Differential competitive resistance to methylating versus chloroethylating agents among five O-alkylguanine DNA alkyltransferases in human hematopoietic cells

P140K-MGMT and G156A-MGMT genes encode two O-benzylguanine–resistant O-alkylguanine DNA alkyltransferase proteins that confer a high degree of Obenzylguanine and 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) or O-benzylguanine and temozolomide resistance to primary hematopoietic cells. In this study, we directly compared these and three other O-benzylguanine– resistant MGMT genes for their ability to protect the human erythroleukemia cell line, K562, using a direct competitive selection strategy to identify the mutation that conferred the greatest degree of protection from Obenzylguanine and either BCNU or temozolomide. MFG retroviral vector plasmids for each of these mutants [G156A-MGMT (ED50 for O -benzylguanine, 60 Mmol/L); and P140K-MGMT, MGMT-2 (S152H, A154G, Y158H, G160S, L162V), MGMT-3 (C150Y, A154G, Y158F, L162P, K165R), and MGMT-5 (N157T, Y158H, A170S; ED50 for benzylguanine, >1,000 Mmol/L)] were mixed, and the virus produced from Phoenix cells was transduced into K562 cells. Stringent selection used high doses of O-benzylguanine (800 Mmol/L) and temozolomide (1,000 Mmol/L) or BCNU (20 Mmol/L) administered twice, and following regrowth, surviving clones were isolated, and the MGMT transgene was sequenced. None of the mutants was lost during selection. Using temozolomide, the enrichment factor was greatest for P140K-MGMT (1.7-fold). Using BCNU selection, the greatest enrichment was observed with MGMT-2 (1.5-fold). G156A-MGMT, which is the least O-benzylguanine–resistant MGMT gene of the mutants tested, was not lost during selection but was selected against. The optimal mutant MGMT useful as a drug resistance gene may depend on whether a methylating or chloroethylating agent is used for drug selection. [Mol Cancer Ther 2006;5(1):121–8] Introduction O-benzylguanine, an inhibitor of the DNA repair protein, O-alkylguanine DNA alkyltransferase (AGT) plus 1,3-bis (2-chloroethyl)-1-nitrosourea (BCNU) or O-benzylguanine plus temozolomide have increased therapeutic efficacy over thechemotherapeutic agentsaloneinanimal tumorxenograft models (1–8). Thus, these combinations are currently being evaluated in clinical trials with cutaneous lymphomas, malignant melanoma, brain neoplasms, multiple myeloma, and gastrointestinal carcinomas (9–12). However, the damage caused to the bone marrow by these alkylating agents manifests in the short term as myelosuppression or pancytopenia, the dose-limiting toxicity in clinical trials, and has the potential to induce therapy-related myelodysplasia or acute leukemia (13, 14). Mutations in human MGMT have been developed, which retain AGT activity but are O-benzylguanine resistant. Based on the crystal structure of the AGT active site pocket, it is apparent that these mutants affect the size, hydrophobicity, and flexion of the active site pocket, resulting in exclusion of the benzyl moiety from the active site or a reduction in its ability to undergo an alkyl transfer reaction with the active site cysteine. G156A-MGMT , located within the DNA-binding wing flanking the alkyl-binding pocket (15), is moderately O-benzylguanine resistant (EC50 = 60 Amol/L). It distorts the adjacent loop (residues 158–160) that constitutes one wall of the benzyl-binding pocket (16, 17). The P140K-MGMT mutant alters the floor of the alkyl-binding pocket, excluding the benzyl group and Received 7/11/05; revised 10/7/05; accepted 10/28/05. Grant support: USPHS grants RO1CA73062, RO1ES06288, UO1CA75525, and P30CA43703. