- Email: [email protected]
Verapamil and cyclosporin A modulate doxorubicin toxicity by distinct mechanisms
Cancer Letters, 57 (1991) 209-218 Elsevier Scientific Publishers Ireland
209 Ltd.
Verapamil and cyclosporin distinct mechanisms Y. Shoji”, ‘Department
A modulate
M.H. Fishera, A. Periasamyb, of Pharmacology
Uniuersity
of North
Carolina,
(Received (Accepted
17 January 1991) 15 February 1991)
and *Department Chapel
Hill,
NC
B. Hermanb
doxorubicin
and R.L. Julianoa
of Cell Biology and Anatomy, 27599
toxicity by
Lineberger
Cancer
Research
Center,
(U.S.A.)
Summary
obserued without changes in net accumulation of doxorubicin or intracellular distribution, and
Cyclosporin A has been reported to enhance the sensitivity of cells displaying multi-
without enhanced doxorubicin induced DNA breakage. These results suggest that cyclosporin A can modulate doxorubicin cytotoxicity by means other than interference with the P-glycoprotein drug efflux system.
ple drug resistance to anthracyclines. However, the mechanism of action of cyclosporin A in modulating drug resistance is still controoersial. This study compares the effects of cyclosporin A and verapamil on doxorubicin resistance in Chinese hamster ovary cells (CHRC5) using several criteria including in vitro cytotoxicity, drug accumulation, intracellular distribution by video microscopy, and nuclear DNA damage. Our results demonstrate that uerapamil modulation of doxorubicin resistance was paralleled by cellular accumulation of doxorubicin, altered intracellular distribution of doxorubicin from cytoplasm to nucleus, and an increase in the
formation of doxorubicin related DNA strand breaks. In contrast, the modulating effect of cyclosporin was qualitatively different. High concen&aHons of cyclosporin (5 pg/ml) increased doxorubicin accumulation and caused partial redistribution to the nucleus. However, with low concentrations of cyclosporin (1 pg/ml) increased doxorubicin sensitivity was
Correspondence to: R.L. Juliano, Department of Pharmacology, CB#7365, University of North Carolina, Chapel Hill, NC 27599,
U.S.A.
0304-3835/91/$03.50 Published and Printed
0 1991 Elsevier Scientific in Ireland
Publishers
Keywords: cyclosporin distribution
multi-drug resistance; doxorubicin; A;
verapamil; subcellular
Introduction Resistance to anti-tumor drugs is a major limitation to effective cancer chemotherapy. One of the most widely studied manifestations of this problem is the multi-drug resistant (MDR) phenotype which involves gene amplification and over-expression of a 170 kD cell surface glycoprotein (P170) which can act as a drug efflux pump [l-3]. In cells expressing MDR, passive influx of lipophilic cationic anti-tumor drugs such as anthracyclines or vinca alkaloids is followed by an energy dependent drug efflux mediated via the P170 transport ATPase; this reduces cellular drug accumulation and confers a drug resistant state, A variety of pharmacological agents, not anti-tumor drugs in their own right, are known Ireland Ltd
210
to modulate the MDR phenotype and to increase cellular sensitivity to the anthracyclines and certain other anti-tumor agents. Thus verapamil and related compounds [4], [6], phenothiazines [5], local anesthetics lysosomatropic agents [7] and cyclosporins [8--101 have all been reported to enhance the sensitivity of MDR cells to anthracyclines. Perhaps the best studied case is that of verapamil. This compound seems to interact directly with the P170 drug efflux transporter and to inhibit its function [l]. This leads to inaccumulation of anthracyclines, creased altered intracellular distribution, and increases in the formation of anthracycline related DNA strand breaks [ 11,121. In contrast to the rather clear understanding of effects of verapamil on the MDR phenotype, our picture of the effects of cyclosporins is It has been shown that rather confusing. cyclosporin A as well as non-immunosuppressive cyclosporin analogs can modulate drug in a number of cell lines resistance [8-10,13-171. It is also clear that cyclosporin can bind to the P170 drug efflux pump [18]. However, it is not clear that interaction with the P170 is the sole means by which cyclosporins can modulate drug resistance. Thus studies with lymphoma cells and with Ehrlich ascites cells have failed to correlate cyclosporin modulation of doxorubicin resistance with increases in cellular doxorubicin acwhile in other studies cumulation [17,19], such a correlation was observed [ZO].It is particularly interesting to note the divergent results even when the same cell type (CHO cells) w.as used, with one group indicating a strong effect of cyclosporin A on daunorubicin accumulation [ 141 and another group failing to find such an effect on doxorubicin accumulation [15]. In the studies reported below we have used several techniques (cytotoxicity tests, drug uptake studies, analysis of intracellular drug distribution using digital imaging, and measurement of DNA strand breakage) to explore the relationship between cyclosporin A effects on the cellular accumulation and intracellular distribution of doxorubicin, and the
ability of cyclosporin A to modulate doxorubicin resistance. We compared the effects of cyclosporin A to the better known actions of verapamil. We find that while high concentrations of cyclosporin A can increase uptake of doxorubicin by MDR cells, lower concentrations of cyclosporin seem to modulate doxorubicin resistance by a mechanism which does not involve alterations in drug accumulation or intracellular distribution. Materials
and Methods
Drugs
Adriamycin (ADR) (doxorubicin hydrochloride 50 mg, Adria Lab., Columbus, OH), was dissolved in 0.9% NaCl solution as a stock solution at a concentration of 5 mM and stored at - 20°C until use. Verapamil hydrochloride injection solution (2.5 mg/ml, Abbott Lab., North Chicago, IL) was used as a stock solution. Cyclosporin A (Sandoz Research Institute, Hanover, NJ) was prepared at 10 mg/ml in ethanol. Each drug was diluted to the appropriate concentration with saline or with a-MEM containing 10% FCS. Cell lines The Chinese hamster ovary (CHO) cell lines used in this study were the adenosine, thymidine, and glycine requiring auxotroph parental line AUXBl and the multiple drug resistant mutant CHRC5. These cell lines were kindly provided by Dr. V. Ling (Ontario Cancer Institute, Toronto, Ontario, Canada). These cell lines were routinely maintained by serial culture in o-MEM supplemented with 10% FCS and 1% antibiotics at 37°C in a 5% CO2 atmosphere. Both cell lines were found to be free of Mycoplasma. Under these conditions cells were exponentially growing with doubling times of 16 h for AUXBl and 18 h for CHRC5. Cytotoxicity
studies
Cytotoxicity studies were performed as described by Carmichael et al. [21]. Briefly, 2 ml of cells were plated (at 4 x lo3 cells/ml) in 24-well tissue culture plates (Corning).
