Basolateral uptake of organic anions in renal proximal tubule cells is indirectly coupled to the Na+ gradient through Na+-dicarboxylate cotransport and organic anion/dicarboxylate exchange. One member of the organic anion transporter (OAT) family, Oat1, is expressed in the proximal tubule and is an organic anion/dicarboxylate exchanger. However, a second organic anion carrier, Oat3, is also highly expressed in the renal proximal tubule, but its mechanism is unclear. Thus we have assessed Oat3 function in Xenopus laevis oocytes and rat renal cortical slices. Probenecid-sensitive uptake ofp-aminohippurate (PAH, an Oat1 and Oat3 substrate) and estrone sulfate (ES, an Oat3 substrate) in rat Oat3-expressing oocytes was significantly trans-stimulated by preloading the oocytes with the dicarboxylate glutarate (GA). GA stimulation of ES transport by oocytes coexpressing rabbit Na+-dicarboxylate cotransporter 1 and rat Oat3 was significantly inhibited when the preloading medium contained Li+ or methylsuccinate (MS) or when Na+ was absent. All these treatments inhibit the Na+-dicarboxylate cotransporter, but not rat Oat3. Li+, MS, and Na+ removal had no effect when applied during the ES uptake step, rather than during the GA preloading step. Concentrative ES uptake in rat renal cortical slices was also demonstrated to be probenecid and Na+ sensitive. Accumulation of ES was stimulated by GA, and this stimulation was completely blocked by probenecid, Li+, MS, taurocholate, and removal of Na+. Thus Oat3 functions as an organic anion/dicarboxylate exchanger that couples organic anion uptake indirectly to the Na+ gradient.
- proximal tubule
- estrone sulfate
active excretory transport by the kidney is an important determinant of the effects of therapeutics and toxic chemicals. Such transport is mediated by multiple organic cation and organic anion transporters (OATs), primarily in the proximal segment of the nephron. The first step in renal organic anion secretion, i.e., uptake at the basolateral membrane of renal proximal tubule cells, is mediated, uphill, and Na+ dependent (for a review, see Refs.38 and 47). Mechanistically, this step is driven by indirect coupling to the Na+ gradient through Na+-dicarboxylate cotransport and organic anion/dicarboxylate exchange (tertiary active transport; Fig.1) (34, 36, 37, 42). There is some uncertainty about the nature of the final step in organic anion secretion, i.e., efflux across the apical membrane.
Oat1 was the first renal OAT to be cloned (25, 41, 49). Not only was Oat1 localized to the basolateral membrane of renal proximal tubules (45, 51), but also it was shown to support organic anion/dicarboxylate exchange (49). Thus it was assumed that Oat1 was indeed the transporter that drove the classic, Na+-dependent renal organic anion system. However, basolateral organic anion transport appears to be considerably more complex and may also be driven by one of the other OATs that have been cloned recently and partially characterized: Oat2, Oat3, and Oat4 (10, 23, 40, 43). Together with Oat1, they comprise a subfamily of the amphiphilic solute transporter family (Slc22a) within the major facilitator superfamily (7,44, 47, 52).
The mechanisms by which Oat2, Oat3, and Oat4 couple transport to cellular metabolism have not been defined. This lack of understanding is particularly important for Oat3, inasmuch as it is the most highly expressed OAT in kidney and choroid plexus (28, 46). Recent immunolocalization studies detected the rat and human Oat3 orthologs (rOat and hOAT, respectively) in the basolateral membrane of proximal tubules (9, 22, 28). Functional analysis of an Oat3 knockout mouse model confirmed that murine Oat3 was present in the basolateral membrane of proximal tubule cells and in the apical membrane of choroid plexus epithelial cells (i.e., the correct locations to mediate concentrative organic anion uptake) and that the transporter mediates a substantial fraction of organic anion uptake across the basolateral membrane of the renal proximal tubule and at the apical membrane of choroid plexus (46). When Oat3 was initially cloned and expressed in Xenopus laevis oocytes, experiments indicated that Oat3-mediated organic anion uptake was neither coupled to the Na+ gradient nortrans-stimulated by dicarboxylates (9, 23). No mechanism explaining how Oat3 could mediate uphill OA transport was demonstrated, and it was argued that Oat3 is not energetically coupled to metabolism but, rather, is simply a facilitated-diffusion carrier (9, 23). However, facilitated diffusion cannot drive organic anion uptake in the face of opposing electrical and concentration gradients, as required in the renal proximal tubule and choroid plexus.
