GYNECOLOGIC

ONCOLOGY

46,

82-87 (1992)

Potentiation of Cisplatin Cytotoxicity in Human Ovarian Carcinoma Cell Lines by Trifluoperazine, a Calmodulin Inhibitor RAYMOND P. PEREZ, LAURA M. HANDEL, AND THOMAS C. HAMILTON Department

of

Medical

Oncology,

Fox Chase

Cancer

Center,

7701 Burholme

Avenue,

Philadelphia,

Pennsylvania

19111

Received October 11, 1991

Chemotherapyfor ovarian cancer is frequently limited by cisplatin (CDDP) resistance.EnhancedDNA repair is oneof several mechanisms which may cooperateto produceresistancein human ovarian carcinomacell lines. Publishedreports suggestthat calmodulin inhibitors, such as trifluoperazine (TFP), may inhibit one or more stepsin DNA repair. The effects of TFP alone or in combinationwith CDDP were determinedby clonogenicassay of six humanovarian carcinomacell lines,derived from untreated patients (someof which were selectedfor cisplatin resistancein vitro) and from patients clinically refractory to cisplatin-based chemotherapy.TFP produceddose-dependentcytotoxicity in all cell lines. In addition, TFP (10 PM) produced approximately two-fold enhancementof CDDP cytotoxicity in three of the six cell lines (A2780, 2780~CPS,and 2780430). TFP and CDDP had additive or synergisticcytotoxicity in four of the six cell lines by medianeffectsanalysis,while clear antagonismwasapparent in the remaining cell lines. Theseresultssuggestthat TFP may enhanceCDDP cytotoxicity in some,but not all, human ovarian carcinoma cell lines. The potential utility of trifluoperazine in ovarian cancer, either alone or in combination with cisplatm, remainsto be defined in xenograft modelsand in clinical trials. 0 1992 Academic

Press, Inc.

Resistance to chemotherapy significantly limits the efficacy of treatment for ovarian cancer. Although platinum-based combination chemotherapy is associated with a 60-80% clinical objective response rate [l] in patients with advanced-stage disease, the 5-year survival for this population is approximately lo-20%. Virtually all of these patients die with chemotherapy-refractory cancer. These sobering observations provide substantial impetus for studies of cellular resistance mechanisms and strategies for their modulation. Enhanced repair of DNA damage induced by cisplatin or alkylating agents is one of several potentially significant resistance mechanisms [2]. We have observed enhanced repair of cisplatin-induced DNA damage, commensurate

with the degree of cisplatin resistance, in human ovarian carcinoma cell lines [3]. Similar observations have been made by other investigators, in the same [4,5] and in other resistant cell lines [6]. Cisplatin resistance in vitro was partially reversible with aphidicolin, an inhibitor of DNA polymerases (Y and y [3]. These polymerases participate in the synthesis of a new DNA strand to replace excised segments of damaged DNA. Katz et al. [6] confirmed and extended our findings by demonstrating partial reversal of cisplatin resistance by aphidicolin in a cisplatinresistant variant of the 2008 human ovarian cancer cell line. These observations suggest that inhibition of DNA repair may be a potentially useful therapeutic strategy in cisplatin-refractory patients. DNA repair is a complex, multistep process, involving recognition of DNA damage, incision of the damaged segment, synthesis of a DNA new strand, and ligation of the new strand to the existing strand. Several lines of evidence suggest that calmodulin inhibitors may inhibit DNA repair and enhance cytotoxicity of DNA-damaging drugs. Potentiation of bleomycin-induced DNA damage and cytotoxicity by calmodulin antagonists has been reported in Chinese hamster ovary cells [7] and murine L1210 leukemia [8] cells. Other investigators have reported that calmodulin inhibitors may enhance the cytotoxicity and DNA damage produced by streptozotocin [9], decarbazine [lo], actinomycin D, mitoxantrone, mAMSA [ll], methotrexate, 5-fluorodeuroxyuridine [12], and doxorubicin [13]. Finally, Kikuchi et al. reported enhancement of cisplatin cytotoxicity by the napthanesulfonamide calmodulin inhibitors W5 and W7 in human ovarian cancer cell lines in vitro [14,15] and in a xenograft model [16]. Trifluoperazine, a widely used clinical antipsychotic agent, is a calmodulin inhibitor which reportedly interferes with the incision step of DNA damage [17]. Its

