(CANCER RESEARCH 51. 2848-2853. June 1, I991|

Combined Effects of 1,25-Dihydroxyvitamin D3 and Platinum Drugs on the Growthof MCF-7 CellsY. L. Cho,1 C. Christensen, D. E. Saunders,2 W. D. Lawrence, G. Deppe, V. K. Malviya, and J. M. MaloneDepartments of Obstetrics and Gynecology /C. C., G. D., V. K. M., J. M. M., D. E. S.¡and Pathology ¡W.D. L.], H'ayne Stale University School of Medicine, Detroit,

Michigan 48201, and Department of Obstetrics and Gynecology [Y. L. C.], Kyung-pook National University Hospital, Taegu, Republic of Korea 700-412

ABSTRACT

The effects of 1,25-dihydroxyvitamin D3 11.25(<>l1),!),] and platinumtreatments (both singly and combined) on the growth inhibition of MCF-7 cells, an epithelial cell line shown to possess specific receptors for1,25(OH)2D,, were evaluated. The inhibitory effects of 1,25(OH)2D, andplatinum on MCF-7 cell proliferation in vitro were time and dose related.The data showed that 10 UMand 100 MM1.25«>lI ),!>, inhibited MCF-7cell growth by 10.8 ±2.4% and 34.9 ±0.5% (mean ±SE), respectively.The degrees of growth inhibition induced by 0.2 to 200 ¿ig/mlof cis-diammine-l,l-cyclobutane dicarboxylatoplatinum(II) (carboplatin) wereslightly less than those induced by 0.02 to 20 fig/ml of c/i-diamminedi-chloroplatinum(ll) (cisplatin). The combined administration of 10 MMand 100 UM1,25(OH)2D3 with either carboplatin (200 to 0.2 fig/ml) orcisplatin (20 to 0.02 M«/ml)was evaluated. Addition of 1.25«)I I) l>, tothe platinum resulted in marginal to marked enhancement of growthinhibition over that observed with either platinum alone. The strength ofthese interactions varied inversely with the dose of the platinum drugs.Evaluation of drug interactions with isobolograms showed that at near-serum levels, carboplatin or cisplatin interacted synergistically with1,25(OH)2D., to inhibit MCF-7 cell growth. Our findings suggest potential usefulness in combining 1.25«)lI) I),, a biological modifier, withcytotoxic agents for the treatment of malignant disease.

INTRODUCTION

The most biologically active metabolite of vitamin D3,1,25(OH)2D.,,' is a potent mineralotropic steroid hormone

whose primary target tissues include intestine, kidney, and bone(1). The actions of 1,25(OH)2D., in target tissues are mediatedthrough a specific, high-affinity receptor (1, 2). Recently, specific cellular receptors for 1,25(OH)2D,, which are indistinguishable from those in documented target tissues of 1,25(OH)2Dj, have been identified in a variety of tumors and tumorcell lines, including malignant breast (3), epidermal (4), thyroid(5), hematopoietic (6, 7), and retinoblastoma (8) cells. Thissuggests a biological role for l,25(OH)2D.i in human malignancy. In vitro and in vivo evidence suggests that 1,25(OH)2D,can influence the growth and differentiation of malignant cells,including cells from a human malignant melanoma cell line (4),animal and human myeloid leukemia cell lines (9-13), breastcarcinoma cell lines (7, 14-16), a retinoblastoma cell line (8),patients with myeloid leukemia (17), and in vivo growth ofhuman cancer solid tumor xenografts (18). Furthermore, administration of l,25(OH)2Dj and l-«-hydroxyvitamin D., tomice inoculated with myeloid leukemia cells was shown toprolong survival (19). Also, 1,25(OH)2D, was found to inhibit

Received 8/24/90; accepted 3/13/91.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1Y. L. C. was a Visiting Scholar at Wayne State University. Detroit. Ml while

this work was being carried out.2To whom requests for reprints should be addressed.1The abbreviations and trivial names used are: 1,25(OH)2D3. 1.25-dihydroxy-

vitamin D3; cisplatin. m-diamminedichloroplalinumfll): carboplatin. r/j-diam-mine-1,1-cyclobutane dicarboxylatoplatinum(Il): DMEM, Dulbecco's modifiedEagle's medium with HAM's F-12 nutrient mixture; HEPES, 4-(2-hydroxyethyl)-1-pipcrazineethanesulfonic acid; LD¡0.drug dose that causes 20rt lethality (LD50

defined similarly).

the promotional phase of 7,12-dimethylbenz[i*]anthracene-in-

duced skin carcinogenesis in female mice (20). More recently.Hellström and coworkers reported high therapeutic responseswith combinations of various differentiation inducers, includinga vitamin D3 metabolite, in patients with acute leukemia (21,22). These studies make a strong case for the potential of1,25(OH)2D, as a biological agent for single-agent therapy orin combination with other agents in the treatment ofmalignancies.

The MCF-7 cell line was originally established from the

pleural effusion of a patient with breast carcinoma (23). Sincedemonstration of cytosol receptors for estrogen in 1973, theMCF-7 cell line has been used extensively as a model forstudying hormone-responsive/dependent growth and the treatment of hormone-dependent tumors (24). In 1979, a specificreceptor for l,25(OH)2Di was demonstrated in cultured MCF-7 cells (3), and the addition of 1,25(OH)2D., to the culturemedium caused a dose-dependent decrease in their rate ofgrowth (7, 14-16). Therefore, MCF-7 cells were selected as a

model system for studying potential interactions between1,25(OH)2D, and other antineoplastic drugs.

