In the present study we isolated two splice variants of organic anion transporting polypeptide 3A1 (OATP3A1_v1 and OATP3A1_v2) from human brain. OATP3A1_v2 lacks 18 amino acids (aa) at the COOH-terminal end (692 aa) but is otherwise similar in sequence to OATP3A1_v1 (710 aa). OATP3A1_v1 exhibits a wide tissue distribution, with expression in testis, various brain regions, heart, lung, spleen, peripheral blood leukocytes, and thyroid gland, whereas OATP3A1_v2 is predominantly expressed in testis and brain. On the cellular and subcellular levels OATP3A1_v1 could be immunolocalized in testicular germ cells, the basolateral plasma membrane of choroid plexus epithelial cells, and neuroglial cells of the gray matter of human frontal cortex. Immunolocalization of OATP3A1_v2 included Sertoli cells in testis, apical and/or subapical membranes in choroid plexus epithelial cells, and neurons (cell bodies and axons) of the gray and white matter of human frontal cortex. The rodent ortholog Oatp3a1 was also widely distributed in rat brain, and its localization included somatoneurons as well as astroglial cells. Transport studies in cRNA-injected Xenopus laevis oocytes and in stably transfected Chinese hamster ovary FlpIn cells revealed a similar broad substrate specificity for both splice variants. Transported substrates include prostaglandin (PG)E1 and PGE2, thyroxine, and the cyclic oligopeptides BQ-123 (endothelin receptor antagonist) and vasopressin. These studies provide further evidence for the involvement of OATPs in oligopeptide transport. They specifically suggest that OATP3A1 variants might be involved in the regulation of extracellular vasopressin concentration in human brain and thus might influence the neuromodulation of neurotransmission by cerebral neuropeptides such as vasopressin.
organic anion transporting polypeptides (Oatps/OATPs) are multispecific sodium-independent transport proteins that are expressed in a variety of tissues in vertebrate animal species (12). They comprise at least 36 members in human, rat, and mouse. On the basis of their phylogenetic relationships, all human and rodent Oatps/OATPs so far identified have been classified within the OATP/SLCO superfamily of solute carriers into six families (families OATP1 to OATP6; gene family symbols SLCO1 to SLCO6) and 13 subfamilies (subfamilies OATP1A/SLCO1A to OATP6D/SLCO6D) (11). In general, OATPs exhibit a broad substrate specificity and a wide tissue distribution, although both of these properties may be narrower for selective individual carriers. The best-characterized OATPs with the broadest substrate spectrum belong to the subfamilies OATP1A and OATP1B (11). They transport a wide variety of amphipathic organic solutes including bile salts, organic dyes, steroid conjugates, thyroid hormones, anionic oligopeptides, several drugs, and xenobiotics (12). The exact transport mechanism(s) of the OATPs has not yet been worked out. However, studies with rat Oatps suggest that they act as organic anion exchangers. For example, rat Oatp1a1 (previously called Oatp1) has been shown to mediate substrate uptake in exchange for bicarbonate (30) or glutathione (21), and rat Oatp1a4 (previously called Oatp2) also uses intracellular glutathione conjugates as a driving force (22). OATP1A2 (previously also called OATP-A) is mainly expressed in brain, although its mRNA has also been identified in a variety of other tissues including kidney, small intestine, liver, lung, and skeletal muscle (9, 18). OATP1B1 (previously called OATP-C, LST-1, or OATP2) and OATP1B3 (previously called OATP8) (1, 2, 14, 16, 17) are predominantly, if not exclusively, expressed in the liver. The other human OATPs so far identified have rather limited substrate specificities. OATP1C1 (previously called OATP-F) is a high-affinity thyroxine (T4) transporter that is predominantly expressed in brain and testis (28). OATP4C1 transports cardiac glycosides, thyroid hormones, methotrexate, and cAMP and is mainly expressed in the kidney (25). OATP2A1 (previously called PGT) transports prostaglandins (PGs) and is expressed in almost every organ of the human body (31). OATP2B1 (previously called OATP-B) mediates transport of sulfated steroid conjugates and also, at low pH, of bile acids and pravastatin (15, 18, 26). It is expressed in numerous tissues including liver, placenta, intestine, and brain (15, 19, 26, 33, 35, 36). OATP4A1 (previously called OATP-E) transports mainly thyroid hormones and exhibits a wide tissue distribution including heart, placenta, lung, liver, skeletal muscle, kidney, pancreas, brain, testis, and small intestine (8).
Finally, OATP3A1 (previously called OATP-D) has been identified and characterized mainly as a PG transporter with a very broad tissue expression profile (3, 35). A unique feature of human OATP3A1 is its 97% amino acid sequence identity with its rat and mouse orthologs (3, 11). This high conservation during evolution as well as its wide tissue distribution indicate that the SLCO3A1/Slco3a1 gene products might serve important physiological functions in mammalian species. In this study we have identified two alternatively spliced variants of OATP3A1, which have been called OATP3A1_v1 (= original OATP3A1) and OATP3A1_v2. The two OATP3A1 variants are compared with respect to their tissue distribution, cellular expression in brain, and functional transport properties. Interestingly, the two splice variants show a distinct cerebral expression pattern in human brain slices, while their substrate specificities are similar and include prostaglandin PGE1 and PGE2, T4, and the oligopeptides BQ-123 (endothelin receptor antagonist) and vasopressin.
