Molecular Cloning and Functional Characterization of a Polyspecific Organic Anion Transporter from Caenorhabditis elegans1

Experimental Procedures

Materials.

PAH (p-[glycyl-2-³H(N)]aminohippuric acid; specific radioactivity, 40 Ci/mmol), TEA ([ethyl-1-¹⁴C]tetraethylammonium bromide; specific radioactivity, 55 mCi/mmol), [³H]MPP⁺ (1-methyl-4-phenylpyridinium ion; specific radioactivity, 60 Ci/mmol), [³H]MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; specific radioactivity, 80 Ci/mmol), [³H]choline (specific radioactivity, 85 Ci/mmol), and [³H]cimetidine (specific radioacitivity, 18.2 Ci/mmol) were obtained from American Radiolabeled Chemicals, Inc. (St. Louis, MO). Unlabeled anions and cations were obtained from Research Biochemicals International (Natick, MA) or Sigma Chemical Co. (St. Louis, MO). Cell culture media and Lipofectin were from Life Technologies (Rockville, MD). Restriction enzymes were from Promega (Madison, WI). Magna nylon transfer membranes were purchased from Micron Separations, Inc. (Westboro, MA). HeLa cells were obtained from American Type Culture Collection, Inc. (Rockville, MD). The Ready-to-Go oligolabeling kit used in the preparation of cDNA probes for library screening was from Amersham Pharmacia Biotech (Piscataway, NJ).

DNA Sequencing.

Sequencing by the dideoxy chain termination method was performed by Taq DyeDeoxy terminator cycle sequencing with an automated Perkin-Elmer Applied Biosystems 377 Prism DNA sequencer (Perkin-Elmer, Norwalk, CT).

Reverse Transcription Polymerase Chain Reaction (RT-PCR) and Probe Preparation.

Total RNA was isolated from C. elegans with Trizol reagent (Life Technologies) according to manufacturer’s protocol. Poly(A)⁺ RNA was then prepared by affinity chromatography on an oligo dT-cellulose column (Life Technologies). The integrity of the RNA sample was checked by formaldehyde-agarose gel electrophoresis.

A pair of PCR primers specific for the putative C. elegansorganic anion transporter encoded by the gene in theT01B11.5 locus was designed: 5′-CATCATCGTCGTCTAGCTCC-3′ (forward primer) and 5′-TCTCCAATCTTGACAAAGCC-3′ (reverse primer). RT-PCR was conducted using C. eleganspoly(A)⁺ RNA with Geneamp RNA-PCR kit (Perkin-Elmer). A single product was obtained with an estimated size of 450 base pair (bp), as predicted by the primers. The PCR product was genecleaned and cloned into pGEM-T vector (Promega). The cDNA insert was sequenced by the dideoxynucleotide chain termination method for confirmation of its identity.

Screening of cDNA Library.

The SuperScript Plasmid System (Life Technologies) was used to establish the directional cDNA library using the poly(A)⁺ RNA isolated from C. elegans (Fei et al., 1998). The cDNA probe obtained by RT-PCR was labeled with [α-³²P]dCTP using the Ready-to-Go oligolabeling kit (Pharmacia). The C. elegans cDNA library was screened with the probe under medium stringency conditions. Hybridization was carried out at 65°C for 20 h in a solution containing 5 × SSPE (1× SSPE = 0.15 M NaCl, 10 mM NaH₂PO₄, and 1 mM EDTA), 5 × Denhardt’s solution, 0.5% SDS, and 100 μg/ml of denatured salmon sperm DNA. Posthybridization washing was done as described earlier (Kekuda et al., 1998; Wu et al., 1999), which involved extensive washes with 3 × SSPE/0.5% SDS at room temperature. Positive clones were identified and the colonies purified by secondary screening.

Functional Expression of CeOAT1 in HeLa Cells.

Functional analysis of the cloned cDNA was carried out by heterologous expression in human HeLa cells using the vaccinia virus expression technique (Blakely et al., 1992). The cloned cDNA is present in pSPORT vector under the control of T7 promoter. A recombinant vaccinia virus carrying the gene for T7 RNA polymerase mediates the expression of the cDNA in mammalian cells. Transport of [³H]PAH or other radiolabeled compounds in HeLa cells expressing the CeOAT1 cDNA was measured as described previously (Kekuda et al., 1998; Wu et al., 1999). The transport buffer was 25 mM HEPES/Tris (pH 7.5) supplemented with 140 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl₂, 0.8 mM MgSO₄ and 5 mM gluose. When Na⁺-free buffer was used, N-methyl d-glucamine chloride was used to replace NaCl in the transport buffer. Cells transfected with pSPORT vector alone were used to determine endogenous transport activity. The transport activity in cDNA-transfected cells was adjusted for the endogenous transport activity to calculate cDNA-specific activity.