211
Following overnight incubation, cells were treated by adding 20 ~1 of drug solutions to 2 ml of medium for 3 h. Thereafter the wells were washed twice with prewarmed PBS, followed by the addition of 2 ml of drug-free medium; the cells were then incubated for 3 more days. The cell surviving fraction was determined by the thiazolyl blue dye assay, measuring absorbance at 540 nm with an Automated Microplate Reader (Biotek Instruments, Inc., Winoski, VT). All experiments were carried out in quadruplicate. The effects of modulators on resistance to doxorubicin was evaluated by treating cells with graded concentrations of doxorubicin (0- 1 PM for AUXB 1, O-50 PM for CHRC5) and 6.6 PM verapamil or 1.0 or 5.0 pg/ml cyclosporin A concomitantly. Alternatively modulation was analyzed by determining of the effect of a fixed concentration of doxorubicin (1.0 PM for CHRC5) in the presence of variable amounts of verapamil or cyclosporin. Drug accumulation Two milliliters of cell suspensions at a concentration of 1 x lo6 cells/ml were incubated in multi well plates overnight in a-MEM containing 10% FCS at 37°C in a 5% CO2 atmosphere. Cells were treated with 10 PM doxorubicin, in the presence or absence of verapamil or cyclosporin A for 1 h in complete medium. The reaction was terminated by the addition of chilled PBS at 4°C. The cells were then washed in PBS at 4”C, detached with a rubber policeman, and harvested by centrifugation in PBS. The cell-associated doxorubicin was extracted in 0.3 N HCI in 50% ethanol and the fluorescence intensity was determined spectrofluorometrically using 470 nm as the excitation and 590 nm as the emission wavelengths. Cell-associated doxorubicin was expressed as nmoles/106 cells and was derived from standard curves prepared with doxorubicin alone [22]. Cell viabilities were confirmed by trypan blue staining. Cell
culture
for
Cells (AUXBl
digital
uideo
imaging
and CHRC5)
were cultured
for experiments concerning the subcellular distribution of doxorubicin on coverslips in (YMEM containing 10% FCS (2 x lo5 cells/2 ml/well/coverslip for overnight, 1 x lo5 cells/2 ml/well/coverslip for 2 nights culture). Thereafter 100 ~1 of doxorubicin or 100 ~1 of modulator solution were added to each well at appropriate concentrations. Drug treatment was carried out at 37°C in 5% CO2 for 1 h. The modulators did not change the external pH of the medium. AUXBl were treated with 10 PM doxorubicin and CHRC5 were treated with 50 FM doxorubicin to compensate for the differences in total drug accumulated by these cell types. These concentrations gave the same total fluorescence in both cell lines (data not shown). After finishing the drug treatment, coverslips were washed with buffer containing 10 mM HEPES (4-(2-hydroxyethyl)-l-piperazine ethane sulfonic acid), 140 mM NaCI, 5 mM KCl, 1 mM CaC&, 1 mM M&I, and 10 mM glucose (pH 7.4). Coverslips were mounted in special chamber designed for the During digital video imaging system. fluorescence microscopy, HEPES buffer was used to prevent shifts in external pH which could change the doxorubicin distribution pattern. video imaging system Experiments on the subcellular distribution of doxorubicin were performed using the MDVM (multiparameter digitized video microscopy) imaging system described elsewhere [23,24]. The MDVM system basically consists of an epi-fluorescence microscope (IM-35; Carl Zeiss, Inc., Thornwood, and neutral density filter NY) > excitation wheels, video camera and a digitizing board interfaced to the computer. The excitation wavelength was provided by a 100-W mercury vapor lamp and passed through computer controlled interference and neutral density filter wheels. A band pass filter of 515-560 nm, a long pass filter of 590 nm, and a chromatic beam splitter of 580 nm were used to visualize the subcellular distribution of doxorubicin. The microscope stage was maintainDigital
212
ed at 37°C by an air curtain incubator. The fluorescent image of the cell was projected by a rifle-telescope attached to the microscope on the face plate of the ISIT (intensified siliconintensified target) camera (model 66; MTIDage, Michigan city, IN). Then the image was digitized by the imaging board (512 x 512 pixel, 8 bit, model IP512, Imaging Technology, MA) interfaced to the PDP 11/23 Computer (Digital Equipment Corp., Maynard, MA). A 63 x oil immersion plan-Neofluar objective (NA = 1.25) was used to image the fluorescent image and 1.5 x was used at the rifle telescope. During the experiment, the live cell fluorescence was averaged, background subtracted and stored to the disk as a 256 x 256 pixel image. This stored image was used to analyze the data. Both fluorescence and phase contrast images were recorded, with the phase contrast image used to confirm the location of the nucleus. Alkaline elution Doxorubicin induced single stranded breaks were estimated by the alkaline elution technique as described by Kohn [25], with the following specifics. Drug treated cells were radiolabeled with 0.02 pCi/ml [2-‘4C]thymidine (spec. act., 53 mCi/mmol; Amersham Corporation, Arlington Heights, IL), and the control cells with 0.1 &i/ml [methyl-3H]thymidine (specific activity, 20 Ci/mmol; NEN Research, Boston, MA) for 24 h prior to the addition of drug(s). Aliquots of 5 x lo5 ceils were applied to 2.0 pm polyvinyl chloride filters (Omega Specialty Instrument Co., Chelmsford, MA) at 4°C. The lysis solution consisted of 3 ml 2% sodium dodecyl sulfate/O.025 M EDTA in HsO. One milliliter of a 0.5 mg/ml solution of Proteinase K (Boehringer-Mannheim, Indianapolis, IN) was used to eliminate DNA-protein crosslinks, and alkaline elution was in TPAH (tetrapropylammonium hydroxide), pH 12.1 [25]. Scintillation counting in ScintiVerse BD (Fisher Scientific) was performed using a BetaTrac 6895 liquid scintillation counter. The results at each drug concentration represents data from
four independent experiments. CHRC5 cells were exposed to 1.0 PM doxorubicin in the presence or absence of cyclosporin A or to 100 PM doxorubicin . Drug exposures were for 3 h. Results Sensitization of resistant cells to doxorubicin The modulating effect of verapamil or of cyclosporin A on the in vitro cytotoxicity of doxorubicin toward parental AUXBl cells or resistant mutant CHRC5 cells was studied. Figure 1 shows the results typically obtained when cells were exposed for 3 h to various concentrations of doxorubicin in the presence of fixed concentrations of verapamil or cyclosporin A; Table I presents a summary of several experiments. The ICss (dose giving 50% growth inhibition) of doxorubicin alone was 448 nM for AUXBl and 28040 nM for CHRC5. Verapamil as well as cyclosporin A enhanced the cytotoxicity of doxorubicin in both cell lines. In AUXBl, verapamil (6.6 PM)
100
00
60
40
20
o.Ol
.l
10
1
Doxorublcin
1
0
(pm)
on doxorubicinFig. 1. The effects of modulators induced growth inhibition in CHRC5. Cells were exposed to various concentrations of doxorubicin in the presence of 6.6 FM verapamil or 1.0 rg/ml cyclosporin A for 3 h. After the drug treatment, surviving fractions were determined by thiazolyl blue dye assay. (- o -), doxorubicin alone; (- l -), doxorubicin + 6.6pM verapamil; (- A -), doxorubicin + 1.0 pg/ml cyclosporin A. Bars were S.D.
213
Table
1.
Doxorubicin-induced
growth inhibition:
effects of modulators
in
CHRC5 and AUXBl
Drug
ADR ADR + 6.6 pM verapamil ADR + 1.0 pg/ml cyclosporin
I&
A
cells
(r-M)
AUXBl
CHRC5
448 123 127
28040” 820 1759
32310b 800 2230
The results of (a) and (b) were derived from separate experiments. Cells were treated with drugs for 3 h at 37°C. Surviving fractions were determined by thiazolyl blue dye assay. IC,, was calculated from surviving fractions. ADR, doxorubicin.
reduced the ICsO of doxorubicin (3.6-fold) as did cyclosporin A (1 pg/ml). In contrast, in the resistant CHRC5 cell line, verapamil (6.6 PM) greatly altered doxorubicin-induced cytotoxicity, reducing the IC& by 34-fold. In CHRC5, cyclosporin A decreased the I&,, of doxorubicin by approximately 16-fold. In another experiment, the effect of various concentrations of verapamil or cyclosporin A on a non-toxic concentration of doxorubicin was tested. In Fig. 2, it can be seen that 6.6 PM verapamil provided the maximal effect for this drug on doxorubicininduced cytotoxicity. Cyclosporin A at 1 pg/ml produced a substantial increase in doxorubicin toxicity but higher concentrations (2-5 pg/ml) were required for maximal effect. When tested alone neither verapamil nor cyclosporin A was inhibitory to CHRC5 cells. These studies confirmed the fact that both drugs were able to potentiate doxorubicin inin multiple drug resistant duced toxicity CHRC5 cells. slightly
Cellular accumulation of doxorubicin Cellular doxorubicin accumulation in both AUXBl and CHRC5 cell lines was determined fluorometrically. As shown in Fig. 3, accumulation of doxorubicin in AUXBl was 5 times greater than in CHRC5 in the absence of modulators. After 1 h exposure to various concentrations of verapamil, doxorubicin uptake increased in a dose dependent manner in both cell lines. However, the increase in doxorubicin uptake was much greater in the multi
iii, ii
~(+,
_
0-l
.
0
I
1
.
I
2 Cyclorporin
.
I
3 A
.
I
4
.
I
5
7
6
@g/ml)
1
-8 0
; Verapamll
1'0
? I!
(PM)
Fig. 2. Dose response effects of modulators and doxorubicin-induced cytotoxicity. CHRC5 cells exposed to 1.0 PM doxorubicin and different concentrations of verapamil or cyclosporin A for 3 h. Surviving fractions were determined by thiazolyl blue dye assay. Bars were S.D. ( -), modulator alone; ( + ), combination of doxorubicin and modulator.
