INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

DI- AND TRICARBOXYLIC ACIDS are actively taken up by the proximal tubule cells from both peritubular fluid and glomerular filtrate (16). The transport processes have been assumed to be mediated by two distinct Na⁺-dicarboxylate pathways, the luminal and the basolateral Na⁺-dicarboxylate transporters (1, 18, 23). The primary role of these two Na⁺-dicarboxylate transporters is the supply of tricarboxylic acid cycle intermediates such as -ketoglutarate and citrate, which serve one of the important energy sources of the cells, to the proximal tubule cells. For example, citrate, which is taken up by the proximal tubule cells, can provide for up to 10% of renal oxidative metabolism in human kidneys (16).

The other role of Na⁺-dicarboxylate transporters is to sustain the outwardly directed dicarboxylate gradient that drives the organic anion excretion from the proximal tubule. A variety of endogenous and exogenous organic anions, including hippurate, urate, prostanoids, cyclic nucleotides, and a number of clinically important drugs, are secreted from the proximal tubule. The uptake of these organic anions from the peritubular fluid is mediated by the p-aminohippurate (PAH) transporter (5, 12). Recent studies have revealed that the PAH transporter at the basolateral membrane of the proximal tubule is an organic anion/dicarboxylate exchanger (12, 13, 15). Dicarboxylates taken up by the proximal tubule cells drive organic anion uptake via the PAH transporter; however, the relative contribution of the two Na⁺-dicarboxylate transporters (luminal and basolateral) remains to be elucidated.

The Na⁺-dicarboxylate transporter may also be related to urinary tract stone formation. Citrate, which is filtered through the glomerulus, chelates calcium ions and prevents the precipitation of calcium salts in the urine (11). In fact, hypocitraturia is present in ~30% of patients with recurrent stone disease (11). Thus the concentration of citrate in the urine, which is regulated by the luminal Na⁺-dicarboxylate transporter, is closely related to the risk of urolithiasis.

Luminal and basolateral Na⁺-dicarboxylate transporters have been investigated not only from these physiological and pathophysiological perspectives but also for the study of Na⁺-dependent transport processes. Studies using membrane vesicles indicated the electrogenic properties (22), substrate selectivities (1, 14, 17, 18, 23), and stoichiometry (Na⁺ to substrate coupling ratio) (3, 19, 20) of Na⁺-dicarboxylate transporters.

Recently, Na⁺-dicarboxylate transporter (NaDC-1) was isolated from the rabbit kidney using the expression cloning technique (6). NaDC-1 mediated di- and tricarboxylates uptake in a sodium-dependent manner. From its functional characteristics (6) and the results of Western blot analyses (8), NaDC-1 has been predicted to be the luminal Na⁺-dicarboxylate transporter.

In the present study, we isolated a rat renal Na⁺-dicarboxylate transporter (rNaDC-1) and determined its membrane localization by immunohistochemistry. rNaDC-1 is localized exclusively in the luminal membrane of the proximal tubule cell. Furthermore, we analyzed the transport properties of rNaDC-1 by electrophysiological and transport experiments in Xenopus laevis oocytes. Transport of di- and tricarboxylates via rNaDC-1 is electrogenic. From the comparison of the transport rate and the evoked current of citrate, we investigated the species of citrate which is transported by rNaDC-1.

The nucleotide sequence reported in this study has been submitted to GenBank/EBI data bank with accession number AB001321.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Construction of cDNA library and isolation of rNaDC-1. A nondirectional cDNA library was constructed from 3 µg of rat kidney poly(A)⁺ RNA using Superscript Choice system (Life Technology) and was ligated to phage vector ZipLox EcoR I arms (Life Technology). A PCR product corresponding to the nucleotides 1323-1763 of rabbit sodium-dicarboxylate cotransporter (NaDC-1) (6) was labeled with [³²P]dCTP by random priming (T7 QuickPrimer Kit, Pharmacia) and used for the screening of a rat renal cDNA library. Replicated filters of a phage library were hybridized overnight at 37°C in a hybridization solution [50% formamide, 5× standard saline citrate (SSC), 3× Denhardt's solution, 0.2% SDS, 10% dextran sulfate, 0.2 mg/ml denatured salmon sperm DNA, 2.5 mM sodium pyrophosphate, 25 mM MES, and 0.01% antifoam B, pH 6.5]. The filters were finally washed at 37°C in 0.1× SSC/0.1% SDS. cDNA inserts in positive ZipLox phage were recovered into plasmid pZL1 by in vitro excision. Only one clone (clone 6), which we designated rNaDC-1 (rat sodium-dicarboxylate transporter 1), was further subcloned into pBluescript II SK (Stratagene).

