Serotonin (5-hydroxytryptamine, 5-HT), one of best known neurotransmitters, modulates neural activities and a wide range of neuropsychological processes [1]. The first step in 5-HT synthesis is the catalysis of tryptophan to 5-hydroxytryptophan (5-HTP) by tryptophan hydroxylase (TPH); this is the rate-limiting step in 5-HT synthesis. Two TPH enzymes are known; TPH1 is found in several tissues, while TPH2 is a brain-specific enzyme. The enzyme 5-HTP decarboxylase next converts 5-HTP to 5-HT [2]. In the brain, 5-HT is taken and stored from cytoplasm to synaptic vesicles by vesicular monoamine transporters. 5-HT is released into the synapse through a Ca²⁺-dependent mechanism, and its reuptake from the synapse is induced by the serotonin transporter (SERT). SERT is the principal site of action of many antidepressants (mainly selective serotonin reuptake inhibitors, SSRI; serotonin norepinephrine reuptake inhibitors, SNRI; tricyclic antidepressants, TCA) and represents a primary target of interest in antidepressant pharmacogenetics. Interestingly, alteration of tryptophan metabolism elicited by proinflammatory cytokines has recently gained attention as a new concept to explain the etiological and pathophysiological mechanisms of major depression [3-5].

5-HT plays an important role in the regulation of many pivotal functions, including emotion, mood, cognition, sleep, circadian rhythm, motor function, reproductive behaviors, thermoregulation, and endocrine functions, as well as in pathological states such as depression, anxiety, Alzheimer's diseases, schizophrenia, drug addiction, autism, and obesity [6-8]. Therefore, the components of the 5-HT system have developed as important therapeutic targets in the clinic. 5-HT mediates its diverse physiological responses through its receptors. Based on structural, biochemical, and pharmacological differences, 5-HT receptors (5-HTR) are classified into 7 distinct receptor families: 5-HT₁, 5-HT₂, 5-HT₃, 5-HT₄, 5-HT₅, 5-HT₆, and 5-HT₇ receptors. The complexity of the 5-HT system is further increased by alternative splicing and mRNA editing of several 5-HT receptors [9, 10]. With the exception of the 5-HT₃ receptors (consists of the 5-HT_3A, 5-HT_3B, 5-HT_3C, 5-HT_3D and 5-HT_3E receptors), which are ligand-gated ion channels, all 5-HT receptors are G-protein-coupled receptors (GPCR), transmitting their signals via G-proteins (Table 1) [11].

As a brief introduction to each 5-HT receptor (5-HTR), the 5-HT₁R and 5-HT₅R are negatively coupled to adenylate cyclase via Gi/o proteins, and they induce inhibition of cAMP formation. The 5-HT₁R family comprises the subtypes 5-HT_1A, 5-HT_1B, 5-HT_1D, 5-HT_1E, and 5-HT_1F receptors, which are known to play roles in the brain as well as the heart and gastrointestinal tract. The 5-HT₅R family consists of 2 subtypes: the 5-HT_5A and 5-HT_5B. Although the 5-HT_5AR has been found on neurons and neuronal-like cells of the carotid body, both receptors are limited in distribution to the CNS [12]. However, the 5-HT₅R family has not been extensively characterized pharmacologically. The 5-HT₂R couples to Gq/11 proteins and activates phospholipase C leading to the production of inositol-1,4,5-trisphosphate, intracellular Ca²⁺ release, and protein kinase C activation. The 5-HT₂R family currently comprises 3 receptor subtypes, 5-HT_2A, 5-HT_2B and 5-HT_2C, and has important roles in the brain, heart, gastrointestinal tract, platelets, and smooth muscle [11, 13-15]. The 5-HT₃R family members are non-selective cation ligand-gated ion channels. Their effects are excitatory, and they exist mainly in the peripheral nervous system, particularly on nociceptive afferent neurons and on autonomic and enteric neurons. 5-HT₃R antagonists (e.g. ondansetron and tropisetron) are used predominantly as anti-emetic drugs [16]. The 5-HT₄R, 5-HT₆R, and 5-HT₇R are positively coupled to adenylate cyclase via GS proteins and they elevate cAMP formation. The 5-HT₄R consists of multiple subtypes (5-HT_4A, 5-HT_4B, 5-HT_4C, 5-HT_4D, 5-HT_4E, 5-HT_4F, 5-HT_4G, 5-HT_4H, and 5-HT_4HB receptors), and the 5-HT₇R consists of 4 subtypes (5-HT_7A, 5-HT_7B, 5-HT_7C, and 5-HT_7D receptors) [9, 17]. The 5-HT₄R and 5-HT₇R play roles in the CNS as well as other systems. The 5-HT₆R is known to have important functions specifically in the CNS due to its exclusive distribution in the CNS [6, 11, 18].

