made from irradiation-sensitive materials, such as photographic films. Dosimeters are placed in various packages to determine the dose received by the product during treatment.

Microbiological Principles

The main purpose of gamma irradiation of food is to inactivate spoilage and pathogenic microorganisms. The cell’s DNA is the primary target of ionizing radiation; therefore, the smaller the genome size, generally the greater is the resistance of the microorganism to ionizing radiation (139). Bacteria generally are more radiation resistant than many eukaryotic organisms, including plants and animals. There is considerable variability in radiation resistance among bacterial species (3, 106), probably due to differences in ability to repair DNA damage. Gram-negative bacteria are generally more radiation sensitive than gram-positive bacteria. Vegetative cells of spore-forming bacteria are more sensitive than the spores themselves (139). Deinococcus, Enterococcus, Lactobacillus, and cyanobacteria are some of the radiation-resistant bacteria, whereas Shewanella oneidensis and Pseudomonas putida are very sensitive (28). Deinococcus radiodurans is probably the most radiationresistant bacterium known. It can grow while exposed to 50 Gy of gamma radiation per hour. D. radiodurans can survive up to 10 kGy of radiation, a dose that causes approximately 100 DNA double-strand breaks per genome. On the contrary, S. oneidensis is killed by exposure to 0.07 kGy, a dose that causes less than 1 doublestrand break per genome. Radiation doses that are lethal to the most resistant species can be more than 200 times higher than the doses that are lethal to the most sensitive bacteria (13). Radiation resistance of bacteria and fungi seems to be comparable (111). Compared with bacteria, viruses have greater resistance to radiation (17). Damage to DNA is often cited as the cause of cell death by ionizing radiation (28). This damage is caused by reactive oxygen species (ROS) that are generated

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

754 during radiation. These ROS include hydroxyl radical (·OH) and superoxide anion (O2·–). Bacteria vary in their capacity to contain ROS and ability to repair DNA damage; hence, they exhibit different levels of tolerance to radiation. Recovery from radiation damage is influenced by the mineral contents of the bacterial cell. It was observed that gamma-radiation-resistant bacteria, including D. radiodurans, accumulate high levels of intracellular manganese and low levels of iron compared with radiation-sensitive bacteria. Bacteria with low intracellular manganese and high iron contents encounter more Fe-mediated oxidative damage of proteins during irradiation. Resistance of D. radiodurans to gamma radiation is dependent on the concentration of manganous chloride [Mn(II)]. Therefore, it was proposed that Mn(II) accumulation facilitates recovery from radiation injury. The same research group (28) also noted that D. radiodurans has one of the highest catalase activities reported for any bacterium. These authors hypothesized that catalase serves to remove H2O2 generated by nonenzymatic Mn(II)-based dismutation of superoxide anion. Transcriptional profile analysis of several radiationresistant bacteria revealed that ribosomal proteins, transcription factors, and major chaperones are generally highly expressed in these bacteria (40). Based on these findings, many antioxidant enzymes and proteases play important roles in resisting irradiation. Bacteria may exhibit cross-protection toward gamma radiation and other deleterious factors. Radiation-resistant bacteria can withstand desiccation and other kinds of radiation such as UV light (40).

Factors Influencing Irradiation Efficacy •

•

Process temperature. The lethal effects of irradiation can also be influenced by process temperature. The resistance of microorganisms to irradiation increases with decreasing temperatures (73, 133). This has been attributed to the reduced aw and decreased mobility of free radicals. It is advantageous to irradiate some foods in the frozen state. Freezing minimizes the effect of radiation on chemical changes in food (84). The physical state of water, the product temperature, and the initial freezing point of the food matrix can play a role in determining radiation efficacy (125). Packaging material and headspace. Irradiated foods typically are prepackaged in a polymer material prior to irradiation. The packaging materials themselves are subject to changes after exposure to radiation. Hence, food processors must ensure the integrity of

SMP_Food Microbiology_CH29.indd

754

the packaging material during and after the treatment. Microbial lethality due to irradiation is significantly greater in food exposed to air than in vacuum-packaged or MAP food. The lethal effect of ionizing radiation on microbial cells increases in the presence of oxygen (132).

NONTHERMAL PROCESSING TECHNOLOGIES Processors are interested in identifying minimal preservation technologies to produce foods that are microbiologically safe yet have desirable freshlike quality attributes. Emerging preservation methods that use a lethal treatment other than heat can potentially meet these goals. Examples of such technologies include ultrahigh pressure, pulsed electric field (PEF), pulsed light, and ultrasound. Two of these nonthermal preservation technologies are emphasized in this chapter.

High-Pressure Processing

This technology involves treating food with pressures in the range of 100 to 600 MPa. Treating food at 400 to 600 MPa at or near ambient temperature inactivates a variety of bacteria, yeasts, molds, and viruses, but bacterial spores survive the process. Combining these pressures with temperatures in the range of 90 to 120°C enhances the lethal effect of heat against spores, and hence commercial sterilization is possible. This combination approach is often described as pressure-assisted thermal processing. In-depth reviews summarizing the basics of this technology and its food-processing applications are available (18, 23, 29, 48). Treatment of foods by high-pressure treatment is primarily governed by three basic principles (36). •

•

•

Le Chatelier’s principle. Phenomena such as phase change, molecular reconfiguration, and chemical reaction that are accompanied by a decrease in volume are enhanced by pressure. Isostatic principle. Pressure is transmitted throughout the entire sample in a uniform and quasi-instantaneous manner. Unlike thermal processing, pressure treatment can be considered independent of product volume and geometry. However, some researchers caution that pressure nonhomogeneity can occur, especially during treatment of large solid foods (81). Microscopic ordering principle. Increasing the pressure while maintaining a constant temperature increases the degree of ordering of molecules of a given substance (51). Therefore, it is important to consider the possibility of synergistic or antagonistic reactions

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation

755

when foods are processed by combinations of pressure and heat. Food materials containing air pockets may be deformed by pressure treatment due to differences in compressibility between the air and the food material.

High-Pressure Equipment

High-pressure treatment can be industrially applied in either a batch or semicontinuous mode. The major components of high-pressure processing equipment include (i) a pressure vessel for holding the prepackaged sample (a vessel jacket may be included for temperature control), (ii) closures at the top or bottom of the pressure vessel, (iii) a yoke that retains the top and bottom closures in place during treatment, (iv) pump intensifiers for generating the desired pressure, (v) a monitoring and control system, and (vi) a producthandling device for loading and unloading the food. The vessel can be mounted vertically, horizontally, or in a tilted position, with the configuration of choice depending on the space constraints in the food-processing facility. Water, food-grade propylene glycol, silicone oil, sodium benzoate solutions, ethanol solutions, and castor oil are some of the commonly used pressuretransmitting fluids (134). During batch processing, the product is prepackaged and then preconditioned at the desired initial temperature. The product is then loaded in the pressure vessel and the vessel is closed. The product is processed at the predetermined pressure and temperature conditions for a prespecified holding time, followed by decompression and unloading of the treated product. A semicontinuous system consisting of three vessels can be configured in such a way that while one vessel discharges the processed product, the second vessel is undergoing a compression cycle and the third vessel is being loaded with fresh product (8).

Thermal Behavior of Materials during HighPressure Processing

During application of pressure, compression heating increases the temperature of the treated food. A typical pressure-temperature curve for a food subjected to high-pressure treatment is shown in Fig. 29.6. The compressibility of the substance, thermal properties, initial sample temperature, and target pressure can influence the magnitude of the temperature change under pressure. Among the food constituents, water has the lowest heat of compression (3°C per 100 MPa at 25°C). Nonpolar fats and oils with long-chain fatty acids have higher heat of compression values, i.e., up to 9°C per 100 MPa (92, 107).

SMP_Food Microbiology_CH29.indd

755

Figure 29.6  Typical pressure-temperature history of a highpressure process. P, pressure; T, temperature; t, time. (Adapted from reference 135.) doi:10.1128/9781555818463.ch29f6

Critical Process Parameters

Pressure level, process temperature, and treatment time are the main process parameters that influence the antimicrobial efficacy of high-pressure processing (121). Microbial inactivation by high pressure is influenced by the composition, pH, and aw of the treated food. Certain food constituents (e.g., proteins and lipids) potentially have a protective effect on pressure-treated microorganisms (18). The presence of solutes such as sugars and salts at high concentration reduces the aw of the product; this may cause cell shrinkage and thickening of the cell membrane, thus reducing the cell size and membrane permeability and fluidity. These changes protect microbial cells against the lethal effect of pressure. Reducing aw may also increase the resistance of bacterial spores to pressure by increasing the osmotic dehydration of the spore core (23). High pressure can temporarily alter product pH during the treatment. This may decrease the product’s pH while the product is under pressure (23). For example, the pH of apple juice decreases under pressure by 0.2 units per 100-MPa increase (51).

Process Uniformity

It is commonly assumed that applying pressure to food is quasi-instantaneous and uniform. However, some researchers observed potential pressure nonuniformity, especially in large solid foods (81). A pressure drop of 9 MPa between the surface and product interior of a ham product has been observed. This nonuniformity may

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

756 result in a differential microbial reduction throughout the food, thereby leading to product underprocessing.

Microbial Inactivation Efficacy, Kinetics, and Mechanisms

Resistance to high pressure (i.e., barotolerance) varies considerably among microorganisms. Gram-positive bacteria tend to be more pressure resistant than gramnegative bacteria (121). Within the same species, cells in the stationary phase of growth have a higher pressure resistance than those in the exponential phase (9). Additionally, intraspecies barotolerance variability has been well documented. When 17 E. coli strains were individually pressure treated at 500 MPa, the populations inactivated ranged from 0.6 to 3.4 log CFU/ml (77). Bacterial spores are highly resistant to pressure treatment (89). Inactivation of bacterial spores is possible when heat (often greater than 100°C) is combined with the high-pressure treatment (8). Microorganisms are inactivated by high-pressure treatment in a dose-dependent manner (Fig. 29.7). However, high-pressure-induced inactivation does not always follow first-order kinetics (100). When the log number of survivors is plotted against treatment time, an initial linear decrease is observed, followed by a phase of decreasing killing rate, i.e., tailing (99). The concept of “tailing” kinetics was described earlier in this chapter, under thermal treatments. D-value calculations for food preservation are based on the assumption that first-order reaction kinetics are applicable to microbial inactivation data. Use of D values calculated from the initial linear part

of a nonlinear survival curve can underestimate the treatment time and result in insufficient inactivation (93). Tailing should be taken into account when optimizing high-pressure-processing parameters for food applications. The cell membrane is generally recognized as one of the primary targets of pressure treatment (22). High pressure increases the permeability of cell membranes, which leads to leakage of ATP or UV-absorbing materials, loss of osmotic responsiveness, and increased uptake of fluorescent dyes (14, 122, 126). Pressure decreases cell membrane fluidity by increasing the packing density of lipid molecules and inducing phase separation between lipid and membrane proteins (19, 61). Consistent with this proposition, researchers determined that adaptation of barophilic (i.e., pressureloving) deep-sea bacteria to high pressure involved a shift of membrane lipid composition from saturated to unsaturated fatty acids (151). Another study revealed a relationship between loss of membrane integrity and pressure-mediated lethality in exponential-phase cells, whereas cell membranes resealed after depressurization in stationary-phase cells (96). Analysis of gene transcription in E. coli O157:H7 in response to sublethal pressure revealed that there were more than 100 genes responding to the treatment (77). Some of the genes are involved in stress response, the thioldisulfide redox system, iron-sulfur cluster maintenance, and spontaneous mutation. Pressure treatment may damage proteins containing an iron-sulfur cluster, thereby leading to the release of iron, which mediates oxidative stress.

Figure 29.7  Changes in Listeria monocytogenes populations that were prepared under similar conditions and processed with high pressure (400 MPa and 24 ± 1°C) (left) or pulsed electric field (30 kV/cm and 22°C) (right). (Adapted from reference 83.) doi:10.1128/9781555818463.ch29f7

SMP_Food Microbiology_CH29.indd

756

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation

PEF Processing

PEF processing is designed to treat pumpable food and involves application of a short burst of high voltage (10 to 70 kV/cm) to a product placed between two electrodes. The treatment time is on the order of microseconds to milliseconds. The primary purpose of PEF processing is to inactivate vegetative bacteria in foods. Other potential applications include enhancing diffusion and osmotic dehydration as well as modifying tissue microstructure (95). In-depth reviews on the basics and technology applications of PEF are available (11, 66, 152, 154).

PEF Systems

PEF is primarily designed as a continuous process, but batch systems also are available for use in laboratory research. The major components of continuous PEF equipment include a high-voltage power supply, an energy storage capacitor system, a treatment chamber, a pulse generator, fluid-handling accessories, a cooling system, and monitoring systems (11). Energy from the power supply is stored in a capacitor and discharged through the food material using high-voltage switches to generate the desired electric field. Sets of electrodes deliver the high voltage into the food that is flowing in the treatment chamber. Different treatment chamber designs are reported in the published literature. Parallel, coaxial, and colinear configurations are some of the commonly reported designs (153). Different pulse orientations and shapes can be applied to food. PEF equipment can be designed to deliver monopolar or bipolar pulses that are square or exponential-decay in shape. Square bipolar pulses may be more lethal to treated microbes and more energy efficient than other forms (153). The temperature of the PEF-treated food increases as a result of electrical resistance heating. To minimize any undesirable thermal effect on the treated food, cooling steps are included during and after product treatment. After the product passes through a number of PEF treatment chambers and cooling loops, the product is aseptically packaged and cooled further in preparation for storage.

Critical Process Parameters

Unlike thermal processing, where process temperature and holding time are monitored, the PEF process involves the control of an array of parameters. Some of the PEF processing parameters are related to equipment and others are specific to the food. Equipment parameters include electric field strength, treatment temperature, flow rate or treatment time, pulse shape, pulse width, frequency, and pulse polarity. Food composition, pH, and electrical conductivity are parameters of importance to PEF processing.

SMP_Food Microbiology_CH29.indd

757

757

Factors Influencing Process Uniformity during PEF

Air bubbles are poor conductors of electric fields; these may distort the electric field and create treatment nonuniformity. Additionally, nonhomogenous liquid product may have heterogeneous dielectric properties, which lead to dielectric breakdown. Some of these problems can be eliminated by shortening the pulse duration, applying vacuum degassing to the food prior to treatment, and pressurizing the liquid food during the treatment. Process uniformity may be adversely affected by the heat generated during the application of electric pulses. Heat can influence electrical conductivity, thermal conductivity, and viscosity of the fluid food. Precautions should be taken to avoid electrolysis and potential release of metals from the electrodes into the food materials (124).

Microbial Inactivation Efficacy, Kinetics, and Mechanisms

The efficacy of PEF treatment depends largely on the type of microorganism. Gram-positive bacteria with thicker and more rigid cell envelopes are more resistant to PEF treatment than are gram-negative bacteria (41). Cell size may explain some of the differences in the susceptibility of microorganisms to PEF; lethality is greater for larger than for smaller cells (148). Greater doses of PEF are needed to inactivate rod-shaped cells than spherical cells (46). Bacteria in the stationary phase of growth are more resistant to PEF treatment than those in the exponential phase (148). Comparable lethality has been observed for PEF-treated bacteria and fungi, but bacterial spores are highly resistant to PEF treatment (69). The physical and chemical properties of food influence microbial inactivation by PEF. Food properties with the greatest influence on PEF efficacy are conductivity (ionic strength), pH, and aw. The effects of these factors have been studied on various microorganisms (2, 144, 149). Greater rates of inactivation were observed in lower-conductivity liquid. The influence of pH on PEF-induced microbial inactivation depends on the target microorganism (97). An acidic environment (pH 3.8) enhances the inactivation of L. monocytogenes by PEF treatment. To the contrary, S. enterica serovar Senftenberg was more resistant to PEF treatment at pH 3.8 than at pH 7.0 (148). Low aw generally confers a protective effect against PEF treatment (148). Inactivation of microorganisms by PEF depends on treatment severity. Higher field strength and longer treatment time increase the degree of microbial inactivation. It is desirable to achieve a linear decrease in a microbial population with an increase of process

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

758 intensity, but this first-order inactivation kinetics often is not achieved (Fig. 29.7). It is generally accepted that PEF treatment kills cells primarily by inducing high transmembrane potential (57). Once the potential across the membrane exceeds a critical value (~1 V), pores are formed, leading to increased membrane permeability (136). Intracellular compounds, such as ATP and UV-absorbing substances, are released into the extracellular environment after microorganisms were exposed to a lethal PEF treatment (7). In addition to membranes, cell proteins can be affected by PEF. Expression of molecular chaperones (GroEL, GroES, DnaK, and DnaJ, which are known as heat shock proteins) by L. monocytogenes is suppressed after PEF treatment (70). However, less repression was observed in a PEF-resistant L. monocytogenes strain than in a sensitive strain.

Redefining Pasteurization In response to the successful introduction of a number of alternative processing technologies, the National Advisory Committee on Microbiological Criteria for Foods has developed a new definition for pasteurization. According to the new definition, pasteurization is “any process, treatment, or combination thereof, that is applied to food to reduce the most resistant microorganism(s) of public health significance to a level that is not likely to present a public health risk under normal conditions of distribution and storage” (90). High-pressure and PEF processing, gamma and UV irradiation, and other processes are recognized as technologies that potentially satisfy the new definition of pasteurization.

References   1. Agar, I. T., J. Streif, and F. Bangerth. 1997. Effect of high CO2 and controlled atmosphere (CA) on the ascorbic and dehydroascorbic acid content of some berry fruits. Postharvest Biol. Technol. 11:47–55.   2. Alvarez, I., J. Raso, A. Palop, and F. J. Sala. 2000. Influence of different factors on the inactivation of Salmonella senftenberg by pulsed electric field. Trends Food Sci. Tech. 55:143–146.   3. Anellis, A., D. Berkowitz, and D. Kemper. 1973. Comparative resistance of nonsporogenic bacteria to low-temperature gamma irradiation. Appl. Environ. Microbiol. 25:517–523.   4. Annous, B. A., L. A. Becker, D. O. Bayles, D. P. Labeda, and B. J. Wilkinson. 1997. Critical role of anteiso-C15:0 fatty acid in the growth of Listeria monocytogenes at low temperatures. Appl. Environ. Microbiol. 63:3887–3894.

SMP_Food Microbiology_CH29.indd

758

  5. Anonymous. 2003. General Standard for Irradiated Foods. CODEX Stan 106-1983, Rev. 1-2003. Codex Alimentarius Commission, Rome, Italy. www.codexalimentarius.org/ download/standards/16/CXS_106e.pdf.   6. AquaLab. 2011. Water activity for product safety and quality. Decagon Devices, Inc., Pullman, WA. www. aqualab.com/education/water-activity-for-productsafety-and-quality/.   7. Aronsson, K., U. Rönner, and E. Borch. 2005. Inactivation of Escherichia coli, Listeria innocua and Saccharomyces cerevisiae in relation to membrane permeabilization and subsequent leakage of intracellular compounds due to pulsed electric field processing. Int. J. Food Microbiol. 99:19–32.   8. Balasubramaniam, V. M., and D. Farkas. 2008. Highpressure food processing. Food Sci. Technol. Int. 14:413–418.   9. Balasubramaniam, V. M., E. Y. Ting, C. M. Stewart, and J. A. Robbins. 2004. Recommended laboratory practices for conducting high-pressure microbial inactivation experiments. Innov. Food Sci. Emerg. Technol. 5:299–306.   10. Barbosa-Cánovas, G. V., A. J. Fontana, S. J. Schmidt, and T. P. Labuza (ed.). 2007. Water Activity in Foods: Fundamentals and Applications. Blackwell Publishing, Ames, IA.   11. Barbosa-Cánovas, G. V., M. M. Gongora-Nieto, U. R. Pothkamury, and B. G. Swanson. 1999. Preservation of Foods with Pulsed Electric Fields. Academic Press, New York, NY.   12. Beeby, M., B. D. O’Connor, C. Ryttersgaard, D. R. Boutz, L. J. Perry, and T. O. Yeates. 2005. The geno­ mics of disulfide bonding and protein stabilization in thermophiles. PLoS Biol. 3:e309.   13. Belgian Nuclear Research Centre. 2010. Investigating the molecular mechanisms of radiation resistance in bacteria. Belgian Nuclear Resarch Centre, Brussels, Belgium. https://www.sckcen.be/en/yop/view/293.   14. Benito, A., G. Ventoura, M. Casadei, T. Robinson, and B. Mackey. 1999. Variation in resistance of natural isolates of Escherichia coli O157 to high hydrostatic pressure, mild heat and other stresses. Appl. Environ. Microbiol. 65:1564–1569.   15. Bereksi, N., F. Gavini, T. Benezech, and C. Faille. 2002. Growth, morphology and surface properties of Listeria monocytogenes Scott A and LO28 under saline and acid environments. J. Appl. Microbiol. 92:556–565.   16. Beuchat, L. R., and J. I. Pitt. 2001. Detection and enumeration of heat-resistant molds, p. 217–222. In F. P. Downes and K. Ito (ed.), Compendium of Methods for the Microbiological Examination of Foods. American Public Health Association, Washington, DC.   17. Bidawid, S., J. M. Farber, and S. A. Sattar. 2000. Inactivation of hepatitis A virus (HAV) in fruits and vegetables by gamma irradiation. Int. J. Food Microbiol. 57:91–97.   18. Black, E. P., P. Setlow, A. D. Hocking, C. M. Stewart, A. L. Kelly, and D. G. Hoover. 2007. Response of spores to high-pressure processing. Compr. Rev. Food Sci. Food Safety 6:103–119.

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation   19. Braganza, L. F., and D. L. Worcester. 1986. Structural changes in lipid bilayers and biological membranes caused by hydrostatic pressure. Biochemistry 25:7484–7488.   20. Bringert, G. 2004. Thermal sterilization process validation ritical to product safety. September:22–26. www. pharmpro.com/PDFs/pp49gei.pdf.   21. Brown, A. D. 1976. Microbial water stress. Bacteriol. Rev. 40:803–846.   22. Casadei, M. A., P. Mañas, G. Niven, E. Needs, and B. M. Mackey. 2002. Role of membrane fluidity in pressure resistance of Escherichia coli NCTC 8164. Appl. Environ. Microbiol. 68:5965–5972.   23. Cheftel, J. C. 1995. Review: high pressure, microbial inactivation and food preservation. Food Sci. Technol. Int. 1:75–90.   24. Church, I. J., and A. L. Parsons. 1995. Modified atmosphere packaging technology—a review. J. Sci. Food Agric. 67:143–152.   25. Code of Federal Regulations. 2011. Poultry Products Inspection Regulations. Title 9, Part 381:66. Office of the Federal Register, Washington, DC.   26. Code of Federal Regulations. 2011. Thermally Processed Low-Acid Foods Packaged in Hermetically Sealed Containers. Title 21, Part 113. Office of the Federal Register, Washington, DC.   27. Condon, S., and F. J. Sala. 1992. Heat resistance of Bacillus subtilis in buffer and foods of different pH. J. Food Prot. 55:605–608.   28. Daly, M. J., E. K. Gaidamakova, V. Y. Matrosova, A. Vasilenko, M. Zhai, A. Venkateswaran, M. Hess, M. V. Omelchenko, H. M. Kostandarithes, K. S. Makarova, L. P. Wackett, J. K. Fredrickson, and D. Ghosal. 2004. Accumulation of Mn(II) in Deinococcus radiodurans facilitates gamma-radiation resistance. Science 306:1025–1028.   29. Doona, C. J., and F. E. Freeherry (ed.). 2007. High Pressure Processing of Foods. Wiley-Blackwell, Hoboken, NJ.   30. Drummond, L., and D.-W. Sun. 2010. Effect of chilling and freezing on safety and quality of food products, p. 295–322. In E. Ortega-Rivas (ed.), Processing Effects on Safety and Quality of Foods. CRC Press, New York, NY.   31. Earle, R. L., and M. D. Earle. 1983. Unit Operations in Food Processing, Web edition. New Zealand Institute of Food Science and Technology, Palmerston North, New Zealand. www.nzifst.org.nz/unitoperations/.   32. Edgey, M., and A. D. Brown. 1978. Response of xerotolerant and nontolerant yeasts to water stress. J. Gen. Microbiol. 104:343–345.   33. Eichner, K., and M. Karel. 1972. Influence of water content and water activity on the sugar-amino browning reaction in model systems under various conditions. J. Agric. Food Chem. 20:218–223.   34. El-Shenawy, M. A., A. E. Yousef, and E. H. Marth. 1989. Radiation sensitivity of Listeria monocytogenes in broth or in raw ground beef. Lebenson. Wiss. Technol. 22:378–390.   35. Erichsen, I., and G. Molin. 1981. Microbial flora of normal and high pH beef stored at 4°C in different gas environments. J. Food Prot. 44:866.

SMP_Food Microbiology_CH29.indd

759

759   36. Farkas, D. F., and D. G. Hoover. 2000. High pressure processing. J. Food Sci. 65(Suppl.):47S–64S.   37. Fellows, P. J. 2002. Food Processing Technology: Principles and Practice, 2nd ed. Woodhead Publishing, Ltd., Boca Raton, FL.   38. Food Safety and Inspection Service. 2010. Safe food handling: freezing and food safety. Food Safety and Inspection Service, U.S. Department of Agriculture, Washington, DC. www.fsis.usda.gov/factsheets/focus_ on_freezing/index.asp#3.   39. Gamble, H. R. 2011. Trichinae: pork facts—food quality and safety. Animal and Plant Health Inspection Service, U.S. Department of Agriculture, Washington, DC. www. aphis.usda.gov/vs/trichinae/docs/fact_sheet.htm.   40. Gao, N., B.-G. Ma, Y.-S. Zhang, Q. Song, L.-L. Chen, and H.-Y. Zhang. 2009. Gene expression analysis of four radiation-resistant bacteria. Genomics Insights 2:11–22.   41. García, D., N. Gómez, J. Raso, and R. Pagán. 2005. Bacterial resistance after pulsed electric fields depending on the treatment medium pH. Innov. Food Sci. Emerg. Technol. 6:388–395.   42. Gill, C. O., and G. Molin. 1991. Modified atmospheres and vacuum packaging, p. 172–199. In N. J. Russell and G. W. Gould (ed.), Food Preservatives. Blackie, Glasgow, United Kingdom.   43. Goresline, H. E., and N. W. Desrosier. 1959. Preservation of foods by food irradiation. Am. J. Public Health 49:488–492.   44. Gould, G. W. 1989. Drying, raised osmotic pressure and low water activity, p. 97–117. In G. W. Gould (ed.), Mechanisms of Action of Food Preservation Procedures. Elsevier Applied Science, London, United Kingdom.   45. Gould, G. W. 2000. Emerging technologies in food preservation and processing in the last 40 years, p. 1–11. In G. V. Barbosa-Cánovas and G. W. Gould (ed.), Innovations in Food Processing. Technomic Publishing Co., Inc., Lancaster, PA.   46. Heinz, V., I. Álvarez, A. Angersbach, and D. Knorr. 2002. Preservation of liquid foods by high intensity pulsed electric fields-basic concepts for process design. Trends Food Sci. Technol. 12:103–111.   47. Heldman, D. R., and R. W. Hartel. 1998. Principles of Food Processing. Aspen Publishers, Gaitherburg, MD.   48. Hendrickx, M. E. G., and D. Knorr. 2001. Ultra High Pressure Treatments of Food. Kluwer Academic/Plenum Publishers, New York, NY.   49. Hengge, R. 2008. The two-component network and the general stress sigma factor RpoS (sigma S) in Escherichia coli. Adv. Exp. Med. Biol. 631:40–53.   50. Herbert, R. A., and J. P. Sutherland. 2000. Chill storage, p.101–121. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Food, vol. 1. Aspen Publishers, Gaithersburg, MD.   51. Heremans, K. 1995. High pressure effects on biomolecules, p. 81–98. In D. A. Ledward, D. E. Johnston, R. G. Earnshaw, and A. P. M. Hasting (ed.), High Pressure Processing of Foods. Nottingham University Press, Loughborough, United Kingdom.

Manila Typesetting Company

11/08/2012 07:11AM

Preservatives and Preservation Methods

760   52. Herendeen, S. L., R. A. Vanbogelen, and F. C. Neidhardt. 1979. Levels of major proteins of Escherichia coli during growth at different temperatures. J. Bacteriol. 139:185–194.   53. Hollywood, N. W., Y. Varabioff, and G. E. Mitchell. 1991. The effect of microwave and conventional cooking on the temperature profiles and microbial flora of minced beef. Int. J. Food Microbiol. 14:67–76.   54. Holsinger, V. H., K. T. Rajkowski, and J. R. Stabel. 1997. Milk pasteurisation and safety: a brief history and update. Rev. Sci. Tech. 16:441–451.   55. Hubalek, Z. 2003. Protectants used in the cryopreservation of microorganisms. Cryobiology 46:205–229.   56. Hui, Y. H., C. Clary, M. M. Farid, O. O. Fasina, A. Noomhorm, and J. Welti-Chanes. 2008. Food Drying Science and Technology: Microbiology, Chemistry and Applications. DEStech Publications, Inc., Lancaster, PA.   57. Hülsheger, H., J. Potel, and E. G. Niemann. 1983. Electric field effects on bacteria and yeast cells. Radiat. Environ. Biophys. 22:149–162.   58. Ingram, M., and B. M. Mackey. 1976. Inactivation by cold, p. 111–151. In F. A. Skinner and W. B. Hugo (ed.), Inhibition and Inactivation of Vegetative Microbes. Society of Applied Bacteriology Symposia Series, no. 5. Academic Press, New York, NY.   59. Institute of Food Technologists. 2000. Kinetics of microbial inactivation for alternative food processing technologies. A report of the Institute of Food Technologists for the Food and Drug Administration of the U.S. Department of Health and Human Services, IFT/FDA Contract No. 223-98-2333. Food and Drug Administration, Silver Spring, MD. www.fda.gov/Food/ ScienceResearch/ResearchAreas/SafePracticesforFood Processes/ucm100158.htm.   60. International Commission on Microbiological Specifications for Foods. 1996. Microorganisms in Foods 5. Microbiological Specifications of Food Pathogens. Blackie Academic & Professional, London, United Kingdom.   61. Ishii, A., T. Oshima, T. Sato, K. Nakasone, H. Mori, and C. Kato. 2005. Analysis of hydrostatic pressure effects on transcription in Escherichia coli by DNA microarray procedure. Extremophiles 9:65–73.   62. Isom, L. L., K. S. Khambatta, J. L. Moluf, D. F. Akers, and S. E. Martin. 1995. Filament formation of Listeria monocytogenes. J. Food Prot. 58:1031–1033.   63. James, C., C. Vincent, T. I. de Andrade Lima, and S. J. James. 2006. The primary chilling of poultry carcasses, a review. Int. J. Refrig. 29:847–862.   64. Kaplan, M. M., M. Abdussalam, and G. Bijlenga. 1962. Diseases transmitted through milk, p. 11–76. In World Health Organization (ed.), Milk Hygiene. World Health Organization monograph series no. 48. World Health Organization, Geneva, Switzerland.   65. Karel, M., and D. B. Lund. 2003. Physical Principals of Food Preservation, 2nd ed. Marcel Dekker, New York, NY.   66. Knorr, D., M. Geulen, T. Grahl, and W. Sitzmann. 1994. Food application of high electric field pulses. Trends Food Sci. Technol. 5:71–75.

SMP_Food Microbiology_CH29.indd

760

  67. Labuza, T. P., A. Kaanane, and J. Y. Chen. 1985. Effect of temperature on the moisture sorption isotherms and water activity shift of two dehydrated foods. J. Food Sci. 50:385–391.   68. Labuza, T. P., N. D. Heidelbaugh, M. Silver, and M. Karel. 1971. Oxidation at intermediate moisture contents. J. Am. Oil Chem. Soc. 48:86–90.   69. Lado, B. H., and A. E. Yousef. 2002. Alternative food preservation technologies: efficacy and mechanisms. Microb. Infect. 4:433–440.   70. Lado, B. H., J. A. Bomser, C. P. Dunne, and A. E. Yousef. 2004. Pulsed electric field alters molecular chaperone expression and sensitizes Listeria monocytogenes to heat. Appl. Environ. Microbiol. 70:2289–2295.   71. Leaper, S., and P. Richardson. 1999. Validation of thermal process control for the assurance of food safety. Food Control 10:281–283.   72. Lin, C.-C., B.-K. Wu, and D.-K. Lin. 1968. Spoilage bacteria in canned foods: I. Flat sour spoilage bacteria in canned asparagus and the thermal death time. Appl. Microbiol. 16:45–47.   73. López-González, V., P. S. Murano, R. E. Brennan, and E. A. Murano. 1999. Influence of various commercial packaging conditions on survival of Escherichia coli O157:H7 to irradiation by electron beam versus gamma rays. J. Food Prot. 62:10–15.   74. Mackey, B. M., C. A. Miles, S. E. Parsons, and D. A. Seymour. 1991. Thermal denaturation of whole cells and cell components of Escherichia coli examined by differential scanning calorimetry. J. Gen. Microbiol. 137:2361–2374.   75. Mahmoud, B. S. 2010. The effects of X-ray radiation on Escherichia coli O157:H7, Listeria monocytogenes, Salmonella enterica and Shigella flexneri inoculated on whole Roma tomatoes. Food Microbiol. 27:1057–1063.   76. Mahmoud, B. S., G. Bachman, and R. H. Linton. 2010. Inactivation of Escherichia coli O157:H7, Listeria monocytogenes, Salmonella enterica and Shigella flexneri on spinach leaves by X-ray. Food Microbiol. 27:24–28.   77. Malone, A. S., Y.-K. Chung, and A. E. Yousef. 2006. Genes of Escherichia coli O157:H7 that are involved in high-pressure resistance. Appl. Environ. Microbiol. 72:2661–2671.   78. Marth, E. H. 1998. Extended shelf life refrigerated foods: microbiological quality and safety. Food Technol. 52:57–62.   79. Mattick, K. L., F. Jorgensen, J. D. Legan, M. B. Cole, J. Porter, H. M. Lappin-Scott, and T. J. Humphrey. 2000. Survival and filamentation of Salmonella enterica serovar Enteritidis PT4 and Salmonella enterica serovar Typhimurium DT104 at low water activity. Appl. Environ. Microbiol. 66:1274–1279.   80. McLean, R. J. C., and M. A. C. McLean. 2010. Microbial survival mechanisms and the interplanetary transfer of life through space. J. Cosmol. 7:1802–1820.   81. Minerich, P. L., and T. P. Labuza. 2003. Development of a pressure indicator for high hydrostatic pressure processing of foods. Innov. Food Sci. Emerg. Technol. 4:235–243.

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation   82. Mohácsi-Farkas, C., J. Farkas, L. Mészáros, O. Reichart, and E. Andrassy. 1999. Thermal denaturation of bacterial cells examined by differential scanning calorimetry. J. Therm. Anal. Calorim. 57:409–414.   83. Mohamed, H. M. H., B. H. S. Diono, and A. E. Yousef. Structural changes in Listeria monocytogenes treated with gamma radiation, pulsed electric field and ultrahigh pressure. J. Food Safety, in press.   84. Molins, R. A. (ed.). 2001. Food Irradiation: Principles and Applications. John Wiley & Sons, New York, NY.   85. Montville, T. J., and K. R. Matthews. 2008. Food Microbiology: an Introduction, 2nd ed. ASM Press, Washington, DC.   86. Morales-Castro, J., and L. A. Ochoa-Martinez. 2010. Safety and quality effects in food stored under modified atmosphere conditions, p. 253–294. In E. Ortega-Rivas (ed.), Processing Effects on Safety and Quality of Foods. CRC Press, New York, NY.   87. Mudgett, R. E. 1985. Dielectrical properties of foods, p. 15–37. In R. V. Decareau (ed.), Microwaves in the Food Processing Industry. Academic Press, Orlando, FL.   88. Multon, J. L., and H. Bizot. 1978. Intermediate moisture foods and water activity determination. Ann. Nutr. Aliment. 32:631–654. (In French.)   89. Nakayama, A. Y. Yano, S. Kobayashi, M. Ishikawa, and K. Sakai. 1996. Comparison of pressure resistances of spores of six Bacillus strains with their heat resistances. Appl. Environ. Microbiol. 62:3897–3900.   90. National Advisory Committee on Microbiological Criteria for Foods. 2006. Requisite scientific parameters for establishing the equivalence of alternative methods of pasteurization. J. Food Prot. 69:1190–1216.   91. National Food Processors Association. 1982. Thermal processes for low-acid foods in metal food containers. Bulletin 26-L, 12th ed. National Food Processors Association, Washington, DC.   92. Nguyen, L. T., and V. M. Balasubramaniam. 2011. Fundamentals of food processing using high pressure, p. 3–19. In H. Q. Zhang, G. V. Barbosa-Cánovas, V. M. Balasubramaniam, C. P. Dunne, D. F. Farkas, and J. T. C. Yuan (ed.), Nonthermal Processing Technologies for Food. IFT Press, Wiley-Blackwell, West Sussex, United Kingdom.   93. Noma, S., D. Kajiyama, N. Igura, M. Shimoda, and I. Hayakawa. 2006. Mechanisms behind tailing in the pressure inactivation curve of a clinical isolate of Escherichia coli O157:H7. Int. J. Food Microbiol. 109:103–108.   94. Oliveira, M. E. C., and A. S. Franca. 2002. Microwave heating of foodstuffs. J. Food Eng. 53:347–349.   95. Ortega-Rivas, E. 2011. Critical issues pertaining to application of pulsed electric fields in microbial control and quality of processed fruit juices. Food Bioprocess. Technol. 4:631–645.   96. Pagán, R., and B. Mackey. 2000. Relationship between membrane damage and cell death in pressure-treated Escherichia coli cells: differences between exponentialand stationary-phase cells and variation among strains. Appl. Environ. Microbiol. 66:2829–2834.

SMP_Food Microbiology_CH29.indd

761

761   97. Pagán, R., S. Condon, and J. Raso. 2005. Microbial inactivation by pulsed electric fields, p. 45–68. In G. V. BarbosaCánovas, M. S. Tapia, and M. P. Cano (ed.), Novel Food Processing Technologies. CRC Press, Boca Raton, FL.   98. Palaniappan, S., S. K. Sastry, and E. R. Richter. 1992. Effects of electroconductive heat treatment and electrical pretreatment on thermal death kinetics of selected microorganisms. Biotechnol. Bioeng. 39:225–232.   99. Patterson, M. F. 2005. A review: microbiology of pressuretreated foods. J. Appl. Microbiol. 98:1400–1409. 100. Patterson, M. F., M. Quinn, R. Simpson, and A. Gilmour. 1995. Sensitivity of vegetative pathogens to high hydrostatic pressure treatment in phosphate-buffered saline and foods. J. Food Prot. 58:524–529. 101. Pflug, I. J. 1998. Microbiology and Engineering of Sterilization Processes, 9th ed. Environmental Sterilization Laboratory, Minneapolis, MN. 102. Pitt, J. I. 1975. Xerophilic fungi and the spoilage of food of plant origin, p. 273–307. In R. B. Duckworth (ed.), Water Relations of Foods. Academic Press, London, United Kingdom. 103. Pitt, J. I., and B. F. Miscamble. 1995. Water relations of Aspergillus flavus and closely related species. J. Food Prot. 58:86–90. 104. Pitt, J. I., and A. D. Hocking. 2009. Fungi and Food Spoilage. Springer, New York, NY. 105. Qi, Y., and K. J. Miller. 2000. Effect of low water activity on staphylococcal enterotoxin A and B biosynthesis. J. Food Prot. 63:473–478. 106. Rainey, F. A., K. Ray, M. Ferreira, B. Z. Gatz, M. F. Nobre, D. Bagaley, B. A. Rash, M.-J. Park, A. M. Earl, N. C. Shank, A. M. Small, M. C. Henk, J. R. Battista, P. Kämpfer, and M. S. da Costa. 2005. Extensive diversity of ionizing-radiation-resistant bacteria recovered from Sonoran desert soil and description of nine new species of the genus Deinococcus obtained from a single soil sample. Appl. Environ. Microbiol. 71:5225–5235. 107. Rasanayagam, V., V. M. Balasubramaniam, E. Ting, C. E. Sizer, C. Bush, and C. Anderson. 2003. Compression heating of selected fatty food materials during high-pressure processing. J. Food Sci. 68:254–259. 108. Roberts, T. A., G. Hobbs, J. H. B. Christian, and N. Skovgaard. 1981. Psychrotrophic Microorganisms in Spoilage and Pathogenicity. Academic Press, London, United Kingdom. 109. Rose, D. 1995. Advances and potential for aseptic processing, p. 283–303. In G. W. Gould (ed.), New Methods of Food Preservation. Blackie Academic and Professional, Glasgow, United Kingdom. 110. Ryser, E. T. 2001. Public health concerns, p. 397–545. In E. H. Marth and J. L. Steele (ed.), Applied Dairy Microbiology, 2nd ed. Marcel Dekker, New York, NY. 111. Saleh, Y. G., M. S. Mayo, and D. G. Ahearn. 1988. Resistance of some common fungi to gamma irradiation. Appl. Environ. Microbiol. 54:2134–2135. 112. Saravacos, G. D., and A. E. Kostaropoulos. 2002. Handbook of Food Processing Equipment. Kluwer Academic/Plenum Publishers, New York, NY.

Manila Typesetting Company

11/08/2012 07:11AM

762

Preservatives and Preservation Methods

113. Sastry, S. K. 2008. Ohmic heating and moderate electric field processing. Food Sci. Technol. Int. 14:419–422. 114. Sastry, S. K., and B. D. Cornelius. 2002. Aseptic Processing of Foods Containing Solid Particulates. John Wiley & Sons, New York, NY. 115. Sastry, S. K., A. Yousef, H. Y. Cho, S. R. Unal, S. Salenghe, W. C. Wang, M. Lima, S. Kulshrstha, P. Wongsa-Ngasri, and I. Senoy. 2002. Ohmic heating and moderate electric field (MEF) processing, p. 785– 793. In J. Welti-Chanes, G. V. Barbosa-Cánovas, and J. M. Aguilera (ed.), Engineering and Food for the 21st Century. CRC Press, Boca Raton, FL. 116. Scott, W. J. 1953. Water relations of Staphylococcus aureus at 30°C. Aust. J. Biol. Sci. 6:549–564. 117. Scott, W. J. 1957. Water relations of food spoilage microorganisms. Adv. Food Res. 7:83–127. 118. Setlow, B., and P. Setlow. 1995. Small, acid-soluble proteins bound to DNA protect Bacillus subtilis spores from killing by dry heat. Appl. Environ. Microbiol. 61:2787–2790. 119. Sharma, S. K., P. Christen, and P. Goloubinoff. 2009. Disaggregating chaperones: an unfolding story. Curr. Protein Peptide Sci. 10:432–446. 120. Singh, R. P., and D. R. Heldman. 2009. Introduction to Food Engineering, 4th ed. Academic Press, Burlington, MA. 121. Smelt, J. P. 1998. Recent advances in the microbiology of high pressure processing. Trends Food Sci. Technol. 9:152–158. 122. Smelt, J. P., A. G. Rijke, and A. Hayhurst. 1994. Possible mechanisms of high-pressure inactivation of microorganisms. High Pressure Res. 12:199–203. 123. Sober, J. 2005. Botulism. Clin. Infect. Dis. 41:1167–1173. 124. Sobrino-López, A., and O. Martín-Belloso. 2010. Potential of high-intensity pulsed electric field technology for milk processing. Food Eng. Rev. 2:17–27. 125. Sommers, C., and X. Fan. 2011. Irradiation of ground beef and fresh produce, p. 236–248. In H. Q. Zhang, G. V. Barbosa-Cánovas, V. M. Balasubramaniam, C. P. Dunne, D. F. Farkas, and J. T. C. Yuan (ed.), Nonthermal Processing Technologies for Food. IFT Press, WileyBlackwell, West Sussex, United Kingdom. 126. Somolinos, M., D. García, R. Pagán, and B. Mackey. 2008. Relationship between sublethal injury and microbial inactivation by the combination of high hydrostatic pressure and citral or tert-butyl hydroquinone. Appl. Environ. Microbiol. 74:7570–7577. 127. Splittstoesser, D. F., and C. M. Splittstoesser. 1977. Ascospores of Byssochlamys fulva compared with those of a heat resistant Aspergillus. J. Food Sci. 42:685–688. 128. Splittstoesser, D. F., S. B. Leasor, and K. M. J. Swanson. 1986. Effect of food composition on the heat resistance of yeast ascospores. J. Food Sci. 51:1265–1267. 129. Stumbo, C. R. 1965. Thermobacteriology in Food Processing. Academic Press, New York, NY. 130. Teixeira, A. A. 1992. Thermal process calculations, p. 563– 619. In D. R. Heldman and D. B. Lund (ed.), Handbook of Food Engineering. Marcel Dekker, New York, NY. 131. Thayer, D. W. 2003. Ionizing irradiation, treatment of food. In D. R. Heldam (ed.), Encyclopedia of Agricultural,

Food, and Biological Engineering. Marcel Dekker, New York, NY. Thayer, D. W., and G. Boyd. 1999. Irradiation and modified atmosphere packaging for the control of Listeria monocytogenes on turkey meat. J. Food Prot. 62:1136–1142. Thayer, D. W., G. Boyd, L. Lakritz, and J. W. Hampson. 1995. Variations in radiation sensitivity of foodborne pathogens associated with the suspending meat. J. Food Sci. 60:63–67. Ting, E. 2011. High pressure processing equipment fundamentals, p. 20–27. In H. Q. Zhang, G. V. BarbosaCánovas, V. M. Balasubramaniam, C. P. Dunne, D. F. Farkas, and J. T. C. Yuan (ed.), Nonthermal Processing Technologies for Food. IFT Press, Wiley-Blackwell, West Sussex, United Kingdom. Ting, E., V. M. Balasubramaniam, and E. Raghubeer. 2002. Determining thermal effects in high-pressure processing. Food Technol. 56:31–35. Tsong, T. Y. 1990. Review on electroporation of cell membranes and some related phenomena. Bioelectrochem. Bioenerg. 24:271–295. Tsuchido, T., N. Katsui, A. Takeuchi, M. Takano, and I. Shibasaki. 1985. Destruction of the outer membrane permeability barrier of Escherichia coli by heat treatment. Appl. Environ. Microbiol. 50:298–303. Tucker, G. S. 2001. Validation of heat processes, p .75–90. In P. Richardson (ed.), Thermal Processing Technologies in Food Processing. Woodhead Publications, New York, NY. Urbain, W. M. 1986. Food Irradiation. Academic Press, Orlando, FL. U.S. Food and Drug Administration. 2000. Kinetics of microbial inactivation for alternative food processing technologies—glossary. Food and Drug Administration, Silver Spring, MD. www.fda.gov/food/scienceresearch/ researchareas/safepracticesforfoodprocesses/ucm105794. htm. U.S. Food and Drug Administration. 2009. Process validation. Food and Drug Administration, Silver Spring, MD. www.fda.gov/MedicalDevices/DeviceRegulationandGuid ance/PostmarketRequirements/QualitySystemsRegulation s/MedicalDeviceQualitySystemsManual/ucm122439.htm. U.S. Food and Drug Administration. 2010. Draft guidance for industry: acidified foods. Food and Drug Administration, Silver Spring, MD. www.fda.gov/Food/GuidanceCom plianceRegulatoryInformation/GuidanceDocuments/ AcidifiedandLow-AcidCannedFoods/ucm222618. htm#III-A. U.S. Pharmacopeia. 2011. Biological indicators for sterilization. U.S. Pharmacopeia, Rockville, MD. www. pharmacopeia.cn/v29240/usp29nf24s0_c1035.html. Vega-Mercado, H., U. R. Pothakamury, F.-J. Chang, G. V. Barbosa-Cánovas, and B. G. Swanson. 1996. Inactivation of Escherichia coli by combining pH, ionic strength and pulsed electric fields hurdles. Food Res. Int. 29:117–121. Waite-Cusic, J. G., B. H. Diono, and A. E. Yousef. 2011. Screening for Listeria monocytogenes surrogate strains

SMP_Food Microbiology_CH29.indd

762

132.

133.

134.

135. 136. 137.

138.

139. 140.

141.

142.

143. 144.

145.

Manila Typesetting Company

11/08/2012 07:11AM

29.  Physical Methods of Food Preservation

146. 147. 148.

149.

150.

151.

applicable to food processing by ultrahigh pressure and pulsed electric field. J. Food Prot. 74:1655–1661. Warth, A. D. 1978. Relationship between the heatresistance of spores and the optimum and maximum growth temperatures of Bacillus species. J. Bacteriol. 134:699–705. White, H. 1953. The heat resistance of Streptococcus faecalis. J. Gen. Microbiol. 8:27–37. Wouters, P. C., A. P. Bos, and J. Ueckert. 2001. Membrane permeabilization in relation to inactivation kinetics of Lactobacillus species due to pulsed electric fields. Appl. Environ. Microbiol. 67:3092–3101. Wouters, P. C., N. Dutreux, J. P. Smelt, and H. L. Lelieveld. 1999. Effects of pulsed electric fields on inactivation kinetics of Listeria innocua. Appl. Environ. Microbiol. 65:5364–5371. Yamamori, T., and T. Yura. 1982. Genetic control of heat-shock protein synthesis and its bearing on growth and thermal resistance in Escherichia coli K-12. Proc. Natl. Acad. Sci. USA 79:860–864. Yano, Y., A. Nakayama, K. Ishihara, and H. Satio. 1998. Adaptive changes in membrane lipids of baro-

SMP_Food Microbiology_CH29.indd

763

763

152.

153. 154.

155.

philic bacteria in response to changes in growth pressure. Appl. Environ. Microbiol. 64:479–485. Yousef, A. E., and Q. H. Zhang. 2006. Microbiological and safety aspects of pulsed electric field technology, p. 152–166. In V. K. Juneja, J. P. Cherry, and M. H. Tunick (ed.), Advances in Microbiological Food Safety. American Chemical Society, Washington, DC. Zhang, Q., G. V. Barbosa-Cánovas, and B. G. Swanson. 1995. Engineering aspects of pulsed electric fields pasteurization. J. Food Eng. 25:261–281. Zhang, H. Q., G. V. Barbosa-Cánovas, V. M. Balasubra­ maniam, C. P. Dunne, D. F. Farkas, and J. T. C. Yuan (ed.). 2011. Nonthermal Processing Technologies for Food. IFT Press, Wiley-Blackwell, West Sussex, United Kingdom. Zhuang, H., M. M. Barth, and T. R. Hankinson. 2003. Microbial safety, quality, and sensory aspects of freshcut fruits and vegetables, p. 255–278. In J. S. Novak, G. M. Sapers, and V. K. Juneja (ed.), Microbial Safety of Minimally Processed Foods. CRC Press, Boca Raton, FL.

Manila Typesetting Company

11/08/2012 07:11AM

SMP_Food Microbiology_CH29.indd

764

Manila Typesetting Company

11/08/2012 07:11AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch30

P. Michael Davidson T. Matthew Taylor Shannon E. Schmidt

30

Chemical Preservatives and Natural Antimicrobial Compounds

The overall quality of food products diminishes from the time of harvest or slaughter until they are consumed. Quality loss may result from microbiological, enzymatic, chemical, or physical changes. The consequences of microbiological changes include hazards to consumers because of the presence of microbial toxins or pathogenic microorganisms or economic losses through growth of spoilage microorganisms and resultant offodors, off-flavors, texture problems, discoloration, slime, or haze. Food preservation technologies, some in use since ancient times, protect foods from the deleterious effects of microbial growth and inherent deterioration. Microorganisms can be inhibited or inactivated by heating, chilling, freezing, reduction of water activity (aw), nutrient restriction, acidification, modification of packaging atmosphere, fermentation, or nonthermal treatments (e.g., high pressure, irradiation) or through addition of antimicrobial compounds. Food antimicrobials are chemicals added to or present in foods that retard growth of, or kill, microorganisms. Most food antimicrobials are bacteriostatic or fungistatic at use concentrations and not bactericidal or fungicidal. Therefore, food antimicrobials do not preserve a

food indefinitely. Food antimicrobials are often used in combination with other preservation procedures. Their targets are pathogenic and spoilage microorganisms. Food antimicrobials are legally defined as “preservatives,” although the term can also include other additives that function to preserve the food in question, including antibrowning agents (e.g., citric acid) and antioxidants (e.g., butylated hydroxyanisole) [21 CFR 101.22(a)(5), 21 CFR 70.3(o)(2)]. In this chapter, antimicrobial compounds are divided into two classes: traditional and naturally occurring. Antimicrobials are classified as traditional when they (i) have been used for many years, (ii) are approved by many countries for inclusion as antimicrobials in foods (e.g., lysozyme and lactoferrin, which occur naturally but are regulatory-agency approved), or (iii) are produced by synthetic processes (as opposed to natural extracts; e.g., large-scale fermentation of the bacteriocin nisin from Lactococcus lactis subsp. lactis). Ironically, many synthetic traditional antimicrobials are found in nature; examples include acetic acid from vinegar, benzoic acid from cranberries, and sorbic acid from mountain ash tree berries (rowanberries).

P. Michael Davidson, Department of Food Science and Technology, University of Tennessee, Knoxville, TN 37996-4591. T. Matthew Taylor, Department of Animal Science, Texas A&M University, College Station, TX 77843-2471. Shannon E. Schmidt, Pecan Deluxe Candy Company, Dallas, TX 75212-6308.

765

SMP_Food Microbiology_CH30.indd

765

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

766

FACTORS AFFECTING ACTIVITY The efficacy of food antimicrobials depends on multiple factors related to the food product, its processing and postprocess storage environment, and the microorganisms targeted. Food preservation is best achieved when the microorganisms to be inhibited, antimicrobial type and concentration, food storage conditions, food pH and buffering capacity, and presence of other agents that affect shelf life are known and taken into account. Gould (171) classified factors that affect antimicrobial activity into microbial, intrinsic, extrinsic, and process related. Davidson and Branen (107) subdivided the microbial factors into those associated with the micro­ organisms targeted for inhibition and those related to the antimicrobial compound’s physicochemistry. Microbial factors that affect antimicrobial activity include inherent resistance (vegetative cells versus endospores; strain differences), initial number and growth rate, interactions with other microorganisms (e.g., antagonism, quorum sensing), cellular structure (Gram reaction), life cycle and status (logarithmic growth, sublethal injury), and ability to secrete capsular material and/or form biofilms. Intrinsic factors affecting anti­microbial activity are those associated with a food’s physicochemistry and include its physical structure and the presence of physical barriers to invasion (e.g., membranes, shells, husks, hides); nutrient content and availability (including water and by correlation aw); pH and buffering capacity; oxidation/reduction (redox) potential; and anti­ microbials and other compounds that may interact with antimicrobials, resulting in synergistic or antagonistic effects on microbial inhibition. Extrinsic factors include the temperature and duration of food storage, the gaseous atmosphere and relative humidity of the storage, and the packaging environment. Process-related factors include changes in food composition, shifts in microbiota, changes in microbial populations, and changes in food microstructure and/or moisture that result from processing (e.g., thermal processing and resultant shifts from mixed microbial populations to populations of heat-resistant microorganisms [thermophilic sporeformers, thermoduric organisms, etc.], dehydration). Most of these factors influence antimicrobial efficacy in an interactive manner, indicating that while some will enhance the activity of an antimicrobial, others will detract from it. pH is potentially the most important intrinsic factor that influences the effectiveness of many food anti­microbials, including the organic acids and the antimicrobial peptides (e.g., nisin), although not always through similar mechanisms. Weak organic acids and fatty acids are most effective as food antimicrobials in their undissociated/protonated state. This is because weak acids are

SMP_Food Microbiology_CH30.indd

766

better able to diffuse through the cytoplasmic membrane of a microorganism in their protonated form, a result of the acid species bearing no charge and greater relative hydrophobicity. Another major factor that affects antimicrobial activity is polarity (54). This relates both to the ionization of the molecule and the existence of alkyl side groups or other hydrophobic parent molecules on the antimicrobial. Antimicrobials must be at least partially hydrophobic to attach and pass through the cell membrane but also possess some degree of polarity to solubilize in the aqueous phase in which the microbes exist (329).

TRADITIONAL ANTIMICROBIALS

Organic Acids and Derivatives

Many organic acids are used as food additives, though not all possess antimicrobial activity. Research suggests that the most active are the monoprotic acids, including acetic, lactic, propionic, sorbic, and benzoic acids. Multiprotic organic acids such as citric, malic, tartaric, and fumaric acids possess variable but usually limited activity. Ester derivatives of some weak organic acids are discussed here because they likely have similar mechanisms of microbial inhibition. The antimicrobial effectiveness of organic acids is strongly related to the pH of a food system, and the undissociated form of the acid is largely, though not entirely, responsible for the inhibition of microorganisms. Thus, in selecting an organic acid for use as an antimicrobial food additive, both the product pH and the pKa of the acid must be considered. The use of organic acids as antimicrobials is generally limited to foods with pH <5.5, since most organic acids have pKas of 3.0 to 5.0 (124). The mechanisms of action of organic acids and their derivatives have some common elements. There is little evidence that the organic acids and related esters influence cell wall synthesis in prokaryotes or that they significantly interfere with protein synthesis or genetic processes. As stated previously, organic acids are better able to penetrate the microbial membrane in their undissociated form. Once inside the cell, the acid encounters a near-neutral pH environment and dissociates into the free proton and acid anion, acidifying the cell interior (Fig. 30.1). Bacteria maintain a cytoplasmic pH near neutrality to prevent conformational changes to structural proteins, enzymes, nucleic acids, and phospholipids. Liberated protons must then be actively extruded to the outer environment. This is an energetic process, consuming ATP. According to the chemiosmotic theory, the cytoplasmic membrane is impermeable to protons and their removal requires active transport. The extrusion of protons to the exterior environment creates a charge gradient across the membrane,

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials

767

Developed resistance may be termed tolerance, adaptation, or habituation depending on how the microorganism is exposed to the stress and the physiological conditions that lead to enhanced survival (59, 146). For example, Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes can become more acid tolerant and possibly adapt to other stresses (e.g., heat, osmotic pressure) if subjected to mild acidity before exposure to more acidic conditions (8, 20, 123, 389). In foods, adaptive alteration of cells might be of concern if cells sustain the adaptive response over multiple generations, a phenomenon that has not been reported to occur with the food antimicrobials discussed here (108, 342). Nonetheless, the greater the degree of severity of the antimicrobial challenge, the less likely the microorganism is to survive for extended periods, even if it is adapted (108).

Acetic Acid and Acetates

Figure 30.1  Fate of an organic acid (RCOOH) in a low-pH environment in the presence of a microbial cell. doi:10.1128/9781555818463.ch30f1

resulting in an electrochemical potential called the proton motive force (PMF) (266, 267). Lambert and Stratford (225) determined that yeast cells pump excess protons out of the cell, consuming ATP in the process. However, the intracellular pH may eventually be raised to a point that the cell may be able to resume its growth. The time required to accomplish this intracellular pH increase is dependent on the extracellular pH and inhibitor concentration and is termed the “lag time.” While some have suggested that only the undissociated organic acid has antimicrobial activity, other studies have demonstrated a contribution of the acid anion, albeit a slight one (130). Paul and Hirshfield (311) suggested that perturbation of membrane function by organic acids, leading to interference with membrane-bound protein function, represents a mechanism of microbial inhibition. Another possible mechanism proposed by these and other researchers involves the cytoplasmic accumulation of anion, leading to deterioration of nucleic acid and protein functionality and even increases in osmolarity and metabolic interference (85, 343). Many environmental stressors (e.g., heat, cold, starvation, and low pH) can trigger resistance responses of microorganisms to subsequent stress exposure (48).

SMP_Food Microbiology_CH30.indd

767

Acetic acid (pKa, 4.75; molecular mass, 60.05 Da) (Fig. 30.2), the primary component of vinegar, and its sodium, potassium, and calcium salts, sodium and calcium diacetate, and dehydroacetic acid (methylacetopyranone) are some of the oldest food antimicrobials. Acetic acid is more effective against yeasts and bacteria than against molds. Only acetic acid bacteria (principally Acetobacter spp.), lactic acid bacteria (LAB), and butyric acid anaerobes are tolerant to acetic acid (124). Bacteria inhibited by acetic acid include Bacillus spp.,

Figure 30.2  Organic acids used as antimicrobial food preservatives. doi:10.1128/9781555818463.ch30f2

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

768 Campylobacter jejuni, Clostridium spp., E. coli, L. mono­ cytogenes, Arcobacter butzleri, Yersinia enterocolitica, Pseudomonas spp., Salmonella, and Staphylococcus au­ reus (77, 207, 230, 327, 367, 389). Molds and yeasts are more resistant to acetic acid than are bacteria; sensitive yeasts and molds include Aspergillus, Penicillium, Rhizopus, and some strains of Saccharomyces (124, 217). The general mechanism by which acetic acid inhibits microorganisms is similar to other organic acids as discussed previously. Acetic acid and its salts have shown variable success as antimicrobials in food applications. Acetic acid can increase poultry shelf life when added to cut-up chicken parts in cold water at pH 2.5 (269). Addition of acetic acid at 0.1% to scald tank water used in poultry processing decreases the heat resistance of Salmonella ­enterica serovars Newport and Typhimurium and C. jejuni (296). Lillard et al. (236), however, found that 0.5% acetic acid in scald water had no significant effect on Salmonella, total aerobic bacteria, or Enterobacteriaceae on unpicked poultry carcasses. Birk et al. (41) recently reported a 6.0-log10 reduction of C. jejuni in brain heart infusion and chicken juice during storage at 4°C. Over et al. (302) reported that application of 150 mM acetic acid to raw chicken tissue resulted in an ~2.0-log10 reduction in numbers of Salmonella serovar Typhimurium and E. coli, though levels of L. monocytogenes remained largely unaffected. Carlson et al. (68, 69) produced reductions of E. coli O157:H7 on beef cattle hide swatches ranging from 1.6 to 2.1 log10 CFU/cm2 following application of 10% acetic acid at 55°C. Significant reductions were observed in some cases following application of 10% acetic acid at 23°C, although reductions were greater with increased acid solution temperature (69). Acetic acid has shown variable effectiveness as an antimicrobial for use as a meat decontaminant. Use of 2% acetic acid resulted in reductions in viable E. coli O157: H7 on beef after 7 days at 5°C (367). Levels of E. coli O157:H7 and Salmonella on beef trim exposed to 2 or 4% acetic acid were 2.0 to 2.5 log10 less than numbers of pathogens on controls immediately following acid application (181). These researchers reported that, following refrigeration and 30 days of freezing of ground trim previously exposed to acetic acid, pathogen numbers were maintained at levels similar to those observed just after acid application (181). Acetic acid (0.1 to 2%) added to bread dough inhibited growth of 6.0 log10 CFU of rope-forming Bacillus subtilis per g in wheat bread (pH 5.14) stored at 30°C for >6 days (338). Similarly, acetic acid (0.1% wt/vol; pH 4.8) suppressed growth in nutrient broth of six Bacillus spp. previously identified as being capable of forming bread rope

SMP_Food Microbiology_CH30.indd

768

(310). In addition to these and other similar products, acetic acid is used commercially in baked goods, cheeses, condiments and relishes, dairy product analogues, fats and oils, gravies and sauces, and meats (124). Sodium acetate and other salts of acetic acid have been investigated for their antimicrobial capability in multiple types of food products, predominantly fresh and processed meats, poultry, and seafood. Sodium acetate at 1% extended the shelf life of catfish muscle stored at 4°C by 6 days (212). Al-Dagal and Bazaraa (3) determined that whole or peeled shrimp dipped in a 10% sodium acetate (wt/wt) solution for 2 min had extended microbiological and sensory shelf life compared with controls. Similarly, application of 2.5% (wt/vol) sodium acetate to rainbow trout fillets resulted in a 1.4-log10 reduction in total mesophilic aerobes, whereas equivalent concentrations of sodium lactate and citrate produced reductions of 1.2 log10 CFU/g (211). Dehydroacetic acid has a pKa of 5.27 and is therefore active at higher pH values than acetic acid. It is inhibitory to bacteria at 0.1 to 0.4% and to fungi at 0.005 to 0.1% (124). Zhang et al. (434) determined that sodium dehydroacetate in liquid medium was inhibitory to E. coli at 1,000 μg/ml and S. aureus at 125 μg/ml. Sodium diacetate (pKa, 4.75) is approved for use in processed meat and poultry products by the U.S. Department of Agriculture Food Safety and Inspection Service (9 CFR 424.21) at levels not to exceed 0.25% of the product formulation. It is often used in combination with sodium or potassium lactate (see below) or other antimicrobials in processed meats and poultry to inhibit L. monocytogenes growth following postprocess crosscontamination. Barmpalia et al. (29) formulated frankfurters with 0.25% sodium diacetate, which inhibited L. monocytogenes on surfaces over 40 days of storage at 10°C such that levels of the pathogen were 2.5 log10 less than controls at the end of storage. Incorporation of either 0.097 or 0.19% diacetate in frankfurters containing 0.68 or 1.36% lactate inhibited growth of L. monocytogenes over 120 days of post-processing refrigerated storage, effecting an ~1.0-log10 reduction (317). Application of 22 or 44 ppm lauric arginate ester to the surfaces of frankfurters resulted in additional reductions in the pathogen, decreasing L. monocytogenes populations to ~1.0 log10 CFU/package after 90 days of storage (317). Injection of 0.15% diacetate into beef knuckle meat inoculated with E. coli O157:H7 in addition to NaCl, tripolyphosphate, and lactate, however, did not increase the heat sensitivity of the pathogen as compared with controls (65). Sodium diacetate and sodium lactate act synergistically to inhibit pathogens in media and food. Incorporation

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials of a 3% sodium lactate-sodium diacetate blend into chicken meat extended the lag phase of Salmonella serovar Typhimurium cells by approximately 28 h in cells not previously acid stressed (204). Degnan et al. (111) observed a 2.6-log10 decrease in L. monocytogenes populations in blue crab meat following washing with 25 mM sodium diacetate and storage for 6 days at 4°C. Sodium diacetate is also useful in the baking industry because of its ability to inhibit contaminating microbes without negatively affecting fermentative yeast.

Benzoic Acid and Benzoates

Benzoic acid (pKa, 4.19; molecular mass, 122.12 Da) (Fig. 30.2) and sodium benzoate were the first anti­ microbial compounds permitted in foods by the Bureau of Chemistry, the predecessor of the U.S. Food and Drug Administration (FDA) (86). Benzoic acid occurs naturally in cranberries, plums, prunes, cinnamon, cloves, and most berries. Sodium benzoate is highly soluble in water (66.0 g/100 ml at 20°C), while benzoic acid is much less so (0.27% at 18°C). Eklund (131) determined that while the dissociated and undissociated forms of benzoic acid are both inhibitory to bacterial and fungal microbes, the MICs for the undissociated acid are 15 to 290 times lower. Benzoic acid and the benzoates are used primarily as antifungal agents in food applications. The inhibitory concentration of benzoic acid at pH <5.0 against most yeasts ranges from 20 to 700 μg/ml, while for molds it is 20 to 2,000 μg (86). Some fungi, including Byssochlamys nivea, Talaromyces flavus, Pichia membranaefaciens, and Zygosaccharomyces bailii, are resistant to benzoic acid (86). Yousef et al. (430) observed bacteriostatic effects of benzoic acid in broth at 1,000 to 3,000 ppm as a function of inhibition temperature, although bactericidal effects were not observed at any usage level. Benzoic acid at 0.1% was effective in reducing E. coli O157:H7 populations in apple cider by 3.0 to 5.0 log10 CFU/ml after 7 days at 8°C (78). Incorporation of 0.1% sodium benzoate in pH 3.5 grape juice resulted in reduced time required to achieve a 3.0-log10 reduction of the thermotolerant mold Neosartorya fischeri (320). Application of 5.0% potassium benzoate to surfaces of bologna or ham slices resulted in bacteriostatic inhibition of L. monocytogenes over 48 days of vacuum-refrigerated (10°C) storage (158). Koczon´ (220) reported that multiple benzoate derivatives (Zn-p-iodobenzoate, Zn-phydroxybenzoate, Zn-p-aminobenzoate) inhibited the yeast Pichia anomala by activation of a molecular pump system that resulted in expenditures of large amounts of molecular energy. Most recently, Critzer et al. (98, 99) demonstrated that exposure of E. coli O157:H7 to

SMP_Food Microbiology_CH30.indd

769

769

sodium benzoate at 0.1 to 0.5% resulted in increased expression of multiple gene cassettes associated with antimicrobial resistance responses or microbial survival in stressful environments.

Lactic Acid and Lactates

Lactic acid (pKa, 3.79; molecular mass, 90.08 Da) (Fig. 30.2) is synthesized naturally by LAB during anoxic fermentation. Lactic acid and lactates may function as antimicrobials, acidulants, and as flavorings in food products. Lactates have also aided in retarding lipid oxidation and subsequent off-flavor development in processed meats (306). Many studies have addressed the antimicrobial potential of these compounds for decontamination of meat and poultry carcass surfaces during animal slaughter or on surfaces of fabricated, further-processed meats (392). The U.S. Department of Agriculture allows lactic acid to be applied to carcass surfaces pre- or postchilling (£5.0% acid solution), to subprimal cuts and trimmings (2 to 3% acid at £55°C), and to beef heads/tongues (2 to 2.8% in wash systems) (406). At reduced pH, l-lactate is significantly more inhibitory than d-lactate (260, 261). Sodium lactate (2.5 to 5.0%) inhibits Clostridium botulinum, Clostridium sporogenes, L. monocyto­ genes, Salmonella, S. aureus, Y. enterocolitica, and other spoilage bacteria in various meat and poultry products (188, 204, 291, 294, 357). Blending of lactate and diacetate, alone or in the presence of other antimicrobials such as lauric arginate or pediocin, has proven effective for inhibiting growth of L. monocyto­ genes and Salmonella serovar Typhimurium on f­urtherprocessed meat and seafood products, including frankfurters, smoked salmon, bologna, turkey ham, b­ratwurst, and comminuted beef and poultry (29, 34, 166, 204, 240, 249, 286, 291, 317). Stopforth et al. (375) reported bacteriostatic inhibition of inoculated L. monocytogenes following application of 0.07% lauric arginate ester to surfaces of cooked/cured hams formulated with 1.68% potassium lactate + 0.12% sodium diacetate and stored under vacuum for 90 days at 4°C. Similarly, a mixture of 2.5% sodium lactate and 0.25% sodium acetate inhibited growth of L. monocytogenes in sliced cooked ham and cervelat sausage over 5 weeks of storage at 4°C (42). Sodium or calcium lactate (1 to 4.8%) is effective either alone or in combination with sodium acetate, sodium diacetate, sodium citrate, or buffered sodium citrate in inhibiting C. botulinum and Clostridium perfringens growth in vacuum-packed roast beef, turkey, and/or chicken products, pork, and beef goulash (14, 188, 189, 202, 412). Yarbaeva et al. (429) determined that incorporating 2% potassium

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

770 lactate into beef-containing pastries reduced the C. perfringens population by >2.0 log10 CFU/g after 24 h of postcooking storage at room temperature. Houtsma et al. (188) reported that C. botulinum neurotoxin synthesis was delayed at 15 and 20°C by addition of 2.0 and 2.5% sodium lactate, respectively, while addition of >4% sodium lactate inhibited toxin synthesis altogether in cells held at 30°C. Huang and Juneja (191) determined that the addition of 4.5% sodium lactate did not result in increased sensitivity of E. coli O157: H7 to heating in ground beef. Similar conclusions were reported by others investigating the effect of 4.8% sodium lactate in poultry meat on the heat resistance of Salmonella (272). Conversely, Aymerich et al. (17) concluded that sodium lactate interacted synergistically with high-pressure processing in inactivating L. mono­ cytogenes and Salmonella. Alakomi et al. (2) determined that lactic acid permeabilized the membranes of gram-negative bacteria such as E. coli O157:H7, P. aeruginosa, and S. Typhimurium. Since the lactates have relatively little effect on the pH of most foods, in particular meat and poultry, there is still debate surrounding the exact mechanism of microbial inhibition by these compounds. The addition of lactates can effectively reduce the aw of a food system, resulting in increased microbial inhibition (77, 291, 357). Others have determined that aw reduction by lactates at use concentrations is not sufficient to produce the microbial reductions that have been observed in many studies, indicating rather that the likely mechanism of microbial inhibition is the combined effects of all the mechanisms discussed here (84, 420).

Propionic Acid and Propionates

Up to 1% propionic acid (pKa, 4.87) (Fig. 30.2) is produced by fermentation in Swiss-type, eyed cheeses by the heterofermentative gram-positive bacterium Propioni­ bacterium freudenreichii subsp. shermanii. The degree of antimicrobial activity of the propionates depends on the pH of the food, with the undissociated acid being the more active form. Eklund (131) demonstrated that the undissociated form of propionic acid has 11 to 45 times more antimicrobial efficacy than the dissociated form. Propionic acid and sodium, potassium, and calcium propionates are used primarily as antimycotics, although bacteria and yeasts are also inhibited. The microorganism in bread dough that is responsible for rope formation, B. subtilis, is inhibited by propionic acid at pH 5.6 to 6.0 (124). Propionates (0.1 to 5.0%) retard the growth of bacteria including E. coli, S. aureus, Sarcina lutea, Salmonella, Proteus vulgaris, Lactobacillus plan­ tarum, and L. monocytogenes and of the yeasts Candida

SMP_Food Microbiology_CH30.indd

770

and Saccharomyces cerevisiae (124). Gerez et al. (159) determined that 0.4% calcium propionate was significantly less effective than a starter culture of four species of Lactobacillus in packaged wheat bread. The combination of 0.4% propionate with the starter culture extended the bread’s shelf life from 9 to 24 days (159). Nonetheless, dipping of apple slices in solutions of 0.5, 1.0, or 2.0% calcium propionate did not result in appreciable reductions of E. coli populations on surfaces of inoculated apple slices (175). Golden et al. (168) determined the rates of inactivation of L. monocytogenes by propionic and acetic acids at different pH values, temperatures, and concentrations and observed that propionic acid was slightly more effective in inhibiting the pathogen, reducing the time required to achieve a 4.0-log10 reduction in pathogen populations as a function of acid concentration and pH.

Sorbic Acid and Sorbates

Sorbic acid (Fig. 30.2) was first identified in 1859 by A.W. van Hoffman, a German chemist, from the berries of the mountain ash tree (rowanberry) (374). Sorbic acid is a trans-trans, unsaturated monocarboxylic fatty acid that is slightly soluble in water (0.16 g/100 ml at 20°C). As with benzoic acid, the dissociated acid salt is significantly more water soluble (58.2 g/100 ml at 20°C). As with other organic acids, the antimicrobial activity of sorbic acid is greater when the compound is in the undissociated state. Its pKa is 4.75, indicating that its activity is greatest in foods with a pH value less than 6.0. The undissociated form of the acid is 10 to 600 times more inhibitory than the deprotonated form (130). Sorbates are the best characterized of all food antimicrobials as to their spectrum of action. They inhibit both fungi and bacteria. Foodborne yeasts inhibited by sorbates include species of Brettanomyces, Byssochlamys, Candida, Cryptococcus, Debaryomyces, Hansenula, Pichia, Rhodotorula, Saccharomyces, Sporobolomyces, Torulaspora, and Zygosaccharomyces (152, 374). Food­ borne genera of molds inhibited by sorbates include Alternaria, Aspergillus, Botrytis, Cephalosporium, Fusa­ rium, Geotrichum, Helminthosporium, Mucor, Penicil­ lium, Pullularia, Sporotrichum, and Trichoderma (374, 436). Additionally, sorbic acid and sorbates inhibit the production of mycotoxins by the toxigenic molds Aspergillus flavus, Aspergillus parasiticus, Alternaria alternata, B. nivea, Penicillium expansum, and Penicillium patulum (61, 62, 92, 233, 334, 344). Sorbates can be degraded through decarboxylation, resulting in the formation of 1,3-pentadiene, a compound with a petroleumlike odor, by some species of molds (152, 235).

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials Bacteria inhibited by sorbic acid and sorbates include Acinetobacter, Aeromonas, Alicyclobacillus acido­ terrestris, Bacillus, Campylobacter, E. coli O157:H7, Lactobacillus, Listeria innocua, Pseudomonas, Salmo­ nella, Staphylococcus, Vibrio, and Y. enterocolitica (31, 122, 349, 374). Sorbic acid inhibits primarily catalaseproducin­g bacteria, whereas with some exceptions, catalase-negative bacteria are not inhibited. This enables the use of sorbates in products fermented by LAB. Sorbate inhibits the growth of many spoilage and pathogenic bacteria in or on foods, including A. ac­ idoterrestris and Propionibacterium cyclohexanicum in processed fruit juices (417); S. enterica serovar Enteriditis, L. monocytogenes, and Aeromonas spp. on fresh and processed salt- and freshwater fish (174, 383); E. coli O157:H7 in soft cheese (208); L. monocytogenes and S. aureus on cooked ham (11); E. coli O157:H7, L. monocytogenes, and S. aureus on refrigerated beef round (237); and L. monocytogenes and Salmonella on fresh-cut produce and melons (28, 401). Sorbates are effective anticlostridial and antilisterial agents in cured meats and other meat and seafood products (125, 142, 248, 348, 354). Sorbates prevent the germination of Clostridium spores and formation of C. botulinum neurotoxin in meat, poultry, soy protein frankfurters and emulsions, and bacon (374). Lopez et al. (241) showed that 0.05 to 0.1% potassium sorbate at pH 5.0 could inhibit germination and growth of Bacillus stearothermophilus following heating at 120°C for 1 to 5 min. Smoot and Pierson (370) proposed that sorbate inhibits Bacillus and Clostridium spore germination by competing with l-alanine, a chemical signal triggering the germination process, for binding to spores. In studies with a proteolytic strain of Clostridium, sorbate was determined to function as a protonophore in a pHdependent manner to inhibit growth, amino acid uptake, and dissolution of PMF as measured by decreased flagellar activity and to induce alterations in morphological structures (335, 336). One of the primary cellular targets of sorbic acid in vegetative cells is the cytoplasmic membrane. Sorbic acid inhibits amino acid uptake, which in turn has been theorized to be responsible for dissipating the membrane PMF through depletion (147, 148, 361–363). Ronning and Frank (335) concluded that sorbic acid reduces the cytoplasmic membrane electrochemical potential and PMF, and that sorbic acid-induced PMF dissolution resulted in inhibition of amino acid transport, ultimately inhibiting other internal processes. Eklund (130), however, showed that much higher concentrations of sorbic acid are required for complete loss of DY. Interestingly, research has revealed that sorbic acid does not neces-

SMP_Food Microbiology_CH30.indd

771

771

sarily function exclusively as a weak organic acid antimicrobial, exhibiting membrane fluidization activity in addition to PMF destabilization (50, 376). Sorbate can be applied to foods by direct addition, dipping/immersion, spraying, dusting, or incorporation into packaging films. Bakery products can be protected from fungal growth through the application of 0.05 to 0.1% potassium sorbate either as a spray or by direct addition. Foods in which sorbates are used include beverage syrups (0.1%), cakes and icings (0.05 to 0.1%), cheese and associated products (0.2 to 0.3%), cider (0.05 to 0.1%), dried fruits (0.02 to 0.05%), fruit drinks (0.025 to 0.075%), margarine (0.1%), pie fillings (0.05 to 0.1%), salad dressings (0.05 to 0.1%), and wines (0.02 to 0.04%) (12).

Miscellaneous Organic Acids, Fatty Acids, and Fatty Acid Esters

As mentioned previously, multiple organic acids exhibit various degrees of antimicrobial activity against a variety of foodborne microbes. Examples of such acids not already addressed include citric, malic, tartaric, and fumaric acids and their related salts. Fumaric acid has been used in wines in some warm-weather regions to prevent malolactic fermentation (300). More recently, its use as an antimicrobial has been evaluated on surfaces of fresh produce and in processed juices. Kondo et al. (221) determined that E. coli O157:H7, S. Typhimurium, and S. aureus populations on lettuce were reduced by 1.2 to 1.5 log10 following exposure to 50 mM fumaric acid, although tissue browning was also observed following acid application. Addition of 0.15% fumaric acid and 0.05% sodium benzoate to nonpasteurized apple cider (pH 3.3) reduced E. coli O157:H7 populations by 5.0 log10 after 6 h of incubation at 25°C and subsequent refrigeration (93). Kim et al. (214, 215) determined that 0.5 g fumaric acid/100 ml reduced populations of E. coli O157:H7, S. Typhimurium, and L. monocytogenes on alfalfa and broccoli sprouts by approximately 2.8 to 3.4 log10 CFU/g. Sodium citrate (0.625 to 2.5%) inactivated S. aureus in microbiological medium (231), whereas 26 mM citric acid was inhibitory to growth of L. monocytogenes in culture medium (295). Del Río et al. (116, 117) found that citric acid in culture medium extended the lag phase of foodborne microbes and increased the time required for bacterial populations to achieve stationary phase. Citric acid is inhibitory to Salmonella and E. coli O157:H7 on surfaces of poultry and beef (102, 228, 380). Buffered sodium citrate (³1%) reduced the growth of C. perfringens on beef rounds and loins during prolonged chilling compared with control products

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

772 (390). L. monocytogenes populations on frankfurters dipped in 1% citric acid solution and then stored under vacuum at 5°C for 80 days were ~2.0 log10 less than controls (303). Citric acid inhibits growth of pathogens such as L. monocytogenes and E. coli O157:H7 on fresh and fresh-cut produce (28, 66, 319). Salmonella, Campylobacter, and L. monocytogenes are inhibited on surfaces of meat and poultry (60, 117, 170). Growth and toxin synthesis by molds is inhibited by citric acid, especially for species of Aspergillus (324). As a triprotic organic acid, citric acid functions not only as a weak organic acid, but its salts also function as a buffering agent and chelator (263, 264). Levulinic acid (4-oxopentanoic acid; molecular mass, 116.12 Da) (Fig. 30.2) has received recent attention for its reported antimicrobial efficacy against multiple bacterial pathogens on multiple food surfaces, including meat, poultry, and fresh produce (13). Vasavada et al. (411), in studies incorporating 1.4 and 2.7% levulinic acid in pork and turkey sausages, determined that aerobic plate counts in treated sausages were consistently 1.5 to 2.0 log10 CFU/g less than in untreated controls after 14 days of refrigerated (2°C) storage. S. Enteriditis populations in water held at 21°C were reduced by 3.4 log10 CFU/ml when exposed to 0.3% levulinic acid for 30 min. (437). Salmonella serovar Typhimurium and E. coli O157:H7 populations on romaine lettuce were reduced by 4.8 to 5.0 log10 CFU/g after 5 min of exposure to a combination of 0.5% levulinic acid and 0.05% sodium dodecyl sulfate (437). E. coli O157:H7 and Salmonella on alfalfa seeds treated with 0.5% levulinic acid + 0.05% sodium dodecyl sulfate for 30 min were reduced by 5.0 to 6.0 log CFU/g and germination rates were equivalent to deionized water-treated control seeds (438). In addition to the weak organic acids and the shortchain fatty acids, the medium-chain fatty acids have received increased attention regarding antimicrobial potential. Nakai and Siebert (279) reported that caprylic and caproic acids at 2.83 g/liter were inhibitory to L. monocytogenes and Listeria ivanovii. Although the MIC of caproic acid was 2.83 g/liter for P. aeruginosa, the MIC of caprylic acid was nearly 10-fold higher (279). The MICs for lauric and caprylic acid against Listeria spp. ranged from 0.63 to 1.25 mM and ³5.0 mM, respectively (290). Brandt (52) determined that 25 μg of octanoic (caprylic) acid/ml of culture medium inhibited L. monocytogenes at pH 5.0 and a combination of octanoic acid with acidified calcium sulfate synergistically inhibited the pathogen. Populations of Salmonella inoculated on alfalfa sprouts were reduced by ~1.5 log CFU/g following 10 min of exposure to 75 mM caprylic acid (82). Surface treatment of processed meat and poul-

SMP_Food Microbiology_CH30.indd

772

try products with 1.0% octanoic acid reduced L. mono­ cytogenes populations by 1.9 to 3.6 log CFU (63). E. coli O157:H7 populations in feces-contaminated water treated with 0.5% caprylic acid for 30 min at room temperature were reduced by 4.0 log CFU/ml (439). Many fatty acid esters exhibit antimicrobial activity in foods, with glycerol monolaurate (monolaurin) being one of the most effective (205). Monolaurin is inhibitory toward gram-positive bacteria, including Bacillus, Lactococcus, L. monocytogenes, Micrococcus, and S. aureus, at concentrations of £100 μg/ml but is much less effective against gram-negative bacteria (4, 22, 23, 56, 253, 434). Combining monolaurin with EDTA decreased by at least 50% the concentration of monolaurin required to inhibit L. monocytogenes, E. coli O157:H7, or E. coli O104, whereas no additional antimicrobial effect was observed with P. aeruginosa or Salmonella (56). Foodborne fungi, including species of Aspergillus, Alternaria, Candida, Cladosporium, Penicillium, and Saccharomyces, are inhibited by monolaurin (7, 205). Monolaurin’s antimicrobial activity is reduced in the presence of starch or lipid but is unaffected by protein (434). Monolaurin inhibits cells by destabilizing the cell membrane, including cellular uncoupling of electron transport/energy regulation processes, and can interact synergistically with organic acids such as benzoate and lactate (134, 391). The monoester of caprylic acid (monocaprylin) is active in foods against both grampositive and gram-negative bacterial pathogens (290). E. coli O157:H7 and Salmonella spp. populations in culture medium containing 25 mM monocaprylin at room temperature were reduced within 1 min by approximately 3.0 and 2.0 log10 CFU/ml, respectively (82). Populations of Cronobacter sakazakii in reconstituted infant formula containing 50 mM monocaprylin were reduced to below 1.0 log10 CFU/ml after 24 h at 37, 23, or 4°C (278). Dipping frankfurters in a solution containing 50 mM monocaprylin plus 1.0% acetic acid reduced L. monocytogenes populations by 3.0 log10 CFU/frankfurter during refrigerated vacuum storage (153). Lauric arginate ester (LAE) (Na-lauroyl ethylester) is a cationic surfactant-type food antimicrobial that is effective against Staphylococcus, Bacillus, Micrococcus, L. monocytogenes, Citrobacter, E. coli, Salmonella, and Pseudomonas, as well as yeasts such as Saccharomyces, Zygosaccharomyces, and Brettanomyces species (104, 195, 332, 375). LAE is also very active against L. mono­ cytogenes, as determined by multiple studies on various processed meat products when applied as a postlethality antimicrobial intervention. Brandt et al. (53) determined that the minimum bactericidal concentration of LAE for L. monocytogenes was 12.5 μg/g. Combining

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials LAE with nisin reduced this concentration by five- to eightfold (53). L. monocytogenes populations on frankfurters treated with 5,000 ppm LAE at 4.4 or 23°C were reduced by 1.4 log10 CFU/package (381). Similar reductions of L. monocytogenes on cooked hams were observed when equivalent levels of LAE were sprayed on the surface and products were held at refrigeration or room temperature (382). Surface treatment with 0.07% LAE of cooked cured hams formulated with lactate and diacetate inhibited L. monocytogenes for 95 days of vacuum storage at 4°C (375). Similarly, L. monocy­ togenes was inhibited on frankfurters formulated with lactate and diacetate and surface treated with 22 ppm LAE, then held refrigerated for 156 days under vacuum (255). Porto-Fett et al. (317) determined that applying 22 or 44 ppm LAE to surfaces of frankfurters reduced L. monocytogenes populations by 2.0 to 2.5 log10 CFU, without subsequent regrowth occurring during refrigerated, vacuum-packaged storage. Lauric arginate also has antimicrobial activity against L. monocytogenes in fluid milks and soft Mexican-style cheese (371).

Dimethyl Dicarbonate

Dimethyl dicarbonate (DMDC) (Fig. 30.3) is a colorless liquid that is only slightly soluble in water (3.6%). The compound is very reactive with many substances, including water, ethyl alcohol, alkyl and aromatic amines, and sulfhydryl groups (167). The primary target microorganisms for DMDC are yeasts, including Brettanomyces, Saccharomyces, Zygosaccharomyces, Rhodotorula, Candida, Torulopsis, Torula, Endomyces, Kloeckera, and Hansenula. The compound is also bactericidal at 30 to 400 μg/ml to a number of species, including Acetobacter pasteurianus, E. coli, P. aeruginosa, S. aureus, Lactobacillus spp., and Pediococcus cerevisiae (167). E. coli O157:H7 was inactivated in apple cider containing 0.025% DMDC stored for 12 days at 4°C (144). In apple ciders derived from different apple varieties, 125 ppm and 250 ppm DMDC reduced E. coli O157:H7 populations by 7.7 log10 CFU/ml within 24 h at room temperature (33). DMDC effectively inhibits wine spoilage yeasts, including Brettanomyces, Dekkera, and Schizosaccharomyces species (96, 121, 325). Molds are generally more resistant to DMDC than yeasts. The mechanism by which DMDC inhibits microbes is most

Figure 30.3  DMDC. doi:10.1128/9781555818463.ch30f3

SMP_Food Microbiology_CH30.indd

773

773

likely enzyme inactivation. A related compound, diethyl dicarbonate, reacts with imidazole groups, amines, and thiols in proteins (128), as well as with histidyl groups of proteins (280). This can cause inactivation of microbial lactate dehydrogenase or alcohol dehydrogenase by complexing with His residues in the catalytic sites of these enzymes (167).

Lactoferrin and Lactoferricin

Lactoferrin is the primary iron-chelating protein in milk and colostrum; it is also present in other physiological fluids and polymorphonuclear leukocytes. Lactoferrin (molecular mass, 76.5 kDa) is a glycoprotein that exists in milk primarily as a tetramer with Ca2+ and possesses two iron-binding sites per molecule. For each molecule of Fe3+ bound by the enzyme, one bicarbonate (HCO3–) is required. Lactoferrin must be low in iron saturation and bicarbonate must be present for the protein to exert antimicrobial activity (298). Its exact biological role is unknown; however, it may act as a barrier to infection of the nonlactating mammary gland and aid in protecting the newborn against gastrointestinal infection (386, 387). Although lactoferrin was originally believed to function only through chelation of iron, bactericidal activity has been identified, likely as a result of direct interaction between the protein and microorganisms (139). Lactoferrin is inhibitory to many foodborne microorganisms, including B. subtilis, B. stearothermophilus, Carnobacterium viridans, L. monocytogenes, micrococci, E. coli, and Klebsiella species (6, 270, 298, 413). Payne et al. (312) found that bovine lactoferrin had to be reduced to 18% iron saturation to have bacteriostatic activity against four strains of L. monocytogenes and an E. coli strain at concentrations of 15 to 30 μg/ ml in ultrahigh-temperature-treated (UHT) milk. At 2.5 mg/ml, it exhibited no activity against S. Typhimurium or P. fluorescens and little activity against E. coli O157: H7 or L. monocytogenes (313). Branen and Davidson (56) studied the interactive antimicrobial effects of combinations of lactoferrin with monolaurin, nisin, or lysozyme against L. monocytogenes, E. coli, S. Enteriditis, and P. fluorescens in culture medium and fluid milk. In culture medium, lactoferrin combined with nisin inhibited L. monocytogenes and lactoferrin combined with monolaurin inhibited E. coli O157:H7. None of the antimicrobial combinations with lactoferrin were effective in UHT milk incubated at 25°C, attributed to interactions of the antimicrobials with food components and the storage temperature (56). Restriction of microbial access to nutrients via sequestration of iron is likely part of lactoferrin’s mechanism of inhibition (169). Iron stimulates the growth of

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

774 bacteria, including Clostridium, Escherichia, Listeria, Pseudomonas, Salmonella, Staphylococcus, Vibrio, and Yersinia species (133). Many gram-negative and grampositive bacteria synthesize siderophores, low-molecular-weight secondary metabolites that bind iron tightly, allowing enhanced transmembrane uptake (110). Micro­ organisms with low iron needs, such as LAB, are not sensitive to lactoferrin (304). Combining lactoferrin with EDTA resulted in increased release of lipopolysaccharide (LPS) from the outer membranes of E. coli (136) and P. aeruginosa (339). Lactoferrin causes this LPS release by chelation of cations, including magnesium, calcium, and iron, which stabilize the LPS in the bacterial membrane. Lactoferrin can also detach microbes adhering to various surfaces and inhibit the formation of biofilms by bacteria (299). Lactoferricin B (i.e., hydrolyzed lactoferrin [HLF]) is a small polypeptide (25 residues) obtained from pepsin digestion of lactoferrin (15). Unlike bovine lactoferrin, lactoferricin is inhibitory to multiple genera of the LAB, including Lactobacillus and Pediococcus (138). Incubation of S. aureus, E. coli O157:H7, or enteropathogenic E. coli in 20 to 40 μM HLF resulted in only 10 to 15% growth of targeted microbes compared with untreated controls (145). Jones et al. (200) determined that HLF at concentrations of 1.9 to 125 μg/ml has bacteriostatic and/or bactericidal activity against Shigella, Salmonella, Y. enterocolitica, E. coli O157:H7, S. au­ reus, L. monocytogenes, and Candida. These findings have recently been confirmed and reviewed (162, 242). The addition of the chelator EDTA enhances the activity of HLF in liquid medium, indicating that decreases in the antimicrobial activity of HLF may be in part due to excess cations (55). Venkitanarayanan et al. (413) found that adding HLF to ground beef at 100 μg/g reduced E. coli O157:H7 populations by ~1.0 log10 CFU/g after 5 days at 4°C, whereas the pathogen populations in nonHLF-treated beef decreased by only 0.5 log10 CFU/g. In minimal culture medium, however, a decrease in E. coli O157:H7 populations of nearly 2.0 log10 CFU/ml was observed compared with medium containing no HLF (413). Min and Krochta (265) determined that 10 μg HLF per ml inhibited the spoilage yeast Penicillium com­ mune in minimal culture medium, although the peptide was unable to inhibit the yeast in potato dextrose broth, a more complex and nutrient-rich medium. Similarly, 4.0 and 2.0 mg of lactoferrin hydrolysates per ml were inhibitory to E. coli O157:H7 and L. monocytogenes, respectively, in peptone yeast extract glucose broth; however, adding lactoferrin hydrolysates to UHT milk containing either 10.0 mg/ml EDTA or no EDTA did not inhibit either pathogen (271). Most recently, stud-

SMP_Food Microbiology_CH30.indd

774

ies revealed that while pepsin-digested hydrolysates of lactoferrin (³1 mg/ml) reduced E. coli O157:H7 populations in broth by 2.5 log10 CFU/ml, amidated lactoferrin had greater antimicrobial activity against Serratia liq­ uefaciens than did native lactoferrin or pepsin-digested hydrolysates (115). Naidu (276) patented an antimicrobial system in which lactoferrin is immobilized to food-grade polysaccharides and is dissolved in a citrate-bicarbonate buffer with NaCl, resulting in a product named “activated lactoferrin” (ALf). The ability of E. coli O157:H7 to adhere to beef collagen tissue was inhibited by ALf, and the lag phase of E. coli O157:H7 in culture broth was extended by >17 h (277). However, Heller et al. (183) recently reported that 2% ALf followed by warmed 5% lactic acid was less effective at reducing E. coli O157: H7 populations on beef outside round pieces than was application of 5% warm lactic acid alone.

Lysozyme

Lysozyme (peptidoglycan N-acetylmuramoyl hydrolase; EC 3.2.1.17) is a 14.6-kDa lytic enzyme naturally present in avian eggs, mammalian milk, tears and other secretions, insects, and fish. It is affirmed as GRAS (Generally Regarded as Safe) and approved for direct addition to foods (140). It is the predominant protein in egg albumen, being present at approximately 3.5% (198). Hen egg white lysozyme (HEWL) is the primary antimicrobial substance in albumen, although its activity is enhanced by ovotransferrin, ovomucoid, and alkaline pH (~9.3). Lysozyme possesses greater stability to heat under acidic pH (30 min at 80°C), but its thermal stability is reduced as pH is increased, an interesting phenomenon given the natural alkalinization of eggs during their storage postlaying. The enzyme is most active from 55 to 60°C, but maintains up to 50% of its activity at lower temperatures (10 to 25°C) (198). Lysozyme catalyzes hydrolysis of the a-1,4-glycosidic bond between C-1 of N-acetylmuramic acid and C-4 of N-acetylglucosamine, resulting in cell wall degradation and lysis in hypotonic systems. Lysozyme is most active against gram-positive bacteria, most likely because the peptidoglycan of the cell wall is more exposed. It shows greatest activity against Bacillus (B. coagulans, B. stearothermophi­ lus), Micrococcus, Clostridium tyrobutyricum, and Thermoanaerobacterium thermosaccharolyticum (198). It has strain-specific activity against B. cereus, C. je­ juni, C. botulinum, L. monocytogenes, Lactobacillus spp., P. aeruginosa, and Y. enterocolitica (90, 198, 422). Gram-negative pathogens and spoilage bacteria not sensitive to lysozyme treatment include Aeromonas

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials hydrophila, Brochothrix thermosphacta, E. coli O157: H7, S. Typhimurium, Shigella, and Vibrio cholerae (198). The MIC of lysozyme against fungi including Candida, Sporothrix, Penicillium, Paecilomyces, and Aspergillus was >9,530 μg/ml in potato dextrose agar at pH 5.6 (323). Conversely, the MIC against Fusarium graminearum was only 1,600 μg/ml; the combination of lysozyme with EDTA at an equivalent concentration reduced the MIC to £500 μg/ml. Variation in susceptibility of gram-positive bacteria is likely due to the presence of teichoic acids and other materials that bind the enzyme and the fact that certain species have greater proportions of 1,6- or 1,3-glycosidic linkages in the peptidoglycan, which are more resistant than the 1,4 linkage (397). Some strains of L. monocy­ togenes are inhibited by lysozyme only when it is applied in combination with EDTA (56, 192). Hughey et al. (192) hypothesized that microbial peptidoglycan may be partially masked by other cell wall components and that EDTA enhances penetration of the lysozyme to the peptidoglycan. Conditions of microbial growth can also affect the antimicrobial activity of lysozyme. Johansen et al. (197) found that lysozyme’s antimicrobial activity was enhanced when it was applied to cells grown at 5 rather than at 25°C. Additionally, reducing the pH from 7.2 to 5.5 resulted in significant lengthening of the lag phase of L. monocytogenes cells exposed to 5,000 to 10,000 U lysozyme/ml. Lysozyme’s antimicrobial activity has been reported to be independent of its catalytic activity. Ibrahim et al. (193) determined that irreversibly denatured lysozyme had increased bactericidal activity against S. aureus. Furthermore, the antimicrobial activity of denatured lysozyme was confirmed against E. coli K-12 at 70°C. Enhanced antimicrobial activity was also demonstrated for denatured (80°C, pH 6.0) lysozyme against P. aeru­ ginosa and S. Enteriditis (193). Substituting Asp-52 in the catalytic site with Ser abolished the catalytic activity, but not the antimicrobial activity, of HEWL (194). These studies revealed that lyso­zyme denaturation increases the spectrum of activity of the polypeptide without sacrificing its antimicrobial activity. Gram-negative bacteria are generally insensitive to lysozyme, due to their reduced peptidoglycan content (5 to 10%) and presence of an LPS-covered outer membrane that blocks access of the enzyme (281, 282). Nevertheless, the susceptibility of gram-negative bacteria can be increased by pre- or cotreatment with chelators (18, 44, 45, 71). Chelation of Ca2+ or Mg2+ effectively removes essential cations useful for maintenance of LPS integrity (397). Additionally, gram-negative bacteria may be sensitized to lysozyme following expo-

SMP_Food Microbiology_CH30.indd

775

775

sure to acid, heat, osmotic shock, or dehydration shock (71, 398). The efficacy of lysozyme in inhibiting spoilage and pathogenic microbes on fresh and processed meat has been studied by several researchers. A combination of 250 or 500 ppm lysozyme, 250 or 500 ppm nisin, and 5 mM EDTA suppressed aerobic bacterial mesophile growth on ground ostrich patty surfaces during 7 days of aerobic or vacuum storage at 4°C (256). Applying lysozyme and nisin (25.5 g/liter; 1:3 ratio of lysozyme to nisin) and EDTA to pork bologna surfaces inactivated spoilage microbes and L. monocytogenes. Similarly, applying lysozyme to pork tissue significantly reduced B. thermosphacta and Carnobacterium populations (284). Natress and Baker (283) demonstrated that applying a blend of lysozyme and nisin (3:1; 260 μg/cm2) reduced catalase-negative LAB by 4.1 log10 CFU/cm2 compared with controls following 6 weeks of refrigerated, vacuum-packaged storage. Vacuum storage of pork loins following this antimicrobial application also significantly reduced B. thermosphacta populations (283). Applying 0.5 U lysozyme/cm2 on surfaces of turkey bologna followed by in-package pasteurization (65°C, 32 sec) reduced L. monocytogenes to below detectable limits within 10 weeks of refrigerated vacuum-packaged storage; a combination of lysozyme and nisin inactivated the pathogen more rapidly, reducing Listeria populations to below detection limits after only 2 weeks of storage at 4°C (251). Conversely, adding lysozyme (100 or 200 μg/ml), EDTA (1.0 to 2.5 mg/ml), or a combination of these antimicrobials to UHT milk did not significantly inhibit S. Typhimurium or P. fluorescens but did produce a significant inhibitory effect against L. monocytogenes (313). Likewise, growth of heat-shocked L. monocytogenes cells was significantly inhibited by 100 mg lysozyme/ml of demineralized fluid milk, a­lthough fortification with divalent cations removed the observed inhibition (210). Lysozyme is approved for use in cheeses to prevent late blowing caused by excess gas formed by fermentation by the anaerobe C. tyrobutyri­ cum. Lysozyme is also approved for use in frankfurter casings and in cooked meat and poultry products.

Nitrites

Sodium nitrite (NaNO2) and potassium nitrite (KNO2) have a specialized use in cured meat products. In addition to their antimicrobial properties, nitrites have other functions in cured meats. As nitric oxide (NO), nitrite reacts with the muscle protein myoglobin to form nitrosomyoglobin, providing and stabilizing the characteristic cured meat color. It also contributes to the flavor and texture of cured meats and may also serve as an

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

776 antioxidant. Meat curing is often combined with physical processes such as smoking, drying, heating, and even fermentation. As reviewed by Tompkin (393), the antimicrobial properties of nitrite were not recognized until the 1920s, and it was not until the 1950s that nitrate’s role was resolved as solely a source of nitrite, with no significant antimicrobial activity of its own. The primary antimicrobial application of nitrite is the inhibition of C. botulinum spore germination, growth, and toxin synthesis in cured meats. In association with other curing agents, such as NaCl and acidulants such as ascorbate or erythrobate, nitrite exerts a c­oncentrationdependent antimicrobial effect on clostridial spores. Nitrite does not necessarily inhibit spore swelling, but it inhibits subsequent replication of germinated spores (127). The effect of reduced pH on the antimicrobial activity of nitrites has been repeatedly determined and is well understood (393). Anaerobic conditions enhance the antimicrobial effects of nitrites; the curing agents ascorbate and isoascorbate also increase the inhibitory effects, likely by functioning as reducing agents (331). Roberts and Ingram (330) and Duncan and Foster (127) demonstrated that nitrite addition prior to heating does not increase inactivation of spores but inhibits outgrowth of vegetative cells following heating. More recently, researchers have revealed that clostridial small, acid-soluble proteins can protect spores from nitrite exposure at low concentrations but not at elevated nitrite levels (400 mM) (307). The antimicrobial effects of nitrite on microorganisms other than clostridia are variable. Tsai and Chou (400) reported that 200 mg nitrite per liter at pH 5.0 reduced E. coli O157:H7 populations by 6.5 to 7.0 log10 CFU/ml, whereas >400 mg nitrite per liter reduced the pathogen population to nondetectable levels. Increasing the pH to 7.0 or 8.0 resulted in the loss of nitrite’s antimicrobial activity at levels £1,000 mg/liter. Gill and Holley (164) found that adding 180 mg nitrite per liter of broth (pH 6.0) inhibited the growth of S. Typhimurium, Shewanella putrefaciens, Serratia grime­ sii, E. coli, Lactobacillus curvatus, Leuconostoc mes­ enteroides, S. aureus, and L. monocytogenes. Pathogen inhibition was further enhanced in systems in which nitrite was combined with EDTA or NaCl (164). Gibson and Roberts (160, 161) observed only limited inhibition of C. perfringens, Salmonella, enteropathogenic E. coli, and fecal streptococci with 400 μg/ml of nitrite in culture medium adjusted to pH 5.6 to 6.8 and containing up to 6% NaCl. Some evidence of resistance among clostridia, as well as other bacteria, has been reported. Li and McClane (234) reported that C. perfringens isolates carrying

SMP_Food Microbiology_CH30.indd

776

plasmid-borne enterotoxin-encoding genes were significantly less tolerant to 0.2 to 0.4% NaNO2 than were isolates whose enterotoxin genes were chromosomally located, an interesting phenomenon that may aid in the identification of C. perfringens isolates that pose increased foodborne disease risk. The gram-negative meat spoilage bacterium Acinetobacter was determined to have species-specific tolerance to 0.1 to 0.3% NaNO2 (346). In contrast, nisin-resistant L. monocytogenes and C. botulinum vegetative cells and spores were less resistant to 15 to 165 μg NaNO2 per ml than were wild-type cells of the pathogens (257). The antimicrobial mechanism of nitrite has been studied for more than 70 years; however, the likely targets of clostridial inhibition by nitrite have been elucidated only in the past 30 years. Woods et al. (428) showed that nitrite reduced intracellular ATP and excretion of pyruvate in vegetative cells of C. sporogenes. Since these cells oxidize pyruvate to acetate to produce ATP using the phosphoroclastic system, it was theorized that this enzyme system was inhibited by nitrite. Two enzymes in the system, pyruvate ferredoxin oxidoreductase (PFR) and ferredoxin, were suspected to be susceptible to nitrite. Ultimately, inhibition was determined to follow conversion of nitrite to NO and the subsequent interaction of NO with nonheme, protein-located iron. Woods and Wood (427) subsequently determined that the phosphoroclastic system is also inhibited by nitrite. Carpenter et al. (73) confirmed that nitrite inhibited ferredoxin and PFR in C. botulinum and Clostridium pas­ teurianum when cultures were incubated with 1,000 μg NaNO2 per ml, reporting that in vivo inhibition of ferredoxin activity was greater than that of PFR. McMindes and Siedler (259) observed complete inhibition of C. perfringens PFR by 0.0625 mM NO, whereas addition of 2.5 mM ascorbate enhanced inhibition by NO of clostridial PFR. These authors observed pyruvate decarboxylase inhibition by 0.25 to 0.5 mM NO, suggesting another enzymatic target for NO in the cell (259). Roberts et al. (331) confirmed the inhibitory activity of nitrite on the phosphoroclastic system and the ability of ascorbates to enhance the antimicrobial activity of nitrites. Cui et al. (100) provided additional evidence that NO is primarily responsible for inhibiting Clostridium spp., reporting that while 185 μM NO was required for complete inhibition of C. sporogenes in culture medium, 10 mM nitrite was required for 50% inhibition of C. sporogenes. Suggestions that PFR and ferredoxin represent the primary targets of nitrites for inhibiting clostridia are also supported by the observation that addition of iron to meats containing nitrites results in a reduction of anticlostridial activity (394).

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials Nitrite’s antimicrobial mechanism against nonsporulating bacteria may be different from that against sporeformers. Application of 10 to 25 mM nitrite inhibited both the glucose and proline active transport systems in P. aeruginosa; at levels above 1.0 mM, nitrite inhibited oxidative phosphorylation and respiration in Pseudomonas cells (340). Conversely, L. lactis and Enterococcus faecalis were relatively unaffected by nitrite at levels that significantly inhibited growth of P. aeruginosa (340). Morita et al. (268) hypothesized that similar mechanisms were involved in nitrite-directed inhibition of verotoxin-producing E. coli, observing dose-dependent reductions in ATP synthesis following incubation of cells in increasing levels of NaNO2. Meat products that may contain nitrites include bacon, corned beef, bologna, frankfurters and other luncheon meats, cured hams, fermented sausages, shelfstable canned/cured meats, and some perishable canned/ cured meat (e.g., ham). Nitrite is also used in several fish and poultry products. The concentration of nitrite used in these products varies but is generally limited to 156 μg/g in most products and 100 to 120 μg/g in bacon. Levels of nitrite that can be added to food are federally regulated. Ascorbates accelerate the curing process by reducing NO from NO2 and can also inhibit formation of nitrosamines, which are carcinogens formed by reactions of nitrite with secondary or tertiary amines (393).

para-Hydroxybenzoic Acid Esters

The antimicrobial activity of alkyl esters of p-hydroxybenzoic acid (i.e., parabens) (Fig. 30.4) was first reported in the 1920s. Esterification of the carboxyl group of benzoic acid allows the molecule to remain undissociated at up to pH 8.5, giving the parabens an effective inhibitory range of pH 3.0 to 8.0 (1). In most countries, the methyl, propyl, and heptyl parabens are approved for direct addition to foods as antimicrobials, whereas the ethyl and butyl esters are approved in some countries. The antimicrobial activity of parabens is, in general, directly proportional to the chain length of the alkyl component (106). As the alkyl chain length increases, inhibitory activity generally increases. Parabens are generally more active against fungi than against bacteria, with gram-positive bacteria being more sensitive than gram-negative bacteria (106). The primary mechanism of microbial inhibition likely involves paraben-directed interference with the cytoplasmic membrane. Bredin et al. (57) observed increased potassium efflux in E. coli cells following incubation with propyl paraben, reporting that K+ efflux rates were similar to those observed following application of other antimicrobials known to destabilize cation regulation processes (colicin A, poly-

SMP_Food Microbiology_CH30.indd

777

777

Figure 30.4  Alkyl esters of p-hydroxybenzoic acid (parabens). doi:10.1128/9781555818463.ch30f4

myxin B). Eklund (129, 131) postulated that parabens dissipate the electrochemical gradient and PMF in microbial membranes, although experiments revealed that the DY component of the PMF was not significantly affected by paraben exposure. Bargiota et al. (27) and Juneja and Davidson (203) determined that paraben resistance in gram-positive bacteria is correlated with total lipid content, profiles of headgroups, and fatty acids attached to membrane-located phospholipids. Higher contents of long-chain saturated fatty acids and the presence of anionic headgroups imparted greater paraben resistance compared with unsaturated, branched fatty acidcontaining phospholipids. Researchers theorized that resistance is a result of decreased membrane fluidity and reduced opportunity for parabens to insert their alkyl chain into the membrane (203). Previous research into the resistance of gram-negative bacteria has indicated that the LPS may screen such compounds, limiting their access to cellular membranes (148). Some gram-negative bacteria also possess innate mechanisms of paraben resistance, namely hydrolytic enzymes capable of deesterificatio­n of parabens (409, 410). Additional research has investigated membrane-associated channels and the interactions of these bacterial systems with parabens as a method of explaining the sensitivity of gram-negative bacteria to parabens. Nguyen et al. (289) determined that addition of propyl paraben, and to a lesser extent ethyl paraben, to E. coli cells containing a mechanosensitive channel functioning to maintain osmotic balance induced spontaneous opening of the channel in nonosmotically

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

778 stressful environments, resulting in fatal leakage of cytoplasmic contents. To take advantage of their respective solubility and increased activity, methyl and propyl parabens are normally used in a combination of 2 to 3:1 (methyl to propyl) in foods. The compounds may be incorporated into foods by dissolving in water, ethyl alcohol, propylene glycol, or the food itself. The n-heptyl ester is used in various beers, noncarbonated soft drinks, and fruit-based beverages, primarily to inhibit spoilage fungi. Parabens are used in a variety of foods including baked goods, beverages, fruit products and preserves, fermented foods, syrups, dressings, wines, and fillings.

Sulfites The use of sulfur and its salts as disinfectants dates back at least to the ancient Greeks, with references to its use in Homer’s Iliad (171). The salts of sulfur dioxide (SO2) include potassium sulfite (K2SO3), sodium sulfite (Na2SO3), potassium bisulfite (KHSO3), sodium bisulfite (NaHSO3), potassium metabisulfite (K2S2O5), and sodium metabisulfite (Na2S2O5). As antimicrobials, sulfites are used primarily in fruit and vegetable products to control growth of spoilage and fermentative fungi, acetic acid bacteria, and malolactic bacteria (301). Sulfites also exhibit antioxidant properties and can be used to inhibit enzymatic and nonenzymatic browning in a variety of foods. The most important factor affecting the antimicrobial activity of sulfites is pH. Sulfur dioxide and its associated salts exist as a pH-dependent mixture when aqueously dissolved: – + ¬ 2– + SO2•H2O ¬ ® HSO3 + H ® SO3 + H

Aqueous solutions of sulfur dioxide theoretically yield sulfurous acid (H2SO3), although empirical evidence indicates that the actual form of the molecule is SO2•H2O (173). As the system pH decreases, the concentration of SO2•H2O increases and the bisulfite content decreases. The pKa values for sulfur dioxide, depending upon temperature, are 1.76 to 1.90 and 7.18 to 7.20, and it is the undissociated form of the sulfites or SO2•H2O that exhibits the greatest antimicrobial activity (173, 301). The increased antimicrobial efficacy observed at low pH is likely due to the increased ability of SO2•H2O to passively diffuse through the microbial membrane (337). Like SO2, the sulfites are potent antimicrobials, especially bisulfite. Cellular targets for sulfites that have been identified include enzymes and associated cofactors, ATP, the cytoplasmic membrane, and membrane-

SMP_Food Microbiology_CH30.indd

778

associated proteins (301). In cells of S. cerevisiae, sulfites can access the cytoplasm via passive diffusion and inhibit oxidative or anaerobic consumption of glucose, reportedly by inhibiting the enzyme glyceraldehyde-3phosphate dehydrogenase (185). Bacteria are most sensitive to sulfite activity. Sulfites are used for inhibiting a variety of spoilage microbes, as well as acetic acid bacteria, LAB, and gram-negative enteric pathogens (172). Inhibition of Acetobacter spp. can occur at 100 to 200 μg sulfite per ml, whereas some LAB are inhibited in low-pH foods (£3.5) at levels of 1 to 10 μg/ml (301, 423). Metabisulfite inhibited multiple LAB, including Oenococcus oeni, and various acetic acid bacteria at levels ranging from 12.5 to 100 μg/ml (333). Banks and Board (24) found that growth of Salmonella, E. coli, and Y. enterocolitica was inhibited in sausage with metabisulfite added at levels of 15 to 200 μg/ml (free sulfite). Applying 4.18% sodium metabisulfite to surfaces of Salmonella-inoculated apple slices, followed by 6 h of dehydration, reduced Salmonella populations by 4.3 log10 CFU/g (120). In experiments designed to determine the efficacy of 4.18% sodium metabisulfite for inhibiting L. monocytogenes on peach slices, Listeria populations were reduced by 4.3 log10 CFU/g after the peaches were treated with sulfite and dehydrated for 6 h. Tong and Draughon (395) determined that at pH 4.5 the addition of 0.066% sodium bisulfite eliminated ochratoxin A synthesis by ochra molds in liquid medium at 28°C. Sulfur dioxide and the sulfites are used to control growth of undesirable microorganisms and protect the organoleptic quality of fruits, juices, wines, sausages, fresh shrimp, and pickles. In winemaking, 50 to 100 μg SO2 or sulfite per ml is added to expressed juice prior to fermentation to inhibit wild yeasts and other spoilage microbes (10). Use of sulfites in sausage manufacture delays the onset of spoilage by fungi and inhibits growth of salmonellae during product storage (24). Sulfur dioxide can restore bright colors to foods but may give a false impression of product freshness.

NATURALLY OCCURRING ANTIMICROBIAL COMPOUNDS AND SYSTEMS Many foods contain antimicrobial compounds that can extend the shelf life of the food in the natural state. Many of these compounds have been studied in depth for their antimicrobial activity and utility for food preservation. Nevertheless, several concerns have been raised regarding the use of such compounds as direct food additives. An ideal naturally occurring antimicrobial should be

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials effective enough to be added as a whole food or as an edible component (e.g., an herb or spice). Few, if any, antimicrobials exist in natural sources at concentrations high enough to inhibit microbial growth without prior purification or refining. The addition of some natural substances can lead to adverse changes in the sensory properties of foods. The ultimate challenge is to find a naturally occurring antimicrobial that can be added to a “microbiologically sensitive” food in nonpurified form or as a component of a nonsensitive food. The nonpurified food would have to contain an antimicrobial that is completely nontoxic and highly effective in controlling the growth of contaminating microbes. According to Stopforth et al. (373), significant challenges exist for the industry related to economical production, refinement, and addition of naturally occurring antimicrobials to microbiologically sensitive foods at levels useful for preservation but not high enough to induce negative changes in the sensory characteristics of foods.

Animal Sources Lactoperoxidase

Lactoperoxidase is a glycoprotein enzyme that occurs naturally in raw milk, colostrum, saliva, and other biological secretions (e.g., tears). A single heme group is located at the core of the folded protein with a ferric iron (Fe3+) atom located within (365). The bovine form of the protein contains 612 residues and has a molecular mass of ~78 kDa (including coordinated heme and carbohydrate moieties) (67). Bovine milk contains approximately 10 to 30 mg of lactoperoxidase per liter (424). Lactoperoxidase oxidizes halides (I–, Br–) or thiocyanate (SCN–) in the presence of available hydrogen peroxide (H2O2), ultimately forming hypothiocyanate (OSCN–) and hypothiocyanous acid (HOSCN), both of which possess antimicrobial activity (373). Wilkins and Board (424) provide a detailed review of the sources of SCN–, the intermediate and terminal reaction products produced by the lactoperoxidase system, and the differences between bovine and human lactoperoxidase with regard to milk and colostrum content and their sources and levels of H2O2 required for antimicrobial efficacy. Lactoperoxidase is inhibitory toward both gramnegativ­e and gram-positive bacteria, although gramnegativ­e, catalase-positive bacteria, including salmonellae and other enterics, are more sensitive and can be inactivated by lactoperoxidase in some conditions (223). Catalasenegative bacteria, including the LAB, may be inhibited but are not generally inactivated by lactoperoxidase (352, 424). Carlsson et al. (70) suggested that an oxidoreductase found in some species of Streptococcus counter­acts the activity of

SMP_Food Microbiology_CH30.indd

779

779

lactoperoxidase by oxidizing NADH, consequently reducing OSCN–. De Spiegeleer et al. (118) and Sermon et al. (353) reported that l­actoperoxidase-resistan­t E. coli possesses multiple genetic mutations that lead to a decrease in porins in the outer membrane, disallowing OSCN– access, as well as multiple oxidative stress genes not activated by peroxides but activated by reaction products of the lactoperoxidase system. Lactoperoxidase contributes to the keeping quality of raw and pasteurized milk (19, 32). The combination of pasteurization at 60°C for 15 s with naturally occurring lactoperoxidase in goat milk inhibited the growth of E. coli O157:H7 for 6 hours postprocessing, resulting in a reduction in pathogen populations of 0.7 to 1.2 log10 CFU/ ml compared with goat milk samples receiving the heat treatment but no antimicrobial (309). Adding the lactoperoxidase system to skim milk inoculated with L. mono­ cytogenes and held at 25°C resulted in a 50-h lag phase, compared with an ~3-h lag phase in nontreated control milk (47). The lactoperoxidase system also can inhibit growth of S. aureus, L. monocytogenes, E. coli O157:H7, Salmonella, and Y. enterocolitica in products such as beef, vegetable juices, milk, and liquid whole eggs (135, 396).

Chitosan

Chitosan, (1-4)-2-amino-2-deoxy-b-d-glucan, is a natural component of fungal cell walls and can be derived from chitin, a by-product of shellfish processing. Chitosan comprises a series of polymers with different ratios of glucosamine and N-acetylglucosamine. Chitosan inhibits growth of foodborne fungi and bacteria (119, 222, 326). MICs for bacteria and yeasts vary widely depending on the molecular weight of the polymer, degree of acetylation, pH, temperature, and presence of interfering compounds such as proteins and lipids (222). Nevertheless, when chitosan is combined with other antimicrobials such as organic acids or antimicrobial polypeptides, the MIC of chitosan can be reduced. Zivanovic et al. (440) reported that chitosan dissolved in acetic acid was bactericidal to S. Typhimurium and L. monocytogenes strains, although the rate of inactivation was strain and concentration dependent. Sagoo et al. (345) determined that 0.005% chitosan combined with 0.025% sodium benzoate at pH 4.5 reduced the Saccharomyces spp. population by 2.0 to 3.0 log10 CFU/ ml in 60 min at room temperature. Combining chitosan with divergicin M35, a class IIA bacteriocin fermented by Carnobacterium divergens M35, halved the MICs of antimicrobials against L. monocytogenes (35). In addition to blending chitosan with antimicrobials in foods for microbial inhibition, recent studies have incorporated chitosans in edible films with antimicrobials

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

780 for enhanced inhibition of microorganisms. Combining 2 mg of lysozyme with chitosan films significantly increased the size of the zones of inhibition against E. coli O157:H7 and L. monocytogenes and significantly reduced pathogen populations in culture medium (58). Incorporating garlic essential oil, sorbic acid, or nisin into chitosan films significantly inhibited growth of L. mono­ cytogenes, Bacillus spp., Staphylococcus, Salmonella, and E. coli (318). Park et al. (308) found that chitosanbased coatings containing potassium sorbate applied to strawberries produced significant reductions in Rhizopus and Cladosporium populations compared with controls. Coliform and aerobic bacteria counts were also significantly reduced by chitosan-based antimicrobial films. Chitosan’s antimicrobial activity and mode of action are not completely understood, although the polycationic nature of the polymer, the spacing between moieties, and the polymer’s hydrophilicity have all been identified as influencing the overall activity of the molecule and as contributors to the mechanism of microbial inhibition (222). Additionally, the ability of chitosans to chelate LPS-associated cations has been reported to play a part in antimicrobial activity, particularly in reduced-pH systems (355). Sagoo et al. (345) suggested that such interactions contribute to the inhibitory activity of sodium benzoate for yeasts. It is likely that such interactions would also have a potentiating antimicrobial effect on foodborne pathogens.

This 77- to 80-kDa glycosylated protein comprises 10 to 13% of the total egg white protein content (424). Ovotransferrin has approximately 49% sequence homology with lactoferrin (397). There are two iron-bindin­g sites, and anions such as bicarbonate or carbonate are bound alongside iron. Antimicrobial efficacy must be preceded by iron deficiency and alkaline pH (399). Although gram-negative bacteria are inhibited by ovotransferrin, gram-positive bacteria are generally more sensitive to its activity. Micrococcus and Bacillus species are especially susceptible, as are some yeasts (397, 407). Recent research has sought to enhance the antimicrobial efficacy of ovotransferrin by applying it in foods with other antimicrobials. Ko et al. (219) determined that 20 mg of ovotransferrin per ml in brain heart infusion broth completely inhibited the growth of E. coli O157: H7, as did ovotransferrin combined with 2 to 2.5 mg EDTA per ml. Combined application of ovotransferrin and EDTA on surfaces of commercial hams reduced E. coli O157:H7 populations during 13 days of storage at 10°C, although results were not significantly different from those with untreated controls (219). In experiments designed to determine the effectiveness of similar combinations of antimicrobials against L. monocyto­ genes, 20 mg ovotransferrin per ml produced a similar degree of inhibition of the pathogen in liquid medium but was not effective on surfaces of ham when applied alone or in combination with EDTA or lysozyme (218).

Avidin

Plant Sources Spices and Their Essential Oils

Avidin is a glycoprotein present in egg albumen. Its concentration varies with the hen’s age but is approximately 0.05% of the total albumen protein (397). Avidin has a molecular mass of 66 to 69 kDa and consists of four identical subunits of 128 residues, each comprising b-strands that when folded resolve to a b-barrel arrangement. It is stable to heat and a wide pH range. Avidin binds strongly to the B vitamin and enzyme cofactor biotin (four molecules of biotin bind to one molecule of avidin). While its exact biological function is not known, it is thought that avidin is a nonspecific defense mechanism, inhibiting function of biotinylated enzymes and thereby slowing the growth and spread of microbial contaminants (137, 399). Most recently, avidin has been used in the development of several molecular research tools for the rapid detection and identification of many clinical and foodborne microorganisms. Its high affinity for biotin has been exploited for the development of many acid-hybridizing probes (285).

Ovotransferrin

Another iron-chelating protein, ovotransferrin (also called conalbumin), occurs naturally in egg albumen.

SMP_Food Microbiology_CH30.indd

780

Spices are roots, bark, seeds, buds, leaves, or fruit of aromatic plants and are added to foods as flavoring agents. However, it has been known since ancient times that spices and their essential oils exhibit varying degrees of antimicrobial activity. The earliest report on the use of spices as preservatives was around 1550 b.c., when ancient Egyptians used spices for food preservation and embalming the dead. Clove, cinnamon, oregano, thyme, and to a lesser extent sage and rosemary have the greatest antimicrobial activity among the spices. The major antimicrobial components of clove (Syzygium aromaticum) and cinnamon (Cinnamomum zeylanicum) are eugenol [2-methoxy-4-(2-propenyl)phenol] and cinnamic aldehyde (3-phenyl-2-propenal), respectively (Fig. 30.5). Cinnamon contains 0.5 to 10% volatile oil, of which 75% is cinnamic aldehyde and 8% is eugenol, whereas clove contains 14 to 21% volatile oil, 95% of which is eugenol. Cinnamon and cinnamic aldehyde have antimicrobial activity against the bacteria A. hydrophila, Bacillus spp., C. jejuni, E. coli O157:

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials H7 and other verotoxigenic E. coli, Lactobacillus, L. monocytogenes, salmonellae, Shigella spp., S. aureus, and Streptococcus and the fungi Aspergillus, Candida, Penicillium, and Saccharomyces (16, 103, 150, 151, 199, 274, 405). Clove and eugenol are inhibitory to these and other gram-negative and gram-positive bacteria and foodborne fungi (16, 21, 38, 150, 151, 245, 246, 349). Mytle et al. (275) treated L. monocytogenes-inoculated chicken frankfurters with 1 to 2% clove essential oil and determined after incubation at 5 or 15°C that Listeria counts on franks treated with essential oil were significantly less than those on untreated controls (275). Singh et al. (366), however, found that treatment of franks with 1 ml of clove essential oil via dipping in aqueous solution for up to 10 min did not significantly inhibit L. monocytogenes. Eugenol has antimicrobial activity against pathogens on surfaces of cooked beef, pork, and poultry (178, 179, 372). Adding 0.3% cinnamon extract to apple juice and holding at 8°C reduced numbers of E. coli O157:H7 by ~2.0 log10 CFU/ml; when used in combination with 0.1% sodium benzoate or potassium sorbate, the pathogen populations were reduced to nondetectable levels (78). Alginate coatings containing 0.3 or 0.7% cinnamon extract or 0.5% purified eugenol and malic acid inhibited growth of S. Enteriditis, psychrophilic and mesophilic bacteria, and foodborne yeasts and molds (322). Growth of yeasts and molds was completely inhibited when films containing 0.7% cinnamon or 0.5% eugenol were applied (322).

Figure 30.5  Examples of antimicrobial compounds in spice essential oils. doi:10.1128/9781555818463.ch30f5

SMP_Food Microbiology_CH30.indd

781

781

The antimicrobial activities of oregano (Origanum vulgare) and thyme (Thymus vulgaris) have been attributed to their essential oils, which contain carvacrol [5-isopropyl-2-methylphenol] and thymol [5-methyl-2(1-methylethyl)-phenol], respectively (Fig. 30.5). The essential oils and isolated compounds have demonstrable antimicrobial activity against many bacterial and fungal spoilage and pathogenic foodborne microorganisms, including A. hydrophila, Bacillus, C. jejuni, E. coli, Enterococcus spp., lactobacilli, L. monocytogenes, pediococci, Pseudomonas spp., Salmonella, Shigella, S. aureus, Vibrio parahaemolyticus, Y. enterocolitica, Aspergillus, Candida, Geotrichum, Penicillium, Pichia, Rhodotorula, Saccharomyces, and Schizosaccharomyces pombe (5, 21, 64, 80, 81, 176, 177, 206, 244, 246). Sage (Salvia officinalis) and rosemary (Rosmarinus of­ ficinalis) also have antimicrobial activity (64). Rosemary contains borneol [endo-1,7,7-trimethylbicyclo(2.2.1)heptan-2-ol], pinene, camphene, and camphor, whereas sage contains thujone [4-methyl-1-(1-methylethyl)-bicyclo(3.1.0)-hexan-3-one]. In solid medium, sage and rosemary essential oils were especially inhibitory toward gram-positive and gram-negative bacteria, with zones of inhibition ranging from 12 to 30 mm for 13 different foodborne bacteria (49). Pandit and Shelef (305) reported that rosemary was the most effective of 18 spices added to culture medium to inhibit L. monocytogenes. Conversely, others determined that sage and rosemary were no more effective at inhibiting B. cereus in carrot juice than were oregano, clove, or thyme (408). Gutierrez et al. (177) determined that the MICs of rosemary and sage against Listeria and Staphylococcus spp. ranged from 300 to 10,000 ppm, whereas the MICs against E. coli and Pseudomonas spp. were >10,000 ppm. MICs in broth of rosemary extract were 150 to 600 μg/ml against Leuconostoc, Brochothrix, Carnobacterium, and Lactobacillus. When applied singly or in combination on fresh pork meat, no antimicrobial application significantly reduced foodborne microorganism populations (177). Similar losses of antimicrobial efficacy of sage and rosemary extracts applied in other food systems were reported previously (358). It has been suggested that antimicrobial fractions of essential oils were solubilized in lipid fractions of food products and partitioned away from contaminating microbes, a likely occurrence for many such compounds with similarities in chemical structure (358). Sweet basil (Ocimum basilicum) essential oil has anti­ microbial activity, primarily from linalool and methyl chavicol. Basil essential oil extract is inhibitory to Candida, Mucor, and Penicillium species but has little antibacterial activity (9, 224). Purified methyl chavicol

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

782 (0.1%) applied to lettuce leaves reduced A. hydrophila populations to approximately the same degree as treatment with a 125-ppm chlorine washing solution (418). Essential oils from different varieties of basil inhibit multiple types of gram-positive bacteria, with MICs ranging from 200 to 300 ppm in culture medium (72). Adding basil oil (100 ppm) to nham, a Thai dish, reduced S. Enteriditis to nondetectable levels with no regrowth of the pathogen during refrigerated storage (321). Applying basil essential oil on minced meat reduced foodborne bacterial pathogens by approximately 1.0 log10 CFU/g; however, basil essential oil was no more effective than other spice extracts applied (26). Vanillin (4-hydroxy-3-methoxybenzaldehyde) is a major constituent of the vanilla bean, the fruit of an orchid (Vanilla planifolia, Vanilla pompona, or Vanilla tahitensis). Vanillin is active against molds and some gram-positive bacteria. López-Malo et al. (243) determined that 1,500 ppm vanillin in fruit-based agars (mango, papaya, banana, apple, pineapple) significantly inhibited Aspergillus niger, A. flavus, and A. parasiti­ cus. Vanillin alone or in combination with other antimicrobials preserved strawberry purée against inoculated yeasts and background microbiota (75, 76). Addition of the combination of 900 ppm vanillin and 25 ppm citral to orange juice followed by exposure to mild heating decreased the time required for pasteurization based on a ³5.0-log10 CFU/ml reduction of L. innocua (83). A combined treatment of vanillin (700 ppm) with highhydrostatic-pressure processing (100 to 300 MPa) resulted in bacteriostatic inhibition of B. cereus in liquid whole eggs (316). Other spice essential oils with antimicrobial activity include cilantro (coriander; Coriandrum sativum), fingerroot (Boesenbergia pandurata), lemongrass (Cymbopogon c­itratus), savory (Satureja spp.), and tea tree oil (Melaleuca alternifolia) (64, 246). Relatively little antimicrobial activity has been observed for many other spice-bearing plants, including anise, bay, black pepper, cardamom, cayenne, celery, chili, curry, dill, fenugreek, ginger, juniper oil, mace, marjoram, nutmeg, orris root, paprika, sesame, spearmint, tarragon, turmeric, and white pepper (109). Because of their hydrophobicity, there is a need to increase the concentration of essential oils when used in foods versus microbiological media, especially in foods with substantial lipid content that may affect the partitioning of essential oil components into the lipid phase of the food. Protein and lipid constituents in soft cheese reduce the antimicrobial activity of carvacrol and clove oil against gram-positive and gram-negative bacterial pathogens (369). Recently reported encapsulation technologies can overcome such interactions between

SMP_Food Microbiology_CH30.indd

782

food components and antimicrobial essential oils (155, 156). Most studies on the mechanism of action of spice essential oil components have focused on the effect of the compounds on the cytoplasmic membrane of targeted microorganisms. Ultee et al. (402–404) determined that carvacrol inhibits B. cereus by depleting intracellular ATP, reducing membrane potential, and increasing membrane permeability, ultimately leading to cell death. Similarly, studies have revealed that tea tree oil and b-pinene affect potassium and proton leakage in E. coli and yeasts (64). Dissipation of PMF, ATP synthesis inhibition, and enzyme inhibition have also been observed (97, 165, 252). Fisher and Phillips (143) determined that application of citrus or bergamot essential oils significantly increased membrane permeability with a concomitant loss in intracellular glucose and acidification of cytoplasmic contents.

Onions and Garlic

Probably the best-characterized antimicrobial system in plants is found in the juice and vapors of onions (Allium cepa) and garlic (Allium sativum). Growth and toxin production by many microorganisms are inhibited by onion and garlic, including the bacteria B. cereus, C. botulinum type A, E. coli, L. plantarum, Salmonella, Shigella, and S. aureus and the fungi A. flavus, A. para­ siticus, Candida albicans, Cryptococcus, Rhodotorula, Saccharomyces, Torulopsis, and Trichosporon (39, 94, 95, 246). The major antimicrobial component of garlic is allicin (diallyl thiosulfinate; thio-2-propene-1-sulfinic acid5-allyl ester) (Fig. 30.6) (74). Allicin is formed by the action of the enzyme allinase on the substrate alliin [S-(2-propenyl)-l-cysteine sulfoxide]. The reaction occurs only when garlic cells are disrupted, bringing allinase in contact with its substrate. A similar reaction occurs in onions, except that the substrate is S-(1-propenyl)l-cysteine sulfoxide and one of the major products is thiopropanal-S-oxide. The products responsible for antimicrobial activity also contribute to the flavor and odor of both onion and garlic. In addition to antimicrobial sulfurous compounds, onions contain the phenolic compounds protocatechuic acid and catechol, which may also contribute to the antimicrobial activity of onion essential oils and extracts (416). Allicin is also inhibi-

Figure 30.6  Allicin (diallyl thiosulfinate; thio-2-propene-1sulfinic acid-5-allyl ester). doi:10.1128/9781555818463.ch30f6

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials tory to the naturally occurring microbiota in processed meats and sausages (216, 347). The mechanism of action of allicin is most likely tied to the inhibition of sulfhydryl- and disulfide-containing enzymes (39). Wills (425) determined that 0.5 mM allicin inhibited many sulfhydryl enzymes (alcohol dehydrogenase, choline esterase, choline oxidase, hexokinase, papain, succinate dehydrogenase, urease, and xanthine oxidase) and nonsulfhydryl enzymes (lactate dehydrogenase, tyrosinase, and alkaline phosphatase). Barone and Tansey (30) suggested that allicin inactivated proteins by oxidizing thiols to disulfides and inhibiting the intracellular reducing activity of glutathione and cysteine. Exposure of C. albicans to allicin resulted in a decrease in glutathione content, an increase in reactive oxygen species, and an increase in other oxidative stress-related inhibitors (232).

Hops

Resin from the flowers of the hop vine (Humulus lupulus L.) is used in the brewing industry to impart a desirable bitter flavor to beer. About 3 to 12% of the resin is composed of a-bitter acids including humulone (humulon), cohumulone, and adhumulone; b-bitter acids, including lupulone (lupulon), colupulone, xanthohumol, and adlupulone, are also present (297). Both types of bitter acids possess antimicrobial activity against bacteria and fungi at reduced aw, although the b-acids have received greater attention and been used in more applications in recent years. LAB that spoil beer, including Lactobacillus and Pediococcus spp., are resistant to the antimicrobial effects of humulone, colupulone, and trans-isohumulone, whereas strains that do not spoil beer are sensitive (141, 364). Siragusa et al. (368) determined that constant feeding of 250 ppm lupulone to chickens resulted in >3.0 log10 CFU/g less C. perfringens in cecal contents than in controls. b-Acids (0.03 to 0.5%) inhibited L. monocytogenes in broth and on frankfurter surfaces, suppressing growth of the pathogen at concentrations of ³0.05% under refrigeration conditions (359, 360). There is little research on the mechanism of the antimicrobial activity of hop acids, although multiple studies have revealed that antimicrobial activity is enhanced at reduced temperature (359). Interaction with the cytoplasmic membrane is suggested in light of reduced sensitivity of gram-positive bacteria and fungi when supplemented with lipids (226). Additionally, hop bitter acids inhibit gram-positive beer spoilage microbes by dissipating transmembrane pH gradient (364).

Isothiocyanates

Isothiocyanates (R—N=C=S) are derived from glucosinolates in cells of plants belonging to the fam-

SMP_Food Microbiology_CH30.indd

783

783

ily Brassicaceae (also Cruciferae) (cabbage, kohlrabi, Brussels sprouts, cauliflower, broccoli, kale, horseradish, mustard, turnip, and rutabaga). These compounds are synthesized by the enzyme myrosinase (thioglucoside glucohydrolase) acting on the glucosinolates when plant tissue is injured or mechanically disrupted. In addition to the allyl side group, other isothiocyanate side groups include ethyl, methyl, benzyl, and phenyl. These compounds are potent antimicrobials (113, 114). Isothiocyanates are inhibitory to fungi, yeasts, and bacteria in the range of 16 to 110 ng/ml in the vapor phase (196) and 10 to 600 μg/ml in liquid medium (254), although recent reports indicate that MICs of allyl isothiocyanate (AIT) against meat spoilage bacteria in vitro are much higher (1,000 μg/ml) (350). Inhibitory activity against bacteria varies, but gram-negative bacteria are generally more sensitive than gram-positive bacteria (292, 293, 350). Delaquis and Mazza (113) found that E. coli, L. monocytogenes, and S. Typhimurium populations decreased by 1.0 to 5.0 log10 CFU in the presence of 2,000 μg AIT per ml of air. Further examination of this effect revealed that 1,000 μg AIT per ml of air reduced E. coli O157:H7, S. Typhimurium, and L. monocyto­ genes populations by up to 6.0 log10 CFU (114). AIT can effectively reduce E. coli O157:H7 populations in fresh and further-processed beef (79, 273). Application of 20 μl AIT per liter can inhibit growth of E. coli O157: H7, L. monocytogenes, S. Typhimurium, S. aureus, S. grimesii, and Lactobacillus sakei on surfaces of cooked roast beef slices (419). The antimicrobial mechanisms of the isothiocyanates have not been fully elucidated, although as mentioned above for allicin, interaction with sulfhydryl-containing enzymes is indicated. Lin et al. (238) determined that there was increased cellular leakage from E. coli O157: H7, Salmonella, and L. monocytogenes cells treated with AIT compared with other membrane-active antibiotics. Additionally, exposure to increasing amounts of AIT decreased the enzyme activity in E. coli, presumably due to either overall inactivation of cells or inhibition of enzyme synthesis and function. Brandi et al. (51) determined that inhibition of Salmonella by cauliflower juice was terminated by the addition of cysteine, but sulfhydryl inhibition alone did not fully explain the inhibition. Recent studies have revealed that the antimicrobial activity of AIT is significantly enhanced under acidic conditions, which has been suggested to be a function of greater stability under these conditions (247). Dose-dependent inhibition of thioredoxin reductase, a sulfhydryl enzyme essential in DNA synthesis, was observed by this group, supporting the previous hypotheses related to interactions of AIT with sulfhydryl enzymes (247). While these

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

784 compounds have low sensory detection thresholds, it has been previously suggested that they may still serve as useful food antimicrobials due to the small amounts required for inhibition of targeted microorganisms.

Phenolic Compounds

Simple phenolic compounds include monophenols (e.g., p-cresol), diphenols (e.g., hydroquinone), and triphenols (e.g., gallic acid). Gallic acid occurs in plants as quinic acid esters or hydrolyzable tannins (tannic acid) (186). The use of wood smoke as an antimicrobial treatment is an example of applying simple phenols for food preservation. Smoking of foods such as meats, cheeses, fish, and poultry not only imparts desirable flavors but also has a preserving effect through drying and the deposition of simple phenols. While many chemicals are deposited on smoked foods, the major contributors of flavor and antioxidant and antimicrobial effects are phenol and cresol. Liquid smoke applied to the surface of Cheddar cheese inhibits growth of Aspergillus ory­ zae, Penicillium camemberti, and Penicillium roqueforti (421). Of the eight major phenolic compounds in liquid smoke, isoeugenol is the most effective antifungal compound, followed by m-cresol and p-cresol (421). The phenolic acids, including derivatives of phydroxybenzoic acid (protocatechuic, vanillic, gallic, syringic, and ellagic acids) and o-hydroxybenzoic acid (salicylic acid), may be found in plants and foods derived from plants. Tannic acid inhibits growth of A. hydrophila, Cronobacter sakazakii, E. coli, L. monocy­ togenes, Salmonella, S. aureus, Streptococcus faecalis, and Vibrio spp. (89, 213, 314, 341, 379). The intestinal microorganisms Bacteroides fragilis, C. perfringens, E. coli, and S. Typhimurium are inhibited by both tannic acid and propyl gallate, possibly by different mechanisms, including chelation of iron (88). Hydroxycinnamic acids (HCAs) include caffeic, pcoumaric, ferulic, and sinapic acids. They frequently occur as esters and less often as glucosides. Herald and Davidson (184) demonstrated that ferulic acid at 1,000 μg/ml and p-coumaric acid at 500 and 1,000 μg/ml inhibit the growth of B. cereus and S. aureus, but the compounds were less effective against gram-negative bacteria. Wine-associated LAB were inhibited by 50 mg ferulic acid and 200 mg p-coumaric acid per liter, reportedly as a result of significant membrane damage (154). Lawrence et al. (229) determined that ethanol extracts of Aloe vera produced inhibition zones of 15 to 23 mm against gram-positive bacteria, including Bacillus spp., S. aureus, and Streptococcus pyogenes. The p-coumaric acid and cinnamic acid fractions isolated from these extracts inhibited growth of some Bacillus spp. and S.

SMP_Food Microbiology_CH30.indd

784

pyogenes, although S. aureus was reported to not be susceptible to these purified compounds (229). Many studies of the antimicrobial efficacy of HCAs have focused on their antifungal activity. Chipley and Uraih (87) showed that ferulic acid inhibited by up to 75% the synthesis of aflatoxin B1 and G1 by A. flavus and A. parasiticus. Salicylic and trans-cinnamic acids also inhibited toxin production. The spoilage yeast Dekkera was inhibited by different HCAs in culture medium in a phenolic-dependent, pH-independent manner (182). Cells grown in culture medium with ³1 mM ferulic acid had significant morphological differences compared with controls, with cells of Dekkera anomala being small and those of Dekkera bruxellensis misshapen and wrinkled (182). Alkyl esters of phenolic acids (methyl, ethyl, propyl, and butyl acids) in liquid medium were inhibitory in an alkyl ester-dependent manner to E. coli, B. cereus, L. monocytogenes, Fusarium culmorum, and S. cerevisiae at significantly lower concentrations than the native phenolics (262). The differences in MICs between phenolic acid and related alkyl esters for the fungi are likely associated with differences in microbial membrane lipid content and profiles. Yuvamoto and Said (431) determined that 1 to 10 mM caffeine and caffeic acid inhibited the germination and slowed the growth of Aspergillus nidulans. The furocoumarins are a subclass of the coumarins and are related to the HCAs (46). These compounds, including psoralen and its derivatives, are phytoaxelins in citrus fruits, parsley, carrots, celery, and parsnips. Psoralen (purified, parsley-derived) and derivatives in combination with exposure to long-wave UV light (365 nm) are inhibitory to Listeria spp., E. coli O157:H7, and Erwinia carotovara at 0.8 to 10.0%, depending on the microorganism (250). Exposure of L. innocua, E. coli O157:H7, and S. aureus to psoralen at 10 mg/liter and UV light (365 nm, 15 W) for 2 min reduced cell populations by 99.8, 99.0, and 99.99%, respectively (40). Lactobacillus isolates (L. paracasei and L. plantarum) recovered from feta cheese-brining solutions were reduced by 99.95 and 99.94%, respectively, by exposure to psoralen at 5 mg/ liter and UV for 100 s and were reduced to nondetectable levels after 120 s of exposure (40). Wahba et al. (415) determined that an alcohol extract of parsley (0.78%) was inhibitory to S. aureus in liquid medium. Applying 6% parsley extract to cheese reduced S. aureus populations by approximately 4.0 log10 CFU/g after 2 days at 30°C, although pathogen counts did not significantly change thereafter for up to 15 days (415). These compounds inhibit microbes by interfering with DNA replication through reversible, irradiation-induced cross-linking of the furocoumarin with microbial DNA (40, 91).

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials The flavonoids include catechins, flavones, flavonols, and related glycosides (126). Proanthocyanidins or condensed tannins are polymers of favan-3-ol and are found in apples, grapes, berries, plums, sorghum, and barley, as well as other plants (186). Cushnie and Lamb (101) reviewed the antimicrobial effects of various flavonoids against fungal, viral, and bacterial pathogens, suggesting that inhibition of nucleic acid synthesis and degradative impacts on cellular membranes were essential contributors to the antimicrobial activity of flavonoids. HaraKudo et al. (180) determined the antibacterial effects of methanolic extracts from 33 plants against S. aureus and V. parahaemolyticus. Both antibacterial and antiviral activity has been observed for flavonoid extracts from different teas, propolis, and royal jelly (149, 414). Boban et al. (43) observed the antimicrobial activity of red wine and wines modified by removal of polyphenols (including flavonoids) and alcohol and by pH neutralization. The removal of polyphenols resulted in a significant loss of total flavonoids in wines but did not result in major differences in antimicrobial activity against E. coli, indicating that the polyphenols were not the primary class of compounds responsible for antimicrobial activity. It was suggested that a combination of polyphenolics and reduced pH was required for the optimal antimicrobial activity of wine. Su et al. (377, 378) demonstrated that proanthocyanidins and cranberry juice reduced populations of foodborne enteric virus surrogates, and the addition of 0.15 to 0.60 mg of proanthocyanidins or cranberry juice per ml significantly reduced viral infectivity. The cytoplasmic membrane and membrane-associated proteins are primary targets for the antimicrobial activity of phenolic compounds (106, 373). Inhibition of membrane-associated proteins may be involved in the antibacterial mechanisms, as well as inhibition of protein and nucleic acid synthesis in the cell cytoplasm (46, 101). RicoMuñoz et al. (328) determined that phenolic acids inhibited ATPase in S. aureus. Increased membrane permeability also has been reported, leading to loss of cellular metabolites, K+, intracellular RNA, and amino acids (154).

ENCAPSULATION OF FOOD ANTIMICROBIALS An innovative method for enhancing the efficacy of natural and traditional antimicrobials in foods is to nanoencapsulate the antimicrobial within another food-grade material. Research devoted to nanoencapsulation technologies has increased in recent years, as have studies seeking to apply encapsulated antimicrobials to different food products. Antimicrobial encapsulation may be considered for multiple reasons, including (i) stabilization of

SMP_Food Microbiology_CH30.indd

785

785

the antimicrobial against undesirable or deleterious reactions with food components; (ii) stabilization of volatile antimicrobials (e.g., essential oils) against rapid evaporation; (iii) improvement or slowing of the rate at which an antimicrobial is released into the food, allowing for longer exposure of microbes to the antimicrobial; and (iv) protection of the antimicrobial through processing that might otherwise inactivate the chemical (385). Key issues to consider when determining the utility of encapsulating a food antimicrobial are whether the encapsulating material(s) will interact antagonistically with the antimicrobial, the solubility profiles of the encapsulating material and antimicrobial, the type of encapsulation structure desired, and the potential impacts on product acceptability.

Liposomes and Micelles as Food Antimicrobial Encapsulation Technologies

Liposomes are formed from amphipathic phospholipids, which are naturally occurring or synthetic analogues of natural phospholipids, mixed into an aqueous system. Upon mixing of lipids in water or some other buffer, liposomes form as a thermodynamically favored structure (287, 288). Liposomes are generally spherical in shape and are composed of a lipid bilayer, but they may form other structures, depending on the phospholipids incorporated into the system, the pH, the presence of solvents or alcohols, as well as other factors (227, 426). Liposomes have been used as models for various biological membranes, as delivery systems for many pharmaceuticals, and for the encapsulation and stabilization of different food ingredients such as vitamins, enzymes, or colorants (25, 187, 190, 201, 209, 356, 385). Thapon and Brule (388) encapsulated lysozyme and nisin in liposomes, adding them to cheese to prevent spoilage by anaerobes. Degnan and Luchansky (112) observed antilisterial activity by liposomal pediocin AcH incorporated into slurries of beef tallow and meat slurries. Benech et al. (37) determined that nisin encapsulated into liposomes had antilisterial activity, whereby adding 300 IU of nisin in liposomes per g of cheese reduced L. innocua populations by 3 log10 CFU/g after 6 months of ripening and suppressed L. innocua to below detectable levels following 6 months of ripening of Cheddar cheese (36). Were et al. (422) determined that there were liposome formulation-specific differences in inhibition of L. monocytogenes by encapsulated nisin or lysozyme in liquid medium. Encapsulation of 5.0 to 10.0 IU/ml of nisin and 300 μg/ml of EDTA in liposomes comprising phosphatidylcholine and phosphatidylglycerol significantly lengthened the lag phases of L. monocytogenes and E. coli O157:H7, resulting in bacteriostatic inhibition of both pathogens over 48 h of incubation at 32°C

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

786 (384). Schmidt et al. (351) determined that addition of 50 IU/ml of liposomal nisin to fluid milks lengthened the lag phase of L. monocytogenes Scott A incubated at 5 or 20°C. Adding 100 or 500 μg/ml of nisin in liposomes to fluid milk had a similar bacteriostatic effect on L. monocytogenes when the fluid was held at 6 to 8 or at 30°C (105). Liolios et al. (239) found that encapsulation of carvacrol and thymol in phosphatidylcholineand cholesterol-containing liposomes enhanced their in vitro antimicrobial activity against Staphylococcus spp., Candida, Pseudomonas, and L. monocytogenes. Micelles and reverse micelles comprise amphiphilic detergents that possess both hydrophilic and hydrophobic regions that self-orient in systems so as to afford greatest protection to the hydrophobic portions of the surfactant from interaction with water. In the case of micelles, this generally results in a system in which surfactant tails, being hydrophobic, are buried within the micelle and the hydrophilic headgroups protrude into the water phase. Conversely, a reverse micelle is formed when the tails are hydrophilic and the headgroups are buried within the micelle’s core so as to minimize exposure to water (156). While micelles do not generally offer the encapsulation efficiencies of liposomes, they can be made very small and uniform and can deliver with great efficacy some antimicrobials, in particular the phenolic extracts from spice essential oils (258). Application of 0.5 or 0.9% micelle-encapsulated eugenol to L. monocytogenes and E. coli O157:H7 at pH 5.0 to 7.0 completely inhibited the growth of these pathogens at 22 and 32°C; the growth of L. monocytogenes at 10°C (pH 7.0) was completely inhibited in the presence of 0.2, 0.5, and 0.9% micellar eugenol (156). Likewise, micelles containing 0.3 to 0.4% carvacrol or eugenol completely inhibited the growth of L. monocytogenes and E. coli O157:H7 (155). The addition of eugenol-containing micelles to skim or whole milk inoculated with L. monocy­ togenes or E. coli O157:H7 resulted in bacteriostatic or bactericidal inhibition of the pathogens in a milkfat- and pathogen-specific manner (157). Incubating foodborne pathogen biofilms in the presence of micellar eugenol or carvacrol at 0.3 to 0.9% resulted in strain-specific inhibition of both pathogens, and all but one E. coli O157: H7 strain were rapidly inhibited by all carvacrol micelle treatments (315). Use of a sodium lactate- and monolaurin-containing microemulsion produced concentrationdependent inhibition of B. subtilis in culture medium, with complete inhibition of the bacterium following incubation in an 800-ppm microemulsion (432). Similar results were observed by this group using the same microemulsion, with E. coli O157:H7, S. aureus, Candida, and Saccharomyces populations being reduced to nondetect-

SMP_Food Microbiology_CH30.indd

786

able levels in as little as 60 to 120 min when incubated in the presence of the microemulsion (433, 435).

CONCLUSIONS Traditional antimicrobials and, increasingly, natural food antimicrobials are important tools for preserving food from microbiological spoilage and the growth of pathogens. Many researchers have observed a general trend among consumers asking for or demanding foods with fewer synthetic additives or foods with natural ingredients that are deemed more environmentally friendly. The natural antimicrobials will likely continue to grow in popularity in the future. Given recent changes in U.S. laws and regulations related to the safety of fresh and minimally processed produce, beef and other red meats, and other foods, and the occurrence of multiple foodborne disease outbreaks associated with foods not previously thought to support the growth or survival of some foodborne pathogens, it is likely that new applications of existing food antimicrobials will appear in the scientific literature and marketplace. Nevertheless, as seen with the delivery of antimicrobials to foods through encapsulation, the future of food antimicrobials most likely rests in the development of novel applications of existing antimicrobials, including encapsulation, incorporation into edible polymers, and use of combinations of antimicrobials capable of synergistic inhibition of foodborne microorganisms. Despite the extensive research already completed on the sources of antimicrobials, their spectra of activity, and the levels required for successful inhibition of foodborne bacteria and fungi, more research is still needed to better elucidate the mechanisms of antimicrobial activity of many of the chemicals discussed in this chapter. This is clearly the situation for many of the plant-derived antimicrobials. In addition to validating the activity and elucidating the mechanistic features of antimicrobials, they will have to be proven toxicologically safe. Demonstrating the efficacy of antimicrobial compounds in food products at concentrations that do not have adverse sensory effects, as well as controlling the cost of these interventions, are likely the greatest hurdles to their future application.

References   1. Aalto, T. R., M. C. Firman, and N. E. Rigler. 1953. p-Hydroxybenzoic acid esters as preservatives. I. Uses, antibacterial and antifungal studies, properties and determination. J. Am. Pharm. Assoc. 42:449–458.   2. Alakomi, H.-L., E. Skyttä, M. Saarela, T. MattilaSandholm, K. Latva-Kala, and I. M. Helander. 1999. Lactic acid permeabilizes gram-negative bacteria by dis-

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials

  3.   4.

  5.

  6.   7.

  8.

  9.

  10.   11.

  12.   13.   14.

  15.

  16.   17.

rupting the outer membrane. Appl. Environ. Microbiol. 66:2001–2005. Al-Dagal, M. M., and W. A. Bazaraa. 1999. Extension of shelf life of whole and peeled shrimp with organic acid salts and bifidobacteria. J. Food Prot. 62:51–56. Al-Holy, M. A., L. F. Castro, and H. M. Al-Qadiri. 2010. Inactivation of Cronobacter spp. (Enterobacter sakazakii) in infant formula using lactic acid, copper sulfate and monolaurin. Lett. Appl. Microbiol. 50:246–251. Aligiannis, N., E. Kalpoutzakis, S. Mitaku, and I. B. Chinou. 2001. Composition and antimicrobial activity of the essential oils of two Origanum species. J. Agric. Food Chem. 49:4168–4170. Al-Nabulsi, A. A., and R. A. Holley. 2005. Effect of bovine lactoferrin against Carnobacterium viridans. Food Microbiol. 22:179–187. Altieri, C., D. Cardillo, A. Bevilacqua, and M. Sinigaglia. 2007. Inhibition of Aspergillus spp. and Penicillium spp. by fatty acids and their monoglycerides. J. Food Prot. 70:1206–1212. Álvarez-Ordóñez, A., A. Fernández, M. López, and A. Bernardo. 2009. Relationship between membrane fatty acid composition and heat resistance of acid and cold stressed Salmonella senftenberg CECT 4384. Food Microbiol. 26:347–353. Amber, K., A. Aijaz, X. Immaculata, K. A. Luqman, and M. Nikhat. 2010. Anticandidal effect of Ocimum sanc­ tum essential oil and its synergy with fluconazole and ketoconazole. Phytomedicine 17:921–925. Amerine, M. A., and M. A. Joslyn. 1970. Table Wines: the Technology of Their Production, 2nd ed. University of California Press, Berkeley, CA. Ananou, S., A. Baños, M. Maqueda, M. MartínezBueno, A. Gálvez, and E. Valdivia. 2010. Effect of combined physico-chemical treatments based on enterocin AS-48 on the control of Listeria monocytogenes and Staphylococcus aureus in a model cooked ham. Food Control 21:478–486. Anonymous. 1999. Sorbic acid and potassium sorbate for preserving freshness. Public. ZS-1D. Eastman Chemical Co., Kingsport, TN. Anonymous. 2007. Levulinate validated for use as an antilisterial agent. AMI Foundation News 9:1, 6. Aran, N. 2001. The effect of calcium and sodium lactates on growth from spores of Bacillus cereus and Clostridum perfringens in a ‘sous-vide’ beef goulash under temperature abuse Int. J. Food Microbiol. 63:117–123. Arseneault, M., S. Bédard, M. Boulet-Audet, and M. Pézolet. 2010. Study of the interaction of lactoferricin B with phosopholipid monolayers and bilayers. Langmuir 26:3468–3478. Aureli, P., A. Costantini, and S. Zolea. 1992. Antimicrobial activity of some plant essential oils against Listeria monocytogenes. J. Food Prot. 55:344–348. Aymerich, T., A. Jofre, M. Garriga, and M. Hugas. 2005. Inhibition of Listeria monocytogenes and Salmonella by natural antimicrobials and high hydrostatic pressure in sliced cooked ham. J. Food Prot. 68:173–177.

SMP_Food Microbiology_CH30.indd

787

787

  18. Ayres, H. M., J. R. Furr, and A. D. Russell. 1998. Effect of divalent cations on permeabilizer-induced lysozyme lysis of Pseudomonas aeruginosa. Lett. Appl. Microbiol. 27:372–374.   19. Babu, R., M. L. Varshney, and D. S. Sog. 2004. Pre­ servation of raw milk with lactoperoxidase system. J. Food Sci. Technol. 41:42–46.   20. Bae, Y.-M., and S.-Y. Lee. 2010. Effect of pre-exposure to sodium chloride on the resistance of pathogens to thermal and acid stresses. J. Food Safety 30:1016–1025.   21. Bagamboula, C. F., M. Uyttendaele, and J. Debevere. 2003. Antimicrobial effect of spices and herbs on Shigella sonnei and Shigella flexneri. J. Food Prot. 66:668–673.   22. Bala, M. F. A., and D. L. Marshall. 1996. Testing matrix, inoculum size, and incubation temperature affect monolaurin activity against Listeria monocytogenes. Food Microbiol. 13:467–473.   23. Bala, M. F. A., and D. L. Marshall. 1996. Use of doublegradient plates to study combined effects of salt, pH, monolaurin, and temperature on Listeria monocyto­ genes. J. Food Prot. 59:601–607.   24. Banks, J. G., and R. G. Board. 1982. Sulfite-inhibition of Enterobacteriaceae including Salmonella in British fresh sausage and in culture systems. J. Food Prot. 45: 1292–1297.   25. Banville, C., J.-C. Vuillemard, and C. Lacroix. 2000. Comparison of different methods for fortifying Cheddar cheese with vitamin D. Int. Dairy J. 10:375–382.   26. Barbosa, L. N., V. L. Rall, A. A. Fernandes, P. I. Ushimaru, I. da Silva Probst, and A. Fernandes, Jr. 2009. Essential oils against foodborne pathogens and spoilage bacteria in minced meat. Foodborne Pathog. Dis. 6:725–728.   27. Bargiota, E. E., E. Rico-Muñoz, and P. M. Davidson. 1987. Lethal effect of methyl and propyl parabens as related to Staphylococcus aureus lipid composition. Int. J. Food Microbiol. 4:257–266.   28. Bari, M. L., D. O. Ukuku, T. Kawasaki, Y. Inatsu, K. Isshiki, and S. Kawamoto. 2005. Combined efficacy of nisin and pediocin with sodium lactate, citric acid, phytic acid, and potassium sorbate and EDTA in reducing the Listeria monocytogenes population of inoculated fresh-cut produce. J. Food Prot. 68:1381–1387.   29. Barmpalia, I. M., I. Geornaras, K. E. Belk, J. A. Scanga, P. A. Kendall, G. C. Smith, and J. N. Sofos. 2004. Control of Listeria monocytogenes on frankfurters with antimicrobials in the formulation and by dipping in organic acid solutions. J. Food Prot. 67:2456–2464.   30. Barone, F. E., and M. R. Tansey. 1977. Isolation, purification, identification, synthesis and kinetics of the activity of the anticandidal component of Allium sati­ vum, and a hypothesis for its mode of action. Mycologia 69:793–825.   31. Barreteau, H., L. Mandoukou, I. Adt, I. Gaillard, B. Courtois, and J. Courtois. 2004. A rapid method for determining the antimicrobial activity of novel natural molecules. J. Food Prot. 67:1961–1964.   32. Barrett, N. E., A. S. Grandison, and M. J. Lewis. 1999. Contribution of the lactoperoxidase system to

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

788

  33.

  34.

  35.

  36.

  37.

  38.

  39.

  40.

  41.

  42.

  43.

  44.

the keeping quality of pasteurized milk. J. Dairy Res. 66:73–80. Basaran-Akgul, N., J. J. Churey, P. Basaran, and R. W. Worobo. 2009. Inactivation of different strains of Escherichia coli O157:H7 in various apple ciders treated with dimethyl dicarbonate (DMDC) and sulfur dioxide (SO2) as an alternative method. Food Microbiol. 26:8–15. Bedie, G. K., J. Samelis, J. N. Sofos, K. E. Belk, J. A. Scanga, and G. C. Smith. 2001. Antimicrobials in the formulation to control Listeria monocytogenes postprocessing contamination on frankfurters stored at 4°C in vacuum packages. J. Food Prot. 64:1949–1955. Benabbou, R., A. Zihler, M. Desbiens, E. Kheadr, M. Subirade, and I. Fliss. 2009. Inhibition of Listeria mono­ cytogenes by a combination of chitosan and divergicin M35. Can. J. Microbiol. 55:347–355. Benech, R.-O., E. E. Kheadr, C. Lacroix, and I. Fliss. 2002. Antibacterial activities of nisin Z encapsulated in liposomes or produced in situ by mixed culture during cheddar cheese ripening. Appl. Environ. Microbiol. 68:5607–5619. Benech, R.-O., E. E. Kheadr, R. Laridi, C. Lacroix, and I. Fliss. 2002. Inhibition of Listeria innocua in cheddar cheese by addition of nisin Z in liposomes or by in situ production in mixed culture. Appl. Environ. Microbiol. 68:3683–3690. Bennis, S., F. Chami, N. Chami, T. Bouchikhi, and A. Remmal. 2004. Surface alteration of Saccharomyces cerevisiae induced by thymol and eugenol. Lett. Appl. Microbiol. 38:454–458. Beuchat, L. R. 1994. Antimicrobial properties of spices and their essential oils, p. 167–179. In V. M. Dillon and R. G. Board (ed.), Natural Antimicrobial Systems and Food Preservation. CAB International, Wallingford, United Kingdom. Bintsis, T., E. Litopoulou-Tzanetaki, R. Davies, and R. K. Robinson. 2000. The antimicrobial effects of long-wave ultra-violet light and furocoumarins on some micro-organisms that occur in cheese brines. Food Microbiol. 17:687–695. Birk, T., A. C. Grønlund, B. B. Christensen, S. Knøchel, K. Lohse, and H. Rosenquist. 2010. Effect of organic acids and marination ingredients on the survival of Campylobacter jejuni on meat. J. Food Prot. 73:258–265. Blom, H., E. Nerbink, R. Dainty, T. Hagtvedt, E. Borch, H. Nissen, and T. Nesbakken. 1997. Addition of 2.5% lactate and 0.25% acetate controls growth of Listeria monocytogenes in vacuum-packed, sensory acceptable servelat sausage and cooked ham stored at 4°C. Int. J. Food Microbiol. 38:71–76. Boban, N., M. Tonkic, D. Budimir, D. Modun, D. Sutlovic, V. Punda-Polic, and M. Boban. 2010. Antimicrobial effects of wine: separating the role of polyphenols, pH, ethanol, and other wine components. J. Food Sci. 75: M322–M326. Boland, J. S., P. M. Davidson, B. Bruce, and J. Weiss. 2004. Cations reduce antimicrobial efficacy

SMP_Food Microbiology_CH30.indd

788

  45.   46.

  47.

  48.   49.

  50.

  51.

  52.   53.

  54.   55.

  56.

  57.   58.

  59.

of lysozyme-chelator combinations. J. Food Prot. 67: 285–294. Boland, J. S., P. M. Davidson, and J. Weiss. 2003. Enhanced inhibition of Escherichia coli O157:H7 by lysozyme and chelators. J. Food Prot. 66:1783–1789. Borges, F., F. Roleira, N. Milhazes, L. Santana, and E. Uriarte. 2005. Simple coumarins and analogues in medicinal chemistry: occurrence, synthesis and biological activity. Curr. Med. Chem. 12:887–916. Boussouel, N., F. Mathieu, A.-M. Revol-Junelles, and J.-B. Millière. 2000. Effects of combinations of lactoperoxidase system and nisin on the behaviour of Listeria monocytogenes ATCC 15313 in skim milk. Int. J. Food Microbiol. 61:169–175. Bower, C. K., and M. A. Daeschel. 1999. Resistance responses of microorganisms in food environments. Int. J. Food Microbiol. 50:33–44. Bozin, B., N. Mimica-Dukic, I. Samojlik, and E. Jovin. 2007. Antimicrobial and antioxidant properties of rosemary and sage (Rosmarinus officinalis L. and Salvia officinalis L., Lamiaceae) essential oils. J. Agric. Food Chem. 55:7879–7885. Bracey, D., C. D. Holyoak, and P. J. Coote. 1998. Comparison of the inhibitory effect of sorbic acid and amphotericin B on Saccharomyces cerevisiae: is growth inhibition dependent on reduced intracellular pH? J. Appl. Microbiol. 85:1056–1066. Brandi, G., G. Amagliani, G. F. Schiavano, M. De Santi, and M. Sisti. 2006. Activity of Brassica oleracea leaf juice on foodborne pathogenic bacteria. J. Food Prot. 69:2274–2279. Brandt, A. L. 2009. In vitro inhibition of Listeria mono­ cytogenes by novel combinations of food antimicrobials. M.S. thesis. Texas A&M University, College Station, TX. Brandt, A. L., A. Castillo, K. B. Harris, J. T. Keeton, M. D. Hardin, and T. M. Taylor. 2010. Inhibition of Listeria monocytogenes by food antimicrobials applied singly and in combination. J. Food Sci. 75:M557–M563. Branen, A. L., P. M. Davidson, and B. Katz. 1980. Antimicrobial properties of phenolic antioxidants and lipids. Food Technol. 34:42–53, 63. Branen, J., and P. M. Davidson. 2000. Activity of hydrolyzed lactoferrin against foodborne pathogenic bacteria in growth media: the effect of EDTA. Lett. Appl. Microbiol. 30:233–237. Branen, J. K., and P. M. Davidson. 2004. Enhancement of nisin, lysozyme, and monolaurin antimicrobial activities by ethylenediaminetetraacetic acid and lactoferrin. Int. J. Food Microbiol. 90:63–74. Bredin, J., A. Davin-Régli, and J.-M. Pagès. 2005. Propyl paraben induces potassium efflux in Escherichia coli. J. Antimicrob. Chemother. 55:1013–1015. Brown, C. A., B. Wang, and J.-H. Oh. 2008. Anti­ microbial activity of lactoferrin against foodborne pathogenic bacteria incorporated into edible chitosan film. J. Food Prot. 71:319–324. Buchanan, R. L., and S. G. Edelson. 1999. pH-dependent stationary-phase acid resistance response of enterohe-

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials

  60.

  61.   62.   63.

  64.   65.

  66.

  67.

  68.

  69.

  70.

  71.

  72.

  73.

morrhagic Escherichia coli in the presence of various acidulants. J. Food Prot. 62:211–218. Buchanan, R. L., and M. H. Golden. 1994. Interaction of citric acid concentration and pH on the kinetics of Listeria monocytogenes inactivation. J. Food Prot. 57: 567–570. Bullerman, L. B. 1983. Effects of potassium sorbate on growth and aflatoxin production by Aspergillus parasiti­ cus and Aspergillus flavus. J. Food Prot. 46:940–942. Bullerman, L. B. 1984. Effects of potassium sorbate on growth and patulin production by Penicillium patulum and Penicillium roqueforti. J. Food Prot. 47:312–315. Burnett, S. L., J. H. Chopskie, T. C. Podtburg, T. A. Gutzmann, S. E. Gilbreth, and P. W. Bodnaruk. 2007. Use of octanoic acid as a postlethality treatment to reduce Listeria monocytogenes on ready-to-eat meat and poultry products. J. Food Prot. 70:392–398. Burt, S. 2004. Essential oils: their antibacterial properties and potential applications in foods—a review. Int. J. Food Microbiol. 94:223–253. Byelashov, O. A., J. M. Adler, I. Geornaras, K. Y. Ko, K. E. Belk, G. C. Smith, and J. N. Sofos. 2010. Evaluation of brining ingredients and antimicrobials for effects on thermal destruction of Escherichia coli O157:H7 in a meat model system. J. Food Sci. 75:M209–M217. Cacace, J. E., P. J. Delaquis, and G. Mazza. 2002. Effect of chemical inhibitors and storage temperature on the quality of fresh-cut potatoes. J. Food Qual. 25: 181–195. Cals, M.-M., P. Mailliart, G. Brignon, P. Anglade, and B. R. Dumas. 1991. Primary structure of bovine lactoperoxidase, a fourth member of a mammailian heme peroxidase family. Eur. J. Biochem. 198:733–739. Carlson, B. A., I. Geornaras, Y. Yoon, J. A. Scanga, J. N. Sofos, G. C. Smith, and K. E. Belk. 2008. Studies to evaluate chemicals and conditions with low-pressure applications for reducing microbial counts on cattle hides. J. Food Prot. 71:1343–1348. Carlson, B. A., J. Ruby, G. C. Smith, J. N. Sofos, G. R. Bellinger, W. Warren-Serna, B. Centrella, R. A. Rowling, and K. E. Belk. 2008. Comparison of antimicrobial efficacy of multiple beef hide decontamination strategies to reduce levels of Escherichia coli O157:H7 and Salmonella. J. Food Prot. 71:2223–2227. Carlsson, J., Y. Iwami, and T. Yamada. 1983. Hydrogen peroxide excretion by oral streptococci and effect of lactoperoxidase-thiocyanate-hydrogen peroxide. Infect. Immun. 40:70–80. Carneiro de Melo, A. M. S., C. A. Cassar, and R. J. Miles. 1998. Trisodium phosphate increases sensitivity of gram-negative bacteria to lysozyme and nisin. J. Food Prot. 61:839–844. Carovic´-Stanko, K., S. Orlic´, O. Politeo, F. Strikic´, I. Kolak, M. Milos, and Z. Satovic. 2010. Composition and antibacterial activities of essential oils of seven Ocimum taxa. Food Chem. 119:196–201. Carpenter, C. E., D. S. A. Reddy, and D. P. Cornforth. 1987. Inactivation of clostridial ferredoxin and pyru-

SMP_Food Microbiology_CH30.indd

789

  74.

  75.   76.   77.

  78.

  79.

  80.

  81.

  82.

  83.

  84.

  85.

  86.

  87.

789

vate-ferredoxin oxidoreductase by sodium nitrite. Appl. Environ. Microbiol. 53:549–552. Cavallito, C. J., and J. H. Bailey. 1944. Allicin, the antibacterial principal of Allium sativum. I. Isolation, physical properties and antibacterial action. J. Am. Chem. Soc. 16:1950–1951. Cerrutti, P., and S. M. Alzamora. 1996. Inhibitory effects of vanillin on some food spoilage yeasts in laboratory media and fruit purées. Int. J. Food Microbiol. 29:379–386. Cerrutti, P., S. M. Alzamora, and S. L. Vidales. 1997. Vanillin as an antimicrobial for producing shelf-stable strawberry purées. J. Food Sci. 62:608–610. ˇ Cervenka, L., Z. Malíková, I. Zachová, and J. Vytrˇasová. 2004. The effect of acetic acid, citric acid, and trisodium citrate in combination with different levels of water activity on the growth of Arcobacter butzleri in culture. Folia Microbiol. 49:8–12. Ceylan, E., D. Y. C. Fung, and J. R. Sabah. 2004. Antimicrobial activity and synergistic effect of cinnamon with sodium benzoate or potassium sorbate in controlling Escherichia coli O157:H7 in apple juice. J. Food Sci. 69:M102–M106. Chacon, P. A., R. A. Buffo, and R. A. Holley. 2006. Inhibitory effects of microencapsulated allyl isothiocyanate (AIT) against Escherichia coli O157:H7 in refrigerated, nitrogen packed, finely chopped beef. Int. J. Food Microbiol. 107:231–237. Chami, F., N. Chami, S. Bennis, T. Bouchikhi, and A. Remmal. 2005. Oregano and clove essential oils induce surface alteration of Saccharomyces cerevisiae. Phytother. Res. 19:405–408. Chami, N., S. Bennis, F. Chami, A. Aboussekhra, and A. Remmal. 2005. Study of anticandidal activity of carvacrol and eugenol in vitro and in vivo. Oral Microbiol. Immunol. 20:106–111. Chang, S.-S., M. Redondo-Solano, and H. Thippareddi. 2010. Inactivation of Escherichia coli O157:H7 and Salmonella spp. on alfalfa seeds by caprylic acid and monocaprylin. Int. J. Food Microbiol. 144:141–146. Char, C. D., S. N. Guerrero, and S. M. Alzamora. 2010. Mild thermal process combined with vanillin plus citral to help shorten the inactivation time for Listeria innocua in orange juice. Food Bioprocess. Technol. 3:752–761. Chen, N., and L. A. Shelef. 1992. Relationship between water activity, salts of lactic acid, and growth of Listeria monocytogenes in a meat model system. J. Food Prot. 55:574–578. Cherrington, C. A., M. Hinton, and I. Chopra. 1990. Effect of short-chain organic acids on macromolecular synthesis in Escherichia coli. J. Appl. Bacteriol. 68: 69–74. Chipley, J. R. 2005. Sodium benzoate and benzoic acid, p. 11–48. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. Chipley, J. R., and N. Uraih. 1980. Inhibition of Aspergillus growth and aflatoxin release by derivatives of benzoic acid. Appl. Environ. Microbiol. 40:352.

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

790   88. Chung, K. T., Z. Lu, and M. W. Chou. 1998. Mechanism of inhibition of tannic acid and related compounds on the growth of intestinal bacteria. Food Chem. Toxicol. 36:1053–1060.   89. Chung, K. T., and C. A. Murdock. 1991. Natural systems for preventing contamination and growth of microorganisms in foods. Food Microstruct. 10:361–374.   90. Chung, W., and R. E. W. Hancock. 2000. Action of lysozyme and nisin mixture against lactic acid bacteria. Int. J. Food Microbiol. 60:25–32.   91. Cimino, G. D., H. B. Gamper, S. T. Isaacs, and J. E. Hearst. 1985. Psoralens as photoactive probes of nucleic acid structure and function. Ann. Rev. Biochem. 54:1151–1193.   92. Combina, M., A. M. Dalcero, E. Varsavsky, and S. Chulze. 1999. Effects of food preservatives on Alternaria alternata growth and tenuazonic acid production. Food Addit. Contam. 16:433–437.   93. Comes, J. E., and R. B. Beelman. 2002. Addition of fumaric acid and sodium benzoate as an alternative method to achieve a 5-log reduction of Escherichia coli O157:H7 populations in apple cider. J. Food Prot. 65:476–483.   94. Conner, D. E., and L. R. Beuchat. 1984. Effects of essential oils from plants on growth of food spoilage yeasts. J. Food Sci. 49:429–434.   95. Conner, D. E., L. R. Beuchat, R. E. Worthington, and H. L. Hitchcock. 1984. Effects of essential oils and oleoresins of plants on ethanol production, respiration and sporulation of yeasts. Int. J. Food Microbiol. 1:63–74.   96. Costa, A., A. Barata, M. Malfeito-Ferreira, and V. Loureiro. 2008. Evaluation of the inhibitory effect of dimethyl dicarbonate (DMDC) against wine microorganisms. Food Microbiol. 25:422–427.   97. Cox, S. D., C. M. Mann, J. L. Markham, H. C. Bell, J. E. Gustafson, J. R. Warmington, and S. G. Wyllie. 2000. The mode of antimicrobial action of the essential oil of Melaleuca alternifolia (tea tree oil). J. Appl. Microbiol. 88:170–175.   98. Critzer, F. J., D. H. D’Souza, and D. A. Golden. 2008. Transcription analysis of the stx1, marA, and eaeA genes in Escherichia coli O157:H7 treated with sodium benozate. J. Food Prot. 71:1469–1474.   99. Critzer, F. J., D. H. D’Souza, A. M. Saxton, and D. A. Golden. 2010. Increased transcription of the phosphatespecfic transport system of Escherichia coli O157: H7 after exposure to sodium benzoate. J. Food Prot. 73:819–824. 100. Cui, X., C. L. Joannou, M. N. Hughes, and R. Cammack. 1992. The bacteriocidal effects of transition metal complexes containing the NO+ group on the food-spoilage bacterium Clostridium sporogenes. FEMS Microbiol. Lett. 98:67–70. 101. Cushnie, T. P. T., and A. J. Lamb. 2005. Antimicrobial activity of flavonoids. Int. J. Antimicrob. Agents 26: 343–356. 102. Cutter, C. N., and G. R. Siragusa. 1994. Efficacy of organic acids against Escherichia coli O157:H7 attached

SMP_Food Microbiology_CH30.indd

790

to beef carcass tissue using a pilot scale model carcass washer. J. Food Prot. 57:97–103. 103. Daferera, D. J., B. N. Ziogas, and M. G. Polissiou. 2000. GC-MS analysis of essential oils from some Greek aromatic plants and their fungitoxicity on Penicillium digitatum. J. Agric. Food Chem. 48:2576–2581. 104. Dai, Y., M. D. Normand, J. Weiss, and M. Peleg. 2010. Modeling the efficacy of triplet antimicrobial combinations: yeast suppression by lauric arginate, cinnamic acid, and sodium benzoate or potassium sorbate as a case study. J. Food Prot. 73:515–523. 105. da Silva Malheiros, P., D. J. Daroit, N. Pesce da Silveira, and A. Brandelli. 2010. Effect of nanovesicle-encapsulated nisin on growth of Listeria monocytogenes in milk. Food Microbiol. 27:175–178. 106. Davidson, P. M. 2005. Parabens, p. 291–304. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 107. Davidson, P. M., and A. L. Branen. 2005. Food antimicrobials—an introduction, p. 1–10. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Food, 3rd ed. CRC Press, Boca Raton, FL. 108. Davidson, P. M., and M. A. Harrison. 2002. Resistance and adaptation to food antimicrobials, sanitizers, and other process controls. Food Technol. 56:69–78. 109. Davidson, P. M., and A. S. Naidu. 2000. Phytophenols, p. 266–294. In A. S. Naidu (ed.), Natural Food Antimicrobial Systems. CRC Press, Boca Raton, FL. 110. de Carvalho, C. C., and P. Fernandes. 2010. Production of metabolites as bacterial responses to the marine environment. Mar. Drugs 8:705–727. 111. Degnan, A. J., C. W. Kaspar, W. S. Otwell, M. L. Tamplin, and J. B. Luchansky. 1994. Evaluation of lactic acid bacterium fermentation products and foodgrade chemicals to control Listeria monocytogenes in blue crab (Callinectes sapidus) meat. Appl. Environ. Microbiol. 60:3198–3203. 112. Degnan, A. J., and J. B. Luchansky. 1992. Influence of beef tallow and muscle on the antilisterial activity of pediocin AcH and liposome-encapsulated pediocin AcH. J. Food Prot. 55:552–554. 113. Delaquis, P. J., and G. Mazza. 1995. Antimicrobial properties of isothiocyanates in food preservation. Food Technol. 49:73–84. 114. Delaquis, P. J., and P. L. Sholberg. 1997. Antimicrobial activity of gaseous allyl isothiocyanate. J. Food Prot. 60:943–947. 115. Del Olmo, A., J. Calzada, and M. Nuñez. 2010. Antimicrobial efficacy of lactoferrin, its amidated and pepsin-digested derivatives, and their combinations, on Escherichia coli O157:H7 and Serratia liquefaciens. Lett. Appl. Microbiol. 52:9–14. 116. del Río, E., B. González de Caso, M. Prieto, C. AlonsoCalleja, and R. Capita. 2008. Effect of poultry decontaminants concentration on growth kinetics for pathogenic and spoilage bacteria. Food Microbiol. 25:888–894. 117. del Río, E., R. Muriente, M. Prieto, C. Alonso-Calleja, and R. Capita. 2007. Effectiveness of trisodium phosphate,

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials acidified sodium chlorite, citric acid, and peroxyacids against pathogenic bacteria on poultry during refrigerated storage. J. Food Prot. 70:2063–2071. 118. De Spiegeleer, P., J. Sermon, K. Vanoirbeek, A. Aertsen, and C. W. Michiels. 2005. Role of porins in sensitivity of Escherichia coli to antibacterial activity of the lactoperoxidase enzyme system. Appl. Environ. Microbiol. 71:3512–3518. 119. Devlieghere, F., A. Vermeulen, and J. M. Debevere. 2004. Chitosan: antimicrobial activity, interactions with food components and applicability as a coating on fruit and vegetables. Food Microbiol. 21:703–714. 120. DiPersio, P. A., P. A. Kendall, M. Calicioglu, and J. N. Sofos. 2003. Inactivation of Salmonella during drying and storage of apple slices treated with acidic or sodium metabisulfite solutions. J. Food Prot. 66:2245–2251. 121. Divol, B., P. Strehaiano, and A. Lonvaud-Funel. 2005. Effectiveness of dimethyldicarbonate to stop alcoholic fermentation in wine. Food Microbiol. 22:169–178. 122. Dock, L. L., J. D. Floros, and R. H. Linton. 2000. Heat inactivation of Escherichia coli O157:H7 in apple cider containing malic acid, sodium benzoate, and potassium sorbate. J. Food Prot. 63:1026–1031. 123. Dong, T., and H. E. Schellhorn. 2010. Role of RpoS in virulence of pathogens. Infect. Immun. 78:887–897. 124. Doores, S. 2005. Organic acids, p. 91–142. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 125. Drosinos, E. H., P. N. Skandamis, and M. Mataragas. 2009. Antimicrobial treatment, p. 255–298. In F. Toldrá (ed.), Safety of Meat and Processed Meat. Springer, New York, NY. 126. Duncan, C. E. 2010. Factors influencing the stability and marketability of a novel, phytochemical-rich oil from the açai palm fruit (Euterpe Oleracea, Mart.). Dissertation. Texas A&M University, College Station, TX. 127. Duncan, C. L., and E. M. Foster. 1968. Effect of sodium nitrite, sodium chloride, and sodium nitrate on germination and outgrowth of anaerobic spores. Appl. Microbiol. 16:406–411. 128. Ehrenberg, L., I. Fedorscsak, and F. Solymosy. 1976. Diethyl pyrocarbonate in nucleic acid research. Prog. Nucleic Acid Res. Mol. Biol. 16:189–262. 129. Eklund, T. 1980. Inhibition of growth and uptake processes in bacteria by some chemical food preservatives. J. Appl. Bacteriol. 48:423–432. 130. Eklund, T. 1983. The antimicrobial effect of dissociated and undissociated sorbic acid at different pH levels. J. Appl. Bacteriol. 54:383–389. 131. Eklund, T. 1985. Inhibition of microbial growth at different pH levels by benzoic and propionic acids and esters of p-hydroxybenzoic acid. Int. J. Food Microbiol. 2:159–167. 132. Reference deleted. 133. Ekstrand, B. 1994. Lactoperoxidase and lactoferrin, p. 15–63. In V. M. Dillon and R. G. Board (ed.), Natural Antimicrobial Systems and Food Preservation. CAB International, Wallingford, United Kingdom.

SMP_Food Microbiology_CH30.indd

791

791

134. Elci, S., and N. O. Akpolat. 2003. Effect of glycerol monolaurate and sodium benzoate at different concentrations, temperatures and pH on the growth of Listeria monocytogenes. Biotechnol. Biotechnol. Equip. 17:123–127. 135. Elliot, R. M., J. C. McLay, M. J. Kennedy, and R. S. Simmonds. 2004. Inhibition of foodborne bacteria by the lactoperoxidase system in a beef cube system. Int. J. Food Microbiol. 91:73–81. 136. Ellison, R. T., T. G. Giehl, and F. M. LaForce. 1988. Damage of the outer membrane of enteric Gram-negative bacteria by lactoferrin and transferrin. Infect. Immun. 56:2774–2781. 137. Elo, H. A., S. Räisänen, and P. J. Tuohimaa. 1980. Induction of an antimicrobial biotin-binding egg white protein (avidin) in chick tissues in septic Escherichia coli infection. Experientia 36:312–313. 138. Enrique, M., P. Manzanares, M. Yuste, M. Martínez, S. Vallés, and J. F. Marcos. 2009. Selectivity and antimicrobial action of bovine lactoferrin derived peptides against wine lactic acid bacteria. Food Microbiol. 26:340–346. 139. Farnaud, S., and R. W. Evans. 2003. Lactoferrin—a multifunctional protein with antimicrobial properties. Mol. Immunol. 40:395–405. 140. Food and Drug Administration. 1998. Direct food substances affirmed as generally recognized as safe: egg white lysozyme. Fed. Reg. 63:12421–12426. 141. Fernandez, J. L., and W. J. Simpson. 1993. Aspects of the resistance of lactic acid bacteria to hop bitter acids. J. Appl. Bacteriol. 75:315–319. 142. Fernández-Segovia, I., I. Escriche, A. Fuentes, and J. A. Serra. 2007. Microbial and sensory changes during refrigerated storage of desalted cod (Gadus morhua) preserved by combined methods. Int. J. Food Microbiol. 116:64–72. 143. Fisher, K., and C. Phillips. 2009. The mechanism of action of a citrus oil blend against Enterococcus fae­ cium and Enterococcus faecalis. J. Appl. Microbiol. 106:1343–1349. 144. Fisher, T. L., and D. A. Golden. 1998. Survival of Escherichia coli O157:H7 in apple cider as affected by dimethyl dicarbonate, sodium bisulfite, and sodium benzoate. J. Food Sci. 63:904–906. 145. Flores-Villaseñor, H., A. Canizalez-Román, M. ReyesLopez, K. Nazmi, M. de la Garza, J. Zazueta-Beltrán, N. León-Sicairos, and J. G. M. Bolscher. 2010. Bactericidal effect of bovine lactoferrin, LFcin, LFampin and LFchimera on antibiotic-resistant Staphylococcus au­ reus and Escherichia coli. Biometals 23:569–578. 146. Foster, J. W. 1995. Low pH adaptation and the acid tolerance response of Salmonella Typhimurium. Crit. Rev. Microbiol. 21:215–237. 147. Freese, E. 1978. Mechanism of growth inhibition by lipophilic acids, p. 123–131. In J. J. Kabara (ed.), The Pharmacological Effect of Lipids. American Oil Chemists Society, Champaign, IL. 148. Freese, E., C. W. Sheu, and E. Galliers. 1973. Function of lipophilic acids as antimicrobial food additives. Nature 241:321–327.

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

792 149. Friedman, M. 2007. Overview of antibacterial, antitoxin, antiviral, and antifungal activities of tea flavonoids and teas. Mol. Nutr. Food Res. 51:116–134. 150. Friedman, M., R. Buick, and C. T. Elliott. 2004. Antibacterial activities of naturally occurring compounds against antibiotic-resistant Bacillus cereus vegetative cells and spores, Escherichia coli, and Staphylococcus aureus. J. Food Prot. 67:1774–1778. 151. Friedman, M., P. R. Henika, and R. E. Mandrell. 2002. Bactericidal activities of plant essential oils and some of their isolated constituents against Campylobacter jejuni, Escherichia coli, Listeria monocytogenes, and Salmonella enterica. J. Food Prot. 65:1545–1560. 152. Fujita, K.-I., and I. Kubo. 2005. Naturally occurring antifungal agents against Zygosaccharomyces bailii and their synergism. J. Agric. Food Chem. 53: 5187–5191. 153. Garcia, M., M. A. R. Amalaradjou, M. K. M. Nair, T. Annamalai, S. Surendranath, S. Lee, T. Hoagland, D. Dzurec, C. Faustman, and K. Venkitanarayana. 2007. Inactivation of Listeria monocytogenes on frankfurters by monocaprylin alone or in combination with acetic acid. J. Food Prot. 70:1594–1599. 154. García-Ruiz, A., B. Bartolomé, C. Cueva, P. J. MartínÁlvarez, and M. V. Moreno-Arribas. 2009. Inactivation of oenological lactic acid bacteria (Lactobacillus hilgar­ dii and Pediococcus pentosaceus) by wine phenolic compounds. J. Appl. Microbiol. 107:1042–1053. 155. Gaysinsky, S., P. M. Davidson, B. D. Bruce, and J. Weiss. 2005. Growth inhibition of Escherichia coli O157:H7 and Listeria monocytogenes by carvacrol and eugenol encapsulated in surfactant micelles. J. Food Prot. 68:2559–2566. 156. Gaysinsky, S., P. M. Davidson, B. D. Bruce, and J. Weiss. 2005. Stability and antimicrobial efficiency of eugenol encapsulated in surfactant micelles as affected by temperature and pH. J. Food Prot. 68:1359–1366. 157. Gaysinsky, S., T. M. Taylor, P. M. Davidson, B. D. Bruce, and J. Weiss. 2007. Antimicrobial efficacy of eugenol microemulsions in milk against Listeria mono­ cytogenes and Escherichia coli O157:H7. J. Food Prot. 70:2631–2637. 158. Geornaras, I., K. E. Belk, J. A. Scanga, P. A. Kendall, G. C. Smith, and J. N. Sofos. 2005. Postprocessing antimicrobial treatments to control Listeria monocytogenes in commercial vacuum-packaged bologna and ham stored at 10°C. J. Food Prot. 68:991–998. 159. Gerez, C. L., M. I. Torino, M. D. Obregozo, and G. Font de Valdez. 2010. A ready-to-use antifungal starter culture improves the shelf life of packaged bread. J. Food Prot. 73:758–762. 160. Gibson, A. M., and T. A. Roberts. 1986. The effect of pH, water activity, sodium nitrite and storage temperature on the growth of enteropathogenic Escherichia coli and salmonellae in laboratory medium. Int. J. Food Microbiol. 3:183–194. 161. Gibson, A. M., and T. A. Roberts. 1986. The effect of pH, sodium chloride, sodium nitrite and storage temperature on the growth of Clostridium perfringens and

SMP_Food Microbiology_CH30.indd

792

162.

163. 164.

165.

166.

167.

168.

169.

170.

171. 172.

173. 174. 175.

faecal streptococci in laboratory media. Int. J. Food Microbiol. 3:195–210. Gifford, J. L., H. N. Hunter, and H. J. Vogel. 2005. Lactoferricin: a lactoferrin-derived peptide with antimicrobial, antiviral, antitumor and immunological properties. Cell. Mol. Life Sci. 62:2588–2598. Gill, A. O., and R. A. Holley. 2000. Inhibition of bacterial growth on ham and bologna by lysozyme, nisin and EDTA. Food Res. Int. 33:83–90. Gill, A. O., and R. A. Holley. 2003. Interactive inhibition of meat spoilage and pathogenic bacteria by lysozyme, nisin and EDTA in the presence of nitrite and sodium chloride at 24 °C. Int. J. Food Microbiol. 80:251–259. Gill, A. O., and R. A. Holley. 2006. Inhibition of membrane bound ATPases of Escherichia coli and Listeria monocytogenes by plant oil aromatics. Int. J. Food Microbiol. 111:170–174. Glass, K. A., D. A. Granberg, A. L. Smith, A. M. McNamara, M. Hardin, J. Mattias, K. Ladwig, and E. A. Johnson. 2002. Inhibition of Listeria monocyto­ genes by sodium diacetate and sodium lactate on wieners and cooked bratwurst. J. Food Prot. 65:116–123. Golden, D. A., R. W. Worobo, and C. S. Ough. 2005. Dimethyl dicarbonate and diethyl dicarbonate, p. 305– 326. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. Golden, M. H., R. L. Buchanan, and R. C. Whiting. 1995. Effect of sodium acetate or sodium propionate with EDTA and ascorbic acid on the inactivation of Listeria monocytogenes. J. Food Safety 15:53–65. González-Chávez, S. A., S. Arévalo-Gallegos, and Q. Rascón-Cruz. 2009. Lactoferrin: structure, function and applications. Int. J. Antimicrob. Agents 33:301.e301– 301.e308. González-Fandos, E., B. Herrera, and N. Maya. 2009. Efficacy of citric acid against Listeria monocytogenes attached to poultry skin during refrigerated storage. Int. J. Food Sci. Technol. 44:262–268. Gould, G. W. (ed). 1989. Mechanisms of Action of Food Preservation Procedures. Elsevier Applied Science, London, United Kingdom. Gould, G. W. 2000. The use of other chemical preservatives: sulfite and nitrite, p. 200–213. In B. M. Lund, T. C. Baird-Parker, and G. W. Gould (ed.), The Microbiological Safety and Quality of Food. Aspen Publishers, Gaithersburg, MD. Gould, G. W., and N. J. Russell. 1991. Sulphite, p. 72–88. In N. J. Russell and G. W. Gould (ed.), Food Preservatives. Blackie and Son Ltd., Glasgow, United Kingdom. Gram, L. 1991. Inhibition of mesophilic spoilage Aeromonas spp. on fish by salt, potassium sorbate, liquid smoke, and chilling. J. Food Prot. 54:436–442. Guan, W., and X. Fan. 2010. Combination of sodium chlorite and calcium propionate reduces enzymatic browning and microbial population of fresh-cut “Granny Smith” apples. J. Food Sci. 75:M72–M77.

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials 176. Gutierrez, J., C. Barry-Ryan, and B. Bourke. 2009. Antimicrobial activity of plant essential oils using food model media: efficacy, synergistic potential and interactions with food components. Food Microbiol. 26:142–150. 177. Gutierrez, J., G. Rodriguez, C. Barry-Ryan, and P. Bourke. 2008. Efficacy of plant essential oils against foodborne pathogens and spoilage bacteria associated with ready-to-eat vegetables: antimicrobial and sensory screening. J. Food Prot. 71:1846–1854. 178. Hao, Y. Y., R. E. Brackett, and M. P. Doyle. 1998. Efficacy of plant extracts in inhibiting Aeromonas hy­ drophila and Listeria monocytogenes in refrigerated, cooked poultry. Food Microbiol. 15:367–378. 179. Hao, Y. Y., R. E. Brackett, and M. P. Doyle. 1998. Inhibition of Listeria monocytogenes and Aeromonas hydrophila by plant extracts in refrigerated cooked beef. J. Food Prot. 61:307–312. 180. Hara-Kudo, Y., A. Kobayashi, Y. Sugita-Konishi, and K. Kondo. 2004. Antibacterial activity of plants used in cooking for aroma and taste. J. Food Prot. 67:2820–2824. 181. Harris, K., M. F. Miller, G. H. Loneragan, and M. M. Brashears. 2006. Validation of the use of organic acids and acidified sodium chlorite to reduce Escherichia coli O157 and Salmonella Typhimurium in beef trim and ground beef in a simulated processing environment. J. Food Prot. 69:1802–1807. 182. Harris, V., V. Jiranek, C. M. Ford, and P. R. Grbin. 2010. Inhibitory effect of hydroxycinnamic acids on Dekkera spp. Appl. Microbiol. Biotechnol. 86:721–729. 183. Heller, C. E., J. A. Scanga, J. N. Sofos, K. E. Belk, W. Warren-Serna, G. R. Bellinger, R. T. Bacon, M. L. Rossman, and G. C. Smith. 2007. Decontamination of beef subprimal cuts intended for blade tenderization or moisture enhancement. J. Food Prot. 70:1174–1180. 184. Herald, P. J., and P. M. Davidson. 1983. The antibacterial activity of selected hydroxycinnamic acids. J. Food Sci. 48:1378–1379. 185. Hinze, H., and H. Holzer. 1985. Effect of sulfite or nitrite on the ATP content and the carbohydrate metabolism in yeast. Z. Lebensm. Unters. Forsch. 181:87–91. 186. Ho, C. T. 1992. Phenolic compounds in food. An overview, p. 2-7. In C. T. Ho, C. Y. Lee, and M. T. Huang (ed.), Phenolic Compounds in Food and Their Effects on Health, vol. 2. Antioxidants and Cancer Prevention. American Chemical Society, Washington, DC. 187. Hong, M. S., S. J. Lim, Y. K. Oh, and C. K. Kim. 2002. pH-sensitive, serum-stable and long-circulating liposomes as a new drug delivery system. J. Pharm. Pharmacol. 54:51–58. 188. Houtsma, P. C., A. Heuvelink, J. Dufrenne, and S. Notermans. 1994. Effect of sodium lactate on toxin production, spore germination and heat resistance of proteolytic C. botulinum strains. J. Food Prot. 57:327–330. 189. Houtsma, P. C., J. C. Wit, and F. M. Rombouts. 1996. Minimum inhibitory concentration (MIC) of sodium lactate and sodium chloride for spoilage organisms and pathogens at different pH values and temperatures. J. Food Prot. 59:1300–1304.

