Novel liver-specific organic anion transporter OAT7 that operates the exchange of sulfate conjugates for short chain fatty acid butyrate


  • The nucleotide sequence reported in this paper has been submitted to the GenBank /EBML/DDBJ Data bank with the accession number AB074812.


The liver plays an important role in the elimination of endogenous and exogenous lipophilic organic compounds from the body, which is mediated by various carrier proteins that differ in substrate specificity and kinetic properties. Here, we have characterized a novel member of the organic anion transporter family (SLC22) isolated from human liver. The transporter named organic anion transporter 7 (OAT7/ SLC22A9) showed 35% to 46% identities to those of other organic anion transporters of SLC22 family. When expressed in Xenopus oocytes, OAT7 mediated Na+-independent, high-affinity transport of sulfate-conjugated steroids, estrone sulfate (ES; Km = 8.7 μM), and dehydroepiandrosterone sulfate (Km = 2.2 μM). In addition, OAT7 interacted with negatively charged sulfobromophthalein, indocyanine green, and several sulfate-conjugated xenobiotics. In contrast, glucuronide and glutathione conjugates exhibited no inhibitory effects on OAT7-mediated [3H]ES transport. OAT7-mediated [3H]ES transport was trans-stimulated by three-carbon to five-carbon (C3 to C5) short-chain fatty acids. The efflux of [14C]butyrate (C4) via OAT7 was significantly trans-stimulated by extracellular ES. Furthermore, OAT7 mediated [14C]butyrate uptake and [3H]ES efflux in exchange for extracellular butyrate both in Xenopus oocytes and OAT7-stably expressing cells. OAT7 protein was localized in the sinusoidal membrane of hepatocytes by immunohistochemical analysis. Conclusion: OAT7 is the first liver-specific transporter among members of the organic anion transporters of SLC22 family. Our findings suggest a new class of substrates for organic anion transporters and provide evidence for the transport of anionic substances such as sulfate-conjugates in exchange for butyrate in hepatocytes. (HEPATOLOGY 2007;45:1046–1055.)

The mammalian liver, as well as kidney, is a primary organ responsible for the detoxification and the elimination of endogenous and exogenous amphipathic organic anions such as bile salts, bilirubin, sulfobromophthalein, and numerous drugs and their metabolites. These substances are rapidly taken up by hepatocytes across the sinusoidal plasma membrane.1, 2 They are biotransformed by the activities of cytochrome P450 enzymes (phase I) and converted into anionic conjugates with sulfate, glutathione, glucuronate, or other negatively charged moieties (phase II), and subsequently excreted into the bile or the systemic circulation.3 Sulfate conjugation is particularly important for the metabolism of steroids and bile acids as well as drugs and xenobiotics.4 The import and export processes depend on specific transporter molecules that sit in the plasma membrane of the hepatocytes.

Hepatic transport systems for various bile acids and non-bile acid organic anions were divided into three classes, sodium-dependent and sodium-independent influx systems, and adenosine triphosphate–dependent efflux systems. Recently, specific transporters that mediate organic anion uptake across the sinusoidal membrane have been identified.5 In humans, the sodium-dependent bile acid uptake is mediated by the Na+-taurocholate cotransporting polypeptide.6 For sodium-independent uptake, organic anion-transporting polypeptides (OATPs) of the SLC21 (SLC0) family, such as OATP-B,7, 8 OATP-C,9–11 and OATP8,12, 13 mediate the transport of a variety of organic anions such as sulfobromophthalein, bile salts, and sulfate-conjugated steroid hormones, although their driving forces are still unclear. In addition, organic anion transporter 2 (OAT2) in the SLC22 family mediates the transport of organic anion such as para-aminohippurate,14 prostaglandins,15 zidovudine,16 tetracycline,17 and salicylates18 in a sodium-independent manner. Recently, Kobayashi et al.19 have reported that it also mediates the transport of steroid sulfates as well as several drugs. Members of ABCC subfamily MRP3 (multidrug resistance protein 3), MRP4, and MRP6 accept conjugates with glutathione, glucuronate, or sulfate as substrates.20

