[CANCER RESEARCH 52. 2413-2418, May 1. 1992]

Inhibition of Cancer Cell Growth by Calcium Channel Antagonists in the AthymicMouse1

J. M. Taylor and R. U. Simpson2

Department of Pharmacology [J. M. T., R. U. S.J and The Center for Molecular Genetics ¡R.U. S.J, The University of Michigan, Ann Arbor, Michigan 4SI09-0626

ABSTRACT

The calcium channel antagonists (CCAs) amlodipine, diltiazem, andverapamil inhibited HT-39 human breast cancer cell proliferation in aconcentration-dependent manner. The apparent 50% inhibitory dose values were 1.5 MM for the dihydropyridine amlodipine, 5 n\i for thebenzothiazapine diltiazem, and 10 MMfor the phenylalkylamine verapamil. Amlodipine treatment caused a rapid concentration-dependent decrease of intracellular calcium concentration in the HT-39 cell line.

Addition of 1 MMamlodipine had no effect on intracellular calcium levels,3 MMamlodipine lowered intracellular calcium levels in the HT-39 cells

by 13.7%, and 10 MMamlodipine lowered intracellular calcium levels by33.2%. Also, lowering medium calcium levels from 2.0 HIM to 0.5 MMresulted in a rapid 41.3% decrease in intracellular calcium and a concomitant 60% inhibition of HT-39 cell DNA synthesis. When HT-39 cells

were transplanted into athymic mice, marked hypercalcemia developed.Serum calcium levels from control mice were 8.3 ±0.6 mg/dl (mean ±SE; n = 4); those from tumor-bearing mice were 11.3 ±0.08 mg/dl (mean

±SE; n = 17). Blood calcium levels correlated directly with tumor size(r = 0.91, P < 0.01). We examined the capacity of three CCAs tospecifically inhibit HT-39 tumor growth ¡nvivo. One week after inoculation of HT-39 cells, mice were acclimated to vehicle or 0.1 mg/day

amlodipine, 1.0 mg/day diltiazem, or 1.0 mg/day verapamil, in theirdrinking water, for 7 days. Oral administration of the dihydropyridineamlodipine (0.35 mg/day) for 10 days inhibited HT-39 breast tumor

growth by 83.5 ±20.1% (mean ±SE). Oral administration of diltiazem(3.5 mg/day) inhibited HT-39 breast tumor growth rate by 46.5 ±6.6%over a 2-week measurement period, and verapamil (3.5 mg/day) inhibited

tumor growth rate by 68.2 ±9.7% (mean ±SE). The CCAs had no effecton mouse body weight or gross organ morphology at the concentrationsused. Lack of depolarization-induced calcium fluxes in the HT-39 cellline suggests that these cells do not express voltage-operated calcium

channels. Thus, our study correlates an effect of amlodipine to lowerintracellular calcium levels, by a mechanism not known at present, withits effect to inhibit HT-39 cell proliferation. These findings are important

since they demonstrate that amlodipine and other CCAs with knownpharmacodynamics and side effects act to blunt breast tumor progression

INTRODUCTION

Certain cancers are commonly associated with hypercalcemia, a phenomenon that has been related to the tumor'scapacity to secrete specific hypercalcemia-inducing factors (1,2). Although the benefit derived from this process is unclear,mobilization of calcium could be a mechanism that calcium-dependent cancers use for autostimulation of growth. Calciumis recognized as an important regulator of many essentialcellular functions, and in most proliferating cells calcium actsas a general mitogen to stimulate growth. Other second messengers associated with mitogenic effects include generated

Received 1/21/92; accepted 2/26/92.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.

' This work was supported by grants from NIH (CA-43859), the Children's

Leukemia Foundation of Michigan, and the Breast Cancer Research Institute ofthe University of Michigan.

2To whom requests for reprints should be addressed, at Department of

Pharmacology, University of Michigan Medical School, 6322 Medical ScienceBuilding 1, Ann Arbor, MI 48109-0626.

phospholipids and diacylglycerol. It has been shown that, in thepresence of diacylglycerol, protein kinase C is activated by arise in cytosolic free calcium (3, 4). Once activated, proteinkinase C isoenzymes catalyze the phosphorylation of a numberof cellular proteins necessary for proliferation (3-5). In addition, transient rises in cytosolic calcium have been shown toinitiate activation of the calcium receptor calmodulin, whichmay also play an important role in the regulation of proliferation (3). As reviewed by Metcalfe et al. (6), tumors are generallyrecognized to have unusually high calcium levels. It has beensuggested that the high calcium level is due to either excessiveinflux of extracellular calcium or the ability of neoplastic mitochondria to retain higher concentrations of calcium (6). It isplausible that high levels of intracellular calcium yield increasedactivation of the calcium second messenger system resulting inthe exaggerated growth rate of certain malignant cells.

We previously reported that the effects of extracellular calcium on malignant cell growth are cell type specific (7, 8). Wehave shown that HT-39 and MCF-7 human breast cancer celllines and the HL-60 human promyelocytic leukemia cell lineexhibit calcium-dependent proliferation, while the growth ofthe LI210 leukemia cell line is inhibited by increases in extracellular calcium (7, 8). Okazaki et al. (9) have confirmed theseresults, showing that the HL-60 cell line exhibits extracellularcalcium-dependent growth. In addition, Yoneda et al. (IO)showed that growth of a breast carcinoma cell line (VX2) wastightly regulated by extracellular calcium concentrations. However, others have shown that removal of calcium from themedium did not alter cell growth of certain tumorigenic celllines derived from transformed fibroblasts (11), liver hepatomas(12), mouse embryo 3T3 cells (13), and human ovarian cells(14), whereas, with their untransformed ancestors, removal ofcalcium from the medium inhibited cell growth. Clearly, somebut not all tumor cell growth is dependent on extracellularcalcium.

