5-HT₄ receptors are expressed in the basal ganglia and limbic system

Since its first description 30 years ago (Dumuis, Bouhelal, Sebben, Cory, & Bockaert, Reference Dumuis, Bouhelal, Sebben, Cory and Bockaert1988), the 5-HT₄ receptor has been discovered both in the brain and in various sites in the periphery. Peripherally, 5-HT₄ receptors are found in the heart (Kaumann, Reference Kaumann1990), the intestine (Craig & Clarke, Reference Craig and Clarke1990), the adrenal glands (Idres, Delarue, Lefebvre, & Vaudry, Reference Idres, Delarue, Lefebvre and Vaudry1991) and the bladder (Tonini & Candura, Reference Tonini and Candura1996). In the brain, 5-HT₄ receptors are located post-synaptically and are expressed in the hypothalamus, hippocampus, basal ganglia, amygdala, olfactory bulb, septal area and neocortex (Cai, Flores-Hernandez, Feng, & Yan, Reference Cai, Flores-Hernandez, Feng and Yan2002; Compan et al., Reference Compan, Daszuta, Salin, Sebben, Bockaert and Dumuis1996; Roychowdhury, Haas, & Anderson, Reference Roychowdhury, Haas and Anderson1994; Vilaró, Cortés, & Mengod, Reference Vilaró, Cortés and Mengod2005; Waeber, Sebben, Nieoullon, Bockaert, & Dumuis, Reference Waeber, Sebben, Nieoullon, Bockaert and Dumuis1994). In rodents, 5-HT₄ receptors are mainly situated on gamma-aminobutyric acid (GABA) neurones in the striatum and nucleus accumbens (Compan et al., Reference Compan, Daszuta, Salin, Sebben, Bockaert and Dumuis1996), glutamate pyramidal cells in the medial prefrontal cortex (PFC) and hippocampus (Cai et al., Reference Cai, Flores-Hernandez, Feng and Yan2002; Vilaró et al., Reference Vilaró, Cortés and Mengod2005) and hippocampal granule cells (Vilaró et al., Reference Vilaró, Cortés and Mengod2005). 5-HT₄ receptors have also been described in the rat dorsal raphe nucleus (DRN) in autoradiographic studies (Vilaró et al., Reference Vilaró, Cortés and Mengod2005; Waeber et al., Reference Waeber, Sebben, Nieoullon, Bockaert and Dumuis1994), although 5-HT₄ receptor mRNA has not been found suggesting that the 5-HT₄ receptors are located on DRN afferent neurons rather than on DRN 5-HT cells (Vilaró et al., Reference Vilaró, Cortés and Mengod2005).

Evidence from PET studies in humans confirms that 5-HT₄ receptors have a high density in the caudate, putamen and nucleus accumbens, with lower expression across a broad range of cortical and limbic areas, including the hippocampus and amygdala (Beliveau et al., Reference Beliveau, Ganz, Feng, Ozenne, Hojgaard, Fisher and Knudsen2017). Similarly, the highest levels of human 5-HT₄ receptor mRNA have been found in the caudate and putamen, with lower levels in the amygdala temporal cortex and hippocampus (Medhurst, Lezoualc'h, Fischmeister, Middlemiss, & Sanger, Reference Medhurst, Lezoualc'h, Fischmeister, Middlemiss and Sanger2001). Post-mortem human brain studies have also confirmed a high density of the 5-HT₄ receptor in the basal ganglia (caudate nucleus, putamen, nucleus accumbens, globus pallidus and substantia nigra), as well as the hippocampus (particularly in the CA1 region) and an intermediate density in the neocortex and amygdala (Bonaventure et al., Reference Bonaventure, Hall, Gommeren, Cras, Langlois, Jurzak and Leysen2000; Varnas, Halldin, Pike, & Hall, Reference Varnas, Halldin, Pike and Hall2003). In addition, low levels of 5-HT₄ binding sites have been reported in the human thalamus and raphe nuclei (Beliveau et al., Reference Beliveau, Ganz, Feng, Ozenne, Hojgaard, Fisher and Knudsen2017; Varnas et al., Reference Varnas, Halldin, Pike and Hall2003).

The neurochemical and molecular effects of 5-HT₄ receptor activation

5HT₄ receptors are transmembrane-spanning G protein-coupled receptors, and their activation results in increased neuronal excitability that is mediated through the stimulation of adenylyl cyclase and an increase in cyclic adenosine monophosphate (Bockaert, Claeysen, Compan, & Dumuis, Reference Bockaert, Claeysen, Compan and Dumuis2008; Dumuis et al., Reference Dumuis, Bouhelal, Sebben, Cory and Bockaert1988). In addition, 5-HT₄ receptor activation initiates important G protein-independent pathways, including the activation of tyrosine kinase Src and, in turn, neuronal extracellular-signal-regulated kinase (ERK) (Barthet et al., Reference Barthet, Framery, Gaven, Pellissier, Reiter, Claeysen and Dumuis2007), which is known to play a critical role in hippocampal synaptic plasticity (Norman, Thiels, Barrionuevo, & Klann, Reference Norman, Thiels, Barrionuevo and Klann2000).

