The first evidence of an active transport of serotonin (5-hydroxytryptamine, 5-HT) came from pioneer studies in the late 1950s showing an uptake of 5-HT into the blood platelets from rabbit or guinea pig. A few years later, a similar mechanism was observed in the rat brain (Schanberg 1963). The mechanism of this transport, involving the sodium and potassium ionic gradients, was rapidly elucidated, and the existence of a dedicated transporter protein postulated. In addition, the antidepressant imipramine was shown to bind to this putative transporter to reduce 5-HT uptake into the blood platelet, and, therefore, the serotonin transporter was named the “serotonin transporter-imipramine receptor.” Using the binding properties of imipramine to the serotonin transporter, Jane Talvenheimo and Gary Rudnick have purified the transporter protein from porcine blood platelet preparations (Talvenheimo and Rudnick 1980). Following that, studies have shown that most of the antidepressant drugs bind to the serotonin transporter and inhibit 5-HT uptake, further supporting the 5-HT hypothesis of depression. Since the early 1990s, the gene encoding the serotonin transporter (5-HTT) was identified in rat, human, drosophila, mouse, guinea pig, and cow.
The 5-HTT gene is located on chromosome 10 in rat, chromosome 11 in mouse, chromosome 16 in monkey, and chromosome 17 in humans. It counts 43–54 kbp and contains 14–16 exons. Since only 5-HT neurons express the 5-HT transporter, in situ hybridization experiments have located 5-HTT transcripts in the raphe nuclei. However, a transitory expression of functional 5-HTT has also been observed in non-5-HT neurons in various brain regions, notably in sensory thalamocortical neurons (for review see Gaspar et al. 2003). In adults, 5-HTT expression is restricted to 5-HT neurons, with immunolabeling, radioligand binding, and 5-HT reuptake experiments positioning 5-HTT along 5-HTergic axons that project throughout all brain regions. In the shell of the nucleus accumbens, however, a subset of 5-HT neurons do not express 5-HTT, which confers more resistance to the neurotoxic effects of amphetamines and amphetamine-like derivatives (Brown and Molliver 2000; see “Molecules Targeting the Serotonin Transporter” section below). Ultrastructural studies using electron microscopy have further revealed that the 5-HTT is located on the plasma membrane in both synaptic and non-synaptic sites, suggesting that 5-HT diffuses from sites of release into the synaptic cleft to reach more distant receptors in a paracrine mode of transmission.
The primary binding site for antidepressants is located between TM 1, 3, 6, 8, and 10 which directly block 5-HT binding and subsequent translocation. Site-directed mutagenesis combined to crystallography have identified the residues involved in the binding/translocation of 5-HT throughout the transporter. Interaction of these residues allows the transporter to switch from outward-open conformation that exposes the cone-shaped extracellular vestibule to aqueous solution, providing a pathway for substrates, inhibitors, and ions to reach the central binding site, approximately halfway across the membrane bilayer, to an inward-open conformation, where the extracellular gate closes and the intracellular gate opens to allow for the translocation of 5-HT to the cytoplasm. Interaction of 5-HT with the residues A330, A331, Q332, I108, and R104 in the extracellular vestibule allows for its translocation to the substrate binding site formed by residues F335, D98, and Y176. The closing of the extracellular gate resulting from the interaction of 5-HT with F341 and Y176 occludes the transporter in a conformation that restricts access from both the extracellular and intracellular environments. Subsequent spatial reorganization of the TMs results in movements of TM1b and 6b, which open the transporter in the inward-facing conformation to allow for intracellular transport of 5-HT.
Studies have shown that the 5-HTT plays an important role in the maintenance and the fine regulation of 5-HT homeostasis. Indeed, profound alterations of both intra- and extracellular 5-HT levels have been observed following genetic ablation of the 5-HTT in mice and rats. For instance, 5-HTT−/− mice exhibit a decreased rate of synaptic 5-HT clearance and a fourfold to sixfold increase in basal levels of forebrain extracellular 5-HT, as well as a 60–80% decreased 5-HT tissue content in various brain regions (for review see Torres et al. 2003). As a consequence, 5-HTT mutant mice exhibit hypolocomotion, increased body weight, increased anxiety, reduced aggression, serotonin syndrome-like behaviors, impaired stress response, abnormal development of the thalamocortical and retinal axons, altered control of breathing, and thermoregulation. Although the 5-HTT is the primary target of the most prescribed antidepressants (selective serotonin reuptake inhibitors, SSRIs), the effect of 5-HTT ablation on depression-like behaviors has led to conflicting results, showing either an absence of effect in C57Bl6 background mice or an unsuspected increase in depressive-like behaviors in 129S6 and CD1 background mice (Krishnan and Nestler 2011). Later, the generation of 5-HTT knockout rats has also demonstrated an increased depressive-like behavior resulting from 5-HTT ablation (Krishnan and Nestler 2011).
