Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The retinal pigment epithelium (RPE) is a single cell layer that lies at the back of the vertebrate eye. RPE cells are polarized, i.e., their apical and basolateral membranes are functionally different. Thus, they mediate vectorial transport of endogenous and exogenous compounds, metabolites, ions, and fluid across the RPE, between the neural retina and the choroidal blood supply (Joseph and Miller, 1991). Recent molecular studies have shown expression of specific transporters for monocarboxylic acids (Gerhart et al., 1999), taurine (Vinnakota et al., 1997), peptide/histidine (Yamashita et al., 1997), glucose (Kumagai et al., 1994), and folic acid (Huang et al., 1997) in RPE cells of several species, including humans. However, the capacity of RPE cells to transport drugs and xenobiotics is largely unexplored. The studies presented here focus on the capacity of RPE cells to transport cationic drugs, specifically the cationic calcium channel blocker verapamil. Verapamil has been used in the treatment of arterial hypertension and glaucoma via systemic administration, but cardiovascular side effects posed an important problem (Monica et al., 1983; Psaty et al., 1995). Recently, it was shown in patients with ocular hypertension that topically administered verapamil significantly reduced intraocular pressure without alteration of systemic blood pressure (Santafe et al., 1996; Abreu et al., 1998; Melena et al., 1999). Thus, topical application of verapamil could be effective in the management of ocular hypertension (Santafe et al., 1996; Abreu et al., 1998; Melena et al., 1999) and low-tension glaucoma (Ettl et al., 1993; Netland et al., 1995; Santafe et al., 1996) without the serious cardiovascular side effects that are commonly found with systemic administration of calcium channel blockers (Giacomini et al., 1985; Ettl et al., 1998).

Topically administered verapamil readily penetrates into the anterior chamber of the eye and accumulates in the aqueous humor, which acts as a drug reservoir for distribution to other compartments (Ettl et al., 1998; Siegner et al., 1998), i.e., iris-ciliary body, lens, vitreous humor, and choroid-retina (Bourlais et al., 1998). Direct measurement of the distribution of verapamil after single topical administration in rabbit eyes showed that topical verapamil rapidly entered vitreous humor as well as aqueous humor (Ettl et al., 1998; Siegner et al., 1998). Because the half-life of verapamil in vitreous humor is longer than that in either aqueous humor or serum (Siegner et al., 1998), repetitive administration of topical verapamil may produce a high concentration of verapamil in vitreous humor, from which it may enter the RPE cells. However, there is no information about possible elimination of verapamil from the vitreous humor. To begin to address this issue, we have measured the uptake of verapamil by human RPE cells in culture. These studies indicate that verapamil transport by RPE cells is both carrier-mediated and pH-dependent. Characterization of this transporter indicates that the human RPE cell verapamil transporter is unique, with features unlike previously characterized organic cation transporters.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. [³H]Verapamil (85.0 Ci/mmol), [³H]water (5 mCi/ml), and [¹⁴C]carboxyl-inulin (2.21 mCi/g) were purchased from NEN Life Science Products (Boston, MA). All other chemicals were obtained from commercial sources and were of the highest grade available.

Cell Culture. The human RPE cell line (designated as RPE/Hu) previously isolated by Hu et al. (1982) was used for the majority of the studies reported below. This cell line was isolated from an anonymous donor sample not referable to any patient (Hu et al., 1982). These cells were maintained in Ham's F-12 medium, supplemented with 10% fetal bovine serum in a humidified incubator at 37°C with 5% CO₂. In addition, all experiments were repeated using a commercially available human RPE cell line, ARPE-19, obtained from the American Type Culture Collection (Manassas, VA). ARPE-19 cells were grown in 1:1 Dulbecco's modified Eagle's medium:Ham's F-12 medium supplemented with 3 mM L-glutamine and 10% fetal bovine serum. Cells of either cell line were split 1:20 every 3 to 4 days. For transport experiments, 5 × 10⁵ cells were plated into individual wells (3.5 cm²) and cultured for 2 days before transport measurement. The culture medium was changed daily.

