Journal of Experimental Bontany

Journal of Experimental Botany, Vol. 59, No. 15, pp. 4171–4182, 2008  doi:10.1093/jxb/ern260 This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details)

RESEARCH PAPER

Phytochemical and genetic analyses of ancient cannabis from Central Asia Ethan B. Russo1,2,3,*, Hong-En Jiang4,5, Xiao Li5, Alan Sutton2, Andrea Carboni6, Francesca del Bianco6, Giuseppe Mandolino6, David J. Potter2, You-Xing Zhao7, Subir Bera8, Yong-Bing Zhang5, En-Guo Lu¨ 9, David K. Ferguson10, Francis Hueber11, Liang-Cheng Zhao12, Chang-Jiang Liu4, Yu-Fei Wang4 and Cheng-Sen Li5,13,* 1

Visiting Professor, Institute of Botany, Chinese Academy of Sciences, [email protected] 

2

GW Pharmaceuticals, Porton Down Science Park, Salisbury, Wiltshire SP4 OJQ, UK 

3

Faculty Affiliate, Department of Pharmaceutical Sciences, University of Montana, Missoula, MT, USA

4

Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Beijing  100093, China  5

Bureau of Cultural Relics of Turpan Prefecture, Turpan 838000, Xinjiang, China 

6

CRA-Centro di Recerca per le Colture Industriali, via di Corticella 133, 40128, Bologna, Italy 

7

State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany, Chinese  Academy of Sciences, Kunming 650204, China 

8 9

Department of Botany, University of Calcutta, Kolkata 700019, India  ¨ru  ¨mqi, Xinjiang 830011, China  Xinjiang Institute of Archaeology, 4-5 South Beijing Road, U 

10

Institute of Palaeontology, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria 

11

Department of Paleobiology, Smithsonian Institutions, Washington, DC 20560-0121, USA

12

College of Biological Science and Biotechnology, Beijing Forestry University, Beijing 100083, China 

13

Beijing Museum of Natural History, Beijing 100050, China 

Received 7 August 2008; Revised 24 September 2008; Accepted 25 September 2008

Abstract The Yangha Yanghaii Tombs Tombs near near Turpan Turpan,, Xinjia Xinjiangng-Uig Uighur hur Autonomous Region, China have recently been excavated to reveal the 2700-year-old grave of a Caucasoid shaman whose accoutrements included a large cache of cannabis, superbly preserved by climatic and burial conditions. A multidisciplinary international team demonstrated through botanical examination, phytochemical investigat investigation, ion, and genetic genetic deoxyribonuc deoxyribonucleic leic acid analysis by polymerase chain reaction that this material contained tetrahydrocannabinol, the psychoactive compon component ent of cannab cannabis, is, its oxidat oxidative ive degrad degradati ation on product, product, cannabinol, cannabinol, other metabolit metabolites, es, and its synthetic thetic enzyme, enzyme, tetrahydr tetrahydrocann ocannabinol abinolic ic acid synthase, synthase, as well well as a nove novell gene geneti tic c vari varian antt with with two two sing single le nucleotide polymorphisms. The cannabis was presumably ably empl employ oyed ed by this this cult cultur ure e as a medi medici cina nall or

psycho psychoact active ive agent, agent, or an aid to divina divinatio tion. n. To our knowle knowledge dge,, these these invest investiga igatio tions ns provid provide e the oldest oldest docume documenta ntatio tion n of cannab cannabis is as a pharma pharmacol cologi ogical cally ly acti active ve agen agent, t, and and cont contri ribu bute te to the the medi medica call and and archaeological record of this pre-Silk Road culture. Key words: words: Archaeol Archaeology, ogy, botany, botany, cannabis, cannabis, cannabin cannabinoids oids,, archa archaeob eobota otany, ny, ethnop ethnopha harma rmacol cology ogy,, gen geneti etics, cs, medica medicall history, phytochemistry.

Introduction Uighur farmers cultivating the land at the base of the Huoyan Shan (‘Flaming Mountains’) in the Gobi Desert near Turpan, Xinjiang-Uighur Autonomous Region, China some 20 years ago ago uncove uncovered red a vast vast ancien ancientt cemet cemetery ery (54 000 m 2) that  that  seemingly corresponds to the nearby Aidinghu, Alagou, and

* To whom correspondence should be addressed: E-mail: [email protected]; [email protected] ª

2008 The Author(s).

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( http://creativecommons.org/licenses/by-nc/2.0/uk/ ) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

4172 Russo  et al.

Sube Subeix ixii exca excava vati tion onss (Ma (Ma and and Wang Wang,, 1994 1994;; Chen Chen and and Hiebert, Hiebert, 1995; 1995; Davis-Kimb Davis-Kimball, all, 1998; 1998; Kamberi, Kamberi, 1998; 1998; An, 2008) (see Supplementary Fig. S1 at  JXB online) attributed to the the Gush culturee (later (later render rendered ed Ju ¨shi, ¨shi, or Cheshi Cheshi)) ushı ı  cultur  (Aca (Acade demi miaa Turf Turfan anic ica, a, 2006 2006). ). The The first first writ writte ten n repo report rtss concerni concerning ng this clan, drafted drafted about about 2000 years BP (before presen present) t) in the Chines Chinesee histor historica icall record record,, Hou Hanshu, Hanshu, described nomadic light-haired blue-eyed Caucasians speaking ing an Indo Indo-E -Eur urop opea ean n lang langua uage ge (pro (proba babl bly y a form form of  Tochar Tocharian ian,, an extinc extinctt Indo-E Indo-Euro urope pean an tongue tongue relate related d to Celtic, Italic, and Anatolic (Ma and Sun, 1994). The Gush u shı ı  tend tended ed hors horses es and and graz grazin ing g anim animal als, s, farm farmed ed the the land land and were accomplished archers (Mallory and Mair, 2000). The The site site is cent centra rall lly y loca locate ted d in the the Eura Eurasi sian an land landma mass ss (Fig. 1A, B), 2500 km from any ocean and located in the Ayding Lake basin, the second lowest spot on Earth after  the Dead Sea (Fig. 1A, B). Formal excavations completed in 2003 revealed some 2500 tombs dating from 3200–2000 years BP (Xinjiang Institute of Cultural Relics and Archaeology, 2004). Other evidence from chipped stone tools and other items indicate a possible human presence in the area  for some 10 000–40 000 years (Kamberi, 1998; Academia  Turfanica, 2006). Due to a combination of deep graves (2 m

or more), an extremely arid climate (16 mm annual rainfall), and alkaline soil conditio conditions ns (pH 8.6–9.1 8.6–9.1 (Pan, 1996), 1996), the remarkabl remarkablee preserva preservation tion of the human remains resulted resulted in the the mumm mummifi ifica cati tion on of many many bodi bodies es with withou outt a need need for  for  chemic chemical al method methods. s. Numero Numerous us artefa artefacts cts from from the tombs tombs includ included ed equest equestria rian n equip equipmen mentt and numero numerous us Wester Western n Asian Asian crops crops such such as Capparis spinosa L. (capers) (capers) (Jiang (Jiang et al., 2007), Triticum spp. (wheat), Hordeum spp. (naked barley), barley), and Vitis Vitis vinifera vinifera L. (grapevin (grapevines) es) (Jiang, (Jiang, 2008), 2008), often often centur centuries ies before before their their first first descr descript iption ionss in Easter Eastern n China (Puett, 1998). One tomb, M90 (GPS coordinates: 42 ° 48.395 N, 89° 38.958 E; elevation, elevation, 58 m) (see Supplementa Supplementary ry Fig. S2A, B at  JXB online online), ), contain contained ed the skeleta skeletall remain remainss of a male of high social status of an estimated age of 45 years, whose accoutrements included bridles, archery equipment, a  kongou harp, harp, and other other materi materials als suppor supportin ting g his identi identity ty as a shaman (see Supplementary Figs S3A, B, 4A–C at  JXB online). His burial as a disarticulated skeleton, as opposed to a mumm mummifi ified ed body body as more more freq freque uent ntly ly was was foun found, d, suggested that he probably died in the highlands of the Tian an Shan Heavenly Mountains,’ Mountains,’ or  Ta¨ ngri Tagh in Uighur) Uighur) an (‘Heavenly (Fig. 1), and his bones were later interred at Yanghai, as #

#

Fig. 1. Area maps. (A) Map of Turpan, Xinjiang, China and its location in Central Asia. (B) Map of Yanghai Tombs site and surrounding area  (adapted from Xinjiang Institute of Cultural Relics and Archaeology, 2004).