It is now recognized that cytosolic calcium concentrationsare rapidly modulated by mobilization of calcium from intracellular stores and regulation of the conductance of membranecalcium channels. Calcium channel blockers have been mostconsistently shown to regulate calcium influx by modulatingthe duration of the open/closed period of the voltage-dependentslow calcium channel in the plasma membrane. However, othersites for their actions have been demonstrated (15). We hypothesized that it would be possible to control pharmacologicallythe proliferation of calcium-dependent cancer cells by blockingthe exaggerated effects of calcium through the use of CCAs.1

Previously, we reported that two CCAs, verapamil and diltiazem, block the mitogenic action of calcium on a calcium-dependent breast cancer (HT-39) cell line in vitro (12). In thepresent study, we explored the possibility that HT-39 tumorgrowth in athymic mice could elevate serum calcium levels. In

3The abbreviations used are: CCA, calcium channel antagonist; VOCC. voltage-operated calcium channel; EGTA, ethylene glycol bis (fi-aminoethyl ether)-/VyV,A",A"-tetraacetic acid; DMEM. Dulbecco's modified Eagle medium;HEPES, A'-2-hydroxyethylpiperazine-A/'-ethanesulfonic acid; PCS, fetal calfserum; AM, acetoxymethyl ester; [Ca2*]¡,intracellular calcium concentration.

2413

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

CALCIUM AND CANCER

addition, we tested the effects of the CCAs diltiazem, verapamil,and amlodipine on HT-39 tumor growth in vivo.

MATERIALS AND METHODS

Cell Culture Conditions. HT-39 cells were grown in monolayer culturein DMEM supplemented with 10% PCS, 10,000 units/ml penicillin,and 10 mg/ml streptomycin. Cells were maintained at 37°Cunder a

5% CCK/air atmosphere. The growth medium was replaced every 4days. After reaching confluency, cells were trypsinized and passaged1:4 with fresh medium containing serum.

Measurement of DNA Synthesis. To assess drug effects on cell DNAsynthetic capacity, monolayer cultures of HT-39 cells were trypsinizedwith a trypsin-EDTA solution containing 0.5 g/liter trypsin and 0.2 g/liter EDTA (Sigma) and were subcultured in DMEM with 10% PCS,at densities of approximately IO4 cells/cm2, in 25-cm2 flasks (Costar,

Cambridge, MA). After incubation overnight (16-24 h), medium wasreplaced with DMEM plus 10% PCS and appropriate concentrationsof drug or ethanol vehicle (<0.1%) or with modified Eagle's medium

containing variable concentrations of calcium and 10% calcium-depleted PCS. Ethanol alone, at up to 0.1% (v/v), does not alter HT-39cell growth. We prepared a low-calcium modified Eagle's medium

containing 10% PCS by pretreating the PCS with EGTA, to yield PCShaving calcium levels of 0.5 ¿iM.This routinely required 2.3 mM EGTA.The total calcium content was adjusted in this medium by addition ofsterile 1.0 M CaClj solution. The total calcium concentration wasassessed by colorimetrie quantitation, using a diagnostic kit (Sigma)and an absorption spectrophotometer, as previously described (7). Toassess DNA synthesis, cells were labeled with l ¿iCi/ml[3H]thymidinefor l h at 37'C, as previously described (7). Cells were harvested by

trypsinization, and aliquots were taken for cell number determinationusing a Coulter counter (model ZF; Coulter Electronics). The remainingcells were washed with phosphate-buffered saline (Sigma), solubilizedwith l N NaOH, and neutralized with l N HC1. The remaining acid-insoluble material was precipitated by addition of ice-cold 10% trichlor-acetic acid. The precipitates were collected on glass fiber filters (Whatman GF/C). Protein was measured by the methods of Bradford (16).

Measurement of Cytoplasmic Calcium Concentrations. Intracellularcalcium concentrations were determined by the use of the fluorescentcalcium indicator indo-1/AM. Preliminary experiments using fura-2/AM as the calcium indicator revealed that the CCA amlodipine emitsfluorescence using the excitation and emission wavelengths for fura-2/AM. Cells were suspended in 10 ml buffer A (142 mM NaCl, 4.5 mMKCI, 2.2 mM CaCl2, 3.6 mM NaHCO.,, 1 HIMMgCl2, 0.6 HIMo-glucose,30 mM Na-HEPES buffer, pH 7.4), at a concentration of 1 x IO7cells/ml. Cells were loaded with 1 MMindo-1/AM for 15 min at 37°C.The

loaded cells were then diluted to 40 ml with buffer A, collected bycentrifugation at 500 x g for 60 s, and resuspended in an additional 10ml of buffer A. The centrifugation step was then repeated and cellswere resuspended in 4 ml buffer A. Fluorescent measurements weremade on 1-ml volumes of HT-39 cells that were stirred constantly andmaintained at 37°C.Amlodipine or 0.01% ethanol control was added

