Uptake of the Noncytotoxic Transport Probe Procainamide in the Chinese Hamster Ovary Model of Multidrug Resistance1

(CANCER RESEARCH 52. 3539-3546. July 1. 1992] Uptake of the Noncytotoxic Transport Probe Procainamide in the Chinese Hamster Ovary Model of Multidrug Resistance1

K. V. Speeg, Jr.,2 Catherine deLeon, and William L. McGuiret

Department of Medicine, Divisions of Gastroenterologe/Nutrition [K. V. S., C. D.J and Oncology [W. L. M.], The University of Texas Health Science Center at San Antonio and Audie Murphy VA Hospital, San Antonio, Texas 78284

ABSTRACT to be a normal constituent of cells in organs known to secrete organic cations (i.e., renal proximal tubule, hepatocyte bile Many of the cytotoxic substrates of the multidrug transporter are canaliculus, and adrenal medulla) (7-9), we used noncytotoxic organic cations. Cimetidine, procainamide, and tetraethylammonium bro mide were used in a Chinese hamster ovary model of multidrug resistance, probes of organic cation secretion (10, 11), which is generally to study handling of noncytotoxic cationic transport probes. Cimetidine considered to be driven by electrochemical gradients rather and procainamide, but not tetraethylammonium, accumulated to a greater than by an ATPase (12), to determine whether these probes extent (5-fold) in the sensitive CHOAUXB1 (AB) cell line than in the might be substrates for gp-170 or whether this secretory mech resistant ( 11"( 5 (C5) cell line. Accumulation of both cimetidine and anism could be characterized in MDR cells. procainamide was significantly increased by verapamil in CS but not AB. Procainamide accumulation in both AB and CS was temperature depend ent and occurred by passive diffusion. Diltiazem, nifedipine, rifampin, MATERIALS AND METHODS tamoxifen, rhodamine, and ethidium also increased procainamide accu Materials. ['H]Cimetidine (specific activity, 20 Ci/mmol) was pur mulation in C5 but not AB. Azide in glucose-free medium increased chased from Amersham Corp. (Arlington Heights, IL); ['Hjprocaina- procainamide accumulation in CS, and this was reversed when glucose, mide (specific activity, 48 Ci/mmol) and ['H]TEA (specific activity, 16 but not 3-0-methylglucose, was added. Procainamide efflux rates were Ci/mmol) were custom synthesized by Amersham. Verapamil, dilti- similar in AB and CS and not affected by verapamil or azide. The initial azem, nifedipine, rifampin, tamoxifen, rhodamine, ethidium bromide, rate of procainamide uptake was higher in AB than in CS, and both procainamide, 3-0-methylglucose, and sodium azide were from Sigma verapamil and azide increased the initial rate of procainamide uptake in Chemical Co. (St. Louis, MO). GBSS and glucose-free GBSS were CS. Thus, differences in accumulation of the noncytotoxic transport probe prepared in 10-liter batches, filtered through 0.22-Mm filters, and stored procainamide in the colchicine-sensitive and colchicine-resistant compo in sterile bottles. Phosphate-buffered incubation medium contained (in nents of the Chinese hamster ovary cell line mimic the accumulation of HIM):sodium chloride, 139; dibasic potassium phosphate, 2.5; magne known cytotoxic substrates for the multidrug transporter, such as colchi- sium sulfate, 1.2; calcium chloride, 2.0; glucose, 5.5; L-alanine, 6.0; cine, vinblastine, and doxorubicin. The differential accumulation of pro sodium citrate, 1.0; and sodium lactate, 4.0. The standard pH was 7.4. cainamide is due to differences in rates of drug influx, rather than efflux. Cell Culture. The adenosine-, thymidine-, and glycine-requiring aux- Since procainamide influx is passive and decreased accumulation in the otroph AB and its colchicine-resistant mutant C5 were obtained from resistant line appears to parallel M, 170,000 glycoprotein presence and Dr. Victor Ling of the Ontario Cancer Institute (Toronto, Canada) activity, we would speculate that decreased procainamide accumulation (13). Cells were maintained at 37Â°Cin 5% CO2-95% air, in monolayer may be due to an indirect effect of the M, 170,000 glycoprotein, such as culture, in Mminimum essential medium (with ribonucleosides and its effect on intracellular pH. deoxyribonucleosides) supplemented with 10% fetal bovine serum, 10 mM iV-(2-hydroxyethyl)piperazine-jV'-(2-ethanesulfonic acid) buffer, INTRODUCTION pH 7.4, 0.2% sodium bicarbonate, and 40 Â¿ig/mlgentamicin. Stock cultures of C5 were maintained in the presence of colchicine (1 jig/ml), Emergence of drug-resistant cancer cells during the course of to prevent growth of revenants. C5 grown in colchicine-containing chemotherapy is considered a major cause of treatment failure. medium until the time of experiments gave results identical to C5 In in vitro cancer models, MDR' occurs following the amplified maintained in colchicine-free medium. To determine the toxicity of expression of the mdr gene and the appearance of its product, procainamide or cimetidine on AB and C5, cells were grown for 1 week gp-170, in the plasma membrane (1-3). The drugs to which in medium containing either 100 MMprocainamide or cimetidine (sub culture to confluence). These cultures were without evidence of toxicity. resistance develops include the organic cations actinomycin, The human breast cancer line MDA-MB-231, which is sensitive to doxorubicin, and vinblastine. Evidence suggests that gp-170 Adriamycin, and MDA-A1R, which is resistant to Adriamycin, were acts as an efflux pump (3). Other organic cations, including grown as previously described (14). The human cell line KB-3-1, which calcium channel antagonists, calmodulin inhibitors, rifampin, is sensitive to vinblastine, and KB-V-l, which is resistant to vinblastine, and fluorescent dyes, have been reported to increase accumu were generously provided by Dr. Dan Von Hoff (University of Texas lation of cytotoxic drug in multidrug-resistant cells, and some Health Science Center at San Antonio, TX) and were grown as previ have been demonstrated to increase cytotoxicity, reversing the ously reported (6, 15). resistance phenotype (4-6). Since gp-170 has also been found Uptake and Efflux Experiments. Experiments were in 24-well plates (Costar, Cambridge, MA), using confluent cells. Each data point was Received 10/25/91; accepted 4/23/92. in triplicate and corrected for an identically treated well without cells. The costs of publication of this article were defrayed in part by the payment For uptake experiments, medium was removed and the cell sheet was of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. rinsed with GBSS prior to addition of radioactive drug. Following ' This work was supported by NIH Grant CA30195 and the Veterans Admin incubation, radioactive medium was rapidly aspirated and the cell sheet istration Research Service and was reported, in part, at the meeting of the was rinsed once with 2 ml of cold 0.9% NaCl. Cells were then dissolved American Federation for Clinical Research, May 1988 (34). 2To whom requests for reprints should be addressed, at Department of in l N NaOH, and radioactivity was quantitated in acidified aqueous Medicine/Gastroenterology, University of Texas Health Science Center at San counting solution (Amersham), by liquid scintillation counting. Varia Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78284. tions for individual experiments are given in "Results" and the figure 'The abbreviations used are: MDR, multidrug resistance; CHO, Chinese legends. For efflux experiments, following incubation with radioactive hamster ovary; AB, Chinese hamster ovary cell line CHOAUXB1; C5, Chinese hamster ovary cell line CH"C5; GBSS, Grey's balanced salt solution; TEA, drug to establish equilibrium, medium was rapidly removed, the cell tetraethylammonium; gp-170, M, 170,000 glycoprotein. sheet was rinsed with 2 ml of GBSS, and 1 ml of efflux medium ' Deceased. (generally GBSS) was added. At the appropriate times, cells were 3539

