Upon neuronal stimulation, neurotransmitters are released into the synaptic cleft where they activate both pre- and post-synaptic receptors for propagation of the electrical nerve impulse.
Neurotransmission across a synapse is under tight spatio-temporal regulation by high-affinity membrane transport proteins of the neurotransmitter:sodium symporter (NSS)-family. NSS proteins are localized on the pre-synaptic nerve terminals. They mediate rapid retrieval of the released neurotransmitters from the synaptic cleft. The first insight into the three-dimensional structure of this class of proteins was obtained by crystallization of the prokaryotic NSS member, LeuTAa, isolated from the thermophilic bacterium, Aquifex aeolicus. The first structure revealed a conformation likely representing an intermediate (i.e., the occluded state) between the 'outward-' and 'inward-' facing conformations of LeuTAa. Subsequent crystal structures of LeuTAa bound to tricyclic antidepressants (TCAs) viz., clomipramine, desipramine, imipramine and amitriptyline afforded mehanistic insights into the inhibition of NSS proteins. A major limitaion was noted: the inhibition of LeuTAa by TCAs is non-competitive, but TCAs are competitive inhibitors of the human serotonin transporter (SERT). Hence there is an obvious discrepancy, which questions the validity of the LeuTAa-based model. My thesis is based on the working hypothesis that the discrepancy can be resolved by assuming that TCA's bind in the outer vestibule and that their side chain reaches into the substrate binding site. Because serotonin is larger, this precludes simultaneous binding of serotonin and TCA's. I adressed this discrepancy via two approaches:
(i) competition binding experiments with carbamazepine (i.e., an imipramine analog with a short aliphatic side chain, which was verifed to compete with [3H]imipramine binding to SERT), (ii) site-directed mutagenesis where I generated a diagnostic SERTY95F mutation which greatly reduced the affinity for [3H]imipramine but did not affect substrate binding. Dixon plots revealed that carbamazepine bound simultaneously to SERT in the presence of serotonin, paroxetine or ibogaine. In contrast, the binding of amphetamines, viz.
para-chloroamphetamine (PCA) or methylene-dioxy-methamphetamine (MDMA or, 'ecstasy') and of carbamazepine was mutually exclusive. My observations are consistent with the following model: (i) the tricyclic ring of TCAs resides in the outer vestibule of SERT and the dimethyl-aminopropyl side chain occupies the substrate binding pocket; (ii) binding of amphetamines to SERT creates a structural change in the inner and outer vestibules which precludes simultaneous occupancy of the tricyclic ring in the vestibules, (iii) simultaneous binding of ibogaine (which binds to the 'inward-facing' conformation) and of carbamazepine to SERT is indicative of a second low-affinity binding site in the inner vestibule, consistent with the pseudo-symmetric fold of NSS proteins.