in a 10-Mlvolume after 2-5 min of base-line measurement. The fluo-rometer used was an Amico-Bowman spectrophotofluorometer (Actuation= 331 nm; ^„¡„¡„n= 398 nm) interfaced to both a strip-chart recorderand an LED readout, from which quantitative fluorescence values wereobtained. After measurement of the fluorescence (F), cells were lysedby the addition of 100 MMdigitonin, Tris base was added (pH 8), anda minimum fluorescence value (Fm,„)was obtained after chelation ofcalcium by the addition of 20 mM EGTA. A maximum fluorescencevalue (/"„,„)was then obtained after addition of 20 mM CaCU. Intracellular calcium concentrations ([Ca2+]¡)were calculated from the equation[Ca2+]¡= (F - FmJFm,t - F) x Kd, where Kd is the affinity of indo-1

for calcium (assumed to be 250 DM)(17).Maintenance of Athymic Mice and Experiments to Assess CCA Ef

fects on Tumor Growth. CD-I nu/nu athymic mice (14-16 days of age)were purchased from Charles River and housed in sterile laminar flowrooms. One week after arrival, mice were inoculated with 1 x IO6cells

in 200 n\ RPMI/20 mM HEPES (tumor bearing) or with 200 ii\ RPMI/

20 mM HEPES (control), s.c. into the lower back, using a 22-gaugeneedle. One week after inoculation, mice were initially treated witheither 0.1 mg/day amlodipine, 1.0 mg/day diltiazem, or 1.0 mg/dayverapamil, in their drinking water, to equilibrate the mice to thesedrugs. After an additional 1 week, mice were treated with either 0.35mg/day amlodipine, 3.5 mg/day diltiazem, or 3.5 mg/day verapamil,in their drinking water. The first day of administration of the CCAs atthe final concentrations was designated day 0. This stepped dosageincrease in drug concentration prevented the 20% observed mortalityof mice which occurred when mice were initially treated with the higherfinal concentrations of CCAs. The reason for this 20% mortality rateis not known. According to measurements of water consumption, thefinal concentration of drug received was approximately 7.3 mg/kg/dayfor amlodipine and 73 mg/kg/day for either verapamil or diltiazem.Stock solutions of CCAs were made in 50% ethanol/water and wereadded in appropriate concentrations to drinking water. Tumor-bearingcontrol animals received equal amounts of ethanol (0.1%). Stock solutions were made up fresh, and water was changed every 3 days. Tumorsize was measured with calipers every 3 days for up to 8 weeks. Tumorvolume was calculated as length x width x height and expressed asmm-1. Four to 8 weeks after inoculation, mice were sacrificed, and

tumor size and blood calcium concentrations were determined. Cardiacpunctures were performed to collect blood from each animal. Serumcalcium concentrations were determined by colorimetrie quantitation,using a diagnostic kit (Sigma) and an absorption spectrophotometer,as previously described (11). All procedures using the athymic miceconformed strictly to the guidelines of the University of MichiganLaboratory for Animal Medicine Group, which reviewed ourexperiments.

RESULTS

We tested the effects of CCAs on the proliferation of culturedHT-39 cell lines. HT-39 cells in logarithmic phase growth weretreated with vehicle (<0.1% ethanol) or CCAs (10~5 to 10~8M)

for 48 h, in normal DMEM/10% FCS. The final calciumconcentration in this medium was 2.0 mM. Fig. 1 shows theeffect of amlodipine, diltiazem, and verapamil on HT-39 DNAsynthetic capacity. A concentration-dependent inhibition of cellgrowth was observed for all three CCAs, with significant inhibition of cell proliferation at concentrations as low as 0. l /IMfor amlodipine, 1.0 UMfor diltiazem, and 1.0 /IM for verapamil.The apparent 50% inhibitory dose values were 1.5 /¿Mforamlodipine, 5 /¿Mfor diltiazem, and 10 /¿Mfor verapamil.

We next examined the action of the most potent CCA,

•¿�jr_

<f¡

100

60-

íZ. 40-

20-

10-» 10•¿�' io-*

CCA (M)

10-

Fig. 1. Dose-response curve for the effects of CCAs on HT-39 breast cancercell DNA synthesis. Cultured HT-39 cells in logarithmic phase growth weretreated with 0.05% ethanol vehicle or 10~! to 10~8 M amlodipine (•),diltiazem

(•),or verapamil (A), for 48 h, in DMEM/10% FCS. The final calcium concentration in this medium was 2.0 mM. Incorporation of ['HJthymidine (cpm/Mgprotein) was measured as described in "Materials and Methods." Data arepresented as percentage of inhibition of drug-treated cells, compared to vehicle-treated control cells (mean ±SE of four observations).

2414

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

CALCIUM AND CANCER

amlodipine, on modulation of intracellular calcium levels in theHT-39 cell line. Cells were incubated with 1 JIMindo-1/AM, asdescribed in "Materials and Methods." Fig. 2 shows a repre

sentative tracing of HT-39 cells treated with 1, 3, and 10 pMamlodipine. A concentration-dependent decrease in intracellu-lar calcium was observed. Control tracings with three consecutive additions of vehicle (0.01 % ethanol) had no effect on [Ca2+]