more procainamide in the presence of verapamil (Fig. 2, B and C). The effect was not as striking in the KB line as with the CHO line. In Adriamycin-sensitive and -resistant MDA cells, we found neither a difference in procainamide accumulation nor an effect of verapamil (data not shown). Time Course of | 11|( inieiidiiH' Accumulation. AB and C5 were incubated at 23Â°Cwith 70 n\i ['HJcimetidine, in the absence or presence of 100 ^M verapamil. Both AB and C5 progressively accumulated cimetidine, with the sensitive AB line having approximately 5-fold more at 20 min than did C5 (0.49 versus 0.09 fmol/^g DNA). Verapamil did not signifi cantly influence the accumulation of cimetidine by AB but caused a marked increase in cimetidine accumulation in C5, producing levels similar to those in the AB cell line. Tetraethylarnmonium Bromide and Partition Coefficients. | !l I] TEA, a quaternary amine which is virtually always charged, is secreted by the organic cation secretory mechanism in the renal tubule. AB and C5 were incubated with ['H]TEA at 23Â°C,but we were unable to demonstrate any accumulation in either cell with time or varying concentration, whether in the absence or in the presence of 100 ^M verapamil (data not shown). Parti tioning between octanol and GBSS, pH 7.4, at 23Â°Cfor these Fig. 1. Time course of procainamide accumulation in AB. the cell line sensitive to colchicine ill- or C5, the cell line resistant to colchicine (B). Procainamide (SO MMiwas added in the absence or presence of 100 ,.Mverapamil. in Dulbecco's modified Eagle medium, at 23'C. At the times indicated, radioactivity was quantitated by liquid scintillation counting. Each point is the mean of triplicate samples. SD was <5%. harvested following removal of medium and rinsing with 2 ml of cold 0.9% NaCl. DNA was quantitated fluorometrically (16). Most experi ments were performed on at least three occasions. Statistics. Where appropriate, statistical significance was determined by analysis of variance, with P < 0.05 being considered statistically significant. Procainamide Metabolism. Following incubation with ['Hjprocaina- 0 20 40 60 80 100 120 140 mide (90 ÃŸCi/3ml/flask) at 37Â°Cfor3 h, AB and C5 cells in radioactive MINUTES medium were scraped from the plastic flasks and briefly sonicated. Nonradioactive procainamide (15 ^g) and W-acetylprocainamide (15 Mg)were added to each sample. Three ml of acetonitrile were added to each sample, and the samples were mixed and centrifuged to remove precipitated protein. One hundred Â¡.Iwereinjected onto a /iBondapak phenyl column (Millipore Corp., Milford, MA), and 0.5-min fractions were collected for 17 min. The high performance liquid chromatography mobile phase was that of Lai et al. (17). At a pump speed of 0.9 ml min. .V-acetylprocainamide eluted at 5.4 min and procainamide eluted at 6.4 min. Fractions were counted in a liquid scintillation counter. 20 40 60 80 100 120 140 MINUTES

