A novel human organic anion transporting polypeptide localized to the basolateral hepatocyte membrane

Jörg König, Yunhai Cui, Anne T. Nies, Dietrich Keppler

Abstract

We cloned and expressed a new organic anion transporting polypeptide (OATP), termed human OATP2, (OATP-C, LST-1; symbol SLC21A6), involved in the uptake of various lipophilic anions into human liver. The cDNA encoding OATP2 comprised 2073 base pairs, corresponding to a protein of 691 amino acids, which were 44% identical to the known human OATP. An antibody directed against the carboxy terminus localized OATP2 to the basolateral membrane of human hepatocytes. Northern blot analysis indicated a strong expression of OATP2 only in human liver. Transport mediated by recombinant OATP2 and its localization were studied in stably transfected Madin-Darby canine kidney strain II (MDCKII) and HEK293 cells. Confocal microscopy localized recombinant OATP2 protein to the lateral membrane of MDCKII cells. Substrates included 17β-glucuronosyl estradiol, monoglucuronosyl bilirubin, dehydroepiandrosterone sulfate, and cholyltaurine. 17β-Glucuronosyl estradiol was a preferred substrate, with a Michaelis-Menten constant value of 8.2 μM; its uptake was Na+ independent and was inhibited by sulfobromophthalein, with a inhibition constant value of 44 nM. Our results indicate that OATP2 is important for the uptake of organic anions, including bilirubin conjugates and sulfobromophthalein, in human liver.

  • bilirubin conjugates
  • 17β-glucuronosyl estradiol
  • hepatic transport
  • SLC21A6
  • sulfobromophthalein

the removal of endogenous and xenobiotic substances from blood is one major function of the liver, and a number of different uptake and export systems are involved in this process. In contrast to the canalicular membrane domain, where mostly export pumps are located (22, 41), the basolateral membrane exhibits a variety of uptake transporters (34), in addition to some export pumps (24, 25). On the basis of transport properties and sequence similarities, at least two different families of uptake transporters are present in the basolateral hepatocyte membrane. The Na+-taurocholate cotransporting polypeptides cloned from human (12) and rat (13) mediate the uptake of bile salts in a Na+-dependent manner. In contrast to the Na+-taurocholate cotransporting polypeptides, members of the organic anion transporting polypeptide (OATP)/prostaglandin transporter family differ much more with respect to their substrate specificity and tissue distribution. Until now, eight members of this family have been cloned: human OATP (28), rat OATP1 (15), rat OATP2 (35), rat OATP3 (1), the prostaglandin transporters human PGT (31, 32) and rat PGT (21), and the two kidney organic anion transporters OAT-K1 (36) and OAT-K2 (33). All of these Na+-independent carrier systems exhibit a broad substrate specificity. As determined in several expression systems, transported substrates include 17β-glucuronosyl estradiol (E217βG) (7, 20), sulfobromophthalein (BSP) (15, 19, 28, 39), estrone 3-sulfate, ochratoxin A (23), dehydroepiandrosterone sulfate (DHEAS; 27), and cholyltaurine (19, 28). From all members, rat OATP1 and human OATP are the ones best characterized with respect to their transport properties using the Xenopus laevis oocyte expression models and stably transfected mammalian cells. Rat OATP1 (originally termed OATP) has been localized to the basolateral membrane of hepatocytes as well as to the apical membrane domain of the S3 segment of the kidney proximal tubule epithelia (5). In addition, OATP1 is present in the choroid plexus of rat brain (4). Human OATP is highly expressed in brain, in addition to lung, liver, kidney, and testis (28). Although human OATP was cloned on the basis of sequence information obtained from rat OATP1 and both proteins share a similar substrate specificity, marked differences were found with respect to transport rates and apparent Michaelis-Menten constant (K m) values for different substrates (7, 34). Together with the different expression pattern, these findings suggested that human OATP is related but not orthologous to rat OATP1. Therefore, we searched the expressed sequence tag (EST) library to obtain sequence information on as yet unknown members of the OATP family. We cloned the full-length cDNA encoding a new human OATP-related protein and localized it to the basolateral membrane of human hepatocytes. Two stably transfected mammalian cell lines served to determine transport characteristics of the recombinant protein. On the basis of amino acid sequence identity and the identified substrates, we named this new member of the human organic anion transporter family OATP2 (SLC21A6).