82 0090~8258/92 $4.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

TRIFLUOPERAZINE

POTENTIATES

TABLE 1 Cell Lines

Cell line A2780 2780-CP8 2780-C30 2780-CP70

Treatment status of patient when cell line established Untreated -

Comment “Sensitive” parental line Resistant lines derived from A2780 by intermittent, incremental exposure to CDDP in vitro

OVCAR3

CTX/ADR/CDDP

Clinically resistant

OVCAR-4

CTX/ADR/CDDP

Clinically resistant

effects on cisplatin cytotoxicity in cisplatin-resistant human ovarian carcinoma cell lines have not been previously studied. MATERIALS

AND METHODS

Ovarian carcinoma cell lines. A2780 is an ovarian cancer cell line from an untreated patient, originally derived by Dr. S. Aaronson (NCI, Bethesda, MD) [lS]. 2780CP8, 2780-C30, and 2780-CP70 cell lines were produced by intermittent, incremental exposure of the sensitive parental A2780 cell line to various concentrations of cisplatin. The OVCAR-3 and OVCAR-4 cell lines were developed from patients who were clinically refractory to platinum-based chemotherapy [ 191. The treatment states of patients from whom these cell lines were derived are summarized in Table 1. All cell lines were maintained in RPM1 1640 medium supplemented with 10% FBS (GIBCO, Grand Island, NY), 0.28 units/ml insulin (Squibb-Novo, Inc., Princeton, NJ), 100 pg/ml streptomycin, 100 units/ml penicillin, and 0.3 mg/ml glutamine. Cells were grown at 37°C in a humidified atmosphere of 5% COZ in air. Drugs. Trifluoperazine was purchased from Sigma (St. Louis, MO). Cisplatin was furnished by BristolMyers Oncology Division, Bristol Laboratories (Evansville, IN). The clinical formulation of cisplatin was used. Drugs were reconstituted in sterile water and diluted to their final concentrations immediately before use. Cytotoxicity assay. Cytotoxicity was assayed in a soft agarose bilayer system, as previously described [20]. Briefly, single-cell suspensions in media and agarose to 0.3% (w/v), in the presence or absence of drugs, were plated over chilled 0.6% agarose feeder layers. From 10 to 30,000 cells/ml were plated, depending on the cloning efficiency of the individual cell line, yielding approximately 2000 colonies per 10 cm2 in untreated controls for all cell lines. The endpoint for assays was growth of col-

CISPLATIN

83

CYTOTOXICITY

onies in untreated control plates to a minimum diameter of 60 km. Thus, the time at which colonies were counted was variable in individual cell lines, depending on the growth rate of the cell lines. Colonies were counted with an Artek Omnicon FAS IV Image Analysis System following incubation of the plated cells under routine culture conditions for 7-14 days. Within experiments, the percentage of clonogenic survival was determined as the mean number of colonies from triplicate platings at each drug concentration relative to untreated controls. Data analysis and experimental design. The ICsO’s of the platinum analogues in individual cell lines were determined from survival curves generated for each experiment. Data in tables and figures represent the mean (*SD) of two to four individual experiments, unless otherwise indicated. In drug combination experiments, three to four concentrations of individual drugs and three to four combinations of drugs at a constant molar ratio were tested. The concentration range chosen in each cell line encompassed the IC,,, and I& concentrations, and molar ratio combinations of CDDP and TFP were based on the ratio of IC5,j’s for these drugs. Drug interactions were determined by median effects analysis, using commercially available software [21]. Cytotoxicity data were fitted to regression lines, nd the concentration of each drug which produced a given level of cytotoxicity (fractional effect, Fa) alone or in combination was determined. Typical correlation coefficients for cytotoxicity curves used to generate median effect plots were 20.9. These values were used to calculate the “combination index” for a given fractional effect: Combination