We and others identified another major sodium-independent multispecific organic anion transporter (OAT) group in the SLC22 family that mediates the transport of a wide variety of organic anions.21, 22 Expressed in multiple tissues such as the kidney, placenta, liver, and brain, OATs play important roles in the elimination of drugs and their metabolites. We reported that OAT1-4, expressed in the basolateral or apical membrane of renal tubular cells, mediate the transport of various organic anions. We systematically searched for paralogs of OAT family in the draft human genome databases,23 which led to the identification of several nucleotide sequences similar to those of the known SLC22 members.24 We report on a novel human organic anion transporter (termed OAT7) expressed exclusively in liver and localized on the sinusoidal membrane of hepatocytes. OAT7 interacted with various sulfate conjugates not only of steroid hormones but also of several xenobiotics. Although OAT7 mediated an exchange of organic anions similar to other members of OAT family, OAT7 is unique because it mediated the transport of estrone sulfate (ES) in exchange for four-carbon short-chain fatty acid (SCFA) butyrate.


DHEAS, [3H]ES, [3H]dehydroepiandrosterone sulfate; ES, estrone sulfate; hOAT, human organic anion transporter; MC, monocarboxylate; MRP, multidrug resistance protein; OAT, organic anion transporter; OATP, organic anion transporting polypeptide; RT, reverse transcription; RT, room temperature; SCFA, short-chain fatty acid.

Materials and Methods

Isolation of OAT7 cDNA.

We identified various DNA sequences or fragments that showed homology to those of the genes for human OATs (hOATs)14, 24–27 by searching the publicly available human genome database23 using the BLAST program ( As a consequence, we found a nucleotide sequence that seems to correspond to that of an organic anion transporter of SLC22 family in a BAC clone RP11-151E18 (GenBank accession AP002367) mapped to human chromosome 11q13.1. The nucleotide sequence of the predicted gene matched an expressed sequence tag from the human liver (AA705512). The [32P]dCTP-labeled probe was synthesized from IMAGE clone 462315 (corresponding to AA705512) and used for the screening of a human liver cDNA library, as described.28 The cDNA inserts (named OAT7, AB074812) in positive λZipLox phages were recovered in the expression vector pZL1 (Invitrogen) by in vivo excision.

Northern Blot Analysis.

Hybridization blots that contain polyadenylated RNAs from various human tissues [human 12-lane multiple tissue Northern (MTN) blot, human fetal multiple tissue Northern (MTN) blot, Clontech] were used for the Northern blot analysis for OAT7. Using full-length OAT7 cDNA as a probe, the hybridization was performed as described.26

Reverse Transcription PCR Analysis.

Human Multiple Tissue cDNA Panels I and II were purchased from Clontech. For human adrenal gland and uterus, we obtained BD Premium Total RNA (Clontech) and performed reverse transcription (RT) using SuperScript First-Strand Sythesis System for RT-PCR (Invitrogen). Each cDNA sample (1.5 μl) was amplified as described.29 Primers, used for PCR amplification, are shown in Table 1.

Table 1. PCR Primers Used in This Study
TargetPrimersProduct Size
 Sense primer5′-ACCTGGCCATCGCTGCTG-3′551 bp
 Antisense primer5′-TGTGTGTTGCCCACTCGG-3′ 

Preparation of Anti-OAT7 Antibody and Western Blot Analysis.

Corresponding to the 11 amino acids of the COOH-terminus of OAT7, we generated a rabbit anti-OAT7 polyclonal antibody raised against a keyhole limpet hemocyanin-conjugated synthesized peptide, KQEDPRVEVTQ (542-552 of the amino acid sequence). Western blot using the samples of human liver total protein, purchased from Biochain Institute Inc., was carried out as described.29


Paraffin sections of human liver were obtained from commercial sources (Biochain Institute Inc.). They were incubated with an affinity-purified anti-OAT7 antibody (1:100) overnight at 4°C, as reported.30

Expression and Transport Assays in Xenopus Oocytes.

cRNA synthesis and uptake measurements were performed as described.28 After 2 to 3 days' incubation of oocytes injected with capped OAT7 cRNA (20 ng), uptake experiments were performed at room temperature (RT) in ND96 solution (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, and 5 mM HEPES, pH 7.4). The uptake experiment was initiated by replacing the ND96 solution with that containing radiolabeled [3H]ES, [3H]dehydroepiandrosterone sulfate (DHEAS), [14C]butyrate, [3H]propionate, or [14C]valerate (American Radiolabeled Chemicals) and was terminated by adding ice-cold ND96 solution after 1 hour of incubation.

