Fine-tuning Serotonin_2C Receptor Function in the Brain: Molecular and Functional Implications

Abstract

The serotonin_2C receptor (5-HT_2CR) is a member of the serotonin₂ family of 7-transmembrane-spanning (7-TMS) receptors, which possesses unique molecular and pharmacological properties such as constitutive activity and RNA-editing. The 5-HT_2CR is widely expressed within the central nervous system, where is thought to play a major role in the regulation of neuronal network excitability. In keeping with its ability to modulate dopamine (DA) neuron function in the brain, the 5-HT_2CR is currently considered as a major target for improved treatments of neuropsychiatric disorders related to DA neuron dysfunction, such as depression, schizophrenia, Parkinson’s disease or drug addiction.

The aim of this review is to provide an update of the functional status of the central 5-HT_2CR, covering molecular, cellular, anatomical, biochemical and behavioral aspects to highlight its distinctive regulatory properties, the emerging functional significance of constitutive activity and RNA-editing in vivo, and the therapeutic potential of inverse agonism.

1. Introduction

Serotonin_2C receptors (5-HT_2CRs) are members of the 7-transmembrane spanning (7-TMS or heptahelical) receptor superfamily, frequently referred to as G protein coupled receptors (GPCRs). The 5-HT_2CRs couple to multiple cellular signaling pathways and are involved in the regulation of a variety of physiological functions and behaviors. Increasingly, 5-HT_2CRs are therapeutic targets for conditions such as schizophrenia, anxiety, depression, Parkinson’s disease, drug addiction and obesity. The 5-HT_2CR is the only 7-TMS receptor whose mRNA undergoes adenosine-inosine editing events which change the coding for amino acids located within the putative second intracellular domain (i2) of the receptor. In in vitro studies, RNA editing has a dramatic impact on the signaling characteristics of the 5-HT_2CR. In vivo, RNA editing efficiency differs in different brain regions and in response to various drugs. Further, it has been suggested that alterations in RNA editing of the 5-HT_2CR may be involved in the etiology of different disease states such as schizophrenia and affective disorders.

In this review, we discuss some of the various mechanisms that regulate the function of the 5-HT_2CR and the implications for this regulation on various physiological functions and behaviors controlled by this receptor.

2. Molecular pharmacological aspects in vitro

Like most, if not all, 7-TMS receptors, 5-HT_2CRs couple to multiple cellular effector systems. Perhaps the best studied effector coupled to 5-HT_2CRs is the phospholipase C (PLC) pathway. Albeit somewhat less well studied, other major effectors that are coupled directly to 5-HT_2CRs are the phospholipase A2 (PLA₂) signaling cascade, the phospholipase D (PLD) pathways and extracellular signal-regulated kinase (ERK) (for recent reviews see Leysen, 2004; Raymond et al., 2006; Werry et al., 2006). 5-HT_2CRs couple to PLC via Gq/11 proteins (Chang et al., 2000) and can couple to PLD via Gα13 proteins (McGrew et al., 2002), however, the signaling mediator for PLA₂ activation is unknown. In addition to phospholipid signaling pathways, 5-HT_2CRs also activate desensitization mechanisms, such as G protein coupled receptor kinase (GRK) (Berg et al., 2001b) and arrestin (Marion et al., 2004). The receptor is also known to couple to pertussis toxin-sensitive G proteins (e.g. Gαi/o; Alberts et al., 1999; Cussac et al., 2002; Lucaites et al., 1996) as well as to PDZ domain containing proteins (Backstrom et al., 2000; Becamel et al., 2004). Consequently, the net cellular effect of activation of 5-HT_2CRs is a coalescence brought about by the concurrent activation of several effector pathways within cells.

The second intracellular (i2) domain of 7-TMS receptors plays an important role in receptor function. The highly conserved E/DRY motif is located in i2 at the cytosolic end of transmembrane helix 3 and has been linked strongly to mechanisms of receptor activation and G protein coupling (Flanagan, 2005). Many studies have provided evidence that E/DRY motif and other residues within i2 are involved with direct coupling to G proteins (Burstein et al., 1998; Moro et al., 1993; Sugimoto et al., 2004) or other signaling molecules (Laghmani et al., 2005). In addition, i2 may participate in desensitization mechanisms such as ß-arrestin binding, receptor internalization and downregulation (Marion et al., 2006). Evidence also suggests that residues within i2 regulate the capacity of receptors to isomerize thereby controlling the formation of active receptor conformations and constitutive receptor activity (Alewijnse et al., 2000; Burstein et al., 1998; Flanagan, 2005; Rasmussen et al., 1999; Scheer et al., 2000).

