Serotonin, by activating different receptor subtypes, regulates membrane excitability in the central nervous system in a complex manner (Andrade 1998). The involvement of various voltage-gated ion channels in this action of serotonin receptors has been demonstrated (Carr et al. 2002; Colino and Halliwell 1987; Haj-Dahmane and Andrade 1996; Penington and Kelly 1990). However, application of a low concentration of a 5-HT1A or 5-HT2A/2C agonist alone fails to alter the action potential firing elicited by somatic current injection in prefrontal pyramidal neurons. These results had suggested that the low level of 5-HT1A or 5-HT2A/C activation is not sufficient to trigger the change in voltage-gated channels that affects neuronal excitability. In spite of this, 5HT1A receptor activation can induce a membrane hyperpolarization in cortical neurons (Araneda and Andrade 1991) and in hippocampal neurons (Andrade et al. 1986); this last probably by opening inwardly rectifying potassium channels (Colino and Halliwell 1987) and inhibiting voltage-dependent calcium channels (Penington and Kelly 1990). In contrast, 5-HT2A/2C receptor activation can induce a membrane depolarization in cortical neurons (Aghajanian and Marek 1997; Araneda and Andrade 1991), probably by inhibiting an inwardly rectifying potassium conductance (North and Uchimura 1989) or activating a nonselective cationic current (Haj-Dahmane and Andrade 1996).
Regarding 5-HT2C receptor participation on cellular excitability, it has been observed that iontophoretic application of 5-HT ligands suppresses spontaneous firing of prefrontal cortical neurons (Berg et al. 2008; Bergqvist et al. 1999), suggesting that the 5-HT2C receptor limits the excitability of cortical pyramidal neurons
(Carr et al. 2002). Moreover, it has been observed that the 5HT2C receptor agonist 6-chloro-2-(1-piperazinyl)-pyrazine (MK-212) completely prevents the expression of behavioral sensitization in animals pretreated with 3,4-methylenedioxymetham-phetamine (MDMA) (Ramos et al. 2005). Accordingly, previous studies by Pan and Wang (1991a, b) showed that the inhibition of pyramidal cells in the prefrontal cortex produced by MDMA is mediated mainly through the serotoninergic system. Moreover, it has been hypothesized that lesions of serotonergic terminals induced by repeated MDMA injections could increase the inhibitory effect of serotonin through a hypersensitization mechanism, similarly to the effect of 5,7-DHT lesions (Ashby et al. 1994; Conn et al. 1987; Conn and Sanders-Bush 1987). In this manner, these cells would show higher sensitivity to the inhibitory effects of 5-HT2C receptor stimulation by SCH-23390, which shows high affinity for 5-HT2C receptors (Aguirre et al. 1995). It is also noteworthy that a similar MDMA treatment produces a long-term increase of 5-HT2C receptor mRNA expression in the hippocampus (Yau et al. 1994).
Recently, it was shown that the 5-HT2C receptor is predominantly expressed in the deep layers (layers V/VI) in comparison with those in superficial layers (layers I/II/III) of the medial prefrontal cortex (Liu et al. 2007). The 5-HT2C receptor immunoreactivity was found to be intense in the periphery of cell bodies and processes proximal to cell bodies in primarily round- or fusiform-shaped cells in the rat medial prefrontal cortex. Moreover, 50%t of the 5-HT2C receptor immunoreac-tivity in this region is colocalized with glutamate decarboxylase (GAD) isoform 67 immunoreactivity, a marker of GABA interneurons (Liu et al. 2007). These findings are coincident with previous data demonstrating that 5-HT2C receptor mRNA is present in a subpopulation of GABA-ergic interneurons in the prefrontal cortex (Vysokanov et al. 1998). In addition, serotonergic axons in the monkey prefrontal cortex predominantly synapse on interneurons (Smiley and Goldman-Rakic 1996). Studies have also shown that the local administration of the 5-HT2A/2C receptor agonist DOI within the medial prefrontal cortex increased extracellular GABA levels in this area (Abi-Saab et al. 1999), an effect mediated by the 5-HT2C receptors and not through the 5-HT2A receptor (Liu et al. 2007).
The 5-HT2C receptor immunoreactivity in the prelimbic cortex was localized primarily to parvalbumin-positive interneurons, which include basket and chandelier neurons that innervate the cell body and the initial segment, respectively, of pyramidal neurons (Conde et al. 1994; Gabbott and Bacon 1997). In contrast, some studies have demonstrated that 25-50% of pyramidal neurons in the prefrontal cortex contain mRNA for 5-HT2C receptors (Vysokanov et al. 1998). Thus, the inhibitory synapses formed by parvalbumin-expressing basket and chandelier neurons are proximal to the initial axon segment of pyramidal cell, the sites at which action potentials are generated. As such, a primary function of parvalbumin-posi-tive GABA neurons is to modulate the efferent signaling of pyramidal neurons (Lewis et al. 2005; Markram et al. 2004; Miles et al. 1996). The 5-HT2C receptor-mediated influence on GABA interneurons would be expected to primarily target the basket and chandelier cells that provide the strongest inhibitory effects on the output of cortical pyramidal cells, compared with interneurons that innervate distal dendrites of the pyramidal cells, which probably function to regulate incoming afferent signals to the pyramidal neurons (Miles et al. 1996). The localization of 5-HT2C receptors at the deep layers of the rat medial prefrontal cortex (layers V/VI) suggests that these receptors act to modulate the output of neurons in these layers. Thus, release of GABA via activation of the 5-HT2C receptors on parvalbumin-positive medial prefrontal cortex GABA interneurons would be expected to reduce excitatory glutamate output as well as subsequent dopamine neurotransmission within the mesoaccumbens pathway (Liu et al. 2007).