,. As shown in Fig. 2, addition of l ¿¿Mamlodipine had no effecton [Ca2+]i,while 3 /¿Mamlodipine lowered intracellular calcium

by 15%, to 79.8 HM; 10 UM amlodipine lowered intracellularcalcium by 45%, to 52 HM.Base-line [Ca2+]¡was 94.5 nM in this

tracing. Cumulative results from experiments examining theeffect of amlodipine on intracellular calcium levels in the HT-39 cell line are shown in Table 1. The mean base-line intracellular calcium concentration for all experiments was 81.5 ±9.6DM(mean ±SE). Addition of l UMamlodipine had no effecton intracellular calcium levels, 3 ^M amlodipine lowered [Ca2+]¡in the HT-39 cells by 13.7%, and 10 ^M amlodipine lowered[Ca2+]iby 33.2 ±2.5%. Both 3 and 10 ^M amlodipine significantly (P < 0.01) decreased [Ca2*]¡levels, relative to base linewith no-drug-treated cells. In addition, statistically significantdecrements in intracellular calcium were observed between theadditions of 1 and 3 n\\. as well as between 3 and 10 ¿AIamlodipine (P<0.01).

Growth of the HT-39 cell line is regulated by extracellularcalcium levels (11, 12). The data in Table 2 show the effects ofa 72-h incubation of 3.0, 1.4, and 0.1 HIMcalcium-containingmedium on HT-39 cell proliferation. The medium was preparedas described under "Materials and Methods." Analysis of HT-

39 cell growth revealed that lowering the medium calciumconcentration from 1.4 to 0.1 HIM significantly (P < 0.01)inhibited cell proliferation, by 20 ±0.6% and 34.9 ±2.4%, asmeasured by changes in cell number and ['Hjthymidine incor

poration, respectively. Raising medium calcium from 1.4 to 3.0mM stimulated cell proliferation by 51.8 ±0.89% and stimulated DNA synthesis by 66.4 ±3.8%. At the highest (3.0 mM)concentration of calcium used, precipitation of calcium phosphate or calcium-induced clotting of serum did not occur. Thus,our data are not the result of these possible artifacts (4).Alterations of extracellular calcium relate to changes in intracellular calcium. Reducing extracellular calcium levels to 500nM, by the addition of 2.3 mM EGTA to a medium containing2.2 mM calcium, immediately lowered intracellular calciumlevels in the HT-39 cell line by 41.3%. Importantly, a 48-h

—¿�100

AMLODIPINEI 3uM| lOuMI

Y ! Y

024Time (min)

Fig. 2. Effect of amlodipine on intracellular calcium levels in HT-39 cells. HT-39 cells were loaded with 1 MMindo-1/AM at 37'C for 20 min, as described in"Materials and Methods." Base-line |Ca2*]¡was recorded, followed by sequential

additions of 1, 3, and 10 //M amlodipine. Final ethanol concentration was 0.03%.Base-line [Ca2*),was 94.5 nM. Amlodipine (3 and 10 MM)lowered [Ca2*];to 79.8

nM and 52 nM, respectively.

Table 1 Effects of amlodipine on [Ca2']i in the HT-39 cell line

Base-line [Ca2*]¡was recorded, followed by additions of 1, 3, or 10 MM

amlodipine.Decrease in [Ca2*l¡(%)

Cell line 3 MM 10 MM

HT-39 1.5± l.l(n = 8) 13.7 (n = 2)* 33.2 ±2.5 (n = %)'

" Amlodipine concentration.* P < 0.01, statistically significant difference from 1.0 MMvalue and base-line

[Ça2!.' P < 0.01, statistically significant difference from 3.0 MMvalue and base-line

[Ca2*],

treatment of HT-39 cells with normal DMEM/10% FCS containing 2.3 mM EGTA yielded a final medium calcium concentration of 0.5 fiM and resulted in a 59.7 ±8.9% inhibition ofcell DNA synthesis. These data suggest that modulations ofintracellular calcium either by changes in medium calcium orby CCAs result in an inhibition of HT-39 cell proliferation.

To further understand the mechanism by which amlodipinelowers intracellular calcium, we tested for the presence of L-type calcium channels in the cultured HT-39 cells. One featureof these voltage-gated channels is that their current can beenhanced by depolarizing the plasma membrane by addition of50 mM KC1 to cell suspensions (18). HT-39 breast cancer cellsand SK-N-SH human neuroblastoma cells were loaded with 1MMindo-I/AM, as described under "Materials and Methods."

After a base-line reading on the spectrophotofluorometer wasobtained, the cells were depolarized by the addition of 50 mMKC1. No change was observed in intracellular calcium levels inthe HT-39 cell line (data not shown). In contrast, addition of50 HIM KC1 to SK-N-SH cells, a cell line known to expressVOCCs, caused a rapid increase in [Ca2+]i, from 51.8 nM to

129 nM. Lack of depolarization-induced calcium fluxes in theHT-39 cell line suggests that these cells do not express VOCCs.These data suggest that amlodipine is acting to lower intracellular calcium levels through a mechanism other than its knownaction as a plasma membrane calcium channel antagonist.

Since breast cancer cell growth has been associated withhypercalcemia, we explored the possibility that HT-39 tumorgrowth in athymic mice could elevate serum calcium levels.Eight weeks after inoculation of HT-39 cells into athymic mice,tumor size and blood calcium levels were measured. Fig. 3shows that HT-39 tumors propagated in mice increased serumcalcium levels. Control non-tumor-bearing mice had serumcalcium levels of 8.3 ±0.6 mg/dl (mean ±SE; n = 4), whiletumor-bearing mice had significantly (P < 0.01) higher serumcalcium levels, 11.3 ±0.8 mg/dl (mean ±SE; n = 17). Furthermore, a significant (r = 0.91; P < 0.01) correlation existedbetween the degree of hypercalcemia and the size of the tumorin the animals. There was no significant difference (P > 0.05)in body weight between the control animals (28.4 ±0.77 g) andthe tumor-bearing animals (28.5 ±0.19 g). Necropsy revealedthat the primary tumors had not directly invaded bone, and nométastasesoccurred (data not shown), suggesting that the HT-39 human breast cancer cell line is capable of inducing hypercalcemia by the release of humoral factors.