5-HT₄ receptors also indirectly affect neuronal function through their modulation of neurotransmitter release, including GABA (Bijak & Misgeld, Reference Bijak and Misgeld1997), acetylcholine (Consolo, Arnaboldi, Giorgi, Russi, & Ladinsky, Reference Consolo, Arnaboldi, Giorgi, Russi and Ladinsky1994; Johnson et al., Reference Johnson, Drummond, Grimwood, Sawant-Basak, Miller, Tseng and Siok2012; Siniscalchi, Badini, Beani, & Bianchi, Reference Siniscalchi, Badini, Beani and Bianchi1999), dopamine (Bockaert, Claeysen, Compan, & Dumuis, Reference Bockaert, Claeysen, Compan and Dumuis2004; Bockaert et al., Reference Bockaert, Claeysen, Compan and Dumuis2008; Bonhomme, De Deurwaerdere, Le Moal, & Spampinato, Reference Bonhomme, De Deurwaerdere, Le Moal and Spampinato1995), histamine (Johnson et al., Reference Johnson, Drummond, Grimwood, Sawant-Basak, Miller, Tseng and Siok2012) and serotonin (Ge & Barnes, Reference Ge and Barnes1996; Licht et al., Reference Licht, Kirkegaard, Zueger, Chourbaji, Gass, Aznar and Knudsen2010). Of particular interest is the enhanced release of acetylcholine in the frontal cortex and hippocampus following 5-HT₄ receptor activation, which has been linked to its facilitatory role in memory and cognition (Hagena & Manahan-Vaughan, Reference Hagena and Manahan-Vaughan2017; King, Marsden, & Fone, Reference King, Marsden and Fone2008). In support of this 5-HT₄ receptor agonists have been shown to increase electrically-evoked [3H]choline efflux in guinea pig brain slices in the cerebral cortex, hippocampus and nucleus basalis magnocellularis (NbM) (Siniscalchi et al., Reference Siniscalchi, Badini, Beani and Bianchi1999), and facilitate the release of ACh selectively in the frontal cortex (Consolo et al., Reference Consolo, Arnaboldi, Giorgi, Russi and Ladinsky1994). Importantly, both of these effects are blocked by 5-HT₄ receptor antagonists, which further supports a specific role for the 5-HT₄ receptor in cholinergic neurotransmission. Given that 5-HT₄ receptors are not present on cholinergic basal forebrain neurons, it is thought that the enhancement of ACh release induced by 5-HT₄ receptor agonists is likely to be mediated by GABAergic and/or glutamatergic cell populations (Penas-Cazorla & Vilaro, Reference Penas-Cazorla and Vilaro2015).

Importantly, 5-HT₄ receptors in the PFC control the firing rate of midbrain serotonergic cells. Consistent with this, the basal firing of 5-HT cells in the DRN is reduced by over 50% in 5-HT₄ receptor knockout mice compared with wildtypes (Conductier et al., Reference Conductier, Dusticier, Lucas, Cote, Debonnel, Daszuta and Compan2006). The facilitatory control exerted by 5-HT₄ receptors on serotonergic firing is thought to operate via a positive PFC–DRN feedback loop in which the activation of postsynaptic 5-HT₄ receptors located on medial PFC pyramidal cells has an excitatory effect on serotonin neurons in the DRN dorsal raphe, resulting in an increase in the release of serotonin in target structures such as the PFC and hippocampus. Consistent with this model, overexpression of 5-HT₄ receptors in the mPFC, but not in the striatum or hippocampus, increases DRN 5-HT neuronal activity (Lucas et al., Reference Lucas, Compan, Charnay, Neve, Nestler, Bockaert and Debonnel2005). In addition, pharmacological studies have demonstrated that administration of the selective 5-HT₄ receptor agonists, RS67333 and prucalopride, increases DRN 5-HT cell firing (Faye et al., Reference Faye, Hen, Guiard, Denny, Gardier, Mendez-David and David2019; Lucas et al., Reference Lucas, Compan, Charnay, Neve, Nestler, Bockaert and Debonnel2005; Lucas & Debonnel, Reference Lucas and Debonnel2002; Moha ou Maati et al., Reference Moha ou Maati, Bourcier-Lucas, Veyssiere, Kanzari, Heurteaux, Borsotto and Lucas2016). Intriguingly, this increase in firing can be seen within 30 min of drug administration, suggesting that 5-HT₄ receptor agonists hold potential as rapid onset antidepressant agents (Lucas et al., Reference Lucas, Compan, Charnay, Neve, Nestler, Bockaert and Debonnel2005; Lucas & Debonnel, Reference Lucas and Debonnel2002). This is in stark contrast to the decrease in raphe cell firing that is characteristic of SSRIs, and which has been implicated in their delayed onset of action (Blier, Piñeyro, el Mansari, Bergeron, & de Montigny, Reference Blier, Piñeyro, el Mansari, Bergeron and de Montigny1998). Importantly, the excitatory effect of 5-HT₄ receptor agonists on the firing of midbrain 5-HT cells is abolished by bilateral electrolytic lesions of the mPFC, demonstrating a causal role of the mPFC in the facilitation of 5-HT cell firing by 5-HT₄ receptor agonists (Moha ou Maati et al., Reference Moha ou Maati, Bourcier-Lucas, Veyssiere, Kanzari, Heurteaux, Borsotto and Lucas2016).