Index (CI) = $

1

+ $. 2

D1 and D2 are the doses of drugs 1 and 2, which by themselves produce a given fractional effect (i.e., I&,); d, and d2 are the doses which produce the same fractional effect in combination. Combination indices are generally interpreted as follows: CI = 1 indicates zero interaction (additive cytotoxicity); CI < 1 indicates synergy, and CI > 1 indicates antagonism.

RESULTS Trifluoperazine produced concentration-dependent cytotoxicity in all cell lines. The I&,? of trifluoperazine alone in individual cell lines are summarized in Table 2. The OVCAR-3 cell line was most sensitive to trifluoperazine, while the OVCAR-4 cell line was least sensitive. There did not appear to be any relationship between sensitivity to trifluoperazine and sensitivity to cisplatin (Tables 2 and 3).

84

PEREZ, HANDEL,

AND HAMILTON

TABLE 2 I&, (pkf) of TFP

--f-

Cell line

IGO (PW

A2780 2780-CP8 2780X30 2870-CP70 OVCAR-3 OVCAR-4

13.7 13.8 9.3 7.5 6.0 21.0

e +f + + *

3.2 0.7 0.3 4.4 2.5 0

The effects of trifluoperazine (10 PM) on cisplatin cytotoxicity in the A2780 cell line are shown in Fig. 1. Data for the trifluoperazine-cisplatin combination are plotted relative to a trifluoperazine-only control. Trifluoperazineenhanced cisplatin cytotoxicity in this cell line (dose-modifying factor (DMF) = 2.36, P = 0.003). Similar results were obtained in the 2780-CP8 and 2780-C30 cell lines, but not in the 2780-CP7O,OVCAR-3, and OVCAR-4 cell lines (Table 3). In limited preliminary experiments, less pronounced effects were seen with 3.3 PM trifluoperazine (data not shown), suggesting that the effects of trifluoperazine may be concentration-dependent. Cytotoxicity data for cisplatin and trifluoperazine were subjected to median-effects analysis (Table 4). Drugs were tested at constant molar ratios, based on the cytotoxicity of cisplatin and trifluoperazine alone. These data were not corrected for trifluoperazine cytotoxicity prior to analysis. Effects were variable in different cell lines. Clear synergy was observed in the 2780-CP8 cell line (CI = 0.7284). In addition, the data suggest that trifluoperazine and cisplatin had synergistic cytotoxicity in the A2780 cell line, although these data are somewhat difficult to interpret due to the large amount of experimental variability (SD = 0.3219). This variability derived from a single experiment (CI = 1.0423, compared to CI = 0.5226 and 0.4535 obtained in the other two experiments). TABLE 3 Effect of TFP on CDDP Cytotoxicity Cisplatin IC,, (PM) Cell line A2780 2780-CP8 2780-C30 278OCP70 OVCAR3 OVCAR-4

Relative resistance 1 12.7 13.5 13.8 (1) (1.35)

CDDP 0.26 3.30 35.0 3.6 0.27 0.35

k * k + k ?

CDDP + TFP (10 Pdb 0.07 0.11 -c 0.06 1.3 1.78 + 0.54 4.24 16.25 + 3.89 1.04 3.37 2 0.86 0.04 0.30 k 0.09 0.07 0.42 2 0.02

’ Mean + SD. b Cytotoxicity as a percentage of TFP control. ’ t test. d Dh”F, dose-modifying factor (I& CDDP/IC,

P

DMF”

0.003

2.36

0.068 0.022 ns ns ns

1.85 2.15 -

CDDP + TFP).

CDDP CDDP+TFP

10pM

Percent Clonogenlc SuNival

io

0:2

0:4

0:6~0:6-1:0'

CODP Concentration

1.2'

(PM)

FIG. 1. Effect of 10 PM trifluoperazine (TFP) on cisplatin cytotoxicity in A2780 cells. Points represent means ( f standard deviation) of three to four separate experiments. Within experiments, triplicate platings were done at each drug concentration.