We next tested the effect of CCAs on HT-39 cell growth invivo. One week after inoculation with HT-39 cells, mice weretreated with CCAs in their drinking water, as described under"Materials and Methods." The effect of p.o. administration of

amlodipine on growth of HT-39 tumors is shown in Fig. 4. InFig. 4A, tumor size is shown at day 0 and after 10 days oftreatment with amlodipine (0.35 mg/day) or vehicle (0.1%ethanol). Amlodipine treatment inhibited tumor size by 83.5 ±

2415

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

CALCIUM AND CANCER

Table 2 Effect of extracellular calcium on DNA synthesisCultured logarithmic phase growing HT-39 cells in DMEM/10% PCS were transferred to modified Eagle's medium with 10% calcium-depleted FCS. The final

medium calcium levels were adjusted to 3.0, 1.4, or 0.1 mm. After 3 days, changes in cell number and DNA synthesis were measured. The data presented are themean ±SE of four observations.

3.0 m\i Ca2* 1.4mM Ca!* 0.1 HIMCa2*

Cell no.DNA synthesis (cpm thymi-

dine incorporated/h/10*

cells)

21.04 ±0.31 x 10'9.27 ±0.46 x IO4

13.86 ±0.05 x 10'"5.57 ±0.07 x 104°

ll.09± 0.31 x 10'*4.13 ±0.28 x IO4*

" P< 0.01, statistically significant difference from 3.0 m\i value.* /*< 0.01. statistically significant difference from 1.4 m\i value.

: 4 6 8TUMOR SIZE (cm3)

10

Fig. 3. Relationship of serum calcium concentration to size of HT-39 celltumor propagated in athymic mice. Mice were inoculated with HT-39 cells (4.5x 10' cells/animal) in 200 »\of RPMI/20 mM HEPES (tumor-bearing, n = 17)or with 200 ¡tiof RPMI/20 m\i HEPES (control, n = 4). Eight weeks afterinoculation, tumor size and blood calcium levels were measured. Tumors propagated in mice yielded elevated serum calcium levels (control, 8.3 ±0.6 mg/dl;tumor, 11.3 ±0.8 mg/dl; P< 0.04). A significant (r = 0.9!;/>< 0.01) correlationexists between the severity of hypercalcemia and the size of the tumor in theanimal.

20.1 % (P < 0.01 ). Tumors continued to grow rapidly in vehicle-treated mice, whereas tumor size was unchanged in the amlo-dipine-treated mice. As shown in Fig. 4B, control tumors weresignificantly larger than tumors from the amlodipine group byday 8 (P < 0.05). By day 15, mean tumor volume in the controlmice approached 700 mm3, whereas tumors in the amlodipine-treated mice were 100 mm1. Excised tumors from vehicle-

treated mice (n = 5) weighed 0.67 ±0.25 g (mean ±SE); thosefrom amlodipine-treated mice (n = 3) weighed 0.06 ±0.05 g(P < 0.05). There was no difference in body weight between thetumor-bearing vehicle-treated mice (22.3 ±0.6 g) and tumor-bearing mice treated with amlodipine (22.3 ±1.7g). In anotherexperiment, the effects of amlodipine treatment for a longertime period were examined. On day 0, control tumors were62.7 ±13.3 mm3 and amlodipine-treated tumors were 44.8 ±14.8 mm3. Over a 38-day treatment period, control tumorvolume increased to 1251 ±335 mm3, whereas tumors fromthe amlodipine group were only 321 ±51 mm3 (P < 0.05).

After 38 days the animals were sacrificed, and tumors wereexcised and weighed. In this experiment, tumors from vehicle-treated mice (n = 4) weighed 0.61 ±0.16 g, while tumors fromamlodipine-treated mice (n = 4) weighed 0.18 ±0.03 g (P <0.05). There was no difference in body weight between thetumor-bearing vehicle-treated mice (23.2 ±0.4 g) and tumor-bearing mice treated with amlodipine (22.9 ±1.1 g). We nexttested the effect of the CCAs verapamil and diltiazem on HT-39 tumor growth in vivo. In this experiment, tumor volumemeasurements were started on day 14 and taken every third dayfor 2 weeks. Fig. 5 shows the effect of p.o. administration ofdiltiazem and verapamil on the rate of HT-39 tumor growth inathymic mice over the 2-week measurement period. Significantreductions in the rate of tumor growth were observed with both