We recently found reduced uptake of p-aminohippurate (PAH), the model substrate for the classic renal OAT system, in renal cortical slices from Oat3 knockout mice (46). Because in every species tested, from mammals to invertebrates, PAH uptake by renal tissue is concentrative and almost entirely Na+ dependent, these results argued for a reassessment of the mechanism of Oat3-mediated transport. In the present study, we have assessed the energetics of Oat3 using X. laevis oocytes expressing rOat3 and rat renal cortical slices. Our experiments unequivocally show that Oat3, similar to Oat1, is an organic anion/dicarboxylate exchanger that indirectly couples concentrative organic anion uptake to the Na+ gradient. Thus, on the basis of energetics, transport on Oat3 is indistinguishable from transport on Oat1.
MATERIALS AND METHODS
X. laevis oocytes were obtained from dissected ovaries by collagenase A treatment. Substrate uptake assays were performed 3 days after injection with 20 ng of capped cRNA as previously described (14, 45, 48). Oocytes were randomly divided into experimental groups of four to eight each and incubated for 10 or 60 min at room temperature in oocyte Ringer 2 (OR-2) medium (in mM: 82.5 NaCl, 2.5 KCl, 1 Na2HPO4, 3 NaOH, 1 CaCl2, 1 MgCl2, 1 pyruvic acid, and 5 HEPES, pH 7.6) containing 25 μM [3H]PAH (1 μCi/ml) or 100 nM3H-labeled estrone sulfate (ES; 2 μCi/ml) in the absence or presence of 1 mM probenecid, 5 mM LiCl, or 1 mM methylsuccinate (MS). Individual oocyte radioactivity was measured by liquid scintillation spectroscopy with external quench correction. For experiments involving expression of Oat1 and an Na+-dicarboxylate cotransporter, cRNA synthesized from rOat1 and rabbit Na+-dicarboxylate cotransporter 1 (rbNaDC-1) cDNA (33) were coinjected at a ratio of 2:1. For trans-stimulation studies, oocytes were preloaded by 90 min of incubation in OR-2 medium containing 0–5 mM glutarate (GA) or MS and quickly rinsed in GA- or MS-free medium before initiation of uptake. Na+-free uptake was determined in OR-2 medium isosmotically adjusted using N-methylglucamine.
Rat renal tissue slice preparation and uptake assays were performed according to standard laboratory protocols (35, 37). Animals were euthanized by CO2 inhalation, and kidneys were dissected into freshly oxygenated ice-cold saline. Tissue slices (≤0.5 mm, ∼5–10 mg wet wt) were maintained in ice-cold modified Cross and Taggart saline (in mM: 95 NaCl, 80 mannitol, 5 KCl, 0.74 CaCl2, and 9.5 Na2HPO4, pH 7.4). Substrate (100 nM [3H]ES) uptake was determined after 5–60 min of incubation in the presence and absence of inhibitors or Na+ (Na+-free medium was isosmotically adjusted using N-methylglucamine). After uptake was determined, slices were rinsed, blotted, and weighed and then dissolved in 1 ml of 1 M NaOH, neutralized with 1 ml of 1 M HCl, and assayed by liquid scintillation spectroscopy. Duplicate medium samples (50 μl) were also assayed. Uptakes were calculated as uncorrected tissue-to-medium ratios [i.e., (dpm/mg tissue) ÷ (dpm/μl medium)].
All animal experiments were conducted under protocols approved by the National Institute of Environmental Health Sciences Animal Care and Use Committee.
Means were compared using paired or unpaired Student'st-test. Differences in mean values between control and test groups were considered significant when P ≤ 0.05.
[3H]PAH (4 Ci/mmol) and [3H]ES (40 Ci/mmol) were purchased from PerkinElmer Life Sciences (Boston, MA). Unlabeled PAH, ES, probenecid, MS, GA, and taurocholate (TC) were obtained from Sigma (St. Louis, MO). All other chemicals were obtained from commercial sources and were of reagent grade.
Oat3-mediated organic anion transport in X. laevis oocytes.
