The efficient elimination of endogenous metabolites, drugs, and xenobiotics in the kidneys requires the concerted action of polyspecific transporters like organic anion transporters (OATs¹) and organic cation transporters (OCTs) (for review, see Burckhardt and Pritchard, 2000). The family of organic anion transporters (OAT, SLC22A) was discovered in 1997 with the cloning of the renal rat and flounder OAT1 (rOAT1, Sekine et al., 1997; Sweet et al., 1997; fOAT, Wolff et al., 1997). Meanwhile, OAT1 is the best studied transport protein within this OAT family, with nine actually identified and in part functionally characterized orthologs: human (Reid et al., 1998; Hosoyamada et al., 1999), mouse (Lopez-Nieto et al., 1997), rat (Sekine et al., 1997; Sweet et al., 1997), rabbit (Bahn et al., 2002), pig (Hagos et al., 2002), OK-cell [AJ308236; AJ308236: Didelphis virginiana partial mRNA for putative organic anion transporter (okOAT1 gene), Y. Hagos, M. Sendler, A. Babu, and G. Burckhardt, 2001, unpublished], cattle [AJ549816; AJ549816: Bos taurus mRNA for organic anion transporter 1 (oat1 gene), J. Geyer and E. Petzinger, unpublished], winter flounder (Wolff et al., 1997), and Caenorhabditis elegans (George et al., 1999). OAT1 plays a major role in the renal secretion of preferentially hydrophilic organic anions such as endogenous metabolites, xenobiotics, and drugs (Dresser et al., 2001; Burckhardt et al., 2002; Burckhardt and Burckhardt, 2003). Located on the basolateral side of the proximal tubule cells, it mediates the uptake of these compounds into the cells against a negative membrane potential and in exchange for the Krebs cycle intermediate α-ketoglutarate. Consequently, it is held responsible for the nephrotoxicity of some substances like the endogenous uremic toxins (Deguchi et al., 2004), antiviral drugs (Ho et al., 2000), and cephaloridines (Takeda et al., 1999), as well as the mycotoxin ochratoxin A (Tsuda et al., 1999). Besides kidneys, OAT1 was also detected in other organs such as liver, skeletal muscle, brain, and eye (Cihlar et al., 1999; Pritchard et al., 1999; Bahn et al., 2003), where it may be involved in the secretion of neurotransmitter metabolites.

Recent findings of alternative splice-variants in this SLC22A family of transporters [e.g., the rat organic cation transporter 1 (rOCT1; Zhang et al., 1997), the human OCT2 (Urakami et al., 2002), and the human OAT1 (Bahn et al., 2000)], as well as “single nucleotide polymorphisms” (Leabman et al., 2002)] suggest a potential impact on the pharmacokinetics and substrate selectivity in vivo. Knowledge about the uptake and excretion capacity of a patient for a specific substance becomes increasingly important in terms of future concepts for a specific therapy or an individual dose for drug treatment.

Our previous studies concerning the in vivo expression of hOAT1 in kidney specimens of a patient revealed, besides the known two isoforms hOAT1-1 and hOAT1-2, two new isoforms, hOAT1-3 and hOAT1-4, which possess an additional in-frame deletion of 132 bp (Bahn et al., 2000). In the present report, we determined the expression of human OAT1 isoforms in different patients and investigated their distribution in the kidneys. Additionally, all four hOAT1 isoforms were cloned and functionally characterized. Our results confirm that all four hOAT1 isoforms are constitutively expressed in the kidneys, with hOAT1-2 as the main transcript in kidney cortex. No function was detected for hOAT1-3 and hOAT1-4. A functional comparison of hOAT1-1 and hOAT1-2 showed that they resembled each other in their sensitivities for p-aminohippurate (PAH), urate, and furosemide, indicating that both isoforms contribute to the excretion of these substances.

Materials and Methods

Materials. Materials used included fetal bovine serum, trypsin, PBS (from Invitrogen, Groningen, The Netherlands), buffer ingredients, unlabeled substrates, such as PAH, urate, and furosemide (Sigma-Aldrich, Deisenhofen, Germany), and fluorescein (FL; from Molecular Probes, Leiden, The Netherlands).