To establish the anion exchange characteristic of CeOAT1, the high-affinity Na⁺-coupled dicarboxylate transporter (NaDC3) that we recently cloned from rat placenta (Kekuda et al., 1999) was coexpressed with CeOAT1 in HeLa cells. The dicarboxylate transporter is capable of concentrating the exogenously added dicarboxylates such as α-ketoglutarate in these cells by a Na⁺ gradient-dependent mechanism. The dicarboxylates thus concentrated inside the cells provide the exchange anion for the anion exchange mediated by CeOAT1. In these experiments, the amount of plasmid DNA was kept constant during transfection by substituting pSPORT vector DNA wherever appropriate.

Statistics.

Uptake experiments were done in triplicate and repeated two or three times with separate transfections. Uptake values are given as means ± S.E. of these replicate values. Kinetic analysis was carried out by linear and nonlinear regression methods using the commercially available computer program Sigma Plot (Chicago, IL).

Results

Structural Features of CeOAT1.

Database searches of theC. elegans genomic DNA sequence for genes homologous to the members of the mammalian organic cation/anion transporter family identified a putative candidate. The location of this gene corresponds to the CEL T01B11.5 gene on chromosome III. To clone this C. elegans organic cation/anion transporter, we obtained an RT-PCR product using C. eleganspoly(A)⁺ RNA and primers designed on the basis of the predicted exonic sequences of the gene. The size of the RT-PCR product was ∼0.45 kbp and its identity was confirmed by sequencing. This cDNA was used as the probe to screen a C. elegans cDNA library (Fei et al., 1998) to isolate the full-length clone for structural and functional analysis. The nucleotide sequence (GenBank accession no. AF152095) and the deduced amino acid sequence of CeOAT1 are given in Fig. 1. The cDNA is 1703 bp long with a 1581-bp long open reading frame (including the termination codon). The 5′ and 3′ untranslated regions are 15 bp and 107 bp long, respectively. The 3′ untranslated region contains the poly(A)⁺ tail and the polyadenylation signal (AATAA). The predicted protein consists of 526 amino acids. Hydropathy analysis of the amino acid sequence predicts 12 transmembrane domains. When modeled similar to the known members of the mammalian organic cation/anion transporter superfamily, both the amino terminus and the carboxyl terminus of the CeOAT1 protein are located on the cytoplasmic side of the membrane. There is a long extracellular loop consisting of 57 amino acids between the transmembrane domains 1 and 2. This loop contains one potential site for N-linked glycosylation (Asn-96). The predicted molecular mass of the protein is 59 kDa.

Comparison of amino acid sequences shows that there is a significant homology (∼25% identity) between CeOAT1 and the members of the mammalian organic cation/organic anion transporter family. There is also a significant homology between CeOAT1 and CeOCT1 (24% identity). Among the members of the mammalian organic cation/organic anion transporter family, comparative analysis of amino acid sequence predicts the presence of at least three subgroups within the family. The first group consists of OCT1, OCT2, and OCT3. The second group consists of novel organic cation transporter (OCTN)1 and OCTN2. The third group consists of OAT1, OAT2, and OAT3. Interestingly, CeOAT1 is quite divergent from the mammalian organic cation and anion transporters. It is notable that the sequence similarity between CeOAT1 and mammalian organic anion transporters is not significantly different from the sequence similarity between CeOAT1 and mammalian organic cation transporters. In either case, the amino acid sequence identity remains at ∼25%. Furthermore, alignment of CeOAT1 cDNA sequence with the C. elegans genomic sequence has allowed us to deduce the exon-intron organization of the gene (Fig.2). The CeOAT1 gene consists of 11 exons and 10 introns. The exact location of the first exon in the 5′-untranslated region (the first 9 bp of the cDNA) could not be identified. The exon-intron boundaries conform to the consensus donor/acceptor sequences (gt/ag) for RNA splicing.

Functional Characteristics of CeOAT1.