Sequencing of rNaDC-1. Deleted clones of rNaDC-1 obtained by the Kilo-Sequence Deletion kit (Takara, Japan) or specially synthesized oligonucleotide primers were used for the sequencing of rNaDC-1. Sequencing was performed by the dideoxy chain termination method using Sequenase (version 2.0, Amersham).

cRNA synthesis and uptake experiment in X. laevis oocytes. The plasmid DNA of rNaDC-1 in the vector of pBluescript II SK was used for the in vitro transcription. Capped cRNA was synthesized in vitro using T7 RNA polymerase from the linealized rNaDC-1 plasmid DNA with BamH I. Defolliculated oocytes were injected with 10 ng cRNA of rNaDC-1 and were incubated in Barth's solution containing gentamicin [88 mM NaCl, 1 mM KCl, 0.33 mM Ca(NO₃)₂, 0.4 mM CaCl₂, 0.8 mM MgSO₄, 2.4 mM NaHCO₃, 10 mM HEPES, and 50 µg/ml gentamicin, pH 7.4] at 18°C. After 2-3 days of incubation, uptake experiments were performed in ND96 solution (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES, pH 7.4) containing radiolabeled substrates as indicated in each experiment, at room temperature.

Electrophysiological experiment. After 2-3 days of incubation, oocytes injected with 10 ng cRNA of rNaDC-1 were used for the electrophysiological experiments. The two-microelectrode voltage-clamp method was employed for the measurement of currents. The oocytes expressed with rNaDC-1 were impaled in ND96 solution using two microelectrodes filled with 3 M KCl (holding potential = 60 mV). In the experiment shown in Fig. 4, oocytes were superfused with several 2-ml aliquots of ND96 solution containing 1 mM test substrates, and the elicited currents were recorded. For the kinetics experiments, oocytes were superfused with different concentrations of succinate, glutarate, -ketoglutarate, and citrate.

Measurement of charge to substrate coupling ratio of rNaDC-1. The uptake rates of 1 mM [¹⁴C]citrate in ND96 solution over 1, 3, and 5 min were measured in oocytes expressed with and without rNaDC-1. Just after this uptake experiment, electrophysiological measurements were performed using the same batch of oocytes. The oocytes expressed with rNaDC-1 were impaled using two microelectrodes in the ND96 solution containing 1 mM citrate, and the initial membrane potential was recorded in each oocyte. After washing with ND96 that did not contain citrate, the membrane potential of the oocyte was clamped at the initial potential recorded in each oocyte. Then each oocyte was superfused with ND96 solution containing 1 mM citrate, and the evoked currents were recorded. Eight to ten oocytes were used for both the uptake experiments and the electrophysiological measurements, and the mean values were used for the analysis. The currents evoked by 1 mM citrate were converted to the rate of net charge influx according to the following equation: rate of net charge influx (in mol/min) = 60 × A/F, where F is Faraday's constant (9.65 × 10⁴ C/mol), and A is current.

Northern blot analysis. Three micrograms of poly(A)⁺ RNA prepared from various rat tissues were electrophoresed on a 1% agarose/formaldehyde gel and transferred to a nitrocellulose filter. The filter was hybridized in a hybridization solution, overnight at 42°C, with a full-length rNaDC-1 cDNA that was randomly labeled with [³²P]dCTP. The filter was finally washed in 0.1× SSC/0.1% SDS at 65°C.

Preparation of polyclonal antibodies to rNaDC-1. A peptide consisting of 14 amino acids of the carboxy terminus of rNaDC-1 was synthesized, with a cysteine residue at the COOH terminus. This cysteine-terminating peptide was linked to maleimide-activated keyhole limpet hemocyanin (KLH). Rabbits were immunized with this KLH-linked synthetic peptide.