With respect to serotonin-glutamate interaction, previous studies have shown that the N-methyl-d-aspartate (NMDA) channel receptor is an important target of 5-HT2 and 5-HT1A receptor modulation (Arvanov et al. 1999; Blank et al. 1996; Yuen et al. 2005). In the presence of NMDA, application of low concentrations of 5-HT or 5-HT2A/2C agonists exerts a reducing or enhancing effect, respectively, on the action potential firing, suggesting that NMDA receptor activation provides a "gate" to facilitate the opposing regulation of neuronal excitability by the low level of 5-HT1A or 5-HT2A/2C agonists (Ping et al. 2008). The reason why NMDA facilitates serotonergic regulation of neuronal excitability is likely to be the change of intracellular signaling molecule(s) downstream of Ca2+ flow through NMDA channel receptors (Zhong et al. 2008). Evidence shows that the opposing effects of 5-HT or 5-HT2A/2C on neuronal excitability are mediated by differential regulation of a converging target, extracellular signal-regulated kinase (ERK). Extracellular signal-regulated kinase can be regulated by the protein kinase A (PKA) or the protein kinase C (PKC) cascade (Roberson et al. 1999) downstream of 5-HT1A or 5-HT2A/C receptors since ERK activation can increase the amplitude of backpropa-gating action potentials by phosphorylating dendritic A-type K+ channel Kv4.2 subunits (Yuan et al. 2000). In this way, one possible mechanism underlying the regulation of neuronal excitability by 5-HT-NMDA interactions is through the ERK modification of dendritic K+ channels (Zhong et al. 2008).
Recently, the possible involvement of the regulation of neuronal excitability by 5-HT-NMDA interactions in cognitive and emotional processes has been addressed by examining animals exposed to acute stress, since many mental illnesses are exacerbated by stressful conditions (Arnsten and Goldman-Rakic 1998; Mazure et al. 1995). Accordingly, several lines of evidence have shown that stress interferes with serotonin neurotransmission by changing serotonin release or serotonin receptor functions (Adell et al. 1997; Lowry et al. 2000; Maswood et al. 1998; Tan et al. 2004). For example, in the forced swim test, a behavioral paradigm often used to evaluate antidepressant/anxiolytic efficacy, the effect of 5-HT2A/2C, but not 5-HT1A, on action potential firing is lost in stressed animals, which is associated with the selective loss of 5-HT2A/2C-induced increase of ERK activity. This suggests that 5-HT2A/2C receptors are probably desensitized and inactivated by elevated levels of serotonin in response to stress stimulation (Zhong et al. 2008). The other substrate through which 5-HT2C receptors affect psychological processes is the mesoaccumbens pathway, central in psychological processes including motivation, reward, and mood (Nestler and Carlezon 2006; Pierce and Kumaresan 2006; Salamone 1996) and an important site for the actions of psychostimulants such as cocaine (Filip and Cunningham 2003; Pierce and Kumaresan 2006). Thus, alterations in accumbal dopamine or glutamate levels consequent to stimulation of 5-HT2C receptors in the medial prefrontal cortex may be one mechanism by which 5-HT2C receptors modulate the neurochemical and behavioral effects of psychostimulants (Alex and Pehek 2007).
Intramedial prefrontal infusion of 5-HT2C receptor agonists has been shown to block the hypermotive and discriminative stimulus effects of cocaine (Filip and Cunningham 2003) and block the expression of sensitization (the progressive enhancement of the hypermotive effects of a drug following repeated drug exposure) to MDMA (Ramos et al. 2005). On the other hand, intramedial prefrontal cortical infusion of 5-HT2C receptor antagonists enhanced cocaine-induced hyper-activity and increased recognition of the stimulus effects of cocaine (Filip and Cunningham 2003).
Conversely, a growing body of evidence suggests that 5-HT2A and 5-HT2C receptors have opposing functional roles. For example, 5-HT2C receptors appear to inhibit dopamine release, whereas activation of 5-HT2A receptors enhances it (Di Matteo et al. 2001, 2002; Millan et al. 1998). Moreover, antagonism of 5-HT2C receptors potentiates some of the behavioral effects of cocaine, whereas antagonism of 5-HT2A receptors attenuates both cocaine-induced hypermotility and reinstatement of cocaine-seeking behavior (Cunningham et al. 1992; Fletcher et al. 2002a).
In the next section, the 5-HT2C effects on sustained prefrontal learning tasks will be summarized.
Was this article helpful?