Editing of the 5-HT_2CR mRNA leads to changes in amino acids in i2 starting just 2 residues downstream from the E/DRY motif. mRNA transcripts of the human 5-HT_2CR undergo adenosine-to-inosine editing events at five sites which encompass amino acids 156-160 within the putative second intracellular loop of the encoded human receptor. In human brain, the non-edited receptor contains the amino acids isoleucine, asparagine, and isoleucine (i.e., INI) at positions 156, 158 and 160, respectively. Consequences of RNA editing can produce potentially 24 different isoforms of the human receptor; although RNA actually encoding for about 14 different receptor isoforms has been detected (Fitzgerald et al., 1999). In rat, adenosine-to-inosine editing events occur with the potential for generating 11 different mRNA transcripts predicted to encode 7 different receptor isoforms (Burns et al., 1997).

Importantly, differences in function of some edited receptor isoforms have been reported (for review see Werry et al., 2008). Pharmacological characterization of two fully edited isoforms (5-HT_2C-VGV and 5-HT_2C-VSV) has revealed decreases in agonist affinity (Fitzgerald et al., 1999; Herrick-Davis et al., 1999; Niswender et al., 1999; Quirk et al., 2001), potency (Berg et al., 2001a; Burns et al., 1997; Fitzgerald et al., 1999; Herrick-Davis et al., 1999; McGrew et al., 2004; Price et al., 2001; Wang et al., 2000), ligand-independent (constitutive) receptor activity (Berg et al., 2008; Herrick-Davis et al., 1999; Niswender et al., 1999; Wang et al., 2000) and receptor-arrestin binding and internalization (Marion et al., 2004) as compared with the non-edited receptor (5-HT_2C-INI).

In addition to selectivity for receptor subtypes, agonists have selectivity for different signaling pathways coupled to a single receptor subtype; a process known as ‘functional selectivity’ (Berg and Clarke, 2006; Urban et al., 2007). It has been well-established that 5-HT_2C ligands can selectively regulate signaling pathways at the non-edited 5-HT_2C-INI isoforms (Berg et al., 1998, 2001a; Moya et al., 2007; Werry et al., 2005). We have found that RNA editing results in a significant change in agonist functional selectivity for PLC versus PLA₂ signaling via 5-HT_2C receptors (Berg et al., 2001a; 2008). For the fully-edited isoforms, 5-HT_2C-VSV and 5-HT_2C-VGV, functional selectivity is lost (Berg et al., 2001a). However, at the partially edited receptor, 5-HT_2C-VNI, where a single amino acid substitution of valine for isoleucine occurs at position 156 (I156V), the agonist functional selectivity profile is substantially altered but not abolished (Berg et al., 2008). Interestingly, agonists with greater relative efficacy for PLA₂ vs. PLC at the non-edited receptor lost preferential efficacy toward PLA₂ in cells expressing the 5-HT_2C-VNI receptor (i.e., agonist relative efficacy for PLC and PLA₂ were not different), whereas, agonists that were non-selective for PLC vs. PLA₂ at the 5-HT_2C-INI receptor became selective for PLC at the 5-HT_2C-VNI isoform, due to reduced relative efficacy for PLA₂. However, agonists with preferential activity toward PLC retained their PLC signaling preference. Therefore, agonist functional selectivity toward the PLA₂ signaling cascade appears to be sensitive to effects of RNA editing. Overall, these data suggest that the i2 domain of the 5-HT_2CR plays a major role in providing agonist-specific information to the signal transduction machinery of the cell.

It has been well documented that 5-HT_2CRs exhibit a great deal of ligand-independent activity toward PLC. RNA editing appears to reduce the ability of 5-HT_2CRs to signal constitutively to PLC (Berg et al., 2008; Herrick-Davis et al., 1999; Niswender et al., 1999; Wang et al., 2000). Reduction in constitutive receptor activity could arise from reduced capacity of the edited receptors to isomerize to an active conformation(s) capable of coupling to the Gq-PLC-IP pathway or to reduced G protein coupling of the active receptor conformation(s). Several reports suggest a role for the E/DRY motif (especially the arginine residue) in i2 of 7-TMS receptors in regulating the capacity of receptors to isomerize between inactive and active conformations (Alewijnse et al., 2000; Burstein et al., 1998; Flanagan, 2005; Rasmussen et al., 1999; Scheer et al., 2000). However, amino acids in the i2 region of 7-TMS receptors close to the conserved DRY sequence are also known to be involved in G protein coupling (Burstein et al., 1998; Moro et al., 1993; Sugimoto et al., 2004). Given the proximity of the amino acids altered by RNA editing to the E/DRY region, either mechanism could be responsible for the reduced constitutive activity of the edited 5-HT_2CRs toward PLC.

