Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The cytotoxic effects of methylating and chloroethylating antitumor drugs appear to be mediated principally by alkylation at the O⁶-position of guanine (O⁶-G) in DNA. Tumor cell resistance to these agents is frequently associated with high levels of the DNA repair protein O⁶-methylguanine (O⁶-MeG)-DNA methyltransferase (MGMT) (reviewed in D'Incalci et al., 1988; Pegg, 1990; Pegg et al., 1995), which transfers the alkyl group from O⁶-G to an internal cysteine residue, thus restoring the integrity of the DNA. Because MGMT is inactivated in this process, the extent of O⁶-alkylguanine repair is dependent on the initial levels that are present in cells and on the rate at which new MGMT protein is synthesized (D'Incalci et al., 1988; Pegg, 1990; Pegg et al., 1995). Unrepaired O⁶-chloroethylguanine lesions give rise to lethal DNA interstrand cross-links, whereas for O⁶-MeG, the postreplication mismatch repair (MMR) system is required for cell killing (reviewed in Jiricny, 1996; Modrich, 1997).

One approach to improve the clinical effectiveness of therapeutic methylating or chloroethylating drugs is to reduce MGMT activity in tumor cells. Methylating agents are themselves capable of inducing a rapid depletion of MGMT activity in treated cells as a result of the generation of O⁶-MeG in DNA and enzyme inactivation during the repair process (Zlotogorski and Erickson, 1984; Sio et al., 1992; Lee et al., 1993a, 1994). On this basis, clinical trials involving patient treatment with the methylating agents streptozotocin or dacarbazine [5-(3,3-dimethyl-1-triazeno)imidazole-4-carboxamide (DTIC)] before exposure to chloroethylnitrosoureas have been undertaken (Aamadal et al., 1992; Panella et al., 1992; Lee et al., 1993b). However, these combined chemotherapy protocols, although showing improved clinical response, were associated with increased hematological (Lee et al., 1993b) or unusual pulmonary (Aamadal et al., 1992) toxicity, presumably as a consequence of enzyme inactivation in these sensitive tissues.

Depletion of MGMT activity has been achieved in cultured cells by exposing them to O⁶-benzylguanine (O⁶-BeG), which is an effective substrate for the protein (reviewed in Pegg et al., 1995; Dolan and Pegg, 1997). The sensitivity of tumor cells in vitro (Pegg et al., 1995; Dolan and Pegg, 1997) and of human tumor xenografts (Pegg et al., 1995; Dolan and Pegg, 1997) to O⁶-alkylating agents is markedly increased after exposure to O⁶-BeG. A number of analogs have been assessed for their ability to inactivate MGMT and to sensitize cells and xenografts (Pegg et al., 1995; Dolan and Pegg, 1997, McElhinney et al., 1998). The use of MGMT inactivators as chemotherapy adjuvants that enhance the clinical efficacy of methylating or chloroethylating agents could, however, be limited both by an increase in the sensitivity of normal tissues and by the emergence of tumor cells containing mutant forms of the MGMT protein that are resistant to such agents but still capable of repairing DNA (Crone and Pegg, 1993; Xu-Welliver et al., 1999).

An alternative approach to increasing the therapeutic effectiveness of O⁶-G-alkylating agents might be the use of antitumor agents that are capable of attenuating MGMT activity either by direct interaction with the protein or indirectly via effects at the transcriptional level. This could have the advantage that MGMT depletion would be accomplished by a drug that also had antitumor activity, and therefore synergistic effects with the O⁶-alkylating agent may occur. In this respect, it has been shown that antineoplastic agents such as cyclophosphamide (Lee et al., 1992) or cisplatin (Wang and Settlow, 1989) are capable of inhibiting MGMT activity. Cisplatin is widely used in cancer chemotherapy, alone or in association with other antitumor agents, including DTIC and 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU), in the treatment of malignant melanoma (reviewed in Garbe, 1993). It was therefore of interest to investigate the mechanism by which cisplatin attenuates MGMT activity and to evaluate the kinetics of enzyme depletion and recovery in cells exposed to the drug in combination with an O⁶-G-methylating agent, to provide an experimental basis for the rational clinical use of such combinations.