RESULTS Time Course of I'HlProcainamide Accumulation. AB and C5 were incubated at 23Â°Cwith 50 nivi ['H]procainamide, in the absence or presence of 100 MMverapamil (Fig. 1). AB progres sively accumulated procainamide, reaching 4.5 fmol/Vg DNA at 20 min. The rate and amount of accumulation were not influenced by verapamil (Fig. ÃŒA).C5accumulated procaina mide but reached equilibrium earlier and at a much lower level (0.5 fmol/jig DNA) than AB. Exposure to verapamil caused C5 to accumulate procainamide with a time course similar to that of AB (Fig. IB). To determine whether procainamide was Fig. 2. Time course of procainamide accumulation in KB cells. Procainamide (SO nM) was added in the absence or presence of 20 ^M verapamil, in GBSS, at differentially accumulated in other models of MDR, similar 23'C. At the times indicated, radioactivity was quantitated by liquid scintillation experiments were carried out in the human KB cell line and the counting. A, comparison of procainamide uptake in KB-3-1. the cell line sensitive human breast cell line MDA. KB-3-1 cells (sensitive to vin- to vinblastine. with that in KB-V-1. the cell line resistant to vinblastine; B, effect of verapamil on procainamide uptake in KB-3-1; C, effect of verapamil on blastine) accumulated more procainamide than did vinblastine- procainamide uptake in KB-V-1. Each point is the mean of triplicate samples. resistant KB-V-1 cells (Fig. 2A), and only KB-V-1 accumulated SDwas<5%. 3540

organic cations was as follows: TEA, 0.006; procainamide, 0.275; and cimetidine, 2.101. Effect of Procainamide Concentration on Accumulation. Since procainamide was accumulated to the greatest extent in AB and C5, we used this probe for subsequent characterization. To Â¡0 111 determine whether procainamide accumulation occurred by O passive diffusion or by a carrier-mediated process, AB and C5 were incubated with increasing concentrations of procainamide (14 HM to 10 MM)for 1 to 6 min at 23Â°C(Figs. 3A and 4A). There was a linear increase in cell-associated procainamide with time (r > 0.98) at each medium procainamide concentration in both AB and C5. A plot of initial rates versus procainamide concentration was linear, without evidence of saturation (Figs. 3fi and 4B). Verapamil (10 ÃŸM)was without effect in AB y =. 0.0396 t 1.4088Â« R = 1.00 regardless of procainamide medium concentration but caused a large increase in procainamide accumulation in C5 at all me dium procainamide concentrations (Fig. 5). These findings would be consistent with entry of procainamide by passive diffusion. Furthermore, there was no difference in procaina mide accumulation in AB or C5 when excess cimetidine or TEA was in the incubation medium, as might have been ex 00 0.2 0.4 0.6 0.8 1.2 pected if competition for a common carrier occurred (data not PROCAINAMIDE (, M) shown) (10, 11). Fig. 3. Effect of medium procainamide concentration on accumulation of Effect of Other Calcium Channel Antagonists on Procainamide procainamide in AB cells. Procainamide (14 n\i to 1 /IM) was added for 1 to 6 Accumulation. Since calcium channel antagonists other than min at 23Â°C.Following incubation, radioactivity was quantitated by liquid scin verapamil have also been demonstrated to reverse the MDR tillation counting. /. time course data were analyzed by linear regression, to calculate initial velocities (r > 0.98 for each line). Each point is the mean of quadruplicate samples. SD was s5%. Procainamide concentrations were: D, 0.014 AIM;â€¢¿,0.064AIM;A,0.114 /IM;A, 0.214 /IM;O, 0.514 AIM;and â€¢¿,l/IM. B, initial velocities were linearly correlated with procainamide concentration, consistent with uptake by passive diffusion.

ui in LU .1 O

a i

PROCAINAMIDE (. M> 4 6 8 10 12 Fig. 5. Effect of medium procainamide concentration on accumulation of PROCAINAMIDE (,.M) procainamide in CHO cells. Procainamide (50 nM to 50 AIM)was added for 15 Fig. 4. Effect of medium procainamide concentration on accumulation of min at 23"C, in the absence or presence of 10 AIMverapamil. Following incubation, procainamide in C5 cells. Procainamide (14 nM to 10 /AD was added for 1 to 6 radioactivity was quantitated by liquid scintillation counting. Inset, enlargement min at 23*C. Following incubation, radioactivity was quantitated by liquid scin of the data for 100 nM to 2 AIMprocainamide. Each point is the mean of triplicate tillation counting. I, time course data were analyzed by linear regression, to samples. SD was s5%. A, procainamide alone in AB cells; A, procainamide plus calculate initial velocities (r > 0.98 for each line). Each point is the mean of verapamil in AB cells; D. procainamide alone in C5 cells; â€¢¿.procainamideplus quadruplicate samples. SD was Â«5%.Procainamide concentrations were: D, 0.014 verapamil in C5 cells. In AB cells the presence of verapamil was without effect AIM;A, 0.514 AIM;A, l /IM; â€¢¿,2/AI: O, 5 AIM:and â€¢¿IOAIM.B, initial velocities regardless of procainamide concentration, but in C5 cells the presence of vera were linearly correlated with procainamide concentration, consistent with uptake pamil resulted in greater procainamide accumulation at all procainamide by passive diffusion. concentrations. 3541

measured at the higher temperature. Preincubation with Verapamil and Subsequent Effect. Since 50-40 verapamil did not increase procainamide accumulation at 4Â°C, C5 cells were incubated with 10 ^M verapamil at 4Â°Cfor 10 -30-20 min and then washed repeatedly before incubation with [3H] procainamide at 23Â°C.Similarly treated cells had no verapamil M UJ present or verapamil only present during the 23Â°Cincubation.