In advanced or recurrent malignant diseases including gynecological neoplasms, the development of adjuvant chemotherapy has improved response rates, but the overall impact onsurvival has been minimal. Moreover, chemotherapeutic treatment is complicated by toxicity as well as development of drugresistance (25). Thus, the development of new chemotherapeutic agents and new combination regimens is highly desirable.

In recent years, there has been increased interest in the useof hormonal agents as palliative therapy in breast and gynecological malignancies (26-29). In addition to 1,25(OH)2D„the

effects of other biological response modifiers such as interferon,glucocorticoids, and retinoic acid with or without cytotoxicagents have been investigated (30, 31). One effect of theseagents may be to arrest cells in various phases of the cell cycle,thus leading to a partial cell cycle synchronization and a cyto-static effect. For example, cells can be arrested in G0-Gi bytamoxifen and released using estradiol to cause a synchronousprogression into S phase (32, 33). These cells are 50-fold moresensitive to the growth-inhibitory effects of 5-fluorouracil.

There have been legitimate concerns about possible antagonisticeffects of combining cytostatic agents (hormones and biologicalmodifiers) and agents with cell cycle dependence, such as anti-metabolites and certain alkylating agents (e.g.. cyclophospha-mide) (28, 34). However, cisplatin and. more recently, carboplatin have been shown to be effective against not only rapidlygrowing tumor cells but also slowly growing ones (35), suggesting that a cytostatic agent, such as l,25(OH)2Di, might notantagonize (and may even enhance) the effects of this cytotoxicagent.

This study was undertaken to determine if 1,25(OH)2D.,would enhance the effects of platinum agents in inhibitingMCF-7 cell proliferation in vitro.

2848

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

VITAMIN D. PLATINUM. AND BREAST CANCER

MATERIALS AND METHODS

Cells and Cell Culture. MCF-7 cells (passage 172). obtained from theMichigan Cancer Foundation, Detroit, MI, were grown in monolayerculture in DMEM, containing 5% donor calf serum (Sigma ChemicalCo., St. Louis, MO), 100 units/ml of penicillin. 100 Mg/ml of streptomycin, and 10 Mg/ml of insulin (Sigma). Stock cultures were maintainedin 75-cnr flasks (Corning Glass Works, Corning, NY) and incubatedat 37°C.Cell populations were serially passaged every 3 to 5 days.

Drugs. 1.25(OH);Di was kept dissolved in ethanol and stored inconcentrated solution at -20°C. 1.25(OH)2D, was freshly diluted in

the DMEM before each experiment. The ethanol concentration in thetest conditions never exceeded 0.1%. Carboplatin and cisplatin wereselected as cytotoxic agents and were obtained from Bristol-Myers

(Evansville. IN) as powder. Concentrated stocks of carboplatin andcisplatin were made using sterile distilled water on the day of theexperiment, and dilutions were made with DMEM.

Measurement of Growth Inhibition. For each experiment, cells wereharvested at confluence with 5 ml of a trypsin-EDTA solution containing 0.5 g of trypsin and 0.2 g of EDTA/liter (Sigma), and 2.2 x IO4cells were passed in 1 ml of DMEM into a 24-well plate (Corning) onDay 0 after centrifuging through DMEM at 1000 x g. 4°C.After 24-hincubation at 37°Cin a humidified atmosphere of 5% CO: in air. the

appropriate chemotherapeutic agent and 1.25(OH)2D, were added in 1ml of complete medium on Day 1. The effects of combinations of1.25(OH)2D, with platinum were evaluated by coincubating cells with1.25(OH)2D, combined with either carboplatin or cisplatin at variousconcentrations. After a 3-day incubation, cells were counted by lysingthem and counting nuclei on Day 4, using the method of Butler (36).The medium was removed, and 1 ml of hypotonie buffer [0.01 MHEPES (pH 7.4):1.5 mM MgCI2) was added. The cells were allowed toswell for 5 min. Then, 200 ^1 of lysis solution were added, and the platewas shaken for 10 min. The lysis solution consisted of 3 ml of glacialacetic acid and 5 g of ethylhexadecyldimethylammonium bromide(Eastman Kodak Co., Rochester, NY) in 97 ml of deionized water.Finally, 1 ml of saline:formaldehyde (9 g of NaCl/liter; 12.5 ml of 37%formaldehyde/liter) was added. Complete cell lysis was confirmed visually using an inverted microscope. Nuclei were counted using a ModelZM Coulter Counter with a 100-¿ilaperture (Coulter Electronics. Inc.,Hialeah, FL).

Isobologram Analysis for the Interaction of 1,25(OH)2D.<and Platinum. The isobologram method of analysis allows for rigorous classification of the degree of interaction between two drugs into one of fivecategories within a given range of drug concentrations and molar ratios(37-39). A hypothetical isobologram is shown in Fig. 1 and describedin the following text. The LD2(lsand 95% confidence limits for singleand combined drug effects are determined by computerized regressionanalysis of the dose-response data. The data are graphed by plottingthe LD20 for Drug I with 95% confidence limits on the .v-axis, and byplotting the LD20 for Drug II (plus 95% confidence limits) on the y-axis, and by connecting these two LD2(Ipoints with a diagonal line. Thedegree of drug interaction is determined by calculating the LD2n (and95% confidence limits) of the combination and plotting the result alongan imaginary fixed-drug ratio line extending outward from the origin.The interaction between two drugs is defined as "simple addition" when

the effect of the combined drugs is the sum of the effects of the drugsadministered separately. The most straightforward example of thiswould be the effect obtained if Drug I happened to be the same as DrugII (Fig. 1, Region B). "Synergism" is defined here as a mixture of Drug

I and Drug II giving a greater effect than the sum of their individualeffects and is manifested by the LD:os of the combination being lowerthan those which would be predicted if the drugs acted in a simplyadditive fashion (Fig. 1, Region ,4). When a binary mixture of drugsyields a response which is greater than that obtained with either of thedrugs independently, but is less than the response predicted on the basisof simple addition, the interaction is defined as "infraadditivity" (Fig.