In this study, MGMT activity and MGMT gene expression were investigated in the human leukemic cell line Jurkat, in which the kinetics of MGMT activity depletion and recovery were evaluated in cells exposed to cisplatin alone or to a sequential treatment with cisplatin and temozolomide [TMZ; 8-carbamoyl-3-methyl-imidazo[5,1-d]-1,2,3,5-tetrazin-4(3H)-one]. The Jurkat cell line, which not only expresses high levels of MGMT but also is deficient in MMR activity (Levati et al., 1998), was chosen to avoid, in the treated population, a large proportion of dead and dying cells that might obscure the effects of treatment on MGMT activity and MGMT gene expression. Absence of MMR activity is indeed associated with increased resistance both to TMZ (D'Atri et al., 1998) and to cisplatin (Anthoney et al., 1996; Drummond et al., 1996).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Lines. The human leukemic cell line Jurkat was maintained at 37°C in a 5% CO₂ humidified atmosphere in RPMI 1640 (Hyclone Europe, Cramlington, UK) supplemented with 10% heat-inactivated (56°C, 30 min) fetal calf serum (Hyclone Laboratories, Logan, UT) and 2 mM L-glutamine (Hyclone) [hereafter referred to as complete medium (CM)].

Drugs and Reagents. TMZ was kindly provided by the Schering-Plough Research Institute (Kenilworth, NJ). Because the drug readily decomposes in aqueous solution into 5-(3-methyl-1-triazeno) imidazole-4-carboxamide (MTIC; Stevens et al., 1987), solutions were always prepared fresh by dissolving the drug in RPMI 1640. Cisplatin (Prontoplatamine, 500 µg/ml in saline, pH 3-5) was purchased from Pharmacia & Upjohn (Milan, Italy). All reagents for Northern blot analysis were purchased from Bio-Rad (Hercules, CA), whereas the remainder of the chemicals were obtained from Sigma Chemical Co. (St. Louis, MO).

DNA Platination. Calf thymus DNA (Roche Diagnostics, Monza, Italy) was treated with various concentrations of cisplatin at room temperature for 24 h in TE buffer (Tris-HCl/EDTA), pH 8.0 (20 µg of DNA in a total volume of 1 ml, for each drug concentration), protected from the light. Control DNA was incubated in the same experimental conditions without drug.

Drug Treatment and Cell Growth Measurements. Cells were suspended at 5 × 10⁵ cells/ml in CM or in CM containing the appropriate amount of cisplatin or TMZ. They were incubated at 37°C in a 5% CO₂ humidified atmosphere for 1, 6, or 24 h, in the case of cisplatin, or for 2 h, in the case of TMZ. Cells were then washed twice with PBS, suspended at 5 × 10⁵ cells/ml in CM, and maintained in culture for an additional 48 h. Cell growth and viability were evaluated at the end of drug treatment and after 24 and 48 h of culture in drug-free medium. Cells were manually counted using a hemocytometer, and cell viability was determined by trypan blue exclusion. All determinations were in quadruplicate. To evaluate the combined effect of cisplatin and TMZ on cell growth, cells were incubated in CM or in CM containing cisplatin (6.25 or 12.5 µM) for 24 h and then exposed to TMZ (250 µM) for 2 h. Cells were then washed and kept in culture for an additional 48 h. Cell growth and viability were evaluated as described above.

MGMT Assay. Cells were removed from control or drug-treated cultures at the end of drug treatment (time, 0 h) or after an additional 24 or 48 h of culture in the absence of the drug (time, 24 and 48 h). Cells were washed twice with PBS and stored as pellets (1 × 10⁶ cells) at 80°C until used. MGMT activity was determined by measuring the transfer of [³H]methyl groups from the DNA substrate to the MGMT protein as previously described (Watson and Margison, 1999).

