Antifolate Activity of Pyrimethamine Enhances Temozolomide-Induced Cytotoxicity in Melanoma Cells

Published OnlineFirst May 12, 2009; DOI: 10.1158/1541-7786.MCR-08-0263

Ming Chen, Iman Osman, and Seth J. Orlow

The Ronald O. Perelman Department of Dermatology, Departments of Medicine and Cell Biology, and the New York University Cancer Institute Clinical Cancer Center, New York, New York

Abstract Introduction Most metastatic melanoma patients fail to respond to There is an urgent need to identify new therapeutic targets available therapy, underscoring the need to develop and develop new agents for the treatment of melanoma (1, 2). more effective treatments. We screened 2,000 compounds Although dacarbazine has historically been the standard treat- from the Spectrum Library in human melanoma cell lines to ment for melanoma (3, 4), in recent years, it has been replaced identify compounds that enhanced the cytotoxic effect of by its analogue temozolomide (TMZ), a well-tolerated, orally temozolomide, a drug used to treat metastatic melanoma. bioavailable drug that can efficiently cross the blood-brain bar- Screening was done with the temozolomide-resistant rier. TMZ has proven useful in the treatment of metastatic mel- melanoma cell line SK-MEL-19, and six compounds were anoma to the brain in addition to glioblastoma and anaplastic identified that had little or no inherent cytotoxicity but glioma (5, 6). However, the response rate of melanoma to dacar- significantly enhanced growth-inhibition by temozolomide. bazine or TMZ is only ∼20%. Attempts to improve efficacy of These compounds were tested in five additional melanoma TMZ through combination therapy with drugs such as cisplatin, cell lines. Cell proliferation and death assays were used to docetaxel, thalidomide, IFN, and irinotecan have shown no clear compare the efficacy of single agent temozolomide versus benefits in response rates, median time to progression, or overall combination treatments. Effects of combination treatment on survival (7–12). levels of DNA double-strand breaks, the DNA repair protein Chemical genetics offers a powerful tool for identifying O6-methylguanine-DNA-methyltransferase, apoptosis therapeutic candidates and targets (13–15). We have previously [measured by cleaved caspase-3 and poly(ADP-ribose) used this screening methodology to identify novel compounds polymerase], and cell cycle were examined. Pyrimethamine, and targets involved in the regulation of mammalian pigmenta- an antiparasitic, sensitized melanoma cells to temozolomide. tion (16–19), in the evasion of arsenic trioxide resistance in Temozolomide combined with Pyrimethamine synergistically melanoma cells (20), and in the identification of agents specif- inhibited cell proliferation in melanoma cells with ically cytotoxic to melanoma (21). combination index values of 0.7 or less. In addition, In this study, we extended our chemical genetic approach to combination treatment induced cell cycle arrest and identify new compounds that enhance the cytotoxicity of TMZ increased both DNA damage and apoptosis. The increase in against resistant melanoma. We screened a commercially avail- cell death due to combination treatment was rescued by able library of 2,000 compounds, including natural products leucovorin. Other folate antagonists were also effective and marketed drugs in TMZ-resistant melanoma cells. Of enhancers of temozolomide-induced cytotoxicity, and the the six compounds identified as potential enhancers of TMZ- effects of antifolates were also evident in gliomas. induced cytotoxicity, pyrimethamine (PYR) was deemed the Our screening approach led to the identification of most promising candidate due to its established use as an oral Pyrimethamine, an orally available drug that efficiently treatment for malaria and toxoplasmosis of the central nervous crosses the blood-brain barrier, as a potent enhancer system. PYR synergistically enhances the growth inhibitory and of the efficacy of temozolomide as an antineoplastic apoptosis-inducing effects of TMZ. Moreover, we further show agent via inhibition of folate metabolism. (Mol Cancer Res that the antifolate activity of PYR underlies these effects. Our 2009;7(5):703–12) results suggest that chemical genetics can identify novel drug combinations with potential clinical utility in the treatment of melanoma, and specifically show that the combination of TMZ with antifolates, in particular PYR, results in unexpected, Received 6/5/08; revised 12/18/08; accepted 1/5/09; published OnlineFirst 5/12/09. proapoptotic effects on melanoma. Grant support: A grant from Jacklyn and Miguel Bezos. 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. Results Note: Supplementary data for this article are available at Molecular Cancer Identification of Compounds That Enhance TMZ-Induced Research Online (http://mcr.aacrjournals.org/). Requests for reprints: Seth J. Orlow, Dermatology Room H-100, NYU School Cytotoxicity by Screening the Spectrum Collection Library of Medicine, 560 First Avenue, New York, NY 10016. Phone: 212-263-5245; Among the 6 tested human melanoma cell lines, SK- Fax: 212-263-8752. E-mail: [email protected] Copyright © 2009 American Association for Cancer Research. MEL-19 was identified as the most resistant to TMZ, with doi:10.1158/1541-7786.MCR-08-0263 an IC50 at ∼89 μg/mL. A commercial library of 2,000 drugs

Mol Cancer Res 2009;7(5). May 2009 703 Downloaded from mcr.aacrjournals.org on September 23, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst May 12, 2009; DOI: 10.1158/1541-7786.MCR-08-0263

