INTRODUCTION

The kidney plays an essential role in the elimination of numerous organic anions including endogenous compounds, xenobiotics, and their metabolites (1-3). The proximal tubule cells actively secrete them into the urine. The first step of this secretion is the extraction of organic anion from the peritubular blood plasma by the proximal tubule cells through the basolateral membrane. This basolateral uptake of organic anions has been extensively investigated using para-aminohippuric acid (PAH)¹ as a test substrate. The most striking feature of this organic anion transport system is its extremely wide substrate selectivity, covering not only endogenous anionic substrates but also a number of clinically important drugs (1, 3). Because of its importance in renal physiology and pharmacology, cloning of the organic anion transporter has been attempted by many investigators using different approaches; however, the molecular structure of the responsible transporter has not yet determined.

For the last decade it has been proposed that the basolateral uptake of organic anion is mediated by organic anion/dicarboxylate exchanger (1, 4). According to this model, outwardly directed dicarboxylate gradient is essential to express the transport activity of this exchanger. In the present study, we isolated first rat sodium dicarboxylate transporter (rNaDC-1) and then co-expressed it together with rat kidney poly(A)⁺ RNA in Xenopus oocytes to energize organic anion transport in oocytes. We describe here functional expression cloning of an organic anion/dicarboxylate exchanger (OAT1) and its characteristics as a multispecific organic anion transporter.

EXPERIMENTAL PROCEDURES

A nondirectional cDNA library was prepared from rat kidney poly(A)⁺ RNA using Superscript Choice system (Life Technologies, Inc.) and was ligated to ZipLox EcoRI arms (Life Technologies, Inc.). A polymerase chain reaction product corresponding to nucleotides 1323-1763 of the rabbit sodium dicarboxylate transporter (NaDC-1) (5) was labeled with [³²P]dCTP. A rat cDNA library was screened with this probe at low stringency. Hybridization was done overnight in the hybridization solution at 37 °C, and filters were washed finally at 37 °C in 0.1 × SSC, 0.1% SDS. The hybridization solution contains 5 × SSC, 3 × Denhardt's solution, 0.2% SDS, 10% dextran sulfate, 50% formamide, 0.01% Antifoam B, 0.2 mg/ml denatured salmon sperm DNA, 2.5 mM sodium pyrophosphate, and 25 mM MES, pH 6.5. cDNA inserts in positive ZipLox phage were recovered in plasmid pZL1 by in vitro excision and further subcloned into pBluescript II SK (Stratagene) for sequencing and in vitro transcription.

Xenopus laevis oocyte expression studies and uptake measurements were performed as described elsewhere (6). Defolliculated oocytes were injected with in vitro transcribed cRNA and/or rat kidney poly(A)⁺ RNA as indicated in each experiment. In vitro transcription was done by using T7 RNA polymerase in the presence of cap analog. After incubation of oocytes at 18 °C for 2-3 days, uptake studies were performed in sodium uptake solution (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 5 mM HEPES, pH 7.4) containing radiolabeled substrates as indicated in each experiment.

Four-hundred µg of rat kidney poly(A)⁺ RNA was size-fractionated as described elsewhere (6) using preparative gel electrophoresis (Bio-Rad, model 491 Prep cell). Then we co-injected poly(A)⁺ RNAs of each fraction together with rNaDC-1 cRNA into oocytes. Before the uptake study, the oocytes were routinely preincubated for 2 h in sodium uptake solution containing 1 mM glutarate. [¹⁴C]PAH (50 µM) uptake was measured in sodium uptake solution without glutarate for 1 h. A directional cDNA library was constructed from fractions showing the highest rate of [¹⁴C]PAH uptake (1.8-2.4-kilobase (kb) poly(A)⁺ RNA) using Superscript Plasmid system (Life Technologies, Inc.) and was ligated into the SalI and NotI site of pSPORT 1. Recombinants were electroporated into Electro Max DH10B competent cells (Life Technologies, Inc.). Approximately 500 colonies were grown on nitrocellulose membrane. Plasmid DNA was purified from colonies of each plate. Capped cRNA was synthesized in vitro after linearization of each plasmid DNA with NotI. Then we co-injected cRNA synthesized from each filter together with 2 ng of rNaDC-1 cRNA into oocytes. When [¹⁴C]PAH uptake was detected on a particular group, it was subdivided into several groups and further screened. Eight-thousand clones were screened until a single clone was isolated.