Effect of Cisplatin or Platinated DNA on Recombinant Human MGMT (rhMGMT) Activity In Vitro. rhMGMT protein was overexpressed in the baculovirus/insect cell system and then partially purified as previously described (Lacal et al., 1996). An aliquot (60 fmol by activity) of rhMGMT was exposed either to 1 µg of platinated DNA or directly to graded cisplatin concentrations. Incubation was carried out in the dark in TE buffer, pH 8.0, containing 0.1% BSA, at 37°C for 1 h. Control groups consisted of the rhMGMT protein incubated either in buffer alone or with untreated DNA. The reactions were stopped at 4°C, and dithiothreitol was added to the samples to a final concentration of 3 mM. Aliquots of the samples were then used to evaluate residual MGMT activity as previously described (Watson and Margison, 1999).

Northern Blot Analysis. Total cellular RNA was extracted using the guanidine thiocyanate-phenol-chloroform method (Chomczynski and Sacchi, 1987). Aliquots (15 µg) of total RNA were fractionated by electrophoresis on a formaldehyde-containing 1.2% agarose gel. The integrity of RNA was confirmed by RNA visualization, after staining of the gel with ethidium bromide. RNA was transferred to a nylon membrane (Genescreen Plus; New England Nuclear, Boston, MA) and hybridized at 42°C for 24 h with a ³²P-labeled MGMT or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe. The MGMT probe was a polymerase chain reaction-derived cDNA probe obtained after reverse transcription of the RNA from Molt-4 cells (Lacal et al., 1996). The GAPDH probe was a generous gift from Dr. R. Dalla Favera (Department of Pathology, Columbia University, New York, NY). After washing with 0.1× standard saline citrate (10 mM sodium chloride, 1.5 mM sodium citrate) at room temperature for 30 min, the blotted membranes were exposed to x-ray films (Kodak, Rochester, NY) at 80°C. Bidimensional densitometry of the blots was performed using an Imaging densitometer GS-670 (Bio-Rad, Richmond, CA).

Statistical Analysis. MGMT activity in drug-treated samples was expressed as percentage of residual activity with respect to control groups. The mean values of percentage of residual activity (±S.E.) relative to independent experiments were calculated after angular transformation of the original values. The statistical significance of drug-induced inhibition of MGMT activity was evaluated by the paired Student's t test (PTT) analysis, using the original values of MGMT activity expressed in terms of fmol/mg of protein. The statistical significance of drug-induced cell growth inhibition was evaluated by Student's t test.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of Cisplatin on MGMT Activity of Jurkat Cells. Jurkat cells (mean MGMT activity, 472 ± 14 fmol/mg of protein) were exposed to increasing concentrations of cisplatin for 1, 6, or 24 h, washed, and then assayed for MGMT activity. The effect of cisplatin was dependent on the concentration and duration of exposure to the drug: a 1-h treatment slightly increased MGMT activity at all concentrations tested (data not shown). After a 6-h treatment with cisplatin, a small but significant (P < .01) reduction in MGMT activity was observed but only at 25 µM cisplatin (Fig. 1). In cells exposed to cisplatin for 24 h, a significant (P < .01) inhibition of MGMT activity was already evident at 6.25 µM, and this increased at 12.5 and 25 µM (Fig. 1). Cell viability in the control and drug-treated groups was always comparable and higher than 90% (data not shown). However, cells cultured in the presence of cisplatin for 24 h failed to proliferate, whereas untreated cells doubled in number (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of cisplatin on MGMT activity of Jurkat cells. Cells were exposed to the indicated concentrations of cisplatin for 6 h () or 24 h (), washed, and then used to prepare extracts to be tested for MGMT activity. Data, expressed in terms of percentage of residual MGMT activity in drug-treated groups with respect to controls, represent the mean of independent experiments. In parentheses is shown the number of experiments concerning each cisplatin concentration. Error bars, S.E. Statistically significant differences (P < .01, according to PTT) between MGMT activity of control and drug-treated cells were found in the following conditions: a) 6-h exposure to 25 µM cisplatin and b) 24-h exposure to 6.25, 12.5, and 25 µM cisplatin.