To address the mechanism responsible for reduced constitutive activity of 5-HT_2CR edited isoforms, we compared the Gq-PLC-IP coupling efficiency of the RNA edited isoforms 5-HT_2C-VNI , 5-HT_2C-VSV, and 5-HT_2C-VGV occupied by either the full agonist, 5-HT, or the partial agonists, LSD and DOI, to that of the non-edited 5-HT_2C-INI receptor over a range of receptor densities. The slope of the regression line of the receptor density- response curve is a good measure of the receptor-G protein coupling efficiency (Kang and Leeb-Lundberg, 2002). We found that the slopes of the regression line between receptor density and response to maximal occupancy by agonist (full or partial) for RNA-edited receptors (5-HT_2C-VSV, 5-HT_2C-VGV, 5-HT_2C-VNI) were equal to, or greater than, that for the non-edited 5-HT_2C-INI isoform (Berg et al., 2008). One notable exception was LSD which is known not to activate the 5-HT_2C-VGV isoform (Backstrom et al., 1999; Berg et al., 2001a; Fitzgerald et al., 1999). These results are consistent with the idea that the coupling efficiency of the agonist-occupied receptors to Gq-PLC signaling is not reduced by RNA editing.

To examine the possibility that reduced constitutive activity of RNA edited 5-HT_2C isoforms is due to reduced ability to isomerize, we measured thermal stability of the receptors. Constitutively active receptors with a high capacity to isomerize are structurally unstable as a consequence of reduced stabilizing intramolecular constraints and therefore denature more readily at elevated temperature (Alewijnse et al., 2000; Gether et al., 1997; Samama et al., 1997). Receptor instability can be measured by the kinetics of the loss of ligand binding at elevated temperature (Claeysen et al., 2001). Surprisingly, we found that there was no difference in the thermal stability of the 5-HT_2C-VNI, 5-HT_2C-VSV or 5-HT_2C-VGV isoforms from the highly constitutively active non-edited 5-HT_2C-INI receptor (Berg et al., 2008). This suggests that RNA editing does not alter the isomerization capacity of 5-HT_2C receptors.

If there is no difference between the edited and non-edited 5-HT_2C isoforms for either G protein coupling efficiency or isomerization capacity, how can the difference in constitutive receptor activity be explained? We suggest that the reduced constitutive activity toward PLC of the edited 5-HT_2C isoforms is due to reduced efficiency of G protein coupling of the unoccupied receptor and that agonist occupancy promotes active receptor conformations that differ from that of unoccupied receptors such that edited and non-edited 5-HT_2CRs have an equal ability to activate Gq-PLC-IP signaling in the presence of agonist.

In summary, RNA editing can produce different 5-HT_2CR isoforms with different signaling profiles of agonist-stimulated activity and reduced levels of ligand-independent receptor activity. The dynamic nature of the expression profiles of RNA edited isoforms along with their differential distribution throughout the brain may allow for an exquisite level of fine-tuning of serotonergic neurotransmission via the 5-HT_2CR. Moreover, the exciting possibility of development of drugs which have selectivity for 5-HT_2CR isoforms may provide for enhanced therapeutic benefit with reduced adverse effects.