1
0
Cyclosporln A (pglml)
(D
5 -0 i
0.4
E a 6 d 1 1" 0
0.2
0.6 0.0IL 0
6.6
1.65
Verapamll
(PM )
Fig. 3. Effect of modulators on cellular accumulation of doxorubicin. Cells were treated with 10 PM doxorubicin and modulators at different concentrations at 37°C for 1 h. Cellular uptake of doxorubicin was measured fluorospectrometrically and defined as nmol/106 cells. Bars, S.D. for three independent experiments. (II), AUXBl; (a), CHRC5.
drug resistant CHKC5 line than in AUXBl. Thus 1.65 PM verapamil produced an approximate doubling of uptake while 6.6 ,uM produced a 6-fold increase in doxorubicin uptake. The response to cyclosporin A was quite different; thus a low concentration of cyclosporin (1 pg/ml) had almost no effect on doxorubicin uptake in CHRC5 cells, but a slight effect in AUXBl cells, whereas 5 pg/ml of cyclosporin induced a major (7-fold) increase in doxorubicin uptake in CHRC5 cells. Intracellular distribution of doxorubicin Subcellular distribution patterns of
dox-
orubicin in AUXBl or CHRC5 cells in the presence or absence of modulators are shown in Fig. 4. AUXBl cells treated with 10 PM doxorubicin for 1 h showed obvious drug accumulation into the nucleus, staining of the boundary, and connuclear-cytoplasmic siderable cytoplasmic fluorescence (Fig. 4E). During 30 min in drug-free medium much of the nuclear fluorescence was lost and drug transfered to the cytoplasm and medium (not shown). In CHRC5 cells exposed to 50 PM doxorubicin, fluorescence was primarily confined to the cytoplasm with the nucleus appearing dark (Fig. 4A). The cytoplasmic fluorescence was not uniform but rather seemed to come from clusters of vesicles in the peripheral nuclear area; these structures may represent golgi apparatus and/or mitochondria [26]. Wh en CHRC5 cells were exposed to 6.6 PM for 1 h, there was a dramatic change in the doxorubicin fluorescence pattern with a strong shift to the nucleus (Fig. 4B). CHRC5 cells showed bright fluorescence in the nucleus, apparent staining of chromatin, and clear delineation of the nuclear envelop (arrow); the cytoplasm had very weak homogeneous fluorescence. Figures 4C and 4D show CHRC5 cells treated with low concentrations (1 pg/ml) and high concentration (5 pg/ml) of cyclosporin A, respectively. CHRC5 cells treated with doxorubicin plus low concentrations of cyclosporin A, which significantly increased doxorubicin-induced cytotoxicity, showed the same distribution pattern as cells treated with doxorubicin alone, that is, primarily perinuclear cytoplasmic fluorescence, while the nucleus remains dark. High concentrations of cyclosporin A partially shifted the doxorubicin distribution in CHRC5 from cytoplasm to nucleus. However, this pattern was not exactly the same as for cells treated with verapamil; in the cyclosporin treated cells considerable perinuclear cytoplasmic staining persisted and the nuclear envelop rather than chromatin was stained most intensely. A summary of the doxorubicin fluorescence patterns for various conditions is presented in Table II.
215
Changes of the distribution pattern of doxorubicin in CHRC5 cells treated with verapamil (B) or cyclosporin Fig. 4. A (C,D) CHRC5 cells were exposed for 1 h in complete medium containing 50 PM doxorubicin in presence or absence of modulators. A, doxorubicin (50 PM) alone; B, doxorubicin (50 PM) and 6.6 PM verapamil; C, doxorubicin (50 PM) and 1.0 mg/ml cyclosporin A; D, doxorubicin (50 PM) and 5.0 pg/ml cyclosporin A; E, AUXBl cells treated with doxorubicin (10 PM) alone. Digitized video imaging was performed as described in Materials and Methods. N, nucleus; P, perinuclear cytoplasmic region.
216 Table II.
Typical distribution
pattern
of doxorubicin
in AUXBl
and CHRC5 cells.
Drug
AUXB 1
CH’C5
ADR ADR + 6.6 PM verapamil ADR + 1.0 pg/ml cyclosporin ADR + 5.0 pg/ml cyclosporin
N-l, N-2, C-2 N-l, N-2, C-2 N-2, C-2 ND
c-2 N-l, c-2 N-l,
A A
C-l N-2, C-2
Distribution pattern: N-l, nucleus: N-2, nuclear membrane: C-l, homogeneous cytoplasm: C-2, heterogeneous cytoplasm; ND, not done; ADR, doxorubicin. Cells were treated with ADR alone or combination of ADR and verapamil (6.6 PM) or cyclosporin A (1 or 5 pg/ml) for 1 h. Digitized video microscopy was carried out as described in Materials and Methods.
Nuclear DNA damage As indicated in Fig. 5, treatment with combinations of 1 PM doxorubicin and cyclosporin A failed to increase single strand breaks in
CHRC5 cells while higher a concentration of doxorubicin (100 PM) did increase DNA Even high concentrations of breakage. cyclosporin (5 pg/ml) did not increase doxorubicin induced DNA damage, while cyclosporin itself had no effect on DNA damage. Parallel experiments with doxorubicin and verapamil (data not shown) confirmed previous results by others [12,13] that verapmil treatment markedly increases doxorubicin induced DNA strand breaks in doxorubicin resistant cells. Discussion
o
20
40
60
80
100
Fraction of PHIDNA eluted through the filter
Fis. 5. Single strand DNA breakage by doxorubicin in CHRC5 cells by alkaline elution analysis. CHRC5 cells were exposed to doxorubicin alone or combinations of doxorubicin and cyclosporin A (1 or 5 pg/ml) for 3 h. The alkaline elution assay was performed as described in Materials and Methods. (- l -), no drug; (- q -), doxorubicin 1 FM; (- n -), doxorubicin 100 pM; (- o -) doxorubicin 1 PM and cyclosporin A 1.0 pg/ml; (- A -), doxorubicin 1 PM and cyclosporin A 5 pg/ml.
Cells which express the P-glycoprotein based form of multidrug resistance show reduced accumulation and altered intracellular distribution of anthracyclines [11,27,28]. Our results with CHRC5 cells confirm these previous observations in the CHO cell system and also confirm that verapamil enhances anthracycline accumulation, promotes redistribution of anthracycline from cytoplasm to nucleus and increases anthracycline induced DNA breakage [ll, 121. Thus verapamil seems to potentiate nuclear accumulation and DNA damage in the drug resistant cells. Cyclosporin A has clearly been shown to be capable of interacting with P-glycoprotein [ 181. Nonetheless low concentrations of cyclosporin potentiate the cytotoxicity of doxorubicin without increasing doxorubicin uptake and without causing either a cytoplasm to nucleus re-distribution or an increase in doxorubicin in-
217
duced DNA damage (Figs. 3-5). Our results seem to parallel those reported by Chambers et al. [15], but not those of Silberman et al. [14]. These discrepancies may reflect qualitatively different mechanisms of cyclosporin action at low and high concentrations of the drug. Thus cyclosporin may modulate anthracycline resistance in two distinct ways. At sufficiently high concentrations (5 pg/ml) the drug may block the function of the P-glycoprotein efflux pump, leading to increased anthracycline accumulation in cells, and partial re-distribution to the nucleus. At lower concentrations, cyclosporin A seems to enhance anthracycline toxicity by a mechanism which is unrelated to increases in anthracycline uptake or redistribution. The nature of the modulating effect of low dose cyclosporin is unclear; certainly it is not associated with the drug’s immunosuppressive activity since non-immunosuppressive analogs of cyclosporin A also modulate anthracycline toxicity in CHO cells [15] and other cells [13,17]. Anthracyclines bind strongly to membranes [29] and part of their toxicity may be at the membrane level [30]. Cyclosporins are also highly lipophilic, membrane interactive drugs and may exert physical effects on membranes which serve to potentiate anthracycline induced membrane damage. Another interesting possibility relates to recent observations that ionophoric drugs which change membrane potentials and which interfere with the cell’s ability to regulate cytoplasmic pH are cytotoxic under certain conditions [31,32]. Cyclosporin has been reported to modify membrane potentials [33] and may thus alter intracellular ion gradients. Such changes may alter intracytoplasmic distribution and membrane actions of anthracyclines. In any case it is clear that cyclosporin A can modulate doxorubicin resistance by means other than inhibition of the P170 drug efflux pump.