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Note: A.M. Fontes is currently at the Regional Blood Center of Ribeirão Preto, Ribeirão Preto, SP, 14051-140 Brazil. B. Davis is currently at Virexis, Gaithersburg, Maryland. L. Encell is currently at Tanox, Inc., Department of Molecular Biology, 10301 Stella Link, Houston, TX 77025. Requests for reprints: Stanton L. Gerson, Division of Hematology Oncology, BEB-3 Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106-4955. Phone: 216-368-1177; Fax: 216-368-1166. E-mail: [email protected] Copyright C 2006 American Association for Cancer Research. doi:10.1158/1535-7163.MCT-05-0236 121 Mol Cancer Ther 2006;5(1). January 2006 on July 9, 2017. © 2006 American Association for Cancer Research. mct.aacrjournals.org Downloaded from resulting in a very high degree of O-benzylguanine resistance (EC50 > 1.2 mmol/L; refs. 18, 19). Molecular evolution with selection allowed the identification of multiple MGMT mutants highly resistant to O-benzylguanine, which retain the AGT ability to remove alkyl adducts from the O-guanine position in DNA (20). Three mutants (i.e., MGMT-2, MGMT-3 , and MGMT-5), each containing three to five amino acid substitutions in the region of the active site, were the most resistant to O-benzylguanine and BCNU (EC50 > 2 mmol/L; ref. 20). These mutants, O-benzylguanine – resistant MGMT genes, provide a potent selection strategy for drug resistance gene transfer. We have used this for hematopoietic stem cell selection because two chemotherapeutic agents, BCNU, and methylating agents, such as temozolomide, are stem cell toxins, resulting in cumulative stem cell loss and delayed myelosuppression. We first introduced the use of Obenzylguanine–resistant G156A-MGMT to selectively protect the human CD34 cells and suggested that this selective protection could occur while sensitizing the tumor to chemotherapy using O-benzylguanine (21). Selecting the best O-benzylguanine–resistant MGMT for gene therapy is an important therapeutic decision. These mutants vary in stability, level of O-benzylguanine resistance, and repair capacity for either O-methylguanine or Ochloroethylguanine. The aim of the present study was to compare the ability of five of the most potent MGMT mutants to protect human hematopoietic cells against toxicity from O-benzylguanine + BCNU or O-benzylguanine + temozolomide treatment using a direct competitive selection strategy. Materials andMethods Chemicals BCNU and temozolomide were obtained from the Developmental Therapeutics Branch, National Cancer Institute (Bethesda, MD). The BCNU was solubilized in 100% ethanol before dilution in PBS and used within 10 minutes of reconstitution. Temozolomide was solubilized in DMSO (Sigma, St. Louis, MO) before dilution in PBS and was used within 10 minutes of reconstitution. O-benzylguanine was synthesized by R. Moschel at the Frederick Cancer Research Institute and was solubilized in DMSO. Complete medium for K562 (human chronic myelogenous leukemia cell line) consisted of Iscove’s modified Dulbecco’s medium (Life Technologies/Bethesda Research Laboratories, Gaithersburg, MD) supplemented with 10% fetal bovine serum (Hyclone Laboratories, Logan, UT), 1% L-glutamine plus 1% penicillin/streptomycin (Life Technologies/Bethesda Research Laboratories); and for producer cell line Phoenix, consisted of DMEM (Life Technologies/Bethesda Research Laboratories) supplemented with 10% fetal bovine serum (Hyclone Laboratories), 1% L-glutamine plus 1% penicillin/streptomycin. Human interleukin-3 and human granulocyte macrophage colony-stimulating factor were purchased from R&D Systems (Minneapolis, MN). The monoclonal antibody anti-MGMT (monoclonal antibody clone MT3.1) was purchased from Kamiya Biomedical Co. (Seattle, WA), and R-phycoerythrin–conjugated antibodies were purchased from Caltag Laboratories (Burlingame, CA). The mouse isotype IgG1 was purchased from Becton Dickinson (San Jose, CA). RetroviralVectors Retroviral vector pMFG-G156A was constructed to express cDNA sequence for G156A-MGMT as previously described (15). Retroviral vector pMFG-P140K was constructed to express cDNA sequence for P140K-MGMT by Davis et al. (22) and was originally selected for dual resistance to N-methyl-NV-nitro-N-nitrosoguanidine and O-benzylguanine from a library of AGT mutants containing a random sequence at positions 138 to 140 (18). Retroviral vectors pMFG-MGMT-2, pMFG-MGMT-3 , and pMFG-MGMT-5 , containing the following mutations of the MGMT gene, as previously