Identification of a Novel Voltage-driven Organic Anion Transporter Present at Apical Membrane of Renal Proximal Tubule*

A novel transport protein with the properties of volt-age-driven organic anion transport was isolated from pig kidney cortex by expression cloning in Xenopus laevis oocytes. A cDNA library was constructed from size-fractionated poly(A) (cid:1) RNA and screened for p -amin-ohippurate (PAH) transport in high potassium medium. A 1856-base pair cDNA encoding a 467-amino acid peptide designated as OAT V 1 (voltage-driven organic anion transporter 1) was isolated. The predicted amino acid sequence of OAT V 1 exhibited 60–65% identity to those of human, rat, rabbit, and mouse sodium-dependent phosphate cotransporter type 1 (NPT1), although OAT V 1 did not transport phosphate. The homology of this transporter to known members of the organic anion transporter family (OAT family) was about 25–30%. OAT V 1- mediated PAH transport was affected by the changes in membrane potential. The transport was Na (cid:1) -independ-ent

In kidney, the secretion of organic anions takes place in the renal proximal tubular epithelial cells via at least two steps. The first is the transport of organic anions from the peritubular plasma across the basolateral membrane into the proximal tubular cells. p-Aminohippurate (PAH) 1 has been used as a prototypical substrate for the renal organic anion transport systems (1,2). The transport of PAH across the basolateral membrane of proximal tubular cells against the electrochemical gradient (3) occurs in exchange for intracellular dicarboxylates such as ␣-ketoglutarate (4,5). Two organic anion transporters, OAT1 (6 -8) and OAT3 (9), have been proposed to be responsible for this step. The second step is the exit of organic anions across the apical membrane at the tubular epithelial cells into the urine. Although this process is energetically downhill for organic anions, it has been believed that this process is also mediated by specific transporters. The luminal efflux system for PAH has been investigated mostly using brush border membrane vesicles. In dogs (10 -13) and rats (14), anion exchange mechanisms have been demonstrated for the luminal efflux systems. They mediate probenecid-sensitive electroneutral exchange of anionic compounds including both organic (e.g. PAH, urate, and lactate) and inorganic (e.g. Cl Ϫ , HCO Ϫ 3 , OH Ϫ ) anions. A distinct efflux system involving the voltage-driven transport was demonstrated in pig and rabbit (15)(16)(17). In bovine renal brush border membrane vesicles, a PAH/dicarboxylate exchange system was reported (18). PAH transport at the apical membrane in the OK kidney epithelial cell line was shown to be mediated by a voltage-driven transport system but not by an anion exchange system (19). In the physiological condition, a major component of the exit path of organic anions from renal proximal tubular cells has been proposed to be the facilitated diffusion along the electrochemical potential gradient (2,23). Voltage-driven organic anion transport plays an important role for this step. However, the molecular nature and precise functional properties of these efflux systems are still unknown.
MRP2, an ABC (ATP-binding cassette) transporter, was suggested to be one of the exit paths of PAH in proximal tubular cells. Studies of membrane vesicles from MRP2-expressing HEK or Sf9 cells have indicated that MRP2 transports PAH (20,21) in an ATP-dependent manner. Recently, it was shown that the human sodium-dependent phosphate cotransporter type 1 (NPT1) present at the apical membrane of proximal tubules transports organic anions including PAH. The PAH transport by NPT1 expressed in HEK293 was Cl Ϫ sensitive (22). Although NPT1 is a candidate of the exit path for PAH at the apical membrane of proximal tubules, the reported properties of NPT1, in particular its K ϩ -dependence, is not consistent with those of the voltage-driven PAH transporter at the apical membrane. In the present study, to identify the transporter responsible for the exit of PAH through the apical membrane, we have performed expression cloning using pig kidney cortex poly(A) ϩ RNA and identified a novel transporter present at the apical membrane of proximal tubules that mediates voltagedriven facilitated diffusion.

EXPERIMENTAL PROCEDURES
Expression of Pig Kidney Cortex Poly(A) ϩ RNA-Xenopus laevis oocyte expression studies and uptake measurements were performed as described elsewhere (24,25). Defolliculated oocytes were injected with poly(A) ϩ RNA (50 ng) obtained from kidney cortex of adult female pig (body weight 102 kg). Three days after injection and incubation of oocytes at 18°C, the uptake experiments were performed in high potassium uptake solution (98 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.4) containing 20 M [ 14 C]PAH (1 Ci/ml).