References 1
5
6
7
8
9
10
11
thracycline retention and activity in multiple drug resistant cells. Mol. Pharmacol.. 34, 108-185. Slater, L.M., Sweet, P., Stupecky. M., Wetzel. M. W. and Gupta, S. (1986) Cyclosporin A corrects daunorubicin resistance in Ehrlich ascites carcinoma. Br. J Cancer, 54, 235-239. Coley, H.M., Twentyman, R.R. and Workman, P. (1989) Improved cellular accumulation is characteristic of anthracyclines which retain high activity in multidrug resistant cell lines. alone or combination with verapamil or cyclosporin A. Biochem. Pharmacol., 38. 4467-4475. Twentyman, P.R. (1988) Modification of cytotoxic drug resistance by non-immunosuppressive cyclosporins. Br. J. Cancer. 57, 254-258. Williamham. M.C.. Cornwell, M.M., Cardarelli. C.O.. Gottesman. M.M. and Pastan. I. (1986) Single cell analysis
12
of daunomycin uptake and efflux in multidrug-resistant and -sensitive KB cells: effects of verapamil and other drugs Cancer Res.. 46, 5941-5946. Yin. M.. Bankusli. I. and Rustum. Y.M. (1989)
13
Mechanisms of the in viva resistance to adriamycin and modulation by calcium channel blockers in mice. Cancer Res.. 49. 4729-4733. Hu, X.F., Martin, T. J.. Bell. D.R., de Luise. M. and
14
Acknowledgements This work was supported by NIH grants CA 47044 to R.L.J. and NIH A601218 to B. Herman.
Bradley, G., Juranka, P.F. and Ling, V. (1988) Mechanism of m&drug resistance. Biochim. Biophys. Acta. 948, 87-128. Pastan, I. and Gottesman. M. (1987) Multiple drug resistance in cancer. N. Engl. J. Med., 316. 1338-1393. Beck. W.T. (1987) The cell biology of multiple drug resistance. Biochem. Pharm., 36, 2879-2887. Tsuruo. T., Iida, H., Nojiri. M., Tsukagoshi. S. and Sakurai. Y. (1983) Circumvention of vincristine and Adriamycin resistance in vivo and in vitro by calcium influx blockers. Cancer Res., 43, 2905-2910. Ganapathi. R., Grahowski, D. and Schmidt, H. (1986) Factors governing the modulation of vinca alkaloid resistance inhibitor trifluoperazine. Biochem. Pharmacol., 35. 673-678. Tsuruo, T., Iida, H., Kitatani. Y.. Yokata. S.. Tsukagoshi, S. and Sakurai, Y. (1984) Effects of quinidine and related compounds on cytotoxicity and cellular accumulation of vincristine and Adriamycin in drug-resistant tumor cells. Cancer Res., 44, 4303-4307. Klohs, W.D. and Steinkampf. R.D. (1988) The effects of lysosomotropic agents and secretory inhibitors on an-
15
Zalcberg. J.R. (1990) Combination use of cyclosporin A and verapamil in modulating multidrug resistance in human leukemia cell lines. Cancer Res.. 50. 2953-2957. Silbermann. M.H.. Boersma. A.W.. Janssen, A.L.. Scheper, R.J., Herweijer, H. and Nooter. K. (1989) Effects of cyclosporin A and verapamil on the intracellular daunorubicin accumulation in Chinese hamster ovary cells with increasing levels of drug-resistance. Int. J Cancer. 44. 722-726. Chambers, S.K., Hait. W.N., Kacinski. B.M., Keyes, S.R. and Handschumacher. R.E. (1989) Enhancement of an-
218
thracycline
16
17
18
19
20
21
22 23
24
growth
inhibition
in parent
and
multidrug-
25
resistant Chinese hamster ovary cells by cyclosporin A and its analogues. Cancer Res., 49, 6275-6279. Twentyman, P.R., Reeve, J.G., Koch, G. and Wright, K.A. (1990) Chemosensitisation by P-glycoprotein and CP22(sorcin). Br. J. Cancer, 62, 89-95. Slater, L.M., Sweet, P., Stupecky, M., Wetzel, M.W. and Gupta, S. (1986) Cyclosporin A reverses vincristine and daunorubicin resistance in acute lymphatic leukemia in vitro. J. Clin. Invest., 77, 1405-1407. Goldberg, H., Ling, V., Wong, P.Y. and Skorecki, K. (1988) Reduced cyclosporin accumulation in multidrugresistant cells. Biochem. Biophys. Res. Commun., 152, 552-558.
26
27
28
Meador, J., Sweet, P., Stupecky, M., Wetzel, M., Murray, S., Gupta, S. and Slater, L. (1987) Enhancement of cyclosporin A of daunorubin efficacy in Ehrlich ascites carcinoma and murine hepatoma.129 Cancer Res., 47, 6216-6219. Nooter. K., Oostrum, R., Jonker, R., van Dekken, H.. Stokdijik, W. and van den Engh, G. (1989) Effect of cyclosporin A on daunorubicin accumulation in multidrug resistant P388 leukemia cells measured by real-time flow cytometry. Cancer Chemother. Pharmacol., 23(5), 296-300. Carmichael, J., DeGraff, W.G., Gazdar, A.F., Minna, J.D. and Mitchell, J.B. (1987) Evaluation of a tetrazolium based semiautomated calorimetric assay: assessment of chemosensitivity testing. Cancer Res., 47, 936-942. Daoud, S.S. and Huang, K.C. (1988) Renal handling of Adriamycin. Prog. Clin. Biol. Res., 258, 161-179. DeGuiseppi, J., Inmann. R.. Ishihara, A., Jacobson, K. and Herman, B. (1985) Application of digitized fluorescence microscopy to problems in cell biology. Bio. Tech., 3, 394-403. Lemasters, J.J., DiGuiseppi, J., Nieminen, A.L. and Herman, B. (1987) Blebbing. Free Ca2+ and mitochondria membrane potential preceding cell death in hepatocytes. Nature, 325. 78-81.
Kohn, K.W., Ewig, R.A.G., Erickson, L.C. and Zwelling, L.A. (1981) Measurement of strand breaks and cross-links by alkaline elution. In: DNA Repaier: A Labratory Manual of Research Procedures, pp. 379-401. Editors: E.C. Friedberg and P.C. Hanawalt, New York: Marcel Dekker. Pagano, R.E. and Slight, R.G. (1985) Defining lipid transport pathways in animal cells. Science, 229, 1051-1057. Hindenburg, A.A., Gervasoni, Jr. J.E., Krishna, S., Stewart, V.J., Rosado, M., Lutzky, J., Bhalla, K., Baker, M.A. and Taub, R.N. (1989) Intracellular distribution and pharmacokinetics of daunorubicin in anthracycline-sensitive and-resistant HL-60 cells. Cancer Res.. 49, 4607-4614. Keizer, H.G., Schuurhuis, G.J., Broxterman, H.J., Lankelma, J., Schooner-r, W.G.E. J., van Rijn, F., Pinedo, H.M. and Joenje, H. (1989) Correlation of multidrug resiswith decreased drug accumulation, tance altered subcellular drug distribution, and increased P-glyoprotein
29
30 31
32
33
expression in cultured W-1573 human lung tumor cells. Cancer Res., 49, 2988-2993. Burke, T.G., Thomas, G. and Tritton, T.R. (1985) Location and dynamics of anthracyclines bound to unilamellar vesicles. Biochemistry, 24, phosphatidylcholine 5972-5980. Pratt, W. and Rudden. R.W. (1979) The anticancer drug. Oxford Univ. Press New York: Oxford. Newell, K.J. and Tannock. I.F. (1989) Reduction of intracellular pH as a possible mechanism for killing cells in acidic regions of solid tumors: effects of carbonylcyanide3chlorophenylhydrazone. Cancer Res., 49, 4477-4482. Rotin, D., Steele-Norwood, D., Grinstein, S. and Tanneck, I. (1989) Requirement of the Na+/H + exchanger for tumor growth. Cancer Res., 49, 205-211. Vayuvegula, B.. Slater, L., Meador, J. and Gupta, S. (1988) Correlation of altered plasm membrane potentials, A possible mechanism of cyclosporin A and verapamil reversal of pleiotropic drug resistance in neoplasia. Cancer Chemother. Pharmacol., 22, 163-168.