MATERIALS AND METHODS
Mature Xenopus laevis frogs were purchased from the African Xenopus Facility, Knysna, Noordhoek, Republic of South Africa. They were kept in a constant alternating 12-h light (6:30 AM to 6:30 PM)-dark cycle and kept under standard conditions. All experiments were approved by the local regulatory authorities and were performed according to federal laws.
[prolyl-3,4(N)-3H]-BQ-123 (36 Ci/mmol) and [tyrosyl-3,5(N)-3H]-vasopressin[Arg8] (28 Ci/mmol) were purchased from Amersham Biosciences Europe (Otelfingen, Switzerland). 125I-labeled angiotensin I (5-l-isoleucine) (2,000 Ci/mmol), [18.104.22.168.22.214.171.124-3H(N)]-298-arachidonic acid (100 Ci/mmol), sodium-[1,2,6,7-3H(N)]-dehydroepiandrosterone sulfate (DHEAS, 60–74 Ci/mmol), [tyrosyl-3,5-3H]deltorphin II (2-d-Ala, 34–38.5 Ci/mmol), [3H(G)]digoxin (19–37 Ci/mmol), [tyrosyl-2,6-3H(N)]enkephalin (DPDPE, 34–40 Ci/mmol), ammonium-[6,7-3H(N)]-estrone-3-sulfate (57.3 Ci/mmol), [tyrosyl-2,6-3H]oxytocin (33 Ci/mmol), [5,6-3H(N)]-PGE1 (46–56 Ci/mmol), [5,6,8,11,12,14,15-3H(N)]-PGE2 (200 Ci/mmol), [3H(G)]taurocholic acid (3.5 Ci/mmol), l-[125I]-T4 (969 Ci/mmol), and l-3,5,3′-[125I]triiodothyronine (T3, 779 Ci/mmol) were obtained from Perkin Elmer Life Sciences (Boston, MA). All other reagents were of analytical grade and readily available from commercial sources.
Isolation of OATP3A1_v1 and OATP3A1_v2.
A human adult brain Rapid-screen cDNA library panel (OriGene Technologies, Rockville, MD) was screened by PCR using primers specific for OATP3A1 (GenBank accession no. AB031050) (forward primer 5′-GCTGAGAACGCAACCGTGGTTCC-3′, reverse primer 5′-GACTTGAGTTCAGGGCTGACTGTCC-3′). A single clone was isolated and sequenced with ALF Express (Amersham Biosciences Europe). This single clone contained a complete open reading frame (ORF) starting at the initiation codon and had a 3′-untranslated region (UTR) that was different from the published OATP3A1 sequence (35). This new variant 2 of OATP3A1 was called OATP3A1_v2, whereas the original variant 1 of OATP3A1 was renamed as OATP3A1_v1. To clone the complete ORF of the published OATP3A1_v1, two additional primers were designed. The forward primer covered bases 1927–1946 of the published sequence (5′-TCTCCTCCTTCGTTTGTTGG-3′), and the reverse primer included the stop codon (5′-CTATAAAACGGACTCCATGTT-3′). Total RNA from human kidney was used as a template for RT-PCR, which resulted in a 400-bp piece including the specific 3′-end region of the OATP3A1_v1 ORF.
In an additional PCR reaction, the ORF of OATP3A1_v1 was constructed with the forward primer containing the 5′-end of the ORF without the start codon (5′-CAGGGGAAGAAGCCGGG-3′), the reverse primer including the stop codon (5′-CTATAAAACGGACTCCATGTT-3′), and as templates the two cDNAs covering the whole ORF (i.e., the isolated OATP3A1_v2 and the RT-PCR 400-bp amplicon). PCR was performed with a proofreading DNA polymerase (Pfu DNA polymerase, Stratagene Cloning Systems, La Jolla, CA). The resulting PCR product was cloned into the Xenopus laevis expression vector, which contained the initiation codon and the 5′- and 3′-UTRs of Oatp1a1 (4). This construct was then used to prepare cRNA for oocyte expression experiments (see below). The ORF of OATP3A1_v2 was amplified in a similar way with the following primers: forward 5′-CAGGGGAAGAAGCCGGG-3′, reverse 5′-TCAGCTTCTCACAAAGGAGTCTT-3′.
Multiple tissue expression array.