described, MGMT-2 (S152H, A154G, Y158H, G160S, L162V), MGMT-3 (C150Y, A154G, Y158F, L162P, K165R), and MGMT-5 (N157T, Y158H, A170S), were subcloned into MFG retroviral vector (20). These three MGMT mutant genes were originally selected for dual resistance to N-methyl-NV-nitro-N-nitrosoguanidine and O-benzylguanine from a library of MGMT mutant genes containing a random sequence at positions 150 to 172 (19) and conferred the greatest degree of resistance to O-benzylguanine and BCNU in K562 cells (20). CellTransfections andTransductions DNA from each retroviral vector plasmid was isolated using NucleoBond Plasmid Max kit from Clontech (Palo Alto, CA) and then pooled by mixing 2 Ag of each construct. A total of 10 Ag of retroviral plasmid was transfected into 5 10 Phoenix amphotropic packaging cells (a gift from G. Nolon, Stanford University, Stanford, CA), using a calcium phosphate coprecipitation protocol. To transduce K562 cells, these amphotropic producer cells were grown to confluence, and the virus was harvested in DMEM supplemented with 10% heat-inactivated fetal bovine serum. To remove contaminating producer cells, supernatant was filtered through a 0.45-Am filter (Millipore, Bedford, MA). Polybrene was added at 5.5 Ag/mL. K562 cells were infected twice at intervals of 24 hours. The efficiency of transfection and transduction was measured by flow cytometry. Flow Cytometric Analysis To determine the MGMT expression in transduced cells, 1 10 cells were washed in 2% bovine serum albumin (BSA)/PBS, stabilized for 30 minutes at 4jC using 1% paraformaldehyde (in PBX 1 ), and permeabilized by incubating in 1% Tween 20 (in 2% BSA/PBS) for 30 minutes at 37jC. After permeabilization, cells were washed in 2% BSA/PBS, and nonspecific binding sites were blocked for 30 minutes at 22jC with 10% normal goat serum. AGT antibody mT3.1 (1 Ag) was added, and cells were incubated at 4jC overnight. Cells were washed twice with 4 http:/ www.stanford.edu/group/nolan/protocols/pro_helper_dep.html O-Alkylguanine DNA Alkyltransferases 122 Mol Cancer Ther 2006;5(1). January 2006 on July 9, 2017. © 2006 American Association for Cancer Research. mct.aacrjournals.org Downloaded from 2% BSA/PBS and incubated with secondary antibody (goat anti-mouse IgG-1g phycoerytrin conjugated) for 1 hour at 4jC. After washing, cells were resuspended in 300 AL of 2% BSA/PBS for fluorescence-activated cell sorting analysis. Flow cytometry analysis of 10,000 events was done on a FACScan (Becton Dickinson) running CellQuest data acquisition and analysis software (Becton Dickinson). Light scatter was used for gating on permeabilized cells. Stringent Selection Protocol High-stringency O-benzylguanine and BCNU and Obenzylguanine and temozolomide conditions were used to select O-benzylguanine – resistant MGMT mutant expressing K562 cells. For O-benzylguanine and BCNU selection, transfected K562 cells (30 10) were treated with 800 Amol/L O-benzylguanine for 1 hour followed by treatment with 20 Amol/L BCNU for 2 hours. For Obenzylguanine and temozolomide selection, transfected K562 cells (30 10) were treated with 800 Amol/L Obenzylguanine for 1 hour followed to treatment with 1,000 Amol/L temozolomide for 2 hours. Temozolomide-treated cells were washed in PBS and exposed additionally to 800 Amol/L O-benzylguanine for 16 to 24 hours to provide the maximal toxicity to residual O-methylguanine lesions. Cells treated with both agents were washed with PBS to remove the drugs and left to grow for 2 weeks in the presence of 25 Amol/L O-benzylguanine. This provides ongoing AGT inhibition and is the maximal amount of O-benzylguanine that can be added without slowing cell cycle times (3). After recovery, a second identical treatment was done. Outgrowth before analysis was for 14 days. Analysisof theCytotoxicEffectsofO-Benzylguanine andTemozolomide Colony-Forming Unit Assay. The O-benzylguanine and BCNU and O-benzylguanine and temozolomide resistance conferred to K562 cells by these five O-benzylguanine– resistant MGMT mutants was determined by clonogenic assays. Transfected K562 cells were