Transport Measurement. We measured verapamil uptake in monolayers of cells grown on nonporous plastic support, exposing the apical membrane of the cells to the transport buffer containing [³H]verapamil. Before measuring transport, the culture medium was removed and the cells were washed twice with 5 ml of transport buffer (137 mM NaCl, 5.4 mM KCl, 0.3 mM Na₂HPO₄, 0.4 mM KH₂PO₄, 1.3 mM CaCl₂, 0.8 mM MgSO₄, 5.5 mM glucose, 4.2 mM NaHCO₃, and 10 mM Hepes) at pH 7.4 and replaced with 1 ml of transport buffer containing [³H]verapamil (0.4 µCi/ml). After incubation for specific periods at 37°C, the medium was removed and the cells were rapidly rinsed three times with ice-cold 0.1 M MgCl₂. The cells were dissolved in 2 ml of 1 N NaOH and neutralized with 2 ml of 1 N HCl. Aliquots were removed for protein assay using a Bio-Rad protein assay kit (Bio-Rad Laboratories, Hercules, CA) with bovine serum albumin used as a standard and liquid scintillation counting using 15 ml of Ecolume (ICN Biomedical, Cleveland, OH). The intracellular volume of RPE cells (6.91 ± 1.41 µl/mg of protein) was measured with [³H]water and [¹⁴C]carboxyl-inulin and used for calculation of intracellular concentration of verapamil. [³H]verapamil uptake is expressed as the cell-to-medium ratio.

Estimation of Kinetic Parameters. The kinetic parameters for verapamil uptake by RPE cells were calculated by fitting the data to the following equation:

(1)

where V is the uptake rate of verapamil (pmol/min/mg of protein), S is the verapamil concentration in the medium (µM), K_m is the Michaelis-Menten constant (µM), V_max is the maximum uptake rate (pmol/min/mg of protein), and P_dif is the permeability coefficient (µl/min/mg of protein). Curve fitting was performed by an iterative nonlinear least-squares method using a MULTI program (Yamaoka et al., 1981). The input data were weighted as the reciprocal of the observed values and the Damping Gauss Newton Method was used as the fitting algorithm.

(2)

(3)

Statistics. Each observation was obtained from 6 to 12 individual experiments, each conducted in triplicate using the cell line obtained from Dr. Hu (RPE/Hu). In addition, as noted in the text, experimental findings were verified using the commercially available cell line ARPE-19 (n = 3). Data are presented as mean ± S.E. and the average of replicates were shown in all figures. Differences in mean values were considered to be significant when p < 0.05 and comparison between experimental groups was determined using unpaired Student's t test.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Time Course. Apical uptake of 1 µM [³H]verapamil was determined in RPE cells grown to confluence on solid support. Uptake is expressed as cell-to-medium ratio, i.e., concentration in intracellular water  medium concentration. As shown in Fig. 1, verapamil uptake by the RPE/Hu cell line (Hu et al., 1982) increased rapidly over the initial 5 min and continued to increase for 1 h. The cell-to-medium ratio at 1 min was 14, and reached 42 by 1 h, demonstrating that verapamil is extensively concentrated in RPE cells. A similar result was obtained in ARPE-19 cells (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Time course of verapamil uptake by RPE cells. RPE cells (RPE/Hu) were incubated with 1 µM [3H]verapamil at 37°C and its uptake was measured for periods up to 1 h. Data are expressed as the cell-to-medium ratio. Each data point represents the mean ± S.E. (n = 12).

Concentration Dependence. As shown in Fig. 2A, the initial rate of verapamil uptake (estimated at 1 min) by RPE/Hu cells was a saturable function of substrate concentration in the incubation medium over the range from 0.5 to 250 µM, i.e., a carrier-mediated process. The Eadie-Hofstee plot of these data yielded an apparent K_m of 7.2 ± 0.7 µM and a V_max of 726 ± 140 pmol/min/mg of protein. The passive permeability constant (P_dif) was 28.4 ± 7.4 µl/min/mg of protein. The K_m (3.0 ± 0.7 µM) and V_max (364 ± 45 pmol/min/mg of protein) values in ARPE-19 cells were similar to, but somewhat lower than, the values obtained in the RPE/Hu cells. P_dif in the ARPE-19 cell line (24.9 ± 1.8 µl/min/mg of protein) was almost identical to the value in RPE/Hu cells. As shown in Fig. 2B, at concentrations from 0 to 10 µM, the saturable component of uptake accounted for the bulk of the verapamil uptake by RPE cells; whereas, at higher concentrations diffusion played a much greater role in verapamil entry.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Concentration dependence of verapamil uptake by RPE cells. The initial rate of verapamil uptake (estimated at 1 min) by RPE/Hu cells at various concentrations (0.5-250 µM) was determined at 37°C. A, Eadie-Hofstee plot of these data. Kinetic parameters determined from this analysis were Km = 7.2 µM, Vmax = 726 pmol/min/mg of protein, and Pdif = 28.4 µl/min/mg of protein. The best fit line based on these parameters is shown. B, concentration of total, mediated, and diffusive uptake over the range from 0 to 50 µM. Each of the lines shown was calculated from the kinetic parameters determined from the Eadie-Hofstee analysis. Data are shown as the mean ± S.E. (n = 9).