Molecules Targeting the Serotonin Transporter
In addition, 5-HTT is also the target of amphetamine-derivative anorexigens such as fenfluramine and dexfenfluramine, and psychostimulants such as MDMA (ecstasy). These molecules, called “releasers,” are substrates of the 5-HTT and are transported intracellularly to the vesicles via the vesicular monoamine transporter (VMAT). Consequently, the influx of “releasers” chases the 5-HT out the vesicles and extracellularly through the 5-HTT (Fig. 3c). The structural changes and molecular mechanisms allowing the 5-HTT for reversed 5-HT transport in a channel-like manner are still not understood.
Regulation of 5-HTT Function
Syntaxin 1A is a membrane protein of the SNARE family that interacts with multiple partners involved in neurotransmitter exocytosis. The 5-HTT interacts via its N-terminus (amino acids 11–30) with the syntaxin 1A, and this interaction promotes the addressing of the 5-HTT to the cell surface (Haase et al. 2001; Quick 2003). In addition, interaction with syntaxin 1A is critical for the electroneutrality of 5-HT reuptake, i.e., with a Na+/Cl−/5-HT stoichiometry of 1:1:1. In line with this, studies in Xenopus oocytes co-expressing the 5-HTT and syntaxin 1A have shown that inhibition of 5-HTT/syntaxin 1A interaction causes a substrate-independent entry of Na+, with subsequent alteration of the stoichiometry of 5-HT transport, which becomes electrogenic (Na+/Cl−/5-HT = 7:1:1). This 5-HTT/syntaxin 1A interaction is therefore finely regulated by the calmodulin-dependent protein kinase type 2 (CaMKII). The protein SCAMP2 (secretory carrier-associated membrane protein 2) is involved in the addressing of proteins to the plasma membrane after maturation in the Golgi apparatus. The 5-HTT interacts via its N-terminus with SCAMP2. However, the co-expression of the 5-HTT and SCAMP2 in vitro causes the redistribution of 5-HTT from the plasma membrane to intracellular compartments, reducing 5-HT reuptake activity. Integrin αIIbβ3 is an adhesion protein specific of platelets and is essential for platelet aggregation. In blood platelets, the interaction of 5-HTT C-terminus with integrin αIIbβ3 upregulates platelet aggregation. The activation of integrin αIIbβ3 promotes 5-HTT reuptake activity by increasing its expression at the plasma membrane via a mechanism involving the protein kinase P38-MAPK. The protein Hic-5 is involved in the regulation of the actin cytoskeleton. In vivo, in blood platelets and the CNS, an interaction of Hic-5 with the C-terminal extremity of 5-HTT is observed when the latter is sequestered in lipid rafts. In the presence of 5-HT, interaction with Hic-5 disappears in favor of a redistribution of 5-HTT to the plasma membrane, which increases 5-HT reuptake via a mechanism negatively regulated by the protein kinase C (PKC). The protein PICK1 is also a scaffold protein that interacts with the PDZ domain (postsynaptic density 95, DLG1, ZO1) of numerous membrane proteins. The role of the interaction of the C-terminal domain of 5-HTT with PICK1 has not been elucidated yet. The neuronal enzyme that synthesizes nitrogen monoxide (neural nitric oxide synthase, nNOS) interacts with its PDZ domain to the terminal valine (NAV) motif of 5-HTT C-terminus and decreases its reuptake activity, by reducing its expression at the plasma membrane. Mutagenesis of the NAV into glutamate (NAE) prevents this interaction. By contrast, 5-HT reuptake by the 5-HTT induces the activation of the nNOS bound at its C-terminus and the production of nitric oxide (NO), by a calmodulin-dependent mechanism.
Thus, there is a mutual regulation between the 5-HTT and nNOS: tonic inhibition by their physical interaction and phasic inhibition by NO, probably due to posttranslational modifications such as a protein kinase G (PKG)-dependent phosphorylation. The α-synuclein is a protein associated with the synaptic vesicles and stabilizes the SNARE complex during exocytosis processes. The co-expression of 5-HTT and α-synuclein in heterologous cell lines reduces the expression of 5-HTT at the cell surface and, consequently, reduces 5-HT reuptake. Myosin II is a scaffold protein of PKG which interacts with 5-HTT when it is glycosylated on asparagines 208 and 217. A mutation of these asparagines into glutamines alters 5-HTT homodimerization and its interaction with myosin II, leading to a reduced 5-HTT transport activity and an impaired response to PKG activators. The MacMARCKS protein, a homologous of the MARCKS (myristoylated alanine-rich C-kinase substrate) family, is a substrate of PKC bound to actin filaments and is involved in exocytosis. The interaction of 5-HTT C-terminus with the MacMARKS protein was shown to cause a reduction in 5-HT reuptake activity (probably due to its internalization) and alterations in the internalization induced by PKC activation (Jess et al. 2002). The SEC24 protein is a component of the coat protein complex II (COPII) which promotes the formation of transport vesicles from the endoplasmic reticulum (ER). Interaction of the 5-HTT C-terminus tail has been shown to be necessary for the transport of the 5-HTT out of the ER and its insertion into the plasma membrane.