For determination of the kinetic parameters, the concentrations of ES and DHEAS were varied from 1 to 50 μM, and those of butyrate were 5 to 100 μM. The kinetic parameters for the uptake via OAT7 were determined with the Eadie-Hofstee equation. For the inhibition study, the uptake rates of 1 μM [3H]ES by oocytes injected with water or OAT7 cRNA were measured for 1 hour in the absence or presence of 100 μM test compounds in ND96 solution.

Examination of Trans-stimulatory Effect on OAT7-Mediated Transport.

To study its transport mode, OAT7-expressing and control oocytes were incubated with 100 nM [3H]ES for 90 minutes for pre-loading. After washing, individual oocytes were transferred to ND96 solution or that containing 100 μM or 1 mM unlabeled ES. The value of [3H]ES efflux for 3 hours is shown as the percentage of the preloaded value.

To examine the trans-stimulatory effect on the [3H]ES uptake, 50 nl cold monocarboxylates (MCs) (100 mM) was injected into oocytes expressing OAT7 with a fine-tipped glass micropipette as described.30 Then, individual oocytes were washed with ice-cold ND96 solution twice, and incubated with ND96 at RT for 1 hour with [3H]ES (50 nM).

To examine the trans-stimulatory effect of ES on the [14C]butyrate efflux or that of butyrate on the [3H]ES efflux, 50 nl [14C]butyrate (approximately 1 mM) or [3H]ES (approximately 20 μM) was injected into oocytes expressing OAT7. Then, individual oocytes were washed with ice-cold ND96 solution twice and incubated with ND96 at RT for 2 minutes with or without cold ES (0.5 mM) or cold butyrate (1 mM).

For the uptake and efflux measurements in the current study, 8 to 10 oocytes were used for each data point. We repeated each experiment at least 3 times to confirm the reproducibility. The results from representative experiments are shown in the figures.

Cell Culture, Establishment of S2-OAT7, and Its Uptake Measurement.

S2 cells that stably express OAT7 were established and maintained as described elsewhere.30 OAT7 expression was confirmed by the Western blot using the lysate from S2-OAT7 cells (data not shown).

Uptake experiments were performed as described.31 Organic anion transport in S2-OAT7 cells was estimated by measuring the uptakes of [3H]ES, and [14C]butyrate. S2-OAT7 cells or S2-mock cells that were transfected with pcDNA3.1(+) vector only were plated in 24-well plates (1 × 105 cells) and cultured for 2 days. After the medium was removed, the cell monolayers were washed twice with D-PBS (with 5.5 mM D-glucose), and preincubated for 10 minutes. Then, the monolayer was incubated with 500 μlD-PBS containing 40 nM [3H]ES, or 18 μM [14C]butyrate for 2 minutes at 37°C. Because the uptakes of [3H]ES and [14C]butyrate were linear up to 5 minutes, the initial uptake was assessed as the uptake for 2 minutes (data not shown).

To confirm the results in oocytes, we performed the trans-stimulation study on OAT7-mediated ES efflux. First, the S2-OAT7 or S2-mock cells were preloaded with D-PBS with [3H]ES (20 nM for S2 mock, 3 nM for S2-OAT7) at 37°C for 5 minutes. This procedure gave the equal level of accumulation of [3H]ES in both cells. After washing with D-PBS three times, the cells were incubated in a medium with or without butyrate (100 μM), for 30 seconds at 37°C. Then, the amount of substrate accumulated within the cells and effluxed from the cells was determined by measuring radioactivity.

Statistical Analysis.

All the results from the experiments are expressed as mean ± SEM. Statistical significance was judged from Student t tests. Differences were considered significant at a level of P < 0.05. *P < 0.05, **P < 0.01, ***P < 0.001 versus control for all experiments.


Structural Features of OAT7.

OAT7 cDNA consisted of 2342 base pairs and had an open-reading frame of 1662 base pairs encoding a 554-amino-acid-residue protein. The database search revealed that the identical nucleotide sequence was deposited under the name of SLC22A9/hUST3 (AB062418, GenBank) and hOAT4,14 which were not functionally characterized. Because we and others already identified and functionally characterized organic anion transporters OAT4,25 Oat5,30 and Oat6,32 we renamed this newly functionally characterized transporter as OAT7 to avoid confusion. The amino acid sequence of OAT7 shows 35% to 46% identities to those of hOAT1-4.14, 25–27

Tissue Distribution of OAT7.