209 Ltd.
Verapamil and cyclosporin distinct mechanisms Y. Shoji”, ‘Department
A modulate
M.H. Fishera, A. Periasamyb, of Pharmacology
Uniuersity
of North
Carolina,
(Received (Accepted
17 January 1991) 15 February 1991)
and *Department Chapel
Hill,
NC
B. Hermanb
doxorubicin
and R.L. Julianoa
of Cell Biology and Anatomy, 27599
toxicity by
Lineberger
Cancer
Research
Center,
(U.S.A.)
Summary
obserued without changes in net accumulation of doxorubicin or intracellular distribution, and
Cyclosporin A has been reported to enhance the sensitivity of cells displaying multi-
without enhanced doxorubicin induced DNA breakage. These results suggest that cyclosporin A can modulate doxorubicin cytotoxicity by means other than interference with the P-glycoprotein drug efflux system.
ple drug resistance to anthracyclines. However, the mechanism of action of cyclosporin A in modulating drug resistance is still controoersial. This study compares the effects of cyclosporin A and verapamil on doxorubicin resistance in Chinese hamster ovary cells (CHRC5) using several criteria including in vitro cytotoxicity, drug accumulation, intracellular distribution by video microscopy, and nuclear DNA damage. Our results demonstrate that uerapamil modulation of doxorubicin resistance was paralleled by cellular accumulation of doxorubicin, altered intracellular distribution of doxorubicin from cytoplasm to nucleus, and an increase in the
formation of doxorubicin related DNA strand breaks. In contrast, the modulating effect of cyclosporin was qualitatively different. High concen&aHons of cyclosporin (5 pg/ml) increased doxorubicin accumulation and caused partial redistribution to the nucleus. However, with low concentrations of cyclosporin (1 pg/ml) increased doxorubicin sensitivity was
Correspondence to: R.L. Juliano, Department of Pharmacology, CB#7365, University of North Carolina, Chapel Hill, NC 27599,
U.S.A.
0304-3835/91/$03.50 Published and Printed
0 1991 Elsevier Scientific in Ireland
Publishers
Keywords: cyclosporin distribution
multi-drug resistance; doxorubicin; A;
verapamil; subcellular
Introduction Resistance to anti-tumor drugs is a major limitation to effective cancer chemotherapy. One of the most widely studied manifestations of this problem is the multi-drug resistant (MDR) phenotype which involves gene amplification and over-expression of a 170 kD cell surface glycoprotein (P170) which can act as a drug efflux pump [l-3]. In cells expressing MDR, passive influx of lipophilic cationic anti-tumor drugs such as anthracyclines or vinca alkaloids is followed by an energy dependent drug efflux mediated via the P170 transport ATPase; this reduces cellular drug accumulation and confers a drug resistant state, A variety of pharmacological agents, not anti-tumor drugs in their own right, are known Ireland Ltd
210
to modulate the MDR phenotype and to increase cellular sensitivity to the anthracyclines and certain other anti-tumor agents. Thus verapamil and related compounds [4], [6], phenothiazines [5], local anesthetics lysosomatropic agents [7] and cyclosporins [8--101 have all been reported to enhance the sensitivity of MDR cells to anthracyclines. Perhaps the best studied case is that of verapamil. This compound seems to interact directly with the P170 drug efflux transporter and to inhibit its function [l]. This leads to inaccumulation of anthracyclines, creased altered intracellular distribution, and increases in the formation of anthracycline related DNA strand breaks [ 11,121. In contrast to the rather clear understanding of effects of verapamil on the MDR phenotype, our picture of the effects of cyclosporins is It has been shown that rather confusing. cyclosporin A as well as non-immunosuppressive cyclosporin analogs can modulate drug in a number of cell lines resistance [8-10,13-171. It is also clear that cyclosporin can bind to the P170 drug efflux pump [18]. However, it is not clear that interaction with the P170 is the sole means by which cyclosporins can modulate drug resistance. Thus studies with lymphoma cells and with Ehrlich ascites cells have failed to correlate cyclosporin modulation of doxorubicin resistance with increases in cellular doxorubicin acwhile in other studies cumulation [17,19], such a correlation was observed [ZO].It is particularly interesting to note the divergent results even when the same cell type (CHO cells) w.as used, with one group indicating a strong effect of cyclosporin A on daunorubicin accumulation [ 141 and another group failing to find such an effect on doxorubicin accumulation [15]. In the studies reported below we have used several techniques (cytotoxicity tests, drug uptake studies, analysis of intracellular drug distribution using digital imaging, and measurement of DNA strand breakage) to explore the relationship between cyclosporin A effects on the cellular accumulation and intracellular distribution of doxorubicin, and the
ability of cyclosporin A to modulate doxorubicin resistance. We compared the effects of cyclosporin A to the better known actions of verapamil. We find that while high concentrations of cyclosporin A can increase uptake of doxorubicin by MDR cells, lower concentrations of cyclosporin seem to modulate doxorubicin resistance by a mechanism which does not involve alterations in drug accumulation or intracellular distribution. Materials
and Methods
Drugs
Adriamycin (ADR) (doxorubicin hydrochloride 50 mg, Adria Lab., Columbus, OH), was dissolved in 0.9% NaCl solution as a stock solution at a concentration of 5 mM and stored at - 20°C until use. Verapamil hydrochloride injection solution (2.5 mg/ml, Abbott Lab., North Chicago, IL) was used as a stock solution. Cyclosporin A (Sandoz Research Institute, Hanover, NJ) was prepared at 10 mg/ml in ethanol. Each drug was diluted to the appropriate concentration with saline or with a-MEM containing 10% FCS. Cell lines The Chinese hamster ovary (CHO) cell lines used in this study were the adenosine, thymidine, and glycine requiring auxotroph parental line AUXBl and the multiple drug resistant mutant CHRC5. These cell lines were kindly provided by Dr. V. Ling (Ontario Cancer Institute, Toronto, Ontario, Canada). These cell lines were routinely maintained by serial culture in o-MEM supplemented with 10% FCS and 1% antibiotics at 37°C in a 5% CO2 atmosphere. Both cell lines were found to be free of Mycoplasma. Under these conditions cells were exponentially growing with doubling times of 16 h for AUXBl and 18 h for CHRC5. Cytotoxicity
studies
Cytotoxicity studies were performed as described by Carmichael et al. [21]. Briefly, 2 ml of cells were plated (at 4 x lo3 cells/ml) in 24-well tissue culture plates (Corning).