SMP_Food Microbiology_CH30.indd

793

793

190. Hsieh, Y.-F., T.-L. Chen, Y.-T. Wang, J.-H. Chang, and H.-M. Chang. 2002. Properties of liposomes prepared with various lipids. J. Food Sci. 67:2808–2813. 191. Huang, L., and V. K. Juneja. 2003. Thermal inactivation of Escherichia coli O157:H7 in ground beef supplemented with sodium lactate. J. Food Prot. 66:664–667. 192. Hughey, V. L., R. A. Wilger, and E. A. Johnson. 1989. Antibacterial activity of hen egg white lysozyme against Listeria monocytogenes Scott A in foods. Appl. Environ. Microbiol. 55:631–638. 193. Ibrahim, H. R., S. Higashiguchi, M. Koketsu, L. R. Juneja, M. Kim, T. Yamamoto, Y. Sugimoto, and T. Aoki. 1996. Partially unfolded lysozyme at neutral pH agglutinates and kills gram-negative and gram-positive bacteria through membrane damage mechanism. J. Agric. Food Chem. 44:3799–3806. 194. Ibrahim, H. R., T. Matsuzaki, and T. Aoki. 2001. Genetic evidence that antibacterial activity of lysozyme is independent of its catalytic function. FEBS Lett. 506:27–32. 195. Infante, R., J. G. Dominguez, P. Erra, R. Julia, and M. Prats. 1984. Surface active molecules: preparation and properties of long chain Na-acyl-L-a-amino-w-guanidine alkyl acid derivatives. Int. J. Cosmet. Sci. 6:275–282. 196. Isshiki, K., K. Tokuora, R. Mori, and S. Chiba. 1992. Preliminary examination of allyl isothiocyanate vapor for food preservation. Biosci. Biotechnol. Biochem. 56:1476–1477. 197. Johansen, C., L. Gram, and A. S. Meyer. 1994. The combined inhibitory effect of lysozyme and low pH on growth of Listeria monocytogenes. J. Food Prot. 57:561–566. 198. Johnson, E. A., and A. E. Larson. 2005. Lysozyme, p. 361–388. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 199. Johny, A. K., M. J. Darre, A. M. Donoghue, D. J. Donoghue, and K. Venkitanarayana. 2010. Antibacterial effect of trans-cinnamaldehyde, eugenol, carvacrol, and thymol on Salmonella Enteriditis and Campylobacter jejuni in chicken cecal contents in vitro. J. Appl. Poult. Res. 19:237–244. 200. Jones, E. M., A. Smart, G. Bloomberg, L. Burgess, and M. R. Millar. 1994. Lactoferricin, a new antimicrobial peptide. J. Appl. Bacteriol. 77:208–214. 201. Juliano, R. L. 1981. Liposomes as a drug delivery system. Trends Pharmacol. Sci. 2:39–42. 202. Juneja, V. K. 2006. Delayed Clostridium perfringens growth from a spore inocula by sodium lactate in sousvide chicken products. Food Microbiol. 23:105–111. 203. Juneja, V. K., and P. M. Davidson. 1992. Influence of altered fatty acid composition on resistance of Listeria mono­ cytogenes to antimicrobials. J. Food Prot. 56:302–305. 204. Jung, Y. J., K. J. Min, and K. S. Yoon. 2009. Responses of acid-stressed Salmonella Typhimurium in broth and chicken patties to subsequent antimicrobial stress with e-polylysine and combined potassium lactate and sodium diacetate. Food Microbiol. 26:467–474. 205. Kabara, J. J., and D. L. Marshall. 2005. Medium-chain fatty acids and esters., p. 327–360. In P. M. Davidson,

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

794 J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 206. Kalemba, D., and A. Kunicka. 2003. Antibacterial and antifungal properties of essential oils. Curr. Med. Chem. 10:813–829. 207. Karapinar, M., and S. A. Gonul. 1992. Removal of Yersinia enterocolitica from fresh parsley by washing with acetic acid or vinegar. Int. J. Food Microbiol. 16:261–264. 208. Kasrazadeh, M., and C. Genigeorgis. 1995. Potential growth and control of Escherichia coli O157:H7 in soft Hispanic type cheese. Int. J. Food Microbiol. 25:289–300. 209. Kheadr, E. E., J.-C. Vuillemard, and S. A. El-Deeb. 2002. Acceleration of Cheddar cheese lipolysis by using liposome-entrapped lipases. J. Food Sci. 67:485–492. 210. Kihm, D. J., G. J. Leyer, G.-H. An, and E. A. Johnson. 1994. Sensitization of heat-treated Listeria monocy­ togenes to added lysozyme in milk. Appl. Environ. Microbiol. 60:3854–3861. 211. Kilinc, B., S. Cakli, T. Dincer, and S. Tolasa. 2009. Microbiological, chemical, sensory, color, and textural changes of rainbow trout fillets treated with sodium acetate, sodium lactate, sodium citrate, and stored at 4°C. J. Aquat. Food Prod. Technol. 18:3–17. 212. Kim, C. R., J. O. Hearnsberger, A. P. Vickery, C. H. White, and D. L. Marshall. 1995. Extending shelf life of refrigerated catfish fillets using sodium acetate and monopotassium phosphate. J. Food Prot. 58:644–647. 213. Kim, T. J., W. L. Weng, J. L. Silva, Y. S. Jung, and D. Marshall. 2010. Identification of natural antimicrobial substances in red muscadine juice against Cronobacter sakazakii. J. Food Sci. 75:M150–M154. 214. Kim, Y., M. Kim, and K. B. Song. 2009. Combined treatment of fumaric acid with aqueous chlorine dioxide or UV-C irradiation to inactivate Escherichia coli O157: H7, Salmonella enterica, serovar Typhimurium, and Listeria monocytogenes inoculated on alfalfa and clover sprouts. LWT Food Sci. Technol. 42:1654–1658. 215. Kim, Y. J., M. H. Kim, and K. B. Song. 2009. Efficacy of aqueous chlorine dioxide and fumaric acid for inactivating pre-existing microorganisms and Escherichia coli O157:H7, Salmonella typhimurium, and Listeria monocytogenes on broccoli sprouts. Food Control 20:1002–1005. 216. Kim, Y. J., B. A. Nahm, and I. H. Choi. 2010. An evaluation of the antioxidant and antimicrobial effectiveness of different forms of garlic and BHA in emulsion-type sausages during refrigerated storage. J. Muscle Foods 21:813–825. 217. Kirby, G. W., L. Atkin, and C. N. Frey. 1937. Further studies on the growth of bread molds as influenced by acidity. Cereal Chem. 14:865–878. 218. Ko, K. Y., A. F. Mendonca, and D. U. Ahn. 2008. Effect of ethylenediaminetetraacetate and lysozyme on the antimicrobial activity of ovotransferrin against Listeria monocytogenes. Poult. Sci. 87:1649–1658. 219. Ko, K. Y., A. F. Mendonca, H. Ismail, and D. U. Ahn. 2009. Ethylenediaminetetraacetate and lysozyme im-

SMP_Food Microbiology_CH30.indd

794

proves antimicrobial activities against ovotransferrin against Escherichia coli O157:H7. Poult. Sci. 88:406–414. 220. Koczon´, P. 2009. Growth inhibition mode of action of selected benzoic acid derivatives against the yeast Pichia anomala. J. Food Prot. 72:791–800. 221. Kondo, N., M. Murata, and K. Isshiki. 2006. Efficiency of sodium hypochlorite, fumaric acid, and mild heat in killing native microflora and Escherichia coli O157:H7, Salmonella Typhimurium DT104, and Staphylococcus aureus attached to fresh-cut lettuce. J. Food Prot. 69:323–329. 222. Kong, M., X. G. Chen, K. Xing, and H. J. Park. 2010. Antimicrobial properties of chitosan and mode of action: a state of the art review. Int. J. Food Microbiol. 144:51–63. 223. Kussendrager, K. D., and A. C. M. van Hooijdonk. 2000. Lactoperoxidase: physico-chemical properties, occurrence, mechanism of action and applications. Br. J. Nutr. 84:S19–S25. 224. Lachowicz, K. J., G. P. Jones, D. R. Briggs, F. E. Bienvenu, J. Wan, A. Wilcock, and M. J. Coventry. 1998. The synergistic preservative effects of the essential oils of sweet basil (Ocimum basilicum L.) against acid-tolerant food microflora. Lett. Appl. Microbiol. 26:209–214. 225. Lambert, R. J., and M. Stratford. 1999. Weak-acid preservatives: modeling microbial inhibition and response. J. Appl. Microbiol. 86:157–164. 226. Larson, A. E., R. R. Y. Yu, O. A. Lee, S. Price, G. J. Haas, and E. A. Johnson. 1996. Antimicrobial activity of hop extracts against Listeria monocytogenes in media and in food. Int. J. Food Microbiol. 33:195–207. 227. Lasch, J., V. Weissig, and M. Brandl. 2003. Preparation of liposomes, p. 3­­–30. In V. P. Torchilin and V. Weissig (ed.), Liposomes: a Practical Approach, 2nd ed. Oxford University Press, New York, NY. 228. Laury, A. M., M. V. Alvarado, G. Nace, C. Z. Alvarado, J. C. Brooks, A. Echeverry, and M. M. Brashears. 2009. Validation of a lactic acid- and citric acid-based antimicrobial product for the reduction of Escherichia coli O157:H7 and Salmonella on beef tips and whole chicken carcasses. J. Food Prot. 72:2208–2211. 229. Lawrence, R., P. Tripathi, and E. Jeyakumar. 2009. Isolation, purification and evaluation of antibacterial agents from Aloe vera. Braz. J. Microbiol. 40: 906–915. 230. Lee, S.-Y., and D.-H. Kang. 2009. Combined effects of heat, acetic acid, and salt for inactivating Escherichia coli O157:H7 in laboratory media. Food Control 20:1006–1012. 231. Lee, Y.-L., T. Cesario, J. Owens, E. Shanbrom, and L. D. Thrupp. 2002. Antibacterial activity of citrate and acetate. Nutrition 18:665–666. 232. Lemar, K. M., O. Passa, M. A. Aon, S. Cortassa, C. T. Müller, S. Plummer, B. O’Rourke, and D. Lloyd. 2005. Allyl alcohol and garlic (Allium sativum) extract produce oxidative stress in Candida albicans. Microbiology 151:3257–3265. 233. Lennox, J. E., and L. J. McElroy. 1984. Inhibition of growth and patulin synthesis in Penicillium expansum

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials

234.

235.

236.

237.

238. 239.

240.

241.

242. 243.

244.

245.

246.

by potassium sorbate and sodium propionate in culture. Appl. Environ. Microbiol. 48:1031–1033. Li, J., and B. A. McClane. 2006. Comparative effects of osmotic, sodium nitrite-induced, and pH-induced stress on growth and survival of Clostridium perfringens type A isolates carrying chromosomal or plasmid-borne enterotoxin genes. Appl. Environ. Microbiol. 72:7620–7625. Liewen, M. B., and E. H. Marth. 1985. Growth of sorbate-resistant and -sensitive strains of Penicillium roqueforti in the presence of sorbate. J. Food Prot. 48:525–529. Lillard, H. S., L. C. Blankenship, J. A. Dickens, S. E. Craven, and A. D. Shackelford. 1987. Effect of acetic acid on the microbiological quality of scalded picked and unpicked broiler carcasses. J. Food Prot. 50:112–114. Lim, K., and A. Mustapha. 2004. Effects of cetylpyridinium chloride, acidified sodium chlorite, and potassium sorbate on populations of Escherichia coli O157: H7, Listeria monocytogenes, and Staphylococcus aureus on fresh beef. J. Food Prot. 67:310–315. Lin, C.-A., J. F. Preston, and C.-I. Wei. 2000. Antibacterial mechanism of allyl isothiocyanate. J. Food Prot. 63:727–734. Liolios, C. C., O. Gortzi, S. Lalas, J. Tsaknis, and I. Chinou. 2009. Liposomal incorporation of carvacrol and thymol isolated from the essential oil of Origanum dictamnus L. and in vitro antimicrobial activity. Food Chem. 112:77–83. Lloyd, T., C. Z. Alvarado, M. M. Brashears, L. D. Thompson, S. R. McKee, and M. Berrang. 2009. Control of Listeria monocytogenes in turkey deli loaves using organic acids as formulation ingredients. Poult. Sci. 88:2235–2239. Lopez, M., S. Martinez, J. Gonzalez, R. Martin, and A. Bernardo. 1998. Sensitization of thermally injured spores of Bacillus stearothermophilus to sodium benzoate and potassium sorbate. Lett. Appl. Microbiol. 27:331–335. López-Expósito, I., A. Pellegrini, L. Amigo, and I. Recio. 2008. Synergistic effect between different milk-derived peptides and proteins. J. Dairy Sci. 91:2184–2189. López-Malo, A., S. M. Alzamora, and A. Argaiz. 1995. Effect of natural vanillin on germination time and radial growth of moulds in fruit-based agar systems. Food Microbiol. 12:213–219. López-Malo, A., S. M. Alzamora, and S. Guerrero. 2000. Natural antimicrobials from plants, p. 237–264. In S. M. Alzamora, M. S. Tapia, and A. López-Malo (ed.), Minimally Processed Fruits and Vegetables: Funda­ mental Aspects and Applications. Aspen Publishers, Gaithersburg, MD. López-Malo, A., S. M. Alzamora, and E. Palou. 2002. Aspergillus flavus dose-response curves to selected natural and synthetic antimcrobials. Int. J. Food Microbiol. 73:213–218. López-Malo, A., E. Palou, and S. M. Alzamora. 2005. Naturally occurring compounds—plant sources, p. 429– 452. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL.

SMP_Food Microbiology_CH30.indd

795

795

247. Luciano, F. B., and R. A. Holley. 2009. Enzymatic inhibition by allyl isothiocyanate and factors affecting its antimicrobial action against Escherichia coli O157:H7. Int. J. Food Microbiol. 131:240–245. 248. Lund, B. M., S. M. George, and J. G. Franklin. 1987. Inhibition of type A and type B (proteolytic) Clostridium botulinum by sorbic acid. Appl. Environ. Microbiol. 53:935–941. 249. Maks, N., L. Zhu, V. K. Juneja, and S. Ravishankar. 2010. Sodium lactate, sodium diacetate and pediocin: effects and interactions on the thermal inactivation of Listeria mono­ cytogenes on bologna. Food Microbiol. 27:64–69. 250. Manderfield, M. M., H. W. Schafer, P. M. Davidson, and E. A. Zottola. 1997. Isolation and identification of antimicrobial furocoumarins from parsley. J. Food Prot. 60:72–77. 251. Mangalassary, S., I. Han, J. Rieck, J. Acton, and P. Dawson. 2008. Effect of combining nisin and/or lysozyme with in-package pasteurization for control of Listeria monocytogenes in ready-to-eat turkey bologna during refrigerated storage. Food Microbiol. 25:866–870. 252. Mann, C. M., S. D. Cox, and J. L. Markham. 2000. The outer membrane of Pseudomonas aeruginosa NCTC 6749 contributes to its tolerance to the essential oil of Melaleuca alternifolia (tea tree oil). Lett. Appl. Microbiol. 30:294–297. 253. Mansour, M., and J.-B. Milliere. 2001. An inhibitory synergistic effect of a nisin-monolaurin combination on Bacillus sp. vegetative cells in milk. Food Microbiol. 18:87–94. 254. Mari, M., R. Iori, O. Leoni, and A. Marchi. 1993. In vitro activity of glucosinolate derived isothiocyanates against postharvest fruit pathogens. Ann. Appl. Biol. 123:155–164. 255. Martin, E. M., C. L. Griffis, K. L. S. Vaughn, C. A. O’Bryan, E. C. Friedly, J. A. Marcy, S. C. Ricke, P. G. Crandall, and R. Y. Lary, Jr. 2009. Control of Listeria monocytogenes by lauric arginate on frankfurters formulated with or without lactate/diacetate. J. Food Sci. 74:M237–M241. 256. Mastromatteo, M., A. Lucera, M. Sinigaglia, and M. R. Corbo. 2010. Use of lysozyme, nisin, and EDTA combined treatments for maintaining quality of packed ostrich patties. J. Food Sci. 75:M178–M186. 257. Mazzotta, A. S., K. D. Modi, and T. J. Montville. 2000. Nisin-resistant (Nisr) Listeria monocytogenes and Nisr Clostridium botulinum are not resistant to common food preservatives. J. Food Sci. 65:888–890. 258. McClements, D. J., E. A. Decker, Y. Park, and J. Weiss. 2009. Structural design principles for delivery of bioactive components in nutraceuticals and functional foods. Crit. Rev. Food Sci. Nutr. 49:577–606. 259. McMindes, M. K., and A. J. Siedler. 1988. Nitrite mode of action: inhibition of yeast pyruvate decarboxylase (E.C. 4.1.1.1) and clostridial pyruvate:oxidoreductase (E.C. 1.2.7.1) by nitric oxide. J. Food Sci. 53:917–919, 931. 260. McWilliam Leitch, E. C., and C. S. Stewart. 2002. Escherichia coli O157 and non-O157 isolates are more

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

796

261. 262. 263.

264.

265.

266. 267.

268.

269. 270.

271.

272.

273.

274.

275.

susceptible to l-lactate then to d-lactate. Appl. Environ. Microbiol. 68:4676–4678. McWilliam Leitch, E. C., and C. S. Stewart. 2002. Susceptibility of Escherichia coli O157 and non-O157 isolates to lactate. Lett. Appl. Microbiol. 35:176–180. Merkl, R., I. Hrádková, V. Filip, and J. Šmidrkal. 2010. Antimicrobial and antioxidant properties of phenolic acids alkyl esters. Czech J. Food Sci. 28:275–279. Miller, A. J., and J. E. Call. 1994. Inhibitory potential of four-carbon dicarboxylic acids on Clostridium botuli­ num spores in an uncured turkey product. J. Food Prot. 57:679–683. Miller, A. J., J. E. Call, and R. C. Whiting. 1993. Comparison of organic acid salts for Clostridium botuli­ num control in an uncured turkey product. J. Food Prot. 56:958–962. Min, S., and J. M. Krochta. 2005. Inhibition of Penicillium commune by edible whey protein films incorporating lactoferrin, lactoferrin hydrolysate, and the lactoperoxidase systems. J. Food Sci. 70:M87–M94. Mitchell, P. 1961. Coupling of phosphorylation to electron and hydrogen transfer by a chemiosmotic type of mechanism. Nature 191:144–148. Mitchell, P., and J. Moyle. 1969. Estimation of membrane potential and pH difference across the cristae membrane of rat liver mitochondria. Eur. J. Biochem. 7:471–484. Morita, H., H. Yoshikawa, T. Suzuki, S. Hisamatsu, Y. Kato, R. Sakata, Y. Nagata, and T. Yoshimura. 2004. Anti-microbial action against verotoxigenic Escherichia coli O157:H7 of nitric oxide derived from sodium nitrite. Biosci. Biotechnol. Biochem. 68:1027–1034. Mountney, G. J., and J. O’Malley. 1965. Acids as poultry meat preservatives. Poult. Sci. 44:582–586. Murdock, C. A., J. Cleveland, K. R. Matthews, and M. L. Chikindas. 2007. The synergistic effect of nisin and lactoferrin on the inhibition of Listeria monocytogenes and Escherichia coli O157:H7. Lett. Appl. Microbiol. 44:255–261. Murdock, C. A., and K. R. Matthews. 2002. Antibacterial activity of pepsin-digested lactoferrin on foodborne pathogens in buffered broth systems and ultra-high temperature milk with EDTA. J. Appl. Microbiol. 93:850–856. Murphy, R. Y., T. Osaili, L. K. Duncan, and J. A. Marcy. 2004. Effect of sodium lactate on thermal inactivation of Listeria monocytogenes and Salmonella in ground chicken thigh and leg meat. J. Food Prot. 67:1403–1407. Muthukumarasamy, P., J. H. Han, and R. A. Holley. 2003. Bactericidal effects of Lactobacillus reuteri and allyl isothiocyanate on Escherichia coli O157:H7 in refrigerated ground beef. J. Food Prot. 66:2038–2044. Muthuswamy, S., H. P. V. Rupasinghe, and G. W. Stratton. 2008. Antimicrobial effect of cinnamon bark extract on Escherichia coli O157:H7, Listeria innocua and fresh-cut apple slices. J. Food Safety 28:534–549. Mytle, N., G. L. Anderson, M. P. Doyle, and M. A. Smith. 2006. Antimicrobial activity of clove (Syzygium

SMP_Food Microbiology_CH30.indd

796

276. 277. 278.

279.

280.

281.

282.

283.

284.

285.

286.

287. 288. 289. 290.

aromaticum) oil in inhibiting Listeria monocytogenes on chicken frankfurters. Food Control 17:102–107. Naidu, A. S. 2001. Immobilized lactoferrin antimicrobial agents and the use thereof. U.S. patent 6,172,040. Naidu, A. S. 2002. Activated lactoferrin—a new approach to meat safety. Food Technol. 56:40–45. Nair, M. K. M., J. Joy, and K. S. Venkitanarayana. 2004. Inactivation of Enterobacter sakazakii in reconstituted infant formula by monocaprylin. J. Food Prot. 67:2815–2819. Nakai, S. A., and K. J. Siebert. 2004. Organic acid inhibition models for Listeria innocua, Listeria ivanovii, Pseudomonas aeruginosa, and Oenococcus oeni. Food Microbiol. 21:67–72. Nakanishi, N., F. Takeuchi, S. Y. Park, H. Hori, K. Kiyota, T. Uno, and M. Tsubaki. 2008. Characterization of heme-coordinating histidyl residues of an engineered six-coordinated myoglobin mutant based on the reactivity with diethylpyrocarbonate, mass spectrometry, and electron paramagnetic resonance spectroscopy. J. Biosci. Bioeng. 105:604–613. Nakimbugwe, D., B. Masschalck, G. Anim, and C. W. Michiels. 2006. Inactivation of gram-negative bacteria in milk and banana juice by hen egg white and lambda lysozyme under high hydrostatic pressure. Int. J. Food Microbiol. 112:19–25. Nakimbugwe, D., B. Masschalck, M. Atanassova, A. Zewdie-Bosüner, and C. W. Michiels. 2006. Comparison of bactericidal activity of six lysozymes at atmospheric pressure and under high hydrostatic pressure. Int. J. Food Microbiol. 108:355–363. Natress, F. M., and L. P. Baker. 2003. Effects of treatment with lysozyme and nisin on the microflora and sensory properties of commercial pork. Int. J. Food Microbiol. 85:259–267. Natress, F. M., C. K. Yost, and L. P. Baker. 2001. Evaluation of the ability of lysozyme and nisin to control meat spoilage bacteria. Int. J. Food Microbiol. 70:111–119. Nau, F., C. Guérin-Dubiard, and T. Croguennec. 2007. Avidin, p. 75–80. In R. Huopalahti, R. LópezFandiño, M. Anton, and R. Schade (ed.), Bioactive Egg Compounds. Springer, New York, NY. Neetoo, H., M. Ye, and H. Chen. 2010. Bioactive alginate coatings to control Listeria monocytogenes on coldsmoked salmon slices and fillets. Int. J. Food Microbiol. 136:326–331. New, R. R. C. 1990. Introduction, p. 1–32. In R. R. C. New (ed.), Liposomes: a Practical Approach. Oxford University Press, New York, NY. New, R. R. C. 1990. Preparation of liposomes, p. 33– 104. In R. R. C. New (ed.), Liposomes: a Practical Approach. Oxford University Press, New York, NY. Nguyen, T., B. Clare, W. Guo, and B. Martinac. 2005. The effects of parabens on the mechanosensitive channels of E. coli. Eur. Biophys. J. 34:389–395. Nobmann, P., A. Smith, J. Dunne, G. Henehan, and P. Bourke. 2009. The antimicrobial efficacy and structure activity relationship of novel carbohydrate fatty acid de-

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials rivatives against Listeria spp. and food spoilage microorganisms. Int. J. Food Microbiol. 128:440–445. 291. Nuñez de Gonzalez, M. T., J. T. Keeton, G. R. Acuff, L. J. Ringer, and L. M. Lucia. 2004. Effectiveness of acidic calcium sulfate with propionic and lactic acid and lactates as postprocessing dipping solutions to control Listeria monocytogenes on frankfurters with or without potassium lactate and stored vacuum packaged at 4.5°C. J. Food Prot. 67:915–921. 292. Obaidat, M. M., and J. F. Frank. 2009. Inactivation of Escherichia coli O157:H7 on the intact and damaged portions of lettuce and spinach leaves by using allyl isothiocyanate, carvacrol, and cinnamaldehyde in vapor phase. J. Food Prot. 72:2046–2055. 293. Obaidat, M. M., and J. F. Frank. 2009. Inactivation of Salmonella and Escherichia coli O157:H7 on sliced and whole tomatoes by allyl isothiocyanate, carvacrol, and cinnamaldehyde in vapor phase. J. Food Prot. 72:315–324. 294. O’Bryan, C. A., P. G. Crandall, and S. C. Ricke. 2008. Organic poultry pathogen control from farm to fork. Foodborne Pathog. Dis. 5:709–720. 295. Oh, D. H., and D. L. Marshall. 1994. Enhanced inhibition of Listeria monocytogenes by glycerol monolaurate with organic acids. J. Food Sci. 59:1258–1261. 296. Okrend, A. J., R. W. Johnston, and A. B. Moran. 1986. Effect of acetic acid on the death rates at 52°C of Salmonella newport, Salmonella typhimurium, and Campylobacter jejuni in poultry scald water. J. Food Prot. 49:500–503. 297. Omar, M. M. 1992. Phenolic compounds in botanical extracts used in foods, flavors, cosmetics, and pharmaceuticals, p. 154–168. In C. T. Ho, C. Y. Lee, and M. T. Huang (ed.), Phenolic Compounds in Food and Their Effects on Health, vol 1. Analysis, Occurrence, and Chemistry, American Chemical Society, Washington, DC. 298. Oram, J. D., and B. Reiter. 1968. Inhibition of bacteria by lactoferrin and other iron-chelating agents. Biochim. Biophys. Acta 170:351–365. 299. Orsi, N. 2004. The antimicrobial activity of lactoferrin: current status and perspectives. BioMetals 17: 189–196. 300. Ough, C. S., and R. E. Kunkee. 1974. The effect of fumaric acid on malolactic fermentation in wines from warm areas. Am. J. Enol. Viticult. 25:188–190. 301. Ough, C. S., and L. Were. 2005. Sulfur dioxide and sulfites, p. 143–168. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 302. Over, K. F., N. Hettiarachchy, M. G. Johnson, and B. Davis. 2009. Effect of organic acids and plant extracts on Escherichia coli O157:H7, Listeria monocy­ togenes, and Salmonella Typhimurium in broth culture model and chicken meat systems. J. Food Sci. 74: M515–M521. 303. Palumbo, S. A., and A. C. Williams. 1994. Control of Listeria monocytogenes on the surface of frankfurters by acid treatment. Food Microbiol. 11:293–300.

SMP_Food Microbiology_CH30.indd

797

797

304. Pandey, A., F. Bringel, and J.-M. Meyer. 1994. Iron requirement and search for siderophores in lactic acid bacteria. Appl. Microbiol. Biotechnol. 40:735–739. 305. Pandit, V. A., and L. A. Shelef. 1994. Sensitivity of Listeria monocytogenes to rosemary (Rosmarinus offi­ cianalis L.). Food Microbiol. 11:57–63. 306. Papadopoulos, L. S., R. K. Miller, G. R. Acuff, C. Vanderzant, and H. R. Cross. 1991. Effect of sodium lactate on microbial and chemical composition of cooked beef during storage. J. Food Sci. 56:341–347. 307. Paredes-Sabja, D., D. Raju, J. A. Torres, and M. R. Sarker. 2008. Role of small, acid-soluble spore proteins in the resistance of Clostridium perfringens spores to chemicals. Int. J. Food Microbiol. 122:333–335. 308. Park, S.-I., S. D. Stan, M. A. Daeschel, and Y. Zhao. 2005. Antifungal coatings on fresh strawberries (Fragaria x ananassa) to control mold growth during cold storage. J. Food Sci. 70:M202–M207. 309. Parry-Hanson, A., P. J. Jooste, and E. M. Buys. 2009. The influence of lactoperoxidase, heat and low pH on survival of acid-adapted and non-adapted Escherichia coli O157:H7 in goat milk. Int. Dairy J. 19:417–421. 310. Pattison, T.-L., D. Lindsay, and A. Von Holy. 2003. In vitro growth response of bread-spoilage Bacillus strains to selected natural antimicrobials. J. Basic Microbiol. 4:341–347. 311. Paul, B. and I. Hirshfield. 2003. The effect of acid treatment on survival and protein expression of a laboratory K12 strain Escherichia coli. Res. Microbiol. 154:115–121. 312. Payne, K. D., P. M. Davidson, S. P. Oliver, and G. L. Christen. 1990. Influence of bovine lactoferrin on the growth of Listeria monocytogenes. J. Food Prot. 53:468–472. 313. Payne, K. D., S. P. Oliver, and P. M. Davidson. 1994. Comparison of EDTA and apo-lactoferrin with lysozyme on the growth of foodborne pathogenic and spoilage bacteria. J. Food Prot. 57:62–65. 314. Payne, K. D., E. Rico-Muñoz, and P. M. Davidson. 1989. The antimicrobial activity of phenolic compounds against Listeria monocytogenes and their effectiveness in a model milk system. J. Food Prot. 52:151–153. 315. Pérez-Conesa, D., L. McLandsborough, and J. Weiss. 2006. Inhibition and inactivation of Listeria monocy­ togenes and Escherichia coli O157:H7 colony biofilms by micellar-encapsulated eugenol and carvacrol. J. Food Prot. 69:2947–2954. 316. Pina-Pérez, M. C., A. B. Silva-Angulo, B. MuguerzaMarquínez, D. R. Aliaga, and A. M. López. 2009. Synergistic effect of high hydrostatic pressure and natural antimicrobials on inactivation kinetics of Bacillus cereus in a liquid whole egg and skim milk mixed beverage. Foodborne Pathog. Dis. 6:649–656. 317. Porto-Fett, A. C. S., S. G. Campano, J. L. Smith, A. Oser, B. Shoyer, J. E. Call, and J. B. Luchansky. 2010. Control of Listeria monocytogenes on commercially-produced frankfurters prepared with and without potassium lactate and sodium diacetate and surface treated with lauric arginate using the Sprayed Lethality in Container (SLIC®) delivery method. Meat Sci. 85:312–318.

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

798 318. Pranato, Y., S. K. Rakshit, and V. M. Salokhe. 2005. Enhancing antimicrobial activity of chitosan films by incorporating garlic oil, potassium sorbate and nisin. Food Sci. Technol. 38:859–865. 319. Rahman, S. M. E., Y.-G. Jin, and D.-H. Oh. 2010. Combined effects of alkaline electrolyzed water and citric acid with mild heat to control microorganisms on cabbage. J. Food Sci. 75:M111–M115. 320. Rajashekhara, E., E. R. Suresh, and S. Ethiraj. 1998. Thermal death of ascospores of Neosartorya fischeri ATCC 200957 in the presence of organic acids and preservatives in fruit juices. J. Food Prot. 61:1358–1362. 321. Rattanachaikunsopon, P., and P. Phumkhachorn. 2010. Antimicrobial activity of basil (Ocimum basilicum) oil against Salmonella Enteriditis in vitro and in food. Biosci. Biotechnol. Biochem. 74:1200–1204. 322. Raybaudi-Massilia, R. M., J. Mosqueda-Megar, and O. Martín-Belloso. 2008. Edible alginate-based coating as carrier of antimicrobials to improve shelf-life and safety of fresh-cut melon. Int. J. Food Microbiol. 121:313–327. 323. Razavi-Rohani, S. M., and M. W. Griffiths. 1999. The antifungal activity of butylated hydroxyanisole and lysozyme. J. Food Safety 19:97–108. 324. Reiss, J. 1976. Prevention of the formation of mycotoxins in whole wheat bread by citric acid and lactic acid. Experientia 32:168–169. 325. Renouf, V., P. Strehaiano, and A. Lonvaud-Funel. 2008. Effectiveness of dimethyldicarbonate to prevent Bret­ tanomyces bruxellensis growth in wine. Food Control 19:208–216. 326. Rhoades, J., and S. Roller. 2000. Antimicrobial actions of degradated and native chitosan against spoilage organisms in laboratory media and foods. Appl. Environ. Microbiol. 66:80–86. 327. Ricke, S. C. 2003. Perspectives on the use of organic acids and short chain fatty acids as antimicrobials. Poult. Sci. 82:632–639. 328. Rico-Muñoz, E., E. E. Bargiota, and P. M. Davidson. 1987. Effect of selected phenolic compounds on the membrane-bound adenosine triphosphatase of Staphylococcus aureus. Food Microbiol. 4:239–249. 329. Robach, M. C. 1980. Use of preservatives to control microorganisms in food. Food Technol. 34:81–84. 330. Roberts, T. A., and M. Ingram. 1966. The effect of sodium chloride, potassium nitrate and sodium nitrite on the recovery of heated bacterial spores. J. Food Technol. 1:147–163. 331. Roberts, T. A., L. F. J. Woods, M. J. Payne, and R. Cammack. 1991. Nitrite, p. 89–110. In N. J. Russell and G. W. Gould (ed.), Food Preservatives. Blackie and Son Ltd., Glasgow, United Kingdom. 332. Rodríguez, E., J. Seguer, X. Rocabayera, and A. Manresa. 2004. Cellular effects of monohydrochloride of l-arginine, Na-lauroyl ethylester (LAE) on exposure to Salmonella Typhimurium and Staphylococcus aureus. J. Appl. Microbiol. 96:903–912. 333. Rojo-Bezares, B., Y. Sáenz, M. Zarazaga, C. Torres, and F. Ruiz-Larrea. 2007. Antimicrobial activity of nisin

SMP_Food Microbiology_CH30.indd

798

against Oenococcus oeni and other wine bacteria. Int. J. Food Microbiol. 116:32–36. 334. Roland, J. O., and L. R. Beuchat. 1984. Biomass and patulin production by Byssochlamys nivea in apple juice as affected by sorbate, benzoate, SO2 and temperature. J. Food Sci. 49:402–406. 335. Ronning, I. E., and H. A. Frank. 1987. Growth inhibition of putrefactive anaerobe 3679 caused by stringent-type response induced by protonophoric activity of sorbic acid. Appl. Environ. Microbiol. 53:1020–1027. 336. Ronning, I. E., and H. A. Frank. 1989. Morphological changes in putrefactive anaerobe 3679 (Clostridium sporogenes) induced by sorbate, hydrochloric acid, and nitrate. Can. J. Microbiol. 35:388–398. 337. Rose, A. H., and B. J. Pilkington. 1989. Sulphite, p. 201– 224. In G. W. Gould (ed.), Mechanisms of Action of Food Preservation Procedures. Elsevier Applied Science, London, United Kingdom. 338. Rosenquist, H., and Å. Hansen. 1998. The antimicrobial effect of organic acids, sour dough and nisin against Bacillus subtilis and B. licheniformis isolated from wheat bread. J. Appl. Microbiol. 85:621–631. 339. Rossi, P., F. Giansanti, A. Boffi, M. Ajello, P. Valenti, E. Chiancone, and G. Antonini. 2002. Ca2+ binding to bovine lactoferrin enhances protein stability and influences the release of bacterial lipopolysaccharide. Biochem. Cell Biol. 80:41–48. 340. Rowe, J. J., J. M. Yabrough, J. B. Rake, and R. G. Eagon. 1979. Nitrite inhibition of aerobic bacteria. Curr. Microbiol. 2:51–54. 341. Rúa, J., L. Fernández-Álvarez, M. Gutiérrez-Larraínzar, P. del Valle, D. de Arriaga, and M. R. García-Armesto. 2010. Screening of phenolic antioxidants for their inhibitory activity against foodborne Staphylococcus aureus strains. Foodborne Pathog. Dis. 7:695–705. 342. Russell, A. D. 2005. Mechanisms of action, resistance, and stress adaptation, p. 633–657. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 343. Russell, J. B. 1992. Another explanation for the toxicity of fermentation acids at low pH: anion accumulation versus uncoupling. J. Appl. Bacteriol. 73:363–370. 344. Rusul, G., and E. H. Marth. 1987. Growth and aflatoxin production by Aspergillus parasiticus NRRL 2999 in the presence of potassium benzoate or potassium sorbate and at different initial pH values. J. Food Prot. 50:820–825. 345. Sagoo, S. K., R. G. Board, and S. Roller. 2002. Chitosan potentiates the antimicrobial action of sodium benzoate on spoilage yeasts. Lett. Appl. Microbiol. 34:168–172. 346. Saha, S. C., and B. A. Chopade. 2002. Effect of food preservatives on Acinetobacter genospecies isolated from meat. J. Food Sci. Technol. 39:26–32. 347. Sallam, K. I., M. Ishioroshi, and K. Samejima. 2004. Antioxidant and antimicrobial effects of garlic in chicken sausage. Lebenson. Wiss. Technol. 37:849–855. 348. Samelis, J., J. N. Sofos, M. L. Kain, J. A. Scanga, K. E. Belk, and G. C. Smith. 2001. Organic acids and their salts as dipping solutions to control Listeria monocy­ togenes inoculated following processing of sliced pork

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials

349.

350.

351.

352.

353.

354.

355. 356. 357. 358. 359.

360. 361. 362.

363.

364.

bologna stored at 4°C in vacuum packages. J. Food Prot. 64:1722–1729. Santiesteban-López, A., E. Palou, and A. López-Malo. 2007. Susceptibility of food-borne bacteria to binary combinations of antimicrobials at selected aw and pH. J. Appl. Microbiol. 102:486–497. Schirmer, B. C., and S. Langsrud. 2010. Evaluation of natural antimicrobials on typical meat spoilage bacteria in vitro and in vacuum-packed pork meat. J. Food Sci. 75:M98–M102. Schmidt, S. E., G. Holub, J. M. Sturino, and T. M. Taylor. 2009. Suppression of Listeria monocytogenes Scott A in fluid milk by free and liposome-entrapped nisin. Probiotics Antimicrob. Prot. 1:152–158. Seifu, E., E. M. Buys, and E. F. Donkin. 2003. Effect of the lactoperoxidase system on the activity of mesophilic cheese starter cultures in goat milk. Int. Dairy J. 13:953–959. Sermon, J., K. Vanoirbeek, P. De Spiegeleer, R. Van Houdt, A. Aertsen, and C. W. Michiels. 2005. Unique stress response to the lactoperoxidase-thiocyanate enzyme system in Escherichia coli. Res. Microbiol. 156:225–232. Seward, R. A., R. H. Dielbel, and R. C. Lindsay. 1982. Effects of potassium sorbate and other antibotulinal agents on germination and outgrowth of Clostridium botulinum type E spores in microculture. Appl. Environ. Microbiol. 44:1212–1221. Shahidi, F., J. K. V. Arachchi, and Y.-J. Jeon. 1999. Food applications of chitin and chitosans. Trends Food Sci. Technol. 10:37–51. Sharma, A., and U. S. Sharma. 1997. Liposomes in drug d­elivery: progress and limitations. Int. J. Pharm. 154: 123–140. Shelef, L. A. 1994. Antimicrobial effects of lactates: a review. J. Food Prot. 57:445–450. Shelef, L. A., E. K. Jyothi, and M. A. Bulgarelli. 1984. Growth of enteropathogenic and spoilage bacteria in sagecontaining broth and foods. J. Food Sci. 49:737–740. Shen, C., I. Geornaras, P. A. Kendall, and J. N. Sofos. 2009. Control of Listeria monocytogenes on frankfurters by dipping in hops beta acids solutions. J. Food Prot. 72:702–706. Shen, C., and J. N. Sofos. 2008. Antilisterial activity of hops beta acids in broth with or without other antimicrobials. J. Food Sci. 73:M438–M442. Sheu, C. W., and E. Freese. 1972. Effects of fatty acids on growth and envelope proteins of Bacillus subtilis. J. Bacteriol. 111:516–524. Sheu, C. W., W. N. Konings, and E. Freese. 1972. Effects of acetate and other short-chain fatty acids on sugars and amino acid uptake of Bacillus subtilis. J. Bacteriol. 111:525–530. Sheu, C. W., D. Salomon, J. L. Simmons, T. Sreevalsan, and E. Freese. 1975. Inhibitory effects of lipophilic acids and related compounds on bacteria and mammalian cells. Antimicrob. Agents Chemother. 7:349–363. Simpson, W. J. 1993. Studies on the sensitivity of lactic acid bacteria to hop bitter acids. J. Inst. Brew. 99: 405–411.

SMP_Food Microbiology_CH30.indd

799

799

365. Singh, A. K., N. Singh, S. Sharma, S. B. Singh, P. Kaur, A. Srinivasan, and T. P. Singh. 2008. Crystal structure of lactoperoxidase at 2.4 Å resolution. J. Mol. Biol. 376: 1060–1075. 366. Singh, A. K., R. K. Singh, A. K. Bhunia, and N. Singh. 2003. Efficacy of plant essential oils as antimicrobial agents against Listeria monocytogenes in hotdogs. Lebenson. Wiss. Technol. 36:787–794. 367. Siragusa, G. R., and J. S. Dickson. 1993. Inhibition of Listeria monocytogenes, Salmonella typhimurium and Escherichia coli O157:H7 on beef muscle tissue by lactic or acetic acid contained in calcium alginate gels. J. Food Safety 13:147–158. 368. Siragusa, G. R., G. J. Haas, P. D. Matthews, R. J. Smith, R. J. Buhr, N. M. Dale, and M. G. Wise. 2008. Antimicrobial activity of lupulone against Clostridium perfringens in the chicken intestinal tract jejunum and caecum. J. Antimicrob. Chemother. 61:853–858. 369. Smith-Palmer, A., J. Stewart, and L. Fyfe. 2001. The potential application of plant essential oils as natural food preservatives in soft cheese. Food Microbiol. 18:463–470. 370. Smoot, L. A., and M. D. Pierson. 1981. Mechanisms of sorbate inhibition of Bacillus cereus T and Clostridium botulinum 62A spore germination. Appl. Environ. Microbiol. 42:477–483. 371. Soni, K. A., R. Nannapaneni, M. W. Schilling, and V. Jackson. 2010. Bactericidal activity of lauric arginate in milk and Queso Fresco cheese against Listeria monocy­ togenes cold growth. J. Dairy Sci. 93:4518–4525. 372. Stecchini, M. L., I. Sarais, and P. Giavedoni. 1993. Effect of essential oils on Aeromonas hydrophila in a culture medium and in cooked pork. J. Food Prot. 56:406–409. 373. Stopforth, J. D., P. N. Skandamis, P. M. Davidson, and J. N. Sofos. 2005. Naturally occuring compounds—animal sources, p. 453–505. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 374. Stopforth, J. D., J. N. Sofos, and F. F. Busta. 2005. Sorbic acid and sorbates, p. 49–90. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 375. Stopforth, J. D., D. Visser, R. Zumbrink, L. van Dijk, and E. W. Bontebal. 2010. Control of Listeria mono­ cytogenes on cooked cured ham by formulation with a lactate-diacetate blend and surface treatment with lauric arginate. J. Food Prot. 73:552–555. 376. Stratford, M., and P. A. Anslow. 1998. Evidence that sorbic acid does not inhibit yeast as a classic ‘weak acid preservative.’ Lett. Appl. Microbiol. 27:203–206. 377. Su, X., A. B. Howell, and D. H. D’Souza. 2010. Antiviral effects of cranberry juice and cranberry proanthocyanidins on foodborne viral surrogates—a time dependence study in vitro. Food Microbiol. 27:985–991. 378. Su, X., A. B. Howell, and D. H. D’Souza. 2010. The effect of cranberry juice and cranberry proanthocyanidins on the infectivity of human enteric viral surrogates. Food Microbiol. 27:535–540. 379. Taguri, T., T. Tanaka, and I. Kouno. 2004. Antimicrobial activity of 10 different plant polyphenols against

Manila Typesetting Company

11/08/2012 07:12AM

Preservatives and Preservation Methods

800

380. 381.

382.

383.

384.

385. 386.

387.

388. 389. 390.

391.

392. 393. 394. 395.

bacteria causing food-borne disease. Biol. Pharm. Bull. 27:1965–1969. Tamblyn, K. C., and D. E. Conner. 1997. Bactericidal activity of organic acids against Salmonella typhimurium attached to broiler chicken skin. J. Food Prot. 60:623–633. Taormina, P. J., and W. J. Dorsa. 2009. Short-term bactericidal efficacy of lauric arginate against Listeria monocytogenes present on the surface of frankfurters. J. Food Prot. 72:1216–1224. Taormina, P. J., and W. J. Dorsa. 2009. Inactivation of Listeria monocytogenes on hams shortly after vacuum packaging by spray application of lauric arginate. J. Food Prot. 72:2517–2523. Tassou, C. C., K. Lambropoulou, and G.-J. E. Nychas. 2004. Effect of prestorage treatments and storage conditions on the survival of Salmonella Enteriditis PT4 and Listeria monocytogenes on fresh marine and freshwater aquaculture fish. J. Food Prot. 67:193–198. Taylor, T. M., B. D. Bruce, J. Weiss, and P. M. Davidson. 2008. Listeria monocytogenes and Escherichia coli O157:H7 inhibition in vitro by liposome-encapsulated nisin and ethylene diaminetetraacetic acid. J. Food Safety 28:183–197. Taylor, T. M., P. M. Davidson, B. D. Bruce, and J. Weiss. 2005. Liposomal nanocapsules in food science and agriculture. Crit. Rev. Food Sci. Nutr. 45:587–605. Teraguchi, S., K. Shin, T. Ogata, M. Kingaku, A. Kaino, H. Miyauchi, Y. Fukuwatari, and S. Shimamura. 1995. Orally administered bovine lactoferrin inhibits bacterial translocation in mice fed bovine milk. Appl. Environ. Microbiol. 61:4131–4134. Teraguchi, S., K. Shin, K. Ozawa, S. Nakamura, Y. Fukuwatari, S. Tsuyuki, H. Namihira, and S. Shimamura. 1995. Bacteriostatic effect of orally administered bovine lactoferrin on proliferation of Clostridium species in the gut of mice fed bovine milk. Appl. Environ. Microbiol. 61:501–506. Thapon, J. L., and G. Brule. 1986. Effets du pH et de la forme ionize sur l’affinit lysozymes-caseines. Le Lait 66:19–30. Theron, M. M., and J. F. R. Lues. 2007. Organic acids and meat preservation: a review. Food Rev. Int. 23:141–158. Thippareddi, H., V. K. Juneja, R. K. Phebus, J. L. Marsden, and C. L. Kastner. 2003. Control of Clostridium perfrin­ gens germination and outgrowth by buffered sodium citrate during chilling of roast beef and injected pork. J. Food Prot. 66:376–381. Tokarskyy, O., and D. L. Marshall. 2008. Mechanism of synergistic inhibition of Listeria monocytogenes growth by lactic acid, monolaurin, and nisin. Appl. Environ. Microbiol. 74:7126–7129. Toldrá, F. 2009. Safety of Meat and Processed Meat. Springer, New York, NY. Tompkin, R. B. 2005. Nitrite, p. 169–236. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. Tompkin, R. B., L. N. Christiansen, and A. B. Shaparis. 1978. The effect of iron on botulinal inhibition in perishable canned cured meat. J. Food Technol. 13:521–527. Tong, C.-H., and F. A. Draughon. 1985. Inhibition by antimicrobial food additives of ochratoxin A production

SMP_Food Microbiology_CH30.indd

800

by Aspergillus sulphureus and Penicillium viridicatum. J. Food Prot. 49:1407–1411. 396. Touch, V., S. Hayakawa, S. Yamada, and S. Kaneko. 2004. Effects of a lactoperoxidase-thiocyanate-hydrogen peroxide system on Salmonella enteriditis in animal and vegetable foods. Int. J. Food Microbiol. 93:175–183. 397. Tranter, H. S. 1994. Lysozyme, ovotransferrin and avidin, p. 65–97. In V. M. Dillon and R. G. Board (ed.), Natural Antimicrobial Systems and Food Preservation, CAB Intl., Wallingford, UK. 398. Tranter, H. S., and R. G. Board. 1982. Review: the antimicrobial defense of avian eggs: biological perspective and chemical basis. J. Appl. Biochem. 4:295–338. 399. Tranter, H. S., and R. G. Board. 1984. Influence of incubation temperature and pH on the antimicrobial properties of hen egg albumen. J. Appl. Bacteriol. 56:53–61. 400. Tsai, S., and C. Chou. 1996. Injury, inhibition and inactivation of Escherichia coli O157:H7 by potassium sorbate and sodium nitrite as affected by pH and temperature. J. Sci. Food Agric. 71:10–12. 401. Ukuku, D. O., and W. F. Fett. 2004. Effect of nisin in combination with EDTA, sodium lactate, and potassium sorbate for reducing Salmonella on whole and fresh-cut cantaloupe. J. Food Prot. 67:2143–2150. 402. Ultee, A., M. H. J. Bennik, and R. Moezelaar. 2002. The phenolic hydroxyl group of carvacrol is essential for action against the food-borne pathogen Bacillus cereus. Appl. Environ. Microbiol. 68:1561–1568. 403. Ultee, A., L. G. M. Gorris, and E. J. Smid. 1998. Bactericidal activity of carvacrol towards the food-borne pathogen Bacillus cereus. J. Appl. Microbiol. 85:211–218. 404. Ultee, A., E. P. W. Kets, and E. J. Smid. 1999. Mechanisms of action of carvacrol on the food borne pathogen Bacillus cereus. Appl. Environ. Microbiol. 65:4606–4610. 405. Unlu, M., E. Ergene, G. V. Unlu, H. S. Zeytinoglu, and N. Vural. 2010. Composition, antimicrobial activity and in vitro cytotoxicity of essential oil from Cinnamomum zeylanicum (Lauraceae). Food Chem. Toxicol. 48:3274–3280. 406. U.S. Department of Agriculture, Food Safety and Inspection Service. 2010. Safe and suitable ingredients used in the production of meat, poultry, and egg products. Directive 7120.1, Rev. 2. Food Safety and Inspection Service, Washington, DC. www.fsis.usda.gov/OPPDE/rdad/ FSISDirectives/7120.1Rev2.pdf. 407. Valenti, P., P. Visca, G. Antonini, and N. Orsi. 1985. Antifungal activity of ovotransferrin towards genus Candida. Mycopathology 89:165–175. 408. Valero, M., and M. C. Salmerón. 2003. Antibacterial activity of 11 essential oils against Bacillus cereus in tyndallized carrot broth. Int. J. Food Microbiol. 85: 78–81. 409. Valkova, N., F. Lépine, C. Bollet, M. Dupont, and R. Villemur. 2002. prbA, a gene coding for an esterase hydrolyzing parabens in Enterobacter cloacae and Enterobacter gergoviae strains. J. Bacteriol. 184:5011–5017. 410. Valkova, N., F. Lépine, L. Labrie, M. Dupont, and R. Beaudet. 2003. Purification and characterization of PrbA, a new esterase from Enterobacter cloacae hydrolyzing

Manila Typesetting Company

11/08/2012 07:12AM

30.  Chemical Preservatives and Natural Antimicrobials

411.

412.

413. 414. 415.

416. 417.

418.

419.

420. 421. 422.

423. 424.

425. 426.

the esters of 4-hydroxybenzoic acid (parabens). J. Biol. Chem. 278:12779–12785. Vasavada, M., C. E. Carpenter, and D. P. Cornforth. 2003. Sodium levulinate and sodium lactate effects on microbial growth and stability of fresh pork and turkey sausages. J. Muscle Foods 14:119–129. Velugoti, P. R., L. Rajagopal, V. Juneja, and H. Thippareddi. 2007. Use of calcium, potassium, and sodium lactates to control germination and outgrowth of Clostridium perfringens spores during chilling of injected pork. Food Microbiol. 24:687–694. Venkitanarayanan, K. S., T. Zhao, and M. P. Doyle. 1999. Antibacterial effect of lactoferricin B on Escherichia coli O157:H7 in ground beef. J. Food Prot. 62:747–750. Viuda-Martos, M., Y. Ruiz-Navajas, J. Fernánez-López, and J. A. Pérez-Álvarez. 2008. Functional properties of honey, propolis, and royal jelly. J. Food Sci. 73:R117–R124. Wahba, N. M., A. S. Ahmed, and Z. Z. Ebraheim. 2010. Antimicrobial effects of pepper, parsley, and dill and their roles in the microbiological quality enhancement of traditional Egyptian Kareish cheese. Foodborne Pathog. Dis. 7:411–418. Walker, H. W., and M. A. Stahmann. 1955. Chemical nature of disease resistance in plants. Ann. Rev. Plant Physiol. 6:351–366. Walker, M., and C. A. Phillips. 2008. The effect of preservatives on Alicyclobacillus acidoterrestris and Propionibacterium cyclohexanicum in fruit juice. Food Control 19:974–981. Wan, J., A. Wilcock, and M. J. Coventry. 1998. The effect of essential oils of basil on the growth of Aeromonas hydrophila and Pseudomonas fluorescens. J. Appl. Microbiol. 84:152–158. Ward, S. M., P. J. Delaquis, R. A. Holley, and G. Mazza. 1998. Inhibition of spoilage and pathogenic bacteria on agar and pre-cooked roast beef by volatile horseradish distillates. Food Res. Int. 31:19–26. Weaver, R., and L. A. Shelef. 1993. Antilisterial activity of sodium, potassium or calcium lactate in pork liver sausage. J. Food Safety 13:133–146. Wendorff, W. L., W. E. Riha, and E. Muehlenkamp. 1993. Growth of molds on cheese treated with heat or liquid smoke. J. Food Prot. 56:963–966. Were, L. M., B. Bruce, P. M. Davidson, and J. Weiss. 2004. Encapsulation of nisin and lysozyme in liposomes enhances efficiency against Listeria monocytogenes. J. Food Prot. 67:922–927. Wibowo, D., R. Eschenbruch, C. R. Davis, G. H. Fleet, and T. H. Lee. 1985. Occurrence and growth of lactic acid bacteria in wine: a review. Am. J. Enol. Viticult. 36:302–313. Wilkins, K. M., and R. G. Board. 1989. Natural antimicrobial systems, p. 285–362. In G. W. Gould (ed.), Mechanisms of Action of Food Preservation Procedures. Elsevier Applied Science, London, United Kingdom. Wills, E. D. 1956. Enzyme inhibition by allicin, the active principal of garlic. Biochem. J. 63:514–520. Winterhalter, M., and D. D. Lasic. 1993. Liposome stability and formation: experimental parameters and theories on the size distribution. Chem. Phys. Lipids 64:35–43.

SMP_Food Microbiology_CH30.indd

801

801

427. Woods, L. F. J., and J. M. Wood. 1982. The effect of nitrite inhibition on the metabolism of Clostridium botuli­ num. J. Appl. Bacteriol. 52:109–110. 428. Woods, L. F. J., J. M. Wood, and P. A. Gibbs. 1981. The involvement of nitric oxide in the inhibition of the phophoroclastic system in Clostridium sporogenes by sodium nitrite. J. Gen. Microbiol. 125:399–406. 429. Yarbaeva, S. N., P. R. Velugoti, H. Thippareddi, and J. A. Albrecht. 2008. Evaluation of the microbial quality of Tajik sambusa and control of Clostridium perfringens germination and outgrowth by buffered sodium citrate and potassium lactate. J. Food Prot. 71:77–82. 430. Yousef, A. E., M. A. El-Shenawy, and E. H. Marth. 1989. Inactivation and injury of Listeria monocytogenes in a minimal medium as affected by benzoic acid and incubation temperature. J. Food Sci. 54:650–652. 431. Yuvamoto, P. D., and S. Said. 2007. Germination, duplication cycle and septum formation are altered by caffeine, caffeic acid and cinnamic acid in Aspergillus nidulans. Microbiology 76:735–738. 432. Zhang, H., Y. Shen, Y. Bao, Y. He, F. Feng, and X. Zheng. 2008. Characterization and syergistic antimicrobial activities of food-grade dilution-stable microemulsions against Bacillus subtilis. Food Res. Int. 41:495–499. 433. Zhang, H., Y. Shen, P. Weng, G. Zhao, F. Feng, and X. Zheng. 2009. Antimicrobial activity of a food-grade fully dilutable microemulsion against Escherichia coli and Staphylococcus aureus. Int. J. Food Microbiol. 135:211–215. 434. Zhang, H., H. Wei, Y. Cui, G. Zhao, and F. Feng. 2009. Antibacterial interactions of monolaurin with commonly used antimicrobials and food components. J. Food Sci. 74:M418–M421. 435. Zhang, H., Y. Xu, L. Wu, X. Zheng, S. Zhu, F. Feng, and L. Shen. 2010. Anti-yeast activity of a food-grade dilution-stable microemulsion. Appl. Microbiol. Biotechnol. 87:1101–1108. 436. Zhao, T., M. P. Doyle, and R. E. Besser. 1993. Fate of enterohemorrhagic Escherichia coli O157:H7 in apple cider with and without preservatives. Appl. Environ. Microbiol. 59:2526–2530. 437. Zhao, T., P. Zhao, and M. P. Doyle. 2009. Inactivation of Salmonella and Escherichia coli O157:H7 on lettuce and poultry skin by combinations of levulinic acid and sodium dodecyl sulfate. J. Food Prot. 72:928–936. 438. Zhao, T., P. Zhao, and M. P. Doyle. 2010. Inactivation of Escherichia coli O157:H7 and Salmonella Typhimurium DT104 on alfalfa seeds by levulinic acid and sodium dodecyl sulfate. J. Food Prot. 73:2010–2017. 439. Zhao, T., P. Zhao, J. W. West, J. K. Bernard, H. G. Cross, and M. P. Doyle. 2006. Inactivation of enterohemorrhagic Escherichia coli in rumen content- or fecescontaminated drinking water for cattle. Appl. Environ. Microbiol. 72:3268–3273. 440. Zivanovic, S., C. C. Basurto, S. Chi, P. M. Davidson, and J. Weiss. 2004. Molecular weight of chitosan influences antimicrobial activity in oil-in-water emulsions. J. Food Prot. 67:952–959.

Manila Typesetting Company

11/08/2012 07:12AM

SMP_Food Microbiology_CH30.indd

802

Manila Typesetting Company

11/08/2012 07:12AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch31

31

Thomas J. Montville Michael L. Chikindas

Biological Control of Foodborne Bacteria

This chapter provides an overview of the biologically based preservation technologies termed “biopreservation.” Biopreservation is defined as the use of microorganisms (including bacteriophages), their metabolic products, or both to preserve foods that are not generally considered fermented. The preservative, nutritional, and functional properties of fermented foods are covered in chapters 32 to 38. Acid production by lactic acid bacteria (LAB) in temperature-abused foods (controlled acidification) is covered in the first part of the chapter. Some LAB produce antimicrobial proteins, called bacteriocins, that inhibit spoilage and pathogenic bacteria without changing (e.g., through acidification, protein denaturation, and other processes) the physicochemical nature of the food. The largest section of this chapter deals with bacteriocins. The chapter closes by examining the use of bacteriophages as biocontrol agents.

BIOPRESERVATION BY COntrolled acidification Organic acids inhibit microbial growth, as discussed more fully in chapter 30. While organic acids are usu-

ally added to foods, LAB can produce lactic acid in situ. The controlled production of acid in situ is an important form of biopreservation. Many factors determine the effectiveness of in situ acidification. These include the product’s initial pH, its buffering capacity, target microorganism characteristics, the type and concentration of fermentable carbohydrates, ingredients that influence the viability and growth rate of LAB, and the growth rates of LAB and target pathogens at refrigerated and abuse temperatures (82). Clearly, such applications require customization. The production of bacteriocins, diacetyl, and hydrogen peroxide may also contribute to inhibition caused by culture fermentates. For example, Microgard® is a family of products that are added to much of the cottage cheese produced in the United States (33, 145) as GRAS (Generally Recognized as Safe) food preservatives. These products are made by fermenting milk (Microgard 100®) or dextrose (Microgard 200®) using Propionibacterium freudenreichii subsp. shermanii to produce acetate, propionic acid, and low-molecular-weight proteins such as bacteriocins. Although Microgard® products contain several antimicrobial substances, weak organic acids are believed to be the major contributors to their activity (5, 98).

Thomas J. Montville and Michael L. Chikindas, Department of Food Science, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-8520.

803

SMP_Food Microbiology_CH31.indd

803

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

804 The idea of using LAB to prevent botulinal toxigenesis through in situ acid production dates back to the 1950s. This technology relies on the inability of Clostridium botulinum to grow at pH of <4.6 as a defense against temperature abuse. LAB and a fermentable carbohydrate are added to the food. The LAB grow and produce acid only under conditions of temperature abuse. Under proper refrigeration, LAB cannot grow and no acid is formed. Saleh and Ordal (144) used a “normal cheese culture,” Lactobacillus bulgaricus or Lactococcus lactis, in experiments designed to prevent toxin production in chicken à la king containing a fermentable carbohydrate inoculated with spores of C. botulinum. When incubated at 30°C in the absence of LAB, 16 out of 16 samples rapidly became toxic. In the presence of the normal cheese culture, fewer than 3 of the 16 samples became toxic after 5 days at 30°C. In all samples to which the LAB had been added, including those positive for toxin, the pH was reduced to <4.5. There were no significant differences between cultures, and the inhibition was attributed to acid production. The discovery that carcinogenic nitrosamines are formed from nitrites in cured meats initiated a search for nitrite substitutes. Tanaka et al. (158) used controlled acidification to reduce nitrite concentrations in bacon. When bacon was inoculated with 3 log botulinal spores/g and incubated at 28°C, toxin was produced in 58% of the conventional bacon samples prepared with 120 mg/g nitrite but no sucrose or starter culture. When bacon was prepared with 80 or 40 mg/g nitrite, 0.7% sucrose, and starter cultures, £2% became toxic. The U.S. Department of Agriculture approved the “Wisconsin process” for bacon manufacture in 1986.

BACTERIOCINS Bacteriocins are ribosomally synthesized antimicrobial peptides of bacterial origin that are not lethal to the host. They are produced by virtually all bacterial species. Interest in bacteriocins produced by LAB has grown dramatically and is documented by many major reviews and books (see, e.g., references 42, 49, 69, 77, 105, 118, 135, and 157). Many bacteriocins inhibit foodborne pathogens of serious concern such as Listeria monocytogenes, which is recalcitrant to traditional preservation methods (62). In addition, bacteriocinogenic LAB are associated with, and are used as, starter cultures (e.g., see reference 3). The use of bacteriocins and/or bacteriocin-producing bacteria is attractive to the food industry due to growing consumer demand for natural products and increasing concern about foodborne diseases. However, this interest is tempered by regulatory uncertainty and concerns

SMP_Food Microbiology_CH31.indd

804

that the development of bacteriocin-resistant pathogens might render the technology ineffective (116).

General Characteristics

Bacteriocins produced by LAB are a heterogeneous group of ribosomally synthesized small proteins. They normally act against closely related species, but not against the producer. In this case, “closely related” is a loose term and can cover a wide range of gram-positive bacteria. Chelating agents, hydrostatic pressure, or injury leading to penetrability of the outer membrane can render gram-negative bacteria sensitive to bacteriocins (88, 89, 152). Stress factors that act on microbial cell targets different from the bacteriocin’s target often act synergistically to enhance bacteriocin activity against gram-positive bacteria (115). There are only a few reports on LAB bacteriocins active against gram-negative bacteria (156). Degradative enzymes such as lysozyme are not considered bacteriocins. However, in addition to acting on the membrane of the target cell, some bacteriocins (e.g., colicins produced by Escherichia coli) inhibit protein synthesis, degrade RNA, or have other biological functions (93). The seven characteristics of colicins cited in the Tagg et al. (157) review were initially extrapolated to all bacteriocins. Not all LAB bacteriocins meet all of these criteria, which would restrict the designation “bacteriocins” to only those plasmid-mediated proteins that are bactericidal to a narrow range of closely related bacteria having specific binding sites for that bacteriocin and whose biosynthesis ultimately kills the producing cell. In fact, it was originally suggested by Tagg, and later adopted by several research groups, to use the name “BLIS” (bacteriocin-like inhibitory substances) for all molecules different from “classic” bacteriocins. Bacteriocins differ in their spectra of activity, biochemical characteristics, and genetic determinants (90, 118). Most importantly, due to their unique nature and mechanism of action, bacteriocins have to be distinguished from pharmaceutical drugs (antibiotics) (26). Most bacteriocins are small (3 to 10 kDa), have a high isoelectric point, and contain both hydrophobic and hydrophilic domains. Klaenhammer (90) further classified bacteriocins into four major groups. Class I bacteriocins contain the unusual amino acids dehydroalanine, dehydrobutyrine, lanthionine, and bmethyllanthione (84). These amino acids are produced by posttranslational modification of serine and threonine to their dehydro forms. The dehydro amino acids react with cysteine to form thioether (single-sulfhydryl) lanthionine rings. Bacteriocins containing these lanthionine rings are commonly referred to as lantibiotics. There are

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria many structurally similar lantibiotics. Nisin, undoubtedly the best-characterized LAB bacteriocin (160), is produced in several related forms (see, e.g., reference 40). These forms of nisin have minor differences in their amino acid sequences: for instance, nisin A contains histidine at position 27, whereas nisin Z has an asparagine (114). The non-LAB bacteriocin subtilin, produced by Bacillus subtilis, also contains five lanthionine rings and a conformation similar to that of nisin but has other amino acid substitutions including a carboxy terminus two amino acids shorter than nisin. The 12 amino acids at the amino terminus of nisin and epidermin are similar, but epidermin lacks the central lanthionine ring common to nisin and subtilin and has a cyclized carboxy terminus. Lacticin 481 (130), lactocin S (113), and carnocin (154) are other lantibiotics produced by LAB. The genetics and mode of action of lantibiotics were recently reviewed by several authors (6, 11), and their possible use as lead substances in the next generation of antimicrobial drug discovery was discussed (131). Class II bacteriocins are small heat-stable proteins with a consensus leader sequence containing a GlyGly–1-Xaa+1 cleavage site important for processing the prebacteriocin during export. Class II is subdivided into three subgroups. Bacteriocins active against L. monocytogenes and having a -Tyr-Gly-Asn-Gly-Val-Xaa-Cys amino-terminal consensus sequence are classified in subclass IIa. Pediocin PA-1/AcH, sakacins A and P, leucocin A, bavaricin MN, and curvacin A are members of this subclass (49). Subclass IIb (117) contains bacteriocins such as lactococcin G, lactococcin M, lactacin F, and plantaricins EF and JK, which require two different peptides for activity (4, 109). Bacteriocins in subclass IIc, such as lactacin B, require reduced cysteine for activity. Bacteriocins in classes III and IV differ markedly from other bacteriocins (90). The larger (>30-kDa) heat-labile antimicrobial proteins such as helveticins J and V and lactacins A and B are classified as class III bacteriocins. Class IV bacteriocins such as leuconocin S, lactocin 27, and pediocin SJ-1 have lipid or carbohydrate moieties. The composition and function of the nonprotein portions are largely unknown. It may appear that classes III and IV serve the purpose of “lost and found” baskets due to the pronounced diversity of these groups’ representatives. Certainly, these two groups require further investigation. We find the classification system outlined by Klaenhammer to be the most scrupulous. However, over the years several other classifications of bacteriocins have been suggested, mostly targeted at simplification of the substances’ “taxonomy” and/or focused only on LAB-derived bacteriocins. These novel classifications

SMP_Food Microbiology_CH31.indd

805

805 include the ones from Cotter et al. (29, 30), Heng and Tagg (74), and many others. The most comprehensive overview of the history of bacteriocins’ discovery, research, and classification is presented in the recent review by Desriac et al. (42).

Methodological Considerations

Bacteriocin-producing bacteria are easy to isolate from foods. The methods used for their initial isolation and characterization are relatively simple and well established (103). The most common method for demonstrating bacteriocin production (the Bac+ phenotype) is to overlay a colony of the putative bacteriocin producer with an agar medium containing the bacterium being tested for sensitivity. An inhibition zone with sharp edges in the confluent growth of the target bacterium is presumptive evidence for bacteriocin production. Such zones can also be produced by acid, bacteriophage, hydrogen peroxide, or other nonspecific inhibitors. However, these zones resulting from weak organic acid production by the tested bacterium usually have fuzzy edges, and the zones produced by bacteriophages ob­viously do not have colonies of the producer micro­ organism in their centers. Negative-control experiments are required to confirm the production of bacteriocin. A positive control, confirming the proteinaceous nature of the inhibitor, is a straightforward test and can often be done on the same petri dish. In this case, a hole is punched in the solidified agar close to the colony of bacteria tested for bacteriocin production. This hole is filled with proteolytic enzyme, which, when diffused, will inactivate the bacteriocin, causing a semilunar inhibition zone appearance instead of a halo (see, e.g., Fig. 1 in Parret et al. [127]). Research beyond the initial isolation of the Bac+ bacteria and characterization of the inhibitor as a bacteriocin is more difficult. Some isolates produce distinctive inhibition zones on agar media but have no detectable activity in broth. High (10,000 to >50,000 Arbitrary Units/ml) bacteriocin activities (see below for definitions) in the culture supernatants facilitate bacteriocin purification. The purification usually involves salting out the protein followed by some combination of gel filtration, ion-exchange, and affinity and hydrophobicinteraction chromatography. Amino acid sequences are then determined from the electrophoretically pure protein. As more bacteriocins are being purified to the sequence level, it is becoming common to discover that independently purified bacteriocins are identical (see, e.g., reference 3). The lack of recognized standards for bacteriocin activity is a major impediment to progress in this field.

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

806 Only nisin has an international unit of activity. One gram of a commercially available nisin preparation (Nisaplin) contains 25 mg of pure nisin and is defined as having 1 million International Units (earlier known as Reading Units) of activity. Nisin activity is measured by the well diffusion assay of Tramer and Fowler (161) using Micrococcus luteus ATCC 10420 as the sensitive organism. M. luteus is also used to measure the activity of other peptide antimicrobials such as scorpion defensins (28) and magainins (99). However, some investigators use other indicator strains in assays for other bacteriocins if these strains are more sensitive (generate larger zone sizes) than M. luteus. Bacteriocin activity is estimated from the size of the inhibition zones produced by the diffusion of the bacteriocin in confluent lawns of bacteriocin-sensitive bacteria. The zone sizes obtained in diffusion assays are proportional to the log of the bacteriocin activity. Results are easily misinterpreted when zone sizes are reported without considering the log-linear nature of the assay (27). Arbitrary Units (AU) are more useful than zone sizes as quantitative measurements of bacteriocin activity and are usually determined by assaying twofold serial dilutions of the sample. The reciprocal of the highest dilution producing inhibition becomes the number of AU. This can be divided by the sample volume to yield AU/ml. The arbitrary nature of this system for measuring activity cannot be overemphasized. By virtue of the twofold dilutions, the assay is insensitive to activity differences that are less than twofold. The assumption of linearity with volume used to calculate the number of AU per ml is rarely verified. The choices of indicator bacterium, assay technique, length of incubation time, assay medium, and other factors are so idiosyncratic as to make it virtually impossible to compare the AU derived from different procedures. There is no easy solution to this problem. However, by assaying a known concentration of nisin under the same experimental conditions, an approximation of the relationship between AU and equivalent nisin International Units can be generated. Some investigators use conceptually different assays to generate other “Arbitrary Units” unrelated to those described above. For example, the ability to reduce growth rate, optical density, or viability by 50% can be a measure of bacteriocin activity (75). These assays do not require the 12- to 24-h incubation period of the diffusion assays. Methods of bacteriocin quantification that do not rely on biological activity are more quantitative since they do not depend on the sensitivity of an arbitrarily chosen indicator bacterium. Enzyme-linked immuno-

SMP_Food Microbiology_CH31.indd

806

sorbent assays work well if antibodies to the bacteriocin are available (see, e.g., references 13 and 100). A recombinant construction consisting of the gusA gene under the control of the nisin-regulated nisA promoter (41a, 129) can also be used to determine nisin concentration. Nisin induces activity of the nisA promoter and, subsequently, production of b-glucuronidase, whose activity can be measured. A bioluminescence assay for nisin quantification (122) is based on a recombinant construct consisting of the nisin-regulated promoter and bioluminescence genes luxAB from Xenorhabdus luminescens. L. lactis MG1614 cells transformed with such a recombinant DNA can detect as little as 0.0125 ng of nisin per ml (174).

Bacteriocin Applications in Foods

There are many different ways to use bacteriocins as food preservatives. The first is to add bacteriocins directly to the food for the purpose of inhibiting spoilage or pathogenic bacteria. Nisin is sold as a partially purified product of dairy or nondairy fermentation and marketed by companies under the names Nisaplin and Novasin, among others. The efficacy of pediocin PA-1 addition has also been demonstrated (for a review, see reference 140). In fact, pediocin PA-1 is an active component of the commercially available food preservative Alta 2341® (see, e.g., reference 128). The second way to use bacteriocins is to add bacteriocinogenic cultures to the food (22, 48) or use them as starter cultures that produce the bacteriocin in situ. A third way to use bacteriocins is to facilitate the use of defined starter cultures in fermented foods. The use of defined starter cultures rather than indigenous bacteria improves product quality and consistency. However, unless the indigenous bacteria can be inactivated, they usually predominate over the inocula. Because of this, the benefits of defined starter cultures are most pronounced in the dairy industry, where indigenous bacteria are inactivated during pasteurization. The use of bacteriocin-producing starter cultures in fermented meats and vegetables might inactivate the indigenous bacteria and facilitate the use of defined starter cultures. In recent years, significant attention has been given to the study and design of specific modes of controlled delivery of bacteriocins such as encapsulation, surface binding, etc. These techniques will ultimately allow for specific delivery of the antimicrobial to its site of action and extension of the product’s shelf life (for a review, see reference 7).

Addition of Nisin

Nisin is added to milk, cheese, and dairy products, a variety of canned foods, smoked fish, mayonnaise, and

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria baby foods throughout the world (160). It is GRAS (51) in a variety of applications and is used commercially as an antimastitis teat dip (148). Nisin has potential as a treatment for ulcers, in personal hygiene, and as a general sanitizing agent. When nisin is absorbed onto surfaces, it inhibits listerial growth (35) and prevents biofilm formation (14). Nisin also sensitizes spores to heat, so that the thermal treatment can be reduced (31). This application is not approved in the United States because it might conceal poor process control. Many nisin applications target the inhibition of C. botulinum. Spores of C. botulinum are much less nisin sensitive than are vegetative cells of L. monocytogenes. While nisin at concentrations as low as 200 IU/ml can reduce L. monocytogenes viability by 6 logs, concentrations of up to 10,000 IU/ml are required to obtain similar results with botulinal spores (111). Temperature is the major determinant of the inhibitory action of nisin. Nisin is much less effective at elevated temperatures than at refrigeration temperatures against C. botulinum 56A spores in a model food system (141). Eventually, growth in this system occurs when nisin activity falls below some threshold level, which is lower at decreasing temperatures. For example, botulinal growth occurred when residual nisin concentrations fell below 154 IU/ml at 35°C, but not until the nisin concentration was <12 IU/ml at 15°C. Many other factors influence the sporostatic efficacy of nisin (147, 160). In studies used to support the GRAS affirmation of nisin in pasteurized process cheese, nisin concentrations between 500 and 2,000 IU/ml inhibited botulinal spore outgrowth by 50% in broth, but concentrations of up to 10,000 IU/ml were ineffective in cooked meat medium (146). Nisin at 100 to 250 mg/g allows some salt reduction or increased moisture content in pasteurized process cheese spreads without elevating the risk of botulism (151). Specific phospholipids also decrease the antibotulinal efficacy of nisin (141), and fats in general are considered antagonistic due to their hydrophobicity (160). In most applications, nisin serves as one part of a multiple-barrier inhibitory system. Nisin may be a useful adjunct to modified-atmosphere storage (120). Nisin increases shelf life and delays toxin production by type E strains of C. botulinum in fresh fish packaged in a carbon dioxide atmosphere. However, toxin can sometimes be detected before samples are obviously spoiled (159). The combination of nisin and modified atmosphere to prevent L. monocytogenes growth in pork is more effective than either used alone (50). Nisin (5 mg/ml) added to liquid whole egg prior to pasteurization extends refrigerated shelf life from 6 to 11 days to 17 to 20 days (41).

SMP_Food Microbiology_CH31.indd

807

807 Most research has focused on defining conditions under which botulinal spores are inhibited by nisin. Inhibition of spores of other bacterial species also depends on many factors. The ability of nisin to inhibit thermally stressed Bacillus spores is influenced by the time-temperature combination used to affect the thermal stress, subsequent incubation temperature, pH, and even the type of acidulant used (125). In general, nisin is more effective at lower temperatures, against lower spore populations, and under acidic conditions.

Addition of Pediocin

While inactive against bacterial spores, pediocins also inhibit vegetative cells of L. monocytogenes. European patents cover the use of pediocin PA-1 as a dried powder or culture liquid to extend the shelf life of salads and salad dressing (63) and as an antilisterial agent in foods such as cream, cottage cheese, meats, and salads (160, 168). Pediocins are more effective than nisin in inhibiting bacteria in meat and are even more effective in dairy products. Dipping meat in 5,000 AU/ml crude pediocin PA-1 decreases the viability of attached L. monocytogenes by 100- to 1,000-fold. Pretreating meat with pediocin reduces subsequent L. monocytogenes attachment (119). Pediocin AcH (another name for pediocin PA-1) at 1,350 AU/ml reduces Listeria populations in ground beef, sausage, and other products by 1 to 7 logs. Pediocin AcH is more effective at 4°C than at 25°C against L. monocytogenes in wiener (frankfurter or hot dog) exudate. Emulsifiers such as Tween 80 or the entrapment of the pediocin in multilamellar vesicles increase pediocin’s effectiveness in fatty foods (37–39). In most cases, the bacteriocin rapidly reduces the viability of Listeria and delays growth of the survivors.

Addition of Bacteriocin-Producing Bacteria to Nonfermented Foods

Many antilisterial applications involve the addition of bacteriocinogenic culture rather than the bacteriocin. In wieners held at 4°C, L. monocytogenes grows after a 20- to 30-day lag period and increases from 4 to 6 log CFU/g by the end of 60 days (9). At 7 CFU/g, Bac+ and Bac– derivative strains of Pediococcus acidilactici inhibit L. monocytogenes growth in wieners for 60 days. Inhibition by the Bac– strain was not due to acidification or hydrogen peroxide. However, at low inoculum populations, the Bac+ strain, but not its Bac– derivative, extends the lag period of Listeria. The degree of inhibition increases with decreasing temperatures and, in this case, is greater under anaerobic conditions than aerobic conditions.

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

808 Lactobacillus bavaricus MN (which produces bavaricin MN) inhibits listerial growth in a model gravy system at 4°C, even in the absence of a fermentable carbohydrate (176, 177). The addition of a fermentable carbohydrate, reduction of incubation temperatures, and increased Lactobacillus-to-Listeria inoculation ratios all increase the inhibition. These variables also influence the success of this technology in “sous vide” beef cubes (176, 177). L. bavaricus causes L. monocytogenes viability to decline 10-fold over 6 weeks at 4°C under the most favorable conditions (4°C, high inoculation ratio, and beef packed in gravy containing a fermentable carbohydrate). Under the least favorable conditions (10°C, low inoculation ratio, beef cubes without gravy), listerial growth is inhibited by the bavaricin for 1 week but increases 100-fold in beef without the L. bavaricus. Carnobacterium piscicola LK5 is more effective against L. monocytogenes at 5°C than at 19°C in ultrahigh-temperature-treated milk, dog food made from beef, pasteurized crab meat, creamed corn, and wieners (20). C. piscicola LK5 inhibits L. monocytogenes Scott A, even when present at populations 100-fold higher than C. piscicola.

Use of Bacteriocin-Producing Starter Cultures to Improve Safety of Fermented Foods

If a food is to be fermented by LAB, the use of bacteriocin-producing starter cultures can add value to the product. For example, the presence of a nisin producer among the strains used to make Cheddar cheese increases the shelf life of pasteurized processed cheese in which it is used by 14 to 87 days at 22°C (138). Bacteriocinogenic pediococci appear to be especially effective in fermented meats. In situ pediocin production by P. acidilactici PAC 1.0 during the manufacture of fermented dry sausage reduces L. monocytogenes viability >10-fold relative to the acid-induced decrease caused by a nonbacteriocinogenic control (57). When wildtype Bac+ P. acidilactici H is used to ferment summer sausage, concentrations of pediocin up to 5,000 AU/g are produced, reducing L. monocytogenes viability by 3.4 log CFU/g (84).

Use of Bacteriocinogenic Starter Cultures to Direct the Fermentation of Fermented Foods

The use of undefined indigenous bacteria in some types of fermented foods compromises quality, makes process control difficult, and introduces an uncontrolled manufacturing variable. These problems can be overcome by the use of bacteriocinogenic starter cultures. Their ability to outgrow the indigenous microbiota (73, 132, 133, 172), thereby facilitating the use of defined

SMP_Food Microbiology_CH31.indd

808

starter cultures in unpasteurized foods, is a promising application. There are several novel applications of nisin or nisin-producing strains in fermented foods. The use of a nisin-producing L. lactis paired with nisin-resistant Leuconostoc mesenteroides in sauerkraut fermentations retards Lactobacillus plantarum growth (73). This allows L. mesenteroides to establish itself and results in a higher-quality product. The LAB found in wines are very sensitive to nisin and can be inhibited without inhibiting yeasts or influencing the taste (132, 133). At 100 IU/ml, nisin inhibits the bacteria that, by their malolactic fermentation, spoil wine. If the malolactic fermentation is desired, nisin-resistant Leuconostoc oenos can be added with nisin. This promotes malolactic fermentation and suppresses the indigenous LAB (34).

Genetics of LAB Bacteriocins Location of Bacteriocin Genes

The genetic information encoding bacteriocin production and immunity can be located on plasmids, on the chromosome, or both. For instance, the gene coding for enterocin AS-48RJ, a bacteriocin very similar to the cyclic peptide enterocin AS-48, is chromosomal, whereas genes coding for enterocin AS-4 and related bacteriocins are plasmid mediated (2). The nisin structural gene is located on the chromosome of L. lactis (21, 45, 134). The nisin gene resides within a 70-kb conjugative transposon and is genetically linked to genes coding for sucrose fermentation. Integration of this transposable element into the chromosome of a Nis– strain was observed after conjugation by probing the digested total DNA of transconjugants for different regions of the nisin-coding transposon (78). Other examples of chromosomally located LAB bacteriocins are helveticin J and lactacin B (90). At the same time, many of the class IIa bacteriocins, including the well-characterized pediocin PA-1/AcH, have their genes located on plasmids (58, 106, 170).

Organization of Bacteriocin Operons: a Generic Operon

The DNA sequence and amino acid composition of many bacteriocins are known (for a review, see reference 143). Therefore, a general picture of the genetic organization of bacteriocin genes is emerging. The structural gene encoding bacteriocin production appears to be located in an operon-like structure (90). The organization of a generic operon is shown in Fig. 31.1. It carries the genes coding for the bacteriocin prepeptide (structural gene), immunity, maturation, processing, and export of the bacteriocin molecules as well as genes encoding prod-

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria

809

Figure 31.1  A generic bacteriocin operon. The structural gene (struct) codes for a prepropeptide that is modified and excreted by the processing gene products (P1 and P2) and may be regulated by a signal transduction pathway coded for by reg1 and reg2. For additional abbreviations and explanations, see the text. doi:10.1128/9781555818463.ch31f1

ucts involved in the regulation of bacteriocin biosynthesis. The “real” operons are different from the “generic” model. Not all specific bacteriocin operons contain all of the genes incorporated into the generic bacteriocin gene cluster. The structural gene usually encodes a prepeptide that comprises the precursor of the mature bacteriocin preceded by an N-terminal extension or leader sequence. The secondary structure of this N-terminal extension is predicted to be an a-helix that is cleaved during the maturation or export process. Class II bacteriocins encode a prepeptide containing a consensus sequence (Gly–2-Gly–1) at the cleavage site and present strong homology in their hydrophobicity profile. These leader peptides play a significant role in the recognition of the precursor by the ABC transporters, which affect the proteolytic cleavage of the leader peptides and the export of the “maturated” (processed) bacteriocins across the membrane (166). Most bacteriocin leader sequences, except for a few such as divergicin A (178), lactococcin 972 (101), propionicin T1 (53), and some enterocins (25), do not exhibit characteristics of sec-dependent export proteins. Rather, they may be exported by a sec-independent pathway involving a transport protein encoded within the operon. In addition, the N-terminal extension may help neutralize bacteriocin activity within the producer strain, since a nonprocessed bacteriocin is much less antimicrobial than a mature peptide (136).

SMP_Food Microbiology_CH31.indd

809

The structural genes of many bacteriocins produced by LAB have been described and reviewed (41a, 69, 90). These include the structural genes coding for nisin (nisA), pediocin PA-1 (pedA), lactococcins A and B (lcnA and lcnB), sakacin A (sakA), leucocin A (leuA), lactacin F (lafA and lafX), and lactococcin M (lcnM and lcnN). The immunity gene makes Bac+ cells immune to their own bacteriocins. Immunity is specific and usually coordinated with bacteriocin production. Several immunity genes have been sequenced, and their ability to confer immunity to the producer strains has been proven (1, 76, 94, 164, 169). In the case of nisin, the nisI gene product is predicted to be an extracellular lipoprotein that, by anchoring to the membrane through its lipid moiety, confers immunity to the producer cells. Additional genes (nisE, nisF, and nisG) thought to be involved in immunity are in the nisin gene cluster (150). Class IIa operons for bacteriocins code for 11-kDa immunity proteins, which protect the producer cells. These immunity proteins are highly specific fourhelix bundle cytosolic molecules that establish immunity only to their own bacteriocins or to very closely related ones. The domain that recognizes the C-terminal membrane-penetrating region of class IIa bacteriocins is located in the C-terminal part of immunity proteins (for a review, see reference 56). Processing and export genes code for at least two proteins that ensure that the mature bacteriocin is formed

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

810 and exported from the cytoplasm. One of these proteins is an ATP-binding cassette translocator of the hemolysin B (HlyB) subfamily. These translocators (also known as ABC transporters) utilize ATP hydrolysis as an energy source for protein translocation and are characterized by the presence of two domains: a membrane-spanning domain and a cytoplasmic ATP-binding domain (52). Examples of genes coding for ATP-binding transporter proteins in LAB bacteriocins are nisT (nisin), lcnC (lactococcin A), and pedD (pediocin PA-1) (47, 102, 153). The second protein of the secretion apparatus is a structural homologue of HlyD, an accessory protein that may facilitate transport. The HlyD homologous protein does not seem to be present in the nisin gene cluster. Some of the ABC transporters found in bacteria that produce class II bacteriocins have similar N-terminal amino acid sequences. This domain may contain the proteolytic cleavage site for the N-terminal extension of these bacteriocins. In contrast, the nisin gene cluster contains the nisP gene. Its gene product, a serine protease, is thought to be involved in cleavage of the N-terminal extension of nisin. In addition, the nisin operon contains the genes nisB and nisC, which encode two membraneassociated proteins involved in catalyzing posttranslational modifications of precursor molecules (47, 169).

Quorum Sensing in Regulation of Bacteriocin Production

Some microbial genes are regulated to be expressed when the cell density has reached a certain population size, or quorum (8). This process is called quorum sensing or cell-to-cell signaling and involves synthesis and detection of small diffusible pheromones (5, 162). The observation that some bacteria produce bacteriocins in a cell-density-dependent manner led to a discovery of quorum-sensing involvement in the synthesis of these peptides. Originally it was noticed that when diluted in fresh media, some strains would stop producing bacteriocins. However, the synthesis would reoccur if filtersterilized medium from the same strain of cells of high density was added (46).

Regulation of Lantibiotics

Class I bacteriocins undergo extensive posttranslational modification prior to their secretion from the cell. Biosynthesis of at least some lantibiotics is regulated by quorum sensing. All genes necessary for nisin A production are arranged in one gene cluster. Expression of nisABTCIP is regulated by the PnisA promoter. This operon includes the structural gene for nisin precursor peptide (nisA) and genes necessary for maturation and export of and immunity to nisin (nisBT, nisCP, and nisI, re-

SMP_Food Microbiology_CH31.indd

810

spectively). Regulatory genes (nisRK) and the rest of the immunity genes (nisFEG) are under the control of the promoters PnisR and PnisF, respectively. NisK and NisR are a two-component signal transduction system. When a mature nisin A molecule binds to NisK, the signal is transduced to the response regulator NisR, which subsequently is able to bind to PnisA and PnisF (91). Subtilin has an almost identical regulatory system, with minor differences.

Regulation of Pediocinlike Bacteriocins

The synthesis of many class IIa bacteriocins such as plantaricin from L. plantarum is regulated by peptide pheromones (43). The signal transduction is conveyed via a standard two-component system. This system is sometimes called a three-component system when the structural gene for the pheromone itself is included (46). The quorum-sensing regulation of sakacin K has been studied in detail (18). The pheromone binds to a sensor kinase protein, which activates the response regulator. The response regulator interacts with promoters upstream of the regulatory and transport operons as well as with the promoter of the structural gene itself. The structural gene for the bacteriocin is under the stringent control of this system, while regulatory and transport genes appear to be less responsive to the regulation because the pheromone is exported by the very same ABC transporter (46).

Mechanism of Action against Vegetative Cells

Bacteriocins produced by LAB disrupt the integrity of the cytoplasmic membrane, increasing its permeability to low-molecular-weight compounds. The addition of bacteriocins to vegetative cells results in a rapid and nonspecific efflux of preaccumulated ions, amino acids, and—in some cases (nisin, lactostrepcin 5) but not others (pediocin PA-1)—ATP molecules (23, 24, 165, 171, 175). This increased flux of compounds rapidly dissipates chemical and electrical gradients across the membrane. The proton motive force, which serves as the major energy source for many vital processes, is dissipated within minutes after contact with bacteriocin (16, 17). Bacteriocin-treated cells have decreased levels of intracellular ATP. The loss of ATP can be caused by ATP efflux or hydrolysis. While efflux may be a direct result of membrane disruption, in other cases, such as with pediocin PA-1, hydrolysis can result from a shift in the ATP equilibrium due to Pi efflux or as a futile attempt of the cell to regenerate the proton motive force (1, 175). Ultimately, these changes in permeability render the cell unable to protect its cytoplasm from the environment. This leads to cell inhibition and often death.

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria Given the diversity in the biochemical attributes of bacteriocin molecules and genetic determinants, it is surprising that they all act by the common mechanism of membrane permeabilization. A structural motif similar to other antimicrobial peptides occurring in nature (see below) may explain their interaction with membranes (124). In this regard, most bacteriocins with known sequences are amphiphilic cationic peptides showing ahelix, b-sheet, or, in the case of lantibiotics, screw-like secondary structures. Based on the amphiphilic characteristics of bacteriocins, there are at least two different mechanisms that may explain their membrane-permeabilization action (Fig. 31.2). Bacteriocins may act by a multiple-step poration complex in which bacteriocin monomers bind, insert, and oligomerize in the cytoplasmic membrane to form a pore with the hydrophilic residues facing inward and the hydrophobic ones facing outward. The small size of bacteriocins makes it is unlikely that one molecule could form a pore. Alternatively, bacteriocins may disrupt the membrane integrity by a detergent-like membrane solubilization action. The action of nisin, pediocin PA-1, lactococcins A and B, and lactacin F in vivo is concentration and time dependent, supporting a poration complex mechanism (1, 24, 164, 171, 175). A generalized solubilization mechanism would result in an all-or-none lysis of the cells, with a sudden collapse of the bioenergetic parameters and no saturation kinetics. The addition of bacteriocins

811 to membrane vesicles or liposomes loaded with different-size probes does not always result in leakage of the probe. Rather, leakage depends on both the size of the probe and the amount of bacteriocin added (24, 60). Bacteriocins act similarly in that they permeabilize the cytoplasmic membrane. However, they may require different conditions in the target membrane to establish a successful interaction. In vivo, the lantibiotic bacteriocins act on energized membranes (59). The N-terminal part of the nisin molecule is involved in its insertion into the lipid bilayer, while the C-terminal moiety is involved in the initial interaction with the membrane surface (96). Prior to forming pores, nisin has to bind to the peptidoglycan precursor lipid II, which serves as a nisin-docking molecule in the membrane structure (15). Alternatively, most nonlantibiotic bacteriocins act on nonenergized membranes but seem to require a membrane receptor protein (24, 164, 171). Mutation in the rpoN gene, which codes for s54 in Enterococcus faecalis JH2-2, causes resistance to class IIa bacteriocins (36). This suggests a possible role for the rpoN gene product in the docking and/or receptor recognition of the target cell by a class IIa bacteriocin. Nissen-Meyer et al. (121) examined the sequences of four pediocinlike bacteriocins with different inhibitory spectra to identify regions that determine inhibitory specificity. The specificity of inhibitory activity correlates with the C-terminal sequences (43), suggesting that it determines specificity (reviewed in reference 44). Later, the same group (55) showed that a 15-residue pediocin PA-1 C-terminal derivative inhibits activity of the intact molecule. This suggests that the truncated peptide contains a receptor-binding motif that competes with the native pediocin PA-1 for the same surface receptor on the sensitive cells. The activity of some bacteriocins, including lactacin F, lactococcins M and G, and plantaricins EF and JK, occurs through the action of two peptides. These two-component bacteriocins are also postulated to form pores in the cytoplasmic membrane (90, 92, 93).

Mechanism of Action against Spores

Figure 31.2  Models for pore formation and detergentlike mechanisms of bacteriocin action. doi:10.1128/9781555818463.ch31f2

SMP_Food Microbiology_CH31.indd

811

Most information about the mechanisms of bacteriocin action against spores pertains to nisin. Spore germination and outgrowth is a multistep process detailed in chapter 3. Nisin allows spores to germinate and may act as a progerminant (105) but inhibits outgrowth of the preemergent spore. Heat resistance and refractility under phase microscopy are lost at this point. The growth of vegetative botulinal cells is inhibited at much lower nisin concentrations than those required to inhibit outgrowth of spores. Bacillus species spore coats opened by mechanical pressure are much more sensitive to nisin

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

812 than are spores of species with coats that are opened by lysis (160). At a molecular level, nisin modifies the sulfhydryl groups in the envelopes of germinated spores, presumably because the dehydro residues of nisin act as electron acceptors (67, 68).

Similarity to Other Antimicrobial Proteins

The antimicrobial proteins of LAB are not unique in structure or function. Many higher organisms produce similar cytolytic pore-forming proteins (124), including the cecropins, mellittins, magainins, and defensins. In many cases, these proteins have been better studied than the bacteriocins produced by LAB. A brief review of these antimicrobial proteins provides an opportunity to compare and contrast them with the LAB bacteriocins (110). Colicins produced by E. coli have molecular weights of 35 to 70 kDa and form ionpermeable channels in the cytoplasmic membrane of sensitive bacteria. Cecropins are small (ca. 25-aminoacid) peptides of the moth Hyalophora cecropia that are active against gram-negative and gram-positive bacteria. Cecropins contain amphiphilic helices consistent with the structural requirements for membrane insertion. Mellittin is the major toxic component of bee venom and causes membrane lysis, permeabilization, and inhibition of membrane-bound enzyme systems. It contains 26 amino acids, with the N-terminal region rich in hydrophobic residues and the C-terminal end predominantly hydrophilic. Magainins are antimicrobial peptides of the frog Xenopus laevis containing approximately 23 amino acids that exert microbicidal action by permeabilizing sensitive membranes. Defensins are small antimicrobial peptides found in human, rabbit, guinea pig, and rat phagocytes. These small (29- to 34-amino-acid) peptides also exert their antimicrobial activity by permeabilization or disruption of cell membranes. Clearly, antimicrobial proteins are common in the biosphere and are used by many organisms to protect themselves. While these molecules are the prehistoric immune system for bacteria, they remain the first line of defense for eucaryotes.

Resistance to Bacteriocins

While nisin-resistant starter cultures can be advantageous, the appearance of nisin-resistant pathogens can undermine the use of nisin as an antimicrobial. Genetically stable nisin-resistant L. monocytogenes can be isolated at a frequency of 10–6 (72, 107). Nisinresistant isolates are generated from vegetative cells of Staphylococcus aureus, Bacillus licheniformis, B. subtilis, Bacillus cereus (86, 107), and C. botulinum (104) at similar frequencies. The use of multiple bacteriocins

SMP_Food Microbiology_CH31.indd

812

to overcome this problem has been suggested (51, 71) but is effective only if resistance to each bacteriocin is conferred by different mechanisms. L. monocytogenes resistant to mesenterocin 52, curvaticin 13, or plantaricin C19 can be isolated at frequencies of 10–3 to 10–8. Strains resistant to any of these bacteriocins are cross resistant to the other two. When all three bacteriocins are used together, resistant strains are isolated at frequencies similar to those obtained when the bacteriocins are used alone (137). Isolates resistant to all three of these bacteriocins, however, are not resistant to nisin. Some mechanisms of bacteriocin resistance parallel mechanisms of antibiotic resistance (Table 31.1) (112). This comparison is made to provide a conceptual context for bacteriocin resistance and does not suggest that bacteriocin-resistant pathogens are cross resistant to antibiotics. Mechanisms characterized as “specific” have a biochemical specificity to a particular antimicrobial and cannot generate cross-resistance. General mechanisms of resistance, such as changes in membrane permeability, can be viewed as affecting the intrinsic resistance of the bacteria and might cause cross-resistance to other food preservatives. Membranes in nisin-resistant L. monocytogenes have more straight-chain fatty acids, resulting in a higher phase transition temperature (Tc) than that of the wild-type strain (107). Presumably, this lack of fluidity hinders nisin insertion into the membrane. Membrane fluidity plays an important role in resistance of Listeria to other antimicrobials (87). L. monocytogenes grown in the presence of C14:0 or C18:0 fatty acids has a higher Tc and increased resistance to four common antimicrobials compared with cells grown in the presence of C18:1, which have a lower Tc and are more sensitive to other antimicrobials. A different nisinresistant mutant (65) showed increased expression of a protein with strong homology to the glycosyltransferase domain of penicillin-binding proteins, a histidine protein kinase, a protein of unknown function, and ClpB (putative functions from homology). Increased expression of the putative penicillin-binding protein may affect the cell wall composition and thereby alter the sensitivity to the cell wall targeted by the bacteriocin. This mutant was also characterized by increased sensitivity to b-lactam antibiotics and by a slight decrease in sensitivity to another lantibiotic, mersacidin. For class IIa bacteriocins, decreased mannose-specific phosphoenolpyruvate-dependent phosphotransferase system gene expression, a more positive cell surface, increased lysinylation of membrane phospholipids, and increased membrane fluidity due to unsaturated phosphatidyl glycerol contribute to bacteriocin resistance (163).

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria

813

Table 31.1  Parallel mechanisms of antibiotic and bacteriocin resistancea Example Mechanism of resistance

Specificity

Destruction Modification Altered receptors Membrane composition

Specific or general Specific Specific General

a

Antibiotic b-Lactamases Methylation of aminoglycosides Penicillin-binding proteins Altered membranes in resistant E. coli and bacilli

Bacteriocins Protease, specific bacteriocinase Dehydroreductase (70) Probable, but not reported to date Demonstrated for nisin resistance (88, 90)

Adapted from reference 112.

The issue of potential cross-resistance between bacteriocins and antibiotics has been considered. Bacteriocins are different from antibiotics. They have different mechanisms of action. Therefore, cells respond to different stresses in different ways. This statement is confirmed by the sensitivity of multidrug-resistant bacteria to nisin (149). In order to survive, microorganisms can develop resistance to any antimicrobial substance used to preserve food. However, the use of intelligently designed multiple-hurdle technology not only allows inhibition of bacterial growth but also makes bacteriocin-resistant cells sensitive again (108). In addition, bacteriocinresistant bacteria may be even more sensitive to other hurdles used in food preservation (120). Our understanding of bacteriocin resistance is still incomplete. Under actual application conditions, the frequency of bacteriocin resistance may be much lower than that obtained in optimal laboratory media (72, 104). Nonetheless, it is prudent to conduct additional research on bacteriocin resistance so that it might be advantageously manipulated. Resistance management is already a component in other antimicrobial applications ranging from the use of antibiotics in hospitals to the use of Bacillus thuringiensis toxin as an agricultural insecticide.

Regulatory Status

LAB are GRAS for the production of fermented foods. GRAS status, which is conferred by the U.S. Food and Drug Administration (FDA), is especially desirable because it allows a compound to be used in a specific application without additional regulatory approval. The linkage of GRAS status to a specific application is often overlooked. Thus, the GRAS status of LAB for the production of fermented foods does not automatically make LAB, or their metabolic products, GRAS for uses such as the preservation of foods that are not fermented. Nisin is the only bacteriocin that has GRAS affirmation. The 1988 GRAS affirmation for the use of nisin in pasteurized processed cheese (54) was supported by toxicological data. This affirmation is the foundation for additional GRAS affirmations.

SMP_Food Microbiology_CH31.indd

813

Bacteriocins produced by GRAS organisms are not automatically GRAS themselves. Bacteriocins that are not GRAS are regulated as food additives (160) and require premarket approval by the FDA. A food fermented by bacteriocin-producing starters can be used as an ingredient in a second food product. Its use as an ingredient might coincidentally extend the shelf life of the product without necessitating preservative declarations. However, if the ingredient is added for the purpose of extending shelf life, the FDA would undoubtedly consider it an additive and require both premarket clearance and label declaration. Purified bacteriocins used as preservatives definitely require premarket approval by the FDA. GRAS status for bacteriocins can be based on documented use prior to 1958, a consensus of scientific opinion, or a formal GRAS affirmation from the FDA. The presence of bacteriocins in foods prior to 1958 might be inferred from the ease with which bacteriocinogenic bacteria are isolated from a variety of foods. This suggests that they are long-standing members of the natural microbiota of foods. Furthermore, strains that produce nisin (85, 97, 139) and strains that produce pediocin PA-1/AcH (10, 32, 61, 73) have been independently isolated from foods in different parts of the world. This demonstrates widespread occurrence of bacteriocinogenic bacteria in nature and suggests that bacteriocins have been consumed for decades. While these arguments are reasonable, the FDA might nonetheless require both isolation of the bacteriocinogenic organism and its bacteriocin from food produced prior to 1958 before accepting them as GRAS under the prior-use clause. The international regulation of bacteriocins is complex and beyond the scope of this chapter. Nisin is the only bacteriocin approved internationally for use in foods. The Joint Food and Agriculture/World Health Organization accepted nisin as a food additive in 1969 and set the maximum intake level as 33,000 IU per kg body weight. Based on this, many countries allow nisin in a variety of products, sometimes with no restrictions as to maximum level. In addition to milk, cheese, and

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

814 other dairy products, these uses include canned tomatoes, canned soups, other canned vegetables, mayonnaise, and baby food (167).

USE OF BACTERIOPHAGES FOR BIOPRESERVATION

General Characteristics

There is a long history of using bacteriophages to control pathogenic bacteria. American pharmaceutical companies marketed bacteriophage therapies in the 1930s. The Germans and Soviets used phage therapy to cure dysentery and other infections during World War II (155). With the advent of antibiotics in the 1940s, interest in bacteriophage therapies waned. However, bacteriophages are often considered to be effective antimicrobial weapons, especially where conventional methods (i.e., antibiotics) fail (126). The current specter of widespread antibiotic resistance has revived medical interest (12), and food scientists are also revisiting bacteriophage biopreservation, both to control the shedding of pathogens by animals and as preservatives in food systems (83). Bacteriophages are natural components of food microbiota and are routinely consumed as part of our diet. They are stable over a wide pH range but inactivated at 60 to 75°C. Their pre- and postharvest uses are being championed by companies to reduce animal carriage of Salmonella and E. coli O157:H7 and as a food additive for control of L. monocytogenes, as detailed below. Reviews (19, 66, 81) and a comprehensive book, Bacteriophage in the Control of Food- and Waterborne Pathogens, edited by Sabour and Griffiths (ASM Press, 2010), present the topic in greater detail than can be covered here. Bacteriophages are viruses whose only hosts are bacteria. Their structure consists of a tail that serves as the receptor site for phage adsorption, a head that contains the genetic material, and a sheath. When phages adsorb to the host, they release endolysins that degrade the cell wall. The capsid and sheath then undergo a conformational change that injects the genetic material into the target pathogen. Once the genetic material is in the pathogen, it seizes the growing cell’s metabolic machinery to produce phage components. The phage components self-assemble and are released during a lytic burst that kills the bacterial host. The lytic burst has the effect of dramatically increasing the number of phages available to attach to the remaining pathogens. If high concentrations of bacteriophages attach to the cells, they may release enough endolysin to lyse the cell independent of the lytic cycle. This is called “lysis from without.”

SMP_Food Microbiology_CH31.indd

814

Challenges Associated with Bacteriophage Biopreservation of Food

Using bacteriophages to control foodborne bacteria is fundamentally different from their use to control human infections. Human infections are characterized by a large population of a single bacterial species actively growing in a well-defined ecosystem. Treatment is temporally limited and does not create resistance in the pathogen reservoir. Resistance is created by changes in the pathogen receptor site. The bacterial absorption site for the bacteriophage can be proteins, lipopolysaccarides, or lipoproteins. In theory, changes in these surface receptors could give rise to bacteriophage resistance. However, there are an estimated 108 bacteriophage strains. More than 600 bacteriophage genomes have been sequenced. Ideally, bacteriophages should be used in cocktails containing, perhaps, as many as 30 different viral strains or types. Resistance to one phage type does not affect the efficacy of the remaining phage types, thus allowing for effective control of the targeted bacteria. Foods can contain small populations of several nongrowing pathogens amidst large populations of other bacteria. This is problematic because bacteriophage therapy in humans requires a threshold population (105 to 106 CFU) of actively growing bacteria. Thus, the use of bacteriophages for control of pathogens for which there is “zero tolerance” in food is extremely challenging. A more promising application might be controlling growth of spoilage bacteria when product is temperature abused, in a fashion similar to that of controlled acidification. Control of preslaughter fecal shedding of pathogens, where pathogen populations can reach 105 to 106 CFU per g, is also a possible application. Bacteriophage characteristics affect the efficacy of their application. These include multiplicity of infection (MOI), ability to self-replicate, requirement that the host be growing, and a limited host range. The MOI is the ratio of bacteriophage particles to pathogen cells. Higher MOIs enhance efficacy. Bacteriophages are selfreplicating and can increase the MOI during the lytic cycle. Reports of MOI values in the literature are confusing. They are easy to determine in vitro, in which the number of phages and pathogens are easily controlled. However, in animal systems, the phages can undergo nonspecific absorption to animal tissue or be diluted out by rumen contents. Thus, applications such as the reduction of E. coli O157:H7 (123) that appear very promising in vivo are ineffective in ruminants. Some investigators simply report the initial concentrations of phage and pathogen. Both approaches make it extremely difficult to compare results of different studies.

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria When the MOI is very high (106 to 108 phage/bacterium), the extracellular action of endolysins (lysis from without) can obviate the need for host growth. The mechanism is then strictly enzymatic rather than the result of the lytic cycle, and the MOI does not increase. Lytic bacteriophages have a limited host range. Phage sensitivity is determined by the ability of the bacteriophage to attach to cell receptors and insensitivity to the host restriction/modification system. A restricted host range can be advantageous because pathogens could be killed without destroying desirable fermentative or probiotic bacteria, normal spoilage microbiota, or resident human microbiota. However, multiple strains of bacteriophages would be required to kill multiple pathogens in the same food, or even multiple strains of the same pathogen (127). For example, 30 different O serotypes would have to be recognized to lyse the majority of enteropathogenic and enterotoxigenic E. coli. Changes in surface receptors or restriction/modification systems can make sensitive cells bacteriophage resistant, undermining the efficacy of bacteriophage biocontrol. Resistance might be controlled through the use of multiple-strain cocktails, rotation of the bacteriophage in the cocktails, or multiple hurdles to decrease the development of phage-resistant pathogens (95). Safety considerations are paramount for any of these approvals. The safety of bacteriophages may depend on their widespread occurrence in humans, food, drinking water, and the human gastrointestinal tract and their previous use for medical human therapy.

Applications in Animals

There is tremendous interest in controlling pathogens at the farm or field level. There have been varying degrees of success using bacteriophages to achieve this. Efficacy is affected by MOI, the ability to withstand gastric passage, and dilution in the rumen. It is possible, but far from conclusive, that bacteriophages can be used to control E. coli O157:H7 in cattle. In one study in which calves were inoculated with 108 CFU/ml of E. coli O157:H7 and then given a phage cocktail, there was a significant reduction in the pathogen, but the population rebounded to control levels within 16 h. This was not due to the development of resistant mutants. In another example, the population of E. coli O157:H7 was reduced by 2 to 3 log10 when sheep were treated with 1011 PFU. At this high application level, the reduction was most likely due to autolysin action. The MOI was also a key factor in bacteriophage biocontrol of Salmonella and Campylobacter on chicken skin (64). When applied at an MOI of 1, bacteriophages caused less than a 1-log10 reduction in the pathogen pop-

SMP_Food Microbiology_CH31.indd

815

815 ulation relative to untreated chicken skin. At an MOI of 100 to 1,000, the reduction was up to 100-fold. At an MOI of 105, no pathogens were recovered. Since MOIs of 107 can eliminate even Salmonella strains that can absorb bacteriophages, but are “phage resistant” due to their restriction/modification systems, the mechanism is probably lysis from without. OmniLytics, Inc. (Sandy, UT) received a letter of no objection for the application of bacteriophages to reduce the levels of E. coli O157:H7 and Salmonella spp. in poultry. The effect may be transient and dependent on the type and number of bacteriophage strains used. Bacteriophages have similar efficacy against Campylobacter species (173). Biophage Pharma (Montreal, Canada) has won approval for the use of Coli-Pro to treat E. coli infections in swine and for Salmo-Pro to treat Salmonella infections in poultry.

Applications in Food

In general, bacteriophages are best used for short-term inactivation rather than long-term preservation. When pathogens are present in high numbers, the reduction caused by bacteriophages can be low (typically 2 to 3 log10). There are reports of reductions of only 2 to 3 log10 in ripened cheese, 1 to 2 log10 in milk, and 2 log10 in wounded produce. However, when the initial load of pathogens is low, it can be reduced to undetectable levels. There are other reports of large reductions in specific foods. Repeated applications of 106 PFU of bacteriophage have produced 7-log10 reductions in L. monocytogenes populations in soft cheese. When beefsteak was inoculated with 105 CFU of E. coli O157:H7 and then treated with 108 PFU of bacteriophage, a 4log10 reduction was achieved. An evaluation of bacteriophage control of E. coli O157:H7 illustrates many issues associated with the use of bacteriophages. Single and three-strain phage cocktails reduce viable counts up to 5 log in a broth system at 30°C. At 7°C, there is no effect because the E. coli growth required for phage replication does not occur. At 37°C, phage biocontrol is initially effective but then fails as phage-resistant E. coli begin to dominate the population. These bacteriophage-insensitive mutants occur at a frequency of 10–6, even in multiphage cocktails, but revert to sensitivity after about 50 generations. Thus, if occasional bacteriophage-insensitive E. coli mutants are generated where they are not recycled into the pathogen reservoir (e.g., at the slaughterhouse rather than the farm), resistance should not reduce the efficacy of the control step. When meat inoculated with E. coli O157:H7 (100 CFU/g) was spotted with a phage cocktail (108 PFU), seven out of nine samples were

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

816 negative for E. coli O157:H7 after enrichment. At this high MOI (2 × 106), the lethality may be caused by lysis from without. In 2006, the FDA approved a novel food additive preparation: a cocktail of six bacteriophages active against L. monocytogenes. This foodborne-pathogen-controlling preparation was approved for use on meat cuts, sausages, etc. The EBI Food Safety (now Micreos Food Safety, Wageningen, The Netherlands) Listex P100 preparation received FDA-confirmed GRAS status for use as a food preservative in selected applications such as processed cheese (approved in 2006) and ready-to-eat products (approved in 2007), and is considered to be effective in controlling Listeria in ready-to-eat foods (70).

Conclusions

The use of bacteriophages to control pathogens in food has “shown promise” for decades. But, perhaps due to the difficulty of obtaining reproducible results in foods, they have not gained widespread use. While there is a clear potential for this technology, a number of fundamental scientific issues need to be addressed before bacteriophages become widely used as components of food preservation systems. Like bacteriocins, bacteriophages are not “silver bullets” but need to be used from a perspective that considers the microbial ecology of the food. The unique interplay of the specific bacteriophage with a specific pathogen in a specific food requires optimization of each specific food application and continued vigilance for acquisition of pathogen resistance.

CONCLUSIONS AND OUTLOOK FOR THE FUTURE The biological methods of food preservation covered here mark only the beginning of the biopreservation era in the food industry. Controlled acidification is conceptually straightforward, but its successful application depends on a variety of product-specific factors. This has limited both its commercial use and academic interest in controlled acidification. The use of antimicrobial proteins, in one form or another, is sure to increase. Table 31.2  Analogies between the use of insecticides in

production agriculture and the use of antimicrobials for food safety Control of insects in crops Chemical pesticides Integrated pest management Bioinsecticides Insecticidal plants

Control of bacteria in foods Chemical preservatives Microbial competition Bacteriocins Antimicrobial foods

SMP_Food Microbiology_CH31.indd

816

In production agriculture, B. thuringiensis insecticidal proteins have been applied to plants for the past 30 years. The genes for this protein are now being cloned into the plant itself. By analogy (Table 31.2), the day may come when resistance to pathogens can be genetically engineered into microbially sensitive foods. Research in the authors’ laboratory and preparation of this manuscript were supported by state appropriations, U.S. Hatch Act Funds, and grants from the U.S. Department of Agriculture CSREES NRI Food Safety Program. We acknowledge the contribution of Karen Winkowsi to an earlier version of this chapter.

References   1. Abee, T., F. M. Rombouts, J. Hugenholtz, G. Guihard, and L. Letellier. 1994. Mode of action of nisin Z against Listeria monocytogenes Scott A grown at high and low temperatures. Appl. Environ. Microbiol. 60:1962–1968.   2. Abriouel, H., R. Lucas, N. Ben Omar, E. Valdivia, M. Maqueda, M. Martínez-Cañamero, and A. Gálvez. 2005. Enterocin AS-48RJ: a variant of enterocin AS-48 chromosomally encoded by Enterococcus faecium RJ16 isolated from food. Syst. Appl. Microbiol. 28:383–397.   3. Albano, H., S. D. Todorov, C. A. van Reenen, T. Hogg, L. M. Dicks, and P. Teixeira. 2007. Characterization of two bacteriocins produced by Pediococcus acidilactici isolated from “Alheira,” a fermented sausage traditionally produced in Portugal. Int. J. Food Microbiol. 116:239–247.   4. Allison, G., C. Fremaux, C. Ahn, and T. R. Klaenhammer. 1994. Expansion of bacteriocin activity and host range upon complementation of two peptides encoded within the lactacin F operon. J. Bacteriol. 176:2235–2241.   5. Al-Zoreky, N., J. C. Ayres, and W. Sandine. 1991. Anti­ microbial activity of Microgard® against food spoilage and pathogenic microorganisms. J. Dairy Sci. 74:758–763.   6. Asaduzzaman, S. M., and K. Sonomoto. 2009. Lantibiotics: diverse activities and unique modes of action. J. Biosci. Bioeng. 107:475–487.   7. Balasubramanian, A., L. E. Rosenberg, K. Yam, and M. L. Chikindas. 2009. Antimicrobial packaging: potential vs. reality—a review. J. Appl. Packaging Res. 3:193–221.   8. Bassler, B. L. 2002. Small talk. Cell-to-cell communication in bacteria. Cell 109:421–424.   9. Berry, E. D., R. W. Hutkins, and R. Mandigo. 1991. The use of bacteriocin producing Pediococcus acidilactici to control post processing Listeria monocytogenes contamination of frankfurters. J. Food Prot. 54:681–686.   10. Bhunia, A. K., and M. C. Johnson. 1992. Monoclonal antibody-colony immunoblot method specific for isolation of Pediococcus acidilactici from foods and correlation with pediocin (bacteriocin) production. Appl. Environ. Microbiol. 58:2315–2320.   11. Bierbaum, G., and H. G. Sahl. 2009. Lantibiotics: mode of action, biosynthesis and bioengineering. Curr. Pharm. Biotechnol. 10:2–18.

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria   12. Biswas, B., S. Adhya, P. Washart, B. Paul, A. N. Trostel, B. Powell, R. Carlton, and C. R. Merril. 2002. Bacteriophage therapy rescues mice bacteremic from a clinical isolate of vancomycin-resistant Enterococcus faecium. Infect. Immun. 70:204–210.   13. Bouksaim, M., C. Lacroix, R. Bazin, and R. E. Simard. 1999. Production and utilization of polyclonal antibodies against nisin in an ELISA and for immuno-location of nisin in producing and sensitive bacterial strains. J. Appl. Microbiol. 87:500–510.   14. Bower, C. K., J. McGuire, and M. A. Daeschel. 1995. Suppression of Listeria monocytogenes colonization following adsorption of nisin onto silica surfaces. Appl. Environ. Microbiol. 61:992–997.   15. Breukink, E., I. Wiedemann, C. van Kraaij, O. P. Kuipers, H. Sahl, and B. de Kruijff. 1999. Use of the cell wall precursor lipid II by a pore-forming peptide antibiotic. Science 286:2361–2364.   16. Bruno, M. E. C., A. Kaiser, and T. J. Montville. 1992. Depletion of proton motive force by nisin in Listeria monocytogenes cells. Appl. Environ. Microbiol. 58:2255–2259.   17. Bruno, M. E. C., and T. J. Montville. 1993. Common mechanistic action of bacteriocins from lactic acid bacteria. Appl. Environ. Microbiol. 59:3003–3010.   18. Brurberg, M. B., I. F. Nes, and V. G. Eijsink. 1997. Pheromone-induced production of antimicrobial peptides in Lactobacillus. Mol. Microbiol. 26:347–360.   19. Brüsson, H. 2005. Phage therapy: the Escherichia coli experience. Microbiology 151:2133–2140.   20. Buchanan, R. L., and L. A. Klawitter. 1992. Effectiveness of Carnobacterium piscicola LK5 for controlling the growth of Listeria monocytogenes Scott A in refrigerated foods. J. Food Safety 12:217–224.   21. Buchman, G. W., S. Banergee, and J. N. Hansen. 1988. Structure, expression and evolution of a gene encoding the precursor of nisin, a small protein antibiotic. J. Biol. Chem. 263:16260–16266.   22. Buyong, N., J. Kok, and J. B. Luchansky. 1998. Use of a genetically enhanced, pediocin-producing starter culture, Lactococcus lactis subsp. lactis MM217, to control Listeria monocytogenes in cheddar cheese. Appl. Environ. Microbiol. 64:4842–4845.   23. Chen, Y., and T. J. Montville. 1995. Efflux of ions and ATP depletion induced by pediocin PA-1 are concomitant with cell death in Listeria monocytogenes Scott A. J. Appl. Bacteriol. 79:684–690.   24. Chikindas, M. L., M. J. García-Garcerá, A. J. M. Driessen, A. M. Ledeboer, J. Nissen-Meyer, I. F. Nes, T. Abee, W. N. Konings, and G. Venema. 1993. Pediocin PA-1, a bacteriocin from Pediococcus acidilactici PAC1.0, forms hydrophilic pores in the cytoplasmic membrane of target cells. Appl. Environ. Microbiol. 59:3577–3584.   25. Cintas, L. M., P. Casaus, C. Herranz, L. S. Håvarstein, H. Holo, P. E. Hernández, and I. F. Nes. 2000. Biochemical and genetic evidence that Enterococcus faecium L50 produces enterocins L50A and L50B, the sec-dependent enterocin P, and a novel bacteriocin secreted without an N-terminal extension termed enterocin Q. J. Bacteriol. 182:6806–6814.

SMP_Food Microbiology_CH31.indd

817

817   26. Cleveland, J., T. J. Montville, I. F. Nes, and M. L. Chikindas. 2001. Bacteriocins: safe, natural antimicrobials for food preservation. Int. J. Food. Microbiol. 71:1–20.   27. Cleveland, J., M. Chikindas, and T. J. Montville. 2002. Multimethod assessment of commercial nisin preparations. J. Ind. Microbiol. Biotechnol. 29:228–232.   28. Cociancich, S., M. Goyffon, F. Bontems, P. Bulet, F. Bouet, A. Menez, and J. Hoffman. 1993. Purification and characterization of a scorpion defensin, a 4kDa anti­ bacterial peptide presenting structural similarity with insect defensins and scorpion toxins. Biochem. Biophys. Res. Commun. 194:17–22.   29. Cotter, P. D., C. Hill, and R. P. Ross. 2005. Bacteriocins: developing innate immunity for food. Nat. Rev. Microbiol. 3:777–788.   30. Cotter, P. D., C. Hill, and P. R. Ross. 2006. What’s in a name? Class distinction for bacteriocins. (Author’s reply.) Nat. Rev. Microbiol. doi:10.1038/nrmicro1273-c1.   31. Cruz, J., and T. J. Montville. 2008. Influence of nisin on the resistance of Bacillus anthracis Sterne spores to heat and hydrostatic pressure. J. Food Prot. 71:196–199.   32. Daba, H., C. Lacroix, J. Huang, R. E. Simard, and L. Lemieux. 1994. Simple method of purification and sequencing of a bacteriocin produced by Pediococcus acidilactici UL5. J. Appl. Bacteriol. 77:682–698.   33. Daeschel, M. A. 1989. Antimicrobial substances from lactic acid bacteria for use as food preservatives. Food Technol. 43:164–166.   34. Daeschel, M. A. 1990. Controlling wine malolactic fermentation with nisin and nisin-resistant strains of Leuconostoc oenos. Appl. Environ. Microbiol. 51:601–603.   35. Daeschel, M. A., J. McGuire, and H. Al-Makhlafi. 1992. Antimicrobial activity of nisin adsorbed to hydrophilic and hydrophobic silicon surfaces. J. Food Prot. 55:731–735.   36. Dalet, K., C. Briand, Y. Cenatiempo, and Y. Héchard. 2000. The rpoN gene of Enterococcus faecalis directs sensitivity to subclass IIa bacteriocins. Curr. Microbiol. 41:441–443.   37. Degnan, A. J., N. Buyong, and J. B. Luchansky. 1993. Antilisterial activity of pediocin AcH in model food systems in the presence of an emulsifier or encapsulated within liposomes. Int. J. Food Microbiol. 18:127–138.   38. Degnan, A. J., and J. B. Luchansky. 1992. Influence of beef tallow and muscle on the antilisterial activity of pediocin AcH and liposome-encapsulated pediocin AcH. J. Food Prot. 55:552–554.   39. Degnan, A. J., A. E. Yousef, and J. B. Luchansky. 1992. Use of Pediococcus acidilactici to control Listeria monocytogenes in temperature-abused vacuum-packaged wieners. J. Food Prot. 55:98–103.   40. de Kwaadsteniet, M., K. ten Doeschate, and L. M. Dicks. 2008. Characterization of the structural gene encoding nisin F, a new lantibiotic produced by a Lactococcus lactis subsp. lactis isolate from freshwater catfish (Clarias gariepinus). Appl. Environ. Microbiol. 74:547–549.

Manila Typesetting Company

11/08/2012 07:13AM

818

Preservatives and Preservation Methods

  41. Delves-Broughton, J., G. C. Williams, and S. Williamson. 1992. The use of the bacteriocin, nisin, as a preservative in pasteurized white egg. Lett. Appl. Microbiol. 15: 133–136.   41a. de Ruyter, P. G., O. P. Kuipers, and W. M. de Vos. 1996. Controlled gene expression systems for Lactococcus lactis with the food-grade inducer nisin. Appl. Environ. Microbiol. 62:3662–3667.   42. Desriac, F., D. Defer, N. Bourgougnon, B. Brillet, P. Le Chevalier, and Y. Fleury. 2010. Bacteriocin as weapons in the marine animal-associated bacteria warfare: inventory and potential applications as an aquaculture probiotic. Mar. Drugs 8:1153–1177.   43. Diep, D. B., L. S. Håvarstein, J. Nissen-Meyer, and I. F. Nes. 1994. The gene encoding plantaricin A, a bacteriocin from Lactobacillus plantarum C11, is located on the same transcription unit as an agr-like regulatory system. Appl. Environ. Microbiol. 60:160–166.   44. Diep, D. B., and I. F. Nes. 2002. Ribosomally synthesized antibacterial peptides in Gram positive bacteria. Curr. Drug Targets 3:107–122.   45. Dodd, H. M., N. Horn, and M. J. Gasson. 1990. Analysis of the genetic determinant for the production of the peptide antibiotic nisin. J. Gen. Microbiol. 136:555–556.   46. Eijsink, V. G., L. Axelsson, D. B. Diep, L. S. Håvarstein, H. Holo, and I. F. Nes. 2002. Production of class II bacteriocins by lactic acid bacteria; an example of biological warfare and communication. Antonie van Leeuwenhoek 81:639–654.   47. Engelke, G., Z. Gutowski-Eckel, M. Hammelmann, and K. D. Entian. 1992. Biosynthesis of the lantibiotic nisin: genomic organization and membrane localization of the NisB protein. Appl. Environ. Microbiol. 58:3730–3743.   48. Ennahar, S., O. Assobhel, and C. Hasselmann. 1998. Inhibition of Listeria monocytogenes in a smear-surfac­e soft cheese by Lactobacillus plantarum WHE 92, a pediocin AcH producer. J. Food Prot. 61:186–191.   49. Ennahar, S., T. Sashihara, K. Sonomoto, and A. Ishizaki. 2000. Class IIa bacteriocins: biosynthesis, structure and activity. FEMS Microbiol. Rev. 24:85–106.   50. Fang, T. J., and L. W. Lin. 1994. Inactivation of Listeria monocytogenes on raw pork treated with modified atmosphere packaging and nisin. J. Food Drug Anal. 2:189–200.   51. Farber, J. M. 1993. Current research on Listeria monocytogenes in foods: an overview. J. Food Prot. 56:640–643.   52. Fath, M. J., and R. Kolter. 1993. ABC transporters: bacterial exporters. Microbiol. Rev. 57:995–1017.   53. Faye, T., T. Langsrud, I. F. Nes, and H. Holo. 2000. Biochemical and genetic characterization of propionicin T1, a new bacteriocin from Propionibacterium thoenii. Appl. Environ. Microbiol. 66:4230–4236.   54. Federal Register. 1988. Nisin preparation: affirmation of GRAS status as a direct human food ingredient. 21CFR Part 184. 53:11247–11251.   55. Fimland, G., O. R. Blingsmo, K. Sletten, G. Jung, I. F. Nes, and J. Nissen-Meyer. 1996. New biologically a­ctive hybrid bacteriocins constructed by combining

regions from various pediocin-like bacteriocins: the Cterminal region is important for determining specificity. Appl. Environ. Microbiol. 62:3313–3318. Fimland, G., L. Johnsen, B. Dalhus, and J. NissenMeyer. 2005. Pediocin-like antimicrobial peptides (class IIa bacteriocins) and their immunity proteins: biosynthesis, structure, and mode of action. J. Pept. Sci. 11: 688–696. Foegeding, P. M., A. B. Thomas, D. H. Pinkerton, and T. R. Klaenhammer. 1992. Enhanced control of Listeria monocytogenes by in situ-produced pediocin during dry fermented sausage production. Appl. Environ. Microbiol. 58:884–890. Fremaux, C., Y. Héchard, and Y. Cenatiempo. 1995. Mesentericin Y105 gene clusters in Leuconostoc mesenteroides Y105. Microbiology 141:1637–1645. Gao, F. H., T. Abee, and W. N. Konings. 1991. The mechanism of action of the peptide antibiotic nisin in liposomes and cytochrome c oxidase proteoliposomes. Appl. Environ. Microbiol. 57:2164–2170. García Garcerá, M. J., M. G. L. Elferink, A. J. M. Driessen, and W. N. Konings. 1993. In vitro poreforming activity of the lantibiotic nisin: role of protonmotive force and lipid composition. Eur. J. Biochem. 212:417–422. Garver, K. I., and P. M. Muriana. 1993. Detection, identification and characterization of bacteriocin-producing lactic acid bacteria from retail food products. Int. J. Food Microbiol. 19:241–258. Gandhi, M., and M. L. Chikindas. 2007. Listeria: a foodborne pathogen that knows how to survive. Int. J. Food Microbiol. 113:1–15. Gonzalez, C. F. 1988. Method for inhibiting bacterial spoilage and composition for this purpose. European patent application 88101624. Goode, D., V. M. Allen, and P. A. Barrow. 2003. Reduction of experimental Salmonella and Campylobacter contamination of chicken skin by application of lytic bacteriophage. Appl. Environ. Microbiol. 69:5032–5036. Gravesen, A., K. Sørensen, F. M. Aarestrup, and S. Knøchel. 2001. Spontaneous nisin-resistant Listeria monocytogenes mutants with increased expression of a putative penicillin-binding protein and their sensitivity to various antibiotics. Microb. Drug Resist. 7:127–135. Greer, G. G. 2005. Bacteriophage control of foodborne bacteria. J. Food Prot. 68:1102–1111. Gross, E., and J. L. Morell. 1967. The presence of a dehydroalanine in the antibiotic nisin and its relationship to activity. J. Am. Chem. Soc. 89:2791–2792. Gross, E., and J. L. Morell. 1971. The structure of nisin. J. Am. Chem. Soc. 93:4634–4635. Guder, A., I. Wiedemann, and H. G. Sahl. 2000. Posttranslationally modified bacteriocins—the lantibiotics. Biopolymers 55:62–73. Guenther, S., D. Huwyler, S. Richard, and M. J. Loessner. 2009. Virulent bacteriophage for efficient biocontrol of Listeria monocytogenes in ready-to-eat foods. Appl. Environ. Microbiol. 75:93–100.

SMP_Food Microbiology_CH31.indd

818

  56.

  57.

  58.   59.

  60.

  61.

  62.   63.   64.

  65.

  66.   67.   68.   69.   70.

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria   71. Hanlin, M. B., N. Kalchayan, P. Ray, and B. Ray. 1993. Bacteriocins of lactic acid bacteria in combination have greater antibacterial activity. J. Food Prot. 56:252–255.   72. Harris, L. J., H. P. Fleming, and T. R. Klaenhammer. 1991. Sensitivity and resistance of Listeria monocytogenes ATCC 19115 Scott A and VAL 500 to nisin. J. Food Prot. 54:836–840.   73. Harris, L. J., H. P. Fleming, and T. R. Klaenhammer. 1992. Novel paired starter culture system for sauer­kraut, consisting of a nisin resistant Leuconostoc mesenteroides strain and a nisin-producing Lactococcus lactis strain. Appl. Environ. Microbiol. 58:1484–1489.   74. Heng, N. C. K., and J. R. Tagg. 2006. What’s in a name? Class distinction for bacteriocins. Nat. Rev. Microbiol. doi: 10.1038/nrmicro1273-c1.   75. Herranz, C., Y. Chen, H. J. Chung, L. M. Cintas, P. E. Hernández, T. J. Montville, and M. L. Chikindas. 2001. Enterocin P selectively dissipates the membrane potential of Enterococcus faecium T136. Appl. Environ. Microbiol. 67:1689–1692.   76. Holo, H., O. Nissen, and I. F. Nes. 1991. Lactococcin A, a new bacteriocin from Lactococcus lactis subsp. cremoris: isolation and characterization of the protein and its gene. J. Bacteriol. 173:3879–3887.   77. Hoover, D. G., and L. R. Steenson. 1993. Bacteriocins of Lactic Acid Bacteria. Academic Press, New York, NY.   78. Horn, N., S. Swindell, H. Dodd, and M. Gasson. 1991. Nisin biosyntheis genes are encoded by a novel conjugative transposon. Mol. Gen. Genet. 228:129–135.   79. Reference deleted.   80. Reference deleted.   81. Hudson, J. A., C. Billington, C. Carey-Smith, and G. Greening. 2005. Bacteriophages as biocontrol agents in food. J. Food Prot. 68:426–437.   82. Hutton, M. T., P. A. Chehak, and J. H. Hanlin. 1991. Inhibition of botulism toxin production by Pediococcus acidilactici in temperature abused refrigerated foods. J. Food Safety 11:255–267.   83. Institute of Food Technologists. 2001. Analysis and evaluation of preventive control measures for the control and reduction/elimination of microbial hazards on fresh and fresh-cut produce. Compr. Rev. Food Sci. Food Safety 2(Suppl. 1):13–37.   84. Jack, R. W., and G. Jung. 2000. Lantibiotics and microcins: polypeptides with unusual chemical diversity. Curr. Opin. Chem. Biol. 4:310–317.   85. Jager, K., and S. Harlander. 1992. Characterization of a bacteriocin from Pediococcus acidilactici PC and comparison of bacteriocin-producing strains using molecular typing procedures. Appl. Microbiol. Biotechnol. 37:631–637.   86. Jarvis, B., and J. Farr. 1971. Partial purification, specificity and mechanism of the nisin-inactiviating enzyme from Bacillus cereus. Biochim. Biophys. Acta 227:232–240.   87. Juneja, V. K., and P. M. Davidson. 1993. Influence of altered fatty acid composition on resistance of Listeria monocytogenes to antimicrobials. J. Food Prot. 56:302–305.

SMP_Food Microbiology_CH31.indd

819

819   88. Kalchayanand, N., M. B. Hanlin, and B. Ray. 1992. Sublethal injury makes Gram-negative and resistant Gram-positive bacteria sensitive to the bacteriocins, pediocin AcH and nisin. Lett. Appl. Microbiol. 15:239–243.   89. Kalchayanand, N., T. Sikes, C. P. Dunne, and B. Ray. 1994. Hydrostatic pressure and electroporation have increased bactericidal efficiency in combination with bacteriocins. Appl. Environ. Microbiol. 60:4174–4177.   90. Klaenhammer, T. R. 1993. Genetics of bacteriocins produced by lactic acid bacteria. FEMS Microbiol. Rev. 12:39–86.   91. Kleerebezem, M. 2004. Quorum sensing control of lantibiotic production; nisin and subtilin autoregulate their own biosynthesis. Peptides 25:1405–1414.   92. Kok, J., H. Holo, M. J. van Belkum, A. J. Haandrikman, and I. F. Nes. 1993. Nonnisin bacteriocins in lactococci: biochemistry, genetics and mode of action, p. 121–150. In D. G. Hoover and L. R. Steenson (ed.), Bacteriocins of Lactic Acid Bacteria. Academic Press, New York, NY.   93. Konisky, J. 1982. Colicins and other bacteriocins with established modes of action. Ann. Rev. Microbiol. 36:125–144.   94. Kuipers, O. P., M. M. Beerthuyzen, R. J. Siezen, and W. M. de Vos. 1993. Characterization of the nisin gene cluster nisABTCIPR of Lactococcus lactis: requirement of expression of the nisA and nisI gene for producer immunity. Eur. J. Biochem. 216:281–292.   95. Leverentz, B., W. S. Conway, M. J. Camp, W. J. Janisiewcz, T. Abuladze, M. Yang, R. Saftner, and A. Sulakvelidze. 2003. Control of Listeria monocytogenes on fresh-cut produce by treatment with lytic bacteriophage and a bacteriocin. Appl. Environ. Microbiol. 69:4519–4526.   96. Lins, L., P. Ducarme, E. Breukink, and R. Brasseur. 1999. Computational study of nisin interaction with model membrane. Biochim. Biophys. Acta 1420:111–120.   97. Luchansky, J. B., K. A. Glass, K. D. Harrsono, A. J. Degnan, N. G. Faith, B. Cauvin, G. Bascus-Taylor, K. Arihara, B. Bater, A. J. Maurer, and R. G. Cassers. 1992. Genomic analysis of Pediococcus starter cultures used to control Listeria monocytogenes in turkey summer sausage. Appl. Environ. Microbiol. 58:3053–3059.   98. Lyon, W. J., J. E. Sethi, and B. A. Glatz. 1993. Inhibition of psychrotrophic organisms by propionicin PLG-1, a bacteriocin produced by Propionibacterium thoenii. J. Dairy Sci. 76:1506–1513.   99. Maher, S., and S. McClean. 2006. Investigation of the cytotoxicity of eukaryotic and prokaryotic antimicrobial peptides in intestinal epithelial cells in vitro. Biochem. Pharmacol. 71:1289–1298. 100. Martínez, J. M., M. I. Martínez, C. Herranz, A. Suárez, M. F. Fernández, L. M. Cintas, J. M. Rodríguez, and P. E. Hernández. 1999. Antibodies to a synthetic 1–9-Nterminal amino acid fragment of mature pediocin PA-1: sensitivity and specificity for pediocin PA-1 and crossreactivity against Class IIa bacteriocins. Microbiology 145:2777–2787. 101. Martínez, B., M. Fernández, J. E. Suárez, and A. Rodríguez. 1999. Synthesis of lactococcin 972, a bacteriocin produced

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

820

102.

103.

104. 105.

106.

107.

108.

109.

110.

111. 112. 113.

114.

115.

by Lactococcus lactis IPLA 972, depends on the expression of a plasmid-encoded bicistronic operon. Microbiology 145:3155–3161. Marugg, J. D., C. F. Gonzalez, B. S. Kunka, A. M. Ledeboer, M. J. Pucci, M. Y. Toonen, S. A. Walker, L. C. M. Zoetmulder, and P. A. Vandenbergh. 1992. Cloning, expression, and nucleotide sequence of genes involved in production of pediocin PA-1, a bacteriocin from Pediococcus acidilactici PAC 1.0. Appl. Environ. Microbiol. 58:2360–2367. Mayr-Harting, A., A. J. Hedges, and R. C. W. Berkeley. 1972. Methods for studying bacteriocins, p. 313–342. In J. R. Norris and D. W. Ribbons (ed.), Methods in Microbiology, vol. 7A. Academic Press, New York, NY. Mazzotta, A. S., A. D. Crandall, and T. J. Montville. 1997. Nisin resistance in Clostridium botulinum spores and vegetative cells. Appl. Environ. Microbiol. 63:2654–2659. Mazzotta, A. S., and T. J. Montville. 1999. Characterization of fatty acid composition, germination, and thermal resistance in a nisin resistant mutant of Clostridium botulinum 169B, and the wild-type strain. Appl. Environ. Microbiol. 65:659–664. Miller, K. W., P. Ray, T. Steinmetz, T. Hanekamp, and B. Ray. 2005. Gene organization and sequences of pediocin AcH/PA-1 production operons in Pediococcus and Lactobacillus plasmids. Lett. Appl. Microbiol. 40:56–62. Ming, X., and M. A. Daeschel. 1993. Nisin resistance of foodborne bacteria and the specific resistance responses of Listeria monocytogenes Scott A. J. Food Prot. 11:944–948. Modi, K. D., M. L. Chikindas, and T. J. Montville. 2000. Sensitivity of nisin-resistant Listeria monocytogenes to heat and the synergistic action of heat and nisin. Lett. Appl. Microbiol. 30:249–253. Moll, G. N., E. van den Akker, H. H. Hauge, J. NissenMeyer, I. F. Nes, W. N. Konings, and A. J. Driessen. 1999. Complementary and overlapping selectivity of the two-peptide bacteriocins plantaricin EF and JK. J. Bacteriol. 181:4848–4852. Montville, T. J., and M. E. C. Bruno. 1994. Evidence that dissipation of proton motive force is a common mechanism of action for bacteriocins and other antimicrobial proteins. Int. J. Food Microbiol. 24:53–74. Montville, T. J., A. M. Rogers, and A. Okereke. 1992. Differential sensitivity of Clostridium botulinum strains to nisin. J. Food Prot. 56:444–448. Montville, T. J., K. Winkowski, and R. D. Ludescher. 1995. Models and mechanisms for bacteriocin action and application. Int. Dairy J. 5:797–815. Mørtvedt, C. I., J. Nissen-Meyer, K. Sletten, and I. F. Nes. 1991. Purification and amino acid sequence of lactocin S, a bacteriocin produced by Lactobacillus sake L45. Appl. Environ. Microbiol. 57:1829–1834. Mulders, J. W., I. J. Boerrigter, H. S. Rollema, R. J. Siezen, and W. M. de Vos. 1991. Identification and characterization of the lantibiotic nisin Z, a natural nisin variant. Eur. J. Biochem. 201:581–584. Murdock, C. A., J. Cleveland, K. R. Matthews, and M. L. Chikindas. 2007. The synergistic effect of nisin and lactoferrin on the inhibition of Listeria monocy-

SMP_Food Microbiology_CH31.indd

820

116.

117. 118. 119.

120.

121.

122.

123.

124. 125.

126. 127.

128.

129.

130.

131.

togenes and Escherichia coli O157:H7. Lett. Appl. Microbiol. 44:255–261. Naghmouchi, K., E. Kheadr, C. Lacroix, and I. Fliss. 2007. Class I/Class IIa bacteriocin cross-resistance phenomenon in Listeria monocytogenes. Food Microbiol. 24:718–727. Nes, I. F., and H. Holo. 2000. Class II antimicrobial peptides from lactic acid bacteria. Biopolymers 55:50–61. Nettles, C. G., and S. F. Barefoot. 1993. Biochemical and genetic characteristics of bacteriocins of food-associated lactic acid bacteria. J. Food Prot. 56:338–356. Nielsen, J. W., J. S. Dickson, and J. D. Crouse. 1990. Use of a bacteriocin produced by Pediococcus acidilactici to inhibit Listeria monocytogenes associated with fresh meat. Appl. Environ. Microbiol. 56:2142–2145. Nilsson, L., Y. Chen, M. L. Chikindas, H. H. Huss, L. Gram, and T. J. Montville. 2000. Carbon dioxide and nisin act synergistically on Listeria monocytogenes. Appl. Environ. Microbiol. 66:769–774. Nissen-Meyer, J., H. Holo, L. S. Håvarstein, K. Sletten, and I. F. Nes. 1992. A novel lactococcal bacteriocin whose activity depends on the complementary action of two peptides. J. Bacteriol. 174:5686–5692. Nussbaum, A., and A. Cohen. 1988. Use of a bioluminescence gene reporter for the investigation of Reddependent and Gam-dependent plasmid recombination in E. coli K12. J. Mol. Biol. 203:391–402. O’Flyn, G., R. P. Ross, G. F. Fitzgerald, and A. Coffey. 2004. Evaluation of a cocktail of three bacteriophage for control of Escherichia coli O157:H7. Appl. Environ. Microbiol. 70:3417–3424. Ojcius, D. M., and J. D. E. Young. 1991. Cytolytic pore-forming proteins and peptides: is there a common structural motif? Trends Biochem. Sci. 16:225–229. Oscroft, C. A., J. G. Banks, and S. McPhee. 1990. Inhibition of thermally-stressed Bacillus spores by combinations of nisin, pH and organic acids. Lebenon. Wiss. Technol. 23:538–544. Parfitt, T. 2005. Georgia: an unlikely stronghold for bacteriophage therapy. Lancet 365:2166–2167. Parret, A. H. A., G. Schoofs, P. Proost, and R. De Mot. 2003. Plant lectin-like bacteriocin from a rhizosphere-colonizing Pseudomonas isolate. J. Bacteriol. 185:897–908. Patel, J. R., G. C. Sanglay, and M. B. Solomon. 2009. Control of Listeria monocytogenes on frankfurters with antimicrobials and hydrodynamic pressure processing. J. Muscle Foods 20:227–241. Pavan, S., P. Hols, J. Delcour, M. C. Geoffroy, C. Grangette, M. Kleerebezem, and A. Mercenier. 2000. Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: a tool to study in vivo biological effects. Appl. Environ. Microbiol. 66:4427–4432. Piard, J. C., O. P. Kuipers, H. S. Rollema, M. J. Desmazeaud, and W. M. de Vos. 1993. Structure, organization, and expression of the lct gene for lacticin 481, a novel lantibiotic produced by Lactococcus lactis. J. Biol. Chem. 268:16361–16368. Piper, C., P. D. Cotter, R. P. Ros, and C. Hill. 2009. Discovery of medically significant lantibiotics. Curr. Drug Discov. Technol. 6:1–18.

Manila Typesetting Company

11/08/2012 07:13AM

31.  Biological Control of Foodborne Bacteria 132. Radler, F. 1990. Possible use of nisin in winemaking. I: Action of nisin against lactic acid bacteria and wine yeasts in solid and liquid media. Am. J. Enol. Viticult. 41:1–6. 133. Radler, F. 1990. Possible use of nisin in winemaking. II: Experiments to control lactic acid bacteria in the production of wine. Am. J. Enol. Viticult. 41:7–11. 134. Rauch, P. J. G., and W. M. De Vos. 1992. Characterization of the novel nisin-sucrose conjugative transposon TN5276 and its insertion in Lactococcus lactis. J. Bacteriol. 174:1280–1287. 135. Ray, B., and M. A. Daeschel. 1992. Food Biopreservation of Microbial origin. CRC Press, Boca Raton, FL. 136. Ray, B., R. Schamber, and K. W. Miller. 1999. The pediocin AcH precursor is biologically active. Appl. Environ. Microbiol. 65:2281–2286. 137. Rekhif, N., A. Atrih, and G. Lefebvre. 1995. Selection and properties of spontaneous mutants of Listeria monocytogenes ATCC 15313 resistant to different bacteriocins produced by lactic acid bacteria strains. Curr. Microbiol. 230:827–853. 138. Roberts, R. E., and E. A. Zottola. 1993. Shelf-life of pasteurized process cheese spreads made from cheddar cheese manufactured with a nisin producing starter culture. J. Dairy Sci. 76:1830–1836. 139. Rodríguez, J. M., L. M. Cintas, P. Casaus, N. Horn, H. M. Dodd, P. E. Hernández, and M. J. Gasson. 1995. Isolation of nisin-producing Lactococcus lactis strains from dry fermented sausages. J. Appl. Bacteriol. 78:109–115. 140. Rodríguez, J. M., M. I. Martínez, and J. Kok. 2002. Pediocin PA-1, a wide-spectrum bacteriocin from lactic acid bacteria. Crit. Rev. Food Sci. Nutr. 42:91–121. 141. Rogers, A. M., and T. J. Montville. 1994. Quantification of factors influencing nisin’s inhibition of Clostridium botulinum 56A in a model food system. J. Food Sci. 59:663–668, 686. 142. Reference deleted. 143. Sablon, E., B. Contreras, and E. Vandamme. 2000. Antimicrobial peptides of lactic acid bacteria: mode of action, genetics and biosynthesis. Adv. Biochem. Eng. Biotechnol. 68:21–60. 144. Saleh, M. A., and Z. J. Ordal. 1955. Studies on growth and toxin production of Clostridium botulinum in precooked frozen food. II: Inhibition by lactic acid bacteria. Food Res. 20:340–346. 145. Sani, A. M., M. R. Ehsani, and M. M. Asadi. 2005. Effect of Propionibacterium shermanii metabolites on sensory properties and shelf life of UF-Feta cheese. Nutr. Food Sci. 35:88–94. 146. Scott, V. N., and S. L. Taylor. 1981. Effect of nisin on outgrowth of Clostridium botulinum spores. J. Food Sci. 46:117–120. 147. Scott, V. N., and S. L. Taylor. 1981. Temperature, pH, and spore load on the ability of nisin to prevent the outgrowth of Clostridium botulinum spores. J. Food Sci. 46:121–126. 148. Sears, P. M., B. S. Smith, W. K. Stewart, R. Gonzalez, S. O. Rubino, S. A. Gusik, E. S. Kulisek, S. J. Projan, and P. Blackburn. 1992. Evaluation of a nisin-based germicidal formulation on teat skin of live cows. J. Dairy Sci. 75:3185–3190.

SMP_Food Microbiology_CH31.indd

821

821 149. Severina, E., A. Severin, and A. Tomasz. 1998. Antibacterial efficacy of nisin against multidrug-resistant Gram-positive pathogens. J. Antimicrob. Chemother. 41:341–347. 150. Siegers, K., and K. D. Entian. 1995. Genes involved in immunity to the lantibiotic nisin produced by Lactococcus lactis 6F3. Appl. Environ. Microbiol. 61:1082–1089. 151. Somers, E. B., and S. L. Taylor. 1987. Antibotulinal effectiveness of nisin in pasteurized process cheese spreads. J. Food Prot. 50:842–848. 152. Stevens, K. A., B. W. Sheldon, N. A. Klapes, and T. R. Klaenhammer. 1992. Effect of treatment conditions on nisin inactivation of Gram-negative bacteria. J. Food Prot. 55:763–767. 153. Stoddard, G. W., J. P. Petzel, M. J. van Belkum, J. Kok, and L. L. McKay. 1992. Molecular analyses of the lactococcin A gene cluster from Lactococcus lactis subsp. lactis biovar diacetylactis WM4. Appl. Envrion. Microbiol. 58:1952–1961. 154. Stoffels, G., J. Nissen-Meyer, A. Gudmundsdottir, K. Sletten, H. Holo, and I. F. Nes. 1992. Purification and characterization of a new bacteriocin isolated from a Carnobacterium sp. Appl. Environ. Microbiol. 58:1417–1422. 155. Stone, R. 2002. Bacteriophage therapy: Stalin’s forgotten cure. Science 298:728–731. 156. Svetoch, E. A., B. V. Eruslanov, Y. N. Kovalev, E. V. Mitsevich, I. P. Mitsevich, V. P. Levchuk, N. K. Fursova, V. V. Perelygin, Y. G. Stepanshin, M. G. Teymurasov, B. S. Seal, and N. J. Stern. 2009. Antimicrobial activities of bacteriocins E 50–52 and B 602 against antibioticresistant strains involved in nosocomial infections. Probiotics Antimicrob. Prot. 1:136–142. 157. Tagg, J. R., A. S. Dajani, and L. W. Wannamaker. 1976. Bacteriocins of Gram-positive bacteria. Bacteriol. Rev. 40:722–756. 158. Tanaka, N. E., E. Traisman, M. H. Lee, and R. Cassens. 1980. Inhibition of botulism toxin formation in bacon by acid development. J. Food Prot. 43:450–452. 159. Taylor, L. Y., O. O. Cann, and B. J. Welch. 1990. Antibotulinal properties of nisin in fresh fish packaged in an atmosphere of carbon dioxide. J. Food Prot. 53:953–957. 160. Thompson, L. V., and J. Delves-Broughton. 2005. Nisin, p. 239–275. In P. M. Davidson, J. N. Sofos, and A. L. Branen (ed.), Antimicrobials in Foods, 3rd ed. CRC Press, Boca Raton, FL. 161. Tramer, J., and G. G. Fowler. 1964. Estimation of nisin in foods. J. Sci. Food Agric. 15:522–528. 162. Turovskiy, Y., D. Kashtanov, B. Paskhover, and M. L. Chikindas. 2007. Quorum sensing: fact, fiction and everything in between. Adv. Appl. Microbiol. 62:191–234. 163. Vadyvaloo, V., S. Arous, A. Gravesen, Y. Héchard, R. Chauhan-Haubrock, J. W. Hasting, and M. Rautenbach. 2004. Cell-surface alterations in class IIa bacteriocinresistant Listeria monocytogenes strains. Microbiology 150:3025–3033. 164. van Belkum, M. J., B. J. Hayema, R. E. Jeeninga, J. Kok, and G. Venema. 1991. Organization and nucleotide sequence of two lactococcal bacteriocin operons.

Manila Typesetting Company

11/08/2012 07:13AM

Preservatives and Preservation Methods

822

165.

166.

167. 168.

169.

170.

Cloning of two bacteriocin genes from a lactococcal bacteriocin plasmid. Appl. Envirion. Microbiol. 57:492–498. van Belkum, M. J., J. Kok, G. Venema, H. Holo, I. F. Nes, W. N. Konings, and T. Abee. 1991. The bacteriocin lactococcin A specifically increases the permeability of lactococcal cytoplasmic membranes in a voltageindependent, protein-mediated manner. J. Bacteriol. 173:7934–7941. van Belkum, M. J., R. W. Worobo, and M. E. Stiles. 1997. Double-glycine-type leader peptides direct secretion of bacteriocins by ABC transporters: colicin V secretion in Lactococcus lactis. Mol. Microbiol. 23:1293–1301. Vandenbergh, P. A. 1993. Lactic acid bacteria, their metabolic products and interference with microbial growth. FEMS Microbiol. Rev. 12:221–238. Vandenbergh, P. A., M. J. Pucci, B. S. Kunka, and E. B. Vedamuthu. 1989. Method for inhibiting Listeria monocytogenes using a bacteriocin. European patent application 89101126.6. van der Meer, J. R., J. Polman, M. M. Beerthuyzen, R. J. Siezen, O. P. Kuipers, and W. M. de Vos. 1993. Characterization of the Lactococcus lactis nisin A operon genes nisP, encoding a subtilisin-like serine protease involved in precursor processing, and nisR, encoding a regulatory protein involved in nisin biosynthesis. J. Bacteriol. 175:2578–2588. Van Reenen, C. A., M. L. Chikindas, W. H. Van Zyl, and L. M. Dicks. 2003. Characterization and heterologous expression of a class IIa bacteriocin, plantaricin 423 from Lactobacillus plantarum 423, in Saccharomyces cerevisiae. Int. J. Food Microbiol. 81:29–40.

SMP_Food Microbiology_CH31.indd

822

171. Venema, K., T. Abee, A. J. Haandrikman, K. J. Leenhouts, J. Kok, W. N. Konings, and G. Venema. 1993. Mode of action of lactococcin B, a thiol-activated bacteriocin from Lactococcus lactis. Appl. Environ. Microbiol. 59:1041–1048. 172. Vogel, R. F., B. S. Pohle, P. S. Tichaczek, and W. Hammes. 1993. The competitive advantage of Lactobacillus curvatus LTH 1174 in sausage fermentations is caused by formation of curvacin. Syst. Appl. Microbiol. 16:457–462. 173. Wagenaar, J., M. Van Bergen, M. Mueller, T. Wassenaar, and R. Carlton. 2005. Phage therapy reduces Campylobacter jejuni colonization in broilers. Vet. Microbiol. 109:275–283. 174. Wahlstrom, G., and P. E. Saris. 1999. A nisin bioassay based on bioluminescence. Appl. Environ. Microbiol. 65:3742–3745. 175. Winkowski, K., M. E. C. Bruno, and T. J. Montville. 1994. Correlation of bioenergetic parameters with cell death in Listeria monocytogenes cells exposed to nisin. Appl. Environ. Microbiol. 60:4186–4187. 176. Winkowski, K., A. D. Crandall, and T. J. Montville. 1993. Inhibition of Listeria monocytogenes by Lactobacillus bavaricus MN in meat systems at refrigeration temperatures. Appl. Environ. Microbiol. 59:2552–2557. 177. Winkowski, K., and T. J. Montville. 1992. Use of a meat isolate, Lactobacillus bavaricus MN, to inhibit Listeria monocytogenes growth in a model meat gravy system. J. Food Safety 13:19–31. 178. Worobo, R. W., M. J. van Belkum, H. Sailer, K. L. Roy, J. C. Vederas, and M. E. Stiles. 1995. A signal peptide secretion-dependent bacteriocin from Carnobacterium divergens. J. Bacteriol. 177:3143–3149.

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

VI SMP_Food Microbiology_CH32.indd

823

Manila Typesetting Company

11/08/2012 07:13AM

SMP_Food Microbiology_CH32.indd

824

Manila Typesetting Company

11/08/2012 07:13AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch32

32

Mark E. Johnson James L. Steele

Fermented Dairy Products

The fermented dairy products category contains products with a diversity of flavors, textures, and appearances, all of which are directly dependent on microbial metabolism. The enzymes and metabolites required to produce these products are provided by a diverse set of microorganisms, including molds, yeasts, and bacteria. Of these organisms, homofermentative lactic acid bacteria (LAB) are of the greatest importance, as the manufacture of fermented dairy products is directly dependent on their primary metabolic end product, lactic acid. These organisms are typically the dominant component of the microbiota of these products and are referred to as starter cultures. Yeasts, molds, and several bacterial species, including heterofermentative LAB, may be added to specific products; however, their purpose is not for acid development but for the production of flavor compounds or carbon dioxide. The microorganisms utilized in the manufacture of a ­variety of cheeses and fermented milks common in North America are presented in Table 32.1. It is also common today to use traditional starter bacteria such as strains of Lactobacillus (Lb.) helveticus, not as acid producers as they are in Swiss, Parmesan, and mozzarella cheeses but as flavor enhancers (adjuncts) in

cheeses such as Cheddar cheese and lower-fat versions of several cheese varieties. While the main function of starter LAB is to ferment lactose to lactic acid, it is not their only function. They are also involved in the development of flavor of fermented dairy products, but the relative importance of the starter culture and other added microorganisms varies from product to product. Starter bacteria include both mesophilic (optimal growth at 25 to 30°C) and thermophilic (optimal growth at 37 at 42°C) species. Mesophilic LAB include Lactococcus (Lc.) lactis subsp. lactis and Lc. lactis subsp. cremoris. Thermophilic LAB include Streptococcus thermophilus, Lactobacillus del­ brueckii subsp. bulgaricus, and Lb. helveticus. A protocooperative relationship exists between S. ­thermophilus and starter lactobacilli that results in increased rate and extent of acid production (73). While this may be ­desirable for manufacturers of mozzarella cheese, it may be detrimental in the manufacture of Swiss cheese, for which a slower rate of acid development is desired to prevent the formation of slits. Lb. helveticus strains are being replaced with less-stimulatory Lactobacillus delbrueckii subsp. lactis or Lactobacillus casei strains. Traditionally, the thermophiles were paired (cocci and

Mark E. Johnson, Center for Dairy Research, Department of Food Science, University of Wisconsin–Madison, Madison, WI 53706-1565. James L. Steele, Department of Food Science, University of Wisconsin–Madison, Madison, WI 53706-1565.

825

SMP_Food Microbiology_CH32.indd

825

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

826

Table 32.1  Microorganisms involved in the manufacture of cheeses and fermented milks Product Cheeses Colby, Cheddar, cottage, cream Gouda, Edam, Havarti Brick, Limburger

Lactococcus lactis subsp. cremoris/lactis Lc. lactis subsp. cremoris/lactis Lc. lactis subsp. cremoris/lactis

Camembert Blue

Lc. lactis subsp. cremoris/lactis Lc. lactis subsp. cremoris/lactis

Mozzarella, provolone, Romano, Parmesan Swiss Fermented milks Yogurt Buttermilk Sour cream

Intentionally introduced secondary cultures

Principal acid producer

Streptococcus thermophilus, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus helveticus S. thermophilus, Lb. helveticus, Lb. ­delbrueckii subsp. bulgaricus S. thermophilus, Lb. delbrueckii subsp. bulgaricus Lc. lactis subsp. cremoris/lactis Lc. lactis subsp. cremoris/lactis

rod) to get maximum rates of acid development. Today, single strains of S. thermophilus are used alone in the manufacture of mozzarella cheese. S. thermophilus is also being paired with mesophiles as starters in Cheddar cheese to increase the rate of acid development during the cooking (38°C) step. Mesophiles are also being paired with thermophiles or are used in cheeses in which thermophiles were the traditional starters. In addition to acid development, starters or adjuncts play a role in the hydrolysis of caseins and peptides, which can stimulate the growth of other bacteria. Lactobacilli used in Swiss cheese enhance growth and gas development of propionibacteria (79), and they also can increase the level of succinic acid and flavor development in Swiss and other cheese varieties. The contribution of succinic acid to cheese flavor is, however, not well characterized. In some fermented dairy products, additional bacteria, often referred to as secondary microflora (but essen­tial to flavor development), are added to influence ­flavor and alter texture of the final product. Two LAB, Leuconostoc species and strains of Lc. lactis subsp. lactis capable of metabolizing citric acid (Cit+), are added to produce aroma compounds and carbon dioxide in cultured buttermilk and certain cheeses (Gouda, Edam, blue, and Havarti). When used together, these bacteria make up 10 to 20% of the total starter culture, with Leuconostoc species being present at about three times the population of Cit+ Lc. lactis subsp. lac­ tis. Heterofermentative ­ lactobacilli (Lactobacillus bre­

SMP_Food Microbiology_CH32.indd

826

None Leuconostoc sp., Cit+ Lc. lactis subsp. lactis Geotrichum candidum, Brevibacterium linens, Micrococcus sp. Penicillium camemberti, sometimes B. linens Cit+ Lc. lactis subsp. lactis, Penicillium roqueforti None; animal lipases added to Romano for picante or rancid flavor Propionibacterium freudenreichii subsp. shermanii None Leuconostoc sp., Cit+ Lc. lactis subsp. lactis None

vis, Lactobacillus fermentum, and Lactobacillus kefir) are part of the varied microflora (including several yeast species) found in the more exotic cultured milks such as kefir and koumiss, in which they produce ethanol, carbon dioxide, and lactic acid. They are not used in other fermented dairy products because of the copious quantities of carbon dioxide produced. Propionibacterium freudenreichii subsp. shermanii is added to Swiss-type cheeses, in which it metabolizes l-lactic acid to propionic acid, acetic acid, and carbon dioxide. While these acids are not responsible for the distinctive flavor of Swiss cheese, other metabolic activities of the propionibacteria most certainly contribute to the development of the desired flavor components. The carbon dioxide forms the “eyes” in Swiss-type cheeses. Propionibacteria also ferment citric acid to glutamic acid (21). Other types of secondary microflora include undefined mixtures of yeasts, molds, and bacteria. These microorganisms can be added directly to the milk or are smeared, sprayed, or rubbed onto the cheese surface. This group of microorganisms has extremely varied and complex metabolic activities, their main function being to produce unique flavors. The use of these secondary cultures is usually limited to surface-ripened and mold-ripened cheeses. Yeasts (Debaryomyces, Candida, Yersinia, and Geotrichum species) and bacteria (Brevibacterium linens, Arthrobacter species, and Micrococcus species) are ­employed in the aging of surface-ripened cheeses such as Limburger, Danbo, and Gruyère. Recently, they have

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

827

been used in the production of specialty cheeses (including Cheddar), especially by farmstead cheese makers. Molds (Penicillum camemberti and Penicillium roque­ forti) are used in Camembert and blue-veined cheeses, respectively. Fermented dairy products are not commercially produced in an environment free of contaminating ­microorganisms; rather, contamination is inevitable and ­depending on the organism can have either a positive or negative impact on cheese quality. Typically, the contaminants are LAB and are referred to as nonstarter LAB. Rapid acid development resulting from starter LAB production of lactic acid lowers the pH to <5.3 in cheese and to <4.6 in fermented milk products within 4 to 8 h. After fermentation is complete, only acid-­tolerant bacteria can grow. However, if acid development is slow or if the pH does not decrease sufficiently, contaminants that otherwise would have been inhibited may be able to grow. In some cheeses the pH can increase during ripening and permit the growth of previously inhibited bacteria. Dairy products may contain yeasts, molds, and many different genera of bacteria whose metabolic ­activities destroy quality. Spoilage of dairy products is described in chapter 7.

IMPORTANCE OF THE STARTER CULTURE The key to commercial development of fermented milk products is the consistent and predictable rate of acid development by LAB. The rate and extent of pH ­decrease during manufacture and in the finished product are critical. pH has profound effects on moisture control during cheese manufacture, retention of coagulants, loss of minerals, hydration of proteins, and electrochemical interactions between protein molecules. These, in turn, have consequences for the development of flavor and physical properties (body and texture) of cheese and fermented milks. The reader is referred to other texts (30, 42) for discussions of these complex and interrelated phenomena. In the past, antibiotic residues and overmature starters have been causes of inconsistent acid production. However, these problems have been minimized through monitoring of the milk supply and the use of improved starter media and preparation (86). Today, bacteriophage infection is the most common cause of incon­sistent acid development, causing significant loss of revenue to the cheese industry. In cultured milks, desired flavors are derived directly from the metabolism of starter cultures and deliberately added aroma-producing secondary microflora. Thus, the desired flavor of the product dictates the choice of

SMP_Food Microbiology_CH32.indd

827

microorganisms. With cultured milks and some cheeses such as mozzarella, cream cheese, and cottage cheese, the short time between processing and consumption (1 day to 4 weeks), coupled with refrigerated storage, precludes the development of flavors other than that produced by starter and secondary cultures. In other cheeses, the choice of microorganism(s) depends primarily on the manufacturing protocol such as the temperature to which the product will be subjected, desired rate and extent of acid development, and desired physical properties of the finished product. There is considerable debate as to the exact contribution of the starter culture to flavor development in cheese, especially in cheeses to which no secondary cultures are added. In these cheeses, nonstarter LAB, particularly lactobacilli, are the dominant adventitious microflora during ripening (67), and it is generally believed that they play a significant role in the ripening of these cheeses. The use of nonstarter LAB, especially lactobacilli, continues to be an active research area, and adjunct LAB are commercially available. Unfortunately, the selection of appropriate cultures is a trial-and-error process since agreement on specific compounds that contribute to desired cheese flavor is often lacking. Differences in descriptions of desired flavor arise from the sheer complexity of cheese flavor as well as individual flavor perceptions and taste preferences. It should be understood that flavor development in cheese is a dynamic process and occurs in an environment that is constantly changing. It is not just that cheese develops stronger flavor with age, but that the very nature of the flavor is evolving. The development of flavor in different cheese varieties has been described (31).

LACTOSE METABOLISM Energy transduction and fermentation pathways for carbohydrate metabolism in LAB have been described in detail (32, 68). Lactose, a disaccharide composed of glucose and galactose, is the primary carbohydrate present in milk (45 to 50 g/liter). Lactococci ­translocate lactose into the cell by a phosphoenolpyruvate phosphotransferase system. The lactose is phosphorylated during translocation and then cleaved by phospho-b-galactosidase into glucose and galactose6-phosphate (Fig. 32.1). The glucose moiety enters the glycolytic pathway, and galactose-6-phosphate is converted into tagatose-6-­phosphate via the tagatose pathway. Both sugars are cleaved by specific aldolases into triose phosphates, which are converted to pyruvic acid at the expense of NAD+. For continued energy production, NAD+ must be regenerated. This

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

828

Figure 32.1  Lactose metabolism in homofermentative LAB. doi:10.1128/9781555818463.ch32f1

is usually accomplished by reducing pyruvic acid to lactic acid. S. thermophilus and some thermophilic lactobacilli transport lactose via a lactose-galactose antiport system driven by an electrochemical proton gradient (68). Lactose is not phosphorylated but is cleaved by b-­galactosidase to yield glucose and galactose. The glucose moiety enters the glycolytic pathway, but galactose is ­ excreted from the cells and accumulates in milk or cheese. Thermophilic lactobacilli that do not excrete ­galactose and Lb. helveticus strains able to transport ­excreted ­galactose utilize the Leloir pathway to metabolize galactose. Lb. delbrueckii subsp. bulgaricus and most strains of S. thermophilus cannot metabolize ­galactose. This presents a problem in cheese manufacture since ­residual sugar can be metabolized heterofermentatively by other bacteria. Rapid production of carbon dioxide by heterofermentative bacteria causes cheese to crack and packages to swell. Residual sugar can also react with amino groups and form pink or brown pigments, i.e., Maillard-browning reaction products. Even thermophiles capable of utilizing galactose may not meta­bolize the residual carbohydrate during cold storage (4 to 7°C)

SMP_Food Microbiology_CH32.indd

828

of the cheese. Therefore, lactococci are sometimes included in the starter to ensure that all residual carbohydrate is fermented. In pasta filata cheese manufacture, the curd is heated (52 to 66°C) and molded. This heat treatment may inactivate starter bacteria and prevent further carbohydrate metabolism. It is not known how lactose is transported in cells by Leuconostoc species or heterofermentative lactobacilli; however, lactose is known to be hydrolyzed by b­galactosidase (38). The galactose moiety is transformed into glucose-6-phosphate (Leloir pathway) and, together with glucose, is metabolized through the phosphoketolase pathway (Fig. 32.2). Heterofermentative LAB lack aldolase, but through a dehydrogenation-decarboxylation system, a pentose sugar (xylulose-5-phosphate) and carbon dioxide are formed. Xylulose-5-phosphate is then cleaved by phosphoketolase to yield glyceraldehyde and acetylphosphate. Lactic acid and ethanol, respectively, are formed from these intermediates, faci­ litating the regeneration of NAD+. However, during cometabolism of lactose and citric acid, Leuconostoc species convert acetylphosphate into acetic acid and generate ATP.

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

829

Figure 32.2  Lactose metabolism in heterofermentative LAB. doi:10.1128/9781555818463.ch32f2

Two enzymes, l-lactic acid dehydrogenase and dlactic acid dehydrogenase, are primarily responsible for the conversion of pyruvate to l-lactic acid and d-lactic acid, respectively. Lactococci produce only l-lactic acid, while Lb. delbrueckii subsp. bulgaricus and Leuconostoc species form only d-lactic acid. Other lactobacilli possess both enzymes and produce both d- and l-lactic acid. The key to end-product formation in lactose metabolism is the regeneration of reducing equivalents. Lactococci have the enzymatic potential to produce compounds other than lactic acid to regenerate NAD+, but these activities are not typically expressed under aerobic conditions (29). Oxygen in milk is used as an electron acceptor by LAB through the activity of oxidases and peroxidases (18, 75). As a consequence, hydro­gen from NADH is transferred to oxygen to produce ­ hydrogen peroxide, and NAD+ is regenerated. However, under anaerobic conditions and low sugar levels, lactococci (7, 29, 80) produce formic acid and ethanol (Fig. 32.3) to regenerate NAD+ via pyruvate formate lyase activity. However, Jensen et al. (43) demonstrated that even slight aeration of an otherwise anaerobic fermentation decreases synthesis of pyruvate formate lyase. This ­results in a shift away from production of ethanol and formic acid to acetic acid and carbon dioxide. Heterofermentative LAB convert acetylphosphate into acetic acid rather than ethanol under aerobic conditions and regenerate NAD(P)+ through NAD(P)H oxidases (58, 84). Ethanol has a very high flavor threshold value and would not be expected to contribute directly to flavor

SMP_Food Microbiology_CH32.indd

829

in the amounts produced by homofermentative LAB. However, subsequent esterification of ethanol with short-chain fatty acids yields esters with very low flavor thresholds. These compounds are responsible for the fruity flavor defects of Cheddar cheese (8). Short-chain fatty acids are probably generated by nonstarter LAB or exogenous lipase sources, since starter bacteria have limited lipase activity (45). The potential for production of diacetyl and carbon dioxide from lactose metabolism in lactococci with ­reduced lactic acid dehydrogenase activity has been described (60). Formation of ethanol, acetic acid, and ­formic acid would regenerate NAD+ (Fig. 32.3). As a result of lactose metabolism (and oxidase ­activity), the environment becomes anaerobic and the ­oxidation-reduction potential is reduced. In cheese, further metabolism by nonstarter LAB may be needed to maintain this low potential. It has been postulated that a low oxidation-reduction potential is necessary for the production and stability of reduced sulfur-containing compounds thought to be vital for the development of certain cheese flavors (35, 59). Sugar and citric acid ­metabolism may result in the formation of a-­dicarbonyls such as glyoxal, methylglyoxal, and diacetyl. These com­ pounds readily react with amino acids and produce a myriad of compounds that contribute to cheese flavor (36). Although lactic acid is commonly thought of as the end of fermentation, this is not always the case. Cometabolism of citric acid and lactic acid by facultative lactobacilli produces carbon dioxide and causes blowing of packaged cheese (33). Propionibacteria metabolize lactic acid to acetic and propionic acids. Clostridia also metabolize lactic acid to acetic acid and carbon dioxide. In some cases, lactose is not fermented but is utilized in the production of exopolysaccharides. Under stress, e.g., low pH or low water activity, some lactococci strains do not ferment the galactose moiety of lactose but form exopolysaccharides containing methyl pentoses and galactose (57). S. thermophilus also converts lactose to an exopolysaccharide during stationary phase of growth or low temperatures (34).

PRODUCTION OF AROMA COMPOUNDS Although lactic acid is the main metabolic end product of lactose metabolism in cultured dairy products and is responsible for the acid taste, it is nonvolatile and odorless and does not contribute to aroma. The main volatile flavor components of fermented milks are acetic acid, acetaldehyde, and diacetyl. In yogurt, these volatile compounds are formed by the starter LAB S. thermophi­ lus and Lb. delbrueckii subsp. bulgaricus. Leuconostoc

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

830

Figure 32.3  Pyruvic acid and citric acid metabolism in LAB. CoA, coenzyme A; Tpp, thiamine pyrophosphate. doi:10.1128/9781555818463.ch32f3

species and Cit+ Lc. lactis subsp. lactis are added to produce aroma compounds in buttermilk and some cheese varieties. Literature describing the production of aroma compounds in fermented dairy products has been compiled by Imhof and Bosset (41).

Diacetyl Production

The production of diacetyl, acetic acid, and carbon dioxide from citric acid by LAB has been reviewed by Hugenholtz (39) and is shown in Fig. 32.3. The carbon dioxide produced is responsible for the holes (eyes) in Gouda and Edam cheeses and the effervescent quality of buttermilk. Milk contains 0.15 to 0.2% citric acid, but not all LAB can metabolize it. However, Leuconostoc species, Cit+ Lc. lactis subsp. lactis, and facultative heterofermentative lactobacilli (25, 54, 66) metabolize citric acid. Leuconostoc species and Cit+ Lc. lactis subsp. lactis strains utilize citric acid and lactose ­simultaneously and under certain conditions can derive energy via metabolism of citric acid. Citric acid is transported into the

SMP_Food Microbiology_CH32.indd

830

cell by a citric acid permease (which is plasmid encoded in lactococci and Leuconostoc; 47, 82) and is metabolized to pyruvic acid without generation of NADH. The result is an excess of pyruvic acid that does not have to be reduced to lactic acid to regenerate NAD+; therefore, it is available for other reactions. Citrate metabolism in Leuconostoc species and Lc. lactis generates an electrochemical proton motive force-generating process (4). Ramos et al. (69) appear to have resolved conflicting reports on the pathway leading to the formation of diacetyl. Using 13C nuclear magnetic resonance, they verified that diacetyl formation involves nonenzymatic decarboxylative oxidation of a-acetolactate (an unstable intermediate derived from two molecules of ­pyruvic acid) and that the alternative suggested pathway, via a diacetyl synthase, is highly unlikely. However, it has been suggested that not all the diacetyl produced by Cit+ Lc. lactis subsp. lactis strains can be explained by spontaneous decarboxylation of the a-acetolactate (3). a-Acetolactate can be produced only when pyruvic acid

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

831

accumulates within the cell. Leuconostoc species metabolize citric acid during growth but do not form ­diacetyl until the pH is below 5.4. a-Acetolactate synthase is inhibited at pH 5.4 or higher by many intermediates of lactose metabolism; however, the inhibition is relieved at lower pH values (17, 70). When diacetyl is not formed, Leuconostoc species form lactic acid from pyruvic acid derived from citric acid and regenerate NAD+. Since NAD+ is regenerated, there is less demand to form ethanol from the acetylphosphate that is generated via the phosphoketolase pathway. Consequently, acetic acid is formed with the generation of ATP (Fig. 32.3) and growth is enhanced (16, 72). Diacetyl can be reduced by 2,3-butanediol dehydrogenases (19) to acetoin and 2,3-butanediol, both flavorless compounds. The presence of citric acid inhibits these reactions, but reduction begins when citric acid is exhausted. To ensure residual levels of citric acid in cultured milks, the ratio of starter culture to Cit+ bacteria must be controlled (26). The cultured milk is stirred when the pH is reduced to 4.5 or below. This introduces oxygen and helps to increase and maintain the desired diacetyl content. The introduction of oxygen is required for nonenzymatic oxidative decarboxylation of a-­acetolactate to diacetyl. In addition, LAB produce NADH oxidase, which transfers hydrogen to oxygen and regenerates NAD+. The NADH oxidase activity replaces the role of the 2,3-butanediol dehydrogenases in regenerating NAD+ and allows the accumulation of diacetyl, rather than its reduction to acetoin and 2,3butanediol (5). NADH oxidase is more active at lower temperatures, while dehydrogenases are less active (6). For rapid acidification, non-citrate-metabolizing lactococci must be the dominant acid producers. If not, cometabolism of citric acid and lactose by Cit+ Lc. lactis subsp. lactis would quickly consume the citric acid and result in the reduction of diacetyl to acetoin and 2,3butanediol before pH 4.6 is reached. Hugenholtz (39) described the use of genetic engineering to construct strains of lactococci with elevated levels of diacetyl.

Acetaldehyde Production

There are several metabolic pathways in LAB that can lead to the formation of acetaldehyde (51, 52). This has resulted in some controversy over the primary pathway utilized by LAB. Cleavage of threonine by threonine aldolase to glycine and acetaldehyde has been suggested to be the most important mechanism for acetaldehyde production in yogurt and buttermilk (52, 89). However, using radiolabeled threonine, Wilkins et al. (87) demonstrated that only 2% of the acetaldehyde produced by mixed cultures of Lb. delbrueckii subsp. bulgaricus and

SMP_Food Microbiology_CH32.indd

831

S. thermophilus originated from threonine, even though both bacteria possess threonine aldolase. Bongers et al. (9) developed a genetically modified Lc. lactis strain (added pyruvate decarboxylase activity and overexpressed NADH oxidase activity) that produced a high level of acetaldehyde. Chaves et al. (14) demonstrated that the main pathway for acetaldehyde formation in S. thermophilus is catalyzed by serine hydroxymethyltransferase, which also has threonine aldolase activity. Acetaldehyde is also formed by Cit+ Lc. lactis subsp. lactis strains (46). When the ratio of diacetyl to acetaldehyde in fermented milks is lower than 3:1, a yogurt or green apple flavor defect is observed (55). The defect is due to excess metabolic activity by Cit+ Lc. lactis subsp. lactis. Excessive acetaldehyde in yogurt is the result of overripening and is always associated with high acid content. Prevention of an excessive amount of acetaldehyde may be accomplished by the use of Leuconostoc, which metabolizes acetaldehyde to ethanol. To prevent overripening, the product must be cooled rapidly and stored at lower temperatures. The trend for faster acid development and larger fermentation vessels may limit the ability of the manufacturer to cool the product fast enough to prevent overripening.

PROTEOLYTIC SYSTEMS IN LAB Proteolytic systems in LAB contribute to their ability to grow in milk and are necessary for the development of flavor in ripened cheeses. LAB are amino acid auxotrophs typically requiring several amino acids for growth. Well-characterized examples include strains of lactococci and lactobacilli that require 6 and 15 amino acids, respectively. The quantities of free amino acids present in milk are not sufficient to support the growth of these bacteria to high cell density; therefore, they require a proteolytic system capable of utilizing the peptides present in milk and hydrolyzing milk proteins (aS1-, aS2-, b-, and g-caseins) to obtain essential amino acids. Peptides and amino acids formed by proteolysis may impart flavor directly or serve as flavor precursors in fermented dairy products. Additionally, the resulting flavors may have either positive or negative impacts. The production of high-quality fermented dairy products is dependent on the proteolytic systems of LAB.

Proteolytic Systems and Their Physiological Role

The genetics and biochemistry of LAB proteolytic systems are well defined. This is especially true for the lactococcal system, in which many of the enzymes have been purified and characterized and numerous isogenic

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

832

Figure 32.4  Schematic representation of the lactococcal proteolytic system. PrtP, cell ­envelope-associated proteinase; Opp, oligopeptide transport system; Dtp, di-/tripeptide transport systems; AAT, amino acid transport systems; EP, endopeptidases; AP, aminopeptidases; TP, tripeptidases; DP, dipeptidases. doi:10.1128/9781555818463.ch32f4

strains lacking specific components of the proteolytic enzyme system have been constructed. For extensive reviews of the proteolytic enzyme systems of LAB, see Kunji et al. (50) and Christensen et al. (15). A model of the lactococcal proteolytic enzyme system is presented in Fig. 32.4. While growing in milk, LAB obtain essential amino acids in a variety of ways. They first utilize nonprotein nitrogen sources such as free amino acids and small peptides. Caseins, which compose 80% of all proteins present in milk, become the primary nitrogen source after nonprotein nitrogen is depleted. Proteolytic systems in LAB can be divided into three components, viz, enzymes outside the cytoplasmic membrane, transport systems, and intracellular enzymes. Extensive investigations have revealed that a cell envelope-associated proteinase, designated PrtP, is the only extracellular proteolytic enzyme present in lactococci. The enzyme is a serine-protease that is expressed as a preproproteinase. A signal peptidase removes the signal peptide upon transport across the cytoplasmic membrane. Subsequently, a lipoprotein maturase (PrtM) is thought to cause a conformational change in the proproteinase, resulting in release of the proregion via auto­ proteolysis, and an active PrtP. The activated enzyme remains associated with the cell due to the presence of a C-terminal membrane anchor sequence. Genes encoding lactococcal proteinases with different substrate specificities have been sequenced, and the amino acid residues involved in substrate binding and catalysis have been

SMP_Food Microbiology_CH32.indd

832

determined. A critical feature of the enzyme is its broad cleavage specificity, which results in the release of more than 100 oligopeptides from soluble b-casein, 20% of which are small enough to be transported by the oligopeptide transport system (44). Loss of PrtP, which is typically plasmid encoded in lactococcal strains, results in derivatives capable of reaching only about 10% of the final cell density of the parental strain, indicating that PrtP is essential for growth of lactococci to high cell density in milk. Transport of nitrogenous compounds across the lactococcal cytoplasmic membrane takes place via groupspecific amino acid transport systems, di- and tripeptide transport systems, and an oligopeptide transport system (Opp). Of these systems, Opp is of greatest importance during growth of lactococci in milk. Growth studies with Opp derivatives have shown that Opp is essential for the uptake of PrtP-generated peptides from b-casein and oligopeptides present in the nonprotein nitrogen component of milk (44). This system, which is organized in an operon, is composed of two ATP-binding proteins, two integral membrane proteins, and a ­ substrate-­binding protein. Peptides from 4 to 18 residues can be transported with little specificity for particular side chains (23). Once inside the cell, peptides are hydrolyzed by peptidases. Peptidase classes that have been identified in lactococci include exopeptidases and endopeptidases. The greatest variety of enzymes are from the exopep-

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

833

Table 32.2  Peptidases purified and characterized from

lactococci Peptidase

Abbreviation

X-prolyl dipeptidyl aminopeptidase

PepX

X-Pro↕Y- . . .

Aminopeptidase N

PepX

X↕Y-Z . . .

Aminopeptidase C

PepC

X↕Y-Z . . .

Aminopeptidase A

PepA

Asp(Glu)↕Y-Z . . .

Pyrrolidone carboxylyl peptidase

PCP

pGlu↕Y-Z . . .

Prolyl iminopeptidase

PepI

Pro↕Y-Z . . .

Dipeptidase

PepV

X↕Y

Prolidase

PepQ

X↕YPro

Tripeptidase

PepT

X↕Y-Z

Endopeptidases

PepO, PepF

. . . W-X↕Y-Z . . .

a

Specificitya

↕ indicates which peptide bond is hydrolyzed.

tidase class, which includes aminopeptidases, tripeptidases, and dipeptidases. No carboxypeptidases have been detected in lactococci. Peptidases identified in lactococci and their cleavage specificities are summarized in Table 32.2. This combination of endopeptidases, aminopeptidases, tripeptidases, and dipeptidases converts the transported peptides into free amino acids, which lactococci require for growth. Examination of the growth of strains lacking a single peptidase in milk has revealed that only the PepN single mutant grows significantly more slowly. Characterization of lactococcal derivatives lacking more than one peptidase has ­indicated that, in general, the more peptidases that are inactivated, the slower the strain grows in milk. The only possible ­ exception to this general rule is PepX, which has not been observed to have a significant effect on growth rate in milk (61). The most direct interpretation of these results is that the ­ multiple-peptidase mutants have a reduced ability to ­ obtain essential amino acids from casein-derived oligopeptides, thereby limiting the availability of amino acids for new protein biosynthesis. However, it is also possible that the reduced growth rate in milk observed with multiple-peptidase mutants is ­related to altered regulation of the proteolytic system or a reduced ability to turn over cellular proteins (15). Proteolytic systems of other LAB have not been as extensively studied. However, comparative genomics analysis of a broad range of LAB has demonstrated that, in general, all major components of the lactococcal proteolytic enzyme system have homologs in the LAB examined to date (56). However, Lb. helveticus differs from other LAB in that individual strains may have up

SMP_Food Microbiology_CH32.indd

833

to four distinct cell envelope proteinases (11), while the other characterized LAB have one or no cell envelopeassociated proteinases. These differences in complement of proteinases may explain the variability among Lb. helveticus strains in regard to their functionality in fermented dairy products.

Proteolysis and Cheese Flavor Development

While flavor development in various types of cheese ­remains a poorly defined process, it is generally agreed that proteolysis is essential for flavor development in bacterially ripened cheeses (31, 83). Proteolytic ­enzymes present in this group of products include chymosin, plasmin, and proteolytic enzymes from starter cultures, ­ adjunct cultures, and nonstarter LAB. The specificities and relative activities of the proteolytic enzymes present in the cheese matrix determine which peptides and amino acids accumulate and, hence, how flavor develops. Free amino acids and peptides in the cheese matrix can contribute to cheese flavor either directly or indirectly and with positive or negative effects. Cheese flavor development has been the subject of comprehensive reviews (12, 31, 76, 81). A major negative effect of proteolytic products is bitterness, which is believed to be caused by hydrophobic peptides ranging in length from 3 to 27 amino residues (53). These peptides are believed to be generated from casein principally by the joint action of chymosin and the LAB proteinases (11, 13) and can be hydrolyzed to nonbitter peptides and amino acids by LAB peptidases. Therefore, the accumulation of bitter peptides is dependent on the relative rates of their formation and hydrolysis. A variety of volatile compounds can be derived from catabolism of amino acids (2, 77, 85, 88). Numerous sulfur-containing compounds, particularly methanethiol, are thought to be important in cheese flavor. Methionine is believed to be the precursor to methanethiol; a number of enzymes and/or pathways have been identified in LAB capable of converting methionine to methanethiol (85). Alternatively, amino acid catabolism can give rise to compounds that have a negative impact on cheese flavor. For example, catabolism of aromatic amino acids can give rise to compounds such as indole and skatole, which contribute to “unclean” flavors in cheese (15). Overall, proteolysis is believed to be essential for development of characteristic flavor compounds in bacterially ripened cheeses; however, the specific pathways by which products of proteolysis give rise to beneficial flavor compounds remains unknown. Additionally, other than bitterness, the mechanisms by which proteolysis impacts on the development of undesirable flavor compounds remain poorly defined.

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

834

BACTERIOPHAGES AND BACTERIOPHAGE RESISTANCE Bacteriophage infection may lead to a decrease or complete inhibition of lactic acid production by the starter culture. This has a major impact on the manufacture of fermented dairy products, as lactic acid synthesis is required to produce these products. Additionally, slow acid production disrupts manufacturing schedules and typically results in products that are downgraded to lower economic value. The severe consequences of bacteriophage infection have led to extensive investigations into bacteriophages and the mechanisms by which LAB resist infection. Bacteriophage infection of LAB was first described in lactococcal starter cultures in the 1930s. Since then, the dairy industry has employed improved sanitation regimens, utilized sophisticated starter culture propagation vessels, developed starter culture systems to minimize the impact of phage infection, and isolated and constructed starter strains with enhanced bacteriophage resistance. However, bacteriophage infection of the starter culture has remained a significant problem in the dairy industry. Klaenhammer and Fitzgerald (49) listed four reasons why dairy product fermentations are particularly susceptible to bacteriophage infection: 1. Phage contamination can occur when fermentations are not protected from environmental contaminants and in a nonsterile fluid medium, pasteurized milk. 2. Processing efficiency is easily disrupted because batch culture fermentations occur under increasingly stringent manufacturing schedules. 3. Increasing reliance on specialized strains limits the number and diversity of available dairy starter cultures. 4. Continuous use of defined cultures provides an ever-present host for bacteriophage attack. Historically, most bacteriophage-related problems have occurred with lactococcal starter cultures; however, problems with starter systems that employ S. ther­ mophilus and Lb. delbrueckii subsp. bulgaricus are now also common. It is principally bacteriophage infection of S. thermophilus strains that has been the problem among these cultures (63). For more extensive reviews of bacteriophages and bacteriophage resistance in LAB, readers are referred to other publications (1, 22, 28, 62, 63, 78).

Bacteriophages of LAB

Significant progress has been made in the characterization of bacteriophages from LAB. All of the bacterio-

SMP_Food Microbiology_CH32.indd

834

phages examined contain double-stranded linear DNA genomes with either cohesive or circularly permuted terminally redundant ends. Both lytic and temperate bacteriophages have been characterized. A major outcome of this characterization has been a clear classification, based on DNA homology and morphological studies, of LAB bacteriophages. Complete nucleotide sequence information in now available for representative isolates of most industrially important LAB bacteriophage species. This information has enhanced our understanding of how bacteriophage and host interactions evolve over time and how to construct novel bacteriophage defense mechanisms.

Lactococcal Bacteriophage Resistance Mechanisms

Selective environmental pressure placed on lactococci by bacteriophages over thousands of years has resulted in strains that contain numerous bacteriophage defense mechanisms. The best-characterized bacteriophageresistant strain is Lc. lactis ME2. This strain has been shown to contain at least five distinct phage defense loci, including one that interferes with bacteriophage adsorption, two restriction/modification (R/M) systems, and two abortive bacteriophage infection (Abi) mechanisms (49). These defense loci are encoded by plasmids capable of conjugal transfer, which suggests that genetic exchange between starter cultures has had an important role in the development of bacteriophage-resistant starter cultures. Recombinant DNA techniques have also been employed to construct lactococcal strains with enhanced bacteriophage resistance. These include the use of antisense RNA derived from conserved bacteriophage genes and the cloning of bacteriophage origins of replication on multicopy plasmids. The latter approach is thought to titrate bacteriophage replication factors and result in a bacteriophage defense mechanism similar to Abi. The remainder of this section covers the previously mentioned three naturally occurring bacteriophage ­defense mechanisms.

Interference with Bacteriophage Adsorption

The adsorption of a bacteriophage to a host cell is ­determined by bacteriophage specificity, physicochemical properties of the cell envelope, accessibility and density of bacteriophage receptor material, and electrical potential across the cytoplasmic membrane (74). The complexity of this interaction has facilitated the isolation of bacteriophage-resistant starter cultures by exposing them to bacteriophages and isolating resistant variants. Lactococcal mutants isolated in such a fashion typically have a reduced capacity to adsorb the bacteriophage(s)

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

835

used in the challenge and are referred to as bacteriophage-insensitive mutants. This is a common practice for deriving starter cultures with reduced bacteriophage sensitivity. Researchers have begun to characterize host components required for bacteriophage adsorption. It is now thought that bacteriophages initially interact reversibly with cell envelope-associated polysaccharide and then interact irreversibly with cell membrane protein(s) (64). The best-characterized example of a lactococcal mechanism that interferes with adsorption of bacteriophages has been described in Lc. lactis subsp. cremoris SK110. Reduction in the ability of bacteriophages to adsorb to cells of this strain is due to masking of the phage receptors by a galactose-containing lipoteichoic acid, rather than the absence of receptors. Inhibition of adsorption of lactococcal bacteriophages due to the lack of a membrane-bound protein has also been reported. Characterization of phage-­resistant mutants of Lc. lactis subsp. lactis C2 has revealed that these mutants bind normally with bacteriophages but no plaques are formed. Subsequently, it was demonstrated that a 32-kDa membrane protein essential for bacteriophage infection was lacking in these mutants. Similarly, it has been suggested that a membrane protein is involved in bacteriophage adsorption to Lc. lactis subsp. lactis ML3. It remains to be determined if this protein also has a role in the injection of phage DNA into the cytoplasm (28).

R/M Systems

R/M systems are widely distributed in lactococci and are often plasmid encoded. In fact, it is not uncommon for strains to contain two or more R/M systems, and at least 23 plasmids that encode these systems have been identified. The two components of an R/M system are a site-specific modifying enzyme and a corresponding site-­specific restriction endonuclease. This system enables the cell to differentiate bacteriophage DNA from its own DNA and inactivate foreign DNA by hydrolysis. The typical end result of bacteriophage infection of a culture containing an R/M system is a reduction in the number of progeny bacteriophage produced. The extent of reduction is dependent on both the activity of the R/M system and the number of unmodified restriction endonuclease sites on the bacteriophage genome. It is important to note that bacteriophages that escape restriction give rise to modified progeny phage that are immune to the corresponding restriction endonuclease. Therefore, while this is an important and widely distributed mechanism of bacteriophage defense, it is also very fragile.

SMP_Food Microbiology_CH32.indd

835

Two mechanisms by which bacteriophages have evolved resistance to R/M systems have been identified. Characterization of lactococcal bacteriophages has revealed that they contain far fewer restriction endonuclease sites than expected for genomes of their size, suggesting that evolutionary pressure has selected for bacteriophages with few restriction endonuclease sites. This view is supported by the observation that lactococcal bacteriophages that have recently emerged in the dairy industry are more sensitive to R/M systems. The characterization of bacteriophages that have evolved to overcome a specific R/M system has revealed that they have acquired a functional copy of the modification enzyme from that system. These examples illustrate that bacteriophage-host interactions are continually ­changing, with bacterial strains acquiring new defense mechanisms and bacteriophages evolving mechanisms to overcome them.

Abi Systems

Like R/M systems, Abi systems are widely distributed in lactococci and are frequently plasmid encoded, ­although some are also encoded by episomes. By definition, these mechanisms inhibit bacteriophage infection following adsorption, DNA penetration, and the early stages of the bacteriophage lytic cycle. Few infections successfully release viable progeny, and those that are successful ­result in fewer progeny bacteriophages being released. The end result for the host, even one that does not ­release ­viable bacteriophages, is death. All of the characterized Abi systems have an unusually low G + C content, suggesting that they have recently been acquired by horizontal gene transfer. Frequently, Abi genes are associated with R/M systems. These systems work well together; the R/M system reduces the number of viable bacteriophage genomes, and the Abi system reduces the number of progeny bacteriophages released from infected cells that evade the R/M system.

CRISPRs

Clustered regular interspaced short palindromic repeats (CRISPRs) are a family of DNA repeats that provide immunity against foreign genetic elements. They are composed of short (21- to 48-bp) direct DNA repeats ­interspersed with nonrepetitive spacers of similar length. The nucleotide sequences of the spacers are diverse but typically are identical to either bacteriophages or plasmids that the organism containing the CRISPR is likely to have encountered. CRISPR loci have been detected in S. thermophilus and a number of species of lactobacilli (37). The relationship between spacer sequences and bacteriophage resistance has been well established

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

836 in S. thermophilus (24). Bacteriophage infection of a bacterium containing an active CRISPR results in the addition of a new spacer that is identical to a sequence present on the bacteriophage. The presence of this new spacer results in resistance to infection by the bacteriophage from which it was derived. The mechanism ­responsible for bacteriophage resistance is thought to be an RNA interference system. The application of CRISPR technology holds great promise for the development of bacteriophage-resistant derivatives of S. thermophilus and lactobacilli that contain CRISPRs.

GENETICS OF LAB Research on the genetics of LAB began in the early 1970s. Initially, research focused on plasmids and natural gene transfer systems in lactococci. Interest in the genetics of LAB has expanded rapidly, and there are now hundreds of researchers active in this area. This interest has resulted in the development of a relatively detailed understanding of the basic genetics of these bacteria, natural gene transfer systems, genome evolution, and genetic diversity within a species and the development of tools required for application of recombinant DNA techniques. For a more complete description of the genetics of LAB, the reader is ­referred to other publications (10, 48, 65) and chapter 38.

General Genetics of LAB

Genetic elements of LAB that have been characterized include chromosomes, introns, transposable elements, and plasmids. Chromosomes of LAB are relatively small compared with those of other eubacteria, ranging from 1.8 to 3.4 Mbp (20). As of the time of writing this chapter, there were more than 250 genome sequences available for LAB, including multiple genome sequences for the organisms that typically function as starter cultures for the production of fermented dairy products (www. genomesonline.org). The value of genomic sequences of LAB to basic and applied research on these organisms cannot be overstated. For example, this information allows for the development of a comprehensive view of the metabolic potential of these organisms. Additionally, comparative genomic approaches allow for the investigation of strain-specific phenotypes. Group I and group II introns have been identified in LAB. These elements are ribozymes that catalyze a self-splicing reaction from mRNAs that contain the intron. To date, group I ­introns in LAB have been identified only in bacteriophage genomes. The only group II intron identified in LAB is associated with the plasmid-encoded gene required for conjugal transfer (27). Numerous transposable ele-

SMP_Food Microbiology_CH32.indd

836

ments, genetic elements capable of moving as discrete units from one site to another in the genome, have been described in LAB (10, 63). Insertion sequences, the simplest of transposable elements, are widely distributed in bacteria and have also been identified in all LAB ­examined to date. Their ability to mediate molecular rearrangements and affect gene regulation has had both positive and negative implications for dairy product fermentations. In one case, the incorporation of an insertion sequence into a prophage of Lb. casei resulted in the conversion of a temperate bacteriophage into a virulent bacteriophage. Alternatively, insertion sequencemediated cointegration has played a pivotal role in the dissemination of many beneficial characteristics via conjugation. More complex transposable elements, such as self-transmissible conjugal transposons that code for the production of the bacteriocin nisin, the ability to metabolize sucrose, and a bacteriophage defense mechanism, have also been described in detail (10, 65). Plasmids, i.e., autonomous replicating extrachromosomal circular DNA molecules, have been identified in several LAB. These are of particular importance in lactococci, where they encode ­numerous characteristics essential for dairy product fermentations, including lactose metabolism, proteinase activity, oligopeptide transport, bacteriophage resistance mechanisms, bacteriocin production and immunity, bacteriocin resistance, exopolysaccharide production, and citric acid utilization (65).

Genetic Modification of LAB Using Recombinant DNA Techniques

Experiments employing recombinant DNA techniques have led to most of the recent advances in the understanding of the physiology and genetics of LAB. The power of recombinant DNA approaches is that strains can be constructed that differ in a single defined genetic alteration, e.g., inactivation of a specific gene. By comparing the wild-type culture to its isogenic derivative, the role of that gene in the phenotype being exam­ ined can be unequivocally determined. This general approach has resulted in a detailed understanding of how these bacteria utilize lactose, obtain essential amino acids, produce diacetyl, and resist bacteriophage infection. Additionally, recombinant DNA approaches have been used to construct novel bacteriophage ­resistance mechanisms and to overproduce enzymes of interest in dairy fermentations. In the future, it is likely that numerous commercial strains will be constructed utilizing recombinant DNA techniques. Readers interested in more comprehensive reviews on the physiology and genetics of LAB are referred to other publications (10, 40, 71).

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

837

References 1. Allison, G. E., and T. R. Klaenhammer. 1998. Phage defense mechanisms in lactic acid bacteria. Int. Dairy J. 8:207–226. 2. Ardö, Y. 2006. Flavour formation by amino acid ­catabolism. Biotechnol. Adv. 24:238–242. 3. Aymes, F., C. Monnet, and G. Corrieu. 1999. Effect of aacetolactate decarboxylase inactivation on a-­acetolactate and diacetyl production by Lactococcus lactis subsp. lac­ tis biovar diacetylactis. J. Biosci. Bioeng. 87:87–92. 4. Bandel, M., M. E. Lhotte, C. Marty-Teysset, A. Veyrat, H. Prévost, V. Dartois, C. Diviès, W. N. Konings, and J. S. Lolkema. 1998. Mechanism of the citrate transporters in carbohydrate and citrate cometabolism in Lactococcus and Leuconostoc species. Appl. Environ. Microbiol. 64:1594–1600. 5. Bassit, N., C. Y. Boquien, D. Picque, and G. Corrieu. 1993. Effect of initial oxygen concentration on diacetyl and acetoin production by Lactococcus lactis subsp. lactis biovar diacetylactis. Appl. Environ. Microbiol. 59:1893–1897. 6. Bassit, N., C. Y. Boquien, D. Picque, and G. Corrieu. 1995. Effect of temperature on diacetyl and acetoin production by Lactococcus lactis subsp. lactis biovar diacety­ lactis. CNRZ 483. J. Dairy Res. 62:123–129. 7. Bills, D. D., and E. A. Day. 1966. Dehydrogenase activity of lactic streptococci. J. Dairy Sci. 49:1473–1477. 8. Bills, D. D., M. E. Morgan, L. M. Libby, and E. A. Day. 1965. Identification of compounds responsible for fruity flavor defect of experimental cheeses. J. Dairy Sci. 48:1168–1173. 9. Bongers, R. S., M. H. N. Hoefnagel, and M. Kleerebezem. 2005. High-level acetaldehyde production in Lactococcus lactis by metabolic engineering. Appl. Environ. Microbiol. 71:1109–1113. 10. Broadbent, J. R. 2001. Genetics of lactic acid bacteria, p. 243–299. In E. H. Marth and J. L. Steele (ed.), Applied Dairy Microbiology, 2nd ed. Marcel Dekker, Inc., New York, NY. 11. Broadbent, J. R., H. Cai, R. L. Larsen, J. E. Hughes, D. L. Welker, V. G. De Carvalho, T. A. Tompkins, Y. Ardö, F. Vogensen, A. De Lorentiis, M. Gatti, E. Neviani, and J. L. Steele. 2011. Genetic diversity in proteolytic enzymes and amino acid metabolism among Lactobacillus helveticus strains. J. Dairy Sci. 94:4313–4328. 12. Broadbent, J. R., and J. L. Steele. 2007. Biochemistry of cheese flavor development: insights from geno­mics studies on lactic acid bacteria, p. 177–192. In K. R. Caldwaller, M. A. Drake, and R. J. McGorrin (ed.), Flavor of Dairy Products. American Chemical Society, Washington, DC. 13. Broadbent, J. R., M. Strickland, B. C. Weimer, M. E. Johnson, and J. L. Steele. 1998. Small peptide accumulation and bitterness in Cheddar cheese made from single strain Lactococcus lactis starters with distinct proteinase specificities. J. Dairy Sci. 81:327–337. 14. Chaves, A. C. S. D., M. Fernandez, A. L. S. Lerayer, I. Mierau, M. Kleerebezem, and J. Hugenholtz. 2002. Metabolic engineering of acetaldehyde production by

SMP_Food Microbiology_CH32.indd

837

Streptococcus thermophilus. Appl. Environ. Microbiol. 68:5656–5662. 15. Christensen, J. E., E. G. Dudley, J. R. Pederson, and J. L. Steele. 1999. Peptidases and amino acid catabolism in lactic acid bacteria. Antonie van Leeuwenhoek 76:217–246. 16. Cogan, T. M. 1987. Co-metabolism of citrate and glucose by Leuconostoc spp.: effects on growth, substrates and products. J. Appl. Bacteriol. 63:551–558. 17. Cogan, T. M., R. J. Fitzgerald, and S. Doonan. 1984. Acetolactate synthase of Leuconostoc lactis and its regulation of acetoin production. J. Dairy Res. 51:597–604. 18. Condon, S. 1987. Responses of lactic acid bacteria to ­oxygen. FEMS Microbiol. Rev. 46:269–280. 19. Crow, V. L. 1990. Properties of 2,3-butanediol dehydrogenases from Lactococcus lactis subsp. lactis in relation to citrate fermentation. Appl. Environ. Microbiol. 56:1656–1665. 20. Davidson, B. E., N. Kordias, M. Dobos, and A. J. Hillier. 1996. Genomic organization of lactic acid bacteria. Antonie van Leeuwenhoek 70:161–183. 21. Deborde, C., D. B. Rolin, A. Bondon, J. D. De Certaines, and P. Boyaval. 1998. In vivo nuclear magnetic resonance study of citrate metabolism in Propionibacterium freud­ enreichii subsp. shermanii. J. Dairy Res. 65:503–514. 22. Desiere, F., S. Lucchini, C. Canchaya, M. Ventura, and H. Brüssow. 2002. Comparative genomics of phages and prophages in lactic acid bacteria. Antonie van Leeuwenhoek 82:73–91. 23. Detmers, F. J. M., E. R. S. Kunji, F. C. Lanfermeijer, B. Poolman, and W. N. Konings. 1998. Kinetics and specificity of peptide uptake by the oligopeptide transport system of Lactococcus lactis. Biochemistry 37:16671–16679. 24. Deveau, H., R. Barrangou, J. E. Garneau, J. Labonté, C. Fremaux, P. Boyaval, D. A. Romero, P. Horvath, and S. Moineau. 2008. Phage response to CRISPR-encoded resistance in Streptococcus thermophilus. J. Bacteriol. 190:1390–1400. 25. Díaz-Muñiz, I., and J. L. Steele. 2006. Limiting galactose requirement for citrate utilization by Lactobacillus casei is annulled in Cheddar cheese extract. Antonie van Leeuwenhoek 90:233–243. 26. Driesson, F. M., and Z. Puhan. 1988. Technology of mesophilic fermented milks. Int. Dairy Fed. Bull. 227:75–81. 27. Dunny, G. M., and L. L. McKay. 1999. Group II introns and expression of conjugative transfer functions in lactic acid bacteria. Antonie van Leeuwenhoek 76:77–88. 28. Forde, A., and G. F. Fitzgerald. 1999. Bacteriophage defense systems in lactic acid bacteria. Antonie van Leeuwenhoek 76:89–113. 29. Fordyce, A. M., V. L. Crow, and T. D. Thomas. 1984. Regulation of product formation during glucose or lactose limitation in nongrowing cells of Streptococcus lactis. Appl. Environ. Microbiol. 48:332–337. 30. Fox, P. F. (ed.). 1993. Cheese: Chemistry, Physics and Microbiology, vol. 1 and 2. Chapman and Hall, Ltd., London, United Kingdom. 31. Fox, P. F., J. Law, P. L. H. McSweeney, and J. Wallace. 1993. Biochemistry of cheese ripening, p. 389–438.