Preparation of Size-fractionated Pig Kidney Poly(A) ϩ RNA and Expression Cloning-Four hundred micrograms of pig kidney poly(A) ϩ RNA was size-fractionated using preparative gel electrophoresis (Bio-Rad, model 491 Prep cell) (6, 26 -28). The isolated poly(A) ϩ RNA fractions were examined for [ 14 C]PAH (20 M) transport activity by expression in X. laevis oocytes in the high potassium uptake solution. The positive fraction showing the greatest activity of [ 14 C]PAH uptake was used to construct a directional cDNA library (27)(28)(29). Screening of the cDNA library was performed as described previously (6,(27)(28)(29). cRNA synthesized in vitro from pools of ϳ500 clones was injected into Xenopus oocytes. A positive pool showing the [ 14 C]PAH uptake activity in the high potassium uptake solution was sequentially subdivided and analyzed until a single clone (OAT V 1) was identified. The cDNA was sequenced in both directions by the dye terminator cycle sequencing method using an ABI PRISM 3100 Genetic Analyzer (PE Biosystems).
Functional Characterization-Twenty-five nanograms of OAT V 1 cRNA synthesized in vitro using T7 RNA polymerase from the OAT V 1 cDNA in pSPORT1 linearized with NotI were injected into defolliculated Xenopus oocytes. Two or 3 days after injection the uptake of radiolabeled substrates was measured in various uptake solutions (differing in ion composition) to examine its dependence on sodium or chloride. In sodium-free solution, LiCl or choline-Cl were used to replace NaCl in the standard uptake solution (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.4). For the high potassium uptake solution, NaCl in the standard uptake solution was replaced by KCl. For Cl Ϫ -free medium, sodium gluconate, potassium gluconate, calcium gluconate, and MgSO 4 were used to replace NaCl, KCl, CaCl 2 , and MgCl 2 , respectively. To prepare the uptake solution at pH 5.5 and 6.5, MES was used instead of HEPES for the buffer system (30). The uptake measurements were performed for 60 min at room temperature (22-25°C). The radioactivity was counted by liquid scintillation spectrometry, and the values are expressed as femtomole per oocyte per min.
To examine the effects of the extracellular concentration of K ϩ and Cl Ϫ on the OAT V 1-mediated [ 14 C]PAH uptake, the concentrations of K ϩ or Cl Ϫ were varied over the range of 0 to 100 mM (0, 5, 10, 25, 50, and 100 mM). The concentrations of K ϩ in the uptake solution were varied in Na ϩ -free solution in which [KCl] plus [choline-Cl] was equal to 100 mM. The concentration of Cl Ϫ was varied in the high potassium uptake solution in which [KCl] plus [K-gluconate] was equal to 100 mM.
Kinetic Study-Because the uptake of substrates in high potassium uptake solution was linear longer than 2 h, the uptake was measured for 30 min for kinetic studies. The concentrations of PAH, urate, and estrone sulfate were varied from 10 M to 10 mM, from 10 M to 5 mM, and from 1 to 500 M, respectively. OAT V 1-mediated substrate uptake g of poly(A) ϩ RNA isolated from pig brain, lung, liver, kidney, small intestine, and skeletal muscle were loaded onto 1.0% agarose/formamide gel, transferred to a nylon membrane, and then probed with 32 Plabeled OAT V 1 cDNA fragment. b, Western blot analysis of OAT V 1 in pig kidney and brush border membrane (BBM) of the pig kidney cortex was performed using an anti-OAT V 1 antibody. The absorption test was performed by preincubation of the antibody with OAT V 1 peptides (100 g/ml) (Peptide (ϩ)). The uptake performed in the high potassium (K ϩ ) uptake solution was compared with that measured in the standard uptake solution (Na ϩ ).
was calculated as the difference between the values of uptake into cRNA-injected oocytes and those of control oocytes (without cRNA injection). The K m values were determined with the Eadie-Hofstee equation.
