There are two nomenclatures for GABA transporters (GATs). In the rat/human nomenclature, GABA transporters are called GAT-1, betaine/GABA transporter-1 (BGT-1), GAT-2, and GAT-3 (Borden et al. 1992). In the mouse nomenclature, GATs are called GAT1, GAT2, GAT3, and GAT4 (without hyphen, Liu et al. 1993). They were encoded by the genes SLC6A1 (for GAT-1) and SLC6A11–13 (for GAT-3, BGT-1, and GAT-2, respectively). In this review, the rat/human nomenclature will be used.
The amino acid γ-amino butyric acid (GABA) is the main inhibitory neurotransmitter in the adult brain. The actions of GABA are mediated via two distinct classes of receptors: ionotropic ( GABAA and GABAc) and metabotropic (GABAB) receptors (Farrant and Kaila 2007). For adequate synaptic transmission, the released GABA has to be rapidly sequestered from the synaptic cleft by high affinity GABA transporters. Experiments performed in neuronal and astroglial cell cultures in the 1970s revealed that a number of GABA analogs selectivity inhibited GABA uptake in one or the other culture type. Based on these results it was concluded that neurons and astrocytes express distinct GABA transporters. Molecular biological techniques revealed that most probably four distinct GABA transporters exist (Borden 1996).
Properties and Subtypes of GABA Transporters
Four different GATs have been identified in rodents and humans so far (Borden 1996). These four GATs have different GABA affinities, distinct pharmacological profiles and show a specific cellular distribution.
GAT-1 was the first GABA transporter to be cloned. Data obtained in different expression system (Xenopus oocytes, COS-7 cells, etc.) revealed that GAT-1 has an IC50 of 5–30 μM and is Na+ and Cl− dependent. The expression of GAT-1 mRNA is restricted to the central nervous system, where it was found to be present in all brain regions examined. Immunohistochemical studies showed that all GABAergic neurons (as identified by the expression of GAD67) express GAT-1. In addition, some non-GABAergic neurons and glial cells (for instance, Bergmann glial cells in the cerebellum) also express GAT-1. GAT-1 immunoreactivity is almost exclusively associated with punctate structures resembling axon terminals and fibers. These observations indicate that GAT-1 is a preferentially neuronal GABA transporter and is predominantly synaptically located, suggesting that the main physiological function of GAT-1 is the fast removal of GABA from the synaptic cleft, i.e., termination of GABAergic transmission, and the recycling of GABA to the presynaptic terminal.
GAT-2 shows about 52% amino acid identity with GAT-1. GAT-2 has an IC50 of 17 μM and is also Na+ and Cl− dependent. GAT-2 mRNA is present both in the central nervous system and in a number of nonneural tissues such as kidney, liver, or heart. Although GAT-2 immunoreactivity was mainly detected in the leptomeninges (pia and arachnoid) surrounding the brain, GAT-2 staining was also observed in some neuronal and glial cells. Because GAT-2 positive punctae are mainly localized to glial limitans and outline blood vessels, GAT-2 may regulate GABA levels in the cerebrospinal fluid, thereby regulating GABAergic transmission indirectly.
GAT-3 displays 52% amino acid identity with GAT-1 and 67% amino acid identity with GAT-2. GAT-3 has an IC50 of 33 μM and is also Na+ and Cl− dependent. The expression of GAT-3 mRNA is restricted to the central nervous system. Immunohistochemical studies show that GAT-3 is predominantly expressed in astrocytes. Moreover, GAT-3 is preferentially located in distal astrocytic processes including perivascular end-feet. GAT-3 positive punctae are adjacent to axon terminals making both symmetrical and asymmetrical synapses. These observations suggest that the main physiological function of GAT-3 is a removal of GABA that escaped from the synaptic cleft, i.e., prevention of synaptic crosstalk.
BGT-1 was firstly identified in kidney, where it mediates the accumulation of the osmolyte betaine in renal medullary epithelial cells to make them resistant to hypertonicity. Since this transporter is able to utilize also GABA as substrate it was termed BGT-1 (for Betaine/GABA transporter). Surprisingly, BGT-1 has a higher affinity for GABA (IC50 = 42 μM) than for betaine (IC50 = 400 μM). This is, however, in line with BGT-1 function in the kidney because the concentration of betaine in plasma is about 180 μM, while GABA concentration in plasma is <1 μM. BGT-1 shows 68% and 65% degrees of identity with GAT-2 and GAT-3 (Kempson et al. 2014). BGT-1 is also expressed in the CNS at low levels, with a concentration in leptomeninges, but its function in the brain is uncertain (Kempson et al. 2014).
Despite these evident differences in the cellular and subcellular localization of GATs in the adult CNS, the situations seems to be less clear in the immature brain, where a substantial abundance of GAT-1 in astrocytes and of GAT-3 in neurons was reported (Conti et al. 2004). In rodent neocortex, GAT-1 expression is rather low at birth and reaches adult levels approximately by the end of third postnatal week (Conti et al. 2004). In contrast, GAT-2 and GAT-3 expression is relatively high at birth and reaches the adult levels the first and second first postnatal weeks, respectively (Conti et al. 2004).
