Molecular properties of the SLC13 family of dicarboxylate and sulfate transporters

Introduction

The SLC13 gene family consists of five members in humans, all of which are sodium-coupled transporters for anions such as sulfate or dicarboxylates. SLC13 transporters with related sequences and function are found in other vertebrates, ranging from fish to mammals. These transporters are secondary active transporters that couple multiple sodium ions to the movement of an anion substrate. The SLC13 family transporters are typically found on plasma membranes in mammalian cells. However, the newer members of the family that have been identified in insects and plants do not necessarily couple sodium and can be found in intracellular organelles.

This review is intended to provide an update on developments in the field since the publication of previous reviews on the members of the SLC13 family [44, 45, 47, 53, 54]. The most recent review of the field, by Markovich and Murer [47] focused primarily on cloning of mammalian members of the family, their physiological function and regulation. The present review adds new information published in the last year, as well as more details on structure–function studies, electrophysiological properties and non-vertebrate members of the family.

SLC13 gene family: vertebrates

Humans contain five genes coding for SLC13 family members, all of which are approximately 40%–50% identical in protein sequence. The genes have been named in the order in which they were identified (Table 1). Although some attempts have been made to organize the protein names [47], the issue has not been completely resolved. The suggested use of NaC for the di- and tri-carboxylate transporters [47] is somewhat problematic since the name is already in use for the NaC family of voltage-gated ion channels that includes ENaC. The gene nomenclature also needs revision. The current SLC system is limited because it does not provide a mechanism for including family members from distantly related species, including bacteria and plants, nor does it allow the naming of subfamilies. A more comprehensive nomenclature based on evolution and independent of substrate should be developed, similar to the one proposed for the organic anion transporting polypeptide family, OATP [24]. In order to allow the naming of subfamilies of OATPs, the gene family name was changed from SLC26 to SLCO. A similar change to the SLC13 family nomenclature would allow the inclusion of the newer family members from non-vertebrate species.

SLC13A1: NaS1

The first member of the SLC13 family to be identified was the sodium-coupled inorganic sulfate transporter, originally named NaSi-1 but more recently renamed NaS1 [47]. The human gene name is SLC13A1 (Table 1). NaS1 carries a range of divalent oxyanion substrates such as sulfate ${\text{(SO}}_4^{2-})$, selenate ${\text{(SeO}}_4^{2-})$, and thiosulfate ${\text{(S}}_2{\text{O}}_3^{2-})$ [11, 40]. The transport is sodium dependent, with at least three sodium ions coupled to the transport of each substrate. The K_m for sulfate in the human NaS1 is around 0.4 mM, and the half-saturation constant for sodium (K_Na) is about 30 mM [41]. The K_m for sulfate in rat NaS1 is about 0.7 mM (radiotracer uptakes) and 0.1 mM (substrate-dependent inward currents) [11, 46]. Initially, the NaS1 transporter was assumed to be electroneutral because the Hill coefficient in non-voltage clamped cells was only around 1.4. However, the transporter produces substrate-dependent inward currents and the Hill coefficient measured under voltage clamped conditions is 2.8 [11].

The NaS1 protein is found on the apical membrane in the renal proximal tubule. In mouse and rat, the protein or message is found in both kidney and small intestine, but in humans the transporter may be restricted to the kidney [5, 40, 43]. Antibodies against human NaS1 recognize a glycosylated protein in the kidney membrane vesicles [42] of both pig and rat. Most of the recent studies on NaS1 have focused on the production and characterization of the Nas1 knockout mouse [14, 16, 17]. Mice with a disruption in the Nas1 gene exhibit increased renal excretion of sulfate and a general decrease in growth. These mice also exhibit behavioral abnormalities and memory disturbances.

SLC13A2: NaDC1

The SLC13A2 gene codes for the low affinity Na⁺/dicarboxylate transporter, NaDC1 (Table 1). NaDC1 is found on apical membranes of the cells of the renal proximal tubule and small intestine where it functions to absorb citric acid cycle intermediates from the renal filtrate or the intestinal lumen (see reviews [53, 54]). The preferred substrates of NaDC1 are dicarboxylates such as succinate. Substrates such as citrate that exist mostly as tricarboxylates at physiological pH, are transported in protonated form as citrate²⁻. The K_m for the test substrate succinate varies between 0.3 and 1 mM, depending on the species. Transport is sodium dependent, with three sodium ions coupled to the transport of a divalent anion substrate. This transporter also interacts with lithium, which at low concentrations can compete with sodium to inhibit transport [54].