A fragment of 164 bp of OATP3A1 (bases 1742–1905) was RT-PCR amplified with human kidney RNA and the following primers: forward 5′-GCTGAG AACGCAACCGTGGTTCC-3′, reverse 5′-GACTTGAGTTCAGGGCTGACTGTCC-3′. After cloning and sequencing, the fragment was used to hybridize a Multiple Tissue Expression (MTE) Array (BD Biosciences Clontech, Basel, Switzerland) containing normalized poly(A)+ RNA from different human tissues. After a 30-min prehybridization at 65°C in ExpressHyb (BD Biosciences Clontech) hybridization solution, the blot was hybridized overnight at 65°C (2 × 106 cpm/ml). After five washes for 20 min with 2× SSC-1% SDS at 55°C and two washes of 20 min with 0.1× SSC-0.5% SDS at 55°C, the blot was exposed to autoradiography film for 3 days at −70°C with an intensifying screen.
Determination of mRNAs.
One microgram of total RNA from each tissue (Clontech, Mountain View, CA) was used to generate cDNA with random primers, using the Reverse Transcription System from Promega (Madison, WI). The reaction time was 59 min at 59°C, followed by 5 min at 99°C. The cDNA was stored at −20°C until use. For the detection of the two splice variants, TaqMan real-time PCR with an ABI PRISM 7900 HT detector was performed according to the instructions of the manufacturer (AB Applied Biosystems, Rotkreuz, Switzerland) and following the method outlined in Ref. 32. Primers and probes designed with Primer Express Sequence Detection System Software (AB Applied Biosystems) were OATP3A1_v1: forward GAAAAACTATAAACGCTACATCAAAAACC, probe AGGGTCAGAGTAGAGGCAAAGAAC, reverse GAGGGCGGGCTGAGCACCAGTG; OATP3A1_v2: forward GAAAAACTATAAACGCTACATCAAAAACC, probe GGTTTTCTCAGTCTCAATGTCTTGGT, reverse CGGGCTGAGCACCAGCACAGAG. The probes for the two splice variants were labeled with FAM as reporter dye at the 5′-end and BHQ1 as quencher dye at the 3′-end. The probe for the 18S rRNA was labeled with VIC as reporter dye at the 5′-end and TAMRA as quencher dye at the 3′-end.
Preparation of rabbit polyclonal antibodies and Western blotting.
To produce specific antibodies against the two OATP3A1 splice variants, the following peptides corresponding to the COOH-terminal 15 amino acids of both proteins were purchased from Neosystem (Strasbourg, France): NH2-NLEDHEWCENMESVL-COOH for OATP3A1_v1 and NH2-CPESHSPSEDSFVRS-COOH for OATP3A1_v2. Polyclonal antibodies against both oligopeptides were raised in rabbits as described previously (34). The antibodies were affinity purified with the AminoLink Plus Immobilization Kit (Pierce Biotechnology, Rockford, IL) for OATP3A1_v1 and the SulfoLink Kit (Pierce Biotechnology) for OATP3A1_v2 according to the manufacturer's protocols. The specificity of the two antibody preparations was tested with cell membranes isolated from OATP-expressing cell lines such as Sf9 cells (i.e., OATP1A2, OATP3A1_v1, OATP3A1_v2, OATP1C1), Chinese hamster ovary (CHO) cells (i.e., OATP1B1, OATP1B3, OATP2B1) and human embryonic kidney (HEK) cells (i.e., OATP4A1) (see ⇓Fig. 2). Cell membranes were isolated by differential centrifugation. Confluent cell plates were washed three times with 0.9% NaCl, followed by a washing step with 0.25 mmol/l sucrose. Cells were scraped with a rubber policeman and resuspended in 5 mmol/l sucrose containing 2 mmol/l phenylmethylsulfonyl fluoride (PMSF) and 1 mg/ml antipain-leupeptin. Cells were homogenized and centrifuged for 10 min at 800g. The supernatant was centrifuged for 20 min at 8,500g and the resulting supernatant again for 1 h at 100,000g. The final pellet was resuspended in 250 mmol/l sucrose supplemented with the proteinase inhibitors PMSF and antipain-leupeptin. The protein content of the membrane fraction was determined by bicinchoninic acid assay. Western blots were performed as described previously (10), and reactions were visualized with the ECL Plus Western Blot Detection Kit (Amersham Biosciences).
Paraffin-embedded tissue sections (3 μm) of normal testis and choroid plexus were obtained from autopsy of a 83-yr-old man (Department of Pathology, Canton Hospital, Liestal, Switzerland) and a 64-yr-old woman (Department of Neuropathology, University Hospital, Zurich), respectively. Tissue sections were deparaffinized in xylol and dehydrated through a graded series of ethanol. Endogenous peroxidase was blocked with 0.3% H2O2 diluted in methanol for 15 min. Affinity-purified OATP3A1_v1 and OATP3A1_v2 rabbit antisera were applied at a dilution of 1:200 (Antibody Diluent, Dako, Glostrup, Denmark) at room temperature for 1 h. The specificity of the immunoreactivity was controlled for by preabsorption of the antibodies with OATP3A1_v1 (20 μmol/l) or OATP3A1_v2 (40 μmol/l) antigenic peptides. Sections were stained with an Envision+ peroxidase rabbit kit (K4002, Dako) for 30 min, and then washed three times with phosphate-buffered saline (PBS) and treated with diaminobenzidine (brown) as the chromogen (K3468, Dako) for 5 min. The staining reaction was terminated by washing the sections in distilled water for 5 min. Finally, the sections were counterstained with hematoxylin for 30 s, washed twice in water, dehydrated sequentially in 70%, 80%, 96%, and 100% ethanol (10 s each), incubated in xylol for 1 min, and mounted with commercial mounting medium (Dako).