The protein phosphatases also play an important role in the tonic regulation of the transporters, maintaining them in a non-phosphorylated state. The catalytic subunit of the protein phosphatase PP2A was shown to physically interact with 5-HTT (Samuvel et al. 2005). Inhibition of PP2A results in increased 5-HTT phosphorylation which causes a decrease in 5-HTT reuptake capacity (Ramamoorthy et al. 1998).
A polymorphism has been identified in the promoter region of the 5-HTT (5-HTTLPR). This polymorphism consists of different lengths of the repetitive sequence containing GC-rich, 20–23-bp-long repeat elements in the upstream regulatory region of the 5-HTT gene. A deletion or insertion in the 5-HTTLPR creates a short (S) or a long (L) allele (14- and 16-repeat alleles), respectively, which alters the transcriptional activity of the promoter. The S variant has been reported to be associated with lower basal and transcriptional efficiency of the 5-HTT, which results in lower 5-HT uptake activity as compared to the L variant. The SS variant has been positively associated with a reduced response to SSRIs and a reduced remission after SSRI treatments in depressed patients. Furthermore, this 5-HTTLPR polymorphism has been studied in various psychiatric disorders such as mood disorder, schizophrenia, panic disorder, autistic disorders, and personality traits resulting in positive or negative associations and sometimes conflicting results (Murphy and Moya 2011). In addition, three different alleles, namely, STin2.9, STin2.10, and STin 2.12 have been identified in the second intron of the 5-HTT (STin2). This variable number of tandem repeat (VNTR) polymorphism enhances 5-HTT expression, proportionally to the number of repeat copies of the 16/17 base pair element (12>10>9). It was shown that these different alleles respond differentially to the transcription factors CTCF and YB-1 and correlates with affective disorders (Murphy and Moya 2011). Furthermore, missense mutations such as 5-HTT I425V and G56A were found to produce a gain-of-function as measured by increased [3H] 5-HT uptake. In the case of I425V, an altered regulation of 5-HTT by nitric oxide synthase precursors and cyclic GMP and P38 mitogen-activated protein kinase was also observed. Finally, only a few studies have focused on 5-HTT 3′-UTR variants. 5-HTT 3′-UTR variants play important roles in mRNA translation, localization, and stability. Thus, mutations in the 3′-UTR can affect the termination codon, polyadenylation (polyA) signals, the ratio of multiple polyA signal usage, as well as the secondary structure of the 3′-UTR mRNA. The 5-HTT 3′-UTR contains two polyA sites, located at 567 bp and 690 bp downstream of the stop codon. The more distal of the polyA signals contains a common SNP (rs3813034) that alters the balance of the two polyA forms of SERT such that the T>G allele of rs3813034 leads to an increase of the distal polyA signal. This SNP has been associated with impaired retention of fear extinction memory and increased anxiety/depression. Another SNP (rs1042173), associated with alcohol craving and drinking intensity, was shown to have functional consequences as it increases mRNA and protein levels.
The serotonin transporter 5-HTT or SERT is the main actor of the regulation of serotonin homeostasis both in the periphery (blood platelets) and in the central nervous system. In the central nervous system, the 5-HTT is located on 5-HT neurons at synaptic and extrasynaptic sites. The 5-HTT reuptakes serotonin (5-HT) into 5-HT neurons to terminate its action on 5-HT receptors. Alteration in 5-HTT function has been linked to several neuropsychiatric disorders such as depression, anxiety disorders, and drug abuse. Indeed, the 5-HTT is the target of the most prescribed antidepressants, as well as some amphetamine derivatives. Therefore, the expression and function of the 5-HTT are finely regulated by different mechanisms, including protein interactions and phosphorylation/dephosphorylation processes. Furthermore, several genetic variants of the 5-HTT that lead to functional alterations in 5-HT reuptake have been associated with traits of depression or anxiety, reduced response to antidepressant, and reduced remission following antidepressant treatment.
- Jess U, El Far O, Kirsch J, Betz H. Interaction of the C-terminal region of the rat serotonin transporter with MacMARCKS modulates 5-HT uptake regulation by protein kinase C. Biochem Biophys Res Commun. 2002;294(2):272–9.Google Scholar