To define the functional identity of CeOAT1, the cDNA was expressed in HeLa cells using the vaccinia virus expression technique and the ability of the expressed protein to transport organic cations and organic anions was assessed. Cells transfected with pSPORT vector alone without the cDNA insert were used as control. Because it was not possible to predict whether the cloned cDNA codes for an organic cation transporter or an organic anion transporter based on the amino acid sequence homology between CeOAT1 and mammalian organic cation and anion transporters, we first tested several organic cations as potential substrates for CeOAT1. None of the cations tested (choline, cimetidine, MPP⁺, MPTP, and TEA) was found to be a substrate for CeOAT1. The transport of these organic cations in cells transfected with CeOAT1 cDNA was similar to that in cells transfected with pSPORT vector alone (Fig.3). However, the transport of the organic anion PAH was increased about 3.5-fold in CeOAT1-expressing cells, compared with control cells.

Functional Coupling between CeOAT1 and Mammalian NaDC3.

Because CeOAT1 was found to transport the organic anion PAH, we analyzed whether CeOAT1 functions as an anion exchanger. In mammalian systems, OAT1 is an anion exchanger and is functionally coupled to a high-affinity Na⁺-coupled dicarboxylate transporter (Sekine et al., 1997; Sweet et al., 1997). OAT2 and OAT3 transport organic anions but do not function as anion exchangers even though the exact operational mechanism of these transporters is not known (Sekine et al., 1998; Kusuhara et al., 1999). There is no information available on Na⁺-coupled dicarboxylate transporters in C. elegans. Therefore, we used the Na⁺-coupled dicarboxylate transporter cloned from rat placenta (Kekuda et al., 1999) for assessing the anion exchange function of CeOAT1. We expressed CeOAT1 and rat (r)NaDC3 either individually or together in HeLa cells and measured PAH transport. In addition, we tested the influence of exogenously added α-ketoglutarate (a dicarboxylate) and Na⁺ on PAH transport in cells expressing CeOAT1 and rNADC3 individually or together. As shown in Fig.4A, when measured in the presence of Na⁺, the transport of PAH was increased about 4-fold as a result of expression of CeOAT1. Expression of rNaDC3 alone did not influence PAH transport, demonstrating that PAH is not a substrate for rNaDC3. When CeOAT1 and rNaDC3 were coexpressed, the transport of PAH was increased about 9-fold. This increase was significantly higher than the increase observed when CeOAT1 was expressed alone without rNaDC3. We then assessed the influence of 20 μM α-ketoglutarate, a dicarboxylate substrate for rNaDC3, on CeOAT1-mediated PAH transport (Fig. 4B). Addition of this dicarboxylate did not have any noticeable effect on PAH transport in cells transfected with empty vector, CeOAT1 cDNA, or rNaDC3 cDNA. However, PAH transport in cells coexpressing CeOAT1 and rNaDC3 increased significantly in the presence of exogenously added α-ketoglutarate (464 ± 5 versus 624 ± 17 pmol/10⁶ cells/30 min in the absence and presence of α-ketoglutarate, respectively). There was a 13-fold stimulation of PAH transport in the presence of α-ketoglutarate in cells coexpressing CeOAT1 and rNaDC3, compared with PAH transport in cells transfected with empty vector. This stimulation was significantly higher than the corresponding stimulation (9-fold) observed in the absence of α-ketoglutarate. We then assessed the role of Na⁺ in the transport process. In the absence of Na⁺ and α-ketoglutarate, expression of CeOAT1 increased PAH transport about 4-fold (Fig. 4C), a value similar to that observed in the presence of Na⁺. However, the transport of PAH mediated by CeOAT1 was not increased any further by coexpression of rNaDC3. These data show that PAH transport by CeOAT1 is not Na⁺-dependent and that Na⁺ is necessary for rNaDC3-induced stimulation of PAH transport by CeOAT1. When these transport measurements were made in the absence of Na⁺ but in the presence of α-ketoglutarate, the dicarboxylate did not have any noticeable effect on CeOAT1-mediated PAH transport in cells expressing CeOAT1 alone or together with rNaDC3 (Fig. 4D). These data show that the stimulation of PAH transport caused by α-ketoglutarate in cells coexpressing CeOAT1 and rNaDC3 is a Na⁺-dependent phenomenon. The stimulation of CeOAT1-mediated PAH transport observed even in the absence of exogenous α-ketoglutarate when rNaDC3 was coexpressed with CeOAT1 is most likely due to the endogenous dicarboxylates. These dicarboxylates may leak out of the cells into the culture medium normally and expression of rNaDC3 in the cells may mediate the reuptake of these dicarboxylates and maintain the cell-to-medium concentration gradient for these exchange anions to energize CeOAT1. The role of endogenous dicarboxylates in facilitating the CeOAT1-mediated PAH transport is also evident from the findings that the expression of CeOAT1 alone caused a 3- to 4-fold increase in PAH transport, whether measured in the presence or absence of Na⁺. This increase was observed in the absence of exogenously added α-ketoglutarate. The participation of endogenous dicarboxylates in the CeOAT1-mediated PAH transport under these experimental conditions may also explain why the increase in CeOAT1-mediated PAH transport was relatively small (∼35%), in response to exogenously added α-ketoglugarate in cells coexpressing CeOAT1 and rNaDC3.