Deleted clones obtained by Kilo-Sequence deletion kit (Takara, Japan), or specially synthesized oligonucleotide primers were used for sequencing of rNaDC1 and OAT1 cDNA. rNaDC1 and OAT1 were sequenced by dideoxytermination method using Sequenase version 2.0 (Amersham Corp.) or dye primer cycle sequencing kit (Applied Biosystems).

The functional characterization of OAT1 was analyzed without co-expression of rNaDC-1 except for the experiment of which the results are shown in Fig. 1c. In experiments of Fig. 1, d and e, and Fig. 4, a and b, oocytes expressed with OAT1 were preincubated with 1 mM glutarate for 2 h before uptake studies were performed. For the determination of the sodium dependence of OAT1 transport activity, [¹⁴C]PAH uptake was measured either in sodium uptake solution or in choline chloride uptake solution (96 mM NaCl of sodium solution was replaced by 96 mM choline chloride, and pH was adjusted to 7.4 with Tris). For kinetic analysis, [¹⁴C]PAH influx was measured for 3 min. For inhibition study, 2 µM [¹⁴C]PAH uptake via OAT1 was measured in the absence or presence of 2 mM nonradiolabeled compounds in sodium uptake solution.

Three µg of poly(A)⁺ RNA prepared from various rat tissues were electrophoresed on a 1% agarose/formaldehyde gel and transferred to a nitrocellulose filter. The filter was hybridized at 42 °C overnight in the hybridization solution with full-length OAT1 cDNA labeled with [³²P]dCTP. The filter was washed finally in 0.1 × SSC, 0.1% SDS at 65 °C.

In situ hybridization was performed as described elsewhere (7) with some modifications. Briefly, after perfusion fixation with 4% paraformaldehyde, rat kidney was excised and postfixed in 4% paraformaldehyde. Five-µm cryostat sections of rat kidney were used in situ hybridization. ³⁵S-Labeled sense and antisense cRNA were synthesized from the full-length OAT1 cDNA (in pBlueScript SK) using T7 or T3 RNA polymerase after linearization of plasmid DNA with SpeI or XhoI, respectively. RNA probe was degraded by partial hydrolysis for 45 min. The cryosections were hybridized with the probe overnight in the hybridization solution and washed to a final stringency of 0.1 × SSC at 37 °C for 30 min.

RESULTS

Using an approach based on homology to rabbit sodium dicarboxylate transporter (NaDC-1) (5), we isolated rNaDC-1 (rat sodium-dependent dicarboxylate transporter-1: accession number for the nucleotide sequence of cDNA: AB001321; DDBJ, EBI, and GenBank^TM nucleotide sequence data bases). When expressed in X. laevis oocytes, rNaDC-1 mediates sodium-dependent [¹⁴C]glutarate uptake (Fig. 1a). Then we co-injected X. laevis oocytes with rat kidney poly(A)⁺ RNA and cRNA of rNaDC-1 and preincubated them with 1 mM glutarate to induce generation of an outwardly directed dicarboxylate gradient. These "co-expressed" oocytes exhibited [¹⁴C]PAH uptake, whereas oocytes injected with only rat kidney poly(A)⁺ RNA exhibited no detectable [¹⁴C]PAH uptake (Fig. 1b). Using a functional expression method together with this co-expression system, we isolated a 2294-base pair cDNA encoding an organic anion/dicarboxylate exchanger (OAT1: organic anion transporter 1).

Fig. 1c shows the dependence of OAT1-mediated [¹⁴C]PAH uptake on the intracellular dicarboxylate (glutarate) concentration. The rate of [¹⁴C]PAH uptake by oocytes via OAT1 is increased by preincubation of the oocytes with 1 mM glutarate. When oocytes co-expressing rNaDC-1 and OAT1 are preincubated with glutarate, they showed a further increase in the rate of [¹⁴C]PAH uptake. This trans-stimulative effect of glutarate indicates that OAT1 is an organic anion/dicarboxylate exchanger. As shown in Fig. 1d, replacement of extracellular sodium with choline had no effect on the rate of [¹⁴C]PAH uptake. Because [¹⁴C]PAH (1 mM) uptake via OAT1 was linearly increased until 5 min under this condition (data not shown), we performed the kinetic study for 3 min. [¹⁴C]PAH uptake via OAT1 follows Michaelis-Menten kinetics (Fig. 1e), and the estimated K_m value (14.3 ± 2.9 µM: mean ± S.E., n = 3) is similar to that previously reported for the basolateral organic anion transport system (80 µM) (2).