The 5-HT₆R is one of the most recently discovered 5-HT receptors. The history of 5-HT₆R was started with the finding of a novel 5-HT receptor in the brain since a 5-HT-stimulated adenylate cyclase activity was detected in the striatum which did not fit with any of the known classes of 5-HT receptors. Furthermore, some neuroblastoma cells (NCB-20 and N18TG2) showed 5-HT-stimulated cAMP production that was sensitive to antipsychotics in a manner suggestive of a novel receptor [19, 20]. The rat 5-HT₆R was identified and sequenced by 2 groups in 1993 [21, 22], and in 1996, the human gene was cloned and shown to have 89% sequence homology with its rat equivalent [23]. The recombinant human 5-HT₆R is positively coupled to adenylate cyclase and has pharmacological properties similar to the rat receptor, exhibiting high affinity for several typical and atypical antipsychotics, including clozapine. The 5-HT₆R protein is a glycoprotein comprising 440 amino acids in humans and mice, and 438 amino acids in rats. All known 5-HT₆R homologues have 7 transmembrane domains that form 3 intracellular and 3 extracellular loops [24].

The gene for the human 5-HT₆R maps to the chromosome region 1p35-p36 and has an open reading frame of 1,320 bp [23]. The gene has 3 exons, which are separated by a 1.8-kb intron at position 714 and a second intron of 193 bp at position 873, corresponding to intracellular loop 3 and extracellular loop 3. In contrast to the complexity of the 5-HT receptor isoforms generated by alternative splicing and mRNA editing, as shown in Table 1, the 5-HT₆R has no known isoforms. A non-functional truncated splice variant of the 5-HT₆R has been identified, but it appears to have no physiological significance.

Within the 5-HT₆R gene, there is a silent polymorphism at bp position 267 within a tyrosine codon, where a cytidine is substituted for a thymidine (C267T variant). Based on a number of genetic linkage studies, the distribution of C and T alleles appears to be more or less equal among the general population. Although this polymorphism does not affect the identity of the tyrosine codon, it has been further analyzed for biased distribution in several human diseases. In this regard, several studies have investigated whether 5-HT₆ polymorphisms are associated with brain-related variables, such as neuropsychiatric disorders. Because several antipsychotic agents (notably clozapine) and antidepressants have high affinity for and are antagonists of 5-HT₆Rs, several genetic studies have examined the possible association of 5-HT₆ polymorphisms with schizophrenia and depression [25]. No association of the C267T polymorphism with schizophrenic was found in studies of Japanese and French patients. In addition, when pharmacogenetic studies were undertaken to assess the association between the C267T polymorphism and the response of schizophrenic patients to atypical antipsychotic drugs, no significant association was observed for patients taking clozapine. However, risperidone caused greater improvement in positive symptoms in patients carrying a thymidine substitution at the 267 position [26]. The association of the C267T polymorphism with patients suffering from depression has also been investigated, but failed to show a significant correlation [27]. A subsequent study reported that patients with major depressive disorder carrying the C267T polymorphism showed a better response to antidepressant medications [28]. Several additional genetic studies have been performed to investigate the association of this polymorphism and other 5-HT₆R polymorphisms on with bipolar disorder, Alzheimer's disease, and Parkinson's disease. However, the results of these studies have not been reproducible and their significance remains to be established [25].