3. Functional neuroanatomy of the 5-HT_2CR

Localization of the mRNA which encodes for the 5-HT_2CR is restricted almost exclusively to the central nervous system, with levels undetectable in liver, kidney, intestine, heart, and lung (Julius et al. 1988). The distribution of the 5-HT_2CR protein tracks closely with that of the transcript in regions that receive innervation from 5-HT neurons arising from the midbrain raphe nuclei (Pompeiano et al. 1994). The concordance of mRNA and protein expression (Burnet et al. 1995; Lopez-Gimenez et al. 2001a, 2001b; Mengod et al. 1990a, 1990b; Pompeiano et al. 1994; Wright et al. 1995) suggests predominant postsynaptic localization of these receptors, although the 5-HT_2CR may be presynaptically localized in some brain areas (Lopez-Gimenez et al. 2001a; Mengod et al. 1990a, 1990b; Pompeiano et al. 1994). The functional significance of the 5-HT_2CR within the limbic-corticostriatal circuits is of current interest given the importance of this circuit in psychiatric and neurological disorders including anxiety, depression, drug addiction, obesity, Parkinson’s disease, and schizophrenia. The limbic-corticostriatal circuitry is an integrated collection of nuclei and pathways which functionally connects the prefrontal cortex (PFC), ventral striatum, amygdala, and hippocampus, among other nuclei. This circuit is integral in coordinating reward-related associative learning and motivated behaviors that contribute to multiple aspects of psychiatric disorders (for reviews, see Hyman et al. 2006; Kalivas and Volkow 2005; Kauer and Malenka 2007).

The distinct patterns of expression of the 5-HT_2CR, which is prominently expressed in neurons throughout the limbic-corticostriatal circuit, enables differential modulation of neurotransmission by the 5-HT_2CR. For example, microinjection studies indicate that the 5-HT_2CR in the PFC and the nucleus accumbens (NAc) do not play an active, tonic role in motor control (Filip and Cunningham 2003; Ramos et al. 2005). However, separate populations of the 5-HT_2CR within the PFC and NAc differentially influence the stimulated output of the mesocorticoaccumbens pathway, which is seen as both altered efflux of DA and γ-aminobutyric acid (GABA) neurotransmitters (see below). Microinjection studies employing selective 5-HT_2CR ligands have described an opposing stimulatory and inhibitory influence of 5-HT_2CR in the NAc and PFC, respectively, over several behavioral effects of the psychostimulant cocaine (Filip and Cunningham 2003; Liu et al. 2007) or 3,4-methylenedioxymethamphetamine (MDMA) (Ramos et al. 2005). Thus, a functional equilibrium within the limbic-corticostriatal circuit in vivo is exerted by the 5-HT_2CR.

Both the NAc and the PFC are indeed enriched in 5-HT_2CR expression. The mRNA for the 5-HT_2CR was detected in what appeared to be GABA medium spiny projection neurons in both the core and shell of the NAc, along a rostrocaudal gradient (Eberle-Wang et al. 1997). The 5-HT_2CR mRNA was detected in medium-sized GABA interneurons in layer V of the PFC (Pasqualetti et al. 1999) but only low levels of 5-HT_2CR mRNA was expressed in cortical pyramidal neurons (Lopez-Gimenez et al. 2001a). In keeping with this finding, expression of the 5-HT_2CR protein was recently shown within parvalbumin-containing GABA interneurons localized to the deep layers (V/VI) of the PFC (Liu et al. 2007). While the neurochemistry of the 5-HT_2CR in the NAc and the PFC has been of prominent interest (see section 4), the receptor is also enriched in other nodes of the limbic-corticostriatal circuit, notably the amygdala and hippocampus. Functional 5-HT_2CR protein within the basolateral amygdala is localized to GABA inhibitory interneurons and glutamate neurons (Stein et al. 2000). The 5-HT_2CR mRNA is also expressed in a subset of pyramidal hippocampal cells restricted to the CA3 field of Ammon’s horn (Pasqualetti et al. 1999), while 5-HT_2CR protein was detected in the pyramidal cell layer in both CA1 and CA3 (Clemett et al. 2000). Since the CA3 pyramidal cells project to the CA1, these data suggest that 5-HT_2CR protein in CA1 may be localized to presynaptic axon terminals from CA3 neurons (Pasqualetti et al. 1999). The amygdala 5-HT_2CR appears to be critical in the regulation of stress, fear and anxiety (Campbell and Merchant 2003; Harada et al. 2008; Li et al. 2003) while a population of hippocampal 5-HT_2CR may be involved in regulation of motility and anxiogenesis (Hackler et al. 2007; Stiedl et al. 2007; Whitton and Curzon 1990). Recent observations that activation of the 5-HT_2CR increased hippocampal neurogenesis suggests that this action may underlie learning and memory processes which involve 5-HT (Meneses and Hong 1997; Nic Dhonnchadha and Cunningham 2008) as well as the beneficial effects of serotonergic antidepressants (Banasr et al. 2004).

