The Effect of O-Alkylguanine-DNA Alkyltransferase and Mismatch Repair Activities on the Sensitivity of Human Melanoma Cells to Temozolomide, 1,3-bis(2-Chloroethyl)1- nitrosourea, and Cisplatin

The prognosis of advanced melanoma is generally poor, because this tumor commonly exhibits intrinsic or acquired resistance to chemotherapy. In an attempt to identify the underlying causes of this resistance, we studied the roles played by the DNA repair enzyme O-alkylguanine-DNA alkyltransferase (OGAT) and the mismatch repair (MMR) system in the sensitivity of melanoma cells to temozolomide (TMZ), 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU), or cis-diamminedichloroplatinum(II) (CDDP). To this end, OGAT levels and MMR efficiency of extracts of nine melanoma cell lines and selected clones derived from four of these lines were determined and correlated with the sensitivity of the respective cells to these drugs. The effectiveness of O-benzylguanine (BG), a specific OGAT inhibitor, in potentiating TMZor BCNU-mediated cytotoxicity was also evaluated. Our results demonstrate that MMR efficiency and OGAT levels strongly affect melanoma cell sensitivity to TMZ. In MMR-proficient cells, a direct correlation between OGAT levels and TMZ IC50 values was found. When OGAT activity was inhibited with BG, the sensitivity of these cells to TMZ increased and was then dictated largely by their MMR efficiency. MMR-deficient cells were highly resistant to the drug irrespective of their OGAT levels. Although OGAT activity and MMR status seemed to be the major determinants of melanoma sensitivity to TMZ, this was not the case for BCNU and CDDP; resistance to the latter drugs clearly involves processes other than the two DNA repair pathways analyzed in this study. Management of disseminated melanoma represents one of the most challenging problems in clinical oncology, because both single-agent and combination therapies give disappointing results in the treatment of this form of neoplasia (for review, see Huncharek et al., 2001). Chemoresistance observed in vivo is reflected in vitro by low chemosensitivity of melanoma cell lines (Schadendorf et al., 1994), suggesting the presence of intrinsic cellular resistance mechanisms. Some of these might involve inhibition or dysregulation of apoptosis (for review, see Serrone and Hersey, 1999; Helmbach et al., 2001, 2002),whereas others might be linked to the levels of DNA repair enzymes that either attenuate or potentiate the effect of the drugs. One of the key factors controlling sensitivity to clinical alkylating agents that attack the O position of guanine is the DNA repair protein O-alkylguanine-DNA alkyltransferase (OGAT) (for review, see Pegg, 1990; Pegg et al., 1995; Gerson, 2002). This detoxifying protein removes small alkyl groups from Oguanine in DNA and transfers them to an internal cysteine residue in a stoichiometric and autoinactivating reaction. Thus, tumor cells with high OGAT levels are generally more resistant to temozolomide or 8-carbamoyl-3-methyl-imidazo[5,1-d]1,2,3,5-tetrazin-4(3H)-one (TMZ), 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU), and 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea (CCNU) than OGAT-deficient cell lines (Pegg, 1990; Pegg et al., 1995; Gerson, 2002). Correspondingly, depletion of OGAT activity by the competitive inhibitor O-benzylguanine (BG; for review, see Dolan and Pegg, 1997) increases tumor cell This work was supported by the Italian Ministry of Health. J.J. and G.M. acknowledge the generous support of the Schweizerischer Nationalfonds zur Förderung der wisseschaftlichen Forschung. Article, publication date, and citation information can be found at http://jpet.aspetjournals.org. DOI: 10.1124/jpet.102.043950. ABBREVIATIONS: OGAT, O-alkylguanine-DNA alkyltransferase; TMZ, temozolomide; BCNU, 1,3-bis(2-chloroethyl)-1-nitrosourea; CCNU, 1-(2chloroethyl)-3-cyclohexyl-1-nitrosourea; BG, O-benzylguanine; MMR, mismatch repair; CDDP, cis-diamminedichloroplatinum(II); O-MeG, Omethylguanine; CM, complete medium; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PBS, phosphate-buffered saline; mAb, monoclonal antibody; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]propanesulfonate; STE, short-term exposure; LTE, long-term exposure. 0022-3565/03/3042-661–668$7.