Sixty-minute uptake of 25 μM [3H]PAH or 100 nM [3H]ES was measured in oocytes expressing rOat1 or rOat3 (Fig. 2). Oocytes expressing rOat1 and rOat3 exhibited substantial probenecid-sensitive PAH uptake compared with water-injected control oocytes, demonstrating expression of specific OATs in both groups of injected oocytes (Fig. 2). Despite functional transporter expression, rOat1-expressing oocytes did not take up ES, whereas rOat3-expressing oocytes exhibited significant probenecid-sensitive ES uptake (Fig. 2). As demonstrated previously for the basolateral organic anion/dicarboxylate exchanger Oat1 (14,49), incubating rOat1-expressing oocytes in medium containing 2.5 mM GA for 90 min before substrate exposure (GA preloading) significantly stimulated probenecid-sensitive PAH uptake (Fig.2). Importantly, preloading with GA also significantly stimulated uptake of PAH and ES by rOat3-expressing oocytes (Figs. 2 and3). This trans stimulation of rOat3-mediated ES uptake increased in a GA concentration-dependent manner, with 2.5 mM GA preloading causing a maximal increase over nonpreloaded oocytes (data not shown). In contrast, preincubating rOat3-expressing oocytes with MS, a dicarboxylate that cannot substitute for α-ketoglutarate (α-KG) or GA on Oat1 (11, 18,49), did not increase PAH or ES uptake by rOat3 (Fig.3).
To determine whether the GA trans-stimulation of rOat3-mediated transport observed in Figs. 2 and 3 could be facilitated by energetic coupling to an Na+ gradient, X. laevis oocytes were coinjected with cRNAs for rOat3 and a rabbit Na+-dicarboxylate cotransporter (rbNaDC-1). Previous studies utilizing renal slices and renal basolateral membrane vesicles firmly established that the Na+-dicarboxylate cotransporter (but not the organic anion/dicarboxylate exchanger) is inhibited by 5 mM Li+, 1 mM MS, and Na+ replacement (6,34, 37). We therefore determined the effects of those treatments on the ability of GA preloading to stimulate ES uptake in doubly injected oocytes (Fig. 4). Cells incubated in OR-2 medium containing 100 μM GA for 90 min before ES uptake showed a significant increase in ES accumulation (Fig. 4). Li+ and MS added to the preloading buffer completely blocked the trans-stimulatory effect of GA, and preincubation in Na+-free medium significantly reduced stimulation by GA (Fig. 4). These treatments were without effect when imposed during the 10-min ES uptake period, rather than the preloading period (data not shown). Thus these oocyte experiments demonstrated that Oat3 is an organic anion exchanger that could be coupled to the Na+ gradient through Na+-dicarboxylate cotransport.
ES uptake in rat renal cortical slices.
ES is well established to be an Oat3 (but not an Oat1) substrate (Fig.2) (9, 23, 46). Thus, to determine whether transport on Oat3 was indeed indirectly coupled to Na+ in the kidney, we measured ES accumulation in rat renal cortical slices, a preparation that tracks basolateral transport into the tubules (5,54). Figure 5 shows the time course of 100 nM ES accumulation by renal slices. Uptake was linear for ≥15 min and then appeared to approach steady state during the subsequent 45 min. Uptake was clearly concentrative, e.g., tissue-to-medium ratios at 60 min were ∼22. ES accumulation was markedly inhibited by 1 mM probenecid throughout the time course. Removal of Na+ from the uptake medium was as effective as probenecid in inhibiting ES uptake (Fig.5). Because this result is consistent with tertiary active, indirect coupling to Na+, as previously shown for Oat1 (14, 49), the ability of externally applied GA to stimulate ES uptake by the slices was evaluated. External GA stimulated ES accumulation in a concentration-dependent manner, with peak stimulation at 20 μM GA (Fig. 6 A). This GA concentration was used in all subsequent experiments. As in the oocyte experiments shown above (Figs. 2-4), 5 mM Li+ and 1 mM MS (inhibitors of Na+-dicarboxylate cotransport) significantly inhibited ES uptake in the presence of 20 μM GA, effectively reducing its accumulation to levels observed in the presence of 1 mM probenecid. Removal of Na+ from the uptake buffer (Na+-free) was equally effective (Fig.6 B).