Metabolic Inhibition. When the RPE/Hu cell line was pretreated with the metabolic inhibitors carbonylcyanide-p-(trifluoromethoxy)phenylhydrazone (2 µM), 2,4-dinitrophenol (1 mM), NaN₃ (10 mM), rotenone (30 µM), and HgCl₂ (100 µM) the initial rate of verapamil uptake (1 min) was significantly inhibited in every case (Fig. 3). Verapamil uptake was also reduced when the incubation temperature was decreased to 4°C.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of metabolic inhibitors and low temperature on verapamil uptake by RPE cells. RPE/Hu cells were preincubated with inhibitors for 20 min and the uptake of 1 µM [3H]verapamil was determined for 1 min in the presence or absence of inhibitors at 37°C. The effect of low temperature on verapamil uptake was determined at 4°C. Data are expressed as a percentage of control uptake in the absence of these inhibitors. The data are presented as mean ± S.E. (n = 6). *, significantly different from the control (p < 0.01).

Specificity. To determine the specificity of this transporter, we examined the effect of various potential inhibitors on the 1-min uptake of 1 µM [³H]verapamil by the RPE/Hu cell line (Fig. 4, A and B). All inhibitors were tested at a concentration of 500 µM, except verapamil itself (100 µM) and CSA (10 µM). The organic anion PAH had no effect on RPE cell verapamil uptake. Likewise, a number of organic cations, including tetraethylammonium (TEA), N¹-methylnicotinamide, carnitine, tubocurarine, tetrapentylammonium, cimetidine, and guanidine did not inhibit verapamil transport (Fig. 4A). However, uptake of verapamil was significantly (60-80%) inhibited by several organic cations, including quinacrine, pyrilamine, quinidine, and verapamil itself (Fig. 4, A and B). Since none of these inhibitors decreased verapamil uptake at 4°C, inhibition was primarily directed against transport of verapamil, rather than its binding to RPE cells (Fig. 4B). As shown in Fig. 4C, similar results were obtained in the ARPE-19 cell line. ARPE-19 verapamil uptake was insensitive to PAH, TEA, and cimetidine; whereas, verapamil itself, pyrilamine, and quinidine were effective inhibitors.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Specificity of verapamil uptake by RPE cells. A, RPE/Hu cells were incubated with 1 µM [3H]verapamil in the presence of organic cations or PAH for 1 min at 37°C. Data are expressed as a percentage of control uptake in the absence of inhibitor. Data represent the mean ± S.E. (n = 6). *, significantly different from the control (p < 0.01). B, RPE/Hu cells were incubated with 1 µM [3H]verapamil in the presence of pyrilamine, quinacrine, or quinidine. Conditions tested were 10 µM at 37°C (), 100 µM (250 µM for pyrilamine) at 37°C (), and 100 µM (250 µM for pyrilamine) at 4°C (). *, significantly different from the control (p < 0.01). Data are shown as mean ± S.E. (n = 6). C, ARPE-19 cells were incubated with 1 µM [3H]verapamil in the presence of organic cations or PAH for 1 min at 37°C. Data are expressed as a percentage of control uptake in the absence of inhibitor. Data represent the mean ± S.E. (n = 6). *, significantly different from the control (p < 0.01).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Effect of diphenhydramine on verapamil uptake by RPE cells. RPE/Hu cells were incubated with 1 µM [3H]verapamil in the presence of diphenhydramine (1-250 µM) at 37°C (), and 4°C (). Data are shown as mean ± S.E. (n = 6). All values at 4°C were significantly lower than at 37°C.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Inhibition of RPE cell verapamil uptake by antiglaucoma drugs. RPE/Hu cells () or ARPE-19 cells () were incubated with [3H]verapamil (1 µM) in the presence of 200 µM nifedipine, 100 µM diltiazem, 100 µM timolol, 100 µM propranolol, and 500 µM acetazolamide at 37°C. *, significantly different from the control (p < 0.01). Data points represent the mean ± S.E. (n = 6).