exposed to increasing doses of temozolomide or BCNU, in the presence or absence of 25 Amol/L O-benzylguanine. After treatment, cells were washed in PBS, and 1,000 cells/mL were plated in triplicate in complete methylcellulose medium (0.8% methylcellulose, 1% BSA, 2 mmol/L L-glutamine, 0.1 mmol/L 1-mercaptoethanol, 10 ng/mL human interleukin-3, and 85 ng/mL human granulocyte macrophage colony-stimulating factor in Iscove’s modified Dulbecco’s medium; Stem Cell technologies, Vancouver, British Columbia, Canada). One-milliliter samples of the mixture were plated into 35-mm dishes and grown for 10 days. Colonies containing >50 cells was enumerated and compared with parental K562 cells. PCR Analyses. In four independent experiments, individual colonies picked from methylcellulose cultures were washed with PBS, centrifuged, and then resuspended in 25 to 30 AL of a buffer containing 50 mmol/L Tris-HCl (pH 8), 10 mmol/L EDTA, 100 mmol/L NaCl, and 1% Triton X-100, plus 1 mg/mL proteinase K. The digestion was carried out at 55jC for 2 hours. Proteinase K was inactivated by heating at 95jC for 8 to 10 minutes. Two to five microliters of the digested sample were used for PCR analysis. Amplification was done (after an initial 3 minutes at 95jC) for 35 cycles of 30 seconds at 94jC, 40 seconds at the indicated annealing temperature, and 40 seconds at 72jC. A final extension period of 10 minutes was done. Annealing temperature was 62jC for the MGMT gene and 60jC for the dystrophin gene. Primers for MGMT gene (5V-CTTCACCATCCCGTTTTCCAG-3V and 5VCTGCCAGACCTGAGCTCCCTC-3V) amplify a 316-bp product; dystrophin primers (5V-TCACTTGCTTGTGCGCAGG-3Vand 5V-GAAAAGTGTATATCAAGGCAGCGACGATAA-3V) amplify a 500-bp product. The PCR was done with 40 pmol of each primer, 0.2 mmol/L deoxynucleotide triphosphates, 2 units of Taq polymerase, in a total volume of 50 AL. The PCR products were separated on 1.3% agarose gel and visualized after ethidium bromide staining. The amplified DNA was purified using the QIAquick PCR Purification kit (Qiagen, Valencia, CA) and sequenced to characterize the MGMT gene mutant. Sequence data were analyzed by the MacVector program, version 3.1. The wild-type human MGMT is from Genbank (accession no. NM_002412). Characterization of the Transduced K562 Cell Population Analysis of Individual K562 Clones. Individual colonies of transduced K562 cells were picked from methylcellulose and cultured in Iscove’s modified Dulbecco’s medium as above, and cells were harvested for PCR analysis to evaluate the efficiency of the integration of retrovirus DNA. PCR-positive clones were submitted to fluorescenceactivated cell sorting analysis for AGT protein expression as above and to reverse transcription-PCR for mRNA MGMT expression level. Reverse Transcription-PCR Assay. Total cellular RNA was extracted from 1 10 K562 cells with SV total RNA isolation system (Promega, Madison, WI) according to the manufacturer’s protocol. Reverse transcription of RNA to cDNA was done according to the protocol of the SuperScript First Strand Synthesis System for reverse transcription-PCR (Life Technologies Bethesda Research Laboratories). For detection of MGMT mRNA, each K562 clone, 10% of the reverse transcriptase reaction with or without reverse transcriptase enzyme, was amplified by PCR with the same thermal cycling profile as above, using 25 cycles. Each PCR was carried out in 25 AL and resolved on a 1.3% agarose gel. Results Engineering Erythroleukemia K562 Cells to Constitutively Produce Five Mutant O-Benzylguanine ^ Resistant Human Alkyltransferase Proteins The five O-benzylguanine–resistant MGMT mutants used in this study (Table 1) have different amino acid substitutions. The three mutants containing three to five amino acid changes near the active site and P140K-MGMT have Molecular Cancer Therapeutics 123 Mol Cancer Ther 2006;5(1). January 2006 on July 9, 2017. © 2006 American Association for Cancer Research. mct.aacrjournals.org Downloaded from millimolar EC50 values for O -benzylguanine inactivation, compared with the wild-type value of 0.1 Amol/L