1. RegionC).If combining Drug I with Drug II results in a response which is not

different from Drug II administered alone, then there is "noninteraction" (Fig. 1. Region D). "Antagonism" is defined as Drug I causing

O)

Q

Drug I

Concentration Drug I

Fig. 1. Hypothetical ¡sobologram.This figure was adapted from Church et al.(38). The isobologram is defined as lines connecting equieffective dose pairs inregard to a selected endpoint such as the LD:0. First, the LDJOpoints of Drug Iand Drug II are plotted on the .v- and }'-axes along with their 95ri confidence

limits. A diagonal line is then drawn between the LDMendpoints of these 2 drugs.This diagonal lint' and its 95'V confidence limits (shaded Region B) delineate the

region of equieffectiveness for systematic combinations of the 2 drugs. The 95%confidence intervals for the 2 single drugs are also extended perpendicularly tothe .v- and y-axes to delineate a region of noninteraction (shaded Region D). Ifthe LD;,, for a drug combination falls in Region A. this would indicate synergismor "superadditivity." If the lethal dose for a drug combination falls in Regions B.

C, D, or E. this would indicate simple activity, infraadditivity, no interaction, orantagonism, respectively. The dashed line emanating from the origin gives theloci of all possible results from a particular fixed-ratio combination of the twodrugs. The observed LD;o and its 95% confidence limits from this particular drugcombination are plotted along the dashed line. In this illustration, the drugcombination's LD20 is located in Region A. indicating synergism.

Drug II to elicit a lesser response than would be obtained by Drug IIalone (Fig. 1, Region E).

RESULTS

Dose-Response of 1,25(OH)2D.,, Carboplatin, and Cisplatin onMCF-7 Cells. Initial experiments were carried out to determinethe optimum drug concentration and exposure time whichwould elicit partial responses in MCF-7 cell growth inhibition.This was done in order to allow ample opportunity for combined drug effects to occur. The growth inhibition was assayedas described in "Materials and Methods."

To determine optimum dose and concentration for each drug,cells (2.2 x IO4 cells/well) were incubated with 0.1 to 100 nivi

1,25(OH)2D3, 0.2 to 200 Mg/ml of carboplatin, and 0.02 to 20/¿g/mlof cisplatin for 3 days. The doses of 1,25(OH)2D3 to betested were selected on the basis of previous reports of 1,25(OH):D, on malignant cells in vitro and on the basis of physiological relevance (6-17). Fig. 2A shows that l,25(OH):D.isuppressed MCF-7 cell growth in a dose-dependent manner.Ten n\t and 100 n\i 1,25(OH),D3 inhibited MCF-7 cell growthby 10.8 ±2.4% and 34.9 ±0.5% (mean ±SE), respectively.Carboplatin and cisplatin also inhibited MCF-7 cell growth ina dose-dependent manner. The degrees of growth inhibitioninduced by 0.2 to 200 Mg/ml of carboplatin were slightly lessthan those induced by 0.02 to 20 ng/m\ of cisplatin (Fig. 2, Band C). The doses causing 50% growth inhibition (LD50) ofMCF-7 cells for 3-day treatment were found to be 15 ng/m\ forcarboplatin and 1.3 Mg/ml for cisplatin, respectively, whichwere less than clinical treatment levels (Fig. 2, B and C).

In two separate time-course experiments, time-dependentgrowth inhibition of MCF-7 cells was seen in the 1,25(OH)2D3-and platinum-treated cultures at 2, 4, and 6 days of continuous

2849

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

VITAMIN D. PLATINUM. AM) BREAST CANCER

oE

IliOoce

10°

«H

40

20-

100

0.1 1 10 1001,25(OH) D (nM)

2 3

60

mx 60-

40HUJo£ 20Ho.

om

muo:UJo.

100

80-

60 •

40

20-

0.2 2 20 200

CARBOPLATIN (u.g/ml)

PERCENTINHIBITION,8fi8SgnC

0.02 0.2 2 20

CISPLATIN (ug/ml)

Fig. 2. Dose-response effects of l,25(OH):Di. carboplutin. and cisplatin onthe growth of MCF-7 cells. Cells (2.2 x 10* cells/well) were passed into 24-well

plates on Day 0 and dosed with various concentrations of l,25(OHhDi orplatinum on Day I. After a 3-day incubation, cells were lysed and counted onDay 4 as described in "Materials and Methods." Columns, mean of 4 wells: bars.