Total RNA Extraction and RT-PCR. Twenty to 50 mg of human kidney tissue of 10 different donors was extracted with the SV-total-RNA-isolation system (Promega, Mannheim, Germany) according to the manufacturer's protocol. For the reverse transcription, 0.5 to 2 μg of total RNA, 4 U of Omniscript reverse transcriptase (QIAGEN, Hilden, Germany), and an oligod(T)-anchor primer (5′ GACCACGCGTATCGATGTCGAC (T)₁₈(AGC) 3′), including the standard rapid amplification of cDNA ends (RACE) sequence as a terminal linker, were used in a reaction for 1 h at 37°C. The subsequent standard PCR (94°C, 2 min; 94°C, 30 s; 55°C, 30 s; 72°C, 1 min, for 35 cycles) was performed with 2 to 4 μl of the RT-reaction and with the sense primer (Ex-U: 5′ TTCCACTAGCTTTGCCTACTATG 3′) and the antisense primers (Ex-R: 5′ CTCTTGTGCTGAGGCCTG 3′) in a PTC200 thermal cycler (MJ Research, Biozym Diagnostik, Hess. Oldendorf, Germany). PCR products were visualized after separation on an agarose gel with ethidium bromide using an UV-transilluminator (UVP, Upland, CA). Specificity of hOAT1 splice products was checked by Southern blot.

“Overlap”-PCR. Starting with the hOAT1-2 isoform, we generated two fragments of 1612 bp and 149 bp in a PCR (94°C, 1 min; 94°C, 15 s; 55°C, 15 s; 72°C, 1 min, for 35 cycles) using the primer sets (hOAT1-U: 5′ GGGGATCCATGGCCTTTAATGACC 3′; 39-bp hOAT1-R: 5′ CTGGGATATATCCCTGCTTCTTTCTGAGTGGGGGCCCACCTGCTCTCCAGGTCCTGCAC 3′) and (39-bp hOAT1-U: 5′ GTGGCCCCCCACTCAGAAAGAAGCAGGGATATATCCCAGGAAAGGGAAACAGACGCGAC 3′ and hOAT1-R: 5′ GGTCTAGACCTCAAAATCCATTC 3′), which contained the 39-bp missing fragment as a terminal linker, and 1.2 units of the proofreading polymerase powerscript (PAN Biotech, Aidenbach, Germany). After visualization of the PCR products in an agarose gel by ethidium bromide, they were cut out of the gel and extracted with the Nucleotrap-kit (Macherey-Nagel, Düren, Germany). In a final PCR (94°C, 1 min; 94°C, 15 s; 55°C, 15 s; 72°C, 1 min, for 35 cycles) with the primers (hOAT1-U: 5′ GGGGATCCATGGCCTTTAATGACC 3′; hOAT1-R: 5′ GGTCTAGACCTCAAAATCCATTC 3′), the full-length open reading frame of 1692 bp was amplified. This resulted in the generation of isoform hOAT1-1. These two isoforms were taken as templates to synthesize hOAT1-3 and hOAT1-4, applying the same PCR strategy with the primers (hOAT1-U: 5′ GGGGATCCATGGCCTTTAATGACC 3′; 132-bp hOAT1-R: 5′ GCAGGAGGACAGTGACAGCGCGGATCATTGTGGGATACAG 3′) and (132bp-hOAT1-U: 5′ CTGTATCCCACAATGATCCGCGCTGTCACTGTCCTCCTGC 3′ and hOAT1-R 5′ GGTCTAGACCTCAAAATCCATTC 3′). All four open reading frames hOAT1-1 to hOAT1-4 were cloned with the TOPO-pcDNA3.1-cloning kit (Invitrogen) according to the manufacturer's protocol. Recombinant clones were screened by PCR and sequence-verified.

Sequencing. Positive clones of the final constructs hOAT1-1, hOAT1-2, hOAT1-3, and hOAT1-4 were sequenced with primers derived from the hOAT1-cDNA with an automated sequencer (ABI, Weiterstadt, Germany). Sequence analysis was done using online services (e.g., MAP; Huang 1994).