The ventral tegmental area (VTA) provides an interesting brain nucleus in which to consider the in vivo significance of unique regulatory mechanisms afforded to the 5-HT_2CR. Intra-VTA microinfusion of 5-HT_2CR antagonists did not alter basal nor cocaine-evoked hyperactivity (Filip and Cunningham 2002; McMahon and Cunningham 2001). However, upon activation via intra-VTA 5-HT_2CR agonist microinfusion, engagement of 5-HT_2CR-controlled pathways originating in the VTA suppressed basal motility and limited the extent of hyperactivity evoked by cocaine (Fletcher et al. 2004). As discussed in section 4, these results are supported by a recent study demonstrating that intra-VTA injection of the 5-HT_2CR agonist RO 60-0175 reduced the enhancement of DA outflow in the NAc induced by a systemic injection of cocaine, while intra-VTA administration of the 5-HT_2CR antagonist SB 242084 had no effect (Navailles et al., 2008). These multifaceted effects could potentially be related to the level of constitutive activity of the 5-HT_2CR, the preponderance of partially to fully edited 5-HT_2CR isoforms expressed and/or the dynamic expression patterns of 5-HT_2CR protein within the DA vs. GABA neurons in VTA (Bubar et al. 2005; Bubar and Cunningham 2007). Thus, the VTA provides an interesting locus for analyses of the unique biology of the 5-HT_2CR.

The VTA is localized in the ventral portion of the mesencephalon and is comprised of five subnuclei with distinct afferent and efferent projections (Phillipson 1979; Swanson 1982). The VTA is well known for supplying the major DA innervation for the limbic-corticostriatal circuit and a population of GABA neurons in the VTA send collaterals that synapse locally on DA neurons within the VTA as well as projections that terminate in both the NAc (Van Bockstaele and Pickel 1995) and/or PFC (Carr and Sesack 2000; Steffensen et al. 1998). Protein expression for the 5-HT_2CR was found to be relatively uniform in VTA GABA neurons across the rostrocaudal gradient and among the VTA subnuclei (Bubar and Cunningham 2007; Eberle-Wang et al. 1997). Conversely, this protein was shown to be differentially expressed in VTA DA neurons along distinct rostrocaudal and subnuclear patterns (Bubar and Cunningham 2007). The differences in the 5-HT_2CR distribution within DA and GABA neurons as well as the variations in the proportion of colocalization across subnuclei and rostral-caudal level suggest that discrete populations of the 5-HT_2CR in the VTA may tightly regulate the influence of 5-HT_2CR upon DA (and GABA) neurotransmission. Modulatory neurobehavioral effects could result from a functional balance between both populations of 5-HT_2CR located on GABA and DA neurons in the VTA (Navailles et al., 2008). Although the distribution of the 5-HT_2CR on these two neuronal subtypes appears to vary slightly among the rostral-caudal levels of the various subnuclei, the incidence of co-localization of 5-HT_2CR with DA neurons appears to predominate in several subnuclei, particularly in the middle VTA (Bubar and Cunningham 2007). Further examination into the impact of these different 5-HT_2CR subpopulations through systematic microinfusion studies is necessary to fully understand how the 5-HT_2CR in the VTA regulates activation of the DA mesocorticoaccumbens pathways.

The characteristics of the 5-HT_2CR as a constitutively active receptor has been extensively clarified in vitro (see section 2) and identified as relevant in the functional neurochemistry of the limbic- corticostriatal circuit in vivo (see section 4)(Berg et al. 2005). The VTA 5-HT_2CR that controls accumbal DA release appears to lack constitutive activity (Navailles et al., 2006), which may be driven by region-dependent RNA editing of the 5-HT_2CR (Burns et al. 1997). Recent studies have also found evidence for 5-HT_2CR homodimerization on the plasma membrane (Herrick-Davis et al. 2004, 2005), and endoplasmic reticulum (Herrick-Davis et al. 2006) of living cells suggesting that dimerization may also play an important role in the function of 5-HT_2CR. Although 5-HT_2CR homodimerization is difficult to explore ex vivo or in vivo, functional studies implicate the involvement of 5-HT_2CR homodimerization in ligand binding, signal transduction, and receptor trafficking processes (Herrick-Davis et al. 2004, 2005). The significance of constitutive activity and receptor dimerization in vivo has only recently come under investigation; thus, future investigations will provide a greater understanding of the importance of these processes in the regulation of 5-HT_2CR function in the VTA and elsewhere in the limbic-corticostriatal circuit and the potential relevance for consideration in the development of pharmacotherapeutics for psychiatric diseases. Drug discovery initiatives have identified a number of new 5-HT_2CR binding chemicals and active initiatives are underway to uncover the pharmacology of selective 5-HT_2CR agonists, antagonist, partial agonists and inverse agonists. Allosteric modulators of the 5-HT_2CR (Im et al. 2003) and small molecule inhibitors that disrupt the association of the 5-HT_2CR with its key binding partners (Ji et al. 2006; Sharma et al. 2007) also provide unique approaches to drug discovery in this system.