Short-term 5-HT₄ receptor activation has also been shown to increase hippocampal cell proliferation and the expression of neuroplasticity-related proteins such as CREB and BDNF (Ishizuka, Goshima, Ozawa, & Watanabe, Reference Ishizuka, Goshima, Ozawa and Watanabe2014; Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007; Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012). These mirror those seen with chronic antidepressant administration (Malberg, Eisch, Nestler, & Duman, Reference Malberg, Eisch, Nestler and Duman2000; Warner-Schmidt & Duman, Reference Warner-Schmidt and Duman2006), but occur more rapidly. For example, 3 days administration of the 5-HT₄ receptor agonist, RS67333, increases hippocampal cell proliferation in the dentate gyrus and the expression of neuroplasticity-related proteins, such as BDNF and CREB (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007; Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012). After 7 days of administration the level of these neuroplastic changes are comparable with those seen after 2–3 weeks of SSRI administration (Malberg et al., Reference Malberg, Eisch, Nestler and Duman2000; Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012; Samuels et al., Reference Samuels, Mendez-David, Faye, David, Pierz, Gardier and David2016). Further, long-term 5-HT₄ receptor activation (4 weeks) facilitates the maturation of newborn neurons in the adult rat hippocampus (Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014).

Furthermore, 5-HT₄ receptors are involved in fine-tuning of the key cellular processes underlying synaptic plasticity, long-term potentiation (LTP) and long-term depression (LTD), which is likely to play a key role in their effects on long term hippocampal-dependent memory. Activation of the 5-HT₄ receptor has been shown to prevent LTD in all the main hippocampal subfields (Kemp & Manahan-Vaughan, Reference Kemp and Manahan-Vaughan2005; Twarkowski, Hagena, & Manahan-Vaughan, Reference Twarkowski, Hagena and Manahan-Vaughan2016). The effect of 5-HT₄ receptor activation on LTP is complex and specific to individual hippocampal subfields, which may relate to differences in the expression of different receptor isoforms (Hagena & Manahan-Vaughan, Reference Hagena and Manahan-Vaughan2017). A number of studies have demonstrated that stimulation of 5-HT₄ receptors in freely behaving rats promotes synaptic information-encoding through LTP at both the CA1 region (Kemp & Manahan-Vaughan, Reference Kemp and Manahan-Vaughan2005) and dentate gyrus (Kulla & Manahan-Vaughan, Reference Kulla and Manahan-Vaughan2002; Twarkowski et al., Reference Twarkowski, Hagena and Manahan-Vaughan2016). Conversely, LTP at mossy fibre synapses of the CA3 region is suppressed by 5-HT₄ receptor activation (Twarkowski et al., Reference Twarkowski, Hagena and Manahan-Vaughan2016). The effect of 5-HT₄ receptor activation on LTP has also been shown to be time-dependent; for example, the long lasting sustainment of LTP (i.e. after 48 h) in the dentate gyrus can be impaired by 5-HT₄ receptor stimulation (Twarkowski et al., Reference Twarkowski, Hagena and Manahan-Vaughan2016). Taken together, this pattern suggests that 5-HT₄ receptor activation acts to promote information encoding through LTP in CA1 and dentate gyrus at the expense of information coding through LTD, which is likely to impact upon the content of information encoding (Hagena & Manahan-Vaughan, Reference Hagena and Manahan-Vaughan2017; Twarkowski et al., Reference Twarkowski, Hagena and Manahan-Vaughan2016). The effect of 5-HT₄ receptor activation on dentate gyrus plasticity is particularly interesting in light of recent evidence that LTP induction in the dentate gyrus is sufficient to induce antidepressant-like effects in rat models within a rapid timeframe (Kanzari et al., Reference Kanzari, Bourcier-Lucas, Freyssin, Abrous, Haddjeri and Lucas2018), suggesting that the facilitatory effect of 5-HT₄ receptor agonists on DG plasticity may, at least in part, account for their fast-acting antidepressant-like properties.

Preclinical studies support a role for the 5-HT₄ receptor in the aetiology of depression and anxiety

Preclinical studies provide support for the involvement of the 5-HT₄ receptor in the aetiology of depression. Changes in 5-HT₄ receptor expression have been reported in a number of rodent models of depression, although the direction of effect varies by model. In the olfactory bulbectomy (OBX) and glucocorticoid receptor heterozygous [GR(±)] models, an increase in 5-HT₄ receptor binding has been reported in the ventral hippocampus (OBX) and striatum [GR(±)] (Licht et al., Reference Licht, Kirkegaard, Zueger, Chourbaji, Gass, Aznar and Knudsen2010). In contrast, in the Flinders-sensitive line rat model of depression, a decrease in hippocampal 5-HT₄ receptor binding was reported (Licht et al., Reference Licht, Marcussen, Wegener, Overstreet, Aznar and Knudsen2009). 5-HT₄ receptor mRNA and protein has also been shown to be down-regulated in rats subjected to maternal deprivation (but not chronic unpredictable stress), with a strong correlation between levels of hippocampal 5-HT₄ receptor mRNA and anhedonic-like behaviour in these animals (Bai et al., Reference Bai, Zhu, Zhang, Zhang, Zhang, Xue and Zhang2014).