Cell Culture and Uptake Experiments. The monkey kidney cell line COS 7 was cultivated in plastic flasks or Petri dishes (Sarstedt, Nümbrecht, Germany) in Dulbecco's modified Eagle's medium (Invitrogen) with 580 mg/l glutamine, 110 mg/l Na-pyruvate, and 10% heat-inactivated fetal calf serum in 5% CO₂ at 37°C. Five micrograms of hOAT1-pcDNA3.1 constructs were transiently transfected into COS 7 cells by electroporation (GenePulser II; Bio-Rad, München, Germany) at 250 V and 300 μF. Twenty-four hours after transfection, the cells were plated in six-well plastic dishes (Sarstedt) at a density of 2 × 10⁵ cells/well. Transport assays were performed 48 h posttransfection in buffer (110 mM NaCl, 3 mM KCl, 1 mM CaCl₂, 0.5 mM MgSO₄, 1 mM KH₂PO₄, 10 mM HEPES, and 5 mM glucose at a pH of 7.5). The cells were washed twice with buffer and incubated at room temperature for 5 min in buffer containing 1 μM FL. The incubation was stopped and the extracellular FL was removed by washing the monolayer two to three times with 500 to 1000 μl of ice-cold PBS. Cells were dissolved in 1 ml 0.5 N NaOH. To assess FL accumulation, fluorescence was measured in a fluorescence spectrophotometer (Hitachi, Tokyo, Japan) at 492/512 nm (excitation/emission). The protein content of each well was determined according to the Bradford procedure (Bradford, 1976). For the determinations of the urate, p-aminohippurate, or furosemide concentrations that blocked 50% of FL uptake (IC₅₀), the following equation (eq. 1) was used and fitted by nonlinear regression with SigmaPlot 2001 (SPSS Inc., Chicago, IL). where ν is the rate of FL uptake in the presence of the inhibitor (urate, p-aminohippurate, or furosemide), ν₀ is the rate of FL uptake in the absence of the inhibitor (in this case set to 100%), I is the inhibitor concentration, and h is the Hill coefficient, representing the cooperativity between the tested substances and the transporter.

The Michaelis-Menten kinetics (K_m values) for hOAT1-1 and hOAT1-2 using fluorescein were determined according to eq. 2:

Immunofluorescence. All four hOAT1 isoforms were transiently transfected into COS 7 cells as described above and seeded on coverslips in a 24-well dish at a density of 2 × 10⁵ cells/well. After 48 h, the cells were washed two times with PBS and fixed for 8 min in 3.7% formaldehyde/PBS solution. For the immunoassay, the cells were washed two times with PBS and permeabilized by an incubation for 5 min in permeabilization buffer (50 mM sodium phosphate buffer, pH 7.4, 0.5 M NaCl, 0.3% TX-100). Subsequently, the cells were blocked with blocking solution (33% goat serum, 0.1% TX-100 in PBS) for 30 min and incubated with a polyclonal rabbit anti-OAT1 antibody, OAT11-A (Alpha Diagnostic, San Antonio, TX) in a dilution of 1:20 for 1 h. The first antibody was removed by two washing steps with PBS/0.1% bovine serum albumin. The cells were probed with the second antibody [Alexa fluor 488 goat anti-rabbit IgG (H+L) conjugate (Molecular Probes)] in a dilution of 1:200 PBS/0.1% bovine serum albumin for 1 h. Afterward, the cells were mounted with an 80% glycerol/20 mM NaHCO₃ solution, the coverslip was sealed, and the probes were analyzed with an LSM510-META (Zeiss, Jena, Germany).