SE. The results are representative of 4 experiments. SEs were calculated for alldata points but were too small in some cases to be visualized. A. dose-response to1.25(OH)jD,; B. dose-response to carboplatin; C, dose-response to cisplatin.

exposure (Fig. 3). During continuous treatment with 100 n\il,25(OH)2Di, progressive growth inhibition was observed over4 days of treatment, plateauing at 60% inhibition. Four days ofexposure to 20 ¿ig/mlof carboplatin or 2 ng/m\ of cisplatinresulted in an increase of the extent of growth inhibition, alsofollowed by a plateau during the 4- to 6-day time period. Thus,the pattern of growth inhibition was similar amongl,25(OH)2D.i, carboplatin, and cisplatin from 2 days of treatment to 6 days of treatment. For carboplatin and cisplatin, thedurations of treatment causing 50% inhibition of MCF-7 cellgrowth were 2.6 days and 2.0 days, respectively. The data fromthese experiments were utilized later in setting up experimentsunder optimal conditions for obtaining combined drug effects.

Effects of Combination Treatment with l,25(OH)2D,and Platinum. To assess the effects of combination treatment on thegrowth inhibition of MCF-7 cells by l,25(OH)2Di combinedwith either carboplatin or cisplatin, MCF-7 cells (2.2 x IO4cells/well) were incubated with various concentrations of plati-

DURATION of TREATMENT (DAYS)

Fig. 3. Time-courses of growth inhibition in MCF-7 cells by l.25(OH)2Dj andplatinum agents. MCF-7 cells (2.2 x IO4 cells/well) were incubated in 24-wcllplates with medium containing lOOnM I.25(OH);D, (•).20 pg/ml of carboplatin(O), or 2 (jg/ml of cisplatin (A). The medium was changed every 2 days withrenewal of each drug. At the indicated times, cells were lysed and counted asdescribed in "Materials and Methods." Points, mean of 4 wells. The results are

representative of 2 experiments. SEs were obtained for all data paints but. insome cases, were too small to visuali/c on this graph.

num in the presence or absence of 10 nM and 100 nM1,25(OH)2D, for 3 days and assayed as described in "Materialsand Methods."

The data in the insets to Fig. 4 show that 10 nM and 100 nM1,25(OH)2D, enhanced the platinum-induced inhibition ofMCF-7 cell growth at all platinum concentrations except thevery highest (200 ng/m\ of carboplatin and 20 ¿ig/mlof cisplatin). The ranges of enhancement of growth inhibition by 10 nMl,25(OH)2Dj over that obtained by carboplatin or cisplatinalone were 0% at 200 ¿/g/mlto 190% at 0.2 /¿g/ml(Fig. 4/4)and from 0% at 20 to 2409¿at 0.02 Mg/ml (Fig. 4fi),respectively. At 100 nM 1,25(OH)2D,, the enhancement rangedfrom 3% at 200 ng/m\ to 260% at 0.2 ng/m\ of carboplatin(Fig. 4C) and from 0% at 20 Mg/ml to 270% at 0.02 Mg/ml ofcisplatin (Fig. 4D). Significantly, 1,25(OH)2D, did not reducethe effects of either platinum drug at any of the concentrationstested.

A preliminary assessment of the degree of interaction between1,25(OH)2D, (at both 10 and 100 nM) and platinum drugs wasmade by comparing the summed effects of the combined drugs(Fig. 4, A to D). The effects of simultaneous drug treatment onthe growth inhibition of MCF-7 cells with 10 nM 1,25(OH)2D,and either carboplatin or cisplatin appeared to be greater thansimple addition at lower concentrations of both platinums (0.2Mg/ml of carboplatin; 0.02 ng/ml and 0.2 ng/m\ of cisplatin)and less than simple addition at higher doses (Fig. 4, A and B).The effects of platinum in combination with 100 nMl,25(OH)2D.i in the inhibition of MCF-7 cell growth rangedfrom simple addition to less than simple addition (Fig. 4, Cand D). These experiments suggest that the amount of interaction between 1,25(OH)2D, and the platinum drugs in inhibiting MCF-7 cell growth varies inversely with the concentrationof the drugs.

Isobolographic Analysis of the Effects of Combination Treatment with 1,25(OH)2D, and Platinum. The data in Fig. 5,showing that 1,25(OH)2Di and platinum compounds interactedmost effectively at lower concentrations, were utilized to designan isobologram experiment which would show the greatestamount of drug interaction. Using this criterion, the followingranges were chosen: 10 to 40 nM 1,25(OH)2D,; 0.05 to 0.2 /ug/

2850

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

VITAMIN D. PLATINUM. AND BREAST CANCER

0.2 2 20 200CARBOPLATIN (jiQ/rnl)

0.1 1CISPLATIN bj

0.02 0.2 2CISPLATIN big/ml)

20

CARBOPLATIN Oifl/ml) CISPLATIN (iiu/ml)

Fig. 4. Combined effects of 1.25(OH)2D3 and platinum on the growth of MCF-7 cells. MCF-7 cells (2.2 x IO4cells/well) were split into 24-well plates on Day 0

and dosed with various concentrations of either carboplatin or cisplatin in thepresence or absence of 10 n.Mor 100 nM 1,25(OH)2D3 on Day 1. After a 3-dayincubation, cells were lysed and counted on Day 4 as described in "Materials andMethods." The stackeü-columngraphs show the calculated inhibition of cell

growth obtained by adding the percentage of inhibition obtained by separateincubations with l.25(OH)2D3 (D) and carboplatin or cisplatin (•).The actual(observed) inhibition (D) of cell growth is plotted next to the stacked columnsdepicting the effects of the individual drugs. Inset shows dose-response curves ofcarboplatin and cisplatin plus (+1)) and minus (-D) 10 n\t and 100 nM1.25(OH)2D3. Columns, mean of 4 wells: bars. SE. The results are representativeof 4 experiments. As before. SEs were determined for all observations. A, 10 nM1.25(OH)2D3 and carboplatin: B. 10 nM I.25(OH)2D3 and cisplatin; C. 100 nM1,25(OH)2D3 and carboplatin: D. 100 nM 1.25(OH)2D3 and cisplatin.