4. Neurochemical and functional aspects in vivo: focus on the 5-HT/DA interaction

Since the discovery and identification of 5-HT_2CRs in the mammalian brain (for review see Barnes and Sharp, 1999), along with their dense localization in brain dopaminergic regions (Clemett et al., 2000; Eberle-Wang et al., 1997; Pompeiano et al., 1994), much attention has been devoted at studying their functional role in the modulation of DA ascending pathway activity, i.e. the nigrostriatal and the mesocorticolimbic systems. Indeed, the 5-HT_2CR is actually considered as a pivotal pharmacological target for improved treatments of neuropsychiatric disorders related to DA neuron dysfunction, such as schizophrenia, depression, Parkinson’s disease or drug addiction (Bubar and Cunningham, 2006; Giorgetti and Tecott, 2004; Meltzer et al., 2003; Millan, 2005; Schapira et al., 2006; Wood et al., 2001).

It is now clearly established that central 5-HT_2CRs exert tonic and phasic inhibitory controls on DA neuron function in vivo. This was first suggested by electrophysiological studies with non selective 5-HT_2CR compounds (Prisco et al., 1994), and then confirmed by several electrophysiological and biochemical studies using more selective 5-HT_2CR antagonists and agonists (for review see Alex and Pehek, 2007). Thus, the basal firing rate of DA neurons in the substantia nigra pars compacta (SNc) and the VTA as well as the release of DA at terminals within the striatum, the NAc and the PFC, is increased and decreased by the peripheral administration of 5-HT_2CR antagonists and agonists, respectively (De Deurwaerdère and Spampinato, 1999, 2001; Di Giovanni et al., 1999; Di Giovanni et al., 2002; Gobert et al., 2000).

The 5-HT_2CR has been also shown to control activated DA neurons by modulating DA neuronal firing (Pierucci el al., 2004; Porras et al., 2002) and DA release (Di Matteo et al. 2004; Hutson et al., 2000; Lucas et al., 2000; Navailles et al., 2004; Porras et al., 2002). Specifically, studies with drugs which stimulate the release of DA through different cellular mechanisms (morphine, haloperidol, cocaine phencyclidine and amphetamine) led to the proposal that 5-HT_2CR exerts preferential control of DA exocytosis (Navailles et al., 2004; Willins and Meltzer, 1998) likely by regulating DA neuronal firing (Navailles et al., 2004). Indeed, the degree of DA neuronal activity appears as a permissive factor for the modulatory action of 5-HT_2CR agonists and antagonists on DA release (Lucas et al., 2000; Navailles et al., 2004; Pozzi et al., 2002).

Mesencephalic regions containing DA cell bodies (VTA and SN) has been first proposed as a primary site of action for the inhibitory control of the mesocorticolimbic and nigrostriatal DA pathways by 5-HT_2CRs (Di Matteo et al., 2001; Navailles et al., 2004; 2006b). Control of DA neuron activity is classically thought to be indirect and to involve a GABA-DA interface (Di Matteo et al., 2001; Navailles et al., 2004) in accord with the presence of 5-HT_2CR transcript and protein in VTA and SN GABA neurons (Bubar and Cunningham, 2007; Eberle-Wang et al, 1997) and with their ability to modulate GABA function within these brain regions (Bankson and Yamamoto, 2004; Di Giovanni et al., 2001; Invernizzi et al., 2007). However, the absence of effect of intra-VTA administered 5-HT_2CR agonists and antagonists on basal DA release in the NAc (Navailles et al., 2006b, 2008), although indirectly, does not support this view, and further microiontophoretic studies assessing the influence of 5-HT_2C agents on DA neuron firing are warranted to address this issue. Furthermore, the recent finding that DA neurons in the VTA co-express the protein for the 5-HT_2CR (Bubar and Cunningham, 2007; Ji et al., 2006) raises the possibility of direct excitatory control of DA neuron function. Modulatory effects of VTA DA neuron firing and accumbal DA release could result from a functional balance between both populations of 5-HT_2CRs located on GABA and DA neurons in the VTA (Navailles et al., 2008).

