Kinase Activity Assay of a Known Human Abelson Kinase Point Mutant H396P and the Reduction of the Inhibitor Gleevec Efficacy d ue to the Mutation Alyson Brinkman & Kun Lin CHM 33901, Section 14 Purdue University, West Lafayette, IN
Abstract Chronic Myelogenous Leukemia is a bone marrow cancer caused by unregulated Abelson (Abl) kinase activity, a protein tyrosine kinase essential for proper progression of the cell cycle. Imatinib (Gleevec, STI-571) is an inhibitor of Abl kinase a ctivity; however, a few mutations in the Abl domain render Gleevec ineffective. The reduction of efficacy of Gleevec on the H396P Abl kinase mutant was assessed through protein expression in a bacterial system, purification, gel analysis, kinase activity assays and structural analysis. It was hypothesized that the mutant Abl kinase when exposed to Gleevec would show greater activity than the wild type Abl kinase. Inconsistent with this prediction, the kinase activity assay demonstrated a g reater reduction in mutant than WT Abl activity with Gleevec, possibly due to loss of WT Abl activity before the assay, rendering results unreliable. Upon investigation of the protein’s 3D structure, it was suspected that one possible mechanism for the predicted Abl resistance to Gleevec was a point mutation disruption of A-loop conformation, which is essential for substrate binding. The ability of Abl kinase to develop resistance to Gleevec through a single point mutation alludes to the need for drug treatments less susceptible resistance, possibly with a cocktail of drugs. Introduction Protein kinases play a vital role in cellular signaling, stress responses, metabolism, and cell adhesion. Protein tyrosine kinases are especially important in cellular signaling. The disruption in the regulation of the protein tyrosine ac tivities is associated with many diseases, and it could potentially lead to uncontrolled cellular division which could ultimately lead to cancer by dysregulation of proto-oncogene protein products. Disruptions in such kinases occur as mutations, such as with a point mutation or a reciprocal translocation. In the case of Chronic Myelogenous Leukemia (CML), a reciprocal translocation occurs between chromosome 9, encoding Abl tyrosine kinase, and chromosome 22, the Breakpoint Cluster Region (BCR), resulting in a fused BCR-Abl gene. This ge ne encodes for a unique constitutively active fusion protein that lacks auto-inhibition of the kinase, and uncontrolled signalling and cell division results (1). The inhibitor imatinib (Gleevec, STI-571, [Novartis]) showed tremendous success in treating CML and other diseases involving Abl mutation and was widely used (2). As kinase inhibitors are competitors to ATP, a contemporary research issue is iden tifying an inhibitor to specifically inhibit Abelson tyrosine kinase (3). Understanding of the structural features of the active site of Abl kinase is conducive to creating specific agents to pharmacologically block the kinase domain. The Abelson kinase consists of an N- and a P-lobe, a hinge region to link them. The P-loop binds ATP, and the C-lobe contains the activation A-loop. This A-loop is the primary source of regulation for the kinase, which inhibits or permits the binding of substrates and interactions with competitive inhibitors. If the A-loop is in its open conformation, the enzyme is available to bind substrate. If closed, substrate binding is blocked (4). The specific structure, showing binding to Imatinib and specific point mutations known to cause Abl dysregulation is shown in Figure 1 (5). Figure 1. The structure of tyrosine Abl kinase, showing Imatinib binding and specific point mutations.