Paraffin-embedded brain sections (5 μm) of human frontal cortex and hippocampal formation were obtained from the Department of Neuropathology, University Hospital Zurich. The specimens were from two patients, a 39-yr-old woman and a 79-yr-old man, without any evidence for central nervous disease. The postmortem periods until sampling of the brain tissue were between 30 and 48 h. The sections were first dewaxed with xylol and hydrated with descending serial ethanol solutions. They were then treated with 0.1 mol/l HCl for 10 min, followed by 0.1% saponin in PBS for 30 min. To eliminate endogenous peroxidase activity, the sections were incubated with 1.5% H2O2 for 10 min. Afterwards, sections were incubated overnight at 4°C with affinity-purified antibodies to OATP3A1_v1 (1:300) or OATP3A1_v2 (1:30) diluted in 50 mmol/l Tris-100 mmol/l NaCl (Tris-saline, pH 7.4) containing 2% normal goat serum and 0.05% Triton X-100. After being washed in Tris-saline, the sections were stained by the Avidin: Biotinylated enzyme Complex immunoperoxidase method with the Vectastain Elite Kit (Vector Laboratories, Burlingame, CA). Diaminobenzidine hydrochloride was used as the chromogen.
Double-immunofluorescence staining of rat brain tissue sections.
Rats were anesthetized with pentobarbital sodium (40 mg/kg) and perfused through the ascending aorta with 4% paraformaldehyde and 15% saturated picric acid in 0.15% phosphate buffer. Brains were removed, postfixed in the perfusion solution at 4°C for 2 h, and cryoprotected in 30% sucrose overnight at 4°C. Sections of 40 μm were cut with a sliding microtome and collected in PBS. They were incubated overnight at 4°C in PBS containing 2% normal goat serum, 0.2% Triton X-100, and the affinity-purified OATP3A1_v1 antibody (1:500) combined with a mouse monoclonal antibody either to glial fibrillary acidic protein (GFAP, 1:5,000, CHEMICON, Temecula, CA) or to 2′,3′-cyclic nucleotide 3′-phosphodiesterase (1:100, CHEMICON). After being washed in PBS, the sections were incubated at room temperature with affinity-purified secondary goat antibodies (Jackson Immunoresearch, West Grove, PA) labeled with Cy2 (1:100) and Cy3 (1:300) diluted in the same buffer as the primary antibody. Sections were washed several times with PBS before being coverslipped with Immu-mount (Shandon, Pittsburgh, PA) and analyzed by confocal laser microscopy (MRC 600 confocal imaging system, Bio-Rad Laboratories, Richmond, CA) with a Zeiss Axioplan microscope (Oberkochen, Germany) at the Laboratory of Electron Microscopy of the University of Zurich.
Stable expression and immunofluorescent detection of OATP31_v1 and OATP3A1_v2 in CHO FlpIn cells.
CHO FlpIn Cells were grown in Ham's F-12 medium supplemented with 10% fetal calf serum, 2 mmol/l l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C with 5% CO2 and 95% humidity. The cells were transfected with OATP3A1_v1 and OATP3A1_v2 cDNAs that were first subcloned into the pIRESneo2 expression vector (CLONTECH Laboratories) with EcoRI and NotI together with a Kozak sequence in front of their start codon (Kozak-ORF). These OATP3A1_v1 and OATP3A1_v2 Kosak-ORFs were then cut out of the pIRESneo2 expression vector with NotI and BfrI (AflII). After gel purification the cDNA probes were directly subcloned into the NotI/BfrI-digested and dephosphorylated pcDNA5/FRT expression vector of the Flp-In System (Invitrogen Life Technologies, Paisley, UK). Both constructs were controlled by digestion with BglII (2.5 h, 37°C) and sequencing of both cDNA strands. The constructs were then introduced into the CHO FlpIn cells according to the manual of the Flp-In system. Stably transfected CHO FlpIn cells were selected after 24 h by adding hygromycin (500 μg/ml). Wild-type cells were grown in the presence of zeozin (100 μg/ml). For immunofluorescence, wild-type and stably transfected CHO FlpIn cells were grown to confluence on coverslips, treated with 5 mmol/l sodium butyrate for 24 h (27), and then fixed directly in the culture dishes (room temperature, 30 min) with 200 mmol/l PBS supplemented with 2% paraformaldehyde, 0.1% glutaraldehyde, and 0.5% Triton X-100. The cells were then washed 2 × 10 min with PBS and incubated for 5 min with ammonium chloride (0.25% wt/vol) in PBS. Subsequently, the cell membranes were permeabilized for 10 min with saponin (0.1% wt/vol) in PBS (PBS-saponin). After blocking with gelatin (2% wt/vol) in PBS-saponin for 1 h, the coverslips were washed twice with PBS-saponin for 10 min and once more with PBS-saponin containing 1% (wt/vol) IgG-free bovine serum albumin (BSA) (PBS-saponin-BSA) (10 min). The coverslips were then incubated with the purified OATP3A1_v1 or OATP3A1_v2 antisera, diluted 1:50 in PBS-saponin-BSA for 2 h at room temperature. After three washing steps with PBS-saponin-BSA, the cells were incubated for 1 h with the secondary antibody [Cy3-coupled goat anti-rabbit F(ab′)2 fragment] diluted 1:500 in PBS-saponin-BSA. Finally, the coverslips were washed sequentially with PBS-saponin-BSA (10 min), PBS-saponin (10 min), and PBS (10 min). After mounting with Glycergel (DakoCytomation, Glostrup, Denmark) immunofluorescence was analyzed with confocal laser scanning microscopy using a Leica TCS 4D microscope (Leica, Wetzlar, Germany) (34).