MATERIALS AND METHODS

Materials.

Pepstatin, leupeptin, aprotinin, agar, FCS, and the protein standard mixture (relative molecular weight of 26,600–180,000) for the SDS-PAGE were from Sigma (Deisenhofen, Germany). Lysozyme and ampicillin were from Boehringer Mannheim (Mannheim, Germany); agarose was from Roth (Karlsruhe, Germany). RNase inhibitor (RNAguard), Stratascript Moloney murine leukemia virus reverse transcriptase, and restriction endonucleases were from Stratagene (Amsterdam, The Netherlands). Marathon-Ready cDNA, Advantage cDNA polymerase mix, and the human 12-lane multiple tissue Northern blot (MTN) and β-actin primers were from Clontech (Heidelberg, Germany). [3H]E217βG (2.0 TBq/mmol), [3H]cholyltaurine (taurocholate) (0.13 TBq/mmol), and [3H]DHEAS (0.59 TBq/mmol) were obtained from DuPont NEN (Boston, MA). [3H]monoglucuronosyl bilirubin (MGB; 0.55 TBq/mmol) was synthesized as described earlier (16, 18).

Antibodies.

The ESL antibody was raised in rabbits against the 21 amino acids at the carboxy terminus of the deduced OATP2 sequence (ESLNKNKHFVPSAGADSETHC). The peptide was synthesized automatically, coupled to maleimide-activated keyhole limpet hemocyanin, and the rabbits were immunized with this conjugate. The mouse monoclonal antibody OKT9 against the human transferrin receptor (42) was a kind gift of Dr. R. Tauber (Berlin, Germany). The mouse monoclonal antibody against desmoplakin (cocktail) was purchased from Progen (Heidelberg, Germany). The mouse monoclonal antibody CD26 against human dipeptidylpeptidase IV was from Pharmingen (San Diego, CA). Cy2-conjugated goat anti-rabbit and Cy3-conjugated goat anti-mouse antibodies were from Dianova (Hamburg, Germany).

Cloning of the human OATP2 cDNA.

On the basis of the sequence information of a 392-bp cDNA clone (EMBL/GenBank accession number H62893), which exhibits 70.4% identity to human OATP over 240 aligned base pairs, one reverse primer was designed and subjected to a 5′ rapid amplification of cDNA ends (RACE) reaction using the Marathon-Ready cDNA kit (Clontech). In detail, a gene-specific primer was designed (oOATP2.rev 5′-CCATGAAGAAATGTGGCAAAGCA-3′) and the PCR was performed in a volume of 50 μl containing 5 μl Marathon-Ready cDNA, 0.2 μM sense primer AP1 (delivered with the kit), 0.2 μM anti-sense primer oOATP2.rev, 5 μl of 10× PCR buffer (400 mM tricine-KOH, pH 9.2, 150 mM potassium acetate, 35 mM magnesium acetate, 750 mg/l BSA), 0.2 mM deoxynucleoside triphosphates, and 1 μl of 50× Advantage cDNA polymerase mix (Clontech) [50% glycerol, 40 mM Tris ⋅ HCl, pH 7.5, 50 mM KCl, 25 mM (NH4)2SO4, 0.1 mM EDTA, 5 mM β-mercaptoethanol, 0.25% Thesit, 1.1 μg/μl TaqStart antibody, and KlenTaq-1 DNA polymerase] under the following PCR cycling conditions: 2 min denaturation at 94°C followed by 5 cycles of 10-s denaturation at 94°C, annealing/elongation for 3 min at 70°C, 5 cycles with 10-s denaturation at 94°C, 3 min annealing/elongation at 68°C, and, subsequently, 25 cycles with 10-s denaturation at 94°C and annealing/elongation for 3 min at 66°C. The reaction was finished by 10 min at 72°C. The amplified fragment was subcloned into pCR2.1.TOPO (Invitrogen, BV, Groningen, The Netherlands) and sequenced. On the basis of this sequence information, a forward primer was designed in the 5′-untranslated region in front of the ATG codon and used in a 3′-RACE reaction to amplify the completeOATP2 cDNA. This 3′-RACE reaction was also performed using the Marathon-Ready cDNA kit according to the manufacturer's instructions with the gene-specific OATP2 primer OATP2.5′ for (5′-TTGTTTCAAACTGAGCATCAACAAC-3′) under the same PCR cycling conditions as described for the 5′-RACE. The amplified fragment of ∼2.7 kbp was subcloned into the vector pCR2.1.TOPO (Invitrogen), resulting in the plasmid pOATP2.TOPO, which was sequenced by 4Base Lab (Reutlingen, Germany). For subcloning of theOATP2 cDNA into the expression vector pcDNA3.1(+) (Invitrogen), the plasmid pOATP2.TOPO was digested with BstX I and theOATP2 fragment was cloned into the BstX I-digested and dephosphorylated vector pcDNA3.1(+), resulting in the plasmid pOATP2.31. On completion of the plasmid, the correctness of the restriction sites and the orientation of the cDNA were verified by sequencing.