-10--i-, Preliminary incubations at 4Â°Cwith ['Hjverapamil demon strated that the washing technique removed >99% of the vera pamil, to a level which did not increase procainamide accumu lation in our hands (data not shown). Procainamide accumula tion in C5 increased significantly (approximately 4-fold) with 0 1 5 10 50 100 both verapamil treatments (Fig. 8). These results demonstrate uM AGENT that preincubation with verapamil at 4Â°Cis sufficient for the

Fig. 6. Effect of varying concentrations of verapamil, diltiazem, and nifedipine on procainamide accumulation in C5. Cells were incubated for 5 min with 0-100 UMverapamil (â€¢),diltiazem (D). or nifedipine (D), with 50 n\i procainamide. Each value is the mean of triplicate samples. SD was <5%.

Table 1 Effect of organic cations on procainamide uptake by CHO cells AB or C5 cells were incubated with 50 UM ['H]procainamide for 15 min at 23Â°C,inthe absence or presence of added drug. Radioactive medium was removed and cells were washed prior to quantitation of cell-accumulated procainamide. uptake DNA)AB23.79(fmol/Mg CompoundNone (MM)10 Â±3.01 Â±0.52 CONTROL -fVERAPAMIL Rifampin 23.80 Â±1.51 9.90 Â±2.63Â° Rifampin 100 27.11 Â±1.76 12.71 Â±0.47Â° Tamoxifen 10 23.21 Â±2.71 8.46 Â±0.21" Tamoxifen 100 24.72 Â±2.68 11.84 Â±0.34Â° Rhodamine 10 23.59 Â±0.76 8.15 + 0.12Â° Rhodamine 100 23.61 Â±1.29 14.00+ 1.14Â° Ethidium 10 23.30 Â±1.8 8.37 Â±0.97Â° EthidiumConcentration 100Procainamide25.18 Â±1.6C55.89 11.3 Â±0.13Â° Â°P< 0.05 venus control (no added compound). o Z phenotype, C5 was incubated with [lH]procainamide for 6 min at pH 7, with increasing concentrations of verapamil, diltiazem, or nifedipine (Fig. 6). All of the calcium channel antagonists CONTROL tVERAPAMIL Fig. 7. Effect of temperature on verapamil effect in CHO cells. Cells were increased procainamide accumulation, but with different poten incubated with 50 n\i procainamide for 5 min at either 23'C (â€¢)or 4"t ' (Q). in cies. The dose resulting in a 50% maximal effect was <1 /Â¿M the absence or presence of 100 MMverapamil. A, AB cells; B, C5 cells. Each value for verapamil, 3 I*Mfor diltiazem, and 25 UMfor nifedipine. is the mean Â±SI) of triplicate samples. Effect of Other Organic Cations on Procainamide Accumula tion. We tested some of the other organic cations which have been reported to increase accumulation of cytotoxic agents in resistant cells of several MDR models, to assess their effects on procainamide accumulation. AB and C5 were incubated with ['H]procainamide in the absence or presence of rifampin, tamoxifen, rhodamine, or ethidium. None of these compounds at the concentrations tested had an effect on procainamide accu mulation in AB, but each enhanced procainamide accumulation in C5 and this appeared to be related to dose (Table 1). Effect of Temperature on Procainamide Accumulation and Verapamil Effect. AB and C5 were incubated with pH]procainamide for 5 min in the absence or presence of 100 UMverapamil, at either 4Â°Cor 23Â°C,to determine whether the effect of verapamil on procainamide accumulation in C5 was tempera VERAPAMIL PRESENT IN THE 1st INCUBATION (4"C) ture dependent. In AB cells, much more procainamide accu OR THE 2nd INCUBATION (23'C) mulated at the higher temperature but verapamil was without Fig. 8. Effect of preincubation temperature on verapamil effect in CHO cells. effect at either temperature (Fig. 7/4). In C5 cells, there also Cells were incubated with or without 10 MMverapamil for 10 min at 4'C. Cells were then washed extensively, first with ice-cold medium and then with medium was more accumulation of procainamide at the higher temper at 23'C. Procainamide (50 n\i) was added for 10 min at 23'C, in the absence or ature and verapamil significantly increased procainamide ac presence of 10 MMverapamil. Each value is the mean Â±SD of triplicate samples. cumulation only at the higher temperature (Fig. IB). AB accu In AB cells, neither preincubation nor concomitant incubation with verapamil affected procainamide accumulation. In C5 cells, preincubation with verapamil mulated only 10% of the procainamide measured at the higher at 4'C resulted in a similar increase in procainamide accumulation as with temperature, whereas C5 accumulated 50% of the amount concomitant incubation. 3542