Transport experiments in Xenopus laevis oocytes.
Capped cRNA was synthesized from the OATP3A1_v1- and OATP3A1_v2-containing X. laevis expression vectors (see above) with the mMESSAGE mMACHINE T7 Kit (Ambion, Austin, TX) for both splice variants from NotI-linearized cDNA. Oocytes were prepared as described previously (13). After an overnight incubation at 18°C, viable oocytes were selected and injected with 50 nl of water or 5 ng of cRNA dissolved in the same volume of water. After 3 days of incubation at 18°C with a daily change of the incubation solution, uptake of radiolabeled substrates was measured at 25°C in 100 μl of uptake solution (mmol/l: 100 NaCl, 2 KCl, 1 MgCl2, 1 CaCl2, and 10 HEPES adjusted to pH 7.5 with Tris) as previously described (29). For kinetic analysis, the Michaelis-Menten constant (Km) was calculated by nonlinear regression analysis (Systat version 8.0, SPSS, Chicago, IL).
Transport experiments in OATP3A1_v1- and OATP3A1_v2-transfected CHO FlpIn cells.
CHO FlpIn cells were grown in Ham's F-12 medium supplemented with 10% fetal calf serum, 2 mmol/l l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C in the presence of 5% CO2 and 95% humidity. Wild-type cells were grown in the presence of zeozin (100 μg/ml), and stably transfected cells were selected with hygromycin (500 μg/ml). Substrate uptake into OATP3A1_v1- and OATP3A1_v2-expressing CHO FlpIn cells was determined as described previously (5). For all transport experiments, expression of OATP3A1_v1 and OATP3A1_v2 was induced by preincubation of the cells for 24 h in the presence of 5 mmol/l sodium butyrate (27). The uptake medium consisted of (mmol/l) 116 NaCl, 5.3 KCl, 1 NaH2PO4, 0.8 MgSO4, 5.5 d-glucose, and 20 HEPES (pH 7.4). Uptake experiments were performed in 800 μl of solution containing the labeled substrates and, where required, additional unlabeled compound to reach the indicated concentrations. Transport was stopped with 2 ml of ice-cold buffer (mmol/l: 116 choline Cl, 5.3 KCl, 1 KH2PO4, 0.8 MgSO4, 5.5 d-glucose, and 20 HEPES), followed by three additional washes. The cells were solubilized in 1 ml of 1% Triton X-100, and the radioactivity was measured by liquid scintillation counting. Specific OATP3A1 variant uptake was determined by subtracting values from identical experiments conducted in wild-type CHO FlpIn cells.
Isolation of the two splice variants OATP3A1_v1 and OATP3A1_v2 from human brain.
Using specific primers to isolate OATP3A1 from human brain, we identified two splice variants of OATP3A1. OATP3A1_v1 corresponded to the previously reported OATP3A1 cDNA sequence (GenBank accession no. AB031050) (3, 35) containing an ORF of 2,133 bp that encodes a protein of 710 amino acids (Fig. 1A). Compared with the OATP3A1 sequence reported by Tamai and coworkers (35), our OATP3A1_v1 variant contains a single Y202I substitution. The variant OATP3A1_v2 was identical in sequence to OATP3A1_v1 up to nucleotide 1998 (amino acid 666). Thereafter, the COOH termini of the two splice variants differed in sequence (Fig. 1A). Furthermore, OATP3A1_v2 was 18 amino acids shorter (ORF: 2079 bp, 692 amino acids) than OATP3A1_v1 (Fig. 1A). A sequence similar to OATP3A1_v2, but with an R168L substitution, has also been deposited in GenBank (GenBank accession no. BC000585). Thus the two OATP3A1 variants most likely represent two splice variants of the same SLCO3A1 gene and differ in their COOH-terminal ends.