211
Following overnight incubation, cells were treated by adding 20 ~1 of drug solutions to 2 ml of medium for 3 h. Thereafter the wells were washed twice with prewarmed PBS, followed by the addition of 2 ml of drug-free medium; the cells were then incubated for 3 more days. The cell surviving fraction was determined by the thiazolyl blue dye assay, measuring absorbance at 540 nm with an Automated Microplate Reader (Biotek Instruments, Inc., Winoski, VT). All experiments were carried out in quadruplicate. The effects of modulators on resistance to doxorubicin was evaluated by treating cells with graded concentrations of doxorubicin (0- 1 PM for AUXB 1, O-50 PM for CHRC5) and 6.6 PM verapamil or 1.0 or 5.0 pg/ml cyclosporin A concomitantly. Alternatively modulation was analyzed by determining of the effect of a fixed concentration of doxorubicin (1.0 PM for CHRC5) in the presence of variable amounts of verapamil or cyclosporin. Drug accumulation Two milliliters of cell suspensions at a concentration of 1 x lo6 cells/ml were incubated in multi well plates overnight in a-MEM containing 10% FCS at 37°C in a 5% CO2 atmosphere. Cells were treated with 10 PM doxorubicin, in the presence or absence of verapamil or cyclosporin A for 1 h in complete medium. The reaction was terminated by the addition of chilled PBS at 4°C. The cells were then washed in PBS at 4”C, detached with a rubber policeman, and harvested by centrifugation in PBS. The cell-associated doxorubicin was extracted in 0.3 N HCI in 50% ethanol and the fluorescence intensity was determined spectrofluorometrically using 470 nm as the excitation and 590 nm as the emission wavelengths. Cell-associated doxorubicin was expressed as nmoles/106 cells and was derived from standard curves prepared with doxorubicin alone [22]. Cell viabilities were confirmed by trypan blue staining. Cell
culture
for
Cells (AUXBl
digital
uideo
imaging
and CHRC5)
were cultured
for experiments concerning the subcellular distribution of doxorubicin on coverslips in (YMEM containing 10% FCS (2 x lo5 cells/2 ml/well/coverslip for overnight, 1 x lo5 cells/2 ml/well/coverslip for 2 nights culture). Thereafter 100 ~1 of doxorubicin or 100 ~1 of modulator solution were added to each well at appropriate concentrations. Drug treatment was carried out at 37°C in 5% CO2 for 1 h. The modulators did not change the external pH of the medium. AUXBl were treated with 10 PM doxorubicin and CHRC5 were treated with 50 FM doxorubicin to compensate for the differences in total drug accumulated by these cell types. These concentrations gave the same total fluorescence in both cell lines (data not shown). After finishing the drug treatment, coverslips were washed with buffer containing 10 mM HEPES (4-(2-hydroxyethyl)-l-piperazine ethane sulfonic acid), 140 mM NaCI, 5 mM KCl, 1 mM CaC&, 1 mM M&I, and 10 mM glucose (pH 7.4). Coverslips were mounted in special chamber designed for the During digital video imaging system. fluorescence microscopy, HEPES buffer was used to prevent shifts in external pH which could change the doxorubicin distribution pattern. video imaging system Experiments on the subcellular distribution of doxorubicin were performed using the MDVM (multiparameter digitized video microscopy) imaging system described elsewhere [23,24]. The MDVM system basically consists of an epi-fluorescence microscope (IM-35; Carl Zeiss, Inc., Thornwood, and neutral density filter NY) > excitation wheels, video camera and a digitizing board interfaced to the computer. The excitation wavelength was provided by a 100-W mercury vapor lamp and passed through computer controlled interference and neutral density filter wheels. A band pass filter of 515-560 nm, a long pass filter of 590 nm, and a chromatic beam splitter of 580 nm were used to visualize the subcellular distribution of doxorubicin. The microscope stage was maintainDigital
212
ed at 37°C by an air curtain incubator. The fluorescent image of the cell was projected by a rifle-telescope attached to the microscope on the face plate of the ISIT (intensified siliconintensified target) camera (model 66; MTIDage, Michigan city, IN). Then the image was digitized by the imaging board (512 x 512 pixel, 8 bit, model IP512, Imaging Technology, MA) interfaced to the PDP 11/23 Computer (Digital Equipment Corp., Maynard, MA). A 63 x oil immersion plan-Neofluar objective (NA = 1.25) was used to image the fluorescent image and 1.5 x was used at the rifle telescope. During the experiment, the live cell fluorescence was averaged, background subtracted and stored to the disk as a 256 x 256 pixel image. This stored image was used to analyze the data. Both fluorescence and phase contrast images were recorded, with the phase contrast image used to confirm the location of the nucleus. Alkaline elution Doxorubicin induced single stranded breaks were estimated by the alkaline elution technique as described by Kohn [25], with the following specifics. Drug treated cells were radiolabeled with 0.02 pCi/ml [2-‘4C]thymidine (spec. act., 53 mCi/mmol; Amersham Corporation, Arlington Heights, IL), and the control cells with 0.1 &i/ml [methyl-3H]thymidine (specific activity, 20 Ci/mmol; NEN Research, Boston, MA) for 24 h prior to the addition of drug(s). Aliquots of 5 x lo5 ceils were applied to 2.0 pm polyvinyl chloride filters (Omega Specialty Instrument Co., Chelmsford, MA) at 4°C. The lysis solution consisted of 3 ml 2% sodium dodecyl sulfate/O.025 M EDTA in HsO. One milliliter of a 0.5 mg/ml solution of Proteinase K (Boehringer-Mannheim, Indianapolis, IN) was used to eliminate DNA-protein crosslinks, and alkaline elution was in TPAH (tetrapropylammonium hydroxide), pH 12.1 [25]. Scintillation counting in ScintiVerse BD (Fisher Scientific) was performed using a BetaTrac 6895 liquid scintillation counter. The results at each drug concentration represents data from
four independent experiments. CHRC5 cells were exposed to 1.0 PM doxorubicin in the presence or absence of cyclosporin A or to 100 PM doxorubicin . Drug exposures were for 3 h. Results Sensitization of resistant cells to doxorubicin The modulating effect of verapamil or of cyclosporin A on the in vitro cytotoxicity of doxorubicin toward parental AUXBl cells or resistant mutant CHRC5 cells was studied. Figure 1 shows the results typically obtained when cells were exposed for 3 h to various concentrations of doxorubicin in the presence of fixed concentrations of verapamil or cyclosporin A; Table I presents a summary of several experiments. The ICss (dose giving 50% growth inhibition) of doxorubicin alone was 448 nM for AUXBl and 28040 nM for CHRC5. Verapamil as well as cyclosporin A enhanced the cytotoxicity of doxorubicin in both cell lines. In AUXBl, verapamil (6.6 PM)
100
00
60
40
20
o.Ol
.l
10
1
Doxorublcin
1
0
(pm)
on doxorubicinFig. 1. The effects of modulators induced growth inhibition in CHRC5. Cells were exposed to various concentrations of doxorubicin in the presence of 6.6 FM verapamil or 1.0 rg/ml cyclosporin A for 3 h. After the drug treatment, surviving fractions were determined by thiazolyl blue dye assay. (- o -), doxorubicin alone; (- l -), doxorubicin + 6.6pM verapamil; (- A -), doxorubicin + 1.0 pg/ml cyclosporin A. Bars were S.D.
213
Table
1.
Doxorubicin-induced
growth inhibition:
effects of modulators
in
CHRC5 and AUXBl
Drug
ADR ADR + 6.6 pM verapamil ADR + 1.0 pg/ml cyclosporin
I&
A
cells
(r-M)
AUXBl
CHRC5
448 123 127
28040” 820 1759
32310b 800 2230
The results of (a) and (b) were derived from separate experiments. Cells were treated with drugs for 3 h at 37°C. Surviving fractions were determined by thiazolyl blue dye assay. IC,, was calculated from surviving fractions. ADR, doxorubicin.
reduced the ICsO of doxorubicin (3.6-fold) as did cyclosporin A (1 pg/ml). In contrast, in the resistant CHRC5 cell line, verapamil (6.6 PM) greatly altered doxorubicin-induced cytotoxicity, reducing the IC& by 34-fold. In CHRC5, cyclosporin A decreased the I&,, of doxorubicin by approximately 16-fold. In another experiment, the effect of various concentrations of verapamil or cyclosporin A on a non-toxic concentration of doxorubicin was tested. In Fig. 2, it can be seen that 6.6 PM verapamil provided the maximal effect for this drug on doxorubicininduced cytotoxicity. Cyclosporin A at 1 pg/ml produced a substantial increase in doxorubicin toxicity but higher concentrations (2-5 pg/ml) were required for maximal effect. When tested alone neither verapamil nor cyclosporin A was inhibitory to CHRC5 cells. These studies confirmed the fact that both drugs were able to potentiate doxorubicin inin multiple drug resistant duced toxicity CHRC5 cells. slightly
Cellular accumulation of doxorubicin Cellular doxorubicin accumulation in both AUXBl and CHRC5 cell lines was determined fluorometrically. As shown in Fig. 3, accumulation of doxorubicin in AUXBl was 5 times greater than in CHRC5 in the absence of modulators. After 1 h exposure to various concentrations of verapamil, doxorubicin uptake increased in a dose dependent manner in both cell lines. However, the increase in doxorubicin uptake was much greater in the multi
iii, ii
~(+,
_
0-l
.
0
I
1
.
I
2 Cyclorporin
.
I
3 A
.
I
4
.
I
5
7
6
@g/ml)
1
-8 0
; Verapamll
1'0
? I!
(PM)
Fig. 2. Dose response effects of modulators and doxorubicin-induced cytotoxicity. CHRC5 cells exposed to 1.0 PM doxorubicin and different concentrations of verapamil or cyclosporin A for 3 h. Surviving fractions were determined by thiazolyl blue dye assay. Bars were S.D. ( -), modulator alone; ( + ), combination of doxorubicin and modulator.