704 Chen et al.

and natural products was screened to identify novel com- exhibited a dose-dependent inhibition of SK-MEL-19 cell pro- pounds that enhance TMZ-induced growth inhibition in liferation (Fig. 2A and B). Combination treatment yielded SK-MEL-19. Screening followed by 3-(4,5-dimethylthiazol- significantly greater growth inhibition than either agent alone 2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was done (Fig. 2C). The combination index (CI) was used to confirm in 96-well plates at final library compound concentration of and quantify the synergy observed with TMZ and PYR. CI 1 μmol/L, in either the absence or presence of 50 μg/mL values were <0.7 at all doses (range from 25-200 μg/mL; TMZ (Fig. 1A). Six compounds (benzyl isothiocyanate, Fig. 2D), indicating that synergy occurred between TMZ and chlorhexidine, cloxyquin, 3, 4-dimethoxydalbergione, PYR, PYR across a broad range of concentrations. Unless otherwise and triamterene) were identified as potent enhancers of noted, 25 μg/mL TMZ and 0.5 μmol/L PYR were used in sub- TMZ-induced cytotoxicity, with a minimum 50% greater sequent experiments as these represent clinically achievable growth inhibition relative to TMZ alone (Fig. 1B). These concentrations of each agent. positive candidates were reconfirmed by testing at a lower The effect of TMZ/PYR combination treatment was further final concentration of 0.5 μmol/L. PYR was chosen for fur- examined by treating five additional, highly TMZ-resistant ther study due to its long-established and safe use in humans melanoma cell lines (SK-MEL-100, 173, 192, 451-Lu, and as an antimalarial drug. WM278) with either TMZ alone or TMZ/PYR in combination. As shown in Table 1, the addition of 0.5 μmol/L PYR resulted Combination of TMZ and PYRIs Synergistic for Inhibition in a significant decrease in TMZ IC50 in all 6 tested melanoma of Cell Proliferation cell lines (Table 1). Effects of 0.5 μmol/L PYR alone on each We examined the effects of either TMZ alone, PYR alone, cell line had also been included in Table 1. Effects of PYR or combined treatment with both agents (TMZ/PYR) on mela- alone were shown as the percentage of growth inhibition of noma cell proliferation (Fig. 2). Both TMZ and PYR alone PYR versus equimolar DMSO.

FIGURE 1. Identification of compounds that enhance TMZcytotoxicity in melanoma cells by screening the Spectrum Collection library. A. Screening of the Spectrum Collection library to identify novel agents that enhance TMZefficacy in cultured melanoma cells. Twenty-four hours after seeding into 96-well plates, cell cultures in 1 plate was treated with each library compound alone at 1 μmol/L, whereas cells in the parallel plate were treated with a combination of 1 μmol/L library compound and 50 μg/mL TMZ. Screening was done with SK-MEL-19 cells in 96-well plates by MTT assay after a 72 h treatment. Positive candidates were later reconfirmed in triplicate. B. Chemical structures of six hit compounds confirmed enhanced TMZactivity and induced greater inhibition of cell growth.

Mol Cancer Res 2009;7(5). May 2009 Downloaded from mcr.aacrjournals.org on September 23, 2021. © 2009 American Association for Cancer Research. Published OnlineFirst May 12, 2009; DOI: 10.1158/1541-7786.MCR-08-0263

Pyrimethamine Enhances Temozolomide Cytotoxicity 705

FIGURE 2. Combination of TMZand PYR is synergistic for inhibition of cell proliferation. Dose-response diagrams of SK-MEL-19 cells exposed to TMZ (A) or PYR (B) alone. Melanoma cells were exposed to the indicated concentration of TMZor PYR for 72 h. Growth inhibition was evaluated by MTT assay. Survival relative to the DMSO control is presented as mean ± SE of at least three experiments, with viability of untreated cells set at 100%. C. PYR enhances TMZgrowth inhibition in SK-MEL-19. Cells were treated with indicated concentrations of TMZor DMSO, either with or without PYR at 0.5 μmol/L for 72 h. Cell viability was determined by MTT assay. *, P < 0.05; **, P < 0.01, compared with TMZ. D. Synergy analysis was done based on data from C using CalcuSyn software. CI values: 1, additivity; <1, synergism.

TMZ Alters PYR-Induced Cell Cycle Arrest treated with TMZ + DMSO, PYR + DMSO,orTMZ + PYR.No To address whether the antiproliferative effect of TMZ/PYR obvious changes were observed in cells treated with TMZ + combination in melanoma is associated with cell cycle regula- DMSO (Fig. 3). PYR at the final concentration of 0.5 μmol/L tion, DNA cell cycle analysis was done upon SK-MEL-19 cells increased the number of S-phase cells, accompanied by a

Table 1. Effects of PYR Alone, TMZ Alone, and Their Combination in a Panel of Melanoma Cell Lines

Cell Line Growth Inhibitory Effect of PYR (%)* IC50 (μg/mL)

TMZ TMZ/PYR (0.5 μmol/L) P

SK-MEL-19 18.3 ± 7.3 88.5 ± 13.0 13.0 ± 0.1 0.001 SK-MEL-192 33.9 ± 3.0 57.4 ± 13.2 2.7 ± 2.5 0.015 SK-MEL-100 15.4 ± 3.9 62.1 ± 6.3 41.4 ± 6.4 0.047 SK-MEL-1736.2 ± 5.380.0 ± 3.058.4± 1.0 0.013 451-Lu 8.0 ± 7.6 101.8 ± 11.0 78.3± 13.1 0.001 WM278 9.4 ± 4.0 56.9 ± 7.7 39.7 ± 2.1 0.004

*Percentage of growth inhibition versus equimolar DMSO for 0.5 μmol/L PYR.