OAT7 mRNA expression in human adult tissues was investigated by Northern blot analysis under high-stringency conditions (Fig. 1A). The strong signals for OAT7 (4.4, 3.2, and 2.4 kb) were detected only in human liver. Prolonged exposure of the blot showed no additional signals in other tissues tested, indicating that OAT7 is exclusively expressed in human liver. OAT7 is also expressed in human fetal liver (Fig. 1B).

Figure 1.

Tissue distribution of OAT7 in humans. Human multiple tissue blots (Clontech) [(A), adult tissues; (B), fetal tissues] were probed with a [32P]dCTP-labeled OAT7 full-length cDNA and washed under high-stringency conditions. kb, kilobases. (C) Distributions of OAT7 mRNA examined by RT-PCR on human multiple cDNA panels. The PCR products (5 μl) were resolved on a 2% agarose gel.

To further explore its tissue distribution, we carried out PCR analysis using human multiple-cDNA panels (Clontech) as well as the human adrenal gland and uterus cDNA prepared from total RNA (Clontech). As shown in Fig. 1C, the OAT7 transcript was detected exclusively in the liver, confirming its liver-specific distribution in humans.

Protein Expression and Localization of OAT7 in Human Liver.

To determine the native molecular weight and subcellular localization of OAT7 in human liver, we generated a rabbit polyclonal antibody against the carboxyl terminus of OAT7. The reactivity of the antibody with human liver total proteins is illustrated in Fig. 2A. The anti-OAT7 antiserum recognized a band with an apparent molecular mass of approximately 50 kDa. The positive band was affected by the presence of an antigen peptide (100 μg/ml) in the absorption experiment, confirming the specificity of the immunoreactions.

Figure 2.

Expression and localization of OAT7 in human liver. (A) Western blot analysis of the human liver protein (20 μg/lane) using an anti-OAT7 antibody (1:100 dilution) showing a 50-kd band for OAT7. The signals were detected with an ECL plus system (Amersham Pharmacia Biotech). The band disappeared in the absorption test using an antigen peptide (100 μg/ml). (B-D) Immunohistochemical detection of OAT7 protein in human liver. OAT7 was homogenously distributed within the liver acini (B, C). In the high-magnification view (D), OAT7 was mainly located in the sinusoidal plasma membrane of hepatocytes. (E) The absorption test in which the primary antibody was preabsorbed with antigen peptide (100 μg/ml), confirming the specificity of immunoreactions. Original magnifications: ×200 (B, C, and E) and ×400 (D). CV, central vein; PV, portal vein.

We investigated the cellular and subcellular distribution of OAT7 in intact human liver tissue by immunohistochemistry (Fig. 2B-D). OAT7 was detected in hepatocytes (Fig. 2B,C), but not in the epithelium of bile ducts, blood vessels, nor in the interstitial tissues. OAT7 is homogenously distributed across liver acini, including hepatocytes near the central vein and those close to the portal vein (Fig. 2B,C). A high-magnification view (Fig. 2D) showed that the immunoreactivity was detected mainly in the sinusoidal membrane in the hepatocytes. In the absorption experiments in which the tissue sections were treated with the primary antibody in the presence of antigen peptides, the immunostaining was not detected (Fig. 2E), showing the specificity of the immunoreactions.

Transport Properties of OAT7.

Xenopus oocytes injected with OAT7 cRNA exhibited transport activity of radiolabeled sulfates of steroid hormone such as [3H]ES, [3H]DHEAS compared with water-injected control (Fig. 3A,B ). Typical cationic substrates for organic cation transporters, such as tetraethylammonium and choline, or anionic substrates for OATs such as para-aminohippurate, α-ketoglutarate, prostaglandins, cyclic nucleotides, and salicylic acid, were not taken up significantly (data not shown).

Figure 3.

Functional expression of OAT7 in Xenopus oocytes. OAT7 mediated the transport of sulfate esters of steroid hormones. The uptake of radiolabeled [3H]estrone sulfate (ES) (50 nM) (A) and [3H]dehydroepiandrosterone sulfate (DHEAS) (100 nM) (B) by water-injected control oocytes and OAT7-expressing oocytes were determined for 1 hour.