5-HT₄ receptor knock out mice do not show a clear depressed-like phenotype, with no differences in behaviour on the forced swim test, which may be due to the considerable adaptations in the serotoninergic system that occur across development in these animals (Amigo et al., Reference Amigo, Diaz, Pilar-Cuellar, Vidal, Martin, Compan and Castro2016; Conductier et al., Reference Conductier, Dusticier, Lucas, Cote, Debonnel, Daszuta and Compan2006). However, the knock out animals do show some evidence of anhedonic- and apathetic-like behaviour, with a reduction in sucrose consumption and nesting behaviour, and decreased central time in the open field (OF), which is suggestive of increased anxiety (Amigo et al., Reference Amigo, Diaz, Pilar-Cuellar, Vidal, Martin, Compan and Castro2016). Interestingly, a recent study suggests that focal overexpression of the 5-HT₄ receptor in the medial PFC produces an ‘anti-depressed’ and anxiolytic behavioural phenotype, which is consistent with the central role ascribed to descending inputs from the mPFC to the DRN in mediating increased raphe firing (Castello et al., Reference Castello, LeFrancois, Flajolet, Greengard, Friedman and Rebholz2018).

5-HT₄ receptors are involved in the mechanism of action of SSRIs

There is also convincing evidence that 5-HT₄ receptors play a role in the action of SSRIs in animal models. Chronic administration of monoaminergic antidepressants, such as fluoxetine, paroxetine and venlafaxine (but not reboxetine), decreases 5-HT₄ receptor density in the basal ganglia and hippocampus in the rat brain (Licht et al., Reference Licht, Marcussen, Wegener, Overstreet, Aznar and Knudsen2009; Vidal, Valdizan, Mostany, Pazos, & Castro, Reference Vidal, Valdizan, Mostany, Pazos and Castro2009; Vidal, Valdizan, Vilaró, Pazos, & Castro, Reference Vidal, Valdizan, Vilaró, Pazos and Castro2010). In contrast, expression of the 5-HT₄ receptor has been shown to be upregulated after chronic fluoxetine administration in corticostriatal pyramidal cells expressing the adaptor protein p11, located in layer 5 of the cerebral cortex (Schmidt et al., Reference Schmidt, Warner-Schmidt, Otopalik, Pickett, Greengard and Heintz2012).

More direct evidence for a role for the 5-HT₄ receptor in the action of conventional antidepressants comes from studies in which 5-HT₄ receptor ligands have been shown to modify the effects of monoaminergic antidepressants. For example, administration of the 5-HT₄ receptor agonist RS67333 has been shown to augment the acute effect of paroxetine on extracellular serotonin levels in the rat ventral hippocampus (Licht et al., Reference Licht, Kirkegaard, Zueger, Chourbaji, Gass, Aznar and Knudsen2010). Perhaps the most convincing evidence comes from studies which demonstrate that disrupting 5-HT₄ receptor function can abolish the behavioural effects of SSRIs. For example, the 5-HT₄ receptor antagonist GR125487 blocks the effects of fluoxetine on a range of depression- and anxiety-related paradigms in a corticosterone model (CORT) of anxiety/depression (Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014), suggesting that 5-HT₄ receptor activation is necessary for the antidepressant and anxiolytic effects of SSRIs (although see Cryan and Lucki, Reference Cryan and Lucki2000). A similar effect is found in 5-HT₄ receptor knock out mice, where chronic fluoxetine administration has been shown to reverse the behavioural effects of OBX in wild type but not knock out animals (Amigo et al., Reference Amigo, Diaz, Pilar-Cuellar, Vidal, Martin, Compan and Castro2016).

Interestingly, there is also evidence to suggest that the 5-HT₄ receptor contributes to the neurotrophic effects of chronic SSRI treatment. For example, the usual increase in cell proliferation in the dentate gyrus following chronic administration of fluoxetine is not present in 5-HT₄ receptor knock out mice (Imoto et al., Reference Imoto, Kira, Sukeno, Nishitani, Nagayasu, Nakagawa and Segi-Nishida2015), and is partially blocked by 5-HT₄ receptor antagonism (Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014), suggesting that the neurogenic effects of SSRIs may be partly mediated by the 5-HT₄ receptor. Further support for this comes from a study which demonstrates that the induction of immature like granule cells by chronic fluoxetine treatment (‘dematuration’) is blocked in 5-HT₄ receptor knock-out mice (Kobayashi et al., Reference Kobayashi, Ikeda, Sakai, Yamasaki, Haneda, Miyakawa and Suzuki2010).

Short-term administration of 5-HT₄ receptor agonists has antidepressant and anxiolytic-like effects in animals

Studies in rodents have demonstrated that 5-HT₄ receptor agonists exert antidepressant-like effects on behavioural paradigms that mirror those seen with conventional SSRIs, but with a more rapid onset of action (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007; Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014; Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012). Using the classic forced swim paradigm, the 5-HT₄ receptor partial agonists RS67333 and prucalopride have been shown to reduce immobility following acute (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007) and 3–5 day administration (Gomez-Lazaro et al., Reference Gomez-Lazaro, Garmendia, Beitia, Perez-Tejada, Azpiroz and Arregi2012; Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012). These effects parallel those seen with acute SSRI administration, which also induces a reduction in immobility on this task, although the effect of the 5-HT₄ agonists was almost double that seen with acute citalopram (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007).

Short-term administration of 5-HT₄ agonists also induces behavioural changes in paradigms that require chronic administration of SSRIs to shift. For example, 3 days administration of RS67333 was sufficient to increase sucrose consumption in the chronic mild stress model of depression (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007) and in CORT rats (Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012), and decrease hyperlocomotion in the OBX model (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007). Within 7 days of RS67333 a complete reversal of the effects of corticosterone on sucrose consumption was seen, paralleling the effect of 21-day treatment with fluoxetine (Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012), and suggesting a rapid reversal of anhedonic behaviour following 5-HT₄ receptor agonism. These effects are consistent with the neurobiological adaptations that have been reported following 3 days of 5-HT₄ receptor agonism, including desensitisation of 5-HT_1A autoreceptors, increased CREB phosphorylation and increased hippocampal cell proliferation (Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007), and suggest that 5-HT₄ receptor agonists have a unique, rapid-acting profile of effects.