To measure the K i values for the transport, oocytes expressing OAT V 1 were incubated for 30 min in high potassium uptake solution with various concentrations of [ 14 C]PAH with or without addition of inhibitors. The K i values were determined by double-reciprocal plot analysis in which 1/uptake rate of [ 14 C]PAH was plotted against 1/[ 14 C]PAH concentration. The K i values were calculated from the following equation when competitive inhibition was observed: where K a , K t , and [I] are K m of PAH with inhibitor, K m of PAH without inhibitor, and concentration of inhibitor, respectively (31).
Efflux Measurement-Fifty nanoliters of [ 14 C]PAH (2.5 nCi, ϳ1 mM), [ 14 C]urate (2.5 nCi, ϳ1 mM), or [ 3 H]estrone sulfate (25 nCi, ϳ20 mM) was injected into oocytes with a fine-tipped glass micropipette as described elsewhere (32,33). The individual oocytes were incubated in ice-cold standard uptake solution (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.4) for 5 min and then transferred to the standard uptake solution kept at room temperature and incubated at room temperature for 2, 5, 15, 30, 60, 90, 120, or 150 min. Then, the incubation solution was collected to determine the efflux of substrates from the oocytes into bath solution at the end of the incubation period. The radioactivity remaining in the oocytes was also measured. The efflux value was expressed as % radioactivity calculated from the radioactivity in the bathing solution ϫ 100%/(the radioactivity in oocyte plus the radioactivity in bathing solution). The dependence of [ 14 C]PAH efflux on inorganic ions was determined by comparing the [ 14 C]PAH uptake in the standard uptake solution and that in the solutions in which NaCl was replaced by KCl, sodium gluconate, potassium gluconate, LiCl, or choline-Cl as mentioned above. The pH dependence of [ 14 C]PAH efflux was measured in the standard uptake solution with varied pH to 5.5, 6.5, 7.4, and 8.5. The efflux was measured for 15 min over which the linear efflux value was obtained. To examine the trans-stimulation of the efflux of PAH, [ 14 C]PAH-injected oocytes were incubated in the standard uptake solution with or without non-radiolabeled PAH (0.01, 0.1, 1, 5, or 10 mM) for 15 min at room temperature.
For the uptake and efflux measurements in the present study, 8 -10 oocytes were used for each data point. The values are expressed as mean Ϯ S.E. The reproducibility of the results was confirmed by three separate experiments with different batches of oocytes. The results from representative experiments are shown in figures.
Electrophysiological Measurements-The electrical currents induced by 5 mM PAH in both control and OAT V 1-cRNA-injected X. laevis oocytes were recorded during voltage clamp at Ϫ60 to ϩ40 mV in the standard uptake solution (24,34,35). At each step (20 mV interval) of voltage clamp, the baseline electrical current was first determined and then 5 mM PAH was added to the bath medium for 2 min. The solution was then changed back to the standard uptake solution and the record was continued until the electrical current was returned to the baseline level.
Anti-peptide Antibody-A rabbit polyclonal antibody against a keyhole limpet hemocyanin-conjugated synthetic peptide, EVQDWAKER-QNTYL, corresponding to 14 amino acid residues near the carboxyl terminus (454 -467 of the amino acid sequence) of OAT V 1 was generated and affinity purified as described elsewhere (36,37).
Western Blot Analysis-Pig kidney epithelial membranes were prepared as described previously (38) and subjected to SDS-polyacrylamide gel electrophoresis (39). The separated proteins were transferred electrophoretically to a Hybond-P polyvinylidene difluoride transfer membrane (Amersham Biosciences). The membranes were treated with diluted affinity purified anti-OAT V 1 antibody (1:5,000) overnight at 4°C. Thereafter, horseradish peroxidase-conjugated anti-rabbit IgG was used as the secondary antibody (Jackson ImmunoResearch Laboratories, Inc). The signals were detected using the ECL Plus system (Amersham Biosciences). The specificity of immunoreaction was confirmed by an absorption experiment in the presence of antigen peptide (100 g/ml).