Downloaded from cancerres.aacrjournals.org on October 6, 2021. © 1992 American Association for Cancer Research. ORGANIC CATION UPTAKE BY CHO CELLS efflux rate for C5 was 0.272 Â±0.017 fmol/^g DNA/min (/,/, = 2.55 min). The presence of verapamil in the incubation, whether during the uptake period or during the efflux period, had no significant effect on procainamide efflux from either AB (Fig. 9B) or C5 (Fig. 9C). Efflux half-times for AB (in min) (where the first plus or minus refers to the presence or absence of O verapamil in the 90-min uptake incubation and the second plus a or minus refers to the presence or absence of verapamil in the efflux medium) were: -/+, 2.56; +/-, 2.24; and +/+, 2.56. Efflux half-times for C5 (in min) were: -/+, 2.14; +/-, 2.02; and +/+, 1.75. Verapamil did increase procainamide accumu lation in C5 when present during the uptake period, as expected. Verapamil (20 Â¿Ã•M)hadno effect on procainamide efflux from KB-V-1 (data not shown). Effect of Verapamil on Rate of Procainamide Uptake. Initial rates of procainamide accumulation were measured in AB and Ã‡A C5 following incubation with 50 nM ['Hjprocainamide, in the lU _J O absence or presence of 10 nM verapamil. The rate of procaina mide accumulation was 2.6-fold greater in AB than in C5 at early time points (Fig. 10/1). Verapamil was without effect on the rate of procainamide accumulation in AB (Fig. 105) but increased the rate of procainamide accumulation 2.8-fold in C5 cells (Fig. IOC).

g y . 1.6015 t 0.0373X R Â«1.00

o i 6 4

Fig. 9. Effect of verapamil on procainamide efflux from CHO cells. Cells were incubated with 50 nM procainamide, with or without 10 MMverapamil, for 90 100 min at 23'C. Procainamide was removed, cells were rapidly washed, and cells SECONDS were placed in procainamide-free medium with or without 10 MMverapamil. Cells were harvested following an additional wash at the times indicated. A, efflux from AB and CS without exposure to verapamil; lÃ¬.effluxfrom AB with or without exposure to verapamil; C, efflux from C5 with or without exposure to verapamil. Each point is the mean of triplicate samples. SD was <5%. I 6- S subsequent increase in procainamide accumulation in C5 and Â« that verapamil need not be present in the incubation medium containing procainamide. 2- Effect of Verapamil on Procainamide Efflux from CHO Cells. The difference in levels of cytotoxic drug accumulation between sensitive cell lines and resistant cell lines expressing the mdr gene is due to gp-170 acting as an efflux pump. To determine whether the differences in procainamide accumulation found between AB and C5 could be explained by differences in rates of efflux, AB and C5 were incubated with [-'Hlprocainamide for 90 min (a time sufficient to reach equilibrium; data not shown) at 23Â°C.Cell-associated procainamide remaining at various â€¢¿J, III times after medium procainamide removal and cell washing was quantitated. Since removal of medium radioactivity and cell washing take approximately 30 s/well, it is possible to miss very rapid efflux. Therefore, some cells were brought to 4Â°C following the 90-min incubation at 23Â°Cand cells were washed at 4Â°C.The equilibrium value (O min of efflux) was identical whether obtained at 4Â°Cor 23Â°C,making a missed early rapid Fig. 10. Effect of verapamil on rate of uptake of procainamide in CHO cells. Cells were exposed to 50 nM procainamide, with or without 10 MMverapamil, efflux unlikely. Procainamide efflux from AB and C5 was and then harvested at the times indicated.. I. AB (D) and C5 (â€¢)without exposure to verapamil; B. AB with (â€¢)and without (D) verapamil; C, C5 with (â€¢)and biexponential (Fig. 9/1). The efflux rate from 0 to 5 min for AB without (D) verapamil. The lines are linear regressions of the data points. Each was 0.283 Â±0.058 fmol/^g DNA/min (t.â€ž=2.45 min), and the point is the mean of triplicate samples. SD was <5%. 3543