Immunohistochemistry of rNaDC-1. Male Sprague-Dawley rats were anesthetized with ether, and the kidneys were removed. The excised kidneys were fixed by immersion in 10% Formalin neutral buffer (Wako, Japan) for 48 h and then embedded in paraffin. Three-micrometer sections prepared on Silane-coated slide glass were deparaffinized, washed with Tris-buffered saline [50 mM Tris · HCl, pH 7.6, and 150 mM NaCl (TBS); DAKO] and treated for 5 min at 121°C. The sections were incubated in 3% H₂O₂ in methanol for 10 min to eliminate endogenous peroxidase activity. After rinsing in TBS, the sections were treated with 10% goat serum for 15 min. After being washed with TBS, the sections were exposed to antiserum against rNaDC-1 in a dilution of 1:500 for 1 h. After being rinsed in TBS, the sections were incubated with biotinylated rabbit IgG. The sections were then rinsed and incubated for 10 min with peroxidase-labeled streptavidin. After being treated with 3-amino-9-ethylcarbazole for 20 min, the sections were counterstained with hematoxylin and examined under a light microscope.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Isolation and sequence of rNaDC-1. A nondirectional cDNA library of rat kidney was screened using a cDNA fragment of rabbit Na⁺-dicarboxylate transporter (NaDC-1) under a low-stringency condition. After several rounds of screenings, three positive clones were isolated, only one of which (clone 6) was predicted to be a full-length clone. Clone 6 consisted of 2,245 nucleotides with an open-reading frame of 1,761 base pairs. The predicted amino acid sequence of clone 6 shared 73% and 75% identity to rabbit NaDC-1 and human NaDC-1, respectively. Hence, we designated clone 6 as rNaDC-1 (rat NaDC-1). Recently, a putative Na⁺-dicarboxylate transporter, Ri-19, was isolated from rat intestine (4), although its function was not determined. The amino acid sequence alignment of rNaDC-1 (rat), NaDC-1 (rabbit), hNaDC-1 (human), and Ri-19 are depicted in Fig. 1. The nucleotide sequence of Ri-19 is 97% identical to that of rNaDC-1; however, the amino acid sequence identity between rNaDC-1 and Ri-19 is only 79%.

View larger version (86K): [in this window] [in a new window]   Fig. 1.   Amino acid sequence alignment of rNaDC-1 (rat), NaDC-1 (rabbit), hNaDC-1 (human), and Ri-19 (a putative rat intestinal Na+-dicarboxylate transporter). Shaded background indicates amino acids that are common among several transporters. Numbers indicated above aligned sequences refer to the amino acid sequence of rNaDC-1.

Di- and tricarboxylate uptake via rNaDC-1 and inhibition study. Oocytes expressed with rNaDC-1 mediated a high level of uptake of 20 µM [¹⁴C]succinate (243.9 ± 19.8 pmol · oocyte¹ · h¹) compared with control oocytes (0.37 ± 0.04 pmol · oocyte¹ · h¹) (Fig. 2). Replacement of extracellular sodium with choline totally abolished rNaDC-1-mediated uptake of [¹⁴C]succinate (1.06 pmol · oocyte¹ · h¹) (Fig. 2). ¹⁴C-labeled -ketoglutarate, [¹⁴C]glutarate, and [¹⁴C]citrate were also transported by rNaDC-1 (data not shown). To determine the substrate selectivity of rNaDC-1, inhibition studies were performed. As shown in Fig. 3, uptake of 10 µM [¹⁴C]succinate via rNaDC-1 was potently inhibited by 1 mM succinate, fumarate, and malate (more than 90% inhibition), moderately by oxaloacetate and citrate (60-70%), and weakly by isocitrate (10%). No significant inhibition by maleate was observed.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   [14C]succinate uptake in oocytes expressed with rNaDC-1. Uptake of 20 µM [14C]succinate was measured in control oocytes or oocytes expressed with rNaDC-1 during 1 h in presence (Na+) or absence (choline+) of sodium.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Inhibitory effect of di- and tricarboxylates on rNaDC-1-mediated uptake of succinate. Uptake of 10 µM [14C]succinate was measured in oocytes expressed with rNaDC-1 in absence (control) or presence of 1 mM di- and tricarboxylates. Each value is expressed as percent of control.