Immunohistochemistry-Three-micrometer paraffin sections of pig kidney were processed for light microscopic immunohistochemical analysis as described previously (36). The kidney sections were incubated with affinity purified anti-OAT V 1 antibody (1:1,000) at 4°C overnight and treated with Envision (ϩ) rabbit peroxidase (DAKO) for 30 min. The immunoreactions were detected with diaminobenzidine (0.8 mM) (40). For absorption experiments, the serial kidney sections were treated with the primary antibody in the presence of antigen peptide (100 g/ml).

RESULTS
Screening of cDNA Library-Total poly(A) ϩ RNA was isolated from pig kidney cortex and size-fractionated. When expressed in Xenopus oocytes, the fractions corresponding to ϳ1.7 to 2.2 kilobases exhibited the peak activity of the uptake of 20 M [ 14 C]PAH in high potassium uptake solution: 1.5-2.0-fold of the uptake by control water-injected oocytes. These poly(A) ϩ RNA fractions were used as a template for construction of a cDNA library. The library was screened for [ 14 C]PAH uptake by expression in Xenopus oocytes to isolate a 1,856-base pair cDNA clone that encodes a protein designated as OAT V 1 (voltage-driven organic anion transporter 1). The OAT V 1 cDNA contains an open reading frame encoding a putative 467-amino acid protein. Twelve transmembrane domains were predicted on the OAT V 1 amino acid sequence based on SOSUI algorithm (41). The amino acid sequence of OAT V 1 exhibited the highest identity (60 -65%) to that of sodium-dependent phosphate cotransporter type 1 (NPT1) of each species (rat, mouse, rabbit, and human). The homology to amino acid sequences of sodiumdependent phosphate cotransporter type 2 (NPT2) and other OATs is ϳ25-30%.
Tissue Distribution-The Northern blot using the OAT V 1 cDNA fragment as a probe showed a strong signal in pig kidney and liver (Fig. 1a). The OAT V 1 transcript was not detected in brain, lung, small intestine, and skeletal muscle. The major transcript detected in liver and kidney was ϳ3.2 kb. In addition, a faint band at ϳ1.8 kb was also detected in both kidney and liver. The 4.4-kb faint band was detected only in liver.
In Western blot analysis, the antibody raised against the COOH terminus peptide of OAT V 1 recognized a band of 60 kDa in the membrane protein prepared from total pig kidney and the brush border membrane of pig kidney cortex (Fig. 1b). This band was consistent with a predicted molecular mass of OAT V 1 protein (51 kDa). The band disappeared in the presence of antigen peptide in the absorption test, confirming the specificity of the immunoreaction (Fig. 1b).
Transport Activity of OAT V 1-Because the luminal facilitated transport system in pig and rabbit was reported to be markedly dependent on membrane potential (15,16), we examined the effect of raising external K ϩ on the OAT V 1-mediated transport. The elevation of external K ϩ depolarizes the plasma membrane of Xenopus oocyte (42). As shown in Fig. 2 H]estradiol-17␤-glucuronide by OAT V 1 expressing oocytes were significantly higher in high potassium solution (98 mM K ϩ ) compared with those incubated in standard uptake solution (2 mM K ϩ ), whereas control oocytes showed no difference between these conditions, suggesting that OAT V 1-mediated transport is dependent on membrane voltage. The uptake of [ 14 C]PAH was saturable and followed Michaelis-Menten kinetics (Fig. 3a) with a K m value of 4.38 Ϯ 0.96 mM (mean Ϯ S.E. of three separate experiments) (Fig. 3b). K m values for [ 14 C]urate and [ 3 H]estrone sulfate were Ͼ5 and 0.212 mM, respectively (data not shown).