An additional issue in CML research regards the substantial portion of patients developing imatinib resistance in the Abl kinase domain during treatment, rendering the drug ineffective (6, 7). Genomic studies of patients who have developed drug resistance to imantinib show a myriad of point mutations, including H396P, the mutation of interest in this study, that confer drug resistance to the cancer cells, rendering imantinib ineffective. To alleviate the resistance problem with imanitib and further resistance problem, a cocktail o f drugs could be developed to minimize the possibility of resistance development. However, logistical issues such as resource intensiveness (8, 9), low yield (10, 11), and toxicity of a foreign kinase k inase to the host cells (12) have prevented large quantities of active protein for structural studies and activity assays. However, it’s necessary to attain a sizeable amou nt of active protein to carry out activity assays. To alleviate the toxicity of expressing a foreign kinase in the bacterial system, Seeliger et al (13) outlined a system in which they co-expressed YopH tyrosin phosphatase with Abl tyrosin kinase to reduce the toxic effect of Abl kinase. To generate the specific mutant of interest, Abl H396P, in vitro PCR point mutagenesis mutagen esis was adapted from Ho et al. (14) to generate the mutation. Kinase assay using Malachite reagents and coupling c oupling phosphatase was used to assess the k inase activity of WT Abl with and without the presence of imanitib, and activity of Abl H396P mutant with and without the presence of imanitib to examine the extent to which wh ich the kinase activity was rescued by the mutation in the presence of the inhibitor imanitib. Structural studies were also performed with PyMol to assess candidate mechanisms that render the inhibitor ineffective. ineffective. Materials and Methods Site directed mutagenesis: Primers suitable for Abelson (Abl) kinase DNA H396P mutation was designed with Human BCR-Abl sequence with NCBI reference number NM_005157 and the quality of the primers confirmed with the online primer design assistance at (http://www.idtdna.com/calc/analyzer) (http://www.idtdna.com/calc/analyzer) T7 promoter and T7 terminator primers that were upstream and downstream of the target Abl kinase gene were selected and the primer sequences are presented below
H396P Forward Primer Sequence: 5’-CACCTACACAGCCCCTGCTG-3’ H396 P Reverse Primer Sequence: 5’-CAGCAGGGGCTGTGTAGGTG-3’ PCR products were run on a 1% agarose gel, stained with ethidium bromide a nd illuminated with UV transilluminator. Where needed, the PCR products we re purified using QIAquick according to manufacturer’s instructions. The purified PCR products were analyzed using a spectrophotometer at 260/280nm for purity and concentration. Digestion of Plasmid and PCR product: Double digestion was set up with XhoI and XbaI according to manufacturer’s instructions resulting in 1X final CutSmart Buffer concentration; specifically, the 20μL of the final mutant PCR product was used in the 30μL reaction in the mutant digest and a nd the final concentration of the plasmid digest was 0.067μg/μL plasmid in the 30 μL digestion sample. After a 60 minute incubation at 37℃, the reaction was arrested with 6μL of 5X loading dye, the digestion products were then analyzed using gel g el electrophoresis and purified with QIAquick. Ligation, Transformation and Sequencing: Ligation was set up using T4 DNA ligase according to manufacturer’s instructions, resulting in a final concentration of 1X Rapid ligation buffer, bu ffer, 6.67ng/μL of plasmid vector, and 8.7ng /μL of H396P mutant digestion. E. digestion. E. coli DH5a coli DH5a cells were transformed using the ligation product. Briefly,
after 30 minutes of incubation following the ad dition of the ligation product, the cells were heat-shocked in a 42℃ water bath and returned to ice for 2 minutes, SOC media was then added followed by an one-hour incubation at 37℃. The cultures were then plated on an agar plate with 50μg/mL kanamycin and 100μg/mL streptomycin an d re-streaked. The cells were cultured in LB media with 50μg/mL kanamycin and 100μg/mL streptomycin unless otherwise specified. The cultured host cells were then lysed using QIAquick Miniprep Kit according to manufacturer’s instructions, the purified DNA was used as the template for Sanger’s sequencing to confirm successful generation of point mutation, the PCR products were the sequenced at Purdue Sequencing Core. Protein Expression and Purification: BL21DE3 E. BL21DE3 E. coli. cells were co-transformed with WT Abl plasmid and YopH ph osphatase, or with the plasmid containing the point mutation and Yop phosphatase as described above, the cells were cultured for protein purification for kinase assay. When the cells neared the end of the log phase, with an optical density at 600nm between 0.8-1.0, the expression of WT Abl or mutant Abl was induced with IPTG resulting in a final concentration of 0.2mM and incubated at 24℃ overnight in a shaker. Cell samples were taken before induction (hereafter referred to as pre-induction sample) and after induction (hereafter referred to as post-induction post-induction sample). The pelleted cells were then lysed with B-PER detergent along with protease inhibitor cocktail according to manufacturer’s instructions, the hexahistidine-tagged Abl kinase was then purified using a Ni-NTA resin column. The binding, washing and eluting buffers used were as a s the following: Binding Buffer: 50 mM Tris, 300 mM NaCl, with pH adjusted to 7.8 with HCl, Washing Buffer: 50 mM Tris, 300 mM NaCl, 30 mM imidazole, with pH adjusted to 7.8 with HCl Eluting Buffer: 50 mM Tris, 300mM NaCl, 200mM imidazole, with pH adjusted to 7.8 with HCl
Samples were collected at the binding stage, washing stage for SDS PAGE analysis later. The eluted samples were then concentrated using u sing a Millipore 30kDa MWCO centrifugal concentrator according to manufacturer’s instructions using the Binding Buffer described abov e. SDS-PAGE: The concentrated protein samples, along with pre-induction samples, post-induction samples, binding samples, washing samples, elution samples were prepared for SDS PAGE using SDS sample loading buffer resulting in a final concentration of 1X SDS sample buffer. The gels were prepared with a 4% stacking gel (prepared with pH 6.8 Tris HCl buffer) on top of a 10% running gel (prepared with pH 8.8 Tris-HCl buffer), samples were run along with a protein ladder and stained with Coomassie blue overnight. Bradford Assay and Kinase Assay: The concentrations of the protein samples were determined using Bradford Assay with a standard curve generated using Bovine Serum Albumin (BSA) Kinase assay using Malachite green reagents with coupling phosphatase 4 The kinase assay was carried out according to manufacturer’s instructions. The final concentrations of the reagents were as follows: 0.11mM Abltide; 0.2μg/μL coupling phosphatase 4; 0.1μg/μL H396P mutant Abl, or 0.1μg/μL WT Abl, or 1X assay buffer for controls; controls; 0.5μM Gleevec Gleevec or 0.02μL/μL 0.02μL/μL DMSO for negative controls; controls; and 0.2mM ATP or ADP ADP where appropriate.