Potential Dependence. To investigate the influence of membrane potential (PD) on verapamil uptake, PD was altered by varying the extracellular potassium concentration: 0 mM (hyperpolarized), 5 mM (control), or 100 mM (depolarized). Microspectrofluorometry with the potential-sensitive dye 3,3'-dihexyloxacarbocyanine iodide demonstrated that altered medium K⁺ concentration caused the expected changes in PD, i.e., depolarization with high K⁺ and hyperpolarization with low K⁺ (data not shown). Verapamil uptake was not changed by these alterations in potassium concentration in the presence or absence of valinomycin (10.4 ± 0.5, 10.1 ± 1.1, 10.3 ± 0.7, and 11.4 ± 1.2, at 0, 5, 100 mM K⁺, and 100 mM K⁺ plus 5 µM valinomycin, respectively), indicating that verapamil uptake is not dependent on membrane potential.

Proton Dependence. As shown in Fig. 7A, uptake of verapamil was dependent on medium pH, being greatest at alkaline external pH, e.g., uptake at pH 8.4 was about 12 times larger than that at pH 5.4. The uptake of verapamil at each pH was inhibited by 200 µM quinidine. The quinidine-insensitive uptake of verapamil presumably reflects diffusive uptake of the nonionized free base. The quinidine-sensitive uptake, calculated from the difference in uptake in the absence and presence of quinidine, increased 14-fold from pH 5.4 to 8.4. The pH dependence of verapamil uptake in the absence and presence of quinidine was identical in ARPE-19 cells (data not shown).

View larger version (10K): [in this window] [in a new window]   Fig. 7.   pH dependence of verapamil uptake by RPE cells. A, cis-inhibition of verapamil uptake by external protons. Uptake of 1 µM [3H]verapamil (1 min) was determined from media of different pH in the absence () or presence () of 200 µM quinidine. B, trans-stimulation of verapamil uptake by intracellular acidification. RPE cells were preincubated for 15 min with 30 mM NH4Cl in Na+- and HCO3-free Ringer's solution and replaced by [3H]verapamil (1 µM) dissolved in fresh medium. Uptake was measured at 1 min. Data are shown as mean ± S.E. (n = 6). *, significantly different from the uptake at pH 5.4 (A) or control (B) (p < 0.01).

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Effect of medium pH on verapamil efflux from RPE cells. RPE/Hu cells were preloaded with 1 µM [3H]verapamil as described in Experimental Procedures for 60 min. Efflux was initiated by replacement of loading medium with 1 ml of fresh medium at pH 8.4, 7.4, 6.4, and 5.4. Efflux was measured after 1 min. Initial cellular content of [3H]verapamil was calculated as the efflux plus final cell content. Data are shown as mean ± S.E. (n = 9). *, significantly different from the efflux at pH 8.4 (p < 0.01).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Repeated topical administration of verapamil is recommended for management of ocular hypertension and glaucoma (Ettl et al., 1998). After topical administration, significant accumulation of verapamil was found in both aqueous and vitreous humors (Ettl et al., 1998; Siegner et al., 1998). Transport of verapamil by RPE cells in the retina-choroid direction is a potential route for elimination from these compartments. However, RPE cell transport of organic cations, like verapamil, has not been investigated. Nevertheless, it is well documented that positively charged drugs and xenobiotics are well transported by other epithelia, including kidney, liver, intestine, and choroid plexus (Pritchard and Miller, 1993; Koepsell, 1998). From these studies, at least three classes of organic cation transporters have been identified at the molecular level: 1) potential-driven carriers, including OCT1, OCT2, and OCT3, which transport organic cations across the basolateral membrane from the blood to the cells (Grundemann et al., 1994; Kekuda et al., 1998; Okuda et al., 1999; Sweet and Pritchard, 1999); 2) proton/organic cation antiporters, e.g., OCTN1 (Tamai et al., 1997); and 3) ATP-driven pumps, e.g., P-gp (Thiebaut et al., 1989; Cordon-Cardo et al., 1990).