Characteristic of OAT V 1-mediated Organic Anion Transport-The PAH uptake by OAT V 1-expressing oocytes was also measured in various uptake solutions with different ionic contents. Replacement of Na ϩ with Li ϩ or choline ϩ did not affect [ 14 C]PAH uptake (Fig. 4a), indicating that OAT V 1-mediated [ 14 C]PAH transport was not dependent on Na ϩ in the uptake solution. The replacement of Na ϩ with K ϩ or Rb ϩ increased the OAT V 1-mediated transport (Figs. 2 and 4a). Fig. 4b further shows the dependence of OAT V 1-mediated [ 14 C]PAH uptake on K ϩ concentration, indicating that raising extracellular K ϩ increased PAH transport in a concentration-dependent manner. other anions. PAH uptake was higher in OAT V 1-expressing oocytes incubated in potassium gluconate solution than that in KCl uptake solution. PAH uptake was also higher in sodium gluconate, NaBr, or NaF solution than that in NaCl solution. As shown in Fig. 4c, PAH transport was dependent on Cl Ϫ concentration. In the chloride-free solution and the solution containing 5 mM chloride, OAT V 1 expressing oocytes showed a significant elevation of PAH uptake (Fig. 4c). The PAH uptake did not show any remarkable pH dependence within the pH range of 5.5 and 8.5 (Fig. 4d). We also found that urate transport by OAT V 1 showed the same properties as those of PAH transport (data not shown).
Voltage-dependent PAH Transport- Fig. 4 showed that PAH uptake by OAT V 1-expressing oocytes was affected by the changes of K ϩ or Cl Ϫ concentration in the uptake solution. It seemed that PAH uptake is dependent on membrane voltage change generated by the alteration of K ϩ and Cl Ϫ concentrations. To confirm the voltage dependence of PAH transport by OAT V 1, we performed electrophysiological measurement of PAH-induced currents under the voltage clamp condition. PAH application to the bath medium induced outward current in the oocytes expressing OAT V 1, whereas no significant current was elicited by 5 mM PAH in the control oocytes. The PAH-induced current was dependent on membrane voltage (Fig. 5). The electric currents induced by 5 mM PAH were increased when the holding potential was elevated from Ϫ60 to ϩ40 mV as shown in Fig. 5.
Characteristic of OAT V 1-mediated PAH Efflux-According to the voltage-driven transport of PAH, in the physiological condition PAH is supposed to be transported from the intracellular compartment to the extracellular compartment along the electrical gradient. As shown in Fig. 6a, OAT V 1expressing oocytes preloaded with [ 14 C]PAH showed time-dependent efflux of radioactivity when incubated in the standard uptake solution. The voltage dependence of PAH transport was also observed for the efflux experiments. The OAT V 1-expressing oocytes showed a reduction of PAH efflux from the oocytes to the extracellular medium when the oocytes were incubated in high potassium uptake solution, which generates inside positive potential (Fig. 6b). The effect of pH on [ 14 C]PAH efflux was examined in the standard uptake solution with varied pH. The efflux of [ 14 C]PAH from the oocytes expressing OAT V 1 was dependent on extracellular pH. [ 14 C]PAH efflux was elevated by incubating the oocytes in the uptake solution with higher pH, whereas it was reduced by lowering the pH of the uptake solution (Fig. 6c).
To determine whether OAT V 1 is an exchanger or a facilitated transporter, the efflux of radioactivity from the oocytes preloaded with [ 14 C]PAH was compared in the absence and presence of extracellular PAH. The efflux of radioactivity was not affected by the extracellularly applied PAH (Fig. 6d) 1 and 1 mM). The PAH uptake was highly inhibited by estrone sulfate, probenecid, 5-nitro-2-(3-phenylpropylamino)benzoate, bumetanide, and furosemide but not by D-and L-lactate, glutarate, and tetraethylammonium (Fig. 7a). Some sulfate and glucuronide conjugates also inhibited OAT V 1-mediated [ 14 C]PAH uptake. The inhibitory effect on [ 14 C]PAH uptake was stronger for sulfate conjugates than that for glucuronide conjugates (Fig. 5,  b and c). The inhibition of [ 14 C]PAH uptake by unlabeled PAH was shown to be competitive in a double reciprocal plot analysis with a K i value of 0.83 mM. Urate and estrone sulfate also competitively inhibited PAH uptake with K i values of 1.87 mM and 17.13 M, respectively (data not shown). Because OAT V 1 exhibited sequence similarity to NPT1 that was reported to transport both phosphate and organic anions, we examined whether OAT V 1 transports [ 32 P]phosphate. In the condition in which we detected the [ 32 P]phosphate (100 M) uptake by rat NaP i type IIa, Na ϩ /phosphate transporter, and OAT V 1-expressing oocytes did not show a detectable amount of [ 32 P]phosphate uptake (data not shown).