Intriguingly, co-administration of 5-HT₄ receptor agonists with SSRIs prevents the attenuation in raphe firing that is typically seen at the onset of SSRI administration (Lucas et al., Reference Lucas, Du, Romeas, Mnie-Filali, Haddjeri, Piñeyro and Debonnel2010), suggesting that 5-HT₄ receptor agonists may have promise as an augmentation strategy for SSRI treatment. In support of this proposal, co-administration of RS67333 and the SSRI citalopram led to a synergistic inhibition of hippocampal firing, while the same combination potentiated the ability of RS67333 to phosphorylate CREB in the hippocampus and reduce immobility in the forced swim test (Lucas et al., Reference Lucas, Du, Romeas, Mnie-Filali, Haddjeri, Piñeyro and Debonnel2010).

5-HT₄ receptor agonists have also been associated with rapid effects on anxiety. Seven days administration of RS67333 reduces latency to feed in the novelty suppressed feeding (NSF) paradigm, whereas 14 days administration of fluoxetine is required to see the same effect (Pascual-Brazo et al., Reference Pascual-Brazo, Castro, Díaz, Valdizán, Pilar-Cuéllar, Vidal and Pazos2012). A similar effect was seen in the mouse CORT model of anxiety/depression, where 7 day administration of RS67333, but not fluoxetine, reduced anxiety-related behaviour in the OF test and the elevated plus maze (EPM), although the effects of subchronic RS67333 on the NSF were not replicated in this model (Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014). Interestingly, the effects of RS67333 on the OF and EPM tests were also evident in animals who had undergone hippocampal X-irradiation, suggesting that they are not dependent on hippocampal neurogenesis (Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014).

Consistent with a fast onset of action, acute administration of RS67333 is sufficient to induce an anxiolytic-like effect on the elevated zero-maze (Bell, Duke, Gilmore, Page, & Begue, Reference Bell, Duke, Gilmore, Page and Begue2014), and reverse the anxiogenic effects of chronic cannabinoid exposure during adolescence (Abboussi et al., Reference Abboussi, Said, Fifel, Lakehayli, Tazi and El Ganouni2016). A recent report confirmed an anxiolytic-like effect of acute systemic infusion of RS67333 on the EPM, the OF test and the NSF test (Faye et al., Reference Faye, Hen, Guiard, Denny, Gardier, Mendez-David and David2019). Intriguingly, a similar profile was seen when RS67333 was directly infused into the mPFC, and the anxiolytic-like effects of RS67333 were attenuated by optogenetic inhibition of the mPFC terminals in the DRN, suggesting that direct activation of 5-HT₄ receptors in the mPFC is responsible for the fast anxiolytic effects of 5-HT₄ receptor agonists, via the mPFC-DRN circuit (Faye et al., Reference Faye, Hen, Guiard, Denny, Gardier, Mendez-David and David2019).

The effects of 5HT₄ receptor antagonism on animal depression- or anxiety-like behaviour are less well established. Acute administration of the 5-HT₄ receptor antagonists SDZ205-557, GR113808 and SB204070 does not influence anxiety-related behaviour in the light/dark choice test in mice (Costall & Naylor, Reference Costall and Naylor1997). Four weeks administration of the antagonist GR125487 has also been shown not to have an effect on behavioural models of anxiety in a corticosterone model of depression/anxiety, although the antagonist did block the effects of an SSRI in this model (Mendez-David et al., Reference Mendez-David, David, Darcet, Wu, Kerdine-Römer, Gardier and Hen2014). In contrast, two studies have reported an anxiolytic effect of acute administration of SB204070, GR113808 (Silvestre, Fernandez, & Palacios, Reference Silvestre, Fernandez and Palacios1996) and SB207266A (Kennett, Bright, Trail, Blackburn, & Sanger, Reference Kennett, Bright, Trail, Blackburn and Sanger1997; Silvestre et al., Reference Silvestre, Fernandez and Palacios1996) on the EPM test. It is unclear what explains these contradictory findings, although there are differences in dosage, timing of antagonist administration and behavioural outcome measures used in the studies, which may be important.

5-HT₄ Receptor activation has a facilitatory effect on rodent tests of learning and memory