The cause of this variability was not apparent. Additive effects were most probably present in the 2780-C30 and OVCAR-3 cell lines. However, marked antagonism was seen in the 2780-CP70 and OVCAR-4 cell lines. DISCUSSION

Calmodulin is a ubiquitous, multifunctional, calciumdependent regulatory protein [22,23]. It mediates cytoskeletal activity, cell-cycle progression, DNA synthesis, and a host of regulatory phosphorylation reactions. Calmodulin may also influence DNA repair. Chafouleas et al. [7] observed potentiation of bleomycin cytotoxicity in CHO cells treated with the napthanesulfonamide calmodulin inhibitor W13 but not in cells treated with the analogue W12, which has minimal calmodulin antagonist activity. Repair of bleomycin-induced DNA damage, assayed by nucleoid sedimentation, was completely inhibTABLE 4 Anaylsis of the Interaction BetweenTFP and CDDP Cell line A2780 2780~CP8 2780-C30 2780-CP70 OVCAR3 OVCAR4

Molar ratio (CDDP:TFP) 1:lO 1:lO 3:l 1:l 1:lO 1:lO

Combination index (Fa = 0.5) 0.6728 0.7284 1.1250 1.5012 0.9294 1.7838

(k0.3219) (k0.1536) (+-0.1030) (+0.0960) (+0.0021) (kO.0911)

Interaction Additive/synergy Synergy Additive Antagonism Additive Antagonism

TRIFLUOPERAZINE

POTENTIATES

ited by W13, but not W12. Similarly, Lazo et al. [8] noted enhancement of bleomycin cytotoxicity and DNA damage in L1210 cells treated with a variety of structurally unrelated calmodulin inhibitors. Calmodulin inhibitors have been reported to enhance DNA damage and cytotoxic effects produced by a variety of drugs, as noted in the Introduction. The potential roles of calmodulin inhibitors in clinical therapeutics and as potentiators of DNA-damaging drugs have been reviewed in detail [24]. It is reasonable to expect that inhibition of an important cellular regulatory protein, such as calmodulin, would be cytotoxic. In addition, agents which inhibit DNA repair might be expected to enhance the effects of DNA-damaging drugs, such as cisplatin. The present investigations evaluated the cytotoxic effects of trifluoperazine alone and in combination with cisplatin in human ovarian carcinoma cell lines. Trifluoperazine produced concentration-dependent inhibition of clonogenic survival in all cell lines. These results are consistent with earlier observations by Hait and Lee [25], where the IC5,, of trifluoperazine alone against a panel of five murine and human malignant cell lines was in the range of 4-8 PM. Our results are also consistent with those reported by Kikuchi et al., demonstrating that the calmodulin inhibitors W5 and W7 were cytotoxic to different human ovarian carcinoma cell lines [14]. It is also interesting to note that individual cell lines exhibited different sensitivities to trifluoperazine, without an apparent relationship with cisplatin sensitivity. The mechanisms responsible for trifluoperazine cytotoxicity are not clearly established. The effects of trifluoperazine on cisplatin cytotoxicity were initially investigated in individual cell lines using a single, constant trifluoperazine concentration (10 +U). This concentration is the reported I&, of trifluoperazine for other calmodulin-mediated processes, such as calmodulin-dependent phosphodiesterase activation [26]. This concentration appeared to enhance cisplatin cytotoxicity by approximately two-fold in three of the six cell lines tested. However, this simple initial analysis may not adequately describe the interactions between trifluoperazine and cisplatin. Data points plotted relative to a trifluoperazine control will necessarily be shifted by a constant proportion. This manipulation implies that the effects of one agent on the cytotoxicity of another will be constant across the entire dose-effect range, which may not be correct. Therefore, additional experiments were done in which the effects of trifluoperazine, cisplatin, and combinations of these drugs in constant molar ratios were determined. Data were then analyzed by median-effects analysis, which compares the effects of drug combinations to the effects of individual drugs across the entire doseeffect range. The effects of trifluoperazine in combination