Kinetics of MGMT Depletion and Recovery after Exposure to Cisplatin. Jurkat cells were exposed to increasing concentrations of cisplatin for 6 or 24 h, washed, and maintained in culture for additional 48 h. Cell number and viability were evaluated at the end of treatment and after 24 and 48 h of culture in the absence of the drug.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of cisplatin on Jurkat cell growth. Cells were incubated in medium alone () or exposed to cisplatin at 6.25 µM (), 12.5 µM (), or 25 µM () for 6 h (A) or 24 h (B). Cells were then washed, incubated in drug-free medium, and counted after 24 and 48 h of culture. Values represent the mean of five independent experiments. Error bars, S.E. In all cases, the mean number of cells treated with cisplatin was significantly (P < .01, according to Student's t test) lower than that of control.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Kinetics of MGMT depletion and recovery in Jurkat cells exposed to cisplatin. Cells were incubated with cisplatin at 6.25 µM () or 12.5 µM () for 6 h (continuous lines) or 24 h (dotted lines). MGMT activity was evaluated at the end of drug treatment or after an additional 24 or 48 h of culture in drug-free medium. Data, expressed in terms of percentage of residual MGMT activity in drug-treated groups with respect to controls, represent the mean of independent experiments. In parentheses is shown the number of experiments concerning each cisplatin concentration. Error bars, S.E. In all cases, MGMT activity of Jurkat cells cultivated for an additional 24 h in the absence of cisplatin was significantly (P < .01, according to PTT) lower than that detected at the end of drug treatment.

Effect of Cisplatin on MGMT mRNA Levels. Experiments were performed to investigate whether cisplatin-induced depletion of MGMT activity in Jurkat cells could be the result of decreased levels of the corresponding transcript. Cells were treated with 6.25 or 12.5 µM cisplatin for 6 or 24 h, washed, and kept in culture for an additional 48 h. Northern blot analysis of MGMT and GAPDH transcripts was performed on total RNA extracted from cells immediately after the end of drug treatment or after an additional 24 and 48 h of culture without cisplatin. Northern blot analysis in cells exposed to the drug for 24 h and their corresponding controls was performed after the removal of dead cells by Ficoll-Hypaque gradient centrifugation of the population. The hybridization signals were quantified by densitometric analysis of the autoradiograms and normalized in relation to GAPDH. The results of a representative experiment are shown in Fig. 4. Jurkat cells treated with cisplatin 12.5 µM for 6 h (Fig. 4A) showed a slight decrease in MGMT mRNA expression at the end of the incubation period. After 24 h of culture in drug-free medium, the MGMT mRNA level in drug-treated cells was comparable to that of the control.

View larger version (65K): [in this window] [in a new window]   Fig. 4.   Northern blot analysis of MGMT mRNA levels in cisplatin-treated Jurkat cells. Cells were exposed to the indicated concentrations of cisplatin for 6 h (A) or 24 h (B). MGMT mRNA levels were evaluated at the end of treatment (0 h) and after 24 or 48 h of culture in the absence of the drug. Results are from a representative experiment. RNA (15 µg) from each group was electrophoresed and transferred to a nylon membrane. The membrane was hybridized with either the MGMT probe (a) or the GAPDH probe (b). The integrity of RNA was confirmed by RNA visualization, after staining of the gel with ethidium bromide (c). The hybridization signals relative to MGMT were quantified by densitometric scanning of the autoradiograms and normalized in relation to GAPDH. d, the ratios between normalized absorbance values of cisplatin-treated and untreated cells. The first row refers to the Northern blots presented in the figure; the second row refers to an additional independent experiment.