Manila Typesetting Company

11/08/2012 07:13AM

Fermentations and Beneficial Microorganisms

838 In P. F. Fox (ed.), Cheese: Chemistry, Physics and Microbiology, vol. 1. Chapman and Hall, Ltd., London, United Kingdom. 32. Fox, P. F., J. A. Lucey, and T. M. Cogan. 1990. Glycolysis and related reactions during cheese manufacture and ripening. Food Sci. Nutr. 29:237–253. 33. Fryer, T. F., M. E. Sharpe, and B. Reiter. 1970. Utilization of milk citrate by lactic acid bacteria and “blowing” of film wrapped cheese. J. Dairy Sci. 37:17–28. 34. Gancel, F., and G. Novel. 1994. Exopolysaccharide production by Streptococcus salivarius spp. thermophi­ lus cultures. 1. Conditions of production. J. Dairy Sci. 77:685–688. 35. Green, M. L., and D. J. Manning. 1982. Development of texture and flavor in cheese and other fermented products. J. Dairy Res. 49:737–748. 36. Griffith, R., and E. G. Hammond. 1989. Generation of Swiss cheese flavor components by the reaction of amino acids with carbonyl compounds. J. Dairy Sci. 72:604–613. 37. Horvath, P., A.-C. Coûté-Monvoisin, D. A. Romero, P. Boyaval, C. Fremaux, and R. Barrangou. 2009. Comparative analysis of CRISPR loci in lactic acid bacteria genomes. Int. J. Food Microbiol. 131:62–70. 38. Huang, D. Q., H. Prévost, and C. Diviès. 1995. Principal characteristics of b-galactosidase from Leuconostoc spp. Int. Dairy J. 5:29–43. 39. Hugenholtz, J. 1993. Citrate metabolism in lactic acid bacteria. FEMS Microbiol. Rev. 12:165–178. 40. Hutkins, R. W. 2001. Metabolism of starter cultures, p. 207–241. In E. H. Marth and J. L. Steele (ed.), Applied Dairy Microbiology, 2nd ed. Marcel Dekker, Inc., New York, NY. 41. Imhof, R., and J. O. Bosset. 1994. Review: relationships between micro-organisms and formation of aroma compounds in fermented dairy products. Z. Lebensm. Unters. Forsch. 198:267–276. 42. International Dairy Federation. 1988. Fermented Milks: Science and Technology. International Dairy Federation, Brussels, Belgium. 43. Jensen, N. B. S., C. R. Melchiorsen, K. V. Jokumsen, and J. Villadsen. 2001. Metabolic behavior of Lactococcus lactis MG1363 in microaerobic continuous cultivation at a low dilution rate. Appl. Environ. Microbiol. 67:2677–2682. 44. Juillard, V., D. Le Bars, E. R. S. Kunji, W. N. Konings, J.-C. Gripon, and J. Richard. 1995. Oligopeptides are the main source of nitrogen for Lactococcus lac­ tis during growth in milk. Appl. Environ. Microbiol. 61:3024–3030. 45. Kamaly, M. K., and E. H. Marth. 1989. Enzyme activities of lactic streptococci and their role in maturation of cheese: a review. J. Dairy Sci. 72:1945–1966. 46. Keenan, T. W., R. C. Lindsay, M. E. Morgan, and E. A. Day. 1966. Acetaldehyde production by single strain lactic streptococci. J. Dairy Sci. 49:10–14. 47. Kempler, G. M., and L. L. McKay. 1981. Biochemistry and genetics of citrate utilization in Streptococcus lactis subsp. diacetylactis. J. Dairy Sci. 64:1527–1539.

SMP_Food Microbiology_CH32.indd

838

48. Klaenhammer, T., E. Altermann, F. Arigoni, A. Bolotin, F. Breidt, J. Broadbent, R. Cano, S. Chaillou, J. Reutscher, M. Gasson, M. van de Guchte, J. Guzzo, A. Hartke, T. Hawkins, P. Hols, R. Hutkins, M. Kleerebezem, J. Kok, O. Kuipers, M. Lubbers, E. Maguin, L. McKay, D. Mills, A. Nauta, R. Overbeek, H. Pel, D. Pridmore, M. Saier, D. van Sinderen, A. Sorokin, J. Steele, D. O’Sullivan, W. de Vos, B. Weimer, M. Zagorec, and R. Siezen. 2002. Discovering lactic acid bacteria by genomics. Antonie van Leeuwenhoek 82:29–58. 49. Klaenhammer, T. R., and G. F. Fitzgerald. 1994. Bacterio­ phages and bacteriophage resistance, p. 106–168. In M. J. Gasson and W. M. de Vos (ed.), Genetics and Bio­ technology of Lactic Acid Bacteria. Chapman and Hall, Ltd., London, United Kingdom. 50. Kunji, E. R. S., I. Mierau, A. Hagting, B. Poolman, and W. N. Konings. 1996. The proteolytic systems of lactic acid bacteria. Antonie van Leeuwenhoek 70:187–221. 51. Lees, G. J., and G. R. Jago. 1978. Role of acetaldehyde in metabolism: a review. 1. Enzymes catalyzing reactions involving acetaldehyde. J. Dairy Sci. 61:1205–1215. 52. Lees, G. J., and G. R. Jago. 1978. Role of acetaldehyde in metabolism: a review. 2. The metabolism of acetaldehyde in cultured dairy products. J. Dairy Sci. 61:1216–1224. 53. Lemieux, L., and R. E. Simard. 1992. Bitter flavour in dairy products. II. A review of bitter peptides from caseins: their formation, isolation and identification, structure masking and inhibition. Lait 72:335–382. 54. Lindgren, S. E., and L. T. Axelsson. 1990. Anaerobic llactate degradation by Lactobacillus plantarum. FEMS Microbiol. Lett. 66:209–214. 55. Lindsay, R. C., E. A. Day, and W. E. Sandine. 1965. Green flavor defect in lactic starter cultures. J. Dairy Sci. 48:863–869. 56. Liu, M., J. R. Bayjanov, B. Renckens, A. Nauta, and R. J. Siezen. 2010. The proteolytic system of lactic acid bacteria revisited: a genome comparison. BMC Genomics 11:36. 57. Liu, S. Q., R. V. Asmundson, P. K. Gopal, R. Holland, and V. L. Crow. 1998. Influence of reduced water activity on lactose metabolism by Lactococcus lactis subsp. cremoris at different pH values. Appl. Environ. Microbiol. 64:2111–2116. 58. Lucey, C. A., and S. Condon. 1986. Active role of oxygen and NADH oxidase in growth and energy metabolism of Leuconostoc. J. Gen. Microbiol. 132:1789–1796. 59. Manning, D. J. 1979. Chemical production of essential Cheddar flavor compounds. J. Dairy Res. 46:531–537. 60. McKay, L. L., and K. A. Baldwin. 1974. Altered metabolism in a Streptococcus lactis C2 mutant deficient in lactate dehydrogenase. J. Dairy Sci. 57:181–186. 61. Mierau, I., E. R. S. Kunji, K. J. Leenhouts, M. A. Hellendoorn, A. J. Haandrikman, B. Poolman, W. N. Konings, G. Venema, and J. Kok. 1996. Multiple­peptidase mutants of Lactococcus lactis are severely impaired in their ability to grow in milk. J. Bacteriol. 178:2794–2803. 62. Mills, S., R. P. Ross, H. Neve, and A. Coffey. 2011. Bacteriophage and anti-phage mechanism in lactic acid bacteria, p. 165–186. In S. Lahtinen, A. C. Ouwehand,

Manila Typesetting Company

11/08/2012 07:13AM

32.  Fermented Dairy Products

839

S. Salminen, and A. von Wright (ed.), Lactic Acid Bacteria: Micro­biology and Functional Aspects, 4th ed. CRC Press, Boca Raton, FL. 63. Moineau, S. 1999. Applications of phage resistance in lactic acid bacteria. Antonie van Leeuwenhoek 76:377–382. 64. Montville, M. R., B. Ardestani, and B. L. Geller. 1994. Lactococcal bacteriophage require a host cell wall carbohydrate and a plasma membrane protein for adsorption and ejection of DNA. Appl. Environ. Microbiol. 60:3204–3211. 65. Morelli, L., F. K. Vogensen, and A. von Wright. 2004. Genetics of lactic acid bacteria, p. 249–293. In S. Salminen, A. von Wright, and A. Ouwehand (ed.), Lactic Acid Bacteria: Microbiology and Functional Aspects, 3rd ed. Marcel Dekker, Inc., New York, NY. 66. Palles, T., T. Beresford, S. Condon, and T. M. Cogan. 1998. Citrate metabolism in Lactobacillus casei and Lactobacillus plantarum. J. Appl. Microbiol. 85:147–154. 67. Peterson, S. D., and R. T. Marshall. 1990. Nonstarter lactobacilli in Cheddar cheese: a review. J. Dairy Sci. 73:1395–1410. 68. Poolman, B. 1993. Energy transduction in lactic acid bacteria. FEMS Microbiol. Rev. 12:125–148. 69. Ramos, A., K. N. Jordan, T. M. Cogan, and H. Santos. 1994. 13C nuclear magnetic resonance studies of citrate and glucose cometabolism by Lactococcus lactis. Appl. Environ. Microbiol. 60:1739–1748. 70. Ramos, A., J. S. Lolkema, W. N. Konings, and H. Santos. 1995. Enzyme basis for pH regulation of citrate and pyruvate metabolism by Leuconostoc oenos. Appl. Environ. Microbiol. 61:1303–1310. 71. Salminen, S., A. von Wright, and A. Ouwehand (ed.). 2004. Lactic Acid Bacteria: Microbiology and Functional Aspects, 3rd ed. Marcel Dekker, Inc., New York, NY. 72. Schmitt, P., and C. Diviès. 1991. Co-metabolism of citrate and lactose by Leuconostoc mesenteroides subsp. cremo­ ris. J. Ferment. Bioeng. 71:72–74. 73. Sieuwerts, S., D. Molenaar, S. A. F. T. van Hijum, M. Beerthuyzen, M. J. A. Stevens, P. W. M. Janssen, C. J. Ingham, F. A. M. de Bok, W. M. de Vos, and J. E. T. van Hylckama Vlieg. 2010. Mixed-culture transcriptome analysis reveals the molecular basis of mixed-culture growth in Streptococcus thermophilus and Lactobacillus bulgaricus. Appl. Environ. Microbiol. 76:7775–7784. 74. Sijtsma, L., N. Jansen, W. C. Hazeleger, J. T. M. Wouters, and K. J. Hellingwerf. 1990. Cell surface characteristics of bacteriophage-resistant Lactococcus lactis subsp. cremo­ ris SK110 and its bacteriophage-sensitive variant SK112. Appl. Environ. Microbiol. 56:3230–3233. 75. Smart, J. B., and T. D. Thomas. 1987. Effect of oxygen of lactose metabolism in lactic streptococci. Appl. Environ. 53:533–541.

SMP_Food Microbiology_CH32.indd

839

76. Smit, G., B. A. Smit, and W. J. M. Engels. 2005. Flavour formation by lactic acid bacteria and biochemical flavor profiling of cheese products. FEMS Microbiol. Rev. 29:591–610. 77. Smit, G., A. Verheul, R. van Kranenburg, E. Ayad, R. Siezen, and W. J. M. Engels. 2000. Cheese flavour development by enzymatic conversions of peptides and amino acids. Food Res. Int. 33:153–160. 78. Sturino, J. M., and T. R. Klaenhammer. 2004. Bacterio­ phage defense systems and strategies for lactic acid bacteria. Adv. Appl. Microbiol. 56:331–378. 79. Thierry, A., D. Salvat-Brunaud, and J.-L. Maubois. 1999. Influence of thermophilic lactic acid bacteria strains on propionibacteria growth and lactate consumption in an Emmental juice-like medium. J. Dairy Res. 66:105–113. 80. Thomas, T. D., D. C. Ellwoos, and V. M. C. Longyear. 1979. Change from homo- to heterolactic fermentation by Streptococcus lactis resulting from glucose limitation in anaerobic chemostat cultures. J. Bacteriol. 138:109–117. 81. Urbach, G. 1995. Contribution of lactic acid bacteria to flavor compound formation in dairy products. Int. Dairy J. 5:877–903. 82. Vaughan, E. E., S. David, A. Harrington, C. Daly, G. F. Fitzgerald, and W. M. De Vos. 1995. Characterization of plasmid-encoded citrate permease (citP) genes from Leuconostoc species reveals high sequence conversation with the Lactococcus lactis citP gene. Appl. Environ. Microbiol. 61:3172–3176. 83. Visser, S. 1993. Proteolytic enzymes and their relation to cheese ripening and flavor: an overview. J. Dairy Sci. 76:329–350. 84. Warriner, K. S. R., and J. G. Morris. 1995. The effects of aeration on the bioreductive abilities of some heterofermentative lactic acid bacteria. Lett. Appl. Microbiol. 20:322–327. 85. Weimer, B., K. Seefeldt, and B. Dias. 1999. Sulfur metabolism in bacteria associated with cheese. Antonie van Leeuwenhoek 76:247–261. 86. Whitehead, W. E., J. W. Ayres, and W. E. Sandine. 1993. A review of starter media for cheese making. J. Dairy Sci. 76:2344–2353. 87. Wilkins, D. W., R. H. Schmidt, R. B. Shireman, K. L. Smith, and J. J. Jezeski. 1986. Evaluating acetaldehyde synthesis from L-[14C(U)] threonine by Streptococcus thermophilus and Lactobacillus bulgaricus. J. Dairy Sci. 69:1219–1224. 88. Yvon, M., and L. Rijnen. 2001. Cheese flavor formation by amino acid catabolism. Int. Dairy J. 11:185–201. 89. Zourari, A., J. P. Accolas, and M. J. Desmazeaud. 1992. Metabolism and biochemical characteristics of yogurt bacteria. A review. Lait 72:1–34.

Manila Typesetting Company

11/08/2012 07:13AM

SMP_Food Microbiology_CH32.indd

840

Manila Typesetting Company

11/08/2012 07:13AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch33

Fred Breidt Roger F. McFeeters Ilenys Perez-Diaz Cherl-Ho Lee

33

Fermented Vegetables

Historical Perspective The wide variety of fermented foods can be classified by the products of the fermentation, such as alcohol (beer, wine); organic acids, including lactic acid and acetic acid (vegetables, dairy); carbon dioxide (bread); and amino acids or peptides from protein (fish fermentations and others) (35, 56, 96, 97). Food fermentation is one of the earliest technologies developed by humans. Littoral foragers in Asia during the primitive pottery age (8000 to 3000 b.c.) are believed to have fermented vegetables prior to the development of crop-based agriculture (57). Dairy fermentations in the Middle East likely followed the domestication of cattle around this time. It is likely that the first product of fermentation to be discovered was alcohol from fermented fruits. More sophisticated fermentation skills using cereals to make alcohol were developed around 4000 b.c., with beer produced in Egypt and rice wine in northeast Asia (56). Early written references to fermentation technology in Asia are found in the historic Chinese book of poems Shijing (1100 to 600 b.c.), which celebrates “the thou-

sand wines of Yao,” a reference to a kingdom in China from 2300 b.c. It is believed that cucumbers were first fermented around 2000 b.c. in the Middle East. Early written records of cucumber pickles come from paper fragment remains of a play (The Taxiarchs) by the Greek writer Eupolis (429–412 b.c.), and pickles are also mentioned several times in the Christian Bible. Korean-style fermented cabbage, kimchi, is thought to have originated in the primitive pottery age from the natural fermentation of withered vegetables stored in seawater (56). Early references to kimchi include the Korean poem “Gapoyugyeong in Donggugisanggukjip” by Yi Kyu-Bo (1168-1241 a.d.). European-style sauerkraut is thought to have originated in China, and the technology may have been brought to Europe during the Mongol invasion of central Europe in the 13th century. Today, industrial vegetable fermentation is carried out on a massive scale. In the United States, companies producing cucumber pickles can have at one location as many as 1,000 fermentation tanks of 40,000-liter capacity, totaling 40 million liters.

Fred Breidt and Ilenys Pérez-Díaz, U.S. Department of Agriculture, Agricultural Research Service, and Department of Food, Bioprocessing & Nutrition Sciences, NC State University, Raleigh, NC 27603. Roger F. McFeeters, U.S. Department of Agriculture, Agricultural Research Service, and Department of Food, Bioprocessing & Nutrition Sciences, NC State University, Raleigh, NC 27603 (retired). Cherl-Ho Lee, Graduate School of Biotechnology, Korea University, Seoul, 136-701 Korea.

841

SMP_Food Microbiology_CH33.indd

841

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

842

Vegetable Fermentation Overview The primary retail fermented vegetable products produced in the United States and Europe are cucumber pickles, olives, and sauerkraut. In Asia, a variety of fermented vegetable products are available, including pickles and fermented cabbage, notably kimchi in South Korea. The word “pickle” by itself usually refers to a pickled cucumber. The current market for pickled vegetables (fermented and acidified) in the United States is roughly $2 billion. The retail market for cucumber pickles, however, is dominated by acidified, pasteurized, and refrigerated products which are not fermented. In addition to cucumbers, a number of nonfermented pickled vegetable products, mainly acidified peppers, are also popular in the retail market (35). Commercial sale of hamburger dill pickle slices to the food service industry makes up most of the U.S. market for fermented cucumber pickles. Other cucumber pickles include deli-style, half-sour cucumber pickles, which are partially fermented prior to consumption (28). Fermented cabbage, i.e., sauerkraut, was introduced in the United States by immigrants from Germany and other European countries. Although the popularity of pickles and sauerkraut in Europe continues today, consumption has declined in the United States. Commercial production of kimchi in South Korea is increasing, due to the purchase of commercially prepared kimchi by people living in urban areas. Fermented olives are produced primarily in southern Spain, which has the world’s largest export market for table olives (31). Examples of acid-fermented vegetables produced in different regions of the world are listed in Table 33.1 (55). Reports on the microbiology and biochemistry of vegetable fermentations first appeared in the scientific literature in the early 1900s. Early research on the “lactic bacilli” present in fermenting vegetables was done by E. B. Fred at the University of Wisconsin (92, 93). Carl

Pederson, at Cornell University, studied sauerkraut fermentation from the 1930s to early 1970s. He reported on various aspects of the subject, which culminated in a comprehensive review article (86). J. L. Etchells and R. N. Costilow published extensively in the field of pickled vegetables. Included in these studies were the ­development of pasteurization methods (30, 49), investigations of the yeasts that are responsible for spoilage of cucumber pickle products (26, 29), and a ­preservationprediction chart to describe the storage stability of sweet pickles based on salt and sugar concentrations (2). H. P. Fleming and coworkers, along with Costilow, developed the purging technology that is now commonly used in commercial cucumber fermentations (17, 33). Other important developments include methods for controlled fermentations (6, 27), an understanding of the role of the malolactic enzyme of lactobacilli in the production of carbon dioxide during cucumber fermentation (75), the development of a Lactobacillus plantarum strain that lacks the ability to carry out the malolactic reaction (19), and the use of calcium to improve the texture of pickled vegetables (73, 74). The fermentation process for vegetables can result in nutritious foods that may be stored for extended periods, 1 year or more, without refrigeration. Prior to fermentation, fresh fruits and vegetables harbor a variety of microorganisms, including aerobic spoilage microflora such as Pseudomonas, Erwinia, and Enterobacter species, as well as yeasts and molds (83). The cell populations for these bacteria, which spoil the vegetables if allowed to grow, range from 104 to 106 CFU/g. Brining vegetables for fermentation results in the production by lactic acid bacteria (LAB) of organic acids and a variety of antimicrobial compounds (11, 23). LAB are initially present on fresh vegetables in lower numbers, 102 to 103 CFU/g, compared with other mesophilic microorganisms. During fermentation, diffusion of organic acids into the brine, and the low pH that results, influences microbial growth across

Table 33.1  Examples of acid-fermented vegetables produced in different regions of the world Product name

Country

Major ingredients

Microorganisms

Usage

Sauerkraut

Germany

Cabbage, salt

Leuconostoc mesenteroides, Lactobacillus brevis, Lactobacillus plantarum

Salad, side dish

Kimchi

Korea

Korean cabbage, radish, various vegetables, salt

L. mesenteroides, Lb. brevis, Lb. plantarum

Salad, side dish

Dhamuoi

Vietnam

Cabbage, various vegetables

L. mesenteroides, Lb. plantarum

Salad, side dish

Dakguadong

Thailand

Mustard leaf, salt

Lb. plantarum

Salad, side dish

Burong mustasa

Philippines

Mustard

Lb. brevis, Pediococcus cerevisiae

Salad, side dish

SMP_Food Microbiology_CH33.indd

842

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

843

the surface of the vegetable material. As sugars diffuse from the vegetables into the brine, the LAB grow rapidly. Because the LAB are more acid resistant than the spoilage microbiota, they ­dominate brined vegetable fermentations. In the absence of brine, spoilage microbiota are able to grow, unhindered by the metabolic end products of the LAB. The growth of spoilage bacteria results in deterioration of the vegetable material, due to the elaboration of degradative enzymes (proteases, lipases, amylases, nucleases, and others). Leuconostoc mesenteroides and related LAB, inclu­ ding Weissella and other Leuconostoc species, are important in the initiation of the fermentation of many vegetables, i.e., cabbages, beets, turnips, cauliflower, green beans, sliced green tomatoes, olives, and sugar beet ­silages. L. mesenteroides grows more rapidly than most other LAB over a range of temperatures (5 to 35°C) and NaCl brine concentrations (0 to 5%). L. mesenteroides carries out a heterolactic fermentation of vegetable sugars, typically fructose and glucose, and produces ­carbon dioxide and acids (lactic and acetic). The production of acid quickly lowers the pH, thereby inhibiting the ­development of undesirable microorganisms and the ­activity of their enzymes. The carbon dioxide produced replaces air and provides anaerobic conditions favorable for the stabilization of ascorbic acid and the natural color of the vegetables. The high acidity produced by L. mesenteroides and other LAB subsequently inhibits the growth of these heterofermentative ­microbes in favor of more acid-tolerant homofermentative LAB. Homofermentative species, such as Lb.  plantarum, produce exclusively lactic acid from the ­ remaining sugars. In most vegetable fermentations, Lb. plantarum will eventually outcompete other LAB because of its superior acid tolerance (67). In brined vegetables with initial NaCl concentrations of 6 to 12%, such as pickled cucumbers and olives, Lactobacillus species, primarily Lb. plantarum, dominate the fermentation from the start, with little or no ­evidence of heterolactic species present. There are several excellent reviews of cucumber pickle, sauerkraut, kimchi, and olive fermentations (14, 31, 35, 37, 102). Here we present a summary of the commercial production practices, microbiology, and biochemistry of ­ cucumber, cabbage, and olive fermentations.

Cucumber Fermentations In the United States, commercial cucumber (Cucumis sativus) fermentations are commonly done in 30,000- to 40,000-liter, open-top, fiberglass tanks that are ­located out-of-doors so the brine surface is exposed to sunlight.

SMP_Food Microbiology_CH33.indd

843

The UV radiation in sunlight is relied upon to kill aerobic surface yeasts that can metabolize lactic acid produced by the fermentation. Cucumbers are covered with salt brine and held below the brine surface with wooden headboards. Fermentations are typically carried out in brine equilibrated at about 6% NaCl. Calcium chloride (0.1 to 0.4%, equilibrated) is added to the cover brine to maintain the firm, crisp texture of the fermented cucumbers during fermentation and storage (35). Cucumber fermentations typically undergo a homolactic acid fermentation, which does not result in production of carbon dioxide from sugars (glucose and fructose, about 1% each). However, carbon dioxide may be generated from the respiration of cucumbers and by the decarboxylation of malate during the initiation of fermentation (75). Some LAB have an inducible malolactic enzyme that converts malate to lactate and carbon dioxide. The malolactic enzyme reaction occurs intracellularly and results in the uptake of a proton, thereby increasing the internal cell pH. While it is a desirable reaction in winemaking (used for de-acidifying wines), malolactic fermentation in cucumbers may result in the creation of “bloaters,” or fermented cucumbers with undesired internal gas pockets, decreasing the production yield. In an effort to prevent bloater formation, cucumber fermentations are purged with air to remove excess carbon dioxide from the tank (17). Potassium sorbate (~0.04%) or 0.16% acetic acid can be used as processing aids to limit the growth of aerobic microorganisms in air-purged cucumber fermentations, particularly molds and yeasts (42). Excessive growth of aerobic microorganisms may also be controlled by stopping air purging for several hours each day. After fermentation by Lb. plantarum and related LAB, cucumbers may be stored in the fermentation tanks for 1 year or more. In areas where the temperature decreases to below 0°C, the concentration of NaCl is often increased during storage to as high as 10 to 15% to minimize freezing damage and maintain the desirable texture of fermented cucumbers. Prior to sale, cucumbers are washed to remove excess salt and then packed in a variety of containers (plastic pails, pouches, jars) with an appropriate cover liquor. The cover liquor typically contains acetic acid and spices in addition to residual lactic acid. Fermented pickles may be pasteurized, but large containers are not heat treated. Further microbial growth is prevented by the organic acids, low pH, and lack of fermentable sugars. Most commercial cucumber fermentations rely upon growth of the LAB that are naturally present on the surface of cucumbers. However, some processors choose to use starter cultures to enhance product consistency. A commercial starter culture of Lb. plantarum that does

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

844 not decarboxylate malic acid (and hence does not contribute to the formation of bloaters) has been developed (19) and tested to determine the ability of the culture to grow in cucumber fermentations (6). A method for the preparation of a starter culture that meets kosher requirements is also available to processors (88). The initial pH of brined cucumbers is about 6.5. In practice, commercial fermentations may use recycled brine, or acetic acid may be added to brine solutions. The addition of acid can help remove excess CO2 and also help select for the growth of LAB, so the initial pH of commercial fermentations can vary substantially. In addition to lactic acid, the LAB produce a variety of metabolites, e.g., bacteriocins, peroxides, and peptides, that can be inhibitory to other bacteria (23). At the end of the fermentation, there may be 1.5% lactic acid, a pH of 3.1 to 3.5, and little or no residual sugar. In this anaerobic, acidic, high-salt environment lacking sugar, very few microorganisms are capable of growing or surviving, effectively preserving the cucumbers. Occasionally, fermented cucumbers undergo an undesired secondary fermentation during storage, which is characterized by an increase in pH, the disappearance of lactic acid, and the formation of propionic and butyric acids. The incidence of fermented cucumber spoilage tends to increase at the beginning of the spring season, when the ambient temperature increases. The increase in the concentration of propionic and butyric acids causes a malodorous spoilage (32). The microbial ecology of this type of spoilage is currently not well defined but may include the growth of spore-forming bacteria such as clostridia when the pH increases above 4.6. Fermented cucumbers have salt concentrations (6% or greater) that are too high to be used in products for direct human consumption. Hence, prior to packing and distribution, the salt is reduced to about 2% by washing with water. This results in a waste stream with high concentrations of salt plus a high biological oxygen demand from the organic components that are present in the brine and that diffuse out of the cucumbers during the desalting process. To reduce the environmental ­impact, cucumber brine from the desalting process is usually recycled and may be used in subsequent fermentations (72). Prior to recycling, fermentation brines may be processed to remove “softening enzymes,” primarily polygalacturonases (10), which can degrade pectic substances in the cucumber cell wall and soften the fruits.

Cabbage Fermentations Commercial production of fermented cabbage consists primarily of kimchi, made from the Chinese cabbage,

SMP_Food Microbiology_CH33.indd

844

Brassica rapa, in Korea, and sauerkraut, from Brassica oleracea, in the United States and Europe. Sauerkraut fermentations are done in bulk fermentation tanks that may contain 100 tons or more of shredded or chopped cabbage. The cabbage for these fermentations consists of large heads, typically 3.6 to 4.5 kg. The outer leaves and woody core of the cabbage are removed prior to shredding or chopping. This core of the cabbage contains sucrose, which can lead to dextran formation by L. mesenteroides, resulting in a slimy or stringy texture. The shredded cabbage is dry salted as it is conveyed to fermentation tanks. This process results in a brine forming in the fermentation tanks, with an NaCl concentration of about 2 to 3%. During the first 24 to 48 h, carbon dioxide gas and lactic and acetic acids are produced. Some of the excess brine formed from the salted cabbage may be removed from the tanks during the first week of fermentation. Cabbage typically contains 4 to 5% sugar, consisting of about 2.5% glucose and 2% fructose (38). The initial heterolactic stage of the fermentation results in production of both lactic and acetic acids. The volatile acetic acid makes an important contribution to the flavor and aroma of the final product. Heterofermentative microorganisms also use fructose as an electron acceptor, converting it to mannitol (71). After about 1 week of fermentation, the heterofermentative LAB, which may grow to 9 log CFU/ml or greater, die off. They are replaced by the more acid-tolerant homo­fermentative microorganisms. This biphasic pattern of growth and death can be seen by plating total LAB using MRS agar, with anaerobic growth at 30°C (35, 36). High-quality sauerkraut can be produced without a starter culture if the equilibrated NaCl concentration is adjusted to 2% and the temperature is maintained at 18°C (85). The fermentation end products present after both stages of the fermentation can include mannitol and acetic acid (about 1% each) and lactic acid, which may exceed 2%, because sugar is in excess and does not limit the extent of fermentation. For most manufacturers in the United States, sauerkraut may be stored for up to 1 year in fermentation tanks until it is processed for food service or retail sale. While bulk storage is economical, the products may become very sour as lactic acid accumulates. European manufacturers typically package sauerkraut at the end of the heterolactic fermentation stage (about 1 week after the beginning of fermentation) to produce a product with mild acid flavor (35). Spices, wines, and other ingredients may be added to the sauerkraut to augment flavor. The traditional composition of the microbiota present in sauerkraut fermentations was described by Pederson and Albury (86). The report, written at the

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

845

end of Pederson’s career after 40 years of research, described two heterolactic species, L. mesenteroides and Lactobacillus brevis, and two homolactic species, Pediococcus cerevisiae and Lb. plantarum, as the primary bacteria present in the fermentation. In these early studies, microbial ecology data were obtained primarily by microscopic observation of stained cells and biochemical tests using isolated cultures. The advent of molecular techniques ­allowed for a more detailed examination. Organisms that were morphologically similar and possessed similar biochemical pathways were determined to be genetically different. LAB such as Leuconostoc citreum, Leuconostoc argentinum, Leuconostoc fallax, Lactobacillus paraplantarum, Lactobacillus coryniformis, and Weissella spp. were identified in commercial sauerkraut fermentations by analysis of the 16S rRNA sequences from hundreds of commercial isolates obtained from four fermentations (90). Surprisingly, only a few isolates of Lb. brevis and Pediococcus spp., which were considered to be two of the four principal microorganisms present in sauerkraut, were isolated in the study. Substantial variation between fermentations was observed. Heterolactic Weissella predominated in the early stage of one fermentation, not the Leuconostoc species (90). The biochemical changes, for these fermentations were very similar, and the differences in microbiota did not affect the resulting quality of the sauerkraut. Kimchi fermentation is microbiologically similar to sauerkraut fermentation, although the ingredients, flavor, and preparation methods differ. Chinese cabbage and Asian radishes are popular primary ingredients in kimchi fermentations. For cabbage kimchi (known as baechu kimchi), the fresh cabbage is cut in half lengthwise or quartered and initially soaked in brine of 5 to 10% NaCl to wilt the cabbage. The cabbage is then washed and drained. An aqueous paste of ground red pepper is prepared and mixed in with the cabbage leaves. Small amounts of additional ingredients, such as garlic, ginger, and jeotgal, a highly salted (20% NaCl) anchovy product (63), are usually included along with additional vegetable material, such as green onion. After the cabbage and other ingredients are packed into containers (jars, pouches), the final salt concentration is between 3 and 6%. In rural areas, kimchi was traditionally packed into earthen jars and buried in the soil to allow constant temperatures for fermentation and a supply of vegetables through the winter. In urban South Korea today, kimchi is prepared commercially or by individuals using household “kimchi refrigerators.” These are small, programmable-temperature refrigerators that provide an initial 18°C fermentation period of a few days, followed by very cold refrigeration (1 to 2°C). This procedure allows

SMP_Food Microbiology_CH33.indd

845

the initial heterolactic stage of fermentation to occur but delays the onset of the homolactic stage of fermentation, keeping kimchi from becoming too sour. Optimum taste is attained when the pH and acidity reach approximately 4.0 to 4.5 and 0.5 to 0.6%, respectively. The vitamin B content increases during sauerkraut and kimchi fermentations, and vitamin C and A are preserved (35, 56). Many Koreans prefer a lightly fermented, carbonated kimchi, characteristic of the initial, heterolactic stage of fermentation. In addition to cabbage or baechu kimchi, there are many other types of kimchi, depending on the type of vegetables used in the fermentation and the ingredients added. These include white kimchi, which contains no red pepper. Commercially prepared refrigerated kimchi, either factory packaged in pouches (containing a CO2 adsorbent) or freshly made in grocery stores, represents a rapidly growing market in South Korea. As described above for sauerkraut, the traditional kimchi fermentation has a biphasic heterofermentative and then homofermentative pattern of LAB succession, with a few prominent species (L. mesenteroides, P. cerevisiae, Lb. brevis, and Lb. plantarum) (56, 79). More recent studies of kimchi fermentation, including DNAbased, culture-independent methods, have revealed a complex microbiota as described above. A variety of Weissella, Leuconostoc, Pediococcus, and Lactobacillus species, as well as yeasts and Archaea, have been identified. The Archaea are likely found in kimchi due to the presence of extreme halophiles present in jeotgal. Molecular methods used in these studies include microarrays (82), denaturing gradient gel electrophoresis (12), and 16S sequencing (15, 84).

Olive Fermentations Like cucumber and cabbage fermentations, olive fermentation practices are based on traditional methods that have been modified for large-scale commercial production. There are several methods used for processing olives (31). The principal types of products include green table olives, natural black olives in brine, and canned ripe black olives. Green table olives are treated with lye (NaOH) and then washed prior to being brined and fermented. Following fermentation, they may be pitted and stuffed before sale. Natural black olives are prepared by a slow fermentation without lye or further treatment. Ripe black olives are prepared by darkening olives through oxidation in an alkali followed by washing and canning. The preparation of green table olives, commonly sold stuffed with pimento or a variety of other materials, involves treating olives with 1 to 3% NaOH prior

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

846 to fermentation. The addition of a strong base serves multiple purposes. The NaOH treatment helps reduce the natural bitterness of the fruit, due to the degradation of oleuropein (40), and reduces the antimicrobial activity of the phenolic components of olives (76). The NaOH treatment also makes the skin of the olive more permeable, aiding sugar diffusion during fermentation. After NaOH treatment, the olives are washed to remove excess alkali and then brined in 10% NaCl, so the salt equilibrates with the fruit and results in a final concentration of around 5 to 6% for fermentation. The initial pH of the fermentation can be above 7 depending on how much washing was done after the NaOH treatment. As a consequence, the initial microflora during fermentation can include a variety of gram-positive bacilli (Bacillus species) and gram-negative enteric bacteria (Enterobacter, Citrobacter, Klebsiella, and Escherichia). As organic acids accumulate and the pH decreases below 6, the LAB, principally Lb. plantarum, dominate the fermentation to the exclusion of the other gram-positive and gram-negative microbes. Yeast species may also be present (Candida, Pichia, Saccharomyces, and others) (31, 43) and contribute desirable flavor characteristics to the brined olives. As with cucumber fermentation, some purging with air may be done to remove excess CO2 and prevent gas pockets that may form in the fruit. However, this can lead to growth of oxidative yeasts, which consume lactic acid and result in elevated pH and spoilage problems. Natural black olives are also prepared by fermentation but do not receive an NaOH treatment prior to brining. They are picked in a ripened state and have a black color as well as a softer texture than green table olives. Fermentation is a much slower process in black olives because of the lack of NaOH treatment. Antimicrobial phenolic compounds diffuse into the brine, which slows fermentation, and diffusion of sugars is also reduced compared with the NaOH-treated green olives. As a consequence, the fermentation may take months to complete. Commercial “ripe black olives” are also prepared from green or semiripened olives that have been brined without an initial NaOH treatment. Following storage in brine for up to 1 year, the olives are subjected to one or more vigorous oxidation treatments with pressurized air in the presence of 1 to 2% NaOH. This treatment blackens the olives, which are then washed with water to remove NaOH and bring the pH down to around 7 or less. The olives are then canned in a 1 to 3% NaCl brine and processed in a retort to sterilize the fruit. Sterilization is needed for these black olives to prevent botulism, because the pH is significantly above 4.6. Because the color of olives blackened by oxidation

SMP_Food Microbiology_CH33.indd

846

is not stable, iron-containing compounds, such as iron gluconate or iron lactate, can be added to help stabilize the black color (31). Problems associated with the commercial fermentation and processing of olives include disposal of waste NaCl brines and NaOH solutions, as well as ­malodorous spoilage fermentations. As with other vegetable fermentations, waste NaCl brine disposal can cause environmental problems if the scale of the brining operation is large. As a consequence, fermentation brines can be recycled, but this may result in the accumulation of offflavors and contribute to spoilage problems. For olive processing, sodium hydroxide solutions are also reused, although they must eventually be disposed of. To ­reduce disposal problems with NaOH, green table olives may have limited washing, resulting in an initial pH that may be high enough to result in excessive growth of enteric bacteria during the early stage of fermentation. Vaughn and coworkers (89, 102) characterized Bacillus and Clostridium species that cause excess gas production and malodorous (known as zapatera) fermentation. Propionic acid bacteria may also be involved in spoilage fermentations (43).

Bacteriophages in vegetable fermentations Bacteriophages that infect LAB were first identified in 1935 (103) in dairy fermentations. Bacteriophages were not investigated in vegetable fermentations (kimchi and sauerkraut) until recently (105–107). Dairy fermentations are typically carried out with pasteurized milk and a single starter culture or cocktail of a few cultures. If the dairy starter culture(s) fails to ferment milk due to bacteriophage infection, the result may be a costly spoilage problem. Vegetable fermentations, however, do not typically use starter cultures, but if phages are present and inhibit one strain of bacteria, other (resistant) strains of indigenous LAB will grow instead. It is possible, however, that bacteriophages have an impact on microbial succession. Since the initial reports mentioned above, over 100 bacteriophages have been isolated and characterized from cucumber and cabbage fermentations (1, 60, 61, 81, 104). In one large study of commercial sauerkraut fermentations (61), more than 40 phages and LAB were characterized to determine host range. Isolates included phages from both the early (heterolactic) and late (homolactic) stages. Lytic phages active against Lb. plantarum were isolated for up to 60 days after the start of fermentation, when the pH was below 4.0. Interestingly, the host-range data revealed that some phages were capable of attacking more than one

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

847

species. Genome sequence analyses have been done for phages from both cucumber and sauerkraut fermentations (58, 59). Genomic analysis of a sauerkraut phage active against L. mesenteroides has revealed a similar pattern of genome organization to sequenced dairy phages, but phage protein sequences had little similarity to dairy phages (58). The impact of phages on fermentation ecology remains unclear, and this is an area ripe for further research.

Biochemistry of vegetable fermentations Fermentation is by definition an anaerobic process. During the fermentation of cucumbers, cabbage, and olives, glucose and fructose are converted to lactic acid, acetic acid, ethanol, and CO2 by LAB and yeasts. The primary pathway for homofermentative LAB involves the breakdown of one six-carbon sugar (glucose) to give two three-carbon lactic acid molecules. Heterofermentative organisms use a more complex metabolism. Glucose is initially converted to CO2 and a five-carbon sugar phosphate, which is further degraded to lactic acid and a two-carbon compound, ethanol or acetic acid. The details of these metabolic pathways have been previously reported (44). Here we shall focus on the biochemical aspects of vegetable fermentation that relate to product quality. There is continuing research interest in fermentation and storage of vegetables, particularly cucumbers, with reduced salt. Chloride waste from vegetable fermentations could be greatly reduced if the salt required for fermentation and storage could be reduced sufficiently to eliminate the need for a desalting step prior to conversion into final products. The relationship between salt type and concentration has been investigated. Lu et al. (62) studied the effects of replacement of NaCl with different anions and cations on the sugar fermentation in cucumber juice. Interestingly, fructose was deter­ mined to be the preferred sugar for Lb. plantarum, as more fructose than glucose was fermented in almost every ­experiment. Sugar utilization decreased as cation or anion concentrations increased with the addition of different salts (62). Many of the volatile components in cucumbers fermented with Lb. plantarum in 2% NaCl were identified by Zhou and McFeeters (108). Thirty-seven volatile compounds were identified, although for most there was little change as a result of fermentation. The most notable effect of fermentation on cucumber volatiles was the inhibition of production of (E, Z)-2,6-nonadienal and 2-nonenal, the two most important odor impact com-

SMP_Food Microbiology_CH33.indd

847

pounds in fresh cucumbers. Marsili and Miller (65) identified trans- and cis-4-hexenoic acid as the most ­potent odorants that define the characteristic brine aroma of cucumbers fermented commercially in about 6% NaCl. Zhou et al. (109) exposed fermented cucumber slurries with 2% NaCl to oxygen and observed nonenzymatic formation of hexanal plus a series of trans unsaturated aldehydes with five to eight carbon atoms that correlated with the development of oxidized odor intensity of the fermented cucumber tissue. Calcium ­disodium EDTA at a concentration of at least 100 µg/ml protects nonfermented pickles against lipid oxidation and bleaching of pigments in the presence of light (10). However, there was some reduction in firmness retention in pickles when this compound was used. Retention of firmness is a key quality issue in the fermentation and storage of cucumbers and peppers. It has not been possible to assure the retention, in cucumbers fermented in reduced salt, of firmness equivalent to that which can be achieved by fermenting in 6% NaCl and storage in 6% or greater NaCl concentrations (34). However, in the past several years there has been increased understanding of cucumber tissue softening. Fleming et al. (39) determined that calcium is beneficial in maintaining the firmness of fermented cucumbers. Nonenzymatic softening of blanched, acidified cucumber tissue was found to follow first-order kinetics (73). This kinetic behavior made it possible to determine the entropy and enthalpy of activation for nonenzymatic softening of cucumbers, even though the chemical reactions responsible for softening were not known. Both the enthalpy and entropy of activation were high at pH 3.0 in the presence of 1.5 M NaCl. Calcium inhibited cucumber softening because it reduced the entropy of activation so much that the overall free energy of activation was reduced (74). This thermodynamic behavior is more like that which occurs when polymers change conformation, such as occurs in protein denaturation. It is very different from the characteristics observed for acid hydrolysis of pectin (54). Krall and McFeeters (54) found that the rate of acid hydrolysis of pectin was too slow to be the cause of nonenzymatic cucumber tissue softening. McFeeters et al. (69) determined the combined effects of temperature and salt and calcium concentrations on the rate of softening of fermented cucumber tissue. The kinetics of softening for fermented cucumbers did not follow a simple first-order reaction. As with many other plant tissues, cucumbers contain enzymes that can degrade components of the plant cell walls, which may result in changes in texture. Pectinesterase, exopolygalacturonase, and endopolygalacturonase activities have been found in cucumbers

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

848 (4, 70, 91). Pectinesterase removes methyl groups from pectin when cucumbers are fermented or acidified (47, 68, 99). However, it has not been determined if enzymatic hydrolysis of pectin by cucumber polygalacturonases is a significant factor in the softening of fermented cucumbers. Commercially important enzymatic softening of fermented cucumbers has been associated with the introduction of fungal polygalacturonases into fermentation tanks, particularly on flowers attached to small cucumbers (3). Buescher and Burgin (9) developed a sensitive diffusion plate assay to measure polygalacturonase activity in fermentation brines and determined that an alumino-silicate clay can adsorb and remove polygalacturonase activity from fermentation brines that are recycled. In addition to enzymes that degrade pectin, enzymes that may degrade other cell wall polysaccharides in cucumbers have been investigated to a very limited extent. Meurer and Gierschner (78) reported endo-β-1,4­gluconase activity in the cucumber that is inactivated below pH 4.8 and an endoglucomannan-splitting enzyme that retains activity down to pH 4.0 but is inactivated during fermentation. They detected six enzymes in fresh cucumbers that hydrolyze p-nitrophenylglycosides of α-d-galactose, β-d-galactose, β-d-glucose, β-d-xylose, α-d-­mannose, and α-l-arabinose, which were inactivated during fermentation. Enzymes capable of hydrolyzing these synthetic substrates are common in plants; e.g., most of the same enzymatic activities have been found in pears (25), olives (45), and Semillon grapes (98). Maruvada (66) detected the same p-­nitrophenylglycosidases ­observed by Meurer and Gierschner (78) in cucumbers. She determined that the activity of all of these enzymes declined to nondetectable levels during the first week of fermen­ tation in 2% NaCl brines. Fleming et al. (34) combined blanching of fresh cucumbers to partially inactivate enzymes, calcium addition, and rapid fermentation with a malolactic-negative Lb. plantarum strain in order to ferment cucumbers and maintain desirable texture with the NaCl concentration reduced to 4%. Olives are fermented in 5 to 6% salt concentrations, similar to cucumbers (77). As noted above, before fermentation olives are treated with NaOH to remove components that inhibit growth of fermentative bacteria but not yeasts. Fleming et al. (40) and Ruiz Barba et al. (94) suggested that oleuropein and degradation products of oleuropein are the primary components responsible for inhibition of bacteria when olives are not treated with NaOH. However, Medina et al. (76) determined that the dialdehydic form of decarboxymethyl elenolic acid and one isomer of oleoside 11-methyl ester were inhibitory to bacteria instead of oleuropein. Gordal olives,

SMP_Food Microbiology_CH33.indd

848

which have low levels of these phenolic compounds, could be fermented without NaOH treatment, whereas Manzanilla olives, which have high levels of phenolic compounds, could not (77). Cabbage contains a group of glucosinolates that have received considerable attention in recent years due to potential health benefits of some of the degradation products formed during the processing of cabbage. A recent report indicates that a high intake of sauerkraut correlates with a reduced incidence of breast cancer in women (95), although others have been concerned about potentially toxic compounds derived from glucosinolates (20). Tolonen et al. (100) determined isothiocyanates and allyl cyanide to be the predominant ­degradation products of glucosinolates in sauerkraut fermented with and without a starter culture. Only minor amounts of goitrin, a toxic compound, and the beneficial phytochemical sulforaphane nitrile were found in sauerkraut. Tolonen et al. (101) found greater amounts of glucosinolate degradation products in sauerkraut fermented with Lactobacillus sakei than in sauerkraut made with starter cultures consisting of other LAB. Ciska and Pathak (16) reported that ascorbigen, a compound formed from the reaction of a degradation product of indole glucosinolate (glucobrassicin) and ascorbic acid, is the dominant glucosinolate degradation product in sauerkraut. Glucoraphinin present in fresh cabbage was converted to sulforphorane during fermentation, although sulforphorane was a relatively minor glucosinolate degradation product in fermented cabbage. There has been some concern about the formation of biogenic amines in sauerkraut. Kalač et al. (50) reported that tyramine was formed in sauerkraut stored for up to 12 months. Only trace levels of histamine, tryptamine, and spermine were detected. These results were confirmed in a survey of vegetable products (80) in which the concentration of tyramine was found to be 4.9 mg/100 g in canned sauerkraut, virtually the same concentration reported by Kalač et al. (50). These biogenic amine levels would not represent a health risk, with the possible exception of individuals taking medications containing monamine oxidase inhibitors.

GENOMICS OF LAB IN VEGETABLE FERMENTATIONS With the advent of whole-genome sequencing, it has become apparent that the LAB present in vegetable fermentations have relatively small genomes compared with many other mesophilic organisms. It is known that LAB are nutritionally fastidious, and rich media (MRS agar) is used for cultivation (36). Since the sequencing of

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

849

the first LAB, Lactococcus lactis (5), genome data have rapidly accumulated (51, 64). Phylogenetic analysis of sequenced genomes indicates that LAB evolved from ancestral anaerobic organisms with a much more extensive metabolic capacity by a process of gene loss. A phylogenetic tree for LAB has been constructed using several methods, including concatenated protein sequences predicted from the sequenced genomes (64). Because LAB grow in a nutritionally rich environment, there may be a limited need to manufacture the complex biomolecules needed for cell growth. From the available genome data it is evident that the process of gene loss may be ongoing, and hundreds of apparently inactive “pseudogenes” are present in some of the sequenced genomes (64). It is

also clear that LAB have acquired a number of genes, including genes used for transport of nutrients, by horizontal gene transfer. Multilocus sequence typing for a number of Lb. plantarum strains has indicated that recombination events have had a prominent role in generating genetic heterogeneity in these highly commercialized bacteria (22). About 50% of the open reading frames (ORFs) identified in the 2- to 3-Mb genomes of LAB have been assigned putative functions related to the metabolism or transport of amino acids, carbohydrates, and inorganic ions (Table 33.2). About 7% of these ORFs are dedicated to energy production by fermentation of sugars. The Lb. plantarum sequence contains a large

Table 33.2  Distribution of ORFs among Clusters of Orthologous Groups (COG) functional categoriesa Lactobacillus brevis ATCC 367

Leuconostoc ­ esenteroides m ATCC 8293

Pediococcus ­pentosaceus ATCC 25745

Lactobacillus plantarum WCFS1

Energy production and conversion

6.8

6.0

6.7

8.8

Cell cycle control, cell division, ­chromosome partitioning

2.4

2.2

1.8

2.2

Amino acid transport and metabolism

13.6

17.3

10.4

16.1

Nucleotide transport and metabolism

7.2

8.3

8.8

7.5

COG categories

Carbohydrate transport and metabolism

14.4

12.8

17.0

21.9

Coenzyme transport and metabolism

5.5

7.1

4.9

7.8

Lipid transport and metabolism

3.8

3.9

3.9

4.3

Translation, ribosomal structure, and biogenesis

17.4

15.9

17.5

15.4

Transcription

15.7

11.3

14.0

20.2

Replication, recombination, and repair

16.2

11.7

13.2

12.8

Cell wall/membrane/envelope biogenesis

10.8

9.4

11.8

13.2

Cell motility

0.2

0.1

0.3

0.3

Posttranslational modification, protein turnover, chaperones

5.4

5.5

5.2

4.9

Inorganic ion transport and metabolism

10.6

9.0

8.5

11.2

3.1

2.3

1.4

2.6

General function prediction only

32.2

26.9

28.4

38.6

Function unknown

Secondary metabolite biosynthesis, ­transport, and catabolism

20.5

16.4

18.1

20.0

Signal transduction mechanisms

7.8

6.2

6.5

8.4

Intracellular trafficking, secretion, and vesicular transport

2.4

2.2

2.9

2.1

Defense mechanisms

4.1

3.0

4.1

5.4

Extracellular structures

0.0

0.1

0.1

0.3

Other ORF without a category

0.0

0.0

0.0

0.0

a

Values are in percentages.

SMP_Food Microbiology_CH33.indd

849

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

850 number of putative phosphotransferase system (PTS) genes, encoding up to 25 complete transport complexes and several incomplete complexes (53). Pediococcus pentosaceus has a reduced number of putative PTS genes compared with Lb. plantarum. L. mesenteroides apparently contains five functional PTSs and several incomplete systems. Predicted gene products related to pyruvic acid catabolism in L. mesenteroides, Lb. plantarum, and P. pentosaceus are listed in Table 33.3. Genes for lactate dehydrogenases, a malolactic enzyme, and an incomplete citric acid cycle are in all three bacterial genome sequences. The sequence data reveal that Lb. plantarum and L. mesenteroides contain more pyruvate-dissipating enzymes than P. pentosaceus, which may provide a metabolic advantage. This may contribute to the predominant role of Lb. plantarum and L. mesenteroides over P. pentosaceus in a variety of vegetable fermentations. The genome sequences of L. mesenteroides, Lb. plantarum, and P. pentosaceus contain multiple copies of the rRNA clusters, which display minimal or no polymorphism within a given genome. The 16S, 5S, and 23S rRNA sequence copies within each bacterium have 99 or 100% identity over their entire length. With the exception of the cysteine tRNA, most tRNAs are present in these genomes in multiple copies. The L. mesenteroides genome sequence contains four putative rRNA clusters, which are located close to the origin of replication. The Lb. plantarum and P. pentosaceus genome sequences contain five putative rRNA clusters distributed around the genome. The number and location of the rRNA operons may influence reproductive fitness (52), with more operons indicating greater fitness; however, the significance

of the genome arrangement of rRNA operons remains to be determined. LAB isolated from vegetable fermentations frequently contain plasmids. Lb. plantarum WCFS1 contains two small plasmids of approximately 2,000 bp and a larger plasmid of about 36,000 bp (53). Putative functions assigned to the proteins encoded by ORFs in these plasmids include conjugal plasmid transfer. Similarly, L. mesenteroides ATCC 8293 contains a plasmid of ­approximately 37,000 bp, which apparently encodes several mobile genetic elements. Plasmid-borne genes encoding proteins involved in bacteriocin production, lactose utilization, and citric acid utilization have been isolated from several Leuconostoc species; however, none of these functions appears to be present in the ATCC 8293 plasmid. The overall G+C content of these plasmids tends to be lower than the 40% chromosomal G+C content, suggesting that these plasmids were acquired from nonhomologous sources. Most of the ORFs identified in the sequenced plasmids have no assigned functions. Further research remains to be done to exploit the potential for genetic manipulation by conjugative transfer of genes in LAB from vegetable fermentations, although this technology has been extensively exploited with Lactococcus spp. from dairy fermentations (41).

Concluding Remarks As we advance into the second century of research on fermented and acidified foods, researchers are building on the solid foundation laid by those noted above, as well as others. Recent research includes mathematical modeling of bacterial growth and competition (24), the

Table 33.3  Pyruvic acid-dissipating enzymes present in the predominant LAB in fermented vegetables

Pyruvate catabolism-related enzyme(s) Pyruvate oxidase Pyruvate dehydrogenase Pyruvate formate lyase Acetolactate synthase Lactate dehydrogenases Hydroxyisocaproate dehydrogenase Malic enzyme (EC 1.1.1.38) Malate dehydrogenase (EC 1.1.1.40) Pyruvate kinase Oxaloacetate decarboxylate Pyruvate carboxylase a

Lactobacillus brevis ATCC 367

Leuconostoc mesenteroides ATCC 8293

Pediococcus pentosaceus ATCC 25745

Lactobacillus plantarum WCFS1

Xa X –b X X – – X X – –

X X X X X X – X X – X

X X X – X – – X X – X

X X X X X X X – X – X

X, enzyme is present. –, enzyme is absent.

b

SMP_Food Microbiology_CH33.indd

850

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

851

molecular ecology of vegetable fermentations (12, 15, 82, 84, 90), closed-tank fermentation technology to reduce salt waste (34), the use of clays to remove softening enzymes from recycled brines (10), studies on the sensory perception of pickled vegetable products (48), and studies on the safety of acidified foods (7, 8). Preservation of vegetables by fermentation or by the addition of acids without fermentation (21, 46, 87, 88) is based upon achieving acid concentrations to decrease pH values low enough (pH 3 to 4) to prevent growth of most microorganisms. Hence, sour flavor is important and a characteristic component of the flavor profile of the vegetable products produced by these processes. An understanding of the biological mechanisms of sour taste perception has lagged behind progress on the other four basic tastes. However, there has been recent progress regarding both the physiology (13) and chemistry of sour taste (18). The recent advances in genomics, molecular microbial ecology, analytical biochemistry, plant breeding, and fermentation technology reveal a bright future for vegetable fermentation science and applications, enhancements of existing products, and new processing techniques. Principal areas for the application of this technology may include the development of commercial low-salt fermentations and nonfermentation preservation methods that will reduce or eliminate salt wastes. Current industrial fermentation practices are in large part based on traditional practices that have been adapted to a larger scale. The development of low-salt fermentations and the storage of fermented vegetables for commercial use present significant technological hurdles, including the potential need for starter cultures (and the impact of bacteriophages on starter cultures) and for new product-handling equipment. Future products may have nutritional properties superior to those of current products and incorporate nontraditional vegetables. The reasons for developing these products will be the same as they have been for centuries. Fermented vegetable products are microbiologically safe, nutritious, and flavorful; have appealing sensory characteristics; and can be conveniently stored for extended periods without refrigeration. Paper no. FSR10-35 of the Journal Series of the Food, Bio­ processing and Nutrition Sciences Department, North Carolina State University, Raleigh, NC 27695-7624. Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture or North Carolina Agricultural Research Service, nor does it imply approval to the exclusion of other products that may be suitable.

SMP_Food Microbiology_CH33.indd

851

References   1. Barrangou, R., S. S. Yoon, F. Breidt, Jr., H. P. Fleming, and T. R. Klaenhammer. 2002. Characterization of six Leuconostoc fallax bacteriophages isolated from an industrial sauerkraut fermentation. Appl. Environ. Microbiol. 68:5452–5458.   2. Bell, T. A., and J. L. Etchells. 1952. Sugar and acid tolerance of spoilage yeasts from sweet-cucumber pickles. Food Technol. 6:468–472.   3. Bell, T. A., J. L. Etchells, and I. D. Jones. 1950. Softening of commercial cucumber salt-stock in relation to poly­ galacturonase activity. Food Technol. 4:157–163.   4. Bell, T. A., J. L. Etchells, and I. D. Jones. 1951. Pectinesterase in the cucumber. Arch. Biochem. Biophys. 31:431–441.   5. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach, S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res. 11:731–753.   6. Breidt, F., and H. P. Fleming. 1992. Competitive growth of genetically marked malolactic-deficient Lactobacillus plantarum in cucumber fermentations. Appl. Environ. Microbiol. 58:3845–3849.   7. Breidt, F., J. S. Hayes, and R. F. McFeeters. 2007. Determination of 5-log reduction times for food pathogens in acidified cucumbers during storage at 10 and 25°C. J. Food Prot. 70:2638–2641.   8. Breidt, F., Jr., J. S. Hayes, J. A. Osborne, and R. F. McFeeters. 2005. Determination of 5-log pathogen reduction times for heat-processed, acidified foods. J. Food Prot. 68:305–310.   9. Buescher, R. W., and C. Burgin. 1992. Diffusion plate assay for measurement of polygalacturonase activities in pickle brines. J. Food Biochem. 16:59–68.   10. Buescher, R., and C. Hamilton. 2002. Adsorption of polygalacturonase from recycled cucumber pickle brines by Pure-Flo B80 clay. J. Food Biochem. 26:153–156.   11. Caplice, E., and G. F. Fitzgerald. 1999. Food fermentations: role of microorganisms in food production and preservation. Int. J. Food Microbiol. 50:131–149.   12. Chang, H. W., K. H. Kim, Y. D. Nam, S. W. Roh, M. S. Kim, C. O. Jeon, H. M. Oh, and J. W. Bae. 2008. Analysis of yeast and archaeal population dynamics in kimchi using denaturing gradient gel electrophoresis. Int. J. Food Microbiol. 126:159–166.   13. Chang, R. B., H. Waters, and E. R. Liman. 2010. A proton current drives action potentials in genetically identified sour taste cell. Proc. Natl. Acad. Sci. USA 107:22320–22325. www.pnas.org/cgi/doi/10.1073/pnas.1013664107.   14. Cheigh, H. S., and K. Y. Park. 1994. Biochemical, micro­ biological and nutritional aspects of kimchi. Crit. Rev. Food Sci. Nutr. 34:175–203.   15. Chin, H. S., F. Breidt, H. P. Fleming, W. C. Shin, and S. S. Yoon. 2006. Identification of predominant bacterial isolates from the fermenting kimchi using ITS-PCR and partial 16S rDNA sequence analyses. J. Microbiol. Biotechnol. 16:68–76.

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

852   16. Ciska, E., and D. R. Pathak. 2004. Glucosinolate derivatives in stored fermented cabbage. J. Agric. Food Chem. 52:7938–7943.   17. Costilow, R. N., C. L. Bedford, D. Mingus, and D. Black. 1977. Purging of natural salt-stock pickle fermentations to reduce bloater damage. J. Food Sci. 42:234–240.   18. Da Conceicao Neta, E. R., S. D. Johanningsmeier, M. A. Drake, and R. F. McFeeters. 2007. A chemical basis for sour taste perception of acid solutions and fresh-pack dill pickles. J. Food Sci. 72:S352–S359.   19. Daeschel, M. A., R. F. McFeeters, H. P. Fleming, T. R. Klaenhammer, and R. B. Sanozky. 1984. Mutation and selection of Lactobacillus plantarum strains that do not produce carbon dioxide from malate. Appl. Environ. Microbiol. 47:419–420.   20. Daxenbichler, M. E., C. H. VanEtten, and P. H. Williams. 1980. Glucosinolate products in commercial sauerkraut. J. Agric. Food Chem. 28:809–811.   21. De Castro, A., P. García, C. Romero, M. Brenes, and A. Garrido. 2007. Industrial implementation of black ripe olive storage under acid conditions. J. Food Eng. 80:1206–1212.   22. de las Rivas, B., A. Marcobal, and R. Muñoz. 2006. Development of a multilocus sequence typing method for analysis of Lactobacillus plantarum strains. Micro­ biology 152:85–93.   23. DeVuyst, L., and E. J. Vandamme. 1994. Antimicrobial potential of lactic acid bacteria, p. 91–142. In L. DeVuyst and E. J. Vandamme (ed.), Bacteriocins of Lactic acid Bacteria. Blackie Academic and Professional, London, United Kingdom.   24. Dougherty, D. P., F. Breidt, Jr., R. F. McFeeters, and S. R. Lubkin. 2002. Energy-based dynamic model for variable temperature batch fermentation by Lactococcus lactis. Appl. Environ. Microbiol. 68:2468–2478.   25. El-Rayah-Ahmed, A., and J. M. Labavitch. 1980. Cell wall metabolism in ripening fruit. II. Changes in carbohydrate-degrading enzymes in ripening ‘Bartlett’ pears. Plant Physiol. 65:1014–1016.   26. Etchells, J. L. 1950. Salt-tolerant yeasts from commercial cucumber brines. Texas Rep. Biol. Med. 8:103–104.   27. Etchells, J. L., T. A. Bell, H. P. Fleming, R. E. Kelling, and R. L. Thompson. 1973. Suggested procedure for the controlled fermentation of commercially brined pickling cucumbers—the use of starter cultures and reduction of carbon dioxide accumulation. Pickle Pak Sci. 3:4–14.   28. Etchells, J. L., T. A. Bell, and W. R. Moore, Jr. 1976. Refrigerated dill pickles, questions and answers. Pickle Pak Sci. 5:1–20.   29. Etchells, J. L., A. F. Borg, and T. A. Bell. 1961. Influence of sorbic acid on populations and species of yeasts occurring in cucumber fermentations. Appl. Microbiol. 9:139–144.   30. Etchells, J. L., and I. D. Jones. 1942. Pasteurization of pickle products. Fruit Prod. 21:330–332.   31. Fernández, A. G., P. G. Garcia, and M. B. Balbuena. 1995. Olive fermentations, p. 593–627. In H.-J. Rehm and G. Reed (ed.), Biotechnology. VCH Publishers, New York, NY.

SMP_Food Microbiology_CH33.indd

852

  32. Fleming, H. P., M. A. Daeschel, R. F. McFeeters, and M. D. Pierson. 1989. Butyric acid spoilage of fermented cucumbers. J. Food Sci. 54:636–639.   33. Fleming, H. P., J. L. Etchells, R. L. Thompson, and T. A. Bell. 1975. Purging of CO2 from cucumber brines to reduce bloater damage. J. Food Sci. 40:1304–1310.   34. Fleming, H. P., E. G. Humphries, R. L. Thompson, and R. F. McFeeters. 2002. Bag-in-box technology: storage stability of process-ready, fermented cucumbers. Pickle Pak Sci. 8:14–18.   35. Fleming, H. P., K.-H. Kyung, and F. Breidt, Jr. 1995. Vegetable fermentations, p. 629–661. In H.-J. Rehm and G. Reed (ed.), Biotechnology. VCH Publishers, New York, NY.   36. Fleming, H. P., R. F. McFeeters, and F. Breidt. 2001. Fermented and acidified vegetables, p. 521–532. In F. P. Downes and K. Ito (ed.), Compendium of Methods for the Microbiological Examination of Foods, 4th ed. American Public Health Association, Washington, DC.   37. Fleming, H. P., R. F. McFeeters, and M. A. Daeschel. 1992. Fermented and acidified vegetables, p. 929– 952. In C. Vanderzant and D. F. Splittstoesser (ed.), Compendium of Methods for the Microbiological Examination of Foods, 3rd ed. American Public Health Association, Washington, DC.   38. Fleming, H. P., R. F. McFeeters, and E. G. Humphries. 1988. A fermentor for study of sauerkraut fermentation. Biotechnol. Bioeng. 31:189–197.   39. Fleming, H. P., R. L. Thompson, T. A. Bell, and L. H. Hontz. 1978. Controlled fermentation of sliced cucumbers. J. Food Sci. 43:888–891.   40. Fleming, H. P., W. M. Walter, Jr., and J. L. Etchells. 1973. Antimicrobial properties of oleuropein and products of its hydrolysis from green olives. Appl. Microbiol. 26:777–782.   41. Gasson, M. 1990. In vivo genetic systems in lactic acid bacteria. FEMS Microbiol. Rev. 87:43–60.   42. Gates, K., and R. N. Costilow. 1981. Factors influencing softening of salt-stock pickles in air-purged fermentations. J. Food Sci. 46:274–277.   43. Gómez, A. H. S., P. G. García, and L. R. Navarro. 2006. Trends in table olive production, elaboration of table olives. Grasas y Aceites 57:86–94.   44. Gottschalk, G. 1986. Bacterial Metabolism, 2nd ed., p. 208–220. Springer-Verlag, New York, NY.   45. Heredia, A., R. Guillén, A. Jiménez, and J. FernándezBolanos. 1993. Activity of glycosidases during development and ripening of olive fruit. Z. Lebensm. Unters. Forsch. 196:147–151.   46. Holzapfel, W. H., R. Geisen, and U. Schillinger. 1995. Biological preservation of foods with reference to protective cultures, bacteriocins, and food-grade enzymes. Int. J. Food Microbiol. 24:343–362.   47. Hudson, J. M., and R. W. Buescher. 1986. Relationship between degree of pectin methylation and tissue firmness of cucumber pickles. J. Food Sci. 51:138–140, 149.   48. Johanningsmeier, S. D., H. P. Fleming, R. L. Thompson, and R. F. McFeeters. 2005. Chemical and sensory properties of sauerkraut produced with Leuconostoc mesen-

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

853

teroides starter cultures of differing malolactic phenotypes. J. Food Sci. 70:S343–S349.   49. Jones, I. D., J. L. Etchells, M. K. Veldhuis, and O. Veerhoff. 1941. Pasteurization of genuine dill pickles. Fruit Prod. 20:304–305, 316, 325.   50. Kalacˇ, P., J. Spicka, M. Krizek, and T. Pelikanova. 2000. Changes in biogenic amine concentrations during sauerkraut storage. Food Chem. 69:309–314.   51. Klaenhammer, T., E. Altermann, F. Arigoni, A. Bolotin, F. Breidt, J. Broadbent, R. Cano, S. Chaillou, J. Deutscher, M. Gasson, M. van de Guchte, J. Guzzo, A. Hartke, T. Hawkins, P. Hols, R. Hutkins, M. Kleerebezem, J. Kok, O. Kuipers, M. Lubbers, E. Maguin, L. McKay, D. Mills, A. Nauta, R. Overbeek, H. Pel, D. Pridmore, M. Saier, D. van Sinderen, A. Sorokin, J. Steele, D. O’Sullivan, W. de Vos, B. Weimer, M. Zagorec, and R. Siezen. 2002. Discovering lactic acid bacteria by genomics. Antonie van Leeuwenhoek 82:29–58.   52. Klappenbach, J., J. Dunbar, and T. Schmidt. 2000. rRNA operon copy number reflects ecological strategies of bacteria. Appl. Environ. Microbiol. 66:1328–1333.   53. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbring, M. W. E. J. Fiers, W. Stickema, R. M. K. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. M. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. USA 100:1990–1995.   54. Krall, S. M., and R. F. McFeeters. 1998. Pectin hydrolysis: effect of temperature, degree of methylation, pH, and calcium on hydrolysis rates. J. Agric. Food Chem. 46:1311–1315.   55. Lee, C. H. 1994. Importance of lactic acid bacteria in nondairy food fermentation, p. 8–25. In C. H. Lee, J. AdlerNissen, and G. Barwald (ed.), Lactic Acid Fermentation of Non-dairy Food and Beverages. HarnLimWon, Seoul, South Korea.   56. Lee, C. H. 2001. Fermentation Technology in Korea. Korea University Press, Seoul, South Korea.   57. Lee, C. H. 2009. Food biotechnology, p. 85–114. In G. Campbell-Platt (ed.), Food Science and Technology. Wiley-Blackwell, West Sussex, United Kingdom.   58. Lu, Z., E. Altermann, F. Breidt, and S. Kozyavkin. 2010. Sequence analysis of Leuconostoc mesenteroides bacteriophage Φ1-A4 isolated from industrial vegetable fermentation. Appl. Environ. Microbiol. 76:1955–1966.   59. Lu, Z., E. Altermann, F. Breidt, Jr., P. Predki, H. P. Fleming, and T. R. Klaenhammer. 2005. Sequence analysis of the Lactobacillus plantarum bacteriophage FJL-1. Gene 348:45–54.   60. Lu, Z., F. Breidt, Jr., H. P. Fleming, E. Altermann, and T. R. Klaenhammer. 2003. Isolation and characterization of a Lactobacillus plantarum bacteriophage, FJL-1, from a cucumber fermentation. Int. J. Food Microbiol. 84:225–235.   61. Lu, Z., F. Breidt, Jr., V. Plengvidhya, and H. P. Fleming. 2003. Bacteriophage ecology in commercial

SMP_Food Microbiology_CH33.indd

853

sauerkraut fermentations. Appl. Environ. Microbiol. 69:3192–3202.   62. Lu, Z., H. P. Fleming, and R. F. McFeeters. 2002. Effects of fruit size on fresh cucumber composition and the chemical and physical consequences of fermentation. J. Food Sci. 67:2934–2939.   63. Mah, J. H., Y. H. Chang, and H. J. Hwang. 2008. Paenibacillus tyraminigenes sp. nov. isolated from Myeolchi-jeotgal, a traditional Korean salted and fermented anchovy. Int. J. Food Microbiol. 127:209–214.   64. Makarova, K., A. Slesarev, Y. Wolf, A. Sorokin, B. Mirkin, E. Koonin, A. Pavlov, N. Pavlova, V. Karamychev, N. Polouchine, V. Shakhova, I. Grigoriev, Y. Lou, D. Rohksar, S. Lucas, K. Huang, D. M. Goodstein, T. Hawkins, V. Plengvidhya, D. Welker, J. Hughes, Y. Goh, A. Benson, K. Baldwin, J.-H. Lee, I. Díaz-Muñiz, B. Dosti, V. Smeianov, W. Wechter, R. Barabote, G. Lorca, E. Altermann, R. Barrangou, B. Ganesan, Y. Xie, H. Rawsthorne, D. Tamir, C. Parker, F. Breidt, Jr., J. Broadbent, R. Hutkins, D. O’Sullivan, J. Steele, G. Unlu, M. Saier, T. Klaenhammer, P. Richardson, S. Kozyavkin, B. Weimer, and D. Mills. 2006. Comparative genomics of the lactic acid bacteria. Proc. Natl. Acad. Sci. USA 103:15611–15616.   65. Marsili, R. T., and N. Miller. 2000. Determination of major aroma impact compounds in fermented cucumbers by solid-phase microextraction-gas chromatography-mass spectrometry-olfactometry detection. J. Chromatogr. Sci. 38:307–314.   66. Maruvada, R. 2005. Evaluation of the importance of enzymatic and non-enzymatic softening in low salt cucumber fermentations. M.S. thesis. North Carolina State University, Raleigh.   67. McDonald, L. C., H. P. Fleming, and H. M. Hassan. 1990. Acid tolerance of Leuconostoc mesenteroides and Lactobacillus plantarum. Appl. Environ. Microbiol. 56: 2120–2124.   68. McFeeters, R. F., and S. A. Armstrong. 1984. Measure­ ment of pectin methylation in plant cell walls. Anal. Biochem. 139:212–217.   69. McFeeters, R. F., M. Balbuena, M. Brenes, and H. P. Fleming. 1995. Softening rates of fermented cucumber tissue: effects of pH, calcium, and temperature. J. Food Sci. 60:786–788.   70. McFeeters, R. F., T. A. Bell, and H. P. Fleming. 1980. An endo-polygalacturonase in cucumber fruit. J. Food Biochem. 4:1–16.   71. McFeeters, R. F., and K.-H. Chen. 1986. Utilization of electron acceptors for anaerobic mannitol metabolism by Lactobacillus plantarum. Compounds which serve as electron acceptors. Food Microbiol. 3:73–81.   72. McFeeters, R. F., W. Coon, M. P. Palnitkar, M. Velting, and N. Fehringer. 1978. Reuse of Fermentation Brines in the Cucumber Pickling Industry, p. 1–115. EPA600/2-78-207. U.S. Environmental Protection Agency, Washington, DC.   73. McFeeters, R. F., and H. P. Fleming. 1989. Inhibition of cucumber tissue softening in acid brines by multivalent cations: inadequacy of the pectin “egg box”

Manila Typesetting Company

11/08/2012 07:14AM

Fermentations and Beneficial Microorganisms

854 model to explain textural effects. J. Agric. Food Chem. 37:1053–1059.   74. McFeeters, R. F., and H. P. Fleming. 1990. Effect of calcium ions on the thermodynamics of cucumber tissue softening. J. Food Sci. 55:446–449.   75. McFeeters, R. F., H. P. Fleming, and R. L. Thompson. 1982. Malic acid as a source of carbon dioxide in cucumber fermentations. J. Food Sci. 47:1862–1865.   76. Medina, E., M. Brenes, C. Romero, A. García, and A. de Castro. 2007. Main antimicrobial compounds in table olives. J. Agric. Food Chem. 55:9817–9823.   77. Medina, E., C. Gori, M. Servili, A. de Castro, C. Romero, and M. Brenes. 2010. Main variables affecting the lactic acid fermentation of table olives. Int. J. Food Sci. Technol. 45:1291–1296.   78. Meurer, P., and K. Gierschner. 1992. Occurrence and effect of indigenous and eventual microbial enzymes in lactic acid fermented vegetables. Acta Aliment. 21:171–188.   79. Mheen, T. I., and T. W. Kwon. 1984. Effect of temperature and salt concentration on kimchi fermentation. Korean J. Food Sci. Technol. 16:443–450.   80. Moret, S., D. Smela, T. Populin, and L. S. Conte. 2005. A survey on free biogenic amine content of fresh and preserved vegetables. Food Chem. 89:355–361.   81. Mudgal, P., F. Breidt, Jr., S. R. Lubkin, and K. P. Sandeep. 2006. Quantifying the significance of phage attack on starter cultures: a mechanistic model for population dynamics of phage and their hosts isolated from fermenting sauerkraut. Appl. Environ. Microbiol. 72:3908–3915.   82. Nam, Y. D., H. W. Chang, K. H. Kim, S. W. Roh, and J. W. Bae. 2009. Metatranscriptome analysis of lactic acid bacteria during kimchi fermentation with genome-probing microarrays. Int. J. Food Microbiol. 130:140–146.   83. Nguyen-the, C., and F. Carlin. 1994. The microbiology of minimally processed fresh fruits and vegetables. Crit. Rev. Food Sci. Nutr. 34:371–401.   84. Park, J. M., J. H. Shin, D. W. Lee, J. C. Song, H. J. Suh, U. J. Chang, and J. M. Kim. 2010. Identification of the lactic acid bacteria in kimchi according to initial and over­ripened fermentation using PCR and 16S rRNA gene sequence analysis. Food Sci. Biotechnol. 19:541–546.   85. Pederson, C. S., and M. N. Albury. 1954. The influence of salt and temperature on the microflora of sauerkraut fermentation. Food Technol. 8:1–5.   86. Pederson, C. S., and M. N. Albury. 1969. The Sauerkraut Fermentation. Technical Bulletin no. 824. N.Y. State Agricultural Experiment Station, Geneva, NY.   87. Pérez-Díaz, I. M., and R. F. McFeeters. 2008. Microbiological preservation of cucumbers for bulk storage by the use of acetic acid and food preservatives. J. Food Sci. 73:M287–M291.   88. Pérez-Díaz, I. M., and R. F. McFeeters. 2010. Preservation of acidified cucumbers with a natural preservative combination of fumaric acid and allyl isothiocyanate that target lactic acid bacteria and yeasts. J. Food Sci. 75: M204–M208.

SMP_Food Microbiology_CH33.indd

854

  89. Plastourgos, S., and R. H. Vaughn. 1957. Species of Propionibacterium associated with zapatera spoilage of olives. Appl. Microbiol. 5:267–271.   90. Plengvidhya, V., F. Breidt, Z. Lu, and H. P. Fleming. 2007. DNA fingerprinting of lactic acid bacteria in sauerkraut fermentations. Appl. Environ. Microbiol. 73:7697–7702.   91. Pressey, R., and J. K. Avants. 1975. Cucumber polygalacturonase. J. Food Sci. 40:937–939.   92. Preuss, L. M., W. H. Peterson, and E. B. Fred. 1928. Gas production in the making of sauerkraut. J. Ind. Eng. Chem. 20:1187–1190.   93. Priem, L. A., W. H. Peterson, and E. B. Fred. 1927. Studies of commercial sauerkraut with special reference to changes in the bacterial flora during fermentation at low temperatures. J. Agric. Res. 34:79–95.   94. Ruiz Barba, J. L., R. M. Rios Sanchez, C. Fedriani Iriso, J. M. Olias, J. L. Rios, and J. L. Jimenez-Diaz. 1990. Bactericidal effects of phenolic compounds from green olives on Lactobacillus plantarum. Syst. Appl. Microbiol. 13:199–205.   95. Rybaczyk-Pathak, D. 2005. Joint association of high cabbage/sauerkraut intake at 12–13 years of age and adulthood with reduced breast cancer risk in Polish migrant women: results from the U.S. Component of the Polish Women’s Health Study (PWHS), abstr. 3697. Abstr. Am. Assoc. Cancer Res. 4th Annu. Frontiers Cancer Prevention Res., Baltimore, MD, 2 November 2005.   96. Steinkraus, K. H. 1983. Handbook of Indigenous Fermented Foods. Marcel Dekker, New York, NY.   97. Steinkraus, K. H. 1993. Comparison of fermented foods of the East and West, p. 1–12. In C. H. Lee, K. H. Steinkraus, and P. J. A. Reilly (ed.), Fish Fermentation Technology. UNU Press, Tokyo, Japan.   98. Takayanagi, T., T. Okuda, and K. Yokotsuka. 1997. Changes in glycosidase activity in grapes during development. J. Inst. Enol. Viticult. Yamanashi Univ. 32:1–4.   99. Tang, H. C. L., and R. F. McFeeters. 1983. Relationships among cell wall constituents, calcium and texture during cucumber fermentation and storage. J. Food Sci. 48:66–70. 100. Tolonen, M., T., T. Marianne, V. Britta, P. Juha-Matti, K. Hannu, and R. Eeva-Liisa. 2002. Plant-derived biomolecules in fermented cabbage. J. Agric. Food Chem. 50:6798–6803. 101. Tolonen, M., S. Rajaniemi, J.-M. Pihlava, T. Johansson, P. E. J. Saris, and E.-L. Ryhanen. 2004. Formation of nisin, plant-derived biomolecules and antimicrobial activity in starter culture fermentations of sauerkraut. Food Microbiol. 21:167–179. 102. Vaughn, R. H. 1985. The microbiology of vegetable fermentations, p. 49–109. In J. B. Wood (ed.), Microbiology of Fermented Foods, vol. 1. Elsevier Applied Science, Barking, United Kingdom. 103. Whitehead, H. R., and A. G. Cox. 1935. The occurrence of bacteriophages in starter cultures of lactic streptococci. N. Z. J. Sci. Technol. 16:319–320. 104. Yoon, S. S., R. Barrangou-Poueys, F. Breidt, and H. P. Fleming. 2007. Detection and characterization of a lytic

Manila Typesetting Company

11/08/2012 07:14AM

33.  Fermented Vegetables

855

Pediococcus bacteriophage from the fermenting cucumber brine. J. Microbiol. Biotechnol. 17:262–270. 105. Yoon, S. S., R. Barrangou-Poueys, F. Breidt, Jr., T. R. Klaenhammer, and H. P. Fleming. 2002. Isolation and characterization of bacteriophages from fermenting sauerkraut. Appl. Environ. Microbiol. 68:973–976. 106. Yoon, S. S., J. W. Kim, F. Breidt, and H. P. Fleming. 2001. Characterization of a lytic Lactobacillus plantarum bacteriophage and molecular cloning of a lysin gene in Escherichia coli. Int. J. Food Microbiol. 65:63–74.

SMP_Food Microbiology_CH33.indd

855

107. Yoon, S. S., Y. J. Shin, S. Her, and D. H. Oh. 1997. Isolation and characterization of the Lactobacillus plantarum bacteriophage SC921. Korean J. Appl. Microbiol. Biotechnol. 25:96–101. 108. Zhou, A., and R. F. McFeeters. 1998. Volatile compounds in cucumbers fermented in low-salt conditions. J. Agric. Food Chem. 46:2117–2122. 109. Zhou, A., R. F. McFeeters, and H. P. Fleming. 2000. Development of oxidized odor and volatile aldehydes in fermented cucumber tissue exposed to oxygen. J. Agric. Food Chem. 48:193–197.

Manila Typesetting Company

11/08/2012 07:14AM

SMP_Food Microbiology_CH33.indd

856

Manila Typesetting Company

11/08/2012 07:14AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch34

Steven C. Ricke Ok Kyung Koo Jimmy T. Keeton

Fermented Meat, Poultry, and Fish Products

Fermented meat products are defined as meats that are deliberately inoculated during processing to ensure sufficient control of microbial activity to alter the product’s characteristics (9). If fresh meat is not preserved or cured in some manner, it spoils rapidly owing to the growth of indigenous gram-negative bacteria and subsequent putrefaction resulting from their metabolic activities (30). Although some manufacturers still depend upon naturally occurring microflora to ferment meat, most use starter cultures consisting of a single species or multiple species combinations of lactic acid bacteria (LAB) and/or staphylococci that have been selected for metabolic activities especially suited for fermentation in meat ecosystems. Understanding the technological, microbiological, and biochemical processes that occur during meat, poultry, and fish fermentation is essential for ensuring safe, palatable products.

MANUFACTURE OF FERMENTED MEAT AND POULTRY PRODUCTS

Sausage Categories and Meat Fermentation

Dry and semidry sausages represent the largest category of fermented meat products, with many ­present-

34

day processing practices having their origin in the Mediterranean region. Traditionally, dry sausages acquired their particular sensory characteristics from exposure to salt; indigenous gram-positive microorganisms such as LAB, coagulase-negative cocci, including staphylococci and kocuria (Micrococcus spp. that were mostly used for fermented meat were reclassified as Kocuria spp. in 1995 [52]), and yeasts residing on the meat; and the rapid drying conditions that exist in the warm, dry Mediterranean climate. These products were heavily seasoned and stuffed into sausage casings that excluded air, which favored the growth of certain gram-positive bacteria such as LAB. Typically, they were not smoked and preserved after fermentation while the accumulation of lactic acid, other organic acids, carbon dioxide, and alcohols served as the preservatives. The name “sausage” was derived from the Latin term salsus, meaning “salted.” Sausageprocessing practices later spread to northern Europe, and by the Middle Ages, hundreds of varieties of dry and semidry sausages were manufactured across the continent. In contrast to the Mediterranean variety, northern European sausages were prepared during the cold winter months and stored until summer and thus were called summer sausages. These sausages con-

Steven C. Ricke and Ok Kyung Koo, Center for Food Safety, Department of Food Science, University of Arkansas, Fayetteville, AR 72704. Jimmy T. Keeton, Department of Nutrition and Food Science, Texas A&M University, College Station, TX 77843-2253.

857

SMP_Food Microbiology_CH34.indd

857

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

858 tained more water than their Mediterranean counterparts and were lightly spiced, heavily smoked at cool temperatures, and less susceptible to spoilage owing to colder ambient temperatures. Summer sausages are similar to present-day semidry sausages. Examples of the most common dry and semidry sausage varieties are listed in Table 34.1 and are categorized as salamis or cervelats.

Product Categories and Compositional End-Point Characteristics

Compositional characteristics of some dry and semidry sausage categories produced in the United States, as well as some processing criteria (moisture-to-protein ratios [MPRs], pH limits, and ingredients), are defined by the U.S. Department of Agriculture Food Safety and Inspection Service (USDA-FSIS) in the Food Standards

Table 34.1  Categories and origins of selected dry and semidry sausagesa Category

Origin

Dry sausages Salamis (Genoa, Milano, Siciliano)

Italy

Lombardia salami

Italy

Capicola

Italy

Mortadella

Italy

Pepperoni

Italy

Chorizos

Spain or Portugal

D’Arles

France

Lyons

France

Alesandri and Alpino Katenrauchwurst, Dauerwurst, Plockwurst, Zerevelat Semidry sausages Summer sausages (cervelat, farmer cervelat) Holsteiner cervelat

United States Germany

Description and unique characteristics Lean pork (coarse), some with wine or cured beef (fine), garlic, large-diameter casing (beef or hog bungs) with flax twine, some with coating of white mold Coarse cut, high fat content, brandy added, twine wrap over casing Boneless pork shoulder butt combined with red hot or sweet peppers, mildly cured Finely chopped, cured beef and pork with added cubes of backfat, mildly spiced, smoked, encased in beef bladders Cured pork, some beef, cubed fat, red peppers, smalldiameter casing Pork (coarse), highly spiced, hot, small-diameter casing Similar to Italian salamis, coarse, large-diameter casings (hog bungs) All pork (fine), diced fat (fine), spices and garlic, cured, large-diameter casing Similar to Italian salami Dry sausages, smoked, air dried

Generic

Mildly seasoned soft cervelat, beef and pork (coarse), no garlic, cured, small-diameter casing

Germany

Thuringer cervelat

Germany

Gothaer cervelat

Germany

Goteborg cervelat, medwurst Landjaeger cervelat

Sweden

Teewurst, frische Mettwurst Lebanon bologna

Germany

Similar to farmer cervelat, packed in ring-shaped casing Medium dry to soft, tangy flavor, mildly spiced, smoked, some beef and veal added Very lean pork, finely chopped, cured, soft texture, mild seasoning Coarse, salty, soaked in brine before being heavily smoked Small diameter (frankfurter size), pressed flat, smoked, flavored with garlic and caraway seeds Undried, spreadable, smoked

United States

Smoked, coarse, large diameter, very acidic

a

Switzerland

From references 7, 12, 13, 67, 87, and 89.

SMP_Food Microbiology_CH34.indd

858

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products and Labeling Policy Book (111). For example, shelfstable, dry sausage must have an MPR of 1.9:1 or less, unless specified otherwise. Nonrefrigerated, semidry, shelf-stable sausage must have an MPR of 3.1:1 or less and a pH of 5.0 or less, unless it is commercially sterilized or unless specified otherwise. Alternatively, nonrefrigerated, semidry, shelf-stable sausages are those that (i) are fermented to a pH of 4.5 or lower (or the pH may be as high as 4.6 if combined with a water activity [aw] of no higher than 0.91); (ii) are in an intact form or, if sliced, are vacuum packed; (iii) have an internal brine concentration no less than 5%; (iv) are cured with nitrite or nitrate; and (v) are smoked with wood. Examples of the compositional characteristics of dry and semidry fermented meat products are given in Table 34.2.

Manufacturing Procedures and Processing Conditions

Dry and semidry sausage manufacture using starter cultures involves the following basic steps: (i) reducing the particle size of high-quality raw meat trimmings; (ii) incorporation of salt, nitrate (Europe, mostly) or nitrite (United States and Europe), glucose, spices, seasonings, and a specific inoculum selected on the basis of the incubation temperature optimum and level of lactic acid desired (various strains of lactobacilli, leuconostocs, pediococci, and streptococci [10]); (iii) uniformly blending all ingredients and further reducing the particle size; (iv) vacuum stuffing the meat into a semipermeable casing to minimize the presence of oxygen; (v) incubation (ripening) at or near the temperature optimum of the starter culture until a specific pH end point is achieved or until carbohydrate utilization is complete; (vi) heating (usually, but product dependent) of the product to inactivate the inoculum and ensure pathogen destruction; and (vii) drying (aging) the product to the required MPR end point. A generic Table 34.2  Compositions of two types of fermented sausages Value (% [wt/wt])a Parameter Moisture Fat Protein Salt pH Total acidity Yield a

Summer sausage/ Thuringer cervelat

Pepperoni/ hard salami

50 24 21 3.4 4.9 1.0 90

30 39 21 4.2 4.7 1.3 64

Except pH. Adapted from reference 105.

SMP_Food Microbiology_CH34.indd

859

859

scheme for manufacturing dry and semidry sausages using starter cultures is shown in Table 34.3.

Factors Affecting Color, Texture, Flavor, and Appearance of Fermented Meats Raw Meat Tissues

Bacterial contamination of carcasses occurs during the slaughter process, with most microorganisms originating from the hide, skin, and/or intestinal contents. These include predominantly gram-negative bacteria, such as enteric Enterobacteriaceae (including Escherichia coli and Salmonella), Pseudomonas, gram-positive LAB, and staphylococci associated with humans, animals, and the environment (10). Fresh or frozen raw meats to be used for fermented sausages should be chilled to <4.5°C, have small microbial populations, be free of physical and chemical defects, and meet the compositional specifications for the product being manufactured. The predominant bacteria that develop in meats not held under vacuum are typically gram-negative, oxidase-positive, aerophilic rods composed of psychrotrophic pseudomonads along with psychrotrophic Enterobacteriaceae (67). Spoilage is usually associated with gram-negative, proteolytic bacteria that produce putrid, “rotten egg-like” odors and flavors (10). Small numbers of salt-tolerant LAB and other gram-positive bacteria are initially present in meat and become the dominant microflora if oxygen is excluded, as in the case of vacuum-packaged or vacuum-encased meat products. Initial populations of LAB may range from 3 to 4 log CFU/g, but they increase during fermentation to 7 to 8 log CFU/g. Staphylococcaceae and Micrococcaceae also occur in fermented meats and tend to decline as the pH decreases. Some strains are lipolytic and proteolytic and contribute flavor as a consequence of breakdown products. Staphylococci and kocuria reduce nitrate to nitrite to generate nitric oxide, which reacts with myoglobin to produce the characteristic cured color of fermented meats (10). Some LAB cause greening (Lactobacillus viridescens) due to hydrogen peroxide, whereas others cause gas production (Lactobacillus brevis and Leuconostoc mesenteroides) and souring (Brochothrix), with consequent off-flavors and off-odors. Pathogenic bacteria pose a serious health threat if not controlled on raw materials and finished fermented products. During fermented meat production, there is not sufficient heat treatment to kill pathogenic bacteria and the meats are commonly manufactured at temperatures favorable to pathogens. The nonheated products rely on reductions in aw, combined with salt

Manila Typesetting Company

11/08/2012 08:54PM

860

SMP_Food Microbiology_CH34.indd

Table 34.3  Generic manufacturing scheme for dry and semidry fermented sausages with starter culturesa Procedure and conditions

Meat tissue selection

Fresh/frozen meats with low microbial populations; no discoloration; no off-odors; limited age; no DFD tissue; trimmed free of blood clots, glands, sinews, gristle, and bruises; refrigerated to <4.5°C (<40°F) Fresh/tempered –3°C (26°F) meats; coarse grind/ chop/mince lean (1/4–1/2 in. [6.35–12.7  mm]) and fat (1/2–1 in. [12.7–25.4 mm]) meats separately; combine to a specified fat end point; blend with seasonings and cure ingredients to uniformly distribute ingredients; avoid overmixing and excessive protein extraction or fat smearing Rehydrate frozen or lyophilized culture with nonchlorinated, distilled water at ambient temperature <1 h before use; tap water is usually acceptable; if antimicrobial compounds are present (e.g., chlorine) and are in excess, distilled water should be used; add inoculum (high-temperature optimum for smokehouse incubation) to meat batch; blend, but avoid excessive mixing (causes fat smearing and coating of lean particles); fine grind/chop (1/8–3/16 in. [3.2–4.8 mm]) to specified particle size Lactobacillus plantarum (21–38°C [70–100°F]), Pediococcus acidilactici (32–46°C [90–115°F]), Pedicoccus pentosaceus (21–38°C [70–100°F]), Staphylococcus carnosus (21–38°C [70–100°F])

Manila Typesetting Company

Comminution, grinding, blending ingredients; inoculum level at ca. 107 CFU/g

11/08/2012 08:54PM

Ingredients used Salt (2.5–3%) Glucose (0.4–0.8%) Nitrite (<150 mg/kg) Sodium erythorbate (550 mg/kg) Antioxidants (natural or synthetic) Spices (sterilized)

Starter culture (frozen, dehydrated)

Dry sausage (North America)

Dry sausage (Europe)

Same as semidry

Same as semidry

Same as semidry

Same as semidry

Same as semidry, with the following exceptions: use high-temperature inoculum (>32.5°C [>90°F]) for smokehouse incubation and low-temperature inoculum (21.3°C [70°F]) for “green” or “ripening” room incubation

Same as semidry, with the following exceptions: use nitrate (200–600 µg/g) alone or in combination with nitrite (nitrate used mostly in Europe; used only for Lebanon bologna and countrystyle hams in the United States); low-temperature fermentation (21.3°C [70°F]) is used most often Lactobacillus sake, Lactobacillus curvatus, S. carnosus, Staphylococcus xylosus, Lb. plantarum, P. pentosaceus, P. acidilactici, Penicillium spp., Debaryomyces spp. (20–24°C [68–75°F])

Fermentations and Beneficial Microorganisms

Semidry sausage

860

Processing parameter

SMP_Food Microbiology_CH34.indd

Incubation and fermentation (ripening)

Same as semidry

Same as semidry

Low-temperature smokehouse incubation or “ripening” room at 15–26°C (60–78°F) and 90% relative humidity for ca. 72 h to an end-point pH of <4.7; air movement, >1 m/s; smoke application at the end of fermentation if desired

Drying chamber at 12.9–15.7°C (55–60°F) and 65–70% relative humidity for ³12 days (dependent upon sausage diameter) to specified MPR

Drying chamber at 10– 11.2°C (50–52°F) and 68–72% relative humidity for ³21 days (dependent upon sausage diameter) to specified MPR

Low-temperature “ripening” at 26°C (78°F) and 88% relative humidity for 3 days to an endpoint pH of 4.7–4.8; chamber relative humidity held at 5–10% lower than relative humidity within the sausage, or use the following schedule: a. 22.2–23.9°C (72–75°F), 94–95% relative humidity for 24 h b. 20–22.2°C (68–72°F), 90–92% relative humidity for 24 h c. 18.3–20°C (65–68°F), 85–88% relative humidity for 24 h Remains in "ripening" room or moved to drying chamber at 20°C (68°F) and 88% relative humidity for 10 days and then 15.7°C (60°F) and 82% relative humidity for 14 days to a specified end point, or held at 11.7–15.0°C) (53–59°F) at 75–80% relative humidity to specified end point

861

Keep at 2°C (ca. 34°F); subject to vacuum to remove oxygen and encase in fibrous or natural casing; oxygen exclusion accelerates anaerobic fermentation and favors LAB growth and color and flavor development High-temperature smokehouse incubation at 32.5–38.1°C (90–100°F) and 90% relative humidity for ³18 h (dependent upon sausage diameter) to an end-point pH of <4.7; air movement, >1 m/s; smoke at the end of fermentation

Manila Typesetting Company

Drying (aging)

34.  Fermented Meat, Poultry, and Fish Products

Vacuum encasing (stuffing)

Adapted from references 9 and 67.

a

11/08/2012 08:54PM

861

Fermentations and Beneficial Microorganisms

862 and pH, to control pathogens. Organisms that are of greatest concern include Staphylococcus aureus, E. coli O157:H7, Salmonella, Listeria monocytogenes, Campylobacter, and the nematode Trichinella spira­ lis (10). S. aureus is a poor competitor to LAB due to pH sensitivity; however, it produces a heat-stable enterotoxin when cell numbers reach 6 log CFU/g. Thus, its growth must be controlled early during processing using proper sanitation and starter culture dominance to reduce the pH to <5.3 as rapidly as possible. E. coli O157:H7 is of the greatest concern due to its severe symptoms in humans and its association with serious salami-related outbreaks in British Columbia, Canada, in 1999; in Ontario, Canada, in 1998; and in Washington and California in 1994. The initial pHs and aws of fermented meat batters are not sufficient to inactivate E. coli, and the bacteria can become well adapted and resistant to the low pHs and low aws while they persist over time (69). For lethality performance standards, a 5-log reduction of E. coli O157:H7 for fermented meat or poultry supplemented with beef has been suggested. Salmonellae are acid sensitive, heat sensitive, and competitively inhibited by starter cultures; however, they become more heat resistant with drying. Moist heat early during processing, combined with good sanitation, is considered effective for controlling Salmonella. Heating to >68.9°C (155°F), thermal processing as outlined by the USDA (108, 110), or heating at a specified time-temperature combination (109) may be required to ensure the destruction of Salmonella in poultry tissues. For Salmonella, a 6.5- to 7-log reduction for ready-to-eat beef or poultry was suggested for lethality performance standards by the USDA-FSIS. L. monocytogenes can survive at refrigeration temperature, low aw, low pH (as low as 4.1), high salt concentration (10%), and in the presence of nitrite. Therefore, although there have been a small number of outbreaks related to this pathogen, L. monocytogenes needs constant monitoring. L. monocytogenes is also capable of biofilm formation on food contact surfaces, causing post-processing contamination at the food-processing plant. Possible hurdles against L. monocytogenes involve the use of bacteriocin-producing starter cultures, which are considered to have greater effect than reduced pH alone. Following good manufacturing practices (4) and using established fermentation temperature-time guidelines would also ensure inhibition of pathogen growth. Postrigor pH and the residual glycogen concentration in muscle tissues influence the quality of fermented sausages. ATP levels in muscle tissues average 1 µm/g 24 h after slaughter, while pH values range

SMP_Food Microbiology_CH34.indd

862

from 5.5 to 5.7 for beef, 5.7 to 5.9 for pork, and 5.8 to 6.0 for poultry. Pork meat sometimes exhibits pale, soft, and exudative characteristics and has tissue pH values of 5.3 to 5.5, a wet or watery meat surface, and a pale pink color. However, these tissues may be incorporated into dry sausages at concentrations of up to 50% of the meat block without impairing sensory qualities (106). Use of 100% pale, soft, and exudative meats in sausages, however, will likely result in products with a pale, yellowish, cured color; poor waterholding capacity; lower aw; increased susceptibility to oxidative rancidity; and poor (soft, grainy, and noncohesive) textural characteristics. Meats exhibiting the dark, firm, and dry (DFD) condition are characterized by a dry surface, dark red color, and high pH (>6.0). This condition occurs more frequently in beef than in pork. DFD trimmings are not suitable for dry ­sausage because of their excessive water-binding­ capacity and potential for accelerated microbial spoilage.­ Dark red meat from more mature animals, however, may be desirable because of its contribution to product appearance. The use of lamb and mutton meats in fermented sausages is limited (3), but studies indicate that acceptable sausage products can be produced when the meats are combined with appropriate seasonings and limited amounts of mutton fat (14, 117). Fat tissues from beef, lamb, and pork have high proportions of saturated fatty acids and yield products that are firmer and have a more desirable texture than products containing poultry and turkey fats, which have a predominance of polyunsaturated fatty acids. Polyunsaturated fatty acids are more susceptible to auto-oxidation and rancidity, which can lead to the development of off-flavors. Thus, poultry meats may be a less desirable source of fat for fermented sausage formulations because they contain higher levels of polyunsaturated fatty acids. Lean poultry meat, if used, is often supplemented with pork or beef meat to avoid sausages that appear too light in color and to ensure appropriate textural attributes. Poultry sausages are usually lower in fat (15%) than red meat sausages (23 to 45%), initially have a higher pH, and contain more moisture that can affect product uniformity (diameter) during drying. Incorporation of turkey thigh meat into sausages can result in darker red products owing to a higher myoglobin content in the lean tissue, but turkey and chicken have higher proportions of polyunsaturated fatty acids and moisture than do red meats. Slightly larger amounts of fermentable carbohydrate should be used in formulas containing higher-pH meats such as poultry. Mechanically deboned poultry meat is an acceptable meat source for fermented dry

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products sausages when limited to 10% of the meat block (45), as sausages tend to become soft when larger amounts are used.

Ingredients

Incorporation of sodium chloride, sodium or potassium nitrite and/or nitrate, glucose, and homofermentative lactic acid starter cultures (Table 34.3) in sausage formulas dramatically alters the ecology of the culture environment and chemical characteristics of finished products (Table 34.4). Incubation of sausages at the optimum temperature for growth of LAB in a reduced-oxygen environment causes a reduction in pH by the rapid conversion of glucose to lactic acid as a consequence of the exponential growth and subsequent suppression of indigenous microflora, such as psychrotrophic pseudomonads, Enterobacteriaceae, and most pathogens. Pseudomonas species are sensitive to salt, nitrite, elevated incubation temperatures (>37°C), and reduced oxygen tension, while the competitiveness of the Enterobacteriaceae is restricted at reduced oxygen levels, low pH, and in the presence of salt. The glucose content of postrigor meats (4.5 to 7 µmol/g) is not sufficient to significantly reduce pH; therefore, 0.4 to 0.8% fermentable carbohydrate in the form of glucose, corn syrup, sucrose, lactose, maltodextrins, or starch is added to sausage formulas to enable reduction of the pH to 4.6 to 5.0. Approximately 1 oz of glucose per 100 lb of meat (0.62 g of glucose/kg of meat) is required to reduce the pH by 0.1 unit. Carbohydrates such as lactose, raffinose, trehalose, dextrins, and maltodextrins, when used in lieu of glucose, yield less lactic acid owing to incomplete utilization of the substrate (57). In the United States, fermented sausages having final pH values of 4.8 to 5.0 contain approximately 25 g of lactic acid per kg (dry weight), whereas some Italian and Hungarian salamis that are fermented with limited amounts of carbohydrate may undergo a decrease in pH of only 0.5 unit. Nitrates and nitrites (40 to 50 µg/g [minimum]) in fermented sausages react with the heme moiety of myoglobin to facilitate the development of cured color, retard lipid oxidation, inhibit the growth of Clostridium botulinum (through a synergistic relationship with salt), and enhance cured flavor. In the United States, sodium chloride (2.5 to 3.0%) or a combination of sodium and potassium chloride is used with sodium or potassium nitrite (156 µg/g [maximum] going into the product, not to exceed 200 µg/g in the final product). Nitrite serves as the primary curing agent in fermented sausages, but sodium or potas-

SMP_Food Microbiology_CH34.indd

863

863

sium nitrate may be legally added to dry sausages at a maximum level of 1,718 µg/g, calculated as sodium nitrate. In Europe, sodium or potassium nitrate (200 to 600 µg/g), in combination with nitrate-reducing bacteria such as Kocuria varians and Staphylococcus carnosus, is utilized during low-temperature­ fermentation (ripening) to enable metabolism by these acidsensitive, nitrate-reducing bacteria. Sodium ascorbate is often used in combination with nitrates and nitrites in dry sausages, at concentrations of up to 550 and 600 µg/g in the United States and Europe, respectively. Ascorbates, isoascorbates, and erythorbates accelerate the reduction of nitrous acid to nitric oxide, thus enhancing color development, reducing residual nitrite, and retarding the formation of N-nitrosamines, a class of potent carcinogens. Ground pepper, paprika, garlic, mace, pimento, cardamom, red pepper, and mustard are commonly used in fermented sausages, but as with all spices, they should be sterilized to avoid wild fermentations. Red pepper and mustard are known to stimulate lactic acid formation, possibly because of the available manganese in these spices, which enhances the glycolytic enzyme fructose1,6-diphosphate aldolase (67). Garlic, rosemary, and sage contain antioxidant and antimicrobial compounds and may assist in preserving the flavor, color, and microbial shelf life of fermented sausages. Gluconodelta-lactone (GDL), a chemical acidulant, is hydrolyzed to gluconic acid, which is then converted to lactic acid and acetic acid by indigenous lactobacilli. The acidulant is sometimes used at concentrations ranging from 0.25 to 0.5%, although up to 1% is allowed in the United States. At a concentration of 0.25% GDL (45), dry sausage color and consistency are important in conjunction with a rapid decline in the pH caused by natural starter culture fermentation. At higher concentrations, GDL can inhibit the growth of lactobacilli, produce undesirable aromas, and impart a sweet flavor from the unfermented sugar. Smoke contains phenols, carbonyls, and organic acids, which may act to preserve sausage products. Phenols are effective antioxidants and microbial inhibitors, whereas carbonyls contribute a desirable amber color by combining with free amino groups to form brown furfural compounds. Organic acids, such as formic, acetic, propionic, butyric, and isobutyric acids, in smoke inhibit the growth of microorganisms on the surfaces of sausages and promote coagulation of surface proteins. Fermentations fail for various reasons, especially if a product does not reach the appropriate pH within a specified time period. Some of the most common causes include the following: starter culture not added or

Manila Typesetting Company

11/08/2012 08:54PM

864

SMP_Food Microbiology_CH34.indd

Table 34.4  Chemical characteristics of selected fermented sausage products

Category

Final pH

Dry sausages

b

864

Cervelat Capicola German “Dauerwurst” German salami Pepperoni Italian salami, hard or dry Genoa salami Thuringer, dry Semidry sausagesb

MPR

0.5–1.0

<2.3:1

4.7–4.8 4.7–4.8 4.5–4.8 0.8–1.2 4.9 4.9

1.9:1 1.3:1 1.1:1 1.6:1 1.6:1 1.9:1

Moisture loss (%) 25–50 (30)

c

35 30 28 28 8–15 (15)

33–39 46–50 45–50

11/08/2012 08:54PM

4.7–5.1

2.3:1 2.3:1 >2.3:1–3.7:1

4.7

1.0–1.3

2.5:1 2.6:1 2.3:1–3.7:1 3.1:1

10–15 10–15 10–15 10–15

2.04:1 0.75:1 2.1:1

29 >50

1.0

<35

32–38 23–29 25–27 34–35 25–32 32–38

0.79 1.0 0.5–1.3

<5.0

Moisturea (%)

Comments Heat processed (optionald); dried or aged after fermentation for moisture loss; may be smoked. Shelf stable

Heat processedd; typically smo­ ked; packaged after processing and chilling. Keep refrigerated

56–62 41–51 41–52 46–50

28–30

a aw ranges for dry and semidry sausages are <0.85–0.91 and 0.90–0.94, respectively. The European Economic Directive 77/99 requests aw of <0.91 or pH of <4.5 for dry sausages to be shelf stable or a combined aw and pH of <0.95 and <5.2, respectively. b Data from references 1, 9, 59, 77, 86, 89, 102, and 112. c Values in parentheses are averages. d USDA-FSIS Title 9 CFR may be amended to require specified time-temperature heating combinations after fermentation or verification that processing conditions destroy all pathogenic microorganisms.

Fermentations and Beneficial Microorganisms

Manila Typesetting Company

Lebanon bologna Cervelat, soft Salami, soft Summer sausage Thuringer, soft Other (for comparison) Dried beef Beef jerky Air-dried sausage

5.0–5.3

Lactic acid (%)

34.  Fermented Meat, Poultry, and Fish Products ­ ishandled, nonfermentable or insufficient amount of m sugar, antimicrobial agents incorporated into the formulation, antibiotic residues in meat trimmings, or processing temperature/humidity fluctuations.

MANUFACTURE OF FERMENTED FISH PRODUCTS Fermented fish products include a variety of fish sauces, fish pastes, and fish-vegetable blends that have been salted, packed whole in layers, or ground into small particles and then fermented in their own “pickle.” These products are eaten as a proteinaceous staple or condiment in Southeast Asia but are consumed as a condiment in northern Europe (15). Fish fermentation involves minimal bacterial conversion of carbohydrates to lactic acid but entails extensive tissue degradation by proteolytic and lipolytic enzymes derived from viscera and muscle tissues. Low-molecular-weight compounds from fish tissue degradation are the primary contributors to aroma and flavor characteristics of sauces. Indigenous microorganisms, however, do contribute to aroma and flavor but are limited to species tolerant of high salt concentrations (10 to 20%) in the curing brine. Partial tissue hydrolysis is responsible for the unique textural attributes of pastes and fish-vegetable blends.

Fish Sauces

Fish sauces, such as nuoc-mam (Vietnam), patis (Philippines), nam-pla (Thailand), budu (Malaysia), nuocmam-nuoc (Thailand), and shottsuru (Japan), are liquids consumed as a condiment with rice and vary in color from clear brown to yellow-brown. Sauces have a predominantly salty taste and are derived from decanting or pressing fermented fish or shrimp after a 9-month to 1-year fermentation (73). Products fermented over a 1- to 2-year period have a distinctive sharp, meaty aroma and may range in protein content from 9.6 to 15.2%. Commercial fish sauce production begins with layering seine-netted fish, shrimp, or shellfish with salt in concrete vats at an approximate ratio of 3:1 (fish to salt), sealing the vats, allowing supernatant liquor to develop, and carefully decanting this liquid. Enzymatic digestion or fermentation may range from 6 months for small fish to 18 months for larger species. The first liquid removed from the fermenting fish contains an abundance of peptides, amino acids, ammonia, and volatile fatty acids and is considered the highest-quality product. Extracts may be supplemented with caramel, caramelized sugar, molasses, roasted corn, or roasted barley to enhance the color and keeping qualities. For some products, the sauce is

SMP_Food Microbiology_CH34.indd

865

865

ripened in the sun for 1 to 3 months and blended with bacterial by-products from the manufacture of monosodium glutamate. The nitrogen content of the supernatant liquor increases as a result of proteolysis during fermentation. Initially, salt penetrates the tissues by osmosis (0 to 25 days), a protein-rich liquid develops through autolysis (80 to 120 days), and ultimately, the fish tissue is transformed into a nitrogen-containing liquid (140 to 200 days). Proteases such as cathepsin B and trypsinlike enzymes have been shown to increase the soluble protein content of the liquor during the first 2 months, but their activity gradually decreases through an inhibition feedback mechanism with the buildup of amino acids and polypeptides. New polypeptide formation occurs during the last fermentation stage as the level of free amino acids increases. Aseptically produced fish sauces do not have a typical aroma (16), suggesting that some microbial involvement is required for aroma development. Bacterial populations in raw fish, predominantly facultative anaerobes, are high (2.7 × 104 CFU/g) initially but decline (2 × 103 CFU/g) over a 6- to 9-month period. Bacillus, Lactococcus, Kocuria, and Staphylococcus are indigenous microflora, but as fermentation progresses, populations and the number of species decline with the changing brine environment. Crisan and Sands (22) identified Bacillus cereus and Bacillus licheniformis as the dominant bacteria in nam-pla, but after fermentation for 7 months, B. lichenifor­ mis, Bacillus megaterium, and Bacillus subtilis were the dominant species. Micrococcus copoyenes, K. varians, Bacillus pumilus, and Candida claussenii are other halotolerant microorganisms, i.e., those capable of growth in 10% salt brine but not in 20% brine, that have been isolated from sauces. The characteristic aroma and flavor of fish sauce are complex and cannot be attributed to specific volatile fatty acids, peptides, or amino acids derived from bacterial fermentation alone. One theory suggests a complex interaction of enzymatic activity and oxidation during fermentation; however, some evidence exists for bacterial production of volatile fatty acids in fresh fish before salting and in the brief period following salting (17). Dougan and Howard (29) characterized nam-pla aroma as being ammoniacal-trimethylaminic, cheesyethanoic, and n-butanoic and having other aromas attributable to low-molecular-weight volatile fatty acids, meaty ketones, keto acids, and amino acids, especially glutamic acid. The flavor of fish sauce has a strong salt component, with contributions from a combination of monoamino acids and possibly aspartic and glutamic

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

866 acids. Other factors, such as the pH of the brine, fermentation temperature, and salt concentration, also affect flavor.

Fish Pastes

Fish pastes, which are more widely produced than sauces, are consumed raw or cooked as a condiment with rice and vegetables. In some countries, fish pastes may be the primary source of dietary protein for low-income families. A wide variety of fermented fish pastes are produced, and they require shorter processing periods than sauces. These include bagoong, tinabal, and balbakwa (Philippines); pra-hoc (Cambodia); padec (Laos); blachan or bleachon and trassiudang (Malaysia); sidal (Pakistan); and shiokara (Japan) (15). Paste production consists of mixing cleaned, eviscerated whole or ground fish, shrimp, plankton, or squid with salt at a ratio of 3:1 (fish to salt) and then placing them in vats to ferment. Proteolytic enzymes from visceral tissue and, to some extent, bacteria break down the tissue until it attains a pasty consistency. Aging in hermetically sealed containers may follow, or the paste may be hand kneaded and then aged. Pickle (liquid exudates), which forms as a result of the osmotic differential of the brine solution and the fish tissue, is decanted, aged, and consumed as fish sauce. When pickle no longer forms, the fish paste is ready for use or aged further. Typical paste has a salty, cheeselike aroma, but other characteristics vary depending upon the method and region of production. Bacteria do not appear to play a major role in the proteolysis of fish paste but may contribute to aroma and flavor or to spoilage. Populations of 6.5 × 103 CFU/g, representing 40 bacterial species, have been reported to occur in bagoong (20% salt), among which Bacillus, Micrococcus, and Moraxella species were dominant (36). Eighteen strains of bacteria, some of which were halophilic, were found in shiokara (16, 118), and Micrococcus species appeared to be responsible for ripening. Halophilic Vibrio and Achromobacter species and more recently Piscibacillus, Halalkalicoccus, and Salinvibrio species were also isolated. Proteolytic enzymes from fish viscera, stomach, pancreas, and intestine are more active than muscle enzymes and are most often responsible for the release of free amino acids and polypeptides, which are believed to undergo further microbial synthesis to yield specific flavor compounds. Some of the compounds derived from proteolytic degradation include volatile fatty acids (formic, ethanoic, propanoic, isobutanoic, and n-pentanoic acids), ammonia, trimethylamine, and mono- and dimethylamines. Aminobutane and 2-methylpropylamine are thought to result from microbial action. If carbo-

SMP_Food Microbiology_CH34.indd

866

hydrates are added to the raw materials and fermented to alcohol, this may suppress proteolytic and microbial activity. Bogoong is the residue of partially hydrolyzed fish or shrimp (73) having a pH of 5.2 and containing 65 to 68% moisture, 13 to 15% protein, 2% fat, 20 to 30% salt, 1.45% nitrogen, 0.18% volatile nitrogen, 0.015% trimethylamine, 0.011% hydrogen sulfide, and a 35% solids base. Pra-hoc, used in dishes such as soups, may contain as much as 24% protein and 17% salt. Shiokara, a Japanese fish or squid product with or without malted rice, is texturally between a sauce and paste and is characterized as a dark brown liquid containing lumps of solid tissue. In its final form, the squid product contains 74.2% water, 7.8% salt, 11.6% protein, and 8.7% ash.

Fermented Rice and Shrimp or Rice and Fish

Balao balao, a traditional food of the Philippines, is a cooked rice and shrimp product that is fermented at room temperature for 7 to 10 days in a 20% salt brine (73). A similar Filipino fermented product in which fish is substituted for shrimp is known as burong isda, while the Japanese version is called naresushi or funasushi. Bacterial isolates involved in the fermentation of burong have been demonstrated to be capable of starch hydrolysis and are identified (73) as having characteristics similar to Lactobacillus plantarum and Lactobacillus coryneformis subsp. coryneformis, but with the capacity to convert oligosaccharides and reducing sugars to lactic acid. An enzyme with a pH optimum of 4.0 has been isolated from these bacteria and found to hydrolyze amylose to oligosaccharides. However, during the initial stages of fermentation, other bacteria may hydrolyze starch for use by LAB.

STARTER CULTURES IN MEATS The use of starter cultures in fermented meat products is a relatively recent practice compared with their use in fermented dairy foods and alcoholic beverages (63). The rationale for the use of a starter culture in meat fermentation is similar in concept to the use of starter cultures in dairy products. In the United States, lactobacilli, pediococci, and staphylococci are the predominant culture microorganisms, whereas in Europe these genera are most often used in combination with staphylococci and kocuria (10, 11, 63). Proper inoculation and incubation procedures are among the most critical steps for the production of safe, flavorful, uniform, and wholesome fermented sausages. Inoculation is accomplished by one of three methods: (i) natural fermentation, which relies on indigenous microflora in the meat to serve as the in-

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products oculum; (ii) back inoculation, which involves transfer of a portion of raw meat from a previous batch of sausage to the present batch, i.e., transfer from a “mother batch” of raw sausage; and (iii) the use of starter cultures, i.e., inoculation of unfermented meat with a pure strain or strains of LAB. Cultures come in three basic forms: frozen liquid, frozen pellets, or freeze-dried. Frozen liquid has been the most popular form in the United States, but the use of frozen pellets is increasing. Frozen pellets and dry cultures are used more outside the United States. The primary reasons for using starter cultures are to ensure an immediate dominance of lactic acid microflora to inhibit other microorganisms, especially pathogens, and to provide metabolic efficiency for the production of lactic acid or nitrate reduction (10). Use of commercial starter cultures is the predominant method of inoculation in the United States.

Development of Commercial Starter Cultures

In fresh meat, LAB are a minor component of the microflora, but when meat is packaged and stored under vacuum, the resulting microenvironment facilitates their growth. Commercial application of starter cultures in the cheese industry during the 1930s led investigators to identify potential starter cultures for fermented meats in the 1940s. Use of pure cultures in sausages began after several studies in the 1950s demonstrated that LAB are responsible for lactic acid production (25, 74–76). An essential requirement for starter cultures is that they can be produced and preserved in a viable and metabolically active form suitable for commercial distribution. Additional requirements of homofermentative, catalase-positive starter cultures include the production of adequate quantities of lactic acid, the ability to grow at salt concentrations of up to 6%, and the capacity to enhance the flavor of finished sausages without production of biogenic amines and slimes. In the United States, Lactobacillus species were first used as starter cultures for meat fermentations in the temperature range of 20 to 25°C. Original starter cultures were logically derived from the predominant microflora of fermented meat products, but when attempts were made to preserve these cultures via lyophilization, as was routinely done with dairy starter cultures, these strains invariably died. Deibel et al. (26) were able to surmount this problem by using Pediococcus cerevisiae (now called Pediococcus acidi­ lactici), the first commercially available meat starter culture. Although P. acidilactici was not a predominant bacterium in naturally fermented meats, it survived lyophilization, possessed characteristic lactic acid fermentation properties, had a higher optimal growth

SMP_Food Microbiology_CH34.indd

867

867

temperature, and was tolerant of salt concentrations of up to at least 6.5% (9, 11, 26). When Deibel et al. (26) developed a lyophilization procedure for maintaining starter cultures consisting of pediococci, it was considered the best method for distributing a viable culture in a reliable and economical manner (13). Eventually, lyophilization proved commercially problematic because rehydration procedures were time-consuming and introduced unacceptable variation (9). Also, lyophilized cultures exhibited inordinately long lag phases, thus lengthening fermentation times. Implementation of frozen culture technology in the late 1960s, along with improvements in conditions for storing, handling, and shipping, led to commercial acceptance of frozen culture concentrates (13, 32). This approach eliminated the need for a rehydration step and provided larger numbers of viable cells than did lyophilization (83). Widespread use of pure starter culture strains in the United States occurred during the late 1970s and was a consequence of the development of efficient, highvolume dry sausage production technologies that required short ripening times and produced consistently uniform products with low defect levels. Foodborne illness outbreaks associated with coagulase­positive staphylococci also focused attention on the need to control the fermentation process and ensure the production of safe products. In addition to solving these problems, the introduction of frozen culture concentrates also renewed interest in lactobacilli as starter cultures (9, 12). In 1974, Lb. plantarum was patented as a starter culture to be used alone or in combination with P. acidi­ lactici for dry and semidry sausages (9, 33). Currently, the predominant genera of bacteria used either singly or as mixed starter cultures in the United States and Europe are Pediococcus, Lactobacillus, Kocuria, and Staphylococcus (9, 46). These cultures are available as frozen liquid or frozen pellets, in a freezedried form, or in a low-temperature stabilized liquid form (syrup). The frozen liquid and pellet forms are used most often in the United States, whereas pellets and dry cultures are used more worldwide (10). P. aci­ dilactici, Pediococcus pentosaceus, Lb. plantarum, S. carnosus, K. varians, or Nesterenkonia halobia (10; S. I. Curtis, personal communication) are preferred in the United States for their rapid and nearly complete (>90%) conversion of glucose to lactic acid (pH 4.6 to 5.1) at high temperatures (ca. 32°C). Culture type selection in the United States is based on fermentation temperature and the final pH of the product. The final desired pH of fermented meat products depends on the “activity” of the culture, which generally refers to the ability of a culture to reduce the pH of the product under a set of

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

868 defined conditions, especially the desired temperature of the fermentation. The rate of pH decline increases with increasing culture activity. Highly active meat starter cultures are able to decrease the pH in a designed meat system from 5.6 to 4.8 within 8 h of fermentation. P. acidilactici is used in high-temperature fermentation (35 to 46.1°C) and enables rapid pH decline. For low­temperature fermentation (21 to 35°C) and rapid pH decline, P. pentosaceus is preferred, whereas for low temperatures (21 to 35°C) and slow pH decline (slow fermentation), Lb. plantarum strains are preferred. In addition, selected Staphylococcus spp. are added for color and flavor development. To improve the textural properties of fermented sausages, control pathogens, and expand the temperature range for fermentation, mixtures of meat starter cultures can be used. In Europe, less fermentative (pH 5.2 to 5.6) bacteria, e.g., Staphylococcus xylosus, S. carnosus, Staphylococcus simulans, Staphylococcus saprophyticus, and to a lesser extent, Kocuria spp. (71), which have a lower growth temperature optimum (<24°C), are preferred for flavor development and red color enhancement. Coagulasenegative cocci including staphylococci and kocuria, which are commonly employed as starter cultures in the manufacture of European dry sausage, reduce nitrate and nitrite via reductase enzymes. Chemical reduction of nitrate/nitrite to nitric oxide via nitrate and nitrite reductase, followed by reaction with the singular heme moiety of myoglobin, forms dinitrosylhemochrome, which develops into a characteristic pink cure color when heated (78). Some Lactobacillus spp. (Lb. delbrueckii subsp. lactis, Lb. sake, Lb. farciminis, Lb. brevis, Lb. buchneri, and Lb. suebicus) reduce nitrite to nitric oxide in vitro (116). Red color development in meat can also be attributed to bacterial consumption of oxygen. Increased oxygen consumption by rapidly growing facultative anaerobes such as lactobacilli could reduce oxygen tension on meat surfaces. Lactobacillus fermentum forms a physical barrier that would limit access of oxygen to the meat surface underneath. In theory, bacteria initially consume oxygen and reduce the oxygen pressure to levels that allow for metmyoglobin formation (6). However, further consumption of oxygen establishes a low-oxygen environment and allows for metmyoglobin reduction to bright red myoglobin derivatives. Bacterial metabolites or intracellular components from bacterial cells are presumed responsible for metmyoglobin conversion, but the exact mechanism of oxymyoglobin oxidation is not fully understood. P. pentosaceus and P. acidilactici have temperature optima of 32 and 35°C, respectively, whereas lactobacilli have lower optima, in the range of 21 to 24°C. Products that

SMP_Food Microbiology_CH34.indd

868

are fermented rapidly, such as beef sticks and pepperoni, utilize P. acidilactici and P. pentosaceus alone or in combination with Lactobacillus species or Kocuria species, which permits the use of elevated fermentation temperatures ranging from 32 to 46°C. Lactobacillus curvatus, Lb. sake, Lb. plantarum, P. pentosaceus, and P. acidilactici produce bacteriocins, which may find broader application in starter cultures for meat fermentations and are discussed in the following section.

Functional Starter Cultures

More recent trends have been focused on starter cultures not only as fermentation tools but also for functional food purposes, to capitalize on their flavor-enhancing,­ bioprotective, and health-beneficial properties. The flavor of fermented meats is generated primarily from meat tissues through proteolytic and lipolytic enzymes. However, starter cultures also generate flavor compounds during carbohydrate catabolism. While the contribution of LAB may be limited to organic acids, staphylococci can generate aroma components by production of amino acids such as leucine, isoleucine, and valine as well as free fatty acids (34). Umami taste derived from l-glutamate production can satisfy the consumer’s preference for protein-rich foods. Selective strains with interesting flavors can enhance organoleptic characteristics, prevent the formation of off-flavors, as well as reduce ripening time through the actions of aldehyde and alcohols from the breakdown of branchedchain amino acids and methyl ketones from microbial b-oxidation of fatty acids (62). Since fermented meats are processed without heat treatment, addition of bacteriocin-producing strains was suggested to ameliorate safety concerns. Bacteriocins are antimicrobial peptides or proteins that either kill or inhibit the growth of closely related strains, particularly other gram-positive bacteria. Because of their minimal risk to human health, bacteriocins are considered natural food preservatives. Lactobacillus sakei, Lb. curvatus, and Lb. plantarum are bacteriocin-producing strains with antilisterial activity. These starter cultures have several advantages for use in fermented meats, including antimicrobial activity, competition with pathogens and spoilage bacteria for nutrients to increase shelf life, and a reduced rate of meat tissue degradation, in conjunction with a more controlled and standardized fermentation of meat. Studies have revealed that such starter cultures can act independently without affecting the flavor of meat. However, bacteriocin-producing strains target only a limited range of bacteria, consisting mostly of gram-positive bacteria. In addition, the bacteriocin expression level depends on fermentation conditions as well as distribution of the

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products bacteriocin without any hindrance by meat protein and fat that can cause complications. Alternative antimicrobials for other bacteriocins are lysostaphin, which is produced by Staphylococcus simulans bv. staphylolyticus or Penicillium nalgiovense and is active against S. aureus; and reuterin, which is produced by Lactobacillus reuteri and inhibits a wider range of both gram-positive and gram-negative bacteria (62). Introduction of probiotic LAB to fermented meat has also been suggested to provide health benefits for consumers. Probiotic bacteria can be strong competitors during fermentation because they are capable of surviving the conditions of the gastrointestinal tract. The intestinal isolates Lactobacillus rhamnosus FERM P-15120 and Lactobacillus paracasei subsp. paracasei FERM P-15121 exhibit metabolic activities similar to those of commercial starter cultures, and both can inhibit the growth of S. au­ reus (90). However, human clinical studies to determine if they possess health-promoting effects have not been conducted as of yet. Conjugated linoleic acid (CLA) produced by lactobacilli, bifidobacteria, and other LAB in fermented meat is also considered to provide health benefits. CLA has antioxidant and anticancer properties, and CLA-producing starter cultures represent a potential opportunity for producing nutraceutical meat products (107). Carnobacterium piscicola is yet another example of a new functional starter culture, with both its ability to degrade leucine and thereby enhance the flavor characteristics of meat (similar to flavor compound production by staphylococci) and its ability to produce a bacteriocin that is active against L. monocytogenes and Enterococcus. Recent efforts to discover or develop additional functional starter cultures for commercial use have been limited by a lack of efficacy studies in humans and by consumer concerns regarding consumption of genetically engineered bacteria.

Characteristics of Commercial Meat Starter Cultures

Bacterial starter cultures for meat and poultry products in the United States have been selected on the basis of being homofermentative; capable of rapidly converting glucose or sucrose to dl-lactic acid anaerobically, with sustained growth as the pH decreases to 4.5; tolerant of salt brines (NaCl and KCl) up to 6%; and capable of growth in the presence of sodium or potassium nitrate (600 µg/g) or nitrite (150 µg/g). These cultures are aciduric, capable of growth in the range of 21 to 43°C, nonproteolytic, nonlipolytic, inactivated at 60.5 to 63.2°C, resistant to phage infection and mutation, and able to outgrow and/or suppress the growth of pathogens. Specific advantages favoring the use of commercial

SMP_Food Microbiology_CH34.indd

869

869

starter cultures over natural fermentations, i.e., back inoculation, are consistency of the inoculum, reduced risk of bacterial cross-contamination, uniformity of lactic acid development, production of desirable flavor components, predictability of the pH end point (as regulated by the carbohydrate level and incubation temperature), and reduced risk of proteolytic and pathogenic bacterial outgrowth. Other benefits include the acceleration of fermentation to increase commercial plant throughput and the reduction of product defects, such as off-flavors, lack of tangy flavor, excessive softness, crumbly texture, gas pockets, and pinholes, all of which result from the activity of heterofermentative bacteria. An inoculum population of 7 log CFU/g of raw product is sufficient for rapid lactic acid development within 6 to 18 h under controlled temperature, airflow, and relative humidity conditions (10), but the incubation time is also dependent on carbohydrate type and concentration, spice composition, exclusion of oxygen, and product diameter or thickness. In the United States, commercial processors inoculate meats with homofermentative, gram-positive bacteria that have been isolated from and adapted to specific meat products. These inocula (starter cultures) may consist of a single species of Lb. plantarum, Lactobacillus pentosus, Lb. sake, Lb. curvatus, K. varians, P. acidilactici, or P. pentosaceus or, more commonly, a combination of these bacteria. Use of starter cultures ensures the dominance of desirable bacteria, production of acids consisting of >90% lactic acid, and modulation of fermentation based on combinations of inoculum level, glucose concentration, and incubation time/temperature intervals (9, 10). Thus, the combined effects of low pH, increased acidity, concomitant loss of moisture during drying, reduction of aw, concentration of curing salts such as sodium chloride and sodium nitrite, bacterial inhibition of spoilage or pathogenic microorganisms, and heat processing (if applied) preserve fermented meat and poultry products against spoilage by inactivating indigenous tissue and bacterial enzymes. Mold and yeast starter cultures are not widely used in the United States, with the exception of dry sausages produced in the San Francisco area. White or gray molds are typical on casing surfaces of sausages produced in Hungary, Italy, Spain, Greece, Yugoslavia, Romania, Slovakia, and the Czech Republic. Their effect is primarily cosmetic, but it has been reported that the mycelial coat can reduce moisture loss and facilitate uniform drying (9). Catalase produced by molds may serve as an antioxidant by reacting with surface oxygen to prevent it from entering the product, whereas nitrate reductase promotes the development of a red surface (71). Green

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

870 mold on the surface of sausage is typically the result of excessive humidity and sporulation, but Penicillium chrysogenum on French sausage is noted for its ability to produce a preferred ebony color. Molds are capable of decomposing lactic acid, which increases the pH and results in a milder flavor (43). P. nalgiovense is most commonly used for this purpose, but application of P. chrysogenum and Penicillium camemberti may also be used. In Germany, only Penicillium candidum, P. nal­ giovense, and Penicillium roqueforti are approved for application to sausages. The use of nontoxigenic molds may reduce the risk of mycotoxin production by other molds (43, 61). Yeasts such as Candida famata and Debaryomyces hansenii are used alone or in combination with bacterial cultures to produce a powdery surface or a fruity/alcoholic aroma, which may be construed as spoiled if uncontrolled (9). These cultures are added at a population level of 6 log CFU/g and are characterized by a high salt tolerance and growth at low aw, e.g., as low as 0.87 (47). Streptomyces griseus subsp. hutter produces a cellar­ripened sausage aroma and enhances the color because of its nitrate reductase and catalase activities (71).

Classification of Bacterial Starter Cultures for Meat

The genera of bacteria most commonly used for meat starter cultures are Lactobacillus, Pediococcus, Kocuria, and Staphylococcus (9, 46, 53, 102); Pediococcus, Lactobacillus, and other LAB are preferred when acid production is of primary importance. Specific strains of coagulase-negative cocci such as staphylococci and kocuria are used in meat curing when lower incubation temperatures and less acid production are required for flavor development, as found in many European sausages (9, 102). Lactococcus lactis subsp. hordniae and L. lac­ tis subsp. xylosus are now in the genus Lactobacillus, whereas Streptococcus diacetilactis has been classified as a citrate-utilizing strain of L. lactis subsp. lactis (52). Lactobacilli and pediococci are gram-positive, nonspore-forming rods and cocci, respectively, which produce lactic acid as a major end product during the fermentation of carbohydrates (8, 103). Originally classified on the basis of morphology, glucose fermentation pathway, optimal growth temperature, and stereoisomer of lactic acid produced, this phenotypic grouping has remained largely intact but may change considerably as molecular characterization is completed. Research on oligonucleotide cataloging and rRNA sequencing by DNA-rRNA hybridization indicate that all gram-positive bacteria cluster in 1 of the 12 major eubacterial phyla and can be further divided into two main groups or clusters

SMP_Food Microbiology_CH34.indd

870

(8, 68, 102). One cluster, designated the Actinomycetes subdivision, comprises bacteria possessing a mol% G+C of DNA above 53 to 55% and would include meat fermentation genera such as Micrococcus, whereas the low-mol% G+C cluster, designated the Clostridium sub­ division, would include meat fermentation genera such as the staphylococci (8, 101, 102, 115). LAB are thought to form a large related cluster that lies phylogenetically between strictly anaerobic species, such as the clostridia, and facultatively or strictly aerobic staphylococci and bacilli (8, 35, 55, 102). At the species level, lactobacilli fall into three clusters, obligately homofermentative, obligately heterofermentative, and facultatively heterofermentative, which do not appear to correlate with current classification schemes (8, 102). Most of the species involved in or associated with meat fermentations are within the Lactobacillus casei/ Pediococcus subgroup (8). The pediococci are grampositive cocci and are the only LAB capable of dividing in two planes. Consequently, they can appear as pairs, tetrads, or other formations (37, 83, 102). The latest taxonomic classification places pediococci as members of the gram-positive, facultatively anaerobic phylogenetic cluster, which contains 15 genera, including the staphylococci, streptococci, kocuria, and leuconostocs (93). Pediococci are closely aligned with the phylogenetic cluster that includes lactobacilli and leuconostocs (68). Most of the strains designated P. cerevisiae that were used as meat starters have been reclassified as P. acidilactici, and now P. acidilactici, Pediococcus dam­ nosus, Pediococcus parvulus, and P. pentosaceus form an evolutionary group with the Lb. pentosus, Lb. bre­ vis, and Lb. buchneri complex (102). P. pentosaceus has been shown to be in the same evolutionary group with Lb. brevis, Lb. plantarum, and Lb. reuteri (21). The lactobacilli associated with meat fermentations are members of the genus Lactobacillus. They are characterized as gram-positive, non-spore-forming rods that are catalase negative on media not containing blood, are usually nonmotile, usually reduce nitrate, and ferment glucose (48, 56). The genus Lactobacillus currently consists of more than 150 species (72). DNA hybridization and sequencing comparisons indicate that despite the numerous species, as a whole the lactobacilli comprise a well-defined group of bacteria (48). Part of the reason for the large number of species is that the genus Lactobacillus has been studied exhaustively, not only from the perspective of taxonomy and identification but also from the standpoint of nutritional requirements for application in studies on biochemistry and metabolism (95). Most research on meat fermentations has emphasized Lb. plantarum as

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products the predominant species (9, 102). However, attempts to identify lactobacilli isolated from meat are usually less than successful because most of the documented descriptions and schemes of identification are based on isolates from other food sources (86, 92, 96). A series of investigations on atypical lactic acid streptococci isolated from fermented meats resulted in their identification as Lb. sake and Lb. curvatus (50). These species outnumbered the typical lactobacilli, identified as Lb. plantarum, Lb. brevis, Lactobacillus alimen­ tarius, Lb. casei, Lb. farciminis, Weissella viridescens, unspecified leuconostocs, and pediococci, by 1,000fold and were typically psychrotrophic and less acid tolerant (46, 102). Efforts to accurately classify and identify these strains are becoming more important as various isolates of LAB become more commonly used as starter cultures.

Starter Culture Metabolism

The term “fermentation” was first defined by Pasteur as life in environments devoid of oxygen. However, meat fermentations are more accurately defined as bacterial or microbial metabolic processes in which carbohydrates and related compounds are oxidized, with the release of energy, in the absence of oxygen as a final electron acceptor. Thus, by definition, fermentative microorganisms cannot use oxygen as a terminal electron acceptor to generate ATP and must either conduct anaerobic respiration, using more electronegative electron acceptors (carbon dioxide, sulfate, nitrate, or fumarate), or ferment intermediate metabolites formed from the substrate as electron acceptors (42). The bacteria responsible for fermentation are either facultative or obligate anaerobes. As a collective group, the gram-positive acidogenic bacteria (predominantly LAB) include the genera Lactobacillus, Streptococcus, Pediococcus, Leuconostoc, Lactococcus, and Enterococcus, which can metabolize a large number of mono- and oligosaccharides; polyalcohols; aliphatic compounds; mono-, di-, and tricarboxylic acids; and some amino acids, although individual species characteristically have a limited range of carbon and energy sources (65). During fermentation of sausages, members of this group of bacteria are responsible for two basic microbiological processes that occur simultaneously and are interdependent, viz., the production of nitric oxide by nitrate- and nitrite-reducing bacteria and the decrease in pH as a result of anaerobic glycolysis. These two activities are synergistic due to the pH dependency of nitrite and nitrate reduction. The following discussion details reactions that are important in

SMP_Food Microbiology_CH34.indd

871

871

meat, poultry, and fish fermentations and is based on several excellent reviews (8, 55, 63).

Carbohydrate Fermentation Pathways

Acidogenic bacteria ferment indigenous and added carbohydrates to form primarily lactic acid. The formation of lactic acid, which is an anaerobic process, helps to create a reduced-pH environment in the meat, and this in turn contributes to the development of cured meat color. The formation of lactic acid also causes coagulation of meat proteins and, in combination with drying, gives sausages their characteristic firm texture. There are three potential pathways that LAB may use to form lactic acid from carbohydrates. The homofermentative pathway yields 2 moles of lactic acid per mole of glucose; the heterofermentative pathway yields 1 mole of lactic acid, 1 mole of ethanol, 1 mole of acetic acid, and 1 mole of carbon dioxide per mole of glucose; and the bifidum pathway (which is not discussed at length here, since it is generally not found in the bacteria important in meat starter cultures) yields 3 moles of acetic acid and 2 moles of lactate per 2 moles of glucose (42). The homofermentative pathway, or glycolysis, is the primary means of generating lactic acid in meat fermentations. Both P. acidilactici and P. pentosaceus are microaerophilic and, under anaerobic growth conditions, homofermentative, yielding dl-lactic acid (37, 83). Glucose and most other monosaccharides are fermented, and unlike other pediococci, both species can ferment pentoses (37). It is probable that P. acidilac­ tici and P. pentosaceus transport glucose via the phosphoenolpyruvate phosphotransferase system and derive ATP using the Embden-Meyerhof pathway (glycolysis). The Embden-Meyerhof pathway features the formation of fructose-1,6-diphosphate (FDP), which is cleaved by FDP aldolase to form dihydroxyacetonephosphate and glyceraldehyde-3-phosphate. These three-carbon intermediates are converted to pyruvate, which also energetically favors ATP formation via substrate-level phosphorylation at two separate metabolic transformation steps (2 ATP per glucose). Because the resulting NADH formed during glycolysis requires oxidation to regenerate NAD+ and maintain oxidation-reduction balance, pyruvate is reduced to lactate by an NAD+-dependent lactate dehydrogenase. Since lactic acid is virtually the only end product, the fermentation is referred to as a homolactic fermentation (8). Heterofermentation is also characterized by dehydrogenation steps, yielding 6-phosphogluconate, followed by decarboxylation and formation of a pentose-5phosphate. The pentose is cleaved by phosphoketolase to form glyceraldehyde-3-phosphate, which ­eventually

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

872 ­ ecomes lactic acid through glycolysis, and acetyl phosb phate (which is required to maintain electron balance) is reduced to ethanol (8, 42). In theory, homofermentative and heterofermentative bacteria can be distinguished by the pattern of fermentation end products and the presence or absence of FDP aldolase and phosphoketolase (55). Thus, obligately homofermentative species possess a constitutive FDP aldolase and lack phosphoketolase, whereas the reverse is true of obligately heterofermentative species (8, 55, 57). Species of LAB used to ferment meat can be regarded essentially as facultative heterofermenters because they not only possess a constitutive FDP aldolase and consequently use glycolysis for hexose fermentation but also possess a pentose-inducible phosphoketolase. The type of fermentation is dependent on growth conditions. In an ecosystem such as meat, with a wide variety of complex substrates serving as sources of pentoses, organic acids, and other fermentable compounds in addition to hexoses, it is possible for LAB to be homofermentative when using hexoses but heterofermentative when using pentoses and other substrates. Lactic acid is the primary metabolite derived from both homo- and heterofermentation, but during heterofermentation the production of mixtures of lactic acid, acetic acid, ethanol, and carbon dioxide also occurs.

ATP Generation, Pyruvate Metabolism, and Energy Recycling

The most important energy requirements of the bacterial cell are macromolecular synthesis and the transport of essential solutes against a concentration gradient (8). Fermentation that is characteristic of LAB is essentially the oxidation of a substrate to generate energy-rich intermediates that are used in turn for ATP production by substrate-level phosphorylation. Although LAB can tolerate more minimal and broad ranges of internal pH values, there still must be a substantial amount of ATP generated to keep the cytoplasmic pH above a threshold level to offset the extensive external acid production (8). Bacteria have considerable flexibility in the pathways they can use to generate ATP and regulate its distribution. In aerobic bacteria, which rely on an electron transport chain, large quantities of ATP can be generated via a proton motive force across the electron transport chain coupled to an H+-translocating ATP synthase (8). Even though LAB do not possess ATP-generating capabilities via electron transport, lactobacilli do have some metabolic flexibility in altering pyruvate pathways and conserving energy by end-product efflux. Some LAB have alternative pathways of pyruvate metabolism other than direct reduction to lactic acid. Oxidation involves

SMP_Food Microbiology_CH34.indd

872

the formation of NADH from NAD+, which requires regeneration back to NADH, and pyruvate serves as the key intermediate for this by acting as the electron acceptor (8). Pyruvate-formate lyase generates formate and acetyl coenzyme A from pyruvate and coenzyme A, and the acetyl coenzyme A can be used as a precursor for substrate-level phosphorylation, for direct reduction to ethanol, or both. This alteration in pyruvate metabolism has been observed in studies in which lactobacilli shift fermentation patterns and, concomitantly, the amount of potential ATP formed as the growth rate changes (28). In these studies, mixed acid-forming strains of lactobacilli produced less lactic acid and more acetic acid with decreasing growth rates, and for every mole of acetic acid, an extra mole of ATP (2 versus 1 mole of ATP produced per mole of lactic acid) was potentially formed (28). During fermentation, LAB produce sufficient end products in the cytoplasm to result in a very high internal-to-external cell gradient (80). It has been suggested that when a fermentation product such as lactic acid exceeds the electrochemical proton gradient, additional metabolic energy is gained by product efflux through carrier-mediated transport in symport with protons (70, 80, 104). Metabolic energy is conserved because a product gradient is essentially a proton gradient that can be used to directly produce ATP by proton-driven ATPase, and it helps to maintain a proton motive force without consumption of ATP (8, 80).

Nitrate/Nitrite Reduction

Cured meat derives its characteristic pink color from the reaction of myoglobin (Fe2+) or metmyoglobin (Fe3+) with nitric oxide to ultimately form nitric oxide myoglobin, an unstable, bright pink compound (78). Nitric oxide myoglobin is formed directly from the reaction of purple-red myoglobin with nitric oxide. Another proposed route of formation may be through the oxygenation of myoglobin to bright red oxymyoglobin (Fe2+) and subsequent oxidation to brown metmyoglobin (Fe3+). Nitric oxide reduces metmyoglobin to gray-brown nitric oxide metmyoglobin (Fe2+), which transmutates to nitric oxide myoglobin with the loss of oxygen. Nitric oxide myoglobin, when heated, forms nitrosohemochrome (Fe2+), the stable pink pigment present in commercially processed cured meats. Nitrate/nitrite-reducing bacteria transform added nitrates into nitrites and eventually into more reduced forms for reaction with the dominant myoglobin or metmyoglobin pigment forms. Micrococcaceae possess nitrate reductase to convert nitrate to nitrite, which in turn is converted to nitric oxide via nitrous acid under acidic conditions. Nitric oxide is a strong

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products electron donor and rapidly reacts with the heme moiety of myoglobin to form the characteristic pink meat pigment, dinitrosylhemochrome. During fermentation, some strains of LAB may produce metabolites, such as carbon dioxide, acetate, formate, succinate, and acetoin, that cause off-flavors; however, nitrate and nitrite prevent the formation of certain off-flavors from compounds such as formate by their inhibition of pyruvate-formate lyase activity in lactobacilli (46). The formation of nitric oxide from the reduction of nitrates and nitrites is crucial because it reduces nitrosometmyoglobin to nitrosomyoglobin and induces red color formation (63). The advantage of microbial activity is the reduction of nitrates, thus removing excess nitrate/ nitrite from the meat.

Metabolic Activities Important in Commercial Starter Cultures

Strains of P. acidilactici were initially selected for use as starter cultures because they ferment carbohydrates rapidly to lactic acid at higher temperatures and effectively lower the pH of the meat, from a range of 5.6 to 6.2 down to a range of 4.7 to 5.2 (13). Traditionally, semidry sausages (summer sausage, thuringer, and beef sticks) and some types of pepperoni are allowed to undergo greening, i.e., fermentation at 16 to 38°C, which favors the development of indigenous microflora and extends fermentation times (13). Although indigenous microflora have the advantage of imparting unique flavor(s) and other desirable sensory properties to sausages, there is also sufficient variation in lactic acid production to be a considerable disadvantage for large-scale production of sausage. The use of starter cultures consisting of pediococci allows for the inoculation of large numbers of one defined microorganism into the raw meat, followed by a uniform fermentation. More importantly, meat inoculated with P. acidi­ lactici can be incubated at higher temperatures (43 to 50°C), which precludes the growth of most indigenous microorganisms so that flavor and pH characteristics are predictable (13). Consequently, commercial production is enhanced because flavor development can be controlled and the fermentation process hastened. Strains of P. acidilactici that have been selected and developed for use as commercial starter cultures are well suited for the production of semidry sausages, which are fermented and/or smoked at higher temperatures (26 to 50°C). However, when used for production of dry sausages, which require lower fermentation temperatures (15 to 27°C), these strains generally produce lactic acid at a much slower rate (13). Since P. pento­ saceus has a lower optimum growth temperature than

SMP_Food Microbiology_CH34.indd

873

873

P. acidilactici (28 to 32 versus 40°C) (37), it has been promoted as an effective meat starter culture. P. pen­ tosaceus is also more attractive because it has a 25% lower Arrhenius energy of activation for fermentation than that of P. acidilactici (82, 83). Thus, P. pentosa­ ceus has a lower optimum growth temperature and a higher capacity for rapid production of lactic acid than does P. acidilactici.

Metabolic Contributions of Starter Cultures to Sausage Sensory Qualities

Fermentation of sausages most often involves inoculation of ground meats with a starter culture, followed by an incubation period to allow enzymatic conversion of available carbohydrate to approximately equimolar concentrations of dl-lactic acid. Fermentable carbohydrates in the form of residual muscle glucose (0.1 mg/g) and added glucose or sucrose are the primary sources of energy available for metabolism. The decline in pH to 4.5 to 4.7 with the buildup of lactic acid results in a dramatic loss of water-holding capacity as the pH nears the isoelectric point (pI, ~5.1) of myofibrillar proteins. Product texture becomes firmer, density increases with dehydration during aging, and partial denaturation of the myofibrillar proteins occurs due to the presence of lactic acid. Thus, the tangy flavor and chewy texture of fermented sausages are consequences of the dominant fermentation metabolite, lactic acid, and dehydration. Lactic acid is the dominant flavor component in fermented sausage, but spices, salt, sugar, sodium nitrite reaction products, smoke, and meat components such as fatty acids, amino acids, and peptides are also contributors to the flavor profile (67). Secondary flavor contributions from metabolites of lactobacilli and pediococci are minor in fermented sausages produced in the United States; however, natural heterofermentative lactobacilli (Lb. brevis and Lb. buchneri) from back inoculations, when combined with staphylococci and kocuria, give many European sausages their characteristic flavor as a result of volatile acids, alcohols, and carbon dioxide. Micrococcaceae, for example, possess lipolytic and proteolytic enzymes that generate aldehydes, ketones, and short-chain fatty acids that give characteristic aromas and flavors to sausages (67). They also produce catalase, which decomposes hydrogen peroxide. L. mesenteroi­ des and Lb. brevis produce ethanol, acetic acid, lactic acid, pyruvic acid, acetoin (which imparts a nutty flavor and aroma), and carbon dioxide to give an effervescent sensation, but fermentation conditions must be controlled to avoid excessive pinholes, gas pockets, and off-flavors.

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

874 In natural fermentations as well as sausages inocu­ lated with lactobacilli, Lb. sake and Lb. curvatus have been observed to be the dominant bacteria at temperatures of 25°C and below, whereas Lb. plan­ tarum tends to dominate at higher fermentation temperatures (54). Lb. sake, Lb. curvatus, and to a lesser extent Lb. plantarum form dl-lactic acid and have flavin-dependent oxidases capable of forming hydrogen peroxide. Tolerance to peroxide may be one of the factors contributing to their dominance during fermentation. Hydrogen peroxide, if not decomposed, can induce oxidation of unsaturated fatty acids, leading to rancid flavor development, and/or oxidize the heme component of myoglobin, leading to fading or the formation of green, yellow, gray, or other offcolor pigments (71). Lb. curvatus does not possess pseudocatalase or true catalase and may permit the accumulation of hydrogen peroxide. Lb. sake, Lb. plantarum, and pediococci, however, exhibit catalase activity and can therefore decompose hydrogen peroxide formed during fermentation. Sensory analysis of unspiced fermented dry sausages produced using various starter culture combinations (18) has revealed that sausage flavor is influenced by culture composition. Berdague et al. (18) reported that the butter odor of dry sausages was largely dependent upon the degradation of sugars by way of pyruvic acid and that curing and rancid odors were correlated with compounds resulting from lipid oxidation. Their work indicated that S. saprophyticus and Staphylococcus warneri were most associated with butter odor, which in turn was correlated with the presence of acetoin, diacetyl, 1,3-butanediol, and 2,3-butanediol. Distinctive curing odors were associated with combinations of S. carnosus and P. acidilactici, S. carnosus and Lb. sake, and S. carnosus and P. pentosaceus and were correlated with 2-pentanone, 2-hexanone, 2-heptanone, and an unknown compound. A combination of S. sap­ rophyticus and Lb. sake, which produced the most acetic acid, was associated with fruity odors derived from esters and with a less intense rancid odor. Meat proteins are hydrolyzed by endogenous and microbial proteases to yield peptides and amino acids, which in turn are degraded to ammonia and amines, causing a slight rise in the pH. Demeyer and Samejima (27), however, suggested that the major protease activity in fermented meat is derived from the meat enzymes. Amino acid degradation products and nucleotide inosine monophosphate intensify meat flavors and contribute to the overall sausage flavor. Thus, it appears that compounds produced via endogenous proteolysis in combination with starter culture fermentation play

SMP_Food Microbiology_CH34.indd

874

a significant role in the nonacidic flavor and aroma of fermented sausages.

Genetics and Biotechnology of Meat Starter Cultures

To utilize biotechnological approaches with meat starter cultures, genetic transfer systems and mobile genetic elements for carrying a potentially important gene(s) must be identified. The standard approach to gene cloning in LAB is to develop plasmids based either on indigenous cryptic plasmids or heterologous plasmids resistant to a broad range of antibiotics (38). Since Chassy et al. (20) first detected the presence of plasmids in lactobacilli, significant progress has been made, particularly in elucidating genetic systems in dairy lactobacilli and L. lactis (38, 48). Many plasmids have been isolated from meat lactobacilli, and some have been functionally identified. Most Lactobacillus strains have one to several plasmids, with sizes ranging from 1.2 to 150 kb. Characteristics of the plasmids are diverse and depend on the replicons, modes of replication, and functions, including lactose metabolism, bacteriocin production, N-acetylglucosamine­ fermentation, and transport of cysteine (94). Lb. plantarum strains harbor several plasmids, including pMD5057 with tetracycline resistance (24) and pLKA with resistance to phages (31). Metabolic functions in Lb. plan­ tarum and four atypical Lactobacillus species isolated from fresh meat that are plasmid linked include maltose utilization (64) and cysteine metabolism (97). Schillinger and Lücke (92) screened 229 strains of lactobacilli from meat and meat products for antibacterial compounds such as bacteriocins and determined that Lb. sakei exhibited potential plasmid-associated bacteriocin production. Lb. sakei produces lactocin S, sakacin A, and sakacin P, which inhibit L. monocytogenes, S. aureus, and E. coli O157:H7 (19, 98, 100). Pediococcus spp. also contain many different plasmids, although there are relatively fewer vectors used for genetic engineering. Plasmids from P. pentosaceus and P. acidilactici are involved in utilization of raffinose, melibiose, and sucrose (41). In addition, production of the bacteriocin of Pediococcus, pediocin, is plasmid encoded by pSRQ11 and pSMB74 (94). Genetic transfer has been observed in several species of lactobacilli and lactococci, but limited studies have been conducted with species associated directly with meat fermentations. In vivo gene transfer systems have involved either conjugation, which requires cell-to-cell contact (bacterial mating) for transfer of genetic material, or transduction (transfection or phage-mediated genetic exchange), whereas in vitro physiological trans-

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products formation requires uptake of naked DNA by the recipient cell (49, 114). In vitro gene transfer of a tetracycline resistance plasmid from Lb. plantarum to other gram-positive bacteria has been demonstrated (39), and successful in vivo conjugation of a plasmid from Lb. plantarum to Enterococcus faecalis has also been reported (51). Conjugative transfer of plasmids has also been achieved in Lb. curvatus, both from E. faecalis as well as between organisms of the same Lb. curvatus strain, and the stability of the conjugated plasmid was assessed during meat fermentation (113). Further genetic analysis and manipulation of meat lactobacilli have been handicapped by a lack of natural competence, restriction/modification systems, and transformation procedures (48). In order to minimize these problems, other transformation methods, such as glass bead-based transformation (84) and electroporation methods, have been successively used to transform plasmids in Lb. plantarum, Pediococcus, and other gram-positive bacteria (2, 66, 81, 85). Although shuttle vectors have been used to construct and introduce heterologous genes into lactobacilli, they generally contain antibiotic resistance markers, which greatly facilitate the process but may not receive regulatory approval for use in foods (48). Furthermore, considerable instability and loss of function occurs during replication because these plasmids, for the most part, belong to a group that replicates via single-stranded DNA intermediates or rolling circle replication (38, 44). This form of replication is a potential source of plasmid segregational instability. For meat lactobacilli, vectors containing replicons from Lactobacillus, Staphylococcus, or E. coli sequences all contribute to this instability (60, 81). Therefore, plasmids containing a Lactobacillus replicon can be stabilized under selective conditions and used to express genes from multiple copies of the plasmid, but under nonselective conditions they are frequently segregationally unstable and are lost (60, 81). Consequently, efforts have focused on gene cloning strategies that enable heterologous genes to integrate in a stable fashion into the bacterial chromosome (38, 50). Recombination in chromosomes can occur as general recombination, when chromosomal DNA is transferred from one bacterium to the next via conjugation, or transduction, which necessitates extensive sequence homology between externally introduced DNA and chromosomal DNA as well as specific bacterial recombination factors (14). Recombination can also be mediated by transposable genetic elements that involve short target sequences of 10 nucleotide base pairs or less and occur independently of bacterial host function (114). Transposons have been introduced into Lb. curvatus (58), and integration of plasmids by single-crossover events within regions of

SMP_Food Microbiology_CH34.indd

875

875

homology can be accomplished by constructing plasmid suicide vectors that lack the ability to replicate in LAB (38). Scheirlinck et al. (91) used a suicide plasmid to integrate Bacillus stearothermophilus a-amylase genes and Clostridium thermocellum endoglucanase genes into Lb. plantarum. This homologous insertion into the chromosome was made possible by producing plasmid constructs containing an unknown portion of a Lactobacillus sequence as part of the suicide vector to facilitate site-specific integration. Complete replacement of a gene(s) can be achieved with homologous doublecrossover recombination events. An alternative to using homologous sequences is to locate insertion sequences already present on the chromosome that can transpose chromosomal DNA to plasmids and to use these in the construction of integrative vectors (49). Only a few such elements have been identified in meat lactobacilli, and only IS1163 and IS1520, from Lb. sakei, have been isolated, sequenced, and described in any detail (5, 99). Food-grade vectors have been developed to replace antibiotic resistance markers in food applications. Cryptic vectors are useful and safe; however, they can still cause problems with the use of an antibiotic marker. Bacteriocins, including nisin, have been commonly used for selection markers in food-grade vectors (94). P. aci­ dilactici and P. pentosaceus carry plasmids encoding pediocin, which can be applied for the construction of food-grade vectors. While this type of vector has not been developed as of yet, it offers considerable potential for application in fermented meats (40). Development of genetically engineered strains of lactobacilli that have large-scale use in fermented meats or other foods has remained elusive. Part of the problem is that transfer of structural genes to a new host can cause unstable target protein expression depending on the growth conditions (49). Optimal expression of cloned genes requires not just cloning of the structural gene but also efficient promoters, ribosomebinding sites, and termination sites, all of which must be identified, isolated, cloned, and sequenced (49). A vector-free chromosomal integration system is preferable due to the inherent stability of the transferred gene and the simplicity of cloning; however, this system does not produce high enough levels of heterologous gene expression for practical application to fermented meats, can yield unsatisfactory stabilization of the foreign gene, and can cause changes in bacterial physiology (50, 88). One approach to solving this problem involves using the amyL promoter, which codes for production and secretion of a-amylase from B. licheniformis, as a reporter gene for the cloning of expression-secretion sequences from Lb. plantarum

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

876 (50). A possible approach is to minimize variation of cell physiology by integrating the target gene into the prophage sequence, where transcription does not occur in the lysogenic bacteria (88). Inducible systems for nisin-controlled gene expression that utilize inducible promoters linked to promoterless reporter genes have been described for Lactococcus, and these appear to be applicable to expression systems in other LAB (119). To date, 62 LAB genomes (38 Lactobacillus, 9 Lacto­ coccus, 5 Streptococcus thermophilus, 6 Leuc­onostoc, 2 Pediococcus, and 2 Oenococcus) have been sequenced and published (Genomes OnLine Database [GOLD]: www.genomesonline.org/), and more than 100 LAB genomes will be sequenced in the near future. Genetic analysis has revealed alternative utilization of carbon sources for LAB, which fits with the limited sugar availability during meat fermentation. Genomic sequences generated from Table 34.5  Research needs for meat starter culturesa Improving current starter culture technology Substrate specificity of starter culture microorganisms Immobilized cells for meat fermentation Increased rate of lactic acid production (reduction of lag phase) Control of pathogens without heating Use of starter cultures in combination with acidulants, e.g., GDL Identification and testing of nitrite substitutes (development of cured color with C. botulinum outgrowth protection) New cryoprotectant (antifreeze) solutions for frozen starter cultures Development of new pediocins that are more soluble and are not bound by fat, subject to proteolysis, or inactivated by other meat components Production of biogenic amine-free (negative) meat starter culture Development of meat cultures to be grown at higher salt concentrations Production of starter cultures for pathogen and yeast control Rapid detection and identification using molecular detection techniques Improvement of starter cultures for higher level of nitrite reduction and better red color development Probiotic LAB (Lactobacillus acidophilus) applied to meat fermentation Use of cultures to enhance meat product nutrition and quality Enhancement of nutritional quality by in situ production of critical dietary nutrients for selected populations Greater utilization of nontraditional meat sources via fermentation Accelerated curing of traditional meat products (nonnitrite hams, bacon, comminuted meats, meat snacks, natural acidification against C. botulinum) Generation of natural antioxidants in situ (a-tocopherol production by starter cultures) Production of antimycotic agents for mold inhibition Development of starter cultures unique to fish pastes and sauces a

Based on comments from Curtis (personal communication).

SMP_Food Microbiology_CH34.indd

876

Lb. sakei have identified putative osmoprotectant and psychroprotectant proteins and other proteins associated with heme usage and oxidative stress responses. These stress tolerance genes appear to support competitive growth in harsh fermentative environments. Genomics should provide potential targets for strain improvement, thereby identifying strains that can ferment more rapidly and more consistently produce flavor compounds. Genomics may also help to identify important genes associated with development of individual flavors. Production of desired flavor compounds can be achieved by a better understanding of metabolic pathways based on genomics. These studies can also serve as a highly useful tool for identifying bacteriocinproducing strains because the encoding genes are small and diverse in their sequence composition and require other genes for regulation and transport (79). Although genetically modified organisms are not widely accepted, genomics may have a role through high-throughput screening and identification of improved starter cultures. However, studies at the transcriptional and translational level are required to fully understand the functionality of key genes. Application of metabolomic and metagenomic approaches should provide more detailed analysis of the complex nature of bacteria involved in meat fermentation. Fluorescence in situ hybridization approaches to analyze population dynamics and understand the microbial ecology during fermentation should help in developing more consistent fermentation processes. Many opportunities are available for advancing meat starter culture research, and some of the potential research areas are summarized in Table 34.5. Earlier predictions (94, 103) for biotechnological approaches to optimize fermentations are nearing reality.

References 1. Acton, J. C. 1977. The chemistry of dry sausages. Proc. Recip. Meat Conf. 30:49–62. 2. Alegre, M. T., M. C. Rodríguez, and J. M. Mesas. 2004. Transformation of Lactobacillus plantarum by electroporation with in vitro modified plasmid DNA. FEMS Microbiol. Lett. 241:73–77. 3. Al-Sheddy, I. A., D. Y. C. Fung, and C. L. Kastner. 1995. Microbiology of fresh and restructured lamb meat: a review. Crit. Rev. Microbiol. 21:31–52. 4. American Meat Institute Foundation. 1997. Good Manufacturing Practices for Fermented Dry and Semi-Dry Sausage Products. American Meat Institute Foundation, Washington, DC. 5. Ammor, M. S., M. Gueimonde, M. Danielsen, M. Zagorec, A. H. A. M. van Hoek, C. G. de los ReyesGavilán, B. Mayo, and A. Margolles. 2008. Two different tetracycline resistance mechanisms, plasmid-carried tet(L) and chromosomally located transposon-­associated

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products

6.

7.

8.

9. 10.

11. 12.

13.

14.

15.

16. 17. 18. 19.

20.

tet(M), coexist in Lactobacillus sakei Rits 9. Appl. Environ. Microbiol. 74:1394–1401. Arihara, K., H. Kushida, Y. Kondo, M. Itoh, J. B. Luchansky, and R. G. Cassens. 1993. Conversion of metmyoglobin to bright red myoglobin derivatives by Chromobacterium violaceum, Kurthia sp., and Lactobacillus fermentum JCM1173. J. Food Sci. 58:38–42. Association of Food and Drug Officials. 1999. Safe Practices for Sausage Production—Distance Learning Course Manual, version 1. Food Safety and Inspection Service, U.S. Department of Agriculture, Washington, DC. Axelsson, L. T. 2004. Lactic acid bacteria: classification and physiology, p. 1–66. In S. Salminen and A. von Wright (ed.), Lactic Acid Bacteria: Microbiological and Functional Aspects. Marcel Dekker, Inc., New York, NY. Bacus, J. N. 1986. Fermented meat and poultry products. Adv. Meat Res. 2:123–164. Bacus, J. N. 2005. Microbiology—shelf-stable dried meats. Food Safety and Inspection Service, U.S. Department of Agriculture, Washington, DC. www.fsis. usda.gov/PDF/FSRE_SS_5MicrobiologyDried.pdf. Bacus, J. N., and W. L. Brown. 1981. Use of microbial cultures: meat products. Food Technol. 35(1)74–78, 83. Bacus, J. N., and W. L. Brown. 1985. The lactobacilli: meat products, p. 58–71. In S. E. Gilliland (ed.), Bacterial Starter Cultures for Foods. CRC Press, Inc., Boca Raton, FL. Bacus, J. N., and W. L. Brown. 1985. The pediococci: meat products, p. 86–95. In S. E. Gilliland (ed.), Bacterial Starter Cultures for Foods. CRC Press, Inc., Boca Raton, FL. Bartholomew, D. R., and C. I. Osuala. 1986. Acceptability of flavor, texture, and appearance of mutton processed meat products made by smoking, curing, spicing, adding starter cultures and modifying fat source. J. Food Sci. 51:1560–1562. Beddows, C. G. 1998. Fermented fish and fish products, p. 416–440. In B. J. B. Wood (ed.), Microbiology of Fermented Foods, vol. II, 2nd ed. Elsevier Applied Science Publishing Co. Inc., London, United Kingdom. Beddows, C. G., A. G. Ardeshir, and W. Johari bin Daud. 1979. Biochemical changes occurring during the manufacture of budu. J. Sci. Food Agric. 30:1097–1103. Beddows, C. G., A. G. Ardeshir, and W. Johari bin Daud. 1980. Development and origin of the volatile fatty acids in budu. J. Sci. Food Agric. 31:86–92. Berdague, J. L., P. Monteil, M. C. Montel, and R. Talon. 1993. Effects of starter cultures on the formation of flavor compounds in dry sausage. Meat Sci. 35:275–287. Champomier-Vergès, M. C., S. Chaillou, M. Cornet, and M. Zagorec. 2002. Lactobacillus sakei: recent developments and future prospects. Res. Microbiol. 153:115–123. Chassy, B. M., E. Gibson, and A. Giuffrida. 1976. Evidence for extrachromosomal elements in Lactobacillus. J. Bacteriol. 127:1576–1578.

SMP_Food Microbiology_CH34.indd

877

877

21. Claesson, M. J., D. van Sinderen, and P. W. O’Toole. 2008. Lactobacillus phylogenomics—towards a reclassification of the genus. Int. J. Syst. Evol. Microbiol. 58:2945–2954. 22. Crisan, E. V., and A. Sands. 1975. The microbiology of four fermented fish sauces. Appl. Microbiol. 29:106–108. 23. Curtis, S. I. 1995. Personal communication. 24. Danielsen, M. 2002. Characterization of the tetracycline resistance plasmid pMD5057 from Lactobacillus plantarum 5057 reveals a composite structure. Plasmid 48:98–103. 25. Deibel, R. H., and C. F. Niven, Jr. 1957. Pediococcus cerevisiae: a starter culture for summer sausage. Bacteriol. Proc. 1957:14–15. 26. Deibel, R. H., G. D. Wilson, and C. F. Niven, Jr. 1961. Microbiology of meat curing. IV. A lyophilized Pediococcus cerevisiae starter culture for fermented sausages. Appl. Microbiol. 9:239–243. 27. Demeyer, D., and K. Samejima. 1991. Animal biotechnology and meat processing, p. 127–143. In L. O. Fiems, B. G. Cottyn, and D. I. Demeyer (ed.), Animal Biotechnology and the Quality of Meat Production. Elsevier, Amsterdam, The Netherlands. 28. de Vries, W., W. M. C. Kapteijn, E. G. van der Beek, and A. H. Stouthamer. 1970. Molar growth yields and fermentation balance of Lactobacillus casei L3 in batch cultures and continuous cultures. J. Gen. Microbiol. 63:333–345. 29. Dougan, J., and G. E. Howard. 1975. Some flavouring constituents of fermented fish sauces. J. Sci. Food Agric. 26:887–894. 30. Egan, A. F. 1983. Lactic acid bacteria of meat and meat products. Antonie van Leeuwenhoek 49:327–336. 31. Eguchi, T., K. Doi, K. Nishiyama, S. Ohmomo, and S. Ogata. 2000. Characterization of a phage resistance plasmid, pLKS, of silage-making Lactobacillus plantarum NGRI0101. Biosci. Biotechnol. Biochem. 64:751–756. 32. Everson, C. W., W. E. Danner, and P. A. Hammes. 1970. Bacterial starter cultures in sausage products. J. Agric. Food Chem. 18:570–571. 33. Everson, C. W., W. E. Danner, and P. A. Hammes. 1974. Process for curing dry and semidry sausages. U.S. patent 3,814,817. 34. Fadda, S., C. Lopez, and G. Vignolo. 2010. Role of lactic acid bacteria during meat conditioning and fermentation: peptides generated as sensorial and hygienic biomarkers. Meat Sci. 86:66–79. 35. Felis, G. E., and F. Dellaglio. 2007. Taxonomy of lactobacilli and bifidobacteria. Curr. Issues Intest. Microbiol. 8:44–61. 36. Fujii, T., S. D. Basuki, and H. Tozawa. 1980. Microbiological studies on the ageing of fish sauce; chemical composition and microflora of fish sauce produced in the Philippines. Nippon Suissan Gakkaishi 46:1235–1240. 37. Garvie, E. I. 1994. Genus Pediococcus, p. 530. In D. H. Bergey and J. G. Holt (ed.), Bergey’s Manual of

Manila Typesetting Company

11/08/2012 08:54PM

Fermentations and Beneficial Microorganisms

878

38. 39.

40.

41.

42. 43. 44. 45. 46. 47. 48.

49.

50.

51.

52.

Determinative Bacteriology, 9th ed. The Williams and Wilkins Co., Baltimore, MD. Gasson, M. J. 1993. Progress and potential in the biotechnology of the lactic acid bacteria. FEMS Microbiol. Rev. 12:3–20. Gevers, D., G. Huys, and J. Swings. 2003. In vitro conjugal transfer of tetracycline resistance from Lactobacillus isolates to other Gram-positive bacteria. FEMS Microbiol. Lett. 225:125–130. Giacomini, A., A. Squartini, and M. P. Nuti. 2000. Nucleotide sequence and analysis of plasmid pMD136 from Pediococcus pentosaceus FBB61 (ATCC43200) involved in Pediocin A production. Plasmid 43:111–122. Gonzalez, C. F., and B. S. Kunka. 1986. Evidence for plasmid linkage of raffinose utilization and associated a-galactosidase and sucrose hydrolase activity in Pediococcus pentosaceus. Appl. Environ. Microbiol. 51:105–109. Gottschalk, G. 1986. Bacterial Metabolism, 2nd ed. Springer-Verlag, New York, NY. Grazia, L., P. Romano, A. Bagni, D. Roggiani, and G. Guglielmi. 1986. The role of moulds in the ripening process of salami. Food Microbiol. 3:19–25. Gruss, A., and S. D. Ehrlich. 1989. The family of highly interrelated single-stranded deoxyribonucleic acid plasmids. Microbiol. Rev. 53:231–241. Hammer, G. F. 1987. Meat processing: ripened products. Fleischwirtschaft 67:71–74. Hammes, W. P., A. Bantleon, and S. Min. 1990. Lactic acid bacteria in meat fermentation. FEMS Microbiol. Lett. 87:165–174. Hammes, W. P., and H. J. Knauf. 1994. Starters in the processing of meat products. Meat Sci. 36:155–168. Hammes, W. P., and C. Hertel. 2006. The genera Lactobacillus and Carnobacterium, p. 320–403. In M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt (ed.), The Prokaryotes—a Handbook on the Biology of Bacteria: Bacteria, Firmicutes, Cyanobacteria, 3rd ed. Springer-Verlag, New York, NY. Harlander, S. K. 1992. Genetic improvement of microbial starter cultures, p. 20–26. In National Research Council (ed.), Applications of Biotechnology to Traditional Fermented Foods. National Academy Press, Washington, DC. Hols, P., T. Ferain, D. Garmyn, N. Bernard, and J. Delcour. 1994. Use of homologous expression-secretion signals and vector-free stable chromosomal integration in engineering of Lactobacillus plantarum for a-amylase and levanase expression. Appl. Environ. Microbiol. 60:1401–1413. Jacobsen, L., A. Wilcks, K. Hammer, G. Huys, D. Gevers, and S. R. Andersen. 2007. Horizontal transfer of tet(M) and erm(B) resistance plasmids from food strains of Lactobacillus plantarum to Enterococcus fae­ calis JH2-2 in the gastrointestinal tract of gnotobiotic rats. FEMS Microbiol. Ecol. 59:158–166. Jay, J. M. 2003. A review of recent taxonomic changes in seven genera of bacteria commonly found in foods. J. Food Prot. 66:1304–1309.

SMP_Food Microbiology_CH34.indd

878

53. Jay, J. M., M. J. Loessner, and D. A. Golden. 2005. Modern Food Microbiology, 7th ed., p. 175–196. Springer, New York, NY. 54. Kagermeier, A. 1981. Taxonomie und Vorkommen von Milchsaurebakterien in Fleischprodukten. Dissertation. Fakultat fur Biologie, Ludwig-Maximilian-Universität München, Munich, Germany. 55. Kandler, O. 1983. Carbohydrate metabolism in lactic acid bacteria. Antonie von Leeuwenhoek 49:209–224. 56. Kandler, O., and N. Weiss. 1994. Regular, non-sporing gram-positive rods, p. 565–570. In D. H. Bergey and J. G. Holt (ed.), Bergey’s Manual of Determinative Bacteriology, 9th ed. The Williams and Wilkins Co., Baltimore, MD. 57. Klettner, P.-G., and D. List. 1980. Beitrag zum Einfluss der Kohlenhydratart auf den Verlauf der Rohwurstreifung. Fleischwirtschaft 60:1589–1593. 58. Knauf, H. J., R. F. Vogel, and W. P. Hammes. 1989. Introduction of the transposon Tn919 into Lactobacillus curvatus. FEMS Microbiol. Lett. 65:101–104. 59. Languer, H. J. 1972. Aromastoffe in der Rohwurst. Fleischwirtschaft 52:1299–1306. 60. Leer, R. J., N. van Luijk, M. Posno, and P. H. Pouwels. 1992. Structural and functional analysis of two cryptic plasmids from Lactobacillus pentosus MD353 and Lactobacillus plantarum ATCC 8014. Mol. Gen. Genet. 234:265–274. 61. Leistner, L. 1986. Mould-ripened foods. Fleischwirtschaft 66:1385–1388. 62. Leroy, F., J. Verluyten, and L. De Vuyst. 2006. Functional meat starter cultures for improved sausage fermentation. Int. J. Food Microbiol. 106:270–285. 63. Liepe, H. U. 1983. Starter cultures in meat production, p. 400–424. In H.-J. Rehm and G. Reed (ed.), Biotechnol­ ogy, Food and Feed Production with Microorganisms, vol. 5. Verlag Chemie, Weinheim, Germany. 64. Liu, M.-L., J. K. Kondo, M. B. Barnes, and D. T. Bartholomew. 1988. Plasmid-linked maltose utilization in Lactobacillus spp. Biochimie 70:351–355. 65. London, J. 1990. Uncommon pathways of metabolism among lactic acid bacteria. FEMS Microbiol. Lett. 87:103–112. 66. Luchansky, J. B., P. M. Muriana, and T. R. Klaen­ hammer. 1988. Application of electroporation for transfer of plasmid DNA to Lactobacillus, Lactococcus, Leuconostoc, Listeria, Pediococcus, Bacillus, Staphy­ lococcus, Enterococcus and Propionibacterium. Mol. Microbiol. 2:637–646. 67. Lücke, F.-K. 1998. Fermented sausages, p. 441–483. In B. J. B. Wood (ed.), Microbiology of Fermented Foods, vol. II, 2nd ed. Elsevier Applied Science Publishing Co., Inc., London, United Kingdom. 68. Makarova, K. S., and E. V. Koonin. 2007. Evolutionary genomics of lactic acid bacteria. J. Bacteriol. 189:1199–1208. 69. McQuestin, O. J., C. T. Shadbolt, and T. Ross. 2009. Quantification of the relative effects of termperature, pH, and water activity on inactivation of Escherichia

Manila Typesetting Company

11/08/2012 08:54PM

34.  Fermented Meat, Poultry, and Fish Products

70.

71.

72.

73. 74. 75. 76. 77. 78. 79. 80. 81.

82. 83. 84. 85.

86. 87.

coli in fermented meat by meta-analysis. Appl. Environ. Microbiol. 75:6963–6972. Michels, P. A. M., J. P. J. Michels, J. Boonstra, and W. N. Konings. 1979. Generation of electrochemical proton gradient in bacteria by the extrusion of metabolic end products. FEMS Microbiol. Lett. 5:357–364. Mogensen, G. 2004. Starter cultures, p. 191–210. In J. Smith (ed.), Technology of Reduced‑Additive Foods, 2nd ed. Blackie Academic and Professional, Chapman and Hall, New York, NY. National Center for Biotechnology Information Taxonomy Database. Lactobacillus taxonomy. www. ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax. cgi?id=1578. Accessed 13 August 2012. National Research Council (ed.). 1992. Applications of Biotechnology to Traditional Fermented Foods, p. 121–149. National Academy Press, Washington, DC. Niinivaara, F. P. 1955. The influence of pure cultures of bacteria on the maturing and reddening of raw sausage. Acta Agric. Fenn. 85:95–101. Niven, C. F., Jr. 1951. Sausage discolorations of bacterial origin. Bulletin no. 13. American Meat Institute Foundation, Chicago, IL. Niven, C. F., Jr., R. H. Deibel, and G. D. Wilson. 1958. The AMIF sausage starter culture. Circular no. 41. American Meat Institute Foundation, Chicago, IL. Palumbo, S. A., and J. L. Smith. 1977. Lebanon bologna processing. Proc. Recip. Meat Conf. 30:63–68. Pearson, A. M., and W. F. Tauber. 1999. Processed Meats, 3rd ed. Chapman and Hall, New York, NY. Pfeiler, E. A., and T. R. Klaenhammer. 2007. The genomics of lactic acid bacteria. Trends Microbiol. 15:546–553. Poolman, B. 1993. Energy transduction in lactic acid bacteria. FEMS Microbiol. 12:125–148. Posno, M., R. J. Leer, N. van Luijk, M. J. F. van Giezen, P. T. H. M. Heuvelmans, B. C. Lokman, and P. H. Pouwels. 1991. Incompatibility of Lactobacillus vectors with replicons derived from small cryptic Lactobacillus plasmids and segregational instability of the introduced vectors. Appl. Environ. Microbiol. 57:1822–1828. Raccach, M. 1984. Method for selection of lactic acid bacteria and determination of minimum temperature for meat fermentations. J. Food Prot. 47:670–671. Raccach, M. 1987. Pediococci and biotechnology. Crit. Rev. Microbiol. 14:291–309. Rattanachaikunsopon, P., and P. Phumkhachorn. 2009. Glass bead transformation method for Gram-positive bacteria. Braz. J. Microbiol. 40:923–926. Rodriguez, M. C., M. T. Alegre, and J. M. Mesas. 2007. Optimization of technical conditions for the transformation of Pediococcus acidilactici P60 by electroporation. Plasmid 58:44–50. Rogosa, M., and M. E. Sharpe. 1959. An approach to the classification of the lactobacilli. J. Appl. Bacteriol. 22:329–340. Romans, J. R., W. J. Costello, C. W. Carlson, M. L. Greaser, and K. W. Jones. 2000. The Meat We Eat, 14th ed., p. 773–886. Interstate Publishers, Inc., Danville, IL.

SMP_Food Microbiology_CH34.indd

879

879

88. Rossi, F., A. Capodaglio, and F. Dellaglio. 2008. Genetic modification of Lactobacillus plantarum by heterologous gene integration in a not functional region of the chromosome. Appl. Microbiol. Biotechnol. 80:79–86. 89. Rust, R. E. 1976. Sausage and Processed Meats Manufacturing. American Meat Institute, Washington, DC. 90. Sameshima, T., C. Magome, K. Takeshima, K. Arihara, M. Itoh, and Y. Kondo. 1998. Effect of intestinal Lactobacillus starter cultures on the behavior of Staphylooccus aureus in fermented sausage. Int. J. Food Microbiol. 41:1–7. 91. Scheirlinck, T., J. Mahillon, H. Joos, P. Dhaese, and F. Michiels. 1989. Integration and expression of aamylase and endoglucanase genes in the Lactobacillus plantarum chromosome. Appl. Environ. Microbiol. 55:2130–2137. 92. Schillinger, U., and F.-K. Lücke. 1987. Identification of lactobacilli from meat and meat products. Food Microbiol. 4:199–208. 93. Schleifer, K. H. 1994. Gram-positive cocci, p. 527–558. In D. H. Bergey and J. G. Holt (ed.), Bergey’s Manual of Determinative Bacteriology, 9th ed. The Williams and Wilkins Co., Baltimore, MD. 94. Shareck, J., Y. Choi, B. Lee, and C. B. Miguez. 2004. Cloning vectors based on cryptic plasmids isolated from lactic acid bacteria: their characteristics and potential applications in biotechnology. Crit. Rev. Biotechnol. 24:155–208. 95. Sharpe, M. E. 1992. The genus Lactobacillus, p. 1653– 1679. In M. P. Starr, H. G. Trüper, A. Balows, and H. G. Schlegel (ed.), The Prokaryotes—a Handbook on Habitats, Isolation, and Identification of Bacteria, vol. II, 2nd ed. Springer-Verlag, New York, NY. 96. Sharpe, M. E., T. F. Fryer, and D. G. Smith. 1981. Identification of the lactic acid bacteria, p. 233–259. In B. M. Gibbs and F. A. Skinner (ed.), Identification Methods for Microbiologists, part A, 2nd ed. Academic Press, Ltd., London, United Kingdom. 97. Shay, B. J., A. F. Egan, M. Wright, and P. J. Rogers. 1988. Cysteine metabolism in an isolate of Lactobacillus sake: plasmid composition and cysteine transport. FEMS Microbiol. Lett. 56:183–188. 98. Skaugen, M., E. L. Andersen, V. H. Christie, and I. F. Nes. 2002. Identification, characterization, and expression of a second, bicistronic, operon involved in the production of lactocin S in Lactobacillus sakei L45. Appl. Environ. Microbiol. 68:720–727. 99. Skaugen, M., and I. F. Nes. 1994. Transposition in Lactobacilli sake and its abolition of lactocin S production by insertion of IS1163, a new member of the IS3 family. Appl. Environ. Microbiol. 60:2818–2825. 100. Slavica, V. M., D. Obradovic, B. Velebit, B. Branka, S. Marija, and L. Turubatovic. 2010. Antimicrobial properties of indigenous Lactobacillus sakei strain. Acta Vet. (Beogr.) 60:59–66. 101. Stackebrandt, E., and M. Teuber. 1988. Molecular taxonomy and phylogenetic position of lactic acid bacteria. Biochimie 70:317–324.

Manila Typesetting Company

11/08/2012 08:54PM

880 102. 103.

104.

105. 106.

107. 108.

109.

110.

880

Stiles, M. E., and W. H. Holzapfel. 1997. Lactic acid bacteria of foods and their current taxonomy. Int. J. Food Microbiol. 36:1–29. Sybesma, W., J. Hugenholtz, W. M. de Vos, and E. J. Smid. 2006. Safe use of genetically modified lactic acid bacteria in food. Bridging the gap between consumers, green groups, and industry. Electron. J. Biotechnol. 9:424–448. ten Brink, B. R. Otto, U. P. Hansen, and W. N. Konings. 1985. Energy recycling by lactate efflux in growing and nongrowing cells of Streptococcus cremoris. J. Bacteriol. 162:383–390. Terrell, R. N., G. C. Smith, and Z. L. Carpenter. 1977. Practical manufacturing technology for dry and semidry sausage. Proc. Recip. Meat Conf. 30:39–44. Townsend, W. E., C. E. Davis, and C. E. Lyon. 1978. Some properties of fermented dry sausage prepared from PSE and normal pork. In Kongressdocumentation. 24th Europäischer Fleischforscher-Kongress, Kulmbach, Germany. Tyopponen, S., E. Petäjä, and T. Mattila-Sandholm. 2003. Bioprotectives and probiotics for dry sausages. Int. J. Food Microbiol. 83:233–244. U.S. Department of Agriculture, Food Safety and Inspection Service. 1995. Prescribed treatment for pork and products containing pork to destroy trichinae. Code of Federal Regulations, Title 9, Part 318.10. Office of the Federal Register, Washington, DC. U.S. Department of Agriculture, Food Safety and Inspection Service. 1995. Requirements for the production of cooked beef, roast beef, and cooked corn beef. Code of Federal Regulations, Title 9, Part 318.17. Office of the Federal Register, Washington, DC. U.S. Department of Agriculture, Food Safety and Inspection Service. 1999. Appendix A. Compliance guidelines for meeting lethality performance standards for certain meat and poultry products. U.S. Department of Agriculture, Food Safety and Inspection Service, Wash­ ington, DC. www.fsis.usda.gov/oa/fr/95033f-a.htm.

SMP_Food Microbiology_CH34.indd

Fermentations and Beneficial Microorganisms 111.

112. 113.

114.

115. 116.

117.

118. 119.

11/08/2012 08:54PM

U.S. Department of Agriculture, Food Safety and Inspection Service. 2005. Food Standards and Labeling Policy Book. U.S. Department of Agriculture, Food Safety and Inspection Service, Office of Policy, Program and Employee Development, Washington, DC. www .fsis.usda.gov/OPPDE/larc/Policies/Labeling_Policy_ Book_082005.pdf. Vandekerckhove, P., and D. Demeyer. 1975. Die Zusammernstzung belgischer Rohwurst (Salami). Fleischwirtschaft 55:680–682. Vogel, R. F., M. Becke-Schmid, P. Entgens, W. Gaier, and W. P. Hammes. 1992. Plasmid transfer and segregation in Lactobacillus curvatus LTH1432 in vitro and during sausage fermentations. Syst. Appl Microbiol. 15:129–136. von Wright, A., and M. Sibakov. 1993. Genetic modification of lactic acid bacteria, p. 161–198. In S. Salminen and A. von Wright (ed.), Lactic Acid Bacteria. Marcel Dekker, Inc., New York, NY. Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51:221–271. Wolf, G., E. K. Aarendt, U. Pfahler, and W. P. Hammes. 1990. Heme-dependent and heme-independent nitrite reduction by lactic acid bacteria results in different N-containing products. Int. J. Food Microbiol. 10:323–330. Wu, W. J., D. C. Rule, J. R. Busboom, R. A. Field, and B. Ray. 1991. Starter culture and time/temperature of storage influences on quality of fermented mutton sausage. J. Food Sci. 56:919–925. Zenitani, B. 1955. Studies on fermented fish products. I. On the aerobic bacteria in “Shiokara.” Bull. Jpn. Soc. Sci. Fish. 21:280–283. Zhou, X. X., W. F. Li, G. X. Ma, and Y. J. Pan. 2006. The nisin-controlled gene expression system: construction, application and improvements. Biotechnol. Adv. 24:285–295.

Manila Typesetting Company

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch35

Sterling S. Thompson Kenneth B. Miller Alex S. Lopez Nicholas Camu

Cocoa and Coffee

Cocoa and coffee are two of the many foods that rely on a microbial curing process or fermentation for flavor development. The popularity and worldwide appeal of these products are due primarily to their unique flavors and aromas. Although a primary curing process is conducted in the preparation of each product before marketing, fermentation of cocoa is absolutely essential for flavor development whereas with coffee the curing process is less crucial to flavor and more important for the removal of pulp. Consequently, this chapter focuses mainly on the more comprehensive role of fermentation in cocoa curing and to a lesser extent on the role of fermentation in the production of coffee.

COCOA PROCESSING Commercial cocoa is derived from the seeds (beans) of the ripe fruit (pods) of the plant Theobroma cacao, which is native to the Amazon region of South America. It has been used by the Amerindians to produce a beverage since time immemorial and was introduced to Europe in the 15th century by Cortez during the period of discovery and colonization of the Americas. Its popu-

35 larity and demand led to the establishment and spread of rootstock to virtually all of the European colonies located between 15° north and 15° south of the equator with climates that could support cocoa production. Of the Theobroma species, only T. cacao produces beans suitable for chocolate manufacturing. Immediately following harvesting of the ripe fruit, or following a brief storage period, the seeds are removed and subjected sequentially to fermentation and drying processes often referred to as “curing” that are carried out on farms, estates, or cooperatives in the producing countries. The origin of these processes has been lost in antiquity, but it was believed at one time that fermentation was conducted simply to aid in removing the mucilaginous pulp surrounding the seed so as to facilitate drying and storage, as in the case of coffee. This in fact is one purpose of fermentation, but the main reason for fermentation of cocoa is to induce biochemical transformations within the beans that lead to the formation of color, aroma, and flavor precursors of chocolate. Without this step, cocoa beans are excessively bitter and astringent and when processed do not develop the flavor that is characteristic of chocolate. The characteristics and strength

Sterling S. Thompson and Kenneth B. Miller, Microbiology Research Technical Center, Hershey Foods Corporation, 1025 Reese Avenue, Hershey, PA 17033-0805. Alex S. Lopez, 6621 Creeping Thyme Street, Las Vegas, NV 89148. Nicholas Camu, Cocoa Fermentation Department, Barry Callebaut, Aalstersestraat 122, 9280 Lebbeke, Belgium.

881

882

Fermentations and Beneficial Microorganisms

of chocolate flavor are governed primarily by the genetics of the cocoa variety, and the fermentation process releases and develops this flavor potential (47). The inherited characteristics of the bean therefore set a limit to what can be achieved by fermentation. It is impossible to improve genetically inferior material by superior processing techniques; on the other hand, it is quite easy to ruin good-quality cocoa by inadequate curing. The cocoa fruit varies among varieties in size, shape, external color, and appearance. These characteristics have often been used in classifying cocoa, but as far as the flavor quality is concerned, the only really important morphological differences are those that distinguish between the white-seeded Criollo variety of South and Central America and the purple-seeded Forastero variety of the Amazon. The former type is the source of the original “fine” cocoa and has almost disappeared from the market because of its susceptibility to disease; its lower productivity; and its replacement by the hardier, more prolific Forastero variety and varietal crosses, which now account for over 95% of the world production. Hence, the following discussion refers primarily to the processing of Forastero cocoa. Flowers are produced seasonally from cushions that emerge on the bark of the trunk and stems. Fertilized flowers bear fruit 170 days after pollination, a period during which the fruit grows to maturity and changes color from green or dark red-purple to yellow, orange, or red, depending on the variety. The mature fruits are thick walled and contain 30 to 40 beans, each enveloped in a sweet, white, mucilaginous pulp and loosely attached to an axial placenta. Only the beans are used in chocolate manufacturing. For the purpose of describing the curing process, the bean may be envisaged as comprising two main parts, namely, the testa (seed coat), together with the attached sugary, mucilaginous pulp that surrounds it; and the embryo, or the cotyledons, contained within. The mucilage, containing sugars and citric acid (78), serves as a substrate for the microorganisms that are involved in the natural fermentation process; the cotyledons, referred to as the nib in the cured bean, are used in chocolate manufacturing. Processing begins with the harvesting of healthy ripe fruits, an operation carried out over a period of 3 to 4 days at a frequency that varies according to the size of the farm and yield. Fermentation is a batch-type process, and harvesting is conducted to allow for the accumulation of sufficient material for each batch while taking precautions that, in the process, pods do not overripen and the seeds within do not germinate. The pods are usually collected in piles in the field and broken open on site or at the processing plant (fermentary) at the

end of the harvesting operation. The beans are removed manually in some cases; alternatively, on some large estates in West African countries, Mexico, and Brazil, pod breaking and bean extraction are mechanized (53). Once removed from the pods, the seeds are aggregated in heaps or in receptacles of one kind or another and left to ferment for a period of 2 to 8 days. During this interval, microorganisms that are transferred to the seeds from laborers’ hands, fruit surfaces, and containers used in transport and fermentation degrade the cells of the mucilage that surround the bean (57). The collective microbial activity resulting from the accidental inoculation by a multitude of microorganisms is referred to as fermentation or sweating. This process results in the liberation of pulp juices from which alcohols and acids are produced with the evolution of heat. During fermentation, concentrations of ethanol and lactic and acetic acids sequentially increase and decrease. This can result in an excess of acids that remains in the bean at the end of the bean fermentation (79). Together, these factors provoke changes and affect the curing of the bean. The two principal objectives of fermentation are to remove mucilage, thus provoking aeration during fermentation of the beans and facilitating drying later on; and to provide heat and acetic acid necessary for inhibiting germination, which ensures proper curing of the beans (48).

Methods of Fermentation

The manner in which cacao is fermented varies considerably from country to country, and in many instances even adjacent farms may adopt different curing methods. A substantial amount of research has been devoted to fermentation practices. Today, many of the primitive methods such as fermentation in banana leaf-lined holes in the ground, in derelict canoes, and in makeshift banana and bamboo frames are the exception rather than the rule. In general, large farms with adequate production of cacao will opt for permanent facilities specifically constructed for this purpose. In such instances, fermentation is carried out in batteries of wooden or fiberglass boxes. However, most of the world’s cocoa is produced on small farm holdings under very rural conditions, and the relatively small volumes produced do not always merit permanent processing facilities. In this case, cocoa beans are fermented in any convenient receptacle such as fruit boxes, baskets, plastic buckets, and fertilizer bags, or when these are not readily available, the beans are simply piled on a sheet and covered with any handy material. On the whole, however, the majority of the world’s cocoa is fermented on drying platforms, in heaps covered with banana leaves, in baskets, or in an assortment of wooden boxes (24, 40, 68).

35.  Cocoa and Coffee

883

Approximately one-half of the world crop is fermented in some type of box, and the remaining half is fermented by using heaps or other primitive methods.

those used for small heaps. Basket fermentation is often used when fermentation in heaps leaves beans liable to predial larceny.

Fermentation on Drying Platforms

Fermentation in Boxes

Fermentation on drying platforms is practiced in parts of Central America where Criollo cocoa was once grown. Wet cocoa beans are spread directly onto drying platforms where they ferment and dry during the day and are heaped into piles each night to conserve heat and retard the growth of surface molds. Criollo cocoa requires only a short fermentation for flavor development, about 2 to 3 days. Forastero varieties require a fermentation time of 5 to 8 days for the development of flavor (25). Although Criollo cocoa has been largely replaced by Forastero hybrids in these countries, in many instances the old method of fermentation still persists (68). This practice preserves the fine flavor characteristics of Criollo beans; however, it is inappropriate for Forastero varieties, which require longer fermentation times for optimal flavor development. Fermenting cocoa on the drying floor is convenient, but unless properly managed, the process tends to produce underfermented cocoa with the added danger of undesirable mold growth, which will cause the development of off-flavor.

Fermentation in Heaps

Fermentation in heaps is a popular method among smallholding farmers in Ghana and many other African cocoa-producing countries. It is also observed sporadically in the Amazon region of Brazil. This method does not require a permanent structure and is well suited to family holdings with a small production. Judging from Ghanaian cocoa, fermentation in heaps can produce good-quality products. Varying quantities of cocoa beans from 25 to 1,000 kg are heaped in the field on banana or plantain leaves and covered with the same material. The beans are mixed (turned) periodically to ensure even fermentation and to decrease the potential for mold growth. This is often done daily or every other day by forming another heap. Mixing is laborious, and small heaps may not be turned at all. The duration of fermentation can be from 4 to 8 days.

Fermentation in Baskets

Fermentation in baskets is practiced principally by small-scale producers in Nigeria, the Amazon region, the Philippines, and some parts of Ghana. Small lots of cocoa are placed in woven baskets lined with plantain leaves. The surface is covered with plantain leaves, and the leaves are weighted down to hold them in place. The turning procedure and the fermentation are similar to

Fermentation in boxes is considered to be an improvement over other methods. This batch process requires a fixed volume of beans and is the method of choice on large estates. The sizes of the containers vary from region to region, but the design and function are standard. The container, or “sweat box,” may be a single unit or one of a number of compartments within a large box created by subdividing the space into units measuring approximately 1 by 1 by 1 m with either fixed or movable internal partitions. These boxes hold between 600 and 700 kg of freshly harvested (wet) cocoa beans. The box is always raised above ground level and placed over a drain that carries away the pulp juices (sweatings) liberated by the degradation of the mucilage during fermentation. The wooden floor of the box generally has holes or spaces between the boards or slats to facilitate drainage and aeration. Sweat boxes vary considerably in size from that of a small fruit box (0.4 by 0.4 by 0.5 m) to boxes measuring 7 by 5 by 1 m used on some Malaysian estates (36). Large estates and cooperatives often have batteries of 20 to 30 sweat boxes arranged in tiers in three to seven rows, one below the next to facilitate mixing or turning. Mixing is achieved by simply removing a dividing wall and shoveling the beans into the next box or, in the case of the tier design, into the box below. On some Malaysian estates, boxes are built on pallets and a forklift is used to transfer the contents into an empty box. Variations occur not only in the size of the sweat box but also in the type of wood used in construction, in methods of drainage and aeration, and in duration of fermentation. The recommendation is to ferment a 1-m3 volume for 6 to 7 days with two to three mixings during this period. In the majority of cases, boxes are filled to within 10 cm of the top and the surface is covered with a padding of banana leaves or jute sacking to help retain the heat and prevent the surface beans from drying. In some countries, fermentation norms have been modified in an attempt to overcome problems such as acidity by varying the prefermentation treatment and the depth of beans in the sweat boxes (47). In Malaysia, for instance, harvested cocoa pods are stored for up to 15 days before breaking and removing the beans, or the beans are pressed or predried to reduce the pulp volume before fermentation (6, 20). The progress of the fermentation is assessed by the odor and the external and internal color changes in the

884

Fermentations and Beneficial Microorganisms

beans. When the process is judged to be complete, the beans are dried in the sun or in mechanical dryers.

Microbiology of Cocoa Fermentation

Fermentation begins immediately after beans are removed from the pods, as they become inoculated with a variety of microorganisms from the pod surface, knives, laborers’ hands, containers used to transport the beans to the fermentary, dried mucilage on surfaces of the fermentation box (tray, platform, or basket) from the previous fermentation, insects, and banana or plantain leaves (29, 30, 40, 54, 57, 65, 73). It is the pulp surrounding the bean, not the cocoa bean, that undergoes microbial fermentation. Chemical changes take place within the bean as a result of the fermentation of the pulp. The testa of the bean acts as a natural barrier between microbial fermentation activities outside the bean and chemical reactions within the bean. However, there is a migration of ethanol, acetic acid, lactic acid, and water of microbial origin from the outside to the inside of the bean. After the bean dies, soluble bean components leach through the skin and are lost in the drainings. The pulp consists of about 85% water, 2.7% pentosans, 0.7% sucrose, 10% glucose and fructose, 0.6% protein, 0.7% acids, and 0.8% inorganic salts (33), making it a rich substrate for microbial growth. The concentrations of sucrose, glucose, and fructose in the bean are influenced by the age of the pod (74). The initial microbial population is variable in number and type; however, the key groups active during fermentation are yeasts, lactic acid bacteria (LAB), and acetic acid bacteria (AAB) (44, 57, 66, 82). Climatic conditions may influence the sequence of microorganisms involved in the fermentation (82). It is hypothesized that Bacillus species play an important role during the latter stages of the fermentation in Brazil and become the dominant group during drying (82). More research is needed to confirm this hypothesis. More than 100 aerobic sporeforming bacteria were isolated from cocoa bean fermentations in Brazil. Bacteria were identified as the Bacillus species B. subtilis, B. licheniformis, B. firmus, B. coagulans, B. pumilus, B. marcerans, B. polymyxa, B. laterosporus, B. stearothermophilus, B. circulans, B. pasteurii, B. megaterium, B. brevis, and B. cereus. B. subtilis, B. circulans, and B. licheniformis were encountered more frequently than the other Bacillus species during fermentation (82). During fermentation, yeast, LAB, and AAB develop in succession. Species of microorganisms that have been detected in cocoa during fermentation in Ghana, Malaysia, and Belize are listed in Table 35.1. At the onset of fermentation, a pH of 3.4 to 4.0, a sugar content of

10 to 12%, and a low oxygen tension favor the growth of yeasts (74, 81). Yeasts utilize the carbohydrates in the pulp under aerobic and anaerobic conditions and may comprise 40 to 65% of the microflora when the fermentation begins (14, 58). The yeast phase lasts 24 to 48 h, during which populations may increase to 90% of the total microflora. Yeast populations have been determined in several investigations (44, 54, 57, 74). Some yeasts produce various pectinolytic enzymes that degrade the cocoa pulp, thereby aiding in the drainage of juices (27, 66). In addition to metabolizing sugar to produce ethanol, yeasts utilize citric acid, causing the pH to increase (66). All yeast species that contribute to fermentation are not present simultaneously, but the species follow a succession that is influenced by the turning step (aeration) and the fact that fermenting bean masses are not homogeneous (44). Several genera of yeasts are involved in fermentation (Table 35.1). In one study (9), of the 142 yeast genera detected in fermenting bean masses, 105 were asporogenous and 37 were ascosporogenous. Between 48 and 72 h of fermentation, the yeast population begins to decrease so that by the third day it is reduced to 10% of the total microbial population (14). Three factors are responsible for the rapid decline in the dominance of yeasts. First, yeasts rapidly metabolize sucrose, glucose, and fructose in the pulp to form carbon dioxide and ethanol, causing a reduction in energy source. Second, the production of ethanol produces a toxic environment that suppresses yeast growth. For example, Schwan et al. (82) reported that a decline in the population of Kloeckera apiculata was associated with an ethanol concentration of >4%. A small amount of heat is developed simultaneously with ethanol production (4). Third, acetic acid, which is produced from ethanol by the AAB, is also toxic to yeasts. The acetic acid concentration may reach 1 to 2% (15). The anaerobic conditions created by yeasts make the environment suitable for LAB (58). LAB prefer a low oxygen concentration or, if oxygen is present, a high concentration of carbon dioxide (44). Such an environment develops as the pulp collapses and the yeast population decreases. The population of LAB increases ­rapidly, but large numbers may be present for only a brief period (47, 66, 68). Studies conducted by the authors demonstrated that the population of LAB may reach 106 to 107 CFU/g in a typical fermentation in Belize and a high of 20% of the total microflora after 1.5 days in fermentations in Trinidad (73). In Brazil, the population of LAB is about 65% of the total microflora after 14 h of fermentation. The population of LAB remains high for up to 3 days, at which time it decreases to <10% of the total microflora (58).