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 304, No. 2 Copyright © 2003 by The American Society for Pharmacology and Experimental Therapeutics 43950/1038813 JPET 304:661–668, 2003 Printed in U.S.A. 661 at A PE T Jornals on M ay 1, 2017 jpet.asjournals.org D ow nladed from sensitivity to triazene compounds and chloroethylnitrosoureas both in vitro and in vivo (Pegg, 1990; Pegg et al., 1995; Dolan and Pegg, 1997; Gerson, 2002). Experimental evidence obtained during the past few years has implicated the mismatch repair (MMR) system in the cellular response to chemotherapeutic agents. Loss of MMR is associated with resistance to TMZ, 6-thioguanine, cis-diaminedichloroplatinum(II) (CDDP), doxorubicin, and etoposide (for review, see Fink et al., 1998), and with hypersensitivity to CCNU (Aquilina et al., 1998). The MMR system, dedicated to the correction of biosynthetic errors (for review, see Modrich, 1997; Jiricny and Nyström-Lahti, 2000), involves the hMSH2, hMSH3, hMSH6, hMLH1, and hPMS2 polypeptides, as well as other proteins that participate in DNA metabolism (Modrich, 1997; Jiricny and NyströmLahti, 2000). When cells treated with methylating agents undergo DNA replication, unrepaired O-methylguanine (OMeG) residues in the template strand form mismatches with thymines, which serve as substrates for the MMR system. However, because the repair process is directed to the newly synthesized strand and because no better match other than thymine can be found for O-MeG, reiterated futile attempts of the MMR system to correct the O-MeG/T mismatches might lead to cell death (for review, see Karran, 2001). In cells treated with CDDP, the MMR system may promote cell death by preventing recombinational repair of double-strand breaks that arise as secondary lesions of drug-induced DNA modifications (Karran, 2001). In contrast, in cells exposed to CCNU, it has been hypothesized that MMR participates in the removal of the interstrand cross-links generated by unrepaired O-(2-chloroethyl)guanine, thus providing protection against cytotoxicity of the drug (Aquilina et al., 1998). BCNU and CDDP are commonly used in the treatment of melanoma (Huncharek et al., 2001), TMZ is in phase III clinical trials (Middleton at al., 2000), and BG is in phase I/II clinical studies to assess the effect of OGAT inhibition on tumor cells resistant to TMZ or BCNU (Gerson, 2002). However, the relationship between OGAT levels and TMZ or BCNU sensitivity, as well as the therapeutic potential of BG as inhibitor of the enzyme, have to date been investigated in only two melanoma cell lines (Dolan et al., 1991; Wedge et al., 1996a,b) and in a human melanoma xenograft model (Wedge et al., 1997). Moreover, no studies examined the effect of MMR status on the cytotoxicity of TMZ, BCNU, and CDDP in melanoma cells. We set out to readdress this situation by evaluating the influence of OGAT and MMR activities on melanoma cell sensitivity to TMZ, BCNU, and CDDP and on the ability of BG to potentiate the cytotoxic effects of the former two drugs. To this end, we used several MMR-proficient melanoma cell lines and clones, as well as the recently established MMR-deficient human melanoma cell lines PR-Mel, which does not express the hMLH1 and hPMS2 proteins (Alvino et al., 2002), and PD-Mel, which lacks hMSH6. We show that in melanoma cells resistance to TMZ is controlled principally by the OGAT and MMR systems, whereas resistance to BCNU and CDDP involves other factors as well. Materials and Methods Cell Lines and Clones. Nine human melanoma cell lines were used in this study. M14 and M10 (Golub et al., 1976) were kindly provided by Dr. G. Zupi (Istituto Regina Elena, Rome, Italy); LCPMEL, GL-MEL, and PD-Mel (Lacal et al., 2000) were kindly provided by Dr. Guadagni (Istituto Regina Elena); CR-Mel, PR-Mel, and CNMel (Alvino et al., 2002) were established in our laboratory; and SK-Mel-28 was obtained from the American Type Culture Collection (Manassas, VA). LCP-Mel was obtained from a primary melanoma; PR-Mel, M14, and M10 were established from cutaneous metastases; and the remaining cell lines were originated from noncutaneous metastases, including lymph node and organ metastases. Clonal populations were obtained from PR-Mel, M14, LCP-MEL, and GL-MEL lines by limiting dilution. All cell lines and clones were cultured at 37°C in 5% CO2-humidified atmosphere and maintained in RPMI 1640 medium (Hyclone Europe, Cramlington, UK) supplemented with 10% heat-inactivated (56°C, 30 min) fetal calf serum (Hyclone Laboratories, Logan, UT), 2 mM L-glutamine, and antibiotics (Invitrogen, Paisley, Scotland) (referred to as complete medium,