Characterization of the Substrate-Binding Site in GABA Transporters

γ-Aminobutyric acid (GABA) neurotransmission is characterized by the restricted expression of the GABA-synthesizing enzyme glutamate decarboxylase (GAD), the GABA receptors, and the GABA transporters in GABAergic synapses consisting of a presynaptic nerve ending, a postsynaptic entity, and surrounding astrocytes (1). Additionally, the presynaptic nerve ending is characterized by the presence of the vesicles expressing the vesicular GABA transporter (2,3). GABA catabolism, on the other hand, is not restricted to GABAergic synapses because the primary metabolic enzyme GABA transaminase is widely distributed not only in the central nervous system (CNS), but also in many other tissues, including the liver (4,5). Among the different proteins involved in these processes, the receptors, transporters, and the GABA transaminase are all capable of recognizing, and binding the GABA molecule, each with a unique affinity and stringency regarding specificity (6–8). Considering the high degree of flexibility of the GABA molecule, it is not surprising that GABA may be recognized by these different proteins in distinctly different conformations. Thus, the receptors (GABAA) and the transporters have been known for several years to bind GABA in a more extended and somewhat folded conformation, respectively (see KrogsgaardLarsen et al. [9]). Actually, in agreement with this, the GABA analogs of restricted conformation guvacine, nipecotic acid, THPO, and isoguvacine, isonipecotic acid, and THIP (Fig. 1) can be divided in two groups, each reflecting the GABA molecule in a folded and elongated conformation. The three folded analogs are specific ligands for GABA-transporters, whereas the three latter analogs specifically bind to the GABAA receptors with no affinity for the transporters. Hence, these GABA analogs have served as important lead structures in the design of new analogs interacting with either one or the other of the macromolecules constituting important functional components of the GABA synapses (9). The present review is aimed at a further characterization of the molecular structures, which are recognized by the cloned GABA transporters as well as neurons and astrocytes expressing these transporters. For a number of years, it has been a challenge to explain on a rational pharmacological basis how these two cell types can exhibit completely different pharmacological properties regarding inhibitors of GABA transporters, considering the fact that at least the most abundant transporters are expressed in each of these cell types (7,10). This aspect will be discussed extensively in this chapter.