35.  Cocoa and Coffee

885

Table 35.1  Microorganisms isolated from fermenting cocoa beans Microorganism isolated in: Type of microorganism

Ghanaa

Malaysiaa

Lactic acid bacteria

Lactobacillus plantarum Lactobacillus mali Lactobacillus collinoides Lactobacillus fermentum

L. plantarum L. collinoides

Acetic acid bacteria

Acetobacter rancens Acetobacter pasteurianus subsp. ascendens Acetobacter xylinum Gluconobacter oxydans Candida spp. Hansenula spp. Kloeckera spp. Pichia spp. Saccharomyces spp. Saccharomycopsis spp. Schizosaccharomyces spp. Torulopsis spp.

A. rancens Acetobacter lovaniensis

Yeasts

a b

A. xylinum G. oxydans Candida spp. Debaryomyces spp. Hanseniaspora spp. Hansenula spp. Kloeckera spp. Rhodotorula spp. Saccharomyces spp. Torulopsis spp.

Belizeb L. plantarum L. fermentum Lactobacillus brevis Lactobacillus buchneri Lactobacillus cellobiosus Lactobacillus casei pseudoplantarum Lactobacillus delbrueckii Lactobacillus fructivorans Lactobacillus kandleri Lactobacillus gasseri Leuconostoc mesenteroides Leuconostoc paramesenteroides Leuconostoc oenos Acetobacter spp. G. oxydans

Brettanomyces claussenii Candida spp. Candida boidinii Candida cocoai Candida intermedia Candida guilliermondii Candida krusei Candida reukaufii Kloeckera apis Kloeckera javanica Pichia membranaefaciens Saccharomyces cerevisiae Saccharomyces chevalieri Schizosaccharomyces spp. Schizosaccharomyces malidevorans

Data from Carr et al. (14). Unpublished data.

Both homofermentative and heterofermentative LAB occur in cocoa fermentations; however, the majority are homofermentative (82). LAB detected in traditional box fermentations in Brazil were isolated and characterized as homofermentative and heterofermentative. The homofermentative species ­included Lactobacillus plantarum, Lac­ to­bacillus casei, Lactobacillus delbrueckii, Lactobacillus ­acidophilus, Pediococcus cerevisiae, Pediococcus acidilac­ tici, and Lactococcus lactis. The heterofermentative species included Leuconostoc mesenteroides and Lactobacillus brevis (59). Citric acid is metabolized either into acetic acid, carbon dioxide, and lactic acid by heterofermentative

species or into acetylmethylcarbinol and carbon dioxide by homofermentative lactics. LAB may be more important in cocoa fermentation in Brazil, where during the first 48 h of the fermentation their population was consistently larger than the yeast population (58). This population dynamic differs from those of most other fermentations, where yeasts are the dominant microorganism during the first 48 h. If LAB remain a high percentage of the total microbial population during the fermentation, high concentrations of lactic acid will be produced. Since lactic acid is not volatile, it will remain in the chocolate after manufacturing, producing an undesirable chocolate (70, 85, 95).

886

Fermentations and Beneficial Microorganisms

As the beans are turned to aerate the mass, more of the pulp is metabolized and conditions become aerobic. The population of LAB decreases and the population of AAB increases with increased aeration. The population of AAB generally reaches 105 to 106 CFU/g in a typical fermentation in Belize. Members of two genera of AAB, Acetobacter and Gluconobacter, have been isolated from fermenting cocoa beans. Acetobacter species occur more frequently than Gluconobacter species (14, 15, 57). The population of AAB has been observed to comprise 80 to 90% of the total microflora after 2 days in fermentations in Trinidad (73). AAB oxidize ethanol into acetic acid exothermally (4), causing the temperature of the bean mass to rise to 45 to 50°C. Turning the beans periodically facilitates oxidation of the ethanol into acetic acid and conserves the high temperature of the bean mass. When all of the ethanol is oxidized into acetic acid and then carbon dioxide and water, the fermentation subsides and the temperature of the bean mass decreases quickly. During the later stages of fermentation and drying, aerobic spore-forming Bacillus species develop and may become dominant (15, 57). Bacillus species are present during the first 72 h of the fermentation, but during this early stage their population remains constant. They become dominant later in the fermentation, making up >80% of the microbial population (81, 83). Development of Bacillus species in the bean mass is favored by increased aeration, an increased pH (3.5 to 5.0) of the pulp, and an increase in temperature to 45 to 50°C (15, 57, 83). Bacillus species can produce several compounds that may contribute to the acidity and off-flavors of fermented cocoa. The C3, C4, and C5 free fatty acids that are present in the bean mass during the aerobic phase of fermentation may contribute to the development of some of the off-flavors of chocolate (50, 62). The importance of Bacillus species in cocoa bean fermentation is not well established, but they are reputed to produce acetic and lactic acids, 2,3-butanediol, and tetramethylpyrazine, which can affect the flavor of chocolate (50, 83, 96). A key factor that must be considered when deciding on a fermentation scheme is when to remove the beans from their fermentation environment and begin drying. Extending the fermentation can result in undesirable microbial activity, leading to putrefaction and the production of compounds such as butyric and valeric acids that contribute to off-flavors (51). Forsyth and Quesnel (24) suggested that the following factors may collectively indicate when fermentation is optimum: (i) external color of the beans, (ii) time schedule, (iii) decrease in temperature, (iv) bean cut test using the internal color as a ­criterion, (v) aroma of the fermenting mass, and (vi) plumping or swelling of the beans.

A more desirable measurement of optimum fermentation would be a chemical method that is relatively rapid, inexpensive, and easy to perform and interpret. We have observed that the end point of fermentation can be determined by the pH of the beans, provided that a normal temperature curve is established. The minimum pH that gives acceptable cocoa liquor is 5.2; however, the actual fermentation pH may be slightly lower. Other methods to measure the qualitative and quantitative changes that occur during cocoa fermentation have been reviewed by Shamsuddin and Dimick (84).

Fermentation Using Pure-Culture Seeding

Fermentation of cocoa beans continues to be a natural process that is required for the development of chocolate flavor and cocoa aroma. Chocolate flavor is influenced by several factors, in particular the cocoa bean variety, the fermentation and drying processes, and the subsequent thermal processing (roasting) of well-fermented beans. Because cocoa bean fermentations are conducted in the growing regions in an artisan manner on a small scale, or under less optimal conditions on a large scale, the beans may vary in their degree of fermentation, resulting in beans with varying quality. Beans can be underfermented, overfermented, or optimally fermented with the desired sensory attributes. Beans that are underfermented lack cocoa flavor, and beans that are overfermented are high in acidity with other off-flavor notes. Either one of these conditions will lead to low crop value for the farmer. With greater understanding of the fermentation process, it may be possible to manipulate the fermentation, changing it from a natural, nonmediated, and unpredictable process to a controlled process, initiated with an appropriate starter culture, in which the fermentation occurs in a more stable and repeatable manner. The dairy, meat, and alcoholic beverage industries have largely replaced traditional natural fermentations with defined cultures, higher-quality raw materials, strict control of the fermentation process, better handling of the final products, and diversification of the market (45). More than 40 species of microorganisms are known to be present in the natural fermentation process. However, most of these microorganisms are probably not necessary to achieve the natural process. Studies have been conducted to determine the potential application of a defined microbial inoculum to improve the natural fermentation (8, 10, 21, 74–77, 79). Currently, the use of defined starter cultures has been restricted largely to exploratory studies, but several studies have shown promising results, mainly with the use of yeasts for enhanced production of cocoa pulp juice

35.  Cocoa and Coffee (8, 21, 75, 77). However, during these earlier studies, no attempt was made to remove the natural bacteria and natural yeast isolates (e.g., Saccharomyces chevalieri and Candida zeylanoides). For example, yeasts such as Kluyveromyces fragilis selected from culture collections were evaluated without adding bacteria, but there was no evidence that the fermentation could be accelerated or improved with this approach. Samah et al. (76) demonstrated that Gluconacetobacter xylinus subsp. xyli­ nus (formerly Acetobacter xylinum) could be used as the sole inoculum to produce cocoa beans with a higher pH and higher levels of acetic acid but that these beans had lower chocolate flavor than the control beans. In 1998, Schwan (79) used a defined microbial cocktail consisting of a yeast, Saccharomyces cerevisiae var. chevalieri, two LAB species, L. delbrueckii subsp. lactis (formerly Lactobacillus lactis) and L. plantarum, and two AAB species, Acetobacter aceti and Gluconobacter oxydans, to conduct cacao bean fermentations in 200-kg wooden boxes with aseptically prepared cacao beans, inoculated at different times. The fermentation process mimicked exactly the conditions in 800-kg boxes on Brazilian farms. With the zero-time inoculum the fermentation was almost identical to the natural fermentation. The fermentation with the phased added inoculum was similar but was slower in developing and less pronounced, which resulted in a slightly poorer end product. Based on the fermentation process and the quality of the chocolate made by using a defined cocktail inoculum, specific characteristics are required of the micro­ organisms used. The yeast species should be highly ­fermentative and produce pectinolytic enzymes. The LAB should be members of the homo- and heterofermentative groups and be able to metabolize citric acid. The AAB should be able to grow in an environment of up to 6% ethanol and tolerate 45°C and a pH of 3.5 (11). Considering this information, Camu et al. (11–13) initiated a multiphase approach to study the microbial biodiversity and population dynamics of LAB and AAB present in spontaneous Ghanaian cacao bean heap fermentations. A combination of cultivation, independent monitoring of the microbial population dynamics, and metabolite target analyses of fermented cocoa beans was used to understand their influence on the sensory characteristics, taste, and flavor of chocolate. The ultimate goal of the study was to develop a starter culture for a controlled cacao bean fermentation process that will be useful for the production of cocoa and cocoa-related products. The spontaneous Ghanaian cacao bean heap fermentation process had a limited biodiversity and targeted population dynamics of both LAB and AAB used dur-

887 ing the fermentation. Four main clusters were identified among the LAB isolated (L. plantarum, Lactobacillus fermentum, Leuconostoc pseudomesenteroides, and Enterococcus casseliflavus). Only three clusters were identified among the AAB isolated (Acetobacter pas­ teurianus, Acetobacter senegalensis, and Acetobacter ghanensis). Particular strains of L. plantarum, L. fer­ mentum, and A. pasteurianus that originated from the environment (cocoa pod surfaces, baskets, knives, etc.) were well adapted to the environmental conditions prevailing during Ghanaian cacao bean heap fermentation. Apparently, they played a significant role in the cacao bean fermentation process (11, 12). The predominant yeast population of the Ghanaian heap fermentations was restricted to three species (Pichia kudriavzevii, S. cerevisiae, and Hanseniaspora opuntiae) representing 74% of the total isolates (19). These three yeast species seem to be important for the initiation of the cacao bean fermentation process. Ethanol tolerance, pH tolerance, and heat tolerance are the determining factors for the (temporal) yeast distribution during cacao bean fermentations. Yeasts depectinized the pulp and produced ethanol from the sugars. LAB produced lactic acid, ­ acetic acid, ethanol, and mannitol from the sugars and/or citrate; citrate fermentation appeared to be of utmost importance. Its metabolism contributed to strain competitiveness, aroma precursor formation, and pH regulation of the natural environment. A. pasteu­ rianus metabolized ethanol, mannitol, and lactate and converted ethanol into acetic acid lactate and mannitol and acetate into CO2 and H2O (Fig. 35.1). In order to determine the appropriate starter culture organisms to use in the seeding studies, in vitro fermentations were conducted using a cocoa pulp simulation medium. Using specific LAB and AAB cultures isolated from heap fermentations, a series of small-scale fermentations were conducted to determine if the isolates would tolerate ethanol and low pH and would metabolize citric acid (LAB) or oxidize ethanol into acetic acid (AAB). After a series of large-scale fermentation studies were conducted, L. plantarum 80, L. fermentum 222, and A. pasteurianus 386B were selected for inclusion as a starter culture for heap fermentation (42). Six heap fermentation studies were conducted in a natural (farmer’s plantation) and artificial (outside plantation) environment using these isolates. Even though no yeasts were included in the starter culture, a yeast fermentation was observed. No effort was made to decontaminate the surfaces of the pods, a natural source of inocula; however, the starter culture outgrew the natural microflora in every study and dominated the entire fermentation process. The results obtained from

888

Fermentations and Beneficial Microorganisms

Figure 35.1  (A) Course of residual glucose (), fructose (p), and sucrose (*) and of mannitol produced (¢) in the pulp. (B) Course of residual glucose (), fructose (p), and sucrose (*) and of mannitol produced (¢) in the beans. (C) Course of lactic acid produced in the pulp () and in the beans (p). (D) Course of residual citric acid (filled symbols, left axis) and of succinic acid produced (open symbols, right axis) in the pulp ( and ) and in the beans (p and r). (E) Course of acetic acid produced in the pulp () and in the beans (p). (F) Course of ethanol produced in the pulp () and in the beans (p). doi:10.1128/9781555818463.ch35f1

all of the studies indicated faster degradation of citric acid and sugars, causing higher final concentrations of organic acids in the pulp and beans (Fig. 35.2). Camu concluded that it is possible to use starter cultures in cocoa bean heap fermentations to produce reliable and controlled fermentations that will consistently produce high-quality cocoa and chocolate (10). The research conducted by Camu et al. (11, 12) has led to the application of a process that is evaluating the inoculation of cacao beans with a defined microbial population containing isolates consisting of yeasts, LAB, and AAB. Studies have been conducted in West Africa, Southeast

Asia, and South America. The results of these studies are promising. Finished chocolate products have been produced, but the process has not been fully commercialized and research continues. However, a few conclusions have been determined. Use of a defined inoculum will increase the changes that typically occur during the fermentation, the fermentations are more consistent, predictable flavor is developed, the flavor can be customized, and this process is transferable to other cacao-growing regions. There may be some challenges associated with the process, including maintaining the stability of the microorganisms in tropical climates and

35.  Cocoa and Coffee

889

Figure 35.2  Course of sugar consumption and mannitol production in cocoa pulp during Ghanaian cocoa bean heap fermentations inoculated with L. plantarum 80 (A, D), L. fermen­ tum 222 (B, E), or L. plantarum 80 plus L. fermentum 222 (C, F), in combination with A. pasteurianus 386B, performed at the farm site (heap 17, A; heap 18, B; heap 19, C) and the factory site (heap 20, D; heap 21, E; heap 22, F). , residual glucose; p, residual fructose; ¢, mannitol produced. doi:10.1128/9781555818463.ch35f2

expanding the process beyond the small scale to largescale production.

Biochemistry of Cocoa Fermentation

The actual production of chocolate flavor precursors occurs within the cocoa bean and is primarily the result­ of biochemical changes that take place during fermentation and drying. The mode of fermentation and the microbial environment during these stages of cocoa production provide the necessary conditions for complex biochemical reactions to occur. Although flavor compounds such as lactic and acetic acids are produced outside the bean by microbial activity, chocolate flavor development is largely dependent on the enzymatic formation of flavor precursors within the cotyledon that are unique to cocoa. Such classes of compounds include free amino acids, peptides, reducing sugars, and

polyphenols. When fermented, dried cocoa beans containing these flavor precursors are subjected to roasting during chocolate manufacture, a necessary step in flavor development, a series of complex, nonenzymatic browning reactions occurs to produce flavor and color compounds characteristic of chocolate (35, 46, 69, 71, 72). However, if unfermented cocoa beans lacking these precursor compounds are roasted, very little chocolate flavor is produced. It is, therefore, important that these flavor precursor compounds are formed inside the cocoa bean during fermentation. The initiation of fermentation also corresponds to an incipient germination phase that is necessary for ­mobilization of the enzymes and hydration of bean components in preparation for growth. However, the germination phase is undesirable in cocoa beans used to make chocolate.

890

Fermentations and Beneficial Microorganisms

Bean Death

Bean death is a critical event during cocoa fermentation that allows the biochemical reactions responsible for flavor development to occur within the cocoa bean. Although rising temperatures and increasing acetic acid concentrations during fermentation have been implicated in causing seed death (60), more recent data (43) indicate that the production of ethanol during the anaerobic yeast growth phase correlates very closely with the death of the seed. Total inability of the seed to germinate occurs about 24 h after maximum concentrations of ethanol are attained within the cotyledon (Fig. 35.3). As a result, events associated with germination and certain quality defects, e.g., the utilization of valuable seed components such as cocoa butter and the opening of the testa by hypocotyl extension, will not occur. This produces a more stable, desirable end product. From a flavor perspective, events associated with the death of the seed also cause cellular membranes to leak and permit enzymes and substrates to react to form flavor precursor compounds important to chocolate flavor development. As shown in Figure 35.3, the activity of these enzymes in the cotyledon results in significant increases in free amino acid and reducing-sugar (glucose and fructose) contents. While the temperature of beans increases, the concentration of organic acids increases, causing a decrease in pH. All of these factors influence the biochemistry within the bean and have an impact on cocoa flavor and quality.

Environmental Factors

There are several environmental factors, viz., pH, temperature, and moisture, in the fermenting mass that influence cocoa bean enzyme reactions. Each enzyme has an optimum pH at which it is most active, and within a defined range, an enzyme reaction accelerates as temperature increases. In addition, a certain amount of moisture is necessary to allow enzymes and their substrates to react to form products. Significant changes in pH, temperature, and moisture occur during cocoa fermentation and drying processes (Fig. 35.3) that influence the type and quantity of flavor precursor compounds produced by enzymatic action. Moisture content within the cotyledon during fermentation is usually >35% and will permit adequate migration of enzymes and substrates for enzymatic activity. However, once the drying process begins, moisture content gradually decreases, making it increasingly difficult for enzymes and substrates to react. When a moisture content of 6 to 8% is achieved, virtually all enzyme activity ceases. The pH of the unfermented cotyledon is about 6.5 and may decrease to as low as 4.5 by the end of the fer-

Figure 35.3  Physical (A) and chemical (B) changes in cocoa beans during fermentation and drying in Belize. Fermentation was conducted with 2,000 lb of wet cocoa beans from ripe pods in wooden boxes that were turned daily. Drying was conducted in flat-bed dryers indirectly heated with hot air. Data represent results from an average of 11 fermentation trials using composite samples collected daily. (A) Temperature was measured in the whole bean mass. Moisture (%) and pH analyses are based on shell-free cotyledons. (B) Sucrose, glucose (Glc), fructose (Fruc), total amino acid, acetic acid, and ethanol contents (%) were determined by analysis of water extracts from shell-free cotyledon samples. Data are taken from Lehrian (43). doi:10.1128/9781555818463.ch35f3

mentation. This lowering of pH occurs after seed death and is due primarily to diffusion into the bean of organic acids produced by LAB and AAB. It is the growth of these and other microorganisms that also contributes to the increasing temperature of the mass of fermenting beans. Typically, the temperature will rise from 25 to about 50°C, followed by a slight decrease as bacterial growth subsides. An increase in temperature of more than 20°C during fermentation can have a profound impact on enzyme activity. If very little change in temperature occurs, enzyme activity is reduced, resulting in fewer

35.  Cocoa and Coffee

891

flavor precursors and poor chocolate flavor. Likewise, if appropriate amounts of organic acids are not produced during fermentation, the pH of the cotyledon will not be suitable for optimal enzyme activity and the flavor profile of the resulting cocoa will be affected. However, too much acid will produce excessive sourness that can mask the chocolate flavor. Consequently, there is a delicate balance among the length of fermentation, environmental factors, and microbial activity that influences enzyme activity within the cotyledon. Hydrolytic and oxidative enzymes play a major role in reactions that produce flavor precursors. A summary of cocoa bean enzymes and their substrates and optimum pHs is given in Table 35.2 (47). More recently, Hansen et al. (32) studied cocoa bean enzymes and their activities during the fermentation and drying process in an effort to understand their impact on cocoa flavor and quality.

Hydrolytic Enzyme Reactions

Hydrolytic enzymes such as invertase, glycosidases, and proteases have highest activity during the anaerobic phase of cocoa fermentation. The products of these enzyme activities during cocoa fermentation fall into three basic categories: sugars, amino acids and peptides, and cyanidins. Sugars and amino acids and peptides participate in nonenzymatic browning reactions during roasting to form important chocolate flavor precursors, whereas the cyanidins have more of an impact on color development and some minor flavor components. Sucrose is the major sugar in unfermented cocoa beans. It is not a reducing sugar and therefore does not participate in nonenzymatic browning reactions that occur during roasting to contribute to chocolate flavor. However, sucrose is converted to glucose and fructose by invertase during the fermentation process. These reducing sugars represent more than 95% of the total reducing monosaccharides in cocoa beans, and their

concentrations increase almost threefold during fermentation, while sucrose is depleted (Fig. 35.3). Another class of hydrolytic enzymes within the cocoa bean that contributes to both flavor and color during fermentation is the glycosidases. The substrates for these enzymes are the purple anthocyanins located in specialized vacuoles within the cotyledon that are responsible for the characteristic deep purple color of the unfermented bean. The actions of specific glycosidase enzymes begin at seed death and are responsible for cleaving the sugar moieties, galactose and arabinose, attached to the anthocyanins. This results in a bleaching of the purple color of the beans as well as the release of reducing sugars that can participate in flavor precursor reactions during roasting (23). Pigments themselves do not carry any flavor potential (23, 50, 67). Cocoa beans that still contain significant purple color are considered to have been poorly fermented and are less desirable. Although there are small amounts of free amino acids present in unfermented cocoa beans, the total free amino acid pool increases significantly during fermentation due to the action of both endo- and exoproteases on cocoa bean proteins. After seed death occurs, these proteolytic enzymes are free to act on protein substrates within the bean and their activity becomes dependant on pH and temperature. A vicilinlike globular storage protein within the cotyledon is the primary target of these proteolytic enzymes, and ratios of free amino acids and peptides that are unique to cocoa are produced (93). These flavor precursor compounds contribute to the development of cocoa flavor when roasted in the presence of reducing sugars.

Oxidative Enzyme Reactions

Significant oxidative enzyme activity also occurs, being most prevalent late in the aerobic phase of fermentation but continuing well into the drying of cocoa. Polyphenol oxidase is the major oxidase in cocoa and

Table 35.2  Characteristics of the principal enzymes active during the curing of the cocoa beana Enzyme(s)

Location

Invertase

Testa

Substrate(s) Sucrose

Products

pH

Glucose and fructose

4.0

Temp(s) (°C) 52

5.25

37

Reference(s) 49

Glycosidases (β-galactosidase)

Bean

Glycosides (3-β-d-galactosidyl cyanidin and 3-α-l-arabinosidyl cyanidin)

Cyanidin and sugars

3.8–4.5

45

23

Proteases

Bean

Proteins

Peptides and amino acids

4.7

55

7, 61

Polyphenol oxidases

Bean

Polyphenols (epicatechin)

σ-Quinones and σ-diquinones

6.0

31.5, 34.5

a

Information taken from Lopez (47).

63

892

Fermentations and Beneficial Microorganisms

is responsible for much of the brown color that occurs during fermentation as well as some flavor modifications. This enzyme becomes active during the aerobic phase of the fermentation as a result of oxygen permeating the cotyledon. Events that contribute to activity include seed death, subsequent breakdown of cellular membranes, reduction in the amount of seed pulp, and aeration of the bean mass by agitation. Oxygen continues to penetrate the beans during the drying process, enabling polyphenol oxidase activity to continue until rising temperatures and insufficient moisture become inhibiting factors. Catechins and leucocyanidins are the major classes of polyphenols that are subject to oxidation in cocoa beans. Epicatechin comprises more than 90% of the total catechin fraction and is the major substrate of polyphenol oxidase (28). Oxidation of epicatechin during the aerobic phase of fermentation and drying is largely responsible for the characteristic brown color of fermented cocoa beans. Polyphenols in the dihydroxy configuration are oxidized to form quinines, which in turn can polymerize with other polyphenols or form complexes with amino acids and proteins to yield characteristic colored compounds and high-molecular-weight insoluble material. This formation of complexes also has an impact on flavor. The formation of these less soluble polyphenolic complexes reduces astringency and bitterness associated with native polyphenols present in unfermented cocoa (25, 65). In addition, the ability of polyphenols to form complexes with proteins results in the reduction of offflavors associated with the roasting of peptide and protein material (34, 94). Despite the beneficial effects of fermentation on the color and flavor of cocoa, the biochemical changes during fermentation have been shown to decrease the native polyphenol and antioxidant contents (91). In recent years, much attention has been devoted to the potential health benefits of polyphenols and their antioxidant activity in cocoa and chocolate products (41). The nonfat portion of the cocoa bean is a rich source of certain classes of polyphenols, and it has been shown that the cocoa solids contents of chocolate products are highly correlated with the polyphenol and antioxidant contents of these products (37). Therefore, the flavorenhancing properties of cocoa fermentation must be balanced against the polyphenol- and antioxidant-­lowering effects of fermentation. The availability of cocoa beans with various degrees of fermentation provides the chocolate manufacturer with the opportunity to produce chocolate products that meet the diverse needs of the modern consumer.

Flavor and Quality Implications

The ultimate goal of biochemical changes during fermentation is to produce cocoa beans with desirable flavor and color characteristics. Good chocolate flavor potential is achieved by the production of specific amino acids, peptides, and sugars through the action of proteases and invertases on cocoa bean substrates. Proper control of fermentation conditions and microflora ensures that concentrations of organic acids are maintained at reasonable levels to minimize sour and putrid off-flavors while still developing the pH and temperature environment for enzyme-substrate reactions that produce chocolate flavor precursors. Enzyme-mediated conversion of polyphenol materials during fermentation and drying processes reduces astringency and bitterness and produces the desirable brown color typical of properly fermented cocoa beans. The drying process will then preserve the flavor and color characteristics of the beans until they are made into chocolate. Although cocoa has been successfully produced for centuries, flavor characteristics of specific varieties of cocoa are becoming diluted due to the prevalence of genetic hybrids. While these hybrids are being selected for high crop yield and disease resistance, certain flavor attributes are being lost. The actual identities of the flavor precursors responsible for chocolate flavor have not yet been confirmed. However, research by Voigt and Beihl (92) focused on characterizing cocoa bean enzymes and proteins that yield breakdown products unique to cocoa during the fermentation process. This work needs to continue in order to understand flavor development and maintain the high quality of chocolate flavor the consumer expects.

Drying

After the beans are fermented, they have a moisture content of about 40 to 50%, which must be reduced to 6 to 8% for safe storage. A higher final moisture content will result in mold growth during storage. The drying process relies on air movement to remove water. This environment favors aerobic microorganisms that proliferate at rates that decrease with moisture loss. Sun drying is the preferred method, but in regions where harvesting coincides with frequent rainfall, some form of artificial drying is necessary and desirable. In general, sun drying is employed on small farms, whereas large estates may resort to both natural and artificial drying. Sun drying allows a slow migration of moisture throughout the bean, which transports flavor precursors that were formed during fermentation.

35.  Cocoa and Coffee During sun drying, beans are placed on wooden platforms, mats, polypropylene sheets, or concrete floors in layers ranging from 5 to 7 cm thick. The beans are constantly mixed to promote uniform drying, to break agglomerates that may form, and to discourage mold growth. Under sunny conditions, the beans dry in about a week, but under cloudy or rainy conditions, drying times may be prolonged to 3 or 4 weeks, increasing the risk of mold development and spoilage. Various types of artificial dryers employed to overcome the dependence on weather conditions have been described by McDonald et al. (55). Hot air dryers of one form or another, fueled by wood or oil as a source of cheap, readily available energy, are generally employed. The beans may be heated by direct contact with the flue gases; however, the preferred method relies on indirect heating via heat exchangers. Improperly used or poorly maintained heating systems present the danger of contamination with smoke, which results in smoky or hammy off-flavors characteristic of beans from some countries. Platforms, trays, and rotary dryers of various designs, coupled to furnaces, are used, but in every instance, the initial drying must be slow and frequent mixing must be done to obtain uniform removal of water. This method results in volatilization of acids and sufficient time for oxidative biochemical reactions to occur. For this reason, temperatures should not exceed 60°C and drying should take at least 48 h. Elevated temperatures also tend to produce cocoa with brittle shells and cotyledons that crumble during handling. In short, the drying rate should be controlled so as to remove moisture at a rate that will avoid case hardening (rapid drying on the bean surface with moisture retention inside the bean) or excessive mold growth while still allowing sufficient time for biochemical oxidative reactions and loss of acid to occur.

Storage

Due to marketing practices and manufacturing procedures, fermented, cured, dried beans may be stored for periods of 3 to 12 months in warehouses on farms, on wharfs in exporting and receiving countries, and at factories before being processed into chocolate. The efficiency of the drying process will determine the shelf life of the product. Uniformly dried beans with a moisture content of 7 to 8% that are stored at a relative humidity of 65 to 70% will generally maintain that moisture, resist mold growth and insect infestation, and not require repeated fumigation. The cocoa quality can change during storage, depending on temperature, relative humidity, and ventilation conditions. Slow oxidation and acid

893 loss continue to enhance product quality somewhat, but prolonged storage results in a noticeable staling (48).

COFFEE PROCESSING The specific geographic origin of coffee is not known, but it is believed that coffee originated in Ethiopia and spread to the rest of Africa and then to southern America and Asia. Currently, coffee is produced in Brazil, Central America, Colombia, Côte d’Ivoire, Ethiopia, Guatemala, India, Indonesia, Mexico, Uganda, the United States (Hawaii), Vietnam, and other countries. In 2010, production was estimated to be approximately 7.0 million tons. Coffee grows in tropical and subtropical climates at altitudes between 200 and 2,000 m. The plant produces its first fruit (cherries) about 4 to 5 years after sowing. The peak production for plants occurs when they are 8 to 10 years old. Coffee beans are produced by the genus Coffea. More than 40 species are known, but only a few are used to produce coffee. C. arabica, C. canephora, C. robusta, C. liberica, and C. excelsa are the important species. C. arabica and C. canephora are the two predominant species on the world market. C. arabica originated from Brazilian coffee plantations and is cultivated in South America, Central America, and some African countries. C. robusta can be found in Côte d’Ivoire, Cameroon, Uganda, Indonesia, and India. Table 35.3 shows the broad spectrum of countries producing arabica and robusta coffees. The coffee fruit is a fleshy berry approximately the size of a small cherry. The coffee bean is composed of six parts: skin, pulp, mucilage, parchment (outer layer around the bean), silver skin (inner layer), and bean. The pulp layer is approximately 29% of the dry weight of the fruit. It consists of 76% water, 10% protein, 2% fiber, 8% mineral salts, and 4% soluble and insoluble components, such as tannins, pectin, reducing and nonreducing sugars, caffeine, chlorogenic and caffeic acids, cellulose, hemicellulose, lignin, and amino acids (22). The mucilage is 5% of the dry weight of the bean. It contains water; pectin; pectic acid; small quantities of arabinose, galactose, xylose, and rhamnose (2); organic acids (22); and hydrolytic and oxidative enzymes (2, 3). It takes approximately 1 year from the flowering stage for the fruit to reach maturity. As the fruit ripens, it changes color from green to cherry red. The ripe, red coffee cherry consists of two green beans surrounded by the pulp layer that is enclosed in parchment. It is necessary to first remove the parchment and pulp layer to obtain the green coffee beans, which are

Fermentations and Beneficial Microorganisms

894

Table 35.3  Countries and regions producing arabica and robusta coffee beans Countries and regions producing: Arabica coffee

Robusta coffee

Arabica and robusta coffee

Bolivia

Benin

Borneo

Colombia

Central African Republic

Brazil

Costa Rica

Congo

Cameroon

Cuba

Côte d’Ivoire

Ecuador

Dominican Republic

Gabon

India

Guatemala

Ghana

Java

Haiti

Guinea

Madagascar

Honduras

Liberia

Malawi

Jamaica

Nigeria

Philippines

Mexico

Sierra Leone

Sri Lanka

Nicaragua

Togo

Sulawesi (Celebes)

Panama

Sumatra

Paraguay

Tanzania

Peru

Thailand

Puerto Rico

Timor

Salvador

Uganda

Venezuela

Vietnam Zaire Zimbabwe

then dried, roasted, milled, and used for making coffee beverages. After the seeds are separated from the other fruit, two methods are used to process the seed, the dry process and the wet process. The drying step can be natural drying (sun) or artificial drying using static or rotary driers (31). The dry process yields a product that is referred to as “natural coffee.” The percent yield of dried coffee obtained from ripe cherries varies among species; C. arabica produces 12 to 18%, and C. liberica produces 10%. The wet process is a wet fermentation step that yields “washed coffee.” Coffee beans also undergo a fermentation step to prepare the fruit for commercial use (5, 16, 86). Fermentation may involve molds, yeasts, several species of LAB, coliforms, and other gram-negative bacteria. These microorganisms originate from the surface of the fruit and the soil (1, 26, 88). The role of fermentation in coffee production is less critical in the development of flavor than that in cocoa production, although improperly fermented fruit will result in off-flavors. The fermentation of coffee causes the degradation or breakdown of the mucilage around the parchment skin, which gives the bean a better appearance. The natural

enzymes that are present in the mucilage degrade the pectin. With pectin being a large portion of the mucilage (38), the microorganisms responsible for colonization and utilization of the pectin must be capable of producing pectinases. Fermentation is conducted in concrete tanks or bins with depths of 1 to 10 m (18). The tanks are covered with a roof, and the bottom of each tank is sloped to allow the water used during the fermentation to drain out. The length of the fermentation step varies from 24 to 90 h based on the climatic conditions, the altitude, the pollution of the water used in the process, and the maturity of the fruit. An optimal fermentation time is considered to be 16 to 24 h (87). Extending the fermentation time will result in the development of flavor defects and bean discoloration (56, 64). The end of the fermentation is a subjective determination based on when the texture of the mucilaginous tissues changes from viscous to liquid. Once the fermentation is complete, any remaining mucilage is removed by washing. The highest microbial activity occurs during the first 12 to 24 h after the beans are harvested (89). Most of the microorganisms detected during this early stage belong to the genera Enterobacter and Escherichia. During the first 24 h, populations of these bacteria increase

35.  Cocoa and Coffee from 102 to 109 CFU/g. Pectinolytic species of Bacillus, Fusarium, Penicillium, and Aspergillus have also been detected. Only a few yeast species have been identified. The strains of yeast that have been isolated did not have the ability to degrade mucilaginous tissue. Fermentation studies conducted with Kona coffee beans demonstrated that Erwinia dissolvens is the main cause of mucilage decomposition (26). LAB such as L. mesenteroides, L. plantarum, and L. brevis have also been detected and are able to cause decomposition of pectic polymers (39). The total microbial population as well as the types of microorganisms depend on the temperature, pH, moisture, composition of substrate in simple sugars and/or polysaccharides, and maintenance of the devices used (31). During fermentation, there is a slight increase in temperature and the pH of the unfermented coffee bean, which can range from 5.4 to 7.0, decreases to 3.7 (1, 56). The final pH will vary based on the fermentation conditions. The time it takes for submerged fermentations (anaerobic) is always longer because acidification of the medium takes longer (31). Simple sugars are converted to ethanol or organic acids (38, 52). As stated above, there are two methods to process coffee beans, a wet process and a dry process. The following briefly describes the two processes.

Wet Processing

Colombia, Central America, and Hawaii are more likely to use the wet process on arabica coffees (80). Two factors are the key to obtaining a good-quality final product, viz., using a uniform population of mature berries and rapid processing (2, 38). After harvesting, the mature beans are mechanically depulped. This is followed by a fermentation step that is used to convert the remaining mucilage into water-soluble products that are removed by washing prior to final drying. The white, sticky, partially depulped berries are held under water in wooden or concrete bins for 12 to 60 h, depending on environmental conditions, the ripeness of the berries, and the variety being processed. Microbiological by-products are periodically washed away during the fermentation step to avoid the development of excessive off-flavors. Once the sticky pulp layer is converted into water-soluble products, the beans are washed with water and dried in the sun or with a mechanical drier to a moisture content of 11 to 12%. Overfermentation of the berries will cause spoilage and the development of taints (17, 38).

Dry Processing

The natural drying process relies on partial dehydration of the pulp layer while the coffee fruit is ripening on

895 the tree. This method is more likely to be used in Brazil and Ethiopia and for robusta coffee (80). The ripe fruit is harvested (hand or machine stripped) and subjected to additional drying. The mature berries may be spread onto platforms, soil, concrete, or tarmac in layers up to 10 cm thick. Berries are heaped up at night and respread each morning. It can take from 10 to 25 days of sun drying. During this time, a natural microbial fermentation occurs; enzymes are secreted that break down the pulp and mucilage (38). Bacteria, yeasts, and pectinolytic molds are present during the fermentation. However, bacteria are more prevalent than yeasts or molds. Unlike that in cocoa fermentation, there is no defined microbial succession that occurs with coffee maturation and fermentation. Fermentation of the sugars produces alcohol and acetic, lactic, butyric, and other carboxylic acids. Some mucilage will penetrate the coffee bean during drying, and for this reason the natural process produces a light brown bean instead of the blue-green color of wet processed coffee. The resulting dry berry is free of pulp and mucilage but is still surrounded by dry skins, which are mechanically removed. Molds may grow during slow drying, producing off-flavors typical of a butyric fermentation. The coffee beans are dried to a final moisture content of 11 to 12%. As is the case with cocoa beans, the temperature during drying can negatively or positively influence the quality of the final coffee drink.

Use of a Starter Culture in Coffee Fermentation

The rate at which demucilagination occurs can depend on factors such as altitude and temperature. Due to its thicker mucilage, robusta coffee takes longer for complete demucilagination (72 to 100 h) than does arabica coffee (12 to 24 h) (90). Because of this wide variation in time, use of a starter culture may have an impact. In addition to speeding up the fermentation process by reducing the demucilagination time, use of a starter culture is an eco-friendly process, there is no need for external additives and no extra cost, coffee quality will not deteriorate, and it favors microbial diversity (90). Even with the potential to improve the fermentation process, no significant research has been conducted using starter cultures. Unlike the current studies on microbial seeding of cacao discussed above, no research on coffee is focused on moving the fermentation of coffee to a commercialized finished coffee product. In the Philippines and Indonesia, a unique coffee, civet coffee, has been commercialized and is consumed by coffee connoisseurs in the United States, Europe, and East Asia. This coffee is produced following consumption of

896

Fermentations and Beneficial Microorganisms

ripe coffee cherries by the civet, a small, lithe-bodied, nocturnal mammal native to tropical Asia and Africa. The coffee cherries undergo fermentation inside the animal’s intestinal tract, and the animal excretes the hard, indigestible innards of the cherries. These coffee beans are used to make a coffee brew. The exported finished coffee beans can sell for as much as $227 per pound. Recently, Velmourougane et al. (90) conducted studies in India using robusta and arabica fruits. The objective of the research was to study the biochemical and microbial aspects of fermentation, understand and standardize the optimum process, and produce acceptable-quality coffee. The harvested fruit was immersed in water to separate the ripe cherries from the floats and dried cherries. The pulp was removed from the ripe fruit, excess water was removed, the beans were placed in plastic drums, and a starter culture was introduced from previous arabica and robusta batch fermentation. Various amounts of starter cultures were added: 50, 100, 200, and 400 liters/1,000 kg. The pH, temperature, mucilage degradation rate, and microbiological counts were monitored at the beginning and end of the fermentations. Coffee samples were sun dried to 10 to 10.5% moisture. Regardless of the amount of starter culture added, there was little variation in pH: 3.4 to 3.5 for arabica beans and 3.42 to 3.98 for robusta beans. The rate of demucilagination increased with the amount of starter culture added. At 50 liters/1,000 kg and 400 liters/1,000 kg, the times for demucilagination of arabica fruit were 11.5 and 66 h and those for robusta fruit were 10 and 43 h, respectively. Bacteria were the predominant organisms, followed by yeast. LAB were the dominant bacteria present. Fungi were detected at the beginning of the fermentation but in much lower numbers; these numbers significantly increased after 2 days. The researchers concluded that use of a starter culture would be of greater benefit for processing robusta than arabica coffee. The use of a starter culture of 50 to 100 liters/kg will reduce the fermentation time sufficiently without negatively affecting the quality of the coffee. The starter culture that is used should be from a previous fermentation not more than 1 day old. More research on the use of starter cultures is required for this approach to become a reliable, viable option.

Roasting

Roasting is necessary for the preparation of the final bean. The roasting time depends on the quality of the raw coffee. Industrial roasting may range between 90 s and 6 min, while handcraft roasting may range between 18 to 20 min (31). Roasting is conducted using rotary drums at temperatures ranging from 200 to 250°C for

up to 20 min (31). Coffee produced in each country will have its own unique flavor. Therefore, mixtures of various varieties are roasted together to create a good­quality coffee. Roasting is an art, and the final roast quality depends on the roaster’s know-how (31).

CONCLUSION Although serving somewhat different purposes, microbial fermentation plays a critical role in the production of both cocoa and coffee. A common goal of fermentation of both of these products is the breakdown and removal of fruit pulp. This process aids in the proper drying of coffee and in the case of cocoa provides a suitable environment for flavor and color development. As in most natural curing processes, there is a delicate balance between environmental factors and conditions enabling microbial activities, all of which influence the biochemical changes that take place in processed foods. Coffee and cocoa are no exceptions, and it is the proper control of the fermentation process that largely determines the color and flavor qualities of the final products. Consequently, understanding the microbiology and biochemistry of cocoa and coffee production, as well as the factors that influence them, is critical to quality control.

References 1. Agate, A. D., and J. V. Bhat. 1966. Role of pectinolytic yeasts in the degradation of the mucilage layer of Coffea robusta cherries. Appl. Microbiol. 14:256–260. 2. Amorium, H. V., and V. L. Amorium. 1977. Coffee enzymes and coffee quality, p. 27–55. In R. L. Ory and A. J. St. Angelo (ed.), Enzymes in Food and Beverage Processing. ACS Symposium Series, vol. 47. American Chemical Society, Washington, DC. 3. Amorium, H. V., and M. Mello. 1991. Significance of enzymes in non-alcoholic coffee beverage, p. 189–209. In P. F. Fox (ed.), Food Enzymology, vol. 2. Elsevier, Amsterdam, The Netherlands. 4. Anonymous. 1979. The Fermentation of Cocoa. National Union of Cocoa Producers, Villahermosa, Mexico. 5. Arunga, R. 1982. Coffee, p. 259–274. In A. H. Rose (ed.), Economic Microbiology, vol. 7. Fermented Foods. Academic Press, New York, NY. 6. Biehl, B., B. Meyer, G. Crone, L. Pollman, and M. B. Said. 1984. Chemical and physical changes in the pulp during ripening and post-harvest storage of cocoa pods. J. Sci. Food Agric. 48:189–208. 7. Biehl, B., U. Passern, and D. Passern. 1977. Subcellular structures in fermenting cocoa beans—effect of aeration and temperature during seed and fragment incubation. J. Sci. Food Agric. 28:41–52. 8. Buamah, R., V. P. Dzogbefia, and J. H. Oldham. 1997. Pure yeasts culture fermentation of cocoa (Theobroma

35.  Cocoa and Coffee cacao L.): effect on yield of sweatings and cocoa bean quality. World J. Microbiol. Biotechnol. 13:457–462. 9. Camargo, R. J., J. Leme, and A. M. Filho. 1963. General observations on the microflora of fermenting cocoa beans (Theobroma cocoa) in Bahia (Brazil). Food Technol. 17:116–118. 10. Camu, N. 2007. Biodiversity, population dynamics, and metabolite target analysis of Ghanaian cocoa bean heap fermentation processes. Ph.D. thesis. Vrije Universiteit Brussel, Brussels, Belgium. 11. Camu, N., A. González, T. De Winter, A. Van Schoor, K. De Bruyn, P. Vandamme, J. S. Takrama, S. K. Addo, and L. De Vuyst. 2007. Influence of turning and environmental contamination on the dynamics of lactic acid bacteria and acetic acid bacteria populations involved in spontaneous cocoa bean heap fermentation in Ghana. Appl. Environ. Microbiol. 74:86–98. 12. Camu, N., T. De Winter, K. Verbrugghe, I. Cleenwerck, P. Vandamme, J. S. Takrama, M. Vancanneyt, and L. De Vuyst. 2007. Dynamics and biodiversity of lactic acid bacteria and acetic acid bacteria populations involved in spontaneous heap fermentation of cocoa beans in Ghana. Appl. Environ. Microbiol. 73:1809–1824. 13. Camu, N., T. De Winter, S. K. Addo, J. S. Takrama, H. Bernaert, and L. De Vuyst. 2008. Fermentation of cocoa beans: influence of microbial activities and polyphenol concentrations on the flavour of chocolate. J. Sci. Food Agric. 88:2288–2297. 14. Carr, J. G., P. A. Davies, and J. Dougan. 1979. Cocoa Fermentation in Ghana and Malaysia. University of Bristol Research Station, Long Ashton, Bristol, and Tropical Products Institute, London, United Kingdom. 15. Carr, J. G., P. A. Davies, and J. Dougan. 1980. Cocoa Fermentation in Ghana and Malaysia, vol. II. University of Bristol Research Station, Long Ashton, Bristol, and Tropical Products Institute, London, United Kingdom. 16. Castelein, J., and H. Verachtert. 1983. Coffee fermentation, p. 587–615. In H. J. Rehm and G. Reed (ed.), Biotechnology, vol. 5. Chemie Verlag, Weinheim, Germany. 17. Clarke, R. J., and R. Macrae. 1987. Coffee Technology, vol. 2. Elsevier Applied Science, New York, NY. 18. Coste, E. 1959. L’usinage des cafés verts, p. 2–60. In R. Coste (ed.), Les Caféiers et les Cafés dans le Monde, vol. 1. Les Caféiers. Larose, Paris, France. 19. Daniel, H.-M., G. Vrancken, J. F. Takrama, N. Camu, P. De Vos, and L. De Vuyst. 2009. Yeast diversity of Ghanaian cocoa bean heap fermentations. FEMS Yeast Res. 9:774–783. 20. Duncan, R. J. E., G. Godfrey, T. N. Yap, G. L. Pettipher, and T. Tharumarajah. 1989. Improvement of Malaysian cocoa bean flavor by modification of harvesting, fermentation and drying methods—the Sime-Cadbury process. Planter 65:157–173. 21. Dzogbefia, V. P., R. Buamah, and J. H. Oldham. 1999. The controlled fermentation of cocoa (Theobroma cacao L) using yeasts: enzymatic process and associated physico-chemical changes in cocoa sweatings. Food Biotechnol. 13:1–12. 22. Elias, L. G. 1978. Composicioń guimica de la pulpa de café y otros subproductos, p. 19–29. In J. E. Braham and R.

897 Bressan (ed.), Pulpa de Café: Composicion, Technologia y Utilicatioń. INCAP, Panamiá, Panama. 23. Forsyth, W. G. C., and V. C. Quesnel. 1957. Cocoa glycosidase and colour changes during fermentation. J. Sci. Food Agric. 8:505–509. 24. Forsyth, W. G. C., and V. C. Quesnel. 1957. Variations in cocoa preparation, p. 157–168. In Sixth Meeting InterAmerican Cocoa Conference. InterAmerican Cocoa Conference, Balia, Brazil. 25. Forsyth, W. G. C., and V. C. Quesnel. 1963. Mechanisms of cocoa curing. Adv. Enzymol. 25:457–492. 26. Frank, H. A., N. A. Jum, and A. S. Dela Cruz. 1965. Bacteria responsible for mucilage-layer deposition in Kona coffee cherries. Appl. Microbiol. 13:201–207. 27. Gauthier, B., J. Guiraud, J. C. Vincent, J. P. Porvais, and P. Galzy. 1977. Components on yeast flora from the traditional fermentation of cocoa in the Ivory Coast. Rev. Ferment. Ind. Aliment. 32:160–163. 28. Griffiths, L. A. 1957. Detection of the substrate of enzymatic browning in cocoa by a post-chromatographic enzymatic technique. Nature 180:1373–1374. 28. Grimaldi, J. 1954. The Cocoa Fermentation Process in the Cameroon, vol. V. 5RTIC/DOC 24. Bioq. 12. Reunion del Comite Técnico Interamericano del Cocoa. 30. Grimaldi, J. 1978. The possibilities of improving techniques of pod breaking and fermentation in the traditional process of the preparation of cocoa. Café Cacao 22:303–316. 31. Guadalupe, L. G. S., G. Loiseau, and D. Montet. 2008. Fermentation and processing of coffee and cocoa, p. 72–99. In R. C. Ray and O. P. Ward (ed.), Microbial Biotechnology in Horticulture, vol. 3. Science Publishers, Enfield, NH. 32. Hansen, C. E., M. del Olmo, and C. Burri. 1998. Enzyme activities in cocoa beans during fermentation. J. Sci. Food Agric. 77:273–281. 33. Hardy, F. 1960. Cocoa Manual, p. 350. Inter American Institute of Agricultural Science, Turrialba, Costa Rica. 34. Hardy, F., and C. Rodrigues. 1953. A Report on Cocoa Research, 1945–1951, p. 89–91. The Imperial College of Tropical Agriculture, St. Augustine, Trinidad. 35. Hodge, J. 1953. Chemistry of browning reactions in model systems. J. Agric. Food Chem. 1:928–943. 36. Hoi, O. K. 1977. Cocoa bean processing—a review. Planter (Kuala Lumpur) 53:507–530. 37. Hurst, W. J., K. B. Miller, J. Apgar, D. A. Stuart, C. Y. Lee, N. McHale, and B. Ou. 2005. The determination of polyphenols and the correlation to cocoa content in selected U.S. confectionery products. Poster, Cornell Institute of Food Science Symposium on Functional Foods, Bioactive Compounds, and Human Health, 21 May 2005, Ithaca, NY. 38. Jones, K. L., and S. E. Jones. 1984. Fermentations involved in the production of cocoa, coffee and tea, p. 433–446. In M. E. Bushell (ed.), Modern Applications of Traditional Biotechnologies. Elsevier, Amsterdam, The Netherlands. 39. Karan, N. E., and A. Belarbi. 1995. Detection of polygalacturonase and pectin esterases in lactic acid bacteria. World J. Microbiol. Biotechnol. 11:559–563.

898

Fermentations and Beneficial Microorganisms

40. Knapp, A. W. 1937. Cocoa Fermentation. A Critical Survey of Its Scientific Aspects. John Bale Sons and Curnow, London, United Kingdom. 41. Kris-Etherton, P. M., and C. L. Keen. 2002. Evidence that the antioxidant flavonoids in tea and cocoa are beneficial for cardiovascular health. Curr. Opin. Lipidol. 13:41–49. 42. Lefeber, T., M. Janssens, N. Camu, and L. De Vuyst. 2010. Kinetic analysis of strains of lactic acid bacteria and acetic acid bacteria in cocoa pulp simulation media toward development of a starter culture for cocoa bean fermentation. Appl. Environ. Microbiol. 76:7708–7716. 43. Lehrian, D. W. 1989. Recent developments in the chemistry and technology of cocoa processing, p. 22–33. In Y. B. Che Man, M. N. B. Abdul Karim, and B. A. Amabi (ed.), Food Processing Issues and Prospects. Faculty of Food Science and Biotechnology, University Pertanian Malaysia, Serdang, Selangor, Malaysia. 44. Lehrian, D. W., and G. R. Patterson. 1983. Cocoa fermentation, p. 529–575. In H. J. Rehm and G. Reed (ed.), Biotechnology, vol. 5. Chemie Verlag, Weinheim, Germany. 45. Leroy, F., and L. De Vuyst. 2004. Lactic acid bacteria as functional starter cultures for the food industry. Trends Food Sci. Technol. 15:67–78. 46. Lopez, A. S. 1972. The development of chocolate aroma from non-volatile precursors, p. 640–646. In Proceedings of Fourth International Conference on Cocoa Research. Goverment of Trinidad and Tobago, West Indies. 47. Lopez, A. S. 1986. Chemical changes occurring the processing of cocoa, p. 19–53. In P. S. Dimick (ed.), Proceedings of the Cocoa Biotechnology Symposium. Department of Food Science, Pennsylvania State University, University Park, PA. 48. Lopez, A. S., and P. S. Dimick. 1995. Cocoa fermentation, p. 562–577. In H. J. Rehm and G. Reed (ed.), Bio­ technology, vol. 9. Chemie Verlag, Weinheim, Germany. 49. Lopez, A. S., D. W. Lehrian, and L. V. Lehrian. 1978. Optimum temperature and pH of invertase of the seeds of Theobroma cocoa L. Rev. Theobromo (Brazil) 8:105–112. 50. Lopez, A. S., and V. C. Quesnel. 1971. An assessment of some claims relating to the production and composition of chocolate aroma. Int. Chocolate Rev. 26:19–24. 51. Lopez, A. S., and V. C. Quesnel. 1973. Volatile fatty acid production in cocoa fermentation and the effect on chocolate flavor. J. Sci. Food Agric. 24:319–326. 52. Lopez, C. I., E. Bautista, E. Moreno, and E. Dentan. 1989. Factors related to the formation of “over fermented coffee beans” during the wet processing method and storage of coffee, p. 1373–1384. In Proceedings of the 13th International Scientific Colloquium on Coffee. Paipa, Colombia. Assn. Scientifique du Café, Paris, France. 53. Lozano, A. R. 1958. Mechanical pod breakers. Cocoa (Turrialba) 3:35. 54. Martelli, H. L., and H. F. K. Dittmar. 1961. Cocoa fermentation. V. Yeasts isolated from cocoa beans during the curing process. Appl. Microbiol. 9:370–371.

55. McDonald, C. R., R. A. Lass, and A. S. Lopez. 1981. Cocoa drying—a review. Cocoa Growers Bull. 31:5–39. 56. Menchu, J. F., and C. Rolz. 1973. Coffee fermentation technology. Café Cacao 17:53–61. 57. Ostovar, K., and P. G. Keeney. 1973. Isolation and characterization of microorganisms involved in the fermentation of Trinidad’s cocoa beans. J. Food Sci. 38:611–617. 58. Passos, F. M. L., A. S. Lopez, and D. O. Silva. 1984. Aeration and its influence on the microbial sequence in cocoa fermentations in Bahia with emphasis on lactic acid bacteria. J. Food Sci. 49:1470–1474. 59. Passos, F. M. L., D. O. Silva, A. S. Lopez, C. L. L. F. Ferreira, and W. V. Guimarães. 1984. Characterization and distribution of lactic acid bacteria from traditional cocoa bean fermentations in Bahia. J. Food Sci. 49:205–208. 60. Quesnel, V. C. 1965. Agents inducing the death of the cocoa seeds during fermentation. J. Sci. Food Agric. 16:441–447. 61. Quesnel, V. C. 1972. Annual Report of Cocoa Research, p. 48. University of the West Indies, St. Augustine, Trinidad. 62. Quesnel, V. C. 1972. Cacao curing in retrospect and prospect, p. 602–606. In Proceedings of Fourth International Conference on Cocoa Research. Goverment of Trinidad and Tobago, West Indies. 63. Quesnel, V. C., and K. Jugmohunsingh. 1970. Browning reaction in drying cocoa. J. Sci. Food Agric. 21:537. 64. Rodriquez, D. B., and H. A. Frank. 1969. Acetaldehyde as a possible indicator of spoilage in coffee Kona (Hawaiian coffee) J. Sci. Food Agric. 20:15–19. 65. Roelofsen, P. A. 1953. Polygalacturonase activity of yeast, Neurospora and tomato extract. Biochim. Biophys. Acta 10:410–413. 66. Roelofsen, P. A. 1958. Fermentation, drying, and storage of cocoa beans. Adv. Food Res. 8:225–296. 67. Rohan, T. A. 1957. Cocoa preparation and quality. West Afr. Cocoa Res. Inst. 1956–1957:76. 68. Rohan, T. A. 1963. Processing of Raw Cocoa from the Market. FAO Agriculture Studies, no. 60. Food and Agriculture Organization, Rome, Italy. 69. Rohan, T. A. 1969. The flavor of chocolate: its precursors and a study of their reactions. Gordian 69:443–447, 500–501, 542–544, 587–589. 70. Rohan, T. A., and T. Stewart. 1965. The precursors of chocolate aroma: the distribution of free amino acids in different commercial varieties of cocoa beans. J. Food Sci. 30:416–419. 71. Rohan, T. A., and T. S. Stewart. 1967. The precursors of chocolate aroma: production of free amino acids during fermentation of cocoa beans. J. Food Sci. 32:396–398. 72. Rohan, T. A., and T. S. Stewart. 1967. The precursors of chocolate aroma: production of reducing sugars during fermentation of cocoa beans. J. Food Sci. 32:399–402. 73. Rombouts, J. E. 1952. Observations on the microflora of fermenting cocoa beans in Trinidad. Trinidad Proc. Soc. Appl. Bacteriol. 15:103–111. 74. Rombouts, J. E. 1953. Critical review of the yeast species previously described from cocoa. Trop. Agric. (Trinidad) 30:34–41.

35.  Cocoa and Coffee 75. Samah, O. A., M. F. Putih, and J. Selamat. 1992. Biochemical changes during fermentation of cocoa beans inoculated with Saccharomyces cerevisiae (wild strain). J. Food Sci. Technol. 29:341–343. 76. Samah, O. A., M. F. Puteh, J. Selemat, and H. Alimon. 1993. Fermentation products in cocoa beans inoculated with Acetobacter xylinum. ASEAN Food J. 8:22–25. 77. Sanchez, J., G. Daguenet, J. P. Guiraud, J. C. Vincent, and P. Galzy. 1985. A study of the yeast flora and the effect of pure culture seeding during the fermentation process of cocoa beans. Lebensm. Wiss. Technol. 18:69–76. 78. Saposhnikova, K. 1952. Changes in the acidity and carbohydrates during ripening of the cocoa fruit: variations of acidity and weight of seeds during fermentation of cocoa in Venezuela. Agron. Trop. (Maracay) 2:185–195. 79. Schwan, R. F. 1998. Cocoa fermentations conducted with a defined microbial cocktail inoculum. Appl. Environ. Microbiol. 64:1477–1483. 80. Schwan, R. F., and A. E. Wheals. 2003. Mixed microbial fermentations of chocolate and coffee, p. 429–449. In T. Boekhout and V. Robert (ed.), Yeasts in Food: Beneficial and Detrimental Aspects. Woodhead Publishing Ltd. Cambridge, United Kingdom. 81. Schwan, R. F., A. S. Lopez, D. O. Silva, and M. C. D. Vanetti. 1990. Influência da frequência e intervalos de revolvimentos sobre a fermentacao de cocoa e qualidade do chocolate. Rev. Agratrop. 2:22–31. 82. Schwan, R. F., A. H. Rose, and R. G. Board. 1995. Microbial fermentation of cocoa beans, with emphasis on enzymatic degradation of the pulp. J. Appl. Bacteriol. 79:96–107. 83. Schwan, R. F., M. C. D. Vanetti, D. O. Silva, A. S. Lopez, and C. A. de Moraes. 1986. Characterization and distribution of aerobic spore-forming bacteria from cocoa fermentations in Bahia. J. Food Sci. 51:1583–1584. 84. Shamsuddin, S. B., and P. S. Dimick. 1986. Qualitative and quantitative measurements of cocoa bean fermentation, p. 55–78. In P. S. Dimick (ed.), Proceedings of the Cacao Biotechnology Symposium. Department of Food Science, Pennsylvania State University, University Park, PA. 85. Sieki, K. 1973. Chemical changes during cocoa fermentation using the tray method in Nigeria. Int. Chocolate Rev. 28:38–42.

899 86. Sivetz, M. and N. Desrosier. 1979. Coffee Technology, p. 74–116. AVI Publishing Company, Inc., Westport, CT. 87. Suryakantha Raju, K., S. Vishveshwara, and C. S. Srinivasan. 1978. Association of some characters with cup quality in Coffea canephora × Coffea arabica hybrids. Indian Coffee 1978:195–197. 88. Van Pee, W., and J. M. Castelein. 1972. Study of the pectinolytic microflora, particularly the Enterobacteriaceae, from fermenting coffee in the Congo. J. Food Sci. 37:171–174. 89. Vaughn, R. H., R. DeCamargo, H. Falanghe, G. MelloAyres, and A. Serzedello. 1958. Observations on the microbiology of the coffee fermentation in Brazil. Food Technol. 12:57. 90. Velmourougane, K., D. R. Shanmukhappa, K. Ventatesh, C. B. Prakasan, and Jayarama. 2008. Use of starter culture in coffee fermentation—effect on demucilisation and cup quality. Indian Coffee 72:31–34. 91. Villeneuve, F., E. Cros, J. C. Vincent, and J. J. Macheix. 1989. Search for a cocoa fermentation index III— changes in the flavan-3-ols in the bean. Café Cacao 33:165–170. 92. Voigt, J., and B. Beihl. 1995. Precursors of the cocoa specific aroma components are derived from the vicilin class (7S) globulin of the cocoa seeds by proteolytic processing. Bot. Acta 108:283–289. 93. Voigt, J., D. Wrann, H. Heinrichs, and B. Biehl. 1994. The proteolytic formation of essential cocoa-specific aroma precursors depends on particular chemical structures of the vicilin-class globulin of the cocoa seeds lacking in the globular storage proteins of coconuts, hazelnuts and sunflower seeds. Food Chem. 51:197–205. 94. Wadsworth, R. V. 1955. The preparation of cocoa, p. 131–142. In Cocoa Conference. Report of the 1955 London Cocoa Conference, Grosvenor House, 13 to 15 September 1955, London, United Kingdom. 95. Weissberger, W., T. E. Kavanagh, and P. G. Keeney. 1971. Identification and quantification of several non-volatile organic acids of cocoa beans. J. Food Sci. 36:877–879. 96. Zak, D. K., K. Ostovar, and P. G. Keeney. 1972. Implication of Bacillus subtilis in the synthesis of tetramethylpyrazine during fermentation of cocoa beans. J. Food Sci. 37:967–968.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch36

36

Iain Campbell

Beer

This chapter is a general overview of the scientific principles of the brewing industry. More detailed information on the science and technology of beer production is available in textbooks (3, 9, 10, 12, 13). Detailed information on the fermentation aspects of brewing can be found in recent texts (1, 18, 22). The legal definition of beer varies among countries. The strictest definition, as in Germany, limits the ­ingredients to hops, yeast, water, and malt, not necessarily malted barley, although that cereal is understood if no other is specified. In many other countries, sugars or ­ unmalted cereals are permitted as adjuncts. Normally, the upper limit is about 30% of the grist, but in some tropical countries where barley and its malt are unavailable, acceptable beers can be brewed with 100% ­unmalted cereal plus microbial enzymes. Fig. 36.1 shows a general outline of the traditional brewing process. Archaeological evidence has shown that beer has been produced for at least 5,000 years, perhaps as a serendipitous discovery from the baking of bread. Brewing yeast is unable to ferment the starch of cereal grains, so a preliminary germination is required in which the starch and protein are hydrolyzed enzymatically into simple sugars and amino acids, which provide the main

nutrients for the yeasts to carry out the fermentation. This may have happened accidentally with moist grain, and naturally occurring fermentative yeasts would have produced a primitive beer. In subsequent developments over the millennia, barley became the principal cereal for beer production because of its husk, which provides an excellent natural filter for clarification of the extracted wort.

MALTING The first stage of the brewing process is the production of malt, now almost always by specialist maltsters. It is a common mistake to think that during malting the grains convert starch into simple sugars that the yeast can metabolize. In fact, that happens during mashing, and the changes during malting are limited to the breakdown of cell walls and the protein matrix in which the starch granules are embedded, but such modification of the grain is necessary for hydrolysis of the starch during mashing. Not all barley is suitable for malting. One important property is the nitrogen content. Too low a level of nitrogen will restrict yeast growth in the subsequent

Iain Campbell, International Centre for Brewing and Distilling, Heriot-Watt University, Edinburgh EH14 4AS, United Kingdom.

901

SMP_Food Microbiology_CH36.indd

901

Manila Typesetting Company

11/08/2012 07:18AM

Fermentations and Beneficial Microorganisms

902

Figure 36.1  The brewing process. doi:10.1128/9781555818463.ch36f1

f­ ermentation, but normally the problem is too much nitrogen in the barley, more than is necessary for yeast growth, with the resulting excess nitrogen encouraging microbial, particularly bacterial, spoilage of the final beer. Also, grain intended for malting must be carefully dried

and stored to avoid the risk of either dormancy (delayed germination) or death of the barley (3, 7, 14). Another problem associated with insufficient drying is fungal growth and possibly mycotoxin production, although this is more likely to be associated with a long period of wet weather, particularly just before harvest (6). The malting process occurs in three stages: steeping, germination, and kilning. Steeping in water over 24 to 48 h, usually in two stages with an “air rest” between, stimulates the growth of the embryo plant, as would occur if it had been planted in soil. Germination is controlled by temperature, aeration, and humidity to the point where the stem is just about to emerge from the grain, by which time rootlets have already formed. By that stage, the zone of modification of the starch has ­extended almost to the end of the grain opposite from the embryo. In the traditional malting process, grain transferred from the steeping vessel was spread onto the floor in a layer about 0.8 m deep and manually turned daily to prevent the developing rootlets from binding together during the 2 weeks of germination. Since late in the 19th century, many types of mechanical malting devices have been developed to speed up the process and reduce the labor requirements. One of the first, ­invented by Saladin, which still is operated in essentially the same design, is shown in Fig. 36.2. A row of rotating helical turners is carried on a gantry that runs at regular ­intervals along the full length of the germination vessel to turn the germinating grain, which is aerated by the humidified air passing through the slotted floor. At the end of germination, the grain has a moisture content of about 50%, so drying to <6% moisture is required for storage. However, kilning is more complex than simply drying the grain. Malt is not only a source of fermentable sugars and other yeast nutrients; it is also a source of amylolytic enzymes for hydrolysis of any additional unmalted cereal in the recipe. Although

Figure 36.2  Saladin malting system. The row of helical turners moves from one end of the grain bed to the other at intervals during germination to turn the malt and prevent matting of the rootlets. doi:10.1128/9781555818463.ch36f2

SMP_Food Microbiology_CH36.indd

902

Manila Typesetting Company

11/08/2012 07:18AM

36.  Beer

903

these enzymes are moderately heat resistant, they are increasingly inactivated by drying temperatures above 70°C. Also, higher temperatures give darker-colored malt because of browning reactions, first explained by Maillard, between the sugars and nitrogen components of the grain. For some beers, darker malts are desirable, although with the penalty of reduced enzymatic activity for hydrolysis of the starch of cereal adjuncts. The science of malt production is more the concern of the botanist and biochemist, but there are important ­microbiological aspects. First, malt and malt ­ extracts are food products, so high standards of hygiene are ­required. Although some bacterial contamination is inevitable, ­ potentially the most troublesome contaminants are fungi. Certain fungi have specific spoilage or toxigenic effects: the production of mycotoxins or, in the case of some species of Fusarium, of a polypeptide “gushing factor” that, by creating nuclei for development of carbon dioxide bubbles, causes violent frothing of the beer when the bottle or can is opened. However, since the growth of any mold on barley gives an obvious “weathered” appearance and a moldy aroma which persists to the final beer, simple sensory assessment is sufficient for the maltster to decide on acceptance or rejection of a barley delivery. In fact, the maltster normally has no alternative to such an immediate assessment when deciding to accept a load of grain from a waiting truck, since conventional microbiological analysis would require at least 1 week.

MASHING AND WORT PRODUCTION Wort is the sugary solution prepared from malt, either alone or with sugar (e.g., glucose, sucrose, or maltose crystals or syrups) or unmalted cereal adjunct if ­appropriate, after the grist is extracted with warm water. Details of the process vary among breweries, and it is impracticable to deal with all of the possibilities here. It is true that most of these adjuncts are cheaper than malt, but an important reason for their use is to control the color and flavor of the final beer, e.g., by including a small proportion of roasted barley in the recipe. In traditional milling, although the contents of the grains must be ground finely to maximize the yield of fermentable extract, it is important that the outermost layer, the husk, is only cracked rather than ground to powder. In a later stage of the process, the husk acts as a natural filter. Barley has become the dominant cereal of the brewing industry largely because its husk structure is ideal for this purpose. The traditional decoction mashing process for Bavarian and Czech beers has origins predating the invention of

SMP_Food Microbiology_CH36.indd

903

the thermometer. Consistent brewing requires reproducible conditions, and consistent temperatures were achieved by mixing the milled grist with a measured volume of well water, heating a measured proportion to boiling, and returning the boiled slurry to the mash tun. Well water is of constant temperature throughout the year, and water always boils at the same temperature at a given altitude, so with the same volumes each time, mashing temperatures would always be the same. Experience revealed that several steps of increasing temperature were required, which we now know are the optima of α- and β-amylases and protease activity of the malt. This stepwise increase was achieved by a succession of accurately measured decoctions. While double- or triple-decoction mashing undoubtedly produces a good-quality wort, even from poorly modified malts, the process is expensive to operate and now is used only for prestigious products for which the traditional pro­ cess confers sales appeal. Infusion mashing is also a traditional process, associated particularly with British ales and similar beers of Belgium and northern Germany. At its simplest, the mash tun contains a floating bed of grist, suspended on entrapped air, with the wort drawn off at the base continuously replaced by a top spray of hot sparge water until further extraction of sugar is not practical. Originally, infusion mashing operated at a constant temperature of about 70°C (traditionally, the temperature at which the brewer’s reflection was just obscured by the steam beginning to rise from the brewing liquor as it was heated), and well-modified malt was essential (3). Most breweries now use a mashing vessel in which the malt grist is mixed with warm water, heated in steps by using a steam jacket, rather than decoction, through the optimum temperatures of the various mashing ­enzymes. The resulting wort is then clarified in the lauter tun (from the German lauter = clear), which has a slotted metal base on which the bed of husk material functions as a filter, since all cereal solids must be removed before the next stage of the process. When raw unmalted cereal forms part of the recipe, the ground cereal is heated in a cereal cooker at least to the gelatinization temperature of the starch of the cereal (70 to 75°C in the case of maize, a common ingredient), but often to boiling, and then transferred to the mashing vessel to allow hydrolysis of its starch and protein by the malt enzymes. In recent years, many breweries have replaced the lauter tun with a mash filter, in reality a type of filter press (13). As in the lauter tun, the cereal solids constitute the filter bed, but they are supported on vertical polypropylene sheets with numerous fine perforations, much smaller than the slots in the bottom plates of the

Manila Typesetting Company

11/08/2012 07:18AM

Fermentations and Beneficial Microorganisms

904 lauter tun. Therefore, the malt can be ground more finely than for a lauter tun, giving better extraction of nutrients, but the main advantage of the mash filter is that in each of the filter chambers the mash can be squeezed to expel the wort. The process is faster than filtration in a lauter tun, and a much smaller amount of sparge liquor is required. Therefore, stronger worts can be produced, and more quickly.

HOPS AND HOP BOILING Over brewing history, many different herbs have been used as flavoring for beer, but for the past 5 centuries hops have been preferred, almost certainly for microbiological reasons. Certainly it is now recognized that hops have an important antimicrobial, particularly ­antibacterial, effect, and it is presumed that the medieval brewers realized that hopped beers maintained their quality for longer periods of time than did beers with other flavorings. Hops require a temperate climate to grow, but in practice their cultivation is restricted to certain areas, e.g., the county of Kent in Britain and Washington state in the United States. The different hop varieties in common use vary in their content of the bitter acids, resins, and oils that contribute to beer flavor and aroma and in whether they contain seeds. Many brewers believe that hop seeds impart a harsh flavor to beer. The choice of a particular variety, or blend of varieties, is an important part of a beer recipe. Flavor is extracted by boiling, which isomerizes the hop α- and β-acids (see below) and purges the wort of harsh, grainy flavors, with the incidental advantages of sterilizing and concentrating the wort. In appearance, hops resemble small pinecones but have a softer texture. Cone hops are still used by some traditional breweries, but most breweries now use processed hops, either hop extract or pellets prepared from ground cone material. Although various organic solvents were initially used for production of extracts, modern hop extraction technology uses liquid carbon dioxide to avoid potential problems of residual solvent. Also, various pre-isomerized hop products are available, which can be added during the last few minutes of boiling to reduce evaporation losses. Thus, it is interesting that the boiling stage, originally intended for extraction of flavors, is now in many breweries simply a process of biochemical and microbiological stabilization of the wort. To the brewer, the most important hop components are the resins, tannins, and essential oils (21). Typically, these constitute approximately 14, 4, and 0.5%, respec-

SMP_Food Microbiology_CH36.indd

904

tively, of the weight of the mature hop cones. Bitterness of beer is due to the α-acid (humulone) and β-acid ­(lupulone) components of the resins, but not ­ directly: these complex acids are isomerized during hop boiling into the bitter iso-α-acid isohumulone and iso-βacid hulupone, respectively (10, 12, 21). These names are derived from the botanical name for hop, Humulus ­lupulus. Within these general groups of iso-acids are numerous analogs according to the acyl side chains on the resin molecule. Essential oils contribute aroma to the beer, but most of the oil is lost during prolonged boiling. Therefore, aroma hops, as distinct from high-α-acid bitterness hops, must be added late in the boil to ensure maximum aroma ­effect. Also, if sugar crystals or syrups are included in the recipe, this is the appropriate stage for addition. Late addition provides adequate protection against microbial contamination but prevents significant darkening of the wort by caramelization. Hop tannins have little direct influence on flavor but react with malt protein to form “hot break,” a precipitate of protein and tannin complexes and ­insoluble calcium salts and phosphates. This deposit must be ­removed ­before fermentation, since the particulate ­material would ­adversely ­affect fermentation and flavor ­development by the yeast. After traditional hop boiling with cone hops, the hot break is removed from the wort by filtration through the settled bed of spent hops. Pellets or extract do not provide a suitable filter medium, so wort prepared with these products must be clarified by centrifugation. In modern plants, the centrifugal effect is provided by a whirlpool hop separator: the hopped wort is run tangentially into a circular tank, often with a slightly concave or coned base. Hop debris and hot break collect at the center of the resulting vortex, whereas clear wort is drawn from the side of the vessel. The wort, still at approximately 100°C, has to be cooled to 20°C or lower before “pitching” with the yeast inoculum. Less obvious, but nevertheless essential, is the requirement, explained below, to aerate the wort to contain 6 to 8 µg of dissolved oxygen/ml. After cooling and aeration, the wort is ready for the fermentation stage of the process.

FERMENTATION

History

As mentioned above, it is likely that many of the ­earliest beers were fermented by chance inoculation. Such “natural” fermentations are still common in cider and grape wine production, but the only significant example still practiced in the European brewing industry is the

Manila Typesetting Company

11/08/2012 07:18AM

36.  Beer

905

l­ambic beer of Belgium (4). After hop boiling, wort is run into shallow open trays and cooled by evaporation from their large surface area. The inoculum of bacteria and yeasts is partly the resident microbial flora of the brewery equipment, the cooling trays and fermentation vessels in particular, and partly chance arrivals from outside. So there is an element of good luck in producing a beer of good flavor character. The lengthy fermentation involves a succession of yeasts, typically including various Brettanomyces spp., which produce both ethanol and acetic acid, but their limited acid production creates a refreshing sharpness rather than disagreeable sour flavor. Although the fermentation is normally completed by naturally developing Saccharomyces spp., occasionally a cultured brewing yeast may be added to rescue a failing fermentation.

Brewing Yeast (Saccharomyces cerevisiae)

Formerly, the actively fermenting yeasts of the fermentation industries, both culture yeasts and common contaminant “wild yeasts,” were classified as different ­species of Saccharomyces. S. cerevisiae of the traditional northern European ales and similar beers was collected by skimming off the “yeast head” or “top yeast” that rose to the surface of an active fermentation, whereas S. carlsbergensis, “bottom yeast,” which did not form such a head, had to be collected from the bottom of the vessel at the end of fermentation (4). Originally, bottom yeasts were used only for Czech, Danish, and German pilsner-style beers, but now non-head-forming yeasts are more widely used to maximize the useful capacity of enclosed fermentation vessels. Saccharomyces baya­ nus and Saccharomyces uvarum, which are important culture yeasts in the wine industry, are also possible brewery contaminants, as is Saccharomyces diastaticus, an amylolytic yeast. Most of these species are now classified officially by yeast taxonomists as a single species, S. cerevisiae (4), but still it is convenient in the brewing industry to distinguish the different types by their former specific names. It is important to realize that S. cerevisiae is not a true facultative anaerobe like Escherichia coli (18, 22). Although a brewing yeast changes between oxidative and fermentative metabolism according to envi­ ronmental aerobic or anaerobic conditions, it ­ cannot grow indefinitely under anaerobic conditions. As with all eukaryotic cells, yeast cell membranes contain ­unsaturated fatty acids and sterols, which can be synthesized only in the presence of atmospheric oxygen. The amounts of unsaturated fatty acids in malt and sterols naturally present in wort are too low to support yeast growth, hence the requirement for initial

SMP_Food Microbiology_CH36.indd

905

a­ eration of the wort to allow the yeast to synthesize these compounds. “Pitching yeasts” in satisfactory condition must be added in the correct amount to inoculate the fermentation vessel: usually to 1 × 107 to 2 × 107 cells/ml, although measurement of yeast cake or slurry by weight is more convenient on a production scale. In the course of a typical fermentation, the yeast population increases by a factor of about 8, i.e., only three successive cell divisions. Subsequent multiplication is inhibited by the lack of oxygen, unsaturated fatty acids, and sterols, and since aeration late in the fermentation would cause flavor problems, no further cell growth is possible. A brewery fermentation is essentially the lag, log, and early stationary phases of the yeast growth curve (Fig. 36.3). Brewing is unusual in modern biotechnology in reusing the culture from the previous fermentation, although the culture is normally replaced at ­regular intervals, e.g., after 10 successive fermentations; the number varies among breweries. During the lag phase of 6 to 12 h, the yeast utilizes the dissolved oxygen in the wort to restore its unsaturated fatty acid and sterol supply and to adjust from the anaerobic, acid, alcoholic conditions at the end of the previous fermentation to the very different environment of fresh wort. As it is the lag phase, there is no detectable alcohol production or increase in yeast population.

Figure 36.3  Theoretical progress of a typical brewery fermentation, showing changes in the population of S. cerevisiae and concentrations of fermentable sugar, amino nitrogen, and ethanol. The graph shows yeasts in suspension and the start of settling of cells from the beer at the end of fermentation. The time axis is not calibrated, since fermentation rates differ widely among breweries. Other variables are shown as percentages of the initial or final value, expressed as 100%. doi:10.1128/9781555818463.ch36f3

Manila Typesetting Company

11/08/2012 07:18AM

Fermentations and Beneficial Microorganisms

906 Yeast nutrients, including sugars, are taken up from the wort to generate new yeast cells; fermentable sugars also provide the energy for this process. A traditional­-strength wort of 10° Plato, i.e., 10% solute, has a specific­ gravity of 1.040, which is largely due to the ­sugars in solution. Amino acids, vitamins, and inorganic salts ­ account for less than 1% of the dissolved solids. Typically, the approximate sugar composition of such a wort would be as follows: maltose, 4 to 5%; glucose and maltotriose, about 3% in total; and higher dextrins, which are not utilized by brewing yeast, about 2%. However, the various fermentable sugars are not utilized simultaneously: this is related to the transport of nutrients into the cells and varies with different strains. Normally, brewing yeasts transport and metabolize the simplest sugar first, i.e., glucose, and maltose transport begins with synthesis of the maltose permease system only after most of the glucose has been metabolized. For a similar reason, ­utilization of maltotriose begins late in the fermentation, when most of the maltose has been consumed. Note also in Fig. 36.3 that fermentation of sugars continues after yeast growth ceases, due to the ­ continuing action of the relevant enzymes, but more slowly than before. However, without yeast growth and protein synthesis, there is no further requirement for amino nitrogen; in fact, there may be some release of amino nitrogen from the small percentage of dying cells. Ethanol and carbon dioxide are the main products of yeast metabolism, but small amounts of other compounds make an important contribution to the flavor of the beer (1, 18). By-products of the Embden-Meyerhof pathway, the major metabolic route in the anaerobic conditions of fermentation, contribute to flavor, but a major effect is the nitrogen metabolism of yeasts. S. cerevisiae

has a limited range of amino acid permeases and most amino acids for protein synthesis have to be synthesized by the yeast cell. Keto acids, which are intermediates in the biosynthetic pathway, may be converted to higher alcohols (Fig. 36.4), which make a significant contribution to the flavor of beer and indeed all fermented beverages. Also related to transport is a decrease in pH during the fermentation, a result of excretion of H+ during the uptake of nutrients. The pH of cereal wort is normally 5.1 to 5.2 and decreases during fermentation to 3.8 to 4.0, mainly during the period of yeast growth. Acetyl coenzyme A (acetyl-CoA), shown as an intermediate in Fig. 36.4, is only one of the acyl-CoAs that are important in the biosynthetic activities of yeasts. With the lack of oxygen under fermentation conditions limiting yeast growth and therefore reducing demand for these CoA complexes, the acyl groups may have to be recycled, which is most conveniently achieved by esterification with alcohol (13). Since acetyl-CoA and ethanol are present in greatest amounts, ethyl acetate is the principal ester formed during fermentation, but other ester combinations occurring in smaller amounts may have a more powerful effect on beer flavor.

Progress of Fermentation

Traditional fermentation vessels are open rectangular tanks 2 to 3 m in depth (10). Such vessels are particularly useful for top yeasts, which rise to the surface of the fermenting beer, from where they are skimmed off as the inoculum for the next fermentation. Since about 1970 the cylindroconical fermentation vessel (CCFV) (Fig. 36.5) has become the preferred type (12, 13). These vessels are enclosed to reduce the risk of contamination and facilitate recovery of carbon dioxide and have a

Figure 36.4  Formation of higher alcohols (fusel alcohols) as by-products of amino acid biosynthesis. Note the similarity of the reactions pyruvate → acetaldehyde → ethanol and α-keto acid → aldehyde → higher alcohols; both are important for redox balance under anaerobic conditions. doi:10.1128/9781555818463.ch36f4

SMP_Food Microbiology_CH36.indd

906

Manila Typesetting Company

11/08/2012 07:18AM

36.  Beer

907 late in the fermentation encourages the production of ­unwanted diacetyl. Therefore, the solution is to add with the first batch of wort most of the yeast and dissolved oxygen required, perhaps three times as much as that volume of wort would normally require. The rest of the pitching yeast and dissolved oxygen are added with the second batch of wort, and the later wort additions are fermented by the yeast already present. An important property of brewing yeasts is the ability to flocculate, i.e., spontaneously aggregate into clumps, late in the fermentation. Too early flocculation stops the fermentation as the yeasts settle out of partly fermented beer, but nonflocculent yeasts are also troublesome ­because they have to be removed by filtration or centrifugation. Currently, the preferred explanation of flocculation is a lectinlike activity of the cell wall that develops during fermentation, but other factors are also involved, e.g., divalent ions, particularly Ca2+; ­decreasing amounts of flocculation-inhibiting fermentable sugar; and increasing amounts of stimulatory ethanol (19).

Figure 36.5  CCFV. Vigorous circulation of fermenting wort is created by a central upward flow of CO2 bubbles and a downward flow in contact with the cooling jackets. Cooling of the cone section is optional, depending on whether the brewery stores settled yeast before the next fermentation. doi:10.1128/9781555818463.ch36f5

conical lower section to facilitate recovery of bottom yeast, now almost universally used, which settles out late in the fermentation. The shape of the vessel also promotes a vigorous mixing of the wort and yeasts, with a central upward movement due to the bubbles of carbon dioxide and peripheral downward movement next to the cooling panels on the walls, encouraging a faster fermentation. In recent years some of the larger brewery companies have installed very large CCFVs, of capacities of up to 6,000 hl, for large-scale production of their principal brands. Although all CCFVs operate in a similar way, an important difference is that the very large vessels may be four to five times the size of the brewhouse and must be filled in installments. The first batch of wort must be inoculated immediately, as it is dangerously bad microbiological practice to store uninoculated wort. The wort is aerated and the yeast begins to grow, and the second, third, fourth, and possibly even fifth batches of wort are added as available. Unfortunately, there is a biochemical problem. The wort must be aerated because yeasts need oxygen to grow efficiently. An intermediate state of oxygen deficiency causes excessive production of ­esters, especially ethyl acetate. However, oxygen added

SMP_Food Microbiology_CH36.indd

907

POSTFERMENTATION TREATMENTS

Conditioning

Immature “green” beer from the fermentation vessel contains acetaldehyde, diacetyl, and other unwanted by-products of yeast metabolism, which must be removed (10, 18). Although present in only small amounts, these by-products have a marked effect on the flavor, or more correctly the aroma, of the beer. In traditional cask conditioning, the beer, still with about 1% fermentable sugar, undergoes a secondary fermentation in casks, which ­absorbs these undesirable flavor compounds and generates sufficient carbon dioxide for dispense. Similar changes are associated with the traditional low-­temperature (0 to 2°C) secondary fermentation over several months for pilsner and other lager beers (from the German lager = to store). It is now known that long storage at low temperature is not essential; the beneficial flavor changes can be achieved more rapidly by a few days’ storage of the beer in contact with 105 to 106 yeast cells/ml at 15°C, followed by cooling to 0°C only long enough to precipitate “chill haze” material and yeast sediment. It is important to avoid accidental access to air after yeasts have settled out, because diacetyl is formed from an acetolactate precursor by spontaneous chemical reactions in the presence of oxygen and yeast is no longer present to remove the buttery off-flavor of diacetyl.

Filtration

With the exception of cask-conditioned beers, which are clarified by the addition of fining agents, it is common

Manila Typesetting Company

11/08/2012 07:18AM

Fermentations and Beneficial Microorganisms

908 practice now to filter beer to complete clarity. Various designs of filters are in common use, most using either cellulose fibers or diatomite or pumice powder as the filter medium (11). Such materials, which adsorb microbial and inert haze-forming material in the depths of the filter, are not used for sterilization of the beer because sufficiently fine filters would cause unacceptably low flow rates. An efficient filter will reduce the yeast count to <100 cells/liter (whereas at least 108 yeast cells/liter are required to form a visible haze), and inactivation of the remaining yeasts is achieved by pasteurization. However, membrane filtration, which does sterilize beer, is becoming increasingly popular. Pasteurization is unnecessary for sterile-filtered beer, so membrane filtration avoids the substantial energy cost and the potential flavor defects caused by heating the beer. With two filters used in series—a rough prefilter to remove as much particulate material as possible and the second filter, the membrane filter, to sterilize—an acceptable flow rate and long filter life can be achieved.

Pasteurization and Packaging

Draft beer is pasteurized by passage through a heat exchanger at about 70°C before filling into already cleaned and sterilized kegs. Beer for sale in cans or glass bottles is pasteurized after filling. In both systems, the heat treatment is equal in terms of pasteurization units (PU; 1 PU = 60°C for 1 min), but the “tunnel” pasteurizer for bottles and cans uses a lower temperature (typically 62°C) and longer treatment time than the heat ­exchanger. Individual breweries have their own preferred pasteurization treatment. In theory, 5 PU is sufficient to kill the small numbers of brewing yeasts likely to pass through a cellulose filter, but to eliminate the slightly more heat-resistant bacterial or wild-yeast contaminants that may occasionally be present, up to 30 PU may be applied. Since increasing heat treatment may adversely affect flavor, the choice of PU value is a compromise between potential risks of oxidized flavors and microbial spoilage (10, 13). The polyethylene terephthalate (PET) bottles that are increasingly being used for beer cannot withstand pasteurization temperature; hence they must be filled aseptically with membrane­filtered or pasteurized beer.

HIGH-GRAVITY BREWING The basic principle of high-gravity brewing is that it is theoretically possible to double the production of a brewery, without the expense of additional brewhouse or fermentation capacity, by fermenting double-strength wort. The resulting double-strength beer would be ­diluted

SMP_Food Microbiology_CH36.indd

908

to normal strength immediately before packaging. The economic advantages are obvious. At first sight, the ­addition of water to beer might seem rather questionable,­ but that water would have been used anyway in the first stage of the standard process. Unfortunately, the idea is very difficult in practice. In order to match the flavor of high-gravity beer with that of the normal product, the concentrations of everything must be doubled, including malt, hops, yeast, and even the dissolved oxygen concentration in the wort. Addressing the nonmicrobiological aspects first: with standard mash and lauter tuns, doubling the weight of malt does not yield double the extract. Extraction becomes less efficient at higher concentrations, and ­efficient sparging of the grain bed in the lauter tun ­dilutes the wort excessively. Therefore, until the development of a mash filter that can produce high-gravity wort from an all-malt grist, it was necessary to use sugar syrup ­adjuncts to increase the gravity, which is illegal in some countries. There is a similar problem with hops, because extraction becomes progressively less efficient with larger quantities of hops or hop pellets. That particular problem was solved by the use of hop extracts, especially preisomerized extracts that may be added ­either late in hop boiling or at the stage of dilution of high-gravity beer to normal strength. Although the brewer can easily add twice the previous amount of yeast to the double-strength wort, it is difficult to achieve twice the dissolved oxygen concentration. A maximum concentration of 6 to 8 µg of dissolved oxygen/ml is routinely achieved by injecting air into wort between the cooler (after hop boiling) and the fermentation vessel. In the high-gravity process, twice the amount of yeast would require 12 to 16 µg of dissolved oxygen/ml, achieved only by injecting pure oxygen rather than air. As noted earlier, reduced oxygen levels cause excess ester production. A standard wort of specific gravity 1.040 is termed 10° Plato, i.e., it contains 10% sugar. Double-strength wort of 20% sugar is likely to retard yeast growth by ­osmotic stress. A more serious situation occurs at the end of fermentation, when the yeast is recovered from beer of 8% instead of the previous 4% alcohol by ­ volume and its viability may be too low for successful repitching in the next fermentation. A brewery that already has a yeast culture plant, previously used to replace the yeast culture occasionally, could easily prepare a new yeast culture for each fermentation. Since the plant is already available, the additional expense is small in comparison with the savings from high-gravity operation. Finally, the dilution water must be deoxygenated to a level of <0.05 µg of dissolved oxygen/ml to prevent any ­flavor

Manila Typesetting Company

11/08/2012 07:18AM

36.  Beer

909

­ roblems from staling reactions or development of p ­diacetyl. That figure has decreased gradually over the years as improved technology has made such low dissolved oxygen levels achievable. The essential deaeration plant is the only equipment that is required ­specifically for high-gravity operations. The example described above of brewing doublestrength wort has only recently been successfully accomplished in practice. Originally, 1.5 times normal strength was the practical limit, because of flavor problems. Without sufficiently efficient oxygenation of the wort in the first stages of the fermentation, the yeast is unable to grow normally and produces excessive amounts of esters, as indicated in the section above on very large cylindroconical vessels. However, with recent developments in oxygenation and the use of fresh, aerobically grown yeasts for each fermentation, twice the normal strength is now practicable.

MICROBIAL CONTAMINANTS OF THE BREWING PROCESS In comparison with other foods and drinks, beer is a relatively stable product, protected by its alcohol and carbon dioxide contents, anaerobiosis, low pH, and the antibacterial properties of hops, in addition to the heat treatments of wort boiling and beer pasteurization. Even so, beer production carries a risk of microbial contamination at all stages of the process. On barley growing in the field, and on germinating barley during malting, the most important contaminants are molds. The field fungi that develop during cultivation are seldom a problem, and then only after a particularly wet growing season which allows Fusarium to develop sufficiently to cause a problem of “gushing” (6). However, storage fungi, principally Aspergillus and Penicillium spp. which develop on improperly dried barley after harvest, can have serious effects, as mentioned above, in forming mycotoxins. Slow-growing, strictly aerobic molds present no hazard to the brewing process or postfermentation beer. The lactic acid bacteria Lactobacillus and Pediococcus are mentioned below as possible contaminants of fer­ mentation and of finished beer. These occur on the surface of barley grains, and grain dust released into the brewery atmosphere may be the cause of such contamination. However, encouraging the growth of preexisting bacterial flora or deliberately added starter cultures ­during malt steeping and germination can have several beneficial effects. Some lactic acid bacteria are weakly proteolytic, sufficient to inactivate the polypeptide ­ believed to be associated with excessive foaming

SMP_Food Microbiology_CH36.indd

909

(­ gushing) of beer mentioned above. Also, an appropriate blend of “lactic malt” with the normal grist can be used for downward pH adjustment if required. These effects are particularly useful in countries where additives are prohibited: malt is a necessary ingredient, not an additive. Benefits in addition to the flavor contribution of lactate ions include increased amino nitrogen, giving improved yeast nutrition, and reduced glucan, resulting in improved wort filtration. These deliberately grown lactic acid bacteria are not a hazard to subsequent fermentation since most, if not all, are killed by the temperature of kilning. In the brewery itself, yeasts and bacteria are potentially the troublesome contaminants. In beer, its acidic (usually about pH 3.9), alcoholic, anaerobic characteristics, with the additional antibacterial effects of carbon dioxide and hop acids and oils, restrict the range of potential spoilage microorganisms. Yeasts on grain are mainly aerobes that are normally unable to grow in fermenting wort or beer, but some species of the genera Brettanomyces, Candida, Debaryomyces, Pichia, Saccharomyces, Torulaspora, and Zygosaccharomyces have been isolated from instances of beer spoilage (4). Presumably their natural habitat and original source are plants, but once established in a brewery they are difficult to eradicate and are often transferred from one fermentation to the next in the pitching yeast, in increasing numbers each time. Members of the fermentative genera Saccharomyces, Torulaspora, and Zygosaccharomyces and equivalent non-spore-forming Candida spp. may contaminate the fermentation or persist through filtration and pasteurization to produce turbidity and off­flavors in the beer. Most of these contaminants have ­diameters smaller than the 5- to 7-μm diameters of culture yeasts, so filters designed to retain brewing yeasts are less efficient with contaminants. Dekkera and its non-spore-forming anamorph Brettanomyces cause acetic acid spoilage and turbidity. Debaryomyces, Pichia, and equivalent aerobic Candida spp. are oxidative yeasts and so are limited to the early stages of fermentation (where they often grow faster than culture yeasts) or to beer to which air has gained access after fermentation. These yeasts cause turbidity and yeasty or estery off­flavors and in bottled or canned beer often form a surface film fragmenting into flaky particles or a deposit. Only a few species of the lactic acid bacteria Lacto­ bacillus and Pediococcus are sufficiently resistant to the antibacterial properties of hopped beer to grow at all stages of the process. If able to grow, these bacteria cause turbidity and off-flavor, often caused by diacetyl (15). Diacetyl is a strongly flavored minor by-product of their metabolism that is highly valued in the dairy ­ industry

Manila Typesetting Company

11/08/2012 07:18AM

Fermentations and Beneficial Microorganisms

910 as the flavor of butter but is generally considered to be a spoilage fault of beer, especially the more delicately flavored pilsner type. Superattenuation (unwanted continuing alcohol production) by starch-­fermenting lactobacilli and slime formation by pediococci are other ­possible faults. Zymomonas, like the lactic acid bacteria, is capable of growth under anaerobic conditions and grows at all stages of the process, but its most usual effect is in packaged beer, causing further fermentation with turbidity, off-flavors, and often “rotten apple” or dimethyl sulfide aromas. Other possible spoilage bacteria are limited by their properties to specific stages of the process (20). Acetobacter and Gluconobacter are strict aerobes and grow, with characteristic acetic acid production, only in beer accidentally exposed to air. Obesumbacterium, the most important genus of the Enterobacteriaceae so far as beer spoilage is concerned, causes turbidity and off-flavor and often produces indole, phenols, diacetyl, hydrogen sulfide, and dimethyl sulfide. These bacteria grow well in the early stages of fermentation until inhibited by the decreasing pH and increasing ethanol content. Even so, they survive to be transferred with the yeast recovered for the next fermentation. Finally,

Megasphaera (cocci) and Pectinatus (rods) species are recently discovered strictly anaerobic gram-negative bacteria that form acetic, butyric, and propionic acids, ­hydrogen sulfide, dimethyl sulfide, and turbidity and have become troublesome only because of modern advan­ces in maintaining very low dissolved oxygen ­levels in beer. Other bacteria are occasionally isolated during beer sampling, e.g., Bacillus and Micrococcus spp., but these are survivors from earlier contamination and are unable to grow in wort or beer. Their main nuisance effect is the further laboratory testing required to confirm that they are not the more troublesome Lactobacillus or Pediococcus, which are of similar appearance on a Gram-stained film (15). A simple distinguishing test is that the lactic acid bacteria do not form the enzyme catalase (Fig. 36.6).

Isolation of Microbial Contaminants

Culture media for microbiological quality control can be either nonselective or selective. All brewing yeasts and bacteria should grow on nonselective media, usually either malt extract agar or Wallerstein Laboratories nutrient agar, a semisynthetic medium of similar nutrient value but more consistent composition. These media are useful for examination, by plating or membrane filtration, of

Figure 36.6  Identification of bacterial contaminants of the brewing industry. Bacillus, Micrococcus, and Streptococcus may be present but are unlikely to grow in beer, and Megasphaera and Pectinatus grow only under strictly anaerobic conditions. doi:10.1128/9781555818463.ch36f6

SMP_Food Microbiology_CH36.indd

910

Manila Typesetting Company

11/08/2012 07:18AM

36.  Beer

911

samples in which no microorganisms should be found or should be present only in low numbers, e.g., pasteurized beer or swabs or rinsings from recently cleaned and sanitized plant equipment. Any microorganism, whether bacterium, culture yeast, or wild yeast, can be regarded as a contaminant in these situations (5, 15, 20). Selective media are intended to suppress the growth of culture yeasts but allow the growth of any contaminants that may be present. Such media should normally be used to examine samples of pitching yeast or samples taken during or immediately after fermentation in which culture yeast is present. A commonly used selective­ medium­ is lysine agar, a synthetic medium of glucose, inorganic salts, vitamins, and l-lysine as a sole nitrogen source. Saccharomyces spp. are unable to grow, but yeasts of other genera do grow by using the NH2 groups of lysine for nitrogen requirements. Unfortunately, wild strains of S. cerevisiae and other Saccharomyces spp. are particularly likely contaminants but cannot be detected on lysine agar. However, diastatic yeast contaminants, including the former S. diastaticus, can be isolated on another medium based on the same principle, containing starch as a sole carbon source and ammonium sulfate, which all yeasts can utilize as a nitrogen source. The most useful selective agent for detection or counting of bacterial contaminants is the antifungal ­antibiotic cycloheximide (also known as actidione), which can be added to any suitable culture medium but is usually added to Wallerstein Laboratories nutrient agar as “actidione agar” (5). Some wild yeasts are sufficiently resistant to grow, but in general, a more effective way to recover wild Saccharomyces contaminants relies on their sporulation. Sporulation, which is stimulated by starvation conditions, is presumably advantageous in nature, but most brewery culture yeasts, after innumerable generations in their rich culture medium, have lost the ability to form viable spores. Yeast spores, unlike those of bacteria, are only marginally more heat resistant than the vegetative cells, but the difference is sufficient to recover contaminants on nonselective media after heat treatment. Practical details of microbiological analyses are fully explained in analytical manuals (5). In general, complete identification of recovered contaminants is unnecessary. Bacteria and non-Saccharomyces wild yeasts need to be identified to the genus level only, as indicated in Fig. 36.6 and 36.7. Only fermentative yeasts need further identification to determine whether or not they are Saccharomyces. If so, they may be S. cerevisiae, in which case even that identification to the species level is insufficient. It is important to determine whether it is the culture yeast itself, possibly persisting on improperly cleaned fermentor or other equipment surfaces or surviv-

SMP_Food Microbiology_CH36.indd

911

Figure 36.7  Identification of common yeast contaminants of the brewing industry. All genera listed are teleomorphic, i.e., form spores. Anamorphic (non-spore-forming) forms of Dekkera and Hanseniaspora are Brettanomyces and Kloeckera, respectively. The anamorph of all other genera listed is Candida. S. cerevisiae ferments glucose, sucrose, maltose, and raffinose but not lactose. doi:10.1128/9781555818463.ch36f7

ing through filtration and pasteurization. Remedial action in that situation would obviously be different from that required by the discovery of a wild S. cerevisiae strain as a contaminant. A quick and simple test to distinguish industrial yeast strains is their sensitivity or resistance to a range of antifungal compounds; most strains have their own unique pattern. In theory, a small-scale fermentation can distinguish the culture yeast from others, but the delay in obtaining useful results is unacceptably long. With recent advances in genetic methods, DNA subtyping is a potentially useful rapid method to distinguish different strains, but at present it is likely to be used only in the laboratories of the largest brewing companies (8, 15, 16).

Sterilization

In most industrial fermentation processes, steam is the usual method of sterilization of the plant. In the brewery, however, chemical sterilants are preferred, mainly because they can be used at ambient temperature and avoid attemperation problems with adjacent working fermentors (17). Modern breweries use automatic in-place cleaning and sterilization equipment to remove deposits (of yeast and organic soil derived from beer foam) by a powerful jet of water containing detergent, followed by a spray of chemical sanitizer, and finally by a spray of sterile rinse water. Until recently, caustic sterilants containing 2% NaOH and

Manila Typesetting Company

11/08/2012 07:18AM

Fermentations and Beneficial Microorganisms

912 additives were widely used, but acid sterilants based on phosphoric or peracetic (CH3COOOH) acids have become more popular, not least because they are unaffected by carbon dioxide and can be used in closed vessels without the long delay and expense of draining off that valuable gas. Most of the other chemical sanitizers of the food industries are unsuitable for various reasons. Chlorine-based sanitizers are not widely used because of the possibility of flavor problems: residual chlorine reacts with unknown compounds in beer to produce a strong medicinal flavor of organohalogen compounds. Quaternary ammonium compounds and biguanides are also regarded with suspicion. Their protective persistence on treated surfaces even after rinsing, which is so useful in other food industries, is unacceptable in the brewing industry because of the possible adverse effects of residual sanitizer on beer foam.

MODERN DEVELOPMENTS For most of its history, brewing was a small-scale opera­ tion, either a domestic enterprise or one associated with monasteries. Developments over the past 300 years have coincided with increasingly large-scale commercial production and with general improvements in scientific knowledge and technology. Research in the brewing ­industry has improved our understanding of the processes involved, and better-quality malts and beers can be produced more rapidly and efficiently. For example, in 1900 it took at least 14 days to produce a batch of malt, mashing required up to 8 h, ale and lager fermentations could last 7 and 14 days, respectively, and sub­ sequent lager maturation could take many weeks ­longer. A century later these times have been halved, with improvement in consistency and quality, although some traditionalist consumers might disagree. New products offer possibilities for increasing market share. One recent example is diet beer or light (“lite”) beer of lower carbohydrate content and, by implication, lower calorie content. Up to 20% of the carbohydrates of wort can be dextrins, which standard brewing yeasts do not utilize and which thus remain in the beer. Even in countries where the addition of hydrolytic enzymes is not permitted, dextrin-fermenting yeasts can legally be used to ferment that carbohydrate. One possible source of such strains is hybridization between S. diastaticus and brewing yeasts, and successful results have indeed been achieved in that way. Recent genetic engineering research has also produced potentially useful diastatic yeasts, but at present commercial brewers, in Europe at any rate, are reluctant to risk the public disapproval of their use. Another obvious use of diastatic yeasts is to create beers of higher ethanol content by fermentation

SMP_Food Microbiology_CH36.indd

912

of the dextrins (8). In the currently popular “ice beer,” the ethanol content is increased by selectively freezing out part of the water content of the beer, with the incidental benefit of improved flavor and stability due to the chilling process. As another example, there are various possibilities for production of low- or nonalcoholic beers, for which health, road safety, and religious considerations have created a moderate demand in recent years. To have a beer flavor, the product must be fermented first, although usually from a weaker-than-normal wort. Distillation is a possible method for subsequent removal of the ethanol, although with the risk of removal of other flavor compounds as well. Reverse osmosis is much more specific for removal of ethanol from beer. A completely different approach is the use of a yeast unable to ferment maltose, the principal sugar of malt wort, with the advantage that a standard brewery plant is used and ­expensive distillation or reverse osmosis equipment is not required. Although some Saccharomyces spp. are unable to utilize maltose, the best results so far have been with a maltose-negative strain of Schwanniomyces occidentalis, thanks to its acceptable flavor profile. However, the residual maltose may make such a beer too sweet for some palates. Another recent development is the sale of certain brands of beer in clear bottles rather than the brown or green glass bottles that were traditionally used as protection against photochemical reactions which create a “light-struck” stale flavor, even over relatively short ­periods of storage. Although there are other possible staling reactions, e.g., those associated with excessive pasteurization, light-struck hop degradation compounds are particularly important because of their low flavor threshold. New types of hop iso-acids, often tetrahydro isomers, are more stable to light and allow brewers to satisfy the current popular demand for clear glass bottles (2, 21). As with other isomerized hop products, these can be added at any stage between the last few minutes of hop boiling and the final packaging of the beer. Continuous fermentation is not actually a modern development, but recently there has been renewed ­interest in continuous fermentation of beer. Between about 1955 and 1970 there was considerable research on continuous operation of industrial microbiological processes in general, and those in the brewing industry were no exception. At that time some breweries built continuous-fermentation systems based on a succession of two or three stirred vessels, in each of which the ­ homogeneous contents replicated the successive stages of batch ­ fermentation: aerobic growth of the yeast, ­ anaerobic fermentation, and in some plants, a

Manila Typesetting Company

11/08/2012 07:18AM

36.  Beer

913

third ­anaerobic ­vessel for flavor development and yeast separation. Other breweries preferred a single-vessel system, an unstirred tower in which the different levels represented the stages of batch fermentation, and fermented beer overflowed from a yeast separator at the top. Although undoubtedly faster and more economical than previous batch operations, a serious disadvantage of both types of continuous systems was the high cost of construction and operation in comparison with the cylindroconical vessels that were introduced in their modern form during the 1960s. Another problem was the difficulty of replicating in a different fermentation system the flavors of traditional batch beers. Recently, however, there has been renewed interest in continuous fermentation, now using immobilized-cell technology. Although very successful on a laboratory scale, full-size fermentors suffer from damage to the immobilized-cell support by the expansion of bubbles of carbon dioxide as they rise up the necessarily tall column. New types of support currently under development avoid this difficulty, but the greatest success of immobilized-cell systems has been in postfermentation conditioning or the production of low-alcohol beer, processes with less vigorous evolution of CO2 than ­ active fermentation. There is no doubt that yet more novel fermentation methods and products will be developed in the brewing industry in the future. We can also be sure that there will be a continuing demand for traditional beers, and the small breweries specializing in such products are certainly enjoying much success at the present time.

References   1. Boulton, C. A., and D. E. Quain. 2001. Brewing Yeast and Fermentation. Blackwell, London, United Kingdom.   2. Bradley, L. L. 1997. Uses of iso-alpha acids and chemically modified products. Ferment 10:48–50.   3. Briggs, D. E., J. S. Hough, R. Stevens, and T. W. Young. 1981. Malting and Brewing Science, 2nd ed., vol. I. Malt and Sweet Wort. Chapman and Hall, London, United Kingdom.   4. Campbell, I. 2003. Wild yeasts in brewing and distilling, p. 247–266. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY.   5. Campbell, I. 2003. Microbiological methods in brewing analysis, p. 367–392. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY.

SMP_Food Microbiology_CH36.indd

913

  6. Flannigan, B. 2003. The microbiota of barley and malt, p. 113–180. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY.   7. Gibson, G. 1989. Malting plant technology, p. 279–325. In G. H. Palmer (ed.), Cereal Science and Technology. Aberdeen University Press, Aberdeen, United Kingdom.   8. Hammond, J. R. M. 2003. Yeast genetics, p. 67–112. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY.   9. Hardwick, W. A. (ed.). 1995. Handbook of Brewing. Dekker, New York, NY. 10. Hough, J. S., D. E. Briggs, R. Stevens, and T. W. Young. 1982. Malting and Brewing Science, 2nd ed., vol. II. Hopped Wort and Beer. Chapman and Hall, London, United Kingdom. 11. Leeder, G. 1998. Design of a state-of-the-art filter cellar. Ferment 11:108–121. 12. Lewis, M. J., and T. W. Young. 1995. Brewing. Chapman and Hall, London, United Kingdom. 13. Moll, M. 1994. Beers and Coolers. Intercept, Andover, United Kingdom. 14. Palmer, G. H. 1989. Cereals in malting and brewing, p. 61–242. In G. H. Palmer (ed.), Cereal Science and Technology. Aberdeen University Press, Aberdeen, United Kingdom. 15. Priest, F. G. 2003. Gram-positive brewery bacteria, p. 181–217. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY. 16. Schofield, M. A., S. M. Rowe, J. R. M. Hammond, S. W. Molzahn, and D. E. Quain. 1995. Differentiation of yeast strains by DNA fingerprinting. J. Inst. Brewing 101:75–78. 17. Singh, M., and J. Fisher. 2003. Cleaning and disinfection in the brewing industry, p. 337–366. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY. 18. Slaughter, J. C. 2003. Biochemistry and physiology of yeast growth, p. 19–66. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY. 19. Stratford, M. 1992. Yeast flocculation: a new perspective. Adv. Microb. Physiol. 33:1–71. 20. Van Vuuren, H. J. J., and F. G. Priest. 2003. Gram-negative brewery bacteria, p. 219–245. In F. G. Priest and I. Campbell (ed.), Brewing Microbiology, 3rd ed. Kluwer, New York, NY. 21. Verzele, M., and D. C. de Keukeleire. 1991. Developments in Food Science, vol. 27. Chemistry and Analyses of Hop and Beer Bitter Acids. Elsevier, Amsterdam, The Netherlands. 22. Walker, G. M. 1998. Yeast Physiology and Biotechnology. Wiley, Chichester, United Kingdom.

Manila Typesetting Company

11/08/2012 07:18AM

SMP_Food Microbiology_CH36.indd

914

Manila Typesetting Company

11/08/2012 07:18AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch37

Mickey E. Parish Graham H. Fleet

Wine

Winemaking is a bioprocess that has its origins in antiquity. Scientific understanding of the process commenced with the studies of Louis Pasteur, who demonstrated that wines were the product of alcoholic fermentation of grape juice by yeasts (3). Since then, winemaking has developed into a modern, multinational industry with a strong research and development base in the disciplines of viticulture and enology. Viticulture concerns the study of grapes and grape cultivation, and enology covers postharvest processing of the grapes, from crushing through fermentation to packaging and retailing of the wine. The sensory and health sciences are also relevant to the modern wine industry. The appeal of wine is intimately linked to human perceptions of aroma, flavor, and color and the belief that its consumption, in moderation, is beneficial to health (26, 232). While there is evidence that polyphenol antioxidant constituents of wine have in vitro activity related to cardiac health, atherosclerosis, and cancer (69), a review of pertinent research suggests that in vivo health benefits from wine consumption are derived primarily from the ethanol content, while the antioxidant components likely play a smaller role due to rapid conjugation and excretion from the body (129). Further, there are reports that

37 some wines may contain small quantities of ochratoxin A, fumonisins, ethyl carbamate, or biogenic amines, further complicating our understanding of wine’s health effects (45, 152, 168, 181, 203, 261, 289). Research on mitigation mechanisms to control microbial production of these compounds, as well as continued efforts to better understand the benefits of antioxidants and ethanol, may support the public perception that wine is a healthy food when consumed in moderation. Microorganisms are fundamental to the winemaking process. To understand their contribution, it is necessary to know (i) the taxonomic identities of the species and strains associated with the process; (ii) the kinetics of their growth and survival throughout the entire production chain; (iii) the biochemical, physiological, and genomic responses of these species and their effects on the physical and chemical properties of the wine; (iv) the influence of winemaking practices upon the microbial response; and (v) the linkage between microbial action, sensory quality, and consumer acceptability of the wine (109, 112). This chapter focuses on the occurrence, growth, and significance of microorganisms in winemaking. It covers wines produced only from grapes and includes table wines, sparkling wines, and fortified wines.

Mickey E. Parish, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740. Graham H. Fleet, Department of Food Science and Technology, The University of New South Wales, Sydney, New South Wales 2052, Australia.

915

Fermentations and Beneficial Microorganisms

916

THE PROCESS OF WINEMAKING Details of the process of winemaking are beyond the scope of this chapter and are described elsewhere (30, 118, 247). Figure 37.1 outlines the main steps in the production of white and red table wines. This chapter

emphasizes grape wines, although it is recognized that wines from other fruits are regionally popular.

Grapes

Numerous varieties of the classic wine grape, Vitis vinifera, are used in winemaking, with the particular va-

Figure 37.1  Outline of processes for making red and white wines. doi:10.1128/9781555818463.ch37f1

37.  Wine

917

Table 37.1  Main components of grape juicea Substance(s) Glucose Fructose Pentose sugars Pectin Tartaric acid Malic acid Citric acid Ammonia Amino acids (total) Protein Vitamins Anthocyanins Flavonoids and nonflavonoids a

Concn 75–150 mg/ml 75–150 mg/ml 0.8–2 mg/ml 0.1–1 mg/ml 2–10 mg/ml 1–8 mg/ml 0.1–0.5 mg/ml 5–150 μg/ml 150–2,500 μg/ml 10–100 μg/ml Varies 0.5 mg/ml 0.1–1.0 mg/ml

Data obtained from references 30, 239a, and 251.

riety determining the fruity or floral characteristics of the final product. Some main varieties used in whitewine making are Riesling, Traminer, Muller-Thurgau, Chardonnay, Semillon, and Sauvignon Blanc, and those used in red-wine making include Cabernet Sauvignon, Merlot, Cabernet Franc, Pinot Noir, Shiraz, Gamay, Grenache, and Barbera. Other grape species, most notably Vitis labrusca, and interspecies hybrids crossed with Vitis vinifera to produce “French-American hybrids” also are used for wine production, mainly in eastern North America. A few of the more noteworthy hybrids and native species include Seyval Blanc, Vidal Blanc, Vignoles, Norton, Chambourcin, Concord, and Traminette. Muscadine grape wines (Muscadinia rotundifolia) are popular in the southeastern United States and are made with a sweet, fruity style. The grapes are harvested at an appropriate stage of maturity, which determines the chemical composition of the juice extracted from them. Particularly important are the concentrations of sugars and acids that are the major constituents of the juice (Table 37.1) and affect its fermentation properties. Other preharvest conditions that affect the chemical composition of the grape and its juice include climate, sunlight exposure, soil, use of fertilizers, availability of water, vine age, and use of fungicides and insecticides. Traditionally, grapes were harvested manually, but now there is increasing use of mechanical harvesters, which are often operated at night to minimize the temperature of the berries at the time of crushing (60).

Crushing and Prefermentation Treatments

For white wines, the grapes are mechanically destemmed and crushed, and the juice is drained away from the skins. If required, clarification of the juice is done by cold settling, filtration, centrifugation, or combinations

of these methods. Cold settling is generally done at 5 to 10°C for 24 to 48 h with the addition of pectolytic enzymes to help break down grape material. The juice is then transferred into the fermentation tank, where the fermentation may commence naturally or may be initiated by inoculation with selected yeasts (101). Red grapes are mechanically destemmed and crushed, and the juice plus skins (must) are directly transferred into the fermentation tank. Fermentation begins either naturally or after inoculation, and during the first few days, the skins rise to the top of the juice to form a cap. Throughout this early stage, often described as maceration, juice is regularly pumped over the cap. This step extracts purple and red anthocyanin pigments, as well as other phenolic substances, from the grape skins to give color, tannins, and astringent character to the wine. The extraction process is assisted by the production of ethanol during this preliminary fermentation. When sufficient extraction has been achieved, the partially fermented wine is drained and pressed from the skins into another tank for completion of the fermentation. Some variations of this process include thermovinification and carbonic maceration. In thermovinification, the juice plus skins are heated to 45 to 55°C with pumping over to accelerate color and tannin extraction, after which the juice is separated from the skins and transferred into the fermentation tank. In carbonic maceration, uncrushed grapes are placed in a tank that is gassed with carbon dioxide. The temperature is maintained at 25 to 35°C for several days, during which the grapes undergo endogenous metabolism that extracts color and phenolics from the skin. After 8 to 10 days, the grapes are pressed to yield a partially fermented juice (1 to 1.5% ethanol) that is transferred into a tank for subsequent fermentation (29, 30). Other pretreatments of the juice or must include the adjustment of pH and sugar concentration (where permitted), addition of diammonium phosphate or other nutrients to assist yeast growth, addition of sulfur dioxide (50 to 75 μg/ml) as an antioxidant and antimicrobial, and addition of ascorbic acid or erythorbic acid as an antioxidant.

Alcoholic Fermentation

Traditionally, fermentation of the juice was conducted in large wooden barrels or concrete tanks, but most wineries now use stainless steel tanks with facilities for temperature control (principally cooling), cleaning in place, and other modern features for process management (30, 86). However, some premium-quality white wines (e.g., Chardonnay) may be fermented in wooden (oak) barrels. White wines are generally fermented at 10

918

Fermentations and Beneficial Microorganisms

to 18°C for 7 to 14 days or more, where the lower temperature and lower fermentation rate favor the retention of desirable, volatile flavor compounds. Red wines are fermented for about 7 days at 20 to 30°C, where the higher temperature is necessary to extract color from the grape skins. The alcoholic fermentation can be conducted either as an indigenous, or wild, fermentation or as an induced, or seeded, fermentation. With indigenous fermentation, yeasts resident in the grape juice initiate and complete the fermentation. With seeded fermentation, selected strains of yeasts, generally those of Saccharomyces cerevisiae or Saccharomyces bayanus, are inoculated into the juice at initial populations of 106 to 107 cells/ml. Such yeasts are commercially available as active dry preparations and are used extensively throughout the world (74). The advantages and disadvantages of indigenous and seeded fermentations have been well discussed (3, 85, 113, 145, 230). Essentially, seeded fermentations are more rapid and predictable, while indigenous fermentations have more varied outcomes, with the potential of failures but with the prospect of wines with more interesting characters due to contributions from a greater diversity of yeast species and strains. Alcoholic fermentation is considered complete when the fermentable sugars, glucose and fructose, of the juice are completely utilized. Winemakers typically monitor the fermentation by following the reduction of sugars and concomitant increase in ethanol production so that “stuck” or “sluggish” fermentations may be detected and treated in a timely manner. Research on simple, low-cost, and rapid biosensors may ultimately offer the winemaker new tools for simultaneous monitoring of sugars and metabolites (132, 220). The wine is then drained or pumped (racked) from the sediment of yeast and grape material (lees) and transferred into stainless steel tanks or wooden barrels for malolactic fermentation, if desired, and aging. Clarification by filtration or centrifugation may be done at this stage. Leaving the wine in contact with the lees for long periods is not encouraged because the yeast cells autolyze, with the potential of adversely affecting wine flavor and providing nutrients for the subsequent growth of spoilage bacteria (110).

acid bacteria (LAB) resident in the wine are responsible for the malolactic fermentation, but many winemakers now encourage this reaction by inoculation with commercial cultures of Oenococcus oeni, formerly known as Leuconostoc oenos (83). The main reaction is decarboxylation of l-malic acid to l-lactic acid, giving a decrease in acidity of the wine and an increase in its pH by about 0.3 to 0.5 unit. Wines produced from grapes cultivated in cool climates tend to have higher concentrations of malic acid (e.g., >5 g/liter, pH 3.0 to 3.5), which can mask their varietal character. A decrease in acidity by malolactic fermentation gives a wine with a softer, mellower taste. Also, growth of malolactic bacteria in wine contributes additional metabolites that may confer complex and interesting flavor characteristics. Apart from flavor considerations, there are practical reasons for having wines complete malolactic fermentation. Wines that have not undergone malolactic fermentation before bottling risk the natural onset of this reaction at some later stage in the bottle. If this happens, the wine becomes gassy and cloudy and is spoiled. There is also a view that wines with completed malolactic fermentation have greater microbiological stability and are less prone to spoilage by other species of LAB. Fewer nutrients are available for microbial growth, and bacteriocin production by malolactic bacteria may be a further inhibitory factor. Additionally, there is evidence that a successful malolactic fermentation may inhibit growth of Brettanomyces/Dekkera (127). Malolactic fermentation is not necessarily beneficial to all wines. Wines produced from grapes grown in warmer climates have less malic acid (e.g., <2 g/liter, pH >3.5). Further reduction in acidity by malolactic fermentation is deleterious to overall sensory balance, and also, it increases the pHs of the wines to levels at which spoilage bacteria are more likely to grow. However, preventing the natural occurrence of malolactic fermentation in these wines (as might occur after bottling) is an extra technical burden. Consequently, many winemakers prefer to encourage the malolactic fermentation and later adjust wine acidity, if necessary. Nevertheless, there are winemakers who prefer not to have the malolactic fermentation occur in their wines.

Malolactic Fermentation

Most white wines are not stored for lengthy periods after fermentation. If storage is necessary, it is generally done in stainless steel tanks. Some white wines (e.g., Chardonnay) may be aged in wooden barrels. Most red wines are aged for periods of 1 to 2 years by storage in wooden (generally oak) barrels. During this time, chemical reactions that contribute to flavor develop-

It has been known since the early 1900s that wines frequently undergo another biochemical transformation that has been termed the malolactic fermentation (30, 71, 148, 176, 302). The malolactic fermentation commences naturally about 2 to 3 weeks after completion of the alcoholic fermentation and lasts 2 to 4 weeks. Lactic

Postfermentation Processes

37.  Wine ment occur between wine constituents and components extracted from the wood of the barrels (29, 30, 247). Critical points for control during storage and aging are the exclusion of oxygen and the addition of sulfur dioxide to free levels of 20 to 25 μg/ml. These controls are necessary to prevent the growth of spoilage bacteria and yeasts and to prevent unwanted oxidation reactions. Just before bottling, the wines may be stored at 5 to 10°C to precipitate excess tartarate and then clarified by application of one or more processes that include the addition of fining agents (bentonite, albumen, isinglass, gelatin), centrifugation, pad filtration, and membrane filtration. For some white wines with residual sugar, potassium sorbate at up to 100 to 200 μg/ml may be added to control yeast growth (101).

Wine Flavor

The distinctive flavors of wine originate from grape constituents and the processing operations, which include alcoholic fermentation, malolactic fermentation, and aging (54, 247, 277). The grapes contribute many volatile components (e.g., terpenes) that give wines their distinctive varietal, fruity characters. In addition, they contribute nonvolatile acids (tartaric and malic) that affect flavor and tannins (flavonoid phenols) that give bitterness and astringency. The fermentation steps, especially alcoholic fermentation, increase the chemical and flavor complexity by assisting extraction of compounds from the grapes, modifying some grape-derived substances, and producing a vast array of volatile and nonvolatile metabolic end products. Further chemical alterations occur during aging, and enzymes derived from the grapes and excreted by yeasts and malolactic bacteria, as well as those added at prefermentation, might be expected to participate in chemical-flavor transformations. Thus, the final flavor represents contributions from many compounds and cannot be attributed to any one “impact substance” (116).

YEASTS Yeasts are significant in winemaking because they carry out the alcoholic fermentation and they can cause spoilage of the wine. Their autolytic products may affect sensory quality and influence the growth of malolactic and spoilage bacteria (110). By interacting with grape anthocyanins, they can modify wine color (194). Some yeast species associated with grapes (e.g., Metschnikowia, Candida, and Cryptococcus species) have antifungal activity and could naturally control the development of spoilage and mycotoxigenic molds (110, 111). About 25 yeast species are commonly isolated from grapes and

919 wines, with Saccharomyces and Hanseniaspora species being the most prevalent. Cultural methods for their isolation, enumeration, and identification are well described elsewhere (108, 118, 179). However, there is increasing evidence that, at certain stages during the production chain, some yeasts may enter a viable but nonculturable state and require culture-independent molecular methods, such as PCR-denaturing gradient gel electrophoresis, for their detection (87, 199, 200, 227). Nucleic acid-based, molecular methods are widely used to characterize and differentiate wine yeasts and to study the microbial ecosystem in grape musts and wines (18, 87, 179, 185, 258, 310). Based on new DNA sequencephylogenetic criteria, yeast taxonomy is in a phase of transition and revision, and this process includes potential reclassification of many wine yeasts (166, 198).

Origin

Wine yeasts originate from any of three sources, namely, the surfaces of grapes, the surfaces of winery equipment (crushers, presses, fermenters, tanks, pipes, pump barrels, filtration units), and inoculum cultures (111, 115). Grapes are a primary source of yeasts that enter the winery environment. These yeasts probably come from wind-blown soil and vegetation, as well as attack from insects and birds, but the processes by which grape berries become contaminated and colonized by yeasts throughout their cultivation in the vineyard are not well understood (115). Very few yeast cells (10 to 103 CFU/g) are found on immature grape berries, but the numbers increase to 104 to 106 CFU/g as the berries mature to harvest. Unripe grapes harbor a predominance of Rhodotorula, Sporobolomyces, Cryptococcus, and Candida species, along with the yeastlike fungus Aureobasidium pullulans. Most of these species also occur on mature grapes, but at this stage, species of Hanseniaspora (Kloeckera) and Metschnikowia often predominate (57, 115, 160, 227). Damaged grapes with increased availability of fermentable substrates have increased populations of Hanseniaspora and Metschnikowia species as well as other yeasts, including Saccharomyces and Zygosaccharomyces. Saccharomyces yeasts are rarely isolated from healthy, mature grapes by plate culture methods, raising questions about their origin in natural, unseeded wine fermentations (115, 190). Nevertheless, they must be present on grapes at low populations because they are readily isolated by enrichment culture. Moreover, aseptically harvested and crushed grapes naturally ferment and give a dominance of Saccharomyces. Using such methods, followed by molecular characterization of the isolates,

920

Fermentations and Beneficial Microorganisms

many studies have demonstrated significant diversity in the strains of S. cerevisiae obtained from various vineyards and even the same vineyard over several vintages. Some studies suggest that certain strains are unique to particular geographical regions and may contribute to the regional characters, or “terroirs,” of wines, but data supporting this view are not consistent (38, 77, 233, 242, 290, 294). Explanations for these variations probably reside in the facts that the grape surface is a unique phyllospheric habitat and numerous factors are likely to affect its yeast ecology. Such factors include the surface chemistry of the grape and its ability to support microbial attachment and growth; microbial exposure to the natural stresses of temperature, sunlight, irradiation, rainfall, and desiccation; tolerance to chemical inhibitors from the grape itself and from the application of insecticides and fungicides; and interactions with other species (other yeasts, bacteria, and filamentous fungi) (111, 115, 290). The surfaces of winery equipment that come into contact with grape juice and wine are locations for development of the so-called residential or winery yeast flora. The extent of this development depends upon the nature of the surface and the effectiveness of cleaning and sanitizing operations. These surfaces are considered to be the main source of Saccharomyces in wine fermentations. Such sources harbor multiple strains of S. cerevisiae that have accumulated from the grapes and starter cultures used in previous years and carry over into subsequent vintages (52, 58, 77, 108, 141, 256). Starter cultures, if used to inoculate the juice, will be a principal source of yeasts. Presently, various strains of S. cerevisiae and S. bayanus are used, but the future could see the development and use of other species such as those in the genera Hanseniaspora, Candida, and Pichia (36, 49, 50, 145, 270). Commercial yeast preparations used for inoculation are not necessarily pure and may contain a proportion of species other than S. cerevisiae (74). Fermentation studies on microbial interactions among Saccharomyces species, non-Saccharomyces yeasts, and bacteria suggest that such interactions are complex and have significant impact on sensory qualities of the resulting wine.

species throughout the entire course of fermentation. Recently, molecular techniques have made it possible to follow the development of particular strains throughout fermentation. Virtually all ecological studies show that S. cerevisiae and, to a lesser extent, S. bayanus are the principal wine yeasts and predominate during the middle to final stages of fermentation. Nevertheless, there is an important biodiversity of other yeasts that contribute to most fermentations and that can predominate under some conditions (85, 113, 165). Figure 37.2 gives a general representation of the growth of yeasts during the fermentation of grape juice, whether it is conducted through indigenous or inoculated processes. Freshly extracted grape juice harbors a yeast population of 103 to 105 CFU/ml, comprising mostly Hanseniaspora (Kloeckera) species, but species of Candida, Metschnikowia, Pichia, Hansenula, Kluyveromyces, and Rhodotorula also occur. These species are often referred to as the non-Saccharomyces yeasts, or wild flora. The juice will also contain low populations of indigenous Saccharomyces species, depending on the extent of their occurrence on grapes and contamination from equipment used to process the juice. Fermentation is initiated by growth of various species of the nonSaccharomyces yeasts (e.g., Hanseniaspora uvarum, Kloeckera apiculata, Candida stellata, Candida colliculosa, and Metschnikowia pulcherrima) as well as Saccharomyces species. The growth of the nonSaccharomyces species is generally limited to the first 2 to 4 days of fermentation, after which they die off. Nevertheless, they achieve maximum populations of

Growth during Fermentation

Figure 37.2  Generalized growth of yeast species during alcoholic fermentation of wine.  , S. cerevisiae; ·, Kloeckera and Hanseniaspora species; n, Candida species. Variations will occur in the initial and maximum populations for each species; for fermentations inoculated with S. cerevisiae, the initial po­pu­ lation is approximately 106 CFU/ml (113). doi:10.1128/9781555818463.ch37f2

Many studies have described the yeast populations that grow during alcoholic fermentation. Early research gave qualitative descriptions of the main species isolated from different stages (early, middle, and final) of fermentation. Subsequent studies monitored the growth of individual

37.  Wine 106 to 107 CFU/ml before death, and such growth is metabolically significant in terms of substrates utilized (hence not available to S. cerevisiae) and end products released into the wine (47, 206, 250). Also, the dead cells of these yeasts become part of the total yeast pool for subsequent autolysis. Their death is attributed to an inability to tolerate the increasing concentrations of ethanol, which is produced largely by S. cerevisiae, but other factors may be involved. After 4 days or so, the fermentation is completed by S. cerevisiae, which reaches final populations of about 108 CFU/ml. In some cases, S. bayanus, Saccharomyces paradoxus, and Saccharomyces uvarum may be the dominant species, especially for fermentations at lower temperatures (77, 242, 267). This basic ecological profile has been found in wineries throughout the world by using both culture (56, 58, 114, 121, 134, 225, 228, 259) and, now, cultureindependent (51, 200) analytical methods and is readily demonstrated in experimental fermentations (143, 226). It is determined by the relative abilities of the different species to survive, grow, and interact under the stresses initially imposed by the chemical and physical properties of the juice (e.g., low pH and high sugar concentration) and the stresses of the changing environment as fermentation progresses (e.g., increasing ethanol level and anaerobiosis) (15, 42, 143, 197). Cell-cell interactions and quorum-sensing molecules may also be involved (210, 269). In addition to the successional growth of different species throughout fermentation, further underlying ecological complexity is demonstrated by the successional development of strains within a species. This revelation became evident with the use of molecular techniques that have enabled strain differentiation and recognition. As many as 100 genetically distinct strains of S. cerevisiae have been found in some fermentations (52, 84, 121, 228, 254, 256, 268), and it has been demonstrated that strain-strain interactions, determined by metabolic footprinting, influence wine flavor profiles (154). Moreover, strain diversity throughout fermentation has been reported for non-Saccharomyces species (225, 228, 259).

Factors Affecting Yeast Growth during Fermentation

Many intrinsic and extrinsic variables determine the rate and extent of the growth of yeasts during fermentation (Table 37.2) (35, 79, 85, 113, 165). Yeast growth is best measured by plate counts, but carbon dioxide production, as measured by the loss of culture weight, and utilization of juice sugars are also used to monitor fermentation kinetics (107, 108, 118, 226).

921 Table 37.2  Factors affecting the growth of yeasts during

alcoholic fermentation

________________________________________ Grape juice characteristics Sugar concentration Level of assimilable nitrogen pH Amt of fungicide and pesticide residues Content of dissolved oxygen Accumulation of toxic metabolites Processing factors Addition of sulfur dioxide Extent of settling and clarification of juice Addition of yeast nutrients Inoculation with selected yeasts Temperature control Pumping over Biological factors Influences of grape fungi and bacteria Population and composition of indigenous yeasts Presence of killer yeasts

________________________________________ Juice Composition

In most circumstances, grape juices provide all the nutrients and conditions necessary for vigorous and complete fermentation. However, chemical and physical properties of the juice vary according to grape variety, climatic influences, viticulture practices, and maturity at harvest. Relevant properties include the sugar concentration, the amount of nitrogenous substances, concentrations of vitamins, the dissolved oxygen content, the amount of soluble solids, the presence of fungicide and pesticide residues, pH, and the presence of any yeast-inhibitory or -stimulatory substances produced by the growth of molds and bacteria on the grapes (79, 113). The concentrations of fermentable hexoses in grape juice vary between 150 and 300 mg/ml (Table 37.1) and may be as high as 400 mg/ml in juice prepared from grapes infected with Botrytis cinerea (“pourriture noble” or “noble rot”) that are used for production of certain sweet wines (88). The initial sugar concentration will affect growth rates of the different species and strains of yeasts and the extent to which they contribute to the overall fermentation. The growth of C. stellata, Candida zemplinina, and Torulaspora delbrueckii may be favored in juices with higher sugar concentrations (42, 182). Free amino acids and ammonium ions (Table 37.1) are the principal nitrogen sources used by yeasts during fermentation. Most juices contain sufficient nitrogen substrates (>150 mg of assimilable nitrogen/liter) to allow rapid and complete fermentation, but heavily

922 processed or clarified juices may be nitrogen deficient (20, 150). Nitrogen availability in vineyard soils and the use of nitrogen fertilizers can affect the concentration of assimilable nitrogen in the juice and subsequent yeast growth (20, 170). The nitrogen demand by yeasts increases significantly with increasing sugar concentration in the juice and varies with the strain of S. cerevisiae (279). Consequently, supplementation of juices with various yeast foods or diammonium phosphate is a common practice to ensure that nitrogen availability is not a factor that limits yeast growth. Most studies on the nitrogen requirements of wine yeasts have been conducted with S. cerevisiae, and little is known about the nitrogen demands of non-Saccharomyces species or their ability to remove specific nitrogen substrates from the juice before the growth of S. cerevisiae (20, 135). The ability of wine yeasts to utilize grape juice proteins as a source of nitrogen requires further consideration. Strains of S. cerevisiae generally do not produce extracellular proteolytic enzymes (107), but some non-Saccharomyces wine yeasts are proteolytic (e.g., K. apiculata and M. pulcherrima) (36, 43, 107, 121, 274). Grape juices generally contain enough vitamins (inositol, thiamine, biotin, pantothenic acid, and nicotinic acid) to permit maximum growth of S. cerevisiae. Vitamin losses may occur in heavily processed juices, where supplementation can improve yeast growth. Species of non-Saccharomyces yeasts are more demanding of vitamins than S. cerevisiae, and vitamin availability may be a factor that limits their contribution to fermentation. The pH of grape juice varies between 2.8 and 4.0, depending on the concentrations of tartaric and malic acids. Although growth and fermentation rates for S. cerevisiae are decreased as the pH is decreased from 3.5 to 3.0 (42, 147), it is not fully understood how juice pH affects the relative growth rates of the nonSaccharomyces yeasts and their potential to influence alcoholic fermentation. Treatment of grapes with fungicides and pesticides before harvest can give juices that contain residues of these substances. Depending on their concentration and chemical nature, these residues may decrease yeast growth and even change the ecology, thereby leading to slow or incomplete fermentations (25, 31, 113). Conditions that stimulate yeast growth and fermentation include aeration of the juice before or during the early stages of fermentation and the presence of grape solids and particulate materials (31, 35, 79). Different yeast species may selectively adsorb to such particles to form a biofilm of immobilized biomass.

Fermentations and Beneficial Microorganisms

Clarification of Grape Juice

The procedures used to clarify juices, especially for white wine fermentations, will influence the populations of indigenous yeasts in the juice and their potential contribution to the fermentation. Centrifugation and filtration remove yeast cells, thereby decreasing or eliminating the contribution of indigenous species to the fermentation. In contrast, clarification by cold settling presents opportunities for the growth of indigenous yeasts, especially those species or strains that grow well at low temperatures (e.g., K. apiculata) (31, 42, 204, 244).

Sulfur Dioxide

The addition of sulfur dioxide to grapes or juice for controlling oxidation reactions and restricting the growth of indigenous microflora is a well-established practice (251). Nevertheless, strong growth of nonSaccharomyces species occurs during the early stages of most commercial fermentations where the usual amounts of sulfur dioxide (50 to 100 μg/ml) have been added (113). These findings question the efficacy of sulfur dioxide in controlling indigenous yeasts and challenge one of the reasons for using sulfur dioxide in winemaking. Good comparative data on the responses of wine yeasts to this agent under the conditions of winemaking are lacking (96). Although alternatives to the use of sulfur dioxide in wines and must are the subject of research, to date much of the wine industry remains reliant on SO2 (61, 93, 117, 157, 255).

Temperature of Fermentation

Temperature control is an important practice in modern winemaking. It affects the growth rates and metabolic activities of yeasts (42), their ability to tolerate ethanol (122), and their contribution to the fermentation. Fastest yeast growth and fermentation occur at 25 to 30°C, and the ecology of the fermentation follows the pattern outlined in Fig. 37.2. However, when the temperature is decreased below 20°C, there is an increased contribution of the non-Saccharomyces species to the fermentation. Species such as K. apiculata and C. stellata exhibit increased tolerance to ethanol and do not die off as shown in Fig. 37.2. They can produce maximum populations of 107 to 108 CFU/ml, which remain viable until the end of fermentation (98, 147). Moreover, they may have higher growth rates than S. cerevisiae at low temperatures (42). The impact of such ecological shifts on the chemical compositions and sensory qualities of wines has yet to be determined. S. bayanus is more cryophilic than S. cerevisiae and is more likely to be found in fermentations conducted at lower temperature (192, 267).

37.  Wine

Inoculation of the Juice

Perhaps the most significant technological innovation in winemaking during the last 50 years has been the seeding (inoculation) of the juice with selected strains of S. cerevisiae or S. bayanus. These strains have been selected according to criteria that enhance the efficiency of the process and product quality (Table 37.3) (74, 145, 149). Seeding of the fermentation is undertaken with the assumption that the inoculated strain will outcompete and dominate over indigenous strains of S. cerevisiae and non-Saccharomyces yeasts. Although much evidence shows that inoculated strains dominate at the end of fermentation, the view that early growth of the indigenous species is suppressed or insignificant cannot be supported. Growth of the indigenous nonSaccharomyces species, according to Fig. 37.2, still occurs (146), and moreover, indigenous strains of S. cerevisiae may grow despite massive competition from the seeded strain. Indeed, if conditions in the juice do not favor the growth of the seeded strain, indigenous S. cerevisiae may dominate the fermentation, and this domination can be verified by molecular techniques that

Table 37.3  Some properties used to select yeasts for

application in wine fermentationsa

________________________________________ Desirable properties Complete and rapid fermentation of sugars High tolerance of alcohol Resistance to sulfur dioxide Fermentation at low temperatures Production of good flavor and aroma profiles Production of glycerol Production of b-glycosidases Malic acid degradation Killer phenomenon Good sedimentation properties Tolerance of pesticides and fungicides Fermentation under pressure Suitability for mass culture, freeze-drying distribution, and rehydration Undesirable properties Production of sulfur dioxide Production of hydrogen sulfide Production of volatile acidity High level of formation of acetaldehyde, pyruvate, and esters Foaming properties Formation of ethyl carbamate precursors Production of polyphenol oxidase (affects wine color) Inhibition of malolactic fermentation Production of mousy and other taints

________________________________________ a

From references 74 and 149.

923 allow the differentiation of yeast strains. Although there is a high probability that inoculated S. cerevisiae will dominate the fermentation, seeding does not guarantee the dominance of any particular strain or its exclusive contribution to the fermentation (58, 96, 121, 141, 236, 256, 259). Significant factors that affect this outcome are the population of indigenous yeasts already in the juice and the extent to which they have adapted to grow in that juice (226).

Interactions with Other Microorganisms

Various species of molds, acetic acid bacteria (AAB), and LAB naturally occur on grapes and on winery equipment. Conditions that allow their proliferation on the grape or in the juice have the potential to affect yeast growth during alcoholic fermentation and are discussed in later sections. Killer yeasts are certain strains that produce extracellular proteins or glycoproteins, termed killer toxins (zymocins), that destroy other yeasts (264). Usually, strains of one species kill only strains within that species, but killer interactions between different species also occur. Killer toxin-producing strains of S. cerevisiae and killersensitive strains of S. cerevisiae occur as part of the natural flora of wine fermentations. In some wineries, killer strains of S. cerevisiae predominate at the end of fermentation, suggesting that they have asserted their killer property and taken over the fermentation. Killer strains of wine isolates of Candida, Pichia, Hanseniaspora, and Hansenula occur and can assert their killer action against wine strains of S. cerevisiae (140, 305, 307). Expression of the killer phenomenon during wine fermentations is affected by many factors, which include the ethanol concentration, pH, temperature, amount of assimilable nitrogen, presence of fining agents, and relative populations of killer and killer-sensitive strains (113). There are several implications of killer yeasts in winemaking. First, inoculated strains of S. cerevisiae may be destroyed by indigenous killer strains of S. cerevisiae or non-Saccharomyces species, leading to sluggish or stuck fermentation, slower fermentation, or completion of the fermentation by a less desirable species. Second, there may be advantage in conducting the fermentation with selected or genetically engineered killer strains of S. cerevisiae for the purposes of controlling the growth of less desired indigenous species (217). Moreover, strains could be selected or constructed to produce stable, broad-spectrum killer toxins that would protect the wine from infection by spoilage yeasts. Finally, strains might be selected to have immunity against the killing action of indigenous yeasts, thereby giving them a greater chance of dominating the fermentation (264).

924 In recent years, investigations on the interactions of Saccharomyces and non-Saccharomyces microflora during fermentations have led to a reevaluation of the role of non-Saccharomyces yeasts on the sensory quality of the final product (48, 295). It is becoming more apparent that certain non-Saccharomyces yeasts increase wine complexity and enhance wine characteristics (47), although under certain conditions mixed fermentations may show sluggish fermentation behavior (46). Metabolic profiling has also shown that multiple strains of Saccharomyces species within fermentations can interact to affect the sensory profile of a wine (6, 154). Various combinations of yeasts coinoculated into Sauvignon Blanc musts produced wines with large differences in volatile thiols and other flavor compounds and, ultimately, large differences in consumer acceptability (162). While there have been major advances in our understanding of fermentation ecology and biochemistry and of the molecular biology of organisms involved in the fermentation process, further knowledge and understanding may ultimately lead to the routine use of mixed starter cultures with Saccharomyces and non-Saccharomyces components.

Stuck or Sluggish Fermentations

A sporadic but serious problem is the premature cessation of yeast growth and alcoholic fermentation, giving wine with residual, unfermented sugar (>2 to 4 g/liter) and a lower-than-expected concentration of ethanol. Such fermentations are referred to as being stuck or sluggish if they take longer than normal to give low residual sugar (25, 150). Factors considered to cause this problem include excessive clarification and processing of the juice; fermentation temperatures that are too high; juice deficiency in nutrients or growth factors; the presence of fungicide residues; influences from other microorganisms such as molds, AAB, and killer yeasts; ethanol toxicity; and accumulation of medium chainlength fatty acids such as octanoic and decanoic acids that can become toxic to yeast growth. Another consideration is failure in the transport of grape juice sugars into the yeast cell and the multitude of factors that affect expression of the genes responsible for this activity (24, 25). Interventions to overcome stuck fermentations include the addition of nitrogen-containing yeast foods, controlled aeration of the juice or wine, and the addition of yeast cell wall hulls or other bioadsorbents to remove toxic substances (25).

Biochemistry, Physiology, and Genomics

Yeasts utilize grape juice constituents as substrates for their growth, thereby generating metabolic end prod-

Fermentations and Beneficial Microorganisms ucts that are excreted into the wine (30, 85, 149). The main products are carbon dioxide and ethanol and, to a lesser extent, glycerol and succinic acid. In addition, many hundreds of volatile and nonvolatile secondary metabolites are produced in small amounts that, collectively, contribute to the sensory quality of the wine. These substances include a vast range of organic acids, higher alcohols, esters, aldehydes, ketones, sulfur compounds, and amines. The chemical identities of individual substances and their flavor or aroma sensations, sensory thresholds, and concentrations in wines are well documented (54, 116), and the biochemistry of their formation in S. cerevisiae, at least, is well-known (21, 167, 277). However, further studies are needed to determine the metabolic characteristics of the non-Saccharomyces yeasts. The production of these metabolites varies considerably depending on the yeast strain, yeast species, and conditions of fermentation (145, 206, 250). Table 37.4 shows some of the main metabolites produced by yeasts associated with wine fermentations. During wine fermentation, S. cerevisiae responds to the changing environment by sequential expression and regulation of many genes associated with carbohydrate, nitrogen, and sulfur metabolism and genes required to tolerate the stresses of high sugar concentration, low pH, the presence of ethanol, and nutrient deficiency (8, 186, 253, 292, 311). A natural process of yeast strain adaptation and selection probably occurs in wine ecosystems (237). Table 37.4  Some principal compounds produced by yeasts

during alcoholic fermentation of winea Concn in wine (mg/liter)b

Sensory description

Ethanol Propanol Isobutanol Isoamyl alcohol 2-Phenylethyl alcohol Ethyl acetate

100–150 g/liter 9–68 9–174 6–490 4–197

Burning Pungent, harsh Solvent, bitter Malt, burnt, nail polish Floral, rose

23–64

Isoamyl acetate 2-Phenylethyl acetate Acetic acid Succinic acid Acetaldehyde

0.1–3.4 0–18.5

Pineapple, varnish, nail polish Banana, pear Rose, fruity, honey

Compound

Diacetyl Glycerol

100–115 0.5–1.7 10–75 0.1–5 5–15 g/liter

Vinegary, sour Salty, bitter Nutty, pungent, green apple Buttery Slightly sweet

Adapted from references 116, 145, 167, and 277. Unless otherwise noted.

a

b

37.  Wine The first stage of the alcoholic fermentation involves gene expression under conditions of exponential growth in an environment of low pH and relatively high sugar concentration. The second stage involves significant metabolic activity of stationary-phase cells under the added stress of the presence of ethanol and possible nutrient limitation. Many of the secondary metabolites that affect wine flavor are produced during this second stage and reflect the yeast response to stress conditions (15, 149).

Carbohydrates

Glucose and fructose in juice are metabolized by the glycolytic pathway into pyruvate, which is decarboxylated by pyruvate decarboxylase into acetaldehyde. Acetaldehyde is reduced to ethanol by alcohol dehydrogenase. Although most of the pyruvate is converted to ethanol and carbon dioxide, small proportions are converted to secondary metabolites (15, 24, 149, 277). Glycerol, which imparts desirable smoothness and viscosity to the wine, is produced during glycolysis. Its production is increased by the presence of sulfur dioxide, higher incubation temperature, and increased sugar concentration, but there are significant strain and species influences. Transport of sugars into the cell involves several transporter genes, and factors that affect their activity are important rate-limiting steps in the fermentation process (167, 292). Wine yeasts vary in their ability to take up glucose and fructose, and this variation can affect the residual sweetness of the wine (22). The potential for wine yeasts to degrade grape pectins and enhance juice extraction needs more consideration, given that some strains of S. cerevisiae and nonSaccharomyces may produce pectolytic enzymes (43, 107, 239, 274).

Nitrogen Compounds

Wine yeasts can utilize ammonium ions and amino acids in the juice as sources of nitrogen (20, 151). Genes for the transport and assimilation of amino acids are not expressed until after the ammonium ions are utilized (186, 253). S. cerevisiae does not have extracellular proteolytic activity, but some wine species of Hanseniaspora, Kloeckera, Candida, and Metschnikowia produce extracellular proteases that may break down juice proteins (36, 43, 107, 274). The metabolism of nitrogen substrates by yeasts has important implications in winemaking (20, 25, 150, 167, 186, 205, 286). Juices that are limiting in nitrogen content can give stuck or sluggish fermentations and wines with unacceptably high contents of hydrogen sulfide. Metabolism of arginine (the predominant

925 amino acid in grape juices) by S. cerevisiae can lead to the production of urea, which is able to react with ethanol to form ethyl carbamate, a suspected carcinogen. The amount of urea produced depends on many factors, including the concentration of arginine relative to those of other nitrogen substances in the juice and the strain of S. cerevisiae. Strains of S. cerevisiae have been engineered to catabolize urea more efficiently, thereby reducing ethyl carbamate production in wine (64, 68). Metabolism of amino acids by decarboxylation, transamination, reduction, and deamination reactions produces higher alcohols, fatty acids, esters, and carbonyl compounds that affect wine flavor (20, 196, 277, 299). Supplementation of Chardonnay with a high level of ammonium nitrogen alone produces acetic and solvent-type sensory attributes, whereas a combination of ammonium nitrogen with amino acids at the same nitrogen content produces tropical and banana-type descriptors (286). This suggests that nitrogen supplementation using ammonium ions only should be conducted carefully and with knowledge of the amino acid content of the must. Yeasts also release amino acids into the wine. This occurs during the final stages of alcoholic fermentation by mechanisms not fully understood and later when the cells have died and there is autolytic degradation of yeast proteins (44, 151). These amino acids can serve as nutrients for the growth of malolactic bacteria or spoilage bacteria (1, 110). Catabolism of tryptophan resulting in formation of indole and indole-derived products has been linked to a “plastic-like” taint in wines, especially those produced in sluggish fermentations (5, 37).

Sulfur Compounds

Yeasts produce a range of volatile sulfur compounds that, depending on threshold concentrations, have a positive or negative impact on wine flavor (167, 240, 277). The predominant compounds are sulfur dioxide (sulfite), hydrogen sulfide, and dimethyl sulfide, with lesser amounts of other organic sulfites, mercaptans, and thioesters. The production of sulfite and hydrogen sulfide in S. cerevisiae is linked to the biosynthesis of cysteine and methionine by the sulfate reduction pathway (8, 186, 272). The formation of sulfite by S. cerevisiae depends on the strain. Most strains produce less than 10 μg/ml, but some give levels up to 100 μg/ml (251). High-sulfite-producing strains are avoided in winemaking because of the negative effect of sulfite on wine quality and the potential of sulfite to cause allergic reactions in some consumers and to inhibit malolactic bacteria. At concentrations exceeding 50 μg/liter, hydrogen sulfide gives an unpleasant “rotten egg” aroma

926 to wine. Many chemical and biological factors affect the production of hydrogen sulfide during wine fermentation, the most significant of which is the strain of S. cerevisiae. Some strains produce hydrogen sulfide at concentrations exceeding 1 μg/ml. Hydrogen sulfide production is genetically based, but it is also influenced by factors such as the composition of the grape juice and fermentation conditions (272). Elemental sulfur used as a fungicide on grapes prior to harvest, metabisulfite added to the grapes and juice at crushing, and sulfate that occurs naturally in the juice are all significant precursors of hydrogen sulfide in the wine. Deficiencies in assimilable nitrogen and vitamins in the juice can cause hydrogen sulfide production by S. cerevisiae (20, 150, 299). Under these conditions, the intracellular pool of cysteine and methionine is low, allowing the sulfate reduction pathway to operate with consequent production of hydrogen sulfide. If the juice contains an adequate supply of assimilable nitrogen (e.g., ammonium ions, amino acids), cysteine and methionine are produced at concentrations that, through feedback inhibition, decrease the activity of the sulfate reduction pathway and the production of hydrogen sulfide. There is also a view that under nitrogen-limiting conditions, S. cerevisiae degrades intracellular proteins to provide essential amino acids, including cysteine and methionine, from which hydrogen sulfide is formed (150). The potential for the non-Saccharomyces yeasts to produce sulfur compounds during wine fermentations requires study; some species produce hydrogen sulfide at levels similar to those produced by S. cerevisiae (195). The volatile thiols 4-mercapto-4 methylpentan-2one, 3-mercaptohexan-1-ol, and 3-mercapto-hexyl acetate have recently been found in some wines. They have very low perception thresholds (3 to 50 ng/liter) and, at low levels, impart most desirable passion fruit, guava, grapefruit, or box tree aromas to some wines. These compounds occur in grape juice as nonvolatile conjugates to cysteine and are deconjugated into the volatile form by S. cerevisiae and S. bayanas during alcoholic fermentation (255, 256, 284).

Organic Acids

Of the numerous organic acids produced in wine by yeasts, succinic and acetic acids are the most significant (238, 277). Succinic acid has a bitter, salty taste and is produced by S. cerevisiae at concentrations up to 2.0 mg/ml, depending on strain. Lower concentrations are produced by non-Saccharomyces species. The production of succinic acid is not associated with any major defects in wine quality. In contrast, acetic acid becomes detrimental to wine flavor at concentrations exceeding

Fermentations and Beneficial Microorganisms 1.5 mg/ml and may lead to stuck fermentations (25). Most strains of S. cerevisiae produce only small amounts (<0.75 mg/ml) of this acid, but some can produce more than 1.0 mg/ml and are unsuitable for winemaking (149). Factors that limit yeast growth such as low temperature, high sugar concentrations, low pH, deficiency in available nitrogen, and excessive clarification cause increased acetic acid production by S. cerevisiae (149, 277). Candida, Kloeckera, and Hanseniaspora species may produce larger amounts of acetic acid than S. cerevisiae, but there is substantial strain variation in this property (206, 250, 263). The production of lactic acid by wine yeasts is considered insignificant (<0.1 mg/ml), but some species of Saccharomyces, Kluyveromyces, and Candida can produce this acid in amounts of 5 to 10 mg/ml (173, 208). Such strains could be used to increase the acidity of some wines. Although tartaric acid is prevalent in grape juice and wine, it is not metabolized by wine yeasts (238). However, malic acid is partially (5 to 50%) metabolized by S. cerevisiae and other wine yeasts. It is completely degraded by some species of Schizosaccharomyces and some strains of Zygosaccharomyces bailii (124, 238), by which it is oxidatively decarboxylated into pyruvate, which is then converted to ethanol. The possibility of using species of Schizosaccharomyces to deacidify wines in place of the malolactic fermentation has attracted considerable interest, but such use must be carefully controlled as these yeasts can produce off-flavors (124, 265). Yeasts produce small amounts (1 to 15 μg/ml) of free fatty acids in wines (238). Of special note are hexanoic, octanoic, and decanoic acids, which, on accumulation, may become toxic to S. cerevisiae and O. oeni and contribute to stuck fermentations (1, 25, 110).

Monoterpenes

Monoterpenes are important aroma compounds that occur in grapes and confer a variety of desirable, fruity characters (e.g., grape, raspberry, and passion fruit). Most of these monoterpenes occur as nonvolatile precursors that are glycosidically conjugated to glucose or disaccharides. Their potent volatility and fragrance are released on hydrolysis of this linkage. Glycosidase activity produced by yeasts during alcoholic fermentation catalyzes this transformation and can significantly affect wine flavor. The extent of this activity depends on the yeast species and strain, with the non-Saccharomyces yeasts having stronger glycosidase activity than S. cerevisiae (36, 43, 104, 106, 249). Moreover, it is possible that wine yeasts may also synthesize monoterpenes de novo (40).

37.  Wine

Autolysis

The autolytic degradation of yeast cells at the end of alcoholic fermentation and during cellar storage of the wines is often underestimated as a significant biochemical event. During autolysis, yeast proteins, nucleic acids, and lipids are extensively degraded, releasing peptides, amino acids, nucleotides, bases, and free fatty acids into the wine. These products affect wine flavor and serve as nutrients for the growth of bacteria (1, 110). In addition to S. cerevisiae, the non-Saccharomyces species would be involved in autolytic reactions (44, 151, 235, 309).

Yeasts and Flavor Diversity

Individual yeasts produce an array of several hundred flavor metabolites (organic acids, higher alchols, esters, aldehydes, ketones, amines, sulfur compounds, etc.) that, collectively, contribute to the individuality of wine character. This array varies with the yeast species, strains within a species, and conditions of fermentation (99, 145, 167, 206, 223, 250, 277). Moreover, the metabolite profile may vary, depending on whether the yeasts grow in single or mixed cultures (154). Wines produced by indigenous, nonseeded fermentations are often perceived as having more diverse and interesting sensory characters than those produced by inoculation with strains of S. cerevisiae (113, 230). These differences may be attributed to the greater ecological and metabolic spectrum of yeasts associated with nonseeded fermentations (167). Also, the regional character of some wines has been ascribed, in part, to influences of the local yeast flora. Yeast-dependent production of volatile thiols and other flavor compounds during fermentation has been shown to strongly influence the sensory character of Sauvignon Blanc wines (91, 155, 156, 162, 278). However, such relationships are not always consistent or predictable and reflect the challenges that remain in linking yeast ecology and wine chemistry with sensory perception (54, 116).

Spoilage Yeasts

Yeasts can spoil wine at several stages during the process: during alcoholic fermentation and bulk storage in the winery and after packaging (30, 85, 93, 179, 273, 281). Growth of inappropriate species or strains of yeasts during alcoholic fermentation can give an inferior wine (e.g., high in content of esters, acetic acid, and hydrogen sulfide), and the product is spoiled. Wine that is exposed to air (e.g., as that in incompletely filled barrels or tanks) quickly develops a film or surface flora of weakly fermentative or oxidative yeasts of the genera Candida, Pichia, and Hansenula. Particularly significant is Pichia membranifaciens. These species oxidize

927 ethanol, glycerol, and acids, giving wines with unacceptably high levels of acetaldehyde, esters, and acetic acid. Fermentative species that grow in wines during cellar storage and in packaged wines include Z. bailii, Brettanomyces/Dekkera species, and Saccharomycodes ludwigii. In addition to causing excessive carbonation, sediments, and haze, these species produce acid and estery off-flavors. Packaged wines that contain residual sugars are prone to refermentation, especially by S. cerevisiae, causing swelling and explosion of the container (281). Good quality control and assurance practices, including end product specifications (e.g., <1 to 2 viable yeast cells in 200 ml), are essential to prevent spoilage problems (179, 281). Real-time PCR methods are available to quickly detect wine spoilage yeasts (53, 191, 218, 219). Brettanomyces and Dekkera species also cause unpleasant mousy or medicinal taints due to the formation of tetrahydropyridines and volatile phenolic substances such as 4-ethyl guaiacol and 4-ethyl phenol (93, 273, 276). Extensive research has been conducted on the genetic characterization and detection of these organisms, along with control methods, due to their robust survival and growth characteristics in wine and on equipment, their ubiquitous nature in wine environments, and the potent off-flavors they produce (59, 66, 144, 159, 211, 212, 245, 246, 276, 280, 300). Treatment of wine with a low electric current was shown to inactivate these yeasts (180).

Genetic Improvement of Wine Yeasts

A vast biodiversity of wine yeasts naturally occurs in vineyards and wineries, and selection from these reservoirs has generally met the needs of winemaking. Nevertheless, the process of strain selection and development can be accelerated and more specifically targeted through the use of classical and modern genetic improvement technologies (149, 230, 232). The sequence of the S. cerevisiae genome has been determined, leading to continued advances in the elucidation of gene function and genomic interactions (231). Genomic research is providing opportunities to more fully understand cellular mechanisms related to issues such as ethanol tolerance, hexose transport, and strain discrimination (161, 164, 293). These advances have made it possible to genetically engineer desirable characteristics into wine strains of S. cerevisiae (41, 47, 64, 68), and many targets have been identified and investigated to improve wine-processing efficiency, wine quality, and wine appeal (Table 37.5). There is little doubt that successful, technologically advanced strains can be engineered, but their commercialization will depend on safety and

928 Table 37.5  Directions for the genetic improvement of wine

yeasts ________________________________________________________ Characteristics for improved wine flavor and quality More balanced production of specific flavor volatiles (esters, higher alcohols) Increased glycerol production Increased release of grape terpenoids and volatile thiols Decreased production of acetic acid, hydrogen sulfide, and phenolic off-flavors Decreased production of biogenic amines and ethyl carbamate Decreased ethanol production for low-alcohol wines Characteristics for improved process efficiency More efficient utilization of grape juice sugars and amino acids Increased ethanol tolerance Malolactic activity Antimicrobial action (killer toxins and bacteriocins against spoilage yeasts and bacteria) Enzymes (protease, pectinase) to assist clarification Enhanced sedimentation properties ________________________________________________________

environmental approval by government authorities and consumer acceptance of foods and beverages modified by modern biotechnologies. A recent risk assessment on the impact of a genetically engineered strain of S. cerevisiae on the environment concluded that the transgenic strain did not have a selective advantage over the untransformed parental strain and would not predominate if released into the environment (257).

Fermentations and Beneficial Microorganisms tems and is intimately associated with the malolactic fermentation. Consequently, the taxonomy, biochemistry, physiology, and genomics of this species have been extensively studied (11, 176, 201, 291). It is also recognized that other LAB affect the malolactic fermentation (283). Wine LAB are isolated and enumerated by culture on plates of MRS agar supplemented with 10 to 15% (vol/vol) tomato juice or apple juice (108, 118). They are identified by an extensive array of biochemical and physiological tests (39, 83, 92). However, molecular methods based on rRNA gene sequencing, multilocus sequence typing, specific PCR primers, restriction fragment length polymorphism, random amplified polymorphic DNA, and nucleic acid probes are now routinely used for species identification and strain differentiation (27, 75, 137, 222, 308).

Ecology

The LAB of wines originate from the grapes and winery equipment, but inoculation of selected species to conduct the malolactic fermentation is widely practiced. Freshly extracted grape juice contains LAB at populations of 103 to 104 CFU/ml, but the bacteria undergo little or no growth during the alcoholic fermentation and tend to die off because of competition from yeasts (Fig. 37.3). Nevertheless, these bacteria are capable of abundant growth in the juice, and if yeast growth is delayed, they grow and spoil the juice or cause stuck al-

LACTIC ACID BACTERIA LAB are significant in winemaking mainly because they cause spoilage under certain conditions and are responsible for the malolactic fermentation. They also release autolytic products, which may affect flavor, and they may contribute to public health concerns by raising the levels of biogenic amines and ethyl carbamate in wines. Wine LAB have the unique ability to tolerate the stresses of the wine environment, namely, low pH, the presence of ethanol and sulfur dioxide, low temperature, anaerobiosis, and dilute concentrations of nutrients. Their occurrence, growth, and significance in wines have been reviewed previously (11, 16, 71, 148, 172, 176, 302). The relevant species are members of the genera Oenococcus, Leuconostoc, Pediococcus, and Lactobacillus and include O. oeni, Leuconostoc mesenteroides, Pediococcus parvulus, Pediococcus pentosaceus, Pediococcus damnosus (formerly Pediococcus cerevisiae), and various species of Lactobacillus (e.g., Lb. brevis, Lb. plantarum, Lb. fermentum, Lb. buchneri, Lb. hilgardii, and Lb. trichodes). O. oeni is found uniquely in wine ecosys-

Figure 37.3  Growth of LAB during vinification of red wines, pH 3.0 to 3.5. The solid line shows the growth of O. oeni, often the only species present. Occasionally, species of Lactobacillus and Pediococcus develop toward the end of malolactic fermentation or at later stages during conservation (broken line). For wines of pH 3.5 to 4.0, a similar growth curve is obtained but there may be slight growth and death of LAB during the early stages of alcoholic fermentation. Also, there is a greater chance that species of Lactobacillus and Pediococcus will grow and conduct malolactic fermentation. doi:10.1128/9781555818463.ch37f3

37.  Wine coholic fermentation. About 1 to 3 weeks after completion of the alcoholic fermentation, the surviving LAB commence vigorous growth to conduct the malolactic fermentation. Final populations of 106 to 108 CFU/ml are produced. The onset, duration, and ecology of this growth are determined by many factors, which include the properties of the wine, vinification variables, and influences of other microorganisms. Consequently, the natural occurrence of malolactic fermentation and its completion by the preferred species, O. oeni, can be unpredictable (148, 176). Moreover, multiple strains of O. oeni may be associated with the one fermentation (243). Strains of O. oeni obtained from different locations exhibit significant genetic diversity (75). Of the wine properties, pH, the concentration of ethanol, and the concentration of sulfur dioxide have strong influences on the growth of LAB. Different species, and even strains within species, show different responses to these properties (72, 118, 302). Wines of low pH (e.g., pH 3.0), high ethanol content (>12%, vol/vol), and high total sulfur dioxide level (>50 μg/ml) are less likely to support the growth of LAB and may not undergo successful malolactic fermentation. Strains of O. oeni are more tolerant of low pH than those of Leuconostoc, Pediococcus, and Lactobacillus species and generally predominate in wines of pH 3.0 to 3.5. Wines with pH values exceeding 3.5 tend to have mixed microfloras consisting of O. oeni and various species of Pediococcus and Lactobacillus. Species of Pediococcus and Lactobacillus are more tolerant of higher concentrations of sulfur dioxide than is O. oeni and are more likely to occur in wines with larger amounts of this substance (72, 73). Thus, winemaker management of pH and sulfur dioxide content is important if it is desired to have malolactic fermentation conducted solely by O. oeni. Other factors that affect the growth of O. oeni in wine and successful completion of malolactic fermentation include excessive growth of molds and AAB on grapes, yeast species and strains responsible for the alcoholic fermentation, and bacteriophages. Substances produced by the growth of molds or AAB on damaged grapes may either stimulate or inhibit malolactic fermentation (176, 302). During alcoholic fermentation and subsequent autolysis, yeasts release nutrients that encourage the growth of LAB (1, 110). However, some strains of S. cerevisiae produce high concentrations of sulfur dioxide, proteins, and fatty acids (hexanoic, octanoic, decanoic, and dodecanoic) that inhibit malolactic bacteria (1, 110, 176). It is not known how non-Saccharomyces species may affect the development of the malolactic fermentation. Bacteriophages active against O. oeni occur

929 in wines and can interrupt and delay the malolactic fermentation (70, 176). Lysogeny of O. oeni is common, and bacteriophage-resistant strains have been described previously (72, 224). The fate of LAB after malolactic fermentation depends on the wine and winemaking practices, but they may survive in wine for long periods. Because wine pH is increased by malolactic fermentation, the wine becomes a more favorable environment for bacterial growth, and spoilage microorganisms such as Pediococcus and Lactobacillus species may develop, especially in wines of pH 3.5 to 4.0. As noted already, wines after malolactic fermentation may have better microbiological stability. Nevertheless, O. oeni and various species of Lactobacillus and Pediococcus can reestablish growth in such wines (303). The death of O. oeni in wines subsequent to the growth of Pediococcus and Lactobacillus species can occur (73). Strains of Pediococcus, Lactobacillus, and O. oeni produce bacteriocins that are active against one another, and this property probably contributes to the bacterial ecology at this stage (16, 17, 207). Strains of O. oeni that produce bacteriocins with broad-spectrum action against pediococci and lactobacilli would have obvious value in controlling spoilage by these bacteria.

Biochemistry, Physiology, and Genomics

LAB change the chemical composition of wine by utilizing some constituents for growth and generating metabolic end products. Depending upon the species and strains that develop, such changes will have a positive or negative impact on wine quality. Although detailed studies that correlate the growth of LAB with changes in wine composition are few (40), significant experimental research has been done to understand the biochemistry, physiology, and genomics of this growth (11, 16, 148, 172, 176). The genome of O. oeni PSU-1 has been sequenced (201).

Carbohydrates

Wines contain residual amounts of glucose and fructose (0.5 to 1 mg/ml) and smaller amounts (<0.5 mg/ ml) of other hexose and pentose sugars. Concentrations of these sugars decrease with the growth of LAB, but consistent trends have not emerged (73). This finding is consistent with the ability of these bacteria to ferment a wide range of hexose and pentose sugars and the substantial variation among species and strains in conducting these reactions (72, 172, 221, 291). The pathways utilized by LAB for sugar transport and metabolism are well described (172), but it is not known how wine conditions of low pH and high ethanol concentration affect these reactions. Pediococcus species as well as some

930 species of Lactobacillus ferment hexoses by the EmbdenMeyerhof-Parnas pathway (homofermenters), while O. oeni and some other species of Lactobacillus use the hexose monophosphate or phosphoketolase pathway (heterofermenters). The latter pathway is used by all species to metabolize pentose sugars. Fructose is also metabolized into mannitol by O. oeni (248). It is not known if wine LAB produce enzymes that hydrolyze grape juice pectins or other polysaccharides, but such properties occur in other LAB (193). Some strains of O. oeni produce b-1,3-glucanase, capable of lysing the glucans in yeast cell walls (139).

Nitrogen Compounds

The metabolism of nitrogen compounds by LAB can influence wine flavor, protein stability, and public health safety. The concentrations of some wine amino acids (e.g., arginine and histidine) decrease with the growth of LAB, suggesting their utilization as a nitrogen source (73, 148). Arginine is metabolized into ornithine, ammonia, and carbon dioxide by the arginine deiminase pathway (172, 201). Wine LAB, including strains of O. oeni, produce citrulline as an intermediate of this metabolism. Citrulline chemically reacts with ethanol to produce the carcinogen ethyl carbamate. Consequently, there is increasing concern about the levels of ethyl carbamate in wines and the role of malolactic bacteria in contributing to its production (4, 172, 289). The ability to decarboxylate amino acids into amines occurs widely throughout wine LAB, including O. oeni, with histamine, putrescine, tyramine, cadaverine, and phenylethylamine being frequently found in wines (214). While the concentration of biogenic amines in wines is lower than for many other fermented foods, the contribution of LAB to the content of biogenic amines in wines remains an emerging public health issue (168). Interestingly, certain wine-associated LAB are capable of degrading biogenic amines (126). The histidine-decarboxylating genes of some wine LAB have been sequenced, and DNA probes for the rapid detection of amine-producing strains have been developed (138, 177). Research on analytical methods for detection of biogenic amines and amine-producing bacteria is receiving considerable attention (168, 169, 214), and it is reported that multiplex PCR allows for simultaneous detection of four genes related to the production of histamine, tyramine, and putrescine (63). Nitrogen supplementation during alcoholic fermentation may increase the production of biogenic amines in the final wine (7). The production of extracellular proteases and peptidases by wine LAB requires more detailed study. Extracellular protease production by O. oeni has been

Fermentations and Beneficial Microorganisms described previously, and this property may be of nutritional advantage to the organism, as well as decreasing the levels of proteins that contribute to haze and sediments in wine (103).

Organic Acids

The decarboxylation of l-malic acid into l-lactic acid is one of the most significant metabolic reactions conducted by LAB in wines. These bacteria possess mechanisms for the transport of malic acid into the cell and the efflux of lactic acid and also possess the malolactic enzyme for decarboxylation. Purification and properties of the malolactic enzyme, as well as the bioenergetics and genomics of the reaction, have been described previously for several species (11, 148, 176, 201). However, factors that regulate the expression of malolactic activity are not completely understood. Glucose-induced inhibition has been reported previously for some strains of O. oeni (202). The metabolism of tartaric acid is not usually encountered during or after malolactic fermentation, but when it occurs the wine is spoiled (93, 273). Citric acid, which occurs in wine at concentrations up to 0.7 g/liter, can be completely or partially metabolized during malolactic fermentation, depending on the wine pH and species of LAB. Its degradation is frequently correlated with small increases in the concentration of acetic acid and diacetyl, and this observation is consistent with the action of citrate lyase, which is produced by some but not all species of wine LAB (12, 73, 176, 213). The concentrations of fumaric, gluconic, and pyruvic acids can decrease during malolactic fermentation. Gluconic acid, which is significantly increased in wines made from grapes infected with B. cinerea, is metabolized by most LAB except the pediococci (302). The presence of phenolic acids can inhibit the degradation of malic and citric acids by LAB (34). Sorbic acid, which may be added to wines in some countries to control yeast growth, can be metabolized by O. oeni to form 3,5-hexadien-2-ol and 2-ethoxyhexa-3-diene, which cause geranium-like off-flavors (93, 118, 273).

Stress Response

Malolactic bacteria are required to grow and perform under stressful environmental conditions, as mentioned previously. Their ability to tolerate the combined effects of low pH (3.0 to 3.5) and high ethanol concentrations (up to 14 to 15%) is particularly noteworthy (119), and the molecular mechanisms underlying the cellular response to these conditions are attracting significant research (11, 142, 201). Explanations are far from complete but involve multiple adaptive reactions, includ-

37.  Wine ing the generation of a proton motive force through the malolactic reaction, activation of proton-extruding ATPase, synthesis of various stress proteins, and modification of membrane structures. Progress in understanding these mechanisms should lead to strain development and management strategies that promote more effective and efficient malolactic fermentation.

Flavor Enhancement

Much has been written about the impact of malolactic fermentation on wine flavor (71, 148, 176, 277). Malolactic fermentation not only affects the taste of wine through deacidification but also contributes other flavor characteristics (often described as buttery, nutty, fruity, or vegetative) that may enhance or detract from overall acceptability (81). Such changes are related to the wine constituents metabolized by the malolactic bacteria and the nature, concentrations, and flavor thresholds of the products generated. Autolysis of malolactic bacteria may also affect flavor. Connecting sensory impression to flavor substances produced by particular species or strains of malolactic bacteria has proved elusive because of confounding influences of grape variety, yeasts, and vinification variables. Mechanisms advanced to explain the impact of malolactic fermentation on wine flavor include the production of diacetyl, esters, higher alcohols, and sulfur volatiles and the release of glycosydically linked terpenes and other compounds (277). Diacetyl production, which is linked to citrate and sugar metabolism, is well documented. Diacetyl contributes buttery or butterscotch aromas that have a positive impact at concentrations of less than 5 mg/liter, but higher levels are detrimental (12). The concentrations of various esters, organic acids, and higher alcohols change during malolactic fermentation, suggesting a role of bacterial esterases and lipases, but these observations and linkages are not definitive and require more detailed study (71, 193). Volatile thiols, responsible for vegetable and fruit ­aromas in wines, are associated with very low concentrations of substances such as methanethiol, dimethylsulfide, 3-(methysulfanyl)propan-1-ol, 3-(methylsulfanyl) propionic acid, 4-mercapto-4-methylpentan-2-one, and 3-mercaptohexanol. There is preliminary research sug­ gesting that O. oeni and other LAB produce these substances through the metabolism of methionine and cysteine-conjugated precursors in grapes (234). In a similar context, there are increasing data demonstrating that O. oeni and other species possess various glycosidase activities that liberate desirable flavor volatiles from nonvolatile, glycosylated, terpenoid precursors in grapes (84, 136, 184, 288). Glycosidase activity of

931 O. oeni is also thought to play a role in the release of glycosidically bound flavors from oak wood compounds (28, 80), thereby adding to wine complexity.

Spoilage Reactions

Uncontrolled growth of LAB during or after malolactic fermentation results in wine spoilage (30, 85, 93, 118, 273, 302). Wines containing high concentrations of residual glucose and fructose are more likely to support bacterial growth, with the production of unacceptable amounts of acetic acid, d-lactic acid, and carbon dioxide. Mannitol taint is caused by some strains of heterofermentative lactobacilli (e.g., Lb. brevis) due to enzymatic reduction of fructose to mannitol. Excessive production of diacetyl (>5 μg/ml) gives overpowering buttery flavors (12). Certain strains of L. mesenteroides, O. oeni, Lb. hilgardii, Lb. brevis, and Lactobacillus cellobiosus have been implicated in the formation of mousy taints due to the production of acetyltetrahydropyridines from lysine and ornithine metabolism (62). The degradation of glycerol, especially by species of Pediococcus and Lactobacillus, gives acrolien and associated bitterness while reducing the desirable attributes of glycerol (215, 277). Spoilage arising from the degradation of tartaric and sorbic acids has been mentioned already. However, the ability to metabolize glycerol and tartaric acid is not widespread among the LAB. The production of extracellular polysaccharides by some strains of O. oeni and P. damnosus gives unsightly ropiness and increased viscosity that retards processing by filtration (176). Glucan synthase genes responsible for this property have been identified (298). LAB also can produce volatile phenols (65), although their impact on wine quality is less well understood than those produced by Brettanomyces and Dekkera species. Although a PCR-based molecular screening method was developed to detect LAB capable of metabolizing hydroxycinnamic acids (76), further research is needed to clarify the impact of these bacteria on taints from volatile phenols in wines.

Control of Malolactic Fermentation

Control of malolactic fermentation is one of the challenges of modern winemaking (71, 148). Because the natural ­occurrence of malolactic fermentation can be unpredictable, commercial cultures of malolactic bacteria have ­become available for inoculation into the wine to induce this reaction. Generally, these are strains of O. oeni that have been selected for a range of desirable criteria (Table 37.6) (74, 148). Many factors affect the successful induction of malolactic fermentation by inocula, the most important of which include selection of the appropriate strain, proper

932 Table 37.6  Desirable properties of bacteria for use in

malolactic fermentation

________________________________________ Strong malolactic activity under wine conditions Strong ability to grow in wines, including those of low pH (3.0–3.2) and high ethanol content (14%) and those containing sulfur dioxide (50 μg/ml total), and at low temperatures (15–20°C) Resistance to destruction by bacteriophages; nonlysogenic Resistance to fungicide and pesticide residues Production of desirable flavors and no off-flavors Release of conjugated terpenes and volatile thiols Production of bacteriocins effective against spoilage bacteria Nonproduction of biogenic amines and precursors of ethyl carbamate Nonproduction of yeast-inhibitory factors if used before alcoholic fermentation Suitability for mass culture, freeze-drying, distribution, and rehydration

________________________________________ reactivation and preculture of the freeze-dried concentrate, the level of inoculum, and the timing of inoculation (209). Generally, wines are inoculated to give 106 to 107 CFU/ml of bacteria just after alcoholic fermentation is completed. Arguments have been advanced for inoculating malolactic bacteria into the juice either before or simultaneously with the yeast culture. Under these conditions, bacterial cells are not exposed to the stresses of high ethanol concentration, giving a higher probability of successful growth and malic acid degradation. Risks associated with early inoculation of malolactic bacteria include their metabolism of grape juice sugars to yield unacceptable concentrations of acetic acid and interference with yeast growth and alcoholic fermentation (128, 243, 252, 260). Even under optimum conditions, inoculation with malolactic bacteria does not ensure successful completion of the fermentation. In some instances, the particular properties of the wine may not be suitable for growth of the bacterial strain or the mixture of strains inoculated (128). Several biotechnological innovations have been considered to overcome this problem. Bioreactor systems consisting of high concentrations of malolactic bacteria immobilized in beads of alginate, carrageenan, or other support have been developed (71, 86, 163, 183). An alternative system uses high densities (109 to 1010 CFU/ml) of O. oeni retained within a cell recycle membrane bioreactor (123). Under these conditions, cells of malolactic bacteria act as biocatalysts and, in the absence of growth, rapidly convert malic acid into lactic acid in wine that is passed through the reactor on a continuous basis. Such technologies give rapid, continuous deacidification of wines but have not proved commercially successful, mainly because of instability of the malolactic activity of cells in the reactor.

Fermentations and Beneficial Microorganisms The availability of strains of S. cerevisiae that could carry out malolactic fermentation simultaneously with alcoholic fermentation would be most attractive to winemakers. With the use of recombinant DNA technology, strains of S. cerevisiae expressing the malolactic gene of Lactococcus lactis and the malate transport gene of Schizosaccharomyces pombe have been constructed and reported to give good malolactic activity in wine (296, 297). Additionally, a coculture of S. cerevisiae and O. oeni successfully produced the malolactic fermentation (208). Complete prevention of malolactic fermentation is an option preferred by some winemakers. Wines of low pH (<3.2), high ethanol content (>14%), and high sulfur dioxide level (>50 mg/ml) are less prone to malolactic fermentation. The bacteriocin nisin and the antibacterial enzyme lysozyme are effective in preventing the growth of LAB in wine and could be added if permitted (67, 93, 102, 125).

ACETIC ACID BACTERIA AAB cause the vinegary spoilage of wines through the oxidation of ethanol into acetaldehyde and acetic acid (13). In addition, their growth on grapes before fermentation can produce substances that not only affect wine flavor but interfere with the growth of yeasts, leading to stuck fermentations (89, 93, 273). The taxonomy of AAB is complex and has undergone many revisions. Currently, these bacteria are classified into eight genera: Acetobacter, Gluconobacter, Gluconoacetobacter, Acidomonas, Asaia, Kozakia, Swaminathania, and Saccharibacter (271, 287). The main isolates from grapes and wine have been Gluconobacter oxydans, Acetobacter pasteurianus, and Acetobacter aceti, but recently Gluconoacetobacter hansenii and Gluconoacetobacter liquefaciens have been found (94, 130, 131), and additional species are likely to occur. These bacteria are isolated by culturing on glucose-yeast extract, calcium carbonate agar, and Wallerstein Laboratories nutrient agar to which 1 to 2% (wt/vol) of ethanol has been added (14, 89). Some isolates give very weak growth and are difficult to maintain, and there is evidence that they enter a viable but nonculturable state in wine ecosystems (14, 95). The species can be differentiated by key phenotypic tests (89, 94, 95), but PCR-based methods are available for species and strain differentiation (130, 131, 158, 285, 287).

Ecology

Sound, unspoiled grapes harbor low populations of AAB, generally less than 102 CFU/g, with G. oxydans

37.  Wine being the predominant species. Damaged, spoiled grapes and those infected with the mold B. cinerea have higher populations (>106 CFU/g) that are characterized by a mixture of Gluconobacter, Acetobacter, and Gluconoacetobacter species (10, 94, 130, 131). In the absence of yeasts, AAB quickly grow in grape juice, reaching populations of 106 to 108 CFU/ml. The extent of their growth during alcoholic fermentation depends on their initial populations relative to those of yeasts, juice pH, and the concentration of added sulfur dioxide. In juice prepared from sound grapes where the yeast population is 10 to 100 times greater than that of AAB, there appears to be little growth or influence of these bacteria on the alcoholic fermentation (90, 94, 130, 131). However, when initial populations of these bacteria exceed about 104 CFU/ml, they may grow in conjunction with yeasts during the early stages of fermentation. Populations as high as 108 CFU/ml can develop but die off due to the combined influences of ethanol and anaerobosis caused by yeast growth. Nevertheless, acetic acid and other substances produced by AAB become inhibitory to the yeast, causing premature cessation of the alcoholic fermentation (89, 90). There is a notable succession of species, and even strains within species, throughout vinification, reflecting different organism responses to the combined stresses of ethanol, low pH, sulfur dioxide, and oxygen availability. G. oxydans quickly dies off at the beginning of fermentation, giving way to a prevalence of A. pasteurianus, A. aceti, and Gluconoacetobacter species in the later stages (94, 130, 131). At the end of alcoholic fermentation, the population of AAB is generally less than 102 to 103 CFU/ml. However, subsequent transfer of the wine from fermentation tanks to other storage vessels may produce sufficient agitation and aeration to encourage the growth of the survivors to a level of 105 CFU/ml or higher. It is not uncommon to isolate A. pasteurianus or A. aceti from wines during bulk storage in barrels or tanks kept under anaerobic conditions (89, 108, 304). The mechanisms by which these aerobic bacteria survive for long periods under apparently anaerobic or semianaerobic conditions require explanation. Their growth is activated by exposure of the wine to air, and the wine is quickly spoiled (95). Even after bottling, ingress of oxygen through defective corks can initiate their growth and wine spoilage (14). Wine spoilage by AAB is best controlled by good hygienic practices and prevention of wine exposure to air.

Biochemistry

Oxidation of ethanol into acetic acid is a key reaction of AAB. Ethanol is first oxidized by ethanol dehydrogenase

933 into acetaldehyde, which is oxidized by acetaldehyde dehydrogenase into acetic acid. Species of Acetobacter further oxidize acetic acid into carbon dioxide and water by the tricarboxylic acid (TCA) cycle, but this reaction is inhibited in the presence of ethanol. Species of Gluconobacter do not have a fully functional TCA cycle and are unable to completely oxidize acetic acid (78). Sulfur dioxide, which is generally present in wine, can chemically trap acetaldehyde, causing an accumulation of this intermediate at the expense of its further oxidation into acetic acid. Ethanol concentrations above 10% become increasingly inhibitory to the growth of AAB and their ability to oxidize ethanol. Aldehyde dehydrogenase is less stable than ethanol dehydrogenase at high concentrations of ethanol, and such conditions give an increased accumulation of acetaldehyde. Lower oxygen concentrations also favor the accumulation of acetaldehyde. In addition to acetic acid and acetaldehyde, ethyl acetate is another end product that is significant in the vinegary spoilage of wines (89, 90, 93, 118). Acetobacter and Gluconobacter lack a functional Embden-Meyerhof-Parnas pathway and metabolize hexose and pentose sugars by the hexose monophosphate pathway into acetic and lactic acids. However, Acetobacter species give only weak metabolism of sugars due to their decreased ability to phosphorylate these substrates. At pHs below 3.5 or glucose concentrations above 5 to 15 mM, as occur in grapes or grape juice, metabolism of glucose by the hexose monophosphate pathway is inhibited and glucose is directly oxidized into gluconic and ketogluconic acids (10). Consequently, grapes heavily infected with G. oxydans give juices with high concentrations (50 to 70 mg/ml) of gluconic acid. Strains of A. aceti and A. pasteurianus also produce gluconic acid in grape juice but to a lesser extent than G. oxydans (90). Strains of G. oxydans and A. aceti oxidize glycerol into dihydroxyacetone (10). Glycerol is not normally present in grape juice, but its concentration may be significant (up to 20 mg/ml) in juices prepared from grapes infected with B. cinerea or other molds. The metabolism of glycerol by AAB, either on grapes, during the early stages of alcoholic fermentation, or in wine, leads to significant production of dihydroxyacetone that may affect wine quality as well as bind added sulfur dioxide (10, 89, 273). Species of Acetobacter use the TCA cycle for the metabolism of organic acids, causing decreases in concentrations of citric, succinic, malic, tartaric, and lactic acids in wines. Moreover, lactate may be oxidized into acetoin. The metabolism of amino acids and protein by wine AAB requires study. Some AAB produce

934 extracellular cellulose and other polysaccharides that cause ropiness and filtration difficulties (89, 93).

OTHER BACTERIA Most bacteria cannot survive the high ethanol contents and low pHs of wines. Nevertheless, there are occasional reports that Bacillus and other bacterial species can survive and grow in wines and contribute to spoilage (30, 93, 108, 118). Bacillus coagulans, Bacillus circulans, and Bacillus subtilis were isolated from spoiled sweet wines, and Bacillus megaterium was isolated from spoiled brandy. B. coagulans and Bacillus badius isolated from wine corks had limited ability to grow in wines (108, 171, 273). Bacillus thuringiensis is prevalent on wine grapes because of its use in vineyards as a bioinsecticide. It carries over into grape juice and wine, where it remains viable but is unable to grow. It has no impact on the ability of S. cerevisiae to conduct alcoholic fermentation but is inhibitory to O. oeni in agar culture but not liquid culture (9). Juices and wines of high pH (e.g., 4.0) have, on rare occasions, been spoiled by the growth of Clostridium butyricum and have elevated concentrations of butyric, isobutyric, propionic, and acetic acids (273). Species of Actinomyces and Streptomyces have been isolated from wines, corks, and wooden barrels and may contribute to the musty, earthy, or corky taints sometimes found in wines (2, 108, 171).

MOLDS Molds (filamentous fungi) have an impact on wine production at several stages during the chain of operations. Their main influence occurs during grape cultivation, when they cause grape spoilage and loss of yield and produce metabolites that adversely affect wine quality. They also contaminate wooden barrels and corks (108, 110). Although molds are usual contaminants of grape juice, the conditions of anaerobiosis, increasing ethanol concentration, and the presence of sulfur dioxide prevent their growth during fermentation and conservation and in the final product. Grapes and vines harbor a range of mold species, the populations and diversity of which depend on grape variety, degree of berry maturity and physical damage, climatic conditions, and viticultural practices. Infections of the vine by species such as Plasmopara viticola (downy mildew), Uncinula necator (powdery mildew), Phomopsis viticola (cane and leaf spot), and Eutypa lata (dieback) have major effects on the quality and yield of grapes and, eventually, can kill the vine (97, 118). Rotting and spoilage of grape berries before harvest are

Fermentations and Beneficial Microorganisms caused by a variety of species, the principal one being B. cinerea, which causes bunch rot (88). Other species include those of Penicillium, Aspergillus, Rhizopus, Mucor, Alternaria, and Cladosporium, whose presence varies with grape maturity and climatic conditions (181, 261, 262). Grapes that are heavily infected with molds have altered chemical compositions and mold enzymes that adversely affect wine flavor, color, and filterability and the growth of yeasts during alcoholic fermentation (33, 110, 275). Fungal contamination of vines and grapes is controlled by the application of fungicides. However, such use must be carefully managed to minimize residues in the juice and their potential inhibitory effects on the alcoholic and malolactic fermentations (31). Biocontrol of grape molds with appropriate, antagonistic yeasts is a novel direction for minimizing the use of chemical fungicides (111). Ochratoxin A (OTA) and fumonisins are mycotoxins with carcinogenic and immunosuppressive properties. Their discovery in wines has focused attention on grape molds and their public health significance. The presence of OTA in wine is linked mainly to the growth of Aspergillus carbonarius on grapes, although Aspergillus niger and Aspergillus tubingensis also produce OTA to a lesser degree (45, 152, 181, 261). The discovery of OTA in wine has led to the development of detection methods and the investigation of treatments to mitigate its presence (120, 178, 301). Fumonisin B2 and B4 are produced by A. niger and were found in grapes, must, and wines (45, 173–175, 203). In one study, fumonisins were detected in 18 of 77 wine samples from 13 countries, indicating that the mycotoxin is widespread in nature (203). Other mycotoxins of potential concern are patulin and tricothecene (262). However, mycotoxin production on grapes does not necessarily mean that mycotoxins will occur in the wine because they may be inactivated during the fermentation processes or adsorbed onto the cell surfaces of yeasts (19). Although mycotoxins in wines are generally found in low concentrations, emerging government focus on cumulative and chronic exposures to toxins in foods may increase public health concerns about these compounds in the future. Although B. cinerea is well-known for causing bunch rot, its controlled development on grapes gives the distinctive and highly prized botrytized sweet wines: Sauternes, Trockenbeerenauslese, and Tokay (88, 247). Under certain climatic and viticultural conditions, this species parasitizes grape berries without significantly disrupting the general integrity of their skins, causing the so-called pourriture noble, or noble rot. Mold growth on and in the berry leads to its dehy-

37.  Wine dration and the concentration of its constituents. This concentration effect, along with fungal metabolism of grape sugars and acids (especially tartaric), gives a juice that has increased sugar concentration (300 to 400 mg/ml); high concentrations of glycerol (3 mg/ml), other polyols, and gluconic acid; and less tartaric acid. Fermentations of such juices require particular attention as they are prone to become stuck, possibly due to nitrogen deficiency combined with the higher sugar content, and also there is evidence that the mold secretes antiyeast substances (110). B. cinerea also produces various phenolic oxidases and glycosidases that can affect wine color and flavor, and it produces extracellular soluble glucans that block membranes during filtration processes (88). Methods to quantitatively detect B. cinerea have been reported (32, 82). Mold contamination of wine corks and wooden barrels can cause earthy, moldy, or corky taints in wines and rejection of the product (171, 266). Molds use the cork or wood as a growth substrate, generating potent aroma compounds such as trichloroanisoles, 1-octen-3one, geosmin, and guaiacol that are subsequently leached into the wine to cause the taint. Molds isolated from these sources include species of Penicillium, Aspergillus, Trichoderma, Cladosporium, Paecilomyces, and Monilia (2, 266).

SPARKLING WINES OR CHAMPAGNE Sparkling wines contain dissolved carbon dioxide that is produced by secondary fermentation of a base wine in a closed system, such as a bottle or a tank (23, 30, 153). The microbial ecology and biochemistry of base wine production are essentially the same as those of table wine production. Three types of secondary fermentation processes are used. The traditional méthode champenoise requires fermentation in the bottle in which the wine is sold. A base wine (usually white wine) is selected, to which sugar or sugar syrup is added to give a final concentration of about 2.4 mg/ ml. Yeast nutrients (diammonium phosphate and vitamins) may also be added. A suitable strain of S. cerevisiae is inoculated into the wine, which is thoroughly mixed and transferred into the bottles (levurage or tirage). The bottles are sealed with crown caps or corks and then stored at 12 to 15°C for fermentation. The bottles are specially made to withstand the high pressure (about 600 kPa) of carbon dioxide produced during fermentation. Fermentation is completed in 3 to 6 months, after which the wine is aged in the bottles in contact with the yeast lees for at least 6 to 12 months, or longer for premium-quality wines. After storage,

935 the bottles are restacked with their necks downwards and periodically twisted or shaken to facilitate settling of the yeast sediment into the neck. This process is called riddling, or remuage. Finally, the sedimented yeast plug is removed from the bottle neck by a process termed disgorgement (dégorgement). The yeast plug is frozen by dipping the neck of the bottle into calcium chloride solution at –24°C, and then the cork or cap is removed, allowing the internal pressure to force out the yeast plug with little loss of wine. The lost wine is replaced by the addition of base wine (dosage), which may contain a small amount of sugar to give a desired amount of sweetness, and sulfur dioxide. Riddling, disgorgement, and dosage are automated processes in the modern winery. Following these operations, the bottles are corked and an agraffe (wire clamp) is applied. Variations in the méthode champenoise include the transfer method and bulk fermentation, or the Charmat process. In the transfer process, fermentation is conducted in the bottle as already described, followed by a short maturation on lees. The bottle contents are chilled and emptied into a pressurized tank. Dosage is added to the wine, which is filtered to remove the yeasts and bottled. In the Charmat process, the base wine, with added sugar and yeast, is fermented in a pressurized tank. After fermentation, the wine is chilled and clarified by centrifugation before transfer into a second tank (pressurized) containing the dosage. The wine is finally filtered and bottled (23, 153). Special strains of S. cerevisiae are required for the secondary fermentation. Criteria for these strains include the following: give complete sugar fermentation under conditions of low temperature (10 to 15°C), relatively high ethanol concentration (8 to 12%), low pH (as low as 3.0), the presence of up to 20 μg/ml of free sulfur dioxide, low nutrient availability, and increasing pressure of carbon dioxide (up to 600 kPa); flocculate and sediment to facilitate the riddling process; give good flavor; and undergo autolysis during aging on the lees (188, 189). Usually, the yeast is inoculated into the wine base at a population of 1 × 106 to 5 × 106 CFU/ml. The secondary fermentation is completed over the next 30 to 40 days, during which time the yeast grows to a maximum population of about 107 CFU/ml. Subsequently, yeast cells slowly die, and generally, viable cells cannot be detected after 100 days. Autolysis of the yeast then commences (44, 282). Chemical changes occur in two phases, namely, during the secondary fermentation and during the period of aging and yeast autolysis. During secondary

936 fermentation, the added sugar is utilized with the production of carbon dioxide and about 1% ethanol. In the early stages, amino nitrogen is consumed, but after exhaustion of the sugar, yeast cells give a small efflux of amino nitrogen into the wine. Minor changes in the concentrations of glycerol, some organic acids, and some esters have been noted, but the data are not consistent. The yeast cells are operating under extreme environmental conditions, especially increasing concentrations of carbon dioxide and increasing physical pressure, and these circumstances are likely to affect their metabolic behavior. Yeast autolysis is characterized by the degradation of cellular macromolecules and the release of their degradation products, as well as other cell constituents, into the external medium. This process correlates with gradual increases in concentrations of proteins, amino acids, nucleic acids, lipids, free fatty acids, and mannoproteins (originating from the yeast cell wall) in the wine, as well as changes in the concentrations of various esters, higher alcohols, carbonyl compounds, terpenes, and lactones. These changes influence sparkling wine aroma, flavor, and bubble properties (44, 105, 151, 188, 189, 229, 235, 309). Innovations to increase the efficiency of the secondary fermentation and aging processes include the use of immobilized yeast cells (306), killer yeasts (282), and autolytic mutant yeasts (188). Interspecific hybrids obtained through crosses of S. cerevisiae and S. bayanus var. uvarum were highly flocculent, fermented well at both low and high temperatures, and produced sparkling wines comparable to the parent strains (55).

FORTIFIED WINES Fortified wines such as sherry, port, and Madeira have ethanol concentrations of 15 to 22%. This higher ethanol concentration is achieved by the addition of ethanol (usually derived from the distillation of wine products) at certain stages during the process. Details of fortified wine production are given elsewhere (30, 133, 241). With sherry-style wines, an essentially dry white wine base is produced from particular grape varieties. The yeast ecology and the biochemistry of the fermentation are similar to those described already. At the completion of alcoholic fermentation, the wine is fortified and transferred into oak casks for aging and maturation. In the case of oloroso sherries, the wine is fortified to contain 18 to 19% ethanol, and this fortification stops any further microbiological processes. Consequently, subsequent maturation is a chemical process. With the fino sherries, the wine is fortified

Fermentations and Beneficial Microorganisms to contain 15 to 16% ethanol and aged in a series of oak casks by the solera system, in which portions of the wine in each cask are systematically removed and replaced with an amount of younger wine of the same style so that the wine is continuously blended and emerges with a consistent character. The frequency of transfers and amount of wine removed from each cask vary according to the producer. This process encourages the natural formation of a surface film or velum of yeast growth (flor) at the air-wine interface. Essentially, the velum is a wrinkled layer of yeast biomass about 5 to 10 mm thick. It represents a unique ecological niche where yeasts (and probably bacteria as well) have adapted to survive and metabolize in the presence of high concentrations of ethanol. The velum consists mainly of a mixture of distinct, hydrophobic strains of S. cerevisiae, but strains of Torulaspora delbrueckii, Dekkera bruxellensis, Candida cantarellii, and Zygosaccharomyces may also be present (100, 187). There appears to be a definitive evolution of species and strains during the process of velum development, and the ecology may be different at the top (aerobic) and bottom (anaerobic) locations. The oxidative metabolism of the velum decreases the concentrations of wine acids, glycerol, and alcohol and substantially increases the concentration of acetaldehyde. Amino acids (including proline) are assimilated, with the consequent production of higher alcohols. These changes, as well as many other less quantitative reactions conducted by the yeasts, contribute to the unique flavor of fino sherries (187, 216). Port-style wines are prepared from a red wine base, and the Madeira-style wines are produced from a blend of red and white wines. In these processes, fortification with ethanol is done at an appropriate stage during the alcoholic fermentation so that the fermentation is arrested to give a wine with residual unfermented sugar and a desired amount of sweetness. These wines are then subjected to aging and maturation that do not involve any secondary fermentations because of the high concentration of ethanol (133, 241). Despite their high ethanol concentrations, fortified wines may undergo spoilage from ethanol-tolerant strains of yeasts (Zygosaccharomyces bisporus, S. cerevisiae, Rhodotorula, Candida, and Brettanomyces species) and LAB (Lb. hilgardii and other species) (273, 281). The authors gratefully acknowledge Eric Thornber for his assistance with this chapter. The opinions and statements expressed by M.E.P. represent his views only and do not necessarily reflect those of the U.S. Food and Drug Administration.