Northern blot analysis.

The Northern blot analyses were performed using the commercial human 12-lane MTN blot (Clontech). For the β-actin control, a human β-actin cDNA fragment, supplied with the Northern blot, and forOATP2, the 659-bp Hind III restriction fragment (bp 214–872) were used as probes. The membrane was prehybridized for 2 h at 42°C in 10 ml of hybridization buffer [6× sodium chloride-sodium citrate (SSC), 0.5% SDS, 5× Denhardt's solution, 50% formamide, 100 μg/ml denatured salmon sperm DNA] without and hybridized with the labeled DNA fragments for 18 h under the same conditions. Nick translation was performed by use of the Rediprime DNA labeling system (Amersham-Pharmacia, Freiburg, Germany) according to the manufacturer's instructions. The labeled DNA fragments were purified using NucTrap probe purification columns (Stratagene). After hybridization, the membrane was washed once in 2× SSC-0.1% SDS for 20 min at 42°C, once in 1× SSC-0.1% SDS, and once in 0.5× SSC-0.1% SDS. Both washing steps were carried out for 20 min at 55°C. The blot was air dried, and autoradiography was performed at −80°C with an intensifying screen for 24 h (OATP2) and 18 h (β-actin).

DNA sequencing.

With the use of the T7 sequencing kit from Amersham-Pharmacia and [α-35S]dATP, the cDNA clones were sequenced according to the dideoxynucleotide chain termination method of Sanger et al. (37). Dried gels were exposed to Kodak BioMax MR-1 films (Sigma).

Computer analysis.

The HUSAR program (38), based on the Wisconsin Genetics Computer Group program package (10), was used during this study for restriction mapping, sequence analyses, and sequence alignments.

Cell culture and transfection studies.

HEK293 (human embryonic kidney) and Madin-Darby canine kidney strain II (MDCKII) cells were cultured in minimum essential medium (Sigma), supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin, at 37°C and 5% CO2.

Cells were transfected using the polybrene (hexadimethrine bromide) method (8). Briefly, exponentially growing cells were incubated in a 10-cm petri dish with 10 μg plasmid DNA and 30 μg polybrene in 3 ml of complete medium for 8 h under normal culture conditions. Cells were then incubated with 5 ml of 30% DMSO in complete medium at room temperature for 5 min. The DMSO mixture was then removed, and cells were washed twice with complete medium and cultured overnight before starting Geneticin (G418) selection. After 3 wk of G418 selection (600 μg/ml), single colonies were screened for OATP2 expression by immunoblot analysis and immunofluorescence microscopy. Expression of recombinant OATP2 was further enhanced by culturing transfected cells with 10 mM sodium butyrate (9).

Preparation of membrane vesicles.

Sinusoidal membrane vesicles from human liver (6) and crude membrane fractions from cultured cells were prepared as described earlier (9).

Tissue samples and immunofluorescence studies.

Tissue samples for immunofluorescence studies were obtained perioperatively as described recently (25). Moreover, frozen sections for immunofluorescence microscopy were prepared as described in detail (25). All antibodies were diluted with PBS (140 mM NaCl, 10 mM phosphate, pH 7.4) supplemented with 5% FCS at the following dilutions: ESL at 1:100, OKT9 and CD26 at 1:50, anti-desmoplakin (cocktail) at 1:20, and Cy2-conjugated anti-rabbit IgG and Cy3-conjugated anti-mouse IgG at 1:400. Fluorescence microscopy was performed on an Axiovert S100TV microscope (Carl Zeiss, Jena, Germany) equipped with a video camera (Hamamatsu Photonics, Hamamatsu, Japan). Captured files were analyzed with the Openlab imaging software (Improvision, Coventry, UK).