Gene structure of SLCO3A1.
The SLCO3A1 gene is localized on chromosome 15q26. To understand its genomic structures, the two OATP3A1 splice variants were aligned with a human genomic sequence. Table 1 shows the organization of the ORF of the SLCO3A1 gene. All intron/exon boundaries fit with the canonical donor and acceptor consensus motifs. The ORF of OATP3A1_v1 is composed of 10 exons, while the shorter OATP3A1_v2 has 11 exons. The first 9 exons are spliced identically. The difference between the two variants is caused by the use of an additional splice donor site in exon 10, resulting in a different COOH-terminal end of the proteins.
Tissue distribution of OATP3A1_v1 and OATP3A1_v2.
To compare the tissue distribution of the two OATP3A1 splice variants we first performed a commercially available MTE Array analysis. Since the initial primers used (see materials and methods) could not differentiate between the two splice variants, the signals detected in Fig. 1B indicate expression of OATP3A1_v1 and/or OATP3A1_v2 in the various tissues. The latter include especially testis, various brain regions, heart, lung, spleen, peripheral blood leukocytes, and thyroid gland. To differentiate between expression of OATP3A1_v1 and OATP3A1_v2 in various tissues, real-time PCR with total RNAs was performed as described in materials and methods. The data in Table 2 show detection of both splice variants of OATP3A1 in the tissues examined. Specifically, OATP3A1_v1 is generally expressed at higher levels in all investigated tissues compared with OATP3A1_v2. About three more cycles were needed to detect OATP3A1_v2 in lung, thyroid, spleen, and liver, corresponding to an ∼10-fold lower expression level compared with OATP3A1_v1.
Immunolocalization of human OATP3A1_v1 and OATP3A1_v2 in testis and different brain regions.
To investigate the expression and distribution of the two OATP3A1 splice variants in more detail, we performed immunolocalization studies in human testis, choroid plexus, and various brain regions. For this purpose we developed OATP3A1 variant-specific polyclonal antibodies that did not cross-react with any other human OATPs (Fig. 2). In testis, different developmental stages of germ cells were stained by both antibody preparations (Fig. 3, A–F). However, the cellular distribution of OATP3A1_v1 immunoreactivity was more restricted (Fig. 3B) than the expression of OATP3A1_v2, the distribution pattern of which was consistent with possible OATP3A1_v2 expression in Sertoli cells (marked by an arrow in Fig. 3E) as well. Different cellular locations of the two OATP3A1 variants were also detected in the human choroid plexus (Fig. 3, G–M). Thus, while OATP3A1_v1 expression could be clearly localized to the basolateral plasma membrane (indicated by an arrowhead) of choroid plexus epithelial cells (Fig. 3H), OATP3A1_v2 immunoreactivity showed a more intracellular and apically oriented (indicated by an arrow) distribution (Fig. 3L), suggesting its expression at the apical plasma membrane as well as in subapical intracellular vesicular compartments.
In human brain, even more striking differences in the cellular expression pattern of OATP3A1_v1 and OATP3A1_v2 were observed. Gray matter of the frontal cortex of the human brain expressed OATP3A1_v1 very abundantly (Fig. 4A). This gray matter OATP3A1_v1 immunopositivity was not associated with neuronal cell bodies (Fig. 4B). Furthermore, OATP3A1_v1 protein expression could not be detected in the white matter of the frontal cortex (Fig. 4C). In contrast, OATP3A1_v2 immunoreactivity was associated with neuronal bodies and axons in both the gray (Figs. 4, D and E) and the white (Fig. 4F) matter of the human frontal cortex. Control stainings without primary antibody were negative (data not shown).
Since only limited amounts of human brain tissue samples were available, we next investigated the expression of Oatp3a1, which exhibits a 97.6% amino acid sequence identity with the human OATP3A1_v1 (3) also in rat brain. For this purpose, we took advantage of the cross-reactivity of the OATP3A1_v1 antibody with the rat Oatp3a1 homolog. As illustrated in Fig. 5A, Oatp3a1 showed a wide and strong expression in many brain regions including frontal cortex, brain stem, and cerebellum. Interestingly, Oatp3a1 expression was associated with neuronal bodies in a somewhat punctuated manner, suggesting that it might be expressed in somatodendritic and/or somatoastrocytic synapses (Fig. 5, B–E). Staining of the sections in the absence of primary antibody showed no reactivity on corresponding sections (Fig. 5, F–I). Furthermore, while double-immunofluorescence staining indicated close proximity of some fibers stained with the astrocytic marker GFAP and Oatp3a1 (Fig. 5K), virtually no Oatp3a1 was found to be associated with oligodendrocytes (Fig. 5M). Hence, rat Oatp3a1, for which no splice variants could be detected (data not shown), exhibits a wide brain distribution similar to that of the human OATP3A1 variants (Fig. 1B) and shares its association with somatoneurons with human OATP3A1_v2 (Fig. 4, D–F).