CCAs, compared to the rate of tumor growth in vehicle-treatedcontrol mice. Diltiazem inhibited tumor growth rate by 46.5 ±6.6% (P < 0.01). Verapamil inhibited tumor growth rate by68.2 ±9.7% (P < 0.01). The animals were sacrificed, andtumors were excised and weighed. Tumors from vehicle-treatedmice (n = 11) weighed 1.24 ±0.8 g (mean ±SE), while tumorsfrom verapamil-treated mice (n = 13) weighed 0.35 ±0.1 g(mean ±SE; P < 0.05). Although diltiazem treatment significantly slowed tumor growth rate, tumor weight at the end ofthe experiment was 0.57 ±0.2 g (mean ±SE; n = 11), a valuenot significantly different from the vehicle-treated mice tumors(1.24 ±0.8 g). This apparent discrepancy of significant reduction in tumor growth rate but not tumor size at the end of theexperiment is due to the large variability of tumor size on day14 (control, 192.2 ±83.4 mm3; diltiazem, 256.6 ±61.3 mm3;verapamil, 371.0 ±104.1 mm3). There was no difference inbody weight between the tumor-bearing vehicle-treated mice(29.0 ±0.9 g) and tumor-bearing mice treated with verapamil(27.2 ±1.1 g) or diltiazem (27.7 ±1.5 g). In all experimentsusing athymic mice, internal organs were examined by theUniversity of Michigan Laboratory Animal Medicine group;the heart, liver, lungs, and uterus appeared normal.

DISCUSSION

In previous studies, we showed that HT-39 and MCF-7human breast cancer cells and HI 60 human leukemia cellsrespond to higher extracellular calcium with an increase inproliferation, whereas the proliferation of L1210 mouse leukemia cells is inhibited by elevated extracellular calcium (7, 8). Inaddition, we showed that the calcium-sensitive HT-39 cell lineis growth inhibited by the CCAs verapamil and diltiazem, inculture (8). Recently, many studies have exploited CCAs toreverse the multidrug resistance phenomenon commonly observed in cancer therapy (19-21). Interestingly, in support ofour initial report, a study exploring the verapamil-inducedaccumulation of daunomycin noted that verapamil alone killedthe cultured daunomycin-resistant B-30 Chinese hamster ovary

cell line (19).The studies in this report show for the first time that CCAs

alone specifically inhibit breast cancer cell growth in vivo.Amlodipine inhibited tumor growth rate by 106.3 ±14.9% (P< 0.001), diltiazem inhibited tumor growth rate by 46.5 ±6.6%(P < 0.01), and verapamil inhibited tumor growth rate by 68.2±9.7% (P< 0.01). Importantly, amlodipine achieved this effectat 10-fold lower dosages than diltiazem and verapamil. Therewas no change in body weight between the tumor-bearing drug-treated and tumor-bearing vehicle-treated groups, suggestingthat little toxicity was observed at the levels of CCAs administered. In addition, these levels of CCAs inhibited tumor proliferation without gross side effects on normal tissue.

CCA drugs can be divided, on the basis of structure, into2416

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

CALCIUM AND CANCER

ìili

e 200

100-

CONTROL AMLODIPINK

TREATMENT

B 800

fj control

amlortipiiu-

10TI M K (days)

Fig. 4. Effect of amlodipine on HT-39 cell growth in vivo. Athymic mice wereinoculated with HT-39 cells, as described in "Materials and Methods." Mice were

treated with vehicle or 0.1 mg/day amlodipine in their drinking water for 1 week,followed by 2 weeks of treatment with 0.35 mg/day amlodipine.. I, tumor size atthe day of dosage and after 10 days of treatment with amlodipine or vehicle B,tumor development over a 2-week period. Data represent mean ±SE for fiveanimals in the control group and three animals in the amlodipine treatmentgroup. By day 8, control tumors were all significantly larger than those in theamlodipine-treated group.

three subclasses of compounds, the dihydropyridines, the ben-zothiazapines, and the diphenylalkylamines. Each class showsheterogeneity in tissue binding and selectivity of organ effects(22, 23). The tissue variations in sensitivity to different subclasses of CCAs suggest that selective effects of particular CCA-like drugs for inhibiting cancer cell proliferation may also exist.The clinical effects of CCAs have been well characterized. CCAsare currently used therapeutically to control angina, ventriculartachyarrythmia, and hypertension. Also, CCAs are used incancer chemotherapy to reverse the multidrug resistance commonly associated with long term drug therapy. Clinical trialshave determined the pharmacokinetic characteristics, the dosage modifications required in certain disease states, and thedrug interaction profiles for several CCAs (24-27). In contrastto other dihydropyridines, amlodipine has a much longer half-life and greater Unavailability but has a slower onset of action(28). Importantly, p.o. administration of amlodipine (10 nig/kg/day) has been maintained in normotensive rats for up to 30weeks with no apparent toxic effects (29).

Interestingly, our data suggest that the HT-39 cell line doesnot express the dihydropyridine-sensitive VOCC. A functionalcriterion for the VOCC is the detection of depolarization-induced calcium fluxes (18). Using the fluorescent dye fura-2,Sher et al. (18) have shown that an immediate rise in intracel-lular calcium occurs upon depolarization of small lung carcinoma cells with 50 mM KC1. Reversal of the KCl-inducedincrease in intracellular calcium by verapamil and nifedipine

suggests the presence of L-type calcium channels in the smallcell carcinoma cell line (18). In our study, exposure of thehuman neuroblastoma SK-N-SH cell line to 50 mM KC1 causedan immediate rise in [Ca2+]i.However, exposure of the HT-39cell line to 10, 50, or 100 mw KC1 had no effect on [Ca2+]¡.The

G o DAYS CCAs inhibit HT-39 cell growth with the same relative potency•¿�10DAYS as observed for the antagonism of the VOCC, and inhibition of