The 5-HT₆R is expressed earlier in brain development than other 5-HT receptors. High levels of 5-HT₆R are first expressed on embryonic day 12 (E₁₂) in the rat brain, expression decreases slightly on E₁₇, and then remains stable through to adulthood. This expression pattern coincides with the emergence of serotonergic neuron, implying a role for 5-HT₆Rs early in the neuronal growth process involving the serotonergic system [29]. Rat and human 5-HT₆R mRNA is detected in the striatum, amygdala, nucleus accumbens, hippocampus, cortex and olfactory tubercle, but has not been found in peripheral organs. Using the highly specific radiolabelled 5-HT₆R antagonist [¹²⁵I] SB-258585, autoradiographic binding studies in the rat brain show high receptor levels in caudate-putamen, nucleus accumbens, striatum, and olfactory tubercles, and choroid plexus. Moderate recptor levels are seen in the hippocampus, thalamus, cerebral cortex, and frontal and parietal cortex [24, 30]. Similarly, immunohistochemical staining shows high receptor levels in nucleus accumbens, striatum, olfactory tubercles, cortex, hippocampus, and hypothalamus [31, 32]. In in situ hybridization and RT-PCR analyses [22, 31], 5-HT₆R levels exhibit a similar pattern in rats and humans. However, relatively little 5-HT₆R expression has been demonstrated in the mouse, and it is not clear why the mouse 5-HT₆R homolog does not exhibit the widespread brain expression seen in rats and humans. Indeed, many 5-HT₆R antagonists that induce enhanced cognition in rats have very little effect in mice, which may be due to the low expression in mice or to differences in ligand affinity across species [33]. Immunohistochemical staining for the 5-HT₆Rs has revealed that on neurons it is localized on dendrites, cell bodies, and postsynaptic sites, and is expressed in GABAergic, cholinergic, and glutamatergic neurons [24, 32].

Although there are several well-known non-selective 5-HT ligands that bind strongly to 5-HT₆Rs, such as lysergic acid diethylamide (LSD), for many years there were no selective 5-HT₆R agonists or antagonists available. Since the discovery of the human 5-HT₆R by Kohen et al. [23], an increasing number and diversity of selective and novel 5-HT₆R ligands have been developed using 5-HT₆R-specific high-throughput screening technologies [34, 35]. The synthesis of 5-HT₆R ligands, especially 5-HT₆R antagonists, has been very successful, with a number of highly potent ligands being reported.

Although a variety of highly selective 5-HT₆R ligands has been reported, the major efforts have focused on antagonism because of the positive effects of 5-HT₆R antagonists in several animal models, as discussed below. Before the discovery of such 5-HT₆R antagonists, 5-HT₆Rs were known to have high affinity for various atypical antipsychotic drugs and tricyclic antidepressants, but they displayed no clear selectivity [36]. Currently, more than 20 selective 5-HT₆R antagonists have been discovered. The most potent and selective 5-HT₆R antagonists are Ro 04-6790 (displays 100-fold selectivity for 5-HT₆R over other 5-HT receptors), Ro 63-0563 (100-fold selectivity), SB-271046 (50-fold selectivity), SB-258585 (100-fold selectivity), and SB-399885 (200-fold selectivity) [37-39]. Although Ro04-6790 and SB-271046 were the first identified and the most studied 5-HT₆R antagonists, respectively, they have limited capacity to cross the blood-brain barrier and appear to be orally active [37, 39]. Other 5-HT₆R antagonists such as SB-699929, SB-357134, and SB-399885 appear to have better pharmacokinetical and pharmacological profiles than SB-271046 and SB-258585 [40]. AVN-322, BVT-74316, PRX-07034, R-1485, SYN-114, SYN-120, and SUVN-502 are additional 5-HT₆R antagonists that are being developed for the treatment of cognitive disorders and are currently in phase I clinical trials [41]. Several 5-HT₆R antagonists including AVN-211, SAM-531, SB-742457, and SGS-518 have reached phase II clinical trials for cognitive disorders [41]. [¹¹C]-GSK215083 is a radiolabeled 5-HT₆R antagonist being developed as a PET radiotracer for the 5-HT₆R, and is in phase I trials [42].

Compared to the 5-HT₆R antagonists, considerably fewer compounds claim to be selective 5-HT₆R agonists. Examples are 2-ethyl-5-methoxy-N,N-dimethyltryptamine (EMDT), EMD386088, WAY-466, E-6801, LY586713, WAY-208466, WAY-181187, and R-13c [40]. EMD386088 displays 20-fold selectivity for the 5-HT₆R over other 5-HT-binding receptors, including the 5-HT transporter protein and dopamine receptors [43]. R13-c displays 50-fold selectivity over other 5-HT and dopamine receptors [44]. E-6801and E-6837 are potent partial agonists of the 5-HT₆R [45]. Thus, there are few 5-HT₆R agonists, and only WAY-181187 (displays 50-fold selectivity against serotonergic and other receptors) has been characterized and widely used [46, 47]. Recently, a new 5-HT₆R agonist, ST1936, has been reported and compared with the characteristics of WAY-181187 [48].