Figure 4 shows the properties of the OAT7-mediated [3H]ES transport. The cell-associated count of [3H]ES increased linearly up to 3 hours of incubation in OAT7-expressing oocytes, indicating that OAT7 not only binds but also translocates ES across the plasma membrane (Fig. 4A). The uptake rate of ES via OAT7 was not affected by the replacement of extracellular sodium with choline (Fig. 4B), indicating that OAT7-mediated transport is sodium independent. OAT7-mediated ES transport was saturable (Fig. 4C) following Michaelis-Menten kinetics. Eadie-Hofstee plot yielded a Km value of 8.7 ± 1.1 μM for ES (means ± SEM, n = 3). We also measured the kinetic parameters for DHEAS. Its Km value was determined to be 2.2 ± 0.3 μM (mean × SEM, n = 3, data not shown), indicating that OAT7 has high affinity for DHEAS as well as for ES. The trans-stimulatory effect of extracellular ES (100 μM and 1 mM) on the OAT7-mediated efflux of preloaded [3H]ES was examined (Fig. 4D). The [3H]ES efflux was significant and stimulated by extracellular ES in OAT7-expressing oocytes. This result indicates that OAT7 is an exchanger.

Figure 4.

Properties of OAT7-mediated ES transport. (A) Time-dependent uptake of ES by OAT7-expressing oocytes. [3H]ES (50 nM) uptake in control (open circles) and OAT7-expressing oocytes (closed circles) was measured during 3 hours of incubation. (B) Na+-dependency of OAT7-mediated ES transport. The uptake rate of [3H]ES (50 nM) by control (open columns) or OAT7-expressing oocytes (closed columns) was measured in the presence or absence of extracellular Na+. NS, not significant. C, concentration dependence of OAT7-mediated [3H]ES uptake. Uptake rate of ES (1-50 μM) was measured in control and in OAT7-expressing oocytes. OAT7-mediated uptake was calculated by subtracting the uptake in control from that in OAT7-expressing oocytes and was used for the kinetic analysis. Inset: Eadie-Hofstee plot. V, velocity; V/S, velocity per concentration of ES. D, trans-stimulatory effects of extracellular ES on OAT7-mediated [3H]ES efflux. OAT7-expressing (closed columns) and control (open columns) were incubated with 100 M [3H]ES for 90 minutes. Individual oocytes were transferred to the ND96 solution or those containing 100 μM or 1 mM unlabeled ES. The level of [3H]ES efflux for 3 hours is shown as the percentage of the preloaded level.

Substrate Selectivity of OAT7.

To examine substrate selectivity of OAT7, we performed competition experiments in which [3H]ES (1 μM) uptake by OAT7 was measured in the presence or absence of non-radiolabeled organic compounds (100 μM) (Fig. 5). ES and DHEAS significantly inhibited OAT7-mediated [3H]ES uptake (Fig. 5A). We found that sulfobromophthalein and indocyanine green, which are cholephilic organic anions used clinically for assessment of hepatic function, also inhibited OAT7-mediated transport. OAT7 failed to interact with PAH, α-ketoglutarate, nonsteroidal anti-inflammatory drugs, diuretics, and cimetidine that are preferentially recognized by the other members of the SLC22 family. Importantly, OAT7-mediated ES uptake was not affected by probenecid, a representative inhibitor of organic anion transport.33, 34 Figure 5A also illustrates that OAT7 failed to interact with substrates of organic cation transporters, tetraethylammonium, and choline.35 These results demonstrate that OAT7 does not encompass broad substrate specificity, which is distinct from previously cloned OAT/organic cation transporters or OATPs. Bile salts such as cholate and taurocholate did not inhibit ES transport mediated by OAT7, illustrating the clear difference in substrate selectivity between OAT7 and OATPs.

Figure 5.