Evoked currents by di- and tricarboxylates in oocytes expressed with rNaDC-1. Using the two-microelectrode voltage-clamp method, we directly showed that the transport of di- and tricarboxylates via rNaDC-1 is electrogenic. When oocytes expressed with rNaDC-1 were superfused with di- and tricarboxylates, inward currents were elicited (Fig. 4). This current was not observed when control oocytes were superfused by di- and tricarboxylates. The degrees of current evoked by succinate, -ketoglutarate, glutarate, fumarate, and malate were almost equivalent, but those by citrate and isocitrate were smaller (citrate > isocitrate); maleate did not evoke significant current. The degree of current evoked by each substrate is in concurrence with the degree of inhibition of succinate transport via rNaDC-1 by each substrate.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Evoked currents in an oocyte expressed with rNaDC-1 by di- and tricarboxylates. An oocyte expressed with rNaDC-1 was impaled with two microelectrodes in ND96 solution, and membrane potential was clamped at 60 mV. Then, the oocyte was superfused by 2 ml of ND96 solution containing 1 mM substrates, and the elicited currents were recorded.

Kinetics of transport via rNaDC-1. From the electrophysiological experiments, we determined the transport kinetics of di- and tricarboxylates. Figure 5 shows the current-concentration relationship of succinate, -ketoglutarate, and citrate in oocytes expressed with rNaDC-1. The currents evoked by different concentrations of succinate and -ketoglutarate followed the Michaelis-Menten equation, and the calculated K_m values of succinate and -ketoglutarate were 29.4 ± 2.5 and 57.1 ± 7.2 µM (means ± SE), respectively. In contrast, the current-concentration curve of citrate was resolved into a high-affinity (K_m = 319 ± 35.8 µM) and low-affinity (K_m = 11.8 ± 0.52 mM) component.

View larger version (9K): [in this window] [in a new window]   Fig. 5.   Concentration-current relationship of rNaDC-1. Oocytes expressed with rNaDC-1 were superfused with different concentrations of succinate, -ketoglutarate, and citrate, and the evoked currents (in nA) were plotted.

Coupling ratio of substrate to charge. To determine the coupling ratio of substrate to charge influx, we performed the electrophysiological experiments and the uptake measurements at the same time. [¹⁴C]citrate (1 mM) uptakes over 1, 3, and 5 min were measured in oocytes expressed with rNaDC-1 and control oocytes. Uptake of 1 mM citrate via rNaDC-1 increased linearly until 5 min (Fig. 6A). Just after the uptake experiments, the initial inward currents evoked by 1 mM citrate in the same batch of oocytes were recorded. Both experiments were performed at pH 7.4. The uptake rate of 1 mM citrate via rNaDC-1 calculated from the 1-min experiment was 2.05 ± 0.52 pmol · min¹ · oocyte¹ (Fig. 6B). The current evoked by 1 mM citrate was 3.60 ± 0.38 nA, and the charge influx calculated from this value was 2.24 ± 0.24 pmol · min¹ · oocyte¹ (Fig. 6B). This result demonstrates that citrate-to-charge influx ratio is 1:1 at pH 7.4. The results obtained from another experiment using a different batch of oocytes were the similar.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Stoichiometry of charge to citrate influx. A: uptake rate of 1 mM citrate was measured in control oocytes and oocytes expressed with rNaDC-1 during 1, 3, and 5 min. Values are means ± SE (n = 8-10). B: with the same batch of oocytes that were used in the uptake experiments in A, electrophysiological experiments were performed. Currents evoked by 1 mM citrates were recorded, and each value was converted to charge influx rate (in pmol · oocyte1 · min1). Values are means ± SE (n = 8).

Tissue distribution of rNaDC-1. The distribution of rNaDC-1 mRNA was determined using poly(A)⁺ RNA obtained from various tissues of rat. As shown in Fig. 7, a single transcript of rNaDC-1 (2.4 kb) was detected in the small and large intestines and in the kidney (small intestine > large intestine >> kidney). Expression of rNaDC-1 was not found in the other tissues, including the liver. Rabbit NaDC-1 mRNA was strongly expressed both in the small intestine and the kidney and relatively weakly in the liver and lung (6). In contrast, the expression of human NaDC-1 was weak in the kidney compared with that in the small intestine, and no signal was obtained in the liver (7). Thus the tissue distribution of rNaDC-1 is closer to that of human NaDC-1 rather than that of rabbit NaDC-1.