Ancient cannabis  4173

nearby nearby tombs tombs contai contained ned large large timbers timbers of  Picea (spruce) spp. spp. that that grow grow at 3000 3000 m elev elevat atio ion. n. Mode Modern rn Uigh Uighur  ur  past pastor oral alis ists ts foll follow ow a simil similar ar annu annual al migra migrato tory ry path path to summer summer grazing grazing lands some some 60–80 60–80 km distan distantt from from the tomb tombs. s. Near Near the the head head and and foot foot of the the sham shaman an’s ’s bier bier lay lay a large leather basket and wooden bowl (see Supplementary Fig. S5A, B at  JXB online) filled with 789 g of vegetative matte matter, r, init initia iall lly y thou though ghtt to be Coriandru Coriandrum m sativum sativum L. (coriander), but which, after meticulous botanical examination, proved to be Cannabis sativa L. (Jiang et al., 2006). An init initia iall radi radioc ocarb arbon on date date of 2500 2500 year yearss BP has has subsubsequen sequently tly been been correc corrected ted to a calibr calibrate ated d figure figure of 2700 2700 years BP based on additional additional analyses of equestrian equestrian gear and correlation to tree ring data (dendrochronology) in China. While an earlier earlier publicatio publication n (Jiang (Jiang et al., 2006) emphasized morpho morpholog logica icall featur features es in identi identifyin fying g the cannab cannabis, is, the current study used additional botanical, phytochemical, and genetic investigations to demonstrate that this cannabis was psycho psychoact active ive and probab probably ly cultiva cultivated ted for medici medicinal nal or  divi divina nato tory ry purpo purpose ses. s. Great Great care care was take taken n to prev preven ent  t  contamination of the sample throughout the analyses.

Materials and methods Photomicrography Photomicrography methods  Upon courier delivery from China, a polythene bag containing 11 g of ancie ancient nt cannab cannabis is was steril steriliz ized ed with with ethano ethanol, l, handl handled ed with with

laboratory gloves in a laminar-flow hood, and transferred with a clean metal spatula (Fig. 2A). Two levels of light microscopy were used in this this study. study. For the the obser observat vation ionss on the achene acheness (Fig. (Fig. 2D), 2D), a low low power power Brunel Brunel MX3 microsco microscope pe (Chippenh (Chippenham, am, Wiltshire Wiltshire,, UK) was used and a  33 objectiv objectivee utilized utilized in conjunct conjunction ion with an Olympus Olympus SP350 SP350 8 megapi megapixel xel camer camera, a, stere stereo o inser insertt 30 mm lens lens tube, tube, and Photonic PL2000 – double arm cold light source. Greater magnification tion was requi required red for more more detail detailed ed observ observat atio ions ns of trich trichome omess (Fig. 2B, C): a high power stereo light microscope with a Trinocular  Head for camera camera attachme attachment nt (STE UK, Sittingb Sittingbourn ourne, e, Kent, Kent, UK) with an eye piece graticul graticulee for specimen specimen size measurem measurement ent fitted with 34, 310, and 340 objectives. The camera’s 33 optical zoom capability capability provided additional magnification. magnification. The observation observationss on the seed were made on unmounte unmounted d specispecimens. For these, small pieces of plant tissue were placed directly onto the the low-p low-powe owerr micros microsco cope pe plate plate.. When When using using the high high power  power  microscope, samples were dry mounted on a glass slide. To achieve views where large proportions of the material were simultaneously in focus, flat samples specimens (as shown) gave the greatest success. On the low power microscope the seed sample was illuminated with incident light, using a Photonic PL2000 – double arm ‘cold light source’ (Fig. 2D). Some samples, when placed on the high power power micros microscop cope, e, were were also also illumi illuminat nated ed using using the cold cold light  light  sour source ce.. Othe Others rs were were illu illumi mina nate ted d from from belo below. w. When When view viewin ing g samp sample less moun mounte ted d bene beneat ath h a cove coverr slip slip,, it is comm common on to set set up a microscope microscope using the Ko¨hler illuminatio illumination n method (Delly, 1988). This This ensu ensure red d that that ligh lightt from from the the cond conden ense serr lens lens was focuse focused d correctly correctly on the microscope slide. For uncovered specimens, specimens, the cond conden ense serr heig height ht and and aper apertu ture re were were adju adjust sted ed whil whilee view viewin ing g the subject until optimum resolution was achieved. In all cases, the specimens were measured using a graticule within the eyepiece. To enab enable le phot photog ogra raph phss to be take taken n thro throug ugh h the the low low powe power  r  microscope, one eyepiece was replaced with a compatible 30 mm

Photomicro icrogra graphs phs of ancient ancient cannabi cannabis. s. (A) Photog Photograph raph of the whole whole cannabi cannabiss sample sample being being transf transferre erred d in lamina laminarr flow hood. hood. (B) Fig. 2. Photom Photomicrograph of leaf fragment at low power displaying non-glandular and amber sessile glandular trichomes. Note retention of chlorophyll and green colour, scale bar  100 lm. (C) Higher power photomicrograph of a single sessile glandular trichome. At least 4 of its 8 secretory cells are clearly visible on the right, and the scar of attachment to the stype cells in the centre, scale bar  25 lm. (D) Low power photomicrograph of  a cannabis achene (‘seed’) including the base with a non-concave scar of attachment visible, scale bar  1 mm. ¼

¼

¼

4174 Russo  et al. lens tube to which single lens reflex or digital cameras would be atta attach ched ed.. As in ordin ordinary ary phot photog ogra raph phy, y, the the dept depth h of field field is considered to be the distance from the nearest object plain to the farth farthes estt obje object ct plai plain n that that is in focus focus.. When When obje object ctss are are a long long distance from the camera lens the depth of field is large. However, depth decreases as the image comes closer to the lens. When taking photom photomicr icrogra ographs, phs, depth depth of field field is measure measured d in micron micronss (Delly, (Delly, 1988). To maximize the chance of finding substantial areas of tissue simultaneously in focus within this narrow depth of field, multiple samples were laid as flat as possible onto glass slides. In all cases, photom photomicr icrogra ographs phs were were taken taken on a solid solid bench bench and the shutte shutter  r  activated remotely to reduce manually-induced camera-shake. In no instanc instancee was any image image modific modificati ation on techni technique que used used in these photographs.

Phytochemistry methods  Approximately 2 g of the dried plant material was extracted with 200 200 ml meth methan anol ol:c :chl hlor orofo oform rm (9:1 (9:1 v/v) v/v) by sonic sonicat atio ion n at room room temperature temperature (21 °C), the standa standard rd extrac extractiv tivee techni technique que for this this laboratory (GW Pharmaceuticals), a method that recruits >95% of  phytocannabi phytocannabinoid noid content. The solvent layer was then transferred through a paper filter into a rotary evaporator flask. The flask was evaporated to dryness at 40 °C, under under reduced reduced pressur pressure, e, prior prior to resuspension in 4 ml of methanol:dichloromethane (3:1 v/v). This sample was transferred to two autosampler vials to be analysed by GC-FID-MS and HPLC-UV. At all stages, the clean glassware was extracted with the same solvents to ensure that none of the observed peaks peaks would would be a result result of contam contamina inatio tion. n. GC-FID-M GC-FID-MS S analyse analysess were performed on a HP6890 gas chromatograph, chromatograph, coupled coupled to a 5975 inert mass spectrometer spectrometer.. The system was controlled with Agilent  MSD chemstation D.03.00.611. The GC was fitted with a Zebron fused fused silica silica capill capillary ary column column (30 m30.32 0.32 mm inner inner diamet diameter) er) coated with ZB-5 at a film thickness of 0.25 lm (Phenomenex). The oven temperature was programmed from 70 °C to 305 °C at a rate of  5 °C minÀ1. The injector port and the transfer line were maintained at 275 °C and 300 °C, respectively. Helium was used as the carrier  gas at a pressure of 55 kPa. The injection split ratio was 5:1. HPLC profiles were obtained using an Agilent 1100 series HPLC system controlled controlled by Chemstation Chemstation version A09.03 software. Cannabinoid Cannabinoid profiles were generated using a C18 (15034.6 mm, 5 lm) analytical column fitted with a C 18 (1034.6 mm, 5 lm) guard column. The mobile phase consisted of acetonitrile, 0.25% w/v acetic acid and methanol at a flow rate of 1.0 ml min À1 and the column was kept at  35 °C. The UV profiles were recorded at 220 nm.