Effect of organic ions on OAT7-mediated [3H]ES uptake. (A) The inhibition of OAT7-mediated [3H]ES uptake by typical substrates of SLC22 family. OAT7-mediated uptake of [3H]ES (1 μM) was determined in the absence or presence of inhibitors at 100 μM. The values are expressed as percentages of OAT7-mediated [3H]ES uptake in the absence of the inhibitors (control, open column). (B) The inhibition of OAT7-mediated [3H]ES uptake by sulfate-, glucuronate-, and glutathione-conjugates. OAT7-mediated uptake of [3H]ES (1 μM) was determined in the absence or presence of inhibitors at 100 μM. The values are expressed as percentages of OAT7-mediated [3H]ES uptake in the absence of the inhibitors (control, open column).

We further examined the substrate selectivity of OAT7 using various compounds conjugated with sulfate, glucuronide, and glutathione (Fig. 5B). Unlabeled ES, DHEAS, β-estradiol sulfate, 4-methlumbelliferyl sulfate, and minoxidil sulfate showed significant inhibitory effects on OAT7-mediated uptake of [3H]ES, suggesting that OAT7 preferentially interacts with sulfate conjugates. However, p-nitrophenyl sulfate, vinblastine sulfate, and vincristine sulfate did not exhibit apparent inhibitory effects on OAT7. In contrast, none of the glucuronide-conjugates or glutathione-conjugates, substrates of OATPs and MRPs, had apparent inhibitory effects on OAT7.

Interaction of OAT7 with Short Chain Fatty Acids.

Recently, we reported that OAT4, an apical isoform of OAT subfamily, mediates organic anion transport in exchange for 5-carbon (C5) dicarboxylate glutarate.31 This indicated that OAT4 is an organic anion/dicarboxylate exchanger similar to basolateral isoforms OAT128 and OAT3, which have been clarified as organic anion/dicarboxylate exchangers.36 To study the transport mode of OAT7, first we examined the trans-stimulatory effects of various length dicarboxylates on OAT7-mediated ES uptake. Injection of dicarboxylates (100 mM, 50 nl) into the oocytes did not induce any trans-stimulatory effects (data not shown).

We then tested the trans-stimulatory effect of MCs on the uptake and efflux of radiolabeled substrates via OAT7. Among MCs tested, short-chain fatty acids (SCFAs) with 3 to 5 carbons (C3-C5) and nicotinate significantly trans-stimulated the uptake of [3H]ES by the oocytes expressing OAT7 (Fig. 6A). Next we examined the interaction of the MC to OAT7 by inhibition experiments in which the uptake of 1 μM [3H]ES was measured in the presence of various MCs (1 mM) ranging from C2 (acetate) to C6 (caproate),37 as well as lactate and nicotinate (0.1 mM each). OAT7-mediated [3H]ES uptake was inhibited by propionate (C3), butyrate (C4), and valerate (C5) (Fig. 6B).

Figure 6.

Effects of short-chain fatty acids (SCFAs) on OAT7-mediated ES transport. (A) Trans-stimulatory effect of monocarboxylates (MCs) on the uptake of [3H]ES via OAT7. Control (open column) and OAT7-expressing oocytes (closed column) were injected with 100 mM unlabeled MCs indicated, or water and incubated for 1 hour at RT. Then, the oocytes were incubated with [3H]ES (100 nM), and its uptake for 1 hour was determined. (B) inhibition of OAT7-mediated [3H]ES uptake by MCs. [3H]ES concentration used was 1 μM. The inhibitor concentration was 1 mM for SCFAs and 0.1 mM for lactate and nicotinate (closed column). The value was expressed as a percentage of [3H]ES uptake in OAT7-expressing oocytes in the absence of the inhibitor (open column).

Next, we examined OAT7-mediated butyrate transport in Xenopus oocytes to examine its transport characteristics, because butyrate showed highest transport activity via OAT7 among C3 to C5 SCFAs (Fig. 7A). OAT7 exhibited significant transport activity for [14C]butyrate, compared with water-injected control (Fig. 7A). Then we performed the efflux study to measure [3H]ES efflux from the oocytes expressing OAT7. OAT7 mediated [3H]ES efflux in exchange for unlabeled butyrate in the medium (1 mM) (Fig. 7B). Finally, the efflux of [14C]butyrate from the oocytes expressing OAT7 was significantly trans-stimulated by unlabeled ES in the medium (0.1 mM) (Fig. 7C).

Figure 7.