37.  Wine

937

References

  16. Bauer, R., and L. M. T. Dicks. 2004. Control of malolactic fermentation in wine—a review. S. Afr. J. Enol. Viticult. 25:74–88.   17. Bauer, R., H. A. Nel, and L. M. T. Dicks. 2003. Pediocin PD-1 as a method to control growth of Oenococcus oeni in wine. Am. J. Enol. Viticult. 54:86–91.   18. Beh, A. L., G. H. Fleet, C. Prakitchaiwattana, and G. M. Heard. 2006. Evaluation of molecular methods for the analysis of yeasts in foods and beverages, p. 69–106. In A. D. Hocking, J. I. Pitt, R. A. Samson, and U. Thrane (ed.), Advances in Food Mycology. Springer, Berlin, Germany.   19. Bejaoui, H., F. Mathieu, P. Taillandier, and A. Lebrihi. 2004. Ochratoxin A removal in synthetic and natural grape juices by selected enological Saccharomyces strains. J. Appl. Microbiol. 97:1038–1044.   20. Bell, S.-J., and P. A. Henschke. 2005. Implications of nitrogen nutrition for grapes, fermentation and wine. Aust. J. Grape Wine Res. 11:242–295.   21. Berry, D. R., and J. C. Slaughter. 2003. Alcoholic beverages fermentation, p. 25–29. In A. G. H. Lea and J. Piggott (ed.), Fermented Beverage Production, 2nd ed. Kluwer Academic, New York, NY.   22. Berthels, N. J., R. R. Cordero Oteso, F. F. Bauer, J. M. Thevelein, and I. S. Pretorius. 2004. Discrepancy in glucose and fructose utilization during fermentation by Saccharomyces cerevisiae wine yeast strains. FEMS Yeast Res. 4:683–689.   23. Bidan, P., M. Feuillat, and J. Moulin. 1986. Rapport de la France. Les vins Mousseux. Bull. Off. Int. Vin 59:563–626.   24. Bisson, L. F. 1993. Yeasts—metabolism of sugars, p. 55–75. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland.   25. Bisson, L. F. 1999. Stuck and sluggish fermentations. Am. J. Enol. Viticult. 50:107–119.   26. Bisson, L. F., A. L. Waterhouse, S. E. Ebeler, M. Andrew-Walker, and J. T. Lapsley. 2002. The present and the future of the international wine industry. Nature 418:696–699.   27. Blasco, L., S. Ferrer, and I. Pardo. 2003. Development of specific fluorescent oligonucleotide probes for in situ identification of wine lactic acid bacteria. FEMS Microbiol. Lett. 225:115–123.   28. Bloem, A., A. Lonvaud-Funel, and G. de Revel. 2008. Hydrolysis of glycosidically bound flavour compounds from oak wood by Oenococcus oeni. Food Microbiol. 25:99–104.   29. Boulton, R. 2003. Red wines, p. 107–108. In A. G. H. Lea and J. R. Piggott (ed.), Fermented Beverage Production, 2nd ed. Kluwer Academic, New York, NY.   30. Boulton, R. B., V. L. Singleton, L. F. Bisson, and R. E. Kunkee. 1995. Principles and Practices of Winemaking. Chapman and Hall, New York, NY.   31. Cabras, R., and A. Angioni. 2000. Pesticide residues in grapes, wines and their processing products. J. Agric. Food Chem. 48:967–973.

  1. Alexandre, H., P. J. Costello, F. Remize, J. Guzzo, and M. Guilloux-Benatier. 2004. Saccharomyces cerevisiae-Oenococcus oeni interactions in wine: current knowledge and perspectives. Int. J. Food Microbiol. 93:141–154.   2. Alvarez-Rodríguez, M. L., C. Belloch, M. Villa, F. Uruburu, G. Larriba, and J.-J. R. Coque. 2003. Degradation of vanillic acid and production of guaiacol by microorganisms isolated from cork samples. FEMS Microbiol. Lett. 220:49–55.   3. Amerine, M. A. 1985. Winemaking, p. 67–81. In H. Koprowski and S. A. Plotin (ed.), World’s Debt to Pasteur. Alan R. Liss Incorporated, New York, NY.   4. Arena, M. E., and M. C. Manca de Nadra. 2005. Influence of ethanol and low pH on arginine and citrulline metabolism in lactic acid bacteria from wine. Res. Microbiol. 156:858–864.   5. Arevalo-Villena, M., E. J. Bartowsky, D. Capone, and M. A. Sefton. 2010. Production of indole by wineassociated microorganisms under oenological conditions. Food Microbiol. 27:685–690.   6. Aroyo-López, F. N., R. Pérez-Torrado, A. Querol, and E. Barrio. 2010. Modulation of the glycerol and ethanol syntheses in the yeast Saccharomyces kudriavzevii differs from that exhibited by Saccharomyces cerevisiae and their hybrid. Food Microbiol. 27:628–637.   7. Bach, B., S. Colas, L. Massini, L. Barnavon, and P. Vuchot. 2011. Effect of nitrogen addition during alcoholic fermentation on the final content of biogenic amines in wine. Ann. Microbiol. 61:185–190.   8. Backhus, L. E., J. de Risi, P. O. Brown, and L. F. Bisson. 2001. Functional geonomic analysis of a commercial wine strain of Saccharomyces cerevisiae under differing nitrogen conditions. FEMS Yeast Res. 1:111–125.   9. Bae, S., G. H. Fleet, and G. M. Heard. 2004. Occurrence and significance of Bacillus thuringiensis on wine grapes. Int. J. Food Microbiol. 94:301–312.   10. Barbe, J. C., G. de Revel, A. Joyeux, A. Bertrand, and A. Lonvaud-Funel. 2001. Role of botrytized grape micro-organisms in SO2 binding phenomena. J. Appl. Microbiol. 90:34–42.   11. Bartowsky, E. 2005. Oenococcus oeni and malolactic fermentation—moving into the molecular arena. Aust. J. Grape Wine Res. 11:174–187.   12. Bartowsky, E., and P. A. Henschke. 2004. The buttery attribute of wine—diacetyl—desirability, spoilage and beyond. Int. J. Food Microbiol. 96:235–252.   13. Bartowsky, E., and P. A. Henschke. 2008. Acetic acid bacteria spoilage of bottled red wine—a review. Int. J. Food Microbiol. 125:60–70.   14. Bartowsky, E. J., D. Xia, R. L. Gibson, G. H. Fleet, and P. A. Henschke. 2003. Spoilage of bottled red wine by acetic acid bacteria. Lett. Appl. Microbiol. 36:307–314.   15. Bauer, F. F., and I. S. Pretorius. 2000. Yeast stress response and fermentation efficiency: how to survive the making of wine—a review. S. Afr. J. Enol. Viticult. 21:27–51.

938   32. Cadle-Davidson, L. 2008. Monitoring pathogenesis of natural Botrytis cinerea infections in developing grape berries. Am. J. Enol. Viticult. 59:387–395.   33. Calonnec, A., P. Carolaro, C. Poupot, D. Dubourdieu, and P. Dariet. 2004. Effects of Uncinula necator on the yield and quality of grapes (Vitis vinifera) and wine. Plant Pathol. 53:434–445.   34. Campos, F. M., A. R. Figueiredo, T. A. Hogg, and J. A. Couto. 2009. Effect of phenolic acids on glucose and organic acid metabolism by lactic acid bacteria from wine. Food Microbiol. 26:409–414.   35. Cantarelli, C. 1989. Factors affecting the behaviour of yeast in wine fermentation, p. 127–151. In C. Cantarelli and G. Lanzarini (ed.), Biotechnology Applications in Beverage Production. Elsevier Applied Science, London, United Kingdom.   36. Capece, A., C. Fiore, A. Maraz, and P. Romano. 2005. Molecular and technological approaches to evaluate strain biodiversity in Hanseniaspora uvarum of wine origin. J. Appl. Microbiol. 98:136–144.   37. Capone, D. L., K. van Leeuwen, K. H. Pardon, M. A. Daniel, G. A. Elsey, A. D. Coulter, and M. A. Sefton. 2010. Identification and analysis of 2-chloro-6-methylphenol, 2,6-dichlorophenol and indole: causes of taints and off-flavours in wines. Aust. J. Grape Wine Res. 16:210–217.   38. Cappello, M. S., G. Bleve, F. Grieco, F. Dellaglio, and G. Zacheo. 2004. Characterization of Saccharomyces cerevisiae strains isolated from must of grape grown in experimental vineyard. J. Appl. Microbiol. 97:1274–1280.   39. Carr, F. J., D. Chill, and N. Maida. 2002. The lactic acid bacteria: a literature survey. Crit. Rev. Microbiol. 28:281–370.   40. Carrau, F. M., K. Medina, E. Boido, L. Farina, C. Gaggero, E. Dellacassa, G. Versini, and P. A. Henschke. 2005. De novo synthesis of monoterpenes by Saccharomyces cerevisiae wine yeasts. FEMS Microbiol. Lett. 243:107–115.   41. Cebollero, E., D. Gonzalez-Ramos, and R. Gonzalez. 2009. Construction of a recombinant autolytic wine yeast strain overexpressing the csc1-1 allele. Biotechnol. Prog. 25:1598–1604.   42. Charoenchai, C., G. H. Fleet, and P. Henschke. 1998. Effects of temperature, pH and sugar concentration on the growth rates and cell biomass of wine yeast. Am. J. Enol. Viticult. 49:283–288.   43. Charoenchai, C., G. H. Fleet, P. Henschke, and B. E. N. Todd. 1997. Screening of non-Saccharomyces wine yeasts for the presence of extracellular hydrolytic enzymes. Aust. J. Grape Wine Res. 3:2–8.   44. Charpentier, C., and M. Feuillat. 1993. Yeast autolysis, p. 225–242. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland.   45. Chiotta, M. L., A. Susca, G. Stea, G. Mulè, G. Perrone, A. Logrieco, and S. N. Chulze. 2011. Phylogenetic characterization and ochratoxin A—fumonisin profile of black Aspergillus isolated from grapes in Argentina. Int. J. Food Microbiol. 149:171–176.

Fermentations and Beneficial Microorganisms   46. Ciani, M., L. Beco, and F. Comitini. 2006. Fermentation behavior and metabolic interactions of multistarter wine yeast fermentations. Int. J. Food Microbiol. 108:239–245.   47. Ciani, M., and F. Comitini. 2011. Non-Saccharomyces wine yeasts have a promising role in biotechnological approaches to winemaking. Ann. Microbiol. 61:25–32.   48. Ciani, M., F. Comitini, I. Mannazzu, and P. Domizio. 2010. Controlled mixed culture fermentation: a new perspective on the use of non-Saccharomyces yeasts in winemaking. FEMS Yeast Res. 10:123–133.   49. Ciani, M., F. Fatichenti, and I. Mannazzu. 2002. Yeasts in winemaking biotechnology, p. 111–123. In M. Ciani (ed.), Biodiversity and Biotechnology of Wine Yeasts. Research Signpost, Kerala, India.   50. Clemente-Jimenez, J. M., L. Mingorance-Cazorla, S. Martínez-Rodríguez, F. J. L. Heras-Vázquez, and F. Rodríguez-Vico. 2005. Influence of sequential yeast mixtures on wine fermentation. Int. J. Food Microbiol. 98:301–308.   51. Cocolin, L., L. Bisson, and D. A. Mills. 2000. Direct profiling of the dynamics in wine fermentations. FEMS Microbiol. Lett. 189:81–87.   52. Cocolin, L., V. Pepe, F. Comitini, G. Comi, and M. Ciani. 2004. Enological and genetic traits of Saccharomyces cerevisiae isolated from former and modern wineries. FEMS Yeast Res. 5:237–246.   53. Cocolin, L., K. Rantsiou, L. Iacumin, R. Zirconi, and G. Comi. 2004. Molecular detection and identification of Brettanomyces/Dekkera bruxellensis and Brettanomyces/ Dekkera anomalus in spoiled wine. Appl. Environ. Microbiol. 70:1347–1355.   54. Cole, V. C., and A. C. Noble. 2003. Flavor chemistry, p. 393–412. In A. G. H. Lea and J. R. Piggott (ed.), Fermented Beverage Production, 2nd ed. Kluwer Academic, New York, NY.   55. Coloretti, F., C. Zambonelli, and V. Tini. 2006. Characterization of flocculent Saccharomyces interspecific hybrids for the production of sparkling wines. Food Microbiol. 23:672–676.   56. Combina, M., A. Elia, L. Mercado, C. Catania, A. Ganga, and C. Martinez. 2003. Dynamics of yeast populations during spontaneous fermentations of wines from Mendonza, Argentina. Int. J. Food Microbiol. 99:237–243.   57. Combina, M., L. Mercado, P. Borgo, A. Elia, V. Jofré, A. Ganga, and C. Martinez. 2005. Yeasts associated to Malbec grape berries from Mendoza, Argentina. J. Appl. Microbiol. 98:1055–1061.   58. Constanti, M., M. Poblet, L. Arola, A. Mas, and J. M. Guillamón. 1997. Analysis of yeast populations during alcoholic fermentation in a newly established winery. Am. J. Enol. Viticult. 48:339–344.   59. Conterno, L., C. M. L. Joseph, T. J. Arvik, T. HenickKling, and L. F. Bisson. 2006. Genetic and physiological characterization of Brettanomyces bruxellensis strains isolated from wines. Am. J. Enol. Viticult. 57:139–147.   60. Coombe, B. G., and P. R. Dry. 2004. Viticulture, 2nd ed., vol. 1. Resources. Winetitles, Adelaide, Australia.

37.  Wine   61. Costa, A., A. Barata, M. Malfeito-Ferriera, and V. Loureiro. 2008. Evaluation of the inhibitory effect of dimethyl dicarbonate (DMDC) against wine microorganisms. Food Microbiol. 25:422–427.   62. Costello, P. J., and P. A. Henschke. 2002. Mousy offflavour of wine. Precursors and biosynthesis of the causative N-heterocycles 2-ethyl tetrahydropyridine, 2-acetyltetrahydropyridine and 2-acetyl-1-pyrroline by Lactobacillus hilgardii DSM 20176. J. Agric. Food Chem. 50:7079–7087.   63. Coton, M., A. Romano, G. Spano, K. Ziegler, C. Vetrana, C. Desmarais, A. Lonvaud-Funel, P. Lucas, and E. Coton. 2010. Occurrence of biogenic amine-forming lactic acid bacteria in wine and cider. Food Microbiol. 27:1078–1085.   64. Coulon, J., J. I. Husnik, D. L. Inglis, G. K. van der Merwe, A. Lonvaud, D. J. Erasmus, and H. J. J. van Vuuren. 2006. Metabolic engineering of Saccharomyces cerevisiae to minimize the production of ethyl carbamate in wine. Am. J. Enol. Viticult. 57:113–124.   65. Couto, J. A., F. M. Campos, F. R. Figueiredo, and T. Hogg. 2006. Ability of lactic acid bacteria to produce volatile phenols. Am. J. Enol. Viticult. 57:166–171.   66. Couto, J. A., F. Neves, F. Campos, and T. Hogg. 2005. Thermal inactivation of the wine spoilage yeasts Dekkera/ Brettanomyces. Int. J. Food Microbiol. 104:337–344.   67. Daeschel, M. A., D. S. Jung, and B. T. Watson. 1991. Controlling malolactic fermentation with nisin and nisinresistant strains of Leuconostoc oenos. Appl. Environ. Microbiol. 57:601–603.   68. Dahabieh, M. S., J. I. Husnik, and H. J. J. van Vuuren. 2009. Functional expression of the DUR3 gene in a wine yeast strain to minimize ethyl carbamate in Chardonnay wine. Am. J. Enol. Viticult. 60:537–541.   69. Dávalos, A., and M. A. Lasunción. 2009. Healthpromoting effects of wine phenolics, p. 571–592. In M. V. Moreno-Arribas and M. C. Polo (ed.), Wine Chemistry and Biochemistry. Springer, New York, NY.   70. Davis, C. R., N. F. A. Silveira, and G. H. Fleet. 1985. Occurrence and properties of bacteriophages of Leuconstoc oenos in Australian wines. Appl. Environ. Microbiol. 50:872–876.   71. Davis, C., D. Wibowo, R. Eschenbruch, T. H. Lee, and G. H. Fleet. 1985. Practical implications of malolactic fermentation—a review. Am. J. Enol. Viticult. 36:209–301.   72. Davis, C. R., D. Wibowo, G. H. Fleet, and T. H. Lee. 1988. Properties of wine lactic acid bacteria: their potential enological significance. Am. J. Enol. Viticult. 39:137–142.   73. Davis, C. R., D. Wibowo, T. H. Lee, and G. H. Fleet. 1986. Growth and metabolism of lactic acid bacteria during and after malolactic fermentation of wines at different pH. Appl. Environ. Microbiol. 51:539–545.   74. Degré, R. 1993. Selection and cultivation of wine yeast and bacteria, p. 421–447. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland.

939   75. de las Rivas, B., A. Marcobal, and R. Muñoz. 2004. Allelic diversity and population structure in Oenococcus oeni as determined from sequence analysis of housekeeping genes. Appl. Environ. Microbiol. 70:7210–7219.   76. de las Rivas, B., H. Rodríguez, J. A. Curiel, J. M. Landete, and R. Muñoz. 2009. Molecular screening of wine lactic acid bacteria degrading hydroxycinnamic acids. J. Agric. Food Chem. 57:490–494.   77. Demuyter, C., M. Lollier, J-L. Legras, and C. Le Jeune. 2004. Predominance of Saccharomyces uvarum during spontaneous alcoholic fermentation, for three consecutive years, in an Alsatian winery. J. Appl. Microbiol. 97:1140–1148.   78. Deppenmeier, U., M. Hoffmeister, and C. Prost. 2002. Biochemistry and biotechnological applications of Gluconobacter strains. Appl. Microbiol. Biotechnol. 60:233–242.   79. Dequin, S., J. M. Salmon, H.-V. Nguyen, and B. Blondin. 2003. Wine yeasts, p. 389–411. In T. Boekhout and V. Robert (ed.), Yeasts in Foods: Beneficial and Detrimental Aspects, Behr’s-Verlag, Hamburg, Germany.   80. de Revel, G., A. Bloem, M. Augustin, A. Lonvaud-Funel, and A. Bertrand. 2005. Interaction of Oenococcus oeni and oak wood compounds. Food Microbiol. 22:569–575.   81. de Revel, G., N. Martin, L. Pripis-Nicolau, A. LonvaudFunel, and A. Bertrand. 1999. Contribution to the knowledge of malolactic fermentation influence on wine aroma. J. Agric. Food Chem. 47:4003–4008.   82. Dewey, F. M., M. Hill, and R. DeScenzo. 2008. Quantification of Botrytis and laccase in winegrapes. Am. J. Enol. Viticult. 59:47–54.   83. Dicks, L. M. T., F. Dellaglio, and M. D. Collins. 1995. Proposal to reclassify Leuconostoc oenos as Oenococcus oeni (corrig.) gen. nov. Int. J. Syst. Bacteriol. 45:395–397.   84. D’Inecco, N., E. J. Bartowsky, S. Kassara, A. Lante, P. Spettoli, and P. A. Henschke. 2004. Release of glycosidically bound flavour compounds of Chardonnay by Oenococcus oeni during malolactic fermentation. Food Microbiol. 21:257–265.   85. Dittrich, H. H. 1995. Wine and brandy, p. 464–503. In G. Reed and T. W. Nagodawithana (ed.), Biotechnology, 2nd ed., vol. 9. Enzymes, Biomass, Food and Feed. VCH, Weinheim, Germany.   86. Diviès, C. 1993. Bioreactor technology and wine fermentation, p. 449–475. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland.   87. Divol, B., and A. Lonvaud-Funel. 2005. Evidence for viable but non-culturable yeasts in botrytis-affected wine. J. Appl. Microbiol. 99:85–93.   88. Doneche, B. 1993. Botrytized wines, p. 327–351. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland.   89. Drysdale, G. S., and G. H. Fleet. 1988. Acetic acid bacteria in winemaking—a review. Am. J. Enol. Viticult. 39:143–154.   90. Drysdale, G. S., and G. H. Fleet. 1989. The effect of acetic acid bacteria upon the growth and metabolism of

940 yeasts during the fermentation of grape juice. J. Appl. Bacteriol. 67:471–481.   91. Dubourdieu, D., T. Tominaga, I. Masneuf, C. Peyrot des Gachons, and M. L. Murat. 2006. The role of yeasts in grape flavor development during fermentation: the example of Sauvignon blanc. Am. J. Enol. Viticult. 57:81–88.   92. du Plessis, H. W., L. M. T. Dicks, I. S. Pretorius, M. G. Lambrechts, and M. du Toit. 2004. Identification of lactic acid bacteria isolated from South African brandy base wines. Int. J. Food Microbiol. 91:19–29.   93. du Toit, M., and I. S. Pretorius. 2000. Microbial spoilage and preservation of wine: using weapons from nature’s own arsenal—a review. S. Afr. J. Enol. Viticult. 21:74–96.   94. du Toit, W. J., and M. G. Lambrechts. 2002. The enumeration and identification of acetic acid bacteria from South African red wine fermentations. Int. J. Food Microbiol. 74:57–64.   95. du Toit, W. J., I. S. Pretorius, and A. Lonvaud-Funel. 2005. The effect of sulphur dioxide and oxygen on the viability and culturability of a strain of Acetobacter pasteurianus and a strain of Brettanomyces bruxellensis isolated from wine. J. Appl. Microbiol. 98:862–871.   96. Egli, C. M., W. D. Edinger, C. M. Mitrakul, and T. Henick-Kling. 1998. Dynamics of indigenous and inoculated yeast populations and their effect on the sensory character of Riesling and Chardonnay wines. J. Appl. Microbiol. 85:779–789.   97. Emmet, R. W., A. R. Harris, R. H. Taylor, and J. K. McGechan. 1988. Grape diseases and vineyard protection, p. 232–278. In B. G. Coombe and P. R. Dry (ed.), Viticulture, vol. 2. Winetitles, Adelaide, Australia.   98. Erten, H. 2002. Relations between elevated temperatures and fermentative behaviour of Kloeckera apiculata and Saccharomyces cerevisiae associated with winemaking in mixed cultures. World J. Microbiol. Biotechnol. 18:373–378.   99. Estevez, P., M. Luisa Gil, and E. Falque. 2004. Effects of seven yeast strains on the volatile composition of Palomino wines. Int. J. Food Sci. Technol. 39:61–69. 100. Esteve-Zarzoso, B., M. J. Peris-Torán, E. GarcíaMaiquez, F. Uruburu, and A. Querol. 2001. Yeast population dynamics during the fermentation and biological aging of sherry wines. Appl. Environ. Microbiol. 67:2056–2061. 101. Ewart, A. 2003. White wines, p. 89–106. In A. G. H. Lea and J. R. Piggott (ed.), Fermented Beverage Production, 2nd ed. Kluwer Academic, New York, NY. 102. Faia, A. M., and F. Radler. 1990. Investigation of the bactericidal effect of nisin on lactic acid bacteria of wine. Vitis 29:233–238. 103. Farias, M. E., and M. C. Manca de Nadra. 2000. Purification and partial characterization of Oenococcus oeni protease. FEMS Microbiol. Lett. 185:263–266. 104. Fernandez-Gonzalez, M., R. di Stefano, and A. Briones. 2003. Hydrolysis and transformation of terpene glycosides from Muscat must by different yeast species. Food Microbiol. 20:35–41.

Fermentations and Beneficial Microorganisms 105. Feuillat, M. 2003. Yeast macromolecules: origin, composition and enological interest. Am. J. Enol. Viticult. 54:211–213. 106. Fia, G., G. Giovani, and I. Rosi. 2005. Study of b-glucosidase production by wine-related yeasts during alcoholic fermentation. A new rapid fluorimetric method to determine enzymatic activity. J. Appl. Microbiol. 99:509–517. 107. Fleet, G. H. 1992. Spoilage yeasts. Crit. Rev. Biotechnol. 12:1–44. 108. Fleet, G. H. 1993. The microorganisms of winemaking— isolation, enumeration and identification, p. 1–26. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 109. Fleet, G. H. 1999. Microorganisms in food ecosystems. Int. J. Food Microbiol. 50:101–117. 110. Fleet, G. H. 2003. Yeast interactions and wine flavour. Int. J. Food Microbiol. 86:11–22. 111. Fleet, G. H. 2003. Yeasts in fruit and fruit products, p. 267–288. In T. Boeckhout and V. Robert (ed.), Yeasts in Food: Beneficial and Detrimental Aspects. Behr’sVerlag, Hamburg, Germany. 112. Fleet, G. H. 2008. Wine yeasts for the future. FEMS Yeast Res. 8:979–995. 113. Fleet, G. H., and G. M. Heard. 1993. Yeasts—growth during fermentation, p. 27–54. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 114. Fleet, G. H., S. Lafon-Lafourcade, and P. RibereauGayon. 1984. Evolution of yeasts and lactic acid bacteria during fermentation and storage of Bordeaux wines. Appl. Environ. Microbiol. 48:1034–1038. 115. Fleet, G. H., C. Prakitchaiwattana, A. L. Beh, and G. Heard. 2002. The yeast ecology of wine grapes, p. 1–17. In M. Ciani (ed.), Biodiversity and Biotechnology of Wine Yeasts. Research Signpost, Kerala, India. 116. Francis, J. L., and J. L. Newton. 2005. Determining wine aroma from compositional data. Aust. J. Grape Wine Res. 11:114–126. 117. Fredericks, I. N., M. du Toit, and M. Krügel. 2011. Efficacy of ultraviolet radiation as an alternative technology to inactivate microorganisms in grape juices and wines. Food Microbiol. 28:510–517. 118. Fugelsang, K. C., and C. G. Edwards. 2010. Wine Microbiology: Practical Applications and Procedures, 2nd ed. Springer, New York, NY. 119. G-Alegría, E., I. López, J. I. Ruiz, J. Sáenz, E. Fernández, M. Zarazaga, M. Dizy, C. Torres, and F. Ruiz-Larrea. 2004. High tolerance of wild Lactobacillus plantarum and Oenococcus oeni strains to lyophilisation and stress environmental conditions of acid pH and ethanol. FEMS Microbiol. Lett. 230:53–61. 120. Gambuti, A., D. Strollo, A. Genovese, M. Ugliano, A. Ritieni, and L. Moio. 2005. Influence of enological practices on ochratoxin A concentration in wine. Am. J. Enol. Viticult. 56:155–162. 121. Ganga, M. A., and C. Martínez. 2004. Effect of wine yeast monoculture practice in the biodiversity of nonSaccharomyces yeasts. J. Appl. Microbiol. 96:76–83.

37.  Wine 122. Gao, C., and G. H. Fleet. 1988. The effects of temperature and pH on the ethanol tolerance of the wine yeasts, Saccharomyces cerevisiae, Candida stellata and Kloeckera apiculata. J. Appl. Bacteriol. 65:405–410. 123. Gao, C., and G. H. Fleet. 1995. Cell-recycle membrane bioreactor for conducting continuous malolactic fermentation. Aust. J. Grape Wine Res. 1:32–38. 124. Gao, C., and G. H. Fleet. 1995. Degradation of malic and tartaric acids by high density cell suspensions of wine yeasts. Food Microbiol. 12:65–71. 125. Gao, Y. C., G. Zhang, S. Krentz, S. Darius, J. Power, and G. Lagarde. 2002. Inhibition of spoilage lactic acid bacteria by lysozyme during wine alcoholic fermentation. Aust. J. Grape Wine Res. 8:76–83. 126. García-Ruiz, A., E. M. González-Rompinelli, B. Bartolomé, and M. V. Moreno-Arribas. 2011. Potential of wine-associated lactic acid bacteria to degrade biogenic amines. Int. J. Food Microbiol. 148:115–120. 127. Gerbaux, V., C. Briffox, A. Dumont, and S. Krieger. 2009. Influence of inoculation with malolactic bacteria on volatile phenols in wines. Am. J. Enol. Viticult. 60:233–235. 128. Gockowiak, H., and P. A. Henschke. 2003. Interaction of pH, ethanol concentration and wine matrix on induction of malolactic fermentation with commercial “direct inoculation” starter cultures. Aust. J. Grape Wine Res. 9:200–209. 129. Goldberg, D. M. and G. J. Soleas. 2011. Wine and health: a paradigm for alcohol and antioxidants. J. Med. Biochem. 30:93–102. 130. Gonzalez, A., N. Hierro, M. Poblet, A. Mas, and J. M. Guillamon. 2005. Application of molecular methods to demonstrate species and strain evolution of acetic acid bacteria population during wine production. Int. J. Food Microbiol. 102:295–304. 131. Gonzalez, A., H. M. Poblet, N. Rozes, A. Mas, and J. M. Guillamon. 2004. Application of molecular methods for the differentiation of acetic acid bacteria in a red wine fermentation. J. Appl. Microbiol. 96:853–860. 132. Goriushkina, T. B., A. P. Soldatkin, and S. V. Dzyadevych. 2009. Application of amperometric biosensors for analysis of ethanol, glucose and lactate in wine. J. Agric. Food Chem. 57:6528–6535. 133. Goswell, R. W., and R. E. Kunkee. 1977. Fortified wines, p. 478–533. In A. H. Rose (ed.), Economic Microbiology, vol. 1. Academic Press, London, United Kingdom. 134. Granchi, L., M. Bosco, A. Messini, and M. Vincenzini. 1999. Rapid detection and quantification of yeast species during spontaneous wine fermentation by PCR-RFLP analysis of the rDNA ITS region. J. Appl. Microbiol. 87:949–956. 135. Granchi, L. D. Gannucci, C. Viti, L. Giovannetti, and M. Vincenzini. 2003. Saccharomyces cerevisiae biodiversity in spontaneous commercial fermentations of grape musts with “adequate” and “inadequate” assimilable-nitrogen content. Lett. Appl. Microbiol. 36:54–58.

941 136. Grimaldi, A., E. Bartowsky, and V. Jiranek. 2005. A survey of glycosidase activities of commercial wine strains of Oenococcus oeni. Int. J. Food Microbiol. 105:233–244. 137. Guerrini, S., A. Bastianini, G. Blaiotta, L. Granchi, G. Moschetti, S. Coppola, P. Romano, and M. Vincenzini. 2003. Phenotypic and genotypic characterization of Oenococcus oeni strains isolated from Italian wines. Int. J. Food Microbiol. 83:1–14. 138. Guerrini, S., S. Mangani, L. Granchi, and M. Vincenzini. 2002. Biogenic amine production by Oenococcus oeni. Curr. Microbiol. 44:374–378. 139. Guilloux-Benatier, M., M. O. Pageault, A. Man, and M. Feuillat. 2000. Lysis of yeast cells by Oenococcus oeni enzymes. J. Ind. Microbiol. Biotechnol. 25:193–197. 140. Gutiérrez, A. R., S. Epifanio, P. Garijo, P. Lopez, and P. Santamaria. 2001. Killer yeasts: incidence in the ecology of spontaneous fermentation. Am. J. Enol. Viticult. 52:352–356. 141. Gutiérrez, A. R., P. Santamaria, S. Epifania, P. Garijo, and R. Lopez. 1999. Ecology of spontaneous fermentation in one winery during 5 consecutive years. Lett. Appl. Microbiol. 29:411–415. 142. Guzzo, J., M. P. Jobin, F. Delmas, L. C. Fortier, D. Garmyn, R. Tourdot-Maréchal, B. Lee, and C. Diviès. 2000. Regulation of stress response in Oenococcus oeni as a function of environmental changes and growth phase. Int. J. Food Microbiol. 55:27–31. 143. Hansen, E. H., P. Nissen, P. Sommer, J. C. Nielsen, and N. Arneborg. 2001. The effect of oxygen on the survival of non-Saccharomyces yeasts during mixed culture fermentations of grape juice with Saccharomyces cerevisiae. J. Appl. Microbiol. 91:541–547. 144. Hayashi, H., R. Arai, S. Tada, H. Taguchi, and Y. Ogawa. 2007. Detection and identification of Brettanomyces/Dekkera sp. yeasts with a loop-mediated isothermal amplification method. Food Microbiol. 24:778–785. 145. Heard, G. M. 1999. Novel yeasts in winemaking—looking to the future. Food Aust. 51:347–352. 146. Heard, G. M., and G. H. Fleet. 1985. Growth of natural yeast flora during the fermentation of inoculated wines. Appl. Environ. Microbiol. 50:727–728. 147. Heard, G. M., and G. H. Fleet. 1988. The effects of temperature and pH on the growth of yeast species during the fermentation of grape juice. J. Appl. Bacteriol. 65:23–28. 148. Henick-Kling, T. 1993. Malolactic fermentation, p. 289– 326. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 149. Henschke, P. A. 1997. Wine yeast, p. 527–560. In F. K. Zimmermann and K. D. Entian (ed.), Yeast Sugar Metabolism. Technomic Publishing Co., Lancaster, PA. 150. Henschke, P., and V. Jiranek. 1993. Yeasts—metabolism of nitrogen compounds, p. 77–164. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland.

942 151. Hernawan, T., and G. H. Fleet. 1995. Chemical and cytological changes during the autolysis of yeasts. J. Ind. Microbiol. 14:440–450. 152. Hocking, A. D., P. Varelis, J. I. Pitt, S. F. Cameron, and S. L. Leong. 2003. Occurrence of ochratoxin A in Australian wine. Aust. J. Grape Wine Res. 9:72–78. 153. Howe, P. 2003. Sparkling wines, p. 139–156. In A. G. H. Lea and J. R. Piggott (ed.), Fermented Beverage Production, 2nd ed. Kluwer Academic, New York, NY. 154. Howell, K. S., D. Cozzolino, E. J. Bartowsky, G. H. Fleet, and P. A. Henschke. 2006. Metabolic profiling as a tool for revealing Saccharomyces interactions during wine fermentation. FEMS Yeast Res. 6:91–101. 155. Howell, K. S., M. Klein, J. H. Swiegers, Y. Hayasaka, G. M. Elsey, G. H. Fleet, P. B. Høj, I. S. Pretorius, and M. A. de Barros Lopes. 2005. Genetic determin­a­nts of volatile-thiol release by Saccharomyces cerevisiae during wine fermentation. Appl. Environ. Microbiol. 71:5420–5426. 156. Howell, K. S., J. H. Swiegers, G. M. Elsey, T. E. Siebert, E. J. Bartowsky, G. H. Fleet, I. S. Pretorius, and M. A. de Barros Lopes. 2004. Variation in 4-mercapto4-methyl-pentan-2-one release by Saccharomyces cerevisiae commercial wine strains. FEMS Microbiol. Lett. 240:125–129. 157. Izquierdo-Cañas, P. M., E. García-Romero, B. HuertasNebreda, and S. Gómez-Alonso. 2011. Colloidal silver complex as an alternative to sulphur dioxide in winemaking. Food Control 23:73–81. 158. Jara, C., E. Mateo, J. M. Guillamón, M. J. Torija, and A. Mas. 2008. Analysis of several methods for the extraction of high quality DNA from acetic acid bacteria in wine and vinegar for characterization by PCR-based methods. Int. J. Food Microbiol. 128:336–341. 159. Jensen, S. L., N. L. Umiker, N. Arneborg, and C. G. Edwards. 2009. Identification and characterization of Dekkera bruxellensis, Candida pararugosa, and Pichia guilliermondii isolated from commercial red wines. Food Microbiol. 26:915–921. 160. Jolly, N. P., O. P. H. Augustyn, and I. S. Pretorius. 2003. The occurrence of non-Saccharomyces cerevisiae yeast species over three vintages in four vineyards and grape must from four production regions of the Western Cape, South Africa. S. Afr. J. Enol. Viticult. 24:35–42. 161. Karpel, J. E., W. R. Place, and L. F. Bisson. 2008. Analysis of the major hexose transporter genes in wine strains of Saccharomyces cerevisiae. Am. J. Enol. Viticult. 59:265–275. 162. King, E. S., R. L. Kievit, C. Curtin, J. H. Swiegers, I. S. Pretorius, S. E. P. Bastian, and I. L. Francis. 2010. The effect of multiple yeasts co-inoculations on Sauvignon Blanc wine aroma composition, sensory properties and consumer preference. Food Chem. 122:618–626. 163. Kourkoutas, Y., A. Bekatorou, I. A. Banat, R. Marchant, and A. A. Koutinas. 2004. Immobilization technologies and support materials suitable in alcohol beverages production—a review. Food Microbiol. 21:377–397. 164. Kumar, G. R., R. Goyashiki, V. Ramakrishnan, J. E. Karpel, and L. F. Bisson. 2008. Genes required for etha-

Fermentations and Beneficial Microorganisms

165.

166.

167. 168. 169.

170.

171.

172. 173.

174.

175. 176. 177. 178.

179. 180.

nol tolerance and utilization in Saccharomyces cerevisiae. Am. J. Enol. Viticult. 59:401–411. Kunkee, R. E., and L. Bisson. 1993. Wine-making yeasts, p. 69–128. In A. H. Rose and J. S. Harrison (ed.), The Yeasts, 2nd ed., vol. 5. Yeast Technology. Academic Press, London, United Kingdom. Kurtzman, C. P. 2003. Phylogenetic circumscription of Saccharomyces, Kluyveromyces and other members of Saccharomycetaceae and the proposal of the new genera Lachancea, Nakaseomyces, Vanderwaltozyma and Zygotorulaspora. FEMS Yeast Res. 4:233–245. Lambrechts, M. G., and I. S. Pretorius. 2000. Yeast and its importance in wine aroma—a review. S. Afr. J. Enol. Viticult. 21:97–129. Landete, J. M., S. Ferrer, and I. Pardo. 2007. Biogenic amine production by lactic acid bacteria, acetic bacteria and yeast isolated from wine. Food Control 18:1569–1574. Landete, J. M., B. de las Rivas, A. Marcobal, and R. Muñoz. 2011. PCR methods for the detection of biogenic amine-producing bacteria on wine. Ann. Microbiol. 61:159–166. Lee, J., and K. L. Steenwerth. 2011. Rootstock and vineyard floor management influence on ‘Cabernet Sauvignon’ grape yeast assimilable nitrogen (YAN). Food Chem. 127:926–933. Lee, T. H., and R. F. Simpson. 1993. Microbiology and chemistry of cork taints in wine, p. 353–372. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. Liu, S. Q. 2002. Malolactic fermentation in wine—beyond deacidificiation. J. Appl. Microbiol. 92:589–601. Logrieco, A. F., R. Ferracane, G. Cozzi, M. Haidukowsky, A. Susca, G. Mulè, and A. Ritieni. 2011. Fumonisin B2 by Aspergillus niger in the grape–wine chain: an additional potential mycotoxicological risk. Ann. Microbiol. 61:1–3. Logrieco, A., R. Ferracane, M. Haidukowsky, G. Cozzi, A. Visconti, and A. Ritieni. 2009. Fumonisin B2 production by Aspergillus niger from grapes and natural occurrence in must. Food Addit. Contam. 26:1495–1500. Logrieco, A., R. Ferracane, A. Visconti, and A. Ritieni. 2010. Natural occurrence of fumonisin B2 in red wine from Italy. Food Addit. Contam. 27:1136–1141. Lonvaud-Funel, A. 1999. Lactic acid bacteria in the quality improvement and depreciation of wine. Antonie van Leeuwenhoek 76:317–331. Lonvaud-Funel, A. 2001. Biogenic amines in wines: role of lactic acid bacteria. FEMS Microbiol. Lett. 199:9–13. Lopardo, R., M. Musto, F. Cellini, and P. D. Grieco. 2011. Development and application of a duplex PCR for the detection of Aspergillus carbonarius occurring in grapes. Ann. Microbiol. 61:5–9. Loureiro, V., and M. Malfeito-Ferreira. 2003. Spoilage yeasts in the wine industry. Int. J. Food Microbiol. 86:23–50. Lustrato, G., I. Vigentini, A. De Leonardis, G. Alfano, A. Tirelli, R. Foschino, and G. Ranalli. 2010. Inactivation of wine spoilage yeasts Dekkera bruxellensis using low

37.  Wine

181.

182.

183. 184.

185.

186.

187.

188.

189.

190. 191.

192.

193.

194.

electric current treatment (LEC). J. Appl. Microbiol. 109:594–604. Magnoli, C., M. Violante, M. Combina, G. Palacio, and A. Dalcero. 2003. Mycoflora and ochratoxin-producing strains of Aspergillus section Nigri in wine grapes in Argentina. Lett. Appl. Microbiol. 37:179–184. Magyar, I., and T. Tóth. 2011. Comparative evaluation of some oenological properties in wine strains of Candida stellata, Candida zemplinina, Saccharomyces uvarum and Saccharomyces cerevisiae. Food Microbiol. 28:94–100. Maicas, S. 2001. The use of alternative technologies to develop malolactic fermentation in wine. Appl. Microbiol. Biotechnol. 56:35–39. Mansfield, A. K., B. Zoecklein, and R. S. Whiton. 2002. Quantification of glycosidase activity in selected strains of Brettanomyces bruxellensis and Oenococcus oeni. Am. J. Enol. Viticult. 53:303–307. Maqueda, M., E. Zamora, N. Rodríguez-Cousiño, and M. Ramírez. 2010. Wine yeast molecular typing using a simplified method for simultaneously extracting mtDNA, nuclear DNA and virus dsRNA. Food Microbiol. 27:205–209. Marks, V. D., G. K. van der Merwe, and H. J. J. van Vuuren. 2003. Transcriptional profiling of wine yeast in fermenting grape juice: regulatory effect of diammonium phosphate. FEMS Yeast Res. 3:269–287. Martinez, P., M. J. Valcarcel, L. Perez, and T. Benitez. 1995. Metabolism of Saccharomyces cerevisiae flor yeasts during fermentation and biological ageing of fino sherry: by-products and aroma compounds. Am. J. Enol. Viticult. 49:240–250. Martínez-Rodríguez, A., A. V. Carrascosa, J. M. Barcenilla, M. A. Pozo-Bayon, and M. C. Polo. 2001. Autolytic capacity and foam analysis as additional criteria for the selection of yeast strains for sparkling wine production. Food Microbiol. 18:183–191. Martínez-Rodríguez, A. J., M. C. Polo, and A. V. Carrascosa. 2001. Structural and ultrastructural changes in yeast cells during autolysis in a model wine system and in sparkling wines. Int. J. Food Microbiol. 71:45–51. Martini, A., M. Ciani, and G. Scorzetti. 1996. Direct enumeration and isolation of wine yeasts from grape surfaces. Am. J. Enol. Viticult. 47:435–440. Martorell, P., A. Querol, and M. T. FernándezEspinar. 2005. Rapid identification and enumeration of Saccharomyces cerevisiae cells in wine by real-time PCR. Appl. Environ. Microbiol. 71:6823–6830. Masneuf-Pomarède, I., M. Bely, P. Marullo, A. Lonv­audFunel, and D. Dubourdieu. 2010. Reassessment of phenotypic trains for Saccharomyces bayanus var. uvarum wine yeast strains. Int. J. Food Microbiol. 139:79–86. Matthews, A., A. Grimaldi, M. Walker, E. Bartowsky, P. Grbin, and V. Jiranek. 2004. Lactic acid bacteria as a potential source of enzymes for use in vinification. Appl. Environ. Microbiol. 70:5715–5731. Medina, K., E. Boido, E. Dellacassa, and F. Carrau. 2005. Yeast interactions with anthocyanins during red wine fermentation. Am. J. Enol. Viticult. 56:104–109.

943 195. Mendes-Ferreira, A., A. Mendes-Faia, and C. Leão. 2002. Survey of hydrogen sulphide production by wine yeasts. J. Food Prot. 65:1033–1037. 196. Mendes-Ferreira, A., A. Mendes-Faia, and C. Leão. 2004. Growth and fermentation patterns of Saccharomyces cerevisiae under different ammonium concentrations and its implications in winemaking industry. J. Appl. Microbiol. 97:540–545. 197. Mendoza, L. M., M. C. Manca de Nadra, E. Bru, and M. E. Farías. 2009. Influence of wine-related physicochemical factors on the growth and metabolism of non-Saccharomyces and Saccharomyces yeasts in mixed culture. J. Ind. Microbiol. Biotechnol. 36:229–237. 198. Merico, A., P. Sulo, J. Piškur, and C. Compagno. 2007. Fermentative lifestyle in yeasts belonging to the Saccharomyces complex. FEBS J. 274:976–989. 199. Millet, V., and A. Lonvaud-Funel. 2000. The viable but non-culturable state of wine microorganisms during storage. Lett. Appl. Microbiol. 30:136–141. 200. Mills, D., J. Eric, and L. Cocolin. 2002. Yeast diversity and persistence in Botrytis-affected wine fermentation. Appl. Environ. Microbiol. 68:4884–4893. 201. Mills, D. A., H. Rawsthorne, C. Parker, D. Tamir, and K. Makarova. 2005. Genomic analysis of Oenococcus oeni PSU-1 and its relevance to winemaking. FEMS Microbiol. Rev. 29:465–475. 202. Miranda, M., A. Ramos, M. Veiga-da-Cunha, M. C. Loureiro-Dias, and H. Santos. 1997. Biochemical basis for glucose-induced inhibition of malolactic fermentation in Leuconostoc oenos. J. Bacteriol. 179:5345–5354. 203. Mogensen, J. M., T. O. Larsen, and K. F. Nielsen. 2010. Widespread occurrence of the mycotoxin fumonisin B2 in wine. J. Agric. Food Chem. 58:4853–4857. 204. Mora, J., and A. Mulet. 1991. Effects of some treatments of grape juice on the population and growth of yeast species during fermentation. Am. J. Enol. Viticult. 42:133–136. 205. Moreira, N., P. Guedes de Pinho, C. Santos, and I. Vasconcelos. 2011. Relationship between nitrogen content in grapes and volatiles, namely heavy sulphur compounds, in wines. Food Chem. 126:1599–1607. 206. Moreira, N., F. Mendes, P. Guedes de Pinho, T. Hogg, and I. Vasconcelos. 2008. Heavy sulphur compounds, higher alcohols and esters production profile of Hanseniaspora uvarum and Hanseniaspora guilliermondii grown as pure and mixed cultures in grape must. Int. J. Food Microbiol. 124:231–238. 207. Navarro, L., M. Zarazaga, J. Sáenz, F. Ruiz-Larrea, and C. Torres. 2000. Bacteriocin production by lactic acid bacteria isolated from Rioja wines. J. Appl. Microbiol. 88:44–51. 208. Nehme, N., F. Mathieu, and P. Taillandier. 2010. Impact of the co-culture of Saccharomyces cerevisiae– Oenococcus oeni on malolactic fermentation and partial characterization of a yeast-derived inhibitory peptidic fraction. Food Microbiol. 27:150–157. 209. Nielsen, J. C., C. Prahl, and A. Lonvaud-Funel. 1996. Malolactic fermentation in wine by direct inoculation with freeze-dried Leuconostoc oenos cultures. Am. J. Enol. Viticult. 47:42–48.

944 210. Nissen, P., D. Nielsen, and N. Arneborg. 2003. Viable Saccharomyces cerevisiae cells at high concentrations cause early growth arrest of non-Saccharomyces yeasts in mixed cultures by a cell-cell contact mediated mechanism. Yeast 20:331–341. 211. Oelofsel, A., A. Lonvaud-Funel, and M. du Toit. 2009. Molecular identification of Brettanomyces bruxellensis strains isolated from red wines and volatile phenol production. Food Microbiol. 26:377–385. 212. Oelofsel, A., I. S. Pretorius, and M. du Toit. 2008. Significance of Brettanomyces and Dekkera during winemaking: a synoptic review. S. Afr. J. Enol. Viticult. 29:128–144. 213. Olguín, N., A. Bordons, and C. Reguant. 2009. Influence of ethanol and pH on the gene expression of the citrate pathway in Oenococcus oeni. Food Microbiol. 26:197–203. 214. Önal, A. 2007. A review: current analytical methods for the determination of biogenic amines in foods. Food Chem. 103:1475–1486. 215. Pasteris, S. E., and A. M. Strasser de Saad. 2008. Transport of glycerol by Pediococcus pentosaceus isolated from wine. Food Microbiol. 25:545–549. 216. Peinado, R. A., J. A. Mauricio, and J. Moreno. 2006. Aromatic series in sherry wines with gluconic acid subjected to different biological aging conditions by Saccharomyces cerevisiae var capensis. Food Chem. 94:232–239. 217. Pérez-Nevado, F., H. Albergaria, T. Hogg, and F. Girio. 2006. Cellular death of two non-Saccharomyces wine-related yeasts during mixed fermentations with Saccharomyces cerevisiae. Int. J. Food Microbiol. 108:336–345. 218. Phister, T. G., and D. A. Mills. 2003. Real-time PCR assay for detection and enumeration of Dekkera bruxellensis in wine. Appl. Environ. Microbiol. 69:7430–7434. 219. Phister, T. G., H. Rawsthorne, C. M. L. Joseph, and D. A. Mills. 2007. Real-time PCR assay for detection and enumeration of Hanseniaspora species from wine and juice. Am. J. Enol. Viticult. 58:229–233. 220. Piermarini, S., G. Volpe, M. Esti, M. Simonetti, and G. Palleschi. 2011. Real time monitoring of alcoholic fermentation with low-cost amperometric biosensors. Food Chem. 127:749–754. 221. Pimentel, M. S., M. H. Silva, I. Cortes, and A. M. Faia. 1994. Growth and metabolism of sugar and acids of Leuconostoc oenos under different conditions of temperature and pH. J. Appl. Bacteriol. 76:42–48. 222. Pinzani, P., L. Bonciani, M. Pazzagli, C. Orlando, S. Guerrini, and L. Granchi. 2004. Rapid detection of Oenococcus oeni in wine by real-time quantitative PCR. Lett. Appl. Microbiol. 38:118–124. 223. Plata, C., C. Millan, J. C. Mauricio, and J. M. Ortega. 2003. Formation of ethyl acetate and isoamyl acetate by various species of wine yeasts. Food Microbiol. 20:217–224. 224. Poblet-Icart, M., A. Bordons, and A. Lonvaud-Funel. 1998. Lysogeny of Oenococcus oeni (syn. Leuconostoc oenos) and study of their induced bacteriophages. Curr. Microbiol. 36:365–369.

Fermentations and Beneficial Microorganisms 225. Pohve-Jemec, K., N. Cadez, T. Zagorc, V. Bubic, A. Zupec, and P. Raspor. 2001. Yeast population dynamics in five spontaneous fermentations of Malvasia must. Food Microbiol. 18:247–259. 226. Pohve-Jemec, K., and P. Raspor. 2005. Initial Saccharomyces cerevisiae concentration in single or composite cultures dictates bioprocess kinetics. Food Microbiol. 22:293–300. 227. Prakitchaiwattana, C. J., G. H. Fleet, and G. M. Heard. 2004. Application and evaluation of denaturing gradient gel electrophoresis to analyse the yeast ecology of wine grapes. FEMS Yeast Res. 4:865–877. 228. Pramateftaki, P. V., P. Lanaridis, and M. A. Typas. 2000. Molecular identification of wine yeasts at species or strain level: a case study with strains from two vine-growing areas of Greece. J. Appl. Microbiol. 89:236–248. 229. Presa-Owens, C. D. L., P. Schlich, H. D. Davies, and A. C. Noble. 1998. Effect of Methode Champenoise process on aroma of four V. vinifera varieties. Am. J. Enol. Viticult. 49:289–294. 230. Pretorius, I. 2000. Tailoring wine yeast for the new millennium: novel approaches to the ancient art of winemaking. Yeast 16:675–729. 231. Pretorius, I. S., and E. F. Bauer. 2002. Meeting the consumer challenge through genetically customized wine yeast strains. Trends Biotechnol. 20:426–432. 232. Pretorius, I. S., and P. B. Høj. 2005. Grape and wine biotechnology: challenges, opportunities and potential benefits. Aust. J. Grape Wine Res. 11:83–108. 233. Pretorius, I. S., T. J. van der Westhuizen, and O. P. H. Augustyn. 1999. Yeast biodiversity in vineyards and wineries and its importance to the South African wine industry. A review. S. Afr. J. Enol. Viticult. 20:61–76. 234. Pripis-Nicolau, L., G. de Revel, A. Bertrand, and A. Lonvaud-Funel. 2004. Methionine catabolism and production of volatile sulphur compounds by Oenococcus oeni. J. Appl. Microbiol. 96:1176–1184. 235. Pueyo, E., A. Martínez-Rodríguez, M. Polo, G. SantaMaría, and B. Bartolomé. 2000. Release of lipids during yeast autolysis in a model wine system. J. Agric. Food Chem. 48:116–122. 236. Querol, A., E. Barrio, T. Huerta, and D. Ramon. 1992. Molecular monitoring of wine fermentations conducted by active dry yeast strains. Appl. Environ. Microbiol. 58:2948–2953. 237. Querol, A., M. T. Fernández-Espinar, M. L. del Olmo, and E. Barrio. 2003. Adaptive evolution of wine yeast. Int. J. Food Microbiol. 86:3–10. 238. Radler, F. 1993. Yeasts—metabolism of organic acids, p. 165–182. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 239. Radoi, F., M. Kishida, and H. Kawasaki. 2005. Endopolygalacturonase in Saccharomyces wine yeasts: effect of carbon source on enzyme production. FEMS Yeast Res. 5:663–668. 239a.Rankine, B. L. 1989. Making Good Wine. A Manual of Winemaking Practices for Australia and New Zealand. Sun Books, Melbourne, Australia.

37.  Wine 240. Rauhut, D. 1993. Yeasts—production of sulfur compounds, p. 183–223. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 241. Reader, H. P., and M. Dominguez. 2003. Fortified wines: sherry, port and madeira, p. 157–194. In A. G. H. Lea and J. R. Piggott (ed.), Fermented Beverage Production. Kluwer Academic, New York, NY. 242. Redzepović, S., S. Orlić, S. Sikora, A. Majdak, and I. S. Pretorius. 2002. Identification and characterization of Saccharomyces cerevisiae and Saccharomyces paradoxus strains isolated from Croatian vineyards. Lett. Appl. Microbiol. 35:305–310. 243. Reguant, C., R. Carrete, N. Ferrer, and A. Bordons. 2005. Molecular analysis of Oenococcus oeni population dynamics and the affect of aeration and temperature during alcoholic fermentation on malolactic fermentation. Int. J. Food Sci. Technol. 40:451–459. 244. Regueiro, L. A., C. L. Costas, and J. E. L. Rubio. 1993. Influence of viticultural and enological practices on the development of yeast populations during winemaking. Am. J. Enol. Viticult. 44:405–408. 245. Renouf, V., M.-C. Perello, G. de Revel, and A. LonvaudFunel. 2007. Survival of wine microorganisms in the bottle during storage. Am. J. Enol. Viticult. 58:379–386. 246. Renouf, V., P. Strehaiano, and A. Lonvaud-Funel. 2008. Effectiveness of dimethyldicarbonate to prevent Brettanomyces bruxellensis growth in wine. Food Control 19:208–216. 247. Ribéreau-Gayon, P., D. Dubourdieu, B. Donéche, and A. Lonvaud. 2000. Handbook of Enology, vol. 1. The Microbiology of Wine and Vinifications. John Wiley & Sons, Chichester, United Kingdom. 248. Richter, H., I. Hamann, and G. Unden. 2003. Use of the mannitol pathway in fructose fermentation of Oenococcus oeni due to limiting redox regeneration capacity of the ethanol pathway. Arch. Microbiol. 179:227–233. 249. Rodríguez, M. E., C. A. Lopes, M. van Broock, S. Valles, D. Ramón, and A. C. Caballero. 2004. Screening and typing of Patagonian wine yeasts for glycosidase activities. J. Appl. Microbiol. 96:84–95. 250. Romano, P., C. Fiore, M. Paraggio, M. Caruso, and A. Capece. 2003. Function of yeast species and strains in wine flavour. Int. J. Food Microbiol. 86:169–180. 251. Romano, P., and G. Suzzi. 1993. Sulfur dioxide and wine microorganisms, p. 373–394. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 252. Rosi, I., G. Fia, and V. Canuti. 2003. Influence of pH values and inoculation time on the growth and malolactic activity of a strain of Oenococci oeni. Aust. J. Grape Wine Res. 9:194–199. 253. Rossignol, T., L. Dulau, A. Julien, and B. Blondin. 2003. Genome-wide monitoring of wine yeast gene expression during alcoholic fermentation. Yeast 20:1369–1385. 254. Sabate, J., J. Cano, A. Querol, and J. M. Guillamon. 1998. Diversity of Saccharomyces strains in wine fermentation: analysis for two consecutive years. Lett. Appl. Microbiol. 26:452–455.

945 255. Salaha, M.-I., S. Kallithraka, I. Marmaras, E. Koussissi, and I. Tzourou. 2008. A natural alternative to sulphur dioxide for red wine production: influence on colour, antioxidant activity and anthocyanin content. J. Food Comp. Anal. 21:660–666. 256. Santamaría, P., P. Garijo, R. López, C. Tenorio, and A. R. Gutiérrez. 2005. Analysis of yeast population during spontaneous alcoholic fermentation: effect of age of the cellar and practice of inoculation. Int. J. Food Microbiol. 103:49–56. 257. Schoeman, H., G. M. Wolfaardt, A. Botha, P. van Rensburg, and I. S. Pretorius. 2009. Establishing a risk-assessment process for release of genetically modified wine yeast into the environment. Can. J. Microbiol. 55:990–1002. 258. Schuller, D., E. Valero, S. Dequin, and M. Casal. 2003. Survey of molecular methods for typing wine yeast strains. FEMS Microbiol. Lett. 231:19–26. 259. Schutz, M., and J. Gafner. 1994. Dynamics of the yeast strain population during spontaneous alcoholic fermentation determined by CHEF gel electrophoresis. Lett. Appl. Microbiol. 19:253–259. 260. Semon, M. J., C. G. Edwards, D. Forsyth, and C. O. Dinn. 2001. Inducing malolactic fermentation in Chardonnay musts and wines using different strains of Oenococcus oeni. Aust. J. Grape Wine Res. 7:52–59. 261. Serra, R., L. Abrunhosa, Z. Kozakiewicz, and A. Venancio. 2003. Black Aspergillus species as ochratoxin A producers in Portuguese wine grapes. Int. J. Food Microbiol. 88:63–68. 262. Serra, R., A. Braga, and A. Venancio. 2005. Mycotoxinproducing and other fungi isolated from grapes for wine production, with particular emphasis on ochratoxin A. Res. Microbiol. 156:515–521. 263. Shimazu, Y., and M. Watanabe. 1981. Effects of yeast strains and environmental conditions on formation of organic acids in must during fermentation. J. Ferment. Technol. 59:27–32. 264. Shimizu, K. 1993. Killer yeasts, p. 243–264. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 265. Silva, S., F. Ramon Portugal, P. Andrade, M. de Fatima Texeira, and P. Strehaiano. 2003. Malic acid consumption by dry immobilized cells of Schizosaccharomyces pombe. Am. J. Enol. Viticult. 54:50–55. 266. Silva Pereira, C., J. J. Figueiredo Marques, and M. V. San Ramão. 2000. Cork taint in wine: scientific knowledge and public perception—a critical review. Crit. Rev. Microbiol. 26:147–162. 267. Sipiczki, M. 2002. Taxonomic and physiological diversity of Saccharomyces bayanus, p. 53–69. In M. Ciani (ed.), Biodiversity and Biotechnology of Wine Yeasts. Research Signpost, Kerala, India. 268. Sipiczki, M., P. Romano, A. Capece, and M. Parragio. 2004. Genetic segregation of natural Saccharomyces cerevisiae strains derived from spontaneous fermentation of Aglianico wine. J. Appl. Microbiol. 96:1169–1175. 269. Smith, J. L., P. M. Fratamico, and J. S. Novak. 2004. Quorum sensing: a primer for food microbiologists. J. Food Prot. 67:1053–1070.

946 270. Soden, A., I. L. Francis, H. Oakey, and P. A. Henschke. 2000. Effects of co-fermentation with Candida stellata and Saccharomyces cerevisiae on the aroma and composition of Chardonnay wine. Aust. J. Grape Wine Res. 6:21–30. 271. Sokollek, S. J., C. Hertel, and W. P. Hammes. 1998. Cultivation and preservation of vinegar bacteria. J. Biotechnol. 60:195–206. 272. Spiropoulos, A., J. Tanaka, I. Flerianos, and L. F. Bisson. 2000. Characterisation of hydrogen sulfide fermentation in commercial and natural wine isolates of Saccharomyces. Am. J. Enol. Viticult. 51:233–248. 273. Sponholz, W. R. 1993. Wine spoilage by microorganisms, p 395–420. In G. H. Fleet (ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland. 274. Straus, M. L. A., N. P. Jolly, M. G. Lambrechts, and P. van Rensburg. 2001. Screening for the production of extracellular hydrolytic enzymes by non-Saccharomyces wine yeasts. J. Appl. Microbiol. 91:182–190. 275. Stumner, B. E., I. L. Francis, T. Zanker, K. A. Lattey, and E. S. Scott. 2005. Effects of powdery mildew on the sensory properties and composition of Chardonnay juice and wine when grape sugar ripeness is standardized. Aust. J. Grape Wine Res. 11:66–76. 276. Suárez, R., J. A. Suárez-Lepe, A. Morata, and F. Calderón. 2007. The production of ethylphenols in wine by yeasts of the genera Brettanomyces and Dekkera: a review. Food Chem. 102:10–21. 277. Swiegers, J. H., E. J. Bartowsky, P. A. Henschke, and I. S. Pretorius. 2005. Yeast and bacterial modulation of wine aroma and flavour. Aust. J. Grape Wine Res. 11:139–173. 278. Swiegers, J. H., R. L. Kievit, T. Siebert, K. A. Lattey, B. R. Bramley, I. L. Francis, E. S. King, and I. S. Pretorius. 2009. The influence of yeast on the aroma of Sauvignon blanc wine. Food Microbiol. 26:204–211. 279. Taillandier, P., F. P. Portugal, A. Fuster, and P. Strehaiano. 2007. Effect of ammonium concentration on alcoholic fermentation kinetics by wine yeasts for high sugar content. Food Microbiol. 24:95–100. 280. Tessonnière, H., S. Vidal, L. Barnavon, H. Alexandre, and F. Remize. 2009. Design and performance testing of a real-time PCR assay for sensitive and reliable direct quantification of Brettanomyces in wine. Int. J. Food Microbiol. 129:237–243. 281. Thomas, S. D. 1993. Yeasts as spoilage organisms in beverages, p. 517–562. In A. H. Rose and J. S. Harrison (ed.), The Yeasts, 2nd ed., vol. 5. Yeast Technology. Academic Press, London, United Kingdom. 282. Todd, B. E. N., G. H. Fleet, and P. A. Henschke. 2000. Promotion of autolysis through the interaction of killer and sensitive yeasts: potential application in sparkling wine production. Am. J. Enol. Viticult. 51:65–72. 283. Toit, M., L. Engelbrecht, E. Lerm, and S. KriegerWeber. 2011. Lactobacillus: the next generation of malolactic fermentation starter cultures—an overview. Food Bioprocess Technol. 4:876–906. 284. Tominaga, T., R. Baltenweck-Guyot, C. Peyrot des Gachon, and D. Dubourdieu. 2000. Contribution of vol-

Fermentations and Beneficial Microorganisms

285.

286.

287.

288.

289.

290.

291. 292. 293.

294.

295.

296.

297.

atile thiols to the aromas of white wines made from several Vitis vinifera grape varieties. Am. J. Enol. Viticult. 51:178–181. Torija, M. J., E. Mateo, J. M. Guillamón, and A. Mas. 2010. Identification and quantification of acetic acid bacteria in wine and vinegar by TaqMan-MGB probes. Food Microbiol. 27:257–265. Torrea, D., C. Varela, M. Ugliano, C. Ancin-Azpilicueta, I. L. Francis, and P. A. Henschke. 2011. Comparison of inorganic and organic nitrogen supplementation of grape juice—effect on volatile composition and aroma profile of a Chardonnay wine fermented with Saccharomyces cerevisiae yeast. Food Chem. 127:1072–1083. Trcek, J. 2005. Quick identification of acetic acid bacteria based on nucleotide sequences of the 16S-23S DNA internal transcribed spacer region and of the PQQdependent alcohol dehydrogenase gene. Syst. Appl. Microbiol. 28:735–745. Ugliano, M., A. Genovese, and L. Moio. 2003. Hydrolysis of wine aroma precursors during malolactic fermentation with four commercial starter cultures of Oenococcus oeni. J. Agric. Food Chem. 51:5073–5078. Uthurry, C. A., J. A. Suarez Lepe, J. Lombardero, and J. R. Garcia Del Hierro. 2006. Ethyl carbamate production by selected yeast and lactic acid bacteria in red wine. Food Chem. 94:262–270. Van der Westhuizen, T. J., O. P. H. Augustyn, W. Khan, and I. S. Pretorius. 2000. Seasonal variation of indigenous Saccharomyces cerevisiae strains isolated from vineyards of the Western Cape in South Africa. S. Afr. J. Enol. Viticult. 21:10–16. van Vuuren, H. J. J., and L. M. T. Dicks. 1993. Leuconostoc oenos—a review. Am. J. Enol. Viticult. 44:99–112. Varela, C., J. Cardenas, F. Melo, and E. Agosin. 2005. Quantitative analysis of wine yeast gene expression profiles under winemaking conditions. Yeast 22:369–383. Vaudano, E., and E. Garcia-Moruno. 2008. Discrimination of Saccharomyces cerevisiae wine strains using microsatellite multiplex PCR and band pattern analysis. Food Microbiol. 25:56–64. Versavaud, A., P. Courcoux, C. Roulland, L. Dulau, and J. N. Hallet. 1995. Genetic diversity and geographical distribution of wild Saccharomyces cerevisiae strains from the wine-producing area of Charentes, France. Appl. Environ. Microbiol. 61:3521–3529. Viana, F., J. V. Gil, S. Genovés, S. Vallés, and P. Manzanares. 2008. Rational selection of nonSaccharomyces wine yeasts for mixed starters based on ester formation and enological traits. Food Microbiol. 25:778–785. Volschenk, H., M. Viljoen, J. Grobler, F. Bauer, A. Lonvaud-Funel, M. Denayrolles, R. E. Subden, and H. J. J. van Vuuren. 1997. Malolactic fermentation in grape must by a genetically engineered strain of Saccharomyces cerevisiae. Am. J. Enol. Viticult. 48:193–197. Volschenk, H., M. Viljoen-Bloom, R. E. Subden, and H. J. J. van Vurren. 2001. Malo-ethanolic fermentation in grape must by recombinant strains of Saccharomyces cerevisiae. Yeast 18:693–670.

37.  Wine 298. Walling, E., E. Gindreau, and A. Lonvaud-Funel. 2004. A putative glucan synthase gene dps detected in exopolysaccharide producing Pediococcus damnosus and Oenococcus oeni strains isolated from wine and cider. Int. J. Food Microbiol. 98:53–62. 299. Wang, X. D., J. C. Bohlscheid, and C. G. Edwards. 2003. Fermentative activity and production of volatile compounds by Saccharomyces grown in synthetic grape juice media deficient in assimilable nitrogen and/or pantothenic acid. J. Appl. Microbiol. 94:349–359. 300. Wedral, D., R. Shewfelt, and J. Frank. 2010. The challenge of Brettanomyces in wine. LWT Food Sci. Technol. 43:1474–1479. 301. Welke, J. E., M. Hoeltz, H. A. Dottori, and I. B. Noll. 2010. Rapid, simple, and economical method for quantification of ochratoxin A in red wine. J. Planar Chromatogr. 23:116–118. 302. Wibowo, D., R. Eschenbruch, C. Davis, G. H. Fleet, and T. H. Lee. 1985. Occurrence and growth of lactic acid bacteria in wine—a review. Am. J. Enol. Viticult. 36:302–313. 303. Wibowo, D., G. H. Fleet, T. H. Lee, and R. E. Eschenbruch. 1988. Factors affecting the induction of malolactic fermentation in red wines with Leuconostoc oenos. J. Appl. Bacteriol. 64:421–428. 304. Wilkes, K. L., and M. R. Dharmadhikari. 1997. Treatment of barrel wood infected with acetic acid bacteria. Am. J. Enol. Viticult. 48:516–520. 305. Yap, N. A., M. de Barros Lopes, P. Langridge, and P. A. Henschke. 2000. The incidence of killer activity of non-

947

306.

307.

308.

309. 310.

311.

Saccharomyces yeasts towards indigenous yeast species of grape must: potential application in wine fermentation. J. Appl. Microbiol. 89:381–389. Yokotsuka, K., M. Yajima, and T. Matsudo. 1997. Production of bottle-fermented sparkling wine using yeast immobilised in double-layer gel beads or strands. Am. J. Enol. Viticult. 48:471–481. Zagorc, T., A. Maraz, N. Cadez, K. Povhe Jemec, G. Peter, M. Resnik, J. Nemanic, and P. Raspor. 2001. Indigneous wine killer yeasts and their application as a starter culture in wine fermentation. Food Microbiol. 18:441–451. Zavaleta, A. I., A. J. Martínez-Murcia, and F. Rodríguez-Valera. 1997. Intraspecific genetic diversity of Oenococcus oeni as derived from DNA fingerprinting and sequence analyses. Appl. Environ. Microbiol. 63:1261–1267. Zhao, J., and G. H. Fleet. 2003. Degradation of DNA during the autolysis of Saccharomyces cerevisiae. J. Ind. Microbiol. Biotechnol. 30:175–182. Zott, K., O. Claisse, P. Lucas, J. Coulon, A. LonvaudFunel, and I. Masneuf-Pomarede. 2010. Characterization of the yeast ecosystem in grape must and wine using real-time PCR. Food Microbiol. 27:559–567. Zuzuarregui, A., P. Carrasco, A. Palacios, A. Julien, and M. del Olmo. 2005. Analysis of the expression of some stress induced genes in several commercial wine yeasts strains at the beginning of vinification. J. Appl. Microbiol. 98:299–307.

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch38

38

Erika A. Pfeiler Todd R. Klaenhammer

Probiotics and Prebiotics

The probiotic concept While much of food microbiology research is concerned with the study of foodborne pathogens, there is an ­important group of foodborne microorganisms that may have significant benefits for the health of humans and animals. Probiotic (from the Latin and Greek words meaning “for life”) bacteria have long been believed to influence general health and well-being through their association with the gastrointestinal tract (GIT) and its normal microbiota. The concept of probiotics was first popularized at the beginning of the 20th century by the Russian Nobel laureate Elie Metchnikoff (Fig. 38.1). He proposed that a normal, healthy gastrointestinal microbiota in humans and animals provided resistance against “putrefactive” intestinal pathogens (11). He theo­ rized that the intestinal flora influences the incidence and ­severity of enteric infections and ­either enhances or slows atrophy and aging processes. Metchnikoff isolated a Lactobacillus culture from a fermented milk consumed by Bulgarian peasants who were renowned for living long and healthy lives. The culture produced large amounts of lactic acid and survived during ­intestinal ­implantation studies. From these and many ­observations made during

his work on the intestinal microbiota, Metchnikoff suggested that by transforming the “wild population of the intestine into a cultured ­population . . . the pathological symptoms may be ­ removed from old age, and . . . in all probability, the duration of life of man may be considerably increased” (as related by Bibel [11]). The consumption of fermented dairy products (yogurt, kefir, sour cream) became popular in Europe, with the Pasteur Institute supporting the manufacture of a product called Le Ferment, based on the Bulgarian Lactobacillus. Later, Minoru Shirota (ca. 1930 [88]) selected from human feces a Lactobacillus culture, Lactobacillus casei strain Shirota, that survived passage through the GIT. The fermented milk product Yakult is manufactured using a pure culture of strain Shirota. Yakult remains a dietary staple in the Japanese and Korean societies and is now available in Europe, and the company has ­announced plans to begin marketing this product in the United States.

The gastrointestinal microbiota Humans are colonized by thousands of microbial species (73) that exist as commensals, largely on ­mucosal

Erika A. Pfeiler, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740. Todd R. Klaenhammer, Department of Food, Bioprocessing & Nutrition Sciences, North Carolina State University, Schaub Hall, Box 7624, Raleigh, NC 27695-7624.

949

SMP_Food Microbiology_CH38.indd

949

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

950

Figure 38.1  Probiotic pioneers: (A) Elie Metchnikoff (1845–1916), (B) L. casei strain Shirota, and (C) Bifidobacterium species. doi:10.1128/9781555818463.ch38f1

tissues of the nose, mouth, GIT, and vagina (Fig. 38.2). Interestingly, some of the most widely used probiotic species were identified in the analysis depicted in Fig. 38.2 but were minor components of the natural commensal microbiota. Some of the core functions found within the collective gene set of the microbiota encode proteins with functions such as metabolic activities for sugar harvesting and cell surface proteins implicated in adherence to host proteins, ­ including collagen, fibrinogen, and fibronectin. Bacterial populations found throughout the human body are estimated at 1014 cells, 10-fold more cells than the 1013 mammalian cells comprising the human body itself, with most residing in the GIT. It has been historically established that the composition of the GIT microbiota is complex, dynamic, and specific to each host and can change markedly with diet, age, and lifestyle (Fig. 38.3) (62, 93). Studies comparing normally colonized and gnotobiotic animals have established that the gastrointestinal microbiota is responsible for many important properties that affect the metabolism of food and drugs, the renewal of gut epithelial cells, immune system development, heart size, and general behavioral characteristics

SMP_Food Microbiology_CH38.indd

950

(96). The microbiotas of humans, animals, and fowl vary considerably with the architecture of their GITs (93). Species of microorganisms are located at different locations throughout the GIT and include strains that are either harmful or beneficial to the host depending on the circumstances and specific strains involved (Fig. 38.4) (28). For example, there is a clear distinction between Escherichia coli and the fecal coliforms comprising the normal flora and those varieties that are pathogenic and/ or enterotoxigenic, such as E. coli O157:H7. The general impact of the ­microbiota on the host can depend on many factors that affect its composition, including diet, age, exposure to exogenous microorganisms, the genetic makeup of the host, and physiological conditions of the host tissues. The digestive tract is composed of four major categories of microbial populations (24, 94, 95) that are defined as: •

•

autochthonous microbiota: populations of microbes that are present in large numbers and permanently colonize the host normal microbiota: microorganisms that are ­frequently present but can vary in number and be sporadically absent

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

951

Figure 38.2  Relations among the most abundant bacterial species. The network was deduced from the analysis of 155 bacterial species present in at least one individual at a genome coverage of ≥1%. Size of the nodes (circles) indicates species abundance over the cohort; width of the edges (lines connecting the circles) indicates the value of the Pearson correlation coefficient (only the 342 values above 0.4 or below –0.4 out of a total of 11,935 were used for the network). Red arrows identify common lactobacilli also used as probiotic cultures. Adapted from reference 73 with permission from Macmillan Publishers Ltd. doi:10.1128/9781555818463.ch38f2

•

•

pathogens: microorganisms that are periodically acquired but can persist and cause infection and disease allochthonous microbiota: microbes of another origin that are present temporarily (most probiotic cultures are allochthonous [95])

Scientists have long been interested in characterizing the microbiota of the GIT and have used ­ increasingly sophisticated technologies to accomplish this task. While culture-based technologies have long been available to microbiologists, only a small proportion of the GIT microflora is culturable. Molecular biology techniques have proven valuable in the analysis of the GIT microbi-

SMP_Food Microbiology_CH38.indd

951

ota, particularly the use of DNA sequencing to ­identify members of this community. A common ­investigational strategy, sequencing the 16S rRNA genes present in mixed environmental samples of microorganisms, is not unique to the study of the GIT microbiota; however, this strategy has been useful in assessing the diversity of this ecosystem. Molecular technologies, ­ including 16S rRNA sequencing (100), have revealed that individual humans carry their own unique collection of autochthonous strains. While the exact microorganisms vary considerably among individuals, it is clear that each person exhibits a persistent flora that can be recovered repeatedly over extended periods. Severe disturbances

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

952

Figure 38.3  Differences in species of bacteria in human feces of different ages. From reference 62. doi:10.1128/9781555818463.ch38f3

to the GIT, such as antibiotic therapy, enteric infections, or dietary stress, can temporarily disrupt the autochthonous and normal microbiota. However, these populations appear to quickly recover as the host returns to the normal state. Recent studies have revealed an important role of the GIT microbiota in host health and disease, including ­research that determined differences in the compo-

sition of the autochthonous microbiota of obese and lean mice. An examination of these populations using a metagenomic ­ sequencing approach revealed that the ­ microbiota of obese mice was enriched for genes ­involved in ­ energy extraction from food, suggesting a mechanistic link ­ between the microbiota and a physiological condition in the host. Studies in lean versus obese humans have revealed a similar phenomenon (56). Number/gm feces Log10scale

Harmful and Beneficial

Harmful roles

Bacteroides b, d, e, f, h, i

a: Pathogenic b: Production of potential carcinogens c: Production of toxic H2S d: Intestinal putrefaction

Sulfate reducers c

Eubacteria Methanogens g Anaerobic Gram+ bacteria a, b, e Enterobacteriaceae a,e E. coli a, b, d, f

1011

Bifidobacterium e, f, h, i

Lactobacillus f, h, i

Veillonella a Clostridium a Micrococcaceae a Vibrionaceae a Ps aeruginosa a

Beneficial roles e: Competitive bacterial exclusion

1010 10

9

108 10

7

f: Stimulation of immune functions g: Lower gas distension h: Aid in food adsorption/digestion i: Vitamins synthesis

105 103 102

Harmful

Beneficial

Figure 38.4  Predominant colonic microorganisms categorized into potentially harmful or beneficial groups. Adapted from reference 29. doi:10.1128/9781555818463.ch38f4

SMP_Food Microbiology_CH38.indd

952

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

953

Findings such as these highlight the importance of the National Institutes of Health Human Microbiome Project, a multidisciplinary, multicenter ­ effort to characterize the ­microbial communities present in ­humans, ­including the GIT ­ microbiota, with the goal of establishing the role of these communities as well as defining parameters to optimize their contributions to human health (96). Recently, Qin et al. (73) catalogued the fecal microbiota of 124 European individuals and found that between 1,000 and 1,150 species of bacteria were prevalent in the subjects tested. The effects of these ­microorganisms on the host are largely unknown, but this initiative promises to advance research findings in this area. Lactobacillus and Bifidobacterium species are natural residents of the GIT and are generally considered to be beneficial to the health of their hosts and to pose no risk of harmful responses within their normal ecological niches (see Fig. 38.4). Further, lactobacilli have long been associated with food through their role in fermentation and food preservation. While some species of Bacillus and Escherichia have been proposed as probiotics, the preponderance of research in this field has been focused on Lactobacillus and Bifidobacterium. As a result, they are the two major species around which the probiotic concept has evolved. From this platform, the question of probiotics’ ability to positively or negatively influence the autochthonous, normal, or pathogenic microbiota of the GIT can be discussed.

Probiotics and Abiotics

Definitions

In contrast to “antibiotic,” which describes a substance antagonistic to the growth of a microorganism, the term “probiotic” was first coined to describe a substance produced by one microorganism that stimulates the growth of another microorganism (57). Fuller (27) later defined probiotics as “a live microbial feed supplement which beneficially affects the host animal by improving its intestinal balance.” The definition evolved to consider many types of probiotic cultures (mono- and mixedstrain cultures, multiple probiotic species), ­applications (gastrointestinal versus topical), and mechanisms of probiotic activity (live cells, dead cells, and cellular components). While the term “probiotic” does not ­encompass a legal definition in most countries, including the United States, the following definition of probiotics was adopted by a joint United Nations Food and Agriculture Organization/World Health Organization working group in 2002: “live microorganisms which when administered in adequate amounts confer a health

SMP_Food Microbiology_CH38.indd

953

benefit on the host” (41). This is currently one of the most widely recognized definitions of these microorganisms and is applied in this text. There is also interest in the potential health effects of consuming killed bacteria or cellular components. Since the definition of probiotic stipulates use of a live microorganism, these products do not fit this definition but are collectively referred to as “abiotics,” defined as “nonviable probiotic organisms or cellular components thereof that exert beneficial effects on health or wellbeing” (adapted from the definition in reference 89). Abiotics typically consist of probiotic strains that have been inactivated by lysis, heat, or UV irradiation and have demonstrated some effectiveness in downregulating inflammatory response in intestinal epithelial cells (43). The preponderance of research, however, is conducted using live bacterial cells. This chapter focuses primarily on live microorganisms but gives this definition as a point of contrast to probiotics (Fig. 38.5).

Mechanisms of Action

Probiotic microorganisms typically designed for deli­ very in dairy foods are most often members of the Lactobacillus or Bifidobacterium genus (Table 38.1). Nonpathogenic microorganisms that occur within niches in the host gut or tissues, such as Bacillus species, E. coli, and some yeast species, are also used as human and animal probiotics. Interactions of probiotics and gut cells are believed to be mediated by the interaction of bacterial cell wall macromolecules (which can vary from strain to strain, leading to probiotic strain specificity) and pattern recognition receptors on eukaryotic cells (54). While the mechanisms through which probiotics exert their beneficial effects are not entirely understood on a clinical level, an examination of studies indicates three major avenues through which probiotic cultures appear to carry out beneficial activities in the GIT (53, 75, 82). The mechanisms of action of probiotics can thus be summarized in three major categories. 1. Interaction with the immune system. The GIT is a prominent part of the immune system and constantly interacts with consumed foods as well the commensal microbiota. Exposure of certain probiotic strains to the GIT, either in in vitro model systems or in vivo, has been shown to have effects on the immune system. Some probiotics can alter cytokine production by exposed macrophages and dendritic cells and can shift the production of cytokines from the inflammation-inducing interleukin12 (IL-12) pathway to the anti-­inflammatory IL-10 pathway (31, 36, 49, 87).

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

954

Figure 38.5  Heath benefits and suspected mechanisms of probiotics versus abiotics. IgE and IgA, immunoglobulins E and A. doi:10.1128/9781555818463.ch38f5

2. Strengthening the mucosal barrier. The GIT functions as a semipermeable barrier that ­allows the selective passage of certain molecules. Dysfunc­ tional barriers are involved with a number of diseases of the GIT, including inflammatory bowel Table 38.1  Examples of human probiotic species and strains

with research documentationa

Lactobacillus acidophilus (NCFM, SBT-2062, DDS1) Lactobacillus casei (Shirota, CRL431, DN014 001, immunitas) Lactobacillus johnsonii (La1, Lj1) Lactobacillus fermentum Saccharomyces boulardii

Probiotics for Animal Nutrition

Lactobacillus rhamnosus (GG, 271, GR1)

Considerable effort has been directed toward the development of probiotic cultures for use in both agricultural and companion animals. The primary health targets for agricultural animal probiotics are enhancement of animal growth, attainment of weight gains, and reduction in the carriage of human enteric pathogens. Reduced pathogen carriage is of significant interest because control of enteric pathogens at the farm level can greatly reduce the risk of foodborne illness. While antibiotics have been used extensively as prophylactics in animal feeds since the 1950s to improve growth and feed conversion, concerns over the development, transmission,

Lactobacillus plantarum (299V) Lactobacillus reuteri (SD2112) Lactobacillus salivarius (UCC118) Streptococcus thermophilus (1131) Bifidobacterium lactis (BB12) Bifidobacterium longum (SBT-2928, BB536) Bifidobacterium breve Yakult Lactobacillus paracasei (CLR431, F19) Lactobacillus delbrueckii subsp. bulgaricus (2038) a

Compiled from reference 80.

SMP_Food Microbiology_CH38.indd

disease (97). Some probiotic strains can reinforce and repair this barrier by stimulating the production of protective proteins, such as mucins, by ­intestinal epithelial cells (110). 3. Exclusion of pathogens. While specific probiotic strains can inhibit the adherence of some bacterial and viral pathogens in vitro, the mechanisms through which this occurs in vivo are largely unknown. It has been suggested that preemptive binding of probiotics to receptor sites blocks pathogen adhesion. Further, stimulation of host cells to produce mucin and thereby to tighten the mucosal barrier likely acts to block infection by pathogenic species (75, 82).

954

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

955

and spread of antibiotic resistance determinants through this practice have led to its ban by regulatory agencies in some countries (Regulation [EC] No. 1831/2003 of the European Parliament and of the Council on additives for use in animal nutrition) and have revived interest in developing probiotic cultures for livestock and poultry (27). Nearly 40 years ago, Nurmi and Rantala (65) discovered that newly hatched chicks could be protected from Salmonella infection by exposure to a suspension of gut contents from the adult chicken. This concept, termed “competitive exclusion,” was later revived in a probiotic mixture composed of 29 species of nonpathogenic bacteria isolated from the chicken gut. The probiotic mixture is sprayed over the chicks so that as they preen, their intestinal tract is seeded with microbes that occupy this ecological niche to provide colonization resistance against Salmonella, Campylobacter, and Listeria (38, 64, 74). This “seeding” mimics the colonization that occurs naturally in small flocks, in which hens are in close contact with the hatchlings. Generally, it has been determined that undefined competitive ­exclusion culture mixtures (e.g., mixtures derived from gut contents) perform better than simpler defined mixtures (e.g., a single strain or defined mix of strains) (23). Similarly, Zhao et al. (111) specifically targeted colonization of cattle by E. coli O157:H7 by creating a probiotic cocktail of 17 isolates of E. coli and 1 Proteus mirabi­ lis strain. These were isolated from cattle and selected on the basis of their ability to inhibit E. coli O157:H7 in vitro. In challenge studies with E. coli O157:H7, the “probiotic mixture” reduced the level of pathogen carriage in most of the probiotic-treated animals. Bacterial probiotics are generally effective in chickens, pigs, and preruminant calves, whereas fungal probiotics have shown better results in adult ruminants. In the animal probiotic field, host specificity, animal age, and targeted benefit are critically important in selecting cultures for specific probiotic applications. Microorganisms used as probiotics for farm animals (Table 38.2) are typically pure cultures or multiple-strain cultures that act more broadly in multiple hosts under varied conditions (17, 27). There are currently a number of commercially available competitive exclusion products available for agricultural animals. Parameters for studies involving probiotic administration for companion animals have largely focused on cats and dogs and have been similar to those for ­humans. Probiotic feeding studies for companion animals often utilize defined multiple-strain mixes and seek clinical end points similar to those in human studies such as a reduction of diarrhea and treatment of gastrointestinal disorders (109).

SMP_Food Microbiology_CH38.indd

955

Probiotics for Human Health

While substantial research has been conducted to examine the physiological and biochemical characteristics of probiotic microorganisms in culture or model systems, a major area of need for probiotic research involves ­accumulation of evidence that supports the primary claim that probiotics exert a beneficial influence on the human GIT ecosystem. A list of the proposed health outcomes of probiotic consumption and their suspected mechanisms is compiled in Table 38.3. Well-designed studies using carefully selected and defined probiotic cultures are increasingly supporting these health outcomes (94).

Table 38.2  Microorganisms reviewed by the Food and Drug

Administration Center for Veterinary Medicine that were found to present no safety concerns when used in direct-fed microbialsa Aspergillus niger

Lactobacillus curvatus

Aspergillus oryzae

Lactobacillus delbrueckii

Bacillus coagulans

Lactobacillus fermentum

Bacillus lentus

Lactobacillus helveticus

Bacillus licheniformis

Lactobacillus lactis

Bacillus pumilus

Lactobacillus plantarum

Bacillus subtilis

Lactobacillus reuteri

Bacteroides amylophilus

Leuconostoc mesenteroides

Bacteroides capillosus

Pediococcus acidilactici

Bacteroides ruminicola

Pediococcus cerevisiae (damnosus)

Bacteroides suis

Pediococcus pentosaceus

Bifidobacterium adolescentis

Propionibacterium acidipropionici (cattle only)

Bifidobacterium animalis

Propionibacterium freudenreichii

Bifidobacterium bifidum

Propionibacterium shermanii

Bifidobacterium infantis

Saccharomyces cerevisiae

Bifidobacterium thermophilum

Enterococcus cremoris

Lactobacillus acidophilus

Enterococcus diacetylactis

Lactobacillus brevis

Enterococcus faecium

Lactobacillus buchneri (cattle only)

Enterococcus intermedius

Lactobacillus bulgaricus

Enterococcus lactis

Lactobacillus casei

Enterococcus thermophilus

Lactobacillus farciminis (swine only)

Yeast

Lactobacillus cellobiosus Compiled from reference 6.

a

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

956 However, some of these health claims remain controversial, and increasing pressure has appeared for clinical trials that demonstrate the health benefits of probiotics, such as those used to test the effectiveness of drugs and other clinical therapies. However, studies of this type can be inherently limited for three reasons. First, as with other nutritional interventions, the effect of probiotics is likely to be smaller than that of conventional drugs and can be more easily confounded by factors such as age, diet, and lifestyle (76). Second, study results can vary with the strains used, the population level of probiotic cells delivered, the health marker targeted, and the number, age, and condition of the subjects evaluated. Also,

little is known about the minimal dose of probiotic and the frequency of consumption required to elicit physiological effects. Finally, traditional clinical trials require the measurement of validated biomarkers to measure the effectiveness of a treatment. While these biomarkers exist for many disease states, there are no consensus biomarkers available to examine more general effects, such as immune stimulation or improvement of gastrointestinal health. A significant technical challenge to the assessment of probiotic effects involves the administration of a consistent number of live cells in each dose. Differences in dosing levels of probiotic strains can confound the ­noticeable

Table 38.3  Proposed health benefits and mechanisms of probioticsa Proposed health outcome

Suspected mechanisms

Promotion of lactose digestion in lactoseintolerant individuals

Lactase activity from probiotic cultures Lactase released from transient bile-sensitive lactic acid bacteria that lyse in the small intestine

Resistance to enteric pathogens

Colonization resistance Systemic immunity Shortened duration and enhanced recovery from diarrhea Adjuvant increasing secretory antibody production Alteration to unfavorable or antagonistic conditions for pathogens (lowering pH, production of bacteriocins and short-chain fatty acids)

Anticarcinogenic

Antimutagenic activity (binding of mutagens) Lowering procarcinogenic enzyme activities (nitroreductase, β-glucuronidase, azoreductase) of colonic bacteria in humans and animals Reduction in the number of aberrant crypt foci in colon of animals Stimulation of immune function Influence on secondary bile salt concentrations

Reduction of toxic impacts of small bowel bacterial overgrowth

Decrease in the production of toxic metabolites, e.g., dimethylamine, by colonic microbiota

Immune system modulation

Strengthening of nonspecific and antigen-specific defenses against infections and tumors Adjuvant effect on antigen-specific immune responses Lowering of inflammatory responses Amelioration of atopic eczema in infants Influence on cytokine production and developmental pathways for Th1/Th2 development

Treatment of blood lipids, heart disease

Alteration of bile salt hydrolase activity Reduction of cholesterol (mechanism unknown)

Antihypertensive effects

Bacterial peptidase action on milk proteins produces a tripeptide that acts as an angiotensin1-converting enzyme inhibitor to lower blood pressure in hypertensive animals

Reduction of ulcerative Helicobacter pylori activity

Production of inhibitors by lactic acid bacteria, in fermented milks, against H. pylori

Treatment of hepatic encephalopathy

Competitive exclusion or inhibition of urease-producing gut flora

Reduction of urogenital infections

Adhesion to urinary and vaginal cells Competitive exclusion Production of inhibitors (biosurfactants, hydrogen peroxide)

Alteration of gut motility

Unknown

a

Compiled from references 72, 79–81, and 94.

SMP_Food Microbiology_CH38.indd

956

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

957

effects of strains. Levels of cells delivered in probiotic food products can range from 106 to 109 CFU per ml or gram depending on the food carrier and method of elevating the probiotic bacterial concentration. For example, in the manufacture of Sweet Acidophilus milk, Lactobacillus acidophilus NCFM cells are concentrated to 1010 CFU/ml in frozen culture concentrates and then added directly to 2% fluid milk to yield a target concentration approaching 107 CFU/ml. Alternatively, as is the case in some yogurt fermentations, probiotic cultures are added as inoculants and ­allowed to grow during the fermentation phase, reaching levels of 108 to 109 CFU/ml. The proposed target level for daily consumption of a probiotic culture is 108 to 109 CFU per day (up to 109 to 1010 CFU/day if losses are expected during stomach transit) (79). Achieving these levels of viable probiotic cultures in foods can be a challenging technological task. Substantial effort is required, on a strainto-strain-dependent basis, to design the fermentation, concentration, and storage (dried/frozen) conditions that will enhance the viability and functional activity of probiotic cultures. Stress preconditioning can enhance the storage stability of probiotic cultures in harsh environments (36).

Probiotics as Human Therapeutics

There is considerable interest in using probiotics as potential interventions for treatment of diseases, including the treatment of diarrheal conditions, inflammatory bowel disease, allergies, and infections of the upper respiratory tract. For a number of reviews on clinical uses of probiotics, see reference 33. A number of reports ­indicate the successful treatment of diarrhea, particularly in children and in cases of antibiotic-associated ­ diarrhea. Despite these clinical findings, no mechanism of action for probiotics in the treatment of these conditions has been defined (108). Meta-analyses of many potential probiotic treatments by the Cochrane Collaboration (Table 38.4) either have found no beneficial effects of probiotic treatment or have indicated that previous studies were not conducted in a manner that allows definitive interpretation of the data. The delivery of vaccines and other biotherapeutics to the mucosal immune system of the GIT via probiotic microbes is considered beneficial and could increase ­ antigen potency and reduce the potential side ­effects ­incurred through other, more traditional delivery routes. Oral ingestion of proteins (vaccines or ­enzymes) results in denaturation, degradation, and loss of biological ­activity. In contrast, bioactive molecules can be protected by viable or nonviable cells during passage through the stomach and then released into the GIT.

SMP_Food Microbiology_CH38.indd

957

Bioactive molecules targeted for delivery by probiotic lactic acid bacteria include vaccine antigens against viral and bacterial pathogens, digestive enzymes for humans, and growth-promoting enzymes in animals. Probiotics are an attractive choice for mucosal delivery vehicles ­because of their ability to survive GIT passage, their ­record of safety, their nonreactivity to the immune system, and the ease of genetic manipulation (107). Wells et al. (106) originally pioneered the development and use of lactic acid bacteria as vaccine delivery vehicles and have developed recombinant strains designed to treat a number of conditions, including HIV infection. Later, it was demonstrated that a genetically modified Lactococcus lactis designed to produce and secrete the cytokine IL-10 in the GIT could decrease the GIT inflammatory responses of mice with colitis (90). This group also addressed the important issue of biocontainment of the recombinant microorganism by devising an elegant solution of inserting the IL-10 gene into the L. lactis thyA gene, which is responsible for the synthesis of the ­nucleic acid thymidine. With this gene inactivated, L. lactis ­requires an ­exogenous source of thymidine that can be provided in the culture medium but is scarce in the environment, thereby preventing the growth and spread of the microorganism outside controlled conditions (90). This recombinant microorganism has successfully completed phase 1 clinical trials in humans indicating both the feasibility of this type of treatment and the effectiveness of the engineered biocontainment system (14). This treatment has completed phase 2a clinical trials to test, among other end points, efficacy in treating ulcerative colitis, although results have not yet been released. There are many exciting opportunities to employ probiotic bacteria as vehicles to deliver bioactive molecules to targeted locations in the GIT. This is likely to be one of the most important areas for practical ­ application of probiotic cultures that are derived through recombinant DNA technology.

Taxonomy and molecular identification of probiotic strains The correct identification of probiotic strains is critical to a number of activities associated with these microorganisms, including product labeling, safety assessment, scientific communication, and attribution of beneficial effects. Despite scientific advances in bacterial taxonomy and classification that are applied to pathogenic microorganisms, research into the taxonomy of probiotics has not been as well defined, which has led to historical problems of misidentification of probiotic microbes. Strains under investigation were often misidentified or

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

958

Table 38.4  Compilation of Cochrane Collaboration reports on clinical uses of probiotics Year

Title

Reference

2004

Probiotics for treating infectious diarrhea

3

2006

Probiotics for maintenance of remission in Crohn’s disease

77

2007

Probiotics for the prevention of pediatric antibiotic-associated diarrhea

40

2007

Probiotics for induction of remission in ulcerative colitis

60

2007

Probiotics for non-alcoholic fatty liver disease and/or steatohepatitis

58

2007

Probiotics in infants for prevention of allergic disease and food hypersensitivity

67

2008

Probiotics for treatment of Clostridium difficile-associated colitis in adults

71

2007

Probiotics for preventing preterm labor

69

2008

Probiotics for treating eczema

13

2008

Probiotics for induction of remission in Crohn’s disease

16

2008

Methods of preventing bacterial sepsis and wound complications for liver transplantation

32

2008

Interventions for treating collagenous colitis

18

2008

Probiotics for prevention of necrotizing enterocolitis in preterm infants

2

2009

Probiotics for the treatment of bacterial vaginosis

86

2009

The effects of antimicrobial therapy on bacterial vaginosis in non-pregnant women

66

2009

Interventions for prevention of post-operative recurrence of Crohn’s disease

22

2009

Dietary interventions for recurrent abdominal pain (RAP) and irritable bowel syndrome (IBS) in childhood

37

2010

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

34

incorrectly named, leading to health claims that were ­inappropriately inferred on the basis of name recognition. Because of these issues, the bacterial genus and species names found in the probiotic literature prior to 1995 should be viewed with caution. However, the development of molecular biology techniques and a growing interest in the taxonomy of probiotic strains has led to significant improvements in this area. In bacterial taxonomy, a species is considered to be the fundamental unit of classification (26), while a strain is “a set of isolates that when typed, give the same typing result and that can be distinguished from other isolates of the same genus and species” (5). Historically, bacterial species were grouped by the relatedness of their DNA; i.e., if two individual isolates were determined to share 70% or more of their genomic DNA sequence as ­assayed by genome-genome hybridization, they were considered to be in the same species. However, the ­advent of molecular techniques and the availability of efficient and cost-effective commercial sequencing facilities have made bacterial species identification by sequencing and phylogenetic analysis commonplace (26). The availability of molecular tools and analyses to properly identify probiotic species and individual strains during the last decade has helped dispel confu-

SMP_Food Microbiology_CH38.indd

958

sion over species identity and ancestry. The accumulating ­sequence information on rRNAs provides a growing resource for comparative identification of probiotic cultures, both established candidates and potentially new candidates. Phylogenetic analysis can be conducted to varying degrees and combined with other characteristics (phenotypes) as needed to make definitive taxonomic classifications. A number of DNA sequence-based ­typing systems have been used to analyze conserved regions of the rRNA operon or other conserved genes in probiotic cultures (68): •

•

•

•

PCR amplification and sequencing of ~1,500 bp of the 16S rRNA gene PCR amplification and sequencing of ~450 bp of the internal transcribed spacer region between the 16S and 23S rRNA genes (Fig. 38.6) PCR amplification and sequencing of ~50 bp of the variable region of 16S rRNA to identify members of the L. acidophilus complex (51) PCR amplification and sequencing of alternative genes that are universally present and highly conserved, e.g., the recA gene of bifidobacteria (50)

Species of Lactobacillus, a common genus of probiotic microorganisms, are members of the phylum

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

959

Firmicutes, class Bacilli, order Lactobacillales, and family Lactobacillaceae. This genus was first identified by Beijerinck in the early 1900s, and members were first classified by their optimal growth temperatures and sugar fermentation pathways. Later, the species was classified based on physiological assays such as the homo-/heterofermentation profiles of the members (26). However, identification and classification utilizing the sequence of the 16S rRNA is currently the most widely accepted method for species-level typing of microorganisms and offers a fast, reproducible, and inexpensive method for discrimination. Members of Bifidobacterium, another widely studied genus of probiotic microorganisms, belong to the phylum Actinobacteria, class Actinobacteria, order Bifidobacteriales, and family Bifidobacteriaceae. As with the lactobacilli, historical methods of identification of species of bifidobacteria relied on phenotypic differences, including the origin or host from which the species was isolated (e.g., B. infantis), as well as sugar fermentation profiles (102). The taxonomic classification of bifidobacteria was improved through the use of 16S rRNA sequencing. A current list of all recognized bacterial species, including Bifidobacterium and Lactobacillus species, can be found in the Approved Lists of Bacterial Names at www.bacterio.cict.fr. The ability to properly identify probiotic species provides the first giant step toward eliminating any confusion over strain identity and ancestry. The prac-

tice is expected to also uncover new probiotic species that have been hidden below the surface of traditio­ nal taxonomic descriptions. DNA sequence analysis of the rRNA operons of strains loosely classified as L. acidophilus reveals six closely related species that comprise the “Acidophilus” complex: L. ­acidophilus, L. crispatus, L. amylovorus, L. gallinarum, L.  gasseri, and L. johnsonii (Fig. 38.6) (39, 47, 52). All of these are considered to have probiotic potential, but it ­remains to be determined what specific roles and benefits are exerted by each species exclusively or collectively by the group. Methods of strain-level typing for probiotics have also benefited from the use of molecular techniques. DNAbased typing methods, like their counterparts in ­sequence typing, quickly provide reproducible results that are able to be universally shared between laboratories. The Food and Agriculture Organization/World Health Organization guidelines suggest the use of pulsed-field gel electrophoresis and randomly amplified polymorphic DNA (41). However other methods have been suggested, including plasmid profiling, restriction enzyme analysis, and ribotyping (35). Polyphasic taxonomy, which applies different types of information about a microorganism, including genotypic and phenotypic data, can also play a role in probiotic typing (98). Finally, the growing number of microbial whole-genome sequences can contribute to typing through ­genomotyping, or using comparative genomics to identify all differences between two sequenced strains.

L. helveticus 38-40% L. crispatus 35-38% L. acidophilus L. amylovorus 40-41% GC 34-37% L. delbrueckii 49-51%

L. johnsonii 33-35% L. gasseri 33-35%

L. sake 16S rRNA

tRNA

5% estimated sequence divergence 23S rRNA

5S rRNA

spacer

Figure 38.6  Phylogenetic relationships among members of the L. acidophilus complex, representing nine variable regions in the 16S rRNA gene used for phylogenetic identification and analysis. Adapted from references 45 and 99. doi:10.1128/9781555818463.ch38f6

SMP_Food Microbiology_CH38.indd

959

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

960

The Effects of Probiotics on GIT Ecology

• •

A major scientific barrier to the acceptance of probiotics is the inability to demonstrate how a probiotic ­influences the complex commensal microbiota of the GIT and then determine the impact of those specific changes on factors that could affect general health and well-being. Historically, studies examining the entire GIT microbiota relied on traditional culture-based methods, but an overwhelming number of commensal microorganisms are not culturable. This knowledge gap has led to scientific criticism of the probiotic field and a charge to demonstrate benefits with the rigor of pharmaceutical studies by applying new technologies to investigate the direct cause-and-effect relationships of probiotics with the GIT microbiota and their larger effects on health (94). The introduction of molecular technologies to ­microbial ecology provides a set of powerful tools for the analysis of the complex, variable, and dynamic communities of the GIT, as well as the impact of probiotic cultures on those communities. These technologies do not rely on culture-based techniques and can provide a more accurate picture of the microbial composition, diversity, and fluctuation within a community (99). An early example of this technology was illustrated by Suau et al. (91), who sequenced 284 PCR-generated 16S rRNA amplicons cloned from one human fecal sample and reported that only 24% of the sequences correlated with known (culturable) organisms, with 76% of the sequences representing previously unidentified microorganisms. Metagenomic techniques, such as environmental shotgun sequencing, have been increasingly applied to investigation of the GIT microflora. Among the first of these studies was an inventory of the sequences of 16S rRNA from microorganisms found in the GIT. This landmark study identified a tremendous diversity in this ecosystem and revealed that 62% of identified microorganisms had not previously been described (25). Along with advances in examining the overall diversity of GIT microbiota, studies have addressed how probiotic administration affects the commensal microbial population. Widely used among these technologies are genetic-based molecular techniques used to fingerprint specific DNA patterns that are characteristic for a single bacterial strain. Methods applied to identify probiotic cultures from complex samples have been reviewed by O’Sullivan (68) and include: • •

ribotyping analysis of restriction fragment length polymorphisms of genomic DNA using pulsed-field gel electrophoresis (Fig. 38.7)

SMP_Food Microbiology_CH38.indd

960

•

randomly amplified polymorphic DNA 16S rRNA sequencing in situ PCR and fluorescent in situ hybridization ­allowing visualization of a specific probiotic culture, or indicator flora, within mixed microbial populations

The molecular tools available for genomic fingerprinting can unequivocally link a fed probiotic culture with the strain recovered in the GIT or feces (1, 95, 104). These technologies have revealed that when feeding is stopped, the probiotic strain is no longer recovered after about 14 to 25 days. It is widely accepted that probiotics do not permanently colonize the GIT. Instead, continuous delivery would be required to maintain their presence (94). Among successful methods of analysis for the composition of the GIT microbiota, the analysis of rRNA genes and use of in situ probes or denaturing gradient gel electrophoresis can distinguish between different species that may be present. A sample from the GIT is used in a PCR amplification with universal primers that are designed to amplify a variable region of the 16S rRNA that is flanked by sequences conserved among all bacteria (Fig. 38.6). The amplicons are identical in length but vary in their melting properties due to variations in the internal 16S rRNA sequence and %G+C content of the bacterium’s DNA. Analysis under denaturing conditions using temperature or chemicals yields a specific banding profile, with each band representing a 16S rRNA amplicon from a different bac-

During Feeding Before Feeding

2 Weeks After Feeding Halted

Figure  38.7  DNA fingerprint of the predominant Lactobacillus culture isolated from human feces before feeding with a probiotic, after feeding with L. acidophilus, and 2 weeks after feeding was halted. SmaI-digested DNA fragments prepared from individual Lactobacillus colonies were separated on a pulsed-field electrophoresis gel. doi:10.1128/9781555818463.ch38f7

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

961

terium that was originally present in the GIT sample. These DNA amplicon bands can be ­extracted from the gel and ­ sequenced to reveal the identity of the organisms present. The presence and intensity of the bands may also reflect the relative population levels of various species in the sample. Profiles of bacteria from human fecal samples show some common bands among different individuals, reflecting universal and dominant species. There are also unique bands characteristic of an individual’s microbiota at that ­ moment in time (112). Denaturing gradient gel electrophoresis analysis of human fecal samples after feeding Lactobacillus salivarius revealed a markedly stable ­ microbiota that was not modified by probiotic treatment (99). A review by O’Toole and Cooney (70) identifies more recent studies that examined the effects of probiotics on the GIT microbiota. Highlighted among these are studies that revealed no large-scale changes of these communities after probiotic administration. A recent study revealed the ­ effects of consuming milk fermen­ ted with Bifidobacterium animalis subsp. lactis on an ­ulcer­ative colitis mouse model. Along with an improve-

ment in colitis scores among the mice, a metagenomics approach was used to monitor the fecal microbiota of the mice and revealed an ­increase in lactose-consuming and butyrate-producing bacteria. The group hypothesized that consumption of the milk product altered the gastrointestinal microbiota to produce an ­inhospitable environment for colitogenic Enterobacteriaceae (101). Additional work of this nature is needed and requires targeting of specific microbial populations in the GIT that are anticipated to be affected by the introduction of probiotic cultures. The techniques of metagenomics will likely play a large role in the examination of probiotic effects on the GIT microbiota. The documented ability of foods and food-bearing probiotics to modify the intestinal flora will be instrumental to proving or refuting probiotic claims on a mechanistic basis. The developments in molecular techniques over the past ­decade have removed many of the key issues that previously hindered scientific progress in probiotics. Precise methods for the identification, tracking, and analysis of probiotic cultures within complex microbial ecosystems now promise to revolutionize our understanding of the

Table 38.5  Desirable selection criteria for probiotic strainsa Appropriateness Taxonomic identification known by phylogenetic analysis and rRNA sequencing Origin: normal inhabitant of the species targeted and isolated from a healthy individual Safety: nontoxic, nonpathogenic, generally-recognized-as-safe status Technological suitability Amenable to mass production and storage: adequate growth, recovery, concentration, freezing, dehydration, storage, and distribution Viability at high populations (preferred at 107–109 CFU/g) Stability of desired characteristics during culture preparation, storage, and delivery Desirable organoleptic qualities (or no undesirable qualities) when included in foods or fermentation processes Genetically stable, maintains phenotypic properties Genetically accessible for potential modification Competitiveness Capable of survival, proliferation, and metabolic activity at the target site in vivo Resistant to bile Resistant to acid Able to compete with the normal microbiota, including the same or closely related species; potentially resistant to bacteriocins, acid, and other antimicrobials produced by resident microbiota Adherence, colonization, and retention evaluated Performance and functionality Able to exert one or more clinically documented health benefits Antagonistic toward pathogenic or cariogenic bacteria Production of antimicrobial substances (bacteriocins, hydrogen peroxide, organic acids, or other inhibitory compounds) Immunostimulatory Anti-inflammatory Antimutagenic Anticarcinogenic Production of bioactive compounds (enzymes, vaccines, and peptides) a

Adapted from reference 45.

SMP_Food Microbiology_CH38.indd

961

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

962 functional roles and in vivo effects associated with probiotic bacteria.

Criteria for selection of probiotic cultures Different probiotic species and strains within a species exhibit distinctive properties that can markedly affect their survival in foods, survival in the GIT, and other ­important probiotic properties. Therefore, strain selection becomes a critical parameter to ensure both the culture’s stability and probiotic performance. Desirable traits for selection of functional probiotics fall into four basic categories: appropriateness, technological suita­ bility, competitiveness, and performance and functionality (Table 38.5) (45). These criteria were established based on research analyzing microbial performance in a number of conditions, including survival in vitro, industrial considerations (stability, viability, and technological suitability), and the history of safe use of some bacterial species in foods (nonpathogenicity and nontoxigenicity). Some criteria, such as survival during GIT transit, retention in the GIT, anti-inflammatory effects, and immunomodulatory ­ activity, require in vivo analysis by human or animal studies. These studies can be lengthy and ­expensive, and they become more complicated as multiple probiotic strains, singly or in cocktails, are evaluated for functional properties. As a result, efforts have been made to develop in vitro systems that better predict probiotic performance in vivo. Some early

e­ xamples ­include the development of a transit tolerance test (19) that evaluates probiotic survival in simulated gastric juice (pH 2) containing pepsin and sodium chloride. Most (14 of 15 strains) of the probiotic cultures evaluated lost 90% ­viability during the gastric portion of the test but survived well under conditions mimicking those of the small intestine. Gastric transit can be detrimental to delivery of viable probiotic cells to the small intestine. Efforts to protect cultures have included microencapsulation or incorporation within dairy foods, which can provide buffering capacity against gastric exposure (19). A more dynamic in vitro system for screening probiotic cultures has been developed by Marteau et al. (61), with a simulated GIT model composed of major components (stomach, small intestine, and colon), chemical ­constituents (pH, bile, pepsin, nutrients, and water), and peristaltic movement. In this model in vitro system, the survival of lactobacilli and bifidobacteria was similar to the results obtained in human in vivo survival studies. Approaches of this type can also be used to elucidate the mechanisms of the probiotic activities of a strain. Oxalate degradation has been proposed as a probiotic function of some strains, as excess oxalate can lead to kidney stone formation. A continuous culture system was constructed, modeling three sections of the large intestine, and inoculated with human fecal samples. A culture of L. gasseri ATCC 33323 was added to the system, which led to degradation of oxalate in the system, indicating the potential for use of this strain to reduce oxalate levels in the GIT (7, 55).

Figure 38.8  Genome atlas of L. acidophilus NCFM. The atlas represents a circular view of the complete genome sequence of L. acidophilus NCFM. The key on the right describes the single circles in the top-down-outermost-innermost direction, as follows. Circle 1 (innermost), GC-skew. Circle 2, Clusters of Orthologous Groups (COG) classification. Predicted open reading frames (ORFs) were analyzed using the COG database and grouped into five major categories: 1, information storage and processing; 2, cellular processes and signaling; 3, metabolism; 4, poorly characterized; 5, ORFs with uncharacterized COGs or no COG assignment. Circle 3, ORF orientation. ORFs in the sense orientation (ORF+) are shown in blue; ORFs oriented in the antisense direction (ORF–) are shown in red. Circle 4, BLAST similarities. Deduced amino acid sequences compared against the nonredundant (nr) database using gapped BLASTP (4a). Regions in blue represent unique proteins in NCFM, whereas highly conserved features are shown in red. The degree of color saturation corresponds to the level of similarity. Circle 5, G+C content deviation. Deviations from the average G+C content are shown in either green (low-GC spike) or orange (high-GC spike). A boxfilter was applied to visualize contiguous regions of low or high deviations. Circle 6, ribosomal machinery. tRNAs, rRNAs, and ribosomal proteins are shown as green, cyan, and red lines, respectively. Clusters of proteins are represented as colored boxes to maintain readability. Circle 7, mobile elements. Predicted transposases are shown as light purple dots, and phage-related integrases are shown as orange dots. Circle 8, stress response. Genes involved in the general stress response, including chaperones, and genes involved in heat shock, DNA repair, and pH regulation are shown as dark purple dots. Circle 9, peptide and amino acid utilization. Proteases and peptidases are shown as green dots, and non-sugar-related transporters are shown as light blue dots. Circle 10 (outermost), two-component regulators (2CRS). Each 2CRS is represented as a brown dot, consisting of a response regulator and a histidine kinase. In circles 7 to 10 each full dot represents one predicted ORF and stacked dots represent clusters of ORFs. Selected features representing single ORFs and ORF clusters are shown outside of circle 10 with bars indicating their absolute size. The origin and terminus of DNA replication are identified in green and red, respectively. Other features: SlpA and SlpB (S-layer proteins), CdpA (cell division protein [3a]), sugar utilization (sucrose, fructo-oligosaccharide [FOS], trehalose, and raffinose), LacE (phosphotransferase system-sugar transporter), BshA and BshB (bile salt hydrolases), Mub-909 to Mub-1709 (mucus-binding proteins; the numbers correspond to the La numbering scheme), FbpA (fibronectin-binding protein), Cfa (cyclopropane fatty acid synthase), Fibronectin_binding (fibronectin-binding protein cluster), EPS_cluster (exopolysaccharides), Lactacin_B (bacteriocin), pauLA-I to pauLA-III (potential autonomous units), and prLA-I and prLA-II (phage remnants). Reprinted from reference 4. doi:10.1128/9781555818463.ch38f8

SMP_Food Microbiology_CH38.indd

962

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

SMP_Food Microbiology_CH38.indd

963

Manila Typesetting Company

11/08/2012 07:19AM

963

Fermentations and Beneficial Microorganisms

964 Selection criteria that address competitiveness (e.g., adherence to intestinal tissues and mucin and production of antimicrobials) and performance issues (e.g., immunostimulation and anticarcinogenic activities) are more complex because the underlying mechanisms by which probiotics exert these functional effects in vivo are not well understood. Therefore, it is difficult to precisely ­define the microbial feature or collection of characteristics that dictate some of the above selection criteria. While much is not still known, our understanding of probiotic properties has been revolutionized by whole-genome sequencing and functional genomics. Genome-sequencing projects examining the collective attributes of strains with probiotic properties have identified a number of similarities that suggest critical traits for these strains, including an extensive network of glycosyl hydrolases in the bifidobacteria that allow for metabolism of complex, indigestible carbohydrates available in the GIT. Further examination of these types of genes ­delineated this genus into species that colonize the infant GIT because of their ability to break down oligosaccharides present in human milk and those that colonize the adult GIT, which are unable to break down these sugars (85). Complete genome annotation and functional analysis support the identification of key genes that are expected to direct functions important in probiotic activity. For example, within the L. acidophilus genome (Fig. 38.8), a series of surface-associated proteins (e.g., proteins ­associated with mucin binding and fibronectin binding) have been identified and correlated with the bacterium’s ability to adhere to intestinal epithelial cells in vitro (15). Across many Lactobacillus species, a number of gene regions have already been investigated and ­implicated in probiotic activity. Among them are genetic loci linked to bile salt hydrolase activity, acid tolerance, bacteriocin activity, oligosaccharide utilization, oxalate degradation, exopolysaccharide production, and communication with immunomodulatory cells of the host (7, 31, 47, 49). Comparative genomic analysis is a powerful tool for recognizing the relationships between microbes and identifying key similarities and differences. Align­ ment of whole genomes has illustrated the overall synteny ­ between the closely related members of the L. acidophilus complex and their substantial differences from Lactobacillus plantarum (46). Moreover, genomic content can be compared between strains of the same species, allowing for an inventory of genetic content or its deficiencies. Molenaar et al. (63) compared the genomic content of 20 strains using DNA microarrays based on the L. plantarum WCFS1 genome. Remarkably, considerable variation was observed between strains, notably

SMP_Food Microbiology_CH38.indd

964

in properties considered important for probiotic activities such as exopolysaccharide synthesis, bacteriocin production, and carbohydrate utilization. Genomic analysis has revealed some fascinating differences and similarities within these probiotic bacteria (12, 46). First, these bacteria are rich in transporters for uptake of sugars and amino acids, emphasizing their dependence on importing, rather than synthesizing, ­nutrients from the GIT. In this regard, members of the L. acidophilus complex (L. acidophilus, L. gasseri, and L. johnsonii) are largely deficient in their ability to synthesize amino acids and other essential nutrients. This is consistent with the evolution of these bacteria within the small intestine, where nutrients are abundant and can be acquired via transporters. In contrast, Bifidobacterium longum encodes complete pathways for synthesis of all amino acids and nucleotides, reflecting the more nutrientcompetitive environment of the colon, where bifidobacteria naturally reside. In addition, B. longum ­harbors a surprisingly large number of genes predicted to encode proteins required for catabolism of complex carbohydrates and plant polymers (83). Such compounds, like fructo-oligosaccharides, are considered prebiotics that can promote the growth of bifidobacteria and some lactobacilli. For the lactobacilli, evolution within the nutritionally rich environments of milk and the GIT has resulted in genome reduction and loss of some biosynthetic capabilities. An analysis of the genome of L. johnsonii (2 Mbp) highlighted the biosynthetic deficiencies of the members of the L. acidophilus complex as compared with the ­expanded metabolic capacity of L. plantarum (12). The most metabolically diverse bacterium among the probiotic lactobacilli is L. plantarum. Its larger genome (3 Mbp) encodes a relatively large number of carbohydrate transport and catabolism pathways that were clustered in a 600-kb “lifestyle adaptation region.” This metabolic capacity and flexibility explains why this bacterium is found naturally in the GIT and can also dominate microbial populations in fermenting plant materials. As a result of genomics, correlations between gene, function, and mechanisms of probiotic activity are now rapidly emerging in this field, and the potential functionality and safety of probiotic cultures can now be ­rationally considered based on genome content. Among the lactobacilli, carbohydrate metabolism genes, genes involved in bile tolerance, and cell surface proteins are characteristics common among probiotic strains. Other species of lactobacilli, however, which are primarily involved in the fermentation of food products such as cheese and are not reported to have probiotic properties, are markedly reduced in their

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

965

gene content of this type (59, 103). Further approaches to validate the importance of probiotic genes include the use of transcriptomics and directed mutagenesis. One particularly interesting example was employed to examine the effects of L. salivarius UCC118 on infection by Listeria monocytogenes in a mouse model. This study revealed that wild-type L. salivarius prevented colonization, whereas a mutant strain unable to produce the Listeria-active bacteriocin did not prevent colonization (20). Establishing mechanisms that link in vitro phenotypic criteria to in vivo functionality remains one of the major scientific challenges for probiotics in the coming decade. In this regard, this field is poised perfectly to further exploit the recent progress in high-throughput sequencing capacity and functional genomics toward the investigation of these bacteria and their probiotic capabilities (44, 46). There remains a myriad of possible probiotic strains representing a diverse set of phenotypes that are being linked increasingly to a variety of benefits. Defining and screening genetic traits important for functional probiotic activities promises to identify superior strains and allow the construction of strain combinations that elicit unique or multiple effects.

SAFETY Many species of microorganisms have been consumed by humans in various forms and at high concentrations for centuries and are not considered to be pathogenic or toxigenic. Safety assessments of the commonly used and studied probiotic genera Lactobacillus and Bifidobacterium document a low potential for ­ adverse health effects from these species (26). However, some proposed probiotic strains are not limited to these genera; other potential genera include Bacillus, Enterococcus, and Escherichia, which contain species that are important foodborne pathogens. Further, the isolation of novel probiotic strains necessitates an assessment of their safety before they are added to foods and supplement products. Safety review of potential probiotics generally involves, although may not be limited to, the strain-level assessment of four criteria: • • •

•

identity of the microorganism history of safe use of the microorganism fermentation and processing conditions for the microorganism the microorganism’s potential for pathogenicity, toxigenicity, and resistance to clinically relevant antibiotics

Reviews conclude that the infective and toxigenic potential of lactobacilli and bifidobacteria is low (9, 10);

SMP_Food Microbiology_CH38.indd

965

these include one particular study that identified only 1 case of Lactobacillus infection in France per 10 million individuals over the period of a century (26). Of note, however, is that opportunistic infections caused by lactobacilli occur predominantly in individuals who are immunocompromised or elderly. With medical technologies that provide for extended survival of chronically immunocompromised individuals, and an aging Western population, increased scrutiny must be given to strains that may be administered to these individuals. Some intestinal isolates in particular, such as Lactobacillus rhamnosus, have warranted increased surveillance due to a higher correlation rate with secondary infections. Ongoing genome-­sequencing projects have not revealed any identifiable pathogenic determinants for probiotic cultures. However, genome-sequencing projects have identified genes in probiotic species that annotate as antibiotic resistance genes or multidrug resistance transporters (48). This is not surprising, given the widespread use of antibiotics in agriculture and medicine and the ­apparent widespread distribution of antibiotic resistance genes among commensal and food bacteria (105). It will be important to determine whether any antibiotic resistance determinant identified in probiotic cultures is functional or clinically relevant and whether it has any reasonable probability for genetic transfer to other microbes.

PREBIOTICS In addition to the multiple impacts of probiotics, benefits to intestinal health can be realized by feeding ingredients that selectively promote the growth of the existing beneficial microbiota. Growth factors in human breast milk, which stimulate the growth of Bifidobacterium, have long been recognized to change the composition of the intestinal microbiota. Bifidogenic factors are typically complex carbohydrates (e.g., galactosyllactose in breast milk) that are not metabolized by the host or ­microbiota residing in the upper GIT. As a consequence, these factors reach the colon, where they are preferentially metabolized by the resident bifidobacteria. A ­ recent genome-sequencing project identified a number of breast milk carbohydrate metabolism genes present in B. longum subsp. infantis that putatively provide this strain with a competitive advantage in colonization of the infant GIT (84). Stimulation of the natural population levels of beneficial bacteria in this manner ­decreases colonic pH, stimulates mucosal immunity, and retards enteric infections (21). Expansion of this core concept toward the development of food ingredients that are metabolized selectively by one or

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

966 more groups of beneficial intestinal bacteria has been championed by Gibson and Roberfroid (29) and has led to the following definitions: Prebiotic: a nondigestible food ingredient that beneficially affects the host by selectively stimulating the growth and/or activity of one or a limited number of bacteria in the colon Synbiotic: a combination of a prebiotic ingredient with a probiotic culture Considering the specificity and uniqueness of the a­ utochthonous and normal microbiota, the use of prebiotics to stimulate the growth and activity of beneficial bacteria in an individual’s intestinal microbiota is a logical and effective approach to extend probiotic benefits. Food ingredients classified as prebiotics generally exhibit the following characteristics: • •

•

limited hydrolysis and absorption in the upper GIT selective growth stimulation of beneficial bacteria in the colon potential to repress pathogens and limit virulence via a number of processes including attenuation of virulence, immunostimulation, and stimulation of a beneficial flora that promotes colonization resistance

The best-known prebiotics are fructo-oligosaccharides derived from food sources. The largest natural source is inulin recovered by water extraction from the chic-

ory root. Inulin is composed of a glucose-nfructose polymer, with a degree of polymerization (DP, i.e., the number of fructose residues, n) that varies from 2 to 60 (average DP = 10). The bonds are 1-2 osidic linkages. Inulin can also be found in edible plants like onions, asparagus, bananas, wheat, and Jerusalem artichokes. Oligofructoses that contain glucose-nfructose moieties, plus polymeric fructose chains with DP values from 2 to 10, can be synthesized or generated as hydrolysis products of inulin. Most prebiotic compounds are bifidogenic in nature (Table 38.6). However, there are some exceptions. Notable among these is the finding by Kaplan and Hutkins (42) that some lactobacilli residing in the small intestine also use fructo-oligosaccharides. Studies with a variety of bifidogenic factors and prebiotics are providing evidence of health-related effects occurring via the prebiotic and its stimulated microbiota in the areas of colonization resistance, reduction of colon cancer markers in animals (enzymes and aberrant crypt foci), reduction of serum triglycerides, and enhanced adsorption of minerals (calcium, magnesium, iron, and zinc) (reviewed in reference 21). Genome-sequencing projects on probiotic lactobacilli indicate that the ability to metabolize indigestible carbohydrates in the GIT is an important attribute for survival in the GIT (8, 42, 83, 84). The current potential for prebiotics rests largely with compounds that can be extracted from foods

Table 38.6  Prebiotic compounds influencing members of the intestinal microfloraa Prebiotic factor

Origin

Microbes stimulated

Effects

Oligosaccharides

Onion, garlic, chicory root, ­burdock, asparagus, Jerusalem artichoke, soybean, wheat bran

Bifidobacterium species

Increase in number of bifidobacteria, suppression of putrefactive bacteria, prevention of constipation and diarrhea

Fructo-oligosaccharides (inulin, oligofructose)

See above

Bifidobacterium species, Lactobacillus ­acidophilus, Lactobacillus casei, Lactobacillus plantarum

Growth of bifidobacteria and acid promotion

Fructans

Ash-free white powder from tubers of Jerusalem artichoke

Bifidobacterium species

Growth promotion of Bifidobacterium

Human kappa casein and derived glycomacropeptide

Human milk: chymotrypsin and pepsin hydrolysate

Bifidobacterium bifidum

Growth promotion

Stachyose and raffinose

Soybean extract

Bifidobacterium species

Growth factor

Casein macropeptide

Bovine milk

Bifidobacterium species

Growth promotion

Lactitol (4-O-β-d-galactopyranosyl)-d- glucitol

Synthetic sugar alcohol of lactose

Bifidobacterium species

Growth promotion

Lactulose (4-O-β-d-galactopyranosyl)-d- fructose

Synthetic derivative of lactose

Bifidobacterium species

Growth promotion

Compiled from references 30, 78, and 92.

a

SMP_Food Microbiology_CH38.indd

966

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

967

and investigation of their impact on the resident intestinal microbiota and health-based markers. One area of ­exciting ­research is the production of designer prebiotics that may offer multiple activities in retarding undesirable microorganisms, better promoting the native desirable microbiota, or stimulating the growth or activity of synbiotic cultures (28). The following aspects are good targets for development of prebiotics: •

• •

• • • • • •

expand avenues for incorporation into appropriate food vehicles improved stimulation of beneficial microfloras antipathogenic, antiadhesive, and attenuative of pathogenic effects low-dosage forms synthesis from dietary polysaccharides noncariogenic properties good preservative and drying characteristics low caloric value controllable viscosity

•

•

•

•

•

•

stabilization of cultures and components for delivery in probiotic applications use of probiotic lactic acid bacteria for targeted delivery of novel bioactive compounds science-driven implementation of findings in commercial development assured safety of probiotic cultures, components, and food carriers identification of physiologically relevant biomarkers that can assess parameters of probiotic effectiveness (strain, dose, growth, and colonization potential) epidemiological and long-term studies in humans and animals to investigate the impact of probiotic cultures on diseases of longevity, e.g., colon cancer, heart disease, and inflammatory bowel diseases

The views expressed in this chapter do not necessarily reflect those of the U.S. Food and Drug Administration. Research at North Carolina State University on probi­ otic lactobacilli is supported by the N.C. Dairy Foundation, Danisco USA, Inc., and Dairy Management, Inc. Thanks are extended to Rodolphe Barrangou for his creative input into Fig. 38.4 and 38.5. E.A.P. acknowledges the support of the FDA Commissioner’s Fellowship Program.

CONCLUSIONS The probiotic field has offered considerable promise since the early observations by Metchnikoff about the importance of the GIT microbiota in limiting the development of putrefactive organisms and the potential role of lactic acid bacteria in establishing a healthy microbiota. Today, microbiological and molecular methods are providing critical new insights into probiotic bacteria, the normal microbiota, and the interactions between the two that may be responsible for eliciting beneficial outcomes realized in health and well-being. Proving or ­refuting the probiotic concept will be a worthy challenge to the most talented of scientists, due to the complexity of the host, the dynamic interactions within ­microbial ecosystems, and the multifaceted impacts of our associated microorganisms on health and wellbeing. Some of the key issues and important challenges for food microbiologists working on probiotics in the years ahead will be: •

•

•

•

taxonomic identification and genome sequencing of all probiotic cultures used in research, clinical trials, and commercial products definition of the active principles responsible for probiotic and abiotic activities correlation of genus, species, strain, phenotype, and genotype to specific probiotic functions determination of the impact of probiotics on the normal microbiota and associated host tissues

SMP_Food Microbiology_CH38.indd

967

References   1. Alander, M., R. Satokari, R. Korpela, M. Saxelin, T. Vilpponen‑Salmela, T. Mattila‑Sandholm, and A. von Wright. 1999. Persistence of colonization of human colonic mucosa by a probiotic strain, Lactobacillus rhamnosus GG, after oral consumption. Appl. Environ. Microbiol. 65:351–354.   2. Alfaleh, K., and D. Bassler. 2008. Probiotics for prevention of necrotizing enterocolitis in preterm infants. Cochrane Database Syst. Rev. (1):CD005496.   3. Allen, S. J., B. Okoko, E. Martinez, G. Gregorio, and L. F. Dans. 2004. Probiotics for treating infectious diarrhoea. Cochrane Database Syst. Rev. (2):CD003048.   3a. Altermann, E., L. B. Buck, R. Cano, and T. R. Klaenhammer. 2004. Identification and phenotypic characterization of the cell-division protein CdpA. Gene 342:189–197.   4. Altermann, E., W. M. Russell, M. A. Azcarate-Peril, R. Barrangou, B. L. Buck, O. McAuliffe, N. Souther, A. Dobsen, T. Doung, M. Callanan, S. Lick, A. Hamrick, R. Cano, and T. R. Klaenhammer. 2005. Complete genome sequence of the probiotic lactic acid bacterium Lactobacillus acidophilus NCFM. Proc. Natl. Acad. Sci. USA 102:3906–3912.   4a. Altschul, S. F., T. L. Madden, A. A. Schäffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389–3402.   5. Arbeit, R. D., J. C. Arbique, B. Beall, I. A. Critchley, F. J. Marsik, S. Michaud, C. Steward, F. C. Tenover, and D. L. Trees. ·2007. Molecular methods for bacterial

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

968 strain typing; approved guideline MM11-A. Clinical and Laboratory Standards Institute, Wayne, PA.   6. Association of American Feed Control Officials. 2010. AAFCO 2010 Official Publication. Association of American Feed Control Officials, Champaign, IL.   7. Azcarate-Peril, M. A., J. M. Bruno-Bárcena, H. M. Hassan, and T. R. Klaenhammer. 2006. Transcriptional and functional analysis of oxalyl-coenzyme A (CoA) decarboxylase and formyl-CoA transferase genes from Lactobacillus acidophilus. Appl. Environ. Microbiol. 72:1891–1899.   8. Barrangou, R., E. Altermann, R. Hutkins, R. Cano, and T. R. Klaenhammer. 2003. Functional and comparative genomic analyses of an operon involved in fructooligosaccharide utilization by Lactobacillus acidophilus. Proc. Natl. Acad. Sci. USA 100:8957–8962.   9. Bernardeau, M., M. Guguen, and J. P. Vernoux. 2006. Beneficial lactobacilli in food and feed: long-term use, biodiversity and proposals for specific and realistic safety assessments. FEMS Microbiol. Rev. 30:487–513.   10. Bernardeau, M., J. P. Vernoux, S. Henri-Dubernet, and M. Guéguen. 2008. Safety assessment of dairy microorganisms: the Lactobacillus genus. Int. J. Food Microbiol. 126:278–285.   11. Bibel, D. J. 1988. Elie Metchnikoff’s bacillus of long life. ASM News 54:661–665.   12. Boekhorst, J., R. J. Siezen, M. C. Zwahlen, D. Vilanova, R. D. Pridmore, A. Mercenier, M. Kleerebezem, W. M. de Vos, H. Brüssow, and F. Desiere. 2004. The complete genomes of Lactobacillus plantarum and Lactobacillus johnsonii reveal extensive differences in chromosome organization and gene content. Microbiology 150:3601–3611.   13. Boyle, R. J., F. J. Bath-Hextall, J. Leonardi-Bee, D. F. Murrell, and M. L. Tang. 2008. Probiotics for treating eczema. Cochrane Database Syst. Rev. (4):CD006135.   14. Braat, H., P. Rottiers, D. W. Hommes, N. Huyghebaert, E. Remaut, J. P. Remon, S. J. van Deventer, S. Neirynck, M. P. Peppelenbosch, and L. Steidler. 2006. A phase I trial with transgenic bacteria expressing interleukin-10 in Crohn’s disease. Clin. Gastroenterol. Hepatol. 4:754–759.   15. Buck, B. L., E. Altermann, T. Svingerud, and T. R. Klaenhammer. 2005. Functional analysis of putative adhesion factors in Lactobacillus acidophilus NCFM. Appl. Environ. Microbiol. 71:8344–8351.   16. Butterworth, A. D., A. G. Thomas, and A. K. Akobeng. 2008. Probiotics for induction of remission in Crohn’s disease. Cochrane Database Syst. Rev. (3):CD006634.   17. Cacas, I. A. 1998. Lactobacillus reuteri: an effective probiotic for poultry and other animals, p. 475–516. In S. Salminen and A. von Wright (ed.), Lactic Acid Bacteria: Microbiology and Functional Aspects. Marcel Dekker, New York, NY.   18. Chande, N., J. W. McDonald, and J. K. Macdonald. 2008. Interventions for treating collagenous colitis. Cochrane Database Syst. Rev. (2):CD006096.   19. Charteris, W. P., P. M. Kelly, L. Morelli, and J. K. Collins. 1998. Development and application of an in vitro methodology to determine the transit tolerance of potentially probiotic Lactobacillus and Bifidobacterium

SMP_Food Microbiology_CH38.indd

968

species in the upper human gastrointestinal tract. J. Appl. Bacteriol. 84:759–768.   20. Corr, S. C., Y. Li, C. U. Riedel, P. W. O’Toole, C. Hill, and C. G. Gahan. 2007. Bacteriocin production as a mechanism for the antiinfective activity of Lactobacillus salivarius UCC118. Proc. Natl. Acad. Sci. USA 104:7617–7621.   21. Crittenden, R. G. 1999. Prebiotics, p. 141–156. In G. W. Tannock (ed.), Probiotics: a Critical Review. Horizon Scientific Press, Norfolk, United Kingdom.   22. Doherty, G., G. Bennett, S. Patil, A. Cheifetz, and A. C. Moss. 2009. Interventions for prevention of post-operative recurrence of Crohn’s disease. Cochrane Database Syst. Rev. (4):CD006873.   23. Doyle, M. P., and M. C. Erickson. 2006. Reducing the carriage of foodborne pathogens in livestock and poultry. Poultry Sci. 85:960–973.   24. Dubos, R., R. W. Schaedler, R. Costello, and P. Hoet. 1965. Indigenous, normal and autochthonous flora of the gastrointestinal tract. J. Exp. Med. 122:67–76.   25. Eckburg, P. B., E. M. Bik, C. N. Bernstein, E. Purdom, L. Dethlefsen, M. Sargent, S. R. Gill, K. E. Nelson, and D. A. Relman. 2005. Diversity of the human intestinal microbial flora. Science 308:1635–1638.   26. Felis, G. E., and F. Dellaglio. 2007. Taxonomy of lactobacilli and bifidobacteria. Curr. Issues Intest. Microbiol. 8:44–61.   27. Fuller, R. 1999. Probiotics for farm animals, p. 15–22. In G. W. Tannock (ed.), Probiotics: a Critical Review. Horizon Scientific Press, Norfolk, United Kingdom.   28. Gibson, G. R. 1998. Dietary modulation of the human gut microflora using prebiotics. Br. J. Nutr. 80(Suppl. 2): S209–S212.   29. Gibson, G. R., and M. B. Roberfroid. 1995. Dietary modulation of the human colonic microbiota: introducing the concept of prebiotics. J. Nutr. 125:1401–1412.   30. Gomes, A. M. P., and F. X. Malcata. 1999. Bifidobac­ terium spp. and Lactobacillus acidophilus: biological, biochemical, technological and therapeutical properties relevant for use as probiotics. Trends Food Sci. Technol. 10:139–157.   31. Grangette, G., S. Nutten, E. Palumbo, S. Morath, C. Hermann, J. Dewulf, B. Pot, T. Hartung, P. Hols, and A. Mercenier. 2005. Enhanced antiinflammatory capacity of a Lactobacillus plantarum mutant synthesizing modified teichoic acids. Proc. Natl. Acad. Sci. USA 102:10321–10326.   32. Gurusamy, K. S., Y. Kumar, and B. R. Davidson. 2008. Methods of preventing bacterial sepsis and wound complications for liver transplantation. Cochrane Database Syst. Rev. (4):CD006660.   33. Haller, D., J. M. Antoine, S. Bengmark, P. Enck, G. T. Rijkers, and I. Lenoir-Wijnkoop. 2010. Guidance for substantiating the evidence for beneficial effects of probiotics: probiotics in chronic inflammatory bowel disease and the functional disorder irritable bowel syndrome. J. Nutr. 140:690S–697S.   34. Holubar, S. D., R. R. Cima, W. J. Sandborn, and D. S. Pardi. 2010. Treatment and prevention of pouchitis after

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

969

ileal pouch-anal anastomosis for chronic ulcerative colitis. Cochrane Database Syst. Rev. (6):CD001176.   35. Holzapfel, W. H., P. Haberer, R. Geisen, J. Björkroth, and U. Schillinger. 2001. Taxonomy and important features of probiotic microorganisms in food and nutrition. Am. J. Clin. Nutr. 73(2 Suppl.):365S–373S.   36. Hosono, A., J. Lee, A. Ametani, M. Natsume, M. Hirayama, T. Adachi, and S. Kaminogawa. 1997. Characterization of a water-soluble polysaccharide fraction with immunopotentiating activity from Bifidobacterium adolescentis M101-4. Biosci. Biotech. Biochem. 61:312–316.   37. Huertas-Ceballos, A. A., S. Logan, C. Bennett, and C. Macarthur. 2009. Dietary interventions for recurrent abdominal pain (RAP) and irritable bowel syndrome (IBS) in childhood. Cochrane Database Syst. Rev. (1): CD003019.   38. Hume, M. E., D. E. Corrier, D. J. Nisbet, and J. R. DeLoach. 1998. Early Salmonella challenge time and reduction in chick cecal colonization following treatment with a characterized competitive exclusion culture. J. Food Prot. 61:673–676.   39. Johnson, J. L., C. F. Phelps, C. S. Cummins, J. London, and F. Gasser. 1980. Taxonomy of the Lactobacillus aci­ dophilus group. Int. J. Syst. Bacteriol. 30:53–68.   40. Johnston, B. C., A. L. Supina, M. Ospina, and S. Vohra. 2007. Probiotics for the prevention of pediatric antibiotic-associated diarrhea. Cochrane Database Syst. Rev. (2):CD004827.   41. Joint FAO/WHO Working Group. 2002. Guidelines for the Evaluation of Probiotics in Food. Food and Agriculture Organization/World Health Organization, Rome, Italy and Geneva, Switzerland.   42. Kaplan, H., and R. W. Hutkins. 2000. Fermentation of fructooligosaccharides by lactic acid bacteria and bifidobacteria. Appl. Environ. Microbiol. 66:2682–2684.   43. Kataria, J., N. Li, J. L. Wynn, and J. Neu. 2009. Probiotic microbes: do they need to be alive to be beneficial? Nutr. Rev. 67:546–550.   44. Klaenhammer, T. R. 1998. Functional activities of Lactobacillus probiotics: genetic mandate. Int. Dairy J. 8:497–506.   45. Klaenhammer, T. R., and M. J. Kullen. 1999. Selection and design of probiotics. Int. J. Food Microbiol. 50:45–58.   46. Klaenhammer, T. R., and W. M. Russell. 2000. Species of the Lactobacillus acidophilus complex, p. 1151– 1157. In R. K. Robinson, C. Batt, and P. D. Patel (ed.), Encyclopedia of Food Microbiology, vol. 2. Academic Press, San Diego, CA.   47. Klaenhammer, T. R., R. Barrangou, B. L. Buck, M. A. Azcarate-Peril, and E. Altermann. 2005. Genomic features of lactic acid bacteria effecting bioprocessing and health. FEMS Microbiol. Rev. 29:393–409.   48. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. Siezen. 2003. Complete genome sequence of

SMP_Food Microbiology_CH38.indd

969

Lactobacillus plantarum WCFS1 Proc. Natl. Acad. Sci. USA 100:1990–1995.   49. Konstantinov, S. R., H. Smidt, W. M. de Vos, S. C. Bruijns, S. K. Singh, F. Valence, D. Molle, S. Lortal, E. Altermann, T. R. Klaenhammer, and Y. van Kooyk. 2008. S layer protein A of Lactobacillus acidophilus NCFM regulates immature dendritic cell and T cell functions. Proc. Natl. Acad. Sci. USA 105:19474–19479.   50. Kullen, M. J., L. J. Brady, and D. J. O’Sullivan. 1997. Evaluation of using a short region of the recA gene for rapid and sensitive speciation of dominant bifidobacteria in the human large intestine. FEMS Microbiol. Lett. 154:377–383.   51. Kullen, M. J., R. B. Sanozky‑Dawes, D. C. Crowell, and T. R. Klaenhammer. 2000. Use of DNA sequence of variable regions of the 16S rRNA gene for rapid and accurate identification of bacteria in the Lactobacillus acidophilus complex. J. Appl. Microbiol. 89:511–518.   52. Lauer, E., C. Helming, and O. Kandler. 1980. Hetero­ geneity of the species Lactobacillus acidophilus (Moro) Hansen and Moquot as revealed by biochemical characteristics and DNA‑DNA hybridization. Zentbl. Bakteriol. Mikrobiol. Hyg. 1 Abt. Orig. C1:150–168.   53. Lebeer, S., J. Vanderleyden, and S. C. De Keersmaecker. 2008. Genes and molecules of lactobacilli supporting probiotic action. Microbiol. Mol. Biol. Rev. 72:728–764.   54. Lebeer, S., J. Vanderleyden, and S. C. De Keersmaecker. 2010. Host interactions of probiotic bacterial surface molecules: comparison with commensals and pathogens. Nature Rev. Microbiol. 8:171–184.   55. Lewanika, T. R., S. J. Reid, V. R. Abratt, G. T. Macfarlane, and S. Macfarlane. 2007. Lactobacillus gasseri Gasser AM63T degrades oxalate in a multistage continuous culture simulator of the human colonic microbiota. FEMS Microbiol. Ecol. 61:110–120.   56. Ley, R. E. 2010. Obesity and the human microbiome. Curr. Opin. Gastroenterol. 26:5–11.   57. Lilly, D. M., and R. H. Stillwell. 1965. Probiotics: growth promoting factors produced by microorganisms. Science 147:747–748.   58. Lirussi, F., E. Mastropasqua, S. Orando, and R. Orlando. 2007. Probiotics for non-alcoholic fatty liver disease and/or steatohepatitis. Cochrane Database Syst. Rev. (1):CD005165.   59. Makarova, K. S., and E. V. Koonin. 2007. Evolutionary genomics of lactic acid bacteria. J. Bacteriol. 189: 1199–1208.   60. Mallon, P., D. McKay, S. Kirk, and K. Gardiner. 2007. Probiotics for induction of remission in ulcerative colitis. Cochrane Database Syst. Rev. (4):CD005573.   61. Marteau, P., M. Minekus, R. Havenaar, and J. H. H. Huis in’t Veld. 1997. Survival of lactic acid bacteria in a dynamic model of the stomach and small intestine: validation and the effect of bile. J. Dairy Sci. 80:1031–1037.   62. Mitsuoka, T. 1992. The human gastrointestinal tract, p. 69–114. In B. J. B. Wood (ed.), The Lactic Acid Bacteria, vol. 1. The Lactic Acid Bacteria in Health and Disease. Elsevier Applied Science, London, United Kingdom.

Manila Typesetting Company

11/08/2012 07:19AM

Fermentations and Beneficial Microorganisms

970   63. Molenaar, D. F., F. H. Bringel, W. M. Schuren, W. M. de Vos, R. J. Siezen, and M. Kleerebezem. 2005. Exploring Lactobacillus plantarum genome diversity by using microarrays. J. Bacteriol. 187:6128–6136.   64. Nisbet, D. J., G. I. Tellez, V. K. Lowry, R. C. Anderson, G. Garcia, G. Nava, M. H. Kogut, D. E. Corrier, and L. H. Stanker. 1998. Effect of a commercial competitive exclusion culture (Preempt) on mortality and horizontal transmission of Salmonella gallinarum in broiler chickens. Avian Dis. 42:651–656.   65. Nurmi, E., and M. Rantala. 1973. New aspects of Salmonella infection in broiler production. Nature (London) 241:210–211.   66. Oduyebo, O. O., R. I. Anorlu, and F. T. Ogunsola. 2009. The effects of antimicrobial therapy on bacterial vaginosis in non-pregnant women. Cochrane Database Syst. Rev. (3):CD006055.   67. Osborn, D. A., and J. K. Sinn. 2007. Probiotics in infants for prevention of allergic disease and food hypersensitivity. Cochrane Database Syst. Rev. (4):CD006475.   68. O’Sullivan, D. J. 1999. Methods for the analysis of the intestinal microflora, p. 23–44. In G. W. Tannock (ed.), Probiotics: a Critical Review. Horizon Scientific Press, Norfolk, United Kingdom.   69. Othman, M., J. P. Neilson, and Z. Alfirevic. 2007. Probiotics for preventing preterm labor. Cochrane Database Syst. Rev. (1):CD005941.   70. O’Toole, P. W., and J. C. Cooney. 2008. Probiotic bacteria influence the composition and function of the intestinal microbiota. Interdiscip. Perspect. Infect. Dis. 2008:175285.   71. Pillai, A., and R. Nelson. 2008. Probiotics for treatment of Clostridium difficile-associated colitis in adults. Cochrane Database Syst. Rev. (1):CD004611.   72. Puhan, Z. 1999. Effect of probiotic fermented dairy products in human nutrition. Ind. Latte 35:3–11.   73. Qin, J., R. Li, J. Raes, M. Arumugam, K. S. Burgdorf, C. Manichanh, T. Nielsen, N. Pons, F. Levenez, T. Yamada, D. R. Mende, J. Li, J. Xu, S. Li, D. Li, J. Cao, B. Wang, H. Liang, H. Zheng, Y. Xie, J. Tap, P. Lepage, M. Bertalan, J. M. Batto, T. Hansen, D. Le Paslier, A. Linneberg, H. B. Nielsen, E. Pelletier, P. Renault, T. Sicheritz-Ponten, K. Turner, H. Zhu, C. Yu, S. Li, M. Jian, Y. Zhou, Y. Li, X. Zhang, S. Li, N. Qin, H. Yang, J. Wang, S. Brunak, J. Doré, F. Guarner, K. Kristiansen, O. Pedersen, J. Parkhill, J. Weissenbach, MetaHIT Consortium, P. Bork, S. D. Ehrlich, and J. Wang. 2010. A human gut microbial gene catalogue established by metagenomic sequencing. Nature 464:59–65.   74. Radloff, J. 1998. Spray guards chicks from infections. Sci. News 153:196.   75. Rastall, R. A., G. R. Gibson, H. S. Gill, F. Guarner, T. R. Klaenhammer, B. Pot, G. Reid, I. R. Rowland, and M. E. Sanders. 2005. Modulation of the microbial ecology of the human colon by probiotics, prebiotics and synbiotics to enhance human health: an overview of enabling ­science and potential applications. FEMS Microbiol. Ecol. 52:145–152.

SMP_Food Microbiology_CH38.indd

970

  76. Rijkers, G. T., S. Bengmark, P. Enck, D. Haller, U. Herz, M. Kalliomaki, S. Kudo, I. Lenoir-Wijnkoop, A. Mercenier, E. Myllyluoma, S. Rabot, J. Rafter, H. Szajewska, B. Watzl, J. Wells, D. Wolvers, and J. M. Antoine. 2010. Guidance for substantiating the evidence for beneficial effects of probiotics: current status and recommendations for future research. J. Nutr. 140:671S–676S.   77. Rolfe, V. E., P. J. Fortun, C. J. Hawkey, and F. BathHextall. 2009. Probiotics for maintenance of remission in Crohn’s disease. Cochrane Database Syst. Rev. (4): CD004826.   78. Salminen, S., M. Roberfroid, P. Ramos, and R. Fonden. 1998. Prebiotic substrates and lactic acid bacteria, p. 343–358. In S. Salminen and A. von Wright (ed.), Lactic Acid Bacteria: Microbiology and Functional Aspects. Marcel Dekker, New York, NY.   79. Sanders, M. E. 1999. Probiotics—scientific status summary. Food Technol. 53:67–77.   80. Sanders, M. E., and J. Huis in’t Veld. 1999. Bringing a probiotic-containing functional food to market: microbiological, product, regulatory, and labeling issues. Antonie van Leeuwenhoek 76:293–315.   81. Sandholm, T. M., S. Blum, J. K. Collins, R. Crittenden, W. de Vos, C. Dunne, R. Fonden, G. Grenov, E. Isolauri, B. Kiely, P. Marteau, L. Morelli, A. Ouwehand, R. Reniero, M. Saarela, S. Salminen, M. Saxelin, E. Schiffrin, F. Shanahan, E. Vaughan, and A. von Wright. 1999. Probiotics: towards demonstrating efficacy. Trends Food Sci. Technol. 10:393–399.   82. Saxelin, M., S. Tynkkynen, T. Mattila-Sandholm, and W. M. de Vos. 2005. Probiotic and other functional microbes: from markets to mechanisms. Curr. Opin. Biotechnol. 16:204–211.   83. Schell, M. A., M. Karmirantzou, B. Snel, D. Vilanova, B. Berger, G. Pessi, M. C. Zwahlen, F. Desiere, P. Bork, M. Delley, R. D. Pridmore, and F. Arigoni. 2002. The genome sequence of Bifidobacterium longum reflects its adaptation to the human gastrointestinal tract. Proc. Natl. Acad. Sci. USA 99:14422–14427.   84. Sela, D. A., J. Chapman, A. Adeuya, J. H. Kim, F. Chen, T. R. Whitehead, A. Lapidus, D. S. Rokhsar, C. B. Lebrilla, J. B. German, N. P. Price, P. M. Richardson, and D. A. Mills. 2008. The genome sequence of Bifidobacterium longum subsp. infantis reveals adaptations for milk utilization within the infant microbiome. Proc. Natl. Acad. Sci. USA 105:18964–18969.   85. Sela, D. A., and D. A. Mills. 2010. Nursing our microbiota: molecular linkages between bifidobacteria and milk oligosaccharides. Trends Microbiol. 18:298–307.   86. Senok, A. C., H. Verstraelen, M. Temmerman, and G. A. Botta. 2009. Probiotics for the treatment of bacterial vaginosis. Cochrane Database Syst Rev. (4): CD006289.   87. Shida, K., and M. Nanno. 2008. Probiotics and immunology: separating the wheat from the chaff. Trends Immunol. 29:565–573.   88. Shortt, C. 1998. Living it up for dinner. Chem. Ind. 20:300–303.

Manila Typesetting Company

11/08/2012 07:19AM

38.  Probiotics and Prebiotics

971

  89. Shortt, C. 1999. The probiotic century: historical and current perspectives. Trends Food Sci. Technol. 10: 411–417.   90. Steidler, L., W. Hans, L. Schotte, S. Neirynck, F. Obermeier, W. Falk, W. Fiers, and E. Remaut. 2000. Treatment of murine colitis by Lactococcus lactis secreting IL‑10. Science 289:1352–1355.   91. Suau, A., R. Bonnet, M. Sutren, J.-J. Godon, G. R. Gibson, M. D. Collins, and J. Dore. 1999. Direct analysis of genes encoding 16S rRNA from complex communities reveals many novel molecular species. Appl. Environ. Microbiol. 65:4799–4807.   92. Tamine, A. Y. 1997. Bifidobacteria—an overview of physiological, biochemical, and technological aspects, p. 9. In R. Harmink (ed.), Non-digestible Oligosaccha­ rides: Healthy Food for the Colon. Krukkerij Modern, Bennekonm, The Netherlands.   93. Tannock, G. W. 1995. The Normal Microflora. Chapman and Hall, London, United Kingdom.   94. Tannock, G. W. 1999. Probiotics: a Critical Review. Horizon Scientific Press, Norfolk, United Kingdom.   95. Tannock, G. W. 1999. Analysis of the intestinal microflora: a renaissance. Antonie van Leeuwenhoek 76: 265–278.   96. Turnbaugh, P. J., R. E. Ley, M. Hamady, C. M. FraserLiggett, R. Knight, and J. I. Gordon. 2007. The human microbiome project. Nature 449:804–810.   97. Turner, J. R. 2009. Intestinal mucosal barrier function in health and disease. Nat. Rev. Immunol. 9:799–809.   98. Vandamme, P., B. Pot, M. Gillis, P. de Vos, K. Kersters, and J. Swings. 1996. Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiol. Rev. 60:407–438.   99. Vaughan, E. E., H. G. H. J. Heilig, E. G. Zoetendal, R. Satokari, K. Collins, A. D. L. Akkermans, and W. M. de Vos. 1999. Molecular approaches to study probiotic bacteria. Trends Food Sci. Technol. 10:400–404. 100. Vaughan, E. E., M. C. de Vries, E. G. Zoetendal, K. Ben-Amor, A. D. Akkermans, and W. M. de Vos. 2002. The intestinal LABs. Antonie van Leeuwenhoek 82:341–352. 101. Veiga, P., C. A. Gallini, C. Beal, M. Michaud, M. L. Delaney, A. Dubois, A. Khlebnikov, J. E. van Hylckama Vlieg, S. Punit, J. N. Glickman, A. Onderdonk, L. H. Glimcher, and W. S. Garrett. 2010. Bifidobacterium ani­ malis subsp. lactis fermented milk product reduces inflammation by altering a niche for colitogenic microbes. Proc. Natl. Acad. Sci. USA 107:18132–18137.

SMP_Food Microbiology_CH38.indd

971

102. Ventura, M., D. van Sinderen, G. F. Fitzgerald, and R. Zink. 2004. Insights into the taxonomy, genetics and physiology of bifidobacteria. Antonie van Leeuwenhoek 86:205–223. 103. Ventura, M., S. O’Flaherty, M. J. Claesson, F. Turroni, T. R. Klaenhammer, D. van Sinderen, and P. W. O’Toole. 2009. Genome-scale analyses of health-promoting bacteria: probiogenomics. Nat. Rev. Microbiol. 7:61–71. 104. Walter, J., G. W. Tannock, A. Tilsala-Timisjarvi, S. Rodtong, D. M. Loach, K. Munro, and T. Alatossava. 2000. Detection and identification of gastrointestinal Lactobacillus species by using denaturing gradient gel electrophoresis and species-specific PCR primers. Appl. Environ. Microbiol. 66:297–303. 105. Wang, H. H., M. Manuzon, M. Lehman, K. Wan, H. Luo, T. E. Wittum, A. Yousef, and L. O. Bakaletz. 2006. Food commensal microbes as a potentially important avenue in transmitting antibiotic resistance genes. FEMS Microbiol. Lett. 254:226–231. 106. Wells, J. M., K. Robinson, L. M. Chamberlain, K. M. Schofield, and R. W. LePage. 1996. Lactic acid bacteria as vaccine delivery vehicles. Antonie van Leeuwenhoek 70:317–330. 107. Wells, J. M., and A. Mercenier. 2008. Mucosal delivery of therapeutic and prophylactic molecules using lactic acid bacteria. Nat. Rev. Microbiol. 6:349–362. 108. Wolvers, D., J. M. Antoine, E. Myllyluoma, J. Schrezenmeir, H. Szajewska, and G. T. Rijkers. 2010. Guidance for substantiating the evidence for beneficial effects of probiotics: prevention and management of ­infections by probiotics. J. Nutr. 140:698S–712S. 109. Wynn, S. G. 2009. Probiotics in veterinary practice. J. Am. Vet. Med. Assoc. 234:606–613. 110. Yan, F., and D. B. Polk. 2010. Probiotics: progress toward novel therapies for intestinal diseases. Curr. Opin. Gastroenterol. 26:95–101. 111. Zhao, T., M. P. Doyle, B. G. Harmon, C. A. Brown, P. O. Mueller, and A. H. Parks. 1998. Reduction of carriage of enterohemorrhagic Escherichia coli O157:H7 in cattle by inoculation with probiotic bacteria. J. Clin. Microbiol. 36:641–647. 112. Zoetendal, E. G., A. D. L. Akkermans, and W. M. de Vos. 1998. Temperature gradient gel electrophoresis analysis of 16S rRNA from human fecal samples reveals sand host-specific communities of active bacteria. Appl. Environ. Microbiol. 64:3854–3859.

Manila Typesetting Company

11/08/2012 07:19AM

SMP_Food Microbiology_CH38.indd

972

Manila Typesetting Company

11/08/2012 07:19AM

Advanced Techniques in Food Microbiology

VII

SMP_Food Microbiology_CH39.indd

973

Manila Typesetting Company

11/08/2012 07:20AM

SMP_Food Microbiology_CH39.indd

974

Manila Typesetting Company

11/08/2012 07:20AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch39

Grace L. Douglas Erika Pfeiler Tri Duong Todd R. Klaenhammer

Genomics and Proteomics of Foodborne Microorganisms

Foods are teeming with microorganisms that may be innocuous, pathogenic threats, spoilage agents, or beneficial microorganisms driving fermentations or acting as biocontrol agents. Historically, the overarching priorities in this field have been the destruction and control of undesirable organisms in foods and the promotion of growth and activity of desirable microbes. Recommendations from an American Academy of Microbiology Report entitled Research Opportunities in Food and Agriculture Microbiology have emphasized the importance of studying the distribution of microorganisms and microbial communities in foods; understanding the nature, specificity, and adaptation of microorganisms to food environments and their human, animal, and plant hosts; and investigating the impact of production and processing practices on the evolution, persistence, resistance, and flow of microbes in foods (30). More generally stated, the recommendation promotes understanding the interaction of microbes with our foods, with their hosts, and with the processing and storage environments created by mankind. Facing this challenge, genomics and proteomics now underlie a renaissance in food microbiology that will impact every facet of how we will understand and investigate microorganisms associated with our foods. Food

39

microbiologists investigating pathogens, commensals, or starter cultures will rely extensively on DNA and protein sequence information to define genetic content, construct metabolic pathways, and predict microbial substrates and products. In addition, microarrays, next-generation sequencing technologies, and gene expression profiles can now be exploited to investigate the responses of microbes to food environments, processing conditions, host ecosystems, and the presence of other organisms. Emerging technologies in genomics and proteomics will also accelerate food safety screening and detection and identification of pathogens during food outbreaks or acts of bioterrorism involving the food system (17). This chapter will outline the basic concepts underlying genomics, proteomics, and associated technologies. Furthermore, selected examples will illustrate how these approaches have already provided an in-depth understanding of important microbial properties, behavior, processes, or interactions within food environments.

DNA SEQUENCING The heart of all genomics research lies in DNA sequencing. Until the past few years, most genome-scale sequencing

Grace L. Douglas, Erika Pfeiler, Tri Duong, and Todd R. Klaenhammer, Department of Food, Bioprocessing & Nutrition Sciences, North Carolina State University, Raleigh, NC 27695-7624.

975

SMP_Food Microbiology_CH39.indd

975

Manila Typesetting Company

11/08/2012 07:20AM

976

Advanced Techniques in Food Microbiology

was carried out using the chain termination technique developed by Fredrick Sanger in 1974, which along with his work in sequencing a phage genome earned him the Nobel Prize in Chemistry in 1980. This method generally begins with DNA fragments cloned into vectors from which they may be individually amplified. The DNA is replicated using a template strand, a primer, DNA polymerase, and a mix of deoxynucleotide triphosphates (dNTPs). Dideoxynucleotide triphosphates (ddNTPs) are added to the reaction because the absence of a 3' hydroxyl in these nucleotides stops strand growth (Fig. 39.1). Therefore, four separate reactions can be set up using the same template DNA and the four ddNTPs, A, T, G, and C. When the products of these reactions are run on an acrylamide gel, they are separated by size and sequences can then be determined by reading the bands up the gel. While manual sequencing works for short pieces of DNA, sequencing an entire genome requires a more highthroughput approach, and in the 1990s, such a setup was made possible. First, the development of ddNTPs that each have a different fluorescent label made it possible to sequence from a single reaction instead of four. Second, capillary electrophoresis was introduced whereby the reaction products are run through a capillary tube where a scanner reads the fluorescence wavelength coming from each amplicon fragment that passes by. Software programs are then used to correlate the wavelength reading with the appropriate

nucleotide, and also to assemble the sequence from the approximately 600-bp lengths of sequence into an entire genome. This is accomplished by matching overlapping sequences in different contigs and assembling those to create longer and fewer contigs until the genome is closed. Genomes of microorganisms of foodborne significance that have been sequenced are listed in Table 39.1.

NH2 O

dNTP

HO

N

P

N

OH

O

O

N

N

OH NH2 O

ddNTP

HO

P

N

N

OH

O

O

N

N

H

Figure 39.1  dNTP (top) has a 3' hydroxyl present on the deoxyribose that ddNTP (bottom) does not. This stops DNA strand growth because DNA polymerase no longer has a way to connect the bases in the growing strand. doi:10.1128/9781555818463.ch39f1

SMP_Food Microbiology_CH39.indd

976

Next-Generation Sequencing

Over the past few years, high-throughput sequencing technologies that continually decrease the time and cost required for sequencing have emerged. These technologies are capable of massively parallel sequencing, producing millions of reads in one run compared to Sanger sequencing, which produces 96 reads per run. Three next-generation technologies commercially available are the 454 GS FLX developed by Roche, the Solexa GA from Illumina, and SOLiD from Applied Biosystems (57) (Fig. 39.2). Many new sequencing projects are utilizing these technologies. Roche developed the 454 GS20 pyrosequencing platform in 2005, and since then 454 sequence reads have increased from 100 nucleotides to 400 nucleotides with technology upgrades. While this is still shorter than the 600- to 700-nucleotide reads obtained with Sanger sequencing, the large number of high-quality (>99% accurate) reads generated by 454 is significantly greater than that obtained with Sanger sequencing (57, 59). Roche 454 whole-genome sequencing begins with fragmentation of the genome. Adaptors are ligated to each end of the single-stranded fragments, which are then added to beads under dilute conditions promoting attachment of one fragment per bead. The beads are suspended individually in water droplets in a water-in-oil emulsion, where a PCR amplification occurs with primers complementary to the adaptors. After the amplified fragments are denatured to single-stranded DNA, each bead is transferred to its own picoliter-sized well on a fiber-optic slide along with beads carrying enzymes for pyrosequencing (59). Pyrosequencing is accomplished in each well simultaneously by adding one of the four dNTPs at a time. As a nucleotide is incorporated, inorganic pyrophosphate is released and used to produce energy in the form of ATP for photon generation. The photon intensity is detected from each well and correlates to the number of nucleotides incorporated. Excess nucleotides are washed away before the next one is added (57, 59). The two other high-throughput methods provide shorter reads than the 454 system but much higher coverage. Illumina’s Solexa GA platform also utilizes

Manila Typesetting Company

11/08/2012 07:20AM

977

39.  Genomics and Proteomics of Foodborne Microorganisms Table 39.1  Microorganisms of foodborne significance with completed or in-progress genome sequencing projectsa Bacteria

Source

No. of strains sequenced (no. in progress)

Industrial/beneficial food microbes   Bifidobacterium animalis subsp. lactis

Probiotic

3 (1)

  Bifidobacterium longum

Probiotic

2 (2)

  Lactobacillus acidophilus

Probiotic

1 (1)

  Lactobacillus brevis

Starter culture

1

  Lactobacillus casei

Probiotic/dairy starter culture

3

  Lactobacillus delbrueckii subsp. bulgaricus

Dairy starter culture

2 (1)

  Lactobacillus fermentum

Probiotic

1 (2)

  Lactobacillus gasseri

Probiotic

1 (4)

  Lactobacillus helveticus

Dairy starter culture

1 (1)

  Lactobacillus johnsonii

Probiotic

1 (1)

  Lactobacillus plantarum

Probiotic/vegetable starter culture

2 (1)

  Lactobacillus reuteri

Probiotic

2 (5)

  Lactobacillus rhamnosus

Probiotic

2 (2)

  Lactobacillus sakei subsp. sakei

Meat starter culture

1

  Lactobacillus salivarius

Probiotic

1 (1)

  Lactococcus lactis subsp. cremoris

Dairy starter culture

2

  Lactococcus lactis subsp. lactis

Dairy starter culture

2

  Leuconostoc mesenteroides subsp. mesenteroides

Vegetable fermentation

1 (1)

  Oenococcus oeni

Wine fermentation

1 (2)

  Streptococcus thermophilus

Dairy starter culture

3

  Bacillus cereus

Meats, milk, cheese, vegetables, fish

9 (35)

  Campylobacter jejuni

Raw beef, poultry, raw milk, eggs

5 (10)

  Clostridium botulinum

Meats, fish, canned foods

  Clostridium perfringens

Meats

3 (6)

  Escherichia coli O157:H7

Ground beef, raw milk, vegetables

4 (15)

  Listeria monocytogenes

Meats, poultry, milk, cheese, vegetables

  Salmonella enterica subsp. enterica

Meats, poultry, eggs, and milk

  Shigella spp.

Vegetables, meat, poultry, milk, water

8 (38)

  Vibrio cholerae

Shellfish and fish

8 (24)

Foodborne pathogens

a

10 (5)

6 (20) 16 (23)

Reprinted with permission from O’Flaherty and Klaenhammer, 2011 (69). (Source: http://www.ncbi.nlm.nih.gov/genomes/lproks.cgi. August 2010).

ligation of adaptors to fragmented single-stranded DNA, but the fragments are then attached in isolated spots on a solid surface that contains primers complementary to the adapters. After PCR, the solid surface contains millions of isolated colonies of amplified single-stranded DNA. Sequencing takes place by synthesis when all four dNTPs, each with a differ-

SMP_Food Microbiology_CH39.indd

977

ent fluorescent reversible dye terminator, are added simultaneously. Only one dNTP may be added at a time due to the terminator, which is removed after each cycle (11, 57). Read lengths are typically between 35 and 70 nucleotides. The Applied Biosystems SOLiD platform is based on sequencing by ligation (78). First, adaptors are ligated

Manila Typesetting Company

11/08/2012 07:20AM

978

Advanced Techniques in Food Microbiology Single-stranded DNA

Adaptors ligated to DNA

DNA immobilized on bead and amplified in water–oil emulsion

a 454

Picolitre well DNA polymerase and enzymes on beads

Light

ATP

b SOLiD

Random oligonucleotides with known 3’ dinucleotide

Primer to adaptor

• 35–50 nucleotides • 171 million reads • 6,000 Mb

Adaptor sequence

First base

Coding scheme second base

Known base from adaptor

c Solexa GA

Amplified DNA spots

Unamplified immobilized DNA

e Pacific Biosciences

Terminator dNTP

• 50 nucleotides • 30 million reads • 1,500 Mb

DNA polymerase

Immobilized on substrate

d Heliscope

• 250 nucleotides • 400,000 reads • 100 Mb

Single species dNTP

Single species dNTP

• 50 nucleotides • 30 million reads • 1,500 Mb

DNA polymerase

Highly focused detection path

Detection volume Immobilized DNA polymerase

Figure 39.2  (a) Schematic representation of 454 GS FLX pyrosequencing. Oligonucleotide adapters are ligated to fragmented DNA and immobilized to the surface of microscopic beads before PCR amplification in an oil-water droplet emulsion. Beads are isolated in picoliter wells and incubated with dNTPs, DNA polymerase, and beads bearing enzymes for the chemiluminescent reaction. Incorporation of a nucleotide into the complementary strand releases pyrophosphate, which is used to produce ATP. This, in turn, provides the energy for the generation of light. The light emitted is recorded as an image for analysis. (b) SOLiD sample preparation is similar to that of 454 pyrosequencing. After amplification, the beads are immobilized onto a custom substrate. A primer that is complementary to the adapter sequence (green), random oligonucleotides with known 3´ dinucleotides, and a corresponding fluorophore are hybridized sequentially along the sequence, and image data are collected. After five repeats, the complementary strand is melted away and a new primer is added to the adapter sequence, ending at a position one nucleotide upstream of the previous primer. Second-strand synthesis is repeated, allowing two-color encoding and double reading of each of the target nucleotides. Repeats of these cycles ensure that nucleotides in the gap between known dinucleotides are read. Knowledge of the first base in the adapter reveals the dinucleotide, using the color-space scheme. In other words, knowing that the last adapter nucleotide is T and the color is red means that the first base to be sequenced must be A. Knowing that the first base is A and the color is green means that the next base must be C, and so on. (c) For Solexa GA sequencing, adapters are ligated onto DNA and used to anchor the fragments to a prepared substrate. Fold-back PCR results in isolated spots of DNA of a large enough quantity that the amassed fluorophore can be detected. Terminator nucleotides and DNA polymerase are then used to create cDNA. Images are collected at the end of each cycle before the terminator is removed. (d) Heliscope sequencing immobilizes unamplified DNA with ligated adapters to a substrate. Each species of dNTP with a bright fluorophore attached is used sequentially to create second-strand DNA; a “virtual terminator” prevents the inclusion of more than one nucleotide per strand and cycle, and background signal is reduced by removal of “used” fluorophore at the start of each cycle. (e) The new sequencing method developed by Pacific Biosciences occurs in zeptoliter wells that contain an immobilized DNA polymerase. DNA and dNTPs are added for synthesis. Fluorophores are cleaved from the complementary strand as it grows and diffuse away, allowing single nucleotides to be read. Continuous detection of fluorescence in the detection volume and high dNTP concentration allow extremely fast and long reading. Adapted by permission from Macmillan Publishers Ltd., Nature Reviews Microbiology, vol. 57, copyright 2009. doi:10.1128/9781555818463.ch39f2

SMP_Food Microbiology_CH39.indd

978

Manila Typesetting Company

11/08/2012 07:20AM

979

39.  Genomics and Proteomics of Foodborne Microorganisms to the ends of DNA fragments and PCR takes place in a water-in-oil emulsion similar to that of the 454 system. Sequencing is begun with a primer complementary to the adapter. Fluorophores are associated with known 3¢ dinucleotides, which are attached to different combinations of degenerate oligomers. These nonamers hybridize to complementary bases on the fragments and are ligated together. Every fluorophore detected reveals the identity of two bases. The specificity provided by hybridization of a nonamer increases accuracy. After the first read, the process is repeated using a primer 1 base pair upstream of the previous primer. This is repeated until the entire sequence is obtained (57). Sequence assembly from shorter reads is difficult, especially in areas where the sequence is repetitive. The longer reads now provided by 454 are beginning to make the technology more compatible with de novo assembly, while the shorter reads and high coverage obtained with Solexa and SOLiD are more suitable for resequencing, where sequence fragments may be mapped to a reference genome (63). Resequencing has many applications in food microbiology. It is useful to separate species with sequences too similar to differentiate with traditional methods, such as 16S rRNA sequencing and multilocus sequence typing, which rely on differences in 16S rRNA and housekeeping gene sequences, respectively (63). Resequencing a strain with a mutant phenotype for comparison with the wild-type genome enables location of causative genomic mutations more efficiently than with traditional mapping approaches (57). The ability to sequence numerous isolates of the same strain, as performed with 19 isolates of Salmonella enterica serovar Typhi using 454 and Solexa, enables researchers to understand how pathogens are evolving, which may lead to improvements in disease prevention and treatment (41). The large amount of data obtained from the next generation sequencing methods requires new strategies for data management and sequence assembly (63, 76). Several programs have been successfully used to assemble next-generation sequencing data. Newbler, a program utilized with 454 sequence output, has been successfully applied with microorganisms (72). Several programs for assembly of shorter reads have also been developed, such as SSAKE (93), VCAKE (44), and SHARCGS (27). Another program, SHRAP, has been developed to deal with the short reads that are repetitive by localizing the assembly to smaller clones and ordering the clones hierarchically (85).

SMP_Food Microbiology_CH39.indd

979

Single-Molecule Sequencing

Although the three methods described above have increased sequencing ability, the amplification step introduces variability. Additionally, isolates of a single strain may contain single nucleotide polymorphisms (SNPs). Recently, methods that do not require an amplification step have emerged, enabling sequencing of a single molecule and eliminating issues caused by amplification or variability within a strain. Helicos Biosciences developed Heliscope, which utilizes dNTPs attached to very bright fluorophores and is a sequencingby-synthesis method similar to the Solexa platform without amplification. Pacific Biosciences developed a method to sequence fragments in zeptoliter wells with fluorophore-associated dNTPs (57). Methods such as these will soon be common sequencing tools in genomic sciences.

BIOINFORMATICS Once a genome sequence is complete, the next task is to find genes in the sequence and annotate them. Elucidating the content of assembled DNA sequences belongs in part to the field of bioinformatics. While defined as a “scientific discipline that encompasses all aspects of biological information acquisition, processing, storage, distribution, analysis and interpretation” (77), the focus here will be on bioinformatics tools and the process of genome annotation. Genome annotation is the process of analyzing a raw DNA sequence and determining its potential biological significance. In particular, genome annotation can be divided into three stages: nucleotide level, protein level, and process level annotations (84). Nucleotide annotation is the first step in the complete annotation of a genome sequence. Genomic features are located and identified, including genes, untranslated RNAs, and any markers from preexisting physical or genetic maps. The most visible part of this phase is “gene calling.” In prokaryotic genomes, gene calling is primarily the identification of long open reading frames (ORFs), or nucleic acid sequences uninterrupted by stop codons. Simplified, “ORF calling” is the process of performing a six-frame translation of the DNA sequence and identifying all ORFs longer than a chosen threshold. Several software algorithms are available for both eukaryotic and prokaryotic gene calling including GLIMMER (26), Genezilla (58), NCBI ORF finder, GENEMARK.hmm (56), and Genie (52). In addition to gene calling, other genomic features are identified by software packages such as TransTerm (33), which searches for Rho-independent terminators, and RBSfinder, which searches for bacterial ribosomal binding sites upstream of the initiation codon of ORFs.

Manila Typesetting Company

11/08/2012 07:20AM

980

Advanced Techniques in Food Microbiology

After genes are found, protein level annotation attempts to define the protein complement of an organism, name the proteins, and assign them putative functions. Most approaches for computational protein annotation rely upon sequence similarity comparisons with sequences previously annotated in preexisting databases. A typical protein annotation pipeline would search for similarities using BLASTP (protein BLAST) (1) against several databases available through the National Center for Biotechnology Information (NCBI). The BLAST program also has more advanced options, PSI-BLAST and PHI-BLAST, to search for protein similarities, as well as programs that perform similarity searches at the nucleotide level (BLASTN) or that include translation of a sequence in multiple reading frames (BLASTX, TBLASTX, and TBLASTN) (http://www.ncbi.nlm.nih.gov). Another approach is to search against a database of functional domains. One commonly used database is PFAM (9), which contains alignments and Hidden Markov Model profiles for almost 8,000 protein families. Other databases include those available through the Expert Protein Analysis System (ExPASy) Proteomics Server such as Prosite (42) and SWISS-PROT (7). Finally, process level annotation concerns itself with relating the genome to biological processes. Several online databases are available to aid researchers in placing genes and gene products into a cellular context. However, wide variation in terminology can make effective searching difficult, whether by computers or by people. The Gene Ontology (GO) project is developing a standardized vocabulary for describing the functions of genes and their products (6). The GO project attempts to describe proteins on three levels: molecular function, biological process, and cellular component. Molecular function describes activities such as catalytic or binding activity. A biological process is a series of events carried out by one or more groups of molecular functions. Cellular components describe gene products in terms of the cellular structures that they comprise. The Kyoto Encyclopedia of Genes and Genomes (KEGG) (http:// www.genome.ad.jp/kegg/) is an online resource that provides knowledge to help researchers tie together genomic information with pathways and chemical information (45). For example, Fig. 39.3 illustrates the known metabolic pathways for folate metabolism and identifies which genes (denoted by green boxes) are predicted within the genome of Lactococcus lactis (13). Using this type of analysis, this genome revealed components for both folate (vitamin B11) and riboflavin (vitamin B2) biosynthetic pathways. By know-

ing the steps of these pathways, Sybesma et al. (86) were able to apply genetic approaches for metabolic engineering of these pathways and simultaneously increase the overproduction of both vitamins. Genomes can also be displayed graphically using tools that present either intrinsic DNA information such as GC-Skew or annotation data such as ORF location and deduced function. The genome map of Lactobacillus plantarum WCFS1 with the predicted origin of replication at the top is shown in Fig. 39.4 (49). The outer two circles show ORFs encoded on the positive strand in red and ORFs encoded on the negative strand in blue. Additionally, GC-Skew is shown in green and G+C content in black. Finally, the innermost rings show deduced gene function such as prophagerelated functions in green, IS-like elements in purple, rRNA gene operons in black, and tRNA-encoding genes in red. Annotation of the features of a genome can also reveal many interesting structural properties that may be unique to a genome. For example, many bacteria and archaea encode primitive immune systems in short palindromic repeats with nonrepetitive spacer regions in between, called clustered regularly interspaced short palindromic repeats (CRISPR) (82). The spacer regions are generally bacteriophage derived and are observed in strains resistant to phages encoding those sequences (8). CRISPR sequences provide a unique molecular signature that can be used in strain identification and tracking.

SMP_Food Microbiology_CH39.indd

980

COMPARATIVE GENOMICS With the number of completely sequenced bacterial genomes increasing rapidly, one powerful approach to defining unique or conserved gene content and understanding how these microorganisms evolved is comparative genomics, via an in silico analysis. This was beautifully illustrated after genome sequencing of Listeria monocytogenes and Listeria innocua when a comparative analysis of the two genomes revealed extensive colinearity (38). It was found that 270 genes (9.5%) were specific to the L. monocytogenes genome, whereas only 149 (5%) were specific to the L. innocua genome. Many of those genes specific to L. monocytogenes were key to survival in the mammalian gut, including those coding for bile salt hydrolases (BSHs) and acid tolerance via glutamic acid decarboxylases and a virulence locus responsible for infection and motility during pathogenesis (Fig. 39.5). In other foodborne pathogens, similar comparative genomic analyses have revealed a number of pathogenicity islands. The genomic analysis also revealed that patho-

Manila Typesetting Company

11/08/2012 07:20AM

981

39.  Genomics and Proteomics of Foodborne Microorganisms

Figure 39.3  KEGG pathway map and predicted enzymes for folate metabolism in Lactococcus lactis. Pathway intermediates and reaction products are shown with EC numbers for enzymes that catalyze these reactions. Those catalytic activities encoded in the L. lactis genome are highlighted in green. doi:10.1128/9781555818463.ch39f3

genic L. monocytogenes and nonpathogenic L. innocua were likely derived from a common ancestor, with the latter subsequently losing most of the genes directing pathogenesis. Therefore, in silico comparative analysis of complete genomes can reveal important clues

SMP_Food Microbiology_CH39.indd

981

regarding an organism’s genetic content, capabilities, and relationship to other microorganisms. There are a number of software tools available for these comparisons, including the Artemis Comparison Tool (ACT, Sanger Center) and MUMmer (53).

Manila Typesetting Company

11/08/2012 07:20AM

982

Advanced Techniques in Food Microbiology

Figure 39.4  Genome atlas of L. plantarum WCFS1. The predicted origin of replication is shown at the top. The outer to inner circles show (i) positive-strand ORFs (red); (ii) negative-strand ORFs (blue); (iii) GC-skew (green); (iv) G+C content (black); (v) prophage-­related functions (green) and IS-like elements (purple); and (vi) rRNA gene operons (black) and tRNA-encoding genes (red). Reprinted from Kleerebezem et al. (49) with permission. doi:10.1128/9781555818463.ch39f4

Genome sequence information has also provided key insights into the evolution and adaptation of microorganisms to specific foods (28). Streptococcus thermophilus is a lactic acid bacterium used worldwide for the manufacture of yogurt and some hard cheeses, whose estimated global market value is ~40 billion dollars. The natural habitat for this bacterium is raw milk and artisanal dairy starter cultures that have been propagated by cheese makers for centuries. Sequencing the genome of two yogurt strains of S. thermophilus and comparison to other streptococci revealed that the organism had undergone extensive rounds of genome decay, with over 10% of the genes being char-

SMP_Food Microbiology_CH39.indd

982

acterized as nonfunctional “pseudogenes.” Notably, with the exception of lactose metabolism, many of the other gene regions associated with carbohydrate transport and metabolism were also degraded. Genes commonly associated with pathogenic streptococci (e.g., Streptococcus agalactiae and Streptococcus pneumoniae) were either lost or inactivated due to mutations or transposon insertions. Bolotin et al. (12) concluded that S. thermophilus has evolved specifically to grow in milk and in the adaptation process it has eliminated or degraded much of its genomic complement that is not required for growth in this environment.

Manila Typesetting Company

11/08/2012 07:20AM

983

39.  Genomics and Proteomics of Foodborne Microorganisms

Figure 39.5  Comparison of the region containing the virulence gene cluster of L. monocytogenes and the homologous regions of the L. innocua and Bacillus subtilis genomes. Open blue boxes and arrows, orthologs among the three genomes; solid red boxes and arrows, virulence gene cluster; solid yellow boxes and arrows, genes absent from B. subtilis. (A) Scheme generated by GenomeScout (LION Bioscience). (B) Enlargement of the region containing the virulence gene cluster. Reprinted from Glaser et al. (38). doi:10.1128/9781555818463.ch39f5

FUNCTIONAL GENOMICS Once a genome sequence is generated and annotated, an important step is to identify or confirm the function of the genes found therein. While annotating genes on the basis of homology is a good way to begin this task, there are a number of genes in each bacterial genome that are unknown (30 to 40%) or have no homology to any genes found in the databases. Also, it is possible that a gene with a known phenotype in one organism may have a different role in another organism. The discipline of functional genomics deals with defining the roles of genes in their appropriate organisms. This task is usually carried out through two approaches: forward genetics and reverse genetics. Forward genetics deals with identifying the genotype of an organism with a known phenotype or feature. Routinely, attempts are made to create random mutations and screen for de-

SMP_Food Microbiology_CH39.indd

983

rivatives that exhibit a change in the phenotype under investigation. Mutant organisms can be created in different ways. Point mutations, or changes to a single nucleotide, can be created by using mutagens such as ethyl methanesulfonate. This approach is effective but often leads to secondary mutations that exert additional, unrelated changes in phenotype. Another method to create random mutations is via transposable elements, DNA sequences that have the ability to randomly insert into the genome. The advantage of using these elements for mutagenesis is that they can be subsequently located in the genome and the precise position of the insertion mutation can be mapped, facilitating the identification of the genetic loci encoding the phenotype. In contrast, reverse genetics utilizes the known genome sequence of the organism, or at the very least the sequence of the gene of interest. Reverse genetics begins with a directed change to the genotype of an organism and determining the resulting phenotype. Mutations are directed into the targeted DNA sequence using gene inactivation or replacement strategies. Most are based upon integration via homologous recombination of a cloning vector that replicates conditionally (for example, a plasmid that replicates only within a certain temperature range) in the bacterial cell. The vector encodes a selective marker such as an antibiotic resistance gene and a region of homology that is unique to the targeted gene (Fig. 39.6). Once transformed into the bacterial cell, homologous recombination events in the population are expected to occur at frequencies of 10–6 to 10–8. These events can be detected by selecting for antibiotic-resistant derivatives under conditions where plasmid replication cannot occur. Once a potential integrant is isolated, the position of the integration event is confirmed by detecting the junction fragments with PCR, DNADNA hybridization (Southern blotting), or sequencing. Once confirmed, integrants or deletion mutants can be characterized for any changes in phenotype that may elucidate the role of the gene or operon under study. Care must be taken to minimize the downstream or polar effects of integrants; therefore, gene deletions or replacements are the preferred approach, and these require double-crossover events (Fig 39.7). The advantage of this method is that genes are deleted or inactivated and the resulting derivative no longer contains any vector DNA or selection markers (e.g., antibiotic resistance genes). Using targeted knockout or deletion strategies to correlate phenotypes also demands one more step. Using genetic complementation, the original gene is reintroduced into the

Manila Typesetting Company

11/08/2012 07:20AM

984

Advanced Techniques in Food Microbiology understanding microbial metabolism, physiology, and interactions. Understanding the genomic basis for microbial traits can be used to promote positive traits, such as production of flavor-producing compounds in beneficial food microbes (19), or to potentially limit pathogenic traits. Additionally, recombinant techniques have enabled nonnative genes that encode foodprocessing enzymes, such as chymosin and a-amylase, to be cloned and expressed in microorganisms for use in the food industry (70). These techniques are now being expanded to include cloning and expression of antigens in probiotic bacterial strains, taking advantage of their natural adjuvant properties for oral vaccine delivery (65). Foreign genes may even be optimized to the codon usage of the host strain using programs such as JCat (40) and then commercially synthesized to increase compatibility for expression. Genomics provides the information and tools to understand, control, and where appropriate, exploit the benefits of microorganisms in foods. A recent example utilizing both comparative and functional genomics was the discovery of a human mucus-binding pilus protein of Lactobacillus rhamnosus (46). Pili are found in many gram-positive and gram-negative pathogens (50), and their description in L. rhamnosus strain GG (LGG) indicated that probiotic and commensal microbes can employ shared

Targeted Gene PCR Amplified Gene Fragment

Integration Vector

Homologous Recombination Integrant Antibiotic Resistance Gene

Figure 39.6  Plasmid insertion into a gene through homologous recombination for inactivation of gene function. The phenotype of the mutant can then be analyzed to investigate the function of the gene. doi:10.1128/9781555818463.ch39f6

mutant, in trans, and recovery of the original phenotype is expected (37, 67, 81). With knowledge of the sequence of a gene of interest, a gene or operon of interest can also be cloned and overexpressed on high-copy-number plasmids with or without strong promoters. Using gene expression or inactivation strategies, efforts to elucidate or confirm gene function is a very powerful approach to

A

B B

Deletion

C

D A

D

Wild Type

C

Deletion

Figure 39.7  A replacement or deletion mutant can be created by first cloning two noncontiguous portions of a gene into an integration vector. The vector integrates into a targeted gene within one region of homology (black or light gray regions). Excision of the plasmid from the integrant structure can occur in a manner that either resolves the original gene (wild type) or leaves the deleted version. Points of resolution at steps A, B, C, or D will result in various combinations, as illustrated. doi:10.1128/9781555818463.ch39f7

SMP_Food Microbiology_CH39.indd

984

Manila Typesetting Company

11/08/2012 07:20AM

985

39.  Genomics and Proteomics of Foodborne Microorganisms strategies for survival in the gastrointestinal tract. A cluster of pilus-encoding genes (spaCBA) was discovered in LGG after comparative genomic analysis of the probiotic LGG genome and the closely related starter culture L. rhamnosus LC705 genome (46). Pilin proteins in gram-positive bacteria contribute to adhesion to other bacteria and host cells (50). The presence of pili on the cell surface of LGG, with the greatest number clustered at the cell poles, was confirmed by immunogold electron microscopy (Fig. 39.8). Subsequently, the spaC gene, which was predicted to encode the large-sized minor pilin subunit, was inactivated using functional genomic methods. The spaC mutant exhibited a significant reduction in binding to intestinal mucus (46). This study utilized comparative and functional genomics and proteomics to demonstrate the role of SpaC in mucin binding and potentially its importance to the retention of some lactobacilli in the gastrointestinal tract.

DNA MICROARRAYS DNA microarrays allow investigation of biological systems with a genome-wide approach. Their power comes from the exploitation of sequence complementarity between two strands of duplex DNA, providing both sensitivity and specificity (31, 83). While many different microarray platforms are available, all operate under the same principle: sequence-specific nucleic acid probes are immobilized onto a solid substrate and are then used, via hybridization, to probe a pool of labeled target nucleic acid for the presence of these specific sequences. The abundance of the target sequence is then determined from the amount of label detected at the location of a specific probe. The principles behind DNA microarray technology make it very applicable to many different uses that include comparative genomics and global gene expression analysis. Microarrays can be used to determine the presence or absence of similar sequences in two different genomes. In this case, one genome serves as a reference and is spotted onto the array, while a second, test genome is then hybridized against the array. Currently, the main large-scale application for microarrays is gene expression analysis (83), and different microarray solutions exist. In this overview, the “home-baked” cDNA microarray, Affymetrix GeneChip, and Roche NimbleGen platforms will be discussed. The most flexible and accessible platform for academic users seems to be the home-baked cDNA microarray (Fig. 39.9). The fabrication of these arrays begins

SMP_Food Microbiology_CH39.indd

985

Figure 39.8  Identification of pili in L. rhamnosus GG by immunogold electron microscopy. L. rhamnosus GG was grown to stationary phase, treated with anti-SpaC serum, labeled with protein A-conjugated gold particles (10 nm), negatively stained, and examined by transmission electron microscopy. (A) High-resolution electron micrograph showing multiple pili and an isometric bacteriophage (black arrow). Also included is a panel inset adjusted for heightened contrast and darkness to highlight the pilus ultrastructure (white arrow). (B) Electron micrograph showing pili clustered at the cell poles. (Bars: A, 200 nm; B, 500 nm.) Reprinted from Kankainen et al. (46) with permission. doi:10.1128/9781555818463.ch39f8

with probe selection. General probes are PCR amplified from clones from an expressed sequence tag (EST) library or, in the case of microbes, with complete DNA sequence information using specific primers to create gene-specific amplicons. These probes are then purified and spotted individually onto glass slides. In the simplest form of a microarray experiment, the array is used to probe pools of fluorescently labeled cDNA reverse transcribed from total RNAs isolated from reference cells and test cells under some defined condition; for example, cells grown in glucose (reference) versus cells grown in lactose (test). Once the targets have

Manila Typesetting Company

11/08/2012 07:20AM

986

Advanced Techniques in Food Microbiology

Figure 39.9  Microarrays. Probes are PCR amplified from clones in a library or from genomic DNA using gene-specific primers. Individual amplicons are purified and spotted onto glass slides. Total RNA is labeled from both a test sample and a reference sample using fluorescent dyes and allowed to hybridize to the probes on the array. The array is then visualized using a laser scanner that generates color images that are overlaid and compared for intensity and source. Reprinted with permission from Duggan et al. (31). doi:10.1128/9781555818463.ch39f9

been hybridized to the array, it is scanned using a laser scanner, generating an image for each dye used. For example, a common example is to label the reference sample with Cy3, which is represented in a color image as red spots, and label the test sample with Cy5, which is represented in a color image in green. When the two images are overlaid, genes that are expressed more in the reference sample appear as red spots, whereas those that are expressed more in the test sample appear as green spots. In the case of a gene that is equally expressed in the two samples, the gene will appear as a yellow spot. One can then determine the relative message abundance between the two samples based on the signal generated (20, 31). The Affymetrix GeneChip platform is a widely used commercial microarray solution. The Affymetrix platform is based on 20-base oligonucleotides synthesized in situ on a glass surface using photolithography and solid-phase DNA synthesis. An important feature of the GeneChip is the amount of redundancy on the array: for each gene, there are 20 probes representing different portions of the gene. Additionally, for each probe there

SMP_Food Microbiology_CH39.indd

986

is a single-base mismatch probe that is used to subtract any cross-hybridization effects (54). Three major limitations exist when this system is compared to home-baked microarrays: 1. The Affymetrix system is a single-dye system, which excludes the ability to perform competitive hybridization experiments using target cDNA from two conditions. 2. The system is proprietary. Every component required for use of the system, including the reagents, the fluidics station, and scanner, must be purchased. 3. The system is limited to research involving the major model systems. Custom probe array production is available but prohibitively expensive. Roche NimbleGen arrays are made by synthesizing probes on glass slides with a maskless array synthesizer utilizing aluminum mirrors controlled by a computer. The use of the mirrors decreases the cost of producing different microarray designs compared to Affymetrix’s use of chromium masks, which must be fabricated for each design (89). Also, in contrast to the Affymetrix plat-

Manila Typesetting Company

11/08/2012 07:20AM

987

39.  Genomics and Proteomics of Foodborne Microorganisms form, NimbleGen offers high-density gene expression array designs for a wide range of eukaryotes and prokaryotes. The arrays contain 60mer probes, with multiple probes for each transcript, increasing the accuracy of results. The major advantage of microarrays is the amount of throughput. Microarrays have given researchers the ability to assay for the expression of all the genes in a particular genome rather than using a gene-by-gene approach. Using traditional hybridization-based methods such as Northern blotting, researchers were able to assay for expression of only one gene during a particular hybridization experiment. Slot/dot blots were able to increase the number of genes that could be assayed, but researchers were still severely limited in the number of probes that could be blotted on a membrane. With the advent of computer-controlled precision robotics, researchers were able to miniaturize the slot/dot blots while increasing the number of probes arrayed up to 1,000-fold (15). Comparisons of genome content between closely related strains or species can also be accomplished by hybridizations to whole-genome arrays representing the organism when the complete genome sequence is available. Doumith et al. (29) used this approach to create a DNA array of the “flexible” part of Listeria genomes that had been sequenced. Probing that array with the DNA from 113 L. monocytogenes strains, it was determined that 93 of the strains harbored all the previously identified virulence factors. In addition, the method identified 30 genes that were L. monocytogenes specific, which may prove useful for tracking strains in listeriosis outbreaks.

RNA SEQUENCING While microarrays have greatly increased the amount of information known about bacterial transcriptomes, they are limited to use with known sequences and they present technical difficulties arising from cross-hybridization and background levels. Additionally, making comparisons between experiments is problematic, as complex normalization procedures are required. The next-generation sequencing technologies, described previously, have recently been used for RNA sequencing (RNA-Seq). Using next-generation sequencing, RNA may be directly fragmented and sequenced, or it may be reverse transcribed into cDNA and sequenced. The number of reads obtained for each sequence directly correlates with transcript expression (91). RNA-Seq provides a number of advantages over current microarray technology. First, RNA-Seq’s direct sequenc-

SMP_Food Microbiology_CH39.indd

987

ing capabilities eliminate the requirement for a known genome sequence prior to the experiment, whereas prior knowledge of sequence information is essential for microarray design. Additionally, novel transcripts or small regulatory RNA molecules may be identified (22). Each sequence obtained with RNA-Seq is counted, providing sensitive expression level detection over a broad range with minimal background interference, which is not the case with microarrays. The sequences provided by RNA-Seq enable additional information to be obtained in experiments directed at gene expression profiling. For example, while microarrays may be used to discover SNPs through specialized arrays in experiments directed at discovering SNPs, RNA-Seq provides SNP information in all experiments by providing the sequence of each transcript (91). Knowledge of SNPs has been useful for molecular subtyping of L. monocytogenes, which may be applied in foodborne disease investigations (92). Therefore, use of RNA-Seq is increasing and is predicted to eventually replace microarrays (22). However, there are still issues with RNA-Seq. Paramount among these is managing the abundant quantity of data obtained from next-generation sequencing. Programs for mapping short reads generated from RNA-Seq are available but so far have been directed mostly towards eukaryotic genomes (74). Another problem is generated when the short sequence reads obtained match multiple genomic locations. Improvements in next-generation technology should increase read lengths and eventually eliminate this type of problem (91). The mRNA transcripts that are most often of interest during transcriptional profiling account for only about 5% of the total RNA. Most of the RNA pool consists of rRNAs and tRNAs, which need to be depleted before RNA-Seq, otherwise most sequences obtained will be housekeeping RNAs rather than the mRNAs or small noncoding RNAs (sRNA) that are of interest. In eukaryotes, mRNA enrichment is easily accomplished by selection for the 3¢ polyadenylated [Poly(A)] tail that is present only on mRNAs. The lack of this tail on prokaryotic mRNA complicates the process required to obtain useful information from RNA-Seq experiments, but several rRNA and tRNA depletion strategies have emerged for prokaryotic RNA-Seq (55). One group that was interested in sRNA in Vibrio cholerae depleted the rRNA and tRNA by size, first selecting for sRNAs on polyacrylamide gels. The smaller RNA pool was then mixed with oligonucleotides complementary to tRNA and rRNA fractions, and the DNA/RNA hybrids were removed with RNaseH. The subsequent

Manila Typesetting Company

11/08/2012 07:20AM

988

Advanced Techniques in Food Microbiology

RNA-Seq experiment, with the 454 Genome Sequencer 20, enabled the discovery of 627 novel sRNAs in V. cholerae (55). Commercial kits are also emerging to enable enrichment of mRNA. A group interested in the genes in L. monocytogenes that are regulated by the alternative general stress response transcription factor, sigma B, used the Ambion MICROBExpress bacterial mRNA enrichment kit to deplete the rRNA pool prior to RNA sequencing with the Illumina Genome Analyzer. This approach, along with a Hidden Markov Model, enabled identification of several unknown noncoding RNAs as well as promoter sites for the majority of genes dependent on sigma B (71).

ing against protein databases (14). This approach can be readily scaled up and automated so that the mass spectra of hundreds of proteins can be acquired and searched against databases. The tandem mass spectrometric (MS/MS) method is based upon the fragmentation and amino acid sequencing of individual peptides in the protein digest. In this approach, the peptide mixture is ionized directly from the liquid phase using electrospray ionization. The ionized peptides are sprayed into an MS/MS spectrometer, which can then resolve individualized peptides in the mixture. Single peptides are isolated and further resolved into amino- or carboxyterminal-containing fragments. While the MS/MS method is more complex and less scalable than mass peptide fingerprinting, the sequence information acquired using this technique is more specific for the identification of an individual protein rather than a listing of peptide masses. MS/MS data can also be searched against both protein and nucleic acid databases. The primary use of these 2D-PAGE/MS methods is protein expression profiling. Analogous to gene expression profiling using microarrays, proteins from an organism exposed to different conditions are resolved on separate 2-D gels. The gels are subsequently analyzed by computer software to identify changes in protein expression by bacterial cells exposed to different conditions (23, 73). Conventional 2D-PAGE technology relies on comparisons of protein spots from at least two different gels. Due to gel-to-gel variation, it is not possible to directly superimpose the images of two separate gels, adding further complexity to the analysis of these comparisons. The introduction of differential gel electrophoresis has addressed these pitfalls of 2D-PAGE. Analogous to microarrays, 2D differential gel electrophoresis makes use of fluorescent dyes that covalently label all proteins in a protein sample. By labeling protein samples with different fluorescent dyes of nonoverlapping excitation and emission profiles, it is possible to compare the protein expressions from several conditions on the same gel. This method solves many of the issues associated with run-to-run variation of conventional 2D-PAGE (87). Alternative gel-free methods based upon liquid chromatography (LC) have been developed, including direct LC separation of digested peptides coupled to MS/MS (LC/MS/MS), and multidimensional LC coupled to mass spectrometry, called multidimensional protein identification technology (MudPIT) (25, 62, 97). These methods perform the protein separations using LC-based methods, dispensing with issues such as resolvability of proteins and sensitivity of detection in gel-based methods. Additionally, with the availability of robotics for LC methods, automation and increased throughput can be introduced.

PROTEOMICS Microarray technologies provide vast quantities of global gene expression data on the level of mRNA expression. However, it has been determined that in many cases, the abundance of mRNA transcripts does not directly correlate to the final amount of protein expressed (2). Proteomic techniques are required to provide function and quantitative data to complement genomic data. Proteomics can be defined as “the use of quantitative protein level measurements of gene expression to characterize biological processes and decipher the mechanisms of gene expression control” (3). Proteomics has been traditionally associated with the display of a large number of proteins from a cell line or organism using two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). 2D-PAGE is a technique whereby proteins are separated according to isoelectric point by isoelectric focusing (IEF) in the first dimension and according to molecular weight in the second dimension (Fig. 39.10) by sodium dodecyl sulfate-PAGE (68). 2D-PAGE has the ability to resolve a large number of proteins and discern changes in charge and mass due to posttranslational modification. However, it was not until the advent of biological mass spectrometry (MS) that the identity of the resolved proteins could be easily elucidated. Following separation on 2D gels, proteins are stained to allow visualization, excised from the gel as gel fragments, and then digested into peptides with sequence-specific proteases such as trypsin. The resulting peptides can be extracted and sequenced using MSbased methods. There are two main approaches to MS-based protein identification. In peptide mass mapping, the mass spectrum of the digested peptide mixture is acquired using matrix-assisted laser desorption ionization. The resulting mass spectrum serves as a peptide mass fingerprint for the protein, which can then be identified by search-

SMP_Food Microbiology_CH39.indd

988

Manila Typesetting Company

11/08/2012 07:20AM

989

39.  Genomics and Proteomics of Foodborne Microorganisms

Figure 39.10  Proteomic methods. (A) 2D-PAGE-MS. First dimension: proteins are separated based on isoelectric point (pI) using IEF. Proteins migrate along a pH gradient until they reach their pI, at which point they carry no net charge and stop migrating. Second dimension: proteins are further separated according to molecular weight using SDS-PAGE. Gels are stained to identify protein bands. Individual spots (red circles) are excised from the gel, trypsin digested, and sequenced using MS/MS. (B) Protein expression profiling. By overlaying images of 2DPAGE gels, comparisons can be made between the proteomes of different organisms or differences in protein expression of a single organism under different conditions. Downregulated (dotted red circles) and upregulated (blue circles) proteins can be visualized. Printed with permission from Phillips and Bogyo (75). doi:10.1128/9781555818463.ch39f10

PROTEIN-DNA INTERACTIONS In recent years, a technique commonly used in eukaryotes, called chromatin immunoprecipitation (ChIP), has begun to be applied more frequently to studying interactions between DNA and protein in prokaryotes. This technique utilizes formaldehyde to cross-link protein and

SMP_Food Microbiology_CH39.indd

989

DNA under conditions of interest, followed by immunoprecipitation of a chosen protein. Heat is used to reverse the cross-linking and obtain the DNA for either hybridization to a microarray (ChIP-chip) or more recently for DNA sequencing using next-generation technologies (ChIP-seq). This technique enables protein-binding sites

Manila Typesetting Company

11/08/2012 07:20AM

990

Advanced Techniques in Food Microbiology

on DNA to be identified, providing information about DNA-protein interactions, including transcription factor-binding sites and promoter sites under different conditions of interest (90).