Immunoblot analysis.

Membrane fractions were diluted with sample buffer and incubated at 37°C for 30 min before separation on 4% stacking and 10% resolving SDS polyacrylamide gels. Immunoblotting was performed using a tank blotting system from Bio-Rad (Munich, Germany) and enhanced chemiluminescence detection (NEN Life Science Products, Boston, MA). Primary antibody (ESL) was diluted 1:5,000 in 10 mM Tris ⋅ HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween 20. The secondary antibody was a horseradish peroxidase-conjugated goat anti-rabbit IgG (Bio-Rad) used at a 1:2,000 dilution.

Deglycosylation.

Membrane proteins (30 μg) were denatured by incubation with 1% SDS in a total volume of 10 μl at 37°C for 30 min. The denatured proteins were added to 100 μl of digestion buffer (17 mM NaH2PO4, 33 mM Na2HPO4, 0.2 mM NaN3, 5 mM EDTA, and 1.5% N-octylglucoside at pH 7.5) in the presence or absence of 5 units of peptideN-glycosidase F (EC 3.5.1.52). After overnight incubation at 37°C, the samples were analyzed by immunoblot analysis.

Immunofluorescence microscopy of transfected cells.

Transfected HEK293 or MDCK cells were grown polarized on Transwell membrane inserts (pore size of 3 μm; Costar, Cambridge, MA). Sodium butyrate was added to the culture medium 24 h before the experiment (9). After fixation with 4% paraformaldehyde in PBS for 10 min and permeabilization in 1% Triton X-100 in PBS for 10 min, cells were incubated with the polyclonal antibody ESL (diluted 1:50 in PBS) for 30 min at room temperature. Cells were then washed three times with PBS and incubated with Cy2-conjugated goat anti-rabbit IgG (diluted 1:200 in PBS) for 30 min at room temperature. Nuclei were stained with propidium iodide (0.2 μg/ml) added into the dilution of the secondary antibody. Membranes were cut from the inserts and mounted onto slides with 50% glycerol in PBS. Confocal laser-scanning immunofluorescence microscopy was performed using a LSM-410 apparatus from Carl Zeiss as described previously (9).

Transport assays.

Transfected HEK293 cells were seeded in six-well plates (coated with 0.1 mg/ml poly-d-lysine) at a density of 106cells per well and cultured with 10 mM sodium butyrate for 24 h. For uptake studies, cells were first washed with uptake buffer (142 mM NaCl, 24 mM NaHCO3, 5 mM KCl, 1 mM KH2PO4, 1.2 mM MgSO4, 1.5 mM CaCl2, 5 mM glucose, and 12.5 mM HEPES, pH 7.3) and then incubated with 1 ml of uptake buffer containing the tritium-labeled substrate. For inhibition studies, inhibitors were included in the uptake buffer at different concentrations. After incubation for 3 min at 37°C, the substrate was removed, and cells were washed three times with cold uptake buffer before they were lysed with 1 ml of 0.1% SDS in water. The cell-associated radioactivity was determined by transferring 250-μl aliquots of the lysate to scintillation vials and counting radioactivity using a Beckman scintillator (LS 6000IC, Beckman Instruments, Munich, Germany). Protein content was determined according to Lowry using 100 μl of lysate. Uptake was measured at concentrations between 1 μM and 10 μM E217βG for determination of K m and maximal velocity (V max) values.

RESULTS

Cloning of the cDNA encoding human OATP2 (SLC21A6).

A GenBank search against the human OATP cDNA sequence revealed an EST sequence (EMBL/GenBank accession number H62893) that exhibits 70.4% identity over 240 aligned base pairs. On the basis of this sequence information, the complete cDNA encoding the human OATP2 protein was cloned by a combination of a 5′-RACE and a 3′-RACE PCR. The coding region covers 2073 bp, and the region around the ATG translational start codon fulfilled the Kozak consensus sequence (26) with an invariant A at position −4 and an invariant T at position −2. The nucleotide base pair identities of the coding region to human OATP (28), rat OATP1 (15), rat OATP2 (35), and rat OATP3 (1) were 60.7%, 60.1%, 57.6%, and 58.8%, respectively, calculated under default parameters of the BESTFIT program of the HUSAR program package (38).