Preparation of the Membrane Fractions and Western Blot Analysis. Kidney tissue (50 mg, from patients 4 and 7) was ground up in liquid nitrogen with a mortar and pestle and suspended in 1 ml of membrane buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 10 mg/ml phenylmethylsulfonyl fluoride, 4 μg/ml aprotinin, 4 μg/ml leupeptin). After a further homogenization step with an Ultra-Turrax homogenizer, the probes were centrifuged for 10 min at 12000 rpm and 4°C. The pellet was discarded and the supernatant was centrifuged for 30 min at 50,000 rpm and 4°C (Beckman TLX ultracentrifuge, TLA 100.3 rotor; Beckman-Coulter, Fullerton, CA). The resulting pellet (membrane fraction) was resuspended in loading buffer (5 mM Tris-HCl, pH 6.8, 10% glycerol, 1% β-mercaptoethanol, 1% SDS, 0.004% bromphenol blue) and heated (85°C, 10 min), and the protein content was determined according to the method of Bradford (1976). Equal amounts of protein were loaded onto a 10% polyacrylamide gel, size-fractionated for 45 min at 50 mA, and blotted to a polyvinylidene difluoride membrane at 40 mA overnight. The membrane was blocked with 5% dry milk in PBS/0.05% Tween 20 for 1 h at room temperature, and the immunoassay was performed with a polyclonal rabbit anti-OAT1 antibody, OAT11-A (Alpha Diagnostic International, San Antonio, TX) in a dilution of 1:40 overnight at 4°C on an orbital shaker. Afterward, the membrane was washed three times with PBS/0.1% Tween 20 and incubated with the second antibody, an anti-rabbit IgG coupled to horseradish peroxidase, in a dilution of 1:10,000 for 1 h at room temperature. After three more washing steps with PBS/0.1% Tween 20, immunoreactive bands were visualized using the Western blotting detection system (Amersham Biosciences Inc., Freiburg, Germany).

Kinetic and Statistical Analysis. Unless indicated otherwise, data are the mean (±S.E.M.) of three independent experiments with three repeats each. Statistical analysis was performed with Microsoft Excel (Microsoft, Unterschleißheim, Germany) and SigmaPlot 2001 (SPSS Inc.).

Results

hOAT1 Isoform Expression in Human Kidneys. Total RNA of kidney specimens of seven different patients was extracted and the expression of the hOAT1 isoforms was examined by RT-PCR analysis. Applying the primer set Ex-U/Ex-R, we amplified all four isoforms with product sizes of 637 bp (hOAT1-1), 598 bp (hOAT1-2), 505 bp (hOAT1-4), and 466 bp (hOAT1-3) in all investigated patients (Fig. 1A). Most of them, including the commercially available control-mRNA (Fig. 1A, lane C; Ambion, Austin, TX), showed hOAT1-2 as the mainly represented isoform. Interestingly, one mRNA (patient 4) exhibited a comparable expression of the isoforms hOAT1-2, hOAT1-3, and hOAT1-4, indicating individual variation and thus a possible change in the impact of isoform hOAT1-3 and isoform hOAT1-4.

For the analysis of the distribution of hOAT1 isoforms, kidney cortex (C1–C3) and medulla (M1–M3) were prepared from three different patients and analyzed by RT-PCR. An expression of all four isoforms was found in kidney cortex, whereas hOAT1-2 was weak or absent in the medulla (Fig. 1B). hOAT1-3 and hOAT1-4 appeared to be present in equivalent amounts in the medulla and the cortex, documenting that the expression of some isoforms of hOAT1 is not restricted to the proximal tubule as previously shown for OAT1 (Hosoyamada et al., 1999). Additional Western blot analyses of membrane fractions of kidney specimens from patients 4 and 7 revealed two specific bands with a mass of approximately 80 kDa and 70 kDa for both patients (Fig. 1C), comparable to the observed mRNA expression of hOAT1-2 and the new isoforms hOAT1-3 and hOAT1-4 (cf. Fig. 1B). Furthermore, we examined the distribution of hOAT1 in other tissues using “multiple-tissue plates” (RZPD, Berlin, Germany). Initial PCR screenings revealed the expression of all hOAT1 isoforms also in different parts of the brain (data not shown).

Cloning and Functional Comparison of hOAT1 Isoforms. Rapid amplification of cDNA ends (RACE) techniques did not reveal any additional information of a further modification or alteration of hOAT1-3 and hOAT1-4 open reading frames (data not shown). For this reason, we cloned all four hOAT1 isoforms for functional investigations as illustrated in Fig. 2 (for more detailed information, see Bahn et al., 2000) using an overlap-PCR strategy. All hOAT1 isoforms used for functional characterizations were sequence-verified and in part repaired by applying site-directed mutagenesis (Stratagene, Amsterdam, The Netherlands).