Figure 5 describes the dose response for the α-ketoglutarate-dependent effect on CeOAT1-mediated PAH transport in cells coexpressing CeOAT1 and rNaDC3. At lower concentrations (up to 20 μM), α-ketoglurarate stimulated PAH transport in a dose-dependent manner. The maximal stimulation observed was 80%. However, at concentrations above 20 μM, α-ketoglutarate-induced stimulation decreased considerably. The stimulation was only 20% at 100 μM α-ketoglutarate. This biphasic effect of α-ketoglutarate can be explained because of the changes in the magnitude of the concentration gradient that are expected to occur depending on the concentration of the exogenously added α-ketoglutarate. At lower concentrations, rNaDC3 is capable of generating a concentration gradient for the dicarboxylate in the cell-to-medium direction, thus favoring CeOAT1-mediated PAH transport. At higher concentrations, the magnitude of this gradient is expected to be much lower, thus resulting in a much lesser stimulation of PAH transport. Furthermore, because the anion exchange function of CeOAT1 is likely to be bidirectional, excess amounts of α-ketoglutarate in the uptake medium are expected to result in competitive inhibition of PAH influx, thus decreasing the stimulation caused by intracellular α-ketoglutarate. The relative contribution of these two processes to the observed biphasic effect of exogenously added α-ketoglutarate is not known.

We then investigated the specificity of organic anions for rNaDC3-dependent stimulation of CeOAT1-mediated PAH transport (Fig.6). CeOAT1 and rNaDC3 were coexpressed in HeLa cells and PAH transport was studied in the absence and presence of various dicarboxylates and monocarboxylates (20 μM). The stimulation caused by α-ketoglutarate was 60%. Among the other compounds tested, only fumarate showed considerable stimulation (45%). The other dicarboxylates glutarate, succinate, malate, and oxalate, and the monocarboxylate lactate, stimulated PAH transport only to a negligible extent (<10%).

Figure 7 describes the substrate specificity of CeOAT1. These studies were done in HeLa cells coexpressing CeOAT1 and rNaDC3 and the transport of [³H]PAH was measured in the presence of 20 μM α-ketoglutarate. The specificity of CeOAT1 was investigated by assessing the ability of various organic anions (2.5 mM) to compete with [³H]PAH for the transport process. The most effective inhibitors were folate, indomethacin, furosemide, probenecid, and benzyl penicillin (penicillin-G). Unlabeled PAH inhibited [³H]PAH transport by 70%. The inhibition caused by ascorbate, pantothenate, urate, andN⁵-methyltetrahydrofolate was also significant (20–50%).

The saturation kinetics of PAH transport mediated by CeOAT1 was then studied. The transport measurements were made in the presence of 20 μM α-ketoglutarate in cells coexpressing CeOAT1 and rNaDC3. The concentration of PAH was varied in the range of 0.05 to 2 mM. Transport measured in cells transfected with pSPORT vector alone was used to adjust for endogenous transport activity. Figure8 describes the data for CeOAT1-specific transport. The transport was saturable with respect to PAH concentration and the data fit well to a transport model consisting of a single, saturable transport system. The Michaelis-Menten constant (K_t) for the transport process was 0.43 ± 0.07 mM and the maximal velocity (V_max) was 5.2 ± 0.3 nmol/10⁶ cells/30 min.

Discussion

This study represents the first report on the molecular and functional characterization of an organic anion transporter fromC. elegans. CeOAT1 consists of 526 amino acids and exhibits considerably low homology to the members of the mammalian organic cation and anion transporter family. The function of CeOAT1 was characterized by using PAH as the anionic substrate. PAH is a prototypical organic anion that is widely used as a substrate for mammalian organic anion transporters. Among five different organic cations tested (choline, cimetidine, MPP⁺, MPTP, and TEA), none was recognized as a substrate by CeOAT1. All of the mammalian organic cation transporters thus far cloned interact with MPP⁺ and TEA (Koepsell et al., 1999). Similarly, the organic cation transporter cloned from C. elegans(CeOCT1) also interacts with all of the organic cations tested in the current study as potential substrates for CeOAT1 (Wu et al., 1999). These data show that CeOAT1 is an organic anion transporter and not an organic cation transporter.