DISCUSSION Energy-dependent active efflux appears to be the mechanism by which gp-170 renders a cell multiply drug resistant, as shown by many in vitro observations (1-3) and occasionally by direct visualization (20-22). Other explanations have also been sug gested (1). Since gp-170 is a constituent of organs which nor mally engage in organic cation transport (8, 9), well character ized noncytotoxic transport probes (10,11) were used in models of MDR, to determine their handling. These agents have the advantage of being very poorly bound to proteins and cellular constituents or plastic culture dishes and are, thus, easily used in kinetic studies. This study demonstrates that procainamide and cimetidine, but not TEA, accumulate in the colchicine-sensitive AB line to CTL AZ *3oMG,+AZ a greater extent than in the resistant C5 line. Procainamide Fig. 11. Effect of azide on procainamide accumulation in CHO cells. Cells were exposed to 50 n\i procainamide for 15 min, in glucose-free GBSS (CTL), uptake was passive in both AB and C5, and there was no glucose-free GBSS plus 1 mg/ml glucose (+G), glucose-free GBSS plus 1 mg/ml difference in procainamide efflux between the two cell lines. 3-O-methylglucose (+3oMG), glucose-free GBSS plus 0.1 % azide (+AZ), glucose- Influx was greater in the sensitive cell line than in the resistant free GBSS plus both glucose and azide (+G.+AZ), or glucose-free GBSS plus both 3-O-methylglucose and azide (+3oMG,+AZ). Each value is the mean of 2- cell line. Influx was increased by verapamil in the resistant cell 4 experiments with triplicate samples. Exposure to azide resulted in greater line, but verapamil had no effect on procainamide efflux. Influx accumulation of procainamide than in control cells (P < 0.05). The azide effect was reversed by glucose but not by 3-O-methylglucose. was also dependent on energy state, in that influx increased in C5 in the presence of azide in glucose-free medium and could be reversed by addition of glucose but not 3-O-methylglucose. Effect of Azide in Glucose-free Medium on Procainamide Procainamide efflux was not different between AB and C5 in Accumulation. One of the central features suggesting an efflux pump in other models of MDR studied with cytotoxic drugs has been the finding that azide in glucose-free medium increases accumulation in the resistant cell line, suggesting that energy is required to maintain the pump (18, 19). AB and C5 cells were incubated with procainamide, in glucose-free GBSS, alone or in the presence of 1 mg/ml added glucose, 0.1% sodium azide, or both. Procainamide accumulation was essentially the same in AB cells regardless of the addition of glucose, azide, or both (Fig. 11). However, azide increased the amount of pro cainamide accumulated in C5 cells (P < 0.05 versus control) and this was reversed when glucose was added with the azide (Fig. 11), suggesting that net procainamide accumulation might be energy dependent. A similar experiment was also performed with 3-O-methylglucose, a substrate for the glucose transporter but not an energy source. Unlike glucose, 3-O-methylglucose did not reverse the increase in procainamide accumulation in 90 C5 (Fig. 1)). The influence of azide on procainamide efflux 80 - from C5 cells was also studied, to determine whether the azide effect was due to a decrease in efflux rate. C5 cells were 70- 60 - incubated for 30 min with procainamide in GBSS, glucose-free y = 12.7731 + 0.2575X R = 0.98 GBSS, or glucose-free GBSS containing 0.1% azide. Radioac 50- tive medium was removed, cells were washed, and appropriate medium was replaced. Azide-containing medium resulted in a 40- higher zero-time (equilibrium) procainamide accumulation than 30 did glucose-free medium without azide (Fig. 12/1). However, 20- the rates of procainamide efflux were similar, with a half-life 10- y = 14.9099 t 0.1067X R = 0.77 for glucose-free medium of 14.5 min and a half-life for azide- containing medium of 11.5 min. Rates of procainamide accu mulation were also studied, and azide was found to increase procainamide accumulation 2.4-fold (Fig. \2B). Taken to gether, these data suggest that the effect of azide in glucose- Fig. 12. Effect of azide on efflux 1.11and uptake (B) of procainamide in C5. A, cells were incubated for 30 min with 50 n\i procainamide in GBSS (D), glucose- free medium is on procainamide uptake rather than procaina free GBSS (A), or glucose-free GBSS containing 0.1 % sodium azide (â€¢).Procain mide efflux. amide was removed, cells were washed with the respective medium without procainamide, and GBSS, glucose-free GBSS, or glucose-free GBSS containing Lack of Procainamide Metabolism in CHO Cells. Confluent 0.1% azide was replaced. Cells were harvested at the times indicated. Each point T25 flasks of AB and C5 cells and a flask without cells were is the mean of triplicate samples. SD was <5%. Ã„,cells were incubated with 50 incubated with pHJprocainamide for 3 h at 37"C. There was no nM procainamide in glucose-free GBSS with (â€¢)or without (D) 0.1% sodium azide and were subsequently harvested at the times indicated. Lines are linear evidence for metabolism (no new peaks or any decrease in regressions of the data points; each point is the mean of triplicate samples. SD recovery of procainamide) in either AB or C5. was <5%. 3544