Initial functional studies were performed on COS 7 cells transiently transfected with 5 μg of the respective hOAT1 isoform. Fluorescein (1 μM) was taken as a tracer in an uptake assay over 5 min, which is a good compromise between a sufficient signal-to-noise ratio and the linear (initial) phase of fluorescein uptake by hOAT1 (data not shown). hOAT1-1 and hOAT1-2 exhibited nearly the same fluorescein uptake into COS 7 cells, whereas hOAT1-3 and hOAT1-4 did not show any uptake of fluorescein comparable to nontransfected cells (Fig. 3). Since hOAT1-3 or hOAT1-4 may have regulatory effects, we performed cotransfections of hOAT1-1 with each of the other isoforms, taking 2.5 μg of DNA of each isoform for the transfection. A cotransfection of hOAT1-1 with hOAT1-2 revealed an almost identical uptake of fluorescein compared with the value of hOAT1-1 or hOAT1-2 at 5 μg (cf. Fig. 3). hOAT1-3 or hOAT1-4 transfected together with hOAT1-1 led to a reduction of fluorescein transport by 50%, but did not abolish fluorescein uptake, indicating that they do not possess a dominant negative effect.

Immunofluorescence Analysis of hOAT1 Isoform Expression. The fact that hOAT1-3 and hOAT1-4 did not show any function leads to the questions, whether they are expressed as a protein and whether they occur in the membrane. To address these questions, we transiently transfected all four hOAT1 isoforms into COS 7 cells and analyzed hOAT1 protein expression with an OAT1-specific antibody, followed by an immunofluorescence labeling and laser scan microscopy. Since the epitope is located intracellularly, cells had to be permeabilized for labeling, preventing immunodetection of membrane-bound hOAT1. Nevertheless, Fig. 4 indicates that a lack of protein expression is not the reason for the functional defects of hOAT1-3 and hOAT1-4.

Functional Characterization of hOAT1-1 and hOAT1-2. To explore possible differences that may explain the existence of both isoforms, we performed functional comparisons between hOAT1-1 and hOAT1-2. First, we examined the concentration dependence of fluorescein uptake by hOAT1-1 and hOAT1-2. Figure 5 shows the kinetics of hOAT1-1- and hOAT1-2-mediated fluorescein transport in a representative experiment. In three separate experiments on COS 7 cells expressing hOAT1-1, J_max was 76.1 ± 12.2 pmol · mg^-1 protein · 5 min^-1 with a K_m of 11.6 ± 3.7 μM, and in cells expressing hOAT1-2, J_max was 53 ± 10.8 pmol · mg^-1 protein · 5 min^-1 with a K_m of 11.9 ± 6.4 μM. The similar affinities of both transporters for fluorescein made it reasonable to test the interaction of other known substrates with the two hOAT1 isoforms.

Increasing concentrations of unlabeled PAH inhibited fluorescein transport with an IC₅₀ of 16 μM and 10 μM for hOAT1-1 and hOAT1-2 (Fig. 6), respectively. Using increasing concentrations of urate on fluorescein uptake revealed IC₅₀ values of 440 μM for hOAT1-1 and 385 μM for hOAT1-2 (Fig. 7). The diuretic drug furosemide was also tested on the interaction with both hOAT1 isoforms. In this case we detected an IC₅₀ of 14 μM for hOAT1-1 and 20 μM for hOAT1-2, documenting an equivalent affinity for this drug (Fig. 8).