Obenzylguanine (Table 2). There is variability among the mutants in protein stability: G156A-MGMT is the least stable. For transduction into K562 cells, we generated an amphotropic producer cell line by transfecting Phoenix cells with a pool of the five plasmids (pMFG-G156AMGMT , pMFG-P140K-MGMT , pMFG-MGMT-2 , pMFGMGMT-3 , and pMFG-MGMT-5 ) mixed at equimolar ratios. Twenty-four hours after transfection, 24% of Phoenix cells expressed these transgenes as determined by flow cytometry using an anti-MGMT antibody (data not shown). After two cycles of transduction, the percentage of K562 cells expressing AGT by flow cytometry was 11% in two independent experiments (data not shown). MGMTExpressioninClonesof TransducedK562Cells To understand the results of the competitive selection, we determined the proportion of clones expressing MGMT . K562 clones positive for MGMT by PCR were expanded and MGMT expression analyzed by flow cytometric and reverse transcription-PCR assay. Of 14 MGMT colonies analyzed by this method, three expressed high levels of AGT protein (89–96.2%, Fig. 1C–E), one moderate level (30%, Fig. 1F), four low level (4–7%, Fig. 1G–J), and six clones expressed at very low levels or not at all (1–2%, Fig. 1K–P). This analysis was repeated thrice with a total of 48 isolates, and similar results were obtained. Overall, 22% of PCR-positive clones expressed high levels of AGT protein, about 28% expressed low levels, and 43% had undetectable levels of expression. We next asked whether differences in AGT expression were due to transcriptional activity effects or posttranscriptional events. Reverse transcription-PCR analysis of mRNA showed that the highest level of MGMT mRNA was detected in the clones that express high levels of AGT protein (Fig. 2, lanes C1, C37, and C39). Much lower levels of MGMT mRNA (Fig. 2, lanes C5, C18, C20 , and C32) were detected for the clones that express 7% to 30% of AGT protein (Fig. 1, C5, C18, C20 , and C32), whereas the absence of MGMT mRNA (Fig. 2, lanes C4, C6, C22, C24, C26, C27 , and C31) was observed in clones with low or absent AGT protein (Fig. 1, C4, C6, C22, C24, C26, C27 , and C31). Therefore, the absence of AGT protein is due to silencing of transcriptional activity and not due to translational effects. Competitive Selectionof Drug-Resistant K562Clones by Stringent DrugTreatment MGMT-transduced K562 cells were treated with 800 Amol/L O-benzylguanine for 1 hour and either 1,000 Amol/L temozolomide or 20 Amol/L BCNU for 2 hours. Surviving cells were retreated with the same drugs. Two independent experiments were done, and the selection results were pooled. Survival after the first drug selection was about 3% of control, with greater survival after the second treatment. To determine the degree of drug resistance conferred to K562 cells by these mutants after high-stringency selection, we determined the BCNU and temozolomide IC90 by seeding cells into methylcellulose medium and monitored for colony formation. K562 cell cultures containing unselected AGT mutants were 2-fold more resistant to BCNU (IC90 of 22 Amol/L) compared with parental K562 cells (IC90 around 10 Amol/L). After highstringency O-benzylguanine and BCNU selection, K562 cells were 8-fold more resistance to BCNU (BCNU IC90 nearly 80 Amol/L; Fig. 3). Cells treated with BCNU alone had a similar survival curve, presumably because all O-benzylguanine–resistant MGMTs conferred O-benzylguanine resistance to the cell cultures, and a difference between the mutants could not be identified in the survival curves. Cells exposed to temozolomide had increased drug resistance after selection (Fig. 4). Parental K562 cells had a temozolomide IC90 < 100 Amol/L, whereas transduced K562 cells had a temozolomide IC90 of 600 Amol/L. After highstringency O-benzylguanine and temozolomide selection, the temozolomide IC90 increased to >2,000 Amol/L, 20-fold above untransduced cells. When pretreated with 25 Amol/L benzylguanine, there is no significant difference in the temozolomide IC90 value. Table 1. Mutations in O-benzylguanine resistant MGMTs