NaDC1 has a broad substrate specificity, in general the preferred substrates are four carbon dicarboxylates in trans configuration. One measure of substrate specificity that is rarely reported for the SLC13 family is the transport efficiency, estimated from the ratio of V_max/K_m (for radiotracer uptakes) or I_max/K_0.5 (for two-electrode voltage clamp measurements). Typically, K_m values for substrates are compared but those do not always accurately reflect transport efficiency. For example, in the case of hNaDC1, the K_0.5 for succinate is lower than that for fumarate, which might suggest that the preferred substrate is succinate [68]. However, the I_max value for fumarate is much larger than that for succinate, making the efficiency or I_max/K_0.5 ratio greater for fumarate than succinate [68]. The transport efficiency for citrate is usually underestimated since the K_m for citrate is given as total citrate even though the true substrate, citrate²⁻, is only about 7% of the total at pH 7.5. For example, the K_0.5 of total citrate is 6.8 mM in hNaDC1, but the citrate²⁻ K_0.5 is approximately 0.5 mM at pH 7.5. Since the I_max value represents inward current, carried by sodium, this reflects the movement of the true substrate, citrate²⁻. The I_max/K_0.5 ratio for citrate in hNaDC-1 is 1134 nA/mM, followed by fumarate at 365 and succinate at 166 [68]. Therefore, when the I_max/K_0.5 ratios are considered, the efficiency of transporting citrate by NaDC1 is much greater than fumarate or succinate. Because of differences in protein expression between different transporters, the relative efficiencies for different substrates can be compared but not the actual I_max/K_0.5 values unless there is independent evidence that the protein concentrations are the same. In addition to the transport efficiency for a given substrate, the plasma concentrations of substrates are also important determinants of the physiological function of the transporters. In humans, plasma concentrations of substrates such as succinate, α-ketoglutarate and citrate are approximately 5, 25 and 100 μM, respectively [27, 39].

There is relatively little information about the in vivo regulation of NaDC1. Most of the research with the dicarboxylate transporters has focused on functional characterization and structure–function studies. There is evidence that NaDC1 in the kidney is regulated by metabolic acidosis [3]. The trafficking of NaDC1 is also affected by kinases, including SGK-1, that are involved in regulating transporters involved in acid–base balance [6]. The recent development of a knockout mouse model may help clarify the physiological functions of NaDC1. Mice with disruptions in the Slc13a2 gene survive to adulthood with no obvious phenotypic aberrations and exhibit increased urinary excretion of citrate (C.D. Smith, personal communication).

SLC13A3: NaDC3

The high affinity Na⁺/dicarboxylate transporter, NaDC3, is encoded by the SLC13A3 gene (Table 1). The human NaDC3 is 43% identical in protein sequence to hNaDC1 [47]. NaDC3 has a broader tissue distribution than NaDC1 and is found on basolateral membranes in renal proximal tubule cells [28], as well as liver, brain and placenta [12, 37, 56]. The K_m for succinate in NaDC3 is between 10 μM and 140 μM, depending on the species [56, 66]. The substrate specificities of NaDC1 and NaDC3 overlap somewhat but NaDC3 appears to handle a larger range of structures, particularly longer chain or branched dicarboxylates, compared with NaDC1. For example, an important physiological substrate of NaDC3 is α-ketoglutarate [12], whereas NaDC1 has a very low transport capacity for this substrate [68]. NaDC3 handles substrates with substitutions on the third carbon, such as 2,3-dimethylsuccinate or the heavy metal chelators, meso-2,3-dimercaptosuccinate and dimercaptopropane-1-sulfonate [7]. In the brain, NaDC3 mediates the transport of the modified amino acid, N-acetylaspartate, with a relatively high affinity (K_m 60–250 μM) [21]. NaDC3 may also interact with some monovalent organic anions, such as benzylpenicillin [8]. The transport of substrates by NaDC3 is sodium-dependent, probably involving three Na⁺ ions with each substrate, and transport is electrogenic. Similar to NaDC1, the transport by NaDC3 is inhibited by millimolar concentrations of lithium [12, 66]. There is relatively little information on the structure or in vivo regulation of NaDC3, although there has been one study showing that treatment with protein kinase C activators produces endocytosis of the transporter [25]. There is no knockout mouse model yet.