Kinetics of MGMT Activity Depletion and Recovery in Jurkat Cells Exposed to TMZ. Jurkat cells were treated with increasing concentrations of TMZ for 2 h, washed, and cultured in drug-free medium for 48 h. MGMT activity was evaluated at the end of TMZ treatment and after 24 and 48 h of culture in the absence of drug. TMZ induced a rapid depletion of MGMT activity in a concentration-dependent manner (Fig. 5). After drug removal, MGMT activity began to rise and completely recovered within 48 h. No significant inhibition of cell growth was observed at the concentrations of TMZ tested (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Effect of TMZ on MGMT activity of Jurkat cells. Cells were incubated with TMZ at 62.5 (), 125 () 250 (), and 500 () µM for 2 h; washed; and tested for MGMT activity at the end of drug treatment or after 24 or 48 h of culture in medium alone. Data, expressed in terms of percentage of residual MGMT activity in drug-treated groups with respect to controls, represent the mean of independent experiments. In parentheses is shown the number of experiments concerning each TMZ concentration. Error bars, S.E. Statistically significant differences between MGMT activity of control and drug-treated cells were found in the following conditions: a, 125, 250, and 500 µM TMZ at time 0 h; 500 µM TMZ at time 24 h (P < .01, according to PTT); and b, 62.5 µM TMZ at time 0 h; 250 µM TMZ at time 24 h; and 500 µM TMZ at time 48 h (P < .05, according to PTT).

Kinetics of MGMT Activity Depletion and Recovery in Jurkat Cells Exposed to a Combined Treatment with Cisplatin and TMZ. Jurkat cells were cultivated in the absence or in the presence of cisplatin (6.25 and 12.5 µM) for 24 h and then exposed to 250 µM TMZ for 2 h. MGMT activity was evaluated in extracts obtained from cells treated with cisplatin or TMZ, alone or in combination, at the end of the incubation period (time 0) and after 24 and 48 h of culture in the absence of the drugs. Pretreatment of Jurkat cells with cisplatin significantly (P < .01) reduced the rate of MGMT recovery after exposure to TMZ, and 12.5 µM cisplatin was more effective than 6.25 µM (Fig. 6).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Combined effect of cisplatin and TMZ on MGMT activity of Jurkat cells. Cells were treated with 250 µM TMZ for 2 h () or with cisplatin at 6.25 () or 12.5 () µM for 24 h or exposed to 6.25 µM cisplatin followed by 250 µM TMZ () or to 12.5 µM cisplatin followed by 250 µM TMZ (). MGMT activity was evaluated at the end of drug treatment and after 24 or 48 h of culture in medium alone. Data, expressed in terms of percentage of residual MGMT activity in drug-treated groups with respect to controls, represent the mean of three independent experiments. Error bars, S.E. In all instances, MGMT activity recovery in cells exposed to cisplatin plus TMZ was significantly (P < .01, according to PTT) lower than that of cells treated with TMZ alone.

Effect of Cisplatin or Platinated DNA on Activity of rhMGMT Protein. To examine whether inhibition of MGMT activity in cisplatin-treated Jurkat cells might be due to a direct effect of the drug on the MGMT protein, 60 fmol of the purified recombinant enzyme were incubated with cisplatin at concentrations ranging from 6.25 to 100 µM for 1 h and then assayed for residual MGMT activity. Similar experiments were carried out to investigate whether inactivation of MGMT protein could occur through the interaction of the enzyme with platinum adducts in DNA. In this case, the rhMGMT protein (60 fmol) was preincubated with DNA treated with the drug at concentrations ranging from 6.25 to 100 µM. Neither cisplatin nor platinated DNA detectably inhibited the activity of the rhMGMT protein (Fig. 7).