HT-39 cell growth by CCAs can be attenuated with increasedlevels of calcium in the medium (11, 12). This suggests that theHT-39 cell line expresses voltage-operated calcium channelsand that modulation of these calcium channels by CCAs lowersintracellular calcium and affects cell growth. Conversely, a lackof depolarization-induced voltage-operated calcium fluxes inthe HT-39 cells suggests that these cells do not express VOCCs.The possibility exists that the HT-39 cell line expresses analtered dihydropyridine-sensitive calcium channel that is insensitive to KC1 activation. Alternatively, the action of amlodipineto inhibit cancer cell growth and to lower [Ca2+]¡specifically

could be by a mechanism other than its known action to blockvoltage-dependent calcium channels. Stimulation of the sarcoplasmic reticulum Ca2+-ATPase has been reported for CCAs inthe micromolar range (15). Stimulation of a similar Ca2+-

ATPase in the endoplasmic reticulum could account for decreased [Ca2+]i in amlodipine-treated cells. Reports have also

noted that CCAs can accumulate within muscle cells (30). Theability of these drugs to enter muscle cells supports the possibility that CCAs may exert secondary actions on an internalsite of the cell, such as the endoplasmic reticulum, to reduceintracellular calcium levels.

The data presented in this report correlate a lowering ofintracellular calcium levels with an inhibition of DNA synthesis.Thus, amlodipine's growth-inhibitory actions could potentiallyresult from the observed inhibition of [Ca2+]¡.The concentrationof amlodipine required to cause a significant inhibition of [Ca2+]

¡was 3 /tM, and the apparent 50% effective concentration forinhibition of cell growth was 1.5 /¿M.It is presently unclearwhether the 13% decrease in [Ca2+]¡observed after treatmentwith 3 /¿Mamlodipine is sufficient to result in a 55-60%inhibition of DNA synthesis. It is possible that a thresholdconcentration of intracellular calcium must be maintained inthe HT-39 cells and amlodipine treatment causes a population

TUMORGROWTH(%INCREASE/DAY)s

äSSèT*T*

CONTROL DILTIAZEM VERAPAMII,

TREATMENTFig. 5. Effect of diltiazem and verapamil on HT-39 cell growth m vivo.Athymic

mice were inoculated with HT-39 cells, as described in "Materials and Methods."

One week after inoculation, mice were treated with vehicle (n II), diltiazem(1.0 mg/day; n = 11), or verapamil (1.0 mg/day; n = 13), in their drinking water.After an additional I week, mice were treated with vehicle, diltiazem (3.5 mg/day), or verapamil (3.5 mg/day). in their drinking water, on designated day 0. Onday 14, tumor volume measurements were taken every 3 days for 2 weeks. Rateof tumor growth is expressed as a daily growth rate normalized to percentage ofincrease over tumor size on day 14. Data are presented as mean ±SE [control.37.4 ±4.0%; diltiazem, 20.0 ±4.1% («,/><0.01); verapamil, 11.9 ±3.0% (*, P

2417

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

CALCIUM AND CANCER

of cells to fall below that level. Amlodipine may be acting toblock calcium-dependent signal transduction, and this mayblock the mitogenic action of growth factors that utilize thecalcium messenger. Other possible mechanisms of action foramlodipine to inhibit cell growth cannot be ruled out. Asreviewed by Zerning (15), both functional and binding studieshave suggested a number of possible intracellular sites of actionfor CCAs, including protein kinase C, calmodulin, and phos-phodiesterase. Modulation of one of these sites by amlodipinecould also potentially result in decreased proliferation, and weare currently examining the sensitivity of these enzymes toCCAs.

Because of their pronounced inhibitory effects on the proliferation of HT-39 human breast carcinoma cells CCAs mayrepresent an alternative direction for cancer chemotherapy.Presently, there are a number of endocrine and antimetabolitechemotherapeutic agents available for treatment of cancers.However, none of these agents is universally efficacious, andeach of these agents has side effects. Our data suggest thatCCAs have specific actions to inhibit cancer cell growth in vivo,with no observed side effects. Clinical trials for the use of thesedrugs in the treatment of hypertension have determined thatlong term treatment with these agents at the concentrationsused in our study is not associated with significant toxicities.These studies suggest that CCAs alone, as a class of drugs,particularly amlodipine, may be useful in controlling calcium-dependent cancer cell proliferation.

REFERENCES

1. Mundy, G. R.. and Martin, T. J. The hypercalcemia of malignancy: patho-genesis and management. Metabolism, 31: 1247-1277. 1982.

2. Tashjian, A. H., Voelkel, E. F., Lloyd, W., Derynck, R., Winkler, M. E., andLevine, L. Actions of growth factors on plasma calcium. Epidermal growthfactor and human transforming growth faclor-alpha cause elevation of plasmacalcium in mice. J. Clin. Invest., 78: 1405-1409, 1986.

3. Rasmussen, II.. and Barret, P. Q. Calcium messenger system: an integratedview. Physiol. Rev., 64:938-984, 1984.

4. Durham, A. C., and Walton. J. M. Calcium ions and the control of proliferation in normal and cancer cells. Biosci. Rep., 2: 1S-30, 1982.

5. Nishizuka, Y. The role of protein kinase C in cell surface signal transductionand tumor promotion. Nature (Lond.). 308: 693-698, 1984.

6. Metcalfe, J. C., Pozzan, T., Smith, G. A., and Hesketh, T. R. A calciumhypothesis for the control of cell growth. Biochem. Soc. Symp., 45: 1-26,1980.