CISPLATIN

CYTOTOXICITY

85

with cisplatin were variable in individual cell lines (Table 4). Additive or synergistic effects were seen in four of the six cell lines tested, while clear antagonism was seen in two cell lines. The magnitude of potentiation of cisplatin cytotoxicity by trifluoperazine was small (approximately twofold). However, clinical data suggest that low-level cisplatin resistance is relatively common. Approximately 30% of patients who are refractory to conventional-dose cisplatin obtain objective responses following two- to threefold dose escalation [27]. These patients must have had lowlevel resistance in order to respond to dose escalations of this magnitude. Thus, relatively small effects may still be potentially clinically important. In addition, calmodulin regulates a variety of cellular processes. It is not inconceivable that calmodulin inhibitors could produce effects which might offset the cytotoxicity of cisplatin. For example, calmodulin appears to be an important regulator of cell-cycle progression [28,29]. Cell-cycle arrest produced by calmodulin inhibitors might permit some cells to repair and recover from potentially lethal cisplatin damage. This is consistent with data from Pera et al. [30], who demonstrated that survival of cells treated with cisplatin increased when cells were growth arrested by nutritional deprivation following treatment. Moreover, calmodulin inhibitors in general, and trifluoperazine in particular, are not entirely specific. Clearly, growth arrest and competing nonspecific effects could easily have confounded the results obtained in a simple, continuoustreatment cytotoxicity assay. It is therefore possible that our assay underestimated the effects of trifluoperazine. It is also not surprising that different effects were seen in individual cell lines. We have previously observed distinct platinum analogue sensitivity phenotypes in human ovarian carcinoma cell lines [20]. These observations suggest that the determinants of sensitivity to platinum analogues may vary in different cell lines. A corollary of this contention is that biochemical modulation may be expected to benefit some, but not all, patients with cisplatin-resistant disease. Analysis of interactions between drugs in combination is complex. Terminology, methodologies, and criteria for statistical validity are not standardized [31]. We have approached these problems in several ways. First, straightforward definitions of interactions were used. Additive cytotoxicity is defined as being no greater than the effects of individual drugs, while synergy and antagonism describe effects which, respectively, are greater or less than those of additive combinations. Second, the combination index equation used describes a classical isobologram [31]. Isobologram analyses are a widely used and generally accepted method for drug combination analyses. Third, we have attempted to interpret the data conservatively. In a separate publication [32], we have observed some

86

PEREZ, HANDEL.

variability for combination indices derived for sham combinations of platinum analogues, which have additive effects by definition. It is important, therefore, to cautiously interpret small effects, pending the development of valid, formal significance tests for these data. These data should be considered in light of several important caveats. First, these observations are descriptive rather than mechanistic. Although the experiments were initiated on the basis of a hypothetical mechanistic rationale, information regarding specific mechanisms of cytotoxicity cannot be inferred from the results. In addition, trifluoperazine is not a completely specific calmodulin inhibitor. Lack of specificity makes mechanistic interpretation of studies using calmodulin inhibitors difficult. Second, the experimental design does not address the schedule dependence of effects, a potentially important consideration. The potential importance of schedule dependence is underscored by Katz et al. [6], who investigated the effects of the DNA repair inhibitor aphidicolin on cisplatin cytotoxicity in the A2780 and 2008 human ovarian carcinoma cell lines and resistant sublines. Continuous treatment with aphidicolin and cisplatin produced additive effects, whereas intermittent treatment with cisplatin followed by aphidicolin produced marked synergy in these cell lines. Future investigations in our laboratory will assess the effects of trifluoperazine on cisplatin cytotoxicity using an optimized schedule based on data from DNA repair assays. Finally, the clinical significance of in vitro synergy remains to be established. The potential utility of calmodulin inhibitors in ovarian cancer, either alone or in combination with cisplatin, remains to be defined in xenograft models and in clinical trials. ACKNOWLEDGMENTS Dr. Perez is a past American Cancer Society Clinical Oncology Fellow (89-142) and is currently supported by grants from U.S. Bioscience and the Mary L. Smith Charitable Lead Trust (04269-06-J). Work from our laboratory is supported by institutional Grants NIH CA 00927, NIH RR0895 the Pew Charitable Trusts (88-01522OOO), and appropriations from the Commonwealth of Pennsylvania.