Analysis of the deduced amino acid sequence of OATP2.

The open reading frame of 2073 bp encodes a deduced protein of 691 amino acids with a calculated molecular mass of 76,448 Da. The amino acid identity to the previously known human OATP isoform, calculated under default parameters of the BESTFIT program of the HUSAR program package (38), is 44%, and is 44%, 46%, and 46% to the rat orthologs OATP1, OATP2, and OATP3, respectively. An alignment of these four peptide sequences is shown in Fig. 1. All OATP isoforms represent glycoproteins with native molecular masses of ∼80 kDa. A computer-aided transmembrane (TMHMM) analysis (40) based on a CLUSTAL alignment of all OATPs demonstrates that they consist of 12 predicted transmembrane domains (Fig.2) with both ends of the protein located intracellulary. From 11 potential N-glycosylation sites in the predicted OATP2 amino acid sequence, six are located outside in predicted extracellular loops. In comparison, human OATP exhibits eight potential N-glycosylation sites (28) with seven of them located in extracellular loops and three of them at identical positions as the predicted N-glycosylation sites of human OATP2.

Fig. 1.

Alignment of human organic anion transporting polypeptide OATP2 (symbol SLC21A6; EMBL/GenBank accession number AJ132573), human OATP1 (previously termed OATP; SLC21A3; U21943), rat OATP1 (L19031), rat OATP2 (U88036), and rat OATP3 (AF041105). Sequences were aligned by the CLUSTAL program from the HUSAR program package; identity scores relative to OATP2 are calculated by the BESTFIT program out of the same package. Areas highlighted in black indicate the amino acids that are identical to the amino acid of the deduced OATP2 sequence at this position. Numbers indicate the amino acid position corresponding to OATP2. Peptide sequence recognized by the polyclonal antibody ESL is indicated by the dashed line.

Fig. 2.

Probability for transmembrane helices 1 to 12 in human OATP2 (A) and OATP (or OATP1; B) predicted by the TMHMM program of the Center for Biological Sequence Analyses (http://www.cbs.dtu.dk/services/TMHMM-1.0). Numbers in bars represent number of transmembrane helices.

Tissue distribution of human OATP2.

The tissue distribution of human OATP2 was studied by Northern blotting using the human 12-lane MTN blot and an OATP2 cDNA fragment under stringent hybridization and washing conditions. This cDNA fragment shows only 65% identity to the known human OATP, excluding cross-reactivity with this mRNA. In contrast to humanOATP (or OATP1), which is highly expressed in human brain, kidney, liver, and testis (28), strong signals for humanOATP2 were only detected in liver (Fig.3). The length of the detected mRNA species was 2.8 kb, likely to correspond to the fully spliced OATP2mRNA, and 4.5 kb, probably corresponding to a partially or unspliced mRNA. Prolonged exposure of the blot for up to 96 h revealed no additional signals in other tissues, suggesting that humanOATP2 is almost exclusively expressed in human liver. Additional tissues not tested in our Northern blotting but expressingOATP2 may still be identified.

Fig. 3.

Analysis of OATP2 expression in different human tissues studied by Northern blotting. A human 12-lane multiple-tissue Northern (MTN) blot was hybridized using a 659-bp Hind III restriction fragment as probe. For β-actin, the cDNA fragment supplied with the MTN blot was used. OATP2 Northern blot was performed under stringent conditions to prevent detection of other OATP family members.

Stable expression of recombinant OATP2 in mammalian cells.

OATP2 cDNA was transfected into HEK293 and MDCKII cells, and sodium butyrate was used to enhance the expression of recombinant OATP2 in these cells (9). Expression of OATP2 was verified by immunoblot analysis. As a positive control, we used a preparation of human liver basolateral membranes. The polyclonal antibody ESL detected two major bands in human liver samples (Fig. 4), one of ∼84 kDa and one of ∼58 kDa. The specificity of the ESL antibody was demonstrated by the comparison of OATP2-transfected cells with vector-transfected cells (Fig. 4). Immunoreactive bands with a molecular mass of ∼84 kDa were observed using the ESL antibody in both MDCKII and HEK293 cells transfected with OATP2 cDNA. No specific signals were observed in vector-transfected cells (Fig. 4). Deglycosylation experiments using peptide N-glycosidase F demonstrated that the relative molecular mass of the recombinant protein was reduced to ∼58 kDa, indicating that the lower band detected in human liver corresponds to the unglycosylated protein and that the 84-kDa OATP2 is indeed a glycoprotein.