Downloaded from cancerres.aacrjournals.org on October 6, 2021. © 1992 American Association for Cancer Research. ORGANIC CATION UPTAKE BY CHO CELLS the presence of azide in glucose-free medium. cimetidine accumulation is that it occurs secondary to some There was no evidence for an active organic cation transport action of gp-170. MDR cells have an increased cytosolic pH, system, such as that found in the kidney, as an explanation for which is thought to be a consequence of the multidrug trans the difference in procainamide accumulation in AB and C5. porter (28-30). In one report, C5 intracellular pH was 0.17 Procainamide accumulation was by passive diffusion. If accu units higher than that in AB (29), whereas the difference has mulation is totally passive for each of the three probes tested, been reported to be as much as 0.44 pH units in other cells TEA would be excluded from the cells in short term experi (28). Addition of verapamil results in intracellular pH returning ments, since TEA is always charged and very poorly lipid to levels found in sensitive cells. Thus, gp-170 may act as a soluble. Indeed, this was what was found. Horio et al. (23) also proton pump. Drugs such as procainamide and cimetidine reported that TEA had no effect on vinblastine transport by the might be expected to accumulate to a greater degree in cells Madin-Darby canine kidney cell line. Cimetidine, however, with a lower intracellular pH, due to trapping as the cation. would be expected to accumulate to a greater extent than This has been demonstrated for cimetidine and procainamide procainamide, since at pH 7.4 there is approximately 100-fold in rabbit renal brush-border membrane vesicles with an out more uncharged cimetidine and it is 10-fold more lipid soluble. wardly directed proton gradient (31). The procainamide obser This was not the case, suggesting that cimetidine accumulation vations in CHO cells might be explained as a result of the effect may be determined by factors besides lipid solubility and charge. of gp-170 on intracellular pH. Cimetidine would accumulate to Efflux rates for procainamide were not different in AB and C5 a lesser extent, in part due to its lower pK,,. TEA could not and, thus, would not explain the difference in accumulation in accumulate because it is always charged and water soluble and the two cell lines. This finding does not indicate that mediated cannot passively enter the cell to a significant degree. The efflux does not exist, although we were unable to demonstrate variable degree to which intracellular pH is affected in MDR trans-stimulation in either cell line (data not shown). We were cells might explain why we found no difference in procainamide also unable to inhibit overall renal secretion of the MDR accumulation in sensitive and resistant MDA cells and a smaller substrate colchicine with the organic cations TEA or A'-meth- difference in KB cells than in CHO cells. To explain a 2.5-fold ylnicotinamide in vivo (24). difference in initial rates of uptake between AB and C5 or in How, then, can the difference in procainamide and cimetidine C5 in the presence and absence of verapamil, a cytosolic pH accumulation be explained? One possibility is that the mem difference of 0.4 units would be necessary, whereas explanation branes of the two cell lines are intrinsically different, such that of a 5-fold difference in procainamide accumulation would permeability, even if passive, is not the same as was originally require a cytosolic pH difference of 0.7 units (32), which is suggested by Ling and Thompson (13). While this possibility greater than has been previously reported. We do not know the cannot be excluded, it seems inconsistent with the observations cytosolic pH values in the CHO cells in our experiments, nor that procainamide accumulation is dependent on energy state do we know whether procainamide stays in the cytosol or moves (actually increased by energy depletion in the resistant cell line) into other, more acidic, compartments, such as lysosomes (33). and increased in the resistant cell line by verapamil and other agents known to interact with gp-170. REFERENCES Another possibility is that procainamide and cimetidine are substrates for gp-170. There are other examples in the literature 1. Beck, W. T. The cell biology of multiple drug resistance. Biochem. Pharma- in which differences in drug accumulation in the MDR cell col., 36: 2879-2887, 1987. 2. Pastan, I., and Gottesman. M. M. Multiple-drug resistance in human cancer. have been reported to be due to less drug uptake than in its N. Engl. J. Med., 316: 1388-1393. 1987. sensitive counterpart and not due to a significant difference in 3. Gottesman, M. M., and Pastan, 1. The multidrug transporter, a double-edged drug efflux. For example, Fojo et al. (6) found slower drug sword. J. Biol. Chem.. 263: 12163-12166, 1988. 4. Tsuruo, T., lida, H., Tsukagoshi, S., and Sakurai. Y. Overcoming of vincris influx in multiply drug-resistant KB cells, compared to sensitive tine resistance in P388 leukemia in vivo and in vitro through enhanced KB cells, without a difference in rates of drug efflux. Verapamil cytotoxicity to vincristine and vinblastine by verapamil. Cancer Res., 41: 1967-1972, 1981. increased drug influx in the resistant KB cells. Sirotnak et al. 5. Tsuruo, T., lida, H., Tsukagoshi. S., and Sakurai, Y. Increased accumulation (25) reported a 24-fold difference in vincristine influx in resist of vincristine and Adriamycin in drug-resistant P388 tumor cells following incubation with calcium antagonists and calmodulin inhibitors. Cancer Res., ant Chinese hamster lung cells, compared to sensitive cells, but Â«.â€¢4730-4733,1982. only a 2-fold difference in efflux rates between the two cell 6. Fojo, A., Akiyama, S., Gottesman, M. M., and Pastan, I. Reduced drug lines. Danks et al. (26) reported a 10-fold higher rate of influx accumulation in multiply drug-resistant human KB carcinoma cell lines. of VP-16 in sensitive human leukemic cells, compared to re Cancer Res., 45: 3002-3007. 1985. 7. Fojo, A., Ueda, K., Slamon, D. J., Poplack, D. G., Gottesman, M. M., and sistant cells, whereas VP-16 efflux was actually 2.5-fold slower Pastan. I. Expression of a multidrug-resistance gene in human tumors and in the resistant cells. Ramu et al. (27) reported that doxorubicin tissues. Proc. Nati. Acad. Sci. USA, 84: 265-269, 1987. 8. Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman, M. M., Pastan, I., and influx was more rapid in sensitive P388 cells than in resistant Willingham. M. C. Cellular localization of the multidrug-resistance gene P388 cells, whereas there was no difference in doxorubicin product P-glycoprotein in normal human tissues. Proc. Nati. Acad. Sci. USA, efflux between the two cell lines. If procainamide and cimetidine Â«4:7735-7738, 1987. are substrates for gp-170, the mechanism by which drug accu 9. Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman, M. M., Pastan, I., and Willingham, M. C. Immunohistochemical localization in normal tissues of mulation is accomplished would not seem to be due to active different epitopes in the multidrug transport protein PI70: evidence for efflux, since we found no difference between the sensitive and localization in brain capillaries and cross-reactivity of one antibody with a muscle protein. J. Histochem. Cylochem., 37: 159-164, 1989. resistant cells. Verapamil had no measurable effect on procain 10. McKinney, T. D., and Speeg, K. V. Cimetidine and procainamide secretion amide efflux from either AB or C5 cells, whether present during by proximal tubules. Am. J. Physiol., 242: F672-F680, 1982. 11. McKinney, T. D., deLeon, C., and Speeg, K. V. Organic cation uptake by a procainamide uptake or efflux or both. An early, very rapid, cultured renal epithelium. J. Cell Physiol., 137: 513-520, 1988. efflux component was sought but not found, and verapamil had 12. McKinney, T. D. Renal transport of organic anions and cations. In: R. W. the expected enhancing effect on procainamide accumulation Schrier and C. W. Gottschalk (eds.). Diseases of the Kidney, pp. 285-307. Boston: Little Brown, 1988. inC5. 13. Ling, V., and Thompson. L. H. Reduced permeability in CHO cells as a Another possibility for the difference in procainamide and mechanism of resistance to colchicine. J. Cell Physiol., 83: 103-116, 1974. 3545