AND HAMILTON line resistant to cis-diamminedichloroplatinum(I1). Cancer Res. 50, 1863-1866 (1990). 5. Parker, R., Eastman, A., Bostick-Burton, F., and Reed, E. Acquired cisplatin resistance in human ovarian cancer cells is associated with enhanced repair of cisplatin-DNA lesions and reduced drug accumulation. .Z. Clin. Invest. 87, 772-777 (1991). 6. Katz, E., Andrews, P., and Howell, S. The effect of DNA polymerase inhibitors on the cytotoxicity of cisplatin in human ovarian carcinoma cells. Cancer Commun. 2, 159-164 (1990). 7. Chafouleas, J., Bolton, W., and Means, A. Potentiation of bleomycin lethality by anticalmodulin drugs: A role for calmodulin in DNA repair. Science 224, 1346-1348 (1984). 8. Lazo, J., Hait, W., Kennedy, K., Braun, I., and Meandzija, B. Enhanced bleomycin-induced DNA damage and cytotoxicity with calmodulin antagonists. Mol. Phnrmacol. 27, 387-393 (1985). 9. Lonn, U., and Lonn, S. Cytotoxicity, calmodulin, and DNA lesions in cells treated with streptozotocin. Biochem. Pharmacol. 37, 34413446 (1988). 10. Lonn, U., and Lonn, S. W-7, a calmodulin inhibitor, potentiates decarbazine cytotoxicity in human neoplastic cells. Znt. Z. Cancer 39, 638-642 (1987). 11. Ganapathi, R., Grabowski, D., Schmidt, H., Seshardri, R., and Israel, M. Calmodulin inhibitor trifluoperazine selectively enhances cytotoxic effects of strong vs weak DNA binding antitumor drugs in doxorubicin-resistant P388 mouse leukemia cells. Biochem. Biophys. Res. Commun. Wl, 912-919 (1985). 12. Lonn, U., and Lonn, S. Increased growth inhibition and DNA lesions in human colon adenocarcinoma cells treated with methotrexate or 5-fluorodeoxyuridine followed by calmodulin inhibitors. Cancer Res. 48, 3319-3323 (1988). 13. Ganapathi, R., Grabowski, D., Hoeltge, G., and Neelon, R. Modulation of doxorubicin-induced chromosomal damage by calmodulin inhibitors and its relationship to cytotoxicity in progressively doxorubicin-resistant tumor cells. Biochem. Pharmacol. 40, 1657-1662 (1990). 14. Kikuchi, Y., Iwano, I., and Kato, K. Effects of calmodulin antagonists on human ovarian cancer cell proliferation in vitro. Biochem. Biophys. Res. Commun. 123, 385-392 (1984). 15. Kikuchi, K., Miyauchi, M., Kizawa, I., Oomori, K., and Kato, K. Establishment of a cisplatin-resistant human ovarian cancer cell line. J. Natl. Cancer Inst. 77, 1181-1185 (1986). 16. Kikuchi, K., Oomori, K., Hirata, I., Kita, T., Miyamuchi, M., and Kato, K. Enhancement of antineoplastic effects of cisplatin by calmodulin antagonists in nude mice bearing human ovarian carcinoma. Cancer Res. 47, 6459-6461 (1987). 17 Charp, P., and Regan, J. Inhibition of DNA repair by trifluoperazine. Biochim. Biophys. Acta 824, 34-39 (1985). 18. Eva, A., Robbins, K., Anderson, P. Srinivasan, A., Tronick, S., Reddy, E., Ellmore, N., Gallen, A., Lautenberger, J., Papas, T., Westin, E., Wong-Staal, F., Gallo, R., and Aaronson, S. Cellular genes analogous to retroviral one genes are transcribed in human tumor cells. Nature 295, 116-119 (1982). 19. Hamilton, T., Lai, G., Rothenberg, M., Fojo, A., Young, R., and Ozols, R. Mechanisms of resistance to cisplatin and alkylating agents, in Drug resistance in cuncer therapy (R. Ozols, R. Ed.), Kluwer Academic Publishers, Boston, pp. 151-169 (1989). 20. Perez, R., O’Dwyer, P., Handel, L., Ozols, R., and Hamilton, T. Comparative cytotoxicity of CI-973, cisplatin, carboplatin, and tetraplatin in human ovarian carcinoma cell lines. Znt. J. Cancer 48, 265-269 (1991). I,.