Discussion

Alternative splicing of transport proteins has been shown to result in several effects, such as tissue-specific distribution, as described for the human NBC-1A and NBC-1B (Soleimani and Burnham, 2001); different transport protein specificity, noted for the Na⁺-K⁺-Cl^- cotransporter (Plata et al., 2001); different membrane localization and mode of action, recently observed for the sodium/bile acid transporter ASBT (Lazaridis et al., 2000); and negative regulating effects, documented for splice-variants of the rat NaPi-2 (Tatsumi et al., 1998). Recently, we cloned the hOAT1 gene and described the detection of two new splice-variants of hOAT1, called hOAT1-3 and hOAT1-4 (Bahn et al., 2000). The aim of this study was to document the expression of these hOAT1 isoforms in human kidneys and to determine their function in comparison with hOAT1-1 and hOAT1-2. RT-PCR analysis of seven different kidney specimens and one control RNA revealed both isoforms without or with the 39-bp deletion (hOAT1-1 and hOAT1-2), as well as the recently described isoforms with the 132-bp deletion (hOAT1-3 and hOAT1-4) in most of the examined probes. These findings are further supported by similar results of OAT1 isoform expression in different regions of the human brain investigated by PCR-screening of multiple-tissue plates and in the rabbit kidney (data not shown), indicating a tissue- and species-independent role of OAT1 isoforms. In almost all kidney specimens (including rabbit kidney), isoform OAT1-2 is by far the highest expressed variant of OAT1. This may imply that the explored functional characteristics of basolateral organic anion transport, that are attributed to OAT1 activity in vivo (Shimomura et al., 1981; Groves et al., 1998), are based mainly on the expression of isoform OAT1-2 (for a review of OAT1 functional characteristics, see Burckhardt and Burckhardt, 2003). Interestingly, we found that in one investigated case (patient 4), an equal expression of hOAT1-2, hOAT1-3, and hOAT1-4 was visible, reflecting an individual alteration of hOAT1 isoform expression in vivo, possibly resulting in a change of the drug excretion capacity.

To see whether hOAT1-3 and hOAT1-4 are expressed in the same region as hOAT1-1 or hOAT1-2, we determined the renal distribution of hOAT1 isoforms. Three different specimens of kidney cortex and medulla from three different donors were analyzed. RT-PCR results exhibited a clear and high mRNA expression of hOAT1-2 in kidney cortex, whereas it was weak or not detected in the medulla. This is in accordance with the observations of Hosoyamada et al. (1999), who showed that OAT1 protein expression in the human kidney is restricted to the proximal tubule of kidney cortex. In the rat kidney, OAT1 mRNA expression determined by in situ hybridization was confined to the S₂ segment of the proximal tubule (Sekine et al., 1997; Tojo et al., 1999). However, a recent reevaluation of OAT1 protein distribution in the human kidneys by Motohashi et al. (2002) revealed a wider expression of OAT1 along the proximal tubule. hOAT1-3 and hOAT1-4 were amplified in kidney cortex and weakly in the medulla, consistent with the proposed localization of human OAT1 by Motohashi et al. (2002). An additionally performed Western blot of membrane fractions from kidney specimens of patient 4 and patient 7 revealed a main protein product at 80 kDa (as a control, we tested a flag-coupled, heterologously expressed hOAT1, which gave a signal of the same size; data not shown) for patient 7 and a weaker protein product at 70 kDa, consistent with the results of the RT-PCR for the hOAT1 isoforms for this patient. A reduced expression of the 80-kDa band was detected for patient 4, corresponding exactly to the amplified mRNA amount. These data provide a first clue toward an in vivo expression of the newly identified splice-variants hOAT1-3 and hOAT1-4.

A functional characterization of all four hOAT1 isoforms cloned via a PCR-based strategy showed a DNA concentration-dependent uptake of fluorescein for hOAT1-1 and hOAT1-2. hOAT1-3 and hOAT1-4 did not interact with fluorescein or PAH (data not shown), indicating that hOAT1-3 and hOAT1-4 might be nonfunctional, like, for example, the splice-variants of the rat NaPi-2 (Tatsumi et al., 1998), the pig OAT1 isoform (Hagos et al., 2002), or the human OCT1 isoforms (Hayer et al., 1999). To explore a possible regulatory effect of hOAT1-3 and hOAT1-4 that was delineated for the rat NaPi-2 isoforms, we cotransfected isoform hOAT1-1 with equal amounts of each of the other isoforms (2.5 μg of DNA each). hOAT1-1 transfected together with hOAT1-2 revealed a fluorescein uptake comparable with the signal resulting from 5 μg of hOAT1-1. Isoforms hOAT1-3 and hOAT1-4 transfected in combination with hOAT1-1 did not abolish fluorescein uptake, indicating that hOAT1-3 and hOAT1-4 do not have a dominant negative effect on the functional isoforms, as was demonstrated for the human OCT1 isoforms (Hayer et al., 1999). Therefore, we had to prove whether hOAT1-3 and hOAT1-4 are expressed on the protein level and, additionally, whether these two new splice-variants are correctly processed to the membrane. All four isoforms were transiently transfected into COS 7 cells and analyzed by immunofluorescence using a commercially available OAT1 antibody and confocal laser scan microscopy. The results provide evidence that all four cloned hOAT1 isoforms are expressed as proteins in vitro. Moreover, three-dimensional reconstructions of the laser scan pictures for each splice-variant illustrated that there are no differences in protein delivery between the functional (hOAT1-1 and hOAT1-2) and nonfunctional (hOAT1-3 and hOAT1-4) isoforms of hOAT1 (data not shown). Taken together, these data, in combination with the Western blot analysis of kidney specimens, imply that the observed impairment of transport function for hOAT1-3 and hOAT1-4 is not due to a loss of expressed protein.