1
0
Cyclosporln A (pglml)
(D
5 -0 i
0.4
E a 6 d 1 1" 0
0.2
0.6 0.0IL 0
6.6
1.65
Verapamll
(PM )
Fig. 3. Effect of modulators on cellular accumulation of doxorubicin. Cells were treated with 10 PM doxorubicin and modulators at different concentrations at 37°C for 1 h. Cellular uptake of doxorubicin was measured fluorospectrometrically and defined as nmol/106 cells. Bars, S.D. for three independent experiments. (II), AUXBl; (a), CHRC5.
drug resistant CHKC5 line than in AUXBl. Thus 1.65 PM verapamil produced an approximate doubling of uptake while 6.6 ,uM produced a 6-fold increase in doxorubicin uptake. The response to cyclosporin A was quite different; thus a low concentration of cyclosporin (1 pg/ml) had almost no effect on doxorubicin uptake in CHRC5 cells, but a slight effect in AUXBl cells, whereas 5 pg/ml of cyclosporin induced a major (7-fold) increase in doxorubicin uptake in CHRC5 cells. Intracellular distribution of doxorubicin Subcellular distribution patterns of
dox-
orubicin in AUXBl or CHRC5 cells in the presence or absence of modulators are shown in Fig. 4. AUXBl cells treated with 10 PM doxorubicin for 1 h showed obvious drug accumulation into the nucleus, staining of the boundary, and connuclear-cytoplasmic siderable cytoplasmic fluorescence (Fig. 4E). During 30 min in drug-free medium much of the nuclear fluorescence was lost and drug transfered to the cytoplasm and medium (not shown). In CHRC5 cells exposed to 50 PM doxorubicin, fluorescence was primarily confined to the cytoplasm with the nucleus appearing dark (Fig. 4A). The cytoplasmic fluorescence was not uniform but rather seemed to come from clusters of vesicles in the peripheral nuclear area; these structures may represent golgi apparatus and/or mitochondria [26]. Wh en CHRC5 cells were exposed to 6.6 PM for 1 h, there was a dramatic change in the doxorubicin fluorescence pattern with a strong shift to the nucleus (Fig. 4B). CHRC5 cells showed bright fluorescence in the nucleus, apparent staining of chromatin, and clear delineation of the nuclear envelop (arrow); the cytoplasm had very weak homogeneous fluorescence. Figures 4C and 4D show CHRC5 cells treated with low concentrations (1 pg/ml) and high concentration (5 pg/ml) of cyclosporin A, respectively. CHRC5 cells treated with doxorubicin plus low concentrations of cyclosporin A, which significantly increased doxorubicin-induced cytotoxicity, showed the same distribution pattern as cells treated with doxorubicin alone, that is, primarily perinuclear cytoplasmic fluorescence, while the nucleus remains dark. High concentrations of cyclosporin A partially shifted the doxorubicin distribution in CHRC5 from cytoplasm to nucleus. However, this pattern was not exactly the same as for cells treated with verapamil; in the cyclosporin treated cells considerable perinuclear cytoplasmic staining persisted and the nuclear envelop rather than chromatin was stained most intensely. A summary of the doxorubicin fluorescence patterns for various conditions is presented in Table II.
215
Changes of the distribution pattern of doxorubicin in CHRC5 cells treated with verapamil (B) or cyclosporin Fig. 4. A (C,D) CHRC5 cells were exposed for 1 h in complete medium containing 50 PM doxorubicin in presence or absence of modulators. A, doxorubicin (50 PM) alone; B, doxorubicin (50 PM) and 6.6 PM verapamil; C, doxorubicin (50 PM) and 1.0 mg/ml cyclosporin A; D, doxorubicin (50 PM) and 5.0 pg/ml cyclosporin A; E, AUXBl cells treated with doxorubicin (10 PM) alone. Digitized video imaging was performed as described in Materials and Methods. N, nucleus; P, perinuclear cytoplasmic region.
216 Table II.
Typical distribution
pattern
of doxorubicin
in AUXBl
and CHRC5 cells.
Drug
AUXB 1
CH’C5
ADR ADR + 6.6 PM verapamil ADR + 1.0 pg/ml cyclosporin ADR + 5.0 pg/ml cyclosporin
N-l, N-2, C-2 N-l, N-2, C-2 N-2, C-2 ND
c-2 N-l, c-2 N-l,
A A
C-l N-2, C-2
Distribution pattern: N-l, nucleus: N-2, nuclear membrane: C-l, homogeneous cytoplasm: C-2, heterogeneous cytoplasm; ND, not done; ADR, doxorubicin. Cells were treated with ADR alone or combination of ADR and verapamil (6.6 PM) or cyclosporin A (1 or 5 pg/ml) for 1 h. Digitized video microscopy was carried out as described in Materials and Methods.
Nuclear DNA damage As indicated in Fig. 5, treatment with combinations of 1 PM doxorubicin and cyclosporin A failed to increase single strand breaks in
CHRC5 cells while higher a concentration of doxorubicin (100 PM) did increase DNA Even high concentrations of breakage. cyclosporin (5 pg/ml) did not increase doxorubicin induced DNA damage, while cyclosporin itself had no effect on DNA damage. Parallel experiments with doxorubicin and verapamil (data not shown) confirmed previous results by others [12,13] that verapmil treatment markedly increases doxorubicin induced DNA strand breaks in doxorubicin resistant cells. Discussion
o
20
40
60
80
100
Fraction of PHIDNA eluted through the filter
Fis. 5. Single strand DNA breakage by doxorubicin in CHRC5 cells by alkaline elution analysis. CHRC5 cells were exposed to doxorubicin alone or combinations of doxorubicin and cyclosporin A (1 or 5 pg/ml) for 3 h. The alkaline elution assay was performed as described in Materials and Methods. (- l -), no drug; (- q -), doxorubicin 1 FM; (- n -), doxorubicin 100 pM; (- o -) doxorubicin 1 PM and cyclosporin A 1.0 pg/ml; (- A -), doxorubicin 1 PM and cyclosporin A 5 pg/ml.
Cells which express the P-glycoprotein based form of multidrug resistance show reduced accumulation and altered intracellular distribution of anthracyclines [11,27,28]. Our results with CHRC5 cells confirm these previous observations in the CHO cell system and also confirm that verapamil enhances anthracycline accumulation, promotes redistribution of anthracycline from cytoplasm to nucleus and increases anthracycline induced DNA breakage [ll, 121. Thus verapamil seems to potentiate nuclear accumulation and DNA damage in the drug resistant cells. Cyclosporin A has clearly been shown to be capable of interacting with P-glycoprotein [ 181. Nonetheless low concentrations of cyclosporin potentiate the cytotoxicity of doxorubicin without increasing doxorubicin uptake and without causing either a cytoplasm to nucleus re-distribution or an increase in doxorubicin in-
217
duced DNA damage (Figs. 3-5). Our results seem to parallel those reported by Chambers et al. [15], but not those of Silberman et al. [14]. These discrepancies may reflect qualitatively different mechanisms of cyclosporin action at low and high concentrations of the drug. Thus cyclosporin may modulate anthracycline resistance in two distinct ways. At sufficiently high concentrations (5 pg/ml) the drug may block the function of the P-glycoprotein efflux pump, leading to increased anthracycline accumulation in cells, and partial re-distribution to the nucleus. At lower concentrations, cyclosporin A seems to enhance anthracycline toxicity by a mechanism which is unrelated to increases in anthracycline uptake or redistribution. The nature of the modulating effect of low dose cyclosporin is unclear; certainly it is not associated with the drug’s immunosuppressive activity since non-immunosuppressive analogs of cyclosporin A also modulate anthracycline toxicity in CHO cells [15] and other cells [13,17]. Anthracyclines bind strongly to membranes [29] and part of their toxicity may be at the membrane level [30]. Cyclosporins are also highly lipophilic, membrane interactive drugs and may exert physical effects on membranes which serve to potentiate anthracycline induced membrane damage. Another interesting possibility relates to recent observations that ionophoric drugs which change membrane potentials and which interfere with the cell’s ability to regulate cytoplasmic pH are cytotoxic under certain conditions [31,32]. Cyclosporin has been reported to modify membrane potentials [33] and may thus alter intracellular ion gradients. Such changes may alter intracytoplasmic distribution and membrane actions of anthracyclines. In any case it is clear that cyclosporin A can modulate doxorubicin resistance by means other than inhibition of the P170 drug efflux pump.
