468 Sodium cotransporters Ernest M Wright*, Donald DF Loo, Eric Turk and Bruce A Hirayama Recent studies of cloned mammalian sodium cotransporters in heterologous systems have revealed that these integral membrane proteins serve multiple functions as cotransporters, uniporters, channels and water transporters. Some progress has been gained in understanding their secondary structure, but information on helical bundling and tertiary structure is lacking. Site-directed mutagenesis and the construction of chimeras have resulted in the identification of residues and domains involved in ligand binding, and natural mutations have also been found that are responsible for human genetic diseases. Major factors in the short-term regulation of cotransporter function by protein kinases are exocytosis and endocytosis. Addresses Department of Physiology, School of Medicine, University of California at Los Angeles, CA 90095-1751, USA *e-mail: ERNEST@PHYSIOLOGY.MEDSCH.UCLA.EDU Current Opinion in Cell Biology 1996, 8:468-473 © Current Biology Ltd ISSN 0955-0674 Abbreviations EAAT excitatory amino-acid cotransporter GABA T-aminobutyricacid GAT1 GABA cotransporter 1 NET norepinephrine cotransporter PICA protein kinase A PKC protein kinase C SGLT Na+/glucose cotransporter Introduction Sodium cotransporters (symporters) belong to a functional superfamily of membrane proteins responsible for the accumulation of ions and molecules in cells [1]. Since 1992 [2], there has been an expanding quantity of literature on new cotransporter clones, but most of these are isoforms and homologs of existing clones. A small number of new mammalian cotransporters have been isolated by expression cloning, a technique introduced by this laboratory [3], and these include the ileal Na+/bile salt [4], renal Na÷/dicarboxylate [5"], and thyroid Na+/iodide [6"] cotransporters. In this review, we will largely focus on papers describing the structure, function and regulation of mammalian cotransporters, including the Na+/glucose and Na÷/neurotransmitter cotransporter families. Where possible, we will relate these advances to those in other cotransporter families such as the plant and animal H÷ cotransporter families. EVidence for multiple functions of sodium cotransporters The overexpression of cloned sodium cotransporters in Xenopus laevis oocytes and cultured cells, and the use of biophysical techniques, to study their function, has revealed unforeseen properties of this membrane protein class (see Table 1). Apart from their most obvious task in coupling the uphill transport of substrates (e.g. glucose, T-aminobutyric acid [GABA], glutamate, and norepinephrine) to the downhill transport of Na ÷, they also function as uniporters, ion and water channels, and water transporters. The stoichiometry between ion and substrate transport has been established using ra- dioactive tracer, steady-state cotransport current, kinetic and thermodynamic measurements [7",8"-11"]. In the absence of substrates, transport of ions occurs in an uncoupled mode ('leak'), and this is observed as an increase in ion fluxes or currents above the level of the untransfected cell [7"',8",11",12,13,14"']. The leak currents for the Na÷/glucose cotransporter are 5-10% of the maximum sugar-stimulated currents, and they are abolished by a blocker of Na+/glucose cotransport (phlorizin) [7"'] The ion selectivity of this uncoupled current is H+>>Na+>Li+>chotine [7"',12], and the reversal potential (Er) varies with Na ÷ concentration according to the Nernst potential (ENa÷). The kinetics of the Na+-uncoupled current mimic the kinetics of Na÷/glucose cotransport with the same apparent Na + affinity (3-5 mM at -150mV) and Hill coefficient (1.8) (B Hirayama, M Panayotova-Heiermann, unpublished data). Na + leak currents have also been observed with the low-affinity Na+/glucose cotransporter 2 (SGLT2) [15], and H + leak currents have been recorded for the plant H+/glucose and H+/amino acid cotransporters [16,17"], but not the human H+/dipeptide cotransporter [18"]. Uncoupled cation trans- port through cotransporters leads to variations in apparent cation/substrate coupling coefficients [8",19"]. Cation leaks (slippage) have also been reported for the GABA, glutamate, noradrenaline and serotonin cotrans- porters [11",13,14"',20], and, in the case of the GABA carrier GAT1 (GABA cotransporter 1) [21"], unitary cur- rent measurements indicate that the leak occurs through an ion channel. The ion selectivity of the leak in an excitatory amino-acid cotransporter (EAAT)I-2 chimera was K+(1.6)>Na+(1)>Li+(0.8)>choline(0.04) [ 14"]. Evidence has accumulated that the neurotransmitter cotransporters contain ligand-gated ion channels. The initial observation was that the cotransporter-coupled currents were larger than expected from the cotransporter density and turnover number [9",11",13,20,22"]. For the excitatory amino-acid cotransporters EAAT1 and EAAT4 [22",23"], there is compelling evidence for a CI- con- ductance in parallel with Na+/glutamate transport, and for the glutamate transporter in photoreceptors [24"'] noise analysis shows that the conductance is due to a 0.7 picoSiemen (pS) chloride channel. The channel was seen only in the presence of Na + and glutamate, and