Na+ removal and probenecid did not completely inhibit concentrative ES transport by renal slices. To investigate the remaining component of transport further, TC was used as an inhibitor of ES uptake (Fig. 6 C). TC is known to be a substrate for several OATs, including Oat3 and members of the organic anion-transporting polypeptide (Oatp) family (1, 8, 20,53). TC reduced ES uptake in a concentration-dependent manner. At 100 μM, TC inhibited ES uptake to a greater extent than 1 mM probenecid or Na+ removal (Fig. 6 C). When applied together, 1 mM probenecid + 100 μM TC reduced uptake to a greater extent than probenecid alone, indicating the presence of a probenecid-insensitive, TC-sensitive component to ES uptake in renal slices.
Understanding renal OAT function at the molecular level is essential to our ability to improve drug design, delivery and therapeutic strategies and to understand and intervene in disease progression. hOAT1-mediated transport has been implicated in the nephrotoxicity associated with nucleoside phosphonate antiviral (adefovir and cidofovir) therapy (13, 14, 19). Indeed, inhibition of active drug uptake into kidney proximal tubule cells via coadministration of an Oat1 substrate may mitigate their nephrotoxicity without compromising their therapeutic value (29). OAT function may also be a key factor in the progression of disease resulting from exposure to increased levels of toxic endogenous or xenobiotic substances. During chronic renal failure, uremic toxins accumulate in the blood, and recent studies have shown that several uremic toxins are transported by, or are inhibitors of, rOat1 and rOat3 (15, 17, 30, 31). Thus these transporters may control the progression of nephrotoxicity as well as urinary excretion of potentially toxic anionic drugs and metabolites.
Categorizing the contributions of individual OAT family members to overall kidney transport function, although clinically important, has been experimentally difficult, because multiple transporter paralogs (i.e., Oat1, Oat2, Oat3, and Oat4) are expressed in the proximal tubule and appear to be responsible for renal organic anion excretion. However, increased knowledge of individual OAT specificity and subcellular localization is expanding our ability to interpret transport studies along these lines. For example, PAH is transported by Oat1 and Oat3, but not by Oat2 or Oat4 (9, 10, 23, 40, 41, 46,49); renal transport of ES is mediated by Oat3 and Oat4, but not by Oat1 or Oat2 (9, 10, 23, 40, 46); and TC is transported by Oat3, but not by Oat1, Oat2, or Oat4 (9, 10, 40, 46). Thus, on the basis of specificity, Oat2 should not play any role in the renal transport of any of these three substrates. rOat1 (45,51) and, recently, mouse Oat3 (46) have been localized to the basolateral membrane of renal proximal tubule cells. In contrast, hOAT4 has been observed in the apical membrane (3). This suggests that transport of PAH across the basolateral membrane of renal proximal tubule cells (e.g.,in renal cortical slices) would be predominantly mediated by Oat1 and Oat3 and that, for the OATs, uptake of ES and TC would serve as a measure of transport mediated solely by Oat3. In agreement with this interpretation, when expressed in X. laevis oocytes, rOat1 and rOat3 transported PAH (Fig. 2) (23, 49). On the other hand, as shown previously for Oat1 and Oat3 (9, 23, 46), ES is transported by Oat3, but not by Oat1 (Fig. 2). Thus, in oocytes and renal slice [reflects only basolateral uptake by the tubules (5, 54)], measurement of ES transport allows us to focus on Oat3-mediated events.
Previous reports on Oat3 function in the X. laevis oocyte expression system found no evidence for energetic coupling of transport to Na+ or dicarboxylate gradients (9, 23). In the present experiments, careful control of imposed dicarboxylate gradients and coexpression of an Na+-dicarboxylate cotransporter along with Oat3 has revealed indirect coupling to Na+. Our data clearly show that rOat3-mediated uptake of PAH and ES is markedly trans-stimulated by GA (Figs. 2 and3). This is precisely the behavior previously demonstrated for the basolateral organic anion/dicarboxylate exchanger rOat1/hOAT1 (14, 49). In addition, similar to thetrans-stimulatory effect of GA on rOat1/hOAT1, GA stimulation of rOat3 uptake was concentration dependent (data not shown), whereas MS did not stimulate rOat3 uptake (Fig. 3). These findings strongly suggest that Oat3, similar to Oat1, functions as an organic anion/dicarboxylate exchanger and could be energetically coupled to the Na+ gradient through an Na+-dicarboxylate cotransporter that generates the dicarboxylate counterion gradient utilized for organic anion exchange.