A large body of evidence demonstrates a facilitatory effect of 5-HT₄ receptor activation on rodent and primate behavioural tests of cognition, and in particular learning and memory (Hagena & Manahan-Vaughan, Reference Hagena and Manahan-Vaughan2017; King et al., Reference King, Marsden and Fone2008). Early studies using relatively non-selective compounds, such as the mixed 5-HT₄ receptor agonist/5-HT₃ receptor antagonist BIMU1 demonstrated a facilitation of autoshaping (Meneses & Hong, Reference Meneses and Hong1997) and associative memory (Marchetti-Gauthier, Roman, Dumuis, Bockaert, & Soumireu-Mourat, Reference Marchetti-Gauthier, Roman, Dumuis, Bockaert and Soumireu-Mourat1997), which was attributed to 5-HT₄ receptor activation since both effects were blocked by co-administration of a 5-HT₄ receptor antagonist. Subsequent studies using more selective agonists confirmed that acute 5-HT₄ receptor activation improves a wide of range of behavioural tests of learning and memory, including place recognition (Lamirault & Simon, Reference Lamirault and Simon2001), object recognition (Lamirault & Simon, Reference Lamirault and Simon2001; Levallet, Hotte, Boulouard, & Dauphin, Reference Levallet, Hotte, Boulouard and Dauphin2009; Moser et al., Reference Moser, Bergis, Jegham, Lochead, Duconseille, Terranova and Scatton2002), delayed matching-to-sample (Terry et al., Reference Terry, Buccafusco, Jackson, Prendergast, Fontana, Wong and Eglen1998), Morris water maze (Lelong, Dauphin, & Boulouard, Reference Lelong, Dauphin and Boulouard2001), delayed spontaneous alternation (Mohler et al., Reference Mohler, Shacham, Noiman, Lezoualc'h, Robert, Gastineau and Ragozzino2007) and olfactory association learning (Marchetti et al., Reference Marchetti, Chaillan, Dumuis, Bockaert, Soumireu-Mourat and Roman2004; Marchetti et al., Reference Marchetti, Jacquet, Escoffier, Miglioratti, Dumuis, Bockaert and Roman2011) tasks. The positive effect of 5-HT₄ receptor agonism on object recognition is also seen after chronic (14 day) activation of the receptor (Quiedeville et al., Reference Quiedeville, Boulouard, Hamidouche, Da Silva Costa-Aze, Nee, Rochais and Bouet2015). Critically, 5-HT₄ receptor antagonism has consistently been shown to block these agonist-induced pro-cognitive effects, confirming a specific role for the 5-HT₄ receptor in learning and memory (Fontana, Daniels, Wong, Clark, & Eglen, Reference Fontana, Daniels, Wong, Clark and Eglen1997; Lamirault & Simon, Reference Lamirault and Simon2001; Mohler et al., Reference Mohler, Shacham, Noiman, Lezoualc'h, Robert, Gastineau and Ragozzino2007; Moser et al., Reference Moser, Bergis, Jegham, Lochead, Duconseille, Terranova and Scatton2002). Whilst 5-HT₄ receptor activation has particularly strong effects on hippocampal-dependent learning and memory, it also facilitates other forms of cognition, including attention (as measured by the five choice serial reaction time task, Hille, Bate, Davis, and Gonzalez, Reference Hille, Bate, Davis and Gonzalez2008) and social memory (Letty et al., Reference Letty, Child, Dumuis, Pantaloni, Bockaert and Rondouin1997).

The timing of 5-HT₄ receptor activation seems to be critical to the profile of effects on memory processes, with a particularly clear role for the receptor in learning and memory acquisition (Hagena & Manahan-Vaughan, Reference Hagena and Manahan-Vaughan2017). Consistent with this, studies in which an agonist is administered prior to training have typically shown a facilitatory effect of 5-HT₄ receptor activation (Lamirault & Simon, Reference Lamirault and Simon2001; Meneses & Hong, Reference Meneses and Hong1997; Orsetti, Dellarole, Ferri, & Ghi, Reference Orsetti, Dellarole, Ferri and Ghi2003), whereas the effect of post-training agonist administration is less clear, with reports of both enhanced (Lamirault & Simon, Reference Lamirault and Simon2001; Orsetti et al., Reference Orsetti, Dellarole, Ferri and Ghi2003) and impaired (Meneses & Hong, Reference Meneses and Hong1997; Nasehi, Tabatabaie, Khakpai, & Zarrindast, Reference Nasehi, Tabatabaie, Khakpai and Zarrindast2015) memory consolidation.

Interestingly, the cognition enhancing effects of 5-HT₄ receptor activation have also been demonstrated in the chronic corticosterone model of anxiety/depression (Darcet, Gardier, David, & Guilloux, Reference Darcet, Gardier, David and Guilloux2016), suggesting that this may be a promising strategy for treating the cognitive impairments associated with stress-related disorders. Chronic administration of the 5-HT₄ receptor agonist RS67333 restored corticosterone-induced deficits on a broad range of tests of learning and memory, in contrast to chronic fluoxetine, which had a much narrower profile of effects. The 5-HT₄ receptor agonist RS67333 has also been shown to restore emotional memory performance in a passive avoidance test in the Flinders sensitive line rat model of depression (Eriksson et al., Reference Eriksson, Delagrange, Spedding, Popoli, Mathe, Ogren and Svenningsson2012). Together these findings suggest that 5-HT₄ receptor agonists may be a potential target for the improvement of the cognitive and emotional processing deficits associated with depression.