OAT7-mediated butyrate transport. (A) OAT7-mediated uptake of SCFAs, propionate, butyrate, and valerate. The uptake rates of [3H]propionate (200 nM), [14C]butyrate (90 μM), and [14C]valerate (9.3 μM) were measured on control (open column) and OAT7-expressing oocytes (closed column). (B) Trans-stimulatory effect of butyrate on [3H]ES efflux via OAT7. Control (open column) and OAT7-expressing (closed column) oocytes were injected with [3H]ES. After washing, the oocytes were incubated with 1 mM or 0 mM unlabeled butyrate. The level of [3H]ES efflux was determined for 1 hour. (C) Trans-stimulatory effect of ES on the efflux of [14C]butyrate via OAT7. Control (open column) and OAT7-expressing (closed column) oocytes were injected with [14C]butyrate. After washing, the oocytes were incubated with 0.5 mM or 0 mM unlabeled ES. The amount of [14C]butyrate effluxed for 2 minutes was determined.

To confirm these findings, we established a cell lines stably expressing OAT7 (S2-OAT7). The expression of OAT7 was confirmed by Western blot using the crude membrane fractions from S2-OAT7 cells (data not shown). OAT7-mediated uptake of [3H]ES (Fig. 8A) and [14C]butyrate (Fig. 8B) showed saturable kinetics. The Eadie-Hofstee plot yielded Km for ES (40.7 μM) and butyrate (35.7 μM) and Vmax for ES (674 pmol/min/mg protein) and butyrate (42.1 pmol/min/mg protein). The [3H]ES efflux via OAT7 was significantly trans-stimulated by extracellular butyrate (100 μM) (Fig. 8C).

Figure 8.

Transport of ES and butyrate in S2 cells expressing OAT7. Representative kinetic studies of the uptake of [3H]ES (A) and of [14C]butyrate (B) mediated by OAT7 expressed in mouse S2 cells. The cells were incubated for 2 minutes at 37°C. Inset: Eadie-Hofstee plot. V, velocity; V/S, velocity per concentration of ES (A) and butyrate (B). (C) Trans-stimulatory effects of butyrate on the ES efflux via S2-mock (open column) and S2-OAT7 (closed column).


OAT7 is unique among members of the OAT subfamily in several ways. It is substantially diverged in its substrate selectivity from other OAT members that function as multispecific organic anion transporters. Using Xenopus oocytes expressing OAT7, we found that OAT7 preferentially interacts with sulfate conjugates of xenobiotics and steroid hormones but not with other organic anions or organic cations (Fig. 5A). We previously reported that OAT3 and OAT4, expressed in human kidney, mediate the transport of sulfate conjugates. In contrast to OAT7, however, OAT3/4 also transport a wide variety of organic anions including several drugs.25, 26 Recently, Kobayashi et al.19 reported that it also mediates the transport of steroid sulfates as well as several drugs. In addition to its narrow substrate recognition, OAT7-mediated transport was not inhibited by probenecid, which is a typical inhibitor of other members of the OAT subfamily. These observations indicate that OAT7 has unique substrate specificity that was not found in other members of the SLC22 family.

Another marked feature of OAT7 is in its limited expression. Of 18 major tissues examined, only the liver was found to express OAT7 (Fig. 1). This tissue distribution distinguishes OAT7 from other OATs characterized to date, because most OATs are found mainly in kidney and expressed in multiple tissues.21, 22 Amongst OATs, OAT2 is considered to be important for hepatic organic anion handling because of its abundant expression in the liver. It also is expressed moderately in the kidney. Besides OAT2, rat OAT3, expressed in the liver, kidney, and brain, mediates the transport of [3H]ES.38 However, its human and murine orthologs hOAT3/mOat3 are not expressed in the liver.26, 39 OAT7 is, thus, the first liver-specific functional OAT member found in humans.