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Effect of cisplatin or platinated DNA on rhMGMT protein. rhMGMT (60 fmol) was incubated with cisplatin () or platinated DNA () at 37°C for 1 h and then tested for residual activity. Data are expressed in terms of percentage of MGMT activity of treated samples with respect to controls (rhMGMT incubated at 37°C without cisplatin or exposed to untreated DNA). Incubation of the control aliquots at 37°C for 1 h produced the loss of only 5% of the activity, relative to rhMGMT protein stored at 4°C during an identical incubation period. Values represent the mean of four determinations. Error bars, S.E.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Cisplatin is one of the most widely used cancer chemotherapeutic drugs. It is used as single agent or in combination chemotherapy regimens in the treatment of a variety of neoplasias (Reed and Kohn, 1990). In the treatment of malignant melanoma, cisplatin has been combined with DTIC and BCNU (Garbe, 1993), which are known to exert their cytotoxic effects mainly through alkylation of O⁶-G. In this report, we examined the interaction of cisplatin and TMZ on the expression of MGMT, the principal repair pathway for toxic DNA methylation damage. TMZ, a new antitumor triazene compound that is presently under phase II-III clinical investigation (Bleehen et al., 1995; Woll et al., 1995; Bower et al., 1997), spontaneously decomposes in aqueous solution into MTIC, the methylating species that is also generated by host metabolism of DTIC (Stevens et al., 1987). We used the MMR-deficient Jurkat cell line, which is relatively resistant to killing by these agents, to avoid the possible confounding effects that combined toxicities might have on the interpretation of the results.

These results show that cisplatin attenuates MGMT activity in Jurkat cells and that this is dependent on the drug concentration and on the period of exposure to the drug. It is unlikely that this was due to the direct inactivation of MGMT by cisplatin or the indirect inactivation via the formation of platinum adducts in DNA because neither platinum nor platinated DNA was able to inactivate purified rhMGMT in vitro (Fig. 7). The finding that the activity of the rhMGMT protein was not inhibited by platinated DNA contrasts with the results of an earlier report by Wang and Setlow (1989), which suggested that the MGMT protein can be inactivated by reaction with platinum adducts in DNA. However, it must be noted that the previous experiments were performed using HeLa cell extract as source of MGMT and that the cell extract was subjected to a DNase treatment and incubated with platinated DNA for 2 h before testing for residual MGMT activity. Therefore, it cannot be excluded that the apparent discrepancy between these results is due to the different experimental conditions used.

The kinetics of MGMT activity depletion in cells exposed to cisplatin is consistent with the effect of the drug on mRNA levels rather than a direct effect on the protein itself: MGMT activity is maximally reduced 24 h after drug removal and starts to recover slowly by 48 h (Fig. 3). On the other hand, the decrease of MGMT mRNA is maximal at the end of cisplatin treatment, and recovery is already evident after 24 h of cell culture in the absence of the drug (Fig. 4). This finding is consistent with the hypothesis that the decrease in MGMT activity is a consequence of the reduction in the mRNA level and with the reported long half-life of the MGMT mRNA (>10-12 h) and protein (>15-20 h) (Kroes and Erickson, 1995; Pegg et al., 1995).

The mechanism via which cisplatin reduces MGMT mRNA levels in Jurkat cells is not clear. The MGMT protein is expressed constitutively in normal cells and in about 80% of tumors. However, the levels of the enzyme, which appear mainly dependent on the level of MGMT gene transcription (Pegg et al., 1995), are quite different in various cell lines and tissues. The decreased level of MGMT mRNA in Jurkat cells exposed to cisplatin might be related to either reduced gene transcription or enhanced mRNA processing, and this might occur via interactions with any of the factors that result in tissue differences in expression levels. If platinum adducts were produced in the promoter region of the MGMT gene, this could affect the binding of regulatory proteins, perhaps resulting in decreased transcription. Indeed, the MGMT promoter region, which is extremely rich in GC sequences (Nakatsu et al., 1993), might represent a preferential target of cisplatin-induced interstrand cross-links, which occur mainly at the d(GpC) sequences (Lemaire et al., 1985). However, the promoters of other genes are also GC rich, and if this was the case, cisplatin would be expected to attenuate their expression: we have no evidence for this.