ml of cisplatin; and 0.5 to 2 ng/m\ of carboplatin. Fig. 5Ashows the LD2(,s and 95% confidence limits for 1,25(OH)2D.,alone (91 ±10 nM) and for carboplatin alone (4.9 ±0.3 ¿ig/ml), connected by a diagonal line. The LD20and 95% confidencelimits, for the combined drugs, were plotted along an imaginaryline beginning at the origin and extending along the 20:1 fixed-

drug ratio line. The LD20 for the combined drugs (22 ±6 nMfor 1,25(OH)2D, and 1.1 ±0.2 Mg/ml for carboplatin) lay wellbelow the line connecting the single-drug LD20s, indicating that1,25(OH)2D., and carboplatin interacted synergistically (at thechosen concentrations) to inhibit MCF-7 cell growth. Fig. 5Bshows the isobologram for a 200:1 fixed ratio 1,25(OH)2D,:cisplatin dosage combination. The LD20 for1,25(OH)2D, alone was 91 ±10 nM, and the LD20 for cisplatinalone was 0.35 ±0.03 ¿ig/ml.In the combination experiment,the LD20s were 16 ±4 nM for 1,25(OH)2D, and 0.08 ±0.02Mg/ml for cisplatin, which were far below the line connectingthe single-drug LD20s, indicating that 1,25(OH)2D, and cisplatin also synergistically inhibited MCF-7 cell growth at the

chosen concentrations.

DISCUSSION

Recently, the study of growth-inhibitory (cytostatic) orgrowth-stimulatory (sensitizing) drugs combined with standardchemotherapy has been undertaken to determine if these drugsenhance the antineoplastic activity of cytotoxic drugs (27, 28,31). Furthermore, it has been shown that breast carcinomaspossess receptors for 1,25(OH)2D3 and that these cancers aregrowth inhibited by 1,25(OH)2D., (14,15). Also, this laboratoryhas identified receptors for 1,25(OH)2D, in a broad spectrumof gynecological malignancies, such as uterine sarcomas andcommon epithelial ovarian malignancies (40). These data ledto the current study to evaluate the potential interaction between 1,25(OH)2D, and platinum agents in inhibiting thegrowth of a hormone-responsive epithelial carcinoma, MCF-7.

In the present study, it was demonstrated that 1,25(OH)2D,inhibited MCF-7 cell growth in a time- and dose-dependentmanner. Treatment of MCF-7 cells resulted in up to 35%growth inhibition by 3 days and plateaued at 60% inhibition by6 days. These results are consistent with those reported byThomas and Simpson (7) and Simpson and Arnold (15). Thelimited growth inhibition by 1,25(OH)2D, was previouslyshown to be dependent on receptor number per cell and is a

40 BO 120 160

40 80 160

1,25(OH)2D3(nM)

Fig. 5. Isobologram plots of the interactions between 1,25(OH)2D3 and platinum agents in the growth inhibition of MCF-7 cells. Cells were passed into 24-well plates and dosed 24 h later with various concentrations of the drugs. Thecells were exposed to the drugs for 3 days. Cell numbers were determined asdescribed in "Materials and Methods." Single-drug LD2c>swere determined fromincubations with 1 to 100 nM 1.25(OH)2D3 (plotted on the .v-axis of A and A),with 2 to 200 ^g/ml of carboplatin (plotted on the j'-axis of A), or with 0.2 to 20fig/ml of cisplatin (plotted on the .y-axis of B). Effective drug concentrations forthe combination incubation presented here were determined in the Fig. 4 experiment. Combined drug LD20s were determined by incubating with multiple fixed-ratio combinations of 1,25(OH)2D3 (10 to 40 nM) with carboplatin (0.5 to 2 /ig/ml) (A) or of 1,25(OH)2D3 (10 to 100 nM) with cisplatin (0.05 to 0.2 /jg/ml) (B).The combined drug LD20s and 95'r confidence limits are plotted as a single dalapaini (with error bars) along the fixed-drug-ratio line. A, isobologram of1,25(OH)2D3 and carboplatin; B. isobologram of 1,25(OH)2D3 and cisplatin.

2851

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

VITAMIN D. PLATINUM. AND BREAST CANCER

likely explanation for the limited effects of 1,25(OH)2D, shownhere (7, 18).

Cisplatin has proved to be one of the most important cyto-

toxic drugs in the treatment of solid tumors, including gynecological neoplasms, and has become the central drug in manynew combination regimens of adjuvant chemotherapy (25).Carboplatin has emerged as the leading second-generation cis-platin, which is comparable to cisplatin in antitumor activity,but less toxic to the kidneys, gastrointestinal tract, and peripheral nerves (41).