Fig. 4.

Immunoblot analysis of OATP2. Basolateral membranes from human liver (BLM, 5 μg protein) and membrane fractions from Madin-Darby canine kidney strain II (MDCKII) cells (40 μg) transfected with OATP2 (MDCK-OATP2) or vector (MDCK-Co) and from HEK293 cells (20 μg) transfected with OATP2 (HEK-OATP2) or vector (HEK-Co) were separated by SDS-PAGE (10% separating gel). OATP2 was detected by the polyclonal antibody ESL (dilution of 1:5,000) directed against the carboxy terminus of human OATP2.

Immunolocalization of OATP2 in liver and transfectants.

Incubation of cryosections from human liver with the ESL antibody revealed fluorescent staining of the lateral and the basal membrane domains of hepatocytes (Fig. 5, Aand D). No expression of OATP2 was detectable in bile ductular cells (not shown), as examined by comparative localization of cytokeratin 19 as a marker for bile ductular cells (3). Strong staining was observed in all liver specimens studied. Localization of OATP2 in the basolateral membrane was confirmed by double labeling of cryosections with anti-desmoplakin (Fig. 5, E and F) and anti-transferrin receptor antibodies (not shown) as markers for the lateral and basal domains, respectively. OATP2 was absent from the canalicular domain (Fig. 5, A–C) as shown by double labeling of liver sections with the ESL antibody and an antibody directed against dipeptidylpeptidase IV, which is localized to the apical membrane of hepatocytes (14). No plasma membrane staining was observed when cryosections were incubated with the pre-ESL immune serum.

Fig. 5.

Immunofluorescence localization of OATP2 in human liver. Cryosections (3–5 μm) of normal human liver were incubated with the ESL antibody against human OATP2, resulting in a basolateral membrane staining (A and D). Absence of OATP2 from the apical plasma membrane domain was confirmed by double labeling of cryosections with the ESL antibody (A) and the CD26 antibody against dipeptidylpeptidase IV (DPPIV; B); C shows the merged picture. Lateral membrane domains were identified by double labeling with the ESL antibody (D) and desmoplakin (DP; E);F shows the merged picture. Bar = 10 μm.

The cellular localization of recombinant OATP2 was studied using confocal immunofluorescence laser scanning microscopy. In both MDCKII and HEK293 cells, OATP2 was sorted to the plasma membrane (Fig.6). In polarized MDCKII cells, OATP2 localization was restricted to the lateral membrane, consistent with its basolateral localization in hepatocytes (Fig. 5). No staining could be observed in vector-transfected MDCKII and HEK293 cells. The transfected HEK293 cells were used for the functional characterization of OATP2 because of the higher expression level compared with the transfected MDCKII cells.

Fig. 6.

Immunolocalization of recombinant OATP2 in stably transfected cells. HEK293 (HEK-OATP2, A and C) and MDCKII (MDCK-OATP2,B and D) cells transfected with OATP2 cDNA were stained with the polyclonal antibody ESL directed against the carboxy terminus of human OATP2 (green fluorescence). Nuclei were stained with propidium iodide (red fluorescence). A and B: 0.8-μm optical sections in the xy-plane. C and D: vertical sections in the xz-plane indicated by the yellow lines in A and B. In both cell lines, OATP2 was sorted to the plasma membrane (green fluorescence). In polarized MDCK cells, OATP2 staining was restricted to the lateral membrane.

Functional characterization of OATP2.

Several candidate substrates were tested using transfected cells in whole cell uptake assays. E217βG, MGB, cholyltaurine, and DHEAS were characterized as substrates for OATP2 (Table1). Uptake of E217βG mediated by OATP2 was Na+ independent. No significant reduction of the uptake rate could be observed when Na+ was replaced by choline. Uptake of E217βG was temperature dependent. At 4°C, the uptake rate of E217βG was reduced to 7% of that measured at 37°C. OATP2-mediated uptake of E217βG was time dependent (linear up to 5 min) and saturable (Fig. 7). ApparentK m values (8.2 μM) and V maxvalues (48 pmol ⋅ mg protein 1 ⋅ min 1) were determined by double-reciprocal plots according to Lineweaver and Burk (30).