The incubation lasted 10 minutes starting when ATP or ADP was added, then the reaction was arrested with Malachite Reagents.
Data Analysis: The data from Bradford assay and the kinase assay were analyzed using Microsoft Excel.
Results
Fig. 2a First reaction of H396P point mutagenesis mutagenesis products
Fig. 2b Full length point mutagenesis PCR product
Fig. 2c XbaI/XhoI double digest of H396P point mutant
In Figure 2, lanes 3 and 4 contain the samples from the Abl-S417Y PCR reaction 1 and 2. The 1 kb DNA ladder did not actualize in the gel, so the exact length of bands cannot be discerned In Figure 2a, lane 1 contains the 1 kB DNA ladder, and lanes 2 and 3 contain the Abl-S417Y Final PCR reaction, showing a bright band at approximately 1114bp. For Figure 3, lane 2 contains the DNA ladder and lane 4 contains the Abl-S417Y digest with Xbal/Xhol, and a band at approximately 974bp.
Fig. 4a Colonies of E. of E. coli DH5α transformed transformation
Fig. 4b Streaking with the colonies from E. from E. coli DH5α coli DH5α with H396P mutation
Figures 4a and 4b show the original plated, with several colonies, and re-streaked transformed E. coli plates.
Absorbance Absorbance μL of values at values at Average BSA μg 595nm 595nm of A595 added BSA replicate 1 replicate 2 replicates
0
0
0
-0.001
-0.0005
2 5
0.988 2.47
0.063 0.17
0.076 0.175
0.0695 0.1725
10 20
4.94 9.88
0.259 0.426
0.251 0.461
0.255 0.4435
Tab. 1 Bradford assay standard curv absorbance data
Fig. 5 Bradford assay standard curve
5μL of sample 10μL of sample
Abs at 595nm Rep 1 0.453
Abs at 595nm Rep 2 0.421
Average Abs at 595nm 0.437
0.767
0.726
0.7456
Tab. 2 Absorbance values of purified protein samples The data in Table 1 was collected to create the BSA Standard Curve, as shown in Figure 5. The data in Table 2 was collected from Abl kinase protein samples whose absorbances were read at 595nm. The average absorbances were 0.437 for the 5μL sample and 0.7456 for the 10μL sample.
Fig. 6a Fig. 6b SDS-PAGE stained with Coomassie Blue Lanes 1-7 contained protein ladder, pre-induction sample, post-induction sample, binding, washing, elution and concentrated protein sample, respectively. respectively. Bands can be seen around 33kDa and 45kDa. The 33kDa band in lane 5 (washing sample) is lighter than that of the 45kDa band, the opposite trend is seen in lane 6 (elution sample) and in lane 7 (concentrated protein sample). Fig. 6b shows the SDS PAGE of the H396P mutant stained with Coomasie Blue. Lanes 1-7 are protein ladder, pre-induction sample, post-induction sample, binding, washing, elution samples and the concentrated protein sample, respectively. There appears to be no bands in lane 6 (elution sample), however, there are 3 bands in lane 7 (concentrated protein), with a prominent band at 33kDa.
Fig. 7 Kinase Activity Standard Curve
Corrected Specific Activity for Coupling Rate (pmol/min/μg)
% Inhibition with Gleevec Drug
WT Abl without Gleevec
6.4
-
WT Abl with Gleevec
2.5
60.90%
Mutant Abl without Gleevec
60
-
Mutant Abl with Gleevec
11.9
80.20%
Table 3. The results for the kinase assay The standard curve generated with different concentrations of phosphate is presented in Fig. 7. The standard curve is then used to determine the specific kinase activity for coupling rate which is presented in Table 3.