Genetic methods  DNA was extract extracted ed from pulverized pulverized dried leaves, leaves, from two seeds seeds probably belonging to Cannabis spp., and from three seeds probably from other unidentified species. The DNeasy Plant Mini Kit (Qiagen) was was used used,, acco accord rdin ing g to the the Qiag Qiagen en prot protoc ocol ol,, but but with with some some modification to increase the final DNA amount and to avoid external and artificial contamination. For this reason, pre-PCR and post-PCR operatio operations ns were physical physically ly separate separated d and carried carried out in differen different  t  environm environments ents.. Ancient Ancient DNA extracti extraction on and other other pre-PCR pre-PCR works works were performed under a UV-filtered ventilation system and a positive pressure pressure airflow. airflow. Filtere Filtered d pipette pipette tips and sterile sterile tubes tubes and plastic plasticss were always used; gloves, masks, and laboratory coats were always worn. worn. The quali quality ty of DNA obtai obtaine ned d was estim estimate ated d by A260 /  A280 absorb absorban ance ce ratio ratio.. In order order to obtai obtain n the highes highestt possib possible le fideli fidelity ty during PCR synthesis, PCR reactions were performed using the Pwo Master Master readyready-to to-us -usee proofr proofread eadin ing g maste masterr mix (Roch (Rochee Applie Applied d Science Science)) accordin according g to their their protocol protocol.. The primers primers designed designed to test  DNA DNA inte integr grit ity y and and suit suitab abil ilit ity y for for PCR PCR anal analys ysis is and and spec specie iess identification were from the ITS region of nuclear ribosomal DNA (Blattner, 1999), and from a non-coding region of chloroplast DNA

(Taberlet  et al., 1991). The reaction mixtures were subjected firstly to an initial initial heat denatur denaturatio ation n at 94 °C for for 3 min; min; then, then, they they were were subjected to 35 cycles of heat denaturation at 94 °C for 30 s, 1 min of  primer annealing at 55 °C for the ITS region, and 50 °C for cpDNA, and DNA exten extensio sion n at 72 °C for 40 s. Fina Finally lly,, the samples samples were maintained at 72 °C for 5 min for the final extension. PCR reactions were performed performed in an MJ Research Research PTC-100 PTC-100 thermal thermal cycler (MJ Resear Research, ch, USA). USA). The amplifi amplificat catio ion n produc products ts were were separa separate ted d by electrophoresis in a 1.5% agarose gel. The bands were excised and purified with the MinElute Gel Extraction Kit (Qiagen). PCR-purified prod produc ucts ts were were quan quanti tifie fied d and and dire direct ctly ly forw forwar ardd- and and reve revers rseesequenced, using the GenomeLab ä Dye Terminator Cycle Sequencing with a Quick Start Kit on a CEQ8000 Genetic analyser (Beckman Coulter) Coulter).. Primer Primer sequenc sequences es were identifi identified ed and removed removed manuall manually, y, and database searches were performed with the BLASTN algorithm al., 1990). (Altschul et al. 1990). The The sequen sequences ces resul results ts proved proved that that the the pulverize pulverized d dried dried tissue tissue was from Cannabis sativa L., despite our  observati observation on in the mixed sample of some small seeds of different  different  species, species, removed removed before before the DNA extracti extraction; on; no differen differences ces were observed between the sequences obtained and those deposited at the NCBI NCBI genegene-ban bank k (for (for THC THCAA-and and CBDACBDA-syn syntha thases ses,, GeneBa GeneBank nk accession numbers E55108/GI 18529739 and E33091/GI 18623981). By contrast, no amplification was obtained from DNA extracted from seeds of both cannabis and the other, unidentified species. The allelic status at a single locus, B, known to be the major gene determining al., 2003 the CBD/TH CBD/THC C ratio ratio in canna cannabi biss (de Meije Meijer  r  et al. 2003), ), was was investigated in the ancient material. The primer pairs described (de Meijer  et al., 2003) are not sufficiently associated with the chemotype (Pacifico et al., 2006), 2006), and the sequence sequence-bas -based ed primers primers describe described d al., 2006) therein (Pacifico (Pacifico et al. 2006) faile failed d to yield yield any ampli amplific ficati ation on,, probably due to the limited integrity of DNA from ancient cannabis tissues, which did not sustain the amplification of a 1100 Da DNA fragment fragment.. Therefor Therefore, e, three three different different primer primer pairs pairs (Fw150 (Fw1503Rev328, Fw1663Rev318, Rev318, and Fw154 Fw1543Rev318) Rev318) were used. These These primers primers were designed designed on two conserved conserved small small regions regions of a zone varying varying between the known sequences of THC and CBD alleles. When tested on fresh cannabis tissues, these primers were demonstrated to be able to amplify both alleles (PCR and sequences data not shown). Using differ differen entt prime primerr pair pair combi combina natio tions, ns, the risk risk of a no-ma no-match tch or  a mismatch because of possible mutations in the 3 end of primer  region was overcome. The primer sequences are listed in Supplementary Fig. S8 at  JXB online. All reaction mixtures were subjected first  to heat heat denatu denaturat ratio ion n at 94 °C for for 3 min min and and then then to 35 cycle cycless consisting of heat denaturation at 94 °C for 15 s, primer annealing at  54 °C for 30 s, and DNA extension at 72 °C for 1 min. Finally, the samples were maintained at 72 °C for 5 min for the final extension of  DNA. PCR products products were separate separated d by electro electrophor phoresis esis in a 1.5% 1.5% agarose gel. The bands were excised and purified with MinElute Gel Extraction Kit (Qiagen). PCR-purified products were quantified and directly sequenced in forward and reverse, using the GenomeLabä Dye Dye Term Termin inat ator or Cycl Cyclee Sequ Sequen enci cing ng with with Quic Quick k Star Startt Kit Kit on a CEQ8000 Genetic analyser (Beckman Coulter). #

Results Microscopic botanical analysis  Gross examination of the 11 g sample of cannabis provided by the Chinese Chinese Academy Academy of Science Sciencess reveale revealed d loose loose dry veget vegetat ativ ivee mater materia iall (Fig. (Fig. 2A). 2A). The The impre impress ssio ion n that that the the vegeta vegetativ tivee materia materiall had been been lightly lightly pounded pounded was supsupported by examination of the wooden bowl, whose internal surface was worn smooth, apparently from use as a mortar  (see Supplementary Fig. S5B at  JXB online). The cannabis

Ancient cannabis  4175

retained a surprisingly green colour in its leafy parts and disp displa laye yed d visib visible le glan glandul dular ar tric tricho homes mes (Fig (Fig.. 2B), 2B), the the phytochemical factory of the plant and site of manufacture of cannabinoid cannabinoidss and terpenoids (Potter, 2004; McPartland McPartland and Russo, 2001; Kim and Mahlberg, 2003). However, the ancient ancient sample lacked the typical typical cannabis cannabis odour. Microscopic examination confirmed the presence of intact sessile tric tricho homes mes with with an amber amber tint tint (Fig (Fig.. 2B), 2B), whil whilee high higher  er  resolu resolutio tion n docume documente nted d the retenti retention on of visibl visiblee secreto secretory ry cells cells within within the trichom trichomes es (Fig. (Fig. 2C). Achenes Achenes (‘seeds (‘seeds’) ’) averaged 2.2–3.6 mm in length (Jiang et al., 2006), were ligh lightt in colo colour ur with with some some stria striatio tions ns,, but but demon demonstr strat ated ed rough, non-concave fruit attachment (Fig. 2D), all traits of  domestication (Schlumbaum et al., 2008) 2008) associ associated ated with cultiv cultivate ated d cannab cannabis is strain strainss (Vavilo (Vavilov, v, 1926). 1926). In contras contrast, t, achene acheness of wild strains strains are typical typically ly smaller smaller and darker  darker  with concave attachment attachment zones that favour shattering and easy spread (Vavilov, 1926). Germination was attempted attempted with with 100 100 ache achene ness in compo compost, st, but but no emer emergen gence ce was observed after 21 d.