706 Chen et al.

FIGURE 3. TMZ/PYR induces cell cycle arrest in melanoma cells. Cells were incubated in either medium alone, medium containing TMZ (25 μg/mL) plus DMSO, PYR (0.5 μmol/L) plus DMSO, or TMZplus PYR combination for 72 h before harvesting. The harvested cells were stained with propidium iodide and analyzed with fluorescence-activated cell sorting to determine the proportion of cells in each phase of the cell cycle. A. Representative flow cytometry histograms obtained from treated cells. B. Quantitation of cell population distribution over the various stages of the cell cycle. *, P < 0.05; **, P < 0.01, compared with control (CTR).

corresponding decrease in the proportion of cells in G0-G1. Ac- sub-G1 peaks that are indicative of apoptosis. These data cumulation of cells in the G2 phase and S phase was observed suggest that both cell cycle arrest and cell loss due to apoptosis in the cells treated with the combination of both compounds on could be responsible for the antiproliferative effects of TMZ/ day 3(Fig. 3).Cell accumulation in S phase and G 2 phase after PYR. TMZ + PYR treatment suggests that a significant proportion of The concentration of TMZ (25 μg/mL or ∼100 μmol/L) cells was unable to complete DNA synthesis correctly and cell used in this study was relatively low and close to clinically division was interrupted. At the same time, DMSO + PYR and achievable serum levels during chemotherapy (100 μmol/L; TMZ + PYR treatments induced the formation of hypodiploid ref. 22). At a 25 μg/mL concentration, TMZ had no effect on

Pyrimethamine Enhances Temozolomide Cytotoxicity 707

cell cycle progression in melanoma cells under the test condi- Antifolate Activity of PYREnhances TMZ Efficacy in tions we used. Melanoma and Glioma Cells Among the six “hits” identified from library screening, both PYR and triamterene are known to have some antifolate activ- PYREnhances Cell Death and Apoptosis Induced by ity. To investigate whether PYR enhances TMZ efficacy TMZ through its activity as an antifolate, 10 additional antifolate To investigate whether treatment with PYR can enhance the compounds were tested to determine whether they, too, en- induction of SK-MEL-19 cell death by TMZ, trypan blue stain- hanced TMZ-induced growth inhibition. Because TMZ is a ing was used to determine the ratio of dead cells versus total standard chemotherapy drug used to treat gliomas, the effects cells. At clinically relevant concentrations, TMZ induced death of antifolates were also tested in the TMZ-resistant glioma in ∼15% of cells (Fig. 4A), whereas PYR at 0.5 μmol/L had cell line LN-18. Similar synergistic effects on inhibition of only a limited effect on cell death. However, the combination of cell growth were observed with all antifolates studied when TMZ/PYR induced an ∼2-fold increase in cell death in SK- combined with TMZ in both melanoma and glioma cells MEL-19 (Fig. 4A). These results are consistent with the (Fig. 6A). For example, the IC of TMZ was reduced almost MTT assay data (Fig. 2C) and suggest that the TMZ/PYR com- 50 7-fold (from ∼89 to ∼13 μg/mL) in SK-MEL-19 cells when bination treatment significantly enhances the inhibition of they were cotreated with each of the 5 different antifolates melanoma cell growth and induction of cell death by TMZ. [etoprine (DDEP), metoprine (DDMP), cycloguanil, methotrex- To determine whether TMZ, PYR, or TMZ/PYR induce ate (MTX), and PYR; Fig. 6A]. The CI values were calculated cell death by the same pathway, cellular levels of activated using CalcuSyn software and are summarized in Table 2. The caspase-3and PARP were assessed by immunoblot analysis data suggest that combinations of TMZ and antifolates are with an antibody against the cleaved (active) form of the en- mostly synergistic in terms of cell growth inhibition in both zymes. No obvious changes were observed in cells treated with melanoma and glioma cell lines. TMZ alone, whereas TMZ/PYR treatment increased the level Leucovorin (N-5-formyltetrahydrofolate, LV, folinic acid), a of cleaved caspase-3and cleaved PARP in SK-MEL-19 cells reduced form of folate, is used widely to specifically reverse the (Figs. 4B and 6D). These results suggest that the proapoptotic toxic effects of antifolates (26). To further confirm that PYR- effect of TMZ/PYR combination treatment in melanoma cells enhanced TMZ efficacy in melanoma cells was due to the anti- is mediated by a caspase-dependent pathway. folate activity of PYR, 10 μmol/L LV was added to cells treated with PYR alone, TMZ alone, and TMZ/PYR in combination. TMZ/PYRTreatment Increases DNA Damage in Melanoma LV reversed PYR-induced cytotoxicity but not TMZ-induced Cells DNA damage is a well-characterized, initial upstream event in apoptotic cell death (23). Phosphorylation of histone H2AX is one of the earliest responses to strand breakage and is accepted as an early marker for DNA double strand breaks (DSB; ref. 24). We assessed levels of phosphorylated histone H2AX to determine whether TMZ/PYR combination treatment caused more DSBs. The topoisomerase inhibitor camptothecin, which induces DSBs by stalling DNA replication forks, was used as positive control, whereas untreated cells served as negative control. A trace level of phosphorylated histone H2AX was observed in untreated cells and cells treated with TMZ alone (Fig. 5). PYR treatment induced a modest increase in the level of phosphorylated histone H2AX, suggesting that PYR induces cell death through its activity as an antifolate. A strong increase in the level of phosphorylated histone H2AX was observed in cells treated with TMZ/PYR (Fig. 5). These results show that TMZ/PYR combination treatment generates more DSBs, which further induces cell death. Because the O6-meG lesion induced by TMZ treatment can be directly removed by the DNA repair protein O6-methylguanine-DNA-methyltransferase (MGMT), the cellular level of MGMT is one of the main contributors to TMZ resistance (25). Western blot analysis was done to determine whether TMZ, PYR, and TMZ/PYR treatments alter MGMT protein levels. No alteration in MGMT protein level was observed FIGURE 4. PYR increases cell death and apoptosis induced by TMZin in cells treated with either agent alone or in combination SK-MEL-19. Melanoma cells were exposed to the indicated concentration (Supplementary Fig. S1), suggesting that the synergistic, of TMZ, PYR, or both for 72 h. A. Cell death evaluated by trypan blue dye exclusion. *, P < 0.05, compared with 2 or 3. B. Cleaved caspase-3 was antiproliferative effect exerted by the TMZ/PYR combination determined by Western blot analysis. Levels of actin protein served as is inde‐pendent of effects on MGMT levels. loading control. Densitometry values for cleaved caspase-3 are included.