In clinical and in vivo studies, improved median survival ordisease-free interval with combination chemotherapy or che-mohormonal therapy has led to possibly erroneous conclusionsthat these drugs act in an additive or synergistic fashion. However, the nonlinear response curve of most drugs makes itvirtually impossible to accurately demonstrate drug interactionssimply by comparing the sum of individual drug effects withthe actual observed combined effects (42). Isobologram analysisis a superior method because it concisely determines druginteractions independently of the shape of the dose-responsecurves, thus avoiding artifactual conclusions. Also, isobologramanalysis is able to concisely categorize the degree of interactionbetween two drugs, regardless of whether the drugs act throughsimilar or different mechanisms (37, 42).

The most important impact of this investigation is the observation that l,25(OH)2Di enhances the growth-inhibitory effectsof cisplatin and carboplatin of MCF-7 cells in a manner whichvaries inversely with the concentration of the platinum drugs.Comparison of growth inhibition by platinum alone versusl,25(OH)2D.i and platinum (Fig. 3) showed a substantial enhancement of growth inhibition at clinically relevant concentrations of cisplatin (0.02 to 2.0 /ug/ml) and carboplatin (0.2 to 20fig/ml). Furthermore, isobologram analysis of combination effects at drug levels approaching those attainable in the plasmaof patients (0.05 to 0.2 Mg/ml of cisplatin and 0.5 to 2 iig/m\of carboplatin) showed 1,25(OH)2D., and each of the platinumdrugs to interact synergistically. At 10-fold above attainableserum levels (20 ng/m\ of cisplatin and 200 Mg/ml of carboplatin) (41), the platinum drugs were extremely cytotoxic, andtherefore no enhancement of growth inhibition by 1,25(OH)2D,was observed. In contrast, a separate isobologram experiment,with different fixed ratio levels and higher drug concentrations,showed that the interactions between 1,25(OH)2D, and bothplatinums were infraadditive (data not shown). Thus synergismis seen at low platinum concentrations, while infraadditivity isseen at high platinum concentrations. These results are consistent with other findings which show that the degree of druginteraction is dependent upon drug concentrations and drugratios and indicate that conclusions must be specific to thecircumstances examined and cannot be generalized to concentrations and ratios not tested (42).

Studies cited in the "Introduction" suggested that synchro

nization of cells followed by treatment with a cell cycle-dependent cytotoxic agent resulted in an enhanced response. In HL60leukemic and T 47D breast cancer cells, 1,25(OH)2D, has beenshown to cause a dose-dependent accumulation in G0-G| andtherefore might tend to drive responsive cell populations towardsynchronization (43, 44). It is therefore possible that this phenomenon plays a role in the enhancement of platinum effectsby 1,25(OH)2D, reported herein. Additional studies will berequired to determine if this hypothesis is true.

Recently, the combination of 1,25(OH)2D3 and cisplatin hasbeen investigated in HL60 human leukemic cells (45). The

binding of cisplatin and DNA was increased more than 10-foldby 100 n\i 1,25(OH)2D.,. The increase in cisplatin-DNA bindingwas proportional to l,25(OH)2D,-induced differentiation. Thebound platinum was in 2 pools, one that was easily dissociatedfrom nuclei and one that was tightly bound, presumably toDNA. These data suggest that 1,25(OH)2D, altered chromatinstructure in a manner which made the DNA more accessible tocisplatin and thereby enhanced cisplatin binding. The resultsfrom HL60 cells raise the possibility that, in the current study,the enhancement of platinum-induced growth inhibition wascaused by 1,25(OH)2D., creating a more open chromatin structure, thus facilitating binding of the platinum drugs to DNA.

In conclusion, this study showed that the combination treatment of 1,25(OH)2D.1 and platinum produced effects rangingfrom infraadditivity to synergism in inhibiting the growth ofMCF-7 cells. These findings suggest potential usefulness incombining l,25(OH)2D.i, a biological modifier, with cytotoxicagents for the treatment of malignant disease.

ACKNOWLEDGMENTS

The authors are grateful to Nanci Wappler for assistance in dataanalysis and Dr. Michael Church for contributing his expertise inisobologram analysis.

REFERENCES

1. Haussler, M. R., and McCain. T. A. Basic and clinical concepts related tovitamin D metabolism and action. N. Engl. J. Med., 297: 974-983, 1977.

2. Manolagas. S. C. Vitamin D and its revelance to cancer. Anticancer Res., 7:625-638. 1987.

3. Eisman. J. A., Martin. T. J., Maclntyre. I., and Moseley, J. M. 1,25-Dihydroxyvitamin-D receptor in breast cancer cells. Lancet, 2: 1335-1336,1979.

4. Colston, K., Colston, M. J., and Feldman. D. 1,25-Dihydroxyvitamin D3 andmalignant melanoma: the presence of receptors and inhibition of cell growthin culture. Endocrinology. 108: 1083-1086, 1981.

5. Freake, H. C., and Maclntyre, I. Specific binding of 1,25-dihydroxycholecal-ciferol in human medullary thyroid carcinoma. Biochem. J., 206: 181-184.1982.

6. Amento. E. P., Bhalla, A. K., Kurnick. J. T.. Kradin. R. L., Clemens, T. L.,Holick. S. A., Holick. M. F., and Krane, S. M. 1,25-Dihydroxyvitamin D,induces maturation of the human monocytecell line U937, and, in associationwith a factor from human T-lymphocytes. augments production of themonokine. mononuclear cell factor. J. Clin. Invest., 73: 731-739. 1984.

7. Thomas, G. A., and Simpson. R. U. High performance liquid chromatogra-phy analysis of 1,25-dihydroxyvitamin D3 receptor in malignant cells. Correlation of effects on cell proliferation and receptor concentration. CancerBiochem. Biophys.. 8: 221-234. 1986.