Phytochemical analysis  Phytoche Phytochemical mical and genetic genetic teams were initially initially blinded to one one anot anothe her’ r’ss resu result lts. s. The The extr extrac acti tion on of 2 g of plan plant  t 

material produced 67.9 mg of solids after the removal of  solvents. Using high performance liquid chromatography (HPLC) (HPLC),, the larges largestt cannab cannabino inoid id peak peak was cannab cannabino inoll (CBN) at 7.4 min, but concentration levels were very low, averaging 0.007% w/w. CBN is an oxidative breakdown product THC, generated non-enzymatically, with increasing ing age age (Bre (Brenn nnei eise sen, n, 2007 2007). ). Ther Theree were were also also peak peakss corresponding to expected retention times for cannabidiol (CBD) at 4.9 min and cannabichromene (CBC) at 12 min (Fig (Fig.. 3). 3). Both Both are are phyt phytoc ocan anna nabi bino noid idss resu result ltin ing g from from alternativ alternativee enzymatic enzymatic pathways than that yielding THC (de Meijer  et al., 2003). There were very few peaks in the first 20 min of the gas chromatogram where mono- and sesq sesqui uite terp rpen enes es elut elutee (Fig (Fig.. 4). 4). This This lack lack of terp terpen enoi oid d volatiles supports the physical observation that the plant  material material lacked lacked the herbal herbal smell traditiona traditionally lly associate associated d with cannabis cannabis (McPartlan (McPartland d and Russo, 2001). 2001). Shown in Fig. Fig. 5A–C, 5A–C, (and (and in Supple Supplemen mentar tary y Fig. Fig. S7A, S7A, B at  JXB onli online ne)) are are brea breakd kdow owns ns of subsub-re regi gion onss of the the gas gas chro chroma mato togr gram am.. The The majo majorr peak peakss in the the 13–3 13–30. 0.5 5 min min region are free fatty acids (see Supplementary Fig.S7A at   JXB online). The largest peak identified as palmitic acid was the most abundant in the sample. Methyl and propyl cann cannab abin inoi oids ds elut eluted ed in the the 27–3 27–30 0 min min regi region on and and the the

Complete high performance liquid chromatography chromatography (HPLC) of ancient cannabis. Fig. 3. Complete

Complete gas chromatography chromatography-flame -flame ionization detection (GC-FID) of ancient cannabis. Fig. 4. Complete

4176 Russo  et al.

Fig. 5. Gas chromatography of ancient cannabis subsections. (A) GC of the 30–34 min region demonstrates several phytocannabinoids: cannabidiol (CBD), cannabichromene (CBC), cannabicyclol (CBL), and cannabinavarin (CBNV). (B) GC of the 34–36.3 min region displays the highest peak, cannabinol (CBN), the direct non-enzymatic oxidative metabolite of THC, with possible cannabielsoin (CBE) at 34.2 min. (C) GC of the 36.3–40.5 min region displays cannabitriol (CBO) a THC degradant, and CBN variants (see text).

peaks marked as 286 Da and 302 Da all had MS spectra  consis consisten tentt with with propyl propyl cannab cannabino inoids ids.. There There were were two phthal phthalate ate peaks peaks at approx approxima imatel tely y 23.5 23.5 min (belie (believed ved to have have origin originate ated d from from the polyth polythene ene bags in which which the

samples were supplied). A number of phytocannabinoids were were iden identi tifie fied d in the the 30–3 30–34 4 min min regi region on (Fig (Fig.. 5A) 5A) including including cannabidi cannabidiol ol (CBD), cannabich cannabichromen romenee (CBC), cannab cannabicy icyclo cloll (CBL, (CBL, a heat-g heat-gene enerat rated ed artefa artefact ct of CBC

Ancient cannabis  4177

(Brenneisen, 2007), and cannabinavarin (CBNV, a propyl analogue of CBN). In the 34–36.3 min region (Fig. 5B), apar apartt from from cann cannab abin inol ol (CBN) (CBN),, the the larg larges estt indi indivi vidu dual al phytoc phytocann annabi abinoi noid d compon component ent,, there there were were at least least four  four  peaks peaks of 330 Da with with cannab cannabiel ielsoi soin n (CBE, (CBE, an artefa artefact  ct  derived from CBD (Brenneisen, 2007) a likely identification of the peak at 34.2 min. In the 36.3–40.5 min region (Fig. 5C), the known THC degradant cannabitriol (CBO) (Brenneisen, 2007) was seen, as well as a series of peaks with with spec spectr tral al simi simila lari riti ties es to CBN, CBN, thre threee of whic which h are are tent tentat ativ ivel ely y iden identi tifie fied d by the the NIST NIST data databa base se as eith either  er  hydroxyl- or oxo-CBN. The last region (42–50 min; see Supp Supple leme ment ntar ary y Fig. Fig. S7B S7B at  JXB online), online), contained contained phytoster phytosterols ols and triterpene triterpene alcohols alcohols with beta-sito beta-sitostero steroll the most abundant compound. Mass Mass spec spectr traa (MS) (MS) of sele select cted ed phyt phytoc ocan anna nabi bino noid idss corresponding to the above are displayed (Fig. 6). Values are all all in agre greemen ementt with with thos thosee in NIST NIST and and GW Pharmaceutical databases. Ther Theree was was a very very smal smalll peak peak dete detect cted ed at the the corr correc ect  t  rete retent ntio ion n time time in the the samp sample le for for TH THC, C, but but the the spec spectr tra  a  could not confirm its identity.

Genetic analysis  Beca Becaus usee of the the uniq unique ue degr degree ee of pres preser erva vati tion on of the the cannab cannabis, is, a geneti geneticc analys analysis is was undert undertake aken. n. The two ancient DNA sequences determined were labelled China  F and China F(h). Alignment Alignment of these paleo-sequenc paleo-sequences es (excluding the primers’ region, in red in Fig. 7A) with the presently available databases demonstrated:

(i) China China F (Fig. (Fig. 7A) is identi identical cal 134/13 134/134 4 nucleo nucleotid tidee agreem agreement ent to other other deposi deposited ted sequen sequences ces:: AB2128 AB212841, 41, AB212839, AB212839, AB212836, AB212836, AB212833, AB212830, AB212830, all belonging to tetrahydrocannabinolic acid synthases (THCAsynthases), a species-specific genetic region (Schlumbaum et al., 2008) from Cannabis sativa L. (ii) China F(h) (Fig. 7A) is a new variant, variant, not previously previously presen presentt in the geneti geneticc databa databases ses (submi (submitte tted d to NCBI, NCBI, GenBank accession number EU839988), showing a maximum identi identity ty of 132/13 132/134 4 nucleo nucleotid tides es with with the aboveabovementioned sequences and with China F. Utiliz Utilizing ing BLA BLASTX STX,, i.e. i.e. perfor performin ming g search searches es throug through h amino acid translation, it was again shown that the China  F(h) amino acid sequence is not registered in the database, and this is an obvious but necessary confirmation of the origin originali ality ty of this this varian variantt of the THCA syntha synthase se allele allele.. These results also prove that both sequences encode for  THCA synthase, the biosynthetic enzyme for THCA that  decarboxy decarboxylates lates via heat or ageing ageing to yield yield psychoact psychoactive ive THC (Russo, 2007). Direct comparison of the two ancient  sequences sequences,, identified identified the nature nature of the small differences differences observed: the samples have two ‘mutations’ (highlighted

in yellow yellow in Fig. Fig. 7A), 7A), which which can be consid considere ered d transtransversions: from guanine to cytosine, and from cytosine to adenine. The first of these two nucleotide substitutions is syno synony nymo mous us,, i.e. i.e. it does does not not chan change ge the the amin amino o acid acid sequen sequence, ce, while while the second second one is a non-sy non-synon nonymo ymous us subs substi titu tuti tion on,, lead leadin ing g to a seri serine ne-t -thr hreo eoni nine ne exch exchan ange ge (hig (highl hlig ight hted ed in ligh lightt blue blue in Fig. Fig. 7B) 7B) in the the enco encode ded d amino amino acid acid sequen sequence; ce; these these two amino amino acids, acids, howeve however, r, have have simila similarr physic physico-c o-chem hemica icall proper propertie ties. s. No CBDA syntha synthase, se, the biosyn biosynthe thetic tic enzyme enzyme for CBD (de Meijer  Meijer  et al., 2003), was identified in the sample.