708 Chen et al.

ditional compounds in the same class can be tested, providing useful information regarding structure-activity relationships. Our screen of the Spectrum library was the first to identify PYR, a lipophilic dihydrofolate reductase inhibitor with clinical efficacy as an antimalarial drug, as a compound that significantly enhanced TMZ-induced cytotoxicity in melanoma cells. At their clinically achievable concentrations, TMZ and PYR showed synergistic activity in both melanoma and gliomas cells. In addition, PYR also enhanced cell killing induced by DTIC, a metabolite of TMZ that acts via the same FIGURE 5. TMZ/PYR combination treatment increases DNA damage mechanism (data not shown). However, given that none of in SK-MEL-19. Western blot showing the increase in H2AX phosphoryla- other antimalarial compounds we tested (including artemisi- tion (gamma-H2AX) after treatment with TMZ(25 μg/mL) and PYR (0.5 μmol/L) for 2 h in SK-MEL-19 cells. Camptothecin (CPT)treatment nin, chloroquine, primaquine, quindine, and quinine) could at 10 μmol/L served as a positive control (+). Negative control (−) is lysate mimic the effect of PYR on enhancing the TMZ or DTIC of untreated cells. Levels of actin protein served as loading control. efficacy in melanoma cells, we reasoned that PYR does not enhance TMZ efficacy through its activity as an antimalarial compound. cytotoxicity (Fig. 6B). Furthermore, LV rescued both the inhi- PYR is structurally related to folate and is known as an ef- bition of cell growth (Fig. 6B) and the increase in apoptotic fective dihydrofolate reductase inhibitor, particularly in proto- death induced by TMZ/PYR as measured by the proteolytic zoa such as malaria and toxoplasma. Dihydrofolate reductase cleavage of PARP by caspase-3(Fig. 6D). Lastly, effect of inhibitors (or antifolates) have been studied for many years LV on reversing the growth inhibition caused by TMZ/PYR as anticancer and antineoplastic agents. The disruption of folic in combination is dose dependent (Fig. 6C). Taken together, acid metabolism has long been known to inhibit cell growth. these results confirmed that PYR enhances TMZ-induced cyto- Folic acid is essential for the de novo synthesis of the nucleo- toxicity through its function as an antifolate, most likely by vir- side thymidine, which is required for DNA synthesis. Thus, tue of its action as a competitive inhibitor of dihydrofolate antifolates have greater, selective toxicity on rapidly dividing reductase. cells such as tumor cells. MTX, the most studied classic antifolate, provided the first cure in choriocarcinoma more than six decades ago and is effective in many malignancies (35). Anti- Discussion folate drug development has then focused on agents to over- Many attempts have been made to enhance TMZ efficacy come MTX resistance and reduce the toxicity. Currently, through combination treatments with limited success (7–12). second or third generation antifolates were tailored to have Agents tested in combination with TMZ included those that in- more efficient uptake or better polyglutamylation. Among terfere with DNA repair pathways such as the MGMT inhibitor them, the lipophilic antifolate ALIMTA (pemetrexed) is regis- O6-benzylguanine (27) or the base excision repair inhibitor tered for mesothelioma and non–small cell lung cancer. The li- methoxyamine (28, 29). Alternatively, TMZ has been com- pophilic antifolate PYR exerts strong proapoptotic activity in bined with cytotoxic agents, radiotherapy, immunotherapy, addition to its antiprotozoal effects. Several PYR analogues, and antiangiogenic agents in clinical trials. However, compared such as DDEP and DDMP, have also been investigated as anti- with the effect of TMZ alone, no clear benefits have been tumor agents (36). shown from these combinations (30). It has been reported that PYR can induce apoptosis and S Previous studies have shown that TMZ at high concentra- phase accumulation in various cell lines (37–40). PYR has tions can induce cell death and apoptosis in both melanoma been shown to reduce melanoma growth in a severe combined and glioma cells (31, 32) and, specifically, that TMZ induces immune-deficient mouse model (40). Consistent with previous G2 arrest in glioma cells (33). However, a recent report sug- reports (39, 40), we observed that PYR at a concentration of gests that TMZ induces senescence but not apoptosis in human 0.5 μmol/L increased the number of cells in S phase (Fig. 3). melanoma cells (34). In this study, we used a clinically relevant Folate deficiency in cell culture has been reported to induce concentration of TMZ and showed that it had little effect on an excess of strand breaks in DNA (41, 42), and preliminary apoptosis, DNA damage, or cell cycle arrest in melanoma cells. data indicate that genome-wide DNA stand breaks are related These results reflect the very modest clinical performance of to folate status (43–45). In this study, we observed an in- TMZ as a chemotherapeutic agent. crease in DSBs (Fig. 5) in cells treated with PYR/TMZ com- To enhance TMZ efficacy in melanoma, we used a chemical pared with each agent alone. These results suggest that PYR genetic approach to identify novel combinations by screening enhances TMZ efficacy through its antifolate activity. We fur- the Spectrum library. The advantages of this library are 3-fold. ther confirmed this mechanism of action by showing that a First, clinical development is shortened as many compounds in number of other antifolates also enhanced inhibition of cell the library have been used for other indications in humans. Sec- growth induced by TMZ and that the effects of antifolates ond, based on the known function and mechanism of action of could be rescued by LV (Fig. 6). these compounds, one can readily test whether the chemosen- MGMT plays a crucial role in the repair of DNA damage in- sitization effects of these agents are carried out via the same duced by TMZ treatment because it directly removes O6meG mechanism. Finally, after identification of a hit compound, ad- (24). The expression level of MGMT is also susceptible to