OAT1 cDNA consists of 2294 nucleotides, and contains an open reading frame encoding a 551-amino acid residue protein (Fig. 2a). Kyte-Doolittle hydropathy analysis (8) of OAT1 predicts 12 putative membrane-spanning domains (Fig. 2b). Four N-glycosylation sites are predicted in the first hydrophilic loop. There are four putative protein kinase C-dependent phosphorylation sites in the hydrophilic loop between sixth and seventh transmembrane domains.

Under high stringency Northern blot analysis of poly(A)⁺ RNA from rat various tissues, a strong 2.4-kb mRNA band and two bands corresponding to longer transcripts (3.9 and 4.2 kb) were detected predominantly in the kidney (Fig. 3a). Upon longer exposure, a faint 2.4-kb mRNA band was detected in the brain. No hybridization signals were obtained with mRNA isolated from other tissues. In the kidney, expression of OAT1 mRNA is strong in the cortex and outer medulla (cortex > outer medulla) and very weak in the inner medulla.

In situ hybridization of rat kidney coronal sections (Fig. 3b) revealed that OAT1 mRNA is expressed in renal cortex and outer medulla, especially in the medullary rays of the cortex. Expression of OAT-1 was not found in the inner medulla. No significant hybridization signal was detected in control experiments using sense OAT1 cRNA as a probe (data not shown). This overall pattern of in situ hybridization suggests that OAT1 is most strongly expressed in the middle portion of the proximal tubule (S2). This intrarenal localization is in good agreement with the previous studies, which demonstrate that the highest PAH transport activity were seen in S2 of rat proximal tubule (1).

To evaluate the substrate selectivity of OAT1, we examined the levels of the inhibition of OAT1-mediated [¹⁴C]PAH uptake by various compounds (Fig. 4a). cis-Inhibitory effects were observed for structurally unrelated drugs. Cephaloridine (-lactam antibiotic), nalidixic acid ("old" quinolone), furosemide and ethacrynic acid (diuretics), indomethacin (nonsteroidal anti-inflammatory drug), probenecid (uricosuric drug), and valproic acid (antiepileptic drug) potently inhibited (>85%) OAT1-mediated [¹⁴C]PAH uptake by the oocytes. An antineoplastic drug, methotrexate, moderately inhibited the [¹⁴C]PAH uptake. Endogenous compounds, such as prostaglandin E₂, cyclic AMP, cyclic GMP, and uric acid also inhibited [¹⁴C]PAH uptake. We examined several radiolabeled compounds in terms of whether they are taken up into oocytes via OAT1. As Fig. 4b shows, [³H]methotrexate, [³H]cAMP, [³H]cGMP, [³H]prostaglandin E₂, [¹⁴C]urate and [¹⁴C]-ketoglutarate were transported into the oocytes. No uptake of [¹⁴C]tetraethylammonium and [³H]taurocholic acid were detected (data not shown).

DISCUSSION

The present study describes the isolation of cDNA encoding a rat renal organic anion transporter, OAT1. The results indicate that OAT1 possesses the same characteristics as the predicted organic anion/dicarboxylate exchanger responsible for multispecific organic anion transport at the basolateral membrane of renal proximal tubules.

The anion exchange model of the renal organic anion transporter has been proposed within the last decade (1, 4). According to this model, organic anions are transported into the cell in exchange for intracellular dicarboxylates, which are subsequently returned into the cell via sodium-dependent dicarboxylate transporter. In the present report, we directly demonstrate the validity of this exchange model. Accumulated glutarate via sodium-dependent dicarboxylate transporter (rNaDC-1) stimulates [¹⁴C]PAH uptake via anion exchanger (OAT1) in Xenopus oocytes. These data explain indirect sodium coupling of renal organic anion transporter.