Most intracranial microinjection studies, but not all, have provided evidence that 5-HT_2CRs present within DA terminal regions are capable of modulating DA neuron activity, by exerting not only inhibitory but also excitatory influences on DA release. That striatal 5-HT_2CRs exert a facilitatory control of DA release in the rat striatum was first reported by Lucas and Spampinato (2000), but not confirmed by subsequent studies (Alex et al., 2005). NAc 5-HT_2CRs have been shown to inhibit (Dremencov et al., 2005), facilitate (Yan, 2000) or not affect (Navailles et al., 2006b, 2008) basal DA release in the NAc. Also, they have been shown to exert concentration-dependent excitatory and inhibitory effects on activated accumbal DA release (Navailles et al., 2008). At variance with the NAc and the striatum, compelling evidence indicates that 5-HT_2CRs localized in the PFC do not modulate basal or activated DA release in this region, either tonically or phasically (Alex et al., 2005; Pehek et al., 2006; Pozzi et al., 2002). However, as in the case of the NAC DA (Navailles et al., 2006b), PFC DA release is sensitive to VTA 5-HT_2C receptor inhibitory modulation (Pozzi et al., 2002). Furthermore, as previously suggested by behavioral investigations (Filip and Cunningham, 2003), recent neurochemical studies from our laboratory have shown that PFC 5-HT_2CRs are able to modulate activated DA release in the NAc. Intra-PFC administration of 5-HT_2CR agonists and antagonists has been shown to facilitate the release of DA induced by cocaine or morphine in the NAc (unpublished observations). Although the neuronal circuits underlying the above reported effects remains to be determined, as discussed elsewhere (Filip and Cunningham 2003; Navailles et al., 2008), 5-HT_2CR-dependent controls of DA release in DA terminal-regions, in keeping with the expression of 5-HT_2CRs on GABA cells (Eberle-Wang et al., 1997; Liu et al., 2007), may involve local GABA circuits and/or negative feedback loops to the VTA and the SN, as well as polysynaptic circuits including glutamate pathways relaying the PFC to the VTA and the NAc (Sesack et al., 2003). Hence, studies with peripheral and intracranial administration of 5-HT_2C agents altogether indicate that the overall inhibitory control of central 5-HT_2CRs on nigrostriatal and mesocorticolimbic DA pathways may be considered as a composite response involving functional balances between excitatory and inhibitory inputs to DA neurons related to different 5-HT_2CR populations located within multiple brain DA areas. Specifically, it appears that, in contrast to striatal and accumbal DA release, PFC DA release is insensitive to local control by 5-HT_2CR.

A main step in the advance of the knowledge of the functional role of the 5-HT_2CR comes from recent microdialysis studies showing that 5-HT_2C constitutive receptor activity participates in the tonic inhibitory control of DA ascending pathways in vivo (De Deurwaerdère et al., 2004). In agreement with the pharmacological characteristics of inverse agonist activity (Berg et al., 2005) and consistent with in vitro studies in Chinese Hamster ovary (CHO) cells expressing 5-HT_2CRs (Berg et al., 2006; De Deurwaerdère et al., 2004), it has been shown that the purported 5-HT_2CR antagonist SB206553 behaves in vivo as an inverse agonist at 5-HT_2CR. Indeed, SB 206553-stimulated DA release is insensitive to the decrease in 5-HT terminal activity induced by either intra-raphe injections of 5,7-dihydroxytryptamine neurotoxin, or by peripheral administration of the 5-HT_1A receptor agonist 8-OH-DPAT (De Deurwaerdère et al., 2004). Also, the 5-HT_2CR antagonists SB242084 and SB 243213 prevent the increase in striatal and accumbal DA release induced by SB 206553 and reverse the decrease in DA release produced by the 5-HT_2CR agonist Ro 60-0175 in both brain regions (Berg et al., 2006; De Deurwaerdère et al., 2004). Thus, these findings altogether indicate that the effect of SB 206553 on in vivo DA release is independent of the changes in extracellular levels of 5-HT, and is likely related to its inverse agonist properties at 5-HT_2CRs to silence their level of constitutive activity in vivo. Interestingly, inverse agonist action of SB 206553 at native 5-HT_2CRs has been recently shown in primary culture of mouse cortical neurons (Chanrion et al., 2008).