Discussion The The resu result ltss pres presen ente ted d coll collec ecti tive vely ly poin pointt to the the most  most  prob probab able le conc conclu lusi sion on whic which h is that that the the Gush u sh culture  ı ı  cultivate cultivated d cannabis cannabis for pharmaceut pharmaceutical, ical, psychoact psychoactive ive or  divinatory purposes. In examining the botanical evidence from from this this ‘old ‘old and cold’ cold’ site site with with its unique unique degree degree of  pres preser erva vati tion on,, the the cann cannab abis is cons consis iste ted d of a proc proces esse sed d (pounded) sample whose seed size, colour, and morphology, at least according to principles of Vavilov (Vavilov, 1926), 1926), sugges suggestt that that it was cultiv cultivate ated d rather rather than than merely merely gather gathered ed from from wild wild plants plants.. The consid considera erable ble amount amount of  cann cannab abis is pres presen entt (789 (789 g) with withou outt any any larg largee stal stalks ks or  branches would logically imply a pooled collection rather  than than one from from a single single plant. plant. Import Important antly, ly, no obviou obviouss male male cannab cannabis is plant plant parts parts (e.g. (e.g. stamin staminate ate flowers flowers,, not  infrequently observed in Indian herbal cannabis, or  bhang (Russo, (Russo, 2007) 2007) were evident, implying their exclusion exclusion or  possib possible le remova removall by human human interv intervent ention ion,, as these these are pharmacologically less psychoactive. The HPLC, GC, and MS analyses confirm the identity of the supplied plant sample as Cannabis sativa L. The predomina predominance nce of CBN indicates indicates that the original plants 9 contained D -tetrahyd -tetrahydrocan rocannabin nabinol ol (THC) as the major  phytocannabinoid constituent. The presence of CBO and numerous numerous CBN-related CBN-related substance substance peaks peaks further further supports supports this this view view.. CBD CBD and and CBC, CBC, toge togeth ther er with with thei theirr know known n thermo thermo-ox -oxida idativ tivee degrad degradati ation on produc products ts CBE and CBL (Bre (Brenn nnei eise sen, n, 2007 2007), ), are are pres presen ent, t, but but the the GC anal analys ysis is would would appear appear to indica indicate te that, that, in both both cases, cases, CBC and CBL are represented in greater quantities, as expected in a high-THC cannabis strain wherein CBD is only a minor  component. In addition, there is a peak for CBNV which confirms that the plant also contained D9-tetrahydrocannabivarin (THCV), a propyl phytocannabinoid. All of these observations are consistent with strains of cannabis with a high high TH THC C cont conten entt and and in an alte altern rnat ativ ivee taxo taxono nomy my sugg sugges ests ts it shou should ld be assi assign gned ed to Cannabis Cannabis indica indica Lamarck (Hillig and Mahlberg, 2004). While While chroma chromatog tograp raphy hy elutio elution n times times may vary vary with with temperature, column type, and other factors, confirmation was evident evident with corrobora corroboratory tory mass spectra spectra values values that 

4178 Russo  et al.

Fig. 6. Mass spectra of ancient cannabis. Subsections Subsections demonstrate demonstrate the phytocannabino phytocannabinoids ids cannabinol cannabinol (CBN), (CBN), cannabidiol cannabidiol (CBD), (CBD), cannabicyclol cannabicyclol (CBL), cannabinolivarin (CBNV), cannabichromene (CBC), cannabielsoin (CBE), 1 -oxcannabinol, -oxcannabinol, and 1 -hydroxycannabinol. #

were identical to those seen daily in assays performed on fresh cannabis extracts in this laboratory. The The pres presen ence ce of so many many reco recogn gniz ized ed cann cannab abin inoi oid d degrad degradants ants is consis consisten tentt with very very old cannab cannabis is sample samples. s. The very low concentration levels measured in the HPLC analysis analysis may indicate that the sample provided provided contained signi signific fican antl tly y more more leaf leaf and and twig twig mater materia iall than than flower  flower  material, rather than being evidence in itself that the sample was of low low pote potenc ncy y origin original ally ly.. This This plan plantt mate materia riall is therefore conclusively cannabis derived from a population

#

of plants within which THC was the dominant cannabinoid. By cont contra rast st,, a samp sample le take taken n from from a mix mix of wild wild-t -typ ypee Cannabis sativa would customarily harbour a more equal mixture of THC and CBD (de Meijer  et al., 2003). It would appear, appear, therefo therefore, re, that that humans humans select selected ed the material material from from plan plants ts on the the basi basiss of thei theirr high higher er than than aver averag agee TH THC C conte content nt.. To elab elabor orat ate, e, a chem chemot otax axon onom omy y of cann cannab abis is prev previo ious usly ly outl outlin ined ed indic indicat ates es thre threee types types (Smal (Smalll and and Beckst Beckstead ead,, 1973): 1973): chemot chemotype ype I (drug) (drug) strain strainss with highhighTHC:CBD THC :CBD ratios ratios,, chemoty chemotype pe II low-THC low-THC,, higher higher-CBD -CBD

Ancient cannabis  4179

Fig. 7. DNA analysis of ancient cannabis. (A) Nucleotide sequences of the wild-type tetrahydrocannabinolic acid synthase, China F, and the mutant  sequence, China F(h), with two single nucleotide polymorphisms highlighted in lower case yellow. (B) Amino acid translation of China F and China  F(h), demonstrating divergence in a change from serine (wild-type) to threonine (mutant), highlighted in blue.

(fibre) (fibre) strain strains, s, and chemot chemotype ype II with with more equal ratios ratios.. THC and CBD produc productio tion n are mediate mediated d by co-dom co-domina inant  nt  alleles BT and BD, respectively (de Meijer  et al., 2003). By comp compar aris ison on,, pool pooled ed samp sample less from from cann cannab abis is field fieldss in Moro Morocc cco o and and Afgha Afghani nista stan n will will norm normal ally ly produ produce ce 25% 25% high-T high-THC HC plants plants,, 25% high-CB high-CBD D plants plants,, and 50% with lower, mixed titres, combining to yield roughly equivalent  amount amountss of the two phytoca phytocanna nnabin binoid oidss (Russo, (Russo, 2007), 2007), a pattern not observed in our specimen. Isotopic analysis of cellulose from this cannabis sample migh mightt conc concei eiva vabl bly y be used used in comp compar aris ison on with with othe other  r  samples in an attempt to establish its geographic origin. While While multimulti-pur purpos posee cannab cannabis is plants plants used used simult simultaaneously for food (seed), fibre (stalks), and pharmaceutical uses uses (flower (flowering ing tops) tops) have have been been recent recently ly report reported ed from from Darc Darchu hula la in far far west wester ern n Nepa Nepall (Cla (Clark rke, e, 2007 2007), ), more more customarily, a given plant is best suited toward a single purpose. Of additional key importance is the absence of  hemp hemp arte artefa fact ctss from from the the Yang Yangha haii Tomb Tombs. s. The The Gush ushı ı  fabric fabricate ated d clothi clothing ng from from wool wool (see (see Supple Supplemen mentar tary y Fig. Fig. S6B at  JXB online) and ropes from Phragmites (reed) spp. fibre fibress (see (see Supp Supple leme ment ntar ary y Fig. Fig. S6C S6C at  JXB online). Wher Wherea eass hemp hemp text textil iles es have have been been coll collec ecte ted d from from the the Northern China Yangshao Culture from 6000–7000 years theirr appe appear aran ance ce in the the west west was was not not docu docume ment nted ed BP, thei before before 2000 2000 years years BP, for for exam exampl ple, e, 1500 1500 year yearss BP in Kucha, 600 km west of Turpan (Mallory and Mair, 2000). Previo Previous us phytoc phytochem hemica icall analys analyses es of antiq antique ue cannab cannabis is preparatio preparations ns have demonstrate demonstrated d THC remnant remnant fingerprints fingerprints

from from 19th 19th centur century y cannab cannabis is prepar preparat ation ionss (Harve (Harvey, y, 1990) 1990) incl includ udin ing g a 140140-ye year ar-o -old ld samp sample le of Squir Squire’ e’ss Extr Extrac act  t  (Harvey, 1985). A study in 1992 reported the presence of  D8-THC (previously termed D6-THC) from burned cannabis that that was was repor reporte tedl dly y inha inhaled led as an aide aide to chil childb dbir irth th in a Judean cave 1700 years BP (Zias et al., 1993), supported by the finding of cannabinoid residues in an adjacent glass vess vessel el (Zia (Zias, s, 1995 1995). ). In the the Musta Mustang ng regio region n of Nepa Nepal, l, mummified human remains of probable Mongolian ancestry have been dated 2200–2500 2200–2500 years BP in associati association on with cannabis, cannabis, probably probably transporte transported d from elsewhere elsewhere (Kno ¨ rzer, 2000; 2000; Alt  et al. al., 2003 2003), ), but but with with insu insuffi ffici cien entt deta detail il to asce ascert rtai ain n its its use. use. Ruden Rudenko ko reco recove vere red d cann cannab abis is seed seeds, s, censer censers, s, and hempen hempen clothi clothing ng in Pazyry Pazyryk, k, Siberia Siberia from from Scythian kurgans (burial mounds) from 2400–2500 years BP (Rudenko, 1970; Brooks, 1998), closely matching Herodotus’ descriptions of funeral rites for that culture (Herodotus, 1998). Sarianidi also claimed cannabis use in the Bactria–  Margiana Margiana Archaeolo Archaeological gical Complex (BMAC) (present (present day Turkmenis Turkmenistan) tan) (Sarianidi (Sarianidi,, 1994, 1994, 1998), 1998), but this interpreta interpreta-tion has been debated (see discussion in Russo, 2007). Another Another independen independentt genetic genetic analysis analysis of this material published published subsequent subsequent to our analysis analysis (Mukherjee (Mukherjee et al., 2008) confirmed the presence of THCA synthase, but not  the single nucleotide polymorphisms. The authors posited a European–Siberian origin for the material. Current genetic data also confirm that the plant material examined examined is Cannabis accord rdin ing g to ITS ITS and and Cannabis sativa sativa L. acco cpDNA analysis. The results also support the hypothesis