Pyrimethamine Enhances Temozolomide Cytotoxicity 709

epigenetic silencing (46). Although our data showed that TMZ/ Many other pathways are involved in repairing DNA dam- PYR treatment induced more DNA damage through the forma- age induced by methylating agents, which including base exci- tion of DSBs (Fig. 5), no changes in MGMT protein level were sion repair, homologous recombination, polymerase bypass, observed in cells treated with PYR, TMZ, or PYR/TMZ. To- and mismatch repair. For example, previous studies have indi- gether, this suggests that in melanoma, the enhancement of cated that inhibition of MGMT increased TMZ sensitivity only TMZ efficacy by PYR is independent of MGMT level. This in mismatch repair–proficient cells but not in mismatch repair– observation was further confirmed by the clinically relevant deficient cells (47). The major pathway and proteins involved finding that PYR-induced sensitization of cells to TMZ treat- in DNA repair are cell type dependent (48). In this study, we ment was observed in both MGMT-expressing and MGMT- show that MGMT is not a target of PYR-induced TMZ sensi- negative cells. tization.

FIGURE 6. PYR enhances TMZefficacy by inhibiting folate metabolism. A. Growth inhibitory effects of PYR, DDEP, DDMP, cycloguanil (CYC), and the classic antifolate MTX on TMZresponse curve were examined by MTT assay in both melanoma and glioma cells. In SK-MEL-19 cells, 0.005 μmol/L PYR, DDEP, DDMP and 0.001 μmol/L MTX showed significant enhancement of TMZ-induced growth inhibition, whereas 0.05 μmol/L PYR, DDEP, DDMP and 0.005 μmol/L MTX showed a similar effect in LN-18 cells. *, P < 0.05; **, P < 0.01. B. Cytotoxic effect of combination of TMZ(25 μg/mL) and PYR (0.5 μmol/L) can be rescued by LV (10 μmol/L) as examined by MTT assay. *, P < 0.05 between PYR and PYR/LV; **, P < 0.01 between TMZ/PYR and TMZ/PYR/LV. C. The effects of LV on reversing the cytotoxicity caused by TMZ/PYR combination are dose dependent. **, P < 0.01 compare to TMZ/PYR/LV when LV is 10 μmol/L. No significant difference compared with TMZ/PYR/LV when LV is 0.5 or 2 μmol/L. D. Levels of PARP and cleaved PARP expression, and levels of actin protein served as loading control.