View this table:
Table 1.

Substrate specificity of human OATP2 (SLC21A6)

Fig. 7.

OATP2-mediated 17β-glucuronosyl estradiol (E217βG) uptake. A: time course of E217βG uptake. For uptake measurement, HEK293 cells transfected with OATP2(HEK-OATP2) or vector (HEK-Co) were incubated with 1 μM E217βG at 37°C. At the indicated time points, intracellular radioactivity was determined as described inmaterials and methods. B: kinetics of OATP2-mediated E217βG uptake was measured for 3 min at the concentrations indicated. Net OATP2-mediated uptake was calculated by subtracting values obtained with HEK-Co cells from those obtained with HEK-OATP2 cells. For determination of the inhibition constant value for sulfobromophthalein (BSP), the same experiment was performed in the presence of 50 nM BSP. Michaelis-Menten constant and maximal velocity values were determined by double-reciprocal plots according to Lineweaver and Burk (30). All data represent means ± SD of 2 experiments performed in triplicate.

Inhibition of OATP2-mediated E217βG uptake.

Cis inhibition studies of E217βG uptake are summarized in Table 2. Extracellular glutathione and 2-oxoglutarate (tested as exchange substrates at concentrations up to 5 mM) had no significant effect on E217βG uptake. The inhibitor with the highest potency detected so far was BSP, with an IC50 value of 50 nM. Complete inhibition was observed at 100 nM BSP (Table 2). As shown in Fig. 7 B, the inhibition of E217βG uptake by BSP was competitive and exhibited an inhibition constant value of 44 nM.

View this table:
Table 2.

Cis inhibition of 17β-glucuronosyl estradiol uptake in human OATP2-transfected HEK293 cells

DISCUSSION

Our study describes a new member of the OATP gene family, which was cloned on the basis of sequence similarities to human OATP. The protein encoded by this gene, now termed human OATP2, symbol SLC21A6, shows between 44% and 46% amino acid identity to the previously known members of the OATP family: human OATP, rat OATP1, rat OATP2, and rat OATP3 (Fig. 1). On the nucleotide level, the identity score increases to between 58% and 60%, as calculated for the coding region of the gene. Analysis of the deduced amino acid sequence revealed several structural similarities between OATP2 and the other members of the OATP family. Human OATP1 (previously termed OATP) has 10–12 putative membrane-spanning domains (28). We compared the putative transmembrane domain organization of human OATP1 and OATP2, calculated by the TMHMM program package (40) (Fig. 2). This analysis revealed a very similar membrane domain organization for both transporters with 12 predicted membrane-spanning domains for each. On the basis of this putative membrane domain organization, OATP1 exhibits seven extracellular N-glycosylation sites and OATP2 exhibits six N-glycosylation sites with three of them at identical positions to those in OATP1. Nevertheless, OATP2 exhibits features that are so far unique in the OATP family. The amino terminus showed an extension of 6 amino acids, and the length of the protein with 691 amino acids in comparison with the typical length of the OATPs of ∼670 amino acids introduced some gaps in the CLUSTAL sequence alignment (Fig. 1). However, this CLUSTAL alignment also demonstrates the high sequence homology of OATP2 to the other OATP family members. On the basis of these sequence similarities and similar structural features and transport properties, we named this protein, encoded by the newly cloned gene human OATP2 as the second human OATP.

By Northern blot analysis we determined the tissue distribution of human OATP2 mRNA (Fig. 3). In contrast to human OATP1and all known rat OATPs, OATP2 mRNA exhibits a predominant and apparently exclusive expression in human liver. No other tissue studied so far revealed a hybridization signal forOATP2 mRNA even after prolonged exposure of the blot. HumanOATP1 is highly expressed in brain, kidney, lung, testis, and liver (28). Rat OATP1 exhibits a high expression in hepatocytes and, in addition, in kidney, brain, skeletal muscle, and colon (4, 15). Rat OATP2 was cloned from rat brain (35) but is also expressed in retina and liver. Rat OATP3 is exclusively expressed in retina and kidney (1). All tissue distribution analyses were performed by Northern blot experiments. Despite the fact that there can be mRNA expression below the detectability of Northern blot analysis, humanOATP2 is the first OATP family member that is expressed only in liver.