4180 Russo  et al.

of the existence of at least two THCA-synthase nucleotide sequences in the ancient plant material examined. One of  these these sequen sequences ces perfec perfectly tly matche matchess the corres correspon pondin ding g sequence of already-known THCA-synthases deposited in GenBank, both as gene and protein sequences; the second sequ sequen ence ce is a nove novell one, one, with with two two sing single le nucl nucleo eoti tide de poly polymo morp rphi hism smss (SNP (SNPs) s) enco encodi ding ng for for a prot protei ein n with with presum presumabl ably y very very simila similarr charac character terist istics ics.. Whethe Whetherr these these two sequen sequences ces coexiste coexisted d in a single single cannab cannabis is plant plant or  a strain heterozygous at the B locus, or belong to different  plants, could not be concluded. THC repres represent entss one of the possib possible le phytoca phytocanna nnabin binoid oid end-products manufactured by cannabis plants; THC (or, in its native form, THCA) is synthesized by a well-characterized enzyme (THCA- or THC-synthase) from a precursor  (CBG (CBG or CBGA CBGA)) comm common on to most most chem chemot otyp ypes es that  that  represents the metabolic ‘switching point’, downstream of  whic which h the the vari variab abil ilit ity y of the the diff differ eren entt chem chemot otyp ypes es is conce concent ntra rate ted. d. The The agen agents ts of such such varia variabi bili lity ty foun found d in cannabis germplasm are exclusively the different synthases, among which THC(A)-synthase is the only one responsible for making that specific cannabinoid, THC. Therefore, the presen presence ce of the allelic allelic varian variantt respon responsib sible le for coding coding the THC(A)-synthase may well be considered to be diagnostic, or at least strongly suggestive of a THC-producing plant. The fossil cannabis plants found were therefore genetically equipped to produce THC. How much THC they actually produ produce ced, d, cann cannot ot of cour course se be spec specifi ified ed beca becaus usee they they depend depend on a number number of anatom anatomical ical,, enviro environme nmental ntal,, and nutritional nutritional factors that remain unknown. unknown. Numerous questions remain. Current data do not permit  it to be ascertained how the cannabis from the tomb was administered. If used orally, perhaps it was combined in some fashion with Capparis spinosa L., as these these plants plants were found together in a nearby but later tomb at Yanghai (Jiang et al., 2007). That date for that tomb was initially reported as 2700 years BP via radiocarbon methods, and since since correc corrected ted to 2200–2 2200–2400 400 years years BP with additiona additionall calibration employing tree ring data. If this cannabis were smoked or inhaled, no mechanism for so doing has been excava excavated ted in the area. The Gush could have have sifted sifted the ushı ı  could  cannabis cannabis through fabric fabric after pounding, pounding, then fumigated it, much as described for the alleged cannabis candidate, the Sumerian A.ZAL.LA, administered medicinally for ‘hand of ghos ghost’ t’(T (Tho homp mpso son, n, 1923 1923,, 1949 1949), ), sinc sincee posi posite ted d as nocturnal nocturnal epilepsy (Russo, 2007; 2007; Wilson Wilson and Reynolds, 1990). 1990). While While this this cultur culturee could could have have arrive arrived d from from the earlier earlier BMAC region region as ‘oasis ‘oasis hoppers’ hoppers’ (Barber, 1999), and and cert certai ain n cult cultur ural al rela relati tion onsh ship ipss are are appa appare rent nt to the the Scyt Scythi hian an cult cultur uree with with resp respec ectt to cann cannab abis is use use and and equestrian equestrian prowess, those those peoples peoples were Iranian Iranian speakers speakers (Mallo (Mallory ry and Mair, Mair, 2000). 2000). In additi addition, on, Gush ushı ı  cultural affinities and burial practices much more closely resemble those of the presumed proto-Tocharian speaking, incenseburn burnin ing g (Kuz (Kuzmi mina na,, 1998 1998)) Afana Afanasi siev evo o peop people less in the the

Yenis Yenisei ei Vall Valley ey to the the nort north h (Anth (Anthon ony, y, 1998 1998,, 2007 2007;; Mallory, 1998; Renfrew, 1998; Mallory and Mair, 2000), whose whose putati putative ve southw southward ard migrat migration ion some some author authoriti ities es have attributed to ‘global cooling’ c. 4000 years BP (Hsu (H su¨ , 1998), 1998), and to their proto-Ind proto-Indo-Eu o-Europe ropean-sp an-speakin eaking g Yamnaya naya fore forebe bear arss furt furthe herr west west,, dati dating ng to 6000 6000 year yearss BP (Mallory, 1989; Anthony, 1998; Winter, 1998). Abundant  myster mysteries ies remain remain as to the origins origins and customs customs of the Gush Additi tion onal al answ answer erss may may accr accrue ue from from futu future re ushı ı . Addi archaeological excavations or human genetic analyses that  elucid elucidate ate relati relations onship hipss with with other other ancien ancientt cultur cultures es and modern peoples of the region. The unique SNPs discovered ered in this this anci ancien entt samp sample le may may yet yet be of crit critic ical al impor importa tanc ncee in trac tracin ing g the the phyl phylog ogen eny y and and geog geogra raph phic ic spread of cannabis and the humans who used it. The excellent preservation of the cannabis from this tomb allowe allowed d an unpr unprec eced eden ented ted leve levell of moder modern n bota botani nica call investig investigatio ation n through through bioche biochemist mistry ry and geneti genetics cs to conconclud cludee that that the the plan plantt was was cult cultiv ivat ated ed for for psyc psycho hoac acti tive ve purpo purpose ses. s. Whil Whilee cult cultiv ivat atio ion n of hemp hemp for for fibre fibre has has been been docum documen ente ted d in East Easter ern n China China from from a much much earli earlier er date date (vide supra Mallory and Mair, 2000), the current findings represent the most compelling physical evidence to date for  the use of cannabis for its medicinal or mystical attributes.

Supplementary data Photog Photograp raphs hs and diagra diagrams ms of the Yangha Yanghaii Tombs Tombs site, site, Tomb Tomb M90 M90 cont conten ents ts incl includ udin ing g fabr fabric ic and and rope ropes, s, and and additiona additionall chromatogr chromatographi aphicc and genetic genetic analysis analysis primer  sequence information are presented in Supplementary Figs S1–S8, available online. Fig. S1. Study site at the Yanghai tombs with Huoyan Shan mountain range in background (photo EBR). Fig. S2. Diagrams of the Yanghai Tombs (adapted from Xinjia Xinjiang ng Instit Institute ute of Cultur Cultural al Relics Relics and Archae Archaeolo ology, gy, 2004, with permission). shaman’s ’s tomb, tomb, M90 [previ [previous ously ly pubpubFig. Fig. S3. The shaman lished in Mandarin (Xinjiang Instgitute of Cultural Relics and Archaeology, 2004), used with permission]. Fig. S4. The shaman’s skull (photos EBR). Container nerss in which which cannab cannabis is was stored stored in Fig. Fig. S5. Contai tomb tomb [previ [previous ously ly publis published hed in Mandar Mandarin in (Xinji (Xinjiang ang Institut stitutee of Cultur Cultural al Relics Relics and Archae Archaeolo ology, gy, 2004] 2004] used used with permission. Re-exc xcav avat atio ion n of Tomb Tomb M90. M90. This This was was Fig. Fig. S6. S6. Re-e undert undertake aken n to re-exa re-examin minee artefa artefacts cts,, measur measuree GPS coordina ordinates tes,, and assess assess enviro environme nment nt condit condition ionss (photo (photoss EBR). Fig. S7. Chromatography subsections from phytochemical analysis. Primer sequen sequences ces employ employed ed in the geneti geneticc Fig. Fig. S8. Primer analysis analysis to amplify amplify THC- and CBD-allele CBD-allele specific fragments and their sequences sequences (5 3 ). #/