Our RT-PCR analysis revealed that both hOAT1-1 and hOAT1-2 are generally coexpressed in the renal cortex. Consequently, it is of interest whether they show any functional differences that may explain their coexistence. To address this question, we measured fluorescein uptake in COS 7 cells transiently transfected with hOAT1-1 or hOAT1-2 and investigated the affinity for fluorescein. Both splice-variants showed the same affinity for fluorescein with a K_m of 11.6 ± 3.7 μM (hOAT1-1) and a K_m of 11.9 ± 6.4 μM (hOAT1-2). Consequently, we tested the interaction of PAH, urate, and the diuretic drug furosemide under identical experimental conditions for both isoforms. PAH as the “classical” substrate for OAT1 caused an IC₅₀ of 16 μM for hOAT1-1 and 10 μM for hOAT1-2. These values fit very well to the already known affinities of human OAT1 for PAH (Hosoyamada et al., 1999; for review, see Burckhardt et al., 2001).

In human beings, fractional excretion of urate is about 10%, supporting the notion that reabsorption dominates secretion (Roch-Ramel and Guisan, 1999). A protein playing a major role in the reabsorption process on the apical side of the proximal tubule cell was recently identified and called URAT1 (Enomoto et al., 2002). The mechanisms of the basolateral entry/exit of urate are still a matter of debate. hOAT1 was shown to transport urate with a calculated K_m of 943 μM (Ichida et al., 2003). Our IC₅₀ values for urate of 440 μM for hOAT1-1 and 385 μM for hOAT1-2 are more than 2 times lower than the reported K_m value. The normal urate serum concentration was noted to be below 300 μM (Roch-Ramel and Guisan, 1999), implying that both isoforms may be involved in urate secretion under normal physiological conditions. Recently, Bakhiya et al. (2003) reported on the interaction of hOAT3, with urate providing an IC₅₀ of 255 μM. These data suggest the possible involvement of hOAT1 and hOAT3 in urate secretion.

The effect of furosemide is based on an active secretion in the proximal tubule via probenecid- and PAH-sensitive transport systems (Bidiville and Roch-Ramel, 1986). For the rat OAT1 it was reported that it translocates furosemide (Uwai et al., 2000). We calculated an IC₅₀ for furosemide of 14 μM for hOAT1-1 and 20 μM for hOAT1-2, illustrating a substantial interaction of the compound with the transport proteins similar to the observed interaction of these splice-variants with PAH. These data suggest that both human isoforms (hOAT1-1 and hOAT1-2) are involved in the secretion of furosemide. This conclusion is supported by the latest findings of Hasannejad et al. (2003), who described an IC₅₀ for furosemide of 18 μM for hOAT1. Moreover, they provide evidence that furosemide is translocated by hOAT1.

In summary, we provide evidence for a general expression of the recently identified splice-variants of hOAT1. All four documented hOAT1 isoforms are expressed on the protein level in vitro and in vivo. Their expression is restricted to the kidney cortex. Functional characterizations of hOAT1-1 and hOAT1-2 illustrate that they may contribute to the same extent to the excretion of the substrates tested. The functional impact of hOAT1-3 and hOAT1-4 on organic anion transport still remains open.