8. Saulenas, A. M., Cohen. S. M.. Key, L. L.. Winter. C.. and Albert, D. M.Vitamin D and retinoblastoma. The presence of receptors and inhibition ofgrowth in vitro. Arch. Ophthalmol.. 106: 533-535. 1988.

9. Abe. E.. Miyaura. C. M., Sakagami, H., Takeda, M., Konno. K., Yamazaki,T., Yoshiki, S., and Suda, T. Differentiation of mouse myeloid leukemia cellsinduced by 1,25-dihydroxyvitamin D,. Proc. Nati. Acad. Sci. USA, 78:4990-4994. 1981.

10. Rigby. W. F. C.. Shen. L.. Ball, E. D., Guyre. P. M., and Fanger, M. W.Differentiation of a human monocytic cell line by 1,25-dihydroxyvitamin D3(calcitriol): a morphologic, phenotypic. and functional analysis. Blood. 64:1110-1115.1984.

11. Weinberg, J. B., Misukonis. M. A., Hobbs, M. M., and Borowitz, M. J.Cooperative effects of gamma interferon and 1-alpha. 25-dihydroxyvitaminDj in inducing differentiation of human promyelocytic leukemia (HL-60)cells. Exp. Hematol., 14: 138-142, 1986.

12. Munker, R., Norman, A., and Koeffler. H. P. Vitamin D compounds. Effecton clonal proliferation and differentiation of human myeloid cells. J. Clin.Invest., 78:424-430. 1986.

13. Ball, E. D.. Howell, A. L., and Shen. L. Gamma interferon and 1,25-dihydroxyvitamin Dj cooperate in the induction of monocytoid differentiation but not in the functional activation of the HL-60 promyelocytic leukemiacell line. Exp. Hematol., 14: 998-1005. 1986.

14. Chouvet. C., Vicard, E.. Devonec, M., and Saez. S. 1,25-DihydroxyvitaminDj inhibitory effect on the growth of two human breast cancer cell lines(MCF-7. BT-20). J. Steroid Biochem., 24: 373-376, 1986.

15. Simpson, R. U.. and Arnold, A. J. Calcium antagonizes 1.25-dihydroxyvitamin Dj inhibition of breast cancer cell proliferation. Endocrinology, 119:2284-2289. 1986.

2852

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

VITAMIN D, PLATINUM. AND BREAST CANCER

16. Colston. K. W., Berger, U., and Coombes. R. C. Possible role for vitamin Din controlling breast cancer cell proliferation. Lancet, /: 188-191, 1989.

17. Koeffler. H. P.. Hirji. K., Itri, L., and the Southern California LeukemiaGroup. 1.25-Dihydroxyvitamin D3: in rivo and in vitro effects on humanpreleukemic and leukemic cells. Cancer Treat. Rep.. 69: 1399-1407. 1985.

18. Eisman. J. A.. Barkla, D. H.. and Tutton, P. J. M. Suppression of m vivogrowth of human cancer solid tumor xenografts by 1.25-dihydroxyvitaminD3. Cancer Res., 47: 21-25, 1987.

19. Honma, Y., Hozumi. M.. Abe, E., Konno, K., Fukushima, M., Hâta,S.,Nishii, V.. DeLuca. H. F.. and Suda, T. 1,25-Dihydroxyvitamin D3 and 1-«-hydroxyvitamin D3 prolong survival time of mice inoculated with myeloidleukem'ia cells. Proc. Nati. Acad. Sci. USA, 80: 201-204, 1983.

20. Chida, K., Hashiba, H., Fukushima, M., Suda, T., and Kuroki, T. Inhibitionof tumor promotion in mouse skin by 1.25-dihydroxyvitamin D3. CancerRes.. 45:5426-5430. 1985.

21. Robert. K. H.. Hellström. E.. Einhorn. S., and Gahrton, G. Acute myelogen-ous leukemia of unfavourable prognosis treated with retinoic acid, vitaminD3. alpha-interferon, and low doses of cytosine arabinoside. Scand. J. Hae-matol. Suppl. 44. .?•*.•61-74. 1986.

22. Hellstrom, E.. Robert, K. H., Gahrton. G., Mellstedt, H., Lindemalm, C,Einhorn. S., Björkholm. M., Grimfors, G., Uden. A. M., Samuelsson. J.,Ost. A.. Killander. A., Nilsson. B.. Winqvist, I., and Olsson, I. Therapeuticeffects of low-dose cytosine arabinoside, alpha-interferon. l-n-hydroxyvi-tamin D,. and retinoic acid in acute leukemia and myelodysplastic syndromes.Eur. J. Haematol., 40: 449-459. 1988.

23. Soule. H. D.. Vazquez. J., Long. A., Albert. S., and Brennan. M. J. A humancell line from a pleural effusion derived from a breast carcinoma. J. Nati.Cancer Inst.. 51: 1409-1416. 1973.

24. Brooks. S. C.. Locke. E. R., and Soule, H. D. Estrogen receptor in a humancell line (MCF-7) from breast carcinoma. J. Biol. Chem.. 248: 6251-6256,1973.

25. Ozols. R. F.. and Young. R. C. Chemotherapy of ovarian cancer. Semin.Oncol.. //: 251-263. 1984.

26. Schwartz, P. E., Keating. G., MacLusky. N.. Naftolin, F., and Eisenfeld, A.Tamoxifen therapy for advanced ovarian cancer. Obstet. Gynecol., 59: 583-588. 1982.