Immunohistochemical Analysis-The immunohistochemical analysis of pig kidney revealed that OAT V 1 protein was localized in the apical membrane of proximal tubules. The immunoreactivity was detected only in the cortex but not in the medulla portion of pig kidney (Fig. 8). The immunostaining at the apical membrane of proximal tubule completely disappeared in the absorption experiments in which the tissue sections were treated with primary antibody in the presence of antigen peptide (Fig. 8c), confirming the specificity of immunoreactions. DISCUSSION In the present study, we identified a novel protein, OAT V 1, from pig kidney cortex. The amino acid sequence of OAT V 1 showed the highest similarity to human sodium/phosphate co- 5. Electrophysiological measurement. The current responses of OAT V 1-expressing oocytes to bath-applied 5 mM PAH. Membrane current was measured in the standard bath solution at steps of holding potential (Ϫ60, Ϫ40, Ϫ20, 0, ϩ20, ϩ40 mV). Five millimolar PAH was applied at each time point after changing the holding potential. The step changes of holding potential were performed after the current was returned to each baseline value. A representative current trace is shown in the upper panel (a). The current-voltage relationships of PAH-induced current were compared between OAT V 1-expressing oocytes and control oocytes (b).
transporter type 1 (human NPT1) that was previously reported to transport phosphate and organic anions. When expressed in Xenopus oocytes, OAT V 1 exhibited the character of a multispecific organic anion transporter. This transporter mediated the transport of PAH, urate, and estrone sulfate (Fig. 2). Various organic anions but not organic cations interacted with PAH transport by OAT V 1 (Fig. 7). We named this transporter OAT V 1 (voltage-driven organic anion transporter 1) because it mediates a facilitated diffusion driven by membrane voltage. The PAH transport was shown to be sodium-independent and chloride-sensitive, as reported for human NPT1 expressed in HEK293 cells (22). In contrast to human NPT1 (43), OAT V 1 expressed in Xenopus oocytes did not show the uptake of phosphate (data not shown). OAT V 1, furthermore, differed from human NPT1 in the dependence on pH and K ϩ . The PAH uptake by human NPT1 expressed in HEK293 cells was increased by lowering extracellular pH, whereas OAT V 1-mediated PAH uptake was not affected by changing the extracellular pH. In contrast to OAT V 1, the response to high K ϩ medium was not observed for human NPT1. Because of these differences in the functional properties of the transporters, we could not, at this moment, conclude whether OAT V 1 is an orthologue or paralogue of human NPT1. Furthermore, even though OAT V 1 exhibited similar substrate selectivity to that of OAT family members, it could not be classified into the OAT family because of its low sequence similarity (25-30%) to members of the OAT family.
Although OAT V 1-mediated PAH transport was Na ϩ -independent, it was affected by altering extracellular K ϩ and Cl Ϫ (Fig. 4). When the extracellular concentration of K ϩ was increased or that of Cl Ϫ was decreased, the PAH uptake mediated by OAT V 1 was elevated. This suggests that the PAH transport was altered by the action of membrane voltage. Based on the Goldman-Hodgkin and Katz equation, , the increase in extracellular K ϩ or decrease in extracellular Cl Ϫ results in the elevation of membrane potential (toward the positive direction). This membrane depolarization is proposed to promote the uptake of negatively charged PAH via OAT V 1 once the net charge of the fully loaded carrier is negative (24). Moreover, the results from electrophysiological measurements showed that the membrane depolarization increased the outward current associated with the PAH uptake into the oocytes expressing OAT V 1 (Fig. 5), supporting the idea that the PAH transport by OAT V 1 is a voltage-driven transport.
To prove that the transport via OAT V 1 is a facilitated transport of organic anions, the properties of the efflux of PAH mediated by OAT V 1 was examined. It was shown in the present study that the efflux of PAH was not stimulated by the application of PAH in the bath solution (extracellular medium), suggesting that PAH transport via OAT V 1 is not trans-stimulated. Therefore, together with the results from electrophysiological measurements demonstrating the voltage dependence of the outward current associated with PAH transport (Fig. 5), it is suggested that OAT V 1-mediated transport involves electrogenic facilitated diffusion driven by membrane voltage, rather than the exchange of substrate organic anions.