GENOMIC AND PROTEOMIC ADVANCES IN FOOD MICROBIOLOGY One use of genomic approaches to food microbiology has been the identification and characterization of genes that confer resistance to stressors such as acid or bile. These genes can be important to understanding how both foodborne pathogens such as Escherichia coli O157:H7 and commensals such as some Lactobacillus spp. survive the rigors of the intestinal tract or how they survive in acidified foods. Bacteria are able to survive in these environments because they induce an acid tolerance response which allows them to adapt. While this physiological phenomenon was identified before the advent of bacterial genomics (36), the availability of sequenced bacterial genomes allows the use of transcriptional (microarrays) and translational (proteomic techniques) analyses to characterize microbial responses to acid and pH and elucidate which genes are expressed and how they are regulated. Probably one of the best-characterized acid resistance mechanisms is the Gad system of E. coli. This system is present in all strains of E. coli that are of public health significance, including O157:H7 (39). The acid resistance of this system is contributed by three proteins encoded by genes gadA, gadB, and gadC. The gadA and gadBC genes exist in different parts of the genome (Fig. 39.11). The GadA and GadB proteins catalyze the exchange of the a-carboxyl of the amino acid glutamate for a proton in the environment. This reaction leads to the creation of a molecule of carbon dioxide and a molecule of g-amino butyric acid (GABA). Consumption of an H+ from the cytoplasm contributes to an increase in the intracellular pH. GadC is a membrane transporter, known as an antiporter, that expels GABA from the cell and imports fresh glutamate (16, 80). These genes were first discovered before the sequence of E. coli was published in 1997, but an understanding of how these genes are regulated and which other organisms use them has been elucidated using the tools of genomics. While the molecular mechanisms of this defense are simple, the regulatory system that controls them is quite complex and not entirely understood. The main player in the activation of the acid resistance system is the GadE protein, a transcriptional activator that binds upstream of gadA and gadBC and directs the binding of RNA polymerase to transcribe these genes (94). The

SMP_Food Microbiology_CH39.indd

990

GadC Antiporter OH+

O +H

3N

H+

O

CO2

+H

3N

GadA and GadB

OH

O OH

glutamate

γ-aminobutyric acid (GABA)

Figure 39.11  GadA and GadB catalyze the exchange of the a-carboxyl of glutamate for a proton in the environment, leading to the creation of a molecule of carbon dioxide and one molecule of GABA. GadC is an antiporter that expels GABA from the cell and imports fresh glutamate. doi:10.1128/9781555818463.ch39f11

first regulatory circuit discovered, the EvgAS circuit, was initially characterized using microarrays. Many known acid resistance genes, including the Gad genes, were induced when the regulator EvgA was overexpressed in E. coli cells. Also, strains overexpressing the protein EvgA had greater acid resistance than wild-type cells during the exponential phase of growth. EvgA is part of a two-component regulatory system in which the bacterial cell senses specific environmental changes (such as a decrease in pH) and responds to them by upregulating genes designed to deal with that particular type of change. A histidine protein kinase gene (EvgS) is the sensing part of this system, and once it receives a signal, it turns on the response regulator (EvgA) that transcriptionally activates specific genes (Fig. 39.12). Using a microarray to compare EvgA overexpression strains to an EvgA deletion strain, several other genes upregulated by EvgA were identified. These genes were ydeO, gadA, gadB, gadC, and gadE. Based on a highly conserved DNA sequence found upstream of the genes upregulated by EvgA, the DNA-binding site for this protein was determined, and the rest of the E. coli genome was then examined for potential binding sites for EvgA. This binding site was present upstream of ydeO,

Manila Typesetting Company

11/08/2012 07:20AM

991

39.  Genomics and Proteomics of Foodborne Microorganisms H+

Cytoplasm

His P ATP Histidine Protein Kinase EvgS

Asp P Response Regulator EvgA

ADP

EvgA

ydeO

YdeO gadE

GadE

GadE gadA

gadB gadC

Figure 39.12  The EvgA/S circuit of acid resistance regulation is dependent on a histidine protein kinase (EvgS), which senses an environmental change causing it to activate its corresponding response regulator (EvgA), which is then able to act as a transcriptional regulator. The regulation follows a pathway to produce GadE, which ultimately induces the transcription of GadA/BC (Fig. 39.11). doi:10.1128/9781555818463.ch39f12

indicating that EvgA acts as a transcriptional activator for this gene. Hence, the first regulatory circuit for the GAD system was discovered using genomic data, microarrays, and bioinformatics to identify key genes and their regulatory sequences (60, 61). While the details of the Gad system were elucidated primarily by experiments in E. coli, genomics and bioinformatics have enabled researchers to identify and study the effects of these genes in other organisms. In L. monocytogenes, for example, the gadA and gadBC genes were identified based on similarity to these genes in E. coli, L. lactis, and Mycobacterium tuberculosis. Upon creating deletion mutants of these genes and testing their survival in synthetic and porcine gastric fluids, it was determined that L. monocytogenes absolutely depends on the GAD system to maintain pH homeostasis (24). In addition to the data generated by using HCl to lower pH, the gadA and gadBC genes have also been shown to be upregulated in E. coli in the presence of acetic acid, an organic acid typical of those encountered in acidified and fermented foods (5). E. coli O157:H7 can survive in low-acid food products, including unpasteurized apple cider, in which the pH can be below 4.0 (43). Glutamate, the proton-accepting amino acid that is

SMP_Food Microbiology_CH39.indd

991

central to this system, is used extensively in food products as a flavor enhancer or as a way to adjust acidity. It is also present in protein-rich foods such as meat and milk. Many vegetables contain large amounts of bound glutamate, and their fermented counterparts have increased levels of free glutamate (24). In this regard, it is likely that foods and food components can precondition some foodborne pathogens for improved survival and pathogenicity, once ingested. While the ability to survive the acidic conditions in the stomach or food products is important to foodassociated microbes, infection in the intestinal tract requires that bacteria possess the ability to withstand additional types of stresses, such as those imposed by bile. Bile acts as a detergent that can dissolve the phospholipid cytoplasmic membranes of bacteria, leading to their death (21). L. monocytogenes is a species that is able to withstand the presence of high concentrations of bile. In fact, it can withstand levels so high that it has the ability to colonize the gall bladder, which is the human body’s primary repository for bile (10). While the mechanisms through which bacteria tolerate bile are not entirely understood, some genes that play important roles have been identified. One of the most studied groups

Manila Typesetting Company

11/08/2012 07:20AM

992

Advanced Techniques in Food Microbiology

comprises­ those encoding BSHs. Bile salts exist primarily as steroid rings attached through an amide bond to an amino acid group, either glycine or taurine (Fig. 39.13). BSHs catalyze the hydrolysis of the conjugated salt to release the amino acid. Disruption of the bsh gene in L. monocytogenes leads to greater sensitivity to bile (10). Three mechanisms through which BSHs contribute to bile tolerance have been proposed: (i) conferring a nutritional advantage on the cell by liberating amino acids; (ii) facilitating incorporation of the cholesterol moiety into the cell membrane, thereby strengthening it against the detergent action of bile; and (iii) acting as a detoxification mechanism against bile salts (10). While these hypotheses are still under investigation, a comparison between the genomes of L. monocytogenes and the nonpathogenic L. innocua revealed that only the pathogen carried the gene for BSH. Bioinformatic analysis of the bsh promoter region revealed that bsh is under the control of the transcriptional regulator PrfA, which controls expression of other known virulence genes in L. monocytogenes (32). A study in mice, comparing wild-type L. monocytogenes to a derivative strain with deletion of the bsh gene (Dbsh), revealed that the wild-type strain could cause significantly more deaths than the Dbsh strain. These studies indicate that BSH contributes to the virulence of L. monocytogenes, but the mechanism remains undefined. The contribution of genome sequencing and

functional genomics has greatly facilitated our understanding of the pathogenicity of L. monocytogenes. The transcriptional regulator PrfA is also regulated by another factor, sigma B, which is the analogue to RpoS found in E. coli and other gram-negative bacteria. Sigma B is the general stress response sigma factor in gram-positive bacteria controlling stress responses in L. monocytogenes, including salt tolerance and growth at low temperatures (34, 47, 48, 66). Tolerance to these two stressors is mediated in much the same way, via the uptake of small organic compounds known as compatible solutes that act to increase osmolarity within the cell without disrupting cellular functions. The osmolytes utilized by Listeria are glycine-betaine, carnitine, and proline, with proline being the only osmolyte that Listeria can synthesize on its own. The cell relies upon osmolyte transporters to import these molecules: Gbu, a transporter for glycine-betaine; OpuC, the carnitine transporter; and BetL, another glycine-betaine uptake system. Mutants for each of these genes have a decreased ability to take up osmolytes and to grow at low temperatures or to tolerate salt (51, 64, 79, 95, 96). Of particular interest is OpuC, which, when mutated, leads to a decrease in the virulence of Listeria. Microarray studies have revealed that OpuC is also the only osmolyte transporter regulated by sigma B, again illustrating that sigma B-controlled genes, like opuC and bsh, are

O CH3 HO CH3

Taurocholic Acid

SO3-

NH

CH3

HO

OH

H2O Bile Salt Hydrolase

Cholic Acid CH3 HO CH3

O H2NCH2CH2

S

O OH

CH3

O-

O HO OH Taurine Figure 39.13  BSHs catalyze the hydrolysis of the peptide bond between the amino acid and the cholesterol-derived backbone of the bile salt. doi:10.1128/9781555818463.ch39f13

SMP_Food Microbiology_CH39.indd

992

Manila Typesetting Company

11/08/2012 07:20AM

993

39.  Genomics and Proteomics of Foodborne Microorganisms involved with both stress tolerance and virulence (4, 18, 35, 79). Trost et al. (88) were also able to characterize the secretory proteome of L. monocytogenes using the complementary techniques of 2D-PAGE/MS and LC/ MS/MS (88). The researchers identified 105 proteins in the culture supernatant, including the eight proteins known to be involved in virulence. Comparison of the secretory proteomes of L. monocytogenes and L. innocua revealed a number of additional proteins unique to L. monocytogenes that may be involved in pathogenicity and/or adaptation to its lifestyle. Additionally, by combining in silico comparative genomic analysis with these proteomic techniques, the authors determined that the differences between the pathogenic and nonpathogenic Listeria species appear most strongly in the secretory proteome.

CONCLUSIONS The rise of “omic” technologies has led to an explosion of knowledge of DNA, protein, and metabolic information about microorganisms associated with our foods. With this information, it is now possible to understand the complete genetic complement of pathogenic, spoilage, and bioprocessing organisms and determine how environmental conditions affect the expression of key genetic traits. Pathogens can be better tracked in foods and outbreaks because of signature sequences revealed in genomes. Pathogenicity islands have exposed the key gene combinations underlying the survival, infection, and disease properties of foodborne pathogens. Different bioprocessing and probiotic strains can be compared for their genetic content and their beneficial capabilities reliably predicted. Foods and their microenvironments could potentially be better designed and formulated in order to minimize expression of undesirable pathogenic traits (e.g., acid tolerance, virulence, and toxin formation) or to optimize expression of beneficial properties in desirable organisms (e.g., cryoprotection, acidification rates, and adherence to intestinal tissues). The nature of food microbiology has changed dramatically from its historical emphasis on microbial phenotypic properties and behavior to a new perspective dominated by genomic and comparative genomic information. The food microbiologists of the future will become increasingly reliant on genomics and the other omics technologies in their efforts to understand and control microorganisms associated with our foods. G.L.D., E.P., and T.D. gratefully acknowledge support by the Functional Genomic Sciences Program and the NIH— Molecular Biotechnology Training Fellowships.

SMP_Food Microbiology_CH39.indd

993

References   1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410.   2. Anderson, L., and J. Seilhamer. 1997. A comparison of selected mRNA and protein abundances in human liver. Electrophoresis 18:533–537.   3. Anderson, N. L., and N. G. Anderson. 1998. Proteome and proteomics: new technologies, new concepts, and new words. Electrophoresis 19:1853–1861.   4. Angelidis, A. S., L. T. Smith, L. M. Hoffman, and G. M. Smith. 2002. Identification of OpuC as a chillactivated and osmotically activated carnitine transporter in Listeria monocytogenes. Appl. Environ. Microbiol. 68:2644–2650.   5. Arnold, C. N., J. McElhanon, A. Lee, R. Leonhart, and D. A. Siegele. 2001. Global analysis of Escherichia coli gene expression during the acetate-induced acid tolerance response. J. Bacteriol. 183:2178–2186.   6. Ashburner, M., C. A. Ball, J. A. Blake, D. Botstein, H. Butler, J. M. Cherry, A. P. Davis, K. Dolinski, S. S. Dwight, J. T. Eppig, M. A. Harris, D. P. Hill, L. Issel-Tarver, A. Kasarskis, S. Lewis, J. C. Matese, J. E. Richardson, M. Ringwald, G. M. Rubin, G. Sherlock, et al. 2000. Gene ontology: tool for the unification of biology. Nat. Genet. 25:25–29.   7. Bairoch, A., B. Boeckmann, S. Ferro, and E. Gasteiger. 2004. Swiss-Prot: juggling between evolution and stability. Brief. Bioinform. 5:39–55.   8. Barrangou, R., C. Fremaux, H. Deveau, M. Richards, P. Boyaval, S. Moineau, D. A. Romero, and P. Horvath. 2007. CRISPR provides acquired resistance against viruses in prokaryotes. Science 315:1709–1712.   9. Bateman, A., L. Coin, R. Durbin, R. D. Finn, V. Hollich, S. Griffiths-Jones, A. Khanna, M. Marshall, S. Moxon, E. L. Sonnhammer, D. J. Studholme, C. Yeats, and S. R. Eddy. 2004. The Pfam protein families database. Nucleic Acids Res. 32:D138–D141. 10. Begley, M., C. G. M. Gahan, and C. Hill. 2005. The interaction between bacteria and bile. FEMS Microbiol. Rev. 29:625–651. 11. Bennett, S. 2004. Solexa Ltd. Pharmacogenomics 5:433– 438. 12. Bolotin, A., B. Quinquis, P. Renault, A. Sorokin, S. D. Erlich, S. Kulakauskas, B. Purnelle, D. Prozzi, K. Ngui, D. Masuy, F. Hancy, S. Burteau, M. Boutry, J. Delcour, A. Goffeau, and P. Hols. 2004. Complete sequence and comparative analysis of the dairy bacterium Streptococcus thermophilus. Nat. Biotechnol. 22:1554–1558. 13. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach, S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res. 11:731–753. 14. Bonk, T., and A. Humeny. 2001. MALDI-TOF-MS analysis of protein and DNA. Neuroscientist 7:6–12. 15. Brown, P. O., and D. Botstein. 1999. Exploring the new world of the genome with DNA microarrays. Nat. Genet. 21:33–37.

Manila Typesetting Company

11/08/2012 07:20AM

994

Advanced Techniques in Food Microbiology

16. Castanie-Cornet, M.-P., T. A. Penfound, D. Smith, J. F. Elliott, and J. W. Foster. 1999. Control of acid resistance in Escherichia coli. J. Bacteriol. 181:3525–3535. 17. Cebula, T. A., S. A. Jackson, E. W. Brown, B. Goswami, and J. E. LeClerc. 2005. Chips and SNPs, bugs and thugs: a molecular sleuthing perspective. J. Food Prot. 68:1271–1284. 18. Cetin, M. S., C. Zhang, R. W. Hutkins, and A. K. Benson. 2004. Regulation of transcription of compatible solute transporters by the general stress sigma factor, sigma B, in Listeria monocytogenes. J. Bacteriol. 186:794–802. 19. Chaves, A. C., M. Fernandez, A. L. Lerayer, I. Mierau, M. Kleerebezem, and J. H.ugenholtz. 2002. Metabolic engineering of acetaldehyde production by Streptococcus thermophilus. Appl. Environ. Microbiol. 68:5656–5662. 20. Cheung, V. G., M. Morley, F. Aguilar, A. Massimi, R. Kucherlapati, and G. Childs. 1999. Making and reading microarrays. Nat. Genet. 21:15–19. 21. Coleman, R., P. J. Lowe, and D. Billington. 1980. Membrane lipid composition and susceptibility to bile salt damage. Biochim. Biophys. Acta 559:294–300. 22. Coppee, J. Y. 2008. Do DNA microarrays have their future behind them? Microbes Infect. 10:1067–1071. 23. Cordwell, S. J., A. S. Nouwens, and B. J. Walsh. 2001. Comparative proteomics of bacterial pathogens. Proteomics 1:461–472. 24. Cotter, P. D., C. G. M. Gahan, and C. Hill. 2001. A glutamate decarboxylase system protects Listeria monocytogenes in gastric fluid. Mol. Microbiol. 40:465–475. 25. Delahunty, C., and J. R. Yates III. 2005. Protein identification using 2D-LC-MS/MS. Methods 35:248–255. 26. Delcher, A. L., D. Harmon, S. Kasif, O. White, and S. L. Salzberg. 1999. Improved microbial gene identification with GLIMMER. Nucleic Acids Res. 27:4636–4641. 27. Dohm, J. C., C. Lottaz, T. Borodina, and H. Himmelbauer. 2007. SHARCGS, a fast and highly accurate short-read assembly algorithm for de novo genomic sequencing. Genome Res. 17:1697–1706. 28. Douglas, G. L., and T. R. Klaenhammer. 2010. Genomic evolution of domesticated microorganisms. Annu. Rev. Food Sci. Technol. 1:397–414. 29. Doumith, M., C. Cazalet, N. Simoes, L. Frangeul, C. Jacquet, F. Kunst, P. Martin, P. Cossart, P. Glaser, and C. Buchrieser. 2004. New aspects regarding evolution and virulence of Listeria monocytogenes revealed by comparative genomics and DNA arrays. Infect. Immun. 72:1072–1083. 30. Doyle, M., L. Jaykus, and M. Metz. 2005. Research Opportunities in Food and Agriculture Microbiology. American Academy of Microbiology, Washington, DC. http://academy.asm.org/index.php/colloquium-program/ browse-all-reports/240-research-opportunities-in-foodand-agriculture-microbiology 31. Duggan, D. J., M. Bittner, Y. Chen, P. Meltzer, and J. M. Trent. 1999. Expression profiling using cDNA microarrays. Nat. Genet. 21:10–14. 32. Dussurget, O., D. Cabanes, P. Dehoux, M. Lecuit, C. Buchrieser, P. Glaser, and P. Cossart. 2002. Listeria monocytogenes bile salt hydrolase is a PrfA-regulated vir-

ulence factor involved in the intestinal and hepatic phases of listeriosis. Mol. Microbiol. 45:1095–1106. 33. Ermolaeva, M. D., H. G. Khalak, O. White, H. O. Smith, and S. L. Salzberg. 2000. Prediction of transcription terminators in bacterial genomes. J. Mol. Biol. 301:27–33. 34. Ferreira, A., D. Sue, C. P. O’Byrne, and K. J. Boor. 2003. Role of Listeria monocytogenes sigma B in survival of lethal acidic conditions and in the acquired acid tolerance response. Appl. Environ. Microbiol. 69:2692–2698. 35. Fraser, K. R., D. Sue, M. Wiedmann, K. Boor, and C. P. O’Byrne. 2003. Role of sigma B in regulating the compatible solute uptake systems of Listeria monocytogenes: osmotic induction of opuC is sigma B dependent. Appl. Environ. Microbiol. 69:2015–2022. 36. Gale, E. 1946. The bacterial amino acid decarboxylases. Adv. Enzymol. 6:1–32. 37. Gibson, G., and S. Muse. 2002. A Primer of Genome Science. Sinauer Associates, Inc., Sunderland, MA. 38. Glaser, P., L. Frangeul, C. Buchrieser, C. Rusniok, A. Amend, F. Baquero, P. Berche, H. Bloecker, P. Brandt, T. Chakraborty, A. Charbit, F. Chetouani, E. Couve, A. de Daruvar, P. Dehoux, E. Domann, G. Dominguez-Bernal, E. Duchaud, L. Durant, O. Dussurget, K.-D. Entian, H. Fsihi, F. G.-D. Portillo, P. Garrido, L. Gautier, W. Goebel, N. Gomez-Lopez, T. Hain, J. Hauf, D. Jackson, L.-M. Jones, U. Kaerst, J. Kreft, M. Kuhn, F. Kunst, G. Kurapkat, E. Madueno, A. Maitournam, J. M. Vicente, E. Ng, H. Nedjari, G. Nordsiek, S. Novella, B. de Pablos, J.-C. Perez-Diaz, R. Purcell, B. Remmel, M. Rose, T. Schlueter, N. Simoes, A. Tierrez, J.-A. Vazquez-Boland, H. Voss, J. Wehland, and P. Cossart. 2001. Comparative genomics of Listeria species. Science 294:849–852. 39. Grant, M. A., S. D. Weagant, and P. Feng. 2001. Glutamate decarboxylase genes as a prescreening marker for detection of pathogenic Escherichia coli groups. Appl. Environ. Microbiol. 67:3110–3114. 40. Grote, A., K. Hiller, M. Scheer, R. Munch, B. Nortemann, D. C. Hempel, and D. Jahn. 2005. JCat: a novel tool to adapt codon usage of a target gene to its potential expression host. Nucleic Acids Res. 33:W526–W531. 41. Holt, K. E., J. Parkhill, C. J. Mazzoni, P. Roumagnac, F. X. Weill, I. Goodhead, R. Rance, S. Baker, D. J. Maskell, J. Wain, C. Dolecek, M. Achtman, and G. Dougan. 2008. High-throughput sequencing provides insights into genome variation and evolution in Salmonella Typhi. Nat. Genet. 40:987–993. 42. Hulo, N., C. J. Sigrist, V. Le Saux, P. S. LangendijkGenevaux, L. Bordoli, A. Gattiker, E. De Castro, P. Bucher, and A. Bairoch. 2004. Recent improvements to the PROSITE database. Nucleic Acids Res. 32:D134–D137. 43. Jay, J. M. 2000. Modern Food Microbiology, 6th ed. Aspen, Gaithersburg, MD. 44. Jeck, W. R., J. A. Reinhardt, D. A. Baltrus, M. T. Hickenbotham, V. Magrini, E. R. Mardis, J. L. Dangl, and C. D. Jones. 2007. Extending assembly of short DNA sequences to handle error. Bioinformatics 23:2942–2944. 45. Kanehisa, M., and S. Goto. 2000. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28:27–30.

SMP_Food Microbiology_CH39.indd

994

Manila Typesetting Company

11/08/2012 07:20AM

995

39.  Genomics and Proteomics of Foodborne Microorganisms 46. Kankainen, M., L. Paulin, S. Tynkkynen, I. von Ossowski, J. Reunanen, P. Partanen, R. Satokari, S. Vesterlund, A. P. Hendrickx, S. Lebeer, S. C. De Keersmaecker, J. Vanderleyden, T. Hamalainen, S. Laukkanen, N. Salovuori, J. Ritari, E. Alatalo, R. Korpela, T. MattilaSandholm, A. Lassig, K. Hatakka, K. T. Kinnunen, H. Karjalainen, M. Saxelin, K. Laakso, A. Surakka, A. Palva, T. Salusjarvi, P. Auvinen, and W. M. de Vos. 2009. Comparative genomic analysis of Lactobacillus rhamnosus GG reveals pili containing a human- mucus binding protein. Proc. Natl. Acad. Sci. USA 106:17193–17198. 47. Kazmierczak, M. J., S. C. Mithoe, K. J. Boor, and M. Wiedmann. 2003. Listeria monocytogenes sigma B regulates stress response and virulence functions. J. Bacteriol. 185:5722–5734. 48. Kim, H., K. J. Boor, and H. Marquis. 2004. Listeria monocytogenes sigma B contributes to invasion of human intestinal epithelial cells. Infect. Immun. 72:7374–7378. 49. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. USA 100:1990–1995. 50. Kline, K. A., K. W. Dodson, M. G. Caparon, and S. J. Hultgren. 2010. A tale of two pili: assembly and function of pili in bacteria. Trends Microbiol. 18:224–232. 51. Ko, R., L. Tombras Smith, and G. M. Smith. 1994. Glycine betaine confers enhanced osmotolerance and cryotolerance on Listeria monocytogenes. J. Bacteriol. 176:426–431. 52. Kulp, D., D. Haussler, M. G. Reese, and F. H. Eeckman. 1996. A generalized hidden Markov model for the recognition of human genes in DNA. Proc. Int. Conf. Intell. Syst. Mol. Biol. 4:134–142. 53. Kurtz, S., A. Phillippy, A. Delcher, M. Smoot, M. Shumway, C. Antonescu, and S. Salzberg. 2004. Versatile and open software for comparing large genomes. Genome Biol. 5:R12. 54. Lipshutz, R. J., S. P. Fodor, T. R. Gingeras, and D. J. Lockhart. 1999. High density synthetic oligonucleotide arrays. Nat. Genet 21:20–24. 55. Liu, J. M., J. Livny, M. S. Lawrence, M. D. Kimball, M. K. Waldor, and A. Camilli. 2009. Experimental discovery of sRNAs in Vibrio cholerae by direct cloning, 5S/tRNA depletion and parallel sequencing. Nucleic Acids Res. 37:e46. 56. Lukashin, A. V., and M. Borodovsky. 1998. GeneMark. hmm: new solutions for gene finding. Nucleic Acids Res. 26:1107–1115. 57. MacLean, D., J. D. Jones, and D. J. Studholme. 2009. Application of ‘next-generation’ sequencing technologies to microbial genetics. Nat. Rev. Microbiol. 7:287–296. 58. Majoros, W. H., M. Pertea, and S. L. Salzberg. 2004. TigrScan and GlimmerHMM: two open source ab initio eukaryotic gene-finders. Bioinformatics 20:2878–2879. 59. Margulies, M., M. Egholm, W. E. Altman, S. Attiya, J. S. Bader, L. A. Bemben, J. Berka, M. S. Braverman, Y. J.

SMP_Food Microbiology_CH39.indd

995

Chen, Z. Chen, S. B. Dewell, L. Du, J. M. Fierro, X. V. Gomes, B. C. Godwin, W. He, S. Helgesen, C. H. Ho, G. P. Irzyk, S. C. Jando, M. L. Alenquer, T. P. Jarvie, K. B. Jirage, J. B. Kim, J. R. Knight, J. R. Lanza, J. H. Leamon, S. M. Lefkowitz, M. Lei, J. Li, K. L. Lohman, H. Lu, V. B. Makhijani, K. E. McDade, M. P. McKenna, E. W. Myers, E. Nickerson, J. R. Nobile, R. Plant, B. P. Puc, M. T. Ronan, G. T. Roth, G. J. Sarkis, J. F. Simons, J. W. Simpson, M. Srinivasan, K. R. Tartaro, A. Tomasz, K. A. Vogt, G. A. Volkmer, S. H. Wang, Y. Wang, M. P. Weiner, P. Yu, R. F. Begley, and J. M. Rothberg. 2005. Genome sequencing in microfabricated high-density picolitre reactors. Nature 437:376–380. 60. Masuda, N., and G. M. Church. 2002. Escherichia coli gene expression responsive to levels of the response regulator EvgA. J. Bacteriol. 184:6225–6234. 61. Masuda, N., and G. M. Church. 2003. Regulatory network of acid resistance genes in Escherichia coli. Mol. Microbiol. 48:699–712. 62. McCormack, A. L., D. M. Schieltz, B. Goode, S. Yang, G. Barnes, D. Drubin, and J. R. Yates III. 1997. Direct analysis and identification of proteins in mixtures by LC/MS/ MS and database searching at the low-femtomole level. Anal. Chem. 69:767–776. 63. Medini, D., D. Serruto, J. Parkhill, D. A. Relman, C. Donati, R. Moxon, S. Falkow, and R. Rappuoli. 2008. Microbiology in the post-genomic era. Nat. Rev. Microbiol. 6:419–430. 64. Mendum, M. L., and L. T. Smith. 2002. Characterization of glycine betaine porter I from Listeria monocytogenes and its roles in salt and chill tolerance. Appl. Environ. Microbiol. 68:813–819. 65. Mohamadzadeh, M., T. Duong, T. Hoover, and T. R. Klaenhammer. 2008. Targeting mucosal dendritic cells with microbial antigens from probiotic lactic acid bacteria. Expert Rev. Vaccines 7:163–174. 66. Nadon, C. A., B. M. Bowen, M. Wiedmann, and K. J. Boor. 2002. Sigma B contributes to PrfA-mediated virulence in Listeria monocytogenes. Infect. Immun. 70: 3948–3952. 67. Nelson, D., and C. Michael. 2000. Lehninger Principles of Biochemistry, 3rd ed. Worth Publishers, New York, NY. 68. O’Farrell, P. H. 1975. High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250:4007–4021. 69. O’Flaherty, S., and T. R. Klaenhammer. 2011. The impact of omic technologies on the study of food microbes. Annu. Rev. Food Sci. Technol. 2:353–371. 70. Olempska-Beer, Z. S., R. I. Merker, M. D. Ditto, and M. J. DiNovi. 2006. Food-processing enzymes from recombinant microorganisms—a review. Regul. Toxicol. Pharmacol. 45:144–158. 71. Oliver, H. F., R. H. Orsi, L. Ponnala, U. Keich, W. Wang, Q. Sun, S. W. Cartinhour, M. J. Filiatrault, M. Wiedmann, and K. J. Boor. 2009. Deep RNA sequencing of L. monocytogenes reveals overlapping and extensive stationary phase and sigma B-dependent transcriptomes, including multiple highly transcribed noncoding RNAs. BMC Genomics 10:641.

Manila Typesetting Company

11/08/2012 07:20AM

996

Advanced Techniques in Food Microbiology

72. Orsi, R. H., M. L. Borowsky, P. Lauer, S. K. Young, C. Nusbaum, J. E. Galagan, B. W. Birren, R. A. Ivy, Q. Sun, L. M. Graves, B. Swaminathan, and M. Wiedmann. 2008. Short-term genome evolution of Listeria monocytogenes in a non-controlled environment. BMC Genomics 9:539. 73. Pandey, A., and M. Mann. 2000. Proteomics to study genes and genomes. Nature 405:837-846. 74. Pepke, S., B. Wold, and A. Mortazavi. 2009. Computation for ChIP-seq and RNA-seq studies. Nat. Methods 6:S22–S32. 75. Phillips, C. I., and M. Bogyo. 2005. Proteomics meets microbiology: technical advances in the global mapping of protein expression and function. Cell. Microbiol. 7:1061–1076. 76. Pop, M., and S. L. Salzberg. 2008. Bioinformatics challenges of new sequencing technology. Trends Genet. 24:142–149. 77. Rajpal, D. K. 2005. Understanding biology through bioinformatics. Int. J. Toxicol. 24:147–152. 78. Shendure, J., G. J. Porreca, N. B. Reppas, X. Lin, J. P. McCutcheon, A. M. Rosenbaum, M. D. Wang, K. Zhang, R. D. Mitra, and G. M. Church. 2005. Accurate multiplex polony sequencing of an evolved bacterial genome. Science 309:1728–1732. 79. Sleator, R. D., C. G. M. Gahan, and C. Hill. 2001. Identification and disruption of the proBA locus in Listeria monocytogenes: role of proline biosynthesis in salt tolerance and murine infection. Appl. Environ. Microbiol. 67:2571–2577. 80. Smith, D., T. Kassam, B. Singh, and J. F. Elliott. 1992. Escherichia coli has two homologous glutamate decarboxylase genes that map to distinct loci. J. Bacteriol. 174:5820–5826. 81. Snyder, L., and W. Champness. 2003. Molecular Genetics of Bacteria, 2nd ed. ASM Press, Washington, DC. 82. Sorek, R., V. Kunin, and P. Hugenholtz. 2008. CRISPR—a widespread system that provides acquired resistance against phages in bacteria and archaea. Nat. Rev. Microbiol. 6:181–186. 83. Southern, E., K. Mir, and M. Shchepinov. 1999. Molecular interactions on microarrays. Nat. Genet. 21:5–9. 84. Stein, L. 2001. Genome annotation: from sequence to biology. Nat. Rev. Genet. 2:493–503. 85. Sundquist, A., M. Ronaghi, H. Tang, P. Pevzner, and S. Batzoglou. 2007. Whole-genome sequencing and assembly with high-throughput, short-read technologies. PLoS One 2:e484. 86. Sybesma, W., C. Burgess, M. Starrenburg, D. van Sinderen, and J. Hugenholtz. 2004. Multivitamin production in Lactococcus lactis using metabolic engineering. Metab. Eng. 6:109–115.

87. Tonge, R., J. Shaw, B. Middleton, R. Rowlinson, S. Rayner, J. Young, F. Pognan, E. Hawkins, I. Currie, and M. Davison. 2001. Validation and development of fluorescence two-dimensional differential gel electrophoresis proteomics technology. Proteomics 1:377–396. 88. Trost, M., D. Wehmhoner, U. Karst, G. Dieterich, J. Wehland, and L. Jansch. 2005. Comparative proteome analysis of secretory proteins from pathogenic and nonpathogenic Listeria species. Proteomics 5:1544–1557. 89. Venkatasubbarao, S. 2004. Microarrays—status and prospects. Trends Biotechnol. 22:630–637. 90. Wade, J. T., K. Struhl, S. J. Busby, and D. C. Grainger. 2007. Genomic analysis of protein-DNA interactions in bacteria: insights into transcription and chromosome organization. Mol. Microbiol. 65:21–26. 91. Wang, Z., M. Gerstein, and M. Snyder. 2009. RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet. 10:57–63. 92. Ward, T. J., T. F. Ducey, T. Usgaard, K. A. Dunn, and J. P. Bielawski. 2008. Multilocus genotyping assays for single nucleotide polymorphism-based subtyping of Listeria monocytogenes isolates. Appl. Environ. Microbiol. 74:7629– 7642. 93. Warren, R. L., G. G. Sutton, S. J. Jones, and R. A. Holt. 2007. Assembling millions of short DNA sequences using SSAKE. Bioinformatics 23:500–501. 94. Weber, H., T. Polen, J. Heuveling, V. F. Wendisch, and R. Hengge. 2005. Genome-wide analysis of the general stress response network in Escherichia coli: sigma(S)-dependent genes, promoters, and sigma factor selectivity. J. Bacteriol. 187:1591–1603. 95. Wemekamp-Kamphuis, H. H., R. D. Sleator, J. A. Wouters, C. Hill, and T. Abee. 2004. Molecular and physiological analysis of the role of osmolyte transporters BetL, Gbu, and OpuC in growth of Listeria monocytogenes at low temperatures. Appl. Environ. Microbiol. 70:2912–2918. 96. Wemekamp-Kamphuis, H. H., J. A. Wouters, R. D. Sleator, C. G. M. Gahan, C. Hill, and T. Abee. 2002. Multiple deletions of the osmolyte transporters BetL, Gbu, and OpuC of Listeria monocytogenes affect virulence and growth at high osmolarity. Appl. Environ. Microbiol. 68:4710–4716. 97. Wolters, D. A., M. P. Washburn, and J. R. Yates III. 2001. An automated multidimensional protein identification technology for shotgun proteomics. Anal. Chem. 73:5683–5690.

SMP_Food Microbiology_CH39.indd

996

Manila Typesetting Company

11/08/2012 07:20AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch40

E. Van Derlinden L. Mertens J. F. Van Impe

40

Predictive Microbiology

In predictive microbiology, the focus is on the quantitative description and prediction of the behavior (growth, survival, and inactivation) of pathogenic and spoilage microorganisms in food products. By combining microbial knowledge, experimental data and mathematical techniques, predictive microbiology generates models to describe and predict microbial behavior in food products (120). Since a series of major food poisoning outbreaks in the 1980s, there has been a remarkable increase of interest in predictive microbiology (121). The implementation of predictive models has already resulted in improved control of food safety and spoilage, for example, by quantifying the effect of storage and distribution on microbial proliferation via the hazard analysis and critical control point (HACCP) system. Recently, predictive microbiology has been accepted as a tool to define safety of (certain) food products in Europe. Predictive models are also being applied in software packages (e.g., ComBase in the United Kingdom and United States and Sym’Previus in France), useful in both academic and industrial environments. In addition, predictive models can be an essential tool for risk control in the optimization of food engineering processes.

In the early stages of predictive microbiology, most studies were related to pathogenic bacteria causing foodborne diseases. In the last 2 decades, the usefulness of predictive models to monitor food spoilage has been recognized and studies have emerged with respect to the modeling of spoilage organisms, mainly yeasts (9, 30, 145). More recently, the concepts of predictive microbiology are also being explored for beneficial microflora such as lactic acid bacteria, which are often used as starter cultures (for an example, see reference 108). Most first-generation predictive models focus on simplicity and general applicability and can be classified as “black box” or “gray box” models, with the main emphasis on the description of the general (population level) microbial behavior as a response to the environment. Their validity to describe pure cultures in simple, liquid media under moderate environmental conditions is widely illustrated and accepted. However, experiments have shown that extrapolation to more complex systems (e.g., structured foods and cocultures) and more stressing environmental conditions (e.g., acid and heat) is generally inappropriate unless validated. A first section of this chapter focuses on modeling trends up to now. The classical primary and ­secondary

E. Van Derlinden, L. Mertens, and J. F. Van Impe, BioTeC—Chemical and Biochemical Process Technology and Control, Department of Chemical Engineering, KU Leuven, B-3001 Leuven, Belgium.

997

SMP_Food Microbiology_CH40.indd

997

Manila Typesetting Company

11/08/2012 07:21AM

998

Advanced Techniques in Food Microbiology

model approach, used to describe growth and inactivation, is discussed as well as probabilistic models, used to describe the growth/no growth (G/NG) boundary. In the following section, contemporary and future modeling trends are listed and the extension of existing models is discussed, including (i) the trend for the incorporation of multiple environmental factors and (ii) the incorporation of the specific aspect of food structure. To move from the macroscopic to the meso- and microscopic levels, the concepts of metabolic networks and individual-based models (IbM) have been introduced. Also, a short overview of mesoscopic models, i.e., models that describe the dynamics of the population as a combination of different compartments, is given. The last section deals with the transfer of predictive microbiology as a tool for food safety and food quality from academia to industry. Specifically, a series of software tools is listed.

NG models are particularly relevant to pathogenic microorganisms with low infectious doses, as their ability to initiate growth implies that they have the potential to be harmful to the consumer (154). With respect to food poisoning, a similar approach can be adopted to develop toxin/no toxin production models (16, 111).

FROM THE PAST TO THE PRESENT The first publications in the domain of predictive microbiology focused mainly on kinetic models describing the growth and inactivation dynamics of single species exposed to a single environmental condition. This type of model can be referred to as a macroscopic model, in which the overall population dynamics is described as a function of time and intrinsic and/or extrinsic conditions. Ever since their introduction, kinetic models have found wide acceptance because they perform well under conditions that permit rapid population development. However, care should be taken that predictions from kinetic models, due to their semimechanistic or empirical basis, are not made beyond the interpolation region. Since no growth conditions are usually omitted from the model fitting process, conditions close to the G/NG boundary, which are often of industrial interest, may not lie within this region (22). Moreover, when a microbial population experiences progressively harsher conditions and moves towards conditions that will eventually preclude growth, variability increases significantly, and kinetic models will fail to provide accurate descriptions (40). Under these circumstances, it is more useful to consider the probability that growth is likely to occur at all rather than the growth rate (154). This type of approach is now widely used to develop probabilistic models that define combinations of environmental factors representing the boundary between growth and no growth, which are, in fact, a quantitative description of the hurdle concept (121, 122, 150, 183). In practice, G/

SMP_Food Microbiology_CH40.indd

998

Deterministic Models

Primary models quantify microbial cell counts as a function of time. Primary models have been developed for both growth and inactivation. The most general expression for microbial behavior as a function of time, given homogeneous environmental conditions, reads as follows (26): dNi (t ) = µ éë Ni (t ), < env (t ) >, < phys (t ) >, < P (t ) >, dt < S (t ) >, < N j (t ) >, �ùû × Ni (t ) In this equation, Ni(t) (in CFU/ml) represents the cell density of species i at time t (h), and µ (1/h) is the overall specific evolution rate. Growth is sustained when µ is positive, and when µ is negative, cells are inactivated. The magnitude of the growth or inactivation rate is determined mainly by (i) the microbial environment, i.e., the physicochemical properties [env(t)]; (ii) the physiological state of the cells [phys(t)]; (iii) the concentration of the metabolic products [P(t)]; (iv) the availability of the substrate [S(t)]; and (v) interactions with other species [Nj(t)]. To be able to completely predict the microbial behavior, secondary models, i.e., models that can describe the evolution of the influencing factors as a function of time, have to be incorporated.

Growth

In the classical approach, bacterial growth is modeled autonomously by a logistic equation and the model does not include substrate consumption or growth inhibition by toxic products. Initially, growth curves were most often described by an adapted version of the Gompertz model (for an example, see reference 207) or by the logistic model. The main disadvantage of these models is the lack of biologically significant parameters. Also, due to the static nature of these models, implementation under time-varying conditions is not possible. In the beginning of the 1990s, additional primary models were published. Generally, these models focus on the description of the three main phases of growth, i.e., lag, exponential growth, and stationary phase. Baranyi and Roberts (20) added an additional model

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

999

block to the logistic equation to enable description of the lag phase. Rosso (158) also extended the logistic model by imposing an initial phase in which the cell number remains constant. Hills and Wright (89) built a new primary model that described both the cell numbers in a population and the population’s mass. In 1997, Buchanan et al. (33) presented a simple growth model, i.e., a sequence of three straight lines, each representing one of the phases of growth. The disadvantage of both the Rosso and the Buchanan models is their static character such that implementation under dynamic conditions is limited. In contrast, the dynamic model of Baranyi and Roberts consists of two ordinary differential equations enabling simulation under continuously changing conditions when combined with a secondary model(s) (see below). Although the model is not without limitations, the Baranyi and Roberts growth model is among the most used in predictive microbiology because it has biologically interpretable parameters and can be used under dynamic conditions (109). Generally, all these models do not include mechanistic knowledge to describe the transitions between the different growth phases. Stationary phase, for example, is described by including the asymptotic value, i.e., the maximum cell number, as a parameter. Van Impe et al. (191) introduced a novel class of macroscopic growth models which do take (micro)biological phenomena governing the microbial growth process into account. Their work focused on the transition of the exponential growth phase to the stationary phase, which can be induced through toxic product accumulation and/or substrate exhaustion. This modeling approach was also employed to describe the more complex case study of coculture inhibition of Listeria innocua mediated by lactic acid production of Lactococcus lactis (146).

Inactivation

Initially, heat and other forms of inactivation were assumed to follow first-order kinetics in analogy with a chemical reaction, and microbial inactivation dynamics were monitored via the D (decimal reduction time) and z (heat resistance) parameters. However, when applying milder decontamination processes like mild heat treatments, it was observed that the survival curves of some bacteria are not log-linear but can have a range of different forms (e.g., sigmoid, convex, or concave). As a response to this, different static and dynamic inactivation models have been developed. To describe the effect of decontamination procedures, the Chick-Watson (197) model or the Hom model (93) have been applied. These models can describe survival curves that show a shoulder and/or a tailing phase. The disadvantage of

SMP_Food Microbiology_CH40.indd

999

these models is that their parameters are without any biological interpretation. Overall, these models are only rarely applied. A nice overview of primary models often applied in predictive microbiology is given by Geeraerd et al. (70). In addition, a novel model structure that can accurately describe sigmoid inactivation curves, using parameters that have a biological interpretation, is presented. Models most often used to describe microbial inactivation are the adapted Weibull model, the model of Geeraerd et al. (70), the logistic model, and the adapted Gompertz model (207). During the last decades, few new model structures have been published. New model structures relate mainly to adaptations of existing models. An additional block for the Geeraerd inactivation model was presented by Valdramidis et al. (185) to describe microbial inactivation dynamics under dynamic temperature conditions that induce heat adaptation. In 2005, Albert and Mafart (6) published an adapted version of the Weibull model. Gil et al. (74) created a dynamic version of the Gompertz model by differentiating with respect to time. To describe the efficiency of decontamination processes, dose-response models can also be applied. This type of model describes the microbial response as a response to different levels or doses of the selected decontaminant or procedure, that is, they quantify the fraction of surviving cells as a function of time and exposure level. The population level is typically sigmoidal. Below the noninhibitory concentration, only a limited effect can be observed. Gradually increasing the stress level reduces the fraction of surviving cells until the MIC, i.e., the level above which no survival is possible, is reached. Examples of dose-response models are the Fermi equation (91, 143, 144) and the Hülsheger et al. model (96), developed to describe pulsed-electric-field decontamination. Examples of implementations can be found in references 59 and 170. In a next step, secondary models that describe the influence of changing environmental conditions on primary models, i.e., on their parameters, are developed. Secondary models fall into two groups. Some models, i.e., square root models (SQRT) and cardinal parameter models, include (some) biologically or graphically interpretable parameters and can be extended towards more environmental factors via a multiplicative approach (for examples, see references 152, 153, 159, and 160). Furthermore, they are parsimonious and have a high fitting quality. On the other hand, response surface models and neural networks are completely empirical and as such do not presume an a priori knowledge of the underlying relationship (for examples, see references 22 and 69).

Manila Typesetting Company

11/08/2012 07:21AM

1000

Advanced Techniques in Food Microbiology

Secondary models for the growth rate Secondary models developed in the early stages of predictive microbiology focused mainly on the effect of temperature, and to a lesser extent on pH and water activity (aw). A myriad of models have been developed to describe the effect of temperature on the microbial growth rate, e.g., the model of Hinshelwood (90), the Schoolfield et al. model (166), SQRT-based models (152, 153), and the cardinal temperature model with inflection (160). Afterwards, both SQRT and cardinal parameter models were constructed considering multiple environmental factors, based on the gamma concept. In this concept, it is assumed that different influencing factors act independently such that they can be used in a multiplicative way in the secondary models. McMeekin et al. (119) and Miles et al. (131) extended the SQRT model with a factor to describe the effect of aw on the growth rate of Staphylococcus xylosus and Vibrio parahaemolyticus, respectively. An additional SQRT factor for pH was also included by Adams et al. (2). At the end of the 1990s, additional SQRT model factors were developed for specific intrinsic or extrinsic factors, e.g., lactic acid (149) and carbon dioxide (47). Similarly, a series of cardinal models that include multiple influencing factors have been developed. There are cardinal models that describe the effect of temperature, pH, and/or organic acids on the microbial growth (15, 159, 160) or the influence of temperature, salt, and/or aw on the growth of molds (43, 161, 163). Also, response surface/polynomial models and, to a lesser extent, artificial neural networks have been constructed to describe the microbial growth response as a function of time and environmental conditions. Response surface models or polynomial models in predictive microbiology are typically equations including quadratic or cubic terms that model the relation between primary (log- or ln-transformed) model parameters (e.g., y = duration of lag/shoulder phase, inactivation/ growth rate) and several environmental factors (e.g., f­1 and f2), including their interactions.

Moreover, due to the large number of parameters, polynomial models can be overparameterized and, as such, model not only the underlying microbial response but also the associated experimental errors (22). To prevent this phenomenon, response model structures should be of a limited order, i.e., preferably a second order. According to Baranyi et al. (22), the model performance can be further improved by incorporating expected properties into the model structure (e.g., convexity or extremes). To find a good balance between a model that is complex enough to fit the data well and a model that is simple enough that overfitting is avoided, several model selection procedures can be used. These procedures are based on the selection of the most significant terms of the model equation. (i) The forward selection procedure starts from a simple model, e.g., a model that contains only one intercept. It adds terms to this model until further additions do not improve the fit. In each step, it selects the most significant terms of the terms eligible for entry (see, for example, reference 38). (ii) The backward elimination procedure begins with a complex model, e.g., the complete model, and sequentially removes terms until further deletion leads to a significantly poorer fit. In each step, it selects the term that least affects the goodness of fit of the model. For examples, see references 164 and 204. (iii) The stepwise selection procedure combines the previous two procedures. It starts from a simple model and adds the most significant terms. When the added term causes another term in the model to become insignificant, the latter is removed. The stepwise selection procedure terminates if no further term can be added to the model or if the term just entered into the model is the term removed in the subsequent backward elimination step. When performing these selection procedures, it is important to take into account the model hierarchy. For instance, it is not advisable to add squared and interaction terms before the corresponding main terms have been added. Similarly, it is inappropriate to remove a main term if the model has squared and interaction terms involving that main term. An example of the stepwise procedure implementation can be found in reference 132. When applying one of the above methods to optimize the structure of the response surface model, it is important to remember that different procedures can result in different models, no one of which is always best (4). In addition, to overcome the problem of model overfitting, the constrained polynomial approach, i.e., a combination of the flexibility of a polynomial model with knowledge of the microbial behavior, can be implemented (71). Finally, response surface model struc-

ln(y) = a0 + a1·f1+ a2·f2 + a3·f1·f2 + a4·f12 + a5·f22 Response surface models are easy to implement, enable fast model parameter identification, and require no a priori knowledge about the relation between microbial behavior and the environment. However, polynomial models can be characterized by a small interpolation region, i.e., smaller than the region given by the endpoints of the experimental setup (22). In addition, the determination of the true interpolation region is not straightforward.

SMP_Food Microbiology_CH40.indd

1000

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1001

tures are only applicable for the selected microbiological study, i.e., food (model) systems and strain(s). Examples of implementations to describe growth kinetics can be found in references 34, 35, 58, and 155. Artificial neural networks (ANN) are data-driven, nonlinear modeling techniques based on the principles of the human nervous system and consist of numerous processing units that are interconnected. ANN have been recognized to be more flexible and to have higher accuracy than the traditional modeling approaches (24). For implementations on modeling growth dynamics, see references 67, 68, and 100.

Secondary models for inactivation To describe the effect of temperature on the microbial inactivation rate, mainly the Arrhenius and the Bigelow model are applied. The original Arrhenius model was developed, based on thermodynamics, to describe chemical reaction kinetics. With respect to microbial growth, the Arrhenius equation assumes a linear relation between the growth rate (m) and (1/T). However, this relationship exists only for a limited temperature range, i.e., the normal physiological range. At temperatures below or above the normal physiological range, the evolution rate can differ significantly from the value calculated with the Arrhenius model. This implies that the energy of activation is not constant, but instead its magnitude increases with the reciprocal of the absolute temperature (47, 120). To model the temperature effect at more extreme temperatures, modifications to the Arrhenius model have been made (47, 90, 166). In addition, further model extensions have been developed to include additional processing conditions, e.g., pH (37, 48). The Bigelow model was specifically developed to describe the effect of temperature on the microbial inactivation rate, with D (decimal reduction time) and z (the change of temperature required to achieve a 10-fold change in D) as important parameters. Especially the D parameter is an often-applied parameter to describe the heat resistance properties of specific strains. The Bigelow-type model has been extended to describe treatments that include additional physicochemical factor(s) or treatments next to temperature, e.g., for pH (113), aw (65), and ultrasound processing (3). More recently, a series of response surface models have been constructed to describe microbial inactivation kinetics as a function of the selected process treatment, e.g., oxidizing water (56), high hydrostatic pressure (28), essential oil (172), organic acids (196), and ultrasound and irradiation processing (201). Implementations of ANN for inactivation can be found in references 60 and 87.

SMP_Food Microbiology_CH40.indd

1001

Secondary models for the lag phase duration Swinnen et al. (177) developed a secondary model for the duration of the intermediate lag time, induced by abrupt upward temperature shifts. Hereto, the duration of the lag time was plotted as a function of the absolute temperature difference. Different model structures were evaluated to describe the emerging relation. Implementations of response surface models can be found in references 1, 17, 35, 97, and 141. References using neural networks are almost nonexistent. Predictive modeling and data transformations Within the domain of predictive microbiology, different data transformations have been suggested to stabilize the variance. The most applied transformations are (i) the (natural) logarithm and (ii) the square root. Generally, data transformations are applied to obtain homogeneously distributed variances when applying regression analysis, i.e., model fitting procedures. When describing microbial growth or inactivation, the cell number as a function of time is presented as the logarithm (e.g., Gompertz model) or natural logarithm (e.g., Baranyi and Roberts model) to stabilize the variance. As a result, measurement variability can be assumed as normally distributed with a mean of zero. For secondary models, the growth rate as a function of the selected environmental condition(s) is often transformed using square root also to obtain normally distributed data uncertainty and variability. According to Alber and Schaffner (5), the logarithmic transformation is more appropriate than the square root transformation, for growth rates as a function of temperature. On the contrary, Dantigny and Bensoussan (46) advised not to use a logarithmic transformation to stabilize the variance of the radial growth rates (expressed as mm/day) of molds. Instead, a square root transformation was more appropriate. Occasionally, square root transformations have also been applied on G/NG data (35). According to Zwietering et al. (208), no transformation is needed when studying the maximum cell count as a function of environmental factors. They observed that a square root transformation is the most appropriate for the growth rate while a logarithmic transformation is more suited for the duration of the lag phase.

Probabilistic Modeling

Generally, probabilistic models describe the chance that a certain event (e.g., growth, survival, or toxin production) occurs given the intrinsic and extrinsic factors. In the domain of predictive microbiology, a probabilistic modeling approach is applied mainly (i) for the characterization of the G/NG boundary; and (ii) to quantify

Manila Typesetting Company

11/08/2012 07:21AM

1002

Advanced Techniques in Food Microbiology

the chance of microbial survival, recovery or spoilage after certain processing treatments. Both topics are usually described using polynomial models. To a lesser ­extent, probabilistic models are also applied to describe toxin/no toxin production or the time to toxin production in foods (16, 57, 66, 111). Kinetic growth models, as described in the above section, are most often deterministic, i.e., the model outcome is precisely determined via the relations and model parameter values taken. Starting from identical conditions, the model will always produce the same outcome. Biological variability and/or model parameter ­uncertainty can be taken into account using Monte Carlo simulations. In this procedure, microbial ­dynamics are simulated, at least 1,000 times, using model ­parameters that are randomly sampled from the normal distribution of the model parameter (its mean and standard deviation). This way, small differences in single-cell ­dynamics due to biological variability and uncertainty due to measurement errors and model incompleteness are considered. However, this approach still assumes that each cell within the population has the same global dynamics, i.e., all cells are assumed to grow. When conditions become more severe, heterogeneity within a population is more expressed and differences in response between cells within a population can be observed, e.g., some cells can initiate growth, while others remain dormant or even die. This random distribution of microbial responses can be described by probabilistic models in which the individual characteristics are no longer assumed to be identical within a population.

lenge tests, and the evaluation and optimization of the HACCP system. Probabilistic models can be found throughout the domain of predictive microbiology. Initially, G/NG models were developed for pathogens (specifically with a low infectious dose), as the smallest outgrowth can result in foodborne diseases. Later on, applications also shifted towards food spoilage microorganisms. Also, a variety of environmental conditions have been included. Initially, models mainly tackled the most straightforward factors like temperature, pH/acid concentrations, and aw. Recently, also more (product) specific aspects have been taken into account, like modified atmosphere and nitrites. Various approaches have been considered to build G/NG models. In general, they can be classified in two groups (84), deterministic and probabilistic. Determin­ istic G/NG models are based on the assumption that the growth boundary is abrupt and can be represented by a single line. They are developed based on kinetic models with the G/NG boundary set at the conditions where the growth rate is equal to zero. The major disadvantage of this model type is that it does not take into consideration synergistic effects. Approaches based on the concept of Minimum Convex Polyhedron are also considered deterministic. Probabilistic G/NG models quantify the likelihood that microorganisms grow as a function of environmental conditions. The G/NG region is a transition zone where the growth probability increases from 0 to 100% when going from the detrimental to the more favorable environmental conditions. Important groups of probabilistic G/NG models are logistic regression models and ANN. Most G/NG models found in the literature are logistic regression-type models in which a simple polynomial relation is used to quantify the chance of growth given the selected intrinsic and extrinsic conditions. In addition to logistic regression, the advantages of implementing ANN-based models to describe the G/NG boundary have recently been recognized. The stochastic G/NG boundary can be described by a probabilistic neural network, which combines statistical theory aspects with traditional neural networks. More information about probabilistic neural networks can be found in reference 86. Implementations of neural networks to describe ­microbial G/NG regions can be found, for instance, in references 61 and 85. As with response surface models, logistic regression models are highly sensitive to data overfitting. As a consequence, it is advised to implement one of the model selection procedures explained above (forward selection

Modeling the G/NG Region

G/NG models describe the interface between growthsupporting environmental conditions and the no-growth region. The microbial growth range is determined by all environmental factors. For many situations, it has been observed that the growth range with respect to one factor is reduced when a second factor is at a suboptimal or detrimental level. As such, a combination of rather mild stresses, which have separately only a limited effect, can significantly suppress the microbial outgrowth. This concept is referred to as the hurdle concept and is very often used by the food industry (104). Frequently applied stress factors are temperature, acid/pH, aw, modified atmosphere, and preservatives. The implementation of these models enables the development of more natural food products in which microorganisms are not able to grow (based on the hurdle concept). They can also be of help for the determination of shelf life, the selection of chal-

SMP_Food Microbiology_CH40.indd

1002

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1003

or backward selection or stepwise selection) to end up with a model that contains only significant model terms, i.e., a model that mainly fits the observed microbial dynamics. Due to the purely data-driven approach of G/NG models, the quality of the experimental data is of ­utmost importance, i.e., the accuracy of the G/NG boundary is determined by the number of experimental data collected. Especially under conditions approaching the G/NG boundary, a higher number of repetitions will yield a more accurate approximation of the growth probability. As a consequence, highly accurate models ­require a demanding experimental scheme; for example, a 5% growth probability requires at least 20 replicates. In practice, the experimental scheme selected should ­include a balance between practicality and accuracy (for examples, see references 129 and 195). G/NG models are empirical models that do not ­include any mechanistic information about the underlying relationship between the microbial behavior and the chemical and physical food characteristics. Their transferability is specifically limited because of three factors. (i) The general mechanisms behind the synergistic relation between different environmental conditions are not yet fully understood. As a result, G/NG models are valid only for the specific environment and microorganism for which they have been constructed. (ii) G/NG boundaries are known to depend on the initial inoculation level. A decrease in the inoculum size lowers the growth probability (140). Possibly, this can be ­explained by the distribution of physiological cell states as is observed for initial lag times (18). (iii) The predicted response is highly related to the time span considered for defining the observed growth or no growth. When times beyond the experimental range are considered, growth at nonsupporting conditions or a higher growth percentage can be observed as a result of the resuscitation of ­injured cells. To summarize, an experimental scheme will always be a trade-off between what is practically feasible and what is relevant for the food industry, e.g., shelf life.

PRESENT AND FUTURE RESEARCH TRENDS Most existing primary and secondary models enable an accurate description of microbial dynamics under (nonstressing) dynamic conditions for liquid systems. However, over the last decades, it has been widely recognized that these models can fail when applied to real food products and under more realistic, more stressing conditions. Baranyi and Roberts (21) divided the sources of errors in predictive microbiology models into five groups:

SMP_Food Microbiology_CH40.indd

1003

1. Homogeneity error: models do not account for the heterogeneity of foods, strains within a microbial species, and individual cells within a population. 2. Completeness error: models are a simplification since they only include a limited number of environmental factors. 3. Model function error: models are only an approximation of reality because of their empirical nature. 4. Measurement error: limitations in measurement methods lead to inaccuracies in raw data. 5. Numerical procedure error: numerical procedures used for model fitting and evaluation lead to errors. It is unavoidable that a certain model function error remains, i.e., a model will always be a simplified representation of the true event. However, the lack of model correctness can be minimized by reducing the completeness error. To date, most studied (model) systems consider rather simple liquid systems, taking into account only a limited number of intrinsic and/or extrinsic factors (e.g., temperature, pH, and aw). However, morecomplex elements, like the high number of influencing factors, the physicochemical properties of the food structure (and spatial heterogeneity), the presence of background flora and microbial competition, and stress adaptation, are rarely included. As a result of the above-mentioned shortcomings and because predictive models are developed mainly on the basis of experimental data acquired in perfectly controlled laboratory environments, industrial applications of predictive models are not recommendable before a proper validation study, using independent experimental data, has been conducted.

An Eye on Additional Model Factors Extended Secondary Models

Towards the beginning of this century, more-complex secondary growth rate models, i.e., models containing more than three factors, were developed. Devlieghere et al. (54) developed a model that includes temperature, aw, CO2, and sodium lactate to describe the dynamics of the spoilage organism Lactobacillus sakei. The model constructed by Ross et al. (156) to describe the growth of Escherichia coli includes temperature, aw, pH, and lactic acid. Augustin and Carlier (11) added some model factors (based on the concept published by Dalgaard [44]) to the extended cardinal parameter model (describing T, aw, and pH) to include qualitative factors and inhibitory substances, such as phenol, sodium nitrite, sodium lactate, sodium benzoate, and potassium sorbate. Model

Manila Typesetting Company

11/08/2012 07:21AM

1004

Advanced Techniques in Food Microbiology

parameters were defined to enable description of the growth rate and lag time of Listeria monocytogenes. The above models do not include possible synergistic effects between the considered environmental conditions. In the last decade, discussion has emerged with respect to the validity of the gamma concept. According to Bidlas and Lambert (27), the gamma hypothesis is a valid approach to describe the effect of multiple environmental factors on the microbial growth rate. For E. coli, for instance, they showed that sodium acetate and potassium sorbate worked independently at pH values between 5.5 and 6.5. However, many researchers have observed that synergistic effects between certain environmental factors exist. In such case, different intrinsic or extrinsic properties of the food (model system) no longer act independently but rather synergistically. It can be claimed that the gamma hypothesis is no longer valid when the value of typical model parameters (Tmin, pHmin, etc.) changes when other environmental factors become more stressing. Most often, synergistic interactions are observed when approaching the G/NG boundary. For instance, pHmin was shown to increase with decreasing temperature (149). Also, aw,min has been observed to be higher at lower temperatures (162, 178) and at lower pH values (115). An increased antimicrobial effect of nisin against L. monocytogenes was ­observed when combined with CO2 (136). With respect to organic acids, often used as preservatives, interactions have been observed with environmental factors, e.g., with temperature and/or pH (95, 186). Coroller et al. (41) observed synergistic effects between different ­organic acids (e.g., propionic acid, acetic acid, citric acid, and lactic acid). Starting from the observations of synergistic interactions, model correction factors that take into account the interactions between influencing factors have been constructed. In 2000, Augustin and Carlier (12) adapted a previously published multiplicative secondary model to describe independently the effects of environmental factors on the growth rate of L. monocytogenes (11) (see above) by including interactions between the environmental factors. Additional terms were constructed from the G/NG boundary observed for the different combinations of environmental factors. By considering the interactions, model prediction accuracy was improved significantly, i.e., the percentage of fail-safe and fail-dangerous growth predictions was decreased. To enable accurate description of the growth of L. innocua as a function of temperature, pH, and organic acids, Le Marc et al. (105) proposed a model expansion that includes the interaction between these factors in the neighborhood of the growth limits. Cardinal parameter

models were combined for the factors T, aw, and pH. Terms based on the MIC were added for the different organic acids. Interaction terms were based on the assumption that each environmental factor contributes to the interactions relative to their individual effect on the growth rate. By including the interactions, growth rates were described accurately not only under moderate conditions but also in the proximity of the G/NG boundary. Coroller et al. (41) constructed model correction factors that enable to quantify the synergistic effect of different weak acids. Here, model factors were added to the cardinal parameter model with interactions as developed by Le Marc et al. (105) (describing the effect of pH and aw), which also take into account the concentration of the undissociated form of the weak acids and the interactions between the different weak acids. Augustin et al. (13) published an extended model that can describe the growth rate of L. monocytogenes in dairy products, meat, and seafood. This model is a cardinal parameter model that includes model terms describing interactions between the considered factors. The interaction terms are based on the assumptions of Le Marc et al. (105). The effects of temperature, aw, pH, (sodium) nitrite, phenol, CO2, and lactic acid are taken into account. According to these authors, models that include interactions perform significantly better near the G/NG boundaries. To accurately describe the dynamics of L. monocytogenes in meat products, an ANN model was developed by the Danish Meat Research Institute (DMRI) (83). This model includes the effects of temperature, pH, sodium chloride in the water phase, acetic acid and lactic acid in the water phase, sodium nitrite added to the product, and CO2 in the packaging atmosphere, and interactions between the factors. This ANN was the basis for the software tool developed by the DMRI (see “Other software tools” below). An extended model that accurately describes the ­dynamics of L. monocytogenes in seafood and meat products was developed (124, 125). This model incorporates the effect of temperature, aw, pH, phenol, nitrite, CO2, lactic acid, acetic acid, citric acid, benzoic acid, sorbic acid, and diacetate. Possible interactions between the considered factors are described using the approach of Le Marc (105). The Seafood Safety and Spoilage Predictor web-based software is constructed around this model (see “Other software tools” below).

SMP_Food Microbiology_CH40.indd

1004

From Liquid Environments to Structured Food (Model) Systems

The (secondary) models discussed above, even though highly complex, do not consider the growth morphology,

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1005

i.e., data extracted from experiments in liquid environments and real food products are considered together. As such, no difference is made between growth in liquid, i.e., planktonic growth, and growth in solid(like) systems, i.e., growth as colonies. However, food structure is an important factor influencing microbial dynamics in real food products. Food structure affects microbial behavior mainly by the (limited) distribution of water, nutrients, and metabolites and the restricted mobility of the bacteria (198). The constrained growth in a structured, possibly heterogeneous, medium induces additional stress, which results in lower growth rates and/or smaller habitat domains. Measurements of the rate of colonial growth on surfaces showed a decreased growth rate, and comparisons of the growth rates of Salmonella enterica serovar Typhimurium affected by increasing salt or sucrose followed the order broth > immersed colonies > surface colonies (31). Dens and Van Impe (51) presented a model that takes into account the variability of microbial growth with respect to space. The model describes two phenomena: (i) local evolution of biomass and (ii) transfer of biomass through the solid(like) medium. Theys et al. (179) studied the effect of gelatin concentration (0%, 1%, and 5%) and the growth dynamics of Salmonella Typhimurium in tryptic soy broth as a function of pH and aw at 20°C. The gelatin systems were characterized via rheological measurements. It was observed that growth rates decreased drastically when going from 0% to 1% of gelatin. In contrast, only a minimum effect emerged when the gelatin concentration was further increased to 5%. The difference in growth dynamics was attributed mainly to the difference in oxygen concentration between the (stirred) liquid system and the static gelatin environment. A secondary model was built, taking into account the gelatin concentration. In a next step, the validity of this model was confirmed for Salmonella Typhimurium grown in milk and a cheese model system, and for other microorganisms, i.e., Listeria innocua, L. monocytogenes, and Lactococcus lactis (180). Other studies have confirmed that growth in/on surfaces is slower than growth in liquid media (7, 101, 126, 198). In contrast, Theys et al. (182) observed an opposite effect closer to the G/NG boundary of Salmonella Typhimurium as a function of aw/NaCl and pH. At 1% and 5% gelatin, the G/NG boundary was situated under more stressing conditions. A similar observation­ was made by Mertens et al. (129), who noticed that the growth probability of Zygosaccharomyces bailii increased under stressing conditions (pH, aw at 22°C and 30°C) when moving from a liquid to a gelled environment.

SMP_Food Microbiology_CH40.indd

1005

Summarizing, it can be stated that the solid(like) versus liquid character of a food (model) system can (i) decrease the growth rate and/or decrease the growth region or (ii) increase the growth (probability) under more stressing conditions or (iii) have only a limited effect. As such, experimental data obtained from a liquid environment do not guarantee a model that is valid in a solid(like) environment. For instance, Schvartzman et al. (169) compared the G/NG boundary of L. monocytogenes in milk, broth, and a lab-scale cheese. These authors concluded that models obtained for the liquid systems were insufficient to describe the behavior of the pathogen in the structured cheese matrix. The fact that most predictive models do not take into account the more solid(like) character of food products, compared to the liquid properties of the lab media, can be considered one of the major shortcomings of predictive microbiology. Up to now, predictive models that take into account the effect of structure on the studied microbial dynamics are scarce. During the last decade, a myriad of studies have been performed to evaluate microbial dynamics in real food products instead of typical laboratory broth systems. These experiments guarantee that the constructed model is accurate for the selected food product(s). However, this approach yields models that are valid only for these specific foods and have a very limited transferability to other products. In contrast, the use of food model systems enables identification of the parameters and/or aspects that determine the difference in dynamics. In practice, this can be achieved through the careful selection and development of experimental food model systems. For instance, Mertens et al. (128) constructed a model system that represents acidified sauces like ketchup and dressings. More than 10 different gelling agents were evaluated to finally obtain a xanthan gum and carbopol-based system that mimics the rheological character of the selected food products and that is transparent, which enables the use of optical density as a measurement technique. Noriega et al. (138) deve­ loped a synthetic meat system by adding k-carrageenan to a broth that contains meat extract, proteose peptone, tryptone, and glucose. Jeanson et al. (99) combined a retentate, obtained from ultrafiltrated milk, and the coagulant agent Maxiren to build a cheese model matrix. Overall, systematic studies investigating the effect of structure on microbial dynamics are limited. One of the most important reasons is the impracticality of microbial studies in structured environments (128). In order to evaluate the effect of the structured matrix in a systematic and consistent way, the experimental setup must enable accurate control and sampling. The gel ­cassette

Manila Typesetting Company

11/08/2012 07:21AM

1006

Advanced Techniques in Food Microbiology

system developed by the Institute of Food Research (United Kingdom) is a frequently used experimental setup (101, 126, 184). Other researchers use petri dishes (7, 176). In these systems, gelatin or agar is mainly used as the gelling agent. Sporadically, other gelling agents have been used (14, 94, 128, 134). To date, the structure of food model systems has been quantified mostly by using the concentration of the gelling agent. This is, however, a very impractical variable, as it does not allow comparisons between different gelling agents and extrapolation to real food products. As an alternative, the use of rheological measurements was introduced as a more general measurement of food structure ­characteristics (128, 129). A significant difference between liquid versus solid systems emerges at the level of cell versus population dynamics. In solid(like) environments, populations no longer grow planktonically but appear as colonies. Cell numbers and population numbers differ depending on the colony studied. Existing models are generally based on the assumption of a homogeneous system expressed as CFU/ml. However, homogenization of solid(like) systems is not a true representation of the possible heterogeneous character of the populations, e.g., differences in colony size and distribution. On the contrary, a single sample might not be a good representation due to the heterogeneous properties of the studied food (model) system. Few researchers have focused on describing the dynamics of colonies as a function of time and environmental conditions. Skandamis et al. (174) studied colony ­dynamics of E. coli as a function of temperature and pH. Cell number was monitored using plate counts. Colony surface area as a function of time was obtained from the pixels of ­microscopic images. Both cell count and colony surface area were fitted with the Baranyi and Roberts model, growth rates for cell count and colony surface were ­obtained, and a ­relation between the two variables was constructed. Theys et al. (181) also ­related colony volume to cell number. Using microscopic ­ images and image analysis software, the colony size of Salmonella Typhimurium as a function of time could be monitored. Simultaneously, cell number was determined via plate counting. It was observed that during the stationary phase the colony volume still increased although the cell number remained approximately constant. Based on this observation, the existence of a dead fraction was shown, and its proportion as a function of time was described.

investigated, not only because of their ability to inhibit outgrowth of pathogens and spoilage microorganisms in fermented foods but also for their potential to act as protective cultures in minimally processed foods (108). Malakar et al. (114) concluded that interactions between microorganisms can be ignored in food products under most conditions. They observed that interactions are relevant only when high concentrations of background flora (i.e., higher than 108 CFU/ml) are present. This corresponds with the Jameson hypothesis (73), i.e., initially both populations grow as though the other is not present. Interactions appear only when one of the strains approaches its stationary phase. Generally, this hypothesis is the basis for the modeling of microbial interactions. Specific models have been developed to describe the effect of protective lactic acid bacteria on the growth, inhibition, and inactivation of pathogens as a function of the lactic acid produced, i.e., the pH and lactic acid concentration (98, 193, 194). Inspired by these models, Van Impe et al. (191) introduced a novel class of macroscopic predictive microbial growth models that do take (micro)biological phenomena governing the microbial growth process into account. Their work ­focused on the transition of the exponential growth phase to the stationary phase, which is induced through an ­increasingly toxic product accumulation and/or substrate exhaustion. This modeling approach was extended to describe the more complex case study of coculture inhibition of L. innocua mediated by lactic acid production of L. lactis (146). Mejlholm and Dalgaard (123) studied the growth of lactic acid bacteria as a function of environmental conditions relevant for seafood products and their ­inhibiting effect on L. monocytogenes. In a first step, the dynamics of lactic acid bacteria as a function of ­ environmental conditions like aw and pH were characterized based on information extracted from the literature, experiments in laboratory media, and real seafood. In a second step, this model was combined with the model describing the growth dynamics of L. monocytogenes in seafood to quantify the interactions. In this model, the interaction is modeled as follows: when the lactic acid bacteria reach their stationary phase, growth of Listeria is also stopped. Similarly, growth of lactic acid bacteria stops when L. monocytogenes reaches its maximum cell count. For many cases, this approach, based on the Jameson effect, seems to yield a good description of the relation between these two microorganisms. Despite these efforts, interactions between microorganisms, especially in cases where microbial cells are immobilized within the food matrix such as in cheese,

Modeling Microbial Interactions

Microbial interactions occur in many real food systems. In this context, lactic acid bacteria are increasingly being

SMP_Food Microbiology_CH40.indd

1006

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1007

remain poorly documented, and implementation of these phenomena into predictive models and tools is rarely performed. Antwi et al. (7, 8) extended the models of Vereecken et al. (193, 194) to coculture studies in a gelatin gel matrix. With respect to real food products, the literature shows that several studies have been performed for cold-smoked salmon (49, 123). Guillier et al. (82) modeled the competitive growth between L. monocytogenes and the biofilm microflora of smear cheese wooden shelves. Le Marc et al. (106) modeled the interaction between a starter culture and growth of Staphylococcus aureus in milk.

Towards Meso- and Microscopic Models

Unexpected cell dynamics are often observed, particularly during the lag phase, in heterogeneous environments and/or under stressing conditions. These cannot be explained using the macroscopic approach generally applied in predictive microbiology. In the approaches presented above, microbial growth is modeled from a macroscopic viewpoint by describing the dynamics of a population parameter (e.g., total cell number) as a function of time (e.g., the model of Baranyi and Roberts) and environmental conditions. In the typical growth and inactivation models, parameters are mainly assumed to be deterministic, i.e., one typical value represents the typical population dynamics, and individual cell variability is not taken into account. However, the applicability and reliability of existing models under more realistic conditions will definitively be improved by considering variability between cells within a population. By looking inside the black box, underlying mechanisms can be unraveled to finally increase overall model validity. In the last decade, a quest for more mechanistically based predictive models has started (32, 118, 190). Fundamental microbiological research, in general, is conducted at three levels, i.e., the macroscopic, the mesoscopic, and the microscopic level. At the macroscopic level (see models presented above), the overall population characteristics and dynamics are studied. Macroscopic predictive models describe growth and inactivation behaviors of populations. For process control, monitoring, and optimization purposes, macroscopic models are preferred because they have a rather simple structure, i.e., a limited number of model components and parameters. In the mesoscopic level studies, small populations, e.g., subpopulations within a population or colonies in structured environments, are the target. Due to environmental or population heterogeneity, differences in the microbial response are observed and all cells—or their dynamics—can no longer be assumed to

SMP_Food Microbiology_CH40.indd

1007

be identical. Examples of more mesoscopic models can be found in the work of McKellar (117) and Skandamis et al. (173). To completely unravel mechanisms underlying the specific microbial response to, e.g., stressing environments or environmental gradients, information is collected at a microscopic level, i.e., a cellular or even an intracellular level. In the last decade, for instance, many researchers focused on the intracellular stress response, e.g., the heat shock response of E. coli (10, 39, 202). It can be expected that with the level of detail included within the model, the complexity of the model structure and the number of related model parameters will increase exponentially. Mechanistically inspired studies have been conducted with respect to lag phase dynamics. Mainly, the focus is on the individual cell dynamics and how they relate to the overall population behavior (19, 116). Stochastic modeling approaches and IbM techniques have been applied to describe these dynamics. When microbial populations are exposed to severe stress (e.g., heat and acid), sigmoidal growth curve patterns are often disturbed. These unexpected growth curves can be attributed to microbial population heterogeneity induced by the environmental conditions. Attempts have been made to include population heterogeneity in the modeling of microbial kinetics (135, 188). A description of the intracellular dynamics is possible using systems biology (see “Systems Biology—Metabolic Networks” below). In this discipline, the cell is regarded as a system and the overall cell function is determined by the interactions between the different components of the system. When single cells are the center of attention, individual-based modeling (IbM) or agent-based modeling approaches are appropriate (see “Individual-Based Modeling” below). Heterogeneous or subpopulationtype models, presented in “Systems Biology—Metabolic Networks,” use different compartments to describe the dynamics of cells, which can be assumed to have the same growth properties (e.g., subpopulations).

Systems Biology—Metabolic Networks

The intrinsic complexity of biochemical processes, which consist of extensive reaction networks with numerous metabolites, is not reflected by the simple macroscopic level models, classically used in predictive microbiology. However, while more knowledge about the underlying mechanisms of biochemical processes becomes available, new opportunities arise, for instance by using microscopic level metabolic network models to build next-generation predictive models. Systems biology can contribute to the mechanistic character of ­predictive models, either using a top-down approach or a bottom-up approach.

Manila Typesetting Company

11/08/2012 07:21AM

1008

Advanced Techniques in Food Microbiology

The bottom-up approach starts with collecting information at the lowest level considered, i.e., the intracellular environment. From this information, specific trends are extracted and new relations and models can be built. In the top-down approach, the starting point is the available knowledge. New information gathered at the (intra)cellular level is in this case used to complement the existing knowledge. In the case of predictive microbiology, the information obtained from the systems biology approach will lead to mechanistically inspired macroscopic models with improved validity under more realistic, often stressing, conditions. Metabolic flux analysis, which starts from a metabolic network describing the cellular dynamics, stands for an excellent tool to gain in-depth insight (i.e., at the intracellular level) on the impact of process conditions on (the fluxes in) the cell metabolism and growth dynamics. Relevant (extracellular) process conditions and key metabolic reactions/ pathways can be identified, which is valuable information in the development of predictive models for more complex and realistic situations. Exploitation of metabolic flux analysis as a technique to develop accurate mathematical models in the field of predictive microbiology is a largely unexplored domain.

The generic relation between the macroscopic reaction rates is described as follows:

The basics of metabolic networks The metabolic network-based modeling approach will ultimately link microscopic level information, enclosed in the metabolic network, with macroscopic level models, which—in the end—are mechanistically inspired but still rather simple to use in practice. Macroscopic level.  At the macroscopic level, the most important variables defining the microbial dynamics are the population density, N (CFU/ml), the concentration of limiting substrate(s), CS , and the concentration of metabolic product(s), CP , with m, s, and p the specific rates for cell growth, substrate consumption, and product formation, respectively. The effect of intrinsic and/or extrinsic conditions [indicated by “(×)” in the equations below], like temperature and aw, on the microbial behavior is included in the specific rates using specific kinetic models like the cardinal temperature model with inflection (160) or Monod-type relations. dN = µ () × ×N dt dCS = -σ () × × CS dt dCP = π () × × CP dt

SMP_Food Microbiology_CH40.indd

1008

σ () × =

µ () × π () × + +m YXS YPS

where YXS and YPS are the yield coefficients of biomass on substrate and product on substrate, respectively, and m is the maintenance factor.

Microscopic level.  The intracellular dynamics can be characterized by the intracellular metabolites concen­ trations, c, the intracellular reaction fluxes, i.e., the cellspecific reaction rates, v, and the stoichiometry matrix, S. dc = S×v - µ ×c dt The stoichiometric matrix contains in the rows the network components, while the columns show how these components interact. Each value in the matrix describes a specific reaction with a negative value relating to reaction substrates, while a positive value is related to reaction products. This is illustrated in Fig. 40.1 for the minimal network describing the relation between three components Ma, Mb, and Mc. Multiplying the stoichiometric matrix, S, with the intracellular fluxes, v, renders the series of linear relations describing the metabolic network (Fig. 40.1).

Linking the microscopic and macroscopic levels.  The coupling between the macroscopic and microscopic levels is made at the level of the specific reaction rates: (i) the macroscopic level rates m, s, and p depend on the intracellular metabolites concentrations, c, and (ii) the intracellular metabolic fluxes, v (i.e., the cell-specific reaction rates), depend on the extracellular concentrations such as CS and CP, physicochemical environmental conditions such as pH, temperature, and aw, and the physiological state(s) of the cells.

Metabolic network model reduction Metabolic networks are a blueprint of all the reactions that occur inside a microbial cell during the biochemical process. As a result of the complexity of the microscopic level models, problems arise due mainly to insufficient calculation power and a shortage of intracellular experimental data. Especially the latter limitation is the major challenge when solving the metabolic network. Metabolic networks generally tend to be underdetermined, i.e., the number of unknown variables/fluxes in

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1009

Metabolic network

Stoichiometric modeling

ν4

ν1

Ma

Constraint-based model

Mb

ν2

ν5

General equaon

dM i = S ·ν − m · x dt

ν3 Mc

Steady-state

S ·v =0

Mass balance

ν6

Other constraints

Stoichiometric matrix

−1 S= 1 0

1 0

0 1

−1 −1

−1 0 0 −1 0

0

0 0 1

e.g.

νi ≥0

(irreversibility, capability, measured fluxes)

Figure 40.1  Schematic representation of the stoichiometric modeling framework (112). doi:10.1128/9781555818463.ch40f1

the network n is significantly smaller than the measured components m. For application of metabolic network models in the development of predictive models, these models first of all have to be simplified.

Steady-state hypothesis.  A first simplification is obtained by imposing the steady-state constraint yielding a subspace of possible solutions. Under the assumptions that (i) pseudosteady state holds for the intracellular metabolites, c, i.e., globally no mass is consumed or formed, and (ii) the dilution term (−mc) can be neglected compared to the fluxes affecting the same metabolite, the microscopic level mass balances reduce to the so-called general equation (for an example, see reference 112): S · n = 0 The steady-state hypothesis assumes that intracellular reactions proceed much faster than macroscopic population cell dynamics. This general equation only restricts flux distributions that can be achieved by a specific metabolic network to feasible flux distributions, as the general equation is underdetermined (n fluxes are involved in m metabolites, with n logically larger than m). Multiple steady-state conditions can be feasible. Further model reduction can be obtained based on (i) knowledge (e.g., flux limits, measured fluxes, and kinetic relations); and (ii) mathematical techniques, for example, balanced truncation (88, 110), intrinsic low-dimensional manifolds (206), and singular perturbation (187, 203).

FBA.  Even though the metabolic network model has been reduced in complexity by the steady-state and

SMP_Food Microbiology_CH40.indd

1009

knowledge-based assumptions, most often the abovementioned general equation remains underdetermined. A large collection of possible solutions still exists. Flux balance analysis (FBA) is an excellent tool for optimization to find an acceptable solution within the possible steady-state conditions. Within this FBA approach, an optimal solution for the network can be found by optimizing a specific objective function (see, among others, references 36, 92, and 167). Commonly implemented objective functions are (i) maximization of the biomass (29, 171), (ii) maximization of ATP (151), (iii) product maximization (192), and (iv) maximization of the growth rate (42, 165, 205). Following this FBA method, the natural regulations and dynamics of the pathways in the metabolic network are not considered, i.e., the fluxes within the network are predicted while optimizing the objective function. This means that the main assumption of FBA, i.e., a cell has evolved to achieve an optimal behavior, has an important drawback: the optimal solution may not correspond to the actual flux distribution.

Linking the metabolic network with a macroscopic model In the next step, the optimization problem is hierarchically decomposed in two levels: the optimal behavior of a population (CX) and the optimal cell behavior (c). A well-posed optimization problem consists of three ingredients: (i) the mathematical model, (ii) the set of physical, biochemical and practical constraints, and (iii) the performance criterion (cost index), which is a mathematical translation of the objective to be optimized (e.g., maximize the specific growth rate).

Manila Typesetting Company

11/08/2012 07:21AM

1010

Advanced Techniques in Food Microbiology

For the implementation of systems biology in the domain or predictive microbiology, the major future challenge is to harmonize the objective function used at the macroscopic (population) level with the objective function used at the microscopic (cell) level. At the population level, additional factors will play a role in growth dynamics compared to the cell level. Those factors include inter- and intraspecies interactions, food (model system) structure, and biological variability. As a result, the population objective might differ from the cell objective.

rules constituting the model reflect the behavior of the individual cells, such as nutrient consumption, biomass growth, cell division, movement, differentiation, communication, maintenance, and death. These rules are based on theory or prior knowledge. Often, these rules are referred to as hypotheses. Frequently, phenomena that are not evident or self-contained in the input rules of the model are observed. These phenomena are called emergent behaviors (81). By simulating a large number of such cell units as a function of time, the population dynamics can be derived indirectly from the integration of individual cell characteristics. The underlying IbM principles together with the ease of gradual introduction of more complexity make IbM the perfect framework for (i) analysis and study of systems dynamics (e.g., individual and population behavior or emergent behavior); (ii) hypothesis testing; (iii) simulation experiments; (iv) parameter estimation; and (v) experiment design. All functionalities of IbM lead to a more detailed understanding of the individual’s and the population’s behavior. Gradual introduction of more complexity at the individual level makes it easier to analyze the model and investigate thoroughly the effect of individual aspects added to the model. Furthermore, IbM allows introduction of behavior at the level at which it originates, i.e., the individual cell level. This is the case for, e.g., individual variability and spatial effects. Individual variability is incorporated by using random variables drawn from an experimentally derived statistical distribution, as illustrated in references 63 and 130. The introduction of a range of randomness and the consideration of a high number of individuals interacting independently with the environment lead to an adequate representation of reality and a better understanding of cellular metabolism (25). Even spatial effects can be relatively easily incorporated, thereby taking local interactions into account. Kreft et al. (102) reproduced the growth of E. coli cells in a colony by introducing spatial effects in their IbM BacSim. Analysis of systems dynamics will result in a more detailed understanding of the individual’s behavior. Advances in single-cell research have led to more mechanistic knowledge about individual behavior. The emerging field of “-omics” research, i.e., metabolomics, proteomics, and genomics, delivers a huge amount of information concerning different components of the cell. This knowledge can be incorporated in IbMs as hypotheses (expressed in a set of different mathematical equations and/or rules). This way, the possible link between systems biology and IbM is obvious, although one needs to bear in mind that -omics experiments are holistic, i.e., analyze total biomass without differentiat-

Individual-Based Modeling

Whereas the classical macroscopic modeling approach is proven valid for many conditions, the assumption that all cells within a population are identical is too simplified when studying specific microbial phenomena. For instance, it is widely proven that genetically identical cells within a population show a significant variety in the time before their first cell division after exposure to (moderate) stressing conditions. Under more extreme stressing conditions, differences in cell responses can even range from inactivation to survival to growth. Moreover, differences in single cell responses can also be due to spatial differences within the food (model systems), e.g., local nutrient gradients. IbM provides an excellent framework to model such complex microbial dynamics. By considering the individual cell as the basic modeling unit, it is more natural to include (i) mechanistic knowledge of the behavior and interactions of the microbial cell, (ii) individual biological variability, and (iii) interactions between the cells and their environment. Thus, IbM allows considering aspects usually ignored in macroscopic models, i.e., intercell heterogeneity, interactions with the external environment, and interactions among individuals. Furthermore, the individual-based approach can provide a bridge between knowledge at smaller scales and observed population dynamics. IbM is therefore an essential tool for microbiology (62).

Principles and functionalities of IbMs Population dynamics are determined by the behavior of the individuals in the population. It is therefore appropriate to model microbial dynamics in terms of individual cells (102). IbMs are simulation models that treat individuals as unique and discrete entities that have at least one property in addition to age that changes during the life cycle (81). Nevertheless, all individuals are of the same type (e.g., bacteria) and have the same potential regarding state and behavior. Using simple rules or models, the cell’s behavior and its interactions (among cells and with their environment) are characterized. The

SMP_Food Microbiology_CH40.indd

1010

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1011

ing between individual cells. Practically, a change in hypotheses concerning, e.g., microbial behavior, may lead to significantly different macroscopic (population) behavior. This can lead to discrimination between hypotheses and thereby elimination of impossible mechanisms and investigation of true mechanisms (for an example, see reference 52). Since obtaining values of individual parameters can be difficult, or even unfeasible, estimating these parameters from population dynamics can provide the answer. Population dynamics data, especially regarding microorganisms, are abundantly available. Also, some experimental studies in microbiology are difficult to perform and analyze, and often they are simply unfeasible. In all these cases, IbM simulations can contribute to experimental research by means of virtual experiments (142).

IbM and (predictive) microbiology To date, IbM research in predictive food microbiology has been dominated by the application of BacSim (102) and individual discrete simulations (INDISIM) (77). Kreft et al. (102) developed the BacSim IbM starting from the IbM environment Gecko and used this tool initially to evaluate how a single E. coli cell evolves into a colony. Later on, Kreft and coworkers adapted the BacSim software to describe the process of biofilm formation (103) and the dynamics (growth and migration) of Salmonella Enteritidis in eggs (80). In 2004, the BacSim environment was also applied to describe the microbial stationary phase in terms of substrate consumption and metabolite production (175). Later on, Dens et al. used BacSim to study the relation between the lag phase, induced by changes in the medium and/or the temperature, and the microbial cell division ­mechanism (52, 53). INDISIM has been designed to study the growth and behavior of bacterial colonies, starting from single cells (77). The INDISIM environment was used to study biomass distributions and the relation between the colony growth rate and the local nutrient concentration and temperature. In 2005, INDISIM was extended to INDISIM-SOM to study the microbial activity of two strains (ammonifier microorganisms and nitrifier bacteria) in soil, and more specifically, the effect on mineralization and immobilization of carbon and nitrogen and the nitrification process (76). In 2006, the INDISIM tool was applied to study the microscopic causes of the lag phase, i.e., to evaluate which cellular properties contribute the most to the lag phenomenon (147, 148). In a next step, this research group developed INDISIM-YEAST, which focuses on the cell division and metabolism of yeast strains taking into account the substrate uptake,

SMP_Food Microbiology_CH40.indd

1011

metabolism, budding reproduction, and typical viability properties of yeast strains in a homogeneous liquid environment (75). Later on, this environment was used to study the lag phase and growth initiation of yeast and see how cell age and inoculum size affects industrial fermentations like brewing (78). IbM models have also been applied to study microbial dynamics under specific circumstances, e.g., in the presence of specific antimicrobials (133) and biological wastewater treatment (200). Although IbM has proven to be a worthy modeling tool for microbial dynamics, implementation is limited because IbMs are more complex than classical macroscopic models, which makes them difficult to understand and analyze. Practically, IbMs should be regarded as complementary to traditional modeling techniques. Knowledge acquired by IbM is a good starting point for the development of new macroscopic models or the improvement/extension of existing models, finally leading to more mechanistic predictive models.

Compartment-Based Models

Cell-to-cell variability within a genetically homogeneous microbial population is a widely observed phenomenon that often results in improved survival under sublethal stresses, e.g., antibiotics, acid stress, and osmotic stress. Basically, cells can respond to stress in three ways: (i) survive (cell number remains constant or decreases slowly), (ii) grow (cell number increases), or (iii) inactivate (cell number decreases rapidly). Compartment- or subpopulation-based models are based mainly on this concept. The co-occurrence of two subpopulations was claimed to explain the tailing in inactivation curves with respect to, e.g., antimicrobial agents (55), osmotic stress (50), and pressure (137). After an initial shoulder period, during which the bacteria are possibly affected but not yet killed, inactivation starts. The largest fraction of the inoculated cells are severely damaged and inactivate rather fast. A smaller proportion of cells, however, are more resistant and are not inactivated, or are inactivated at a much lower rate. Tailing in inactivation curves results from the survival of the latter, more stress-­resistant subpopulation. Based on the assumption of the two coexisting subpopulations, Whiting (199) developed a logistics-based inactivation model that could describe inactivation curves with shoulder and tailing phenomena. Analogous heterogeneous modeling approaches were applied by others, such as Fujikawa and Itoh (64) and Geeraerd et al. (70). McKellar (117) used a two-compartment model to describe the initial microbial lag phase and ­subsequent

Manila Typesetting Company

11/08/2012 07:21AM

1012

Advanced Techniques in Food Microbiology

growth of L. monocytogenes. The model is based on the assumption that at the beginning the inoculum consists of two compartments, i.e., a growing and a nongrowing compartment. A parameter that defines the proportion of initially growing cells was included. A Gompertz model was used to describe growth of the growing compartment. The number of nongrowing cells remained constant throughout the whole experiment, which implies that growth is determined only by the growing fraction. Nikolaou and Tam (135) presented a new model that describes the heterogeneous response to antimicrobials. The following equations were used to describe a population in which some cells can resist the stress and grow while others fail to overcome the stress and die.

FROM ACADEMIA TO INDUSTRY: SOFTWARE TOOLS

dN = K g N (t ) - r éëC (t )ùû N (t ) ������������� ��������� dt physiological growth rate

kill rate due to agent

The left term of the equation describes the growth of the surviving cells via the growth rate, Kg. The right term represents the inactivating cells via the kill rate r[C(t)], which is determined by the strain and antimicrobial selected. The combination of these two terms, i.e., subpopulations, enables to predict the microbial response to antimicrobial agents. Van Derlinden et al. (188, 189) observed that growth of E. coli K-12 in brain heart infusion broth at superoptimal temperatures (45°C and 46.5°C), especially close to the maximum growth temperature, is disturbed, i.e., a short growth phase is followed by an inactivation phase and a second growth phase, after which the stationary phase was reached. It was hypothesized that microbial cells can evolve in two different ways: (i) some cells are more resistant to the temperature stress and continue multiplication, and (ii) some cells are inactivated in response to the high temperature. A two-subpopulation model, i.e., considering a growing stress-resistant subpopulation and an inactivating stress-sensitive subpopulation, was developed to describe the disturbed growth kinetics. The heterogeneous model combines the primary model of Baranyi and Roberts, which describes growth, with a simple log-linear inactivation model. The effect of temperature on the maximum specific growth and inactivation rate is modeled with the cardinal temperature model with inflection (160) and the Bigelow model, respectively. Mathematical simulations with this model were able to describe the disturbance of the exponential phase rather accurately.

SMP_Food Microbiology_CH40.indd

1012

According to Membré and Lambert (127), current applications of predictive microbiology as a tool to control and improve food safety and food quality can be situated within three domains: product innovation, operational support, and incident support. 1. Product innovation: predictive models show (the tendency) of the effect of changes in the product formulation with respect to the outgrowth of food pathogens and food spoilage microorganisms. 2. Operational support: for instance, in combination with HACCP, predictive models can render quantitative information at the different critical control points. 3. Incident support: predictive models can quantify the effect of product abuse, e.g., during storage in the supermarket, on the consumer. Generally, predictive modeling partly replaces challenge tests, i.e., model simulations can replace the trialand-error approach currently applied. Although these advantages are straightforward, implementation, particularly in small food industry firms, is limited due possibly to (i) limited research and development divisions and/or expenditures, (ii) insufficient mathematical background, and (iii) lack of microbial knowledge that guarantees a correct interpretation of simulation results. In recent years, a series of companies like Nestlé, Unilever, and Purac have developed tailor-made models focusing on specific classes of food product(s). However, a general transfer of predictive models from science to industry has not happened yet. In the last 2 decades, food research has acknowledged this limited implementation of predictive models in industry. As a consequence, many research groups or consortia have worked on the development of software tools focusing on applications in the food industry. Different approaches have been taken: freely accessible versus purchased tools, general approach versus product specificity, etc. Below, recently developed software packages are listed.

ComBase

ComBase (www.combase.cc) is a free online database that combines over 40,000 data sets on growth, survival, or inactivation of microorganisms in food model systems collected from literature or provided by research institutions. The ComBase modeling tool is an online freeware that can be used to evaluate the growth, inactivation, and survival of pathogens and, to a lesser extent, spoilage organisms (23). The major-

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1013

ity of models describe the growth and/or survival of pathogens as a function of temperature (constant or dynamic), pH, and aw/salt concentration. In total, 12 pathogens can be evaluated. For some organisms, the effect of CO2, nitrite, etc., is also documented. The predictive modeling toolbox includes three different modeling tools each yielding graphical and numerical presentation of the simulated response. •

•

•

ComBase Predictor includes multiple growth models and thermal death models for predicting the response of foodborne pathogenic and spoilage microorganisms in response to intrinsic/extrinsic environmental factors. Perfringens Predictor yields a prediction of Clostridium perfringens growth during the cooling of meats. Values for the meat pH and the salt concentration can be specified by the user. For the Perfringens Predictor, an Excel add-in version can be downloaded. DMFit enables modeling of microbial growth and inactivation curves where a linear phase is preceded and followed by a stationary phase. This will yield values (and corresponding estimation errors) for the lag/ shoulder time, growth/inactivation rate, and the initial and final cell counts.

Sym’Previus

Sym’Previus (www.symprevius.org) is a French initiative that unites several French research units. The Sym’Previus package, which can be purchased online, includes the following: (i) a database with growth and inactivation responses of microorganisms in foods and (ii) predictive models for growth and inactivation of pathogenic bacteria and some spoilage microorganisms. The database is a combination of data acquired from publications, industry, and research facilities. The Sym’Previus website also lists publications relating to the modeling of the dynamics of food pathogens and food spoilage microorganisms. More information about the (startup) of the software tool and models incorporated can be found in Leporq et al. (107).

Seafood Safety and Spoilage Predictor

Seafood Safety and Spoilage Predictor (SSSP) can be downloaded without charge from http://sssp.dtuaqua. dk/. The SSSP software can be used to describe or evaluate the dynamics of spoilage bacteria (mainly lactic acid bacteria) and/or pathogens (mainly L. monocytogenes). Over the years, the models for L. monocytogenes have been updated and adapted, resulting in an extended model that includes a high number of intrinsic and extrinsic factors: temperature, aw/NaCl, pH,

SMP_Food Microbiology_CH40.indd

1013

organic acids (lactic acid, acetic acid/diacetate, sorbic acid, citric acid, and benzoic acid), nitrite, phenol, and CO2. The simultaneous dynamics of the spoilage organisms (lactic acid bacteria) and possible effect on the Listeria response is also monitored. For more information about the kinetic model, see Mejlholm and Dalgaard (124). Not only does the most recent version contain a kinetic model, but also the G/NG boundary of Listeria can be simulated. Originally, as the name indicates, the tool was developed for seafood. More recently, the models have also been successfully validated for meat, poultry, and nonfermented dairy products (125).

Pathogen Modeling Program

Pathogen Modeling Program (PMP) (http://portal. arserrc.gov/) is a free software package that includes mainly models built on the basis of experimental data from liquid microbiological media and food products to describe the effects of multiple variables on the growth, inactivation, or survival of foodborne pathogens. Most models included in PMP are growth models, which take into account the effects of temperature, aw, pH, atmosphere, and occasionally additives like nitrite. Inactivation is modeled as a function of temperature, aw/NaCl, pH, nitrite, and lactic acid. Also included in the PMP software are cooling models, which quantify the effect of the selected cooling profile on the growth of Clostridium botulinum and C. perfringens. The PMP was one of the first predictive microbiology software packages and is among the most widely used.

GInaFiT

GInaFiT is a freeware add-in for Microsoft Excel and brings together a series of primary predictive models that can describe different shapes of inactivation curves (72). In total, nine different models are applied for the description of nine shapes of survival curves: (i) classical log-linear curves, (ii) curves displaying a shoulder before the log-linear decrease, (iii) curves displaying a tail after a log-linear decrease, (iv) survival curves displaying both a shoulder and a tail, (v) concave curves, (vi) convex curves, (vii) convex/concave curves followed by tailing, (viii) biphasic inactivation kinetics, and (ix) biphasic inactivation kinetics preceded by a shoulder. In addition to the obtained parameter values, the standard errors of the estimated parameter values, the sum of squared errors, the mean sum of squared errors and its root, the R2, and the adjusted R2 are given. GInaFiT can be downloaded from the KULeuven/ BioTeC homepage (http://cit.kuleuven.be/biotec/).

Manila Typesetting Company

11/08/2012 07:21AM

1014

Advanced Techniques in Food Microbiology

Other Software Tools

E. coli Fermented Meat Model

The software packages above are the most widely ­implemented within the domain of predictive microbiology. Below, some additional, smaller software tools, most often built for a more specific case study, are presented.

Opti-Form Listeria Control Model

Opti-Form Listeria Control Model (www.purac.com) contains a model that predicts growth of L. monocytogenes in/on both uncured and cured cooked meat products. Specifically, this software tool shows the efficiency of the preservatives developed by Purac for the control of L. monocytogenes in combination with other intrinsic (pH and moisture) and extrinsic (temperature) conditions and enables optimization of their concentrations in relation to the food product characteristics. The software can be requested from the Purac Company.

DMRI

The Danish Meat Research Institute (DMRI) built an online software tool that can describe the outgrowth of Listeria in meat products as a function of temperature, pH, NaCl, sodium lactate, sodium acetate, sodium nitrite, and CO2. The model behind this software is an artificial neural network and was built from data obtained from experiments performed on solid surfaces and ready-to-eat meat products, both with various intrinsic and extrinsic conditions (83). This software is available to companies that are members of the Danish Meat Association.

Refrigeration Index Calculator

With the Refrigeration index calculator, developed by Meat & Livestock Australia Limited (http://www. foodsafetycentre.com.au/refrigerationindex.php), the user can see the (qualitative) effect of the temperature on the growth of E. coli in different types of meats, e.g., during cooling of these products (156). In addition to temperature, the predictive model takes into account pH, aw, and lactate concentration.

The E. coli fermented meat model is an Excel add-in that can be used to evaluate the killing of E. coli during meat fermentation or in fermented meat products (http://www.foodsafetycentre.com.au/fermenter.php). The models were constructed based on experimental data obtained from real meat products or systems with similar conditions, i.e., low aw or pH, or both.

Fish Shelf Life Prediction Program

Fish Shelf Life Prediction Program (FSLP) (www.azti.es) is a software tool that can be used to evaluate the shelf life of an aquaculture product (farmed fresh turbot) as a function of the storage temperature. In ­addition to the evolution of the spoilage organism, this package also shows the changes in the sensory quality (food quality).

Shelf Stability Predictor

The Shelf Stability Predictor was developed by the Center for Meat Process Validation of the University of Wisconsin (http://meathaccp.wisc.edu/ST_calc.html). The software includes models that can be used to predict the outgrowth of L. monocytogenes and S. aureus in ready-to-eat meat products. In relation to the given pH and aw of the food (model) system, the software tool gives the growth probability and indicates whether this corresponds to a safe or an unsafe food product.

Websim-MILQ

The Websim-MILQ Web tool was developed within a European Union research project (MILQ-QC-TOOL). The software can be applied to optimize thermal processes in the dairy industry. More information can be found in reference 168 and on the website of the NIZO Food Research Institute (www.nizo.com).

References

Risk Ranger

The Risk Ranger software is also an Australian tool and can be downloaded from http://www.foodsafetycentre. com.au/riskranger.php. This software focuses on the risk related to the presence of different pathogens for different food products, each with their own intrinsic properties. It can help the user take the correct measures to reduce and/or prevent foodborne illnesses. More information can be found in reference 157.

SMP_Food Microbiology_CH40.indd

1014





1. Abou-Zeid, K. A., T. P. Oscar, J. G. Schwarz, F. M. Hashem, R. C. Whiting, and K. Yoon. 2009. Development and validation of a predictive model for Listeria monocytogenes Scott A as a function of temperature, pH, and commercial mixture of potassium lactate and sodium diacetate. J. Microbiol. Biotechnol. 19:718–726. 2. Adams, M. R., C. L. Little, and M. C. Easter. 1991. Modeling the effect of pH, acidulant and temperature on the growth rate of Yersinia enterocolitica. J. Appl. Bacteriol. 71:65–71. 3. Adekunte, A., V. P. Valdramidis, B. K. Tiwari, N. Slone, P. J. Cullen, C. P. O’Donnell, and A. Scanell. 2010. Resistance of Cronobacter sakazakii in reconstituted powdered infant formula during ultrasound

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology



4.



5.



6.



7.



8.



9.

10. 11. 12.

13.

14. 15.

16.

17.

1015

at controlled temperatures: a quantitative approach on microbial responses. Int. J. Food Microbiol. 142:53–59. Agresti, A. 2002. Categorical Data Analysis. John Wiley & Sons, New York, NY. Alber, S. A., and D. W. Schaffner. 1992. Evaluation of data transformations used with the square root and Schoolfield models for predicting bacterial growth rate. Appl. Environ. Microbiol. 58:3337–3342. Albert, I., and P. Mafart. 2005. A modified Weibull model for bacterial inactivation. Int. J. Food Microbiol. 100:197–211. Antwi, M., A. H. Geeraerd, K. M. Vereecken, R. Jenné, K. Bernaerts, and J. F. Van Impe. 2006. Influence of a gel microstructure as modified by gelatin concentration on Listeria innocua growth. Innov. Food Sci. Emerg. Technol. 7:124–131. Antwi, M., K. Bernaerts, J. F. Van Impe, and A. H. Geeraerd. 2007. Modelling the combined effects of structured food model system and lactic acid on Listeria innocua and Lactococcus lactis growth in mono- and coculture. Int. J. Food Microbiol. 120:71–84. Arroyo López, F. N., M. C. Durán Quintana, and A. Garrido Fernández. 2007. Modelling of the growth–no growth interface of Issatchenkia occidentalis, an olive spoiling yeast, as a function of the culture media, NaCl, citric and sorbic acid concentrations: study of its inactivation in the no growth region. Int. J. Food Microbiol. 117:150–159. Arsène, F., T. Tomoyasu, and B. Bukaua. 2000. The heat shock response of Escherichia coli. Int. J. Food Microbiol. 55:3–9. Augustin, J. C., and V. Carlier. 2000. Mathematical modelling of the growth rate and lag time for Listeria monocytogenes. Int. J. Food Microbiol. 56:29–51. Augustin, J. C., and V. Carlier. 2000. Modelling the growth rate of Listeria monocytogenes with a multiplicative type model including interactions between environmental factors. Int. J. Food. Microbiol. 56:53–70. Augustin, J. C., V. Zuliani, M. Cornu, and L. Guillier. 2005. Growth rate and growth probability of Listeria monocytogenes in dairy, meat and seafood products in suboptimal conditions. J. Appl. Microbiol. 99:1019–1042. Babbar, S. B., and R. Jain. 2006. Xanthan gum: an economical partial substitute for agar in microbial culture media. Curr. Microbiol. 52:287–292. Bajard, S., L. Rosso, G. Fardel, and J. P. Flandrois. 1996. The particular behaviour of Listeria monocytogenes under sub-optimal conditions. Int. J. Food Microbiol. 29:201–211. Baker, D. A., and C. Genigeorgis. 1990. Predicting the safe storage of fresh fish under modified atmospheres with respect to Clostridium botulinum toxigenicity by modeling length of the lag phase of growth. J. Food Prot. 53:131–140. Bang, W. S., H. J. Chung, S. S. Jin, T. Ding, I. G. Hwang, G. J. Woo, S. D. Ha, G. J. Bahk, and D. H. Oh. 2008. Prediction of Listeria monocytogenes growth kinetics in sausages formulated with antimicrobials as a

SMP_Food Microbiology_CH40.indd

1015

18. 19. 20. 21. 22.

23. 24. 25.

26.

27.

28.

29. 30.

31. 32.

33.

function of temperature and concentrations. Food Sci. Biotechnol. 17:1316–1321. Baranyi, J. 1998. Comparison of stochastic and deterministic concepts of bacterial lag. J. Theor. Biol. 192:403–408. Baranyi, J., S. M. George, and Z. Kutalik. 2009. Parameter estimation for the distribution of single cell lag times. J. Theor. Biol. 259:24–30. Baranyi, J., and T. A. Roberts. 1994. A dynamic approach to predicting bacterial growth in food. Int. J. Food Microbiol. 23:277–294. Baranyi, J., and T. A. Roberts. 1995. Mathematics of predictive food microbiology. Int. J. Food Microbiol. 26:199–218. Baranyi, J., T. Ross, T. A. McMeekin, and T. A. Roberts. 1996. Effects of parametrization on the performance of empirical models used in ‘predictive microbiology’. Food Microbiol. 13:83–91. Baranyi, J., and M. Tamplin. 2004. ComBase: a common database on microbial responses to food environments. J. Food Prot. 67:1967–1971. Basheer, I. A., and M. Hajmeer. 2000. Artificial neural networks: fundamentals, computing, design, and application. J. Microbiol. Methods 43:3–31. Bermúdez, J., D. López, J. Valls, and J. Wagensberg. 1989. On the analysis of microbiological processes by Monte Carlo simulation techniques. Comput. Appl. Biosci. 5:305–312. Bernaerts, K., E. Dens, K. Vereecken, A. H. Geeraerd, A. R. Standaert, F. Devlieghere, J. Debevere, and J. F. Van Impe. 2004. Concepts and tools for predictive modeling of microbial dynamics. J. Food Prot. 67:2041–2052. Bidlas, E., and R. J. W. Lambert. 2008. Quantification of hurdles: predicting the combination of effects—interaction vs. non-interaction. Int. J. Food Microbiol. 128:78–88. Bover-Cid, S., N. Belletti, M. Garriga, and T. Aymerich. 2011. Model for Listeria monocytogenes inactivation on dry-cured ham by high hydrostatic pressure processing. Food Microbiol. 28:804–809. Boyle, N. R., and J. A. Morgan. 2009. Flux balance analysis of primary metabolism in Chlamydomonas reinhardtii. BMC Syst. Biol. 3:4. Braun, P., and J. P. Sutherland. 2004. Predictive modelling of growth and enzymatic synthesis and activity by a cocktail of Yarrowia lipolytica, Zygosaccharomyces bailii and Pichia anomala. Food Microbiol. 21:459–467. Brocklehurst, T. F., G. A. Mitchell, and A. C. Smith. 1997. A model experimental surface for the growth of bacteria on foods. Food Microbiol. 14:303–311. Brul, S., F. I. C. Mensonides, K. J. Hellingwerf, and M. J. Teixeira de Mattos. 2008. Microbial systems biol­ ogy: new frontiers open to predictive microbiology. Int. J. Food Microbiol. 128:16–21. Buchanan, R. L., R. C. Whiting, and W. C. Damert. 1997. When is simple good enough: a comparison of the Gompertz, Baranyi, and three-phase linear models for fitting bacterial growth curves. Food Microbiol. 14:313–326.

Manila Typesetting Company

11/08/2012 07:21AM

1016

Advanced Techniques in Food Microbiology

34. Buchanan, R. L., L. K. Bagi, R. V. Goins, and J. G. Philips. 1993. Response surface models for the growth kinetics of Escherichia coli O157:H7. Food Microbiol. 10:303–315. 35. Buchanan, R. L., and J. G. Philips. 2000. Updated models for the effects of temperature, initial pH, NaCl, and NaNO2 on the aerobic and anaerobic growth of Listeria monocytogenes. Quant. Microbiol. 2:103–128. 36. Burgard, A., and C. Maranas. 2002. Optimizationbased framework for inferring and testing hypothesized metabolic objective functions. Biotechnol. Bioeng. 82:670–677. 37. Cerf, O., L. R. Davey, and A. K. Sadoudi. 1996. Thermal inactivation of bacteria—a new predictive model for the combined effect of three environmental factors: temperature, pH and water activity. Food Res. Int. 29:219–226. 38. Chorin, E., D. Thuault, J. J. Cléret, and C. M. Bourgeois. 1997. Modelling Bacillus cereus growth. Int. J. Food Microbiol. 38:229–334. 39. Chung, H., W. Bang, and M. Drake. 2006. Stress response of Escherichia coli. Compr. Rev. Food Sci. F 5:52–64. 40. Cole, M. B., M. V. Jones, and C. Holyoak. 1990. The effect of pH, salt concentration and temperature on the survival and growth of Listeria monocytogenes. J. Appl. Bacteriol. 69:63–72. 41. Coroller, L., V. Guerrot, V. Huchet, Y. Le Marc, P. Mafart, D. Sohier, and D. Thuault. 2005. Modelling the influence of single acid and mixture on bacterial growth. Int. J. Food Microbiol. 100:167–178. 42. Covert, M., C. Schilling, and B. Palsson. 2001. Regulation of gene expression in flux balance models of metabolism. J. Theor. Biol. 213:309–325. 43. Cuppers, H. G. A. M., S. Oomes, and S. Brul. 1997. A model for the combined effects of temperature and salt concentration on growth rate of food spoilage molds. Appl. Environ. Microbiol. 63:3764–3769. 44. Dalgaard, P. 1995. Modelling of microbial activity and prediction of shelf life for packed fresh fish. Int. J. Food Microbiol. 26:305–317. 45. Dalgaard, P., O. Mejlholm, and H. H. Huss. 1997. Application of an iterative approach for development of a microbial model predicting the shelf-life of packed fish. Int. J. Food Microbiol. 38:169–179. 46. Dantigny, P., and M. Bensoussan. 2008. The logarithmic transformation should be avoided for stabilising the variance of mould growth rate. Int. J. Food Microbiol. 121:225–228. 47. Daughtry, G. J., K. R. Davey, and K. D. King. 1997. Temperature dependence of growth kinetics of food bacteria. Food Microbiol. 14:21–30. 48. Davey, K. R. 1993. Extension of the generalized chart for combined temperature and pH. LWT Food Sci. Technol. 26:476–479. 49. Delignette-Muller, M. L., M. Cornu, R. Pouillot, and J. B. Denis. 2006. Use of Bayesian modelling in risk assessment: application to growth of Listeria monocytogenes and food flora in cold-smoked salmon. Int. J. Food Microbiol. 106:195–208.

50. den Besten, H. M. W., C. J. Ingham, J. E. T. van Hylckama Vlieg, M. M. Beerthuyzen, M. H. Zwietering, and T. Abee. 2007. Quantitative analysis of population heterogeneity of the adaptive salt stress response and growth capacity of Bacillus cereus ATCC 14579. Appl. Environ. Microbiol. 73:4797–4804. 51. Dens, E., and J. Van Impe. 2001. On the need for another type of predictive models in structured foods. Int. J. Food Microbiol. 64:247–260. 52. Dens, E. J., K. Bernaerts, A. R. Standaert, J.-U. Kreft, and J. F. Van Impe. 2005. Cell division theory and individual-based modeling of microbial lag. Part II. Modeling lag phenomena induced by temperature shifts. Int. J. Food Microbiol. 101:319–332. 53. Dens, E. J., K. Bernaerts, A. R. Standaert, and J. F. Van Impe. 2005. Cell division theory and individual-based modeling of microbial lag. Part I. The theory of cell division. Int. J. Food Microbiol. 101:303–318. 54. Devlieghere, F., A. H. Geeraerd, K. J. Versyck, H. Bernaert, J. F. Van Impe, and J. Debevere. 2000. Shelf life of modified atmosphere packed cooked meat products: addition of Na-lactate as a fourth shelf life determinative factor in a model and product validation. Int. J. Food Microbiol. 58:93–106. 55. Dhar, N., and J. D. McKinney. 2007. Microbial phenotypic heterogeneity and antibiotic tolerance. Curr. Opin. Microbiol. 10:30–38. 56. Ding, T., Q. L. Dong, S. M. E. Rahman, and D. H. Oh. 2011. Response surface modeling of Listeria monocytogenes inactivation on lettuce treated with electrolyzed oxidizing water. J. Food Process Eng. 34:1729–1745. 57. Dodds, K. L. 1989. Combined effect of water activity and pH on inhibition of toxin production by Clostridium botulinum in cooked, vacuum-packed potatoes. Appl. Environ. Microbiol. 55:656–660. 58. Dong, Q., K. Tu, L. Guo, H. Li, and Y. Zhao. 2007. Response surface model for prediction of growth parameters from spores of Clostridium sporogenes under different experimental conditions. Food Microbiol. 24:624–632. 59. Donsì, G., G. Ferrari, and P. Maresca. 2003. On the modelling of the inactivation kinetics of Saccharomyces cerevisiae by means of combined temperature and high pressure treatments. Innov. Food Sci. Emerg. Technol. 4:35–44. 60. Esnoz, A., P. M. Periago, R. Conesa, and A. Palop. 2006. Application of artificial neural networks to describe the combined effect of pH and NaCl on the heat resistance of Bacillus stearothermophilus. Int. J. Food Microbiol. 106:153–158. 61. Fernández-Navarro F., A. Valero, C. Hervás-Martínez, P. A. Gutiérrez, R. A. García-Gimeno, and G. ZureraCosano. 2010. Development of a multi-classification neural network model to determine the microbial growth/no growth interface. Int. J. Food Microbiol. 141:203–212. 62. Ferrer, J., C. Prats, D. Lopez, and J. Vives-Rego. 2009. Mathematical modelling methodologies in predictive food microbiology: a SWOT analysis. Int. J. Food Microbiol. 134:2–8.

SMP_Food Microbiology_CH40.indd

1016

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1017

63. Francois, K., F. Devlieghere, M. Uyttendaele, A. R. Standaert, A. H. Geeraerd, P. Nadal, J. F. Van Impe, and J. Debevere. 2006. Single cell variability of L. monocytogenes grown on liver paté and cooked ham at 7ºC: comparing challenge test data to predictive simulations. J. Appl. Microbiol. 100:800–812. 64. Fujikawa, H., and T. Itoh. 1996. Tailing of thermal inactivation curve of Aspergillus niger spores. Appl. Environ. Microbiol. 62:3745–3749. 65. Gaillard, S., I. Leguérinel, and P. Mafart. 1998. Model for combined effects of temperature, pH and water activity on thermal inactivation of Bacillus cereus spores. J. Food Sci. 63:887–889. 66. Garcia, D., A. J. Ramos, V. Sanchis, and S. Marín. 2009. Predicting mycotoxins in foods: a review. Food Microbiol. 26:757–769. 67. García-Gimeno, R. M., C. Hervás-Martínez, E. BarcoAlcalá, G. Zurera-Cosano, and E. Sanz-Tapia. 2003. An artificial neural network approach to Escherichia coli O157:H7 growth estimation. J. Food Sci. 68: 639–645. 68. García-Gimeno, R. M., C. Hervás-Martínez, R. RodríguezPérez, and G. Zurera-Cosano. 2005. Modelling the growth of Leuconostoc mesenteroides by artificial neural networks. Int. J. Food Microbiol. 105:317–332. 69. Geeraerd, A., C. Herremans, C. Cenens, and J. F. Van Impe. 1998. Application of artificial neural networks as a non-linear modular modeling technique to describe bacterial growth in chilled food products. Int. J. Food Microbiol. 44:49–68. 70. Geeraerd, A. H., C. H. Herremans, and J. F. Van Impe. 2000. Structural model requirements to describe microbial inactivation during a mild heat treatment. Int. J. Food Microbiol. 59:185–209. 71. Geeraerd, A. H., V. P. Valdramidis, F. Devlieghere, H. Bernaert, J. Debevere, and J. F. Van Impe. 2004. Development of a novel approach for secondary modelling in predictive microbiology: incorporation of ­microbiological knowledge in black box polynomial modelling. Int. J. Food Microbiol. 91:229–244. 72. Geeraerd, A. H., V. P. Valdramidis, and J. F. Van Impe. 2005. GInaFiT, a freeware tool to assess non-log-­linear microbial survivor curves. Int. J. Food Microbiol. 102:95–105. 73. Ghanou Besse, N., N. Audinet, L. Barre, A. Cauquil, M. Cornu, and P. Colin. 2006. Effect of the inoculum size on Listeria monocytogenes growth in structured media. Int. J. Food Microbiol. 110:43–51. 74. Gil, M. M., T. R. S. Brandão, and C. L. M. Silva. 2006. A modified Gompertz model to predict microbial inactivation under time-varying temperature conditions. J. Food Eng. 76:89–94. 75. Ginovart, M., and J. C. Canadas. 2008. INDISIMYEAST: an individual-based simulator on a website for experimenting and investigating diverse dynamics of yeast populations in liquid media. J. Ind. Microbiol. Biot. 35:1359–1366. 76. Ginovart, M., D. López, and A. Gras. 2005. Individualbased modelling of microbial activity study miner-

SMP_Food Microbiology_CH40.indd

1017

77. 78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

alization of C and N and nitrification process in soil. Nonlinear Anal. Real World Appl. 6:773–795. Ginovart, M., D. López, and J. Valls. 2002. INDISIM, an individual-based discrete simulation model to study bacterial cultures. J. Theor. Biol. 214:305–319. Ginovart, M., C. Prats, X. Portell, and M. Silbert. 2011. Exploring the lag phase and growth initiation of a yeast culture by means of an individual-based model. Food Microbiol. 28:810–817. Giuffrida, A., D. Valenti, G. Ziino, B. Spagnolo, and A. Panebianco. 2009. A stochastic interspecific competition model to predict the behaviour of Listeria monocytogenes in the fermentation process of a traditional Sicilian salami. Eur. Food Res. Technol. 228:767–775. Grijspeerdt, K., J.-U. Kreft, and W. Messens. 2005. Individual-based modelling of growth and migration of Salmonella enteritidis in hens’ eggs. Int. J. Food Microbiol. 100:323–333. Grimm, V., T. Wyszomirski, D. Aikman, and J. Uchmanski. 1999. Individual based modelling and ecological theory: synthesis of a workshop. Ecol. Model. 115:275–282. Guillier, L., V. Stahl, B. Hezard, E. Notz, and R. Briandet. 2008. Modelling the competitive growth between Listeria monocytogenes and biofilm microflora of smear cheese wooden shelves. Int. J. Food Microbiol. 128:51–57. Gunvig, A., J. Blom-Hanssen, T. Jacobsen, F. Hansen, and C. Borggaard. 2007. A predictive model for growth of Listeria monocytogenes in meat products with seven hurdle variables, p. 197–200. In Proceedings of the 5th International Conference on Predictive Modelling in Foods. Agricultural University of Athens, Athens, Greece. Gysemans, K. P. M., K. Bernaerts, A. Vermeulen, A. H. Geeraerd, J. Debevere, F. Devlieghere, and J. F. Van Impe. 2007. Exploring the performance of logistic regression model types on growth/no growth data of Listeria monocytogenes. Int. J. Food Microbiol. 114:316–331. Hajmeer, M. N., and I. A. Basheer. 2003. A hybrid Bayesian-neural network approach for probabilistic modeling of bacterial growth/no-growth interface. Int. J. Food Microbiol. 82:233–243. Hajmeer, M., and I. Basheer. 2002. A probabilistic neural network approach for modeling and classification of bacterial growth/no-growth data. J. Microbiol. Methods 51:217–226. Hajmeer, M., I. Basheer, and D. O. Cliver. 2006. Survival curves of Listeria monocytogenes in chorizos modeled with artificial neural networks. Food Microbiol. 23:561–570. Härdin, H., and J. van Schuppen. 2006. System reduc­ tion of nonlinear positive systems by linearization and truncation, p. 431–438. In Positive Systems—Proceedings of the Second International Multidisciplinary Sympo­ sium on Positive Systems: Theory and Applications. Lecture Notes in Control and Information Sciences, vol. 341. Springer, Berlin, Germany. Hills, B., and K. Wright. 1994. A new model for bacterial growth in heterogeneous systems. J. Theor. Biol. 168:31–41.

Manila Typesetting Company

11/08/2012 07:21AM

Advanced Techniques in Food Microbiology

1018 90. Hinshelwood, C. N. 1947. The Chemical Kinetics of the Bacterial Cell. Clarendon Press, Oxford, England. 91. Ho, S. Y., and G. S. Mittal. 2001. Non-thermal microbial inactivation in waste brine using high-voltage low-energy electrical pulses. Innov. Food Sci. Emerg. Technol. 2:251–259. 92. Holzhütter, H.-G. 2004. The principle of flux minimization and its application to estimate stationary fluxes in metabolic networks. Eur. J. Biochem. 271:2905–2922. 93. Hom, L. W. 1972. Kinetics of chlorine disinfection in an ecosystem. J. Sanitary Eng. Div. 98:183–194. 94. Hotchin, J. E. 1955. Use of methyl cellulose as substitute for agar in tissue-culture overlays. Nature 175:352–355. 95. Houtsma, P. C., M. L. Kant-Muermans, F. M. Rombouts, and M. H. Zwietering. 1996. Model for the combined effects of temperature, pH, and sodium lactate on growth rates of Listeria innocua in broth and bologna-type sausages. Appl. Environ. Microbiol. 62:1616–1622. 96. Hülsheger, H., J. Potel, and E.-G. Niemann. 1981. Killing of bacteria with electric pulses of high field strength. Radiat. Environ. Biophys. 20:53–65. 97. Hwang, C. A., and M. L. Tamplin. 2005. Modeling the lag phase and growth rate of Listeria monocytogenes in ground ham containing sodium lactate and sodium diacetate at various storage temperatures. J. Food Sci. 72:M246–M253. 98. Janssen, M., A. H. Geeraerd, F. Logist, Y. De Visscher, K. M. Vereecken, J. Debevere, F. Devlieghere, and J. F. Van Impe. 2006. Modelling Yersinia enterocolitica inactivation in coculture experiments with Lactobacillus sakei as based on pH and lactic acid profiles. Int. J. Food Microbiol. 111:59–72. 99. Jeanson, S., J. Chadoeuf, M. N. Madec, S. Aly, J. Floury, T. F. Brocklehurst, and S. Lortal. 2011. Spatial distribution of bacterial colonies in a model cheese. Appl. Environ. Microbiol. 77:1493–1500. 100. Jeyamkondan, S., D. S. Jayas, and R. A. Holley. 2001. Microbial growth modelling with artificial neural networks. Int. J. Food Microbiol. 64:343–354. 101. Koutsoumanis, K. P., P. A. Kendall, and J. N. Sofos. 2004. A comparative study on growth limits of Listeria monocytogenes as affected by temperature, pH and aw when grown in suspension or on a solid surface. Food Microbiol. 21:415–422. 102. Kreft, J.-U., G. Booth, and J. W. T. Wimpenny. 1998. BacSim, a simulator for individual-based modelling of bacterial colony growth. Microbiology 144:3275–3287. 103. Kreft, J. U., C. Picioreanu, J. W. T. Wimpenny, and M. C. M. Van Loosdrecht. 2001. Individual-based modeling of biofilm. Microbiology 147:2897–2912. 104. Leistner, L. 2000. Basic aspects of food preservation by hurdle technology. Int. J. Food Microbiol. 55:181–186. 105. Le Marc, Y., V. Huchet, C. M. Bourgeois, J. P. Guyonnet, P. Mafart, and D. Thuault. 2002. Modelling the growth kinetics of Listeria as a function of tempera-

SMP_Food Microbiology_CH40.indd

1018

106.

107.

108.

109. 110. 111.

112. 113.

114.

115.

116.

117. 118.

119.

120.

ture, pH and organic acid concentration. Int. J. Food Microbiol. 73:219–237. Le Marc, Y., L. Valik, and A. Medvedova. 2009. Modelling the effect of the starter culture on the growth of Staphylococcus aureus in milk. Int. J. Food Microbiol. 129:306–311. Leporq, B., J.-M. Membré, C. Dervin, P. Buche, and J. P. Guyonnet. 2005. The ‘Sym’Previus’ software, a tool to support decisions to the foodstuff safety. Int. J. Food Microbiol. 100:231–237. Leroy, F., B. Degeest, and L. De Vuyst. 2002. A novel area of predictive modelling: describing the functionality of beneficial microorganisms in foods. Int. J. Food Microbiol. 73:251–259. Li, H., G. Xie, and A. Edmondson. 2007. Evolution and limitations of primary mathematical models in predictive microbiology. Br. Food J. 109:608–626. Liebermeister, W., U. Bauer, and E. Klipp. 2005. Biochemical network models simplified by balanced truncation. FEBS J. 272:4034–4043. Lindroth, S. E., and C. A. Genigeorgis. 1986. Probability of growth and toxin production by nonproteolytic Clostridium botulinum in rockfish stored under modified atmospheres. Int. J. Food Microbiol. 3:167–181. Llaneras, F., and J. Picó. 2008. Stoichiometric modelling of cell metabolism. J. Biosci. Bioeng. 105:1–11. Mafart, P., and I. Leguérinel. 1998. Modeling combined effects of temperature and pH on heat resistance of spores by a linear-Bigelow equation. J. Food Sci. 63:6–8. Malakar, P. K., G. C. Barker, M. H. Zwietering, and K. van ’t Riet. 2003. Relevance of microbial interactions to predictive microbiology. Int. J. Food Microbiol. 84:263–272. McClure, P. J., T. M. Kelly, and T. A. Roberts. 1991. The effects of temperature, pH, sodium chloride and sodium nitrite on the growth of Listeria monocytogenes. Int. J. Food Microbiol. 14:77–92. McKellar, R. 2001. Development of a dynamic continuous-discrete-continuous model describing the lag phase of individual bacterial cells. J. Appl. Microbiol. 90:407–413. McKellar, R. C. 1997. A heterogeneous population model for the analysis of bacterial growth kinetics. Int. J. Food Microbiol. 36:179–186. McMeekin, T. A., J. Bowman, O. McQuestin, L. Mellefont, T. Ross, and M. Tamplin. 2008. The future of predictive microbiology: strategic research, innovative applications and great expectations. Int. J. Food Microbiol. 128:2–9. McMeekin, T. A., R. E. Chandler, and P. E. Doe. 1987. Model for combined effect of temperature and salt concentration/water activity on the growth rate of Staphylococcus xylosus. J. Appl. Bacteriol. 62:543–550. McMeekin, T. A., J. N. Olley, T. Ross, and D. A. Ratkowsky. 1993. Predictive Microbiology: Theory and Application. Research Studies Press Ltd., Baldock, England.

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

1019

121. McMeekin, T. A., J. Olley, D. A. Ratkowsky, and T. Ross. 2002. Predictive microbiology: towards the interface and beyond. Int. J. Food Microbiol. 73:395–407. 122. McMeekin, T. A., and T. Ross. 2002. Predictive micro­biology: providing a knowledge-based framework for change management. Int. J. Food Microbiol. 78:133–153. 123. Mejlholm, O., and P. Dalgaard. 2007. Modeling and predicting the growth of lactic acid bacteria in lightly preserved seafood and their inhibiting effect on Listeria monocytogenes. J. Food Prot. 70:2485–2497. 124. Mejlholm, O., and P. Dalgaard. 2009. Development and validation of an extensive growth and growth boundary model for Listeria monocytogenes in lightly preserved and ready-to-eat shrimp. J. Food Prot. 72:2132–2143. 125. Mejlholm, O., A. Gunvig, C. Borggaard, J. BlomHanssen, L. Mellefont, T. Ross, F. Leroi, T. Else, D. Visser, and P. Dalgaard. 2010. Predicting growth rates and growth boundary of Listeria monocytogenes—an international validation study with focus on processed and ready-to-eat meat and seafood. Int. J. Food Microbiol. 141:137–150. 126. Meldrum, R. J., T. F. Brocklehurst, D. R. Wilson, and P. D. G. Wilson. 2003. The effects of cell immobilization, pH, and sucrose on the growth of Listeria monocytogenes Scott A at 10ºC. Food Microbiol. 20:97–103. 127. Membré, J. M., and R. J. W. Lambert. 2008. Application of predictive modelling techniques in industry: from food design up to risk assessment. Int. J. Food Microbiol. 128:10–15. 128. Mertens, L., A. H. Geeraerd, T. D. T. Dang, A. Vermeulen, K. Serneels, E. Van Derlinden, A. M. Cappuyns, P. Moldenaers, J. Debevere, F. Devlieghere, and J. F. Van Impe. 2009. Design of an experimental viscoelastic food model system for studying Zygosaccharomyces bailii spoilage in acidic sauces. Appl. Environ. Microbiol. 75:7060–7069. 129. Mertens, L., E. Van Derlinden, T. D. T. Dang, A. M. Cappuyns, A. Vermeulen, J. Debevere, P. Moldenaers, F. Devlieghere, A. H. Geeraerd, and J. F. Van Impe. 2011. On the critical evaluation of growth/no growth assessment of Zygosaccharomyces bailii with optical density measurements: liquid versus structured media. Food Microbiol. 28:736–745. 130. Métris, A., Y. Le Marc, A. Elfwing, A. Ballagi, and J. Baranyi. 2005. Modelling the variability of lag times and the first generation times of single cells of E. coli. Int. J. Food Microbiol. 100:13–19. 131. Miles, D. W., T. Ross, J. Olley, and T. A. McMeekin. 1997. Development and evaluation of a predictive model for the effect of temperature and water activity on the growth rate of Vibrio parahaemolyticus. Int. J. Food Microbiol. 38:133–142. 132. Molina, M., and L. Giannuzzi. 1999. Combined effect of temperature and propionic acid concentration on the growth of Aspergillus parasiticus. Food Res. Int. 32:677–682. 133. Murphy, J. T., and R. Walshe. 2007. Micro-gen: an agent-based model of bacteria-antibiotic interactions

SMP_Food Microbiology_CH40.indd

1019

134. 135.

136.

137.

138. 139. 140.

141.

142. 143. 144. 145. 146.

147.

148.

in batch culture, p. 239–242. In Proceedings of the Annual European Simulation and Modelling (ESM 2007). Eurosis, Ostend, Belgium. Nene, Y. L., V. K. Sheila, and J. P. Moss. 1996. Tapioca—a potential substitute for agar in tissue culture media. Curr. Sci. 70:493–494. Nikolaou, M., and V. H. Tam. 2005. A new modeling approach to the effect of antimicrobial agents on heterogeneous microbial populations. J. Math. Biol. 52:154–182. Nilsson, L., and Y. Chen, M. L. Chikindas, H. H. Huss, L. Gram, and T. J. Montville. 2000. Carbon dioxide and nisin act synergistically on Listeria monocytogenes. Appl. Environ. Microbiol. 66:769–774. Noma, S., D. Kajiyama, N. Igura, M. Shimoda, and I. Hayakawa. 2006. Mechanisms behind tailing in the pressure inactivation curve of a clinical isolate of Escherichia coli O157:H7. Int. J. Food Microbiol. 109:103–108. Noriega, E., A. Laca, and M. Díaz. 2009. Listeria growth under diffusional limitations in synthetic meats. Int. J. Food Sci. Technol. 44:725–734. Nyström, T. 2004. Stationary-phase physiology. Annu. Rev. Microbiol. 58:161–181. Parente, E., M. A. Giglio, A. Ricciardi, and F. Clementi. 1998. The combined effect of nisin, leucocin F10, pH, NaCl and EDTA on the survival of Listeria monocytogenes in broth. Int. J. Food Microbiol. 40:65–75. Park, S. Y., J. W. Choi, J. Yeon, M. Jeong Lee, D. H. Chung, M. G. Kim, K. H. Lee, K. S. Kim, D. H. Lee, G. J. Bahk, D. H. Bae, K. Y. Kim, C. H. Kim, and S. D. Ha. 2009. Predictive modeling for the growth of Listeria monocytogenes as a function of temperature, NaCl and pH. J. Microbiol. Biotechnol. 15:1323–1329. Peck, S. L. 2004. Simulation as experiment: a philosophical reassessment for biological modeling. Trends Ecol. Evol. 19:530–534. Peleg, M. 1995. A model for microbial survival after exposure to pulsed electric field. J. Sci. Food Agric. 67:93–99. Peleg, P., M. D. Normand, and E. Damru. 1997. Mathematical interpretation of dose-response curves. Bull. Math. Biol. 59:747–761. Pin, C., J. Sutherland, and J. Baranyi. 1999. Validating predictive models of food spoilage organisms. J. Appl. Microbiol. 87:491–499. Poschet, F., K. M. Vereecken, A. H. Geeraerd, B. M. Nicolaï, and J. F. Van Impe. 2005. Analysis of a novel class of predictive microbial growth models and application to coculture growth. Int. J. Food Microbiol. 100:107–124. Prats, C., D. López, A. Giró, J. Ferrer, and J. Valls. 2006. Individual-based modelling of bacterial cultures to study the microscopic causes of the lag phase. J. Theor. Biol. 241:939–953. Prats, C., A. Giró, J. Ferrer, D. López, and J. VivesRego. 2008. Analysis and IbM simulation of the stages in bacterial lag phase: basis for an updated definition. J. Theor. Biol. 252:56–68.

Manila Typesetting Company

11/08/2012 07:21AM

1020

Advanced Techniques in Food Microbiology

149. Presser, K. A., D. A. Ratkowsky, and T. Ross. 1997. Modelling the growth rate of Escherichia coli as a function of pH and lactic acid concentration. Appl. Environ. Microbiol. 63:2355–2360. 150. Presser, K. A., T. Ross, and D. A. Ratkowsky. 1998. Modelling the growth limits (growth/no growth interface) of Escherichia coli as a function of temperature, pH, lactic acid concentration, and water activity. Appl. Environ. Microbiol. 64:1773–1779. 151. Ramakrishna, R., J. S. Edwards, A. Mcculluch, and B. O. Palsson. 2001. Flux balance analysis of mitochondrial energy metabolism: consequences of systemic stoichiometric constraints. Am. J. Physiol. Regul. Integr. Comp. Physiol. 280:R695–R704. 152. Ratkowsky, D. A., R. K. Lowry, T. A. McMeekin, A. N. Stokes, and E. Chandler. 1983. Model for bacterial culture growth rate throughout the entire biokinetic temperature range. J. Bacteriol. 154:1222–1226. 153. Ratkowsky, D. A., J. Olley, T. A. McMeekin, and A. Ball. 1982. Relationship between temperature and growth rate of bacterial cultures. J. Bacteriol. 149:1–5. 154. Ratkowsky, D. A., and T. Ross. 1995. Modelling the bacterial growth/no growth interface. Lett. Appl. Microbiol. 20:29–33. 155. Razavilar, V., and C. Genigeorgis. 1998. Prediction of Listeria spp. growth as affected by various levels of chemicals, pH, temperature and storage time in a model broth. Int. J. Food Microbiol. 40:149–157. 156. Ross, T., D. A. Ratkowsky, L. A. Mellefont, and T. A. McMeekin. 2003. Modelling the effects of temperature, water activity, pH and lactic acid concentration on the growth rate of Escherichia coli. Int. J. Food Microbiol. 82:33–43. 157. Ross, T., and J. Sumner. 2002. A simple, spreadsheetbased, food safety risk assessment tool. Int. J. Food Microbiol. 77:39–53. 158. Rosso, L. 1995. Modélisation et microbiologie prévisionnelle: elaboration d’un nouvel outil pour l’agroalimentaire. Ph.D. thesis. Université Claude Bernard, Lyon, France. 159. Rosso, L., J. R. Lobry, S. Bajard, and J. P. Flandrois. 1995. Convenient model to describe the combined effects of temperature and pH on microbial growth. Appl. Environ. Microbiol. 61:610–616. 160. Rosso, L., J. R. Lobry, and J. P. Flandrois. 1993. An unexpected correlation between cardinal temperatures of microbial growth highlighted by a new model. J. Theor. Biol. 162:447–463. 161. Rosso, L., and T. Robinson. 2001. A cardinal model to describe the effect of water activity on the growth of moulds. Int. J. Food Microbiol. 63:265–273. 162. Salter, M. A., D. A. Ratkowsky, T. Ross, and T. A. McMeekin. 2000. Modelling the combined temperature and salt (NaCl) limits for growth of a pathogenic Escherichia coli strain using nonlinear logistic regression. Int. J. Food Microbiol. 61:159–167. 163. Sautour, M., P. Dantigny, C. Divies, and M. Bensoussan. 2001. A temperature type model for describing the relationship between fungal growth and water activity. Int. J. Food Microbiol. 67:63–69.

164. Schaffner, D. W., W. H. Ross, and T. J. Montville. 1998. Analysis of the influence of environmental parameters on Clostridium botulinum time-to-toxicity by using three modeling approaches. Appl. Environ. Microbiol. 64:4416–4422. 165. Schilling, C., M. Covert, I. Famili, G. Church, J. Edwards, and B. Palsson. 2002. Genome-scale metabolic model of Helicobacter pylori 26695. J. Bacteriol. 184:4582–4593. 166. Schoolfield, R. M., P. J. H. Sharpe, and C. E. Magnuson. 1981. Non linear regression of biological temperaturedependent rate models based on absolute reaction rate theory. J. Theor. Biol. 88:719–731. 167. Schuetz, R., L. Kuepfer, and U. Sauer. 2007. Systematic evaluation of objective functions for predicting intracellular fluxes in Escherichia coli. Mol. Syst. Biol. 3:119. 168. Schuytser, M. A., J. Straatsma, P. M. Keijzer, M. Verschueren, and P. De Jong. 2008. A new web-based modelling tool (Websim-MILQ) aimed at optimisation of thermal treatments in the dairy industry. Int. J. Food Microbiol. 128:153–157. 169. Schvartzman, M. S., X. Belessi, F. Butler, P. Skandamis, and K. Jordan. 2010. Comparison of growth limits of Listeria monocytogenes in milk, broth and cheese. J. Appl. Microbiol. 109:1790–1799. 170. Shamsi, K., C. Versteeg, F. Sherkat, and J. Wan. 2008. Alkaline phosphatase and microbial inactivation by pulsed electric field in bovine milk. Innov. Food Sci. Emerg. Technol. 9:217–233. 171. Shimizu, H., Y. Shinfuku, M. Sono, C. Furusawa, and T. Hirasawa. 2008. Metabolic flux balance analysis of an industrially useful microorganism Corynebacterium glutamicum by a genome-scale reconstructed model. In Proceedings of the 3rd International Conference on Bio-Inspired Models of Network, Information and Computing Systems, article no. 17. ICST, Brussels, Belgium. 172. Skandamis, P. N., and G. J. E. Nychas. 2000. Development and evaluation of a model predicting the survival of Escherichia coli O157:H7 NCTC 12900 in homemade eggplant salad at various temperatures, pHs, and oregano essential oil concentrations. Appl. Environ. Microbiol. 66:1646–1653. 173. Skandamis, P. N., K. W. Davies, P. J. McClure, K. Koutsoumanis, and C. Tassou. 2002. A vitalistic approach for non-thermal inactivation of pathogens in traditional greek salads. Food Microbiol. 19:405–421. 174. Skandamis, P. N., T. F. Brocklehurst, E. Z. Panagou, and G. J. E. Nychas. 2007. Image analysis as a mean to model growth of Escherichia coli O157:H7 in gel cassettes. J. Appl. Microbiol. 103:937–947. 175. Standaert, A. R., F. Poschet, A. H. Geeraerd, F. V. Uylbak, J. U. Kreft, and J. F. Van Impe. 2004. A novel class of predictive microbial growth models: implementation in an individual-based framework, p. 183–188. In Proceedings of the 9th International Symposium on Computer Applications in Biotechnology (CAB9). 176. Stecchini, M. L., M. Del Torre, I. Sarais, O. Saro, M. Messina, and E. Maltini. 1998. Influence of structural

SMP_Food Microbiology_CH40.indd

1020

Manila Typesetting Company

11/08/2012 07:21AM

40.  Predictive Microbiology

177.

178.

179.

180.

181.

182.

183.

184. 185.

186.

187.

188.

189.

1021

properties and kinetic constraints on Bacillus cereus growth. Appl. Environ. Microbiol. 64:1075–1078. Swinnen, I. A. M., K. Bernaerts, K. Gysemans, and J. F. Van Impe. 2005. Quantifying microbial lag phenomena due to a sudden rise in temperature: a systematic macroscopic study. Int. J. Food Microbiol. 100:85–96. Tapia de Daza, M. S., Y. Villegas, and A. Martinez. 1991. Minimal water activity for growth of Listeria monocytogenes as affected by solute and temperature. Int. J. Food Microbiol. 14:333–337. Theys, T. E., A. H. Geeraerd, A. Verhulst, K. Poot, F. Van Bree, F. Devlieghere, P. Moldenaers, D. Wilson, T. Brocklehurst, and J. F. Van Impe. 2008. Effect of pH, water activity and gel microstructure including oxygen profiles and rheological characterization, on the growth kinetics of Salmonella Typhimurium. Int. J. Food Microbiol. 128:67–77. Theys, T. E., A. H. Geeraerd, and J. F. Van Impe. 2009. Evaluation of a mathematical model structure describing the effect of (gel) structure on the growth of Listeria innocua, Lactococcus lactis and Salmonella Typhimurium. J. Appl. Microbiol. 107:775–784. Theys, T. E., A. H. Geeraerd, F. Devlieghere, and J. F. Van Impe. 2009. Extracting information on the evolution of living- and dead-cell fractions of Salmonella Typhimurium colonies in gelatin gels based on microscopic images and plate-count data. Lett. Appl. Microbiol. 49:39–45. Theys, T. E., A. H. Geeraerd, F. Devlieghere, and J. F. Van Impe. 2010. On the selection of relevant environmental factors to predict microbial dynamics in solidified media. Food Microbiol. 27:220–228. Tienungoon, S., D. A. Ratkowsky, T. A. McMeekin, and T. Ross. 2000. Growth limits of Listeria monocytogenes as a function of temperature, pH, NaCl, and lactic acid. Appl. Environ. Microbiol. 66:4979–4987. Tsigarida, E., I. Boziaris, and G. J. E. Nychas. 2003. Bacterial synergism or antagonism in a gel cassette system. Appl. Environ. Microbiol. 69:7204–7209. Valdramidis, V. P., A. H. Geeraerd, and J. F. Van Impe. 2007. Stress-adaptive responses by heat under the microscope of predictive microbiology. J. Appl. Microbiol. 103:1922–1930. Valero, A., E. Carrasco, F. Pérez-Rodriguez, R. M. García-Gimeno, and G. Zurera. 2006. Growth/no growth model of Listeria monocytogenes as a function of temperature, pH, citric acid and ascorbic acid. Eur. Food Res. Technol. 224:91–100. Van Breusegem, V., and G. Bastin. 1991. Reduced order dynamical modelling of reaction systems: a singular perturbation approach, p. 1049–1054. In Proceedings of the 30th IEEE conference on decision and control. IEEE, Washington, DC. Van Derlinden, E., K. Bernaerts, and J. F. Van Impe. 2009. Unraveling E. coli dynamics close to the maximum growth temperature through heterogeneous modeling. Lett. Appl. Microbiol. 49:659–665. Van Derlinden, E., K. Bernaerts, and J. F. Van Impe. 2010. Quantifying the heterogeneous heat response of

SMP_Food Microbiology_CH40.indd

1021

190.

191.

192. 193.

194.

195.

196.

197. 198.

199. 200.

201.

202.

E. coli under dynamic temperatures. J. Appl. Microbiol. 108:1123–1135. Van Impe, J. F., A. M. Cappuyns, and E. Van Derlinden. 2009. Towards a next generation of predictive models based on systems biology tools. In Proceedings of the 6th International Conference on Predictive Modelling in Foods. ICPMF, Dublin, Ireland. Van Impe, J. F., F. Poschet, A. H. Geeraerd, and K. M. Vereecken. 2005. Towards a novel class of predictive microbial growth models. Int. J. Food Microbiol. 100:97–105. Varma, A., B. Boesch, and B. Palsson. 1993. Biochemical production capabilities of Escherichia coli. Biotechnol. Bioeng. 42:59–73. Vereecken, K. M., and J. F. Van Impe. 2002. Analysis and practical implementation of a model for combined growth and metabolite production of lactic acid bacteria. Int. J. Food Microbiol. 73:239–250. Vereecken, K. M., F. Devlieghere, A. Bockstaele, J. Debevere, and J. F. Van Impe. 2003. A model for lactic acid-induced inhibition of Yersinia enterocolitica in mono- and coculture with Lactobacillus sakei. Food Microbiol. 20:701–713. Vermeulen, A., K. P. M. Gysemans, K. Bernaerts, A. H. Geeraerd, J. Debevere, F. Devlieghere, and J. F. Van Impe. 2009. Modelling the influence of the inoculation level on the growth/no growth interface of Listeria monocytogenes as a function of pH, aw and acetic acid. Int. J. Food Microbiol. 135:83–89. Virto, R., D. Sanz, I. Álvarez, S. Condón, and J. Raso. 2006. Application of the Weibull model to describe inactivation of Listeria monocytogenes and Escherichia coli by citric and lactic acid at different temperatures. J. Sci. Food Agric. 86:865–870. Watson, H. E. 1908. A note on the variation of the rate of disinfection with change in the concentration of the disinfectant. J. Hyg. 8:536–542. Wilson, P. D. G., T. F. Brocklehurst, S. Arino, D. Thuault, M. Jakobsen, M. Lange, J. Farkas, J. W. T. Wimpenny, and J. F. Van Impe. 2002. Modelling microbial growth in structured foods: towards a unified approach. Int. J. Food Microbiol. 73:275–289. Whiting, R. 1993. Modeling bacterial survival in unfavorable environments. J. Ind. Microbiol. 12:240–246. Xavier, J. B., M. K. De Kreuk, C. Picioreanu, and M. C. M. van Loosdrecht. 2007. Multi-scale individualbased model of microbial and bioconversion dynamics in aerobic granular sludge. Environ. Sci. Technol. 41:6410–6417. Ye, S.-Y., Y.-X. Qiu, X.-L. Song, and S.-C. Luo. 2009. Optimization of process parameters for the inactivation of Lactobacillus sporogenes in tomato paste with ultrasound and 60Co-g irradiation using response surface methodology. Radiat. Phys. Chem. 78: 227–233. Yuk, H.-G., and D. L. Marshall. 2003. Heat adaptation alters Escherichia coli O157:H7 membrane lipid composition and verotoxin production. Appl. Environ. Microbiol. 69:5115–5119.

Manila Typesetting Company

11/08/2012 07:21AM

1022

Advanced Techniques in Food Microbiology

203. Zagaris, A., H. Kaper, and T. Kaper. 2004. Analysis of the computational singular perturbation reduction method for chemical kinetics. J. Nonlinear Sci. 14:59–91. 204. Zaika, L. L., E. Moulden, L. Weimer, J. G. Phillips, and R. L. Buchanan. 1994. Model for the combined effects of temperature, initial pH, sodium chloride and sodium nitrite concentrations on anaerobic growth of Shigella flexneri. Int. J. Food Microbiol. 23:345–358. 205. Zheng, H., H. Zhou, T. Shen, and B. Rui. 2009. Flux balance analysis within physiologically feasible region, p. 1– 4. In Proceedings of the 3rd Conference on Bioinformatics and Biomedical Engineering. IEEE, Washington, DC.

206. Zobeley, J., D. Lebiedz, J. Kammerer, A. Ishmurzin, and U. Kummer. 2005. A new time dependent complexity reduction method for biochemical systems. Trans. Comput. Syst. Biol. 3880:90–110. 207. Zwietering, M., I. Jongenburger, F. Rombouts, and K. van ’t Riet. 1990. Modeling of the bacterial growth curve. Appl. Environ. Microbiol. 56:1875–1881. 208. Zwietering, M. H., H. G. A. M. Cuppers, J. C. de Wit, and K. van ’t Riet. 1994. Evaluation of data transformations and validation of a model for the effect of temperature on bacterial growth. Appl. Environ. Microbiol. 60:195–203.

SMP_Food Microbiology_CH40.indd

1022

Manila Typesetting Company

11/08/2012 07:21AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch41

Juliana M. Ruzante Richard C. Whiting Sherri B. Dennis Robert L. Buchanan

41

Microbial Risk Assessment

In the context of food safety risk analysis, risk is a func­ tion of the probability or likelihood of an adverse health effect and the severity of that effect consequential to the consumption of a food containing the hazard(s) (19). Assessment of the risk (known as risk assessment), risk management, and risk communication constitute the three components of risk analysis. In this chapter, we will introduce the basic concepts of microbial risk assess­ ment and provide an overview of the methodology and applications to food safety.

RISK ANALYSIS Risk analysis is used to develop an estimate of risks, to identify and implement measures to control the risks, and to communicate with stakeholders about the risks and measures applied (19). It is an integral part of U.S. and international decision making in food safety. Risk analysis has improved the decision-making process by formalizing the structure and the management of the analysis and by bringing together science and policy in an effective manner (11). In addition, it has improved

food safety and public health by providing a general framework that helps establish science-based targets to reduce foodborne illness; plan, implement, and monitor interventions; establish priorities; and target messages to the most at-risk populations (11). The conceptual framework for risk analysis has undergone significant advances, much of them via international bodies (the Codex framework is shown in Fig. 41.1, but see also www.foodrisk.org for more details and other risk analy­ sis frameworks proposed by different international or­ ganizations and by the U.S. Federal Government). The process of risk analysis is initiated by the risk manager. Risk managers are the people who need the risk assess­ ment to assist them in making a decision about the safety of a food, to estimate the impact of mitigation strate­ gies, or to address other safety-related questions. Risk managers set the objectives for the risk assessment and pose specific questions to the risk assessment team. The risk managers in many organizations also assemble the risk assessment team and provide staff, funds, and time for the risk assessors to conduct the assessment. The risk management process might also include managing

Juliana M. Ruzante, Pew Health Group, The Pew Charitable Trust, Washington, DC. Richard C. Whiting, Exponent, Inc., Center for Chemical Regulation and Food Safety, Bowie, MD. Sherri B. Dennis, Center for Food Safety and Applied Nutrition, Food and Drug Administration, College Park, MD. Robert L. Buchanan, Center for Food Safety and Security Systems, University of Maryland, College Park, MD.

1023

SMP_Food Microbiology_CH41.indd

1023

Manila Typesetting Company

11/08/2012 08:36PM

1024

Advanced Techniques in Food Microbiology

Figure 41.1  Codex Alimentarius risk analysis framework. doi:10.1128/9781555818463.ch41f1

communication and transparency by creating outside expert panels, conducting public ­ meetings, requesting independent reviews of the finished risk assessment, and disseminating the results in appropriate forums. The last may consist of preparing an interpretive sum­ mary to accompany the detailed risk assessment and disseminating the risk assessment through publications and public presentations. The activities and responsi­ bilities of risk management will vary with the different types of organizations; for example, private industries might not need to assemble public meetings or peer review panels, while this is often required for govern­ mental agencies. The second component of risk analysis, risk assess­ ment, is the process of collecting and evaluating data, constructing models, calculating the requested values, and interpreting the results. The activities are specifi­ cally directed by the questions posed by risk managers. Risk assessors conduct the assessment and present their analysis back to the risk managers. The risk managers then decide on a course of action (which may include making no changes), organize the response, and order a follow-up evaluation of the effectiveness of the action. It is very important that risk managers respect the risk assessors’ role in evaluating and selecting scientific data, creating distributions and models, and conducting the calculations and that they not impose any bias upon the risk assessment. Risk assessment plays an essential role in international trade. According to the World Trade Organization, Sanitary and Phytosanitary Agreement (WTO/SPS), countries can impose food safety standards that are more stringent than international standards in order to protect human, animal, and plant health; however, these measures must be based on risk assessments and must follow the principles established by international organizations (i.e., Codex Alimentarius Commission)

SMP_Food Microbiology_CH41.indd

1024

(30, 66). Under the WTO/SPS, risk assessments can also be used to demonstrate equivalency in cases where dif­ ferent interventions are used to achieve the same appro­ priate level of protection (ALOP) (66). At the national level, risk assessment should be an integral part of the national food safety system. The Executive Branch of the U.S. Government currently works under Executive Order 12866 (1993), which articulates the need for Federal agencies to base new regulations on appropri­ ate assessments of risk to support the impact and ef­ fectiveness of any proposed regulation. This is also a recommendation from the Codex Alimentarius to other countries when they are formulating policies in public health and food safety. Risk assessment also provides a systematic way to determine which steps or parameters in a process have a greater effect on the risk; it can help estimate the impact of mitigations and identify knowl­ edge gaps; and it allows industry to innovate by help­ ing set up hazard analysis and critical control points (HACCP) plans, performance objectives, and criteria to meet the ALOP. With the increase in the use of risk assessments in defining regulations and implementing policies that may have impact both at the national and the global levels, risk assessments have been the target of considerable sci­ entific, political, and public scrutiny (1), and therefore, the process needs to be carefully planned, transparent, and well communicated across the different stakeholder groups. Lastly, risk communication refers to the sharing of goals, information, and data between risk assessors and managers in addition to other stakeholders such as the public and industry. Risk communication plays an impor­ tant role in ensuring that the results and other important details of a risk assessment are clearly understandable and communicated to all parties; it is essential in ensuring transparency in risk analysis. Experience has shown that

Manila Typesetting Company

11/08/2012 08:36PM

1025

41.  Microbial Risk Assessment regular communication between all parties is necessary to ensure that the risk assessment will have maximum value to the risk managers and that risk management decisions will be better understood and accepted by the public and various stakeholders; therefore, risk commu­ nication should encompass both risk management and risk assessment as proposed by the Codex risk analysis framework (Fig. 41.1). A risk assessment initiated in 2009 by the FDA and the Food Safety and Inspection Service of the U.S. Department of Agriculture to evaluate the risk of liste­ riosis associated with consumption of products sliced or packaged in retail delis (e.g., deli meats, cheese, and deli salads) illustrates the value of involving stakeholders throughout the process of conducting a risk assessment. A public meeting was held to present the background, approach, scope, and data needs for the interagency risk assessment (22). A major data need for this risk assess­ ment was an understanding of food worker behavior including the sequence of actions and the frequency of contact with objects and deli products. Industry assisted in conducting an observation study of food safety prac­ tices in retail deli departments (37). Involving stakehold­ ers early, beginning with the risk assessment design, is anticipated to increase transparency, reduce data gaps, and increase acceptance and understanding of the risk assessment results by the public. According to Hallman (26), “risk communication is not a substitute for poor risk assessment or manage­ ment, nor should it be used as a type of public relations designed to placate the public after an incident involving illness or contamination; instead, it should be an inte­ gral part in each step of risk analysis process.” In this chapter, we will focus on the risk assessment portion of risk analysis, but additional information on the principles of risk analysis and managing risk assess­ ments can be found elsewhere (3, 4, 7, 13, 14, 44).

STEPS OF A MICROBIAL RISK ASSESSMENT Microbial and chemical risk assessments have very similar steps and components; the major difference lies with the methodology used in each one. Often, the ad­ verse consequences of microbial hazards are acute ill­ ness (single exposure), while concerns about chemicals are often (but not exclusively) for chronic exposure. A distinct consideration when developing a microbial risk assessment model is the need to account for the changes in the concentration and prevalence of the hazard, as microbes grow and/or numbers decline throughout the food supply chain (e.g., from farm to fork). In contrast, chemicals usually are diluted or degraded over the life of

SMP_Food Microbiology_CH41.indd

1025

the food, and they seldom increase, and if they do, not by the orders of magnitude that bacteria can. Numerous papers and guidelines have described the basic approaches and methods for conducting microbial risk assessments (29, 40, 49, 59, 61). According to the Codex Alimentarius, the four steps of a risk assessment are hazard identification, exposure assessment, hazard characterization, and risk characterization (7).

Hazard Identification

Hazard identification is the collection of information on the pathogen, food process, risk factors, and disease that is relevant to the risk assessment. It is a descrip­ tive part of the risk assessment report, and it establishes the scope of the risk assessment in terms of the specific food(s), hazard(s), and host (populations of concern) that are addressed.

Exposure Assessment

Exposure assessment is the determination of the prob­ ability of consuming the pathogen and the cell numbers consumed. For a food process, this includes initial con­ tamination frequency and pathogen cell numbers, the effect of any dilution, mixing, or inactivation processes, the amount of growth during transportation and stor­ age, the frequency of recontamination, and handling practices by the food handler or preparer (42). For mi­ crobial hazards, modeling the changes in growth, sur­ vival, or inactivation of populations during each event from raw material to consumption typically is the largest and most complex part of the risk assessment. Predictive microbiology methods and tools are used in character­ izing the changes in microbial populations that occur in foods in each step of the model (42) (see chapter 40, “Predictive Microbiology”). After the final concentra­ tion of the microbial hazard is estimated, it is then com­ bined with food consumption patterns in the exposed population to determine exposure. The low accuracy in determining the bacterial numbers in food and also in estimating the food intake among the different popula­ tion groups is one of the main challenges and sources of uncertainty when calculating the exposure assessment (68). Pathogens are usually present in very low concen­ trations, below the detection limit of most available di­ agnostic tests, which makes detection and quantification difficult, if not impossible (68).

Hazard Characterization

Hazard characterization, or dose-response relationship, describes the nature and extent of the adverse health ef­ fects to individuals from consuming a specified number of the pathogen. It includes consideration of the virulence

Manila Typesetting Company

11/08/2012 08:36PM

1026

Advanced Techniques in Food Microbiology

of the pathogen strain and susceptibility of individual human subgroups to illness. Many foodborne pathogens are opportunistic and affect individuals with impaired immune systems or other vulnerabilities. Children, the elderly, pregnant women, individuals with diseases who are undergoing medical treatments that suppress the im­ mune system, and individuals with other underlying dis­ eases are more vulnerable to infectious microorganisms (e.g., Salmonella and Listeria monocytogenes) and in­ fectious-toxigenic microorganisms (e.g., Escherichia coli O157:H7). The food matrix may also affect the dose-re­ sponse relationship by neutralizing stomach acidity and thereby allowing more pathogens to enter the intestinal tract (9). The adverse-response parameter (biological end points) of interest for the dose-response behavior should be specified by risk managers. One might be interested only in the mortality caused by the hazards, but most often the response parameters consist of a range of morbidities, outcomes, hospitalizations, and deaths and might even include associated long-term sequelae such as Guillain-Barré syndrome for Campylobacter infec­ tions or hemolytic uremic syndrome for E. coli O157: H7 infections. The dose-response relationship is typically repre­ sented by a sigmoid curve with a plot of the log dose versus the probability of illness (28, 41). This is com­ monly interpreted as being indicative of the existence of a threshold; however, this is not a valid assumption because threshold and nonthreshold models can pro­ duce a sigmoid curve when plotted on a log dose versus the probability of illness scale (16). Data for the doseresponse relationship come from animal studies, human volunteer feeding studies, and epidemiologic data from investigations of illnesses (33); each of these have con­ siderable uncertainties due to the inherent variability in the pathogen, the host, and the food vehicle (5, 68). Animal models and healthy human responses do not always correlate well with responses in the susceptible human population. Volunteer studies or clinical human exposure studies are not possible with highly virulent pathogens and cannot involve individuals from suscep­ tible populations because of the possibility of a serious or fatal outcome (5). An epidemiologic study for an outbreak of L. monocytogenes in a Finnish hospital il­ lustrates the collection of data necessary to determine human susceptibilities (38). In this study, the pathogen populations in the food consumed, amount of food con­ sumed, health status of individuals who became ill, and numbers of people who became ill and did not become ill were determined. These types of data from outbreaks help calculate the probabilities of developing illness

and are helpful in providing points for plotting a doseresponse curve; unfortunately, the dose consumed by individuals and the exposed population are often un­ known (68). Three mathematical functions have received the most frequent use for modeling the dose-response relationship, but none is preeminent for all pathogens (28). These in­ clude the exponential, beta-Poisson, and Weibull-gamma models, which are 1-, 2-, and 3-parameter models, re­ spectively. All three are sigmoid, nonthreshold models but have significant differences when extrapolated to low doses. Unfortunately, it is extremely difficult to col­ lect data at low doses and low probabilities of illness to precisely fit a model. As nonthreshold models, each as­ sumes that a single cell of a pathogen has some prob­ ability of causing illness. When all of these models are plotted on a log dose-log probability of illness basis to show the probability ranges at which an ALOP and food safety objective (FSO) would be set, the models are lin­ ear. This means that in the linear portion of the model, a 10-fold increase in dose will have a 10-fold increase in the probability of causing illness. On a population basis, increasing the frequency of contamination 10-fold (all else equal) will increase the number of illnesses in a population 10-fold. Despite the difficulties in determining dose-response relationships for different pathogens and populations, these relationships are beginning to be described (35). Dose-response models have been published for L. mono­ cytogenes (17, 21), E. coli O157:H7 (25, 45, 51), Salmonella (15, 36, 43), Vibrio parahaemolyticus (20), and Cryptosporidium and Giardia (46, 54). Recently, a key events dose-response framework has been pro­ posed to better utilize current and future data to gain insight regarding dose-response relationships based on understanding the critical biological events that occur between exposure and disease (5). This approach may help understand the data needs and improve the doseresponse models, especially for low-dose exposures that rely heavily on assumptions and extrapolations to cope with limitations in knowledge.

SMP_Food Microbiology_CH41.indd

1026

Risk Characterization

The final step of a risk assessment is risk characteriza­ tion. It combines the exposure assessment for the foods of concern with the specific dose-response relationship (hazard characterization) of the pathogen and popula­ tion to calculate an estimate of risk. At this step, the evaluation of different scenarios is also presented and discussed. Using these four steps, a risk assessment links each of the events in the food production chain to the total

Manila Typesetting Company

11/08/2012 08:36PM

1027

41.  Microbial Risk Assessment number of people exposed at consumption and then to public health (probability of illness). Prior to the adop­ tion of risk assessment approaches, the impact of each processing step, e.g., pasteurization, could be evaluated only in isolation, separate from the possible changes in the concentration of the pathogen that might occur throughout the food chain. While it was possible to un­ derstand the impact of a treatment on the concentration of the hazard in the food at that point in the food supply chain, it was not possible to translate that change with the likelihood of a reduction in the potential for human illness at the point of consumption of the food. Hence, it was difficult to evaluate the relative impact of differ­ ent intervention strategies taken at different steps in the food process; for example, it was difficult to determine whether reducing the initial contamination levels or lowering a storage temperature would be more effec­ tive in reducing the number of illnesses. The linkage to public health also allows the ability to establish a safety objective for the food process and determine criteria to know whether a particular control step is adequate. Decisions of whether, for example, a pasteurization step requires inactivation of 5- or 7-log CFU/g reduction of a pathogen or whether the air temperature in a cooling step should be 4.4 or 7.2°C (40 or 45°F) cannot be made without a risk assessment. Therefore, risk assessments help evaluate the efficacy of potential risk mitigations and determine requirements for the performance of each control step; the critical control points of the HACCP plan are set to achieve these requirements (50). Although most of the microbial risk assessments are done addressing a particular pathogen and commodity, risk assessments can be done to help risk managers rank risks in a more scientific manner. The quantitative risk assessment for L. monocytogenes in ready-to-eat foods is an example of a model that was developed and ranks 23 different ready-to-eat categories of commodities ac­ cording to their public health impact (21). Risk assess­ ments such as this are extremely helpful in establishing priorities and optimizing resources. The FAO/WHO model for Cronobacter sakazakii (18) also produces a relative ranking when comparing different intervention strategies with a baseline scenario. Relative rankings can be extremely helpful because assumptions and limita­ tions tend to be consistent across the different scenarios, which minimizes the impact of uncertainty and variabil­ ity within the risk assessment. Economic analyses have not traditionally been consid­ ered part of the risk assessment process. However, deci­ sions that risk managers frequently need to make include an economic, cost-benefit, or most-cost-effective mitiga­ tion component (31, 63). When an economic analysis

SMP_Food Microbiology_CH41.indd

1027

follows a risk assessment, it is important that the risk assessment questions and outputs (answers) be appro­ priately specified in the beginning of the risk assessment process so results are suitable for the economic analysis.

MICROBIAL RISK ASSESSMENT METHODOLOGY Risk assessments can be either qualitative or quan­ titative. Risk assessment also can focus on a specific segment(s) of the food chain (e.g., chicken slaughter) or encompass the entire food continuum (i.e., farm to fork). The risk management question will dictate the ap­ proach and spectrum of the risk assessment.

Qualitative Risk Assessments

Qualitative risk assessments express risk as a grade, a category, or a score rather than as a number as in a quantitative risk assessment. Although qualitative risk assessments lack numerical precision, they remain valu­ able and appropriate assessments. Qualitative risk as­ sessment uses a detailed and transparent narrative to describe the hazard and the risk pathway, to identify data (including uncertainty and variability), and to com­ bine information in a logical manner to provide an esti­ mate of the risk (64). Qualitative risk assessment tends to be faster and can be easier to understand than quan­ titative risk assessment, as it does not involve complex mathematical and computational calculations; however, evaluating the impact of qualitative factors in the out­ come can be problematic (64). Consequently, a qualita­ tive risk assessment may be the first iteration and the prelude (and justification) for a subsequent quantita­ tive risk assessment. Qualitative risk assessments are frequently used to assess the risk of imported products, which in many instances need to be done more quickly and with limited data. According to the FAO, qualitative risk assessment de­ scribes “A risk assessment based on data which, while forming an inadequate basis for numerical risk estima­ tions, nonetheless, when conditioned by prior expert knowledge and identification of attendant uncertainties, permits risk ranking or separation into descriptive cat­ egories of risk” (19). A generic approach to qualitative risk assessment is outlined below (32). 1. Statement of scope: Specify the risk that is to be evaluated, including the nature of the hazard and potential routes of exposure. That is, identify the question(s) to be answered by the qualitative assessment.

Manila Typesetting Company

11/08/2012 08:36PM

1028

Advanced Techniques in Food Microbiology

2. Collection of data: Obtain data to qualitatively characterize the occurrence of the event and the likely consequence of this occurrence. The more complete and representative the data, the more re­ alistic the risk assessment will be. It is often men­ tioned that qualitative risk assessments are used either when data are lacking or when the explicit mathematical relationship between the variables and the resulting risk is unknown. This is not nec­ essary true, as both qualitative and quantitative estimates of risk can be severely affected by lack of relevant data and information on the mathemati­ cal relationship between variables (64). 3. Event tree: Develop an event tree to better illus­ trate the sequence of events (flow) associated with the different risk scenarios to be evaluated. Event trees help to understand how the probabilities of different events affect the probability of the final outcome. The extent of the event tree will be dic­ tated by the statement of scope. 4. Deduce probability: Through a formal and structured process, risk assessors assign “categorical probabili­ ties” (e.g., unlikely/rare/likely) for each event of the risk scenario described in the event tree. Categorical probabilities for each event comprising the exposure scenario are then combined to arrive at a categorical estimate of the overall probability. One way to assign a product probability to the combination of events having categorical probabilities is to use a table in which the probability combinations are assigned to a summary category (an example is given in Table 41.1). Categorical probabilities and their products are not standardized and should be clearly defined in the risk assessment by the risk assessor or agency. This process needs to be well documented and trans­ parent so that it is defensible and reproducible. 5. Deduce severity: Severity may be estimated like probability, with the difference being that instead of combining a series of events weighted by prob­

ability, the overall severity is obtained by combin­ ing various aspects of the expected consequences, weighted by severity. Again, the rules for perform­ ing this operation are defined by risk assessors or agencies and should be transparent, and all terms should be carefully defined. 6. Evaluate magnitude of risk: Probability and se­ verity are combined to express the overall risk; as before, this can be done through tables that com­ bine the scores into a single measure (an example is given in Table 41.2). Assigning labels to the probabilities and severity can be subjective, lack precision and transparency, and be hard to reproduce. In addition, the risk characterization step in a qualitative risk assessment, where categorical prob­ ability and severity are logically combined to estimate the magnitude of risk, can be problematic, leading to errors of probability logic; and when there are multiple path­ ways or too much uncertainty, it can even be impossible (64). In these situations, it is more appropriate for the risk characterization to be a narrative summarizing the major stages of the risk assessment than try to make logical de­ ductions analogous to mathematical calculations (64). Qualitative risk assessments can also be entirely de­ scriptive and not include the tables of probabilities men­ tioned before. In a textual narrative, the risk assessor addresses the questions posed by the risk manager and identifies the hazard, risk pathway, uncertainties, and variabilities, providing an estimate of risk in a trans­ parent manner. Assumptions and conclusions need to be logically articulated in the text. The case study by Wooldridge (65) is a good example of a textual narra­ tive of a qualitative risk assessment.

Quantitative Risk Assessment

Quantitative risk assessment can add many layers of de­ tail and understanding to a risk assessment. These are Table 41.2  Individual and combined risk estimates for intro­

duction and transmission of highly pathogenic avian influ­ enza virus subtype H5N1 in 1-km buffer zones surrounding compartmentalized poultry farms in Thailanda

Table 41.1  Example of rules for assigning categorical

probabilities from two-component probabilities

Probability of human exposure to a given bacterium Amt of food commodity being consumed Amt of food commodity contamination

High

Medium

Low

High

H

H

M

Medium

H

M

L

Low

M

L

L

SMP_Food Microbiology_CH41.indd

1028

Pathways for introduction

Probability Exposure and Overall risk of release consequence

Wild birds

Very low

Medium

Very low

Live poultry

Very low

Very high

Very low

Free-ranging ducks

Very low

Very low

Negligible

Fertilizers

Very low

Very low

Negligible

Humans

Very low

Low

Negligible

Data from reference 34.

a

Manila Typesetting Company

11/08/2012 08:36PM

1029

41.  Microbial Risk Assessment more appropriate for the more complex and controver­ sial analyses. Quantitative risk assessments inevitably involve the use of mathematical models. These models often involve the development of a system (or systems) of equations and calculations and may include computer simulations. Mathematical models encompass a huge variety of model types. Two common model types are determinis­ tic and stochastic models. Deterministic models assume that inputs to a model are known and fixed values with no randomness. Stochastic or probabilistic models in­ corporate uncertainty and variability.

Deterministic Models

In a deterministic model, the outcomes are precisely de­ termined through known relationships among the com­ ponents of the model, the model parameters. A given set of point values (e.g., parameter mean) in a deterministic model will always produce the same output; there is no consideration of any random variation or uncertainty in the system. For example, a deterministic model might use 10% as the prevalence of a pathogen in food X, and it will not account for the fact that in reality the prevalence of this pathogen ranges from 5 to 20% according to the season. Therefore, a given input (e.g., 10%) will always produce the same amount of contaminated product in a lot of 1,000 items of food X, for example. In order to incorporate some variation or uncertainty into deterministic models, risk assessors often calcu­ late risk using different scenarios that assume different inputs such as means, medians, minimums, and/or max­ imums. For example, in our hypothetical example, we could developthree “what if” scenarios: (i) a most-likely scenario, where 10% is the mean prevalence of the pathogen of interest; (ii) a worst-case scenario, where 20% represents the maximum prevalence; and (iii) a best-case scenario, where the pathogen’s prevalence is the lowest (5%). Selecting specific values to insert into a deterministic model yields potentially useful informa­ tion for bounding risk estimates. Although comparison between these different scenarios can be helpful, it is very difficult to estimate the probability of their occurrence. Deterministic models are often easier to explain to risk managers, as they do not incorporate uncertainty and probabilities of outcomes; however, those are extremely important elements for risk managers and decision mak­ ers, who might not want to act on worst-case scenario risk estimates that might have a very low probability of occurring. A good example of a deterministic microbial risk as­ sessment model has been published (12) and is avail­

SMP_Food Microbiology_CH41.indd

1029

able for download at FoodRisk.org (http://foodrisk.org/ exclusives/sQMRA/). The model calculates the number of pathogens from retail to consumption and estimates the number of human illnesses. It was developed for use in Microsoft Excel and includes a kitchen crosscontamination module and a preparation (heating) mod­ ule as well as a dose-response relationship. It is easy to use and guides the user through the basic steps of a risk assessment, making it easier to understand how all four components of a risk assessment come together to esti­ mate the risk. In addition, with this model, one can esti­ mate the relative importance of exposure for the differ­ ent transmission routes (e.g., cross-contamination and different preparation methods) at the population level, which illustrates how risk assessments can be useful in identifying effective points of control.

Stochastic or Probabilistic Models

Most parameters such as serving size, pathogen concen­ tration and growth, storage temperatures, and storage times have a range or distribution of values. These vari­ ables are better described as distributions so a realis­ tic range and frequency of values can be represented. In instances where there is lack of data, when we do not know, for example, the prevalence of a pathogen in a particular commodity or the temperature at which a pathogen becomes inactive, we can express our beliefs about the true value using a probability distribution. In the previous example, the prevalence of the pathogen of interest in food X could be defined as a probability distribution that would represent the range of values from 5 to 20%. How to select the appropriate prob­ ability distribution to represent the uncertainty in input variables will not be covered in this chapter, but more information can be found elsewhere (58). In stochastic models, scientific data are used to generate and define probability distributions for the individual events of the risk scenario. They are then combined to determine the probability distribution of an adverse outcome. The Monte Carlo method is one of the numerical techniques for calculating probabilities. In a Monte Carlo simulation, first we define the possible distribu­ tion for each input value, and then we construct the probability distribution for the model output variables by randomly selecting values for input variables deter­ mined by their distribution and performing the opera­ tions on them according to the model’s equations, as shown in Fig. 41.2. We recalculate a model repeatedly (e.g., 10,000 times) and each time (iteration), a randomly selected value from each input probability distribution is selected for the cal­ culation. Thus, we are using all valid combinations of

Manila Typesetting Company

11/08/2012 08:36PM

1030

Advanced Techniques in Food Microbiology

Figure 41.2  Schematic representation of the Monte Carlo analysis. doi:10.1128/9781555818463.ch41f2

possible inputs to simulate all possible outcomes. When repeated many times, some output values are generated more often than the others because they result from the combinations of inputs that occur more often according to their probability distribution. The results of a Monte Carlo simulation are the likelihood of any outcome oc­ curring and the ranges of possible outcomes that could occur, that is, distributions of possible outcomes. This range of possible outcomes is one of the major advan­ tages of stochastic models because it allows risk manag­ ers to evaluate less likely events and decide whether their occurrences are acceptable or not. In deterministic modeling, each variable within a model is assigned a “best guess” estimate. Various com­ binations of each input variable are manually chosen (such as best case, worst case, and most likely case), and the results are recorded for each so-called “what if” sce­ nario. Running the Monte Carlo simulation is similar to running hundreds or thousands of individual “whatif” analyses for the model but with the added advan­ tage that the “what-if” scenarios are generated with a frequency proportional to the probability they have of occurring. Because of the complexity and large amount of calculating, tabulating, and graphing required, sto­ chastic models rely on specialized software such as R, @Risk, Analytica, and Crystal Ball. There are several stochastic risk assessment models that have been published and are available to the pub­ lic. FoodRisk.org hosts a model for L. monocytogenes in ready-to-eat foods, developed by the University of Tasmania (48) (http://foodrisk.org/exclusives/models/AU_ listeria.cfm), and also links to other models developed by governmental agencies (http://foodrisk.org/resource_types/ tools/FSIS_RA_Models/) and international organizations (http://www.mramodels.org/ESAK/ModelSummary.aspx) (18).

and scope but also on the availability of data, time, and resources. A qualitative risk assessment might be the preferred approach when there is a lack of quantitative data representing the probability and severity compo­ nents, when there are many disparate factors influencing one or both components, when more complex mathe­ matical and computational skills are not available, for screening exercises, or when the risk is familiar and frequently recurs. An initial simple deterministic model may provide insights that are adequate to support the decisions that prompted the risk managers to seek a risk assessment (56). Semiquantitative risk assessments are also an op­ tion to be considered. A good example of a spread­ sheet-based, semiquantitative risk assessment tool is Risk Ranger (47). The spreadsheet converts qualitative inputs into numerical values to calculate risk estimates and a relative risk ranking score. This tool has been used to assess and rank 10 seafood-hazard combinations (53). Another example is the Produce Risk Ranking Tool (RRT), in which a risk score is developed for pro­ duce-hazard pairs, based on a number of different risk criteria (2). The RRT is available at http://foodrisk. org/exclusives/RRT/. Whether qualitative, semiquantitative, or quantita­ tive, assumptions and approaches ought to be properly documented; uncertainties and variability have to be clearly articulated so assessments are transparent and can be reproduced. The systematic approach of the methods for the three types of risk assessments provides insights into the decision-making process being evalu­ ated and identifies data gaps that can help guide future research. A good, reliable risk assessment is directly re­ lated to the quality and availability of data as well as to the technical skills of the risk assessment team.

When To Use a Qualitative or a Quantitative Approach

VARIABILITY AND UNCERTAINTY

Whether to utilize a qualitative or a quantitative ap­ proach depends mainly on the risk manager’s question

SMP_Food Microbiology_CH41.indd

1030

Variability and uncertainty are two terms often used and discussed in risk analysis. In the area of risk analysis and modeling, variability represents the real differences

Manila Typesetting Company

11/08/2012 08:36PM

1031

41.  Microbial Risk Assessment that exist between different samples; for example, differ­ ent lots of raw materials are not all contaminated at the same level. Additional data or better measurements will not minimize or eliminate variability. Uncertainty, on the other hand, comes from sparse data, errors in sam­ pling and testing, and insufficient knowledge about the systems being modeled. Additional replications, more accurate and precise research, and more knowledge can reduce uncertainty. Most data sets have a combination of variability and uncertainty; in other words, there is uncertainty in the distributions applied to the assess­ ment of variability. It is important that uncertainty and variability be differentiated in a risk assessment model so their impact on the outcome can be determined and taken into account when selecting risk mitigation strat­ egies. In some instances, additional resources may be devoted to improve the estimates (reduce uncertainty) if that parameter is found to have an important impact on the output. One way to determine the impacts of uncertainty and variability in the design of a quantitative model is to develop a “two-dimensional risk assessment model” or “second-order model” that separates the two by assign­ ing an uncertainty distribution to the parameters of the variability distributions. Therefore, a nested set of distri­ butions describe each factor, i.e., the initial distribution is the variation, and each parameter is characterized by an uncertainty distribution. To conduct and analyze a twodimensional risk assessment, a value for each uncertainty distribution is selected according to its distribution (se­ lected randomly or by Latin hypercube sampling). With

this specific set of uncertainty values defining the varia­ tion distributions, the risk assessment is conducted, and an output distribution is obtained from the iterations. Then, another set of uncertainty values are selected to de­ fine the variation distributions, and the risk assessment is rerun to obtain another output distribution. Figure 41.3 illustrates seven outputs of a risk assessment, with each sigmoid curve being the cumulative plot of the variation in the output using one set of uncertainty values. The placement of the curve indicates the general level of risk; the 5, 50, and 95% levels can be readily observed. The median of the curves is approximately −5 log10 CFU/g. The steepness of the curve reflects the variation; a rela­ tively steep curve has low variation compared to a slop­ ing curve. The variation in the risk assessment has values ranging from approximately −8 to −2 log10 CFU/g. The spread of the different curves is an indication of the impact of the uncertainties. If the uncertainty is relatively small, the curves are close together; if they are widely spread relative to the slopes of the individual curves, then the un­ certainty is high compared to the variation. In Fig. 41.3, the uncertainty is about 3 log10 CFU/g. Differentiating between uncertainty and variability in the model is im­ portant so research can be targeted more efficiently to areas where there is a potential to improve the outcome. Understanding the degree of uncertainty is also needed to objectively discuss the degree of precaution that may need to be incorporated into subsequent risk manage­ ment decisions. It is the risk assessors’ task to address (model) uncer­ tainty in model inputs and the risk manager’s task to

Figure 41.3  Example of seven potential risk assessment outputs. doi:10.1128/9781555818463.ch41f3

SMP_Food Microbiology_CH41.indd

1031

Manila Typesetting Company

11/08/2012 08:36PM

1032

Advanced Techniques in Food Microbiology

interpret uncertainty in model outputs. To do this ef­ fectively, risk managers must understand the significant uncertainties and their implications for the risk assess­ ment and any risk management measures. In this regard, stochastic models offer an advantage over deterministic and qualitative methods because they provide a distri­ bution of outcomes that is defined by the uncertainty and variability of the input parameters as discussed above. It is typical to find that on average food is safe and processes are under control; however, it is the tails of the distribution (the less likely events) that need to be carefully evaluated by risk managers in order to im­ prove food safety.

sis have been described (68). The results, usually dis­ played in the form of tornado graphs, are a result of statistical analysis (stepwise least-square regression or rank order correlation) done using the input and out­ put data generated during the Monte Carlo iterations. An example of a tornado graph was taken from the V. parahaemolyticus in raw oysters risk assessment done by the FDA and is shown in Fig. 41.4 (20). The re­ sults from a sensitivity analysis are extremely helpful in identifying points in the systems where additional data collection is most useful (high impact of uncertainty for a parameter) and where mitigation strategies aimed at impacting risk could be most efficient (high impact of a variability parameter). For example, when analyzing results from a risk as­ sessment, it may be apparent that the average lot or serving is acceptable but a small percentage may exceed the standard, and the process would be judged unac­ ceptable. Sensitivity analyses can help identify the spe­ cific parameters that make the greatest contribution to the overall variability and/or uncertainty. Additional data can then be collected to refine the distributions, thereby improving the estimates of the outputs. For in­ stance, the warehouse storage times for different lots of a refrigerated ready-to-eat food may not be accurately known. With new information, the risk assessment can be recalculated, and the process may be determined to be acceptable without changing the process. If a portion

SENSITIVITY AND SCENARIO ANALYSIS The output of a risk assessment model, the “risk esti­ mate,” is only one piece of information that may be presented in a risk assessment report. Often, the risk man­ agers are more interested in the results from a sensitivity and scenario analysis that show the relative importance of different parameters and determine the impact on the output from changes in various input parameters. Sensitivity analysis is used to determine the degree of influence that a given input has upon the value of the output. In this chapter, we will focus on stochastic sensitivity analysis; however, less detailed methods to conduct deterministic and worst-case sensitivity analy­

Figure 41.4  Tornado graph of influential variability parameters on log10 risk of V. para­ haemolyticus illness per serving of raw oysters from the Gulf Coast (Louisiana) winter harvest (20). doi:10.1128/9781555818463.ch41f4

SMP_Food Microbiology_CH41.indd

1032

Manila Typesetting Company

11/08/2012 08:36PM

1033

41.  Microbial Risk Assessment of the process lots is still unacceptable, then a sensitiv­ ity analysis for variation identifies the parameter(s) that significantly contributes to the higher-risk portions of the lot. When production schedules and warehouse in­ ventory control are not optimized, for example, differ­ ent lots may have different storage times. Better controls could reduce the variation in storage times, particularly avoiding lots held for extensive periods of time. Reducing variation would likely improve the consistency of the sensory and nutritional qualities of the product as well as the microbial consistency. With variation reduced, the risk assessment is rerun. If the process still has an unacceptable portion of trials that exceed the standard, the process must be changed to lower the absolute value of the output distribution. In this example, production and warehousing schedules would need to be changed so storage times are reduced for all lots. By this iterative process of risk assessments and sensitivity analyses, the design of a food process can be improved until it meets the desired safety standard. When interpreting results from a sensitivity analysis, it is important to recognize that models are a simplifica­ tion of biological processes. When model uncertainty is high (i.e., the biological process is not well known, or the model lacks an important step), this representation of reality might be compromised, and therefore, results from the sensitivity analysis will not correlate well with the actual biological process and can be misleading. Model uncertainty also affects the magnitude of the out­ put to the extent that the model could be invalid. Scenario analysis is also another important tool that risk assessors can use to explore different “what-if” sce­ narios and help in decision making. Different scenarios can be analyzed by changing the input parameter values

and observing changes in output distributions. These scenarios can be used to estimate the impact of regula­ tory or process changes or to improve the understanding of how various factors in a complex process interact. As examples, the impact of processing changes on pub­ lic health have been estimated by scenario analysis for Salmonella in eggs (62) and L. monocytogenes in readyto-eat foods (6). Scenario analysis can also be a good method to assess other types of uncertainty, such as model uncertainty and systems uncertainty, which can­ not be evaluated through Monte Carlo simulation and are not associated with whether imperfect or insufficient data were collected and used to characterize the param­ eters (parameter uncertainty). By comparing the results from the different scenarios, one can gain knowledge on the robustness and confidence of the model (27). An example provided by Havelaar et al. (27) evaluated dif­ ferent dose-response relationships for E.coli O157:H7 in steak tartare and its impact on the number of illnesses in children under 15 years old. They concluded that the exposure model was quite robust, but the estimate of the public health risks was highly uncertain, and there­ fore, the model could be useful for decisions on reducing exposure but not to decide on an ALOP.

RISK ASSESSMENT AND RISK MANAGEMENT METRICS To integrate risk management practices at the produc­ tion level and risk assessment concepts, a series of terms have been developed (8, 14, 24, 52, 57, 60, 67). Figure 41.5 illustrates a simple food process in which an indi­ vidual serving starts as raw ingredients with a certain frequency and level of contamination, is pasteurized to

Figure 41.5  Example of the application of risk management metrics in a hypothetical food process. Trans., transportation. doi:10.1128/9781555818463.ch41f5

SMP_Food Microbiology_CH41.indd

1033

Manila Typesetting Company

11/08/2012 08:36PM

1034

Advanced Techniques in Food Microbiology

reduce pathogen numbers, is stored under conditions in which surviving pathogens may grow, is consumed, and may result in illness. The ALOP is the public health goal for the food that the process is designed to meet. ALOPs can be expressed as the number of cases per year in a country or as the probability of illness per serving for a specified popula­ tion. The ALOP is set via a societal process by risk man­ agers; it is a level that the public will tolerate in order to have the foods it desires. However, it is necessary to communicate to the public that risks cannot be zero and ALOPs must be achievable by the food industry. ALOP is useful as a target for public health policy, but it is very difficult to use it as a way to guide implementation of control measures throughout the food production chain. To reduce this gap, the International Commission on Microbiological Specifications suggested the term “food safety objective” (FSO), which was later adopted by the Codex Alimentarius Commission for Food Hygiene. FSO is defined as the maximum frequency and/or con­ centration of a microbiological hazard in a food at the time of consumption that meets the ALOP. The doseresponse relationship for the selected susceptible human population used in a microbial risk assessment will help determine the numbers of a pathogen consumed with a serving that has a risk equal to or less than the ALOP. This maximum number of pathogens and/or the frequency of contamination is the FSO, and this be­ comes the microbial target that a food process should be designed and controlled to achieve. Factoring in the serving size will convert the FSO into a CFU per gram basis. Because it is not feasible to determine or regu­ late microbial populations at the time of consumption, the maximum numbers of the pathogen can be specified at earlier points in the processing chain. These are the performance objectives (PO). There can be several POs in a food process, and the FSO can be viewed as the PO at consumption. For foods that support growth, al­ lowances for growth between manufacture or retail and consumption are necessary. This requires designation of appropriate home and food service storage temperatures and times. These need to be based on reasonable levels of abuse, as most foods that allow pathogen growth would not have counts below the FSO after extreme storage temperatures and/or times. The selection of an FSO (and corresponding ALOP) then allows calculation of microbial levels at various steps in the food process. A more stringent FSO will require a comparable reduc­ tion in initial contamination, a decrease in growth, or an increase in an inactivation step. For an inactivation pro­ cess step, the decrease that should be achieved is termed the performance criterion. Performance criteria could

also be established for a storage period in which a 1-log increase, for example, would be the maximum allowed. If a reduction is necessary, several inactivation processes could achieve a specified performance criterion. Thermal inactivation is the traditional inactivation process, but high pressure, UV light, or pulsed electrical fields are ex­ amples of other processes that could be employed with juices, for example. There are usually multiple combina­ tions of parameters for each process that could achieve the performance criteria, e.g., different time-temperature combinations that achieve a 7-log reduction by thermal inactivation. The specific operating parameters chosen to achieve the performance criteria are the process cri­ teria. These parameters then become the critical control points in the HACCP plan that achieves control over the safety of the process. The application of risk assessment techniques to better link microbiological criteria to pub­ lic health goals is an area that is actively being pursued by regulatory agencies, international intergovernmental organizations (i.e., Codex Alimentarius), and the food industry. The link between a deterministic risk assessment and FSO is pretty straightforward as described above but might not be entirely realistic because determinis­ tic models do not account for uncertainty and variabil­ ity, which are inevitable when conducting a microbial risk assessment. These limitations and how to fine-tune FSOs with quantitative microbial risk assessment have been previously discussed by Havelaar et al. (27) and are not addressed in this chapter.

SMP_Food Microbiology_CH41.indd

1034

FINAL CONSIDERATIONS Validation and documentation are two elements in risk assessment that are often overlooked during the risk as­ sessment process but nonetheless are extremely impor­ tant for its credibility and robustness. Risk assessments should be validated using different data sets. Although strongly desired, validation might not be practical in many instances. Ideally, models would be validated using high-quality, reliable data sets that were not used in the model. Nonetheless, as new data are produced, models can be validated and reevalu­ ated. In fact, the risk analysis process as a whole should be an iterative process, being reevaluated periodically as data gaps diminish, new interventions are developed, and more information is made available. Independent of the type of risk assessment, it should be well documented so others are able to follow the steps, to understand the assumptions and methodology, and to reproduce the results if necessary. By ensuring reproducibility, one ensures transparency. Models de­

Manila Typesetting Company

11/08/2012 08:36PM

1035

41.  Microbial Risk Assessment veloped using spreadsheets are often difficult to follow; however, efforts to address this weakness must be made when building the risk assessment model. Assumptions, uncertainty, and variability in the data and models and how those were treated as well as their implications must be clearly described and articulated in the final re­ port. In addition, the data sources utilized must be ref­ erenced and potential biases addressed. An interpretive summary or nontechnical report may also need to be written to address risk managers and other stakeholders who may not have the scientific or mathematical back­ ground to fully understand the results of the risk assess­ ment. Highly technical information should be described in appendices, since extremely detailed information of the methods used may reduce the understanding of the risk assessment report. Biases are not as commonly discussed in microbial risk assessments as issues with uncertainty and variabil­ ity are. More specifically, biases in prevalence estimates attributable to the use of imperfect tests are rarely dis­ cussed or incorporated into food safety microbial risk assessments. It may be due to the fact that diagnostic test sensitivity and specificity are not always available to be able to adjust the test-based (apparent) prevalence to the true prevalence. For an alternative view and critique of methods to adjust apparent prevalence to true preva­ lence, see Gardner (23). Many of the data parameters that are needed to model microbial growth, a food process, and/or consumer be­ havior have not been measured in a survey or scientific study. Examples might include the time during which a product is being transported, the temperature in a refrig­ erated processing facility, and many consumer food han­ dling practices. For these parameters, expert opinion will be needed. This involves soliciting input on parameters from authorities on such topics. Various protocols for ask­ ing questions and estimating the parameters’ distribution can be used to obtain as unbiased and accurate an esti­ mate as possible (55, 58). For example, Bayesian methods have been used to estimate the uncertainty and variability in the prevalence of Bacillus cereus spores (39) and L. monocytogenes (10) in specific food-related environments or processing conditions, as well as diagnostic test sensi­ tivity and specificity to adjust apparent prevalence to true prevalence (23). Expert elicitation is a valuable resource when conducting qualitative or quantitative risk assess­ ment; however, elicitations must be carefully designed to avoid biased opinions and to ensure the right balance of expertise and proper analysis of the results (64). With the increased need for science and risk-based ap­ proaches to improve food safety and public health, risk assessments methods have been extensively cited as an ap­

SMP_Food Microbiology_CH41.indd

1035

propriate approach to address these new challenges. The systematic and structured approach of risk assessments allows a better understanding of the food production sys­ tems, effectively identifies data gaps and research needs, and most of all, is a powerful tool for decision makers. We acknowledge the Joint Institute for Food Safety and Applied Nutrition (JIFSAN) at the University of Maryland and JIFSAN instructors Charles Yoe and Greg Paoli for al­ lowing us to use part of the JIFSAN risk assessment training materials in this chapter. At the time this chapter was written, J.M.R. was at the Joint Institute for Food Safety and Applied Nutrition, University of Maryland, College Park.

References   1. Abt, E., J. V. Rodricks, J. I. Levy, L. Zeise, and T. A. Burke. 2010. Science and decisions: advancing risk assess­ ment. Risk Anal. 30:1028–1036.   2. Anderson, M., L. Jaykus, S. Beaulie, and S. Dennis. 2011. Pathogen-produce pair attribution risk ranking tool to prioritize fresh produce commodity and pathogen combi­ nations for further evaluation (P3ARRT). Food Control. 22:1865–1872.   3. Buchanan, R. L., and S. B. Dennis. 2001. Microbial risk assessment—a tool for regulatory decision making. J. Assoc. Food Drug Officials 65:36–46.   4. Buchanan, R. L., S. B. Dennis, and M. Miliotis. 2004. Initiating and managing risk assessments within a risk analysis framework: FDA/CFSAN’s practical approach. J. Food Prot. 67:2058–2062.   5. Buchanan, R. L., A. H. Havelaar, M. A. Smith, R. C. Whiting, and E. Julien. 2009. The key events doseresponse framework: its potential for application to food­ borne pathogenic microorganism. Crit. Rev. Food Sci.

Nutr. 49:718–728.

  6. Carrington, C. D., S. B. Dennis, R. C. Whiting, and R. L. Buchanan. 2004. Putting a risk assessment model to work: Listeria monocytogenes ‘what if’ scenarios. J. Assoc. Food Drug Officials 68:5–19.   7. Codex Alimentarius Commission. 2010. Working prin­ ciples for risk analysis for application in the framework of the Codex Alimentarius, p. 85–93. In Codex Alimentarius Procedural Manual, 19th ed. Secretariat of the Joint FAO/ WHO Food Standards Programme, FAO, Rome, Italy.   8. Cole, M. B., and R. B. Tompkin. 2005. Microbiological performance objectives and criteria, p. 673–695. In J. N. Sofos (ed.), Improving the Safety of Fresh Meat. Woodhead Publishing, Cambridge, United Kingdom.   9. D’Aoust, J. 1977. Salmonella and the chocolate industry. J. Food Prot. 40:718–727. 10. Delignette-Muller, M. L., M. Cornu, R. Pouillot, and J.-B. Denis. 2006. Use of Bayesian modeling in risk assessment: application to growth of Listeria monocytogenes and food flora in cold-smoked salmon. Int. J. Food Microbiol. 106:195–208. 11. Dennis, S. B., J. Kause, M. Losikoff, D. L. Engeljohn, and R. L. Buchanan. 2008. Using risk analysis for microbial food safety regulatory decision making, p. 137–176. In

Manila Typesetting Company

11/08/2012 08:36PM

1036

Advanced Techniques in Food Microbiology

D. W. Schaffner (ed.), Microbial Risk Analysis of Foods. ASM Press, Washington, DC. 12. Evers, E. G., and J. E. Chardon. 2010. A swift quan­ titative microbiological risk assessment (sQMRA) tool. Food Control 21:319–330. 13. FAO/WHO. 1997. Risk Management and Food Safety. Report of a Joint FAO/WHO Consultation. FAO Food and Nutrition Paper No. 65. FAO, Rome, Italy. 14. FAO/WHO. 2002. Principles and Guidelines for Incorpo­ rating Microbiological Risk Assessment in the Development of Food Safety Standards, Guidelines and Related Texts. Report of a Joint FAO/WHO Consultation, Kiel, Germany. FAO, Rome, Italy. 15. FAO/WHO. 2002. Risk Assessment of Salmonella in Eggs and Broiler Chickens. Microbiological Risk Assessment Series No. 2. FAO, Rome, Italy. 16. FAO/WHO. 2003. Hazard Characterization for Pathogens in Food and Water. Microbiological Risk Assessment Series No. 3. FAO, Rome, Italy. 17. FAO/WHO. 2004. Risk Assessment of Listeria mono­ cytogenes in Ready-To-Eat Foods. Microbiological Risk Assessment Series No. 5. FAO, Rome, Italy. 18. FAO/WHO. 2004. Enterobacter sakazakii and Other Microorganisms in Powdered Infant Formula. Microbiolog­ ical Risk Assessment Series No. 6. FAO, Rome, Italy. 19. FAO/WHO. 2006. Food Safety Risk Analysis: a Guide for National Food Safety Authorities. FAO, Rome, Italy. 20. FDA. 2005. Quantitative Risk Assessment on the Public Health Impact of Pathogenic Vibrio parahaemolyticus in Raw Oysters. Food and Drug Administration, Washington, DC. http://www.fda.gov/Food/ScienceResearch/Research­ Areas/RiskAssessmentSafetyAssessment/ucm050421.htm 21. FDA/Food Safety and Inspection Service. 2003. Quantitative Assessment of Relative Risk to Public Health from Foodborne Listeria monocytogenes among Selected Categories of ReadyTo-Eat Foods. Food and Drug Administration, Washington, DC. www.foodsafety.gov/~dms/lmr2-toc.html 22. Federal Register. 2009. Notice. June 9, 2009. Fed. Regist. 74:7276–2727. 23. Gardner, I. A. 2004. An epidemiologic critique of cur­ rent microbial risk assessment practices: the importance of prevalence and test accuracy data. J. Food. Prot. 67:2000–2007. 24. Gorris, L. G. M. 2005. Food safety objective: an in­ tegral part of food chain management. Food Control 16:801–809. 25. Haas, C. N., A. Thayyar-Madabusi, J. B. Rose, and C. P. Gerba. 2000. Development of a dose-response relation­ ship for Escherichia coli O157:H7. Int. J. Food Microbiol. 56:153–159. 26. Hallman, W. K. 2008. Communicating about micro­ bial risks in foods, p. 205–262. In D. W. Schaffner (ed.), Microbial Risk Analysis of Foods. ASM Press, Washington, DC. 27. Havelaar, A. H., M. J. Nauta, and J. T. Jansen. 2004. Fine-tuning food safety objectives and risk assessment. Int. J. Food Microbiol. 93:11–29.