Fig. 8 PyMol Elemental View of H396P Mutation in Abl Kinase protein This figure demonstrates the H396P mutation as seen using the PyMOL online program. The Abl kinase structure is shown in green, with the A-loop highlighted in red. The H396P mutation is highlighted in yellow. Gleevec (orange, in spherical view) is shown in its normal binding position to the Abl kinase.
Discussion The purpose of this study was to compare the activities of wild type Abl kinase and Abl kinase mutant in the absence and presence of the drug Gleevec, which is a competitive inhibitor in the ATP binding site. Structural studies were performed with PyMol to assess the structural changes as a result of the mutation that confer con fer Abl H396P resistance to Gleevec. The gel electrophoresis results of the first round of PCR p roducts in the point mutagenesis process are shown in Figure 2. Bands at 742bp and a band at 395bp were expected in lane 3, and lane 4, respectively. However, though the relative positions are what is expected based on the length of the PCR products, the 1 kb DNA ladder was absent from our gel, so the exact base pair length of our bands could not be discerned. The missing ladder was due to improper casting of the agarose gel: the agarose gel had partially solidified and stuck to the walls of the flask before pouring, so the wells were too shallow to hold the full volume v olume of ladder loaded. Fig. 2a shows the full length PCR product following the point mutagenesis. A band at 1114bp was expected and seen in lane 2, demonstrating the successful production of full length Abl kinase mutant through PCR. Fig. 3 shows the third electrophoresis gel, from which the mutant Abl DNA was excised and purified, an Xbal/Xhol double digest to in order to later intergrate the newly generated H396P mutation into the pET28a plasmid to be expressed. A band at 974bp was expected for our gene of interest, and was evident on the gel as a DNA fragment between the known DNA ladder fragments of 850bp and 1000bp. Fig. 4a shows colonies of E. of E. coli DH5α transformed with the point mutation plasmid once the length of the XbaI/Xhol digest had been confirmed with gel electrophoresis. Not pictured includes the negative control, which contained only the E. coli cells and, without a plasmid to confer co nfer antibiotic resistance to the cells, did not yield any bacterial growth. The positive control, containing both the E. coli cells and WT plasmid yielded the expected growth. The plate of re-streaking of the transformed colonies is shown in Fig. 4b. The Bradford assay standard curve data and the standard curve are shown in (Table 1) and (Fig. 5). The absorbance values of the sample purified protein were determined by
comparing the absorbance values against the standard curve are shown in (Tab. 2).Using the standard curve, the concentration of the protein sample was determined to be 2.9mg/mL. SDS PAGE results are shown in (Fig. 6a) and (Fig. 6b). The 45kDa band in lane 5 (washing) is much darker than the lane at 33kDa in (Fig. 6a), consistent with what’s expected. The co-expressed YopH phosphatase is responsible for the 45kDa band, during the washing phase, the majority of YopH is expected to be eluted out of the Ni-NTA column, leaving most of the Hexahistagged Abl behind. Lane 6 (elution) and lane 7 (concentrated sample) show reuslts consistent with this interpretation, the 45kDa bands in those lanes are much lighter than the 33kDa bands. However, in the mutant SDS PAGE, the 33kDA bands are very faint in comparison to the 45 kDa bands in lanes 5, 6, and 7 (washing, elution, concentrated sample, respectively), and are relatively dark in lanes 3 and 4, (post-induction and binding samples, respectively). One possibility is that most of the Abl present in the cell lysate was washed out of the Ni-NTA column. Two explanations could explain this: one being the binding time was not sufficiently long before the column was allowed to run, the other being the binding buffer or the Ni-NTA column had some amounts of imidazole present, making the resin inaccessible to the protein to be bind. The kinase assay was performed to determine the effects of the H396P mutation on o n the activity of Abl kinase in the absence and presence of Gleevec, as compared to WT Abl. The standard curve as seen in Fig. 7 was used to quantify WT and mutant kinase activity with or without Gleevec. As seen by Table 3, the percent inhibition of WT Abl kinase with Gleevec was 60.90%, while the inhibition with mutant Abl k inase was 80.20%. These results are not consistent with what was predicted. It was hypothesized that the mutant Abl kinase when exposed to Gleevec would