Transport functions of human OATP3A1_v1 and OATP3A1_v2.
To compare their transport functions, the two OATP3A1 splice variants were expressed in X. laevis oocytes and in CHO FlpIn cells (Fig. 6). The two heterologous expression systems were used in parallel because 1) the frog oocyte system has proven successful in the delineation of the substrate specificity of most Oatps/OATPs so far identified (11), 2) mammalian cell lines have resulted in higher expression levels and higher transport activities for OATP1C1 in previous studies (25), and 3) we wanted to obtain a control for the potential interference with endogenous transport systems in the expression systems used. In both systems the prototypic Oatp/OATP substrates taurocholate and estrone-3-sulfate were not transported by OATP3A1_v1 and OATP3A1_v2 (Tables 3 and 4). The same was also true for digoxin and DHEAS. However, expression of both OATP3A1 splice variants induced 1.3- to 1.8-fold higher uptakes of the two PG derivatives PGE1 and PGE2 compared with water-injected oocytes (Table 3) or wild-type CHO FlpIn cells (Table 4). Thereby, OATP3A1-induced PGE1 and PGE2 uptakes were saturable and yielded apparent Km values of 101 ± 52 and 219 ± 137 nmol/l for OATP3A1_v1- and 218 ± 266 and 371 ± 155 nmol/l for OATP3A1_v2-mediated PGE1 or PGE2 uptake, respectively (Fig. 7). In addition, induction of uptake was also observed for T4 in OATP3A1_v1- and OATP3A1_v2-expressing CHO FlpIn cells (Table 4), showing evidence of saturability with increasing concentrations (data not shown), indicating that T4 is also a substrate of both OATP3A1 variants. Finally, several oligopeptides such as deltorphin II, BQ-123, and vasopressin (but not DPDPE) were taken up at higher rates into OATP3A1_v1- and OATP3A1_v2-expressing cells than into wild-type CHO FlpIn cells (Table 4). Hence, similar to other members of the Oatp/OATP superfamily, the two human OATP3A1 variants can also accept certain linear and cyclic oligopeptides as transport substrates.
We have isolated two splice variants of OATP3A1 from a human brain cDNA library. The two splice variants were called OATP3A1_v1 and OATP3A1_v2. They differ in their COOH-terminal amino acid sequences (Fig. 1A) and exhibit, at least in part, distinct cellular and subcellular distribution in the human body (Figs. 1, 3, and 4; Table 2). Thus, while OATP3A1_v1 exhibits a rather wide tissue distribution (Fig. 1B; Table 2), expression of the variant OATP3A1_v2 mRNA is generally lower (Table 2) and its protein is expressed predominantly in brain and testis (Table 2). Furthermore, in choroid plexus epithelial cells OATP3A1_v1 is expressed on the basolateral (Fig. 3H) and OATP3A1_v2 rather on the apical (Fig. 3L) plasma membrane domains. Also, in brain parenchyma OATP3A1_v1 is expressed in neuroglial cells of gray matter (Fig. 4, A–C), whereas OATP3A1_v2 was found to be associated with neurons in both gray and white matter (Fig. 4, D–F). Interestingly, rat Oatp3a1, which exhibits a 97.6% amino acid sequence identity with human OATP3A1_v1, also showed a wide expression pattern in rat brain, mimicking expression characteristics of both OATP3A1_v1 and OATP3A1_v2 in human brain (Fig. 5). On the functional level, OATP3A1_v1 and OATP3A1 exhibited similar multispecific transport activities for various substrates including PGE1 and PGE2, T4, and the oligopeptides BQ-123 and vasopressin (Tables 3 and 4; Fig. 7).
OATP3A1_v1 corresponds in sequence to one of the previously described OATP3A1s (OATP-D) (3), but differs from another published sequence (35) by a single tyrosine to isoleucine mutation at position 202. Our study confirms that OATP3A1_v1 exhibits a broad tissue distribution, with predominant expression in testis, brain, and vascular tissue, and that its preferred substrates include PGE1 and PGE2. When fine-localized in human tissues by immunohistochemistry with variant-specific antibodies, the tissue distribution of OATP3A1_v1 was found to be only partially consistent with the cerebral expression of the homologous rat Oatp3a1, which in addition to some expression in astrocytes is also expressed in somatoneurons. Thus rat Oatp3a1 combines the cerebral expression patterns of OATP3A1_v1 and OATP3A1_v2 and thus qualifies as a precursor of both human splice variants. Functionally, OATP3A1_v1 was found to transport, in addition to PGE1 and PGE2, also T4 and the cyclic oligopeptides BQ-123 and vasopressin. While transport of T4 and BQ-123 is shared by other members of the OATP/SLCO superfamily (11, 28, 34), the cyclic nonapeptide vasopressin has so far not been identified as an OATP substrate. However, transport activity was rather small, suggesting that vasopressin might not necessarily represent a physiological substrate of OATP3A1_v1. Also, and interestingly, the closely related oligopeptide oxytocin was not transported by the OATP3A1 variants.