Further support for the role of 5-HT_2C constitutive receptor activity in the control of midbrain DA neuron excitability in vivo comes from recent studies with the antipsychotic drugs haloperidol and clozapine which show that 5-HT_2CR inverse agonists can have different effects than 5-HT_2CR antagonists in vivo (Navailles et al., 2006a). Indeed, the increase in accumbal and striatal DA release induced by haloperidol is potentiated by the 5-HT_2CR inverse agonist SB 206553 but unaltered by the 5-HT_2CR antagonists SB 242084 and SB 243213. Conversely, the effect of clozapine, which is known to be a strong 5-HT_2CR inverse agonist in vitro (Berg et al., 1999; 2006; Herrick-Davis et al., 2000), is unaffected by SB 206553 but blocked by SB 242084 and SB 243213. These findings, indicating that 5-HT_2C constitutive receptor activity participates in the dopaminergic effects of the antipsychotic drugs clozapine and haloperidol, suggest besides that clozapine modulates subcortical DA release by acting as a 5-HT_2CR inverse agonist in vivo.

Interestingly, intracranial microinjection studies have also shown that the control exerted by 5-HT_2C constitutive receptor activity on DA neurons occurs in a brain region-dependent manner and that the NAc may represent a primary site of action for the regulatory effects of constitutive receptor activity on the mesoaccumbens DA pathway (Navailles et al., 2006b). Intra-VTA injections of the 5-HT_2CR antagonists SB 242084 and/or SB 243213 prevent the decrease in accumbal DA outflow induced by peripheral administration of the 5-HT_2CR agonist Ro 60-0175, but do not affect the increase in DA outflow induced by the peripheral administration of the 5-HT_2CR inverse agonist SB 206553. Intra-NAc infusions of SB 242084, as in the case of its peripheral administration (De Deurwaerdère et al., 2004), block both Ro 60-0175- and SB 206553-induced changes of DA outflow. Thus, whereas VTA and NAc 5-HT_2CRs both participate in the inhibitory control exerted by 5-HT_2CR agonist on accumbal DA release, 5-HT_2CRs in the NAc are primarily involved in the tonic inhibitory control exerted by the constitutive activity of central 5-HT_2CR. In accord with this conclusion, intra-NAc, but not intra-VTA, administration of SB 206553 increases basal DA release in the NAc (Navailles et al., 2006b). The observed region-dependent effect of the inverse agonist SB 206553 could be related to different levels of 5-HT_2CR constitutive activity in the VTA and the NAc which may be related to the pre-mRNA editing of the 5-HT_2CR. Indeed, as discussed elsewhere (Navailles et al., 2006b), region-dependent RNA editing of 5-HT_2CR (Burns et al., 1997) may represent a mechanism generating receptor populations with different levels of constitutive activity (Niswender et al., 1999).

In summary, the findings reported above provide updated insights into the dominant role of the 5-HT_2CR in the regulatory neurochemistry of central DA neuron function. The 5-HT_2CR appears to possess a unique ability to tonically regulate DA release by combined actions involving the effects of endogenous 5-HT and constitutive receptor activity at different 5-HT_2CR populations present in multiple brain regions and thus may provide an exclusive functional basis for the fine-tuning of midbrain DA neuron excitability by the 5-HT_2CR in the living brain.

5. Concluding remarks

Twenty-five years after its initial identification in the central nervous system, a sizeable body of evidence has clearly demonstrated the pivotal role of the 5-HT_2CR in the regulation of neuronal network excitability. The malleability of signaling processes relative to expression of edited 5-HT_2CR isoforms throughout the brain (Schmauss, 2005), together with the relevance of the 5-HT_2C constitutive receptor activity in regulating physiological systems in vivo may provide an exceptional molecular basis for fine-tuning of 5-HT neurotransmission via the 5-HT_2CR in the living brain.

The 5-HT_2CR is thought to be implicated in the pathophysiology of several neuropsychiatric disorders (schizophrenia, depression, anxiety, sleep disorders, drug addiction, obesity), and it is actually considered as a major pharmacological target for the development of improved treatments of these diseases. In this context, it is noteworthy that the discovery of drugs with inverse agonist properties at 5-HT_2CR allows for an additional dimension for control of 5-HT_2C receptor activity, and has greatly increased the richness of our pharmacological tools. However, the benefits of inverse agonist rather than antagonist properties at the 5-HT_2CR remains to be established, and, in the coming years, further experimental and clinical evaluations are needed for a better understanding of the functional significance of constitutive receptor activity and the therapeutic potential of inverse agonism in vivo.