Among the mammalian organic anion transporters, OAT1 is an anion exchanger, whereas OAT2 and OAT3 are not. The operational mechanism of OAT2 and OAT3 remains unknown at present. The current study shows that CeOAT1 is an anion exchanger. Therefore, CeOAT1 is functionally similar to mammalian OAT1. The anion exchange property of CeOAT1 was investigated in the present study by coexpressing this transporter with the rat Na⁺-coupled dicarboxylate transporter. The functional link between CeOAT1 and rNaDC3 is schematically represented in Fig. 9. We have shown previously that rat NaDC3 transports dicarboxylates such as α-ketoglutarate in a Na⁺ gradient-dependent manner with a Na^+/dicarboxylate stoichiometry of 3:1 (Kekuda et al., 1999). When expressed heterologously in HeLa cells, rNaDC3 mediates Na⁺-dependent concentrative transport of α-ketoglutarate (and other dicarboxylic substrates) in these cells. The resultant cell-to-medium concentration gradient for α-ketoglutarate enchances CeOAT1-mediated PAH transport because the dicarboxylate provides the exchange anion for the entry of PAH. The results of the present study provide strong support for the functional coupling between CeOAT1 and rNaDC3. A similar mechanism operates in the mammalian kidney where OAT1 and a high-affinity Na⁺-coupled dicarboxylate transporter, both expressed in the basolateral membrane of the tubular epithelial cells, work together in the active uptake of PAH from the blood into the cells (Pritchard and Miller, 1993; Ullrich, 1994). It is very likely that the CeOAT1 operates in a similar manner in association with a dicarboxylate transporter natively expressed in this organism. The molecular identity and the functional characteristics of the C. elegansdicarboxylate transporter, however, remain unknown.

Because α-ketoglutarate is an excellent substrate for rat NaDC3 and also because this dicarboxylate stimulates PAH transport via CeOAT1, it is likely that α-ketoglutarate is an exchange anion for CeOAT1. Similarly, fumarate also appears to be accepted by CeOAT1 as an exchange anion. On the other hand, glutarate, succinate, and malate are also excellent substrates for rat NaDC3 (Kekuda et al., 1999), but these dicarboxylates exhibit relatively much lower stimulatory effect on PAH transport mediated by CeOAT1. This suggests that glutarate, succinate, and malate are poor substrates for CeOAT1 as exchange anions. This is interesting because glutarate as well as α-ketoglutarate are accepted as substrates by mammalian OAT1, but succinate and fumarate are not (Shimada et al., 1987; Pritchard, 1988,1990). However, CeOAT1 accepts fumarate but not glutarate as a substrate. Apparently, even though there is considerable similarity in the transport mechanism between CeOAT1 and mammalian OAT1, these two transporters differ significantly in their substrate specificity. Oxalate and lactate are not transported by rat NaDC3; therefore, the lack of stimulatory effects of these anions on CeOAT1-mediated PAH transport does not permit us to draw any conclusions regarding whether or not these anions are accepted as substrates by CeOAT1. It can be argued that even though α-ketoglutarate as well as fumarate stimulate CeOAT1-mediated PAH transport in cells coexpressing CeOAT1 and rNaDC3, only fumarate is accepted as an exchange anion by CeOAT1 because of the possible intracellular conversion of α-ketoglutarate to fumarate. This is, however, unlikely because succinate can also be converted to fumarate intracellularly, but this dicarboxylate, although an excellent substrate for rNaDC3, does not stimulate CeOAT1-mediated PAH transport. It is therefore likely that both α-ketoglutarate and fumarate are accepted as exchange anions by CeOAT1 and that intracellular metabolism of dicarboxylates does not play any significant role in the observed effects of these compounds on CeOAT1-mediated PAH transport.

Both the CeOAT1 and mammalian OAT1 are multispecific, based on the broad specificity of organic anions that can compete with PAH for transport via these transporters. Penicillin-G, indomethacin, probenecid, and furosemide are excellent substrates for C. elegans and mammalian OAT1s. The proposed function of OAT1 in mammalian systems is in the elimination of various xenobiotics. This is supported by the expression of OAT1 in the kidney, an organ that plays an important role in xenobiotic elimination. C. eleganspossesses an excretory system consisting of four cells that functions in the elimination of xenobiotics analogous to the function of the kidney in animals. It is likely that the OAT1 described here is expressed in this excretory system in C. elegans and functions in the elimination of organic anions. However, the physiological significance of the differences in the apparent substrate specificity between CeOAT1 and mammalian OAT1 in relation to the handling of endogenous and exogenous organic anions in the C. elegans versus mammalian organisms is not readily apparent.