710 Chen et al.

Table 2. Summary of CI Values for TMZ and Antifolates were seeded at 3,000 cells per well in 96-well plates and al- lowed to attach overnight. Library compounds were then added Cell Line Dose CI to cell cultures at a final concentration of 1 μmol/L, either alone μ MTX/TMZ DDEP/TMZ PYR/TMZ or in combination with 50 g/mL TMZ. After 72 h, cellular proliferation was measured. The plates were read with a micro- SK-MEL-19 12.5 0.292 0.284 0.33 plate reader (Bio-Rad madel 550) by measuring the absorbance 25 0.395 0.5 0.536 of converted MTT at 490 nm. 50 0.566 0.731 0.728 100 0.767 0.686 0.685 Drugs and Reagents. TMZ (NSC 362856) and the lipophil- 150 1.028 0.684 0.629 ic folate analogues DDMP, DDEP, and cycloguanil were kindly LN-18 25 0.6731.087 0.569 50 0.688 0.787 0.687 provided by the National Cancer Institute. PYR, MTX, and LV 75 0.729 0.77 0.779 were purchased from Sigma-Aldrich Chemical Company, Inc. 100 0.655 0.776 0.75 All drugs were dissolved in DMSO. 150 0.737 0.778 0.731 Cell Proliferation Assay and Viability Assay Effect of treatment with TMZ and/or PYR on cell growth The chemical genetics approach we used to identify com- and cytotoxicity in melanoma and glioma cells was assessed pounds that can overcome TMZ resistance may prove useful using colorimetric MTT assay (CellTiter 96 AQueous nonra- in identifying additional compounds to improve the efficacy dioactive cell proliferation assay; Promega), according to man- of other chemotherapeutic agents. For instance, we recently ufacturer's instruction. The absorbance of control cells exposed used a similar approach to overcome arsenic trioxide resistance to vehicle alone defined 100% cell growth, and the effect of in melanoma cells (20). A chemical genetics approach has also drugs on cellular proliferation was expressed as a percentage been used successfully to identify benzimidazoles, a class of of cell growth relative to control cells. Cell viability was also structurally related, tubulin-disrupting drugs, as agents prefer- determined by trypan blue dye exclusion method. entially toxic to melanoma cells compared with melanocytes (21). In addition, this approach could be used to reveal addi- Synergy Analysis of Combined Drug Effects tional novel mechanisms and targets for cancer treatment. In Drug synergy was determined by CI methods derived from particular, our findings strongly suggest that targeting folate Chou-Talalay equations (49) using the CalcuSyn software (Bio- metabolism, especially through dihydrofolate reductase inhibi- soft). Using cell proliferation assay and computerized software tion, may prove to be an effective strategy to improve TMZ data, CI values are generated between two drugs. A CI of 1 efficacy in melanoma. indicates an addictive effect between two agents, whereas a PYR is bioavailable after oral administration and has also CI of <1 indicates synergy. been safely administered to malarial patients for prolonged per- iods with limited side effects. Moreover, PYR differs from the Western Blot Analysis and Antibodies classic antifolates such as MTX in that it is a lipophilic antifo- Cells were harvested in extraction buffer [1% Triton X-100, late, allowing it to diffuse readily across the cell membrane 50 mmol/L Tris, 2 mmol/L EDTA, and 150 mmol/L NaCl (pH without a transporter, and to easily cross the blood-brain barri- 7.5)] containing complete protease inhibitor mixture (Roche). er. Taken together, these characteristics make PYR an attractive The lysates were centrifuged at 14,000 rpm for 10 min at candidate for improving the efficacy of TMZ in melanoma 4°C. Bio-Rad protein assay reagent (Bio-Rad) was used to (including melanoma metastatic to the brain). measure the protein concentrations. Proteins (20 μg) were sep- arated by 8%, 12%, or 15% SDS-polyacrylamide (SDS-PAGE) gel and transferred to Immobilon-P membranes (Millipore). Im- Materials and Methods munoreactive bands were visualized using ECL-plus detection Screening of Spectrum Library in Melanoma Cells reagent and X-OMAT processing. Densitometry values were Tumor Cell Lines. Human malignant melanoma cell lines calculated using ImageQuant TL software. (SK-MEL-19, SK-MEL-100, SK-MEL-173, and SK-MEL- Antibodies. Anti–phospho-H2A.X (Ser 139) antibody 192) were a gift from Dr. Alan Houghton (Memorial Sloan- was purchased from Upstate Cell Signaling Solutions, anti- Kettering Cancer Center, New York, NY). Cells were cultured MGMT antibody was from Chemicon, and antibodies against in DMEM supplemented with 10% FCS, 2 mmol/L L-gluta- cleaved capase-3, PARP, and β-actin were from Cell Signaling mine, and 1% penicillin-streptomycin at 37°C. Melanoma cell Technology. lines (451-Lu and WM278) and glioma cell line LN-18 were obtained from the American Type Culture Collection. 451Lu Cell Cycle Analysis and WM278 were cultured in T2% tumor medium consisting Cell cycle distribution was determined by staining DNA of MCDB 153medium supplemented with 5 μg/mL insulin, with propidium iodide (Sigma). Briefly, cells were treated with 2% FCS, and 20% L-15 Leibovitz medium. TMZ plus DMSO, PYR plus DMSO, or TMZ/PYR combina- Library Screening. The library used in this study was The tion for 72 h and then harvested. Cells were then washed and Spectrum Collection (MicroSource Discovery Inc.). The fixed in 70% ethanol on ice for 30 min. After centrifugation, 2,000 compounds in this library are marketed drugs, other bi- the cell pellets were washed and resuspended in phosphate- ologically active small molecules, or natural products (supplied citrate buffer. Cells were then treated with RNase and stained at a concentration of 10 mmol/L in DMSO). SK-MEL-19 cells with propidium iodide. DNA content was analyzed on a