References 1
5
6
7
8
9
10
11
thracycline retention and activity in multiple drug resistant cells. Mol. Pharmacol.. 34, 108-185. Slater, L.M., Sweet, P., Stupecky. M., Wetzel. M. W. and Gupta, S. (1986) Cyclosporin A corrects daunorubicin resistance in Ehrlich ascites carcinoma. Br. J Cancer, 54, 235-239. Coley, H.M., Twentyman, R.R. and Workman, P. (1989) Improved cellular accumulation is characteristic of anthracyclines which retain high activity in multidrug resistant cell lines. alone or combination with verapamil or cyclosporin A. Biochem. Pharmacol., 38. 4467-4475. Twentyman, P.R. (1988) Modification of cytotoxic drug resistance by non-immunosuppressive cyclosporins. Br. J. Cancer. 57, 254-258. Williamham. M.C.. Cornwell, M.M., Cardarelli. C.O.. Gottesman. M.M. and Pastan. I. (1986) Single cell analysis
12
of daunomycin uptake and efflux in multidrug-resistant and -sensitive KB cells: effects of verapamil and other drugs Cancer Res.. 46, 5941-5946. Yin. M.. Bankusli. I. and Rustum. Y.M. (1989)
13
Mechanisms of the in viva resistance to adriamycin and modulation by calcium channel blockers in mice. Cancer Res.. 49. 4729-4733. Hu, X.F., Martin, T. J.. Bell. D.R., de Luise. M. and
14
Acknowledgements This work was supported by NIH grants CA 47044 to R.L.J. and NIH A601218 to B. Herman.
Bradley, G., Juranka, P.F. and Ling, V. (1988) Mechanism of m&drug resistance. Biochim. Biophys. Acta. 948, 87-128. Pastan, I. and Gottesman. M. (1987) Multiple drug resistance in cancer. N. Engl. J. Med., 316. 1338-1393. Beck. W.T. (1987) The cell biology of multiple drug resistance. Biochem. Pharm., 36, 2879-2887. Tsuruo. T., Iida, H., Nojiri. M., Tsukagoshi. S. and Sakurai. Y. (1983) Circumvention of vincristine and Adriamycin resistance in vivo and in vitro by calcium influx blockers. Cancer Res., 43, 2905-2910. Ganapathi. R., Grahowski, D. and Schmidt, H. (1986) Factors governing the modulation of vinca alkaloid resistance inhibitor trifluoperazine. Biochem. Pharmacol., 35. 673-678. Tsuruo, T., Iida, H., Kitatani. Y.. Yokata. S.. Tsukagoshi, S. and Sakurai, Y. (1984) Effects of quinidine and related compounds on cytotoxicity and cellular accumulation of vincristine and Adriamycin in drug-resistant tumor cells. Cancer Res., 44, 4303-4307. Klohs, W.D. and Steinkampf. R.D. (1988) The effects of lysosomotropic agents and secretory inhibitors on an-
15
Zalcberg. J.R. (1990) Combination use of cyclosporin A and verapamil in modulating multidrug resistance in human leukemia cell lines. Cancer Res.. 50. 2953-2957. Silbermann. M.H.. Boersma. A.W.. Janssen, A.L.. Scheper, R.J., Herweijer, H. and Nooter. K. (1989) Effects of cyclosporin A and verapamil on the intracellular daunorubicin accumulation in Chinese hamster ovary cells with increasing levels of drug-resistance. Int. J Cancer. 44. 722-726. Chambers, S.K., Hait. W.N., Kacinski. B.M., Keyes, S.R. and Handschumacher. R.E. (1989) Enhancement of an-
218
thracycline
16
17
18
19
20
21
22 23
24
growth
inhibition
in parent
and
multidrug-
25
resistant Chinese hamster ovary cells by cyclosporin A and its analogues. Cancer Res., 49, 6275-6279. Twentyman, P.R., Reeve, J.G., Koch, G. and Wright, K.A. (1990) Chemosensitisation by P-glycoprotein and CP22(sorcin). Br. J. Cancer, 62, 89-95. Slater, L.M., Sweet, P., Stupecky, M., Wetzel, M.W. and Gupta, S. (1986) Cyclosporin A reverses vincristine and daunorubicin resistance in acute lymphatic leukemia in vitro. J. Clin. Invest., 77, 1405-1407. Goldberg, H., Ling, V., Wong, P.Y. and Skorecki, K. (1988) Reduced cyclosporin accumulation in multidrugresistant cells. Biochem. Biophys. Res. Commun., 152, 552-558.
26
27
28
Meador, J., Sweet, P., Stupecky, M., Wetzel, M., Murray, S., Gupta, S. and Slater, L. (1987) Enhancement of cyclosporin A of daunorubin efficacy in Ehrlich ascites carcinoma and murine hepatoma.129 Cancer Res., 47, 6216-6219. Nooter. K., Oostrum, R., Jonker, R., van Dekken, H.. Stokdijik, W. and van den Engh, G. (1989) Effect of cyclosporin A on daunorubicin accumulation in multidrug resistant P388 leukemia cells measured by real-time flow cytometry. Cancer Chemother. Pharmacol., 23(5), 296-300. Carmichael, J., DeGraff, W.G., Gazdar, A.F., Minna, J.D. and Mitchell, J.B. (1987) Evaluation of a tetrazolium based semiautomated calorimetric assay: assessment of chemosensitivity testing. Cancer Res., 47, 936-942. Daoud, S.S. and Huang, K.C. (1988) Renal handling of Adriamycin. Prog. Clin. Biol. Res., 258, 161-179. DeGuiseppi, J., Inmann. R.. Ishihara, A., Jacobson, K. and Herman, B. (1985) Application of digitized fluorescence microscopy to problems in cell biology. Bio. Tech., 3, 394-403. Lemasters, J.J., DiGuiseppi, J., Nieminen, A.L. and Herman, B. (1987) Blebbing. Free Ca2+ and mitochondria membrane potential preceding cell death in hepatocytes. Nature, 325. 78-81.
Kohn, K.W., Ewig, R.A.G., Erickson, L.C. and Zwelling, L.A. (1981) Measurement of strand breaks and cross-links by alkaline elution. In: DNA Repaier: A Labratory Manual of Research Procedures, pp. 379-401. Editors: E.C. Friedberg and P.C. Hanawalt, New York: Marcel Dekker. Pagano, R.E. and Slight, R.G. (1985) Defining lipid transport pathways in animal cells. Science, 229, 1051-1057. Hindenburg, A.A., Gervasoni, Jr. J.E., Krishna, S., Stewart, V.J., Rosado, M., Lutzky, J., Bhalla, K., Baker, M.A. and Taub, R.N. (1989) Intracellular distribution and pharmacokinetics of daunorubicin in anthracycline-sensitive and-resistant HL-60 cells. Cancer Res.. 49, 4607-4614. Keizer, H.G., Schuurhuis, G.J., Broxterman, H.J., Lankelma, J., Schooner-r, W.G.E. J., van Rijn, F., Pinedo, H.M. and Joenje, H. (1989) Correlation of multidrug resiswith decreased drug accumulation, tance altered subcellular drug distribution, and increased P-glyoprotein
29
30 31
32
33
expression in cultured W-1573 human lung tumor cells. Cancer Res., 49, 2988-2993. Burke, T.G., Thomas, G. and Tritton, T.R. (1985) Location and dynamics of anthracyclines bound to unilamellar vesicles. Biochemistry, 24, phosphatidylcholine 5972-5980. Pratt, W. and Rudden. R.W. (1979) The anticancer drug. Oxford Univ. Press New York: Oxford. Newell, K.J. and Tannock. I.F. (1989) Reduction of intracellular pH as a possible mechanism for killing cells in acidic regions of solid tumors: effects of carbonylcyanide3chlorophenylhydrazone. Cancer Res., 49, 4477-4482. Rotin, D., Steele-Norwood, D., Grinstein, S. and Tanneck, I. (1989) Requirement of the Na+/H + exchanger for tumor growth. Cancer Res., 49, 205-211. Vayuvegula, B.. Slater, L., Meador, J. and Gupta, S. (1988) Correlation of altered plasm membrane potentials, A possible mechanism of cyclosporin A and verapamil reversal of pleiotropic drug resistance in neoplasia. Cancer Chemother. Pharmacol., 22, 163-168.