SLC13A4: NaS2

NaS2, originally named SUT-1, is a sodium-dependent sulfate transporter found predominantly in human placenta, high endothelial venules and testis [22, 48]. In rats the message is also found in brain and liver [15]. The sequence of hNaS2 is 49% identical to that of hNaS1 [47]. The functional properties of NaS2 are very similar to those of NaS1: the K_m for sulfate in NaS2 is approximately 0.4 mM and the K_Na is around 20 mM [48]. In rat NaS2, the K_m for sulfate is 1.3 mM [15]. Although it has not been examined in detail, the range of substrates handled by NaS2 also appears to be very similar to the substrate specificity of NaS1. To date there is relatively little information on NaS2, and it is not known whether there are any functional differences between NaS2 and NaS1.

SLC13A5: NaCT

The most recent addition to the human SLC13 family is the Na⁺-citrate transporter, NaCT. This transporter was first identified in brain, but NaCT message is also found in liver and testis [32, 33]. The sequence of the human NaCT is 54% identical to that of hNaDC1 and 47% identical to hNaDC3 [47]. A splice variant of NaCT, missing 43 amino acids due to the loss of exon 2, has been reported in human brain, but the function and tissue distribution of this variant is unknown (Genbank AJ489980). In preliminary studies, we have detected a similar splice variant in mouse brain by RT-PCR (unpublished observations).

NaCT is a sodium-dependent transporter for citrate and dicarboxylates. In the rat and human transporters, the K_m for total citrate is lower than for dicarboxylates such as succinate. In mouse NaCT, the K_m is the same for citrate and succinate, around 38 μM [31]. However, the citrate V_max is larger than the succinate V_max, thus the transport efficiency with citrate is approximately twice that of succinate [31]. There are large species differences in citrate affinity, for comparison the K_m for citrate in human NaCT is around 0.6 mM [32]. There is some evidence that NaCT transports citrate as a trivalent anion coupled with four sodium ions. NaCT has a pH optimum of seven for citrate transport [31], compared with the pH profiles of NaDC1 and NaDC3, which show the highest transport of citrate at acidic pH, in accordance with the highest concentration of protonated citrate [62]. In sodium-activation experiments, the Hill coefficient of NaCT is greater than three, suggesting that more than three sodium ions are involved in activating transport [33]. In the rat NaCT, the charge to substrate ratio was determined to be 1:1 for citrate and 2:1 for succinate, again suggesting that four sodium ions may be coupled to the transport of trivalent citrate or divalent succinate [31]. NaCT exhibits species differences in interaction with lithium [34]. The rat NaCT is inhibited by lithium, similar to other dicarboxylate transporters, but the human NaCT is stimulated by lithium. The addition of lithium appears to increase the efficiency of transporting citrate by the human NaCT, mediated by a combination of increased V_max and decreased K_m for citrate [34].

SLC13 family: electrical properties

The members of the SLC13 family found in vertebrates are electrogenic transporters that couple multiple Na⁺ ions to the movement of a divalent or trivalent anion substrate. The electrical properties of many of the members of the SLC13 family have been examined by measuring whole-cell currents using the two-electrode voltage clamp technique. At least three different types of currents have been associated with the expression of SLC13 transporters.

Structure–function studies

Earlier studies, summarized in previous reviews, have shown that carboxy terminal half of the SLC13 transport proteins is required for substrate recognition [47, 54]. For example, chimeras made between NaS1 and NaDC1 have the substrate specificity of the donor of transmembrane helices (TM) 7–11. Furthermore, residues involved in binding substrates and cations, either directly or indirectly, have also been identified in TM 7–11 [54].

SLC13 gene family: other species

Several new members of the SLC13 family have been identified in insects, nematode worms, plants and bacteria. These transporters exhibit a greater variety of cellular distribution and transport mechanism compared with the vertebrate transporters.

Conclusion

The SLC13 family is relatively small, comprising only five genes in mammals, and with only two categories of substrates, either sulfate or di-/tricarboxylates. Considerable advances have been made in this area in recent years, although some unanswered questions remain including: what is the structure of the proteins, how do the transporters work, how are they regulated, and what are their physiological functions? The addition of newer members of the family, such as Indy, has helped to identify potential functions, including the possible involvement of dicarboxylate transporters in metabolic regulation and lifespan determination. The development of knockout mouse models should also help to identify physiological roles of these transporters. Finally, the identification of homologs from bacteria should allow the over expression and purification of the proteins for structural analysis.