In contrast to the clear pro-cognitive effects of 5-HT₄ receptor activation, the effect of 5-HT₄ receptor antagonism is less well characterised and findings are mixed. Whilst the antagonists SDZ205557 and RS67532 have been shown to impair olfactory association memory formation and passive memory formation (Galeotti, Ghelardini, & Bartolini, Reference Galeotti, Ghelardini and Bartolini1998; Marchetti, Dumuis, Bockaert, Soumireu-Mourat, & Roman, Reference Marchetti, Dumuis, Bockaert, Soumireu-Mourat and Roman2000), other studies report no effect of 5-HT₄ receptor antagonism on memory processes (Moser et al., Reference Moser, Bergis, Jegham, Lochead, Duconseille, Terranova and Scatton2002; Orsetti et al., Reference Orsetti, Dellarole, Ferri and Ghi2003). It has been suggested that this may be due to a low level of behaviourally-mediated exogenous serotonin release, which would lessen the effects of receptor blockade compared with receptor activation (Orsetti et al., Reference Orsetti, Dellarole, Ferri and Ghi2003). Likewise, 5-HT₄ receptor knock-out mice do not have clear behavioural deficits in learning and memory (Segu et al., Reference Segu, Lecomte, Wolff, Santamaria, Hen, Dumuis and Compan2010). However, these animals do show an impairment in long-term spatial memory on the Morris water maze following scopolamine administration at a dose ineffective in wild type controls, suggesting that developmental adaptations in the cholinergic system may compensate for the absence of 5-HT₄ receptors under baseline conditions (Segu et al., Reference Segu, Lecomte, Wolff, Santamaria, Hen, Dumuis and Compan2010).

The memory enhancing properties of 5-HT₄ receptor activation are often attributed to a facilitation of central cholinergic activity. This is supported by studies demonstrating that 5-HT₄ receptor agonists enhance the brain ACh output under both baseline conditions (Consolo et al., Reference Consolo, Arnaboldi, Giorgi, Russi and Ladinsky1994) and during mnemonic demands (Mohler et al., Reference Mohler, Shacham, Noiman, Lezoualc'h, Robert, Gastineau and Ragozzino2007), and that 5-HT₄ receptor agonists reverse the cognitive impairments induced by muscarinic antagonists (Cachard-Chastel et al., Reference Cachard-Chastel, Devers, Sicsic, Langlois, Lezoualc'h, Gardier and Belzung2008; Fontana et al., Reference Fontana, Daniels, Wong, Clark and Eglen1997; Lo et al., Reference Lo, De Maeyer, Vermaercke, Callaerts-Vegh, Schuurkes and D'Hooge2014; Marchetti-Gauthier et al., Reference Marchetti-Gauthier, Roman, Dumuis, Bockaert and Soumireu-Mourat1997; Moser et al., Reference Moser, Bergis, Jegham, Lochead, Duconseille, Terranova and Scatton2002). For example, acute 5-HT₄ receptor activation reverses the impairments in spatial learning and memory that are induced by atropine and scopolamine on the Morris water maze (Fontana et al., Reference Fontana, Daniels, Wong, Clark and Eglen1997; Lo et al., Reference Lo, De Maeyer, Vermaercke, Callaerts-Vegh, Schuurkes and D'Hooge2014) and the deficits in passive avoidance behaviour that are induced by scopolamine on the passive avoidance test (Galeotti et al., Reference Galeotti, Ghelardini and Bartolini1998; Lo et al., Reference Lo, De Maeyer, Vermaercke, Callaerts-Vegh, Schuurkes and D'Hooge2014). More specifically, local infusion of the 5-HT₄ receptor agonist RS67333 into the NbM enhances the acquisition and consolidation of place recognition memory, which has been suggested to be due to a potentiation of the activity of the cholinergic NbM-cortical pathways (Orsetti et al., Reference Orsetti, Dellarole, Ferri and Ghi2003; although see Penas-Cazorla & Vilaro, Reference Penas-Cazorla and Vilaro2015). Further support for the idea that the memory-enhancing effects of 5-HT₄ receptor activation are mediated via an interaction with cholinergic systems comes from studies that demonstrate that acetylcholinesterase inhibitors (AChEI) potentiate the effects of 5-HT₄ agonists on cognition (Cachard-Chastel et al., Reference Cachard-Chastel, Lezoualc'h, Dewachter, Deloménie, Croes, Devijver and Gardier2007; Lamirault, Guillou, Thal, & Simon, Reference Lamirault, Guillou, Thal and Simon2003; Mohler et al., Reference Mohler, Shacham, Noiman, Lezoualc'h, Robert, Gastineau and Ragozzino2007; Moser et al., Reference Moser, Bergis, Jegham, Lochead, Duconseille, Terranova and Scatton2002). For example, sub-efficacious doses of the 5-HT₄ receptor agonist prucalopride and the AChEI donepezil reverse scopolamine-induced deficits in spatial learning and memory when given in combination (Cachard-Chastel et al., Reference Cachard-Chastel, Lezoualc'h, Dewachter, Deloménie, Croes, Devijver and Gardier2007).

Establishing the effects of 5-HT₄ receptor activation in humans

Taken together, this preclinical evidence suggests that the 5-HT₄ receptor may be a promising target for the treatment of depression and cognitive disorder in humans. However, there have been relatively few studies in humans to establish the translation of these effects, and to characterise the profile of 5-HT₄ receptor activation on human cognition and emotion.