View larger version (63K): [in this window] [in a new window]   Fig. 7.   Northern blot analysis of rNaDC-1. Three micrograms of poly(A)+ RNAs from various tissues of rat were electrophoresed and blotted on to the nitrocellulose membrane. Membrane was hybridized with full-length cDNA of rNaDC-1 labeled with 32P and was washed under high-stringency conditions.

Membrane localization of rNaDC-1. Rat kidney sections were stained using polyclonal antibodies raised against the 14 polypeptides at the COOH terminus of rNaDC-1. Under low magnification (Fig. 8A), positive staining was found in the outer stripe of outer medulla and in cortex. In cortex, ~50% of proximal tubules seems to be stained, and proximal tubules just below the capsule are also clearly stained by the antibodies against rNaDC-1 (Fig. 8B). These results suggest that rNaDC-1 is localized at proximal tubule S2 and S3. As shown in Fig. 8C, only the luminal membrane of the proximal tubule cell was positively stained by the antibodies for rNaDC-1. No positive signal was obtained on the basolateral membrane of the proximal tubule cells. Staining of the luminal membrane was not observed when the kidney sections were treated with the serum of rabbits not immunized with rNaDC-1 (data not shown).

View larger version (85K): [in this window] [in a new window]   Fig. 8.   Immunohistochemistry of rNaDC-1. Three-micrometer sections of rat kidney were prepared and stained by antiserum raised against rNaDC-1 as described in METHODS. A: in the kidney, rNaDC-1 located in the outer stripe of the outer medulla and in cortex (×4). B: rNaDC-1 also localizes in the superficial cortex (×10). C: positive staining is only detected in the luminal membrane and not in the basolateral membrane (×40).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In this study, we isolated a rat renal Na⁺-dicarboxylate transporter (rNaDC-1) and investigated its localization and functional characteristics.

Recently, a putative rat intestinal Na⁺-dicarboxylate transporter (Ri-19) was isolated (4). Although the nucleotide sequences of rNaDC-1 and Ri-19 are nearly identical (97% identity), the amino acid sequence of rNaDC-1 exhibits only 79% identity to that of Ri-19. This discrepancy is attributed to several deletions and insertions of nucleotides in the coding region of Ri-19, which result in open-reading frame shifts. The partial amino acid sequences of rNaDC-1 (⁷⁸V-¹²¹P and ²⁵¹Q-³²⁷V), which are different from those of Ri-19, are both well conserved among rNaDC-1 (rat), NaDC-1 (rabbit), and hNaDC-1 (human) (Fig. 1). From these observations, we consider at this point that both Ri-19 and rNaDC-1 are encoded by the same gene, and Ri-19 is not a splice variant.

The membrane localization of rNaDC-1 is an important issue that must be elucidated. Previous physiological studies using membrane vesicles demonstrated that there exist two distinct Na⁺-dicarboxylate transporters in the proximal tubule cells (18, 23): one, a luminal Na⁺-dicarboxylate transporter with a low affinity to succinate, and the other, a basolateral Na⁺-dicarboxylate transporter with a high affinity to succinate. Previous physiological experiments have demonstrated that citrate transport via luminal Na⁺-dicarboxylate transporter was greatly increased when the extracellular pH was lowered; in contrast, citrate transport via basolateral Na⁺-dicarboxylate transporter is not affected by extracellular pH (23). The K_m value of succinate to rabbit NaDC-1 is high (450 µM), and citrate transport via NaDC-1 was increased when the extracellular pH was lowered (6). From these functional characteristics, NaDC-1 has been postulated to be the luminal Na⁺-dicarboxylate transporter. Western blot analyses also suggest that NaDC-1 is located in the luminal membrane (8). In the present study, we directly demonstrated that rNaDC-1 is a luminal Na⁺-dicarboxylate transporter by immunohistochemical analysis. Antisera raised against the COOH terminus of rNaDC-1 only stained the luminal membrane of the proximal tubule. We detected no positive signals on the basolateral membrane. The data obtained from the inhibition study (Fig. 3) and the electrophysiological experiment (Fig. 4) are consistent with the substrate specificity of the rat luminal Na⁺-dicarboxylate transport system (14). Tricarboxylic acid cycle intermediates present in the plasma are freely filtered through the glomerulus, and only 3-35% are found in the urine (16). The rest are reabsorbed by proximal tubule cells. rNaDC-1 is considered to play an important role in the uptake of these di- and tricarboxylates from the glomerular filtrate. Our results also suggest that the basolateral Na⁺-dicarboxylate transporter is actually a molecule different from the luminal Na⁺-dicarboxylate transporter. Until now, several organic anion transporters in proximal tubule cells have been cloned (13, 17). Isolation and characterization of basolateral Na⁺-dicarboxylate transporter is required for the clarification of the organic anion transport in the proximal tubule.