28. Holcomb, D. L., M. A. Smith, G. O. Ware, Y.-C. Hung, R. E. Brackett, and M. P. Doyle. 1999. Comparison of six dose-response models for use with food-borne pathogens. Risk Anal. 19:1091–1100. 29. Hoornstra, E., and S. Notermans. 2001. Quantitative microbiological risk assessment. Int. J. Food Microbiol. 66:21–29. 30. Horton, L. R. 2001. Risk analysis and the law: inter­ national law, the World Trade Organization, Codex Alimentarius and national legislation. Food Additives Contaminants 18:1057–1067. 31. Ivanek, R., Y. T. Gröhn, L. W. Tauer, and M. Wiedmann. 2004. The cost and benefit of Listeria monocytogenes food safety measures. Crit. Rev. Food Sci. Nutr. 44:513–523. 32. Joint Institute for Food Safety and Applied Nutrition. 2009. Quantitative Risk Assessment Methods. Joint Institute for Food Safety and Applied Nutrition, College Park, MD. 33. Jones, R. C., S. I. Gerber, P. S. Diaz, L. L. Williams, S. B. Dennis, E. S. Parish, and W. S. Paul. 2004. Intensive investigation of bacterial foodborne disease outbreaks: proposed guidelines and tools for the collection of doseresponse data by local health departments. J. Food Prot. 67:616–623. 34. Kasemsuwan, S., C. Poolkhet, T. Patanasatienkul, N. Buameetoop, M. Watanakul, K. Chanachai, K. Wongsatha­ pornchai, R. Métras, C. Marcé, A. Prakarnkamanant, J. Otte, and D. Pfeiffer. 2009. Qualitative risk assessment of the risk of introduction and transmission of H5N1 HPAI virus for 1-km buffer zones surrounding com­ partmentalised poultry farms in Thailand. http://www. hpai-research.net/docs/Working_papers/wp7_2009.pdf. Accessed April 2011. 35. Kothary, M. H., and U. S. Babu. 2001. Infective dose of foodborne pathogens in volunteers: a review. J. Food Saf. 21:49–73. 36. Latimer, H. K., L. Jaykus, R. A. Morales, P. Cowen, and D. Crawford-Brown. 2001. A weighted composite doseresponse model for human salmonellosis. Risk Anal. 21:295–305. 37. Lubran, M. B., R. Pouillot, S. Bohm, E. M. Calvey, J. Meng, and S. Dennis. 2010. Observational study of food safety practices in retail deli departments. J. Food Prot. 73:1849–1857. 38. Lyytikäinen, O., T. Autio, R. Maijala, P. Ruutu, T. Honkanen-Buzalski, M. Miettinen, M. Hatakka, J. Mikkola, V.-J. Anttila, T. Johansson, and L. Rantala. 2000. An outbreak of Listeria monocytogenes sero­ type 3a infections from butter in Finland. J. Infect. Dis. 181:1838–1841. 39. Malakar, P. K., G. C. Barker, and M. W. Peck. 2004. Modeling the prevalence of Bacillus cereus spores dur­ ing the production of a cooked chilled vegetable product. J. Food Prot. 67:939–946. 40. McKellar, R. C., and X. Lu. 2004. Primary models, p. 21– 62. In R. C. McKellar and X. Lu (ed.), Modeling Microbial Responses in Food. CRC Press, Boca Raton, FL. 41. Moon, H., J. J. Chen, D. W. Gaylor, and R. L. Kodell. 2004. A comparison of microbial dose-response models fitted to human data. Regul. Toxicol. Pharmacol. 40:177–184.

SMP_Food Microbiology_CH41.indd

1036

Manila Typesetting Company

11/08/2012 08:36PM

1037

41.  Microbial Risk Assessment 42. Nauta, M. J. 2002. Modelling bacterial growth in quanti­ tative microbial risk assessment: is it possible? Int. J. Food Microbiol. 73:297–304. 43. Oscar, T. 2004. Dose-response model for 13 strains of Salmonella. Risk Anal. 24:41–49. 44. Powell, D. A. 2000. Food safety and the consumer— perils of poor risk communication. Can. J. Anim. Sci. 80:393–404. 45. Powell, M. R., E. Ebel, W. Schlosser, M. Walderhaug, and J. Kause. 2000. Dose-response envelope for Escherichia coli O157:H7. Quant. Microbiol. 2:141–163. 46. Robertson, L. J., J. D. Greig, B. Gjerde, and A. Fazil. 2005. The potential for acquiring cryptosporidiosis or giardiosis from consumption of mung bean sprouts in Norway: a preliminary step-wise risk assessment. Int. J. Food Microbiol. 98:291–300. 47. Ross, T., and J. Sumner. 2002. A simple, spreadsheetbased, food safety risk assessment tool. Int. J. Food Microbiol. 77:39–53. 48. Ross, T., S. Rasmussen, A. Fazil, G. Paoli, and J. Sumner. 2009. Quantitative risk assessment of Listeria monocytogenes in ready-to-eat meats in Australia. Int. J. Food Microbiol. 131:128–137. 49. Schaffner, D. W. 2004. Models—what comes after the next generation?, p. 303–312. In R. C. McKellar and X. Lu (ed.), Modeling Microbial Responses in Food. CRC Press, Boca Raton, FL. 50. Serra, J. A., E. Domenech, I. Escriche, and S. Martorell. 1999. Risk assessment and critical control points from the production perspective. Int. J. Food Microbiol. 46:9–26. 51. Strachan, N. J. C., M. P. Doyle, F. Kasuga, O. Rotariu, and I. D. Ogden. 2005. Dose response modeling of Escherichia coli O157 incorporating data from foodborne and envi­ ronmental outbreaks. Int. J. Food Microbiol. 103:35–47. 52. Stringer, M. 2005. Food safety objectives—role in mi­ crobiological food safety management. Food Control 16:775–794. 53. Sumner, J., and T. Ross. 2002. A semi-quantitative seafood safety risk assessment. Int. J. Food Microbiol. 77:55–59. 54. Teunis, P. F. M., and A. H. Havelaar. 2002. Risk as­ sessment for protozoan parasites. Int. Biodeterioration Biodegradation 50:185–193. 55. Van der Fels-Klerx, H. J., R. M. Cooke, M. N. Nauta, L. H. Goossens, and A. H. Havelaar. 2005. A structured expert judgment study for a model of Campylobacter transmission during broiler-chicken processing. Risk Anal. 25:109–124. 56. van Gerwen, S. J. C., M. C. te Giffel, K. van ‘t Riet, R. R. Beumer, and M. H. Zwietering. 2000. Stepwise quantita­

SMP_Food Microbiology_CH41.indd

1037

tive risk assessment as a tool for characterization of micro­ biological food safety. J. Appl. Microbiol. 88:938–951. 57. van Schothorst, M. 2005. A proposed framework for the use of FSOs. Food Control 16:811–816. 58. Vose, D. 2008. Risk Analysis, 3rd ed. Wiley & Sons, New York, NY. 59. Voysey, P. A., and M. Brown. 2000. Microbiological risk assessment: a new approach to food safety control. Int. J. Food Microbiol. 58:173–179. 60. Walls, I., and R. L. Buchanan. 2005. Use of food safety objectives as a tool for reducing foodborne listeriosis. Food Control 16:795–799. 61. Whiting, R. C., and R. L. Buchanan. 2007. Progress in micro­ bial modeling and risk assessment, p. 953–969. In M. P. Doyle and L. R. Beuchat (ed.), Food Microbiology: Fundamentals and Frontiers, 3rd ed. ASM Press, Washington, DC. 62. Whiting, R. C., A. Hogue, W. D. Schlosser, E. D. Ebel, R. Morales, A. Baker, and R. McDowell. 2000. A risk as­ sessment for Salmonella Enteritidis in shell eggs. J. Food Sci. 65:864–869. 63. Williams, R. A., and K. M. Thompson. 2004. Integrated analysis: combining risk and economic assessments while preserving the separation of powers. Risk Anal. 24:1613–1623. 64. Wooldridge, M.  2008. Qualitative risk assessment, p. 1–28.  In D. W. Schaffner (ed.), Microbial Risk Analysis of Foods. ASM Press, Washington, DC. 65. Wooldridge, M.  1999. Qualitative risk assessment for antibiotic resistance. Case study: Salmonella ty­ phimurium and the quinolone/fluoroquinolone class of antimicrobials (annex IV). In The European Agency for the Evaluation of Medicinal Products (EMEA) (ed.), Antibiotic Resistance in the European Union Associated with the Therapeutic Use of Veterinary Medicines: Report and Qualitative Risk Assessment by the Committee for Veterinary Medicinal Products. EMEA, London, United Kingdom. 66. World Trade Organization. 2010. The WTO agreement series: sanitary and phytosanitary agreement. http://www. wto.org/english/res_e/booksp_e/agrmntseries4_sps_e.pdf. Accessed January 2010. 67. Zwietering, M. H., C. M. Stewart, R. C. Whiting, and the International Commission on Microbiological Specifications for Foods. 2010.  Validation of control mea­ sures in a food chain using the FSO concept.  Food Control 21:1716–1722. 68. Zwietering, M. H., and S. J. C. van Gerwen. 2000. Sensitivity analysis in quantitative microbial risk assess­ ment. Int. J. Food Microbiol. 58:213–221.

Manila Typesetting Company

11/08/2012 08:36PM

SMP_Food Microbiology_CH41.indd

1038

Manila Typesetting Company

11/08/2012 08:36PM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch42

Robert L. Buchanan E. Noelia Williams

Hazard Analysis and Critical Control Point System: Use in Managing Microbiological Food Safety Risks The wide array of potentially hazardous microorganisms that can occur in the cornucopia of foods consumed by humans represents a significant challenge that has to be met by food producers, manufacturers, distributors, marketers, and consumers on a daily basis if consumers are going to have predictable access to a safe, nutritious, affordable, and dependable food supply. Managing the microbiological risks associated with foods requires an ongoing balancing of the need to protect public health, consumers’ food preferences, and the nonsterile nature of the food supply. The complexity of food production, processing, and preparation requires a systematic approach that simultaneously provides a common framework for managing microbial food safety risks and the flexibility and practicality needed to deal with the diversity of hazards, ingredients, and technologies. This is achieved almost globally with the application of two systems: good hygienic practices (GHPs) and hazard analysis critical control point (HACCP) (15, 29). The traditional approach to managing food safety GHPs is a compendium of “best practices” that provide

42

general guidance on practices and protocols that should be incorporated in the production, processing, distribution, marketing, and preparation of foods. This can take the form of consensus guidance developed by the different segments of the food industry (e.g., Industry Handbook for Safe Processing of Nuts [22]), consensus standards developed through public/private deliberations (e.g., Conference on Food Protection input into the U.S. Food Code), and requirements developed by national food safety regulatory bodies such as the FDA good manufacturing practices (GMPs) (21CFR110) (19). These are the general practices and procedures expected of essentially any food facility in the areas of sanitary conditions and practices, storage conditions, personnel hygiene, facilities’ construction, pest control, etc. These practices may be sector specific or commodity specific such as GMPs for food processing or good agricultural practices for primary production. On an international basis, the Codex Alimentarius Commission provides a general guideline for GHPs through its International Code of Hygienic Practice (15) as well as a series of commodity-specific and pathogen-specific

Robert L. Buchanan, Center for Food Safety and Security Systems, College of Agricultural and Natural Resources, University of Maryland, College Park, MD 20742. E. Noelia Williams, Department of Nutrition and Food Science, College of Agricultural and Natural Resources, University of Maryland, College Park, MD 20742.

1039

SMP_Food Microbiology_CH42.indd

1039

Manila Typesetting Company

11/08/2012 07:22AM

1040

Advanced Techniques in Food Microbiology

codes of hygienic practice, such as the ones for fresh produce (11) and Listeria monocytogenes (13), respectively. The ability to successfully implement GHPs is considered fundamental to the safe production and processing of foods and a prerequisite to the subsequent implementation of HACCP. This will be discussed in more detail later. Long experience with the production and processing of foods has taught that GHPs may not be sufficient to effectively manage food safety risks. This, in part, reflects the complexity of food systems and the general nature of GHPs. GHP programs address the large number of parameters and conditions that can impact product safety but do not prioritize them. Furthermore, because they are controls common to almost all food facilities, they are generally not applied in a facility-specific manner. However, the failures and deviations that lead to a loss of control resulting in food safety issues are not random events but instead are predominately associated with a limited number of specific steps or practices that are often unique to specific food facilities or products. Thus, in addition to GHPs there is a need for a tool that (i) helps food manufacturers identify the steps in the production and marketing of foods that have the greatest likelihood of being associated with food safety incidents if they are not adequately controlled and (ii) provides an extra degree of care to ensure that these sensitive steps are adequately controlled. This is the role that HACCP was designed to address.

were influenced by several aerospace engineering design criteria, particularly the U.S. Air Force Manual 80-3, “Handbook of Instructions for Aerospace Personnel and Subsystem Designers” (25). Current HACCP concepts reflect much of this early “systems thinking,” but these earlier criteria had to be adapted and simplified to meet the needs and capabilities of the food industry. The Pillsbury Company met this challenge and developed the initial framework that allowed them to fulfill the systems management approach required to meet the microbiological criteria established by the space foods program. The HACCP concepts that emerged and were ultimately expanded to include all types of foods and hazards focused on a more qualitative approach to hazard identification, identifying CCPs, and establishing critical limits (CLs). However, at its heart the basic approach is FMEA, i.e., identifying how the food safety system is likely to fail and then devising ways to ensure that such failures are prevented. These concepts formed the foundation for development of the HACCP system for food safety control. The Pillsbury Company began to adopt this CCP approach internally for the control of microbiological, chemical, and physical hazards in their commercial products, particularly after they experienced problems related to the presence of glass in farina (33, 45). Dr. Howard Bauman presented the HACCP concept to the public in 1971 at the first National Conference on Food Protection (30). This presentation included the introduction of three principles: (i) conduct a hazard analysis, (ii) identify critical control points, and (iii) establish monitoring procedures (35). This stimulated a substantial amount of interest on the part of the FDA, which was dealing with outbreaks of botulism in canned food at the time. The FDA invited the Pillsbury Company to present a training session on the use of this new approach, and it appears that the first use of the term HACCP was at that training session (33). The influence of HACCP concepts is clearly evident in the FDA requirements for the production of low-acid canned foods (18). Through experience with this new management system, Pillsbury subsequently adopted two additional principles: (iv) establish corrective actions to be taken when deviations occur at a CCP and (v) establish CLs for the required level of control that should be achieved at each CCP (36). Two more principles were added in 1997 by the National Advisory Committee on Microbiological Criteria for Foods (NACMCF) (29): (vi) establish procedures for verification to confirm that the HACCP system is working effectively and (vii) prepare documentation concerning all procedures and records appropriate to these principles and their application.

ORIGINS OF HACCP As part of a national effort to achieve manned space flight, scientists from the U.S. Army Natick Laboratories and the National Aeronautics and Space Administration joined forces with their contractor, the Pillsbury Company, to develop foods that could nutritionally and safely sustain the astronauts (37). At that time, food safety and quality systems were generally based on end product testing. However, the limitations of end product testing, such as the need for a significant quantity of samples to provide a high degree of assurance, and the reactive nature of this approach to hazard control were unlikely to achieve the 100% assurance goal for all foods produced by the project. It became evident that a different approach based on designing safety into the formulation, manufacturing, and packaging of the products would be needed to achieve this goal. The influence of systems engineering approaches, such as failure mode and effect analysis (FMEA) and “HAZOP” in the development of HACCP is readily apparent (37, 45). This is not surprising, since all activities associated with the manned space flight programs

SMP_Food Microbiology_CH42.indd

1040

Manila Typesetting Company

11/08/2012 07:22AM

1041

42.  Using HACCP in Food Safety Risk Management The introduction of HACCP in the early 1970s generated a great deal of interest but only limited application by the industry. This, in part, reflects a lack of practical instructional materials and the propensity to identify an excessive number of CCPs. This resulted in plans that were too complex to be practical. However, during the late 1980s and early 1990s, HACCP began to reemerge as an important risk management system for foods (5). The approach received a substantial reinvigoration when a Subcommittee of the Food Protection Committee of the National Academy of Sciences issued a report on microbiological criteria that included an endorsement of HACCP (27). This report also resulted in the formation of the NACMCF in 1988. One of the early activities of NACMCF was the development of materials to promote a better understanding and adoption of HACCP. NACMCF initial guidance documents were released in 1989, 1992, and 1994 and then updated in 1997 (29). Internationally, the utilization of HACCP was also advanced by the publication of a book by the International Commission on Microbiological Specification for Foods (ICMSF) on the application of HACCP (23). The concepts put forth by both NACMCF and ICMSF were captured by Codex Alimentarius when the Codex Alimentarius Commission updated their basic code of hygienic practice by the inclusion of a HACCP annex in 2003 (15). The adoption of HACCP in the United States was greatly accelerated when regulatory agencies began to move from traditional inspectional approaches to regulatory systems based on the adoption of HACCP by different segments of the food industry. The NACMCF played a significant initiating role in its 1990 report on safeguarding ready-to-eat meat and poultry products with extended refrigerated shelf lives after the emergence of L. monocytogenes as a foodborne pathogen (28). Subsequent adoption of HACCP-based regulations for seafood and juices by the FDA and for meat, poultry, and egg products by the U.S. Department of Agriculture/ Food Safety and Inspection Service (USDA/FSIS) ensured that HACCP became an integral part of U.S. food safety systems. This is expected to expand to virtually universal coverage of foods with the pending implementation of the Food Safety Modernization Act of 2011. While the Act does not specifically specify HACCP, the description in the Act indicates a system based on a hazard analysis and subsequent implementation of preventive controls, a system that appears to have all the key components of a HACCP system. Paralleling these U.S. Government initiatives was an international movement towards a food safety assurance and regulatory scheme based on the principles

SMP_Food Microbiology_CH42.indd

1041

of HACCP. For example, in April 2004, the European Union adopted several new regulations on the hygiene of foods, including 852/2004/EC, which mandates that all food business operators implement procedures based on HACCP principles. While the supporting guidance document (17) underlying 852/2004/EC expresses concepts similar to those in the 1997 NACMCF document (29), there are some differences in specific approaches and recommendations. Other governmental authorities across the globe, such as those of Canada (9), Australia, and Japan, have adopted or are adopting HACCP-based food safety control systems. The FAO/WHO, both directly and through Codex Alimentarius, have promoted the adoption of HACCP-based food safety systems by developing countries.

HACCP OVERVIEW The systematic approach envisioned with HACCP involves four phases: (i) preliminary activities, (ii) the identification of hazards that are “reasonably likely” to occur during the manufacture and distribution of a food if not controlled (i.e., hazard analysis), (iii) the identification of the parameters, conditions, or circumstances that must be controlled to prevent the hazard from occurring (i.e., identification of CCPs and establishment of CLs), and (iv) the implementation of protocols and procedures to ensure that the controls are functioning as intended (i.e., monitoring, corrective action verification, and documentation). Within these four phases, adherence to seven principles is recognized as being needed to achieve consistent control (29). 1. 2. 3. 4. 5. 6. 7.  

Conduct a hazard analysis Determine the CCPs Establish CLs Establish monitoring procedures Establish corrective actions Establish verification procedures Establish record-keeping and documentation procedures

It is worth noting that these seven principles represent an evolving conceptual framework. Principles 1, 2, and 4 were part of the initial program formulated by Pillsbury (35). Principles 3 and 5 were subsequently added by Pillsbury as they gained more experience with the system (37). Finally, principles 6 and 7 were added by NACMCF (29). While still a subject of debate, the HACCP process is clearly a risk analysis approach to managing food safety risk. The hazard analysis can be appropriately viewed as a qualitative risk assessment (or possibly a risk profile),

Manila Typesetting Company

11/08/2012 07:22AM

1042

Advanced Techniques in Food Microbiology

though there are some who have argued that a hazard analysis is distinctly different from a risk assessment (see below). The remaining six principles are clearly risk management-related activities. The principles adopted by the NACMCF (29) will be the primary focus of the discussion in this chapter. As mentioned above, there are variations in the principles outlined by other groups (e.g., ICMSF [24]), nations, and official intragovernmental bodies (15). Furthermore, there is substantial variation in the interpretation of the principles by different individuals and organizations. This has led to substantial variation in the resultant guidance and regulatory programs that are based on the implementation of HACCP programs. It has also led to substantial discussion about the applicability of HACCP in different segments of the farm-to-table continuum (see below). However, there is surprising agreement on the basic concepts, which is reflected in the almost global acceptance of HACCP as the “gold standard” for food safety risk management. The overarching message to users is that HACCP should not focus too tightly on doctrine but instead should focus on the application of FMEA concepts within a commonsense framework. However, this may mean that when employing HACCP concepts to segments outside the food manufacturing environment the application of HACCP principles may need to be modified somewhat. This subject is discussed further later in the chapter. While the systematic approach to managing risks envisioned in HACCP could be potentially used to address both safety and quality issues, it has been strongly recommended by multiple organizations and governments that implementation of HACCP in a food facility be limited to safety, with an emphasis on significant hazards. This “food safety only” focus has been adopted by U.S. regulatory agencies in their use of HACCP as a regulatory framework (e.g., seafood and juices by FDA and meat and poultry by USDA). This focus on a “food safety only” philosophy for HACCP has also been adopted internationally. A possible international exception is the Codex’s guidance document on HACCP (15) indicating that the focus should be on ensuring the safety and suitability of foods for human consumption. However, in practice, HACCP is adopted through the development of international codes of practice that overwhelmingly focus on safety. Multiple reasons for restricting this risk management tool to safety concerns have been described by various experts and advisory panels. The first is that quality is typically a negotiable parameter, with foods of lower quality being reflected in the price paid by the purchaser. Conversely, safety should be nonnegotiable; if a food does not meet the standards for safety, it should not

enter commerce. While this differential may not always be as straightforward as the statement implies, it does communicate the concept that safety should not be a variable that can be controlled by manipulating the price of the food. Another reason for not combining quality and safety programs is that it is likely to dilute safety efforts and lead to confusion on the part of business operators on the relative importance of safety versus quality issues. Combining the two would greatly expand the potential complexity of HACCP plans, leading to HACCP plans with excessive numbers of CCPs. Thus, while the general systematic approach encompassed in the seven HACCP principles is pertinent to quality issues, if implemented to manage quality issues, it should be done through a separate program that is not labeled as a HACCP plan. In keeping with its origins, the basic premise underlying HACCP is the prevention or elimination of hazards during production through effective system design and management rather than relying solely on inspection or testing of finished products. This is not to imply that testing finished products and other aspects of the manufacturing process do not play a role in HACCP systems (e.g., as a verification tool). Instead, it is to emphasize that the main focus of HACCP is prevention through process control. As described in the chapter on microbiological standards (chapter 6) and elsewhere (24), testing is inherently probabilistic in nature; it is only as effective as the sampling plan and the microbiological methods employed and is affected by the nature and distribution of the contamination. By contrast, if the hazards of concern are properly identified and effective control measures are available and followed, HACCP provides greater confidence in the safety of each unit produced than can be obtained through any practical sampling and testing program. The degree of confidence associated with individual HACCP plans is, in part, dependent on the degree of control implemented by the control measures built into the systems, which, in turn, is established by the CLs established for CCPs. The degree of control (and thus confidence) is dependent on the “stringency” built into the HACCP system. Stringency is not an absolute, and there are often tradeoffs between product safety and product quality (e.g., increased cooking times and temperatures increase microbial inactivation but may adversely affect quality attributes and retention of thermally labile nutrients). One of the key roles of regulatory agencies in HACCP is to articulate the degree of control expected so that food manufacturers can implement HACCP plans that are appropriately stringent. It is important to emphasize that HACCP programs are not a zero-defect proposition. Every process, includ-

SMP_Food Microbiology_CH42.indd

1042

Manila Typesetting Company

11/08/2012 07:22AM

1043

42.  Using HACCP in Food Safety Risk Management ing the manufacturing of foods, has its inherent capabilities and variability. HACCP works best when the operator has specific knowledge of both, and can thereby understand the process parameters within which the system is achieving the level of control intended. For example, the processing of juice to achieve a 5-D inactivation of key pathogenic microorganisms has been highly effective in ensuring the production of safe products. This treatment was based on data indicating that the level of contamination with human pathogens is <1 CFU/ml. Thus, a 5-log reduction would reduce that level to <1 CFU/100,000 ml. However, if the untreated juice had pathogen levels of 106 CFU/ml, it is obvious that the treatment would be insufficient to achieve the level of control intended. In some cases, the degree of treatment in comparison to the inherent levels of the contaminant is so large as to effectively be equivalent to the elimination of the hazard (e.g., thermal processing of low-acid canned foods); however, for most food safety systems the overall impact is control, not elimination of a hazard.

PRELIMINARY STEPS A number of preliminary steps are necessary in order to acquire the key resources and information needed to initiate of the development of a HACCP program (29). Key preliminary steps that will typically precede the conduct of the hazard analysis include (i) gaining management support, (ii) assembling the HACCP team, (iii) describing the food and the method of its manufacture and distribution, (iv) identifying the intended use and consumers of the food, (v) assembling available information on the past microbiological and epidemiologic profiles of similar products and technologies, (vi) developing a flow diagram, and (vii) verifying that flow diagram. A similar series of preliminary steps are identified in the Codex Alimentarius HACCP guidelines (15). It is also generally assumed that in order to achieve success in HACCP, a business operator must first be able to effectively and consistently implement all necessary GHPs. The criticality of gaining management support for HACCP programs cannot be overemphasized. Without a long-term commitment, the time and effort required to develop and implement such a program cannot be sustained, particularly when decisions related to process deviations require actions that may negatively impact productivity or profitability. The option to not employ HACCP has increasingly become untenable, as regulatory agencies have adopted a HACCP framework as a requirement for producing and marketing foods. This has been further reinforced by food marketers build-

SMP_Food Microbiology_CH42.indd

1043

ing into their purchase contracts the need for suppliers to have active HACCP programs. Similarly, insurance companies may require as a condition for being insured that the food manufacturer include HACCP as one of the company’s “due diligence” activities. The development and implementation of an effective HACCP plan require expertise in foodborne hazards, detailed knowledge of the product and its manufacture, information on past performance of the system, the source and history of the ingredients, the intended use and potential abuse of the product, the likely conditions of transport/distribution/marketing, and any information pertaining to special subpopulations at increased risk (e.g., individuals with increased susceptibility or medical conditions and individuals with food allergies). This broad range of information needed typically requires the formation of a multidisciplinary “HACCP team.” The team will typically include individuals from multiple disciplines and backgrounds such as engineering, food production, sanitation, quality assurance, food safety microbiology, and food toxicology. Knowledge of the specific facilities, products, practices, and processes being considered is a must. Access to such a broad knowledge base in food safety can be a substantial challenge to smaller food companies. It is possible to augment a team with missing expertise by enlisting the help of external subject matter experts. However, active involvement by facility employees is critical in terms of both the quality of information considered and the knowledge gained by the company as a result of the in-depth review of their products and processes. The educational insight gained as a result of developing a HACCP program is often an underappreciated benefit of implementing HACCP.

PRINCIPLE 1: CONDUCT A HAZARD ANALYSIS Once the preliminary steps have been completed, the next phase is to conduct a hazard analysis. According to the NACMCF, “The purpose of the hazard analysis is to develop a list of hazards that are of such significance that they are reasonably likely to cause injury or illness if not effectively controlled. Hazards that are not reasonably likely to occur would not require further consideration within a HACCP plan” (29). The NACMCF defines a hazard (for HACCP) as “a biological, chemical, or physical agent that is reasonably likely to cause illness or injury in the absence of its control” (29). NACMCF also defined a hazard analysis as “the process of collecting and evaluating information on hazards associated with the food under consideration to decide which are significant and must be addressed in the

Manila Typesetting Company

11/08/2012 07:22AM

1044

Advanced Techniques in Food Microbiology

HACCP plan.” NACMCF envisioned a two-step process that involves an initial listing of all potential hazards, followed by a hazard evaluation to reduce that list to only the “significant hazards.” The hazard evaluation considers the likelihood of the occurrence hazard and the severity of the consequences if not controlled. The NACMCF document makes a point of stating that the hazard evaluation is separate and distinct from a risk assessment. This seems at odds with the basic definition of risk, i.e., the probability and severity of a hazard yielding an adverse effect (see chapter 41). The HACCP team is effectively asked to do a qualitative risk assessment, often without the benefit of the formal techniques for weighing of quality and consistency of the evidence. This has the potential of decisions related to significant hazards that are uncertain, inconsistent, biased, insufficiently supported by the science, or nontransparent. This is one of the reasons it is critical that the HACCP team fully articulate and document the basis for their decisions. In particular, the HACCP team needs to ensure that they consistently evaluate the risk based on the likely impact in the absence of control. The difference in interpretation is readily apparent if one considers a low-acid canned food. With the controls commonly used in commercial operations, the history of safety for this class of products might lead an uninformed HACCP team member to misinterpret the epidemiological data to indicate that Clostridium botulinum is not a hazard that is reasonably likely to occur. Obviously, this conclusion is dramatically different from the one that would be reached if the product was not adequately processed (i.e., in the absence of control). Based on the results of this two-step hazard analysis, the hazards that should be addressed in the HACCP plan, along with control measures for each hazard, are identified. For a hazard to be listed among the significant hazards that should be addressed in the HACCP plan, the hazard should be of such a nature that its prevention, elimination, or reduction to acceptable levels is essential to the production of a safe food. HACCP does not provide an excuse for food operators to ignore low-risk hazards or to ignore existing legal requirements, including those that define adulteration. Likewise, food operators should not ignore the basic quality assurance concept of continuing improvement. Operations must be in compliance with all legal requirements. Low-risk hazards should not be dismissed but must be reviewed periodically to determine if they are adequately controlled by the company’s management systems and prerequisite programs, such as GMP programs or the control measures that have been put into place for identified significant hazards. In evaluating

low-risk hazards, it may be helpful to consider classes of hazards that have characteristics similar to those being controlled by a significant hazard. For example, the usually low-risk hazard Yersinia pseudotuberculosis might be adequately controlled if control measures for L. monocytogenes are in place. It is often the proper functioning of prerequisite programs that ensures that a potential hazard is not reasonably likely to occur. Most authorities agree that if a food operation does not have sound prerequisite programs that comply with GMPs and other regulatory requirements, HACCP will have little chance for success. It is clear, however, that the NACMCF recognized that not all biological, chemical, or physical properties that may cause a food to be unsafe for consumption are necessarily significant enough to warrant being addressed within a HACCP plan. Making the decision as to whether or not a hazard is reasonably likely to present a significant consumer safety problem is at the heart of the hazard assessment.

SMP_Food Microbiology_CH42.indd

1044

Conducting a Hazard Analysis

As described above, NACMCF (29) advises that the process of conducting a hazard analysis involves two stages: hazard identification and hazard evaluation.

Hazard Identification

Hazard identification is sometimes likened to a brainstorming session. During this stage, the HACCP team reviews the ingredients used in the product, the activities conducted at each step in the process, and the equipment used to make the product. The team also considers the method of storage, distribution, and the intended use and potential misuse of the product by consumers. Based on this review, the team develops a list of potential biological, chemical, and physical hazards that may be introduced, increased (e.g., pathogen growth), or controlled at each step described on the flow diagram. During hazard identification, the HACCP team focuses on developing an inclusive list of “potential” hazards associated with each ingredient, each process step under direct control of the food operation, and the final product. While the focus will be largely on the operations under the direct control of the food operator, the prudent operator will consider the status of the food chain both before and after the product passes through their facilities and operations. Knowledge of any adverse health-related events historically associated with the product or similar products is important in this exercise. It is also important to be specific about the hazard; rather than listing the hazard as “pathogens,” the specific pathogen should be listed (e.g., Salmonella,

Manila Typesetting Company

11/08/2012 07:22AM

1045

42.  Using HACCP in Food Safety Risk Management spores of Clostridium perfringens), especially when the controls may be different for different hazards, e.g., heat treatment versus refrigeration.

Potential biological hazards A biological hazard is one that, if not properly controlled, is reasonably likely to result in foodborne illness. The primary microorganisms of concern are pathogenic bacteria, such as C. botulinum, L. monocytogenes, Salmonella enterica, and Staphylococcus aureus. It is also important not to limit this step to the identification of potential bacterial hazards only. Other biological agents such as viruses, protozoa, toxic algae, and parasites may represent significant hazards in some foods or ingredients. The potential biological hazards associated with a product (and each ingredient) and process should be listed in preparation for evaluation. Potential chemical hazards As with biological hazards, the HACCP team should identify all potential chemical hazards associated with the production of the food before evaluating the significance of each. This includes chemicals from different origins such as naturally occurring toxic compounds, environmental contaminants, ingredients, compounds resulting from microbiological contamination, antinutrients, and processing-generated toxicants. Potential contamination with radionuclides would also be considered in this category. The list of potential chemical hazards should include chemicals that may render a food unsafe for consumption by a small percentage of the population that is particularly sensitive to a specific chemical. For example, sulfiting agents used to preserve fresh leafy vegetables, dried fruits, and wines have caused allergylike reactions in sensitive individuals. Other examples of chemical hazards to be considered include aflatoxin and other mycotoxins, fish and shellfish toxins, and scombrotoxin (histamine) from the decomposition of certain types of fish. These hazards represent a class in which the ultimate source of the hazard is microbiological in origin, and their control is likely to involve both controlling the chemical agent and maintaining conditions that limit the microorganism’s access to and growth in foods or ingredients. Allergenic ingredients are increasingly being identified as significant hazards as governments attempt to control consumer exposure to specific food allergens (e.g., peanuts, tree nuts, seafood, and milk) for which a significant portion of the population are allergic. Potential physical hazards The HACCP team should also identify any potential physical hazards associated with the finished product.

SMP_Food Microbiology_CH42.indd

1045

Foreign objects that are capable of injuring the consumer represent potential physical hazards. Examples of such potential hazards include glass fragments, pieces of wood or metal, and plastic. During the hazard evaluation stage, differentiation may be made between foreign objects that are esthetically unpleasant and those that are capable of causing injury. The HACCP team should define the performance criteria (size, shape, etc.) of physical hazards capable of causing injury. For example, a Public Health Hazard Analysis Board at FSIS concluded that bone particles measuring less than 10 mm do not pose a significant hazard to consumers (2). In addition, FDA’s current Compliance Policy Guides Manual states that a product containing a hard or sharp foreign object that measures 7 mm to 25 mm in length should be considered adulterated (3). This was based on a thorough review of 25 years of data by FDA indicating that objects measuring less than 7 mm pose little risk unless a food is intended for a special risk group such as infants or the elderly (31). In recent years, this evaluation has also been likely to consider the size and shape of the product in relation to choking risks.

Hazard Evaluation

The second stage of the hazard analysis, the hazard evaluation, is conducted after the list of potential biological, chemical, or physical hazards is assembled. During the hazard evaluation, the HACCP team decides which of the potential hazards present a significant risk to consumers. Based on this evaluation, a determination is made as to which hazards must be addressed in the HACCP plan. This includes consideration of whether the hazard is adequately controlled by GHPs or other prerequisite programs, i.e., whether the hazard requires additional controls that are not provided by GHPs. During the evaluation of each potential hazard, the food, its method of preparation, transportation, and storage, and the likely consumers of the product should be considered. The team must consider the influence of the likely food preparation practices and methods of storage and whether the intended consumers are particularly susceptible to a potential hazard. According to NACMCF (29), each potential hazard should be evaluated based on the severity of the potential hazard and its likely occurrence. Severity is the seriousness of the consequences (potential illness or injury) resulting from exposure to the hazard. Considerations of severity (e.g., the magnitude and duration of illness or injury and the potential for sequelae) can be helpful in understanding the public health impact of the hazard. Consideration of the likely occurrence of the hazard in the food as eaten is usually based upon a combination of

Manila Typesetting Company

11/08/2012 07:22AM

1046

Advanced Techniques in Food Microbiology

experience, epidemiologic data, and information in the technical literature. When conducting the hazard evaluation, it is helpful to consider the likelihood of occurrence and severity of the potential consequences if the hazard is not properly controlled. The consideration of severity can probably be enhanced by reviewing available risk assessments that are increasingly available for foodassociated hazards. The focus of the risk-based decision criteria for determination of significant hazards is the potential public health impact in the absence of controls. However, this also implies consideration of how often the controls employed are likely to fail or whether conditions are such that controls are insufficient a portion of the time (e.g., predictable periods of excess levels of a hazard). Examples of how hazard evaluations can be conducted are provided by NACMCF (29) and ICMSF (24). Hazards identified as significant in one operation or facility may not be significant in another operation producing the same or a similar product. This may happen because of differences in sources of supply, product formulation, differences in production methods, etc. For example, due to differences in equipment and/or effective maintenance programs, the probability of metal contamination may be high in one facility but remote in another. The key to this determination, however, is the proper conduct and documentation of a hazard analysis. Vast differences in HACCP plans should not be observed for two facilities that manufacture the same items in a very similar manner. The facility- and food-specific nature of hazards is also why expert panels have consistently concluded that, while generic HACCP plans are a useful tool for assisting HACCP teams in their initial thinking, such generic plans are not a substitute for the individualized plans for each food facility. When a team has determined that a potential hazard is not reasonably likely to occur, this decision should be made on the basis of all available evidence from the literature and, when available, from the plant’s own historical records. There are instances in which the hazard analysis for a specific plant/food may be at odds with the regulatory requirements for specific pathogens. In these instances, the HACCP team may have to include the regulated hazard in their final HACCP plan or seek an “official exception” from the regulatory agency. A summary of the HACCP team’s deliberations and the rationale developed during the hazard analysis must be kept for future reference. This summary should include the conclusions of the likelihood of occurrence and severity of hazards developed during the hazard evaluation. This information will be useful during validation, future reviews, and updates of the hazard analysis and the HACCP plan. It is important that hazard analyses be periodically

reviewed and updated to take into account new scientific information, the emergence of new threats, and changes in supply chains and processing technologies.

SMP_Food Microbiology_CH42.indd

1046

Identification of Control Measures

Upon completion of the hazard analysis, each hazard associated with each step in the production of the food should be listed along with the means that will be used to control the hazard. The term “control measure” was substituted for “preventive measure” in the NACMCF HACCP document because not all hazards can be prevented, but control is usually possible (29). The NACMCF defines control measures as “any action or activity that can be used to prevent, eliminate or reduce a significant hazard.” NACMCF did recognize that the latter form of control may include control measures that prevent or limit increases in the levels of a hazard but may not actually reduce the levels (e.g., adequate refrigeration that prevents growth of a target pathogen during the cooling of a product). The level of control that can be expected of control measures has been an area of confusion and debate since the inception of HACCP. Some feel that only techniques that result in a large reduction or elimination of the hazard should be included. However, there are entire classes of foods for which current processing capabilities are limited to relatively small reductions (e.g., fresh-cut produce). This led ICMSF (24) to propose two forms of CCPs: CCP1, which provides a definitive reduction in a hazard, and CCP2, which controls a hazard but does not eliminate it. While discussed extensively, this approach has not seen wide application, nor has it been adopted into regulatory frameworks. There may be several control measures available, but usually only one is selected as a CCP (e.g., a metal detector is often selected as a CCP, even though magnets and screens on the line may help prevent metal in the product). However, more than one control measure may be required to control a specific hazard. For example, foods whose microbiological safety is ensured by application of a hurdle technology typically rely on multiple control measures that involve a synergy being achieved through the manipulation of factors such as water activity, pH, storage temperature, and modified atmosphere packaging. In this instance, a CCP would of necessity involve multiple control measures that are applied at multiple locations along the flowchart.

Influence of Prerequisite Programs on the Hazard Analysis

As noted earlier in this chapter, food operation management must have a firm foundation in prerequisite programs before considering the implementation of a

Manila Typesetting Company

11/08/2012 07:22AM

1047

42.  Using HACCP in Food Safety Risk Management HACCP plan. One of the most difficult decisions that a HACCP team will face is determining whether a potential hazard can be managed under existing programs, such as GHP compliance or some other prerequisite program, or if the hazard should be managed in the HACCP plan. Prerequisite programs typically include objectives other than food safety, and it may not be easy to associate performance of a prerequisite program element, e.g., pest control or chemical storage programs, with specific production lots or batches (38). Consequently, it is usually more effective to manage nonfood safety issues and low-risk hazards within a quality system rather than including their performance and control as part of the HACCP plan. Personally, we find the key question to contemplate is whether the degree of attention paid to an activity or unit operation as part of a GAP/GMP/ GHP program is sufficient to control a hazard. If the answer is no, then the increased focus of a HACCP program is needed. Nevertheless, prerequisite programs play an important role in controlling potential health hazards. For example, supplier control programs, sanitation programs, pest management programs, and chemical control programs can be used to minimize potential health hazards such as pesticide residue, plant sanitation, or ingredient/ formulation errors. Similarly, physical hazards in many food processes can be minimized by preventive maintenance programs and by upstream control devices such as sifters or magnets. In contrast, potential acute health hazards such as the presence of Salmonella in precooked, ready-to-eat foods are usually controlled via a HACCP system in cases where definitive CCPs are available to eliminate or control the hazard. Deviations from compliance in a HACCP system normally result in action against the product, such as evaluation of product to determine appropriate disposition (reworking, destruction, etc.). This is a key consideration that can aid in distinguishing between control points within prerequisite programs and CCPs that should be included in a HACCP plan. The decision as to whether a hazard warrants control within a HACCP program will depend for the most part on the HACCP team’s evaluation of the risk to consumers that would result from a failure to control the hazard. The use of prerequisite programs in addressing certain potential hazards has been recognized by regulatory agencies. Both USDA-FSIS and FDA consider sanitation standard operating procedures to be prerequisite programs for products produced under the HACCP regulations. In 2003, FSIS issued a Directive on the presence of foreign material in meat or poultry products, which notes that an establishment can determine that foreign mate-

SMP_Food Microbiology_CH42.indd

1047

rial is a food safety hazard that is not reasonably likely to occur as the result of prerequisite programs (21). FSIS also recognized that certain aspects of L. monocytogenes control may be addressed in a prerequisite program in its interim final rule on control of L. monocytogenes in ready-to-eat meat and poultry products (20). The importance of prerequisite programs is also recognized internationally. The Codex Alimentarius Commission specified that HACCP should be implemented after the GHP outlined in the Commission’s Recommended International Code of Practice—General Principles of Food Hygiene have been “…well established, fully operational and verified…” (15). When a hazard is considered to be “not reasonably likely to occur” due to the operation of a prerequisite program, it is particularly important to document the HACCP team deliberations and the rationale developed during the hazard analysis. It is important to monitor the prerequisite program to ensure it is carried out properly, as lack of compliance with the program can bring into question the validity of the hazard analysis.

PRINCIPLE 2: DETERMINE THE CCPs Once the significant hazards and control measures are identified, CCPs within the process scheme are identified. A control point (CP) is any point, step, or procedure at which biological, physical, or chemical factors can be controlled. A critical control point (CCP) is defined by the NACMCF (29) as “a step at which control can be applied and [where control] is essential to prevent or eliminate a food safety hazard or reduce it to an acceptable level.” As previously mentioned, ICMSF (24) had previously defined two types of CCP based on the degree of assurance that could be achieved, i.e., elimination versus control to an acceptable level. NACMCF (29) felt that this could be avoided by inclusion of the phrase “reduce it to an acceptable level” in the definition, thereby avoiding the need to differentiate two types of CCPs. Each significant hazard identified during the hazard analysis should be addressed by at least one CCP. Examples of CCPs may include cooking, chilling, product formulation control, application of a bactericidal rinse, a decontamination step, etc. While measures needed for adequate control of a hazard should be reflected in the HACCP plan, in order to keep HACCP programs plant-friendly and sustainable, CCPs should not be redundant. Redundant CCPs typically add little to the margin of safety but will add to the record keeping and administrative burden of a firm’s management structure. They will also add significant cost without concomitant benefit. Experience has

Manila Typesetting Company

11/08/2012 07:22AM

1048

Advanced Techniques in Food Microbiology

shown that HACCP plans that are unnecessarily cumbersome will not be supported over extended periods. A related issue is the number of CCPs identified in a HACCP plan. If the hazard evaluation phase of the hazard analysis is unable to eliminate a large percentage of the potential hazards as being “not reasonably likely to occur,” it is highly likely that the subsequent identification of CCPs will lead to an excessive number of CCPs being identified. While there is no specific maximum or minimum number of CCPs that must be identified, if dozens of CCPs are identified, one should question whether (i) the hazard analysis should be revisited or (ii) the production of the food product is so complex that it may not be possible to safely manufacture the product. As an example of this concept, a firm may have multiple magnets and metal detectors in a line to protect production equipment from damage and consumers from harm. If the product is passed through a metal detection/reject device after it is in its final package, the last detector will typically be regarded as the CCP for this hazard, whereas up-line metal detectors or magnets will be considered control points.

to the “performance objective” defined as part of the Codex Alimentarius framework for establishment of microbiological food safety risk management metrics (12). Each CCP will have one or more control measures that must be properly applied to ensure prevention, elimination, or reduction of significant hazards to acceptable levels. Each control measure will have an associated CL that serves as the decision criterion for acceptable versus unacceptable performance for the control measure. As such, product manufactured during a period when the CL is not within the required range would be considered as being produced during a period of process deviation. At a minimum, such an event requires an investigation and evaluation before the food is released into commerce. Ideally, CLs are based on objective, quantifiable metrics such as temperature, time, physical dimension, humidity, moisture level, water activity, pH, titratable acidity, salt concentration, available chlorine, viscosity, and presence or concentration of preservatives. As HACCP has evolved, it has been determined that attributes may also be established as CLs, e.g., the CL for a metal detector may be specified as “on and functioning.” The CL for delivery of the time component of a thermal process in a continuous system may be a specific flow rate for a liquid as defined by a pump setting and correct holding-tube length and diameter for a liquid or by a belt speed for a conveyor for solid foods going through a continuous oven.

Decision Trees and Their Use

A number of CCP decision trees have been developed to assist firms in the identification of CCPs, such as NACMCF (29) and Codex Alimentarius (15) (Fig. 42.1). However, such tools should be applied with a degree of caution. Decision trees should generally be applied after the hazard analysis so that only the significant hazards are being considered. By applying the questions to potential hazards that are not reasonably likely to cause illness or injury, the HACCP team will unintentionally identify CCPs that are not related to controlling product safety. Another common problem is that no decision tree will give correct answers for all applications. For example the decision tree employed for microbiological hazards may not be equally amenable for consideration of chemical hazards. NACMCF (29) concluded that “Although application of the CCP decision tree can be useful in determining if a particular step is a CCP for a previously identified hazard, it is merely a tool and not a mandatory element of HACCP. A CCP decision tree is not a substitute for expert knowledge.”

PRINCIPLE 3: ESTABLISH CLs A critical limit is defined as “a maximum and/or minimum value to which a biological, chemical or physical parameter must be controlled at a CCP to prevent, eliminate or reduce to an acceptable level the occurrence of a significant food safety hazard” (29). A CL is analogous

SMP_Food Microbiology_CH42.indd

1048

Setting CLs

Each firm must ensure that the CLs identified in its HACCP plan are adequate for the intended purpose. Hence, the firm or its process authority should base CLs on a benchmark or performance standard that the treatment should achieve. There are many possible sources for these benchmarks. They may be derived from regulatory standards and guidelines (when they relate directly to a health hazard), literature surveys, experimental studies, or expert elicitations such as process authorities. For example, for juice products U.S. regulatory authorities require a treatment targeted to achieve at least a 5-log10 reduction of the pathogen of concern. Hence, a heat treatment could be judged adequate if it achieves this performance criterion. The firm should have adequate scientific studies to validate the CLs it has selected and the proper delivery of the control measure to each unit of product. The normal variation in process delivery should be considered when establishing a CL, and a firm should consider setting target or operational values such that corrective actions are initiated before the CL is exceeded. Validation, which includes assurance that the CLs are well founded as well as effectively and consistently de-

Manila Typesetting Company

11/08/2012 07:22AM

1049

42.  Using HACCP in Food Safety Risk Management

Q1: Does a control measure exist for the iden�fied hazard?

YES

Modify step, process, or product

NO Is control at this step necessary for safety?

NO

Not a CCP

YES

STOP

Q2: Does this step eliminate or reduce the likely occurrence of the hazard to an acceptable level?

NO YES

Q3: Could contamina�on with the iden�fied hazard occur in excess of the acceptable level or could it increase to an unacceptable level?

YES

NO

Not a CCP

STOP

Q4: Will a subsequent step eliminate the iden�fied hazard or reduce its likely occurrence to an acceptable level?

YES

Not a CCP

STOP

NO

Cri�cal Control Point

Figure 42.1  Example of a decision tree used to assist in the identification of CCPs. Adapted from reference 29. doi:10.1128/9781555818463.ch42f1

livered, is often discussed in conjunction with principle 6, verification. However, it is a separate activity, and a number of experts have recommended that validation should be included as an eighth principle. This will be discussed in more detail below. The establishment of CLs is easiest if there are consensus agreements on the level of control that is needed for specific hazards in the final product. For example, the requirement that powdered infant formula be free of Salmonella based on the examination of 60 25-g samples provides a performance objective against which the individual microbiological CCPs can be benchmarked for their contribution to meeting this end product CL. It becomes much more difficult to develop risk-based, evidenced-based CLs that can be scientifically supportable in the absence of such risk management metrics. During the past decade, there has been a great deal of activity in the development of risk assessment and risk management tools that can be used to more effectively

SMP_Food Microbiology_CH42.indd

1049

link food safety systems to likely public health outcomes, but they have only recently been applied to HACCP systems (see below).

Microbiological Criteria as CLs

Just as microbiological testing is generally not a good tool for monitoring a CCP, microbiological criteria are typically not sufficient to serve as CLs in a HACCP program. Because HACCP targets process control, factors that lend themselves to “real-time” monitoring and quick feedback should be identified as control measures. Thus, microbiological testing against a microbiological criterion is generally limited to HACCP verification. Furthermore, microbiological criteria are a means of determining the likelihood that the process is delivering product that is achieving a specified level of control with a certain degree of confidence. As will be discussed later, in today’s modern risk management approach to microbiological food safety this involves the establishment of

Manila Typesetting Company

11/08/2012 07:22AM

1050

Advanced Techniques in Food Microbiology

food safety objectives, performance objectives, and process criteria. These, in turn, serve as the basis for relating microbiological criteria to public health protection. There will be instances when the only option open to a processor is to hold an ingredient lot and perform microbiological testing before release. This may arise when a particular ingredient can be obtained only from sources where little control is exercised over factors that affect the contaminants/pathogens associated with the ingredient. In this case, a CCP may be located at receiving, where the incoming ingredient would be sampled for analysis and placed in controlled storage until the results of the analysis are available. In this instance, microbiological testing is a control measure, the running of the test is the CCP, and the microbiological criterion it is being tested against is the CL. The same approach may be used with specific high-risk shelf-stable products for which delay in release is not a significant factor. For example, in addition to testing for nutrient content, each lot of powdered infant formula is tested for Salmonella and Cronobacter spp. prior to release. This reflects the lack of technologies that can provide commercially sterile product in a powdered form, the potentially low infectious dose for both pathogens, the rapid growth of both microorganisms once the product is rehydrated, and the elevated susceptibility of the target population. Again, in this example the microbiological testing is used as a control measure and not a verification or monitoring tool.

toring is used to determine when a deviation occurs at a CCP, i.e., exceeding the CL, which will result in the initiation of appropriate corrective actions. Third, it provides the basis for written documentation of process control for use in verifying that the HACCP plan has been followed. When establishing monitoring activities, the firm should specify

PRINCIPLE 4: ESTABLISH MONITORING PROCEDURES Monitoring is a planned sequence of observations or measurements to assess whether a CCP is under control and to produce an accurate record for future use in verification. What is monitored typically is a measurement or observation of the physical factor identified as the control measure for a significant hazard. Examples of monitoring activities include the following: measuring temperature tracking elapsed time, e.g., time at a specific   temperature • sampling product and determining pH • determining moisture level or water activity • •

Monitoring serves three main purposes. First, monitoring tracks the system’s operation in a manner essential to food safety management. If monitoring detects a trend towards loss of control, i.e., approaching a target level, then action can be taken to bring the process back into control before a deviation occurs. Second, moni-

SMP_Food Microbiology_CH42.indd

1050

• • •

•

•

what control measure(s) is being monitored how often the monitoring needs to be conducted what procedures will be followed to collect data and what methods and equipment will be used who will be responsible for performing each monitoring activity who is responsible for taking action if the CL is exceeded.

The frequency of monitoring activities should be consistent with the needs of the operation in relation to the variation inherent in the control step. The frequency of monitoring should also be adequate to ensure proper control and detect deviations in a timely manner. There are distinct advantages to integrating the frequency of monitoring with a product coding system that is designed to prevent excessive amounts of product from being involved in a corrective action if problems arise and a CL is exceeded. For example, if a product must reach a certain temperature during a cooking step and that temperature is checked only once per hour, then if the temperature drops below the CL, all product produced after the last check must be reworked or destroyed. Less of the product would be involved if checks were more frequent or if a method of continuous monitoring with an automatic line shutdown or diversion were used. Control measures that can be monitored on a continuous basis are preferred, in part, because such control measures are more likely to include continual recording and electronic monitoring with automated trend analysis. Such monitoring can provide advance warning to allow for correction of a problem before exceeding a CL. In the example above, constant monitoring of temperature with an alarm to warn of a trend toward a CL being exceeded may avoid a deviation or at least would minimize the amount of product subject to the established corrective action. Regardless of whether monitoring data are collected automatically or manually, they should be recorded in a manner that allows the data to be arrayed over time. There are a variety of “control chart” techniques and related statistical process control (SPC) tools that are readily adaptable for use with HACCP systems. Certain characteristics, however, cannot be measured on a continuous basis or may not need continuous

Manila Typesetting Company

11/08/2012 07:22AM

1051

42.  Using HACCP in Food Safety Risk Management monitoring. Products that are mixed in a batch, such as some acidified products, do not need continuous monitoring of pH. A control measure like water activity in a food cannot easily be monitored continually. Such items will be monitored through a planned sequence of sampling and testing of appropriate frequency to document control. One of the underlying concepts of HACCP is to promote awareness of food safety and individual responsibility at the line level of food operations. One way to promote this is to involve line operators in monitoring activities. In many operations, monitoring activities are assigned to quality control when they could be accomplished just as effectively by line operators, with the extra benefit of line worker involvement. The role of quality control personnel may be more appropriate in verification, or “checking the checker.” Those responsible for monitoring a control measure at a CCP must • • •

•

be trained in the appropriate monitoring techniques understand the importance of monitoring accurately report/record results of the monitoring activity immediately report deviations from CLs so that corrective actions can be taken

This person may often be given authority to shut down the line when a deviation occurs. Integral to effective monitoring is the reliability of the data being acquired. Often, the parameters to be monitored are indirect measures of the attribute or condition that needs to be controlled. For example, if the microbiological safety of a ready-to-eat product is dependent on the extent of cooking it receives in a heating tunnel, the underlying processing criterion is to subject all units of the food to a time-temperature combination that will ensure that a population of a target microorganism is reduced by a specified number of log cycles (e.g., cooking to achieve a 5-log reduction of Salmonella populations). However, the actual parameters that are likely to be measured are the temperature of the oven (possibly at multiple locations in the heating chamber) and the conveyor speed, which is used to measure the duration of the heating step. CLs for these physical factors must be set in a manner that ensures that the time-temperature combination needed to achieve the specified inactivation is in turn ensured. If this firm chooses to use equipment that provides continuous monitoring (recording) of time and temperature, then the operator’s duty is to ensure that the monitoring equipment is working and to obtain a manual reading periodically as a verification activity. If there is a deviation from either the time or temperature component of the cooking step, the operator would

SMP_Food Microbiology_CH42.indd

1051

initiate the proper corrective action. One of the potential advantages of “smart monitoring systems” is that the data collected for HACCP monitoring can also serve as a feedback loop that allows the equipment to automatically adjust one or both of the control parameters (i.e., chamber temperature or conveyor speed) so that the equipment stays within its optimal range.

PRINCIPLE 5: ESTABLISH CORRECTIVE ACTIONS The HACCP system is designed to identify situations in which health hazards are reasonably likely to occur and to implement strategies to prevent or minimize their occurrence. However, ideal circumstances do not always prevail. When there is a deviation from established CL limits, corrective actions must be taken. Corrective actions, in the HACCP context, are procedures to be followed when a deviation, i.e., failure to meet a CL, occurs so that (i) no food produced during deviation is released until its safety is verified, and (ii) the system is returned to its “in control” state. Corrective actions should be developed and adopted ahead of time so that operators can quickly and appropriately respond in order to •

•

• •

fix or correct the cause of noncompliance to ensure that the process is brought under control and the CLs are being met hold the product produced when the deviation occurred and determine the disposition of noncompliant product, ensuring that no product that may be harmful is released take steps to prevent reoccurrence determine whether adjustments in the HACCP plan are needed

Because of the variations in CCPs for different foods and the diversity of possible deviations, corrective action plans that are specific to the particular operation must be developed. A corrective action plan should be developed for deviations that may occur at each CCP. Individuals responsible for ensuring that the appropriate corrective actions are being implemented should have a thorough understanding of the process, the product, and the HACCP plan. Records of the corrective actions that have been taken must be maintained. Long-term solutions should be sought when a particular CL appears to be violated routinely. Such repeated events should trigger a review of the HACCP plan as well as the specific CCP and its CL. The method of achieving process control should also be reviewed to determine what improvements are needed to reduce the frequency of deviations. Root cause analyses are helpful in evaluating

Manila Typesetting Company

11/08/2012 07:22AM

1052

Advanced Techniques in Food Microbiology

underlying long-term issues that lead to process deviations. Production and quality assurance staff should consider periodic exercises that allow them to practice their incident response procedures for process deviation.

trol needed to safely produce a food product. Validation is initially conducted when a new product is being introduced and tends to be focused on the collection and evaluation of scientific and technical information to determine if the HACCP plan, when properly implemented, will effectively control the significant hazards. With SPC approaches, this will typically involve the conduct of process capability studies that determine the performance, variability, and reliability of the food safety system. While most experts consider validation as a distinctly different activity from verification and monitoring, the last two activities are often the source of information that determines when a system needs to be revalidated. As mentioned above, differing results with monitoring and verification activities could signal a need to reevaluate and possibly revalidate the system of control measures. Similarly, seasonal or geographic differences in performance may indicate that the original process control study was not of sufficient duration to identify additional parameters that are not adequately controlled in the current version of the HACCP plan. HACCP plans need to be revalidated whenever significant changes occur in product formulations, equipment, processing procedures, or sourcing of raw ingredients. In addition, it is appropriate to reassess a HACCP plan periodically. U.S. regulations for seafood, juice, meat, and poultry require this to be done at least annually.

PRINCIPLE 6: ESTABLISH VERIFICATION PROCEDURES There are several HACCP-related activities that are conducted under the label of verification. Verification is the use of methods, procedures, or tests, in addition to those used in monitoring, to determine if the HACCP plan is being followed, whether the records of monitoring activities are accurate, and whether the control measures are functioning as intended. Verification is a second level of review, beyond the primary review by line personnel who are actually conducting the monitoring. This secondary level of review is typically the responsibility of quality control personnel and consists, in part, in verifying that records are being kept accurately and that monitoring activities are being properly conducted. This may include observations of monitoring activities or conducting an independent test. Verification sampling and testing are conducted where appropriate. In fact, most testing of products for microorganisms or other hazards that cannot be completed in real-time is conducted as a verification activity. Verification often involves a third level of review, whereby outside audit teams are used to review all aspects of the HACCP plan, monitoring procedures, record-keeping practices, etc. Verification activities should be performed in conjunction with a review of monitoring data. If monitoring data indicate that the food safety system is “in control,” but verification activities indicate a control failure, then a root cause analysis should be conducted to determine if the HACCP plan has failed to identify a CCP or if changes over time have led to the need for additional control measures. Repeated incidences of verification failures despite “in control” monitoring data are often a signal that a HACCP plan may need to be revalidated (see below). Verification activities include the periodic calibration of the equipment and instruments used to ensure the safety of a food product. Thus, temperature-measuring devices, water activity meters, pH meters, scales, etc., need to be periodically calibrated. As we move toward increased use of automated systems, verification activities will increasingly need to include periodic review of software systems. As discussed earlier, validation is a process distinct from verification that determines whether the HACCP plan is capable of consistently delivering the level of con-

SMP_Food Microbiology_CH42.indd

1052

PRINCIPLE 7: ESTABLISH RECORD-KEEPING AND DOCUMENTATION PROCEDURES Record keeping is integral to maintaining a HACCP system. Without effective record keeping and review, the HACCP system will not be maintained as an ongoing practice. Records provide the basis for management assessment of the safety of food products and for documenting the safety of products to clients, customers, and regulatory agencies. Records provide a means to trace the production history of foods, to document that CLs were met, and to prove that any needed corrective actions were carried out.

HACCP Plan Documents

The HACCP plan and associated support documents should be on file at the food establishment. Generally, the records utilized in the HACCP system will include the following.

The HACCP Plan • •

Listing of the HACCP team and assigned responsibilities Description of the product and its intended use

Manila Typesetting Company

11/08/2012 07:22AM

1053

42.  Using HACCP in Food Safety Risk Management •

•

• •

• • • •

Block flow diagram for the entire manufacturing process indicating CCPs Hazards associated with each CCP and control measures CLs for each CCP Monitoring procedures (who, what, when, where, and how) Corrective action plans for deviations from CLs Record-keeping procedures Procedures for verification of the HACCP system Records of validation studies

Records Obtained during the Operation of the Plan •

•

• •

Records of data collected by monitoring of control measures at CCPs Corrective action records and accompanying investigational data and analysis Records of verification activities Control charts derived from the various data sets arrayed for trend analysis

Management of the records associated with HACCP plans can be substantial and a challenge to many smaller food operators. Use of computer-based record-keeping systems would seem to be ideal for organizing and archiving data and for providing a platform for subsequent analyses. Potentially, data collection from automated monitoring systems could be directly deposited into a HACCP informatics system using appropriate laboratory information management systems (LIMS) technologies. However, this has to be counterbalanced against ensuring that the data being archived are reviewed by the people who need the information to make informed decisions. The most effective use of HACCP record keeping is when its utility ensures that HACCP is an active risk management system and not an exercise that is carried out and then forgotten until there is an emergency.

HACCP LIMITATIONS While HACCP has been accepted worldwide and has become a mandatory food safety requirement in many countries, it could be improved. As a food safety risk management system, it has a number of significant limitations that were problematic when the system was originally devised and remain so today. The lack of progress in addressing these limitations may in part reflect a general lack of HACCP evolution over the past 50 years. This, in part, may be the consequence of HACCP being adopted as a regulatory framework by a number of countries,

SMP_Food Microbiology_CH42.indd

1053

as well as the focus for international harmonization to facilitate global food trade. The purpose of this section is not to dwell on the limitations of a highly successful system but instead to point out where the innovations in this and other food safety systems are likely to be forthcoming in the next decade.

Hazard Analysis Limitations

As emphasized above, the key to a successful HACCP program is the identification of “significant hazards” that must be controlled. However, since the inception of HACCP there has been a dramatic shift from hazardbased food safety systems to risk-based systems. This, in part, is a direct consequence of implementing HACCP. It also embodies the significant advances in both chemical and microbiological food safety risk assessments that have occurred during the past decade. The dilemma is that the current hazard-based approach to HACCP does not allow a direct linkage between HACCP performance and improvements in public health. Thus, it is very difficult to reach consensus on the stringency that is needed to achieve public health goals. This, in turn, makes it difficult to reach consensus on the magnitude of CLs and the frequency and stringency of monitoring and verification activities. While a number of practitioners of HACCP have asserted that the hazard analyses done within a HACCP plan are distinctly different from a risk assessment, the food safety risk assessment community has consistently indicated that in fact the hazard analysis should be considered a qualitative risk assessment. This is particularly true for the selection of “significant hazard,” for which HACCP guidelines specifically require consideration of both probability and severity of the hazard. These two parameters are the exact attributes that are considered in a risk assessment. It is not surprising that for the past decade various risk assessors have consistently recommended the incorporation of risk assessment techniques into the HACCP hazard analyses (6–8, 34, 41, 42, 46–48) in order to provide a much stronger evidence-based approach to defining what is meant by “reasonably likely to occur.” When the conceptual advances in risk analysis and its international adoption for the development of regulatory standards are compared to the largely nontransparent role that assessing risk plays in determining significant hazards, it is apparent that there is a need to reassess principle 1 so that it better reflects the current state of the science.

Managing Risks

The focus of HACCP has been predominately associated with the food manufacturing phase of the food chain.

Manila Typesetting Company

11/08/2012 07:22AM

1054

Advanced Techniques in Food Microbiology

The application of HACCP principles to the other segments has been generally considered less successful. For example, there has been ongoing debate on whether HACCP can be successfully applied at the farm level (10, 36). However, the basic system concepts underlying HACCP should be no less valid. The difficulty may revolve around the attributes of these other segments. First, the control measures available are generally less definitive than the ones relied upon in the manufacturing environment. Control is typically dependent on multiple parameters, none of which has a definitive reduction step. Accordingly, the HACCP plans on the farm or in food service operations become less clear-cut in terms of establishing definitive CCPs as opposed to a series of control points. However, risk assessment and systems modeling techniques should be able to assess the relative importance of a series of control points, thereby establishing a means of assessing and prioritizing these points as critical control measures. Second, the use of systems modeling should also be able to achieve a much higher degree of evidence to support the selection of CCPs than the current system of using a qualitative hazard analysis and simple decision trees. For example, Domenech et al. (16) applied the use of CCP effectiveness analyses within a risk assessment framework to more effectively identify CCPs and provide analyses needed to make informed decisions regarding CLs. Research currently under way in our laboratory suggests that employing relatively simple product pathway risk assessment models in conjunction with sensitivity analyses and what-if scenarios can assist in the identification of CCPs and the establishment of CLs, respectively. One of the strengths of HACCP programs is the focus on the development of plans that are specific for each food/facility. However, it also raises the challenge for manufacturers, marketers, and regulatory agencies on how to compare the performances of different operations. The emergence of quantitative food safety risk analysis approaches has begun to lead to the development of tools for establishing the risk-based estimates of the equivalence of different food operations. This is in large part an outcome of the introduction of food safety objectives/performance objectives concepts by ICMSF, the further exploration of the concepts by the Joint FAO/WHO Expert Meetings on Microbiological Risk Assessment (JEMRA), and the adoption of these microbiological risk management metrics by the Codex Alimentarius Commission. The ability to link the performance of a HACCP plan to achievement of a risk-based level of control is being actively explored by industry, governments, academics, and intergovernmental agencies. Some of the recent examples of how food safety ob-

jectives concepts are being used include how to evaluate the stringency needed for GHP/HACCP programs for L. monocytogenes in ready-to-eat foods (13, 32), Salmonella in poultry meat (26) and beef (44), Cronobacter spp. in powdered infant formulae (14), and C. botulinum in commercially sterile foods (1).

SMP_Food Microbiology_CH42.indd

1054

Managing HACCP Data

As discussed above, managing the data and records generated by HACCP programs can be a major challenge. However, considering the expense associated with the generation and archiving of the data, there are often minimal attempts to mine the data for additional information, i.e., if the monitoring or verification activity is satisfactory, it is recorded, archived, and ignored unless there is a problem later. However, these data can provide a great deal more information if properly analyzed and arrayed. HACCP systems are ideal for application of SPC, a fact recognized by HACCP researchers from its early adoption from FMEA. It has also been recognized by USDA and FDA, which based a number of required or recommended microbiological verification tools on SPC principles. This area has continued to be of interest to some academic researchers, who envisioned the application of various SPC tools to enhance the collection and analysis of HACCP data. For example, Tokatli et al. (43) demonstrated the utility of multivariate process monitoring and fault diagnosis techniques to HACCP programs involving food pasteurization processes. Augustin and Minvielle (4) evaluated the use of various control charts for application to microbiological contamination of pork meat cuts. Srikaeo and Hourigan (40) demonstrated that SPC techniques could be used to enhance the validation of CCPs related to shell egg washing. Srikaeo et al. (39) used SPC techniques to examine biscuit baking and found that a number of the parameters that influenced the adequacy of the baking process were not in control. Despite the extensive use of SPC in food production and food quality, the utilization of SPC to food safety in general and HACCP in particular has been limited. This, in part, appears to be related to three factors. First, HACCP training is often limited to a description of the principles and simple examples and seldom delves into the techniques for data analysis. Second, there is a general avoidance of statistical analysis and mathematics on the part of food operators and nonengineering food safety scientists, despite the increased availability of these tools. Third, the first two factors appear to contribute to a general lack of user-friendly tools that would help small- to medium-sized food operators enhance their HACCP programs through the application

Manila Typesetting Company

11/08/2012 07:22AM

1055

42.  Using HACCP in Food Safety Risk Management of SPC techniques. Addressing these factors may require a long-term endeavor but can be readily initiated by direct inclusion of SPC techniques into HACCP education programs. Likewise, the provision of simple data informatics tools by regulatory agencies may be the easiest route for encouraging the use of SPC techniques by small food operators. To put this in perspective, one only has to think of the impact that tax preparation software has had here in the United States.

OTHER FOOD SAFETY SYSTEMS While HACCP is the primary food safety risk management system, we have seen the adoption of other systems. The one that has received the most attention recently has been the quality and management systems being developed by the International Organization for Standardization (ISO). They have recently become much more active in food safety through their issuance of ISO 22000:2005: Food Safety Management Systems and ISO 22002-1:2009: Prerequisite programs (PRPs) on Food Safety—Part 1: Food Manufacturing. While both programs are similar to HACCP, there are differences in details. However, ISO appears to be having difficulty in reconciling the approach in the food systems with the risk management approaches articulated in ISO 31000: Risk Management— Principles and Guidelines and ISO 31010:2009: Risk Management—Risk Assessment Techniques (42), which may reflect consideration of the HACCP limitations described above. The other area that is receiving increased attention is the use of a systems approach to food safety issues. The basic concept underlying “system thinking” is that for complex systems (like food production) understanding the component parts of the system can be best understood in the context of their relationships among themselves and with other systems. These techniques have been successfully employed in highly complex industries such as the aerospace products, global communications networks, automobile manufacturing, and finance. Interest in applying these approaches to food safety systems and food in general appears to be increasing. Considering the complexity of food safety in the global marketplace, it is likely that these tools will begin to be used by the food industry.

References 1. Anderson, N. M., J. W. Larkin, M. B. Cole, G. E. Skinner, R. C. Whiting, L. G. M. Gorris, A. Rodriguez, R. Buchanan, C. M. Stewart, J. H. Hanlin, L. Keener, and P. A. Hall. 2011. Food safety objective approach

SMP_Food Microbiology_CH42.indd

1055

for controlling Clostridium botulinum growth and toxin production in commercially sterile foods. J. Food Prot. 74:1956–1989. 2. Anonymous. 1995. Bone Particles and Foreign Material in Meat and Poultry Products (a Report to the Food Safety and Inspection Service). Public Health Hazard Analysis Board, U.S. Department of Agriculture—Food Safety and Inspection Service, Washington, DC. 3. Anonymous. 1999. Foods—adulteration involving hard or sharp foreign objects. Food and Drug Administration Compliance Policy Guides Manual—Update No. 12, chap. 5, section  555.425.  http://www.fda.gov/ora/compliance_ref/ cpg/cpgfod/cpg555-425.htm. 4. Augustin, J.-C., and B. Minvielle. 2008. Design of control charts to monitor the microbiological contamination of pork meat cuts. Food Control 19:82–97. 5. Buchanan, R. L. 1990. HACCP: a re-emerging approach to food safety. Trends Food Sci. Technol. 1:104–106. 6. Buchanan, R. L. 1995. The role of microbiological criteria and risk assessment in HACCP. Food Microbiol. 12:421–424. 7. Buchanan, R. L. 2010. Understanding and managing food safety risks. Food Saf. 16(6):24–31. 8. Buchanan, R. L., and R. C. Whiting. 1998. Risk assessment: a means for linking HACCP plans and public health. J. Food Prot. 61:1531–1534. 9. Canadian Food Inspection Agency. 2010. Food Safety Enhancement Program Manual. http://www.inspection. gc.ca/english/fssa/polstrat/haccp/manue/tablee.shtml. Accessed 12 December 2011. 10. Cerf, O., E. Donnat, and the Farm HACCP Working Group. 2011. Application of Hazard Analysis—Critical Control Point (HACCP) principles to primary production: what is feasible and desirable? Food Control 22:1839–1843. 11. Codex Alimentarius Commission. 2003. Code of Hygienic Practice for Fresh Fruits and Vegetables. CAC/RCP 532003 (Revised 2010). World Health Organization/ Food and Agriculture Organization of the United Nations, Rome, Italy. 12. Codex Alimentarius Commission. 2007. Principles and Guidelines for the Conduct of Microbiological Risk Man­ agement (MRM). CAC/GL 63-2007 (Revised 2008). World Health Organization/Food and Agriculture Organization of the United Nations, Rome, Italy. 13. Codex Alimentarius Commission. 2007. Guidelines on the Application of General Principles of Food Hygiene to the Control of Listeria monocytogenes in foods. CAC/GL 61-2007 (Revised 2009). World Health Organization/ Food and Agriculture Organization of the United Nations, Rome, Italy. 14. Codex Alimentarius Commission. 2008. Code of Hygienic Practice for Powdered Formulae for Infants and Young Children. CAC/RCP 66-2008. World Health Organization/ Food and Agriculture Organization of the United Nations, Rome, Italy. http://www.codexalimentarius.net/download/standards/11026/CXP_066e.pdf. Accessed 30 August 2011. 15. Codex Alimentarius Commission. 2009. Recommended International Code of Practice General Principles of Food

Manila Typesetting Company

11/08/2012 07:22AM

1056

Advanced Techniques in Food Microbiology

Hygiene. CAC/RCP 1-1969, Revision 4 (2003). In Codex Alimentarius Food Hygiene Basic Texts, 4th ed. World Health Organization/ Food and Agriculture Organization of the United Nations, Rome, Italy. 16. Domenech, E., I. Escriche, and S. Martorell. 2008. Asssessing the effectiveness of critical control points to guarantee food safety. Food Control 19:557–565. 17. European Commission Health and Consumer Protection Directorate-General. 2005. Guidance Document on the Implementation of Procedures Based on the HACCP Principles, and on the Facilitation of the Implementation of the HACCP Principles in Certain Food Businesses. European Commission, Brussels, Belgium. http://ec.europa. eu/food/food/biosafety/hygienelegislation/guidance_doc_ haccp_en.pdf. Accessed 12 December 2011. 18. FDA. 1979. 21CFR113—Thermally Processed Low-Acid Foods Packaged in Thermetically Sealed Containers. 44 Federal Register 16215, March 16, 1979. FDA, Washington, DC. 19. FDA. 2011. Current Good Manufacturing Practice in Manufacturing, Packing, or Holding Human Food. 21CFR100 (Revised as of April 1, 2011) http://www. accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch. cfm?CFRPart=110&showFR=1 Accessed 30 October 2011. 20. Food Safety and Inspection Service, U.S. Department of Agriculture. 2003. Control of Listeria monocytogenes in ready-to-eat meat and poultry products; final rule. Fed. Regist. 68:34208–34254. 21. Food Safety and Inspection Service, U.S. Department of Agriculture. 2003. FSIS Directive 7310.5. Presence of Foreign Material in Meat and Poultry Products. http://www. fsis.usda.gov/Frame/FrameRedirect.asp?main=http://www. fsis.usda.gov/OPPDE/rdad/FSISDirectives/7310.5.htm 22. Grocery Manufacturers Association. 2010. Industry Handbook for the Safe Production of Nuts. Grocery Manufacturers Association, Washington, DC. 23. International Commission on Microbiological Specifi­ cations for Foods. 1988. HACCP in Microbiological Safety and Quality. Blackwell Scientific Publications, Oxford, England. 24. International Commission on Microbiological Specifi­ cations for Foods. 2002. Microorganisms in Foods 7: Microbiological Testing in Food Safety Management. Kluwer Academic/Plenum Publishers, New York, NY. 25. Lachance, P. A. 1971. Development of stored food and water systems, p. 205–228. In Environmental Biology and Medicine, vol. 1. Gordon and Breach Science Publishers, New York, NY. 26. Membre, J.-M., J. Bassett, and L. G. M. Gorris. 2007. Applying the food safety objective and related standards to thermal inactivation of Salmonella in poultry meat. J. Food Prot. 70:2036–2044. 27. National Academy of Sciences/National Research Council. 1985. An Evaluation of the Role of Microbiological Criteria for Foods and Food Ingredients. National Academy Press, Washington, DC. 28. National Advisory Committee on Microbiological Criteria for Foods. 1990. Recommendations for Refrigerated Foods

Containing Cooked, Uncured Meat and Poultry Products That Are Packaged for Extended Refrigerated Shelf Life and That Are Ready-To-Eat or Prepared with Little or No Additional Heat Treatment. http://www.fsis.usda. gov/regulations/National_Advisory_Committee_on_ Microbiological/index.asp#prior Accessed 3 November 2011. 29. National Advisory Committee on Microbiological Criteria for Foods. 1998. Hazard analysis and critical control point principles and applications guidelines. J. Food Prot. 61:762–775. 30. National Conference on Food Protection. 1971. Proceedings of the National Conference on Food Protection. Department of Health Education and Welfare, Public Health Service, Washington, DC. 31. Olsen, A. R. 1998. Regulatory action criteria for filth and other extraneous materials. I. Review of hard or sharp foreign objects as physical hazards in food. Regul. Toxicol. Pharmacol. 28:181–189. 32. Perez-Rodriguez, F., E. C. D. Todd, A. Valero, E. Carrasco, R. M. Garcia, and G. Zurera. 2006. Linking quantitative exposure assessment and risk management using the food safety objective concept: an example with Listeria monocytogenes in different cross-contamination scenarios. J. Food Prot. 69:2384–2394. 33. Ross-Nazzal, J. 2007. From farm to fork: how space food standards impacted the food industry and changed food safety standards, p. 219–236. In S. J. Dick and R. D. Launius (ed.), Societal Impact of Spaceflight. National Aeronautics and Space Administration, Washington, DC. 34. Serra, J. A., E. Domenech, I. Escriche, and S. Martorell. 1999. Risk assessment and critical control points from a production perspective. Int. J. Food Microbiol. 46:9–26. 35. Sperber, W. H. 2005. HACCP and transparency. Food Control 16:505–509. 36. Sperber, W. H. 2005. HACCP does not work from farm to table. Food Control 16:511–514. 37. Sperber, W. H., and R. F. Stier. 2009. Happy 50th birthday to HACCP: retrospective and prospective. Food Saf. December 2009/January 2010:42, 44–46. 38. Sperber, W. H., K. E. Stevenson, D. T. Bernard, K. E. Deibel, L. J. Moberg, L. R. Hontz, and V. N. Scott. 1998. The role of prerequisite programs in managing a HACCP system. Dairy Food Environ. Sanit. 18:418–423. 39. Srikaeo, K., J. E. Furst, and J. Ashton. 2005. Characterization of wheat-based biscuit cooking process by statistical process control techniques. Food Control 16:309–317. 40. Srikaeo, K., and J. A. Hourigan. 2002. The use of statistical process control (SPC) to enhance the validation of critical control points (CCPs) in shell egg washing. Food Control 13:263–273. 41. Sumner, J., T. Ross, and L. Abouch. 2004. FAO Application of Risk Assessment in the Fish Industry. FAO Fisheries Technical Paper 442. FAO, Rome, Italy. 42. Surak, J. G., and G. Gonzalez. 2011. Food safety and risk assessment. Food Saf. 17(5):16–19. 43. Tokatli, F. (K.), A. Cinar, and J. E. Schlesser. 2005. HACCP with multivariate process monitoring and fault

SMP_Food Microbiology_CH42.indd

1056

Manila Typesetting Company

11/08/2012 07:22AM

1057

42.  Using HACCP in Food Safety Risk Management diagnosis techniques: application to a food pasteurization process. Food Control 16:411–422. 44. Tuominen, P., J. Ranta, and R. Maijala. 2007. Studying the effects of POs and MCs on the Salmonella ALOP with a quantitative risk assessment model for beef production. Int. J. Food Microbiol. 118:35–51. 45. Wallace, C. A., W. H. Sperber, and S. E. Mortimore. 2011. Food Safety in the 21st Century: Managing HACCP and Food Safety throughout the Global Supply Chain. WileyBlackwell, Chichester, United Kingdom.

SMP_Food Microbiology_CH42.indd

1057

46. Weingold, S. E., J. J. Guzewich, and J. K. Fudala. 1994. Use of foodborne disease data for HACCP risk assessment. J. Food Prot. 57:820–830. 47. Whiting, R. C., and R. L. Buchanan. 1997. Predictive microbiology, HACCP and risk assessment, p. 105–112. Proc. Int. Symp. Predictive Microbiol. Applied Chilled Food Preservation. Quimper, France. 48. Wilson, J. D. 1997. Needs for risk analysis under HACCP. ORACBA News (USDA) 2(4):1–4.

Manila Typesetting Company

11/08/2012 07:22AM

SMP_Food Microbiology_CH42.indd

1058

Manila Typesetting Company

11/08/2012 07:22AM

Food Microbiology: Fundamentals and Frontiers, 4th Ed. Edited by M. P. Doyle and R. L. Buchanan © 2013 ASM Press, Washington, D.C. doi:10.1128/9781555818463.ch43

Peter Gerner-Smidt Eija Hyytia-Trees Timothy J. Barrett

43

Molecular Source Tracking and Molecular Subtyping

Bacterial subtyping may be defined as the characterization of bacteria below the species (or subspecies) level. Such characterization will allow bacterial isolates to be placed into groups that are more or less similar to one another based on one or more (usually many) characteristics and may also reveal how such groups relate to each other. Subtyping can thus be used to study the population structure of a particular bacterial species, to determine the possible evolution of the subject microorganism, or to study the molecular epidemiology of a microbe. The types of methods used for subtyping and the approaches to data analysis and interpretation may vary greatly with the reason for specific subtyping. This chapter will focus almost entirely on subtyping for molecular epidemiology. The term “molecular epidemiology,” in the context of foodborne bacteria, is usually applied to the subtyping of bacteria that cause foodborne disease and the ways in which such subtyping data contribute to understanding the transmission of those bacteria to humans. Molecular epidemiology can be applied to identifying the source of a particular outbreak or to a broader understanding of the role of certain foods or processes in outbreak-related or sporadic infections.

REASONS FOR PERFORMING MOLECULAR SUBTYPING Perhaps the most easily appreciated reason for molecular subtyping is to facilitate the identification and investigation of foodborne disease outbreaks. By identifying isolates with the same molecular subtype, one can determine which isolates are most likely to have a common source. When isolates of foodborne pathogens with the same subtype are found to cluster temporally, especially if they also cluster geographically, it may indicate an outbreak of foodborne disease. Such cluster detection is the principal focus of the PulseNet system (98). PulseNet is described in more detail in a later section of this chapter. Once clusters are recognized, epidemiologic investigations can determine if a cluster actually represents an outbreak. Molecular subtyping can further facilitate epidemiologic investigations by determining which infections, of all those occurring at the same time as the outbreak, are most likely to be outbreak related (those sharing the same subtype or very similar ones). Subtyping contributes to tracking the source of an outbreak in two ways: by facilitating the epidemiologic investigation and by matching the subtype of isolates from food products with those from patients.

Peter Gerner-Smidt, Eija Hyytia-Trees, and Timothy J. Barrett, Enteric Diseases Laboratory Branch, Centers for Disease Control and Prevention, 1600 Clifton Rd., Atlanta, GA 30333.

1059

SMP_Food Microbiology_CH43.indd

1059

Manila Typesetting Company

11/08/2012 07:22AM

1060

Advanced Techniques in Food Microbiology

There are many reasons for performing molecular subtyping other than cluster detection and outbreak investigation. Public health laboratories use subtyping to identify new and emerging bacterial pathogens, such as Salmonella serotype Typhimurium DT104, which was identified in the 1990s. Food processing plants may use subtyping for tracking microorganisms through the plant to determine where they enter into or reside within the plant and to monitor efforts to eliminate them. The types of subtyping methods used depend on the microorganism under study and the specific application (Ta­ ble 43.1). Subtyping methods will be described in detail in the next section, and specific applications will be discussed in following sections. Regardless of the subtyping method being used, the importance of considering subtyping data in the context of other available data cannot be overemphasized. Factors such as the discriminative power of the subtyping method, the genetic diversity of the microbes being analyzed, and the occurrence and distribution of common molecular subtypes must all be considered in interpretation of subtyping results. Whenever possible, interpretation should be made in the context of epidemiologic and environmental investigations. Subtyping can only indicate which isolates are more likely or less likely to be related to an outbreak, food, or environmental source. Subtyping data alone cannot prove such a link.

poses for which molecular methods are now used. In fact, most of them are still used prior to, or in conjunction with, molecular methods. Serotyping of Salmonella is an excellent example. Serotyping is so widely practiced that microbiologists tend to think of it more as identification than as subtyping, but serotyping is itself a powerful method of subtyping, with more than 2,500 named serotypes of Salmonella. With relatively rare serotypes, serotyping alone may provide a sufficient level of subtyping. For more common serotypes, phage typing systems with good strain discrimination have been developed, but these systems are serotype specific and are available in few laboratories, so molecular methods are generally used to provide further strain discrimination. Although molecular methods typically provide greater strain discrimination than phenotypic methods, this is not always the case, and it is only one reason why molecular methods are generally preferred. Perhaps the most important reason for using molecular methods is that they do not require specialized reagents or expertise and can thus be performed in almost any laboratory with basic molecular biology capabilities. This ease of performance is also the most obvious drawback to molecular subtyping, as it allows every laboratory to perform and interpret subtyping according to their own criteria, making interlaboratory comparisons more difficult unless laboratories can agree to standardized protocols.

SUBTYPING METHODS IN COMMON USE

Phenotypic Methods

Although the focus of this chapter is on molecular methods, it is important to consider them in the context of earlier phenotypic methods such as serotyping, phage typing, biotyping, and antimicrobial susceptibility typing. Most of these phenotypic methods have long and successful histories of use in subtyping for the same pur-

Plasmid Profile Analysis

One of the first molecular methods used for strain identification or source tracking is plasmid profile analysis. Plasmids are small pieces of extrachromosomal DNA that are typically circular and supercoiled. Plasmids range in size from a few hundred to several hundred thousand base pairs and are present in most bacterial species. Plasmid DNA is extracted using a method that will separate plasmid from chromosomal DNA, and the plasmids are then separated by agarose gel electro-

Table 43.1  Properties of methods commonly used for molecular subtyping of foodborne pathogens Method

Strain discrimination

Plasmid profile PFGE REP/ERIC PCR PCR-RFLP AFLPb MLSTb MLVAb

Moderate High Moderate Low to moderate High Moderate High

a b

Interlaboratory reproducibility

Supply costs

Specialized equipment

Moderate Moderate Low High Moderate High High

Low Moderate Low Low High High High

No Yes No No Yes Yes Yes

Automatable No No No/yesa No Yes Yes Yes

“No” refers to the standard method, “yes” refers to the DiversiLab system. Comments refer to method as performed using automated DNA sequencing equipment and software.

SMP_Food Microbiology_CH43.indd

1060

Manila Typesetting Company

11/08/2012 07:22AM

Unambiguous output No No No/yesa No Yes Yes Yes

43.  Molecular Source Tracking and Molecular Subtyping phoresis. How the plasmids travel in the gel depends on both size and conformation. Figure 43.1 shows a plasmid profile analysis of nine strains of ceftriaxoneresistant Salmonella. Sensitivity can be increased by restriction digestion of the plasmid DNA so that different plasmids of the same approximate size can be distinguished. One of the first successful uses of plasmid profile analysis in epidemiologic investigations occurred in 1981, when exposure to marijuana was associated with a cluster of Salmonella enterica serotype Muenchen infections (100). Early uses of plasmid profile analysis included identifying outbreak-associated strains of Salmonella serotype Typhimurium (48, 80) and linking patient and hamburger isolates of Escherichia coli O157:H7 (86). Plasmid analysis continues to be useful today. During an outbreak of Salmonella serotype Typhimurium DT104 in England and Wales in 2000, all isolates analyzed by pulsed-field gel electrophoresis (PFGE) had the same profile, but 67% of the isolates could be classified as outbreak associated based on plasmid analysis (50). One major drawback of plasmid profiling is that bacteria may gain or lose plasmids during the course of an outbreak, resulting in different plasmid profiles and confounding interpretation.

Figure 43.1  Example of plasmid profile analysis. Each lane contains plasmid DNA extracted from an isolate of ­ceftriaxoneresistant Salmonella. In this instance, plasmid DNA was separated by PFGE rather than standard gel electrophoresis for better separation of large plasmids. doi:10.1128/9781555818463.ch43f1

SMP_Food Microbiology_CH43.indd

1061

1061

RFLP Methods

DNA may be digested by restriction endonucleases, which are enzymes that cut the DNA at specific nucleotide sequences (the recognition sequence). The number and size of the resulting DNA fragments (restriction fragments) are unique to a particular strain. Other strains will yield different restriction fragment patterns. This is the principle behind all restriction fragment length polymorphism (RFLP)-based methods. In the initial methods, the DNA was digested with high-frequency cutting enzymes like EcoRI or HindIII and the resulting fragments were separated by standard agarose gel electrophoresis (95). The resulting restriction patterns were very complex due to the huge number of fragments obtained, resulting in poor resolution of individual fragments and difficulty in interpreting results. To overcome this problem, several variations on the basic method were developed, of which PFGE still remains critical for surveillance and outbreak investigations of foodborne diseases.

PFGE

The problem of excessively complex RFLP patterns was solved by decreasing the number of restriction fragments by using infrequently cutting enzymes. This way, the actual number of fragments obtained is reduced to a manageable number. While this strategy is obvious, the size of the resulting fragments prevents them from being separated by standard gel electrophoresis. The development of PFGE in 1983 (92) eliminated this drawback. During the following few years, several different types of PFGE were devised, including transverse alternating field electrophoresis (38), field inversion gel electrophoresis (12), orthogonal field alternation gel electrophoresis (12), and contour-clamped homogeneous electric field electrophoresis (15). Although these methods differ considerably in the physical way in which fragments are separated, they all depend on changing the direction of current flow (pulsing) over a gradient of time intervals between changes. The contour-clamped homogeneous electric field technology is probably the only one used today. Since the early 1990s, PFGE has been the “gold standard” for molecular subtyping and source tracking for most foodborne bacteria, as well as many other bacterial agents. PFGE is so widely known and accepted that a lengthy list of applications would be superfluous. The application of PFGE to specific foodborne pathogens is described below. As with any subtyping method, PFGE can present problems in interlaboratory reproducibility and interpretation of results. Such problems can never be totally eliminated, but they can be minimized by

Manila Typesetting Company

11/08/2012 07:22AM

1062

Advanced Techniques in Food Microbiology

using carefully standardized protocols and image analysis software (2, 98). Figure 43.2a provides an example of a PFGE gel, and Fig. 43.2b provides a dendrogram of the same gel data created using BioNumerics software (Applied Maths, Austin, TX).

comparisons. Two such methods, based on the presence of repeated elements, were reported by Versalovic et al. (105). These authors found that amplification of repetitive extragenic palindromic (REP) elements or enterobacterial repetitive intergenic consensus sequences could discriminate between laboratory strains of E. coli, suggesting that these methods could be used for subtyping foodborne pathogens. Methods targeting repetitive elements are collectively known as rep-PCR. Although rep-PCR methods have not become as widely used as other typing methods such as PFGE, they do have the advantages of being relatively inexpensive and easy to perform. They have been used to track fecal pollution by subtyping E. coli isolates in diverse settings such as dairy and swine production systems (65) and Apalachicola Bay (81). As with other subtyping methods, successful source tracking can be improved with the use of appropriate statistical methods (45). The biggest drawback to rep-PCR may be difficulties with intraand interlaboratory reproducibility. This problem may have been solved by the introduction of a commercial semiautomated platform for rep-PCR, the DiversiLab System (bioMérieux, Marcy l’Etoile, France). In this system, quality-controlled reagents are provided in a kit format with Internet-based computer-assisted analysis, reporting, and data storage (46). Although the system does not provide the same high discriminatory power as PFGE, it seems to be gaining acceptance in the food microbiology community (4, 88). The second category of PCR-based methods combines PCR with restriction digestion: either PCR amplification followed by restriction digestion of the amplicon or restriction digestion of the genome followed by selective amplification of certain restriction fragments. In the first approach, the target locus is typically 1 to 2 kilobases, and digestion results in several fragments that are then separated by gel electrophoresis, forming a characteristic RFLP profile. A specific example of the utility of such a PCR-RFLP in subtyping Campylobacter is discussed in detail in the Campylobacter section. The second approach is exemplified by amplified fragment length polymorphism (AFLP), described by Vos et al. in 1995 (106). The most widely used version of AFLP involves digestion of genomic DNA with two restriction enzymes, one of which cuts the DNA very frequently (MseI or TaqI are often used) while the other cuts with only average frequency (such as EcoRI). Specific double-stranded oligonucleotide adaptors allow ligation to create new fragments while preventing religation to recreate the original restriction sites. The resulting new fragments are then amplified by PCR with adaptorspecific primers. These primers contain extensions of 1 to 3

PCR-Based Methods

Although many different PCR-based subtyping methods have been reported, most methods fall into one of two categories. The first type of the PCR-based subtyping methods involves using primers homologous to known, conserved, multicopy sequences. In this case, differences in amplicon lengths are the basis of subtype

Figure 43.2  PFGE analysis of seven isolates of Salmonella Berta. (a) Raw data: lanes 2 to 4 and 6 to 9 contain BlnI digests of S. Berta genomic DNA, and lanes 1, 5, and 10 contain XbaI digests of a molecular weight standard strain. (b) Analyzed data: dendrogram showing relatedness of S. Berta isolates produced using BioNumerics software. doi:10.1128/9781555818463.ch43f2

SMP_Food Microbiology_CH43.indd

1062

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping bp at the 3¢ end. Each 1-­nucleotide ­extension will be selective for ­approximately one-fourth of the newly ligated fragments. The number of fragments being generated in AFLP can thus be adjusted by the number of nucleotides added to the adaptor-specific primers. A target range of 40 to 200 fragments is typically sought. Separation of the resulting fragments in agarose or acrylamide gels is possible, but the resulting patterns are complex and difficult to interpret. Most AFLP performed for the purpose of bacterial subtyping is done using capillary electrophoresis on automated DNA sequencers. Applications of AFLP to specific foodborne bacteria are discussed below. Savelkoul et al. have provided a general review of AFLP as applied to bacterial, plant, and animal genetics (91).

DNA Sequence-Based Methods MLST

Multilocus sequence typing (MLST) was one of the first subtyping or strain-tracking methods that were based on DNA sequences rather than DNA fragment sizes. All of the RFLP-based methods described above suffer, to different degrees, from the same problems. In addition to the problems of interlaboratory variability in methods and interpretation, RFLP methods have further shortcomings. Fragments from different test strains that appear to be the same size on a gel may originate from different parts of the genome with completely different genetic content. Fragments that are very close in size may not be separated during electrophoresis and thus not recognized. The reasons for differences in patterns are not known, and strains with fewer differences are not necessarily more closely related phylogenetically than strains with more differences. DNA sequence-based subtyping yields unambiguous data that can be readily compared between laboratories and can also be used to quantify differences between test strains. MLST was derived from the earlier technology of multilocus enzyme electrophoresis, which is a phenotypic method that detects changes in certain housekeeping loci by charting variation in the electrophoretic mobility of their gene products (93). Only changes that result in a different electrophoretic mobility are detected. MLST is a more sensitive method that detects all nucleotide changes by direct sequencing. MLST is typically performed by sequencing 450- to 500-bp internal fragments of seven housekeeping genes. Since these housekeeping genes are under little selective pressure, the accumulation of changes occurring by chance is rather low, making MLST an ideal tool for understanding the evolution of microorganisms or strains. However, this

SMP_Food Microbiology_CH43.indd

1063

1063

lack of rapid change in the housekeeping genes makes MLST a less-than-ideal method for investigating outbreaks or conducting traceback studies. An MLST typing scheme has been described for Campylobacter (22), and MLST has been applied to other foodborne bacteria as described below. In order to increase the discriminatory power, MLST schemes targeting fasterevolving sequences such as virulence-associated genes or temperate phages have been developed (87, 116). These methods are also described in more detail in the following sections.

VNTR-Based Typing, Including Multiple Locus VNTR Analysis (MLVA)

The recent sequencing of several complete bacterial genomes has revealed that much of the bacterial genome consists of repeated short nucleotide sequences. The repeats may vary in size, location, and complexity. Repeats can be classified according to their structure, such as inverted, dyad, or direct repeats. In bacteria, the repeat units are typically 1 to 10 nucleotides, although repeating units of up to 100 bp are recognized. These repeating arrays are commonly called variablenumber tandem repeats (VNTRs). Variability occurs both in the sequence of the repeating unit and in the number of repeats between strains. Slipped strand mis­ pairing is the molecular model that best explains the variability of short sequence repeats (102). Stretches of relatively short arrays of repeat units, when being copied by DNA polymerase, may engage in illegitimate base pairing. This forces the polymerase to introduce or delete individual repeat units. VNTRs are often found in noncoding regions controlling gene expression, but they are also found within open reading frames (ORFs) (104, 114). The sequence of the repeat unit and the number of copies repeated are characteristic of each VNTR. For highly clonal microorganisms, there may be little variability for a given VNTR, and multiple loci must be examined. This approach has been termed multiple-locus VNTR analysis (MLVA). In some ways, MLVA may be analogous to MLST, whereby multiple loci must be sequenced to provide sufficient strain discrimination. As typically performed, MLVA determines the number of repeats at a VNTR locus by the size of a PCR amplicon using high-resolution capillary electrophoresis rather than by actual DNA sequencing. In Fig. 43.3a and b, the two isolates differ at three of the four sites shown, as determined by the fragment sizes shown in the electropherograms. The dendrogram in Fig. 43.3c shows the relationship of 10 E. coli O157:H7 isolates based on alleles (copy numbers) at eight sites.

Manila Typesetting Company

11/08/2012 07:22AM

Advanced Techniques in Food Microbiology

1064

Figure 43.3

SMP_Food Microbiology_CH43.indd

1064

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping

1065

Figure 43.3 (Continued.)  Example of MLVA of E. coli O157:H7 isolates. (a and b) Electropherograms from automated DNA sequencer showing different fragment sizes at three of four sites; (c)  dendrogram generated with BioNumerics software showing relatedness of isolates based on data at eight VNTR loci. doi:10.1128/9781555818463.ch43f3

The benefits of MLVA include high throughput, technical simplicity, and objectivity of the data. The interlaboratory reproducibility of the data is excellent as long as the same capillary electrophoresis platform is used. However, there are some significant sizing discrepancies between capillary electrophoresis platforms from different manufacturers due to differences in dye and polymer chemistries employed by each platform. In order to facilitate data comparisons between laboratories with different platforms, fragment-sizing data need to be normalized either by using an allelic ladder during electrophoresis (21) or by employing conversion tables in the data analysis (51). The assays also are highly species and even serotype specific, and prior knowledge about the sequence is required for the protocol development. One of the first, and perhaps one of the most important, applications of MLVA was the subtyping of the highly monomorphic potential agent of bioterrorism Bacillus anthracis (54). It has also proven useful for subtyping two other potential agents of bioterrorism for which other methods had not proven adequate, Yersinia pestis (64) and Francisella tularensis (29). In recent years, MLVA has also been widely used to subtype foodborne microorganisms either as a primary method or as a complementary technique to PFGE. These applications will be discussed below.

DNA Microarrays

A DNA microarray is a molecular platform in which a few hundred to thousands of specific DNA oligonucleotides or short sequences (capture molecules) are bound

SMP_Food Microbiology_CH43.indd

1065

to a matrix. The array is used to detect specific DNA sequences (target sequences) in a test sample of DNA from the microbe being investigated by hybridization to capture molecules followed by detection of the arraybound test DNA. Two types of arrays, planar and liquid, are used. Analysis with planar microarrays is typically performed by deposition of DNA probes complementary to the genome targets of interest on a glass slide (solid matrix). Alternatively, PCR probes may be attached to microspheres that are internally dyed with two spectrally distinct fluorophores and combined to create a suspension array (liquid matrix), such as those seen with the Luminex platform (1). Probes are typically synthetically produced oligonucleotides (9 to 100 bp) or PCR products (100 to 1,000 bp). They can target ORFs amplified from the sequenced reference isolate or short intergenic oligonucleotides based on the available sequence (37). An ORF-based DNA array can detect losses of entire ORFs but does not detect minor deletions, point mutations, deletions restricted to intergenic spaces, genetic rearrangements, deletions of homologous repetitive elements, or gene insertions. Short oligonucleotide arrays are more precise for the detection of shorter nucleotide polymorphisms but require a larger number of probes, at a higher cost. Such oligonucleotide arrays have been developed by several companies including Affymetrix (Santa Clara, CA), Operon (Huntville, AL), Agilent Technologies (Santa Clara, CA), and NimbleGen (Madison, WI). DNA microarrays represent an attractive platform for use in bacterial identification and subtyping because

Manila Typesetting Company

11/08/2012 07:22AM

1066

Advanced Techniques in Food Microbiology

it allows for rapid and accurate analysis of large numbers of different DNA molecules. In addition, the basic design of microarrays (i.e., detection of signal associated with sample DNA bound to fixed DNA probes), eliminates the positional variation associated with gelbased methods, such as PFGE, and makes the analysis of results easier to automate and standardize. Limitations of DNA array technology include the initial high cost for the synthesis and spotting of target-specific primers and for the fluorophores used in labeling the reactions. Ambiguities in the interpretation of the ratios of hybridization and cross-hybridization to analogous targets are also important limitations of the technique. Interlaboratory reproducibility of DNA array-based methods remains still unclear, since multilaboratory validation studies are largely missing. Even though planar arrays allow for the analysis of hundreds of thousands of targets at the same time, sample throughput is an obvious bottleneck of the technology. If the total number of probes can be limited to a few hundred, suspension arrays can offer a higher sample throughput platform with lower cost. Applications of the DNA microarray technology include pathogen identification, detection of virulence and antimicrobial resistance markers, binary typing (detection of presence or absence of genes), single nucleotide polymorphism (SNP) typing, and genome comparisons. Specific examples of these applications to foodborne microbes will be given below.

(Illumina, Inc., San Diego, CA), and the SOLiD instrument (Life Technologies, Foster City, CA). The 454 GSFLX technology is based on pyrosequencing on solid support and can produce read lengths of up to 330 bp (69). The Solexa (6) and SOLiD (103) technologies are based on sequencing by synthesis with cyclic reversible terminators and massively parallel sequencing by ligation, respectively, and can currently generate read lengths of 50  bp (SOLiD) to 100 bp (Solexa) (73). While short reads are useful for detection of substitution polymorphisms against a reference genome, they are more difficult to use for de novo assembly of a new genome, especially if the genome being investigated contains large sections of repeated sequences. At this time, whole-genome sequencing has limited use in comparing large numbers of bacterial isolates of the same species. Therefore, its utility for subtyping purposes has yet to be truly evaluated. Other possible applications of whole-genome sequencing will be discussed in “Future of Molecular Typing of Foodborne Pathogens” below.

Whole-Genome Sequencing

Steady advances in sequencing technology have allowed the transition from labor-intensive, conventionally automated Sanger sequencing of single strains to comparison of multiple strains within the same species or subtype. Sequencing projects on this scale allow the detection of novel chromosomal alterations between isolates at the single nucleotide level, such as SNPs, repetitive elements, recombination, and insertions and deletions. Recent advances in automation and informatics have reduced the amount of time required for generation of whole-genome sequences from years to months or even weeks. Included in these improvements are processes that eliminated the need for subcloning DNA fragments prior to sequencing (25, 30) and completely different second-generation sequencing and imaging technologies (72, 101). In addition, detailed physical maps generated through optical scanning serve as a guide to allow rapid sequence assembly, characterization of gaps, and validation of the finished sequence (56). Some of the next­generation technologies commercially available today include the 454 GS-FLX instrument (Roche Applied Science, Basel, Switzerland), the Solexa 1 G analyzer

SMP_Food Microbiology_CH43.indd

1066

SNPs

A simplified approach to sequence-based subtyping is the detection of SNPs. An SNP is a change in a single nucleotide in one sequence relative to another caused by nucleotide mutation, horizontal gene transfer, or intragenic recombination events. Some SNP systems have been developed as a means of simplifying MLST schemes. They offer phylogenetic information like the MLST schemes on which they are based but with less discrimination and are also incapable of identifying new allelic types. Such SNP schemes are therefore generally not suitable for laboratory-based surveillance focusing on cluster detection (7, 49). However, by selecting SNPs from more variable regions, it has been possible to discriminate within highly clonal bacteria such as B. anthracis and Mycobacterium tuberculosis (35, 84). SNP discovery can be accomplished by sequencing a number of defined target genes with a high rate of polymorphism, such as those associated with antibiotic resistance (58). A more universal approach to identify likely SNPs is to use data provided by genome sequencing studies to generate a resequencing microarray containing probes corresponding to each of the nucleotides of potential gene SNP targets (118). Likely polymorphic loci are identified when DNAs from multiple strains are hybridized with the array. Since the mutation rate in ORFs in some highly clonal organisms is slow due to selective pressure, targeting intergenic regions for detection of SNPs seems to be a better strategy (2). Specific examples on utilizing SNPs for characterization of foodborne microbes will be given below.

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping APPLICATION OF MOLECULAR SUBTYPING METHODS TO SPECIFIC FOODBORNE BACTERIA

Salmonella

Bacteria of the genus Salmonella are so widespread in animals, humans, and the environment that virtually every commonly used subtyping method has been applied to the molecular epidemiology of these bacteria. Although it is not generally thought of as a subtyping method, serotyping of Salmonella is actually a very sensitive subtyping method, with more than 2,500 recognized serotypes. With rare serotypes, simply determining the serotype may be sufficient to at least suggest a source or epidemiologic connection. For example, an outbreak of Salmonella serotype Hartford infections occurred among visitors to a theme park in Orlando, FL, in 1995 (18). Serotype Hartford is a sufficiently rare serotype that further subtyping was not needed to suggest an association with drinking orange juice at the park. The association was further strengthened by the isolation of another rare Salmonella serotype, Gaminara, from the implicated brand of orange juice as well as from a person with a serotype Hartford coinfection. Ironically, the PFGE subtypes of serotype Gaminara from the patient and the orange juice were different, which could have led to erroneous conclusions if the epidemiologic evidence had not been considered. Since maintaining the wide panel of diagnostic antisera that is required to perform serotyping is time-­consuming and difficult, several molecular methods have been developed to predict the serotype. These methods may be grouped into two types: methods that target the ­serotypedetermining genes (rfb, fliC, and fljB), and surrogate methods that provide subtypes that may correlate with the serotype but do not target the serotype-­determining genes. If a method that targets the serotype-­determining genes is used, the system will be compatible with the Kauffman-White scheme, which contains the phenotypic descriptions of all Salmonella serotypes (42). This is a major advantage because serotypes determined by the molecular method will be compatible with historical data sets based on the Kauffmann-White Scheme, and serotypes that were not included in the initial validation of the method will be reliably determined if probes to the serotype antigens expressed by the test strain are present in the assay. Two protocols based on PCR amplification of the serotype-determining genes with subsequent detection of the targets using a DNA array have been published (33, 70, 115). Figure 43.4 illustrates the principle and output of molecular serotyping using a suspension microarray-based assay and conversion of the

SMP_Food Microbiology_CH43.indd

1067

1067

raw data into antigenic formula. The surrogate methods are methods that were developed for other purposes but happen to correlate with the serotype, e.g., PFGE (121) or rep-PCR (112), or methods that target non-serotypespecific sequences present in some but not all serotypes (57). The surrogate methods will reliably identify a serotype only if its molecular profile is already present in the correlation database. Any isolate with an unrecognized profile by a surrogate method will have to be confirmed by a method that utilizes the Kauffmann-White scheme, i.e., traditional serotyping or its aforementioned molecular counterpart. For the more common serotypes such as Enteritidis and Typhimurium, serotyping alone typically does not provide sufficient strain discrimination for source tracking. Some of the earliest descriptions of molecular strain typing involved the application of plasmid profile analysis to Salmonella serotypes Muenchen (100), Typhimurium (48), and Enteritidis (107). Though less widely used today, plasmid profile analysis of Salmonella isolates may still be useful (83). The most widely used subtyping method for Salmonella today, and clearly the current gold standard, is PFGE. PFGE has been successfully used for identifying and investigating outbreaks of serotype Typhimurium (5) and many other serotypes (39). In addition to outbreak investigations, PFGE has been successfully applied to tracking Salmonella in production environments. Liljebjelke et al. used PFGE to reveal that isolates of serotypes Typhimurium and Enteritidis from processed poultry carcasses from one farm were indistinguishable from isolates obtained in the poultry house environment and at a company breeder farm, suggesting vertical as well as horizontal transmission (59). Although PFGE has proven very useful in subtyping Salmonella, other methods have been developed and evaluated. Kotetishvili et al. used MLST to subtype a collection of Salmonella isolates, finding a larger number of subtypes by MLST than by PFGE (55). This study appeared promising but was not designed to evaluate whether MLST could be used to distinguish between outbreak and nonoutbreak isolates of a specific serotype. Such a study was done by Fakhr et al., who determined that MLST was not as discriminating as PFGE when applied only to serotype Typhimurium isolates (28). An interesting variant of MLST, based on sequences of stable temperate phages in serotype Typhimurium, was reported by Ross and Heuzenroeder (87). The initial study revealed that this approach provided greater strain discrimination than standard MLST. In recent years, MLVA has been increasingly used to subtype Salmonella serotypes Typhimurium (63, 84) and

Manila Typesetting Company

11/08/2012 07:22AM

Advanced Techniques in Food Microbiology

1068

Figure 43.4  Molecular identification of serotype in Salmonella by the Luminex platform. The data represent identification of serotype based on reactivity with DNA probes corresponding to specific serotype antigens coupled to fluorescent microspheres. The results are expressed as the ratio of median fluorescent intensity for test versus negative control samples (P/N ratio). (a) Results of the O antigen assay; probes detect sequences specific for common O groups or for serotype Paratyphi A; (b) results of the H antigen assay; probes detect specific H epitopes; (c) interpretation of data based on the Kauffmann-White scheme. (Courtesy J.  McQuiston, CDC.) doi:10.1128/9781555818463.ch43f4

Enteritidis (9, 66). In Denmark and Norway, MLVA has been included in the routine surveillance of Salmonella serotype Typhimurium since 2004 and has proven to be superior to PFGE in both surveillance and outbreak detection (10, 62). Dyet et al. investigated the ability of MLVA to discriminate between different serotype Typhimurium phage types in New Zealand (27). They observed a good discriminatory power among DT104 isolates but limited discrimination among phage types DT101, DT160, and RDNC (reacts but does not conform to a recognized Typhimurium phage type). PulseNet USA also routinely uses MLVA as a complementary technique to PFGE to further characterize PFGE-defined clusters of serotypes Typhimurium and Enteritidis with common PFGE patterns and temporarily associated PFGE-defined clusters with closely related patterns (13, 14). A few other DNA microarray applications in addi­tion to molecular serotyping have also been re-

SMP_Food Microbiology_CH43.indd

1068

cently developed for characterization of Salmonella. A ­ series of universal microarrays were developed by U.S. Department of Agriculture (USDA) scientists for detecting antimicrobial resistance genes in bacterial pathogens (35, 36, 60). The latest version has 1,269 oligonucleotide probes to detect genes commonly associated with antimicrobial resistance and multidrugresistant plasmids of the H1 and IncA/C lineages. An older version of the antimicrobial resistance chip, which consisted of 775 resistance gene probes, was used to study 34 Salmonella isolates from a turkey production facility (120). Pelludat et al. (82) used DNA fragments derived from subtractive hybridization experiments, many of them targeting noncoding regions, and sequences available on the public data banks to construct a prototype binary typing array with 83 targets for Salmonella serotype Typhimurium. The results revealed that arrays with only a limited number of

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping t­ argeted probes may be a valuable tool in discriminating closely related strains.

STEC O157 and Non-O157 STEC

One of the first applications of PFGE to foodborne disease investigations occurred during the large ­hamburgerassociated outbreak of E. coli O157:H7 in the western United States in 1993 (3). The demand for PFGE subtyping to support investigations of E. coli O157:H7 outbreaks was the driving force behind the creation of PulseNet, the National Molecular Subtyping Network for Foodborne Disease Surveillance (98). PFGE continues to be the standard method for subtyping E. coli O157 in PulseNet. A standardized PFGE protocol for the top six most prevalent serotypes of non-O157 Shiga toxin-producing E. coli (STEC) was released by PulseNet in 2009 (http://pulsenetinternational.org/ downloads/pfge/5%201_5%202_5%204_PNetStand_ Ecoli_with_Sflexneri.pdf). The relative clonality of the bacterium limits the applicability of MLST (79), and methods such as AFLP, while showing some potential utility, have received little attention (119). Foley et al. compared PFGE, MLST, and rep-PCR for typing isolates from food, cattle, and humans and concluded that PFGE was the method of choice (34). Multiple partially overlapping MLVA protocols for subtyping STEC O157 have been reported in recent years (39, 63, 78). All reports indicate the ability of MLVA to discriminate between apparently unrelated isolates with indistinguishable PFGE patterns. In the United States, MLVA data has been used in outbreak investigations since 2008 to narrow the case definition and strengthen the association with the vehicle, particularly when the outbreak strain displays a common PFGE pattern and the presumptive exposure is a common one, such as ground beef (13). Also, Dyet et al. reported a better epidemiologic concordance of MLVA data than of PFGE data when they used the PulseNet standardized MLVA protocol to characterize 118 isolates from New Zealand (27). A more generic MLVA protocol covering all pathogenic E. coli strains was recently published by Lindstedt et al. (61). However, this assay appears to have limited discriminatory power, particularly among non-O157 STEC isolates (11). Several technologies have been recently used to identify possibly useful SNPs in STEC O157 genomes. Zhang et al. (118) identified 906 SNPs in 523 chromosomal genes and the large virulence plasmid pO157 of 11 STEC O157 strains of human origin by using comparative genome sequencing microarrays. However, over one-half of the SNPs were present only in the two atypical strains that were included in the study and are

SMP_Food Microbiology_CH43.indd

1069

1069

therefore unlikely to be informative for more common STEC O157 strains. Jackson et al. (52) used a DNA tiling array that interrogated the genomes of 44 strains of human, animal, and food origins by targeting 1,000 bp at each of the 60 evenly spaced locations. These analyses revealed 164 SNPs in the 1% of the genome examined and also provided evidence of the genetic mosaicism among E. coli genomes, as witnessed by the nonuniform distribution of SNPs. Clawson et al. (17) used high-throughput 454 system sequencing of pooled STEC O157 DNA of human and cattle origin from a total of 193 strains. They detected more than 16,000 putative polymorphisms, of which 178 were selected for further study. A minimal set of 32 tag SNPs was capable of identifying 42 unique genotypes among 227 epidemiologically unrelated STEC O157 strains. Although PFGE diversity (154 patterns) surpassed the SNP genotype diversity overall, 10 PFGE patterns each occurred with multiple strains having different SNP genotypes. Overall, much of the SNP research has focused on SNP discovery. The only downstream application published so far is that by Manning et al. (68). They used real-time PCR to target 32 tag SNPs that identified 39 genotypes among 528 clinical isolates and also separated them into nine distinct clades. Interestingly, differences in the severity of the type of clinical disease were observed between the clades.

Listeria monocytogenes

For the past several years, PFGE has also been the gold standard for molecular subtyping of L. monocytogenes. It is the standard method for PulseNet and has contributed to identifying clusters of L. monocytogenes infections and tracking the source of outbreaks (41, 90). It has also been used to track L. monocytogenes strains in processing plants. Gudmundsdottir et al. used PFGE to identify potential sources of contamination of coldsmoked salmon in a processing plant in Iceland (43). Multiple partially overlapping MLVA protocols have been reported for subtyping L. monocytogenes, but none of them appear to provide discriminatory power comparable to that of PFGE, particularly among lineage I isolates (10, 76, 96). MLST was also reported to identify a larger number of subtypes of L. monocytogenes than did PFGE, but this study did not determine the epidemiologic utility of the method (85). Zhang et al. developed a novel approach to subtyping L. monocytogenes based on sequence differences in three virulence genes and three virulence-associated genes (116). They referred to the new method as multivirulence-locus sequence typing (MVLST) because of its similarity to MLST. The MVLST method was able

Manila Typesetting Company

11/08/2012 07:22AM

1070

Advanced Techniques in Food Microbiology

to differentiate strains that had indistinguishable PFGE patterns, but the epidemiologic significance was unclear. Zhang and Knabel later reported an abbreviated version of MVLST in which first a multiplex PCR assay identified L. monocytogenes serotypes 1/2a and 4b and then sequencing of two PCR products provided interstrain discrimination (117). This method is less sensitive than the method based on sequencing of six loci, but it is faster and less expensive in both reagent cost and analyst time. Recently, researchers at the USDA Agricultural Research Service have developed and implemented a sequencing-based subtyping method for surveillance of L. monocytogenes in food (26, 108, 109). The first version of this assay, called multilocus genotyping (MLGT) (26) targeted lineage I strains and identified 60 SNPs, indels, or truncations in 22 genes distributed across the Listeria genome. Multiplex PCR amplification of the targeted genes is followed by the detection of SNPs by allele-specific primer extension with probes attached to a liquid microarray. The method will identify strains belonging to the known epidemic clones, ECI, ECIa, and ECII. Subsequently, the method was expanded with two more MLGT assays targeting lineage II and lineage III and IV strains, respectively (108). In the lineage II assay, seven amplification primer sets and 64 probes are used, and in the lineage III/IV assay five amplification primer sets and 51 probes are used. The MLGT method also detects mutations leading to truncation of the internalin A gene (inlA), which is associated with decreased virulence of the strains. MLGT was more discriminatory for the subtyping of isolates from patients with sporadic listeriosis than MVLST, but slightly less so than PFGE. Most recently, Ward et al. (110) developed a targeted multilocus genotyping assay for rapid identification of the lineage, the serogroup, and the epidemic clone of L. monocytogenes using the same approach as for MLGT. The assay targets six genomic regions with 30 probes. This method does not detect the haplotype-like MLGT and is less discriminatory but is nevertheless useful for rapidly identifying strains of public health importance in support of USDA’s risk-based inspection programs. Borucki et al. (8) have described a 629-probe microarray from a shotgun library that was as discriminatory as PFGE for subtyping of L. monocytogenes in their initial testing. Another microarray derived from a comparative genomic study with 409 probes from the genomes of two L. monocytogenes strains and one Listeria innocua strain also has shown promise (24). Despite the wide array of new methods being reported, PFGE remains the method of choice for highly discriminatory subtyping of Listeria.

Campylobacter jejuni

SMP_Food Microbiology_CH43.indd

1070

Unlike the situation for Salmonella, STEC, and L. monocytogenes, there is no consensus on the method of choice for subtyping C. jejuni. The plasticity of the Campylobacter genome and the relative lack of identifiable outbreaks have led some authors to question the utility of real-time molecular typing (47, 74) for outbreak detection. PFGE was suggested as an epidemiologic tool for C. jejuni as long ago as 1991 (113). It was used to discriminate between outbreak and sporadic isolates of C. jejuni within a single serotype as early as 1994 (97) and continues to be used today (53). Despite these successes, several investigators have expressed concern about the effect of genomic instability on PFGE type. There is clear evidence that genomic rearrangement and genetic exchange between strains alters PFGE patterns in vivo in poultry (19, 111), although at least some strains appear to be genetically stable (67). This lack of stability makes it more difficult to use PFGE to trace the source of C. je­ juni contamination or to establish links between animal sources and human infections. Unfortunately, PFGE is not the only method affected by genetic instability. Another early molecular typing method for C. jejuni was based on amplification of the flagellin gene flaA, followed by restriction digestion of the amplicon and separation of the fragments by gel electrophoresis (77). The sensitivity and specificity of this method were improved by sequencing of the flaA gene, and typing was further simplified by focusing on the short variable region within the gene (71). Both of these approaches are affected by recombination within the flagellin locus (44) and are generally not as sensitive as other available methods. MLST has also been reported to be useful, especially when combined with antigen gene sequence data from targets such as flaA (16, 89), flaB, and porA (23). Some alleles have host specificity, and for that reason MLST has emerged as a tool for microbiological source attribution of campylobacteriosis (20) and for study of the population dynamics and geographic spread of Campylobacter (75, 94).

FUTURE OF MOLECULAR TYPING OF FOODBORNE PATHOGENS With the increasing international trade of food and food animals, a foodborne infection in one country may have its origin in another country, even one on another continent. It is therefore crucial that molecular subtyping methods for foodborne pathogens be harmonized worldwide to facilitate the rapid comparison of strains

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping isolated in different countries. This method harmonization for comparison is best done in the framework of surveillance networks. Such networking systems have been in place for the last decade in Europe in the Salmnet and Enter-net networks (31, 32), in the United States with PulseNet (98), and most recently with the extension of the PulseNet network internationally (99). The WHO Global Foodborne Infections Network (formerly Global Salm-Surv) also focuses on surveillance of foodborne infections internationally, building capacity through training workshops (http://www.who.int/gfn/en/). The future success of these networks will largely depend on the development of new subtyping methods. An ideal subtyping method should type all strains, discriminate between all isolates that are epidemiologically unrelated while identifying all related isolates, and be 100% reproducible, universally applicable, and in­expensive. The results obtained should be definitive, fully portable, and easy to interpret. New methods should correlate with previously used methods so that the data obtained with the new methods can be related to historical data obtained with the previous methods. It is evident that no currently used subtyping method fulfills all these requirements, and it is likely that none ever will. Nevertheless, new methods are constantly being developed and evaluated and bring the science closer to the ideal. In recent years, the main focus of subtyping method development has been on DNA sequence-based methods. Sequence-based approaches to subtyping of bacteria, such as MLVA, are already being widely implemented in the surveillance of foodborne infections. They are attractive for several reasons. Not only do sequence-based methods have the potential to obtain information of utility to surveillance, but by targeting virulence or antimicrobial susceptibility genes it may also be possible to obtain results that are clinically useful. Unlike with restriction pattern-based techniques, differences in the patterns obtained by target-specific techniques can also often be quantified, i.e., it is possible to give a good estimate of the likelihood that one strain is a derivative of another. Even though some sequence-based methods, such as planar microarrays and whole-­genome sequencing, are still prohibitively expensive for routine surveillance, others like MLVA and suspension arrays have decreased in price, and the per-sample cost is now comparable to or even cheaper than that of more conventional methods. Instrumentation for capillary electrophoresis and suspension microarrays is also nowadays widely available in public health laboratories. Because of their specificity, many sequence-based methods may potentially also be used for subtyping microorganisms without obtaining a

SMP_Food Microbiology_CH43.indd

1071

1071

pure culture. This is an important consideration, since nonculture diagnostics is a trend that has already begun and will likely spread due to the fact that it is more costeffective for hospitals to outsource their diagnostics to commercial laboratory diagnostics companies that use mainly amplification-based tests than it is to maintain their own laboratory with culture capabilities. Sequencebased methods may also potentially be partially or fully automated, thus increasing the throughput, reducing the risk of human errors, and possibly reducing the cost of the surveillance. However, commercial availability of such methods is unlikely before the method formats have been agreed upon internationally, either by agreement or simply through widespread use. International subtyping networks such as PulseNet, Enter-Net, and Salm-Net will play central roles in establishing the preferred technology. Work is currently under way within the framework of PulseNet International to harmonize MLVA protocols for STEC O157, Salmonella, and L. monocytogenes. However, it is unlikely that MLVA will replace PFGE as a gold standard subtyping method for foodborne bacteria because the protocols are highly ­serotype specific. As mentioned above, much of the recent work on SNPs has focused on SNP discovery. Practical downstream applications are still largely lacking. Within PulseNet USA and the CDC National Enteric Reference Laboratories, a project is under way to develop a molecular genotyping assay for the clinically most important STEC serotypes. This assay will utilize the suspension microarray technology (Luminex) and will include targets for the top 11 clinically most important O serogroups, the top H types associated with the O groups, virulence factors such as the Shiga toxin genes and their subtypes, and SNPs that were identified in earlier projects (17, 118). Comprehensive downstream applications that will identify multiple pathogens/serotypes and give an estimate of the virulence potential of the target microbe and at least a presumptive genotype all in one or very few PCRs are likely to emerge in the next few years. As whole-genome sequencing becomes increasingly more affordable and rapid with the introduction of ­second-generation sequencing technologies, it will become feasible to generate whole-genome sequences in real time during high-priority outbreak investigations. Gilmour et al. used high-throughput 454 sequencing to rapidly provide the genome sequence of the primary outbreak strain and a strain with a variant PFGE pattern during a large multiprovince listeriosis outbreak that was associated with consumption of ready-to-eat meats (40). Within 3 days after the project commencement, they had draft genome sequences available for ­ comparative

Manila Typesetting Company

11/08/2012 07:22AM

1072

Advanced Techniques in Food Microbiology

a­ nalysis, which enabled them to determine the evolutionary lineages and unequivocally define the full breadth of the genetic variation between the two PFGE-defined outbreak strains. The identified novel markers (indels, prophages, and SNPs) were then used for a rapid assessment of genetic relationships of additional clinical, food, and environmental isolates recovered during the outbreak investigation. However, the feasibility of using whole-genome sequencing as a routine subtyping method has not yet been evaluated. Thorough validation studies using well-defined sets of historical isolates are required to evaluate the level of natural variation in the genome sequence that occurs as a result of in vitro culture manipulations (e.g., single colony picks from a pure culture or serial culture passages) and in vivo during short-term, single-point source outbreaks and longterm, reoccurring single-source outbreaks. The level of expected variation among isolates not related to one another needs to be defined too. Tedious validation studies as described above are necessary in order to properly analyze and interpret whole-genome sequence data from the context of molecular epidemiology. The production of millions of short sequence reads has also challenged the infrastructure of existing information technology systems in terms of data transfer, storage, and quality control and computational analysis to assemble, align, and annotate data. A major investment in bioinformatics will be required before whole-genome sequencing becomes a practical alternative, as currently most public health laboratories and even many reference laboratories do not have capabilities to perform whole-genome sequence data analysis. Predicting the future is always perilous, especially in a field where new technology can suddenly open unforeseen possibilities. It is highly likely that PFGE will remain the gold standard method for subtyping foodborne bacteria for at least the next 5 years because the process of developing and validating new methods is time­consuming and a method that would be as universally applicable as PFGE is yet to emerge. MLVA will play an important role as a complementary second-generation technique with the most important pathogens such as STEC O157 and Salmonella serotypes Typhimurium and Enteritidis. Newer sequence-based third-generation applications targeting multiple species or multiple serotypes within species will continue to emerge and be implemented. While the feasibility of using whole-genome sequencing as a routine subtyping technique remains to be seen, the data generated by it most certainly will provide useful information for future downstream applications. Whole-genome sequencing of strains with solid epidemiologic background information will unveil novel

markers for virulence factors, antimicrobial resistance, and source attribution. While it seems highly likely that some type of sequenced-based subtyping will ultimately replace all the current standards, it is difficult to predict exactly what the format of the new technology will be. Whatever advances new technology provides, it must be remembered that subtyping data will continue to be just one part of the picture and cannot alone replace epidemiologic and environmental investigations.

SMP_Food Microbiology_CH43.indd

1072

References 1. Amon, J. J., R. Devasia, G. Xia, O. V. Nainan, S. Hall, B. Lawson, J. S. Wolthuis, P. D. Macdonald, C. W. Shepard, I. T. Williams, G. L. Armstrong, J. A. Gabel, P. Erwin, L. Sheeler, W. Kuhnert, P. Patel, G. Vaughan, A. Weltman, A. S. Craig, B. P. Bell, and A. Fiore. 2005. Molecular epidemiology of foodborne hepatitis A outbreaks in the United States, 2003. J. Infect. Dis. 192:1323–1330. 2. Barrett, T. J., P. Gerner-Smidt, and B. Swaminathan. 2006. Interpretation of pulsed-field gel electrophoresis patterns in foodborne disease investigations and surveillance. Foodborne Pathog. Dis. 3:20–31. 3. Barrett, T. J., H. Lior, J. H. Green, R. Khakhria, J. G. Wells, B. P. Bell, K. D. Greene, J. Lewis, and P. M. Griffin. 1994. Laboratory investigation of a multistate food-borne outbreak of Escherichia coli O157:H7 by using pulsed-field gel electrophoresis and phage typing. J. Clin. Microbiol. 32:3013–3017. 4. Ben-Darif, E., E. De Pinna, E. J. Threlfall, F. J. Bolton, M. Upton, and A. J. Fox. 2010. Comparison of a semiautomated rep-PCR system and multilocus sequence typing for differentiation of Salmonella enterica isolates. J. Microbiol. Methods 81:11–16. 5. Bender, J. B., C. W. Hedberg, D. J. Boxrud, J. M. Besser, J. H. Wicklund, K. E. Smith, and M. T. Osterholm. 2001. Use of molecular subtyping in surveillance for Salmonella enterica serotype Typhimurium. N. Engl. J. Med. 344:189–195. 6. Bentley, D. R., S. Balasubramanian, H. P. Swerdlow, G.  P. Smith, J. Milton, C. G. Brown, K. P. Hall, D.  J. Evers, C. L. Barnes, H. R. Bignell, J. M. Boutell, J. Bryant, R. J. Carter, R. Keira Cheetham, A. J. Cox, D.  J. Ellis, M. R. Flatbush, N. A. Gormley, S. J. Humphray, L.  J. Irving, M. S. Karbelashvili, S. M. Kirk, H. Li, X. Liu, K.  S. Maisinger, L. J. Murray, B. Obradovic, T. Ost, M.  L. Parkinson, M. R. Pratt, I. M. Rasolonjatovo, M.  T. Reed, R. Rigatti, C. Rodighiero, M. T. Ross, A. Sabot, S. V. Sankar, A. Scally, G. P. Schroth, M. E. Smith, V. P. Smith, A. Spiridou, P. E. Torrance, S. S. Tzonev, E.  H. Vermaas, K. Walter, X. Wu, L. Zhang, M.  D. Alam, C. Anastasi, I. C. Aniebo, D. M. Bailey, I.  R. Bancarz, S. Banerjee, S. G. Barbour, P.  A. Baybayan, V. A. Benoit, K. F. Benson, C. Bevis, P.  J. Black, A. Boodhun, J. S. Brennan, J. A. Bridgham, R. C. Brown, A. A. Brown, D. H. Buermann, A. A. Bundu, J.  C. Burrows, N. P. Carter, N. Castillo, E. C. M. Chiara, S. Chang, R. Neil

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping

7.

8.

9.

10.

11.

12. 13.

14.

15. 16.

Cooley, N. R. Crake, O. O. Dada, K. D. Diakoumakos, B. Dominguez-Fernandez, D. J. Earnshaw, U. C. Egbujor, D. W. Elmore, S. S. Etchin, M. R. Ewan, M. Fedurco, L. J. Fraser, K. V. Fuentes Fajardo, W. Scott Furey, D. George, K. J. Gietzen, C.  P. Goddard, G. S. Golda, P. A. Granieri, D.  E. Green, D. L. Gustafson, N. F. Hansen, K. Harnish, C.  D. Haudenschild, N. I. Heyer, M. M. Hims, J. T. Ho, A. M. Horgan, et al. 2008. Accurate whole human genome sequencing using reversible terminator chemistry. Nature 456:53–69. Best, E. L., A. J. Fox, J. A. Frost, and F. J. Bolton. 2005. Real-time single-nucleotide polymorphism profiling using Taqman technology for rapid recognition of Campylobacter jejuni clonal complexes. J. Med. Microbiol. 54:919–925. Borucki, M. K., S. H. Kim, D. R. Call, S. C. Smole, and F. Pagotto. 2004. Selective discrimination of Listeria monocytogenes epidemic strains by a mixed-genome DNA microarray compared to discrimination by pulsedfield gel electrophoresis, ribotyping, and multilocus sequence typing. J. Clin. Microbiol. 42:5270–5276. Boxrud, D., K. Pederson-Gulrud, J. Wotton, C. Medus, E. Lyszkowicz, J. Besser, and J. M. Bartkus. 2007. Comparison of multiple-locus variable-number tandem repeat analysis, pulsed-field gel electrophoresis, and phage typing for subtype analysis of Salmonella enterica serotype Enteritidis. J. Clin. Microbiol. 45:536–543. Bruun, T., G. Sorensen, L. P. Forshell, T. Jensen, K.  Nygard, G. Kapperud, B. A. Lindstedt, T. Berglund, A. Wingstrand, R. F. Petersen, L. Muller, C. Kjelso, S.  Ivarsson, M. Hjertqvist, S. Lofdahl, and S. Ethelberg. 2009. An outbreak of Salmonella Typhimurium infections in Denmark, Norway and Sweden, 2008. Euro Surveill. 14(10):pii 19147. Bustamante, A. V., A. M. Sanso, P. M. Lucchesi, and A. E. Parma. 2011. Genetic diversity of O157: H7 and non-O157 verocytotoxigenic Escherichia coli from Argentina inferred from multiple-locus variablenumber tandem repeat analysis (MLVA). Int. J. Med. Microbiol. 300:212–217. Carle, G. F., M. Frank, and M. V. Olson. 1986. Electro­ phoretic separations of large DNA molecules by periodic inversion of the electric field. Science 232:65–68. Centers for Disease Control and Prevention. 29 October 2007, posting date. Investigation of outbreak of human infections caused by Salmonella I 4,[5],12:i:-. http://www.cdc.gov/salmonella/4512eyeminus.html. Accessed September 14, 2012. Centers for Disease Control and Prevention. 2009. Multistate outbreak of Salmonella infections associated with peanut butter and peanut butter-containing products—United States, 2008-2009. MMWR Morb. Mortal. Wkly. Rep. 58:85–90. Chu, G., D. Vollrath, and R. W. Dacis. 1986. Separation of large DNA molecules by contour-clamped homogeneous electric fields. Anal. Biochem. 234:1582–1585. Clark, C. G., L. Bryden, W. R. Cuff, P. L. Johnson, F. Jamieson, B. Ciebin, and G. Wang. 2005. Use of the Oxford multilocus sequence typing protocol and ­sequencing of the flagellin short variable region to char-

SMP_Food Microbiology_CH43.indd

1073

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

1073

acterize isolates from a large outbreak of waterborne Campylobacter sp. strains in Walkerton, Ontario, Canada. J. Clin. Microbiol. 43:2080–2091. Clawson, M. L., J. E. Keen, T. P. Smith, L. M. Durso, T. G. McDaneld, R. E. Mandrell, M. A. Davis, and J. L. Bono. 2009. Phylogenetic classification of Escherichia coli O157:H7 strains of human and bovine origin using a novel set of nucleotide polymorphisms. Genome Biol. 10:R56. Cook, K. A., T. E. Dobbs, G. Hlady, J. G. Wells, T. J. Barrett, N. D. Puhr, G. A. Lancette, D. W. Bodager, B. L. Toth, C. A. Genese, A. K. Highsmith, K. E. Pilot, L. Finelli, and D. L. Swerdlow. 1998. Outbreak of Salmonella serotype Hartford infections associated with unpasteurized orange juice. JAMA 280:1504–1509. de Boer, P., J. A. Wagenaar, R. P. Achterberg, J. P. van Putten, L. M. Schouls, and B. Duim. 2002. Generation of Campylobacter jejuni genetic diversity in vivo. Mol. Microbiol. 44:351–359. de Haan, C. P., R. I. Kivisto, M. Hakkinen, J. Corander, and M. L. Hanninen. 2010. Multilocus sequence types of Finnish bovine Campylobacter jejuni isolates and their attribution to human infections. BMC Microbiol. 10:200. de Valk, H. A., J. F. Meis, S. Bretagne, J. M. Costa, B. A. Lasker, S. A. Balajee, A. C. Pasqualotto, M. J. Anderson, L. Alcazar-Fuoli, E. Mellado, and C. H. Klaassen. 2009. Interlaboratory reproducibility of a microsatellite-based typing assay for Aspergillus fumig­ atus through the use of allelic ladders: proof of concept. Clin. Microbiol. Infect. 15:180–187. Dingle, K. E., F. M. Colles, D. R. Wareing, R. Ure, A. J. Fox, F. E. Bolton, H. J. Bootsma, R. J. Willems, R. Urwin, and M. C. Maiden. 2001. Multilocus sequence typing system for Campylobacter jejuni. J. Clin. Microbiol. 39:14–23. Dingle, K. E., N. D. McCarthy, A. J. Cody, T. E. Peto, and M. C. Maiden. 2008. Extended sequence typing of Campylobacter spp., United Kingdom. Emerg. Infect. Dis. 14:1620–1622. Doumith, M., C. Cazelet, N. Simoes, L. Frangeul, C. Jacquet, F. Kunst, P. Martin, P. Cossart, P. Glaser, and C. Buchrieser. 2004. New aspects regarding evolution and virulence of Listeria monocytogenes revealed by comparative genomics and DNA arrays. Infect. Immun. 72:1072–1083. Dressman, D., H. Yan, G. Traverso, K. W. Kinzler, and B. Vogelstein. 2003. Transforming single DNA molecules into fluorescent magnetic particles for detection and enumeration of genetic variations. Proc. Natl. Acad. Sci. USA 100:8817–8822. Ducey, T. F., B. Page, T. Usgaard, M. K. Borucki, K. Pupedis, and T. J. Ward. 2007. A single-nucleotidepolymorphism-based multilocus genotyping assay for subtyping lineage I isolates of Listeria monocytogenes. Appl. Environ. Microbiol. 73:133–147. Dyet, K. H., I. Robertson, E. Turbitt, and P. E. Carter. 2011. Characterization of Escherichia coli O157:H7 in New Zealand using multiple-locus variable-number ­tandem-repeat analysis. Epidemiol. Infect. 139:464–471.

Manila Typesetting Company

11/08/2012 07:22AM

Advanced Techniques in Food Microbiology

1074 28. Fakhr, M. K., L. K. Nolan, and C. M. Logue. 2005. Multilocus sequence typing lacks the discriminatory ability of pulsed-field gel electrophoresis for typing Salmonella enterica serovar Typhimurium. J. Clin. Microbiol. 43:2215–2219. 29. Farlow, J., K. L. Smith, J. Wong, M. Abrams, M. Lytle, and P. Keim. 2001. Francisella tularensis strain typing using multiple-locus, variable-number tandem repeat analysis. J. Clin. Microbiol. 39:3186–3192. 30. Fedurco, M., A. Romieu, S. Williams, I. Lawrence, and G. Turcatti. 2006. BTA, a novel reagent for DNA attachment on glass and efficient generation of solidphase amplified DNA colonies. Nucleic Acids Res. 34: e22. 31. Fisher, I. S. 1999. The Enter-net international surveillance network—how it works. Euro Surveill. 4:52–55. 32. Fisher, I. S. 1995. Salm-Net: a network for human Salmonella surveillance in Europe. Euro Surveill. Sep:7–8. 33. Fitzgerald, C., M. Collins, S. van Duyne, M. Mikoleit, T. Brown, and P. Fields. 2007. Multiplex, bead-based suspension array for molecular determination of common Salmonella serogroups. J. Clin. Microbiol. 45:3323–3334. 34. Foley, S. L., S. Simiee, J. Meng, D. G. White, P. F. McDermott, and S. Zhao. 2004. Evaluation of molecular typing methods for Escherichia coli O157:H7 isolates from cattle, food, and humans. J. Food Prot. 67:651–657. 35. Frye, J. G., T. Jesse, F. Long, G. Rondeau, S. Porwollik, M. McClelland, C. R. Jackson, M. Englen, and P. J. Fedorka-Cray. 2006. DNA microarray detection of antimicrobial resistance genes in diverse bacteria. Int. J. Antimicrob. Agents 27:138–151. 36. Frye, J. G., R. L. Lindsey, G. Rondeau, S. Porwollik, F. Long, M. McClelland, C. R. Jackson, M. D. Englen, R. J. Meinersmann, M. E. Berrang, J. A. Davis, J.  B. Barrett, J. B. Turpin, S. N. Thitaram, and P. J. Fedorka-Cray. 2010. Development of a DNA microarray to detect antimicrobial resistance genes identified in the National Center for Biotechnology Information database. Microb. Drug Resist. 16:9–19. 37. Garaizar, J., A. Rementeria, and S. Porwollik. 2006. DNA microarray technology: a new tool for the epidemiological typing of bacterial pathogens? FEMS Immunol. Med. Microbiol. 47:178–189. 38. Gardiner, K., W. Laas, and D. Patterson. 1986. Fractio­ nation of large mammalian DNA restriction fragments using vertical pulsed-field gradient gel electrophoresis. Somat. Cell. Mol. Genet. 12:185–195. 39. Gerner-Smidt, P., K. Hise, J. Kincaid, S. Hunter, S. Rolando, E. Hyytia-Trees, E. M. Ribot, and B.  Swaminathan. 2006. PulseNet USA: a five-year update. Foodborne Pathog. Dis. 3:9–19. 40. Gilmour, M. W., M. Graham, G. Van Domselaar, S. Tyler, H. Kent, K. M. Trout-Yakel, O. Larios, V. Allen, B. Lee, and C. Nadon. 2011. High-throughput genome sequencing of two Listeria monocytogenes clinical isolates during a large foodborne outbreak. BMC Genomics 11:120.

SMP_Food Microbiology_CH43.indd

1074

41. Graves, L. M., S. B. Hunter, A. R. Ong, D. SchoonmakerBopp, K. Hise, L. Kornstein, W. E. DeWitt, P. S. Hayes, E. Dunne, P. Mead, and B. Swaminathan. 2005. Microbiological aspects of the investigation that traced the 1998 outbreak of listeriosis in the United States to contaminated hot dogs and establishment of molecular subtyping-based surveillance for Listeria monocy­ togenes in the PulseNet network. J. Clin. Microbiol. 43:2350–2355. 42. Grimont, P. A. D., and F.-X. Weill. 2007. Antigenic formulae of the Salmonella serovars, 9th ed. WHO Collaborating Centre for Reference and Research on Salmonella, Institut Pasteur, Paris, France. 43. Gudmundsdottir, S., B. Gudbjornsdottir, H. L. Lauzon, H. Einarsson, K. G. Kristinsson, and M. Kristjansson. 2005. Tracing Listeria monocytogenes isolates from cold-smoked salmon and its processing environment in Iceland using pulsed-field gel electrophoresis. Int. J. Food Microbiol. 101:41–51. 44. Harrington, C. S., F. M. ThomsonCarter, and P. E. Carter. 1997. Evidence for recombination in the flagellin locus of Campylobacter jejuni: implications for the flagellin gene typing scheme. J. Clin. Microbiol. 35:2386–2392. 45. Hassan, W. M., S. Y. Wang, and R. D. Ellender. 2005. Methods to increase fidelity of repetitive extragenic palindromic PCR fingerprint-based bacterial source tracking efforts. Appl. Environ. Microbiol. 71:512–518. 46. Healy, M., J. Huong, T. Bittner, M. Lising, S. Frye, S. Raza, R. Schrock, J. Manry, A. Renwick, R. Nieto, C. Woods, J. Versalovic, and J. R. Lupski. 2005. Microbial DNA typing by automated repetitive-sequence-based PCR. J. Clin. Microbiol. 43:199–207. 47. Hedberg, C. W., K. E. Smith, J. M. Besser, D. J. Boxrud, T. W. Hennessy, J. B. Bender, F. A. Anderson, and M. T. Osterholm. 2001. Limitations of pulsed-field gel electrophoresis for the routine surveillance of Campy­ lobacter infections. J. Infect. Dis. 184:242–244. 48. Holmberg, S. D., I. K. Wachsmuth, F. W. HickmanBrenner, and M. L. Cohen. 1984. Comparison of plasmid profile analysis, phage typing, and antimicrobial susceptibility testing in characterizing Salmonella Typhimurium isolates from outbreaks. J. Clin. Micro­ biol. 19:100–104. 49. Hommais, F., S. Pereira, C. Acquaviva, P. EscobarParamo, and E. Denamur. 2005. Single-nucleotide polymorphism phylotyping of Escherichia coli. Appl. Environ. Microbiol. 71:4784–4792. 50. Horby, P. W., S. J. O’Brien, G. K. Adak, C. Graham, J. I. Hawker, P. Hunter, C. Lane, A. J. Lawson, R. T. Mitchell, M. H. Reacher, E. J. Threlfall, and L. R. Ward. 2003. A national outbreak of multi-resistant Salmonella enterica serovar Typhimurium definitive phage type (DT) 104 associated with consumption of lettuce. Epidemiol. Infect. 130:169–178. 51. Hyytia-Trees, E., P. Lafon, P. Vauterin, and E. M. Ribot. 2010. Multilaboratory validation study of standardized multiple-locus variable-number tandem repeat analysis protocol for Shiga toxin-producing Escherichia coli O157: a novel approach to normalize fragment

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

size data between capillary electrophoresis platforms. Foodborne Pathog. Dis. 7:129–136. Jackson, S. A., M. K. Mammel, I. R. Patel, T. Mays, T. J. Albert, J. E. LeClerc, and T. A. Cebula. 2007. Interrogating genomic diversity of E. coli O157:H7 using DNA tiling arrays. Forensic Sci. Int. 168:183–199. Karagiannis, I., T. Sideroglou, K. Gkolfinopoulou, A. Tsouri, D. Lampousaki, E. N. Velonakis, E. V. Scoulica, K. Mellou, T. Panagiotopoulos, and S. Bonovas. 2010. A waterborne Campylobacter je­ juni outbreak on a Greek island. Epidemiol. Infect. 138:1726–1734. Keim, P., L. B. Price, A. M. Klevytska, K. L. Smith, J. M. Schupp, R. Okinaka, P. J. Jackson, and M. E. HughJones. 2000. Multiple-locus variable-number tandem repeat analysis reveals genetic relationships within Bacillus anthracis. J. Bacteriol. 182:2928–2936. Kotetishvili, M., O. C. Stine, A. Kreger, J. G. Morris, Jr., and A. Sulakvelidze. 2002. Multilocus sequence typing for characterization of clinical and environmental Salmonella strains. J. Clin. Microbiol. 40:1626–1635. Latreille, P., S. Norton, B. S. Goldman, J. Henkhaus, N. Miller, B. Barbazuk, H. B. Bode, C. Darby, Z. Du, S. Forst, S. Gaudriault, B. Goodner, H. GoodrichBlair, and S. Slater. 2007. Optical mapping as a routine tool for bacterial genome sequence finishing. BMC Genomics 8:321. Leader, B. T., J. G. Frye, J. Hu, P. J. Fedorka-Cray, and D. S. Boyle. 2009. High-throughput molecular determination of Salmonella enterica serovars by use of multiplex PCR and capillary electrophoresis analysis. J. Clin. Microbiol. 47:1290–1299. Levy, D. D., B. Sharma, and T. A. Cebula. 2004. Single-nucleotide polymorphism mutation spectra and resistance to quinolones in Salmonella enterica serovar Enteritidis with a mutator phenotype. Antimicrob. Agents Chemother. 48:2355–2263. Liljebjelke, K. A., C. L. Hofacre, T. Liu, D. G. White, S. Ayers, S. Young, and J. J. Maurer. 2005. Vertical and horizontal transmission of Salmonella within integrated broiler production system. Foodborne Pathog. Dis. 2:90–102. Lindsey, R. L., J. G. Frye, P. J. Fedorka-Cray, T. J. Welch, and R. J. Meinersmann. 2011. An oligonucleotide microarray to characterize multidrug resistant plasmids. J. Microbiol. Methods 81:96–100. Lindstedt, B. A., L. T. Brandal, L. Aas, T. Vardund, and G. Kapperud. 2007. Study of polymorphic ­variablenumber of tandem repeats loci in the ECOR collection and in a set of pathogenic Escherichia coli and Shigella isolates for use in a genotyping assay. J. Microbiol. Methods 69:197–205. Lindstedt, B. A., M. Torpdahl, E. M. Nielsen, T. Vardund, L. Aas, and G. Kapperud. 2007. Harmo­ nization of the multiple-locus variable-number tandem repeat analysis method between Denmark and Norway for typing Salmonella Typhimurium isolates and closer examination of the VNTR loci. J. Appl. Microbiol. 102:728–735.

SMP_Food Microbiology_CH43.indd

1075

1075

63. Lindstedt, B. A., T. Vardund, L. Aas, and G. Kapperud. 2004. Multiple-locus variable-number tandem-repeats analysis of Salmonella enterica subsp. enterica serovar Typhimurium using PCR multiplexing and multicolor capillary electrophoresis. J. Microbiol. Methods 59:163–172. 64. Lowell, J. L., D. M. Wagner, B. Atshabar, M. F. Antolin, A. J. Vogler, P. Keim, M. C. Chu, and K. L. Gage. 2005. Identifying sources of human exposure to plague. J. Clin. Microbiol. 43:650–656. 65. Lu, Z., D. Lapen, A. Scott, A. Dang, and E. Topp. 2005. Identifying host sources of fecal pollution: diversity of Escherichia coli in confined dairy and swine production systems. Appl. Environ. Microbiol. 71:5992–5998. 66. Malorny, B., E. Junker, and R. Helmuth. 2008. Multilocus variable-number tandem repeat analysis for outbreak studies of Salmonella enterica serotype Enteritidis. BMC Microbiol. 8:84. 67. Manning, G., B. Duim, T. Wassenaar, J. A. Wagenaar, A. Ridley, and D. G. Newell. 2001. Evidence for a genetically stable strain of Campylobacter jejuni. Appl. Environ. Microbiol. 67:1185–1189. 68. Manning, S. D., A. S. Motiwala, A. C. Springman, W. Qi, D. W. Lacher, L. M. Ouellette, J. M. Mladonicky, P. Somsel, J. T. Rudrik, S. E. Dietrich, W. Zhang, B. Swaminathan, D. Alland, and T. S. Whittam. 2008. Variation in virulence among clades of Escherichia coli O157:H7 associated with disease outbreaks. Proc. Natl. Acad. Sci. USA 105:4868–-4873. 69. Margulies, M., M. Egholm, W. E. Altman, S. Attiya, J. S. Bader, L. A. Bemben, J. Berka, M. S. Braverman, Y. J. Chen, Z. Chen, S. B. Dewell, L. Du, J. M. Fierro, X. V. Gomes, B. C. Godwin, W. He, S. Helgesen, C. H. Ho, G. P. Irzyk, S. C. Jando, M. L. Alenquer, T. P. Jarvie, K. B. Jirage, J. B. Kim, J. R. Knight, J. R. Lanza, J. H. Leamon, S. M. Lefkowitz, M. Lei, J. Li, K. L. Lohman, H. Lu, V. B. Makhijani, K. E. McDade, M. P. McKenna, E. W. Myers, E. Nickerson, J. R. Nobile, R. Plant, B. P. Puc, M. T. Ronan, G. T. Roth, G. J. Sarkis, J. F. Simons, J. W. Simpson, M. Srinivasan, K. R. Tartaro, A. Tomasz, K. A. Vogt, G. A. Volkmer, S. H. Wang, Y. Wang, M. P. Weiner, P. Yu, R. F. Begley, and J. M. Rothberg. 2005. Genome sequencing in microfabricated high-density picolitre reactors. Nature 437:326–327. 70. McQuiston, J. R., R. J. Waters, B. A. Dinsmore, M. L. Mikoleit, and P. I. Fields. 2010. Molecular determination of H-antigens of Salmonella using a microsphere based liquid array. J. Clin. Microbiol. 49:565–573. 71. Meinersmann, R. J., L. O. Helsel, P. I. Fields, and K.  L. Hiett. 1997. Discrimination of Campylobacter jejuni isolates by fla gene sequencing. J. Clin. Microbiol. 35: 2810–2814. 72. Metzker, M. L. 2005. Emerging technologies in DNA sequencing. Genome Res. 15:1767–1776. 73. Metzker, M. L. 2010. Sequencing technologies—the next generation. Nat. Rev. Genet. 11:31–46. 74. Michaud, S., S. Menard, and R. D. Arbeit. 2005. Role of real-time molecular typing in the surveillance of

Manila Typesetting Company

11/08/2012 07:22AM

Advanced Techniques in Food Microbiology

1076

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

Campylobacter enteritis and comparison of pulsed-field gel electrophoresis profiles from chicken and human isolates. J. Clin. Microbiol. 43:1105–1111. Miller, W. G., M. D. Englen, S. Kathariou, I. V. Wesley, G. Wang, L. Pittenger-Alley, R. M. Siletz, W. Muraoka, P. J. Fedorka-Cray, and R. E. Mandrell. 2006. Identification of host-associated alleles by multilocus sequence typing of Campylobacter coli strains from food animals. Microbiology 152:245–255. Murphy, M., D. Corcoran, J. F. Buckley, M. O’Mahony, P. Whyte, and S. Fanning. 2007. Development and application of multiple-locus variable number of tandem repeat analysis (MLVA) to subtype a collection of Listeria monocytogenes. Int. J. Food Microbiol. 115:187–194. Nachamkin, I., K. Bohachick, and C. M. Patton. 1993. Flagellin gene typing of Campylobacter jejuni by restriction fragment length polymorphism analysis. J. Clin. Microbiol. 31:1531–1536. Noller, A. C., M. C. McEllistrem, and L. H. Harrison. 2004. Genotyping primers for fully automated multilocus variable-number tandem repeat analysis of Escherichia coli O157:H7. J. Clin. Microbiol. 42:3908. Noller, A. C., M. C. McEllistrem, O. C. Stine, J. G. Morris, Jr., D. J. Boxrud, B. Dixon, and L. H. Harrison. 2003. Multilocus sequence typing reveals a lack of diversity among Escherichia coli O157:H7 isolates that are distinct by pulsed-field gel electrophoresis. J. Clin. Microbiol. 41:675–679. Olsvik, O., H. Sorum, K. Birkness, K. Wachsmuth, M. Fjolstad, J. Lassen, K. Fossum, and J. C. Feeley. 1985. Plasmid characterization of Salmonella Typhimurium transmitted from animals to humans. J. Clin. Microbiol. 22:336–338. Parveen, S., R. L. Murphree, L. Edmiston, C. W. Kaspar, K. M. Portier, and M. L. Tamplin. 1997. Association of multiple-antibiotic-resistance profiles with point and nonpoint sources of Escherichia coli in Apalachicola Bay. Appl. Environ. Microbiol. 63:2607–2612. Pelludat, C., R. Prager, H. Tschape, W. Rabsch, J.  Schuchhardt, and W. D. Hardt. 2005. Pilot study to evaluate microarray hybridization as a tool for Salmonella enterica serovar Typhimurium strain differentiation. J. Clin. Microbiol. 43:4092–4106. Pezzoli, L., R. Elson, C. L. Little, H. Yip, I. Fisher, R. Yishai, E. Anis, L. Valinsky, M. Biggerstaff, N.  Patel, H. Mather, D. J. Brown, J. E. Coia, W. van Pelt, E. M. Nielsen, S. Ethelberg, E. de Pinna, M. D. Hampton, T. Peters, and J. Threlfall. 2008. Packed with Salmonella—investigation of an international outbreak of S. Senftenberg infection linked to contamination of prepacked basil in 2007. Foodborne Pathog. Dis. 5:661–668. Ramisse, V., P. Houssu, E. Hernandez, F. Denoeud, V.  Hilaire, O. Lisanti, F. Ramisse, J. D. Cavallo, and G. Vergnaud. 2004. Variable number of tandem repeats in Salmonella enterica subsp. enterica for typing purposes. J. Clin. Microbiol. 42:5722–5730. Revazishvili, T., M. Kotetishvili, O. C. Stine, A. S. Kreger, J. G. Morris, Jr., and A. Sulakvelidze. 2004. Comparative analysis of multilocus sequence typing

SMP_Food Microbiology_CH43.indd

1076

86.

87.

88.

89.

90.

91.

92.

93.

94.

95. 96.

97.

and pulsed-field gel electrophoresis for characterizing Listeria monocytogenes strains isolated from environmental and clinical sources. J. Clin. Microbiol. 42:276–285. Riley, L. W., R. S. Remis, S. D. Helgerson, H. B. McGee, J. G. Wells, B. R. Davis, R. J. Hebert, E. S. Olcott, L. M. Johnson, N. T. Hargrett, P. A. Blake, and M. L. Cohen. 1983. Hemorrhagic colitis associated with a rare Escherichia coli serotype. N. Engl. J. Med. 308:681–685. Ross, I. L., and M. W. Heuzenroeder. 2005. Discri­ mination within phenotypically closely related definitive types of Salmonella enterica serovar Typhimurium by the multiple amplification of phage locus typing technique. J. Clin. Microbiol. 43:1604–1611. Roussel, S., B. Felix, C. Colaneri, M. L. Vignaud, T. T. Dao, M. Marault, and A. Brisabois. 2010. Semi­automated repetitive-sequence-based polymerase chain reaction compared to pulsed-field gel electrophoresis for Listeria monocytogenes subtyping. Foodborne Pathog. Dis. 7:1005–1012. Sails, A. D., B. Swaminathan, and P. I. Fields. 2003. Utility of multilocus sequence typing as an epidemiological tool for investigation of outbreaks of gastroenteritis caused by Campylobacter jejuni. J. Clin. Microbiol. 41:4733–4739. Sauders, B. D., E. D. Fortes, D. L. Morse, N. Dumas, J. A. Kiehlbauch, Y. Schukken, J. R. Hibbs, and M.  Wiedmann. 2003. Molecular subtyping to detect human listeriosis clusters. Emerg. Infect. Dis. 9:672–680. Savelkoul, P. H. M., H. J. M. Aarts, J. deHaas, L. Dijkshoorn, B. Duim, M. Otsen, J. L. W. Rademaker, L. Schouls, and J. A. Lenstra. 1999. Amplified-­fragment length polymorphism analysis: the state of an art. J.  Clin. Microbiol. 37:3083–3091. Schwartz, D. C., W. Saffran, J. Welsh, R. Haas, M. Goldenberg, and C. R. Cantor. 1983. New techniques for purifying large DNAs and studying their properties and packaging. Cold Spring Harb. Symp. Quant. Biol. 47(Pt. 1):189–195. Selander, R. K., D. A. Caugant, H. Ochman, J. M. Musser, M. N. Gilmour, and T. S. Whittam. 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51:873–884. Sheppard, S. K., F. Colles, J. Richardson, A. J. Cody, R. Elson, A. Lawson, G. Brick, R. Meldrum, C. L. Little, R. J. Owen, M. C. Maiden, and N. D. McCarthy. 2010. Host association of Campylobacter genotypes transcends geographic variation. Appl. Environ. Microbiol. 76:5269–5277. Southern, E. M. 1979. Analysis of restriction-fragment patterns from complex deoxyribonucleic acid species. Biochem. Soc. Symp. 44:37–41. Sperry, K. E., S. Kathariou, J. S. Edwards, and L. A. Wolf. 2008. Multiple-locus variable-number tandemrepeat analysis as a tool for subtyping Listeria monocy­ togenes strains. J. Clin. Microbiol. 46:1435–1450. Suzuki, Y., M. Ishihara, M. Saito, N. Ishikawa, and T. Yokochi. 1994. Discrimination by means of pulsed-field

Manila Typesetting Company

11/08/2012 07:22AM

43.  Molecular Source Tracking and Molecular Subtyping

98.

99.

100.

101. 102. 103.

104. 105.

106.

107.

108.

109.

gel electrophoresis between strains of Campylobacter jejuni Lior type 4 derived from sporadic cases and from outbreaks of infection. J. Infect. 29:183–187. Swaminathan, B., T. J. Barrett, S. B. Hunter, R. V. Tauxe, and the CDC PulseNet Task Force. 2001. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg. Infect. Dis. 7:382–389. Swaminathan, B., P. Gerner-Smidt, L. K. Ng, S. Lukinmaa, K. M. Kam, S. Rolando, E. P. Gutierrez, and N. Binsztein. 2006. Building PulseNet International: an interconnected system of laboratory networks to facilitate timely public health recognition and response to foodborne disease outbreaks and emerging foodborne diseases. Foodborne Pathog. Dis. 3:36–50. Taylor, D. N., I. K. Wachsmuth, Y. H. Shangkuan, E. V. Schmidt, T. J. Barrett, J. S. Schrader, C. S. Scherach, H. B. McGee, R. A. Feldman, and D. J. Brenner. 1982. Salmonellosis associated with marijuana: a multistate outbreak traced by plasmid fingerprinting. N. Engl. J. Med. 306:1249–1253. Tomkinson, A. E., S. Vijayakumar, J. M. Pascal, and T. Ellenberger. 2006. DNA ligases: structure, reaction mechanism, and function. Chem. Rev. 106:687–699. Torres-Cruz, J., and M. W. van der Woude. 2003. Slippedstrand mispairing can function as a phase variation mechanism in Escherichia coli. J. Bacteriol. 185:6990–6994. Valouev, A., J. Ichikawa, T. Tonthat, J. Stuart, S.  Ranade, H. Peckham, K. Zeng, J. A. Malek, G. Costa, K. McKernan, A. Sidow, A. Fire, and S. M. Johnson. 2008. A high-resolution, nucleosome position map of C. elegans reveals a lack of universal sequencedictated positioning. Genome Res. 18:1051–1063. van Belkum, A. 1999. Short sequence repeats in microbial pathogenesis and evolution. Cell. Mol. Life Sci. 56:729–734. Versalovic, J., T. Koeuth, and J. R. Lupski. 1991. Distri­ bution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res. 19:6823–6831. Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. van de Lee, M. Hornes, A. Frijters, J. Pot, J. Peleman, M.  Kuiper, et al. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res. 23:4407–4414. Wachsmuth, I. K., J. A. Kiehlbauch, C. A. Bopp, D.  N. Cameron, N. A. Strockbine, J. G. Wells, and P. A. Blake. 1991. The use of plasmid profiles and nucleic acid probes in epidemiologic investigations of foodborne, diarrheal diseases. Int. J. Food Microbiol. 12:77–89. Ward, T. J., T. F. Ducey, T. Usgaard, K. A. Dunn, and J. P. Bielawski. 2008. Multilocus genotyping assays for single nucleotide polymorphism-based subtyping of Listeria monocytogenes isolates. Appl. Environ. Microbiol. 74:7629–7642. Ward, T. J., P. Evans, M. Wiedmann, T. Usgaard, S. E. Roof, S. G. Stroika, and K. Hise. 2010. Molecular and

SMP_Food Microbiology_CH43.indd

1077

110.

111.

112.

113.

114.

115.

116. 117.

118.

119.

120.

121.

1077

phenotypic characterization of Listeria monocytogenes from U.S. Department of Agriculture Food Safety and Inspection Service surveillance of ready-to-eat foods and processing facilities. J. Food Prot. 73:861–869. Ward, T. J., T. Usgaard, and P. Evans. 2010. A targeted multilocus genotyping assay for lineage, serogroup, and epidemic clone typing of Listeria monocytogenes. Appl. Environ. Microbiol. 76:6680–6684. Wassenaar, T. M., B. Geilhausen, and D. G. Newell. 1998. Evidence of genomic instability in Campylobacter jejuni isolated from poultry. Appl. Environ. Microbiol. 64:1816–1821. Wise, M. G., G. R. Siragusa, J. Plumblee, M. Healy, P. J. Cray, and B. S. Seal. 2009. Predicting Salmonella enterica serotypes by repetitive sequence-based PCR. J. Microbiol. Methods 76:18–24. Yan, W., N. Chang, and D. E. Taylor. 1991. Pulsedfield gel electrophoresis of Campylobacter jejuni and Campylobacter coli genomic DNA and its epidemiologic application. J. Infect. Dis. 163:1068–1072. Yeramian, E., and H. Buc. 1999. Tandem repeats in complete bacterial genome sequences: sequence and structural analyses for comparative studies. Res. Microbiol. 150:745–754. Yoshida, C., K. Franklin, P. Konczy, J. R. McQuiston, P. I. Fields, J. H. Nash, E. N. Taboada, and K. Rahn. 2007. Methodologies towards the development of an oligonucleotide microarray for determination of Salmonella serotypes. J. Microbiol. Methods 70:261–271. Zhang, W., B. M. Jayarao, and S. J. Knabel. 2004. Multi-virulence-locus sequence typing of Listeria mono­ cytogenes. Appl. Environ. Microbiol. 70:913–920. Zhang, W., and S. J. Knabel. 2005. Multiplex PCR assay simplifies serotyping and sequence typing of Listeria monocytogenes associated with human outbreaks. J. Food Prot. 68:1907–1910. Zhang, W., W. Qi, T. J. Albert, A. S. Motiwala, D.  Alland, E. K. Hyytia-Trees, E. M. Ribot, P. I. Fields, T. S. Whittam, and B. Swaminathan. 2006. Probing genomic diversity and evolution of Escherichia coli O157 by single nucleotide polymorphisms. Genome Res. 16: 757–767. Zhao, S. H., S. E. Mitchell, J. H. Meng, S. Kresovich, M. P. Doyle, R. E. Dean, A. M. Casa, and J. W. Weller. 2000. Genomic typing of Escherichia coli O157:H7 by semi-automated fluorescent AFLP analysis. Microbes Infect. 2:107–113. Zou, W., J. G. Frye, C. W. Chang, J. Liu, C. E. Cerniglia, and R. Nayak. 2009. Microarray analysis of antimicrobial resistance genes in Salmonella enterica from preharvest poultry environment. J. Appl. Microbiol. 107:906–914. Zou, W., W. J. Lin, S. L. Foley, C. H. Chen, R. Nayak, and J. J. Chen. 2010. Evaluation of pulsed-field gel electrophoresis profiles for identification of Salmonella serotypes. J. Clin. Microbiol. 48:3122–3126.

Manila Typesetting Company

11/08/2012 07:22AM

SMP_Food Microbiology_CH43.indd

1078

Manila Typesetting Company

11/08/2012 07:22AM

Index

A

Abiotics, 953 Abortive bacteriophage infection (abi) systems, 834–835 AbrB protein, in sporulation, 50 Abrin, intentional contamination with, 94 Absence, microbiological criteria established as, 85 acaA gene, Listeria monocytogenes, 506 Accessibility, in CARVER+Shock strategy, 98–100 Accessory cholera enterotoxin (Ace), 412 Acetaldehyde in fermented milk products, 827–831 in fruit and vegetable treatment, 194 in winemaking, 925, 933 Acetic acid antimicrobial action, 767–769 in bakery products, 216 in cocoa fermentation, 882, 884, 886 Escherichia coli O157:H7 effects, 290 in meat products, 132 in metabolism, 7 in milk fermentation, 829–831 in muscle food decontamination, 123 in muscle food spoilage, 117, 118 in poultry processing, 137

in vegetable fermentation, 843, 847 in winemaking, 926, 931 Acetic acid bacteria, in cocoa fermentation, 884–888 Acetobacter antimicrobial action on, 778 as beer contaminant, 910 in cocoa fermentation, 885, 886 in wine, 932–934 Acetobacter aceti in cocoa fermentation, 887 in winemaking, 932–933 Acetobacter ascendens, 885 Acetobacter ghanensis, 887 Acetobacter loviensis, 885 Acetobacter pasteurianus, 932–933 antimicrobial action on, 773 in cocoa fermentation, 887 Acetobacter rancens, 885 Acetobacter senegalensis, 887 Acetobacter xylinum, 885 Acetoin in milk fermentation, 831 in muscle food spoilage, 117 in wine, 933–934 a-Acetolactate synthase, 831 Acetone, in muscle food spoilage, 117 Acetyl coenzyme A, in brewing, 906

Achromobacter, 154, 205, 866 Acid(s), see also specific acids organism growth and, see also Acid tolerance response Campylobacter, 265 Clostridium botulinum, 452–453 Escherichia coli, 290 hepatitis A virus, 631 noroviruses, 632 organic, 766–773 Salmonella, 229–231 Yersinia, 344 Acid tolerance response, 14, 32 in antimicrobial resistance, 33 Campylobacter, 265 Clostridium botulinum, 804 Clostridium perfringens, 467 Escherichia coli, 28 genomics, 990–991 Salmonella, 233–234, 768 Acidic calcium sulfate, for meat products, 132 Acidification, in food processing, controlled, 803–804 Acidified sodium chlorite, for poultry processing, 137, 277 Acidomonas, in wine, 932 Acid-shock proteins, 14

1079

SMP_Food Microbiology_Index.indd

1079

Manila Typesetting Company

11/08/2012 07:24AM

Index

1080 Acinetobacter antimicrobial action on, 771, 776 in muscle foods, 126, 127 in nuts, 205 in poultry, 134 in seafood, 154, 155 Acinetobacter baumannii, 115 Acinetobacter hydrophila, 145 Acinetobacter lwoffii, 145 Aconitine, intentional contamination with, 94 Acquired immunodeficiency syndrome, see Human immunodeficiency virus infection Acremonium, 205 Acrolein, in wine, 931 ActA protein, Listeria monocytogenes, 525, 527–528 “Actidione agar,” 911 Actin, Listeria monocytogenes, 527–528 Actin rockers, Shigella, 386 Actinomycetes, in wine spoilage, 934 Activated lactoferrin, 774 Active packaging meat products, 134 muscle foods, 125 N-Acyl homoserine lactones, 11–12, 325 Ada protein, in acid tolerance response, 14 Adaptation, to antimicrobials, 32–34 Additive effects, in hurdle technology, 17–18 Adenoviruses, 620, 632 Adenylate cyclase Vibrio cholerae, 411–412 Vibrio vulnificus, 425 Adherence factors Escherichia coli, 288 Vibrio cholerae, 415 Adhesins, Yersinia enterocolitica, 355–357 Adhesion Campylobacter, 271 Cronobacter, 329 Listeria monocytogenes, 524 Adipose tissue, spoilage, 126 Adlupulone, 783 ADP-ribosylation factors, Vibrio cholerae, 412 Aeration, in winemaking, 922 Aerobactin, 324, 390, 429 Aerobic bacteria, metabolism, 3 Aerobic exposure, Campylobacter, 265 Aerobic packaging, for meat, 132–133 Aerococcus, in seafood, 155 Aeromonas antimicrobial action on, 771 bioluminescence, 12 in milk, 173 in modified-atmosphere packaging, 747 in muscle foods, 119, 126 in poultry, 139 in seafood, 141, 149, 150, 153 Aeromonas hydrophila antimicrobial action on, 775, 780–781 refrigeration effects, 746 versus Vibrio furnisii, 427 Affymetrix GenChip platform, 985–986 Afl genes, Aspergillus, 600 Aflatoxins, 205, 598–603 control, 601–602 detection, 599, 601

SMP_Food Microbiology_Index.indd

distribution, 601 ecology, 600 genetics, 600 sources, 598–600 structures, 598 synthesis, 598–599 toxicity, 600 types, 598 aggR gene, Escherichia coli, 288 Agitation, heat transfer and, 740 Agmatine, in poultry spoilage, 139 Agr protein, Staphylococcus aureus, 554–556 Agriculture, antimicrobial usage in, 25–26 Agrobacterium, 153 AI-1 protein, Vibrio cholerae, 416 Aichi virus, 620 AIDS, see Human immunodeficiency virus infection Ail protein, 351 Air chilling, poultry carcasses, 135–136 Airline food Salmonella outbreaks, 240 Shigella outbreaks, 383 Akabane virus, 121 l-Alanine, in spore germination, 61 Alaria americana, 698, 706 Albert-Mafart model, 999 Albert-Schaffner model, 1001 Alcaligenes in muscle foods, 126 in seafood, 155 Alcohol(s), see also Ethanol in milk spoilage, 177 in muscle food spoilage, 117 Aldolase, in fermentation, 4–5 Alicyclobacillus, 70, 71 heat resistance, 65 in juices, 72 Alicyclobacillus acidoterrestris, antimicrobial action on, 771 Alkaline phosphatase, as indicator, 89 Allicin, 782–783 Allochthonous microflora, 951 Allyl cyanide, in vegetable fermentation, 848 Allyl isocyanate, 783 Almonds contamination, 206, 209–211, 235–236 microflora, 204–205 processing, 204, 212–213 Salmonella in, 235–236 Aloe vera extract, 784 Alpha amanitin, intentional contamination with, 94 Alta 2341 bacteriocin product, 806 Alternaria antimicrobial action on, 770, 772 in cheese, 180 in fruits and vegetables, 188, 194 in grains, 213 in nuts, 205 in wine spoilage, 934 Alternaria alternata, 192–193, 195 Alternaria citri, 192 Alteromonas putrefaciens, 145 Amanitin, intentional contamination with, 94 Ami protein, Listeria monocytogenes, 524

1080

Manila Typesetting Company

Amino acids as bacteriocins, 804–805 in beer, 906 in cheese flavor, 833 in fish fermentation, 865, 866 in milk fermentation, 831–833 in muscle food spoilage, 117 in wine, 925, 930 Aminobutane, in fish fermentation, 866 Aminoglycosides, 20–21 for Cronobacter infections, 331 resistance to, 245 Aminopeptidases, in fermented dairy products, 832–833 Ammonia, in muscle food spoilage, 117 Amoebiasis, 727 Amoxicillin, 21 Amoxicillin-clavulanic acid, 28 Ampicillin, 21 for Cronobacter infections, 326–327, 331 resistance to, 28, 29, 120, 384–385 Amplification fragment length polymorphism, 267–268, 1062–1063 Anaerobic organisms, see also specific organisms metabolism, 3, 5–6 in normal microflora, 952 Anaerobic respiration, 5 Angiostrongylus, 698, 702 Angiostrongylus cantonensis, 704–705 Angiostrongylus costaricensis, 703, 704 Animals, see also Meat and meat products; Poultry antimicrobials in, 25–26 resistance development, 244–245; see also Antimicrobial resistance bacteriophages for, 815 feed for bovine spongiform encephalopathy agent in, 656–664 Escherichia coli in, 293 Fusarium toxins in, 608–609 probiotic cultures in, 955–956 Salmonella in, 235 as reservoirs bovine spongiform encephalopathy agent, see Bovine spongiform encephalopathy Campylobacter, 269–271 Clostridium botulinum, 442–443 Cryptosporidium, 715–716 Cyclospora, 720 Escherichia coli, 291–293 Giardia, 724–725 Listeria monocytogenes, 510 monitoring, 586 Salmonella, 234–236 sheep liver fluke, 702–703 Staphylococcus aureus, 557 Trichinella, 673–682 Yersinia, 345–346 Anisakis, 698–700 Ankylosing spondylitis, after Salmonella infection, 244 Anoxybacillus flavithermus, in dairy product spoilage, 72 Antagonism, in hurdle technology, 18 Anthrax, see Bacillus anthracis

11/08/2012 07:24AM

Index

1081

Antibiosis, in fruit and vegetable spoilage control, 196 Antibiotics Resistance Genes Database, 23–24 Antibodies norovirus, 627–628 rapid detection assays based on, 680, 684–685 Taenia saginata, 686 Trichinella, 679–680 Antigen(s) Cronobacter, 312–313 Escherichia coli, 287 Salmonella, 227–228 Staphylococcus aureus, 548–551 Antigen-presenting cells, 551–552 Antimicrobial(s) bactericidal, 20, 22 versus bacteriocins, 804, 813 bacteriostatic, 20 for disinfection, see Disinfectants and disinfectant action food, see Food antimicrobials in food products, 25–26 mechanisms of action of, 21–22 natural, 32 types, 20–22 Antimicrobial resistance, 19–44 alternate metabolic pathways in, 23 amoxicillin-clavulanic acid, 28 ampicillin, 28–29 amplification, 30–31 Bacillus licheniformis, 37 Bacillus subtilis, 22 Campylobacter, 25, 27, 31, 35, 37, 120, 273–274 Carnobacterium, 34 ceftiofur, 28, 35 ceftriaxone, 28 cephalosporins, 27, 120 cephalothin, 28, 34 chloramphenicol, 28, 34, 120 chlortetracycline, 29 ciprofloxacin, 27–28 Clostridium botulinum, 34 commensal bacteria, 34–35 control, 35–37 coselection in, 33–34 cross-resistance, 33–34 definition, 32 development, 24–25 disinfectant, 32 dissemination, 30–31 economics of, 19 enrofloxacin, 27 Enterobacteriaceae, 33–35 Enterococcus, 28, 30–31, 34–35 Enterococcus faecalis, 31, 34, 120 Enterococcus faecium, 25, 31, 120 erythromycin, 27 Escherichia coli, 33, 35, 291 Escherichia coli O157:H7, 25, 28–29, 31, 33 evolution, dissemination, 30–31 fluoroquinolones, 27, 37 food chain impact, 25–31 genetic causes, 24–25 gentamicin, 28 global trends, 35–37

SMP_Food Microbiology_Index.indd

Haemophilus influenzae, 22 inactivation mechanism in, 23 innate, 22, 28 kanamycin, 28 lactic acid bacteria, 34–35 Lactobacillus plantarum, 31, 34 Lactococcus, 32–34 Listeria monocytogenes, 28, 31–32 mechanisms, 22–24, 245 Megasphaera elsdenii, 31 methicillin, 28 monensin, 29 nalidixic acid, 28 nisins, 32, 34 pathogens containing, 26–31 penicillins, 28 persistence, 30–31 Pseudomonadales, 37 Pseudomonas, 30, 34 quinupristin-dalfopristin, 31 recommendations for, 37–38 Salmonella, 25–27, 31, 35, 172, 244–245 Salmonella enterica, 26, 29, 33–34, 120, 172 sanitizer, 32 Shigella, 27–28, 384–385 Staphylococcus aureus, 19, 29, 32 Streptococcus mutans, 34 Streptococcus thermophilus, 30 streptomycin, 27, 120 sulfamethazine, 29 sulfamethoxazole, 28, 120 sulfisoxazole, 28 terminology, 19–20 tetracyclines, 27–28, 120 trimethoprim-sulfamethoxazole, 28 tylosin, 27, 29 usage in food, 25–26, 31–33, 120–121 vancomycin, 19, 120 Vibrio, 31 virginiamycin, 29 Yersinia, 25, 31 Antisera Salmonella, 228 Vibrio, 407 Antitoxin, Clostridium botulinum, 448 Ants, helminth transmission, 705 Antwi model, 1007 Apicomplexa, 713–721 Appendicitis, versus Yersinia infection, 345 Appertization, 63, 753 Apple cider, Escherichia coli O157:H7 in, 298 Applied Biosystems SOLiD system, 976–978 Appresorium formation, 191 Appropriate level of protection, 1024, 1034 Aquaculture antimicrobial use in, 26, 34–35 description, 141–142 safety challenges in, 120 Salmonella in, 234–235 shrimp, 148–149 Arbitrarily primed PCR, 229 Arbitrary Units, bacteriocin activity, 804 Arboviruses, 121 Archaea, in vegetable fermentation, 845 Arcobacter, 119, 263 Arcobacter butzleri, 768 Arginine, in wine, 925, 930

1081

Manila Typesetting Company

Arginine decarboxylase system, for acid resistance, 14 Arginine-dependent system, Escherichia coli, 290 Aroma compounds in cocoa products, 886–887 in fermented dairy products, 829–831 in fish sauces, 865 Arrhenius model, 1001 Arthrobacter, 826 Arthrobacter viscosus, 312 Arthropathy Salmonella, 244 Shigella, 387 Yersinia, 344, 362–363 Artificial intelligence, for muscle food spoilage, 118 Artificial neural networks, 118, 999, 1001, 1004 ARTMAP, for muscle food spoilage, 118 Asaia, in wine, 932 Ascaris lumbricoides, 698, 707 Ascaris suum, 707 Ascorbic acid in meat fermentation, 863 in vegetable fermentation, 848 in winemaking, 917 Ascorbigen, 848 Aseptic methods, 64 Aseptic packaging, 745 Aseptic processing, 744–745 Asian taeniasis, 689–690 Aspergillus antimicrobial action on, 768, 770, 772, 775, 781 as beer contaminant, 909 in cheese, 180 in coffee fermentation, 895 colors, 599–601 enumeration, 601 in grains, 213 identification, 601 water activity requirements, 750 in wine spoilage, 934–935 Aspergillus carbonarius, 603–606, 934 Aspergillus flavus aflatoxins, 598–603 antimicrobial action on, 770, 778, 782, 784 characteristics, 598 detection, 601 genome, 612 growth, 598–599 in nuts, 205, 209 water activity requirements, 750 Aspergillus glaucus (Eurotium), 205, 750 Aspergillus hydrophila, 784 Aspergillus minisclerotigenes, 599 Aspergillus nidulans, 784 Aspergillus niger, 603–604 antimicrobial action on, 782 in fermentation, 5–6 in fruits and vegetables, 193, 196 fumonisin in, 608 genome, 612 in winemaking, 934 Aspergillus nomius, 209, 599 Aspergillus ochraceus, 603–604, 750 Aspergillus oryzae, 784

11/08/2012 07:24AM

Index

1082 Aspergillus parasiticus aflatoxins, 598–603 antimicrobial action on, 770, 782 growth, 598–599 in nuts, 209 water activity requirements, 750 Aspergillus steynii, 603 Aspergillus tubingensis, 934 Aspergillus wenti, 5 Asporogeny, 48–52 Assault tactic, Escherichia coli, 290 Astronaut food, 1040–1041 Astroviruses, 620, 623, 625 Asymmetric sporulation, 48 Atmosphere bacterial growth and control, 747 fruits, 193–194 poultry packaging, 139 vegetables, 193–194 modified, see Modified-atmosphere packaging ATP (adenosine triphosphate) in bacteriocin action, 810 in meat fermentation, 871–872 in microbial metabolism, 3–7 in organic acid action, 766–767 atp gene, Salmonella, 233 ATPase, in metabolism, 5, 7 Attachment and effacing lesions, Escherichia coli, 288, 299–300 Attiéké, 318 Attributes sampling plans, 84 Augustin-Carlier model, 1003–1004 Aureobasidium pullulans in fruit and vegetables spoilage control, 196 in wine, 919 aut gene, Listeria monocytogenes, 525 Auto protein, Listeria monocytogenes, 525 Autochthonous microbiota, 951 Autoimmune sequelae, infections Campylobacter, 273 Salmonella, 244 Shigella, 384 Yersinia, 344, 362–364 Autoinduction, Vibrio cholerae, 13, 416 Autolysis, yeast, 927, 935 Autoxidation, in muscle food spoilage, 117 Avidin, 780 Azoxystrobin, 194 Aztreonam, 21

B

Bacillary dysentery, see Shigella, disease Bacillus antimicrobial action on, 768, 771–772, 775, 780–781, 784 as beer contaminant, 910 in cocoa fermentation, 886 in coffee fermentation, 895 in fermented fish products, 865, 866 in fruits and vegetables, 190 heat resistance, 58 mechanisms of action, 953 in milk, 173 in muscle foods, 119, 126 in nuts, 205 as probiotics, 73

SMP_Food Microbiology_Index.indd

psychrotrophic, 71 in seafood, 155 spores, 45, 48, 53, 71 bacteriocin action on, 807 macromolecules, 53 public health significance, 67 water content, 59–60 in vegetable fermentation, 846 in wine spoilage, 934 Bacillus anthracis, 491–492 enterotoxin, 498 intentional contamination with, 91, 94 spores, 46, 52, 63, 65, 67–68 subtyping, 1065–1066 transmission, 121 Bacillus badius, 934 “Bacillus botulinus,” 442 Bacillus brevis, 884 Bacillus cereus, 491–502 antimicrobial action on, 775, 781, 782, 784 bacteriocin action on, 812 in bakery product spoilage, 72 characteristics, 491–492 in cocoa fermentation, 884 in dairy product spoilage, 72 disease, 491, 493–494 emetic toxin, 495, 498–499 enterotoxins, 67–68, 495–499 in fermented fish products, 865 in grains, 217 growth, 68, 746 heat resistance, 65 hemolysins, 495–499 infectious dose, 494–495 in milk, 178–179, 743 in muscle foods, 119 in nuts, 208 osmoregulation, 15 pathogenicity, 495–499 reservoirs, 492 in seafood, 155 spores, 52, 62, 68, 498 susceptible population, 494–495 toxins, 62–63, 67–68, 495–499 virulence factors, 495–499 Bacillus cinerea, 192–196 Bacillus circulans, 934 in cocoa fermentation, 884 in milk, 179 Bacillus coagulans, 884 antimicrobial action on, 775 in canned foods, 72 heat resistance, 742 heat treatment, 749 in meat spoilage, 72 in milk, 179 spores, 71 in wine spoilage, 934 Bacillus cytotoxicus, 494 Bacillus firmus, 884 Bacillus laterosporus, 884 Bacillus licheniformis antimicrobial resistance, 37 bacteriocin action on, 812 in cocoa fermentation, 884 in fermented fish products, 865 heat resistance, 65, 742 in meat fermentation, 875–876

1082

Manila Typesetting Company

in milk, 179 spores, 65, 68 Bacillus macerans, 884 Bacillus megaterium bacteriology, 67–68 in cocoa fermentation, 884 in fermented fish products, 865 heat resistance, 65 in milk, 179 spores, 62 in wine spoilage, 934 Bacillus mycoides, 178, 491–492 Bacillus nivea, 770 Bacillus pallidus, 179 Bacillus pasteurii, 884 Bacillus polymyxa, 65, 178, 884 Bacillus pseudomycoides, 491–492 Bacillus pumilus, 65, 68, 72, 865, 884 Bacillus sporothermodurans, 179 Bacillus stearothermophilus antimicrobial action on, 771, 773, 775 in cocoa fermentation, 884 in meat fermentation, 875 spores, 71 Bacillus subtilis antimicrobial action on, 768, 773, 786 antimicrobial resistance, 22 bacteriocin action on, 812 in bakery product spoilage, 72 in cocoa fermentation, 884 in fermented fish products, 865 in grains, 216 heat resistance, 65 heat treatment, 749 as indicator, 724–725 in milk, 179–180 spores, 48 chemical resistance, 57–58 coat, 46 formation, 49–51 germination, 60–62 heat resistance, 58 illness related to, 68 macromolecules, 53–54 public health significance, 65 resistance, 55–60 studies, 46 subtilin production, 805 temperature effects, 17 viable but nonculturable, 10 in wine spoilage, 934 Bacillus thermoacidurans, 65, 71 Bacillus thuringiensis, 491–492, 498 enterotoxins, 497 spores, 52, 67 toxin, 813 in winemaking, 934 Bacillus weihenstephanensis, 491–492, 495, 498 Bacitracin, 21 Backward elimination procedure, in models, 1000 BacSim model, 1011 Bacteremia Vibrio alginolyticus, 430 Vibrio hollisae, 428–429 Vibrio vulnificus, 422–423 Yersinia enterocolitica, 345 Bacterial overgrowth, probiotics for, 956

11/08/2012 07:24AM

Index

1083

Bacterial translocation, Cronobacter, 329–330 Bacteriocin(s), 804–814 activity, 805–806 Campylobacter, 277 characteristics, 804–805 for cheese defects, 180 classification, 804–805 in fermented meat products, 119 food applications, 806–808 genetics, 808–810 in hurdle technology, 18 Listeria monocytogenes inhibition, 510 in meat fermentation, 869 mechanism of action, 810–811 methodological considerations, 805–806 versus other antimicrobial proteins, 811–812 regulatory status, 813–814 resistance to, 812–813 spore cell effects, 811–812 starter cultures, 808 structures, 804–805 vegetative cell effects, 810–811 Bacteriocinlike inhibitory substances (BLIS), 804 Bacteriophages in antimicrobial resistance, 25 in biofilms, 13 in biopreservation, 814–816 Campylobacter, 277 cocktail, 816 Cronobacter, 317 Escherichia coli, 293 lactic acid bacteria, 834–836 in vegetable fermentation, 846–847 Bacteriostatic antimicrobials, 20, 22 Bacteroides antimicrobial resistance, 31 in normal microflora, 952 Bacteroides fragilis, 632, 784 Bactofugation, 180 Bagoong, 866 Bakery products, 215–217 Bacillus subtilis in, 72 Salmonella in, 243 spoilage, 72 Balantidium coli, 714, 726–727 Balao balao, 866 Balbakwa, 866 Bambermycin, 22 Baranyi-Roberts model, 998–999, 1001, 1006, 1012 Barrier films, muscle food packaging, 125 Baseline, of microbiota, 88 Basil, 781–782 Baskets, cocoa fermentation in, 883 Bavaricins, 510, 805 Baylisascaris procyonis, 698, 708 Bean death, in cocoa fermentation, 890 Beef tapeworm (Taenia saginata), 682–686, 689–690 Beer bottles for, 912 conditioning, 907 definition, 901 filtration, 907–908 history, 901 hop flower acids in, 783

SMP_Food Microbiology_Index.indd

light, 912 low- and non-alcoholic, 912 pasteurization and packaging, 908 production, see Brewing spoilage, 903 water removal from, 912 Beetles, helminth transmission, 705 Behavioral factors, in foodborne illness, 586–587 Bench trim, 130 Benchmark, for control measures, 88 Benzimidazoles, for fruits and vegetables, 194 Benzoic acid and benzoates antimicrobial action, 769 in apples, 193 for meat products, 132 BetAB protein, Campylobacter, 265 Betaine transport, in osmoregulation, 15 b-Lactam antibiotics, 20–21 b-Lactamase action, 23 Cronobacter, 331 inhibitors, 20, 23 Bidlas-Lambert model, 1004 Bifidobacterium antimicrobial resistance, 30, 34–35 health benefits, 953 prebiotic stimulation, 965 probiotic action, 966 taxonomy, 959 Bifidobacterium animalis subsp. lactis, 961 Bifidobacterium infantis, 959 Bifidobacterium longum, 964 Bifidobacterium longum subsp. infantis, 965 Bigelow model, 1001, 1012 Bile salt hydrolase, Listeria monocytogenes, 529 Bile tolerance, 990–993 Binary fission, 7 Biochip (microarray) technology, see DNA microarray analysis Bioenergetics, 6–7 Biofilms, 13–14 in aquaculture, 142 Cronobacter, 325 muscle surfaces, 114–115 Vibrio cholerae 408–409, 416 Yersinia enterocolitica, 353 Biogenic amines in muscle food spoilage, 116–118, 139, 142–148 in winemaking, 930 Bioinformatics, in genomics, 979–980 Biological control, fruit and vegetable spoilage, 196 Biological hazards, 119, 1045 Biological hurdles, meat processing, 122 Bioluminescence, 12 Biopreservation, see also Bacteriocin(s) bacteriophages, 814–816 controlled acidification, 803–804 definition, 803 Biopsy, muscle, in trichinellosis, 679–680 Biosensor tests, for muscle food spoilage, 118 Bioterrorism, see Intentional contamination Biotin, avidin binding to, 780 Biotracing, 122

1083

Manila Typesetting Company

Biotyping, Campylobacter, 267–269 Bismuth sulfate agar, Salmonella, 226 Bitter acids, hop flower, 783 Bitterness, in milk spoilage, 174 Blachan, 866 blaCMY-2 gene, in antimicrobial resistance, 27, 35 BLAST software, 980 Bleach treatment, bovine spongiform encephalopathy, 654 Bleachon, 866 BLIS (bacteriocinlike inhibitory substances), 804 Blood transfusions, bovine spongiform encephalopathy agent in, 659 Blown pack spoilage, 127 Bluetongue virus, 121 Bogoong, 866 Boiling water, in poultry processing, 136 Bolton broth, Campylobacter, 274–276 Bone taint, in muscle food spoilage, 127 Borneol, 781 Botrytis, 11, 190, 770 Botrytis allii, 196 Botrytis cinerea, 921 in fruits and vegetables, 188, 190–191 in wine spoilage, 934–935 Bottoms-up approach, for models, 1008 “Botulinum cook,” 64 Botulism, see Clostridium botulinum Bovine ephemeral fever, 121 Bovine spongiform encephalopathy, 130, 651–674 agent bodily distribution, 654–656 characteristics, 652–653 contamination with, 654, 657–658 detection, 664–665 in muscle foods, 120 prevalence, 665 screening for, 665 stability, 654 animal studies, 655 atypical, 653 in cattle, 656–660 characteristics, 651, 659–660 epidemiology, 657 history, 651 in humans, 660, 663–665 incubation period, 653 natural history, 660 outbreaks, 656–659 pathogenesis, 654–656 pathology, 660 regulations, 660–664 related prion diseases, 651 sporadic, 665 transmission, 654, 656, 658, 663 typical (classic), 652–653 Boxes, cocoa fermentation in, 883–884 Brain lesions, bovine spongiform encephalo­ pathy, 653, 655–656, 658–660, 664 Brazil nuts, organisms in, 207 Breast milk, powdered formula mixed with, 315–316, 318 Brettanomyces, 770, 909, 927 antimicrobial action on, 772–773 in wine spoilage, 931, 936 Brettanomyces claussenii, 885

11/08/2012 07:24AM

Index

1084 Brevibacterium in nuts, 205 in seafood, 155 Brevibacterium linens, 826 Brewing, 901–913 continuous, 912–913 equipment sanitizing in, 911–912 fermentation, 904–907 high-gravity, 908–909 history, 901 hops in, 783, 904 malting, 901–903 mashing, 903–904 microbial contaminants, 909–912 post-fermentation treatments, 907–908 wort production, 903–904 yeast for, 905–906 Brilliance CampyCount agar, 276 Brochothrix, 781, 859 Brochothrix thermosphacta antimicrobial action on, 775 in muscle foods, 115–118, 126, 127 in poultry, 138, 139 Browning, during meat cooking, 133 Brucella in milk, 172, 743 in muscle foods, 119 Bsh protein, Listeria monocytogenes, 529 BtlA protein, Listeria monocytogenes, 529 Buchanan model, 999 Budu, 865 Burkholderiales, antimicrobial resistance, 37 Burong isda, 866 Burong mustasa, 842 2,3-Butanediol, in milk fermentation, 831 Butter as growth medium, 172 Listeria monocytogenes in, 515 pH, 172 rancidity, 176 salted, 172 Buttiauxella, in muscle foods, 115 Butyl paraben, 777–778 t-Butylhydroperoxide, spore resistance to, 57–58 Butyric acid, in cocoa fermentation, 886 Byssochlamys, 71, 770 Byssochlamys nivea, 769

C

C1 esterase inhibitor, Escherichia coli, 301 Cabbage, fermentation, 844–848 CadA protein, Shigella, 391 Cadaverine, in muscle food spoilage, 117, 139, 142, 144–145 CadF protein, Campylobacter, 271 Cadherins, Listeria monocytogenes, 521 Caffeic acid, 784 Caffeine, 784 CAI-1 protein, Vibrio cholerae, 416 Calcium in fruit and vegetable spoilage control, 194 in lipase production, 174 in protease production, 175 Yersinia, 362 Yersinia enterocolitica dependency, 355 Calcium hypochlorite, in grain milling, 214 Calcium lactate, 769

SMP_Food Microbiology_Index.indd

Calcium propionate antimicrobial action, 770 for cereal products, 215 Caliciviridae, 620–621; see also Noroviruses Caliciviruses, in seafood, 152 Camphene, 781 Camphor, 781 Campy-Cefex agar, modified, Campylobacter, 275–276 Campylobacter, 263–286 antimicrobial action on, 771 antimicrobial resistance, 25–27, 31, 35, 37, 120, 273–274 bacteriology, 265–266 characteristics, 263–265 culture, 274–276 decimal reduction time, 265 detection, 274–276 disease, 272–274 autoimmune sequelae, 273 characteristics, 271–274 epidemiology, 263, 578–580, 588–589 outbreaks, 267, 269–271, 585 surveillance for, 585 diversity, 266–267 in fermented meat products, 862 flagella, 271 genomics, 266–267, 271–272 growth, 265 host associations, 267 immunity, 273 infective dose, 272 in milk, 172 in muscle foods, 114, 119, 120, 123 pathogenicity, 271–272 photoluminescence, 12 in poultry, 137, 140 probiotic protection from, 955 radiation susceptibility, 265 reservoirs, 269–271 in seafood, 140, 153 source associations, 267–269 subtyping, 267, 1062–1063 taxonomy, 264 thermotolerant, 264 toxins, 272 viable but nonculturable, 10 virulence factors, 271–272 Campylobacter coli, 264 antimicrobial resistance, 27, 274 culture, 276 disease, 272–274 environmental susceptibility, 265 in muscle foods, 126 pathogenicity, 271 in seafood, 153 toxins, 272 Campylobacter concisus, 264 Campylobacter curvus, 264 Campylobacter fetus, 264 Campylobacter gracilis, 264 Campylobacter hominis, 264 Campylobacter hyointestinalis, 153, 264 Campylobacter jejuni antimicrobial action on, 768, 772, 775, 780–781 antimicrobial resistance, 27, 273–274

1084

Manila Typesetting Company

bacteriology, 265–266 in biofilms, 13–14 characteristics, 263–265 disease, 272–274 autoimmune sequelae, 273 characteristics, 271–274 epidemiology, 263 outbreaks, 267, 269–271 diversity, 266–267 environmental susceptibility, 265 genomics, 266–267 growth, 265 host associations, 267 immunity, 273 invasion, 271 in milk, 743 in muscle foods, 126 in poultry, 136, 138 reservoirs, 269–271 in seafood, 153 source associations, 267–269 subspecies, 264 subtyping, 267, 1070 toxins, 272 viable but nonculturable, 265 virulence factors, 271–272 Campylobacter lari, 264 environmental susceptibility, 265 pathogenicity, 271 Campylobacter rectus, 264 Campylobacter showae, 264 Campylobacter sputorum, 264 Campylobacter upsaliensis, 264–265 pathogenicity, 271 in seafood, 153 Campylobacteraceae, 263–264 Cancer esophageal, 609 liver, 600 probiotics in, 956 Candida antimicrobial action on, 770, 773–775, 781, 786 as beer contaminant, 909 in cheese, 180 in cocoa fermentation, 885 in milk fermentation, 826 in poultry, 138 in vegetable fermentation, 846 in wine spoilage, 936 in winemaking, 919–920, 923, 925–927 Candida albicans, 782–783 Candida boidinii, 885 Candida cantorelli, 936 Candida claussenii, 865 Candida cocoai, 885 Candida colliculosa, 920 Candida famata, 870 Candida guilliermondii, 885 Candida intermedia, 885 Candida krusei, 885 Candida oleophila, 196 Candida pseudotropicalis, 15, 750 Candida reukaufii, 885 Candida saitoana, 196 Candida stellata, 920, 921 Candida zemplinina, 921 Candida zeylanoides, 887

11/08/2012 07:24AM

Index

1085

Canned foods Clostridium botulinum in, 449–451 low-acid, preservation, 63–65 sporeformers in, 63–65, 70–72 Capillaria philippinensis, 698, 701–702 Capillary electrophoresis, 976 Caproic acid, 772 Caprylic acid, 317–318, 772 Capsular polysaccharides, Campylobacter, 266, 272 Capsule, Vibrio vulnificus, 424 Carbadox, 22 Carbapenems, 20 Carbohydrates, fermentation in fish, 865 in meat products, 871–872 in winemaking, see Winemaking, fermentation in Carbon, sources in milk, 170 Carbon dioxide in brewing, 906 in cocoa fermentation, 883–884 in milk fermentation, 826–829 in muscle food packaging, 125 in packaging, 193, 747 scavengers of, for active packaging, 134 in vegetable fermentation, 847 in winemaking, 917, 925, 931, 935–936 Carbon monoxide packaging, 133–134 Carbonic maceration, grape juice, 917 Carbonyl cyanide m-chlorophenylhydrazone, 22 Carboxyhemoglobin, 133–134 Carcass decontamination meat, 123, 130–131 poultry, 136–138 Cardinal parameter models, 999, 1000, 1003–1004 Carnobacterium antimicrobial action on, 775, 781 in seafood, 155 Carnobacterium divergens antimicrobial action on, 779 in muscle foods, 115 Carnobacterium maltaromaticum, 115 Carnobacterium piscicola, 510, 808, 869 Carnobacterium viridans, 773 Carriage, see also Colonization Escherichia coli, 292 Listeria monocytogenes, 513–514 Shigella, 380 Staphylococcus aureus, 557 Vibrio cholerae, 408 Carvacrol, 781, 782, 786 CARVER+Shock strategy, for intentional contamination, 98–100 Carvone, for fruits and vegetables, 194 Case-control studies, in epidemiology, 578, 585 Casein degradation, 174 hydrolysis, 831 as prebiotic, 966 proteolysis, 170, 832 Cashews, organisms in, 207 Cat tapeworms, 705 Catabolism glucose, 3–5 pentose, 4–5