27. Markaverich. B. M.. Medina. D.. and Clark. J. H. Effects of combinationestrogen:cyclophosphamide treatment on the growth of the M XT transplant-able mammary tumor in the mouse. Cancer Res.. 43: 3208-3211, 1983.

28. Kiang. D. T.. Gay. J.. Goldman. A., and Kennedy. B. J. A randomized trialof chemotherapy and hormonal therapy in advanced breast cancer. N. Engl.J. Med.. 313: 1241-1246. 1985.

29. Santen, R. J.. Manni. A.. Harvey. H.. and Redmond. C. Endocrine treatmentof breast cancer in women. Endocrinol. Rev., //: 221-265, 1990.

30. Goldman, R. Synergism and antagonism in the effects of 1,25-dihydroxyvi-tamin Dj, retinoic acid, dexamethasone. and a tumor-promoting phorbolester on the functional capability of P388D1 cells: phagocytosis and trans-

glutaminase activity. Cancer Res., 45: 3118-3124. 1985.31. Marth, C., Helmberg, M., Mayer, I., Fuith, L. C., Daxenbichler, G., and

Dapunt, O. Effects of biological response modifiers on ovarian carcinomacell lines. Anticancer Res.. 9: 461-468, 1989.

32. Benz, C.. Cadman. E., Gwin. J.. Wu, T., Amara, J.. Eisenfeld, A., andDannies, P. Tamoxifen and 5-fluorouracil in breast cancer: cytotoxic syner-gism in vitro. Cancer Res.. 43: 5298-5303. 1983.

33. Clarke. R., Lippman, M. E., and Dickson, R. B. Mechanisms of hormoneand cytotoxic drug interactions in the development and treatment of breastcancer. Prog. Clin. Biol. Res.. 322: 243-278. 1990.

34. Christensen, C.. Deppe. G., Saunders, D., and Malviya, V. The use of ahuman endometrial carcinoma cell line (RL-95) for in vitro testing of chem-otherapeutic agents. Gynecol. Oncol., 28: 25-33. 1987.

35. Rosenberg, B. Fundamental studies with cisplatin. Cancer (Phila.), 55: 2303-2316. 1985.

36. Butler, W. B. Preparing nuclei from cells in monolayer cultures suitable forcounting and for following synchronized cells through the cell cycle. Anal.Biochem.. 141: 70-73, 1984.

37. Gessner. P. K. A straightforward method for the study of drug interactions:an isobolographic analysis primer. J. Am. Coll. Toxicol., 7:987-1012, 1988.

38. Church. M. W.. Dintcheff. B. A., and Gessner. P. K. The interactive effectsof alcohol and cocaine on maternal and fetal toxicity in the Long-Evans rat.Neurotoxicol. Teratol., 10: 355-361, 1988.

39. Carter. W. H., and Wampler, G. L. Review of the application of responsesurface methodology in the combination therapy of cancer. Cancer Treat.Rep.. 70: 133-140, ¡986.

40. Saunders. D. E., Christensen, C., Lawrence, W. D., Malviya. V. K., Malone,J. M., Jr., and Deppe, G. 1,25-Dihydroxyvitamin D3 receptor in gynecologicneoplasms. Proceedings of the 37th Annual Meeting for the Society ofGynecologic Investigation, Abstract 457, St. Louis, MO, March 1990.

41. Shinkai, T., Saijo. N.. Eguchi. K., Sasaki, Y., Tamura, T., Sakurai, M., Suga,J.. Nakano, H.. Nakagawa, K., Hong. W. S., and Nakajima. T. Cytogeneticeffect of carboplatin on human lymphocytes. Cancer Chemolhcr. Pharmacol.,21: 203-207, 1988.

42. Berenbaum, M. C. What is synergy? Pharmacol. Rev., 41: 93-141, 1989.43. Djulbegovi. B., Christman, S. E., Evans, G., and Moore, M. Studies of the

effect of 1,25-dihydroxyvitamin D3 on the proliferation and differentiationof the human promyelocytic leukaemia cell line HL-60. Biomed. Pharma-cother., 40:407-416, 1986.

44. Eisman. J. A.. Sutherland. R. L., McMenemy, M. L., Fragonas, J. C.,Musgrove, E. A., and Pang, G. Y. Effects of 1,25-dihydroxyvitamin D3 oncell-cycle kinetics of T 47D human breast cancer cells. J. Cell. Physiol., 138:611-616. 1989.

45. Gaczynski, M., Briggs, J. A., Wedrychowski, A.. Olinski. R., Uskokovic, M.,Lian, J. B., Stein, G. S., and Briggs, R. C. cis-Diamminedichloroplatinum(II)cross-linking of the human myeloid cell nuclear differentiation antigen toDNA in HL-60 cells following 1,25-dihydroxyvitamin D3-induced monocytedifferentiation. Cancer Res., 50: 1183-1188, 1990.

2853

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1991;51:2848-2853. Cancer Res   Y. L. Cho, C. Christensen, D. E. Saunders, et al.   Drugs on the Growth of MCF-7 Cells

and Platinum3Combined Effects of 1,25-Dihydroxyvitamin D

  Updated version

  http://cancerres.aacrjournals.org/content/51/11/2848

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/51/11/2848To request permission to re-use all or part of this article, use this link

on April 22, 2020. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from