Previous experiments using voltage-sensitive dyes have indicated that Na⁺-dicarboxylate transport across the brush-border membrane is electrogenic (22). Influxes of several sodium ions and polyvalent anionic substrates occur simultaneously during the transport of di- and tricarboxylates via the Na⁺-dicarboxylate transporters. Therefore, the electrogenic properties of rNaDC-1 depend on 1) the stoichiometry of Na⁺ to substrates and 2) the electrical species of substrates (neutral, monovalent, divalent, or trivalent). By comparing the influx of [²²Na⁺] with [¹⁴C]succinate into the rabbit luminal membrane vesicles, Wright et al. (21) concluded that Na⁺-to-succinate coupling ratio was 3 in the luminal Na⁺-dicarboxylate transporter. In the present study, we confirmed that Na⁺-dependent transport of di- and tricarboxylates via rNaDC-1 is electrogenic. Furthermore, we determined the species of citrate that is transported by rNaDC-1. Although citrate exists predominantly in a trivalent form at pH 7.4 (16), the transportable species of citrate must be an electrically neutral, monovalent, or divalent; if citrate is transported in the trivalent form with three sodium ions, then the transport process should not be electrogenic. The electrogenic transport of citrate via rNaDC-1 can be explained also by a symport of four sodium ions with a trivalent citrate molecule; however, this mode of transport seems to be unlikely. By converting the inward current to the charge influx, we demonstrated that one charge movement occurs along per molecule of citrate taken up. Assuming that Na⁺-to-citrate coupling ratio is 3, this result indicates that citrate is transported by rNaDC-1 in a divalent form. As mentioned above, the predominant form of citrate at pH 7.4 is trivalent, and under more acidic conditions, the predominant form is divalent. Our results explain why the uptake of citrate by luminal Na⁺-dicarboxylate transporter increased when the extracellular pH was lowered. When the extracellular pH decreased, the proportion of transportable species of citrate (divalent form) increases. As a result, transport of citrate via rNaDC-1 increases.

The kinetics data obtained from the electrophysiological experiments indicate that succinate, -ketoglutarate, and glutarate are relatively high-affinity substrates to rNaDC-1 (K_m = 29.4, 57.1, and 78.3 µM, respectively). The K_m value for succinate is not identical to those for rabbit NaDC-1 (0.5 mM) and human NaDC-1 (0.8 mM). This difference is considered to be in part due to the different measurement systems used in each experiment. Kinetics data of human and rabbit NaDC-1 were determined by uptake experiments in oocytes using radiolabeled substrates over 60 min. In the present study, we measured the initial currents from the electrophysiological experiments, which correspond to the initial velocity of transport. The K_m value of succinate (100 µM) obtained from the membrane vesicle studies using electrosensitive dyes (22) is similar to that of our result. Furthermore, in the present experiment, the membrane potential of oocytes was fixed at 60 mV. It was reported that the K_m value of succinate decreased as the membrane potential became more negative in rabbit renal brush-border membranes (19). Thus the K_m value of succinate obtained in the present study is considered to be lower than those of human and rabbit NaDC-1, which were measured without fixing membrane potential. The other possible reason for this difference in affinity is species difference. Although K_m values for succinate (29.4 µM) and glutarate (78.3 µM) are similar for rat NaDC-1, K_m values of glutarate for rabbit NaDC-1 (7.3 mM) and human NaDC-1 (6.3 mM) are 10 times higher than those of succinate (9). Not only glutarate but also -ketoglutarate show different affinity between rat and rabbit NaDC-1. In the experiment using COS cells expressing rabbit NaDC-1, 1 mM -ketoglutarate showed no inhibitory effect on succinate transport via NaDC-1 (10). Thus the transport properties of rNaDC-1 are different from those of rabbit and human NaDC-1.