Stoichiometry and Kinetics of GABA Transporters
For neuronal cells, the concentrations of molecules involved in GABA transport can be found in the literature: [Na+]i = 15 mM, [Cl−]i = 7 mM, [GABA]i = 2 mM, [Na+]o = 150 mM, [Cl−]i = 135 mM, [GABA]i = 0.1 μM. Surprisingly, these values predict the reversal potential for GABA transport of about −60 mV, i.e., unexpectedly close to the resting potential (Attwell et al. 1993; Richerson and Wu 2003). Only if membrane potential is more negative than reversal potential, GABA transporters will operate in the inward direction, i.e., will take up GABA. But when membrane potential is more positive than the reversal potential, which can e.g., occur during excitatory synaptic transmission, GABA transporters will operate in the reverse mode and will release GABA. Moreover, the elevations of either [Na+]i (for instance, as consequence of AMPA/NMDA receptor activation) or [Cl-]i (for instance, as consequence of GABAA receptor activation) will increase the reversal potential of GABA transport therefore favoring its switch to reversed mode LIT . This suggests that synaptic activity can modulate GAT-mediated uptake/release and thus extracellular GABA levels. However, such a synaptic modulation is only possible if GATs are fast enough to follow these dynamic processes. Although GATs expressed in Xenopus oocytes demonstrated rather slow kinetics (one cycle per about 100 ms), recent data indicate that the turnover rate is faster than previously believed, taking about 10 ms per translocation at 37 °C. Moreover, recent study in both artificial expression system and in hippocampal neurons showed that GAT-1 is fast enough to be reversed during an action potential (Wu et al. 2007).
Modulation of GABA Transport
In addition to the modulation via membrane voltage and intracellular ions, functional modulation of GATs occurs through a variety of second messengers such as kinases, arachidonic acid, and pH (Beckman and Quick 1998). These factors may act directly on the transporter protein or by modulating the interaction with other proteins. Interestingly, not only GAT kinetics is regulated but also the number of GATs expressed on the surface. GATs could be internalized and reinserted into the plasma membrane within minutes and, moreover, GAT-1 trafficking resembles the cycling of neurotransmitter-filled synaptic vesicles (Deken et al. 2003). Such redistribution of GAT-1 protein between the plasma membrane and the cytoplasm is regulated by different G-protein coupled receptors via protein kinase C (PKC). PKC activation leads to GAT-1 internalization, while PKC inhibition increases the surface expression of GAT-1 (Beckman et al. 1999). The surface expression of GAT-1 is also dependent on the availability of its substrate. GAT-1 stimulation by GABA or other transported substrates increases GAT-1 surface expression, whereas GAT-1 inhibitors that are not substrates induce GAT-1 internalization. Thus, the number of GAT-1 on the cell surface and/or synapse is fine-tuned by GABA itself (Bernstein and Quick 1999).
GABA Transporters in Diseases
A variety of reports suggest an essential function of GATs in the etiology of neuronal diseases like epilepsy (Richerson and Wu 2004), schizophrenia (Schleimer et al. 2004), dyskinesia (Son et al. 2014), or anxiety disorders (Thoeringer et al. 2009). There is a strong link between GAT function and epilepsy, although the functional implications of GAT in epilepsy are far from being completely understood. During epileptic activity, the enhanced neuronal activity promotes the reversal of GATs and thus a nonvesicular GABA release that can dampen neuronal excitation. Accordingly the inhibition of GATs attenuate epileptic seizures, therefore GAT inhibitors are used for antiepileptic medication. In addition, alterations in the expression of GATs, in particular of GAT-1, has been reported in epileptic animals, where an up- and downregulation of different GATs has been described (Richerson and Wu 2004). The alteration in the GAT expression levels may either enhance nonvesicular GABA release under epileptic conditions, thus contributing to homeostatic inhibition, but can also aggravate seizures. These complex alterations in epileptic tissue complicate a pharmacological interaction with GATs (Allen et al. 2004; Conti et al. 2004).
A beneficial effect of GATs has been suggested for brain ischemia. Under ischemic condition, the GATs reversed due to a massive increase in the intracellular Na+ concentration under this condition, with lead to the nonsynaptic release of GABA. Although this increased GABA concentration may promote a neuroprotective effect, it can also contribute to cell swelling, thus aggravating excitotoxic insults. Recent reports indicate that this GABA release occurred mainly via neuronal GAT-1, while glial GAT-3 does not contribute. In this respect it is also interesting that ischemic conditions stimulate the synthesis of GABA, thus additionally increasing the driving force for reversed GABA transport (Allen et al. 2004).
GABA transporters belong to the family of Na+-Cl−-dependent neurotransmitter transporters and mediate the electrogenic symport of 1 molecule GABA with 1 Cl− and 2 Na+ ions. Due to its transport stochiometry, the regular inward transport direction can be easily reversed under various physiological and pathophysiological conditions, thereby mediating nonvesicular GABA release. Four different GATs have been described so far: GAT-1 is mainly expressed in GABAergic neurons of all brain regions and mediate the fast removal of GABA from the synaptic cleft. GAT-2 is expressed in neuronal and nonneuronal tissues and may regulate GABA levels in the cerebrospinal fluid. GAT-3 expression was restricted to the central nervous system where it was described predominantly in astrocytes and is most probably involved in the regulation of GABA at extrasynaptic sites. The betaine/GABA transporter BGT-1 was expressed in nonneuronal as well as neuronal tissues and its function in the CNS is poorly understood.
- Thoeringer CK, Ripke S, Unschuld PG, Lucae S, Ising M, Bettecken T, Uhr M, Keck ME, Mueller-Myhsok B, Holsboer F, Binder EB, Erhardt A. The GABA transporter 1 (SLC6A1): a novel candidate gene for anxiety disorders. J Neural Transm. 2009;116:649–57. doi: 10.1007/s00702-008-0075-y.PubMedCrossRefPubMedCentralGoogle Scholar