Pyrimethamine Enhances Temozolomide Cytotoxicity 711

cytofluorimeter by fluorescence-activated cell sorting analysis 21. Doudican N, Rodriguez A, Osman I, Orlow SJ. Mebendazole induces apoptosis via Bcl-2 inactivation in chemoresistant melanoma cells. Mol Cancer Res (FACScan). 2008;6:1308–15. 22. Hammond LA, Eckardt JR, Kuhn JG, et al. A randomized phase I and phar- Statistical Analysis macological trial of sequences of 1,3-bis(2-chloroethyl)-1-nitrosourea and temo- Data were expressed as means ± SD. Statistical analysis was zolomide in patients with advanced solid neoplasms. Clin Cancer Res 2004;10: done using the Student's Paired t test. The criterion for statis- 1645–56. tical significance was established as a P value of <0.05. 23. Lips J, Kaina B. DNA double-strand breaks trigger apoptosis in p53-deficient fibroblasts. Carcinogenesis 2001;22:579–85. 24. Jackson SP. Sensing and repairing DNA double-strand breaks. Carcinogen- Disclosure of Potential Conflicts of Interest esis 2002;23:687–96. No potential conflicts of interest were disclosed. 25. Gerson SL. Clinical relevance of MGMT in the treatment of cancer. J Clin Oncol 2002;20:2388–99. 26. Keshava C, Keshava N, Whong WZ, Nath J, Ong TM. Inhibition of meth- References otrexate-induced chromosomal damage by folinic acid in V79 cells. Mutat Res 1. Atkins MB. The treatment of metastatic melanoma with chemotherapy and 1998;397:221–8. – biologics. Curr Opin Oncol 1997;9:205 13. 27. Dolan ME, Pegg AE. O6-benzylguanine and its role in chemotherapy. Clin 2. Balch CM, Soong SJ, Gershenwald JE, et al. Prognostic factors analysis of Cancer Res 1997;3:837–47. 17,600 melanoma patients: validation of the American Joint Committee on Can- 28. Fishel ML, He Y, Smith ML, Kelley MR. Manipulation of base excision re- – cer Melanoma Staging System. J Clin Oncol 2001;19:3622 34. pair to sensitize ovarian cancer cells to alkylating agent temozolomide. Clin Can 3. Markovic SN, Erickson LA, Rao RD, et al. Malignant melanoma in the 21st Res 2007;13:260–7. century, part 2: staging, prognosis, and treatment. Mayo Clin Proc 2007;82: 29. Yan L, Bulgar A, Miao Y, et al. Combined treatment with temozolomide and – 490 513. methoxyamine: blocking apurininc/pyrimidinic site repair coupled with targeting 4. Wolchok JD, Saenger YM. Current topics in melanoma. Curr Opin Oncol topoisomerase IIα. Clin Cancer Res 2007;13:1532–9. – 2007;19:116 20. 30. Quirt I, Verma S, Petrella T, Bak K, Charette M. Temozolomide for the 5. Mutter N, Stupp R. TMZ: a milestone in neuro-oncology and beyond? Expert treatment of metastatic melanoma: a systematic review. Oncologist 2007;12: Rev Anticancer Ther 2006;6:1187–204. 1114–23. 6. Weller M, Steinbach JP, Wick W. TMZ, a milestone in the pharmacotherapy of 31. D'Atri S, Tentori L, Lacal PM, et al. Involvement of the mismatch brain tumors. Future Oncol 2005;1:747–54. repair system in temozolomide-induced apoptosis. J Mol Pharmacol 1998; – 7. Bafaloukos D, Gogas H, Georgoulias V, et al. Temozolomide in combination 54:334 41. with docetaxel in patients with advanced melanoma: a phase II study of the Hel- 32. Roos WP, Batista LF, Naumann SC, et al. Apoptosis in malignant glioma lenic Cooperative Oncology Group. J Clin Oncol 2002;20:420–5. cells triggered by the temozolomide-induced DNA lesion O6-methylguanine. On- – 8. Atkins MB, Gollob JA, Sosman JA, et al. A phase II pilot trial of concurrent cogene 2007;26:186 97. biochemotherapy with cisplatin, vinblastin, temozolomide, interleukin, and 33. Hirose Y, Berger MS, Pieper RO. Abrogation of the Chk1-mediated G IFN-α 2B in patients with metastatic melanoma. Clin Cancer Res 2002;8: (2) checkpoint pathway potentiates temozolomide-induced toxicity in a p53- 3075–81. independent manner in human glioblastoma cells. Cancer Res 2001;61: – 9. Hwu WJ, Krown SE, Menell JH, et al. Phase II study of temozolomide plus 5843 9. thalidomide for the treatment of metastatic melanoma. J Clin Oncol 2003;21: 34. Mhaidat NM, Zhang XD, Allen J, Avery-Kiejda KA, Scott RJ, Hersey P. 3351–6. Temozolomide induces senescence but not apoptosis in human melanoma cells. – 10. Silvani A, Eoli M, Salmaggi A, et al. Phase II trial of cisplatin plus temozo- Br J Cancer 2007;97:1225 33. lomide, in recurrent and progressive malignant glioma patients. J Neurooncol 35. Hertz R, Li MC, Spencer DB. Effect of methotrexate therapy upon 2004;66:203–8. choriocarcinoma and chorioadenoma. Proc Soc Exp Biol Med 1956;93: – 11. Gogas H, Polyzos A, Stavrinidis I, et al. Temozolomide in combination with 361 6. celecoxib in patients with advanced melanoma. A phase II study of the Hellenic 36. Murphy ML, Ellison RR, Karnofsky DA, Burchenal JH. Clinical effects of Cooperative Oncology Group. Ann Oncol 2006;17:1835–41. the dichloro and mono-chlorophenyl analogues of diamino pyrimidine: antago- – 12. Fountzilas G, Karkavelas G, Kalogera-Fountzila A, et al. Post-operative nists of folic acid. J Clin Invest 1954;33:1388 96. combined radiation and chemotherapy with temozolomide and irinotecan in pa- 37. James SJ, Pogribny IP, Pogribna M, Miller BJ, Jernigan S, Melnyk S. Me- tients with high-grade astrocytic tumors. A phase II study with biomarker evalu- chanisms of DNA damage, DNA hypomethylation, and tumor progression in the ation. Anticancer Res 2006;26:4675–86. folate/methyl-deficient rat model of hepatocarcinogenesis. J Nutr 2003;133: – 13. Stockwell BR. Chemical genetics: ligand-based discovery of gene function. S3737 40. Nat Rev Genet 2000;1:116–25. 38. Lin H, Chen C, Tsai W, Yen J, Liu H. In vitro folate deficiency induces ap- 14. Schreiber SL. The small-molecule approach to biology. Chem Eng News optosis by a p53, Fas (Apo-1, CD95) independent, bcl-2 related mechanism in 2003;81:51–61. phytohaemagglutinin-stimulated human peripheral blood lymphocytes. Br J Nutr 2006;95:870–8. 15. Smukste I, Stockwell BR. Advances in chemical genetics. Annu Rev Geno- 39. Huang RF, Ho YH, Lin HL, Wei JS, Liu TZ. Folate deficiency induces a cell mics Hum Genet 2005;6:262–86. cycle-specific apoptosis in HepG2 cells. J Nutr 1999;129:25–31. 16. Williams D, Jung DW, Khersonsky S, Heidary N, Chang YT, Orlow 40. Giammarioli AM, Maselli A, Casargrande A, et al. Pyrimethamine induces SJ. Identification of compounds that bind mitochondrial F1F0 ATPase by apoptosis of melanoma via a caspase and cathepsin double-edged mechanism. screening a triazine library for correction of albinism. Chem Biol 2004; Cancer Res 2008;68:5291–300. 11:1251–9. 41. Branda RF, Blickensderfer DB. Folate deficiency increases genetic damage 17. Ni-Komatsu L, Leung JK, Williams D, et al. Triazine-based tyrosinase in- caused by alkylating agents and γ-irradiation in Chinese hamster ovary cells. hibitors identified by chemical genetic screening. Pigment Cell Res 2005;18: Cancer Res 1993;53:5401–8. 447–53. 42. Courtemanche C, Huang AC, Elson-Schwab I, Kerry N, Ng BY, Ames BN. 18. Snyder JR, Hall A, Ni-Komatsu L, Khersonsky SM, Chang YT, Orlow SJ. Folate deficiency and ionizing radiation cause DNA breaks in primary human Dissection of melanogenesis with small molecules identifies prohibition as a reg- lymphocytes: a comparison. FASEB J 2004;18:209–11. ulator. Chem Biol 2005;12:477–84. 43. Lathrop-Stern L, Selhub J, Choi SW, Mason JB. Genomic DNA stand breaks 19. Ni-Komatsu L, Orlow SJ. Chemical genetic screening identifies tricyclic in peripheral leukocytes are correlated with RBC folate concentrations in human compounds that decrease cellular melanin content. J Invest Dermatol 2008;128: subjects [abstract]. FASEB J 2000;14:A294. 1236–47. 44. Duthie SJ, Hawdon A. DNA instability (strand breakage, uracil misincor- 20. Bowling BD, Doudican N, Manga P, Orlow SJ. Inhibition of mitochondrial poration, and defective repair) is increased by folic acid depletion in human lym- protein translation sensitizes melanoma cells to arsenic trioxide cytotoxicity via a phocytes in vitro. FASEB J 1998;12:1491–7. reactive oxygen species dependent mechanism. Cancer Chemother Pharmacol 2008;63:37–43. 45. Yoshioka A, Tanaka S, Hiraoka O, et al. Deoxyribonucleoside