rNaDC-1 may be useful to clarify the pathophysiological state underlying nephrolithiasis. Citrate forms complexes with ionic calcium and decreases the free calcium concentration in urine. Thus the citrate in urine inhibits renal stone formation (11). Low urinary excretion of citrate is associated with renal stone formation, and oral administration of citrate has been shown to be effective for protection against nephrolithiasis (11). The reabsorption rate of citrate from the urine by the proximal tubule cells, which is dependent on the transport activity of rNaDC-1, affects the incidence of urolithiasis. Several conditions (e.g., potassium depletion, starvation, and administration of acetazolamide) have been associated with decreased citrate excretion, which may reflect the upregulation of Na⁺-dicarboxylate transporter.

At the basolateral membrane of the proximal tubule cells, there exists another organic anion transporter, the PAH transporter. PAH transporter is a multispecific organic anion transporter that plays a central role in the excretion of a variety of endogenous and exogenous organic anions from the proximal tubule (5, 12, 17). Previous studies have suggested that PAH transporter acts as an organic anion/dicarboxylate exchanger (12, 15). When the membrane vesicles or isolated proximal tubules were incubated with low concentrations of dicarboxylates in the medium containing sodium ion, the uptake rate of PAH increased significantly. This phenomenon was not observed in the absence of sodium or when the concentrations of extracellular dicarboxylate were high. The stimulatory effect of low concentrations of dicarboxylates was explained as follows; dicarboxylates were first taken up by the sodium-dicarboxylate transporter, and from the intracellular side dicarboxylates stimulated the uptake of PAH via organic anion/dicarboxylate exchanger. Therefore Na⁺-dicarboxylate transporters and PAH transporter have been considered to be functionally related. As already mentioned, there exist two distinct sodium-dicarboxylate transporters, the luminal and the basolateral Na⁺-dicarboxylate transporters. When we consider the fact that luminal Na⁺-dicarboxylate transporter possesses a high transport capacity for dicarboxylates, the luminal Na⁺-dicarboxylate transporter (rNaDC-1) may largely contribute to the PAH uptake at the basolateral membrane. Recently, we isolated PAH transporter (OAT1) from a rat kidney cDNA library using expression cloning method and demonstrated that if both OAT1 and rNaDC-1 were expressed in the same oocyte and preincubated with glutarate, then PAH uptake increased markedly (13). Thus at least in the nonpolar cell, PAH transporter (OAT1) cooperates with rNaDC-1. In the present study, it was revealed that rNaDC-1 is localized in S2 and S3 of proximal tubule cells. OAT1 seems predominantly expressed in S2 (13). Therefore, at least in rat, rNaDC-1 could stimulate PAH transport via OAT1 in S2. In fact, Dantzler and Evans (2) reported that incubation of isolated rabbit proximal tubule S2 with -ketoglutarate or glutarate from the luminal side showed stimulatory effect on net PAH secretory transport, although the incubation from luminal side was less effective compared with that from the basolateral side. Further studies are required to clarify the low stimulatory effect of dicarboxylates from luminal side on the PAH uptake via PAH/dicarboxylate exchanger.

In conclusion, we isolated and characterized the rat Na⁺-dicarboxylate transporter (rNaDC-1). rNaDC-1 is localized at the luminal membrane of S2 and S3. rNaDC-1 mediates di- and tricarboxylate transport in a sodium-dependent manner, and the transport via rNaDC-1 is electrogenic. Citrate is transported by rNaDC-1 in the divalent form at physiological pH, and this may be the reason for the increase of citrate transport across the luminal membrane when the extracellular pH was lowered.