The newly identified human OATP3A1_v2 variant is generated by the use of an additional splice donor site in exon 10 of the SLCO3A1 gene on chromosome 15q26. OATP3A1_v2 differs from OATP3A1_v1 in several aspects including 1) a shorter COOH-terminal end amino acid sequence (minus 18 amino acids), 2) more restricted tissue distribution with predominant expression in brain and testis, 3) apical and/or subapical expression in human choroid plexus epithelial cells, and 4) expression in somatoneurons of grey and white matter of human frontal cortex. In contrast to these differences in expression, transport kinetics and spectrum of transport substrates are similar between the two OATP3A1 splice variants. Thus the transporter variants transport the prostaglandins PGE1 and PGE2, T4, and the oligopeptides BQ-123 and vasopressin to a similar extent and with similar affinities. Only arachidonic acid was slightly better transported by OATP3A1_v2 than by OATP3A1_v1 (Table 4). While the definitive transport mechanism(s) as well as the role of potential driving force(s) are not yet firmly established, evidence is accumulating that OATPs might act as anion exchangers. For example, intracellular bicarbonate (30) and intracellular pH (23) have been shown to be important in substrate uptake by rat Oatp1a1. Others have shown that transport activity of OATP2B1 is stimulated by a lowering of the extracellular pH (15) concomitant with an increase of the Vmax (26), which points to a role of bicarbonate as a counterion during the OATP-mediated organic anion transport. Furthermore, rat Oatp1a1 utilizes glutathione as a driving force (21), while rat Oatp1a4 mediates exchange of substrates against glutathione and glutathione conjugates (22). Rat Oatp1a1 and Oatp1a4 can be transstimulated by substrates (22), further supporting the concept that OATPs act as anion exchangers. These data suggest that OATP3A1_v1 and OATP3A1_v2 can functionally complement each other at the various expression sites. For example, since OATPs function as bidirectional organic anion exchangers, the polarized surface expression of the two OATP3A1 splice variants in human choroid plexus epithelial cells could account for vectorial transcellular substrate transport from blood into cerebrospinal fluid or vice versa. This assumption, however, remains to be verified by functional transport studies in isolated human choroid plexus epithelial cells.
The present study demonstrates that the spectrum of transport substrates of OATP3A1_v1 and OATP3A1_v2 is not restricted to PGs as previously suggested (3), but includes in addition some cyclic oligopeptides such as the endothelin antagonist BQ-123 and vasopressin, albeit at low transport rates. Together with the previous findings of OATP-mediated transport of other neuropeptides such as, for example, cholecystokinin octapeptide and various opioid peptides (9), the data suggest that both OATP3A1 splice variants might be involved in neuropeptide transport in human brain. Although the latter suggestion is not consistent with the traditional view that, in contrast to classic neurotransmitters, no reuptake mechanisms exist for neuropeptides in neuronal synapses, neuropeptides (including vasopressin) that are released from neuronal membrane sites seem to persist in the extracellular fluid for relatively long periods of time and are thus able to diffuse considerable distances, indicating that their degradation by peptidases is less efficient than generally assumed (20). Hence, it is possible that the neuromodulatory role of vasopressin (and other neuropeptides) in classic neurotransmitter actions is regulated by OATP3A1-mediated vasopressin transport through neuronal reuptake of vasopressin itself and/or of its degradation products (20). In this regard it is noteworthy that neurotoxicity of the cyclic oligopeptide microcystin is dependent on OATP1A2-mediated brain uptake of microcystin (6). Furthermore, the proton-dependent di- and tripeptide transporter PEPT2 was recently identified in astrocytes of rat cerebral cortex, where it is supposed to play a role in the removal of degraded peptides from the extracellular space (7). And finally, the organic cation transporter 3 (OCT3/Slc22a3), which is widely expressed in brain and exhibits transport functions similar to those of the OATPs, has been shown to exert an important function in central regulation of salt and water intake (37). Hence, it is quite possible that OATP3A1 transporters play a similar important role in the central regulation of vasopressin-dependent physiological processes, especially if one considers the high conservation of the OATP3A subfamily during evolution (11). Obviously, this latter assumption requires further experimental validation, for example, by testing the influence of a physiological in-to-out glutathione gradient on vasopressin transport.
In conclusion, the present study has identified two splice variants of human OATP3A1. The data suggest that the two OATP3A1 variants exhibit similar transport functions, but exhibit distinct cellular and subcellular expression in testis, choroid plexus, and human brain frontal cortex. Although the findings indicate an involvement of OATP3A1_v1 and OATP3A1_v2 in neuroglial and neuronal transport of neuropeptides such as vasopressin, the suggested role of the OATP3A1 variants in the modulation of neurotransmission by neuropeptides requires further investigation.
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.
- Copyright © 2007 the American Physiological Society