7. Simpson, R. U., and Arnold, A. J. Calcium antagonizes 1,25-dihydroxy-vitamin D, inhibition of breast cancer cell proliferation. Endocrinology, 119:2284-2289, 1986.

8. Simpson, R. U., and Taylor, J. M. Regulation of cancer cell proliferation bycalcium antagonists. Exp. Oncol. (Life Sci. Adv.), 7: 81-89, 1988.

9. Okazaki, T., Mochizuki, T., Tashima, M., Sawada, H., and Uchino, H. Roleof intracellular calcium ion in human promyelocytic leukemia III 60 celldifferentiation. Cancer Res., 42:6059-6063, 1986.

10. Yoneda, T., Kitamura, M., Ogawa, T., Aya, S., and Sakuda, M. Control ofVX2 carcinoma cell growth in culture by calcium, calmodulin, and prosta-glandins. Cancer Res., 45: 398-405, 1985.

11. Balk, S. D., Polimeni. P. I., Hoon, B. S., Lestourgeon, D. N., and Mitchell,R. S. Proliferation of Rous sarcoma virus-infected, but not of normal, chickenfibroblasts in a medium of reduced calcium and magnesium concentration.Proc. Nati. Acad. Sci. USA, 76: 3913-3916, 1979.

12. Swierenga, S. H., Whitfield, J. F., and Morris, H. P. The reduced extracellularcalcium requirement for proliferation by neoplastic hepatocytes. In Vitro,/* 527-535, 1978.

13. Paul, D., and Ristow, H. J. Cell cycle control by Ca2* ions in mouse 3T3cells and in transformed 3T3 cells. J. Cell. Physiol., 98: 31-39, 1979.

14. Chan, T. C. K. Calcium-independent growth of human ovarian carcinomacells. J. Cell. Physiol.. 141: 461-466, 1989.

15. Zerning, G. Widening potential for Ca3* antagonists: non-L-type Ca2* channel interaction. Trends Pharmacol. Sci., //: 38-44, 1990.

16. Bradford, M. M. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding.Anal. Biochem., 72: 248-254, 1976.

17. Grynkiewicz, G., Poenie, M., and Tsien, R. Y. A new generation of Ca2*

indicators with greatly improved fluorescent properties. J. Biol. Chem., 260:3340-3350. 1985.

18. Sher, £.,Pandiella, A., and dementi, F. Voltage-operated calcium channelsin small cell lung carcinoma cell lines: pharmacological, function, and inimunological properties. Cancer Res., 50: 3892-3896, 1990.

19. Cano-Gauci, D. F., and Riordan, J. R. Action of calcium antagonists onmultidrug resistant cells: specific toxicity independent of increased cancerdrug accumulation. Biochem. Pharmacol., 36: 2115-2123, 1987.

20. Fojo, A., Akiyama, S.. Gottesman. M., and Pastan. I. Reduced drug accumulation in multiple drug-resistant human KB carcinoma cell lines. CancerRes., 45: 3002-3007. 1985.

21. Zacharski, L. R. Rationale for treatment of cancer with calcium antagonists:a review. J. Med. Chem., 19: 145-177, 1988.

22. Janis, R. A., and Triggle, D. J. New developments in Ca2* channel antagonists. J. Med. Chem., 26: 775-785, 1983.

23. Gould, R. J., Murphey, K. M., and Snyder, S. H. Tissue heterogeneity ofcalcium channel antagonist binding sites labeled by 'H-nitrendipine. Mol.Pharmacol., 25: 235-241, 1984.

24. Reiter, M. J., Shand, D. G., Aanonsen. L. M., Wagoner, R., McCarthy, E.,and Pritchett, E. L. Pharmacokinetics of verapamil: experience with a sustained intravenous infusion regimen. Am. J. Cardio!., SO:716-721, 1982.

25. Eichelbaum, M., and Echizen, H. Clinical pharmacology of calcium antagonists: a critical review. J. Cardiovasc. Pharmacol., 6: S963-S967, 1984.

26. Kann, J., Krol, G. J., Raemsch, K. D., Burkholder, D. E., and Levitt, M. J.Bioequivalences and metabolism of nitrendipine administered orally tohealthy volunteers. J. Cardiovasc. Pharmacol., 6: S968-S973, 1984.

27. Piepho. R. W. Individualizaron of calcium entry-blocker dosage for systemichypertension. Am. J. Cardiol.. 56: 105H-111H, 1985.

28. Surges, R. A., and Dodd. M. G. Amlodipine. Cardiovasc. Drug Rev., 8: 25-44, 1990.

29. Nayler, W. G. The effect of amlodipine on hypertension induced cardiachypertrophy and reperfusion induced calcium overload. J. Cardiovasc. Pharmacol., 12: 541-544. 1988.

30. Pang, D. C., and Sperelakis, N. Nifedipine. diltiazem, bepridil and verapamiluptakes into cardiac and smooth muscles. Eur. J. Pharmacol., H": 199-207,

1983.

2418

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

1992;52:2413-2418. Cancer Res   J. M. Taylor and R. U. Simpson  Antagonists in the Athymic MouseInhibition of Cancer Cell Growth by Calcium Channel

  Updated version

  http://cancerres.aacrjournals.org/content/52/9/2413

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/52/9/2413To request permission to re-use all or part of this article, use this link

Research. on January 27, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from