#

Ancient cannabis  4181

Acknowledgements The authors are grateful to the Chinese Academy of Sciences and GW Pharma Pharmaceu ceutic ticals als for support support of the projec project. t. Kim Laught Laughton on facili facilitat tated ed commun communica icatio tion n and logist logistics ics between between the Chines Chinesee auth authori oriti ties es and and the the Briti British sh Home Home Office Office for expo export rtat atio ion n and and import importati ation on of the ancien ancientt cannab cannabis. is. Daniel Daniel Adams, Adams, Laura-J Laura-Jane ane Everit Everitt, t, and Helen Helen Keogh Keogh perform performed ed phytoc phytochem hemica icall analyt analytica icall preparation, preparation, supervised supervised by Peter Gibson. Ying Li provided translation and logistical support to EBR during field work in Xinjiang. Gregory Gerdeman is thanked for his helpful review of the article, as are the anonymous anonymous reviewers reviewers for their suggestions. suggestions. No competing competing financial interests were operative in this study. Author contributions : EBR proposed and co-ordinated the current  study, study, engaged engaged in field field work, work, and wrote the article article drafts. drafts. HEJ performed performed the majority majority of the background research and was actively actively engaged engaged in current investigations investigations.. AS performed performed the phytochemical phytochemical analysis and wrote the pertinent methods and results sections. AC, FDB, FDB, and and GM perf perfor orme med d the the gene geneti ticc anal analysi ysiss and and wrote wrote the the pert pertin inen entt meth method odss and and resu result ltss sect sectio ions ns.. DJP DJP co-o co-ord rdin inat ated ed the handling handling of the ancient ancient cannab cannabis is in the UK, perfor performed med the microphotog microphotography, raphy, and wrote the pertinent pertinent methods and results. EGL, XL, DKF, FH, YBZ, YFW, LCZ, and CJL were all engaged in earlier investigations in relation to this study. YXZ analysed the phytochemistry of the cannabis sample and SB and his colleagues analysed analysed the phytochemist phytochemistry ry and genetics genetics of the cannabis cannabis sample inde indepe pend nden entl tly. y. CSL CSL conc concei eive ved d the the conc concep eptt of stud studyi ying ng the the archaeologi archaeological cal cannabis cannabis samples samples by multidiscipl multidisciplinary inary methods, and organized, organized, co-ordinated co-ordinated and supervised supervised all aspects aspects of the current  study and its performance. performance.

References Selected treasures of the Turfan relics. Academia Turfanica. Turfanica. 2006. Selected Turpan, China: Academia Turfanica. Alt KW, Burger J, Simons A, et al . 2003. Climbing into the past: Journall of  first Himala Himalayan yan mummie mummiess discov discovere ered d in Nepal. Nepal. Journa   Archaeological Science 30, 1529–1535. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment alignment search tool. Journal of Moleculr Biology 215, 403–410. An Z. 2008. Cultural complexes of the Bronze Age in the Tarim Basin and surrounding areas. In: Mair VH, ed. The Bronze Age and Early Iron Age people peopless of Eastern Eastern Central Central Asia, Vol. Vol. I. Washington, DC: Institute for the Study of Man, 45–62. Anthony DW. DW. 1998. The opening of the Eurasian steppe at 2000 Bronze Age Age and and Earl Earlyy Iron Iron Age Age BCE. BCE. In: Mair Mair VH, ed. The Bronze   peopl peoples es of Eastern Eastern Centra Centrall Asia Asia, Vol. Vol. I. Wash Washin ingt gton on,, DC: Institute for the Study of Man, 94–113. wheel, and langua language: ge: how Anthony DW. DW. 2007. The horse, the wheel,  Bronze-Age riders from the Eurasian steppes shaped the modern world . Princeton, Princeton, NJ; Oxford: Princeton Princeton University University Press. ¨ ru¨ mchi mc hi . New York: WW Barber EJW. 1999. The mummies of U  Norton & Company. Blattner Blattner FR. 1999. Direct amplification amplification of the entire ITS region from poorly poorly preserv preserved ed plant plant materi material al using using recomb recombina inant nt PCR. PCR.  Biotechniques 27, 1180–1186. analysis of phytocannabi phytocannabinoids noids Brenneisen R. 2007. Chemistry and analysis and other  Cannabis constituents. In: Elsohly M, ed. Marijuana and the cannabinoids. Totowa, NY: Humana Press, 17–49. Brooks EB. 1998. Textual evidence for 04c Sino-Bactrian contact. In: Mair VH, ed. The Bronze Age and Early Iron Age peoples of   Eastern Central Asia, Vol. II. Washington, DC: Institute for the Study of Man, 716–726.

Chen K-T, Hiebert FT. 1995. The late prehistory of Xinjiang in relation to its neighbors. Journal of World Prehistory 9, 243–300. Clarke Clarke RC. 2007. Traditional Traditional Cannabis cultivation cultivation in Darchula  Darchula  Journal of Industrial  Industrial  Distri District, ct, Nepal: Nepal: seed, seed, resin resin and textil textiles. es. Journal  Hemp 12, 19–42. Davis-Kimball J. 1998. Tribal interactions between the Early Iron Age nomads nomads of the souther southern n Ural Ural steppe steppes, s, Semire Semirechi chiye, ye, and Xinjiang. In: Mair VH, ed. The Bronze Age and Early Iron Age   peo peopl ples es of East Eastern ern Cent Central ral Asia Asia, Vol. Vol. I. Wash Washin ingt gton on,, DC: Institute for the Study of Man, 238–263. de Meijer EP, Bagatta M, Carboni A, Crucitti P, Moliterni VM, 2003. The inheri inheritan tance ce of chemic chemical al Ranalli Ranalli P, Mandolino Mandolino G. 2003. phenotype in Cannabis sativa L. Genetics 163, 335–346. Delly JG. 1988. Photography through the microscope. Rochester, NY: Eastman Kodak. Harvey DJ. 1985. Examination of a 140 year old ethanolic extract  of  Cannabis: identification of new cannabitriol homologues and the ethyl homolo homologue gue of cannab cannabino inol. l. In: Harvey Harvey DJ, Paton Paton W, Marihuana ‘84. Proc Naha Nahass GG, eds. eds. Marihuana Procee eedi ding ngss of the the Oxford Oxford symposium on Cannabis. Oxford, UK: IRL Press, 23–30. Harvey Harvey DJ. 1990. 1990. Stabil Stability ity of cannab cannabino inoids ids in dried dried sample sampless of  Journa Jou rnal l of EthnoEthnocannab cannabis is dating dating from around around 1896–19 1896–1905. 05.  pharmacology 28, 117–128. histories. Oxford Oxford [Engla [England] nd];; New York: York: Herodotus. 1998. The histories Oxford University Press. Hillig KW, Mahlberg PG. 2004. A chemotaxonomic analysis of  cannab cannabino inoid id variati variation on in Cannabis (Cannabaceae). American   Journal of Botany 91, 966–975.  ¨  ¨ KJ. 1998 Hsu 1998.. Did Did the the Xinj Xinjia iang ng Indo Indo-E -Eur urop opea eans ns leav leavee their home because of global cooling? In: Mair VH, ed. The   Bronze Bronze Age and Early Iron Age peoples peoples of Eastern Central  Central   Asia, Vol. II. Washington, DC: Institute for the Study of Man, 683–696. Jiang HE. 2008. [Wo guo zao qi pu tao zai pei de shi wu zheng ju: Tulufan Yanghai mu di chu tu 2300 nian gian de pu tau teng. (in Mandarin)]. Earlier physical evidence of viticulture in China: the vinifera L. in the disco discove very ry of a grap grapev evin inee of  Vitis vinifera the Turp Turpan an Archaeobotanical ical studies studies in Yang Yangha haii tomb tombss 2300 2300 year yearss old. old. Archaeobotan severa severall importa important nt sites sites of Xinjia Xinjiang, ng, China China. Beijin Beijing, g, China: China: Institute of Botany, Chinese Academy of Sciences and Academia  Turfanica, 6–21. Jiang Jiang HE, Li X, Fergu Ferguson son DK, Wang YF, YF, Liu CJ, Li CS. Capparis spinosa 2007. The discovery of  L. (Capparidaceae) (Capparidaceae) in the Yanghai Yanghai Tombs (2800 years years BP), NW China, and its Ethnopharmacology 113, medicinal medicinal implications. implications. Journal of Ethnopharmacology 409–420. Jiang Jiang HE, Li X, Zhao YX, YX, Fergu Ferguson son DK, Hueber Hueber F, Bera S, Wang YF, Zhao LC, Liu CJ, Li CS. 2006. A new insight into Cannabis Cannabis sativa (Cannabacea (Cannabaceae) e) utilization utilization from 2500-year-ol 2500-year-old d Yanghai Tombs, Xinjiang, China. Journal of Ethnopharmacology 108, 414–422. Kamberi Kamberi D. 1998. Discovery Discovery of the Ta¨klimakanian ¨klimakanian civilizatio civilization n during during a century century of Tarim archeologica archeologicall exploratio exploration n (c. 1886–1996). In: Mair VH, ed. The Bronze Age and Early Iron Age peoples of   Eastern Central Asia, Vol. II. Washington, DC: Institute for the Study of Man, 785–811. Kim ES, Mahlberg PG. 2003. Secretory vesicle formation in the Cannabis sativa L. secret secretory ory cavity cavity of glandu glandular lar tricho trichomes mes of  Cannabis (Cannabaceae). Molecules and Cells 15, 387–395.  ¨rzer  ¨ rzer K-H. 2000 2000.. 3000 3000 year yearss of agri agricu cult ltur uree in a vall valley ey of  Kno Vegetation History and Archaeobotan Archaeobotanyy 9, the High Himalayas. Himalayas. Vegetation 219–222. Kuzmina Kuzmina EE. 1998. 1998. Cultur Cultural al connec connectio tions ns of the Tarim Tarim Basin Basin people and pastoralists of the Asian steppes in the Bronze Age. In: Mair VH, ed. The Bronze Age and Early Iron Age peoples of 