14. Schneider, S. L., Fuqua. S. A., Speeg, K. V., Tandon, A. K., and McGuire, studied in vivo. J. Pharmacol. Exp. Ther.. 261: 50-55, 1992. W. L. Isolation and characterization of an Adriamycin-resistant breast tumor 25. Sirotnak, F. M., Yang, C-H., Mines, L. S., Oribe, E., and Biedjer. J. L. cell line. In Vitro Cell. Dev. Biol., 26: 621-628, 1990. Markedly altered membrane transport and intracellular binding of vincristine 15. Akiyama, S., Fojo, A., Hanover. J. A., Pastan. I., and Gottesman, M. M. in multidrug-resistant Chinese hamster cells selected for resistance to Vinca Isolation and genetic characterization of human KB cell lines resistant to alkaloids. J. Cell. Physiol., 126: 266-274. 1986. multiple drugs. Somat. Cell Mol. Genet.. //: 117-126, 1985. 26. Danks, M. K., Yalowich, J. C, and Beck, W. T. Atypical multiple drug 16. Kapuscinski. J., and Skoczylas, B. Simple and rapid fluorometric method for resistance in a human leukemic cell line selected for resistance to teniposide DNA microassay. Anal. Biochem., 83: 252-257, 1977. (VM-26). Cancer Res., 47: 1297-1301, 1987. 17. Lai, C, Kamath. B. L.. Look, Z. M., and Yacobi, A. Determination of 27. Ramu, A.. Pollard, H. B.. and Rosario. L. M. Doxorubicin resistance in P388 procainamide and A'-acetylprocainamide in biological fluids by high-pressure leukemia: evidence for reduced drug influx. Int. J. Cancer, 44: 539-547, liquid chromatography. J. Pharm. Sci., 69:982-984, 1980. 1989. 18. Skovsgaard, T. Mechanisms of resistance to daunorubicin in Ehrlich ascites 28. Keizer, H. G., and Joenje, H. Increased cytosolic pH in multidrug-resistant tumor cells. Cancer Res., 38: 1785-1791, 1978. human lung tumor cells: effect of verapamil. J. Nail. Cancer Inst., 81: 706- 19. Inaba, M.. Kobayashi. H., Sakurai. Y., and Johnson, R. K. Active efflux of 709, 1989. daunorubicin and Adriamycin in sensitive and resistant sublines of P388 29. Boscoboinik. D., Gupta, R. S., and Epand, R. M. Investigation of the leukemia. Cancer Res., 39: 2200-2203, 1979. relationship between altered intracellular pH and mullidrug resistance in mammalian cells. Br. J. Cancer. 61: 568-572, 1990. 20. Yalowich, S., and Taub, R. N. Differences in daunomycin retention in sensitive and resistant P388 leukemic cells as determined by digitized video 30. Thiebaut, F., Currier, S. J.. Whitaker, J., Haugland, R. P., Gottesman. M. fluorescence microscopy. Cancer Res., 43: 4167-4171, 1983. M.. Pastan, I., and Willingham. M. C. Activity of the multidrug transporter results in alkalinization of the cytosol: measurement of cytosolic pH by 21. Willingham. M. C., Cornwell, M. M., Cardarelli, C. O.. Gottesman. M. M., microinjection of a pH-sensitive dye. J. Histochem. Cytochem., 38: 685- and Pastan, I. Single cell analysis of daunomycin uptake and efflux in multidrug-resistant and -sensitive KB cells: effects of verapamil and other 690, 1990. 31. McKinney, T. W., and Kunnemann, M. E. Cimetidine transport in rabbit drugs. Cancer Res., 46: 5941-5946, 1986. renal cortical brush-border membrane vesicles. Am. J. Physiol., 252: F525- 22. Neyfakh, A. A. Use of fluorescent dyes as molecular probes for the study of F535. 1987. multidrug resistance. Exp. Cell Res., 174: 168-176, 1988. 32. Weiner, I. M. Transport of weak acids and bases. In: J. Orloff (ed.). Handbook 23. Ilorio. M., Pastan, I., Gottesman, M. M., and Handler, J. S. Transepithelial of Physiology: Renal Physiology, pp. 521-554. Bethesda, MD: American transport of vinblastine by kidney-derived cell lines. Application of a new Physiological Society. 1973. kinetic model to estimate in situ A, of the pump. Biochim. Biophys. Acta, 33. Poole, B.. and Ohkuma, S. Effect of weak bases on the intralysosomal pH in 1027: 116-122, 1990. mouse peritoneal macrophages. J. Cell Biol.. 90: 665-669, 1981. 24. Speeg. K. V., Maldonado. A. L.. Liaci, J., and Muirhead, D. The effect of 34. Speeg. K. V.. and McGuire, W. L. Use of noncytotoxic transport probes in cyclosporine on colchicine secretion by the kidney multidrug transporter a model of multiple drug resistance. Clin. Res., 36: 500A, 1988.

3546

Downloaded from cancerres.aacrjournals.org on October 6, 2021. © 1992 American Association for Cancer Research. Uptake of the Noncytotoxic Transport Probe Procainamide in the Chinese Hamster Ovary Model of Multidrug Resistance

K. V. Speeg, Jr., Catherine deLeon and William L. McGuire

Cancer Res 1992;52:3539-3546.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/13/3539

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

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

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/52/13/3539. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.