Taken together, the high affinity of the 5-HT₆R for atypical antipsychotic drugs and tricyclic antidepressants, and its abundant distribution in the brain (cortex, hippocampus, striatum, and hypothalamus) imply that the 5-HT₆R plays important roles in the CNS and in the etiology of neurological diseases. The 5-HT₆R shares a signaling mechanism with 5-HT₄R and 5-HT₇R in that they are the three 5-HT receptors positively coupled to Gs proteins, inducing cAMP production through stimulation of adenylate cyclase activity. However, since the 5-HT₆R is almost exclusively expressed in the brain compared with the expression patterns of the 5-HT₄R and 5-HT₇R, recently developed selective 5-HT₆R ligands may represent attractive new therapeutic options for several types of diseases.

As a member of the Gs-GPCR family, it is well known that engaengagement of the 5-HT₆R activates cAMP signaling pathways through adenylate cyclase stimulation. In addition, Svenningsson et al. [64] reported that the activation of 5-HT₆Rs increases phosphorylation of dopamine- and cAMP-regulated phosphoprotein of molecular weight 32,000 (DARPP-32) by protein kinase A. However, there are still insufficient studies on the mechanisms of 5-HT₆R-mediated signal transductions to understand the receptor's various roles in physiological and pathological states in the CNS, including Alzheimer's disease, depression, cognition, and obesity. This is mainly due to the lack of pharmacological tools able to selectively activate 5-HT₆Rs in the CNS. The selective 5-HT₆R agonists have only recently been developed and characterized [47, 48, 65]. This lack of appropriate tools has also contributed to the inconsistent observations on the pharmacological and neurochemical effects of 5-HT₆R antagonists.

To identify the mechanism of 5-HT₆R function and its cellular mechanisms in the CNS, we employed a yeast two-hybrid screening system on a human brain cDNA library, with the 5-HT₆R intracellular loop 2 (iL2), intracellular loop 3 (iL3), and the carboxyl terminus as bait (Fig. 1). We first reported that Fyn, a member of the Src family of non-receptor protein-tyrosine kinases, is bound to the carboxyl terminus of the 5-HT₆R [32]. The expression of Fyn increases 5-HT₆R activity by increasing receptor surface expression without changing total cellular expression of 5-HT₆R protein. Reciprocally, the activation of 5-HT₆Rs also increases Fyn phosphorylation at Tyr-420. Phosphorylation at Tyr-420 was blocked when the 5-HT₆R-Fyn interaction was blocked by overexpression of the Fyn SH3 domain, the best-characterized domain for Fyn-mediated protein-protein interactions. We further demonstrated that the activation of 5-HT₆Rs activated extracellular signal regulated kinase1/2 (ERK1/2) activity through Fyn- and PKA-dependent pathway. We recently showed that 2 selective 5-HT₆R agonists, ST1936 and WAY-181187, also increased Fyn phosphorylation [48]. Because Fyn is known to be involved in Alzheimer's disease through modulation of the microtubule-associated tau and amyloid-β proteins [66, 67], our observations may provide a cellular mechanism for 5-HT₆R-mediated cognition and mood changes in the brain. It is also interesting to note that the 5-HT₆R agonist, LY-586713, increases expression of cortical and hippocampal BDNF which could mediate its pro-cognitive effect; however, the cortical increase in BDNF was not antagonized by SB-271046 [50], suggesting the increased BDNF expression is mediated through a different mechanism.

Our group has also characterized a second 5-HT₆R-interacting protein: Jun activation domain-binding protein-1 (Jab1, Fig. 1). We recently discovered a novel interaction between human 5-HT₆R and Jab1, and we observed Jab1-mediated modulation of the membrane expression and activity of 5-HT₆Rs [68]. In addition, we found that 5-HT₆Rs affect the cytosolic and nuclear distribution of Jab1 as well as the interaction between Jab1 and c-Jun, a target protein downstream of Jab1. Furthermore, we demonstrated that 5-HT₆Rs and Jab1 play important roles under conditions of in vitro hypoxia and in vivo cerebral ischemia. A recent study has suggested that Jab1 is involved in the onset of neuronal diseases such as Alzheimer's disease and Parkinson's disease through interaction with the endoplasmic reticulum stress transducer IRE1 [69]. Therefore, these data provide new insights into the physiological roles of 5-HT₆Rs and Jab1 in the CNS at both the molecular and cellular levels. In addition to Fyn and Jab1 binding proteins, we are currently investigating 2 other proteins as candidate 5-HT₆R-binding proteins.