4182 Russo  et al.   Eastern Central Asia, Vol. I. Washington, DC: Institute for the Study of Man, 63–93. Ma Y, Sun Y. 1994. The Western Regions under the Hsiung-Nu and the Han. In: Harmatta J, Puri BN, Etamadi GF, eds. History of civili civilizat zation ionss of Central Central Asia, Asia, Vol. The deve develo lopm pmen entt of  Vol. II. The sedentary sedentary and nomadic nomadic civilizatio civilizations: ns: 700 BC to AD 250. Delhi: Delhi: Motilal Banarsidass Publishers, 227–246. 1994. The culture culture of the Xinjiang Xinjiang region region.. In: Ma Y, Wang B. 1994. Harmatta J, Puri BN, Etemadi GF, eds. History of civilizations of  Central Asia, Vol. II. The development of sedentary and nomadic civilizatio civilizations: ns: 700 BC to AD 250. Delhi: Delhi: Motila Motilall Banars Banarsida idass ss Publishers, 209–225. search of the Indo-Eu Indo-Europ ropean eans. s. Languag Language, e, Mallory JP. JP. 1989. In search archaeology and myth. London: Thames and Hudson. Mallory JP. 1998. A European perspective on the Indo-Europeans in Asia. In: Mair VH, ed. The Bronze Age and Early Iron Age   peopl peoples es of Eastern Eastern Centra Centrall Asia Asia, Vol. Vol. I. Washi Washing ngto ton, n, DC: DC: Institute for the Study of Man, 175–201. Mallory JP, Mair VH. 2000. The Tarim mummies: ancient China and the mystery of the earliest peoples from the West . New York: Thames & Hudson. McPa McPart rtla land nd JM, JM, Russ Russo o EB. EB. 2001 2001.. Cann Cannab abis is and and cann cannab abis is Journal of Cannabis Cannabis extracts: greater than the sum of their parts? Journal Therapeutics 1, 103–132. Mukherjee A, Roy SC, Bera SD, Jiang HE, Li X, Li CS, Bera S. 2008. Results of molecular molecular analysis of an archaeological archaeological hemp Cannabis sativa L.) (Cannabis L.) DNA samp sample le from from Nort North h West West Chin China. a. Genetic Genetic Resources Resources and Crop Evolution Evolution 55, 481–485. Pacifi Pacifico co D, Misell Misellii F, Michel Micheler er M, Carbon Carbonii A, Ranall Ranallii P, marker-assisted d selection selection of  Mandolino Mandolino G. 2006. Genetics and marker-assiste Cannabis sativa L. Molecular Molecular Breeding Breeding 17, the chemot chemotype ype in Cannabis 257–268. Pan Pan B. 1996. 1996. Turpan Turpan Eremop Eremophyt hytee Botani Botanicc Garden, Garden, Academ Academia  ia  Sinica, China. Botanic Gardens Conservation News 2, 1–2. Potter D. 2004. Growth and morphology of medicinal cannabis. Medicinal uses In: In: Guy Guy GW GW,, Whit Whittl tlee BA, BA, Robs Robson on P, eds. eds. Medicinal of cannabis and cannabinoids . London: London: Pharmaceut Pharmaceutical ical Press, Press, 17–54. Puett M. 1998. China in early Eurasian history: a brief review of  recent scholarship on the issue. In: Mair VH, ed. The Bronze Age and Early Iron Age peoples peoples of Eastern Eastern Central Central Asia Asia, Vol. II. Washington, DC: Institute for the Study of Man, 699–715. Renfrew C. 1998. The Tarim Basin, Tocharian, and Indo-European origins: a view from the West. In: Mair VH, ed. The Bronze Age

and Early Iron Age people peopless of Eastern Eastern Central Central Asia, Vol. Vol. I. Washington, DC: Institute for the Study of Man, 202–212. Rudenko SI. 1970. Frozen tombs of Siberia; the Pazyryk burials of   Iron Age horsemen. Berkeley: Berkeley: University of California California Press. Russo EB. 2007. History of cannabis and its preparations in saga, science and sobriquet. Chemistry and Biodiversity 4, 2624–2648. Sarianidi V. 1994. Temples of Bronze Age Margiana: traditions of  ritual architecture. Antiquity 68, 388–397. Margiana a and protozo protozoroas roastri trism sm. Athen Sarianidi Sarianidi V. 1998. Margian Athens, s, Greece: Kapon Editions. Schlumbaum A, Tensen M, Jaenicke-Despre Jaenicke-Despre´ s V. 2008. Ancient  Vegetation History and ArchaeArchaeplant DNA in archaeobotan archaeobotany. y. Vegetation obotany 17, 233–244. Smal Smalll E, Beck Beckst stea ead d HD. HD. 1973. 1973. Cannab Cannabino inoid id phenot phenotype ypess in Cannabis sativa. Nature 245, 147–148. Taberle aberlett P, Gielly Gielly L, Pautou Pautou G, Bouvet Bouvet J. 1991. Universal Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17, 1105–1109. Thompson RC. 1923. Assyrian medical texts from the originals in the British Museum. London: Oxford University Press. dictionary of Assyrian Assyrian botany. London: Thompson Thompson RC. 1949. A dictionary British Academy. Studiess on the origin origin of cultiv cultivate ated d plants plants. Vavilov avilov NI. 1926. Studie Leningrad: Leningrad: Institut Institut de Botanique Botanique Applique Applique´e ´e et d’Ame´lioration ´lioration des Plantes. Plantes. Wilson JV, Reynolds EH. 1990. Texts and documents. Translation and analysis analysis of a cuneif cuneiform orm text forming forming part of a Babylo Babylonia nian n treatise on epilepsy. Medical History 34, 185–198. Winter W. 1998. Lexical archaisms in the Tocharian languages. languages. In: Bronze Age and Early Iron Age people peopless of  Mair VH, ed. The Bronze   Eastern Central Asia, Vol. I. Washington, DC: Institute for the Study of Man, 347–357. Xinjiang Institute of Cultural Relics and Archaeology. 2004. Tu lu fan kao gu xin shou huo: Shanshan Xian Yanghai mu di fa jue  jian  jian bao. bao. [in Mandar Mandarin][ in][New New result resultss of archae archaeolo ologic gical al work in Turpan: excavation of the Yanghai Graveyard.] Tu lu fan Xue yan   jiu [in Mandarin] [Turfanological Research] 1, 1–66. Zias J. 1995. Cannabis sativa (Hashish) (Hashish) as an effective effective medication in antiquity: antiquity: the anthropologi anthropological cal evidence. In: Campbell Campbell S, Green archaeo eolo logy gy of deat death h in the the ancie ancient nt near near east  east . A, eds. eds. The archa Oxford, UK: Oxbow Books, 232–234. Zias Zias J, Stark Stark H, Sellgm Sellgman an J, Levy Levy R, Werker erker E, Breuer Breuer A, Mechoulam R. 1993. Early medical use of cannabis. Nature 363, 215.