We found that PAH efflux by OAT V 1 expressed in oocytes was increased by elevating extracellular pH and decreased by lowering extracellular pH (Fig. 6c). This may be consistent with the idea of PAH/OH Ϫ exchange that has been proposed for the PAH exit path through the apical membrane of renal proximal tubules (12,13). We attempted to clarify whether OAT V 1 me-diates the PAH/OH Ϫ exchange by incubating the oocytes expressing OAT V 1 in the medium containing sodium acetate or NH 4 Cl that causes intracellular acidification (44 -46). PAH uptake was, however, not altered significantly by these maneuvers (data not shown). There are two possible explanations for this observation; the first is that OAT V 1 exclusively mediates facilitated transport and a PAH/OH Ϫ exchange mode is not involved for OAT V 1. The second is that the voltage-driven PAH transport is much more predominant than the PAH/OH Ϫ exchange mode, so that it masked the PAH/OH Ϫ exchange mode in the experimental conditions. Further investigation is necessary to understand the action of extracellular pH on PAH efflux via OAT V 1.
OAT V 1 mediated the transports of PAH, urate, sulfate, and glucuronide conjugates of steroid hormones in a sodium-independent manner. In addition, OAT V 1 interacted with various anionic compounds such as nonsteroidal anti-inflammatory drugs and diuretics. Therefore, OAT V 1 is a multispecific organic anion transporter, overlapping substrate selectivity with members of the OAT family. Most of the substrates of OAT1 (6, 7, 47) except ␣-ketoglutarate and glutarate interacted with OAT V 1. In addition, the substrates of OAT3 (48) and OAT4 (49) mostly interacted with OAT V 1. It is interesting to know what structural trait is responsible for such similar substrate selectivity among these structurally distinct organic anion transporters. For OAT1, the presence of a negatively charged moiety and an appropriately sized hydrophobic core is proposed to be an important requirement to be a substrate (50,51). Such a simple requirement would also be applicable to OAT V 1. Consistent with this, sulfate conjugates inhibited OAT V 1-mediated uptake, whereas Na 2 SO 4 did not show any effects (Fig. 7). The presence of the anionic moiety and the backbone hydrophobic core is proposed to be an important requirement to be accepted by the binding site of OAT V 1.
Based on the voltage-driven facilitated transport mechanisms of OAT V 1 and its localization at the apical membrane of proximal tubules, OAT V 1 is presumed to be well suited as an exit path for organic anions via the apical membrane of renal proximal tubules. Organic anions would be transported across apical membrane via OAT V 1 along the electrochemical potential gradient. In fact, it is reported that PAH and organic anions driven by membrane voltage are secreted from the epithelial cells into the proximal tubule lumen. OAT V 1 is proposed to play an important role for this process. It is well known that PAH and many endogenous and exogenous organic anions are taken up from the blood across the basolateral membrane into proximal tubule cells via organic anion trans-porters OAT1 and OAT3 in exchange for intracellular dicarboxylates (6 -9). OAT1 is highly expressed at the basolateral side of the S2 segment of the proximal tubule (40), whereas OAT3 is expressed throughout the proximal tubule (52). In this study, OAT V 1 was shown to be expressed in the apical membrane of all the segments of the proximal tubule. Therefore, we propose that organic anions are taken up by the epithelial cells of renal proximal tubules via OAT1 and OAT3 and then exit from the epithelial cell via OAT V 1. Large organic anions with bulky moieties that are not substrates of OAT V 1 may be transported out of the epithelial cells via MRP2, an ABC transporter, using energy from ATP hydrolysis. It was previously reported that the secretion of PAH via the apical membrane of the proximal tubule might also be mediated by a PAH/glutarate (18) or PAH/OH Ϫ exchanger (12,13). Therefore, in addition to OAT V 1, other unidentified transporters may also participate in the excretion of organic anions through the epithelium of the renal proximal tubules.
We have identified a novel Na ϩ -independent voltage-driven organic anion transporter, OAT V 1. It localizes at the apical membrane of the proximal tubule. This transporter might play an important role in the renal excretion of drugs, xenobiotics, and their metabolites in the form of organic anions.