REFERENCES 1. Ozols, R., and Young, R. Chemotherapy of ovarian cancer, Semin. Oncol. 11, 2.51-263 (1984). 2. Perez, R., Hamilton, T., and Ozols, R. Resistance to alkylating agents and cisplatin: Insights from ovarian carcinoma model systems. Pharmacol. Ther. 48, 19-27 (1990). 3. Masuda, H., Ozols, R., Lai, G-M., Fojo, A., Rothenberg, M., and Hamilton, T. Increased DNA repair as a mechanism of acquired resistance to cis-diamminedichloroplatinum(I1) in human ovarian carcinoma cell lines. Cancer Res. 48, 5713-5716, (1988). 4. Masuda, H., Tanaka, T., Matsuda, H., and Kutsaba, I. Increased removal of DNA-bound platinum in a human ovarian cancer cell

TRIPLUOPERAZINE

POTENTIATES

21. Chou, J., and Chou, T. Dose-effect analysis with microcomputers, Biosoft, Cambridge, UK (1987). 22. Klee, C., and Vanaman, T. Calmodulin, Adv. Protein Chem. 35, 213-321 (1982). 23. Means, A., VanBerkum, M., Bagchi, I., Lu, K., and Rasmussen, C. Regulatory functions of calcodulin. Pharmacol. Ther. 50, 255270 (1991). 24. Rosenthal, S., and Hait, W. Potentiation of DNA damage and cytotoxicity by calmodulin antagonists. Yale J. Biol. Med. 61, 3949 (1988). 25. Hait, W., and Lee, G. Characteristics of the cytotoxic effects of the phenothiazine class of calmodulin antagonists. Biochem. Pharmacol. 34, 3973-3978 (1985). 26. Weiss, B., Prozialeck, W., Cimino, M., Barnette, M., and Wallace, T. Pharmacological regulation of calmodulin. Ann. N. Y. Acad. Sci. 356, 319-345 (1980).

CISPLATIN

CYTOTOXICITY

87

27. Ozols, R., Ostchega, Y., Myers, C., and Young, R. High dose cisplatin in hypertonic saline in refractory ovarian cancer. J. Clin. Oncol. 3, 1246 (1985). 28. Rasmussen, C., and Means, A. Calmodulin is involved in regulation of cell proliferation. EMBO J. 6, 3961-3968 (1987). 29. Rasmussen, C., and Means, A. Calmodulin is required for cellcycle progression during GI and mitosis. EMBO J. 8,73-82 (1989). 30. Pera, M., Rawlings, C., and Roberts, J. The role of DNA repair in the recovery of human cells from cisplatin toxicity. Chem. Biol. Interact. 37, 245-261 (1981). 31. Berenbaum, J. What is synergy? Pharmacol. Rev. 41, 93-141 (1989). 32. Perez, R., Perez, K., Handel, L., and Hamilton, T. In vitro interactions between platinum analogues in human ovarian carcinoma cell lines. Cancer Chemother. Pharmacol. 29, 430-434 (1992).