It has recently been shown that transient transfection of wild-type p53 into the p53-null cell line SAOS-2 suppresses MGMT promoter activity in a reporter gene system (Harris et al., 1996). The same authors have demonstrated that in fibroblasts transduced with a wild-type p53-adenoviral vector, endogenous MGMT decreases 24 h after transduction and is no longer detectable 5 days later (Harris et al., 1996). Cisplatin has been shown to increase the level of p53 protein in several cell lines (Anthoney et al., 1996), and it could therefore be hypothesized that MGMT mRNA reduction in Jurkat cells treated with the drug is mediated by p53. However, this hypothesis appears to be weakened by the finding that the Jurkat cell line harbors a mutated form of p53 (Iwamoto et al., 1996).

Regardless of the mechanism responsible for MGMT inhibition in cisplatin-treated cells, our data could provide a rational basis for the reported use of cisplatin in association with BCNU and/or triazene compounds in melanoma treatment. A significant inhibition of MGMT activity occurs in cells exposed to 12.5 µM cisplatin for 24 h. This concentration and time of exposure can be considered achievable in patients. Indeed, after the administration of the standard dose of 100 mg/m², the peak plasma concentration of cisplatin is 3.8 µg/ml (Untch et al., 1994; Andreotti et al., 1995) (i.e., 12.5 µM). Moreover, the half-life of the drug is 24 h or longer (Reed and Kohn, 1990).

The finding that MGMT activity is reduced no more than 50% in Jurkat cells exposed to cisplatin suggests that the in vivo effectiveness of sequential cisplatin and BCNU, which is given in a single dose, may depend on the basal MGMT levels of the tumor cells: for maximal enhancement of activity by cisplatin, MGMT activity must be reduced to below a level that ablates efficient repair of O⁶-chloroethylguanine. Indeed, a previous report showed that in HT-29 cells, which express a high level of MGMT, sensitization to BCNU requires a substantial reduction (>50%) in MGMT activity (Pieper et al., 1991). In contrast, in the case of DTIC and TMZ, which are administered for 3 and 5 days, respectively, preexposure of tumor cells to cisplatin could also be effective in cancers showing very high MGMT activity. In TMZ-treated cells, MGMT activity is markedly reduced at the end of drug exposure, as a consequence of the repair of O⁶-meG, but recovers almost completely within 24 to 48 h. Based on this, it is possible to hypothesize that in vivo, the efficacy of triazene compounds is limited not only by the high basal level of MGMT in the tumor but also by the rate of MGMT resynthesis after drug-induced depletion. The recovery of MGMT on each consecutive drug administration may allow efficient repair of O⁶-meG, and such recovery has been shown in peripheral lymphocytes of patients given DTIC (Lee et al., 1993a) or TMZ (Lee et al., 1994). When Jurkat cells were exposed to cisplatin before TMZ, the recovery of MGMT activity is significantly delayed, and after 24 h of culture in the absence of drugs, it is only 20 to 40% of the control (Fig. 6). Therefore, the administration of cisplatin 24 h before DTIC or TMZ could render tumor cells more susceptible to the cytotoxic effect of this agents by reducing both the basal level and the recovery of MGMT activity. Indeed, we have shown in an earlier study that a combined treatment of MMR-proficient leukemic blasts with TMZ and noncytotoxic concentrations of cisplatin resulted in synergistic inhibitory effects on proliferation (Piccioni et al., 1995).

In conclusion, this report demonstrates that cisplatin is able to decrease MGMT levels in Jurkat cells, probably via the inhibition of MGMT gene transcription. Moreover, preexposure of Jurkat cells to cisplatin markedly reduces the rate of MGMT recovery after inactivation by TMZ treatment. This suggests that the clinical efficacy of triazene compounds might be improved by combination with cisplatin using appropriate doses and schedules of administration.