OAT7 is localized at the sinusoidal membrane of hepatocytes (Fig. 2) and mediates the bidirectional transport of [3H]ES in exchange for SCFA, particularly butyrate (Figs. 6 and 7). Taken together, these results allow us to propose a model whereby OAT7 contributes to the hepatic uptake and/or efflux of organic anions dependent on the metabolic status of SCFA in the liver (Fig. 9). SCFAs such as acetate (C2), propionate (C3), and butyrate (C4) are generated at high levels in the colon by bacterial fermentation and absorbed into colonic epithelial cells.40 They are largely consumed by colonic cells because SCFAs are the preferred metabolic fuel in colonocytes. Remaining SCFAs are then transported to the liver via the portal vein and are taken up from blood into the hepatocytes. In humans, the butyrate concentration in the portal vein is 28.8 μM and that in the hepatic vein is 12 μM.40 Metabolism of butyrate by liver involves its conversion to butyryl-CoA, longer-chain fatty acid acetyl-CoA, or to ketone bodies.41 Because SCFAs are weak anions, they may require a specific transport protein to cross the cell membrane.41 Recently, the transport processes for SCFAs in the apical membrane of large intestine epithelial cells has been identified as a sodium-coupled monocarboxylate cotransporter (SLC5A8), which was originally known as a tumor suppressor down-regulated in colon cancer.42 The primary role of this transporter in colon is to supply SCFAs to colon epithelial cells from the apical side to support their function. Sodium-coupled monocarboxylate cotransporter is reported to also be expressed in the small intestine and kidney but not in the liver. OAT7 is, therefore, the first transporter that could contribute to the hepatic uptake of SCFAs, particularly butyrate. Taking the role of hepatocytes for butyrate metabolism into account, OAT7 likely functions as an entrance pathway for butyrate at the sinusoidal membrane of hepatocytes under the physiological condition.

Figure 9.

Proposed model of the estrone sulfate/butyrate exchange at the sinusoidal membrane via OAT7. Estrone sulfate (ES) formed in the hepatocyte is exported into the blood plasma in exchange for butyrate using inwardly directing butyrate concentration gradient via OAT7 in the sinusoidal membrane. OAs: organic anions. For further explanation, see text.

This transport direction, the exit of ES in exchange for butyrate entering hepatocytes via OAT7, seems appropriate for steroid hormone metabolism in liver (Fig. 9). ES is the most abundant circulating estrogen, at concentrations approximately 10-fold higher than unconjugated estrone. ES exhibits a much longer half-life than the parent estrogens. Sulfated steroid hormones are widely believed to serve an important biological role as steroid hormone precursors or reservoirs for steroid hormone-responsive tissues.43 Liver is proposed to be a major site of steroid sulfation. ES, produced by the liver cytosolic sulfotransferase, is secreted not only into the bile but also into the systemic circulation.3 This feature of OAT7 as an ES/butyrate exchanger suggests its different physiological role from OAT2, which also resides on the basolateral membrane (He X et al., manuscript in preparation). Because OAT2 is organic anion/dimethyldicarboxylate exchanger, as Kobayashi et al.19 reported, it functions as an uptake pathway for organic anions driven by an outwardly directed dicarboxylate gradient.

Many primary active and secondary active transporters have been identified in liver, and their transport properties have been analyzed. Some of them are thought to be responsible for sinusoidal transport of sulfate conjugates in human liver. These include OAT2, OATPs,5 and MRP4.20 In addition to their transport mode, the affinity for ES is different between OAT2 and OAT7. In OAT7, Kms for ES are 8.7 μM (oocytes) and 40.7 μM (S2-OAT7) possibly because of the difference of expression systems, that is, mammalian expression system versus Xenopus oocytes expression system. In contrast, we could not determine the Km for OAT2 in S2-hOAT2 cells because of its low affinity against ES (Anzai N et al., unpublished observation). Among members of OATP family, OATP-B,7, 8 OATP-C,9–11 and OATP812, 13 mediate the transport of steroid sulfate and function as an entrance pathway in their vectorial transport in combination with canalicular efflux transporter MRP2.20 The apparent Km for ES and DHEAS obtained in this study were comparable to those of OATPs. However, OAT7 is distinct from OATPs in its substrate selectivity. Bile acids such as cholate and taurocholate, several glucuronide conjugates, and glutathione conjugates, which are preferentially transported by OATPs, are not transported by OAT7 (Fig. 5B). Among sinusoidal MRPs, MRP4 mediates the transport of steroid sulfates44 as well as other organic anions such as prostagladins45 and cyclic nucleotides.46 Thus, the narrow substrate selectivity of OAT7 might explain the efficient translocation of certain sulfate conjugates without interference from the other anionic compounds such as bile salts.


We thank Dr. A. Bahn for helpful discussions. The anti-OAT7 polyclonal antibody was supplied by Transgenic Inc., Kumamoto, Japan.