To directly test the indirect coupling hypothesis, ES uptake was measured in oocytes coexpressing rbNaDC-1 and rOat3 (Fig. 4). Again, preloading with GA stimulated uptake. However, as expected for the coupled system, preloading in the presence of Li+ or MS (inhibitors of renal Na+-GA cotransport) or removal of Na+ blocked stimulation by preventing uptake of GA by rbNaDC-1 during the preincubation step (Fig. 4). None of these treatments was inhibitory when applied during ES uptake (data not shown). Thus, when GA entry was prevented by inhibition of the Na+-GA cotransporter, rOat3-mediated ES uptake was not stimulated by external GA. These findings in X. laevisoocytes demonstrate that rOat3 (similar to Oat1) is an organic anion/dicarboxylate exchanger that can be driven by indirect coupling to the Na+ gradient.
The same indirect coupling mechanism was demonstrated for the Oat3 substrate ES at the basolateral membrane of proximal tubular cells. ES uptake by rat renal cortical slices in vitro was clearly Na+ dependent (Fig. 5). Moreover, as seen in rOat3-expressing oocytes, ES uptake was stimulated by addition of external GA, and this stimulation was prevented by Li+, MS, and probenecid and by Na+ removal (Fig. 6B). This pattern of stimulation and inhibition is that expected for organic anion transport indirectly coupled to the Na+ gradient through Na+-dicarboxylate cotransport. It is the same pattern seen previously for renal uptake of the Oat1 and Oat3 substrate PAH (34, 35, 37, 49).
These experiments also demonstrate that Oat3 is not the only transporter responsible for ES uptake by rat renal cortical slices. Results from inhibition experiments using another Oat3 substrate, TC, indicate the presence of an Na+-independent, probenecid-insensitive organic anion transporter (i.e.,non-OAT). Another family of transporters, Oatp (Slc21a), mediate Na+-independent, probenecid-insensitive transport of ES and TC (1, 2, 8, 20, 21, 32). Of these family members, Oatp1 is expressed on the apical membrane in the kidney, whereas expression of Oatp2, Oatp3, and Oatp4 was not detected in the kidney (1, 4, 21, 24, 53). The remaining member cloned thus far, Oatp5, is highly expressed in the kidney; however, it has not been localized or functionally characterized (12, 24). Thus this residual transport activity may be attributable to Oatp5.
In conclusion, we demonstrate here the first data identifying the mechanism of organic anion uptake by Oat3. When studied in the X. laevis oocyte expression system or in rat renal cortical slices, rOat3 proved to be an organic anion/dicarboxylate exchanger that could be energetically coupled to the Na+ gradient through Na+-dicarboxylate cotransport. Thus, on the basis of energetic considerations (not specificity), Oat1 and Oat3 appear to be indistinguishable. This conclusion has important implications in several areas. First, it will be important to reevaluate past conclusions concerning the relative roles of these two transporters (and the other OATs) in transport of endogenous metabolites, toxins, and therapeutic drugs and in drug-drug or drug-xenobiotic interactions. Second, renal transport of small organic anions (PAH and fluorescein) is regulated by multiple signals, including hormones and protein kinases (27, 39, 55). Because transport was Na+ dependent, it was assumed that the target of regulation was Oat1. However, Oat1 and Oat3 appear to respond to protein kinase C activation (26, 50). Thus, in light of the present findings, it will be important to determine in intact renal tissue how signaling affects each of the transporters. Finally, the present results raise the following question: Do other members of the OAT family of transporters utilize a similar mechanism? Although this question has not been addressed directly, Oat2, which appears to be localized to the basolateral membrane in the kidney and liver, transports dicarboxylates, suggesting that it too may be an organic anion/dicarboxylate exchanger (16, 40, 43). It remains to be demonstrated that this is indeed the case.
We thank Laura Hall for assistance with the oocyte expression experiments.
↵* D. S. Sweet and L. M. S. Chan contributed equally to this study.
Address for reprint requests and other correspondence: J. B. Pritchard, Laboratory of Pharmacology and Chemistry, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC 27709 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published December 17, 2002;10.1152/ajprenal.00405.2002
- Copyright © 2003 the American Physiological Society