Relationship of RPE Verapamil Transport to Known Organic Cation Transporters. Verapamil uptake by confluent monolayers of RPE cells was a saturable function of substrate concentration (Fig. 2). It depended upon metabolic energy, as indicated by the significant inhibition of uptake by metabolic inhibitors or lowered incubation temperature (Fig. 3). These findings indicate the involvement of a carrier-mediated system in the uptake process. To determine whether RPE verapamil transport could be accounted for by known organic cation transporters, the specificity of verapamil transport into RPE cells was assessed (Fig. 4). These results indicate that the specificity of verapamil transport is very different from the cloned organic cation transporters (OCT1, OCT2, OCT3, OCTN1, and OCTN2). For example, TEA, which is well transported by all five transporters (Grundemann et al., 1994; Gorboulev et al., 1997; Kekuda et al., 1998; Wu et al., 1998b; Sweet and Pritchard, 1999; Yabuuchi et al., 1999), did not inhibit RPE verapamil transport. Similarly, cimetidine, which inhibits these transporters (Kekuda et al., 1998; Urakami et al., 1998), did not change verapamil uptake by RPE cells. Furthermore, N¹-methylnicotinamide, guanidine, tetrapentylammonium, and carnitine, each of which is known to be transported by, or to inhibit, one or more of the cloned organic cation transporters (Gorboulev et al., 1997; Zhang et al., 1997; Kekuda et al., 1998; Tamai et al., 1998; Urakami et al., 1998; Okuda et al., 1999; Yabuuchi et al., 1999) did not inhibit verapamil transport by RPE cells, even at 500-fold higher concentration (Fig. 4). Together, these data indicate that RPE cell verapamil transport is mediated by a carrier distinct from the known organic cation transporters listed above. Nevertheless, there are some similarities between RPE cell verapamil transport and transport by some of these carriers. For example, verapamil, quinidine, and pyrilamine are substrates for OCTN1 (Yabuuchi et al., 1999). In addition, quinidine, an effective inhibitor of OCT1 and OCT2, also inhibited verapamil uptake by RPE cells (Fig. 4). The basis for these similarities must await molecular characterization of the RPE carrier. Transport properties of the cloned organic cation transporters and the verapamil transporter observed in the present study are summarized in Table 1.

Driving Forces for RPE Verapamil Transport. Organic cation transporters can be classified by their driving forces. For example, in the kidney, basolateral organic cation transport is driven by inside negative membrane potential; whereas, apical organic cation transport is driven by a pH gradient. To characterize the mechanism of verapamil transport in RPE cells, we examined the effects of membrane potential and pH on verapamil uptake by RPE cells. Verapamil transport was not changed by depolarization of the cells. These results are additional evidence that the RPE verapamil transporter is distinct from the basolateral organic cation carriers (OCT1, OCT2, and OCT3), all of which are dependent on membrane potential (Grundemann et al., 1994; Gorboulev et al., 1997; Kekuda et al., 1998). On the other hand, both total and quinidine-sensitive (mediated) verapamil uptake were strongly cis-inhibited by protons, i.e., by decreased external pH (Fig. 7A). Considering that the pKa value of verapamil is 8.6, the ionized fractions of verapamil at pH 8.4, 7.4, 6.4, and 5.4 were 55, 77, 90, and 96%, respectively. Thus, the decrease in the quinidine-sensitive uptake as external pH was lowered cannot be explained by a decrease in the concentration of ionized verapamil, and must reflect a direct inhibition of the verapamil uptake process. Such an effect could arise from allosteric modulation of transporter activity by protons, as previously shown for the cloned organic cation transporter OCT2 (Sweet and Pritchard, 1999). Alternatively, it could reflect a direct coupling of verapamil uptake to the proton gradient (i.e., verapamil/proton exchange). To distinguish between these two possibilities, the proton gradient was modified in two ways: by increasing intracellular proton concentration (Fig. 7B), or by examining the effect of external proton concentration on the efflux of verapamil (Fig. 8). Under both conditions, transport of verapamil was increased as the proton gradient across the membrane was increased, demonstrating that RPE cell verapamil transport is mediated by H⁺/verapamil antiport. In this regard, transport by RPE cells is similar to TEA transport via OCTN1, the only organic cation transporter thought to be a proton/organic cation exchanger (Yabuuchi et al., 1999). However, the specificity properties discussed above, most notably the inability of TEA to inhibit RPE cell verapamil transport, strongly argue that verapamil transport by RPE cells is not mediated by OCTN1.