On the basis of the deduced amino acid sequence, we designed an antibody, termed ESL, directed against the carboxy terminus of the protein (Fig. 1). The specificity of this antibody was tested by immunoblot analysis (Fig. 4). This antibody served to localize the protein in human liver as well as the recombinant protein expressed in transfected cells. Immunofluorescence microscopy demonstrated that the OATP2 protein is exclusively localized to the basolateral membrane of hepatocytes (Fig. 5). This localization is consistent with the localization described for the other OATP family members (11, 17, 29) except rat OATP1, which also exhibits an additional apical localization in kidney (5) and brain (4).

We used our transfectants for further characterization of the recombinant protein and for its comparison with the native protein with respect to localization and transport properties. Several studies have addressed the transport function and substrate specificity of OATP family members on the basis of transport experiments in Xenopus laevis oocytes (1, 28, 35). In our study, we used mammalian cells stably transfected with the OATP2 cDNA. The use of Xenopus laevis oocytes may be more artificial than studies using mammalian cells expressing the recombinant protein, and kinetic studies carried out in Xenopus laevis oocytes may be affected by the different protein and lipid environments. In OATP2 cDNA-transfected HEK293 cells, the protein was localized to the plasma membrane, whereas, in polarized MDCKII cells transfected with the OATP2cDNA, this plasma membrane localization was restricted to the lateral membrane (Fig. 6). This localization is in accordance with the localization found in human liver (Fig. 5) and established for other OATPs including rat OATP1 (11). Apical staining was observed neither in human hepatocytes nor in the polarized MDCKII transfectants. No expression of OATP2 was detectable in bile ductular cells.

With the use of the HEK293 transfectants, several substances were tested as possible substrates for OATP2. As shown in Table 1, typical organic anions, transported by other members of the OATP family (for review, see Ref. 3), were also substrates for human OATP2. Substances transported by OATP2 included MGB and E217βG. The latter was the preferred substrate, with a K m value of 8.2 μM (Fig. 7). This K m value was in the same range as the one determined for rat OATP1 after stable transfection (11). Other important physiological organic anions such as cholate and BSP were cis inhibitors of E217βG uptake, suggesting that they are also substrates for human OATP2 (Table 2). More studies on additional transport characteristics of human OATP2 are needed.

After completion of our studies and during preparation of this manuscript, a paper was published that describes a new transport protein, termed LST-1, the liver-specific transporter 1 (2). Localization of LST-1 in liver was not determined, but the sequence comparison of OATP2 and LST-1 revealed 99.8% identity on the amino acid level. LST-1 was expressed in Xenopus laevis oocytes and shown to transport organic anions, including DHEAS, E217βG, and thyroid hormones (2).

In conclusion, we have cloned human OATP2 as a new member of the OATP family and localized it to the basolateral hepatocyte membrane.OATP2 mRNA was strongly expressed in liver, and no other tissue tested so far showed detectable OATP2 mRNA expression. The use of stably transfected human embryonic kidney (HEK293) and canine kidney (MDCKII) cells allowed us to study the localization of the recombinant protein and to determine the substrate specificity of this transporter. Transported substrates included typical organic anions such as E217βG, DHEAS, cholyltaurine, and MGB, suggesting the involvement of OATP2 in the hepatocellular uptake of important endogenous organic anions. Further studies are needed to address the full range of substrates of this major hepatic transporter and to characterize the driving force of human OATP2.

Acknowledgments

We thank Dr. H. R. Rackwitz for synthesizing the peptide for antibody production, Dr. H. Spring for expert help in confocal laser scanning microscopy, Ulrike Buchholz for synthesizing 3H-labeled MGB, and Marion Pfannschmidt and Christina Weber for excellent technical assistance.

Footnotes

  • Address for reprint requests and other correspondence: J. König, Division of Tumor Biochemistry, Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany (E-mail:j.koenig{at}dkfz-heidelberg.de).

  • This study was supported in part by grants from the Deutsche Forschungsgemeinschaft through SFB 352 and SFB 601, Heidelberg, by the Forschungsschwerpunkt Transplantation, and by the Fonds der Chemischen Industrie, Frankfurt.

  • 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. §1734 solely to indicate this fact.

REFERENCES

View Abstract