712 Chen et al.

triphosphate imbalances. 5-Fluorodeoxyuridine-induced DNA double strand strong chemoresistance in melanoma cell clones in vitro: role of O6-methyl- breaks in mouse FM3A cells and mechanism of cell death. J Biol Chem guanine-DNA methyltransferase and the mismatch repair system. Int J Oncol 1987;262:8235–41. 2006;29:785–97. 46. Donson AM, Addo-Yobo SO, Handler MH, Gore L, Foreman NK. 48. Waytt MD, Pittman DL. Methylating agents and DNA repair responses: MGMT promoter methylation correlates with survival benefit and sensitivity methylated base and sources of strand breaks. Chem Res Toxicol 2006;19: to temozolomide in pediatric glioblastoma. Pediatr Blood Cancer 2007;48: 1580–94. – 403 7. 49. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the 47. Alvino E, Castiglia D, Caporali S, et al. A single cycle of treatment combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul with temozolomide, alone or combined with O6-benzylguanine, induces 1984;22:27–55.

Antifolate Activity of Pyrimethamine Enhances Temozolomide-Induced Cytotoxicity in Melanoma Cells

Ming Chen, Iman Osman and Seth J. Orlow

Mol Cancer Res 2009;7:703-712. Published OnlineFirst May 12, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-08-0263

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2009/04/30/1541-7786.MCR-08-0263.DC1 http://mcr.aacrjournals.org/content/suppl/2009/05/10/7.5.703.DC1 http://mcr.aacrjournals.org/content/suppl/2009/05/10/7.5.703.DC2

Cited articles This article cites 49 articles, 17 of which you can access for free at: http://mcr.aacrjournals.org/content/7/5/703.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/7/5/703.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/7/5/703. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.