There is some initial evidence to support a role for the 5-HT₄ receptor in the neurobiology of depression. Healthy individuals with a family history of depression have been shown to have lower 5-HT₄ receptor binding in the striatum, and this reduction is most pronounced in people with more than one first degree relative with a history of depression (Madsen et al., Reference Madsen, Torstensen, Holst, Haahr, Knorr, Frokjaer and Knudsen2014). Conversely, an increase in the density and functionality of 5-HT₄ receptors in cortical and striatal areas has been reported in an analysis of post-mortem brains from depressed violent suicide victims (Rosel et al., Reference Rosel, Arranz, Urretavizcaya, Oros, San and Navarro2004). Polymorphisms in the splice variant region of the gene encoding the 5-HT₄ receptor have been reported to be associated with unipolar depression (Ohtsuki et al., Reference Ohtsuki, Ishiguro, Detera-Wadleigh, Toyota, Shimizu, Yamada and Arinami2002), although this was not confirmed in a recent genome wide association study (Wray et al., Reference Wray, Ripke, Mattheisen, Trzaskowski, Byrne, Abdellaoui and Consortium2018). An involvement of the 5-HT₄ receptor in human memory is also supported by a PET study that reported an inverse correlation of 5-HT₄ receptor binding in the hippocampus and performance on the Auditory Verbal Learning Task (Haahr et al., Reference Haahr, Fisher, Holst, Madsen, Jensen, Marner and Hasselbalch2013). Interestingly, the direction of this effect is opposite to the memory enhancing effects of 5-HT₄ receptor agonism seen in preclinical studies, which may be due to the binding potential representing a composite measure of receptor density and affinity and/or that lower 5-HT₄ receptor binding could also be reflective of higher chronic endogenous serotonin levels (Haahr et al., Reference Haahr, Fisher, Jensen, Frokjaer, Mahon, Madsen and Knudsen2014).

These divergent correlational findings highlight the importance of further investigation in humans using an experimental medicine approach. Establishing the cognitive effects of compounds targeting the 5-HT₄ receptor is particularly important in light of the limited predictive validity of animal models and the disappointing failure to predict the efficacy of pro-cognitive and antidepressant drugs in humans from drug discovery studies in rodents (Hyman, Reference Hyman2012; Millan et al., Reference Millan, Agid, Brune, Bullmore, Carter, Clayton and Young2012). We have previously demonstrated that acute administration of clinically active antidepressants in healthy volunteers and depressed patients produces changes in information processing that can be measured using behavioural and neuroimaging paradigms (Godlewska & Harmer, Reference Godlewska and Harmer2020). Importantly, these changes are predictive of later therapeutic effects and these measures can therefore be used as surrogate markers of clinical effects to screen and understand the mechanisms of candidate agents early in development. Testing the effects of compounds targeting the 5HT₄ receptor in humans using this approach could help clarify the role of the 5HT₄ system in cognition and emotion and translate the potentially promising results from preclinical models to clinical application.

Experimental medicine often takes a stepwise approach, first testing the effects of the compound in healthy participants to gain information on relevant dosing, duration of treatment and specific cognitive and emotional processes which are sensitive to this manipulation. This information can then be used to guide the most appropriate design for testing in clinical groups including those with depression, anxiety and cognitive impairment. Again, experimental medicine markers (such as measures of learning or emotional processing) can be useful prior to clinical testing of efficacy (symptom improvement) to characterise likely mechanisms, target and patient groups. Such a stepwise approach may improve decision making about a compound's development and reduce the high failure rate associated with drug development in this area (Wong, Siah, & Lo, Reference Wong, Siah and Lo2019).

Until recently, further elucidation of the role of the 5-HT₄ receptor in humans has been difficult because of the lack of suitable compounds to manipulate 5-HT₄ receptor function. The development of 5-HT₄ receptor agonists for use as antidepressants in humans is complicated by the important role played by the 5-HT₄ receptor in the periphery, in particular in the gastrointestinal tract and the heart, and the potential for peripheral side effects. Indeed, two 5-HT₄ agonists, cisapride and tegaserod, that were developed for the treatment of gastrointestinal problems have both been restricted in use or withdrawn due to concerns about cardiac side effects. However, the particular concern with these drugs was risk of ventricular arrhythmias, which is probably due to their interaction with the human ether-a-go-go related gene potassium channel (De Maeyer, Lefebvre, & Schuurkes, Reference De Maeyer, Lefebvre and Schuurkes2008). Newer, more selective partial agonists, such as prucalopride which is licensed for the treatment of constipation, have good cardiac safety profiles. Prucalopride has good brain penetration (Johnson et al., Reference Johnson, Drummond, Grimwood, Sawant-Basak, Miller, Tseng and Siok2012), has pro-cognitive and antidepressant-like effects in animal models (Cachard-Chastel et al., Reference Cachard-Chastel, Lezoualc'h, Dewachter, Deloménie, Croes, Devijver and Gardier2007; Lucas et al., Reference Lucas, Rymar, Du, Mnie-Filali, Bisgaard, Manta and Debonnel2007) and therefore offers a unique opportunity to investigate the role of the 5-HT₄ receptor in human cognition.

We recently explored the effect of a single dose of prucalopride in healthy volunteers and found consistent pro-cognitive effects across measures of learning memory including declarative memory, reward learning and emotional encoding (Murphy, Wright, Browning, Cowen, & Harmer, Reference Murphy, Wright, Browning, Cowen and Harmer2019). These findings replicate the findings in preclinical studies and support further clinical assessment of these compounds in depression and for disorders involving memory impairment. In line with data in animal models that 5HT₄ agonists may be beneficial in depression, and also in combination with an SSRI, we are currently exploring the effects of a 5HT₄ partial agonist (PF-04995274) on emotional and cognitive function in unmedicated depressed patients (NCT03516604) and patients who have not showed a sufficient response to current SSRI treatment (NCT03515733). This kind of experimental medicine approach can provide early proof-of-concept evidence to support larger clinical trials, and may have value in selecting the best treatment, for the right subgroup of depressed patients, prior to full randomised clinical trials of efficacy.