show greater activity than the wild type Abl kinase. However, upon closer examination of the specific activities, it’s noted that the specific activity of WT Abl kinase was low to begin with, 6.4pmol/min/μg, compared with 60pmol/min/μg of the Abl H396P mutant. It is suspected that the purified WT Abl ha d lost most of its activity before the kinase assay, rendering the results of the kinase assay of the WT Abl unreliable. However, comparing compa ring the activities of the mutant in the presence and absence of the inhibitor Gleevec, Gleevec , it’s noted that the kinase was still active in the presence of the inhibitor, as a reduction of 80.6%, not 100%, of kinase activity was seen. Since the results in the WT Abl kinase assay were not no t reliable due to the possible loss of activity before the assay, it’s hard to compare the results and extrapolate how much of the kinase activity the mutation was able to rescue in the presence of the inhibitor. It’s expected that the mutation moderately reduces the activity of Abl kinase in the absence of the inhibitor Gleevec, with the inhibitor present, the kinase activity of WT Abl is expected to be be reduced to lower than 5% of the kinase activity of WT Abl without the inhibitor. The kinase activity of the H396P mutant is expected to have only reduced to around half of of the kinase activity of that of mutant without the inhibitor present. Structural studies using PyMol shows that H396 is located in the A-loop, within the C-lobe, of Abl kinase. When the H396P mutation arises, the proline residue (which is very rigid and causes a kink in the structure of the protein) creates a conformational change in the Activation loop. This structure of Abl protein with Gleevec bound can be seen in Fig. 8, the structural change destabilizes the active conformation. This destabilization of the conformation prevents proper binding of Gleevec, and lack of inhibition of the kinase. The presence of considerable phosphorylation activity in the mutant kinase assay shows that this mutation does not disrupt other substrate binding in a significant manner, however, the extent to which the
kinase activity is preserved compared to the WT Abl could not be examined due to the unreliability of the data in the WT kinase assay (Tab. 3) The resistance of Abl kinase proteins with point mutations to Gleevec indicates a need for further pharmaceutical development focused on the creation of drugs that aren’t inhibited by mutations such as H396P. These new ne w medications could either be used after Gleevec resistance develops, or in tandem with Gleevec as a cocktail treatment. Such therapies could be paramount for the treatment of cancers such as Chronic Myelogenous Leukemia, a major life threatening disease with few treatments currently on the market. Further research into this topic is critical for thousands of patients annually.
REFERENCES: 1. Elizabeth, T. 5.36 Biochemistry Laboratory, Spring 2009. (Massachusetts Institute of Technology: MIT OpenCourseWare) 2. Noble, M. E., Endicott, J. A., and Johnson, L. N. (2004) Science 303, 1800-1805 3. Krause, D. S. and Van Etten, R. A. (2005) N. (2005) N. Eng. J. Med. 353, 172-187 4. Roumiantsev, S., Shah, N.P., Gorre, M.E., nicoll, J., Brasher, B.B., Sawyers, C.L., Van Etten, R.A. (2012) Clinical resistance to the kinase inhibitor STI-571 in chronic myeloid leukemia by mutation of Tyr-253 in the Abl kinase domain P-loop. Proceedings P-loop. Proceedings of the National Academy of Science of the USA 99(16), 10700-10705 5. Zdrazil, B. and Ecker. G. (2010) Drug Discovery and Development , 361-382 6. Capdeville, R., Buchdunger, E., Zimmermann, J., and Matter, A. (2002) Nat. (2002) Nat. Rev. Drug Discov. 1, 493-502 7. Shah, N. P., Nicoll, J. M., Nagar, B., Gorre, M. E., Paquette, R. L., Kuriyan, J., and Sawyers, C. L. (2002) Cancel Cell 2, 117-125 8. Deininger, M., Buchdunger, E., and Druker, B. J. (2005) Blood (2005) Blood 105, 2640-2653 9. Sicheri, F., Moarefi, I., and Kuriyan, J. (1997) Nature (1997) Nature 385, 602-609 10. Nagar, B., Hantschel, O., Young, M. A., Scheffzek, K., Veach, D., Bornmann, W., Clarkson, B., Superti-Furga, G., and Kuriyan, J. (2003) Cell 112, 859-871 11. Garcia, P., Shoelson, S. E., George, S. T., Hinds, D. A., Goldberg, A. R., and Miller, W. T. (1993) J. (1993) J. Biol. Chem. 268, 25146-25151 12. Williams, D. M., Wang, D., and Cole, P. A. (2000) J. (2000) J. Biol. Chem. 275, 38127-38130 13. Seeliger, M. A., Young, M., Henderson, M. N., Pellicena, P., King, D. S., Falick, A. M., and Kuriyan, J. (2005) Prot. (2005) Prot. Sci. 14, 3135-3139 14. Ho, N. S., Hunt, H. D., Horton, R. M., Pullen, J. K., and Pease, L. R. (1988) Gene, 77, 51-59