To explore the neurochemical mechanisms involved in 5-HT₆R functions, several microdialysis studies have been performed using selective 5-HT₆R ligands. In studies using in vivo microdialysis, increased acetylcholine levels were observed in the rat medial prefrontal cortex after acute administration of the 5-HT₆R antagonist SB-399885 [70] but not in the hippocampus after administration of another antagonist, Ro04-6790 [71]. Another selective 5-HT₆R antagonist, SB-357134, has been reported to increase high KCl-stimulated acetylcholine release in vitro in rat cortical and striatal slices [72]. However, no in vitro or in vivo microdialysis studies have yet been reported on acetylcholine release using selective 5-HT₆R agonists. Glutamate is a major excitatory neurotransmitter in the CNS. 5-HT₆R antagonists have been shown to increase the extracellular concentration of glutamate both in vivo by SB-271046 in the frontal cortex and hippocampus [73], and in vitro by SB-357134 in the cortex and striatum [72]. Interestingly, it was reported that the selective 5-HT₆R agonist WAY-181187 attenuated the stimulated glutamate levels elicited by sodium azide and high KCl in vitro but not in vivo [47]. A recent electrophysiological study using whole-cell patch-clamp recording showed that 5-HT₆R activation by ST1936 inhibits corticostriatal glutamatergic transmission, which was mimicked by a different agonist, WAY-181187 [65]. This finding is consistent with the reported in vitro microdialysis data using WAY-181187 [47].

Woolley et al. [24] reported on the immunohistochemical colocalization of 5-HT₆R with GABAergic neurons in many areas of the cortex, basal ganglia, hippocampus, thalamus, and cerebellum. These co-localization data together with the microdialysis data suggest that 5-HT₆R agonists/antagonists may modulate cholinergic and/or glutamatergic systems via disinhibition of GABAergic neurons. If 5-HT₆Rs do modulate cholinergic and/or glutamatergic systems in this manner, then 5-HT₆R antagonists should decrease GABA release. In fact, no study has yet reported on the modulation of GABA release by 5-HT₆R antagonists using microdialysis or electrophysiological methods. On the other hand, the 5-HT₆R agonist WAY-181187 has been shown to significantly increase extracellular GABA concentrations in the hippocampus, striatum, and amygdala, but had no effect on GABA levels in the nucleus accumbens or thalamus [47]. Another 5-HT₆R agonist, WAY-208466, also preferentially elevated cortical GABA levels following acute and chronic administration, indicating that neurochemical tolerance does not develop following repeated 5-HT₆R stimulation. These in vivo data were also confirmed by in vitro electrophysiological investigations. WAY-181187 increased the frequency of spontaneous inhibitory postsynaptic currents (sIPSC) recorded from hippocampal CA1 neurons [74]. This effect was blocked by the 5-HT₆R antagonist SB-399885, which confirmed it to be mediated by activation of 5-HT₆Rs. Collectively, the results of these microdialysis and electrophysiological experiments suggest that 5-HT₆R agonists/antagonists may modulate cholinergic, glutamatergic, and/or GABAergic systems.

The 5-HT₆R has gained increasing attention over the past decade, and has become a promising target for the treatment of CNS diseases. Currently, consistent effects have been demonstrated with 5-HT₆R antagonists in preclinical models of cognition and 5-HT₆Rs have obvious pharmaceutical potential. Although the majority of 5-HT₆R research has focused on their pro-cognitive effects, the role of these receptors in depression and anxiety has also been postulated. However, the preclinical results are equivocal since blockade and stimulation of 5-HT₆Rs may evoke pro-cognitive, antidepressant-like, or anti-anxiety-like effects. The explanation for these paradoxical effects remains unclear. The function of the 5-HT₆R has been revealed to be much more complex than initially defined. Based on the existing data, and depending on the drug used, different cellular pathways may be activated. However, the full characterization of the functional profile of 5-HT₆Rs is still pending. The drug discovery process may benefit considerably from this complexity, in terms of the quantity and quality of potential new therapeutic molecules. Thus, the functions of 5-HT₆Rs must be studies at both molecular and cellular levels in order to understand their roles in the CNS and to develop novel drug targets for neurological diseases.

Table 1. The classification and their signal pathways of 5-HT receptor subtypes