Unexpectedly, a search of EBI/GenBank^TM data base (January 31, 1997) revealed that the amino acid sequence of OAT1 shows weak (38%) identity to organic "cation" transporter (OCT1) (9). The sequence of OAT1 shows no significant similarity with the members of the inorganic anion exchanger (AE) family. OCT1 is considered to be a facilitated transporter, and substrates transported by OCT1 are different from those of OAT1. Although OAT1 and OCT1 transport substrates having opposite charges, there is a common feature in which these two transporters interact with molecules possessing a certain size of hydrophobic cores. This common aspect may be relevant to the weak homology between OAT1 and OCT1. Structure function analysis of OAT1 and OCT1 may provide some clues as to the sites of proteins that recognize the charge(s) of the substrates.

To date, at least two multispecific transporter families were identified. One is the ABC superfamily, which includes P-glycoprotein (10), a multidrug resistance associated protein (11, 12) and a canalicular multispecific organic anion transporter (cMOAT) (13). P-glycoprotein and multidrug resistance associated protein extrude a range of anticancer drugs from cells via ATP hydrolysis and confer multidrug resistance on cancer cells. cMOAT is considered to mediate hepatobiliary excretion of organic anions; however, its functional characteristics have not yet been analyzed in detail. Another is the oatp (organic anion transporting polypeptide) superfamily, which includes a Na⁺-independent oatp isolated from rat liver (14), prostaglandin transporter (15), and kidney-specific oatp (OAT-K1) (16). oatp mediates transport of bile acids, bromosulfophthalein, and conjugated and unconjugated steroid hormones (14, 17). We found no significant sequence similarity between OAT1 and other multispecific organic anion transporters, including cMOAT, oatp, and prostaglandin transporter. The structural and transport characteristics of OAT1 are distinct from those of the members of these two multispecific transporter superfamilies. Further investigation is required for clarification of the individual roles of the members of these three multispecific transporter superfamilies in drug delivery to and elimination from the body.

Similar to the case in the kidney, anionic substrates should be transported in other organs, including liver and brain. In high stringency Northern blot analysis, however, no signals were detected except for the kidney and the brain. The liver plays a key role in the detoxification of many endogenous and exogenous compounds. Lipophilic xenobiotics are transported into hepatocytes and extensively biotransformed by P450- and transferase-mediated reactions. As a result, negatively charged amphiphilic compounds are produced, which must be secreted from the hepatocytes. Thus, the liver needs multispecific transporter(s) of anionic compounds (18, 19). cMOAT and oatp may play certain roles in organic anion transport in the liver; however, the multiplicity of drug transport has never been fully explained (20, 21). Recently, organic anion/dicarboxylate exchange systems in the liver and brain have been identified (22, 23), and a search of a data base revealed the existence of a liver protein that is structurally related to OAT1 but whose function has not been determined (24). Isoform(s) of OAT1 serving as organic anion transporters possibly exist in several organs. Identification of isoforms of OAT1 in other organs will facilitate the understanding of drug delivery and excretion system of the body and should provide useful tools for developing drugs that show desirable distribution in the body.

The original role of OAT1 is presumed to be mediation of transport of endogenous anionic compounds, such as cyclic nucleotides, prostaglandin E₂, and uric acid. A variety of anionic xenobiotics, such as drugs and food additives, are newly synthesized and administered to the body, which must be excreted. Because of its wide substrate selectivity, OAT1 fits the role of a xenobiotic transporter and can serve as an essential multispecific anionic drug transporter.

Practically, expression of OAT1 in epithelial cell lines is expected to be useful for the development of in vitro assay systems for drug elimination and drug/drug interaction studies. Furthermore, such systems could be used for screening of drugs for nephrotoxicity, which is one of the most critical points in drug administration to the body. Many drugs that can cause acute renal failure, such as -lactam antibiotics, diuretics, and nonsteroidal anti-inflammatory drugs (25), are transported by the organic anion transport system. One reason why such drugs are nephrotoxic may be related to their accumulation via OAT1. Isolation of OAT1, therefore, will facilitate elucidation of the molecular basis of pharmacokinetics and toxicokinetics.