SMP_Food Microbiology_Index.indd

Catalase-peroxidase Escherichia coli, 301 in meat fermentation, 869 Catechins, 784, 892 Cathepsins, in fish fermentation, 865 Cattle Escherichia coli in, 291–292 spongiform encephalopathy, see Bovine spongiform encephalopathy Taenia saginata in, 682–686 Cecropins, 812 Cedecea lapagei, 145 Cedecea neteri, 145 Ceftiofur, 21, 35 Ceftriaxone, 28 Cefuroxime, 21 Cell membrane, pressure effects on, 756 Cell signaling, 11–13 Center for Meat Process Validation, 1014 Centers for Disease Control and Prevention, antimicrobial resistance monitoring, 35 Centrifugation, in winemaking, 922 Cephalosporins, 20–21 for Cronobacter infections, 331 for enteric fever, 244 resistance to, 27, 120 Cephalosporium, 770 Cephalothin, resistance to, 28, 34 Cereals, see Grains and cereal products Cereulide (emetic toxin), 495, 498–499 Cetylpyridinium chloride for meat products, 132 for muscle food decontamination, 123 for poultry processing, 137 CH proteins, Clostridium perfringens, 479, 481–485 Chaetomium, 205 Chagas disease, 714 Chain of infection, 580 Champagne, 935–936 Chaperones, 742 Charcoal cefoperazone deoxycholate agar, modified, Campylobacter, 274–275 Charmat process, 935 Cheese bacteriocins in, 806–807 Clostridium botulinum inhibition, 454 defects, 170, 176–178, 180 manufacture, see Dairy products, fermented microorganisms in Listeria monocytogenes, 509, 514–515 Salmonella in, 236 pH, 172 spoilage control, 180 fermentative, 177–178 molds in, 180–181 proteases in, 174 sporeformers in, 179–180 Chemical action, see Disinfectants and disinfectant action Chemical changes, muscle food spoilage, 116–118 Chemical hazards, 119, 1045 Chemical preservatives, see Food antimicrobials

1085

Manila Typesetting Company

Chemical resistance, spores, 57–58 Chemical treatments for bovine spongiform encephalopathy agent, 654 for Cronobacter, 317–318 for fruits and vegetables, 194 Chemiosmotic theory, 766–767 Chemometrics, for muscle food spoilage, 118 Chemotaxis, Vibrio cholerae, 415 Chicken, see Poultry Chick-Watson model, 999 Chilling systems, 746–747 for fruits and vegetables, 193 for poultry carcasses, 135–136, 138–139 Chinese hamster ovary cell toxins, Vibrio fluvialis, 427 Chip technology, see DNA microarray analysis Chitinase, 195, 408 Chitosan, 195–196, 779–780 ChiY protein, Yersinia enterocolitica, 355 Chloramphenicol, 21, 23 for Cronobacter infections, 331 resistance to, 28, 34, 120, 245 Chloride channels, regulation, cholera toxin and, 411–412 Chlorine, for seafood, 235 Chlorine dioxide for Clostridium botulinum, 452 for Cryptosporidium, 717–718 for fruits and vegetables, 194–195, 633 for Giardia, 724–725 for muscle food decontamination, 123 for poultry processing, 137, 277 Chlorine treatment for Clostridium botulinum, 451 for enteric viruses, 633 for muscle food decontamination, 123 for poultry processing, 137, 277 for Yersinia enterocolitica, 344 Chloroform, spore resistance to, 57–58 Chlortetracycline, 21, 29 Chocolate, see Cocoa and chocolate Choleglobin, 127 Cholera hog (Salmonella enterica serovar Choleraesuis), 225 human, see Vibrio cholerae Cholera gravis, 411 Cholera toxin cellular response to, 411–412 DNA analysis, 407–408 enteric nervous system effects, 411–412 enzymatic activity, 411–412 production, 406 prostaglandin formation, 412 receptor binding, 411–412 structure, 411 Chorioretinitis, Toxoplasma gondii, 722 Chromatin immunoprecipitation, 989–990 Chromatography aflatoxins, 601 fumonisins, 609 for proteomics, 988 Chromogenic media, Cronobacter, 321–323 Chronic wasting disease, 651, 656–666 Chymosin, in cheese flavor, 833

11/08/2012 07:24AM

Index

1086 CiaB protein, Campylobacter, 271 Cilantro, 782 Cinnamaldehyde, 325 Cinnamic acid, 784 Cinnamic aldehyde, 780–781 Cinnamon, 780–781 Ciprofloxacin, 20 for enteric fever, 244 resistance to, 27, 28, 120, 273–274, 385 Cirrhosis, aflatoxin-induced, 600 Citral, in spoilage resistance, 193 Citric acid antimicrobial action, 771–772 Escherichia coli O157:H7 effects, 290 metabolism, in fermented milk products, 829 in milk fermentation, 831 in poultry processing, 137–138 production, 5 in wine, 930 Citrobacter antimicrobial action on, 772 versus Enterobacter, 312 in muscle foods, 117 in vegetable fermentation, 846 Citrobacter koseri, 314 Citrulline, in winemaking, 930 Civet coffee, 895–896 Cladosporium antimicrobial action on, 772, 780 in cheese, 180 in grains, 213 in nuts, 205 in wine spoilage, 934–935 Cladosporium herbarum, 194 Clarithromycin, 22 Claudins, in CPE protein action, 479–481 Clay, for aflatoxin absorption, 603 Climacteric fruits, 189 Climate change, contamination concerns and, 121 Clonorchis sinensis, 698, 700 Clostridium antimicrobial action on, 768, 771, 774 in fruits and vegetables, 190 in meat fermentation, 870 microbiology, 444–446 in milk, 179–180 in muscle foods, 126 neurotoxins, 444–446 in nuts, 205 psychrotropic, 127 in seafood, 155 spores, 45, 52, 53 in vegetable fermentation, 846 Clostridium acetobutylicum, 458 Clostridium algidicarnis, 127 Clostridium algidixylanolyticum, 127 Clostridium argentinense, 455 characteristics, 445 growth, 454 neurotoxins, 442 Clostridium baratii, 62, 441 characteristics, 444–445 disease, 448 growth, 454–455 microbiology, 444–445 neurotoxins, 441–445 public health significance, 65

SMP_Food Microbiology_Index.indd

Clostridium beijerinckii in cheese, 180 in muscle foods, 127 Clostridium bifermentans, 71 Clostridium botulinum, 441–463, 782 acid tolerance, 452–453, 804 antibodies, 448 antimicrobial action on, 454, 769–770, 775–778 antimicrobial resistance, 34 bacteriocin action on, 807, 812 as biological hazard, 1045 in bioterrorism, 441 contamination with, 441 culture, 457 in dairy product spoilage, 72 disease (botulism), 65–66 clinical features, 447–448 diagnosis, 448 historical aspects, 441–442 incidence, 69–70 incubation period, 447 infant, 447–448 inhalational, 441 outbreaks, 69–70, 448–449 prevention, 450–455 treatment, 448 undetermined classification, 441 wound, 446–448 distribution, 441 ecology, 443–444 in fermented meat products, 863 genomics, 66, 458 in grains, 216 groups, 444–445 growth, 66, 73, 452–455, 746 HACCP applied to, 1044 heat resistance, 65, 742 heat treatment, 748–749 historical aspects, 441–442 meat packaging and, 125 microbiology, 444–446 models for, 1013 in modified-atmosphere packaging, 454 in muscle foods, 119 neurotoxins biochemistry, 446 in bioterrorism, 458 chemical structures, 446 dehydration, 452 detection, 448, 457 freezing, 452 genomics, 443–446 heat treatment, 451–452 inactivation, 451–452 lethal dose, 446 pharmacology, 446 predictive modeling, 455–458 production, 445 public health significance, 65–66 safety in handling, 457–458 serotypes, 443 in water, 452 in nuts, 209 oxygen effects on, 453–454 in seafood, 140, 141, 154, 155 sources, 69–70 spores, 49, 63

1086

Manila Typesetting Company

in canned foods, 63–65, 69 chemical composition, 69 chemical treatment, 451 control, 450–451 distribution, 446 ecology, 443–444 in foods, 443–444 heat resistance, 63–65, 68–69, 743, 744 heat treatment, 63, 450–451 inactivation, 450–451 lethality measurement, 741 pressure effects, 451 public health significance, 65–66 pulsed electric field treatment, 451 radiation susceptibility, 451 structure, 69 thermal processing, 64–65 temperature effects, 16, 450–453 toxins, 62–63; see also subhead neurotoxins intentional contamination with, 94 transmission, 121 types, 442–443 vaccines, 458 water activity requirements, 452, 750 Clostridium butyricum, 63, 65, 441 characteristics, 444–445 in cheese, 180 disease, 448 in food spoilage, 70 growth, 454–455 in meat spoilage, 72 microbiology, 444–445 neurotoxins, 441–445 spores, 62, 65, 71 in wine spoilage, 934 Clostridium difficile in dairy product spoilage, 72 genomics, 458 in muscle foods, 119, 129–130 spores, 61–62 Clostridium estertheticum in meat spoilage, 72 in muscle foods, 127 Clostridium frigidicarnus, 127 Clostridium gasigenes in meat spoilage, 72 in muscle foods, 127 Clostridium laramie in meat spoilage, 72 in muscle foods, 127 Clostridium lituseburense, 127 Clostridium pasteurianum, 776 heat resistance, 742 in juices, 72 in meat spoilage, 72 Clostridium perfringens, 465–489 acid response, 467 antimicrobial action on, 769–772, 776, 783, 784 bacteriology, 66–67 b-toxin, 465 characteristics, 465–467 classification, 465–466 CPE protein, 465, 473–485 action, 477–483, 477–485 biochemistry, 477 cell death due to, 481–482

11/08/2012 07:24AM

Index

1087

cellular action, 478–483 complex formation, 479–481 detection, 469 epitopes, 485 evidence for food poisoning correlation, 473 gastrointestinal effects, 477–478 genetics, 473–476 medical applications, 485 plasma membrane permeability, 481–484 receptor for, 478–479 regulation, 476–477 release, 475–477 structure/function relationships, 483–485 synthesis, 476–477 in vaccine preparation, 485 in dairy product spoilage, 72 disease, 66–67 epidemiology, 468 necrotic enteritis versus type A food poisoning, 465 outbreaks, 468–470 prevention, 469 distribution, 66 enterotoxins, 67; see also subhead CPE protein environmental susceptibility, 466–467 genomics, 458, 467, 477–478 in grains, 216 growth, 65–68, 465, 467 heat resistance, 65, 67, 466 identification, 469–470 as indicator, 89 infectious dose, 471 in meat, 467–469 model for, 1013 in muscle foods, 119, 145 in normal microflora, 952 osmoregulation, 15 plasmids, 467, 473–474 refrigeration, 466–467, 469 reservoirs, 467–468 salt effects, 467 in soil, 467 spores, 62, 65, 466–467 heat resistance, 67, 466, 469 public health significance, 66–67 resistance properties, 472–473 salt effects, 467 sporulation, 67, 470 susceptible populations, 471 temperature effects, 466–467 toxins, 62–63, 67, 465–466; see also subhead CPE protein intentional contamination with, 94 typing, 465–466 vaccines, 485 virulence factors, 472–485 CPE protein, see subhead CPE protein heat resistance, 469 water activity requirements, 750 Clostridium sporogenes, 65 antimicrobial action on, 769, 776–777 in canned foods, 72 in cheese, 180 in dairy product spoilage, 72

SMP_Food Microbiology_Index.indd

in food spoilage, 70 heat resistance, 742 heat treatment, 749 spores, 71 Clostridium tertium, 71 Clostridium tetani, 49, 458 Clostridium thermocellum, 875 Clostridium thermosaccharolyticum, 775 Clostridium tyrobutyricum, 65, 180 antimicrobial action on, 775–776 in dairy product spoilage, 72 Cloves, 780–781 Clp proteins, Listeria monocytogenes, 529 Clustered regular interspaced short palindromic repeats (CRISPRs), 835–836, 980 Clustering algorithms, for muscle food spoilage, 118 cly genes, Bacillus thuringiensis, 492 Coagglutination tests, Vibrio, 407 Coagulation, milk, 170, 179 Coamplified internal amplification control, Cronobacter, 314 Cockroaches, helminth transmission, 705 Cocktail, bacteriophage, 816 Cocoa and chocolate, 881–893 fermentation bean composition and, 884 bean death in, 890 biochemistry, 889–892 cultures for, 886–887 drying after, 892–893 environmental factors in, 890–891 flavor and, 890–891 hydrolytic enzyme reactions, 891 methods, 882–884 microbiology, 884–889 objectives, 881–882 oxidative enzyme reactions, 891–892 reasons for, 881–882 storage after, 893 flavor, 886–887 fruit characteristics, 882 harvesting, 882 quality, 892 Salmonella growth in, 232, 243 Coconut contamination, 207, 210 microflora, 204 organisms in, 207 processing, 212 Codex Alimentarus Commission, 81–82, 639, 1023–1024, 1039 CodY protein, in sporulation, 50 Coffee processing, 606, 893–896 Cohort studies, in epidemiology, 578–579 Cold acclimatization proteins, Yersinia, 343 Cold settling, in winemaking, 917 Cold shock proteins, 16–17, 342–343 Cold temperatures, see Freezing; Refrigeration Colicins, 804, 812 Coliforms in cheese, 178 in grains, 214 as indicators, 89 in milk, 178 in seafood, 148–149

1087

Manila Typesetting Company

yellow-pigmented, Cronobacter as, 311; see also Cronobacter Colitis, Escherichia coli, see Escherichia coli, enterohemorrhagic; Escherichia coli O157:H7 Collateral sensitivity, 37 Colletotrichum, 190 Colletotrichum acutatum, 196 Colletotrichum capsici, 192 Colletotrichum gloeosporioides, 188, 191–193, 196 Colletotrichum musae, 188, 191 Colonies, model studies of, 1006 Colonization, see also Carriage Salmonella, 246 Vibrio cholerae, 408, 414–415 Yersinia, 352 ComBase database, 1012–1014 Commensal bacteria, antimicrobial resistance, 29–31, 34–35 Commercial sterilization, 744 Comminuted meat products processing, 131–132 spoilage, 126–127 Comparative genomics, 964, 980–982 Compartment-based models, 1011–1012 Compatible solutes, 992 Competent cells, 25 Competitive exclusion Clostridium botulinum, 454 Cronobacter, 318 Competitiveness, probiotics, 962 Completeness errors, in models, 1003 Conalbumin, antimicrobial action, 780 Conditioning, beer, 907 Conductive heat transfer, 739 Cone hops, 904 Confectionery products, spoilage, 215–216 Conjugated linoleic acid, in meat fermentation, 869 Conjugation, in antimicrobial resistance, 24–25, 30 Conotoxins, intentional contamination with, 94 Constrained growth, in models, 1005 Constrained polynomial approach, in models, 1000–1001 Contact hemolysin, Shigella, 307 Containers, heat transfer and, 740 Contamination intentional, see Intentional contamination nonhomogeneous distribution of, 85–86 unintentional aquaculture ponds, 141 Bacillus cereus, 498 bovine spongiform encephalopathy agent, 654 Clostridium botulinum, 441 coconuts, 207, 210 Cronobacter, 324–328 enteric viruses, 638–639 environmental, 121 fecal, see Fecal contamination in food processing, 86–87 fruits and vegetables, 190–192 versus intentional contamination, 92–97 Listeria monocytogenes, 509–514 meat and meat products, 121, 125–126

11/08/2012 07:24AM

Index

1088 Contamination, unintentional (Continued) milk, 172–173 muscle foods, 121 nuts, 204–208 poultry, 134 public health significance, 92–93 Salmonella, 139–140, 231, 235 seafood, 140 Shigella, 380–383 Staphylococcus aureus, 557 Vibrio cholerae, 410 viruses, indicators for, 631–632 wine, 934 Contracaecum, 698–700 Control measures, 87–88, 122–125 Controlled acidification, 803–804 Controlled atmosphere storage, 747 Controlled atmosphere stunning, in poultry processing, 134 Convective heat transfer, 739 Cook-chill products, 747 Cool storage, see Refrigeration Copepods, helminth transmission in, 704 Core genes, Campylobacter, 266 Coriander, 782 Corn, see Grains and cereal products; Maize Coroller model, 1004 Corrective action establishment, in HACCP, 1051–1052 Cortex-lytic enzymes, in sporulation, 62 Corynebacterium in nuts, 205 in seafood, 155 Corynebacterium glutamicum, 5 Coryneforms, in muscle foods, 126 Coselection, in antimicrobial resistance, 33–34 Coumaric acid, 784 Counterflow water chilling, poultry carcasses, 135 Coxiella burnettii, in milk, 743 Coxsackieviruses, 622 cpe gene, Clostridium perfringens, 467–468, 472–477 CPE protein, see Clostridium perfringens, CPE protein Crab meat, 117, 154–155 Crawfish, 117, 153–154 Cresol, 784 Creutzfeldt-Jakob disease, 130, 651, 660, 665 variant, see Variant Creutzfeldt-Jakob disease Criollo cocoa beans, 882 CRISPRs (clustered regular interspaced short palindromic repeats), 835–836, 980 Criteria, microbiological, see Microbiological criteria Critical control points, see Hazard Analysis and Critical Control Point system Criticality, in CARVER+Shock strategy, 98–100 Crohn’s disease, 129–130 Cronobacter, 311–337 antimicrobial action on, 772 antimicrobial resistance, 330–331 biochemical characterization, 312–313 biological inactivation, 317 characteristics, 312–313

SMP_Food Microbiology_Index.indd

chemical inactivation, 317–318 competitive exclusion, 318 culture, 312–313 discovery, 311 disease age distribution, 322–324 epidemiology, 311 outbreaks, 324, 326–328 symptoms, 322 epidemiology of, 314 in foods, 318–321 historical view, 311–312 inactivation methods for, 315–318 infectious dose, 324, 328 isolation methods, 321–322 opportunistic nature of, 312 pathogenicity, 324–325, 328–330 phenotypic differentiation, 312 phylogenic characterization, 313–315 reservoirs, 318–321 stressed cells, detection, 322 subcultures, 315 temperature effects, 315–316 transmission, 324 virulence factors, 324–325, 328–330 water activity, 316–317 Cronobacter condimentii, 312, 314 Cronobacter dublinensis, 314 Cronobacter malonaticus, 313–314 Cronobacter muytjiensii, 313–314 Cronobacter sakazakii, 312–314, 784 Cronobacter turicensis, 312–314 Cronobacter universalis, 312, 314 Cross-contamination eggs, 140 in food processing, 86–87 muscle foods, 119, 131, 660–661 poultry, 271 Cross-resistance, antimicrobial, 33–34 Cruise ships norovirus outbreak, 626 Shigella flexneri outbreak, 382 Cryptococcus, 770, 782, 919 Cryptosporidium, 714–718 disease, 585 risk assessment, 1026 in seafood, 140 Cryptosporidium canis, 715–716 Cryptosporidium felis, 715–716 Cryptosporidium hominis, 714–716 Cryptosporidium meleagridis, 715–716 Cryptosporidium muris, 715–716 Cryptosporidium parvum, 119–120, 713–718 Crystal inclusion protein, Bacillus thuringiensis, 492 csp genes, Yersinia, 343 Cucumbers, fermentation, see Vegetables, fermented Cultivars, microbial resistance and, 195–196 Cultures Campylobacter, 274–276 Clostridium botulinum, 457 Cronobacter, 312–313, 321–322 microbial, see Microbial growth probiotic, 961–964 Salmonella, 226, 228 Shigella, 378–379

1088

Manila Typesetting Company

starter, see Starter cultures Vibrio, 402 Curdling, milk, 179 Curing agents Clostridium perfringens growth and, 467 fruits and vegetables, 195 Curvacins, 805, 812 Cuticle, as microorganism barrier, 192–193 Cutinase, 191 CwIJ protein, in sporulation, 62 Cyanide, intentional contamination with, 94 Cyanidins, in cocoa fermentation, 891 Cyanobacter, radiation susceptibility, 753 Cyclic adenosine monophosphate, in cholera toxin action, 412–413 Cyclobutane-type TT (thymidine) dimer, in spores, 56–57 Cycloheximide, 911 Cyclopiazonic acid, 598 Cycloserines, 21 Cyclospora, 718–721 Cyclospora cayetanensis, 575, 579–580, 583, 713 Cyst(s) Balantidium coli, 726–727 Entamoeba histolytica, 727 Giardia, 724–725 Cysteine, staphylococcal enterotoxins, 564 Cysticercosis Taenia saginata, 682–686 Taenia solium, 686–689 Cytolethal distending toxin, Campylobacter jejuni, 266, 272 Cytolysin, Vibrio cholerae, 412, 414 Cytoplasm, Listeria monocytogenes movement in, 527–528 Cytoplasmic membrane, antimicrobial action on, 767 Cytotoxin(s) Bacillus cereus, 494–499 Salmonella, 250 Czapek extract agar, for Aspergillus, 598

D

D values (heat resistance), 756 Bacillus, 59–60 Campylobacter, 265 Clostridium botulinum, 68–69 Clostridium perfringens, 67 Cronobacter, 315–316 Escherichia coli, 290–291 in heat treatment, 64–65, 68–69, 740–742 models for, 999–1001 principles, 740–742 Salmonella, 232–233 spores, 58, 64, 66 Vibrio, 405 Dairy products, see also Milk bacteriocins in, 806–807 composition, 172 defects in, 176–178 fermented, 825–839; see also Cheese aroma products, 829–831 bacteriophages, 834–836 defects, 178 flavor, 827, 833 genetics, 836 health benefits, 949 lactose metabolism, 827–829

11/08/2012 07:24AM

Index

1089

microorganisms, 825–827 probiotics in, 949 proteolytic systems, 831–833 starter culture, 825–827 as growth media, 172 lipases in, 175–176 organisms in Anoxybacillus flavithermus, 72 Bacillus cereus, 72, 492, 498 Clostridium, 170, 180–181 Clostridium botulinum, 72, 444 Clostridium difficile, 72 Clostridium perfringens, 72 Clostridium sporogenes, 72 Clostridium tyrobutyricum, 72 coliforms, 170, 178 Geobacillus stearothermophilus, 72 Geobacillus thermoleovorans, 72 Klebsiella pneumoniae, 178 lactic acid bacteria, 170 Lactobacillus, 170, 177–178 Listeria monocytogenes, 507–509, 514–517 molds, 180–181 psychrotrophic, 170, 173–174 Salmonella, 225, 236–237, 578, 585 sporeformers, 178–180 yeasts, 180–181 Yersinia enterocolitica, 346 pH, 172 proteases in, 174–175 public health significance of, 172–173 spoilage, 72, 173–181 fermentative nonsporeformers in, 177–178 psychrotrophic, 173–176 spore-forming bacteria in, 178–180 water activity, 750 Dakguadong, 842 Dalgaard model, 1003 Danish Meat Research Institute, 1004, 1014 Danofloxacin, 21 Dantigny-Bensoussan model, 1001 Danzig detection and diagnostic system, 100–103 Dark, firm, dry (DFD) meats, 126, 862 Darmbrand (necrotic enteritis), Clostridium perfringens, 465 Data interpretation, on antimicrobial resistance, 34–35 Data transformation, models for, 1001 Death, microbial, see Microbial death Death kinetics, 7 Debaryomyces antimicrobial action on, 770 as beer contaminant, 909 in cocoa fermentation, 885 in meat fermentation, 860 in milk fermentation, 826 in poultry, 138 Debaryomyces hansenii, 750 in cheese, 180 in meat fermentation, 870 osmoregulation, 15 Decarboxylation, in biogenic amine formation, 142–143 Decay, see Fruit(s), spoilage; Vegetables, spoilage Decimal reduction dose, in radiation, 753

SMP_Food Microbiology_Index.indd

Decimal reduction times, see D values (heat resistance) Decision trees, critical control point, 1048 Decoction mashing, in brewing, 903 Decontamination in fruit and vegetable processing, 194–195 models for, 999–1001 in muscle food processing, 123, 130–131 in poultry processing, 136–138, 277–278 Deep tissue spoilage, in meat, 127 Defeathering, poultry, 135 Defect rate, 84–85 Defensins, 812 Dehydration body fluids, in cholera, 410–411 foods, 748–752 cocoa beans, 892–893 drying process, 751 freeze-drying, 55, 751–752 malt for brewing, 902–903 microorganisms Clostridium botulinum, 452 in sporulation, 59–60 staphylococcal enterotoxins and, 559 Dehydroacetic acid, 768 Dehydroalanine, as bacteriocin, 804 Dehydrobutyrine, as bacteriocin, 804 Deinococcus radiodurans, 753–754 Dekkera, 909, 927 antimicrobial action on, 773 in wine spoilage, 931 Dekkera anomala, 784 Dekkera bruxellensis, 784, 936 Demeclocycline, 22 Denaturation, staphylococcal enterotoxins, 561 Deoxynivalenol, 610–611 Deoxynucleotide triphosphates, in DNA sequencing, 976–978 Department of Agriculture, antimicrobial resistance monitoring, 35 Depuration enteric viruses, 632 Vibrio, 404 Desalting, in vegetable fermentation, 847 Descriptive epidemiology, 577 Desferrioxamine, Yersinia enterocolitica, 354 Desiccation resistance Salmonella, 231–232 spores, 55 Desulfotomaculum, spores, 45 Desulfotomaculum nigrificans, 64, 65, 72 Detection, in biosecurity, 100–103 Detergents, spore resistance to, 57–58 Deterministic models, 998–1001 Devlieghere model, 1003 DFD meats, 126, 862 Dhamuoi, 842 Diacetoxyscirpenol, 94 Diacetyl in fermented milk products, 829–831 in muscle food spoilage, 117 in wine, 931 Diagnostics, in biosecurity, 100–103 Diammonium phosphate, in winemaking, 917 Diarrhea, see also Enterotoxin(s); Gastroenteritis

1089

Manila Typesetting Company

Bacillus cereus, 491, 493, 495–496 Campylobacter, 272–273 Cryptosporidium, 714–718 Cyclospora, 719–721 Entamoeba histolytica, 727 Escherichia coli, see Escherichia coli, enterohemorrhagic; Escherichia coli O157:H7 Giardia, 714, 725 microsporidia, 723–724 probiotics for, 957 Salmonella, 247 Shigella, 383–384 staphylococcal food poisoning, 559 Trichinella, 679 Yersinia enterocolitica, 344–345 Dicarboximides, for fruits and vegetables, 194 Dicrocoelium, 703 Dicrocoelium dendriticum, 698, 705–706 Dideoxynucleudide triphosphates, in DNA sequencing, 976–978 Differential gel electrophoresis, for proteomics, 988 Digoxin, intentional contamination with, 94 Diguanylate cyclase, Cronobacter, 325 Dihydroxyacetone, in wine, 933 Dimethyl dicarbonate, 773 Dimethyl disulfide, 117 Dimethyl sulfide in muscle foods, 117 in winemaking, 925–926 Dinitrosylhemochrome, 873 Dioxin, unintentional contamination with, 92 Diphyllobothrium, 698 Diphyllobothrium dendriticum, 700 Diphyllobothrium latum, 700 Diphyllobothrium pacificum, 700 Dipicolinic acid, in spores, 46, 48, 56, 59–60 Dipylidium caninum, 698, 705 Direct plating, Campylobacter, 274 Dirithromycin, 22 Discoloration, in muscle food spoilage, 116, 117, 125, 126, 127 Discriminant function analysis, for muscle food spoilage, 118 Disgorgement, in winemaking, 935 Disinfectants and disinfectant action for brewing equipment, 911–912 enteric viruses, 633–634 for fruits and vegetables, 194–195, 633–634 injury, 9–10 resistance to, 32 spores, 57–58 Vibrio vulnificus, 421 viruses, 632 Disk diffusion method, for resistant bacteria, 30 Disulfide bond, staphylococcal enterotoxins, 563–564 Divergins, 809 DMFit model, 1013 DNA heat damage, 738, 742 protein interactions with, 989–990 radiation damage, 753 in spore repair, 60

11/08/2012 07:24AM

Index

1090 DNA (Continued) spores, 53 damage, 55–60 replication in germination, 62 sulfite damage, 778 transmission, 24–25 DNA analysis Listeria monocytogenes, 504–505 probiotics, 958–959 Vibrio cholerae, 407–408 Vibrio vulnificus, 422 DNA microarray analysis, 985–987, 1065–1066 Listeria monocytogenes, 505–506 recent applications, 1068–1069 Vibrio cholerae, 415 DNA sequencing, 952, 975–979, 1064–1066 DNA-binding proteins, for acid resistance, 14 DnaK protein, Listeria monocytogenes, 529, 1006 DnaK protein, Salmonella, 233 Documentation, HACCP system, 1052–1053 Dog tapeworms, 705 Dormancy, of spores, 46, 54, 61–62 Dose-response relationship, in risk assessment, 999, 1026 Double-strength wort, 909 Doubling times, in growth, 7–8 Dough products, 215–217 Doxycycline, 22 DPA (dipicolinic acid), in spores, 46, 48, 56, 60–61 Dracunculus medinensis, 698, 704 Dried foods, see also Infant formula, powdered drying processes for, 751 freeze-drying process, 56, 751–752 Salmonella in, 231–232 sausage, 857–865 Druggan-Forsythe-Iversen medium, 321–322 Dry air heating, nuts, 212 Dry processing, coffee fermentation, 895 Dry slaughter, poultry, 135 Drying cocoa beans, 883, 892–893 coffee beans, 895 Salmonella survival in, 232 Dwarf tapeworm (Hymenolepis nana), 698, 707 Dysentery, bacillary, see Shigella, disease

E

“E” toxin, 549 eae genes, Escherichia coli, 300 EAEC (enteroaggregative Escherichia coli), 288 EAST (enteroaggregative ST) toxin, Escherichia coli, 288 Echinococcus granulosus, 698, 704, 708 Echinococcus multiocularis, 698, 708 Echinostomum, 698, 702, 705 Echoviruses, 622 Ecosystems, foods as, 8 Edema, periorbital, in trichinellosis, 678–679 Effect, in CARVER+Shock strategy, 98–100 Effector proteins, Escherichia coli, 301

SMP_Food Microbiology_Index.indd

Efflux pumps in antimicrobial resistance, 22–23, 245 Cronobacter, 233 ege gene cluster, Staphylococcus aureus, 552 Egg(s) chicken antimicrobials in, 780 pasteurization, 234 processing, detection system for, 102–103 Salmonella in, 139–140, 234, 241–243, 581, 585–586 helminth, see specific helminths Egg Rule, 234 Egg Safety Final Rule, 581 EHEC, see Escherichia coli, enterohemorrhagic; Escherichia coli O157:H7 EIEC (enteroinvasive Escherichia coli), 288 Eimeria, versus Cyclospora, 718–719 El Tor hemolysin, 412, 414 Electric stunning, in poultry processing, 134 Electrical heat treatment, 745 Electrolyzed water for fruit and vegetable treatment, 195 for poultry processing, 136 Electron microscopy, bovine spongiform encephalopathy agent, 664 Electronic nose, for muscle food spoilage, 118 Electronic tongue, for muscle food spoilage, 118 Electrophoresis, see also Pulsed-field gel electrophoresis bovine spongiform encephalopathy agent, 653 capillary electrophoresis, 976 denaturing gradient, 961 in epidemiologic studies, 584 Listeria monocytogenes, 505 in proteomics, 988 in subtyping, 1061–1062 Vibrio cholerae, 407–408 Vibrio vulnificus, 422 ELISA, see Enzyme-linked immunosorbent assay Ellagic acid, 784 Embden-Meyerhof-Parnas pathway, 3–5, 871–872 Emerging Infection Program (CDC), 576 Emetic toxin, Bacillus cereus, 496, 498–499 Encapsulation, food antimicrobials, 785–786 Encephalitozoon, 713, 723 Encephalopathy, bovine spongiform, see Bovine spongiform encephalopathy Endolysins, bacteriophage, 18, 814–815 Endomyces, 773 Endopeptidases, in fermented dairy products, 832–833 Endopolygalacturonase, 847–848 Endospores, 45, 63 Endotoxins, see Lipopolysaccharides Energetics, 6–7 EngY protein, Yersinia enterocolitica, 355 Enrichment methods, for Campylobacter, 274 Enrofloxacin, 21, 27, 274 Entamoeba dispar, 727 Entamoeba histolytica, 727 Enteric fever, 243–244

1090

Manila Typesetting Company

Enteric nervous system, cholera toxin effects on, 411–412 Enteric viruses human, see Human enteric viruses radiation treatment, 633 Enteritis, see Enterocolitis; Gastroenteritis Enter-Net, 1071 Enteroaggregative Escherichia coli (EAEC), 288 Enteroaggregative ST toxin, Escherichia coli, 288 Enterobacter in coffee fermentation, 894–895 in grains, 214 in muscle foods, 119, 145 in vegetable fermentation, 842, 846 Enterobacter aerogenes, 145–147, 312 Enterobacter agglomerans, 116 Enterobacter cloacae, 311–312, 314 Enterobacter sakazakii, moved to Cronobacter, 311; see also Cronobacter Enterobacter sakazakii Plating Medium, 322 Enterobacteriaceae antimicrobial action on, 768 antimicrobial resistance, 33, 34–35 as beer contaminant, 910 bioluminescence, 12 as indicator, 89 in meat, 863 in milk, 173, 177 in muscle foods, 116, 117, 126 osmoregulation, 15 pH effects on, 14 in poultry, 134, 137, 139 water activity requirements, 750 Enterobacterial repetitive intergenic consensus sequences, in PCR, 229 Enterocins, 808, 809 Enterococcus antimicrobial action on, 781 antimicrobial resistance, 28, 30, 31, 34–35 in meat fermentation, 119, 869, 871 in milk, 173, 177 in muscle foods, 127 radiation susceptibility, 753 Enterococcus casseliflavus, 887 Enterococcus faecalis antimicrobial action on, 776 antimicrobial resistance, 31, 34, 120 in meat fermentation, 875 Enterococcus faecium, 25, 31, 120 Enterocolitis Campylobacter, 272–274 Escherichia coli, 303–304 Salmonella, 243, 246 Yersinia enterocolitica, 344–345 Enterocytes, Cronobacter adherence, 328–329 Enterocytozoon, 713, 723–724 Enterocytozoon bienusii, 713 Enterocytozoon intestinalis, 723 Enterohemolysin, Escherichia coli, 301 Enterohemorrhagic Escherichia coli, see Escherichia coli, enterohemorrhagic; Escherichia coli O157:H7 Enteroinvasive Escherichia coli (EIEC), 288 Enteropathogenic Escherichia coli (EPEC), 288

11/08/2012 07:24AM

Index

1091

Enterotoxigenic Escherichia coli (ETEC), 288, 582–583 Enterotoxin(s) Bacillus cereus, 67–68, 495–499 Bacillus thuringiensis, 495, 498 Campylobacter jejuni, 272 cholera, see Cholera toxin Clostridium perfringens, 67; see also Clostridium perfringens, CPE protein Cronobacter, 325 Escherichia coli, 288 Salmonella, 250 Shigella, 386, 390–391 staphylococcal, see Staphylococcal enterotoxins Vibrio alginolyticus, 430 Vibrio cholerae, 412–414; see also Cholera toxin Vibrio fluvialis, 427 Yersinia, 351–352 Enteroviruses, 120, 622 Entner-Doudoroff pathway, 5 Environmental contamination, in muscle food industry, 121 Environmental Health Specialists Network, 587 Environmental shotgun sequencing, 960 Environmental susceptibility, see specific organisms Environmental testing programs, 88 EnvZ protein, Salmonella, 250 Enzyme(s), see also specific enzymes in milk defects, 171 Enzyme immunoassays, Giardia, 725 Enzyme-linked coagulation assay, Salmonella, 229 Enzyme-linked immunosorbent assay bacteriocins, 806 fumonisins, 609 Taenia saginata, 684–685 Trichinella, 680 Eosinophilia Taenia saginata, 681–684 Trichinella, 678–679 EPEC (enteropathogenic Escherichia coli), 288 Epicatechin, 892 Epidemiologic methods case-control study, 578 cohort study, 578–579 food safety system evaluation, 588–589 limitations, 579 surveillance, 581–589 Epidemiology, see also specific diseases analytical, 578–579 chain of infection, 580 description, 576–580 descriptive, 577 Epithelial cells, invasion Campylobacter, 272 Listeria monocytogenes, 524–526 Shigella, 386 Yersinia enterocolitica, 348–349 Epizootic hemorrhagic disease virus, 121 Equilibrium relative humidity, 749 Erwinia in grains, 214 in vegetable fermentation, 842 Erwinia carotovora, 784 Erwinia carotovora subsp. atroseptica, 194

SMP_Food Microbiology_Index.indd

Erwinia chrysanthemi, 191 Erwinia dissolvens, 895 Erythema nodosum, Yersinia-induced, 345 Erythorbates, in meat fermentation, 863 Erythorbic acid, in winemaking, 917 Erythromycin, 21, 22, 27, 273–274 Escherichia antimicrobial action on, 774 in coffee fermentation, 894 in muscle foods, 145–147 in vegetable fermentation, 846 Escherichia coli acid tolerance response, 990–991 antigens, 287 antimicrobial action on, 20, 768, 770, 776, 778, 780–782, 783, 784 antimicrobial resistance, 33, 35 bacteriocin production, 804 classification, 278–289 cold shock proteins, 342–343 colicins, 812 disease, 294–295 enteroadherent (EAEC), 288 enteroaggregative (EAEC), 119, 288 enterohemorrhagic (EHEC), 289; see also Escherichia coli O157:H7 acid tolerance, 290 antimicrobial resistance, 290 characteristics, 289–291 disease, 293–299 genomics, 291 infectious dose, 294 in muscle foods, 128–129 pathogenicity, 299–304 reservoirs, 291–293 temperature effects, 17 enteroinvasive (EIEC), 288 enteropathogenic (EPEC), 288 enterotoxigenic (ETEC), 288, 582–583 epidemiology, 294–295 in grains, 216 growth, 14, 290 heat inactivation, 748 in meat fermentation, 875 mechanisms of action, 953 in milk, 743 models for, 1003–1004, 1007, 1010–1012, 1014 in muscle foods, 114, 123 in normal microflora, 951 in nuts, 205, 210–212 osmoregulation, 15 pathogenic versus normal flora, 951 pathogenicity, 287–288, 951 pathotypes, 287 pH effects, 14 in poultry, 136, 138 pressure effects, 756 versus Salmonella, 247 in seafood, 140, 154 Shiga-toxin producing (STEC), 94, 128–129, 288 sorbitol-fermenting, 289 subtyping, 1062 temperature effects, 16–17 trehalose, 317 viable but nonculturable, 10, 11 virulence factors, 287, 378, 389–391 water activity requirements, 750

1091

Manila Typesetting Company

Escherichia coli O26, 289–291, 298–299 Escherichia coli O45, 290 Escherichia coli O84, 299 Escherichia coli O103, 290–291, 298 Escherichia coli O104, 299 Escherichia coli O104:H4, 288 Escherichia coli O111, 289–291, 294, 298 Escherichia coli O113, 301–302 Escherichia coli O121, 299 Escherichia coli O123, 298 Escherichia coli O124, 288 Escherichia coli O145, 290, 299 Escherichia coli O157:H7 acid tolerance response, 14, 290, 767–768, 990–991 antimicrobial action on, 767, 769–776, 779–781, 783–786, 874 antimicrobial resistance, 25, 28–29, 31, 33, 291 attachment and effacing lesions, 299–300 bacteriophage effects, 813–815 in biofilms, 13 bioluminescence, 12 carriage, 294–295 characteristics, 289–291 disease characteristics, 293–294 epidemiology, 294–295 outbreaks, 294–298, 575, 578–579, 586, 588–589 pathology, 299–304 in fermented meat products, 862 first isolation, 291–292 freeze-thaw resistance, 14 in fruits and vegetables, 191 genomics, 291 geographic distribution, 292, 294–295 in grains, 215 heat resistance, 290–291 herd prevalence rates, 292 high-pressure processing, 124 historical aspects, 289 infectious dose, 294 injury, 9 in milk, 172 in muscle foods, 114, 119, 120, 126, 128–132 normal microflora interactions with, 951 in nuts, 209 pathogenicity, 299–304 plasmids, 301–302 pressure effects, 756 probiotic protection from, 955 quorum sensing, 12 radiation susceptibility, 125, 290–291, 299, 754 reservoirs, 291–293 risk assessment, 1026, 1033 in seafood, 141 seasonality, 292, 295 shedding, 292 Shiga toxins, 288, 302–304 subtyping, 1061, 1063–1064, 1069 supershedding, 292 surveillance for, 577, 584–586 susceptible populations, 295 temperature effects, 17, 290–291 toxins, 288, 302–304

11/08/2012 07:24AM

Index

1092 Escherichia coli O157:H7 (Continued) transmission, 295, 297 viable but nonculturable, 11 Esophageal cancer, 609 Esp proteins, 17, 301 Essential oils antimicrobial action, 780–782 for fruit and vegetable treatment, 194 in hops, 904 in Listeria monocytogenes inhibition, 510 Esters, antimicrobial action, 766–773 Estrogenic activity, zearalenone, 611 ETEC (enterotoxigenic Escherichia coli), 288 Ethanol addition to fortified wines, 936 in brewing, 906 in cocoa fermentation, 882, 884, 886, 890 in coffee fermentation, 895 in dairy product fermentation, 828–829, 831 enteric viruses, 634 in vegetable fermentation, 847 in winemaking, 917–918, 923, 925, 928–930, 933 Ethyl acetate, in beer, 906 Ethyl carbamate, in wine, 925, 930 Ethyl esters, in milk spoilage, 176 Ethyl paraben, 777–778 Ethyl vanillin, in Cronobacter inactivation, 318 Ethylene, in fruit and vegetable ripening, 189 Ethylene oxide, spore resistance to, 57–58 Etp proteins, Escherichia coli, 301 Eubacterium, 952 Eugenol, 780–781, 786 Eupenicillium, spores, 71 European Food Safety Authority Community Report, 28–29, 33 European Union, Hazard and Critical Control Point system, 1041 Eurotium, 16, 205, 750 Eustrongylides, 698, 702 Eutypa lata, 934 EvgA circuits, 990–991 Evisceration, poultry, 135 Evolutionary algorithms, for muscle food spoilage, 118 Evolutionary probability model, Campylobacter, 269 Exopeptidases, in fermented dairy products, 832–833 Exopolygalacturonase, 847–848 Exopolysaccharides Cronobacter, 312, 325 in fermented dairy products, 829 Exosporium, 46, 52 Exotoxin(s), Cronobacter, 324 Expert Committee on Food Additives, aflatoxin, 600 Expert Protein Analysis System Proteomics Server, 980 Exponential phase, microbial growth, 7 Exposure assessment, in risk assessment, 1025 Extended secondary models, predictive models, 1003–1004 Extrinsic factors, in microbial growth, 8, 14

SMP_Food Microbiology_Index.indd

F

“F” type toxin, 549 F value (thermal lethality), 741–742 Facultative heterofermentation, 5 FADH, in tricarboxylic acid cycle, 5 Fasciola hepatica, 698 Fasciolopsis buski, 698 Fat in milk, 169 in muscle foods, 126 Fatal familial insomnia, 651 Fatty acid(s) in fish fermentation, 866 in sausage, 862 in wine, 926 Fatty acid esters, antimicrobial action, 771–773 FbpA protein, Listeria monocytogenes, 524 Feathers, removal, 135 Fecal coliforms, see Coliforms; Escherichia coli Fecal contamination enteric viruses, 623–624 helminths, 698, 707–708 hepatitis A virus, 632 Listeria monocytogenes, 511 noroviruses, 632 Fecal streptococci in muscle foods, 126 in seafood, 154 Feed, animal, see Animals, feed for Feline calicivirus, 630–631, 633–634 Fermentation, 5 bacteriocinogenic starter cultures for, 808 Escherichia coli O157:H7 survival in, 290 pathways, 5 purposes, 841–843 tricarboxylic acid cycle in, 5 Fermented foods beer, 901–913 cocoa, 881–893 coffee, 893–896 Cronobacter survival in, 318 dairy products, see Dairy products, fermented fish, 865–866 meat, see Meat and meat products, fermented vegetables, 841–855 wine, 915–947 Ferredoxin, 467, 776 Ferrioxamine, Yersinia enterocolitica, 354 Ferritin, Salmonella competition with, 251 Ferulic acid, 784 Fibronectin binding outer proteins, Campylobacter, 271 Filamentous cells, Salmonella, 231–232 Filtration beer, 907–908 wine, 922 Fimbrial colonization factors, Escherichia coli, 288 Finfish, see Fish Fingerroot, 782 Firmness, in vegetable fermentation, 847 First-order kinetics, in microbial growth, 7–8 Fish, 141–148 cold-smoked, 140 farming of, see Aquaculture

1092

Manila Typesetting Company

fermented products, 865–866 illness from, 140 organisms in, 141–142 Clostridium botulinum, 442, 444, 449 helminths, 697–702 Listeria monocytogenes, 510, 512–513 microflora, 112–113, 141–142 pathogenic, 141–142 Salmonella, 234–235, 241–242 spoilage, biogenic amines in, 142–148 water activity, 750 Fish paste, 866 Fish sauce, 865–866 Fish Shelf Life Prediction Program, 1014 Fish tapeworms, 700 Fitness cost, in antimicrobial resistance, 36–37 fla genes, Campylobacter, 271 Flagella Campylobacter, 266–267, 271 Vibrio cholerae, 415 Yersinia enterocolitica, 352–353 Flagellin, Campylobacter, 268, 271, 1070 Flat sour defect, milk, 179 Flavobacterium in muscle foods, 126 in nuts, 205 in poultry, 134 in seafood, 153, 155 Flavonoids, 785 Flavor beer, 904 cheese spoilage, 178 cocoa products, 886–887, 890–891 fermented dairy products, 827, 833 fermented meats, 859, 862–864, 873–874 fish sauce, 865–866 muscle food spoilage, 116 sausage, 873–874 wine, 919, 927, 930 Fleas, helminth transmission, 705 Flocculation beer yeasts, 907 in enteric virus purification, 635, 637 Florfenicol, 21 Florid plaques, bovine spongiform encephalopathology, 660 Flour, see Grains Fludioxonil, 194 Flukes intestinal Fasciolopsis buski, 698, 703 Heterophyes heterophyes, 698, 702 liver Clonorchis sinensis, 698, 700 Fasciola hepatica, 698, 702–703 lung (Paragonimus westermani), 701 Fluoroacetic acid, intentional contamination with, 94 Fluorophores, in sequencing, 979 Fluoroquinolones, 21, 27, 37, 245, 273–274 Flux balance analysis, 1009 Food and Agriculture Organization Cronobacter code, 312 microbiological criteria, 83 risk analysis for enteric viruses, 639 Food and Drug Administration Egg Rule, 234 Food Code for Restaurants, 588 low-acid canned food definition, 63

11/08/2012 07:24AM

Index

1093

Food antimicrobials, 765–801 bacteriocins, 804–814 cellular targets, 766 definition, 766 dimethyl dicarbonate, 773 encapsulation, 785–786 factors affecting activity, 766 lactoferrin, 773–774 lysozyme, 774–776 nitrites, 776–777, 804 organic acids and esters, 766–773 parabens, 777–778 sulfites, 778 Food preservatives and preservation methods, see also specific methods; specific organisms acidification, 803–804 antimicrobials, 765–801 naturally occurring, 778–785 traditional, 766–778 appertizing, 63, 753 bacteriocins, 804–814 biologically based, 803–822 chemical, 765–801 naturally occurring, 778–785 traditional, 766–778 cool storage, see Refrigeration dehydration, 748–752 freezing, see Freezing fruits, 193–196 heat, see Heat treatment and inactivation high-pressure, 754–756 irradiation, see Radiation treatment for low-acid canned foods, 63–64 microbiological criteria for, see Microbiological criteria milk, 179 muscle foods, 132 physical, 737–763 pulsed electric field processing, 757–758 quality assurance in, 70 spoilage in, see Food spoilage sporeformer destruction in, 63–65 sterility in, 64 vegetables, 193–196 Food Safety and Inspection Services, 63, 123 Food safety objectives, 70, 82–83, 1026, 1034 Food spoilage, see also Meat and meat products; Vegetables; specific foods, e.g., Fruits after radiation treatment, 754 beer, 903 confectionery products, 215–216 control, see Food preservatives and preservation methods fruits, 187–201 grains, 203, 213, 215–217 kinetics, 7–8 microbiological criteria and, see Microbiological criteria muscle foods, see Muscle foods, spoilage nuts, 203–205 poultry, 138–139 sporeformers, 63 canned foods, 70–72 illness incidence, 69 prevention, 70

SMP_Food Microbiology_Index.indd

vegetables, 187–201 wine, 927, 931 FoodNet (Foodborne Diseases Active Surveillance Network), 131, 575–576, 579, 581, 585–589 Forastero cocoa beans, 882–883 Foreign objects, as hazards, 1045 Forespore, 45, 48–49, 52, 54 Formic acid in milk fermentation, 829 in muscle food spoilage, 118 Forward genetics, 983 Forward selection, in models, 1000 Fourier transform infrared analysis, for muscle food spoilage, 118 Francisella tularensis intentional contamination with, 94 subtyping, 1065 Freeze-drying, 56, 751–752 Freeze-thaw injury resistance to, 14 in yeasts, 9–10 Freezing, 747–748 Anisakis, 699 Campylobacter, 265, 277 Clostridium botulinum, 452 Clostridium perfringens, 466–467 Cronobacter, 315 definition, 745 enteric viruses, 631 fish, 145 Listeria monocytogenes, 506 microbial injury in, 9–10 protection against, 10 Shigella, 381 Taenia saginata, 685 Taenia solium, 689 Toxoplasma, 722 Trichinella, 677, 681–682 Vibrio, 405 Yersinia enterocolitica, 342–343 Freshness, lack of, see Food spoilage Freshtester, for muscle food quality evaluation, 118 fri gene, Listeria monocytogenes, 529 FRNA phage, as viral contamination indicator, 632 Frozen foods microbial radiation resistance in, 754 thawing, 747–748 Fructans, as prebiotics, 966 Fructo-oligosaccharides, as prebiotics, 966 Fructose, in winemaking, 929–930 Fructose-1,6-diphosphate, in meat fermentation, 871–872 Fruit(s), see also Juices antibiotics used on, 25–26 antimicrobial usage in, 25–26 composition, 189 microbial inhibitors in, 189 modified-atmosphere packaging, 747 ochratoxin in, 603 organisms in Aspergillus, 602 Campylobacter, 270 Cyclospora, 720 enteric viruses, 623–625 Escherichia coli, 296, 298 helminths, 698

1093

Manila Typesetting Company

Salmonella, 235–239 Vibrio cholerae, 410 outbreaks related to, 575 refrigeration, 746–747 shelf life, 189 spoilage characteristics, 187–190 contamination source, 190–192 control, 193–196 defense reactions, 192–193, 195 heat-resistant fungi, 71 humidity and, 193–194 mechanisms, 191–192 microorganisms causing, 191–192 in modified atmospheres, 193–194 resistant cultivars, 195–196 temperature effects, 193–194 water activity, 750 for winemaking, see Winemaking, grapes for Fruity flavor, milk, 170 F-specific coliphages, as viral contamination indicator, 632 Fujikawa-Itoh model, 1011 Fumaric acid, antimicrobial action, 771 Fumigation, nuts, 212–213 Fumonisins, 606–610 control, 609–610 processing effects on, 610 regulations, 609 screening for, 609 sources, 606–608 structures, 606 toxicology, 608–609 in winemaking, 934 Funasuchi, 866 Functional genomics, 983–985 Fungi, see also Mold(s); Yeasts; specific organisms antimicrobial action on, 770 growth, 7 heat-resistant, 71 Fungicides for fruits and vegetables, 194 for grains, 213, 215 in wine, 922 fur gene and Fur protein, 14, 233 Furocoumarins, 784 Fusarium, 606–611 antimicrobial action on, 770 as beer contaminant, 903, 909 in cheese, 180 in coffee fermentation, 895 in grains, 213 in nuts, 205 Fusarium culmorum, 610–611, 784 Fusarium graminearum, 610–612, 775 Fusarium moniliforme (verticillioides), 606–609 Fusarium proliferatum, 606–608 Fusarium roseum (graminearum), 610–612, 775 Fusarium solani, 194, 612 Fusarium sulphureum, 194 Fusarium verticillioides (moniliforme), 606–609, 612 FUT32 gene, viruses, 627 Fuzzy logic, for muscle food spoilage, 118 FyuA protein, Yersinia enterocolitica, 354

11/08/2012 07:24AM

Index

1094

G

Gad proteins in acid tolerance, 990–991 Listeria monocytogenes, 529 Galactose metabolism, in milk fermentation, 828 b-Galactosidase, in milk fermentation, 828 Gallic acid, 784 Gamma hypothesis, in models, 1004 Gamma rays, in food preservation, 752–753 Gamma-radiation susceptibility, spores, 56 Gangliosides, in neurotoxin binding, 412 Garlic, 782–783 Gas chromatography, deoxynivalenol, 610–611 Gas composition, see Atmosphere Gastroenteritis, see also Enterocolitis Campylobacter, 271–274 Clostridium perfringens, see Clostridium perfringens, disease Escherichia coli, 287, 293–299 Listeria monocytogenes, 518–520 Salmonella, 243, 246 Shigella, see Shigella, disease Trichinella, 678–679 Vibrio alginolyticus, 430 Vibrio cholerae, 410–411, 577 Vibrio fluvialis, 427 Vibrio furnissii, 427–428 Vibrio hollisae, 428–429 Vibrio parahaemolyticus, 418–419 Vibrio vulnificus, 422–423 viral, see Human enteric viruses; specific viruses Yersinia enterocolitica, 344–345 Gastrointestinal tract normal microbiota, 949–953 prebiotics in, 966 probiotics in, see Probiotics GC-Skew information, 980 Gecko, 1011 Geeraerd model, 999, 1011 Gel cassette model, 1005–1006 Gelling agents, for models, 1006 Gemifloxacin, 20 Gene chip test, see DNA microarray analysis Gene expression, in sporulation, 46, 49–52 Gene Ontology project, 980 General secretory pathway, Escherichia coli, 301 Generally Recognized As Safe (GRAS) substances, lactic acid bacteria, 813–814 “Generation time,” see Doubling times, in growth Genetic competence, in sporulation, 47 Genetic programming, for muscle food spoilage, 118 Genetic transfer, in lactic acid bacteria, 874–875 Genetics, see also specific organism in antimicrobial resistance, 24–25, 30, 34–35 Genome(s) annotation, 979–980 mycotoxins, 611–612 Genome chip test, see DNA microarray analysis Genomes Online Database, 876

SMP_Food Microbiology_Index.indd

Genomics, 975–988; see also specific organisms advances, 990–993 bile tolerance, 991–993 bioinformatics in, 979–980 comparative, 980–982 DNA microarray analysis in, see DNA microarray analysis DNA sequencing, 975–979 functional, 983–985 RNA sequencing, 987–988 software programs for, 979–980 Gentamicin, 20–21 for Cronobacter infections, 331 resistance to, 28 Geobacillus, 45 Geobacillus stearothermophilus, 64, 179 in canned foods, 72 in dairy product spoilage, 72 in food spoilage, 70 heat resistance, 742, 744 heat treatment, 749 Geobacillus thermoleovorans, 72 Geotrichum antimicrobial action on, 770, 781 in cheese, 180–181 in dairy products, 826 Geotrichum candidum, 188, 196 ger genes, in spore germination, 61 Germ cell wall, spore, 53 Germinant receptors, nutrient, 56 Germination in beer production, 901–903 in brewing, 902 in cocoa fermentation, 889 spore, 46, 56, 60–62 Gerstmann-Sträussler-Scheinker syndrome, 651 Ghost spots, in fruit and vegetable spoilage, 191 Giant intestinal fluke (Fasciolopsis buski), 698, 703 Giardia, 714, 724–725, 1026 Giardia agilis, 724 Giardia ardea, 724 Giardia lamblia (intestinalis/duodenalis), 724–725 Giardia muris, 724 Giardia psittaci, 724 Gibberella zeae, 610 GInaFiT software, 1013 GLIMMER software, 979 Global Foodborne Infectious Network, 585, 1071 Globoseries glycolipids, Escherichia coli, 303 GlpA protein, Vibrio cholerae, 414–415 Glucanase, 195 Glucobrassicin, 848 Gluconate, in muscle food spoilage, 118 Gluconic acid, in wine, 930, 933 Gluconoacetobacter, 932–933 Gluconoacetobacter hansenii, 932 Gluconoacetobacter liquefaciens, 932 Gluconobacter, 886, 910, 932–933 Gluconobacter oxydans, 887, 932–933 Gluconobacter xylinus, 885, 887 Glucono-delta-lactone, 863 Glucoraphinin, 848 Glucose

1094

Manila Typesetting Company

in meat fermentation, 863 metabolism, 3–5, 556, 871–872 in muscle food spoilage, 116–117 as spoilage energy source, 116–117 in winemaking, 929–930 Glucose-repressed system, for acid resistance, 14 Glucosinolates, 848 Glutamate, genomics, 991 Glutamate decarboxylase system, for acid resistance, 14 Glutamate-dependent system, Escherichia coli, 290 Glutamic acid, production, 5 Glutaraldehyde, enteric viruses, 634–635 Glyceraldehyde-3-phosphate, in glucose metabolism, 5 Glycerol formation, in winemaking, 923, 931 Glyceryl monolaurate antimicrobial action, 772 Staphylococcus aureus growth and, 556–557 Glycine betaine, Staphylococcus aureus, 557–558 Glycogen, depletion in muscle, 862 Glycolipids, Shiga toxin binding to, 303 Glycolysis in meat fermentation, 871–872 pathways, 3–5 Glycopeptides, 21 Glycosidases in cocoa fermentation, 891 in wine, 926 GMPs, see Good manufacturing practices Gnathostoma spinigerum, 698, 702, 704 Goitrin, in vegetable fermentation, 848 Gompertz model, 999, 1001, 1012 Good agricultural practices aflatoxin control in, 602 Salmonella, 235 Good Agricultural Practices, for enteric virus control, 633 Good hygiene practice, 87, 1039–1040 Good manufacturing practices, 1039–1040 Good manufacturing processes, 235 gpr gene product, in spores, 53 Grains and cereal products, 213–217 aflatoxin in, 598–603 beer production from, 901–907 contamination, 214–217 deoxynivalenol in, 610–611 foodborne illness due to, 216–217 fumonisin in, 606–610 harvesting, 213 importance, 203 infant rice cereal, 318 milling effects on, 214–215 mycotoxins in, 606–610 Aspergillus, 598–603 Fusarium, 606–611 Penicillium, 603–605 organisms in Alternaria, 213 Aspergillus, 213 Bacillus, 216–217 Cladosporium, 213 Clostridium botulinum, 216 Enterobacter, 214

11/08/2012 07:24AM

Index

1095

Erwinia, 214 Escherichia coli, 215–216 Fusarium, 213 Helminthosporium, 213 Klebsiella, 214 Lactobacillus, 215 Micrococcus, 213 Mucor, 205 Pantoea, 214 Penicillium, 213 Rhizopus, 213 Salmonella, 215–216 Serratia, 214 Staphylococcus, 216–217 processing, 214–215 spoilage, 213, 215–217 water activity, 750 Gram-positive bacteria, in muscle foods, 117 Grapes, see Winemaking, grapes for GRAS (Generally Recognized As Safe) substances, lactic acid bacteria, 813–814 Graves’ disease, Yersinia enterocoliticarelated, 362–364 Greening, in muscle food spoilage, 127 Grimontia hollisae, 428 Grocery Manufacturers Association, lowmoisture food task force, 235 GroEL protein, 17 Ground meat, Escherichia coli O157:H7 in, 295–297 Growth, microbial, see Microbial growth Growth models, 998–999 Growth/no growth models, 1002–1003 GrsA protein, Yersinia, 361–362 Guidelines, microbiological criteria, 81 Guillain-Barré syndrome, Campylobacterrelated, 273 Guillier model, 1007 Guinea worm (Dracunculus medinensis), 698, 704 “Guns, gates, and guards,” for food safety, 100 “Gushing factor,” in beer, 903 GyrA protein, in antimicrobial resistance, 27

H

H antigens Escherichia coli, 287 Salmonella, 227–228 Yersinia enterocolitica, 342 HACCP, see Hazard Analysis and Critical Control Point system Haemophilus influenzae, 22 Hafnia, 117 Hafnia alvei, 116, 126, 144 Halalkalicoccus, 866 Halzoun, 706 Hand decontamination, 634–635 Hanseniaspora in cocoa fermentation, 885 in winemaking, 919–920, 923, 925, 926 Hanseniaspora opuntiae, 887 Hanseniaspora uvarum, 920 Hansenula antimicrobial action on, 770, 773 in cocoa fermentation, 885 in winemaking, 920, 923, 927 Hansenula anomala, 15 HapR protein, Vibrio cholerae, 416

SMP_Food Microbiology_Index.indd

Hard scalding, in poultry processing, 135 Harvesting grains, 213 nuts, 204 Hazard(s) biological, 1045 characterization, 1025–1026 chemical, 1045 classification, 85 control, 1046 evaluation, 1043–1046 identification, 1044–1045 in muscle foods, 119 physical, 1045 surveillance for, 586–588 Hazard Analysis and Critical Control Point system, 1039–1057 control measure identification, 1046 corrective action establishment, 1051–1052 critical control point determination, 87, 1047–1048 critical limit establishment, 1048–1050 data management, 1052–1055 hazard analysis, 1043–1047 limitations, 1052–1055 monitoring procedure establishment, 1050–1051 muscle foods and, 119, 130 origin, 1040–1041 overview, 1041–1042 phases, 1041 preliminary steps, 1043 prerequisite programs, 1046–1047 principles, 1041–1042 recognition as best system, 1041 recordkeeping procedure establishment, 1052–1053 risk assessment in, 1034 safety system evaluation, 588 Salmonella processing, 234–235 sporeformers and, 70 surveillance, 576, 579, 581, 586 team approach, 1–43 variations, 1042 verification procedure establishment, 1052 world recognition, 1041–1042 Hazelnuts, organisms in, 207, 209–210 hbl genes, Bacillus cereus, 495–499 HBsu protein, in spores, 53 Heaps, cocoa fermentation in, 883 Heart disorders, probiotics for, 956 Heat shock proteins, 17, 33, 742–743 Escherichia coli, 1007 Salmonella, 233 Yersinia, 363 Heat shock transcription factor, 17 Heat susceptibility and resistance, 742–743; see also D values (heat resistance) antimicrobial resistance and, 33–34 Bacillus cereus, 68–69 bovine spongiform encephalopathy agent, 654 Campylobacter, 265–266 Clostridium botulinum, 63–65, 68–69, 450–451 Clostridium perfringens, 67, 466, 469 Escherichia coli, 290–291 hepatitis A virus, 632–633

1095

Manila Typesetting Company

noroviruses, 631–632 Salmonella, 232–233 spores, 58–60, 743 staphylococcal enterotoxins, 561 Vibrio, 405 Heat transfer, 739–740 Heat treatment and inactivation, 737–745; see also Canned foods; Pasteurization advances in, 744–745 aflatoxins, 599, 602 appertization, 63, 753 biological indicators, 738–739 “botulinum cook,” 64 Clostridium botulinum, 450–452 Cronobacter, 315–316 engineering principles, 739–740 enteric viruses, 631–632 fruits and vegetables, 194, 195 grains, 215 heat transfer in, 740 kinetics, 8, 740–741 for meat products, 132 microbial injury in, 9–10 microwave, 745 milk, 171–172 models for, 999–1001 monitoring, 738 for mycotoxin removal, 610 nuts, 212 ochratoxin, 606 ohmic, 745 pasteurization, see Pasteurization pathogens of concern, 738–739 pH effects, 743 process development, 738 resistance to, see Heat susceptibility and resistance sporeformer destruction in, 63–65, 70–72 sterilization, 743 thermal lethality measurements, 741–742 Trichinella, 681 validation, 738 vegetables, 194 Helicobacter, 119 Helicobacter pylori, 956 Heliscope, 979 Helminth(s), see also specific helminths in fruit, 698 sources, 698 fecal contamination, 698, 707–708 meat, 673–696, 698, 705–706 other invertebrates, 704–705 seafood, 697–702 vegetables, 698, 702–704 water, 704 Helminthosporium, 213, 770 Helveticins, 805, 808 Hemagglutinin Cronobacter, 324 Vibrio cholerae, 414 Hemagglutinin/protease, soluble, 423 Hemagglutinins, 423 Hemolysins Bacillus cereus, 495–499 Campylobacter, 266 Escherichia coli, 288, 301–302 Vibrio cholerae, 414 Vibrio fluvialis, 427 Vibrio hollisae, 429

11/08/2012 07:24AM

Index

1096 Hemolysins (Continued) Vibrio parahaemolyticus, 417–419 Vibrio vulnificus, 425 Hemolytic-uremic syndrome, 128–12, 172, 288 Escherichia coli, 288, 293–294, 303–304 Shigella, 384 Hepatitis A virus, 620 control, 632–635 depuration, 632–633 detection, 635–639 discovery, 621 disease, 629 environmental persistence, 631 epidemiology, 625–626 freezing, 631 genomics, 621 genotypes, 622 heat resistance, 632–633 high-pressure processing, 633 incubation period, 629 in muscle foods, 120 in ready-to-eat foods, 634–635 in seafood, 140, 151–152 serotype, 621–622 stability, 631 transmission, 631 vaccines, 630 Hepatitis E virus, 620, 623 Hepeviridae, 620 Heptyl paraben, 777–778 Herbs, antimicrobial compounds in, 780–782 Heterofermentation, 5 cocoa, 885 meat, 871–872 vegetable, 844–845, 847 Heterolactic organisms, in vegetable fermentation, 844 Heterophyes heterophyes, 698, 702 Hexanal, for fruit and vegetable treatment, 194 Hexenoic acid, in vegetable fermentation, 847 Hidden Markov model profiles, 980, 988 High-barrier films, Clostridium botulinum growth in, 454 High-gravity brewing, 908–909 High-oxygen packaging, muscle foods, 133 High-pathogenicity islands, Yersinia, 353–355 High-pressure processing, 754–756 meat, 124 viruses, 633 Hill-Wright model, 999 Hinshelwood model, 1000 Histamine in muscle food spoilage, 117, 120, 139, 142–148 toxicity, 143–144 Histidine, in winemaking, 930 Histidine kinase receptor, in signal transduction, 11, 13 Histidine protein kinase gene, 990–991 HIV infection, see Human immunodeficiency virus infection hlyA gene, Listeria monocytogenes, 506 H-NS protein, Shigella, 391 Hog cholera (Salmonella enterica serovar Choleraesuis), 225

SMP_Food Microbiology_Index.indd

Holotoxins, Shiga toxins as, 302 Hom model, 999 Homeland security, food safety and, see Intentional contamination Homeostasis, in microbial growth, 14, 17–18 Homeoviscous adaptation, 16 Homofermentation, 5 cocoa, 885 meat, 871 vegetable, 845 Homogeneity errors, in models, 1003 Homogenization, Yersinia enterocolitica, 344 Honey, Clostridium botulinum in, 447, 450 Hops, in brewing, 783 Horizontal gene transmission, 24–25, 30, 35 Hormodendrum, 180 Host association, Campylobacter, 267 Hpr protein, in sporulation, 50 Hülsheger model, 999 Human enteric viruses, 619–649; see also Hepatitis A virus; Noroviruses concentration, 635, 637 control, 632–635 detection, 635–639 diseases clinical features, 626–630 outbreaks, 625–626 pathogenesis, 626–630 treatment, 628–629 environmental persistence, 621, 630–632 epidemiology, 619, 621–623, 625–626 in fruit, 625, 633–634 purification, 635, 637 in ready-to-eat foods, 634–635 risk assessment, 639 in shellfish, 623–625, 632–633 stability, 631 transmission, 619, 623–625, 631 tropism, 621 types, 619–620 in vegetables, 624–625, 633–634 Human immunodeficiency virus infection cryptosporidiosis in, 715–716 cyclosporiasis in, 721 isosporiasis in, 721 listeriosis in, 520 microsporidiasis in, 723–724 Human leukocyte antigens, disease predisposition and, 244, 362–364 Human Microbiome Project, 953 Humidity, fruit spoilage, 193–194 Humulone, 783, 904 Hunt broth, Campylobacter, 276 Hurdle technology, 17–18, 37 Campylobacter, 277 cereal products, 216–217 enteric viruses, 634 meat processing, 122–125 models for, 1002 poultry processing, 136 Hyalocecropia, 812 Hydrogen ion-coupled ion transport system, Salmonella, 233 Hydrogen peroxide for Clostridium botulinum, 451 for enteric viruses, 633–634 for fruit and vegetable processing, 194–195, 633–634

1096

Manila Typesetting Company

for lactoperoxidase activation, 779 for milk, 171 for muscle food decontamination, 123 for poultry processing, 138 spore resistance to, 57–58, 69 Hydrogen sulfide in muscle food spoilage, 117, 126, 127 in wine, 925–926 Hydrolyzed lactoferrin, 774 Hydroprocessing, grains, 215 Hydroquinone, 784 Hydroxybenzoic acid esters, 777–778, 784 Hydroxycinnamic acids, 784 5-Hydroxyeicosatetraenoic acid, in staphylococcal food poisoning, 566 5-Hydroxytryptamine, in staphylococcal food poisoning, 567 Hyl-Oppa protein, Vibrio furnisii, 428 Hymenolepis diminuta, 698, 705 Hymenolepis nana, 698, 707 Hyperosmotic shock, 15 Hypertension, probiotics for, 956 Hypochlorite for Clostridium botulinum, 452 for enteric viruses, 633, 635 for fruit and vegetable processing, 194–195, 633 for grain milling, 214 for poultry processing, 137 spore resistance to, 57–58 Hypothiocyanite, in milk, 171

I

Ice cream, Salmonella in, 236 Ichthyism, 442 Ics proteins, Shigella, 388, 390–391 Identification, muscle food sources, 121–122 Illumina equipment, 976–977, 988 Imalazil, for fruits and vegetables, 194 Immune response, see also Autoimmune sequelae Campylobacter, 273 hepatitis A virus, 629 Immunity genes, bacteriocins, 809 Immunization, see Vaccines Immunoassays Bacillus cereus, 499 Cryptosporidium, 717 Giardia, 725 Immunodeficiency, see also Human immunodeficiency virus infection cryptosporidiosis in, 715–716 listeriosis in, 518–520 microsporidiasis in, 721 probiotics for, 956 Immunodiffusion assays, staphylococcal enterotoxins, 565 Immunoglobulin(s), hepatitis A virus, 629 Immunohistochemistry, bovine spongiform encephalopathy agent, 664 Immunologic tests, bovine spongiform encephalopathy agent, 655 Immunoprecipitation assays, chromatin, 989–990 Immunosuppression, aflatoxins in, 600 Inactivation models, 999–1001 Incident support, models for, 1012 Incubation period, in epidemiologic studies, 580

11/08/2012 07:24AM

Index

1097

Index organisms, 88–89 Indicator(s), microorganisms, 88–89 Indigenous fermentation, 918 INDISIM (individual discrete simulation), 1011 Individual discrete simulations, 1011 Individual-based modeling, 1010–1011 Infant botulism, 447–448, 450 Infant formula, powdered Cronobacter in, 311–312, 314–320, 324, 528 Salmonella in, 236 Infectious dose, see also specific organisms in epidemiologic studies, 580 Inflammatory bowel disease, in Campylobacter infections, 273 Information management (bioinformatics), 979–980 Infusion mashing, 903 Injury, microbial, 9–10 inl gene, Listeria monocytogenes, 520–521, 524–525 Inner forespore membrane, 53 Insect(s) in grains, 213 helminth transmission, 705 Insertion sequences, in antimicrobial resistance, 24 Inspection, for muscle food spoilage, 118 Instantaneous specific growth rate, 7–8 Institute of Food Research (UK), 1006 Integrins in antimicrobial resistance, 24 as invasin receptors, 349–351 Integrons, Salmonella, 245 Intelligent methodologies, for muscle food spoilage, 118 Intelligent packaging systems, 134 Intentional contamination, 91–108 agents for, 93–96 Clostridium botulinum, 441, 458 economic impact, 97 history, 92 in milk fermentation, 827 risk management, 97–103 CARVER+Shock strategy, 98–100 food system interventions for, 100–103 operational, 97–98 Shigella, 381 versus unintentional contamination, 92–97 Internal amplification control Cronobacter, 314 human enteric viruses, 638 Internal transcribed spacer-PCR, Cronobacter, 313–314 Internalin, Listeria monocytogenes, 520–521, 524–525 International Agency for Research on Cancer, mycotoxins, 597 International Commission on Microbiological Specifications for Foods, 81–88 International Commission on Trichinellosis, 681 International Nomenclature Commission for Staphylococcal SAgs, 550–551 Intestinal flukes Fasciolopsis buski, 698, 703 Heterophyes heterophyes, 698, 702

SMP_Food Microbiology_Index.indd

Intimin, Escherichia coli, 300–301 IntL protein, in antimicrobial resistance, 245 Intrinsic factors, in microbial growth, 8, 14 Inulin, as prebiotic, 966 inv gene, in acid tolerance response, 14–15 Invasins, Yersinia, 349–351 Invasion Campylobacter, 271 Cronobacter, 329 Invertase, in cocoa fermentation, 891 Ipa proteins, Shigella, 387–388, 391 ipg genes, Shigella, 387–388 Iron acquisition Salmonella, 251 Vibrio vulnificus, 424 Yersinia, 353–355 in protease production, 175 Iron binding regulatory protein, Salmonella, 233 Irradiation, see Radiation treatment Irrigation water, organisms in, 235 Irritable bowel syndrome, in Campylobacter infections, 273 Iso-a-acid isohumulone, in hops, 904 Isoascorbates, in meat fermentation, 863 Iso-b-acid hulupone, in hops, 904 Isobologram, in hurdle technology, 17–18 Isobutanoic acid, in muscle food spoilage, 117 Isoelectric focusing, 988 Isopentanol, in muscle food spoilage, 117 Isospora belli, 721 Isostatic principles, in pressure treatment, 754 Isothiocyanates, 783–784, 848

J

Jameson hypothesis, 1006 Jasmonates, for fruits and vegetables, 195 Jeotgal, 844 Johne’s disease, 129 Juices Clostridium pasteurianum in, 72 Escherichia coli O157:H7 in, 298 spoilage, 72

K

K antigens Escherichia coli, 287 Salmonella, 228 Kanagawa (thermostable direct) hemolysin, Vibrio, 417, 419 Kanamycin, 20, 28 Karmali agars, Campylobacter, 275 Kauffmann-White scheme, Salmonella nomenclature, 226, 1067 Kempner, on botulism, 442 Kerner’s disease (botulism), 442 Keto-deoxy-phosphogluconate aldolase, 5 Ketolides, 22 Khamira, 318 Kidney disorders hemolytic-uremic syndrome, 288, 293–294 Shigella, 384 Yersinia, 345 Killer yeasts and toxins, in wine, 923–924 Kilning, in brewing, 902 Kimchi, 844–848

1097

Manila Typesetting Company

Kinases, in sporulation, 49 Kinetics cucumber fermentation, 847–848 heat treatment, 8, 740–741 in microbial growth, 7–8 models for, 1002 pressure treatment, 756 temperature effects, 16–17 Klaenhammer classification, bacteriocins, 805 Klebsiella antimicrobial action on, 20, 773 in grains, 214 in milk, 177 in muscle foods, 144 in vegetable fermentation, 846 Klebsiella planticola, 145 Klebsiella pneumoniae in cheese, 178 in muscle foods, 145–147 Kloeckera antimicrobial action on, 773 in cocoa fermentation, 885 in winemaking, 925, 926 Kloeckera apiculata, 884, 920, 922 Kloeckera apis, 885 Kloeckera javanica, 885 Kluyveromyces, 920, 926 Kluyveromyces fragilis, 887 Kluyveromyces marxianus, 180 Koch’s postulates, 63 Kocuria in fish fermentation, 865 in meat fermentation, 857, 867–868 in seafood, 155 Kocuria varians, 863, 867 Kozakia, 932 Kreft model, 1010–1011 Kuru, 651 Kyoto Encyclopedia of Genes and Genomes, 980

L

Lactacins, 808, 809, 811 Lactate(s) antimicrobial action, 769–770 as spoilage energy source, 116 Lactate dehydrogenase, in glycolysis, 4 Lactic acid antimicrobial action, 769–770 in cocoa fermentation, 882, 884 Escherichia coli O157:H7 effects, 290 in fish fermentation, 865 in glycolysis, 4–7 in meat fermentation, 863, 864, 867, 873–874 in meat products, 132 in milk fermentation, 829 in muscle food decontamination, 123 in muscle food spoilage, 117 in poultry processing, 137, 277 in vegetable fermentation, 841–843, 847 in winemaking, 930, 931 Lactic acid bacteria, see also specific bacteria, e.g., Lactobacillus abortive bacteriophage infection systems, 834 antimicrobial action on, 803–804 antimicrobial resistance, 34–35

11/08/2012 07:24AM

Index

1098 Lactic acid bacteria (Continued) aroma compound production, 829–831 bacteriocin production, see Bacteriocin(s) bacteriophages, 834–836 biochemistry, 929–931 in cheese, 177–178 in cocoa fermentation, 884–888 in coffee fermentation, 894, 896 in competitive exclusion, 454 ecology, 928–929 genetics, 836 in grains, 213 GRAS status, 813–814 lactose metabolism, 827–829 in Listeria monocytogenes inhibition, 510 in malolactic fermentation, 928–932 in meat fermentation, 119, 857, 862–863, 867, 870 in milk, 177, 178, 825–836 in muscle foods, 114, 116–118, 127 nonstarter, 827 in poultry, 139 probiotic, see Probiotics production in situ, 803–804 proteolytic systems, 831–833 quorum sensing, 12 recombinant, 836 regulations, 813–814 signal transduction, 13 in starter culture, 825–827 in vegetable fermentation, 841–850 in winemaking, 928–932 Lacticins, 805 Lactitol, 966 Lactobacillaceae, osmoregulation, 15 Lactobacillus antimicrobial action on, 770, 771, 773–775, 781, 783, 784 as beer contaminant, 909–910 catabolism in, 5 in grains, 215 health benefits, 953 in meat fermentation, 867–876 in meat packaging, 125 in milk, 177 in muscle foods, 127, 147 in normal microflora, 952–953 in nuts, 205 radiation susceptibility, 753 in seafood, 155 taxonomy, 958–960 in vegetable fermentation, 843 in winemaking, 928–931 Lactobacillus acidophilus in cocoa fermentation, 885 genomics, 836, 962–964 phylogeny, 959 prebiotic stimulation, 966 probiotic action, 964 quorum sensing and, 12 Lactobacillus alimentarius, 870 Lactobacillus amylovorus, 959 Lactobacillus bavaricus, 510, 808 Lactobacillus brevis in cheese, 177–178 in cocoa fermentation, 885 in coffee fermentation, 895 genomics, 848–850 in meat fermentation, 859, 868, 870, 873

SMP_Food Microbiology_Index.indd

in milk fermentation, 826 in vegetable fermentation, 842, 844, 849–850 in winemaking, 928, 931 Lactobacillus buchneri, 868, 870, 873, 928 in cocoa fermentation, 885 in muscle foods, 145 Lactobacillus bulgaricus, 804 Lactobacillus casei in cocoa fermentation, 885 genomics, 836 in meat fermentation, 870, 871 in milk fermentation, 825 in muscle foods, 115 prebiotic stimulation, 966 Lactobacillus casei pseudoplantarum, 885 Lactobacillus casei strain Shirota, 949 Lactobacillus casei subsp. casei, 177–178 Lactobacillus casei subsp. pseudoplantarum, 177 Lactobacillus casei subsp. rhamnosus, 178 Lactobacillus cellobiosus, 885, 931 Lactobacillus collinoides, 885 Lactobacillus coryneformis, 844, 866 Lactobacillus crispatus, 959 Lactobacillus curvatus antimicrobial action on, 776 in meat fermentation, 860, 868, 869, 871, 874, 875 in muscle foods, 115, 145 Lactobacillus delbrueckii subsp. bulgaricus in cheese, 178 in cocoa fermentation, 885 in milk fermentation, 825, 828–831, 834 Lactobacillus delbrueckii subsp. lactis in cocoa fermentation, 887 in meat fermentation, 868 Lactobacillus farciminis, 868, 871 Lactobacillus fermentum in cocoa fermentation, 885, 887 in meat fermentation, 868 in winemaking, 928 Lactobacillus fructivorans, 885 Lactobacillus gallinarum, 959 Lactobacillus gasseri, 885, 959, 962, 964 Lactobacillus helveticus, 825–826, 828, 833 Lactobacillus hilgardii, 928, 931 Lactobacillus hordniae, 870 Lactobacillus johnsonii, 959, 964 Lactobacillus kandleri, 885 Lactobacillus kefir, 826 Lactobacillus lactis antimicrobial resistance, 34 in meat fermentation, 874 plasmids, 874 Lactobacillus mali, 885 Lactobacillus paracasei, 784 Lactobacillus paracasei subsp. paracasei, 869 Lactobacillus paraplantarum, 844 Lactobacillus pentosus, 869–870 Lactobacillus plantarum antimicrobial action on, 770, 782, 784, 808, 810 antimicrobial resistance, 31, 34 in cocoa fermentation, 885, 887 in coffee fermentation, 895 in fermented fish products, 866 genomics, 848–850, 964, 980

1098

Manila Typesetting Company

in meat fermentation, 860, 868–871, 874–875 metabolism in, 5, 7 in muscle foods, 126 plasmids, 874 prebiotic stimulation, 966 probiotic action, 964 in vegetable fermentation, 842–844, 846, 848–850 in winemaking, 928 Lactobacillus reuteri, 869 Lactobacillus rhamnosus antimicrobial resistant, 34 genomics, 984–985 in meat fermentation, 869 Lactobacillus sakei antimicrobial action on, 783 in meat fermentation, 860, 868, 871, 874, 876 in muscle foods, 116 in vegetable fermentation, 848 Lactobacillus salivarius, 961, 964 Lactobacillus suebicus, 868 Lactobacillus trichodes, 928 Lactobacillus viridescens, 127, 859 Lactobacillus xylosus, 870 Lactocins, 805, 874 Lactococcins, 805, 809, 811 Lactococcus antimicrobial action, 772 antimicrobial resistance, 34 bacteriophage resistance, 834–835 bacteriophages, 831–833 catabolism in, 5 in fermented fish products, 865 in meat fermentation, 871, 876 in milk, 177 in milk fermentation, 829 proteolytic systems, 831–833 restriction/modification systems, 835 Lactococcus lactis, 804, 808 acid tolerance response, 991 antimicrobial action on, 776 antimicrobial resistance, 32 bacteriophage resistance, 834–835 in cocoa fermentation, 885 in dairy products, 180 genomics, 848–850, 980, 981 in milk fermentation, 825–826 models for, 1005–1006 predictive models for, 999 probiotic action, 957 in vegetable fermentation, 849 Lactococcus lactis subsp. cremoris, 825, 835 Lactococcus lactis subsp. lactis in meat fermentation, 870 in milk fermentation, 177, 825–826, 829–831, 834–836 Lactococcus lactis subsp. lactis var. maltigenes, 177 Lactoferricin, antimicrobial action, 773–774 Lactoferrin antimicrobial action, 318, 773–774 in milk, 170–171 muscle food decontamination, 123 Salmonella competition with, 251 Yersinia enterocolitica, 353 Lactoperoxidase, 171, 246–247, 779

11/08/2012 07:24AM

Index

1099

Lactose digestion, probiotics in, 956 metabolism, in lactic acid bacteria, 827–829 in milk, 170 Salmonella utilization, 226 Lactostrepcin, 810 Lactulose, 966 Lag phase, microbial growth, 7, 1001, 1007 Lager tun, in brewing, 903–904 Lagovirus, 621 Lanthionine, as bacteriocin, 804 Lantibiotics, 804–805, 809–810 Larva migrans, visceral, 708 Larvae, helminth, see specific helminths Lasalocid, 21 Late gas defect, in cheese, 179–180 Latex agglutination test, 469 Lauric arginate, 132, 768, 772–773 Lauryl sulfate tryptose broth, modified, Cronobacter, 322–323 Lcr protein, Yersinia, 359 Le Chatelier’s principle, 754 Le Mark model, 1004, 1007 Lecithinase, Clostridium detection, 455 Legislation and regulations aquaculture, 141 bacteriocins, 813–814 bovine spongiform encephalopathy agent, 660–664 muscle food processing, 130–131 muscle food spoilage, 118 Leloir pathway, 828 Lemongrass, 782 Lethality measurements, 741–742 Leucocins, 805, 809 Leuconostoc antimicrobial action on, 781 catabolism in, 5 in cocoa fermentation, 885 in meat fermentation, 871, 876 in milk, 177, 826, 830–831 in muscle food spoilage, 127 in muscle foods, 115 in nuts, 205 in winemaking, 928–929 Leuconostoc argentinum, 844 Leuconostoc brevis, 885 Leuconostoc citreum, 844 Leuconostoc fallax, 844 Leuconostoc mesenteroides, 808 antimicrobial action on, 776 in cocoa fermentation, 885 in coffee fermentation, 895 in fruits and vegetables, 192 genomics, 848–850 in meat fermentation, 873 in muscle foods, 115 in vegetable fermentation, 842–844, 847, 849 in winemaking, 928, 931 Leuconostoc oenos, 808, 885 Leuconostoc paramesenteroides, 885 Leuconostoc pseudomesenteroides, 887 Leucotrienes, in staphylococcal food poisoning, 566 Leukocyanidins, oxidation, 892 Levulinic acid, 772 Lignin, in spoilage resistance, 193

SMP_Food Microbiology_Index.indd

Ligula intestinalis, 698, 700 Linalool, 781 Lincomycin, 21 Lincosamides, 21 Lineage classification, Listeria monocytogenes, 506 Linoleic acid, conjugated, in meat fermentation, 869 Lipases Clostridium detection, 455 in dairy products, 175–176 in milk, 175–176 Lipolysis, milk, 175–176 Lipooligosaccharide, Campylobacter, 266, 272 Lipopolysaccharides (endotoxins) Cronobacter, 324 Salmonella, 244, 250–251 Vibrio cholerae, 415 Vibrio vulnificus, 424–425 Yersinia enterocolitica, 353 Lipoprotein maturase, 832 Liposomes, antimicrobials in, 785–786 Liquid chromatography, in proteomics, 988 Liquid smoke, antimicrobial action, 784 Listeria antimicrobial action on, 781, 784 bacteriocin action on, 808 disease epidemiology, 588 outbreaks, 578 as indicator, 89 in milk, 173 models for, 1013, 1014 in muscle foods, 114 probiotic protection from, 955 species, 504 taxonomy, 504–506 Listeria grayi, 504, 520 Listeria innocua, 504 antimicrobial action on, 771, 782, 784, 785 antimicrobial resistant, 34 bile tolerance, 992–993 genomics, 520, 980 models for, 1004–1006 predictive models for, 999 in seafood, 154 Listeria ivanovii, 504, 772 Listeria marthii, 504, 520 Listeria monocytogenes, 503–545 acid tolerance response, 14, 767, 991–992 adhesion, 524 animal studies, 520–522 antimicrobial action on, 768–775, 778–781, 783–786, 874 antimicrobial resistance, 28, 31, 32, 34 bacteriocin action on, 119, 804–805, 807–808, 812–813 bacteriophages and, 814–816 biochemical identification, 504–506 in biofilms, 13 as biological hazard, 1044–1045 carriage, 513–514 characteristics, 504–506 classification, 504 contamination, 509–512 control, 37, 504 D values, 9

1099

Manila Typesetting Company

in dairy products, 507–509, 514–517 discovery, 503 disease asymptomatic, 520 characteristics, 518–520 epidemiology, 503, 518–519 history, 503 invasive, 519 outbreaks, 503, 514–518, 580, 585 disinfectant effects, 32 environmental susceptibility, 506 epidemic clones, 505–506, 516–517 escape from phagosome, 526–527 factors affecting, 503 food preservation effects on, 510–511 in food processing plants, 512–513 in fruits and vegetables, 513 genomics, 520, 980, 981, 987–988 growth, 506, 746 hazard analysis, 1047 heat resistance, 33–34 heat treatment, 748 high-pressure processing, 124 host response, 529 infectious dose, 519 inhibitors, 134 internalin, 520–521, 524–525 intracellular growth, 526–527 intracytoplasmic movement, 527–528 lineage classification, 506 in meat fermentation, 862, 869 in milk, 172, 743 models for, 1004–1007, 1012, 1014 in muscle foods, 114, 115, 119, 120, 126, 130, 132, 509–510 in nuts, 206–207, 212 osmoregulation, 15 p60 protein, 525 pathogenicity, 520–529 penetration of cell, 524–526 pH effects, 14 in poultry, 137, 138 pressure effects, 757–758 prevalence in foods, 512–513 probiotic action in, 964 proteomics, 992–993 public health significance, 503–504 pulsed-electric field processing, 757 quorum sensing, 12 radiation susceptibility, 124–125 radiation treatment, 511 in ready-to-eat products, 132, 507–510 refrigeration effects, 746 regulations, 512–513 reservoirs, 511–514 risk assessment, 1026–1027, 1030, 1033 in seafood, 140, 141, 154 serotypes, 504–505 spreading, 527–528 stress proteins, 529 subtyping, 504–506 susceptible populations, 519–520 temperature effects, 9–10, 17 transmission, 511, 519 viable but nonculturable, 10 virulence factors, 520–529 water activity requirements, 506, 750 zero tolerance, 513 Listeria seeligeri, 504, 520

11/08/2012 07:24AM

Index

1100 Listeria welshimeri, 504, 520 Listeriolysin O, 526 Liver disorders, see also Hepatitis aflatoxin-induced, 600 probiotics for, 956 Liver flukes, 121 Chinese (Clonorchis sinensis), 698, 700 sheep (Fasciola hepatica), 698, 702–703 LLO (listeriolysin O), 526 Lobsters, 117 Locus of effacement pathogenicity island, Escherichia coli, 289, 299–300 Log phase, microbial growth, 7 Logistic regression models, 1002–1003 Long open reading frames, in genomics, 979 Low-atmosphere pressure stunning, in poultry processing, 134 Low-moisture foods, Salmonella in, 235 Low-temperature preservation, 745–748; see also Freezing; Refrigeration LPXTG proteins, Listeria monocytogenes, 525–526 Lu (meat dish), Trichinella in, 678 Lung fluke (Paragonimus westermani), 701 Lupulone, 783, 904 LuxS protein, quorum sensing, 11–12 Lye, in olive fermentation, 845–846 Lymphoid tissue, Yersinia enterocolitica, 349 Lyophilization (freeze-drying), 56, 751–752 Lysine decarboxylase, Shigella, 391 “Lysis from without,” 814 Lysozyme antimicrobial action, 774–776 encapsulation, 785–786 spore resistance to, 57–58 Lytic enzymes, in spore cortex, 62

M

M cells Shigella, 386 Yersinia enterocolitica, 349, 352 Macadamia nuts, organisms in, 207 McKeekin model, 1000 McKellar model, 1007, 1011–1012 Macracanthorhynchus hirudinaceus, 698, 705 Macrolides, 21, 22 Macroscopic models, 1008–1010 Mad cow disease, see Bovine spongiform encephalopathy Magainins, 812 Magnetic fields, meat processing, 124 Maize aflatoxins in, 598–603 fumonisin in, 606–610 Major histocompatibility complex molecules, staphylococcal enterotoxin binding to, 562–563 Malabsorption, Giardia, 725 Malakar model, 1006 Malic acid, 771, 926, 930 Malolactic fermentation, 842–843, 918, 928–932 Malt extract agar, Aspergillus, 598 Malting, in beer production, 901–903 Malty flavor, milk, 170, 177 Map protein, Escherichia coli, 301 Marinated meat products, 131–132 Mashing process, in brewing, 903–904

SMP_Food Microbiology_Index.indd

Mass spectrometry aflatoxins, 601 Clostridium botulinum neurotoxins, 456 in proteomics, 988 Mast cells, in staphylococcal food poisoning, 566–567 Master bag, for no-oxygen packaging, 133 Mathematical models, see Predictive models Matrix metalloproteinases, Cronobacter, 330 Measurement errors, in models, 1003 Meat and bone meal, bovine spongiform encephalopathy agent in, 656–658, 660–664 Meat and meat products antimicrobial resistance and, 120–121 antimicrobials for, 769 bacteriocins in, 807 bacteriophages for, 815 biofilms on, 114–115 contamination, 111, 125–126 decontamination, 123–125, 131 environmental contamination due to, 121 fermented, 118–119 categories, 857–858 composition, 858–859 factors affecting quality, 859, 862–865 failure, 863, 865 flavor, 859, 862–864 manufacture, 857–865, 859–861 models for, 1014 processes for, 859 production, 118–119 starter cultures, 866–876 fresh, spoilage, 126–127 high-pressure processing, 124 initial microflora, 111–115 inspection procedures, 685–686, 689 irradiation treatment, 124–125 microbial cell attachment, 114–115 modified-atmosphere packaging, 747 nonintact, 131–132 nonthermal treatments, 123–124 organisms in, 128–130 Bacillus cereus, 492 bovine spongiform encephalopathy agent, 130, 651–674 Campylobacter, 269–270 clostridia, 127 Clostridium botulinum, 69–70, 444, 449 Clostridium butyricum, 72 Clostridium gasigenes, 72 Clostridium pasteurianum, 72 Clostridium perfringens, 467–469 Escherichia coli, 291–293, 295–297 Escherichia coli O157:H7, 586, 588 helminths, 673–696, 705–706 Listeria monocytogenes, 509–512 Penicillium, 605 Salmonella, 234–237, 239–241, 245 sporeformers, 64, 72 Taenia, 119, 682–686, 698–690 Toxoplasma gondii, 722 Yersinia enterocolitica, 346, 347 packaging, 125, 132–134 pathogens in, 119–120 preharvest pathogen control, 131 prions in, 130 processed, spoilage, 127–128

1100

Manila Typesetting Company

processing, 122–125, 130–132 regulations, 130–131 safety, 128–130 spoilage, 126–128 chemical changes in, 116–118 comminuted products, 126–127 compounds formed, 116–118 dark, firm, dry (DFD), 126, 862 development, 115–116 pale, soft, exudative (PSE), 126–127 processed products, 127–128 quality evaluation, 118 substrate utilization for, 116–118 types, 116 traceability, 121–122 water activity, 750 Meat & Livestock Australia Limited calculator, 1014 Meat starter cultures, 866–876 characteristics, 869–870 classification, 870–871 development, 867–868 functional, 868–869 genomics, 874–876 metabolism, 871–874 Megasphaera, 910 Megasphaera elsdenii, 31 Mejlholm-Dalgaard model, 1006 Melittin, 812 Membrane, alterations, in antimicrobial resistance, 22 Membrane filtration, for beer, 908 Membrane potential, 6–7 Membrane transport protein, Campylobacter, 266 Meningitis Cronobacter, 322, 326–328 Listeria monocytogenes, 518–519 Merogony, Cryptosporidium, 715 Mertems model, 1005 Mesenterocin, 812 Mesophiles, 16, 58, 65 Mesoscopic models, 1007–1012 Metabolic flux analysis, 1008 Metabolic network models, 1007–1010 Metagonimus yokagawai, 698, 702 Metal detectors, for food screening, 93 Metchnikoff, Elie, 949 Methanethiol, in cheese flavor, 833 Methicillin, resistance to, 29 Methoxicillin, resistance to, 28 Methyl chavicol, 781–782 Methyl esters, in muscle food spoilage, 117 Methyl ethyl ketone, in muscle foods, 117 Methyl paraben, 777–778 3-Methylbutanal, in milk spoilage, 177 Methylbutanol, in muscle food spoilage, 118 Methyllanthione, as bacteriocin, 804 Methylobacillus flagellatus KT, 315 2-Methylpropylamine, in fish fermentation, 866 4-Methyl-umbelliferyl-a-d-glucosidase, 321 Metmyoglobin, 132–133, 872 Metschnikowia, 919–920, 925 Metschnikowia fructicola, 196 Metschnikowia pulcherrima, 920, 922 mgtCB gene, Salmonella, 247 MIC, in hurdle technology, 18 Micelles, antimicrobials in, 785–786

11/08/2012 07:24AM

Index

1101

Microarray analysis, see DNA microarray analysis Microbacterium, in nuts, 205 MICROBExpress kit, 988 Microbial death in heat treatment, 737–745 versus injury, 9–10 kinetics, 7 radiation-induced, see Radiation treatment Microbial growth, 3–18; see also Fermentation; specific organisms and products acid effects on, see Acid(s) anaerobes, 5–6 atmospheric effects, see Atmosphere bioenergetics, 6–7 biofilms, 13–14 cell signaling in, 11–13 in controlled atmosphere, 747 ecosystems, 8 extrinsic factors affecting, 8, 14 first-order kinetics, 7–8 homeostasis and, 14, 17–18 hurdle technology, 17–18 injured cells, 9–10 intrinsic factors affecting, 8, 14 metabolic pathways, 3–7 pH effects on, see pH physiology, 5 quorum sensing in, 11–12 in refrigeration, see Refrigeration salt effects on, see Salt signal transduction in, 13 temperature effects on, see Temperature effects viable but nonculturable cells, 9–11, 408 water activity and, 748–752 zero, see Food preservatives and preservation methods Microbial injury, 9–10 Microbial interaction models, 1006–1007 Microbiological criteria, 81–90 control measures and, 87–88 definitions, 81, 82 elements, 84 establishment, 83–84 history, 81–83 index and indicator organisms, 88–89; see also Indicator(s) microbiologic profile and, 86–87 need for, 81–82 objectives, 83 performance-based, 83 recently issued, 85 risk management metrics for, 82–83 sampling plans, 83–86 types, 85 Microbiological methods, rapid, see Rapid microbiological methods Microbiological risk management metrics, 82–83 Micrococcaceae, 873 Micrococcus, 548 antimicrobial action on, 772, 773, 775, 780 as beer contaminant, 910 in fermented fish products, 866 in grains, 213 in meat fermentation, 857, 870

SMP_Food Microbiology_Index.indd

in milk, 826 in muscle foods, 126 in nuts, 205 in seafood, 155 Micrococcus flavus, 10 Micrococcus luteus, 806 Microgard, 803–804 Micronutrients, in milk, 170 Microscopic models, 1007–1012 Microscopic ordering principle, 754–755 Microscopy, Cyclospora, 719 Microsporidia, 713, 723–724 Microwave heat treatment, 745 Cronobacter, 315 grains, 215 Salmonella, 234 Midges, viruses transmitted by, 121 Miles model, 1000 Milk aseptic packaging, 745 canned condensed, 179 carbon sources in, 170 composition, 169–171 contamination, 172–173 defects, 170, 174, 176–177, 179 canned condensed, 179 control, 179 molds in, 180–181 yeasts in, 180–181 equipment for, psychrotrophic bacteria in, 173 fermented, see Dairy products, fermented as growth medium, 169–172 heat treatment, 171–172 lactoferrin in, 170–171, 773–774 lactoperoxidase in, 171, 779 lipases in, 175–176 microbial inhibitors in, 170–171 micronutrients in, 170 minerals in, 170 nitrogen availability in, 170 organisms in, 743 Aeromonas, 173 Bacillus, 173, 178–179 Bacillus cereus, 492–493, 498 bovine spongiform encephalopathy agent, 656 Brucella, 172 Campylobacter, 172, 269–270 Clostridium, 179–180 coliforms, 178 Enterobacteriaceae, 173, 177 Enterococcus, 173, 177 Escherichia coli, 290–291 Escherichia coli O157:H7, 172 Klebsiella, 177 lactic acid bacteria, 170, 177 Lactococcus, 177 Leuconostoc, 177 Listeria, 173 Listeria monocytogenes, 172, 507–509, 512 molds, 180–181 Mycobacterium, 172 Pediococcus, 177 Pseudomonas, 173–176 psychrotrophic, 170, 173–176 Salmonella, 236–237 Shigella, 383

1101

Manila Typesetting Company

Staphylococcus, 173 Staphylococcus aureus, 547, 557 Streptococcus, 177 yeasts, 180–181 Yersinia enterocolitica, 346–347 pasteurization Escherichia coli, 290–291 Listeria monocytogenes growth after, 507–509, 512 pH, 172 proteases in, 174–175 public health significance of, 172–173 raw, handling, 176 refrigeration, 173 shelf life, 173 spoilage, 173–181 fermentative nonsporeformers in, 177–178 lipases in, 175–176 molds in, 180–181 proteases in, 174–175 psychrotrophic, 173–176 spore-forming bacteria in, 178–180 yeasts in, 180–181 ultra-high-temperature protease defects, 174–175 sporeformers in, 179 Milk lipase, 175 Milling, grains, 214–215 MILQ web tool, 1014 Minerals in milk, 170 in spores, 59 Minimum Convex Polyhedron, 1002 Minnesota Department of Health laboratory surveillance program, 577–578, 584 Minocycline, 22 Mixing, heat transfer and, 740 Mobile gene elements, 24, 291 Model(s), predictive, see Predictive models Model function errors, in models, 1003 Modified-atmosphere packaging, 747 Clostridium botulinum growth in, 454 fruits, 193–194 high-oxygen, 133 muscle foods, 116, 125 nisin in, 807 no-oxygen, 133–134 poultry, 139 Salmonella growth in, 230 vegetables, 193–194 Moisture in cocoa fermentation, 889–890 in sausage, 864 Moisture control agents, for active packaging, 134 Moisture-to-protein ratio, fermented meat products, 858–859 Mold(s), see also specific organisms antimicrobial action on, 768 in cheese, 180–181 in coffee fermentation, 894 in dairy products, 180–181 in grains, 214–217 in meat fermentation, 118–119, 869–870 in milk, 180–181 in muscle foods, 113 in nuts, 204–209 osmoregulation, 15–16

11/08/2012 07:24AM

Index

1102 Mold(s) (Continued) toxigenic, see Mycotoxins in vegetables, 190–192 water activity requirements, 750 in winemaking, 934–935 Molecular analysis, noroviruses, 630 Molecular epidemiology, definition, 1059 Molecular mimicry hypothesis, 363–364 Molecular pattern elicitors, 193 Molecular subtyping, 1059–1077 amplification fragment length polymorphism, 1062–1063 Campylobacter, 267–269 definition, 1059 DNA sequencing, 1064–1066 electrophoresis, 1061–1062 for epidemiology, 584–585 future, 1071–1072 ideal method, 1071 indications for, 1059–1060 methods, 1060–1066; see also specific methods multilocus sequence typing, 1063, 1067, 1069–1070 multiple locus variable-number tandem repeat analysis (MLVA), 1063–1065, 1069, 1071–1072 phenotypic, 1060 plasmid profile analysis, 1060–1061 polymerase chain reaction, see PCR restriction fragment length polymorphism analysis, 1061–1062 Salmonella, 1067–1069 versus serotyping, 1060 single nucleotide polymorphisms, 979, 1066, 1069, 1071 Monascus bisporus, 16 Monensin, 21, 29 Moniliformis moniliformis, 698, 705 Monilinia, 188, 935 Monilinia fructicola, 191, 194, 195 Monilinia fructigena, 192 Monilinia laxa, 195 Monitoring, in HACCP system, 1050–1051 Monkey feeding assay, staphylococcal enterotoxin, 560, 566 Monobactams, 20 Monocaprylin, 317–318, 772 Monolaurin, antimicrobial action, 772 Monoterpenes, in wine, 926 Monte Carlo simulations, in models, 1002 Moraxella in fermented fish products, 866 in modified-atmosphere packaging, 747 in muscle foods, 126, 127 in poultry, 134 in seafood, 154 Morganella, 144, 146 Morganella morganii, 145–148 Most-probable method, Campylobacter, 274 Mother cells, in sporulation, 45–52, 54 Motility Campylobacter jejuni, 271 Vibrio cholerae, 415 Mouse bioassay, for Clostridium botulinum neurotoxin detection, 446, 456 Moxifloxacin, 20 Moxolactam, for Cronobacter infections, 331

SMP_Food Microbiology_Index.indd

mpl genes, Listeria monocytogenes, 528 MRSA (methicillin-resistant Staphylococcus aureus), 29 Mucor antimicrobial action on, 770, 781 in cheese, 180–181 in grains, 213 in nuts, 205 Mucor piriformis, 196 Multiceps multiceps, 698, 708 Multidimensional protein identification technology, 988 Multidrug resistance, 33 Multilayer perception, for muscle food spoilage, 118 Multilocus genotyping, 1070 Multilocus sequence typing, 979, 1063, 1067, 1069–1070 Campylobacter, 267–269 Clostridium perfringens, 467 Listeria monocytogenes, 505 Salmonella, 229 Vibrio, 403 Multiple locus variable-number tandem repeat analysis (MLVA), 1063–1065, 1069, 1071–1072 Multiple-hurdle systems, meat processing, 122–123 Multiplex single nucleotide polymorphism, Listeria monocytogenes, 505 Multiplicity of infections, bacteriophages, 814–815 Multivariate analysis, for muscle food spoilage, 118 Multi-virulence-locus sequence typing, 504, 1070 Murine norovirus, 633–634 Muscle foods, see also Meat and meat products; Poultry; Seafood antimicrobial resistance and, 120–121 biofilms on, 114–115 contamination, 111 decontamination, 123–125 fermented, 118–119, 866–876 initial microflora, 111–114 microbial attachment, 114–115 microbial ecology, 111–115 organisms in, see also Taenia Acinetobacter, 115, 126–127, 145 Aeromonas, 119, 126 Alcaligenes, 126 Alteromonas putrefaciens, 145 Arcobacter, 119 Bacillus, 119, 126 Bacillus cereus, 119 Brochothrix thermosphacta, 115–118, 126–127 Brucella, 119 Campylobacter, 114, 119–120, 123, 126 Carnobacterium, 115 Cedecea, 145 Citrobacter, 117 Clostridium, 126–127 Clostridium botulinum, 119, 125 Clostridium difficile, 119, 129–130 Clostridium perfringens, 119, 145 Cronobacter, 319 Cryptosporidium, 119–120 Enterobacter, 116, 119, 145–147

1102

Manila Typesetting Company

Enterobacteriaceae, 116–117, 126 Enterococcus, 119, 127 Escherichia coli, 114, 119, 123, 128–129 Escherichia coli O157:H7, 114, 119–120, 126, 128–132 Flavobacterium, 126 Hafnia alvei, 117, 126, 144 Helicobacter, 119 helminths, 673–696, 698 Klebsiella, 144–147 Lactobacillus, 115–116, 125–127, 145, 147 Leuconostoc, 115, 127 Listeria, 114 Listeria monocytogenes, 114–115, 119–120, 126, 130, 132 Micrococcus, 126 Moraxella, 126–127 Morganella, 146–148 Mycobacterium, 119, 129 Pasteurella, 115–116, 118, 145 Plesiomonas shigelloides, 141 Proteus, 117, 144–147 Providencia, 155 Pseudomonas, 114–117, 126–127, 145 Pseudomonas fluorescens, 116, 145 Pseudomonas fragi, 116–117 Pseudomonas ludensis, 116 Pseudomonas putida, 116, 145 Pseudomonas putrefaciens, 145 Rahnella, 115 Salmonella, 119–120, 123, 126, 131 Sarcocystis, 119 Serratia, 115–117, 145 Shewanella liquefaciens, 126 Shewanella putrefaciens, 115–118, 126 Shigella, 119 Staphylococcus, 126 Staphylococcus aureus, 119, 126 Staphylococcus xylosus, 145 Streptococcus, 127 Toxoplasma gondii, 119–120 Trichinella spiralis, 119 Vibrio, 120 Vibrio alginolyticus, 145–146 Weisella viridescens, 115 Yersinia enterocolitica, 119–126 packaging, 125 pathogens in, 119–120 pH, 126 processing, microbial control in, 122–125 spoilage in aerobic conditions, 117 in anaerobic conditions, 117 chemical changes in, 116–118 comminuted products, 126–127 compounds formed, 116–118 dark, firm, dry (DFD), 126, 862 development, 115–116 processed products, 127–128 quality evaluation, 118 substrate utilization in, 116–118 types, 116 traceability, 121–122 Mussels, 152–153 Mutations, in antimicrobial resistance, 24–25 Mxi proteins, virulence factors, 387–389, 391

11/08/2012 07:24AM

Index

1103

Myalgia, in trichinellosis, 678–679 Mycobacterium in milk, 172, 743 in muscle foods, 119 Mycobacterium avium subsp. paratuberculosis, 129 Mycobacterium tuberculosis acid tolerance response, 991 subtyping, 1066 Mycocentrospora acerina, 191 Mycotoxins aflatoxins (Aspergillus), 598–603 Aspergillus, 598–606 definition, 597 deoxynivalenol (Fusarium), 610–611 fumonisins, 606–610 Fusarium, 606–611 genomics, 611–612 in grains, 213–214, 598–611 in muscle foods, 120 in nuts, 209–210 ochratoxin A (Aspergillus), 603–606 structures, 597 in wine, 923–924, 934–935 zearalenone (Fusarium), 611 myf genes, Yersinia, 352–353 Myocarditis, Trichinella, 678 Myoglobin in fermented meat products, 872 in muscle foods, 117, 125

N

NAD regeneration, 5–6 in lactose metabolism, 827–829 in meat fermentation, 872 NADH, in tricarboxylic acid cycle, 5 NAG-ST (nonagglutinable Vibrio Shiga toxin), 414 Nahm, Trichinella in, 678 Nalidixic acid, resistance to, 28 Nam-pla, 865–866 Nanoencapsulation, food antimicrobials, 785–786 Nanophyetus salmincola, 698, 702 Naresuchi, 866 NARMS (National Antibiotic Resistance Monitoring System), 27–28, 35, 290 National Advisory Committee on Microbiological Criteria for Foods, 1040–1041, 1043–1046 National Animal Health Monitoring System, 290, 586 National Antibiotic Resistance Monitoring System (NARMS), 27–28, 35, 290 National Antimicrobial Resistance Monitoring System, 129, 245 National Center for Biotechnology Information, 980 National Enteric Reference Laboratories, 1071 National Food Processors Association, Listeria monocytogenes survey, 513 National Health Objectives, 575–576 National Institutes of Health Human Microbiome Project, 953 National Molecular Subtyping Network (PulseNet), 505, 584, 1059, 1070–1072 Natural competence, Campylobacter, 267

SMP_Food Microbiology_Index.indd

Necrotic enteritis, Clostridium perfringens, 465 Necrotizing enterocolitis, Cronobacter, 322, 326–329 Nectria galligena, 193 Neomycin, 20–21 Neonates, listeriosis in, 518, 519 Neorickettsia helmintheca, 702 Neosartorya, 71 Neosartorya fischeri, 769 Nesterenkonia halobia, 867 Nestlé, models developed by, 1012 Neural networks, 118, 999, 1001, 1004 Neurocysticercosis, 687–688 Neurologic disorders, in Escherichia coli infections, 304 Neuronal cell-based assays, Clostridium botulinum neurotoxins, 446, 457 Neuropathy, botulism, 447 Neurotoxins, in botulism, see Clostridium botulinum, neurotoxins Newbler software program, 979 Nham, 782 nhe genes, Bacillus cereus, 492, 496–499 Nicotine sulfate, intentional contamination with, 95 Nikolaou-Tam model, 1012 NimbleGen platform, 985–987 Nisins in active packaging, 134 activity, 805–806 encapsulation, 785–786 food applications, 806–807 genomics, 808–810 in hurdle technology, 18 in Listeria monocytogenes inhibition, 132, 510 lysozyme with, 775 in meat fermentation, 875 mechanism of action, 810–812 regulations, 813–814 resistance to, 32, 34, 812–813 structures, 805 Nitrates, in fermented meat products, 859, 863, 865, 872–873 Nitrites antimicrobial action, 776–777, 804 carcinogen formation from, 143 Clostridium botulinum inhibition, 804 in fermented meat products, 127, 859, 863, 865, 872–873 Nitrogen in muscle food spoilage, 125 in packaging, 747 p-Nitrophenylglycosides, in vegetable fermentation, 848 Nitrosometmyoglobin, 873 Nitrosomyoglobin, 873 Nivalenol, 610 Nixtamalization, for mycotoxin removal, 610 Nonagglutinable Vibrio Shiga toxin, 414 Nonclimacteric commodities, 189 Nonhemolytic enterotoxin, Bacillus cereus, 492, 496–499 Nonintact meat products, 131–132 Nonsterile unit concept, 64 Nonthermal processing, meat, 123–125 Noriega model, 1005

1103

Manila Typesetting Company

Normal microbiota, 949–953 health benefits, 956–957 prebiotic interactions with, 965–966 probiotic interactions with, see Probiotics Norovirus genus, 620 Noroviruses control, 632–635 detection, 63, 635–639 discovery, 621 disease asymptomatic, 627 clinical features, 626–627 epidemiology, 582, 625–626 outbreaks, 626–627 pathogenesis, 628 treatment, 628–629 environmental persistence, 631 epidemiology, 621–623 genotypes, 639 heat resistance, 632 immunity to, 627 in muscle foods, 120 in ready-to-eat foods, 634–635 surveillance, 587 susceptibility to, 627 transmission, 626 vaccines, 628 Northern blot test, CPE protein, 476–477 Norwalk virus, 621, 622 Novobiocin, 22 Nucleic acid-based assays, see also PCR enteric viruses, 637 Nucleotide annotation, 979–980 Numerical procedure errors, in models, 1003 Nuoc-mam, 865 Nuoc-mam-nuoc, 865 Nutrient germinant receptors, 56 Nutrient germinants, 61 Nutrient limitation, in sporulation, 46 Nuts, 203–213 aflatoxins in, 598–603 contamination, 206, 209–210 definition, 203 harvesting, 204 microflora, 204–205 organisms in Achromobacter, 205 Acinetobacter, 205 Acremonium, 205 Alternaria, 205 Aspergillus, 205, 209 Bacillus, 208 Brevibacterium, 205 Chaetomium, 205 Cladosporium, 205 Clostridium, 205, 209 Corynebacterium, 205 Escherichia coli, 205, 209–212 Flavobacterium, 205 Fusarium, 213 Lactobacillus, 205 Leuconostoc, 205 Listeria monocytogenes, 206–207, 212 Microbacterium, 205 Micrococcus, 205 Paecilomyces, 205 Penicillium, 205 Phialophora, 205 Phomopsis, 205

11/08/2012 07:24AM

Index

1104 Nuts, organisms in (Continued) Proteus, 205 Pseudomonas, 205 Rhizopus, 205 Salmonella, 204–212 Staphylococcus, 205–206, 208 Streptococcus, 205 Trichosporon, 205 Trichothecium, 205 Xanthomonas, 205 processing, 204, 210–212 spoilage, 205 storage, 204, 210–212 Nybelinia surmenicola, 698, 702

O

O antigens Cronobacter, 313 Escherichia coli, 287 Salmonella, 227–228 Shigella, 385 Yersinia enterocolitica, 341–342 Obesumbacterium, 910 Ochratoxin A, 603–606, 934 Ochrobacterium, 153 Octanoic acid, 772 Odors muscle food spoilage, 116–118 in poultry spoilage, 138–139 Oenococcus, in meat fermentation, 876 Oenococcus oeni, 918, 928–929, 931 antimicrobial action on, 778 in winemaking, 926–930 Oesophagostomum, 707 Oh-Kang medium, Cronobacter, 321 Ohmic heat treatment, 745 Oil roasting, nuts, 212 Oleuropein, 848 Oligofructose, as prebiotic, 966 Oligonucleotide assays, Cronobacter, 314 Oligonucleotide/oligosaccharide binding fold, staphylococcal enterotoxins, 561 Oligopeptide transport system, in milk fermentation, 832 Oligosaccharides, as prebiotics, 966 Olives, 845–848 Olorosos sherry, 936 omp genes, Salmonella, 250 Omp proteins, Cronobacter, 330 Onions, 782–783 Oocysts Cryptosporidium, 715–718 Cyclospora, 718–720 Isospora, 722 Sarcocystis hominis, 723 Toxoplasma gondii, 721–722 Operational risk management, for biosecurity, 97–98 Operational support, models for, 1012 Opisthorchis, 698, 700 Opsonization, resistance to, Yersinia, 356–357 Opti-Form Listeria Control Model, 1014 Optimization, in models, 1009–1010 opuBB gene, Salmonella, 232 OpuC protein, in bile tolerance, 992 Oregano, 781

SMP_Food Microbiology_Index.indd

Organic acids and esters, see also individual compounds antimicrobial action, 766–773 muscle food decontamination, 123 in poultry processing, 137–138 in wine, 926, 930, 932 Organic farming, antimicrobial resistance and, 25 Organoleptic examination, for intentional contamination detection, 96 Organophosphates, intentional contamination with, 95 Ormetoprim, in aquaculture, 26 Osmolytes, 992 Osmoprotectants, 10 Campylobacter, 265 Salmonella, 232 Osmoregulation, 15–16 Osmotolerance, 556–558, 751 osp genes, Shigella, 388, 390 Outbreaks, see also specific organisms information sources, 588–589 investigation, 576–580 molecular subtyping in, 1059 Outer membrane proteins Cronobacter, 330 Salmonella, 233 Outgrowth, in sporulation, 62 Ovotransferrin, 780 Oxalate degradation, probiotics, 962 Oxidation in cocoa fermentation, 891–892 in microbial metabolism, 3, 5–7 Oxidative stress response regulatory elements, Campylobacter, 265 Oxidative systems for acid resistance, 14 Escherichia coli, 290 Oxidizing agents, spore resistance to, 57–58 Oxoid assay, Bacillus cereus, 499 Oxygen in cell injury, 10 Clostridium botulinum growth and, 453–454 Clostridium perfringens growth and, 466 in muscle food packaging, 133 in muscle food spoilage, 125 in packaging, 133, 193, 747 radiation treatment with, 754 reduced, in packaging, 193 Oxygen scavengers, for active packaging, 134 Oxymyoglobin, 872 formation, 126 preservation, 132–133 OxyR protein, Campylobacter, 265 Oxytetracycline, 21, 22, 25–26 Oysters, 150–152 Ozone for Clostridium botulinum, 451 for Cryptosporidium, 718 for enteric viruses, 633 for fruits and vegetables, 194, 633 for muscle food decontamination, 123 for poultry processing, 136–137

P

p60 protein, Listeria monocytogenes, 525 P value, in epidemiology, 579–580

1104

Manila Typesetting Company

Packaging aseptic, 745 beer, 908 carbon dioxide in, 193, 747 high-pressure treatment after, 754–758 modified-atmosphere, see Modifiedatmosphere packaging muscle foods, 125 poultry, 139 refrigerated cereal products, 216–217 Padec, 866 Paecilomyces, 205, 775, 935 Paemibacillus, spores, 45 Paemibacillus lactis, 179 Paemibacillus macerans, 742 Paemibacillus polymyxia, 742 pag genes, Salmonella, 251 Pale, soft, exudative (PSE) muscle, 126–127, 862 Pantoea, 214 PapR protein, Bacillus cereus, 497–498 Parabens, 777–778 Paragonimus hueitungensis, 701 Paragonimus szechuanensis, 701 Paragonimus westermani, 698, 701 Parasites, see also Helminth(s) protozoa, 713–733 Parechovirus, 620 Partnership for Food Safety Education, 507 Parvoviruses, 623 Pasteur, Louis, 63 Pasteurella, in seafood, 153 Pasteurization beer, 908 definition, 758 description, 743 eggs, 234 Listeria monocytogenes growth after, 507–509, 512 milk, 171–173, 290–291, 507–509, 512–513 nuts, 176 Pastyrma, Trichinella in, 677 Pathogen Modeling Program, 1013 Pathogenicity islands Salmonella, 247 staphylococcal, 552 Yersinia, 348, 353–355 Pathogens, in gastrointestinal tract, 951 Patis, 865 Patulin, 934 PCR (polymerase chain reaction) Asian taeniasis, 689 Bacillus cereus toxins, 498–499 Campylobacter, 276 Clostridium botulinum, 457 Clostridium perfringens toxins, 470 Cronobacter, 313–315, 322–323 Cryptosporidium, 717 Cyclospora, 719 enteric viruses, 637–639 Giardia, 725 hepatitis A virus, 637–639 lactic acid bacteria, 928, 930 Listeria monocytogenes, 505 noroviruses, 630, 637–639 probiotics, 958 Salmonella, 228–229 Shigella, 379

11/08/2012 07:24AM

Index

1105

staphylococcal enterotoxins, 564 subtyping, 1062–1063 Taenia, 684–685, 688 viable but nonculturable cells, 10 Vibrio, 403 Vibrio vulnificus, 422 wine yeasts, 919 Peanuts aflatoxins in, 598–603 contamination, 210 microflora, 205 organisms in, 207 Salmonella in, 209, 580 seed coats, 204 PEB1 protein, Campylobacter, 271 Pecans hydrolysis, 847–848 organisms in, 207 processing, 212–213 Pectinases coffee fermentation, 895 fruit and vegetable spoilage, 192 Pectinatus, 910 Pectinesterase, in vegetable fermentation, 847–848 Pectinolytic enzymes, in cocoa fermentation, 887 Pectins, grape, 925 Pectobacterium carotovora, 188, 190 Pediocins, 805, 809–810, 874 food applications, 807 genomics, 808–810 Listeria monocytogenes inhibition, 510 for meat products, 132 mechanisms of action, 810–811 Pediococcus antimicrobial action on, 774, 781, 783, 874–875 as beer contaminant, 909–910 catabolism in, 5 in meat fermentation, 871, 876 in milk, 177 plasmids, 874 in vegetable fermentation, 844 in winemaking, 928–929, 931 Pediococcus acidilactici, 807, 860, 867–876, 885 Pediococcus cerevisiae antimicrobial action on, 773 bacteriocin action on, 808 in cocoa fermentation, 885 in vegetable fermentation, 842, 844 Pediococcus damnosus, 870, 928, 931 Pediococcus parvulus, 870, 928 Pediococcus pentosaceus genomics, 848–850 in meat fermentation, 860, 867–876 in vegetable fermentation, 849–850 in winemaking, 928 Penicillin(s), 20–21, 28 Penicillin-binding proteins, 20 Penicillium antimicrobial action on, 768, 770, 772, 775, 781 as beer contaminant, 909 in cheese, 180–181 in fruits and vegetables, 188, 190–193, 194 in grains, 213

SMP_Food Microbiology_Index.indd

in meat fermentation, 860 in nuts, 205 in wine spoilage, 934–935 Penicillium camemberti antimicrobial action on, 784 in meat fermentation, 870 in milk fermentation, 827 Penicillium candidum, 870 Penicillium chrysogenum, 870 Penicillium commune, 774 Penicillium digitatum, 194, 195 Penicillium expansum antimicrobial action on, 770 in fruits and vegetables, 192, 194–196 water activity requirements, 750 Penicillium nalgiovense, 869–870 Penicillium nordicum, 603, 605 Penicillium patulum, 750, 770 Penicillium roqueforti antimicrobial action on, 784 in meat fermentation, 870 in milk fermentation, 827 Pentasomids, 706 Pentose, catabolism, 4–5 Pepsin test, Trichinella, 681 Pepsins, in staphylococcal enterotoxin degradation, 561 Peptidases in fermented dairy products, 832–833 lactic acid bacteria, 831–833 Peptidoglycans, in sporulation, 52 Peptococcaceae, in normal microflora, 952 Peracetic acid, in poultry processing, 138 Performance criteria, 83 Performance objectives, 83, 1034 Perfringens Predictor, 1013 Periorbital edema, in trichinellosis, 678–679 Periplasmic binding protein, Campylobacter, 271 Peroxidase, 195 Peroxyacetic acid for enteric viruses, 633 for fruits and vegetables, 195, 633 for muscle food decontamination, 123 Personal hygiene, 587 dairy products, 707 enteric viruses and, 625–626, 634 Pesiticin, 354 Pesticides, in wine, 922 Pezicula alba, 193 Pezicula malicorticis, 196 PFAM database, 980 pH, see also Acid(s) bacterial growth and, 14–15 antimicrobial action, 766 Campylobacter, 265 Clostridium perfringens, 467 Escherichia coli, 14 intracellular pH, 14–15 Listeria monocytogenes, 14, 506 models for, 1000 organic acid antimicrobials, 766–773 Salmonella, 229–231 Shigella, 379, 381 Staphylococcus aureus, 556 in sulfite antimicrobials, 778 Vibrio, 405–406 Vibrio cholerae, 410 Yersinia, 344

1105

Manila Typesetting Company

brewing, 906 cocoa fermentation, 889–891 coffee fermentation, 895 dairy products, 172 fruits, 189 grape juice, 922 heat resistance and, 743 high-pressure processing, 755 meat fermentation, 858–859, 862, 864, 871–872 milk, 172 muscle foods, 126 starter cultures for meat, 867–868 vegetables, 189 viral growth and, 631 winemaking, 929 pH gradient, in metabolism, 6 Phaeneropsolos bonnei, 698, 705 Phages, see Bacteriophages Phagocytosis Cronobacter, 329 Listeria monocytogenes, 524–525 Salmonella, 250 transmission, 380 Yersinia enterocolitica, 360–361 Pharyngitis, Yersinia enterocolitica, 345 Phenolic compounds, 784–785 in meat fermentation, 863 spore resistance to, 57–58 Phenotyping, 1060 Phenylalanine, for ochratoxin toxicity, 685 Phenylalanine ammonia lyase, 195 Phenylethylamine, in seafood spoilage, 142–143 Pheromones, bacteriocins, 810 Phialophora, 205 Phialophora malorum, 190 Philometra, 698, 702 Phoma, 190 Phoma exigua, 194 Phoma glomerata, 180 Phomopsis, 205 Phomopsis curcurbitae, 192 Phomopsis viticola, 934 phoP/phoQ region, Salmonella, 251 PhoPQ, in signal transduction, 14 Phosphatidylinositol phospholipase C, Listeria monocytogenes, 526–527 Phosphofructokinase, in glycolysis, 4 3-Phosphoglyceraldehyde, in glycolysis, 5 3-Phosphoglycerate mutase, in sporulation, 54 3-Phosphoglyceric acid, in spores, 54 Phospholipase Listeria monocytogenes, 526–527 Yersinia enterocolitica, 353 Phosphorylation in glycolysis, 3–6 in sporulation, 49–50 Phosphotransferases, in vegetable fermentation, 850 Photobacterium bioluminescence, 12 in muscle foods, 115, 145 Photobacterium histaminum, 145 Photobacterium phosphoreum, 116, 118, 145 Photoproducts, spore, 56–57 Physical hazards, 119, 1045

11/08/2012 07:24AM

Index

1106 Physicochemical hurdles, meat processing, 122 Phytoalexins, 193, 195, 784 Phytoanticipins, 192–193 Phytophora infestans, 193 Pichia antimicrobial action on, 770, 781 as beer contaminant, 909 in cocoa fermentation, 885 in dairy products, 180 in vegetable fermentation, 846 in winemaking, 920, 923, 927 Pichia anomala, 769 Pichia kudriavzevii, 887 Pichia membranaefaciens, 770, 885, 927 Picking, poultry feathers, 135 Pickles, see Vegetables, fermented Picornaviridae, 620, 622 Picrotoxin, intentional contamination with, 95 Pig-Bel (necrotic enteritis), 465 Pillsbury Company, HACCP, 1040 Pilus assembly, Salmonella, 249 Pinene, 781 Piscibacillus, 866 Pistachios, organisms in, 208 “Pitching yeast,” 905 pKa (dissociation constant), in antimicrobial action, 766–773 Plague (Yersinia pestis), 339–341, 352, 354 Plankton, Vibrio in, 401 Plant products, see Fruit(s); Grains; Vegetables Plantaricins, 805, 811, 812 Plasmid(s) in antimicrobial resistance, 24 Campylobacter, 271–272 Clostridium perfringens, 467, 473–474 Cronobacter, 314–315 Escherichia coli, 301–302 lactic acid bacteria, 836, 874 Shigella, 386 Staphylococcus aureus, 552 virulence, see Virulence plasmids Yersinia, 348 Plasmid profile analysis, 1060–1061 Plasmin, in cheese flavor, 833 Plasmopara viticola, 934 “Plastic-like” taint, in wine, 925 PlcR protein, Bacillus cereus, 497–498 Pleistophora, 713, 723 Plesiomonas, 150 Plesiomonas shigelloides in muscle foods, 119 in seafood, 141 Pneumococcus, 526 Poisoning, intentional, see Intentional contamination Polarity, antimicrobial action and, 766 Poliovirus, 620, 622 Polyacrylamide gel electrophoresis, for proteomics, 988 Polygalacturonase, 192, 847–848 Polymerase chain reaction, see PCR (polymerase chain reaction) Polynomial models, 1000, 1002 Polypeptides, in fish fermentation, 865–566 Polyphenol oxidases, in cocoa fermentation, 891–892

SMP_Food Microbiology_Index.indd

Polyphenols, 785 Polysaccharide capsule Vibrio cholerae, 415 Vibrio vulnificus, 424–425 Polyvinyl packaging, muscle foods, 125, 132–133 Population surveillance, 585–586 Porins, Salmonella, 250 Pork tapeworm (Taenia solium), 686–689 Potassium bisulfite, 778 Potassium lactate, 132, 769–770 Potassium nitrite, 776–777 Potassium propionate, 770 Potassium sorbate, 34 antimicrobial action, 771 for cereal products, 215 for vegetable fermentation, 843 for winemaking, 919 Potassium sulfite, 778 Potentiometry, for muscle food spoilage, 118 Poultry, 134–140 bacteriophages for, 815 contamination, 134 decontamination, 123–124, 136–138 fermented, manufacture, 862–863 initial microflora, 111–115 organisms in, 119–120 Acinetobacter, 134 Aeromonas, 139 Brochothrix thermosphacta, 138–139 Campylobacter, 136, 138, 140, 273, 276–278 Candida, 138 Clostridium perfringens, 469 Debaryomyces, 138 Enterobacteriaceae, 134, 137, 139 Escherichia coli, 136, 138 Flavobacterium, 134 Listeria monocytogenes, 137–138, 509–510 Moraxella, 134 Pseudomonas, 134, 138–139 Pseudomonas fluorescens, 138 Salmonella, 137–140, 234, 239, 245 Salmonella enterica, 136–138 Shewanella putrefaciens, 139 packaging, 139 processing, 134–136, 277–278 safety, 139–140 spoilage, 138–140 compounds formed, 116–118 quality evaluation, 118 substrate utilization for, 116–118 types, 116 staphylococcal endotoxins in, 559 Pourriture noble, 934–935 Pra-hoc, 866 Prawns, see Shrimp Prebiotics, 965–967 Predictive models, 997–1021 compartment-based, 1011–1012 deterministic, 998–1001 errors in, 1003 extended secondary, 1003–1004 future, 1003–1014 growth, 998–999 growth/no growth, 1002–1003 inactivation, 999–1001 lag phase, 1001

1106

Manila Typesetting Company

levels of, 1007–1008 mesoscopic, 1007–1012 metabolic network, 1007–1010 microbial interaction, 1006–1007 microscopic, 1007–1012 past, 998–1003 present, 1003–1014 primary, 998–999 probabilistic, 1001–1003 secondary, 999–1001, 1003–1004 software tools for, 1012–1014 structural food systems, 1004–1006 Pregnancy listeriosis in, 514–516, 518–520 toxoplasmosis in, 722 Preharvest pathogen control, in muscle foods processing, 131 Prepared foods, Clostridium botulinum in, 449–450 Prerequisite programs, 87 Preservation, see Food preservatives and preservation methods; specific methods and organisms Pressure effects bovine spongiform encephalopathy agent, 654 spores, 56 Vibrio, 405 Prestin broth, Campylobacter, 276 prf genes, Listeria monocytogenes, 528–529 PrfA protein, in bile tolerance, 992 prg genes, Salmonella, 251 Principal-component analysis partial least squares models, 118 Principles and Guidelines for the Conduct of Microbiological Risk Management, 82, 85 Principles for the Establishment and Application of Microbiological Criteria for Foods, 81 Prion diseases, 130, 651; see also Bovine spongiform encephalopathy pro genes, Salmonella, 232 Pro transport systems, in osmoregulation, 15 Proanthocyanidins, 785 Probabilistic models, 1001–1003 Probiotics for animals, 293, 954–956 antimicrobial resistance, 34–35 concept, 949 cultures, 961–964 definition, 953 delivery vehicles for, 957 in fermented meat products, 118–119 gastrointestinal ecology, 960–961 genomics, 962–964 health benefits, 955–957 in meat fermentation, 869 mechanisms of action, 953–955 molecular studies, 958–960 normal microflora interactions with, 949–953 quorum sensing and, 12 safety, 965 spores as, 73 taxonomy, 957–959 therapeutic, 957–958 in vaccines, 957

11/08/2012 07:24AM

Index

1107

Process criterion, 82 Produce, see Fruit(s); Vegetables Product criterion, 82 Product innovation, models for, 1012 Proglottids Taenia saginata, 682–683 Taenia solium, 686 Proline, as osmolyte, 992 Proline transport in osmoregulation, 15 Staphylococcus aureus, 558 Propionates, antimicrobial action, 770 Propionibacterium, 177–178 Propionibacterium cyclohexanicum, 771 Propionibacterium freudenreichii subsp. shermanii, 770, 803, 826 Propionic acid, 770, 803 Propionicin, 809 Propyl gallate, 784 Propyl paraben, 777–778 Propylene oxide, nut treatment, 212–213 Prosite software, 980 Prostaglandins in cholera pathogenesis, 412 in staphylococcal food poisoning, 566 Prosthodendrium molenkampi, 698, 705 Proteases in cocoa fermentation, 891 in fish fermentation, 865 in milk, 174–175 staphylococcal enterotoxins resistance to, 561 Protection, from intentional contamination, 97–103 Protein(s) DNA interactions with, 989–990 fruits, 189 milk, 170 vegetables, 189 Protein expression profiling, 988 Protein kinases, in sporulation, 49–50 Proteolysis in fish fermentation, 865–866 lactic acid bacteria, 831–833 in muscle foods, 117, 143 Proteomics, 988, 990–993 Proteus in muscle foods, 117, 145–147 in nuts, 205 Proteus mirabilis, 145–147, 955 Proteus morganii, 144–147 Proteus vulgaris, 145, 770 Proton gradient, 6–7 Proton motive force, 6, 767, 810 Protozoa, 713–733 Providencia in muscle foods, 145 in seafood, 155 PrP protein, 652, 659, 664–665; see also Bovine spongiform encephalopathy, agent PrtM (lipoprotein maturase), 832 PrtP (proteinase), 832 PSE (pale, soft, exudative) muscle, 126–127, 862 Pseudoappendicular syndrome in Campylobacter infections, 273 in Yersinia infections, 345 Pseudomonadales, 37

SMP_Food Microbiology_Index.indd

Pseudomonas antimicrobial action on, 771–772, 781, 786 antimicrobial resistance, 30, 34 bioluminescence, 12 in fruits and vegetables, 194 in meat, 859, 863 metabolism in, 5 in milk, 173–176 in modified-atmosphere packaging, 747 in muscle foods, 114–117, 126, 127, 145 in nuts, 205 osmoregulation, 15 in poultry, 134, 138–139 in seafood, 149, 153–154 in vegetable fermentation, 842 water activity requirements, 750 Pseudomonas aeruginosa, antimicrobial action on, 20, 768, 770, 772, 773, 775–776 Pseudomonas cepacia, 22, 196 Pseudomonas fluorescens antimicrobial action on, 774, 775 in fruits and vegetables, 191 in milk, 173–176 in muscle foods, 116, 145 in poultry, 138 Pseudomonas fragi in milk, 174–176 in muscle foods, 116, 117 Pseudomonas lundensis in milk, 173 in muscle foods, 116 Pseudomonas putida, 753 in milk, 173 in muscle foods, 116, 145 Pseudomonas putrefaciens, 145 Pseudomonas syringae, 196 Pseudoterranova, 698–700 Psoralen, 784 Psychiatric disorders, bovine spongiform encephalopathy, 660 Psychrophiles, 16 Psychrotrophs, 16 in meat, 127 in milk, 170, 173–176 refrigeration, 746–747 Public health significance, of bioterrorism, see Intentional contamination Pullularia, 770 Pulsed electric field processing, 757–758 Pulsed technologies, meat processing, 124 Pulsed-field gel electrophoresis, 1061–1062, 1067, 1069–1070, 1072 Campylobacter, 267–268 Clostridium perfringens, 473 Cronobacter, 315 versus PCR, 229 for surveillance, 584 Vibrio, 403 PulseNet (National Molecular Subtyping Network), 267, 505, 584, 1059, 1070–1072 Purac Company, models developed by, 1012, 1014 Putrefaction, in muscle food spoilage, 139 Putrescine, in muscle food spoilage, 117, 120, 142, 144–145 Pyridine-2,6-dicarboxylic acid, in spores, 46, 48, 56–60

1107

Manila Typesetting Company

Pyrogenic toxins, 551–553; see also Staphylococcal enterotoxins Pyrosequencing, 976–978 Pyrrolnitrin, 196 Pyruvate in glycolysis, 5–6 in meat fermentation, 872 Pyruvate-ferredoxin oxidoreductase, 776–777 pYV plasmid, Yersinia enterocolitica, 348

Q

Quaking-induced conversion test, bovine spongiform encephalopathy agent, 665 Qualitative risk assessment, 1027–1028 Quality assurance indicators for, see Indicator(s) microbiological criteria for, see Microbiological criteria in spore control, 70 Quality evaluation, muscle food spoilage, 118 Quantitative risk assessment, 1028–1030 Quaternary ammonium compounds, resistance to, 32–33 QuIC test, bovine spongiform encephalopathy agent, 665 Quiescence, in fruit and vegetable spoilage, 191 Quinine complexes, in cocoa fermentation, 892 Quinolones, 20, 245 Quinupristin-dalfopristin, resistance to, 31 Quorum sensing, 11–12 in antimicrobial resistance, 33 bacteriocin production, 810 criteria for, 13 spoilage and, 12

R

Radappertization, 754 Radiation, heat transfer in, 739–740 Radiation treatment, 752–754 bovine spongiform encephalopathy agent, 654 Campylobacter, 265 Clostridium botulinum, 69–70, 451 dose requirements, 754 eggs, 234 environmental conditions for, 754 Escherichia coli, 291 fruit, 194 growth kinetics, 8 hepatitis A virus, 633 ionizing, 752–754 killing doses, 753 Listeria monocytogenes, 511 meat, 124–125 microbiological fundamentals, 753–754 resistance to, 754 Salmonella, 235 spores, 56–57 technological fundamentals, 754 undesirable effects, 754 vegetables, 194 Vibrio, 405 Yersinia enterocolitica, 344 Radicidation, 753

11/08/2012 07:24AM

Index

1108 Radiofrequency heat treatment, 745 Radioimmunoassays, staphylococcal enterotoxins, 565 Radurization, 753 Raffinose, as prebiotic, 966 Rahnella, 115 Raman spectroscopy, for muscle food spoilage, 118 Rancidity dairy products, 175–176 fermented meat products, 862 high-oxygen packaging, 133 meat fermentation, 874 milk, 170, 175–176 nuts, 205 Random amplified polymorphic DNA analysis, 229, 960 Rapid microbiological methods, see also specific methods DNA microarray, see DNA microarray analysis enzyme-linked immunosorbent assay, 680 nucleic acid-based, see PCR Raw materials, microbiological criteria, 86–87 RBSfinder software, 979 Reactive arthritis Salmonella, 244 Shigella, 384 Yersinia, 362–364 Reactive oxygen species, in fruits and vegetables, 193 Ready-to-eat foods, organisms in enteric viruses, 625–626, 634–635 enteroviruses, 623–625 Listeria monocytogenes, 507–510 Real-time PCR, Campylobacter, 276 RecA protein, in DNA repair, 60 Recalls, for beef contamination, 131 Receptors, alterations, in antimicrobial resistance, 22 Recognizability, in CARVER+Shock strategy, 98–100 Recombination Campylobacter, 267 lactic acid bacteria, 836, 875 techniques, 984 Recommended International Code of Hygiene Practice for Foods for Infants and Children, 312 Recontamination, muscle foods, 116 Recordkeeping, HACCP system, 1052–1053 Recuperability, in CARVER+Shock strategy, 98–100 “Refermentation,” 11 Refrigeration, 745–747 microorganism growth in, 16 clostridia, 127 Clostridium botulinum, 66, 70, 125, 453 Clostridium perfringens, 67, 466 Listeria monocytogenes, 506 Vibrio, 404–405 milk, 173 unreliability, 804 Refrigeration Index Calculator, 1014 Regulations, see Legislation and regulations Reiter’s syndrome after Salmonella infection, 244 after Shigella infection, 384

SMP_Food Microbiology_Index.indd

Relative humidity, fruit spoilage, 193–194 Relaying, in enteric virus removal, 632 Reoviridae, 620 Repair, after injury, 10 Repeats in toxin (RtxA) toxin, Vibrio cholerae, 414 Repetitive extragenic palindromic elements, in PCR, 229, 1062 Resequencing, in genomics, 979 Reservoirs, for organisms, see also specific organisms Animals as, see Animals, as reservoirs in epidemiologic studies, 580 Residential yeast, in winemaking, 920 Resins, in hops, 904 Resistance antimicrobial, see Antimicrobial resistance to environmental conditions, see specific condition, e.g., Heat susceptibility and resistance Resistance integrons, 24 Resorcinol compounds, in spoilage resistance, 192 Respiration fruits and vegetables, 189 microbial, 5–6 Response regulator, in signal transduction, 13 Response surface models, 999–1001 Restriction digestion, in PCR, 1062 Restriction endonucleases, in milk fermentation, 835 Restriction fragment length polymorphism analysis, 1061–1062 Asian taeniasis, 689 Campylobacter, 268 cholera toxin, 408 Taenia saginata, 684–685 Restriction/modification systems, Lactococcus, 835 Resuscitation, viable but nonculturable cells, 11 Retail Food Program Database of Foodborne Illness Risk Factors, 587–588 Reverse genetics, 983–984 Reverse micelles, 786 Reverse osmosis, for beer, 912 Reverse-transcription PCR, 637–639 Rheumatic disorders, see Autoimmune sequelae Rhizocotonia carotae, 190 Rhizocotonia solani, 190 Rhizopus antimicrobial action on, 768, 780 in fruits and vegetables, 188 in grains, 213 in nuts, 205 water activity requirements, 750 Rhizopus stolonifer, 191, 193–194 Rhodococcus, 155 Rhodotorula, 770, 773, 781, 782, 885, 919–920, 936 Rhodotorula mucilaginosa, 180 Rice, fermented, 866 Ricin, intentional contamination with, 95 Rickettsia conorii, 528 Riddling, in winemaking, 935 Ripening, delaying, 193–196 Risk, of antimicrobial resistance, communication of, 34–35

1108

Manila Typesetting Company

Risk analysis, see also Risk assessment components, 1023–1024 importance, 1023 management, 97–98, 1033–1034 principles, 1023–1025 purpose, 1023 Risk assessment, 1025–1037 for biosecurity, 97–98 CARVER+Shock strategy in, 98–100 communication in, 1024–1025 definition, 1024 description, 1023–1025 economic aspects, 1024 exposure, 1025 food safety objective paradigm, 1034 hazard characterization, 1025–1026 hazard identification, 1025 human enteric viruses, 639 methods, 1027–1030 models for, 1029–1030 operational risk management in, 97–98 qualitative, 1027–1028 quantitative, 1028–1030 risk characterization, 1026–1027 risk management, 1033–1034 sensitivity analysis, 1031–1033 steps, 1025–1027 uncertainty in, 1030–1032 variation in, 1030–1032 Risk characterization, 1026–1027 Risk management, 97–98, 1033–1034 Risk management metrics, 82–83 Risk Ranger software, 1014 RNA, see also tRNA intergenic spacer PCR spores, 54 synthesis, in spore germination, 62 RNA analysis, 987–988 in genomics, 979 probiotics, 952, 960–961 RNA viruses, see also specific viruses PCR, 637–639 RNAIII, staphylococcal, 555–556 Roasting, coffee beans, 896 Roche NimbleGen platform, 985–987 Roche sequencing equipment, 976 Rodent(s), grain contamination by, 213 Rodent lungworm (Angiostrongylus cantonensis), 704 Rodent tapeworms, 705 Rope spoilage, bakery products, 216 Ropiness, in bakery products, 72 Ropy defect, milk, 170, 177 Rosemary, 125, 781 Ross model, 1003 Rosso model, 999 Rot protein, Staphylococcus aureus, 554, 556 Rotaviruses, 620, 622 Rotting, see Fruit(s), spoilage; Vegetables, spoilage Roundworms, see specific species rpo genes in acid tolerance response, 233 bacteriocins, 811 Rsb proteins, Staphylococcus aureus, 556 RsfA protein, in sporulation, 51 RT Freshmeter, for muscle food quality evaluation, 118 RtxA toxin, Vibrio cholerae, 414 Rugose form, Vibrio cholerae, 409

11/08/2012 07:24AM

Index

1109

S

Saccharibacter, 932 Saccharomyces, 779 antimicrobial action on, 768, 770, 772–773, 781, 782, 786 as beer contaminant, 909 in brewing, 905 in cocoa fermentation, 885 in dairy products, 180 in vegetable fermentation, 846 in winemaking, 917–928 Saccharomyces bayanus, 905, 918, 920–922, 936 Saccharomyces carlsbergensis, 905 Saccharomyces cerevisiae antimicrobial action on, 770, 784 in brewing, 905–906, 911 in cocoa fermentation, 885, 887 injury, 9 osmoregulation, 15 quorum sensing and, 12 in winemaking, 918–928, 932, 935–936 Saccharomyces chevalieri, 885, 887 Saccharomyces diastaticus, 905, 911–912 Saccharomyces paradoxus, 921 Saccharomyces uvarum, 905, 921 Saccharomycodes ludwigii, 927 Saccharomycopsis, 885 sae locus, Staphylococcus aureus, 554–555 Safety, see also Hazard Analysis and Critical Control Point system challenges to, 120 Clostridium botulinum handling, 457–458 indicators, see Indicator(s) from intentional contamination, 97–103 CARVER+Shock strategy in, 98–100 interventions for, 100–103 operational risk management in, 97–98 in meat processing, 128–130 microbiological criteria for, see Microbiological criteria in poultry processing, 139–140 probiotics, 964–965 promotion, 119 risk analysis in, see Risk analysis; Risk assessment surveillance in, 581–589 systems for, evaluation, 588–589 traceabiliity for, 121–122 Sage, 781 Sakacins, 805, 809–810, 874 Salicylic acid, 784 Salinomycin, 21 Salinvibrio, 866 Salm-Net, 1071 Salmon poisoning, 702 Salmonella, 225–261 acid tolerance response, 233–234 antigens, 227–228 antimicrobial action on, 20, 767, 768, 770–772, 776, 778–784 antimicrobial resistance, 25, 26–27, 31, 35, 244–245 antisera, 228 in aquaculture, 26 bacteriophage effects, 814–815 biochemical identification, 227–229 as biological hazard, 1044–1045 bioluminescence, 12

SMP_Food Microbiology_Index.indd

colonization, 246 culture, 226, 228 in dairy products, 236–237, 578 desiccation resistance, 231–232 disease, 243–246 autoimmune sequelae, 243–244 epidemiology, 586, 588 outbreaks, 236–243, 583–584 surveillance, 577 symptoms, 243–244 treatment, 243–244 in eggs, 234, 242–243, 585–586 in fermented meat products, 862 in fruits and vegetables, 191 in grains, 215–216 growth, 229–231 hazard analysis, 1047, 1049 heat resistance, 33–34 historical considerations, 225 incubation period, 246 as indicator, 130 infectious dose, 246 injury, 9 iron acquisition, 251 lipopolysaccharide, 244, 250–251 in milk, 236–237, 743 multidrug resistant, 244–245 in muscle foods, 114, 119, 120, 123, 126, 131 nomenclature, 226–228 nonspecific human response, 246–247 in nuts, 204–209, 205–212 osmoregulation, 15 pathogenicity, 246–251 pH effects on, 229–231 physiology, 229–234 porins, 250 in poultry, 137–140 probiotic protection from, 955 reservoirs, 234–236 resistance to, 172 risk assessment, 1026, 1033, 1034 salinity effects on, 230–231 in seafood, 140, 141, 148–149, 151, 154 serologic testing, 228 serovars, 227 siderophores, 251 species, 227 specific human response, 246–247 subgenera, 226 subspecies, 226–227 subtyping, 1060–1061, 1067–1069, 1071 surveillance for, 584–585 survival, in foods, 231 taxonomy, 225–229 temperature effects on, 229–234 toxins, 250 Vi antigen, 250–251 viable but nonculturable, 10 virulence factors, 246–251 virulence plasmids, 247–250 water activity requirements, 750 Salmonella bongori, 226–227, 247 Salmonella enterica, 226–227 antimicrobial resistance, 29, 172 genomics, 247 Salmonella enterica serovar Abortusovis, 247 Salmonella enterica serovar Agona, 239–240

1109

Manila Typesetting Company

Salmonella enterica serovar Anatum, 238, 246 Salmonella enterica serovar Baildon, 238 Salmonella enterica serovar Bareilly, 239, 243 Salmonella enterica serovar Barenderup, 240 Salmonella enterica serovar Barranquilla, 240 Salmonella enterica serovar Berta, 238 Salmonella enterica serovar Blockley, 240–241 Salmonella enterica serovar Bovismorbificans, 238–240 Salmonella enterica serovar Braenderup, 238 Salmonella enterica serovar Champaign, 240–241 Salmonella enterica serovar Chester, 240–241 Salmonella enterica serovar Choleraesuis, 226, 244, 247 Salmonella enterica serovar Cubana, 246 Salmonella enterica serovar Derby, 237 Salmonella enterica serovar Dublin, 244, 247–248 disease, outbreaks, 236, 237 Vi antigen, 250 Salmonella enterica serovar Eastbourne, 242, 246 Salmonella enterica serovar Enteritidis, 238–243 antimicrobial action on, 771–772, 775 antimicrobial resistance, 33–34 disease, 238–243 outbreaks, 236, 237, 579, 581, 584–585 surveillance, 577–578, 586–588 heat treatment, 748 models for, 1011 pH effects, 33–34 in poultry, 138 reservoirs, 234–235 subtyping, 1067–1068, 1072 unintentional contamination with, 92 virulence factors, 247 Salmonella enterica serovar Gallinarum, 225, 247 Salmonella enterica serovar Gaminara, 238, 1067 Salmonella enterica serovar Give, 239 Salmonella enterica serovar Goldcoast, 239–240 Salmonella enterica serovar Hadar, 239, 240 Salmonella enterica serovar Hartford, 238, 1067 Salmonella enterica serovar Heidelberg, 239–241, 246, 585 Salmonella enterica serovar Indiana, 238 Salmonella enterica serovar Infantis, 228 Salmonella enterica serovar Java, 236, 237 Salmonella enterica serovar Javiana, 238, 240, 242, 246 Salmonella enterica serovar Kedougou, 237 Salmonella enterica serovar Kottbus, 238 Salmonella enterica serovar Litchfield, 238 Salmonella enterica serovar Livingstone, 240–241 Salmonella enterica serovar Meleagridis, 238, 246 Salmonella enterica serovar Montevideo, 229, 236–238, 242–243

11/08/2012 07:24AM

Index

1110 Salmonella enterica serovar Muenchen, 238, 1061, 1067 Salmonella enterica serovar Napoli, 243, 246 Salmonella enterica serovar Newport, 26, 27, 238, 246, 768 antimicrobial resistance, 120, 244–245 outbreaks, 236, 241 in seafood, 151 Salmonella enterica serovar Oranienburg, 236, 237, 240–243 Salmonella enterica serovar Paratyphi, 91, 238, 240, 250 Salmonella enterica serovar Poona, 238 Salmonella enterica serovar Pullorum, 225, 247 Salmonella enterica serovar Rissen, 242–243 Salmonella enterica serovar Rubislaw, 240, 242, 246 Salmonella enterica serovar Saintpaul, 238, 242, 246, 579, 585 Salmonella enterica serovar Senftenberg, 757 Salmonella enterica serovar Stanley, 242 Salmonella enterica serovar Strathcona, 239 Salmonella enterica serovar Tennessee, 232, 242 Salmonella enterica serovar Thompson, 238–242 Salmonella enterica serovar Typhi acid tolerance response, 14 disease, 240, 243 genomics, 979 pathogenicity islands, 247 Vi antigen, 250 Salmonella enterica serovar Typhimurium acid tolerance response, 16, 233–234, 768–769 antimicrobial action on, 769–770, 775–776, 783 antimicrobial resistance, 26, 120, 244–245 in biofilms, 13 disease, outbreaks, 236–242, 584 in dry foods, 231–232 genomics, 247 growth, 14, 230 infectious dose, 246 models, 1005–1006 in nuts, 209 pH effects, 14 in poultry, 136–137 quorum sensing, 12 radiation effects, 124–125 in seafood, 154 subtyping, 1060–1061, 1067–1068, 1072 survival, 231–233 taxonomy, 227 type III secretion systems, 247 unintentional contamination with, 92 virulence factors, 247–250 Salmonella enterica serovar Uganda, 239 Salmonella enterica serovar Weltevreden, 240 Salmonella enterica serovar Zanzibar, 246 Salmonella enterica subsp. arizoniae, 226–227 Salmonella enterica subsp. diarizoniae, 227 Salmonella enterica subsp. houtenae, 227 Salmonella enterica subsp. indica, 227 Salmonella enterica subsp. salmae, 230

SMP_Food Microbiology_Index.indd

Salmonella outbreak detection algorithm (SODA), 584 Salt bacterial growth and Clostridium botulinum, 453 Clostridium perfringens, 467 Listeria monocytogenes, 506 Salmonella, 230–231 Shigella, 379, 381 Staphylococcus aureus, 557–558 Yersinia enterocolitica, 344 in butter, 172 in fermented fish products, 865–866 in fermented meat products, 863, 865 in fermented vegetable products, 842–848 Sampling plans, 83–86, 601 Sanger technique, 976 Sanitizers for biofilms, 114–115 resistance to, 32 Sapovirus, 621 SarA protein, Staphylococcus aureus, 554–556 Sarcina lutea, 770 Sarcocystis, 119 Sarcocystis hominis, 713, 723 Sarcocystis subhominis, 713, 723 SASPs (small, acid-soluble proteins), in spores, 46, 53–54, 56–58, 62 Satay, Trichinella in, 678 Sauces, fish, 865–866 Sauerkraut, 843–848 Sausage, 857–865 categories, 857–858 composition, 858–859 factors affecting quality, 859, 862–865 fermented, 118–119 manufacture, 859–861 processes for, 859 starter cultures, 866–876 Sausage “botulus” poisoning, 442 Savory, antimicrobial compounds in, 782 Saxitoxin, intentional contamination with, 95 Scalding, in poultry processing, 135 Schizosaccharomyces, 885 antimicrobial action on, 773 in winemaking, 926 Schizosaccharomyces malidevorans, 885 Schizosaccharomyces pombe, 781, 932 Schoolfield model, 1000 Schvartzman model, 1005 Schwanniomyces occidentalis, 912 Sclerotinia, 190 Sclerotinia sclerotiorum, 190–192, 194 Sclerotium rolfsii, 192 Scolices, Taenia saginata, 682–686 Scombroid poisoning, 120, 139, 143–144, 148 Scrapie, 130, 651–652, 656–657 Scrapie-associated fibrils, 664 SEA enterotoxins, 549, 551–556, 560–563, 565 Seafood, 140–141; see also Fish; Shellfish; Shrimp contamination, 140 farming of, see Aquaculture fermented, 865–866

1110

Manila Typesetting Company

initial microflora, 111–115 organisms in, 119–120 Achromobacter, 154 Acinetobacter, 154–155 Aerococcus, 155 Aeromonas, 141, 149–150, 153 Agrobacterium, 153 Alcaligenes, 155 Bacillus, 155 Bacillus cereus, 119 Brevibacterium, 155 Campylobacter, 140, 153, 270 Carnobacterium, 155 Clostridium botulinum, 140–141, 154–155 Corynebacterium, 155 Cryptosporidium, 155 Escherichia coli, 140, 154 Escherichia coli O157:H7, 141 Flavobacterium, 153, 155 helminths, 697–702 Kocuria, 155 Lactobacillus, 155 Listeria innocua, 114 Listeria monocytogenes, 140–141, 154, 510, 513–515 Micrococcus, 126 Moraxella, 154 Ochrobacterium, 153 Pasteurella, 153 Plesiomonas, 150 Plesiomonas shigelloides, 141 Providencia, 155 Pseudomonas, 149, 153–154 Rhodococcus, 155 Salmonella, 140–141, 148–149, 151, 154, 234–235, 240–242 Salmonella enterica, 151, 154 Staphylococcus, 149, 154 Staphylococcus aureus, 141, 154 Streptococcus faecalis, 154 Vibrio, 119, 140, 150, 152–153, 401–403 Vibrio alginolyticus, 150, 152, 429–430 Vibrio anguillarum, 153 Vibrio cholerae, 141, 148–150, 153, 409–410 Vibrio fluvialis, 426 Vibrio furnissii, 427 Vibrio hollisae, 428–429 Vibrio parahaemolyticus, 141, 148–154, 417 Vibrio vulnificus, 141, 148–153, 421–422 packaging, 125 spoilage compounds formed, 116–118 development, 115–116 quality evaluation, 118 substrate utilization for, 116–118 types, 116 Seafood Safety and Spoilage Predictor software, 1013 SEB enterotoxins, 549, 551–554, 556, 561–563, 565 SEC enterotoxins, 549, 551–554, 556, 560–563, 565–566 Secretor status, noroviruses, 628 Security, intentional contamination protection, 97–103

11/08/2012 07:24AM

Index

1111

SED enterotoxins, 549, 551–552, 560, 565 SEE enterotoxins, 549, 551–552 Seeded fermentation, 918 Seeds, see also Nuts organisms in, 208 SEF enterotoxins, 549 SEG enterotoxins, 550–551 SEH enterotoxins, 563 SEI enterotoxins, 550–551 SEIJ enterotoxin, 552 SEIK enterotoxins, 552 SEIL enterotoxins, 552 SEIM enterotoxins, 552 SEIN enterotoxins, 552 SEIO enterotoxins, 552 SEIP enterotoxins, 552 SEIQ enterotoxins, 552 SEIU enterotoxins, 552 Selective pressure, antimicrobial resistance, 35–37 Semduramicin, 21 Senescence, 189 Sensitivity analysis, in risk assessment, 1030–1032 Sensory analysis, for muscle food spoilage, 118 Sep proteins, Escherichia coli, 301 Septata, 713, 723–724 Septicemia Cronobacter, 322, 326–327, 330 Listeria monocytogenes, 518 Vibrio hollisae, 430 Vibrio vulnificus, 422–423 Septum, in sporulation, 48 Sequential hurdles, meat processing, 122–123 SER enterotoxin, 552 Serine protease, Escherichia coli, 301 Serotyping versus molecular subtyping, 1060 Salmonella, 228 Serratia in grains, 214 in muscle foods, 115, 117, 145 Serratia grimeseii, 783 antimicrobial action on, 776 in muscle foods, 115 Serratia liquefaciens antimicrobial action on, 774 in muscle foods, 116 SES enterotoxin, 552 Sesame seeds contamination, 205, 210 organisms in, 208, 237 SET enterotoxin, 552 set gene, Shigella, 390 Sewage contamination, see Fecal contamination SHARCGS software program, 979 Sharp objects, as hazards, 1045 Sheep liver fluke (Fasciola hepatica), 698, 702–703 Shelf life, see also Food preservatives and preservation methods Clostridium botulinum and, 454 fruits, 189 muscle foods, 122–125 vegetables, 189 water availability and, 748–752

SMP_Food Microbiology_Index.indd

Shelf Stability Predictor, 1014 Shellfish, 149–155 chitosan production from, 779–780 illness from, 140 organisms in enteric viruses, 623–625, 632–633, 638–639 helminths, 697–702, 704–705 hepatitis A virus, 632–633 Listeria monocytogenes, 510 Salmonella, 234–235, 240–242 Vibrio, 401–402 Vibrio alginolyticus, 429–430 Vibrio cholerae, 409–410 Vibrio hollisae, 428–429 she gene, Shigella, 390 Shewanella, bioluminescence, 12 Shewanella liquefaciens, 126 Shewanella oneidensis, 753 Shewanella putrefaciens antimicrobial action on, 776 in muscle foods, 115–118, 126 in poultry, 139 Shiga toxins characteristics, 302–304 Escherichia coli, 302–304 Escherichia coli O157:H7, 288, 302–304 disease, 303–304 genetics, 303 mode of action, 303 receptors for, 303 structures, 302–303 Shigella, 390 Shiga-toxin producing (STEC) Escherichia coli, 128–129, 290 Shigella, 377–399 actin filaments, 528 antimicrobial action on, 20, 774–775, 781, 782 antimicrobial resistance, 27–28, 384–385 characteristics, 377–380 chromosomal virulence loci, 390–391 classification, 377–378 contamination with, 380–383 cultures, 378–379 disease characteristics, 383 complications, 384 diagnosis, 378–379 versus enteroinvasive Escherichia coli infections, 289 epidemiology, 379–380, 588 history, 377 incubation period, 383 outbreaks, 381–383 prevention, 384–385 reservoirs, 380 vaccine, 385 ecology, 378–379 in foods, 380–383 genomics, 386–391 growth, 381, 386 infectious dose, 383–384 intentional contamination with, 91 in muscle foods, 119 pathogenicity islands, 390 plasmids, 387 reservoirs, 380 susceptible populations, 384

1111

Manila Typesetting Company

taxonomy, 377–378 temperature effects, 17 transmission, 380–381 viable but nonculturable cells, 10 virulence plasmids, 385–391 Shigella boydii classification, 377–378 epidemiology, 379 virulence factors, 387 Shigella dysenteriae characteristics, 377–378 disease, 382–385 ecology, 380 epidemiology, 379 in foods, 381 virulence factors, 390 Shigella flexneri characteristics, 377–378 disease characteristics, 384–385 outbreaks, 382–383 ecology, 380 epidemiology, 379–380 in foods, 381 reservoirs, 380 survival, 381 virulence factors, 385–391 Shigella sonnei characteristics, 377–378 disease characteristics, 384–385 outbreaks, 382–383, 584, 585 ecology, 381 epidemiology, 379 in foods, 380 survival, 381 virulence factors, 387 Shiokara, 866 Shock, in CARVER+Shock strategy, 98–100 Shottsuru, 865 SHRAP software program, 979 Shrimp, 148–149 Sidal, 866 Siderophores Salmonella, 251 Vibrio hollisae, 429 Vibrio vulnificus, 424 Yersinia, 353–354 Sig proteins Bacillus subtilis, 51 Clostridium perfringens, 476–477 Sigma factors in antimicrobial resistance, 33 Campylobacter, 265 Clostridium perfringens, 476 Escherichia coli, 290 in osmoregulation, 15 pH and, 14 in sporulation, 46, 49, 51 Staphylococcus aureus, 556 in stress, 992 Signal transduction, 11, 13 Simultaneous hurdles, meat processing, 122–123 Sin proteins, in sporulation, 50 Sinapic acid, 784 Single nucleotide polymorphisms, 979, 1066, 1069, 1071 Single-molecule sequencing, 979

11/08/2012 07:24AM

Index

1112 sip genes, Salmonella, 247 Skandamis model, 1006, 1007 Slaughter bovine spongiform encephalopathy control in, 663–664 Escherichia coli control in, 293 muscle foods, decontamination during, 123 poultry, 135 Trichinella control in, 680–682 Sle proteins, in spore cortex, 62 Slime, in muscle food spoilage, 116, 117, 127 Slot blot test, CPE protein, 476–477 Small, acid-soluble proteins, in spores, 46, 53–54, 56–58, 62, 472–473 Smart packaging, for meat products, 134 Smoked foods, 146–147, 784, 863 Snails, helminth transmission, 700, 704–705 Snow, John, 577 Snow Mountain virus, 627 SODA (Salmonella outbreak detection algorithm), 584 Sodium acetate, 767–769 Sodium benzoate, 769, 771 Sodium bisulfite, 778 Sodium chlorate, 293 Sodium chloride, see Salt Sodium chlorite, acidified, 277 Sodium citrate, 769, 771 Sodium diacetate, 768 Sodium hydroxide, in olive fermentation, 845–846 Sodium hypochlorite, 137 Sodium lactate, 132, 768–770 Sodium metabisulfite, 778 Sodium nitrate, 344 Sodium nitrite, 34, 344, 776–777 Sodium propionate, 770 Sodium sulfite, 778 Sodium tripolyphosphate, 133 Soft scalding, in poultry processing, 135 Softening, in vegetable fermentation, 847 Software, for predictive models, 1012–1014 Soil, microorganisms in antimicrobial resistance, 37 Bacillus cereus, 492 Clostridium botulinum, 443–444 Clostridium perfringens, 467 enteric viruses, 624–625 Listeria monocytogenes, 511 Solera system, 936 Solexa equipment, 976–977, 979 SOLiD system, 976–979 Solvent extraction, enteric viruses, 637 Somatic coliphages, as viral contamination indicators, 632 sop genes, Salmonella, 247 Sorbic acid and sorbates, 770–771 for cheese, 181 for meat products, 132 for wine, 930 Sortases, Listeria monocytogenes, 526 Source attribution, Campylobacter, 267–268 Souring milk, 170, 177 in muscle food spoilage, 116, 127 Sox proteins, Campylobacter, 265 spa genes, Shigella, 387, 389, 391

SMP_Food Microbiology_Index.indd

Spa proteins, 985 Space food, 1040–1041 Sparganosis, 702, 704 Sparganum, 705–706 Specific growth rate, 7–8 Specifications, microbiological criteria, 81 Specified risk materials, bovine spongiform encephalopathy agent in, 653, 663–665 Spectinomycin, 22 Spermidine, in muscle food spoilage, 117, 120, 139, 142, 144 Spermine, in muscle food spoilage, 117, 120, 144 Sphingolipids, fumonisins and, 608–609 Spices antimicrobial compounds in, 780–782 in fermented meat products, 863 Spiral plating, Cronobacter, 322 Spirometra, 698, 705–706 SPIs (Salmonella pathogenicity islands), 247, 249 Spo proteins, in sporulation, 49–52 Spoilage food, see Food spoilage quorum sensing and, 12 Spongiform encephalopathy, bovine, see Bovine spongiform encephalopathy Spore(s), 45–79; see also specific organisms activation, 46, 60 bacteriocin action on, 811–812 chemical composition, 46, 52–54 chemical resistance, 57–58 desiccation resistance, 56 dormancy, 46, 54 in foods, 62–70 bacteriology, 66–67 causing spoilage, 70–72 growth models, 73 illness caused by, 69 low-acid canned, 63–65, 70–72 prevention with quality assurance programs, 70 formation, see Sporulation freezing resistance, 56 germination, 46, 56, 60–62 heat resistance, 58–60, 68–69, 743 injury, 9–10 macromolecules, 53–54 mineralization, 59 outgrowth, 62 photoproducts, 56–57 pressure effects, 56 as probiotics, 73 radiation resistance, 56–57 resistance, 54–60 small molecules, 54 structure, 46, 52–53 UV-radiation resistance, 56–57 water content, 53, 59–60 wild-type, 59 Spore coat, 52, 53, 57–58 Spore core, 48, 53–54, 59, 62 Spore cortex, 52–53, 59, 62 Spore photoproduct lyase, 57 Sporeformers, see also specific organisms bacteriology, 65–68 contamination routes, 71–72 in dairy products, 178–180

1112

Manila Typesetting Company

detection, 72–73 distribution, 46 in food industry, 62–70 in food spoilage, 63, 70–72 growth, 65–68, 73 heat resistance, 64–65 illness related to, 69 in low-acid canned foods, 63–65, 70–72 in milk, 178–180 phylogeny, 45–46 Sporobolomyces, 770, 919 Sporolactobacillus, 45 Sporosarcina, 49 Sporothrix, 775 Sporotrichum, 770 Sporulation, 46–52 biochemical changes during, 48–49 Clostridium perfringens, 470 Cyclospora, 720 gene expression during, 46, 49–52 induction, 46–48 massive, 47 metabolism during, 47 morphologic changes during, 48–49 organisms involved in, see Sporeformers overview, 45–46 physiological changes during, 48–49 repression, 50 spo gene mutants and, 49–52 stages, 48–49 stationary phase, 47 temperature, 58–59 Spraying water, in poultry processing, 136 spv genes, Salmonella, 247–249 Square root models, 999, 1000 srt gene, Listeria monocytogenes, 526 SSAKE software program, 979 Stachyose, as prebiotic, 966 Standard(s), for microbiological criteria, 81 Staphylococcal enterotoxins, 547–568 antigenicity classification based on, 548–551 epitopes, 564–566 detection, 564–566 diversity generation, 552–553 dose required for poisoning, 550 evolutionary aspects, 551–552 family, 547–548 gastrointestinal effects, 566–568 genetics, 551–553 heat resistance, 561 history, 547 intentional contamination with, 94 mechanism of action, 566–568 nomenclature, 548–551 organisms producing, 547–548 production, regulation, 553–557 structure-function associations, 560–566 structures, 560–562 subtyping, 551 superantigenicity, 551–552, 567–568 vaccines, 565–566 variants, 551 virulence factors, 560–568 zinc binding, 562 Staphylococcal pathogenicity islands, 552 Staphylococcus antimicrobial action on, 548, 771, 772, 774, 776, 780, 786

11/08/2012 07:24AM

Index

1113

antimicrobial resistance, 34–35 classification, 548–551 in fermented fish products, 865 in meat fermentation, 867–868, 875 in milk, 173 in muscle foods, 126 nomenclature, 547–548 in nuts, 205 osmoregulation, 15 in seafood, 149, 154 species, 548 structures, 547–548 superantigens, 551–552 taxonomy, 547–548 Staphylococcus aureus, 547–573 antimicrobial action on, 768–770, 773, 779, 781, 783–785, 874 antimicrobial resistance, 19, 29, 32 bacteriocin action on, 812 as biological hazard, 1045 carriage, 557 enterotoxins, see Staphylococcal enterotoxins environmental susceptibility, 554–558 in fermented meat products, 862 food poisoning characteristics, 559 history, 547 incidence, 558–559 outbreaks, 547, 558–559 susceptible population, 560 toxic dose, 560 in grains, 216–217 history, 547 infective dose, 560 in milk, 18 models for, 1007, 1014 in muscle foods, 119, 126 in nuts, 206, 208 osmoregulation, 15 plasmids, 552 quorum sensing, 12 reservoirs, 557–558 resistance in, 34 in seafood, 141, 154 taxonomy, 547–548 temperature effects, 16 toxic shock syndrome, 567–568 toxins, see Staphylococcal enterotoxins virulence factors, 560–568; see also Staphylococcal enterotoxins water activity requirements, 750 Staphylococcus carnosus, 860, 863, 867–868, 874 Staphylococcus chromogenes, 548 Staphylococcus epidermidis, 22, 548 Staphylococcus hyicus, 548 Staphylococcus intermedius, 548 Staphylococcus saprophyticus, 548, 868, 874 Staphylococcus simulans, 868, 869 Staphylococcus xylosus, 868 growth model, 1000 in meat fermentation, 860 in muscle foods, 145 Starter cultures dairy product fermentation, 825–827 lactic acid bacteria in, 825–827 meat fermentation, 860–861, 866–876 winemaking, 923, 932

SMP_Food Microbiology_Index.indd

Static gelatin environment, for models, 1005 Stationary phase acid tolerance response, 233 in growth, 7 in sporulation, 47 Steady-state hypothesis, for models, 1009 Steam muscle food decontamination, 123 in poultry processing, 135, 136 Steam-ultrasound treatment, for Campylobacter, 277 STEC (Shiga-toxin producing Escherichia coli), 128–129, 290 Steeping, in brewing, 902 Stefan-Boltzmann constant, 740 Stepwise and Interactive Evaluation of Food Safety by an Expert System, 70 Stepwise selection, in models, 1000 Sterility, commercial, in food preservation, 64 Sterilization brewing equipment, 911–912 process, 743–744 sporeformer-containing foods, 64 STM proteins, Salmonella, 232 Stochastic models, 1007 Stomatococcus, 548 Storage, see also Refrigeration cocoa beans, 893 fruits, 193–196 nuts, 204, 210–212 in vegetable fermentation, 847 vegetables, 193–196 Streptococcus antimicrobial action on, 779, 781 cytolysins, 526 in meat fermentation, 871, 876 in milk, 177 in muscle food spoilage, 127 in normal microflora, 952 in nuts, 205 Streptococcus agalactiae, 982 Streptococcus diacetilactis, 870 Streptococcus faecalis antimicrobial action on, 784 in seafood, 154 viable but nonculturable cells, 10 Streptococcus lactis, 885 Streptococcus mutans, 34 Streptococcus pneumoniae, 982 Streptococcus pyogenes, 551, 784 Streptococcus thermophilus antimicrobial resistance, 30, 34 cold shock proteins, 16 in dairy products, 180 genomics, 982 in milk fermentation, 825–826, 828–831, 834–836 Streptogramins, 21 Streptomyces, 934 Streptomyces griseus subsp. hutter, 870 Streptomycin, 20–21, 26, 27, 120, 290 Stress lactic acid bacteria, 930–931 models for, 1002, 1007 Stress proteins, 33, 529 Strongyloides, 698 Strongyloides fuelleborni, 706 Strongyloides stercoralis, 706

1113

Manila Typesetting Company

Structural food systems models, 1004–1006 Strychnine, intentional contamination with, 95 Stunning, in poultry processing, 134 Stx family, see Shiga toxins Suberin, in fruits and vegetables, 193 Sublethal levels, to stressors, 32–34 Subpopulations, in compartment-based models, 1011–1012 Substrate utilization, muscle food spoilage, 116–118 Subtilins, 805 Subtyping, see Molecular subtyping Succinic acid formation, in winemaking, 926 Sucrose in cocoa fermentation, 891 Salmonella utilization, 226 Sugars fruits, 189 vegetables, 189 in winemaking, 929–930 Sulfadimethoxine, 21 in aquaculture, 26 Sulfadoxine, 22 Sulfamerizine, 26 Sulfamethazine, 21, 29 Sulfamethoxazole, 22, 28, 120 Sulfasalazine, 22 Sulfides, in muscle food spoilage, 117 Sulfisoxazole, 21, 28, 290 Sulfites, 778, 925–926 Sulfonamides, 21–23 Sulfophorane, in vegetable fermentation, 848 Sulfur dioxide, 778, 917, 919, 922, 925–926, 929 Sulfurospirillium, 263 Superantigens, staphylococcal enterotoxins, 548, 551–552, 567–568 Superdormant spores, 61–62 Surface plating, Cronobacter, 321–322 Surrogate viruses, for culture, 630 Surveillance for foodborne illness, 581–589 methods, 581–586 purposes, 581 for specific pathogens, 583–585 unidentified source, 581–582 population, 585–586 Survivor plots, in heat transfer, 740 Sushi parasite (Anisakis), 698–700 Suspension assay, viruses, 632 Swaminathania, 932 Sweat box processing, cocoa beans, 883–884 Sweet basil, 781–782 Sweet coagulation, 179 Swinnen model, 1001 SWISS-PROt software, 980 Sym’Previus database, 1013 Synbiotics, 966 Synergism, in hurdle technology, 17–18 Systems biology, 1007–1010

T

T-2 toxin, Fusarium, 94, 610 Taenia, in muscle foods, 119 Taenia asiatica, 682, 689–690 Taenia hydatigena, 685, 688 Taenia saginata, 682–686, 698 Asian, 689–690

11/08/2012 07:24AM

Index

1114 Taenia solium, 686–689, 698 Asian, 689–690 Tailing kinetics, pressure treatment, 756 Talaromyces, 71 Talaromyces flavus, 769 Tandem mass spectrometry, in proteomics, 988 Tannic acid, 784 Tannins, 785, 904 Tapeworms, see also Taenia canid, 698, 704, 708 cat, 705 dog, 705 dwarf, 698, 707 fish, 700 rodent, 705 sparganosis, 698, 705–706 Target modification, in antimicrobial resistance, 22 Tartaric acid antimicrobial action, 771 in wine, 926, 931 Tccp protein, Escherichia coli, 301 T-cell receptors, staphylococcal enterotoxin binding to, 552, 563, 567 TCP colonization factor, Vibrio cholerae, 414 Tcp protein, Vibrio cholerae, 416 tdh gene, Vibrio, 414, 417, 419 Tea tree oil, antimicrobial compounds in, 782 Tecra Kit, for Bacillus cereus toxins, 499 Telithromycin, 22 Temperature effects, 16–17; see also Refrigeration; subjects starting with Heat Bacillus subtilis, 17 bacteriocins, 807 biofilms, 13 Campylobacter, 265–266 Clostridium botulinum, 16 Clostridium perfringens, 466–467, 469 cocoa fermentation, 889–890 Cronobacter, 315–316 Escherichia coli, 16–17, 290–291 fruit spoilage, 193–194 high-pressure processing, 755 injury, 9–10 kinetics, 16–17 Listeria monocytogenes, 10, 17, 506 models for, 1000, 1001 monitoring, 1050–1051 radiation treatment, 754 Salmonella, 14, 229–234 Shigella, 17, 381 spores and sporulation, 58–59 Staphylococcus aureus, 16 Streptococcus thermophilus, 16 Vibrio, 17, 404–405 winemaking, 922 Yersinia, 17, 342–344 Temporary adaptation, to antimicrobials, 32–33 Tenderized meat products, 131–132 “Terroirs,” 920 Terrorism, see Intentional contamination tet(O) gene, in antimicrobial resistance, 27, 31 Tetracyclines, 21–23, 27, 28, 120, 245, 290

SMP_Food Microbiology_Index.indd

Tetratricopeptide repeats, Bacillus cereus toxins, 497–498 Tetrodotoxin, intentional contamination with, 95 Thawing, frozen foods, 747–748 Theobromo cacao, 881 Thermal death time, 741–742 Thermal gradient, 739 Thermal processing, see Heat treatment and inactivation Thermoanaerobacterium, 70 Thermoanaerobacterium saccharolyticum, 64, 65 Thermolabile cytotoxic protein, 250 Thermophiles, 16, 58 in dairy product spoilage, 72 heat resistance, 65 Thermostable direct hemolysin, Vibrio, 419 Thermovinification, 917 Theys model, 1005, 1006 Thin-layer chromatography, fumonisins, 609 Thiocyanate, formation in lactoperoxidase action, 171, 779 Thiols, in winemaking, 931 Thiopropanal-S oxide, 782 Thioredoxin reductase, 783 Thiosulfate-citrate-bile salts-sucrose agar, Vibrio, 402 Three-class plans, for sampling, 84 Threonine synthase, Cronobacter, 325 Thujone, 781 Thyme, 781 Thymol, 781 Thyroid disease, Yersinia enterocoliticarelated, 362–364 Tight junction rearrangements, CPE protein, 483–485 Tilmicosin, 21 Tinabal, 866 Tingling throat syndrome, 697 Tir protein, Escherichia coli, 300–301 Tissue culture assays, Listeria monocytogenes, 520 Tomatine, 192 TonB protein, Yersinia enterocolitica, 354 Tongue worms, 706 Top-down approach, for models, 1008 Torrymeter, 118 Torula, 773 Torulaspora, 180, 770, 909 Torulaspora delbrueckii, 921, 936 Torulopsis antimicrobial action on, 773, 782 in cocoa fermentation, 885 ToxA protein, Yersinia enterocolitica, 354 ToxB protein, Escherichia coli, 301 Toxic shock syndrome, Staphylococcus aureus, 547–549, 552, 563, 567–568 Toxin(s), see also Enterotoxin(s); Mycotoxins; specific organisms as chemical hazards, 1045 Toxin-antitoxin stabilization system, 31 Toxocara canis, 698, 707–708 Toxocara felis, 707–708 Toxoplasma gondii, 713, 721–722 in muscle foods, 119, 120 in organic farming, 25 toxR gene, in acid tolerance response, 15 ToxR regulon, Vibrio cholerae, 415–416

1114

Manila Typesetting Company

Traceability, muscle foods, 121–122 Transduction, in antimicrobial resistance, 25 Transferrin Salmonella competition with, 251 Yersinia enterocolitica, 353 Transformation, in antimicrobial resistance, 25 Transfusions, Yersinia enterocolitica transmission in, 345 Transit tolerance test, probiotics, 961 Translocation, bacterial, Cronobacter, 329–330 Transmembrane gradients, 6 Transmembrane intimin receptor, Escherichia coli, 300 Transmissible mink encephalopathy, 130, 651 Transmissible spongiform encephalopathies, 651; see also Bovine spongiform encephalopathy Transmission, see also specific organisms in epidemiology, 580–581 Transposons, in antimicrobial resistance, 24 TransTerm software, 979 Trassiudang, 866 traT gene, Salmonella, 250 Trehalose, Cronobacter, 316–317 Tricarboxylic acid cycle, 5, 47, 933 Trichinella, 673–682 disease clinical manifestations, 678–679 diagnosis, 679–680 epidemiology, 677–678 pathogenesis and pathology, 678 prevalence, 677–678 prevention, 680–682 treatment, 680 freezing, 748 history, 673–674 life cycle, 675–676 species, 674 Trichinella britovi, 674, 676 Trichinella murrelli, 674, 676, 679 Trichinella nativa, 674, 679 Trichinella nelsoni, 674 Trichinella papuae, 674 Trichinella pseudospiralis, 674–675 Trichinella spiralis, 673–674, 676, 680, 698, 862 in muscle foods, 119 Trichinella zimbabwensis, 674–675 Trichinoscopy, 681 Trichoderma, 770, 935 Trichosporon, 205, 782 Trichostrongylus, 698, 704, 707 Trichothecene, 934 Fusarium, 610 Trichothecium, 205 Trichuris suis, 707 Trichuris trichiura, 707 Trichuris vulpis, 707 “Trigger molecule,” in signal transduction, 13 Trimethoprim-sulfamethoxazole, 28, 331 Trimethylamine, in muscle food spoilage, 118 Trisodium phosphate, 633 muscle food decontamination, 123 poultry processing, 277 in poultry processing, 138

11/08/2012 07:24AM

Index

1115

tRNA intergenic spacer PCR, Cronobacter, 313–314 Trophozoites Balantidium coli, 726 Entamoeba histolytica, 727 Giardia, 724–725 Trovofloxacin, 20 Trypanosoma cruzi, 714 Tryptamine, in muscle food spoilage, 142 Trypticase soy agar, Cronobacter, 312 Tryptophan, in wine, 925 Turkey, see Poultry 12D, in thermal processing, 64 Tylosin, 21, 27, 29 Type I secretion system, Salmonella, 249 Type II secretion system, Yersinia, 355 Type III secretion system Escherichia coli, 300 Salmonella, 247 Shigella, 387, 389–390 Vibrio cholerae, 419–420 Yersinia, 355, 357–359 Type IV secretion system, Campylobacter, 271–272 Typhoid fever (Salmonella typhi), 243–244 Tyramine in muscle food spoilage, 117, 120, 139, 142–143 in vegetable fermentation, 848

U

Ultrafiltration, enteric viruses, 637 Ultrasound for Campylobacter, 277 for meat processing, 124 Uncertainty, in risk assessment, 1030–1032 Uncinula necator, 934 Unilever, models developed by, 1012 University of Wisconsin, Shelf Stability Predictor, 1014 Urea, in wine, 925 Urease, Yersinia enterocolitica, 353 Urinary tract infections, probiotics for, 956 UV radiation, 56–57 Aspergillus detection, 602 enteric viruses, 634 fruits and vegetables, 195 Yersinia enterocolitica, 344

V

V antigen, Yersinia enterocolitica, 359 Vaccines Campylobacter, 277 Clostridium botulinum, 458 Escherichia coli, 293 fish pathogens, 26 hepatitis A virus, 630 norovirus, 628 probiotic, 957 Shigella, 385 staphylococcal enterotoxins, 565–566 Taenia saginata, 686 Vacuolating factor, Bacillus cereus, 495 Vacuum packaging, 125, 510, 747 Valeric acid, in cocoa fermentation, 886 Validation, HACCP procedures, 1052 Van Derlinden model, 1012 van Ermengem, on botulism, 442 Van Impe model, 999, 1006

SMP_Food Microbiology_Index.indd

Vancomycin, 19, 21, 120 Vanillin and derivatives, 318, 782 Variability, in risk assessment, 1030–1032 Variable-number tandem repeats, in subtyping, 1063–1065 Variables sampling plans, 84 Variant Creutzfeldt-Jakob disease, 120, 130, 651 agent, bodily distribution, 654–656 characteristics, 660 epidemiology, 658–659 outbreaks, 658–659 prevention, 660–664 VCAKE software program, 979 Vegetables, see also Juices antimicrobial usage in, 25–26 chilling, 746–747 composition, 189 fermented, 841–855 bacteriophages in, 846–847 biochemistry, 847–848 cabbage, 844–848 cucumbers, 843–848 examples, 842 genomics, 848–850 history, 844 olives, 845–848 overview, 841–843 microbial inhibitors in, 189 organisms in Alternaria, 188, 192–194 Aspergillus niger, 193, 196 Aureobasidium pullulans, 196 Bacillus cereus, 492 Bacillus cinerea, 192–196 Botrytis, 188, 190, 196 Campylobacter, 270 Candida, 196 Cladosporium, 194 Clostridium botulinum, 444, 449 Colletotrichum, 188, 190–192, 196 Cronobacter, 319–320 Cyclospora, 717, 718, 720 enteric viruses, 623–625 Escherichia coli, 291, 296–298 Escherichia coli O157:H7, 191 Fusarium, 194 Geotrichum, 188, 196 helminths, 698, 702–703 hepatitis A virus, 625–626 Leuconostoc, 192 Listeria monocytogenes, 513 Metschnikowia, 196 Monilinia, 188, 191–192, 194–195 Mucor, 196 Myocentrospora, 191 Nectria, 193 Pectobacterium carotovora, 188, 190 Penicillium, 188, 190–194 Pezicula alba, 193, 196 Phialophora, 190 Phoma, 190, 194 Phomopsis, 192 Phytophora, 193 Pseudomonas, 191, 194, 196 Rhizocotonia, 190 Rhizopus stolonifer, 191, 193–194 Salmonella, 191, 235–239 Sclerotinia, 190–191, 194

1115

Manila Typesetting Company

Sclerotium, 192 Shigella, 381–383 Vibrio cholerae, 410 Yersinia enterocolitica, 346–348 outbreaks related to, 575 shelf life, 189 spoilage characteristics, 187–190 contamination source, 190–192 control, 193–196 defense reactions, 192–193, 195 humidity and, 193–194 mechanisms, 191–192 microorganisms causing, 191–192 in modified atmospheres, 193–194 resistant cultivars, 195–196 temperature effects, 193–194 water activity, 749 Verification, HACCP procedures, 1052 Vertical gene transmission, 24 Vesivirus, 621 Veterinary Diagnostic Laboratory Reporting System, 586 Vi antigen, Salmonella, 227, 250–251 Viable but nonculturable cells, 9–11 in biofilms, 13 Campylobacter, 265 Salmonella enterica serovar Typhimurium, 11 Vibrio cholerae, 408 in winemaking, 919 Vibrio, 401–439 antimicrobial action on, 771, 774, 784 antimicrobial resistance, 31 biochemical characteristics, 402–403 cultures, 402 depuration, 404 disease, 402–403, 588 enumeration, 403 environmental susceptibility, 404–406 in fermented fish products, 866 habitats, 401 inhibitors, 405–406 isolation, 402 in muscle foods, 120 pressure effects, 405 radiation susceptibility, 405 in seafood, 119, 140, 150, 401–402 species, 401 temperature effects, 404–405 viable but nonculturable, 10 water activity requirements, 750 Vibrio alginolyticus biochemical characteristics, 403 classification, 429 disease, 430 infectious dose, 430 isolation, 402 in muscle foods, 145–146 reservoirs, 429–430 in seafood, 150, 152, 402 virulence factors, 430 Vibrio anguillarum, 153 Vibrio carchariae, 402 Vibrio cholerae, 406–416 Ace toxin, 412 acid resistance, 290 antimicrobial action on, 775 biochemical characteristics, 403, 407

11/08/2012 07:24AM

Index

1116 Vibrio cholerae (Continued) biofilms, 408–409, 416 carriage, 409 cellular response to, 411–412 classification, 406–407 colonization factors, 414–415 cytolysins, 412, 414 disease (cholera), 409–411, 577 DNA analysis, 407–408 El Tor biotype, 406–407 environmental susceptibility, 404–406 enzymatic activity, 412 epidemiology, 404 flagella, 415 genomics, 987–988 hemagglutinins, 412 hemolysins, 414 incubation period, 410–411 infectious dose, 411 intentional contamination with, 91 isolation, 402 lipopolysaccharide, 415 motility, 415 natural habitats, 401 pH effects on, 15 polysaccharide capsule, 415 quorum sensing, 12 regulation, 415–416 reservoirs, 408–409 RtxA protein, 414 rugose form, 409 in seafood, 141, 148–150, 153 serogroups, 406–407 susceptible population, 411 taxonomy, 406–407 temperature effects, 17 toxins, 411–414; see also Cholera toxin ToxR regulon, 415–416 transmission, 409–410 viable but nonculturable, 408 virulence factors, 411–414 water activity requirements, 750 Zot toxin, 412 Vibrio cholerae non-O1, 402 Vibrio cholerae non-O1/non-O139, 407 adherence factors, 415 disease, 411 reservoirs, 408–409 toxins, 415 Vibrio cholerae O1, 406–408 Vibrio cholerae O139 Bengal, 407, 410 Vibrio cholerae O1/O139, 411 Vibrio cincinnatiensis, 153 Vibrio damsela, 402 Vibrio diazotrophicus, 152 Vibrio fetus, see Campylobacter jejuni Vibrio fluvialis, 426–427 biochemical characteristics, 403 classification, 426 disease, 403, 426–427 environmental susceptibility, 404–405 epidemiology, 403–404 infectious dose, 427 isolation, 402 reservoirs, 426 in seafood, 150, 152, 402 susceptible population, 427 virulence factors, 427

SMP_Food Microbiology_Index.indd

Vibrio furnissii biochemical characteristics, 403 disease, 427–428 isolation, 402 reservoirs, 427 in seafood, 402 taxonomy, 427 virulence factors, 428 Vibrio harveyi, 11 Vibrio hollisae, 428–429 biochemical characteristics, 403 isolation, 402 in seafood, 153 Vibrio marinus, 153 Vibrio mediterranei, 152 Vibrio metschnikovii, 150, 402 Vibrio mimicus biochemical characteristics, 403 isolation, 402 natural habitats, 401 in seafood, 150, 402 Vibrio nereis, 152 Vibrio orientalis, 153 Vibrio parahaemolyticus, 416–420 antimicrobial action on, 781, 785 biochemical characteristics, 403 disease, 404, 418–419 environmental susceptibility, 404–406 epidemiology, 403–404 growth, 417–418, 746, 1000 infectious dose, 418–419 isolation, 402 Kanagawa phenomenon-positive (KP+), 417, 419 reservoirs, 417–418 risk assessment, 1026, 1032 in seafood, 141, 148–154, 402 susceptible population, 418–419 viable but nonculturable, 11 virulence factors, 419–420 water activity requirements, 750 Vibrio splendidus, in seafood, 152 Vibrio vulnificus, 420–426 biochemical characteristics, 403 capsule, 424 classification, 420–421 disease, 422–423 environmental susceptibility, 404–405, 421 epidemiology, 403–404 genotypes, 421–422 growth, 421 infectious dose, 423–424 isolation, 402 lipopolysaccharide, 424–425 reservoirs, 421 in seafood, 141, 148–153, 402 siderophores, 424 susceptible population, 423–424 viable but nonculturable, 10–11 Vinegar, antimicrobial action, 767–768 Violet red bile glucose agar, Cronobacter, 321–323 Vip protein, Listeria monocytogenes, 525–526 vir genes Shigella, 386, 388, 390–391 Yersinia, 358–359

1116

Manila Typesetting Company

Vir plasmid, Campylobacter, 271–272 VirF protein, Yersinia, 362 Virginiamycin, 21, 25, 29 Virulence factors, see also specific organisms Bacillus cereus, 495–499 Campylobacter, 271–272 Clostridium perfringens, 472–485 Cronobacter, 324–325, 328–330 Escherichia coli, 287, 291, 301–302 temperature effects, 17 Vibrio parahaemolyticus, 419–420 Virulence plasmids Escherichia coli, 301–302 Salmonella, 247–250 Shigella, 385–391 Yersinia enterocolitica, 355–356, 361–362 Viruses, see Human enteric viruses; specific viruses Visceral larva migrans, 708 Vitamins, in winemaking, 922 Vittaforma, 713, 723 VNTR (variable-number tandem repeats), in subtyping, 1063 Vomiting toxin, Bacillus cereus, 495, 498–499 Vomitoxin, 610 Vomitus, enteric viruses in, 626 Vulnerability, in CARVER+Shock strategy, 98–100

W

Wallemia, in grains, 213 Wallerstein Laboratories nutrient agar, 910–911 Walnuts, organisms in, 208 Washing fruits and vegetables, 633–634 in meat processing, 130 in poultry processing, 136–137 Water in foods, availability, 748–752 in fruit, 188–189 in grain processing, 214 loss of, in fruits and vegetables, 193–194 muscle food decontamination, 123 in nuts, 204 organisms in bovine spongiform encephalopathy agent, 654 Campylobacter, 270 Cryptosporidium, 717 Cyclospora, 719–720 enteric viruses, 623–625, 633–634 Escherichia coli, 291, 293, 296, 298 Fasciola hepatica, 702–703 Giardia, 724–725 helminths, 704 Listeria monocytogenes, 510, 511 Vibrio, 401–402 Vibrio cholerae, 408 Vibrio fluvialis, 427 Vibrio furnissii, 428–429 Vibrio hollisae, 428–429 Vibrio parahaemolyticus, 417–418 Vibrio vulnificus, 421 Yersinia, 346 Yersinia enterocolitica, 346

11/08/2012 07:24AM

Index

1117

poultry decontamination, 136–137 removal, see Dehydration in spores, 53, 59–60 in vegetables, 188–189 Water activity (aw) definition, 748–749 of food, 748–752; see also specific organisms, water activity requirements bakery products, 215 cereals, 203 formula, 749 grains, 215 nuts, 203, 205 osmoregulation and, 15–16 models for, 1000 Water replacement theory, 232 Water vibration, 232 Websim-MILQ Web tool, 1014 Weibull distribution, 64 Weibull model, 999 Weissella, 843 Weissella viridescens, 115, 871 Western blot test, bovine spongiform encephalopathy agent, 653, 664 Wet processing, coffee fermentation, 895 Wheat, see Grains and cereal products Whipworm (Trichuris trichiura), 707 Whiting model, 1011 Whole-genome sequencing, 1066, 1071–1072 Wildlife Escherichia coli in, 292–293 Taenia in, 682, 689–690 Trichinella in, 673–682 Winemaking, 915–947 acetic acid bacteria in, 932–934 antimicrobials in, 785 fermentation in alcoholic, 917–918 juice factors in, 921–922 malolactic, 918, 928–932 organic acid production, 926 sparkling, 935–936 stuck or sluggish, 924 sulfur dioxide in, 922 temperature, 922 yeast growth during, 920–924 flavor, 919, 927, 931 fortified type, 936 grapes for acetic acid bacteria on, 932–934 carbohydrates in, 925 clarification, 922 composition, 921–922 crushing, 917 lactic acid bacteria on, 928–932 nitrogen compounds in, 925 pretreatment, 917 selection, 916–917 sulfur dioxide content, 922 history, 915 lactic acid bacteria for, 928–932 mold contamination in, 934–935 outline of, 916 postfermentation processes, 918–919 red, 917–918 sparkling type, 935–936 spoilage, 927, 931

SMP_Food Microbiology_Index.indd

storage, 918–919 white, 917–918 yeasts for, 919–928 autolysis, 927, 935 biochemistry, 924–927 flavor production, 927 in fortified wine aging, 936 genetic improvement, 927–928 growth during fermentation, 920–924 inoculation, 923 interactions with other microorganisms, 923–924 metabolism, 924–927 origin, 919–920 removal from champagne, 935 selection, 923 spoilage, 927, 931 “Winged-helix DNA binding protein,” Staphylococcus, 555 Wiskott-Aldrich syndrome protein, Escherichia coli, 301 World Health Organization on antibiotic use, 25 Cronobacter code, 312 Global Foodborne Infectious Network, 585 Global Salm-Surv Network, 1071 microbiological criteria, 83 risk analysis for enteric viruses, 639 “Terrorist Threats to Food: Guidance for Establishing and Strengthening Prevention and Response Systems,” 93 World Trade Organization, food safety risk analysis, 1021 Worms, see Helminth(s) Wort, in brewing, 903–904 Wound botulism, 446–448

X

Xanthohumol, 783 Xanthomonas, 205 Xanthomonas campestris, 312 Xeromyces bispora, 750 Xerotolerance, 15, 751 X-rays, in food preservation, 752 Xylose lysine desoxycholate, Salmonella culture, 226

Y

Yad proteins, Yersinia, 351 YadA adhesin, Yersinia enterocolitica, 355–357, 363–364 Yakult, 949 Yarrowia lipolytica, 180 YE36550 protein, Yersinia enterocolitica, 355 Yeasts, see also specific organisms antimicrobial action on, 768 brewing, 901–903, 905–906 in cocoa fermentation, 884–885 in coffee fermentation, 894 in fruit and vegetable spoilage control, 196 in fruits, 190–192 in grains, 214–217 in meat fermentation, 118–119, 869–870 in milk, 180–181 in muscle foods, 113, 127 in nuts, 204–209

1117

Manila Typesetting Company

in vegetables, 190–192 viable but nonculturable, 11 water activity requirements, 750 in wine, 919–928 Yersinia antimicrobial action on, 774 antimicrobial resistance, 25, 31 characteristics, 339–344 classification, 341–342 disease, 588 environmental susceptibility, 342–344 epidemiology, 588 growth, 342–344 history, 339 pathogenicity, 339, 341, 353–354 reservoirs, 345–346 taxonomy, 339, 341–342 tolerance, 343–344 transmission, 341 Yersinia aldovae, 339–340 Yersinia bercovieri, 340, 342, 352, 355 Yersinia enterocolitica, 339–376 Ail protein, 351 “American” strains, 342 antimicrobial action on, 768, 769, 771, 774–775, 778, 779, 781 biochemical identification, 340 in biofilms, 13 biotyping, 341–342, 348 characteristics, 341–344 cold shock proteins, 342–343 disease autoimmune sequelae, 344, 362–364 characteristics, 344–345 incidence, 346 outbreaks, 346–348 pathology, 348–349 enterotoxins, heat-stable, 351–352 environmental susceptibility, 343–344 flagella, 352–353 growth, 343–344, 746 heat resistance, 344 infective dose, 344 invasin, 349–351 iron acquisition, 353–355 lipopolysaccharide, 353 in milk, 743 in modified-atmosphere packaging, 746 in muscle foods, 119, 126 myf fibrillae, 352–353 O antigens, 341–342 pathogenicity, 348–364 pH effects, 14–15 phospholipase, 353 radiation susceptibility, 344 reservoirs, 345–346 in seafood, 141 serotyping, 341–342 subtyping, 341–342 taxonomy, 341–342 temperature effects, 17 transmission, 346 urease, 354 virulence factors, 348–364 virulence plasmid, 355–356, 361–362 YadA adhesin, 355–357 Yop proteins, 355–362 Ysc secretion, 357–359

11/08/2012 07:24AM

Index

1118 Yersinia frederiksenii, 340, 342, 355 Yersinia intermedia, 340, 342, 355 Yersinia kristensenii, 340, 342 Yersinia mollaretii, 339–340, 342, 352 Yersinia nurmii, 339–340 Yersinia pekkanenii, 339–340 Yersinia pestis, 355, 357 characteristics, 339–341 intentional contamination with, 91, 94 subtyping, 1065 Yersinia pseudotuberculosis characteristics, 339–343 cold shock proteins, 342–343 disease, 344–345, 362–363 incidence, 346 outbreaks, 347–348 growth, 345 heat resistance, 344 reservoirs, 345–346 temperature effects, 342–344 transmission, 341 virulence factors, 349–364 virulence plasmid, 355–356 YadA adhesin, 355–357 Yop proteins, 355–362

SMP_Food Microbiology_Index.indd

Yersinia rohdei, 340 Yersinia ruckerei, 339 Yersiniabactin, 354 Ymo proteins, Yersinia, 352 YmoA protein, Yersinia, 352 Yogurt, 178, 831–833 Yop proteins, 355–362 YpkA protein, Yersinia, 360–361 YplA protein, Yersinia enterocolitica, 353 YSA pathogenicity islands, Yersinia enterocolitica, 355 Ysc protein, Yersinia enterocolitica, 357–359 Ysp protein, Yersinia enterocolitica, 355 yst gene, in acid resistance, 14–15 Yst protein, Yersinia, 351–352 Yts1 protein, Yersinia, 355

Z

z value, in heat transfer, 740, 742 Zapatera, 846 Zearalenone, 611 Zero tolerance for Listeria monocytogenes, 513 for muscle food pathogen control, 130 Zinc benzoates, 769

1118

Manila Typesetting Company

Zinc binding, staphylococcal enterotoxins, 562 Zonula occludens toxin (Zot), Vibrio cholerae, 412 Zoonoses Campylobacter, 269–271 helminths, see Helminth(s) Yersinia, 345–346 Zot toxin, Vibrio cholerae, 412 zpx gene, Cronobacter, 330 ZurR protein, Listeria monocytogenes, 529 Zygosaccharomyces, 770, 773, 909–910 osmoregulation, 15 in wine, 919, 936 Zygosaccharomyces bailii antimicrobial action on, 769 models for, 1005 in winemaking, 926–927 Zygosaccharomyces bisporus water activity requirements, 750 in wine spoilage, 936 Zygosaccharomyces microellipsoides, 180 Zygosaccharomyces rouxii, 750 Zymocins, 923–924 Zymomonas, as beer contaminant, 910

11/08/2012 07:24AM