Alzheimer's disease (AD) is a chronic degenerative disease, affecting primarily limbic, paralimbic and neocortical structures, characterized by progressive cognitive decline, behavioral impairment and ultimately death. Neurofibrillary tangles and β-amyloid (Aβ) plaques are the pathological hallmarks of AD, which is also characterized by neuronal degeneration of the cholinergic system in brain regions involved in cognitive functions (Kang et al., 1987). There is evidence that intracerebroventricular (i.c.v.) infusion of Aβ causes brain dysfunctions similar to those of AD, as evidenced by neurodegeneration and impairment of learning and memory in rodents (Maurice et al. 1996; Mazzola et al. 2003; Yamada et al. 1999). Neuronal degeneration induced by Aβ affects subcortical nuclei modulating various physiological processes and behaviors, such us attention and vigilance, mood and aggression, learning and memory (Lyness et al., 2003). Various neurotransmitters are involved in the synaptic connections of these nuclei, including acetylcholine (ACh), norepinephrine (NA), dopamine (DA) and serotonin (5-HT) (Curcio and Kemper 1984; Yamamoto and Hirano 1985). It is accepted that the ACh system alterations are among the most relevant neurochemical changes in AD, which are related with a cognitive impairment (Bierer et al., 1995). In order to evaluate the importance of ACh deficit in AD, most experimental models of AD are based on surgical or chemical lesions of the basal forebrain, followed by impairments of the ACh pathways and of performance in learning and memory tasks (Cummings, 2000). Moreover, in addition to the ACh deficit, disruption of several other neurotransmitter systems has been reported in AD, including NA, DA and 5-HT systems (Gottfries 1990; Meltzer et al. 1998).#

The 5-HT system plays an important role in multiple brain disorders and serotonin 5-HT receptors have been associated with changes in certain behavioral functions, such as in locomotion, feeding, aggression, social behavior and anxiety (Barnes and Sharp, 1999). A growing body of evidence suggests that the 5-HT system plays a role in cognitive processes particularly in learning and memory. Particularly, serotonin 5-HT1A/1B/1D/1E/1F, 5-HT2A/2B/2C, 5-HT3/3B, 5-HT4, 5-HT5A/5B, 5-HT6 and 5-HT7 receptors show a regional and cellular distribution within the central nervous system (CNS) in brain areas associated to learning and memory processes. Moreover, different markers of 5-HT neurotransmission, including 5-HT receptors are affected by aging and AD (Altman and Normile 1987; Buhot et al. 2000; Meneses 1999 2002 2004; Perez–Garcia and Meneses 2005).#

Actually, neurochemical and behavioral studies as well as on brain distribution support a major role for the serotonin 5-HT4 receptor subtype in learning and memory. The serotonin 5-HT4 receptor is a G-protein-coupled, 7-transmembrane domain protein, positively linked to the activation of adenylate cyclase (Hoyer et al., 2002). The highest densities of serotonin 5-HT4 receptor within the CNS are located in limbic regions, including hippocampus, frontal cortex and amygdale, regions of the brain related to cognitive functions (Eglen et al. 1995; Medhurst et al. 2001; Waeber et al. 1993). Interestingly, to further support the potential role of serotonin 5-HT4 receptors in the cognitive processing, there is evidence for a loss of these receptors in the cortex and the hippocampus of human brains with AD (Reynolds et al., 1995). Recently, Spencer et al. (2004) and Manuel-Apolinar et al. (2005) have shown that serotonin 5-HT4 receptors remain functional in the presence of excess of Aβ peptide and they are subjected to expression changes in diverse brain areas during memory consolidation. Several studies indicate that various serotonin 5-HT4 receptor agonists ameliorate cognitive deficits, such as that observed during AD (Bockaert et al. 2004; Langlois and Fischmeister 2003; Letty et al. 1997; Orsetti et al. 2003). To support the hypothesis of a role for serotonin 5-HT4 receptors, these effects were counteracted by selective serotonin 5-HT4 antagonists, such as GR125487 and RS67532 (Fontana et al. 1987; Galeotti et al. 1998; Lelong et al. 2003; Letty et al. 1997).#

SL65.0155 is a benzodioxanoxadiazolone compound with high affinity and good selectivity for human serotonin 5-HT4 receptors. Unlike other 5-HT4 agonists, this compound lacks cardiovascular and gastrointestinal effects, in line with its antagonistic activity in isolated peripheral tissues, while acts at central serotonin 5-HT4 receptors as an agonist, improving learning and memory performance. Furthermore, this compound and the acetyl-cholinesterase inhibitor rivastigmine have a synergic effect in cognitive performance of aged rats, indicating that serotonin 5-HT4 receptor activation could be useful as add-on therapy in AD (Moser et al., 2002).#

This prompted us to further study the possible pro-cognitive effect of the selective serotonin 5-HT4 partial agonist, SL65.0155 in rodents subjected to several models of amnesia and tested in different cognitive tasks.#

As described in Fig. 1, the i.c.v. injection of BAP 1–42 (400 pmol/mouse) 14 days prior to the learning trial induced a reduction of latency to re-enter the dark box in comparison to i.c.v. injection of the vehicle alone (control vs. BAP vehicle-treated group). This effect was observed in both retention tests, made 1 [F(3,24)=57.52; P<0.001] and 7 [F(3,24)=4.089; P<0.001] days after the learning trial. Repeated administration of the serotonin 5-HT4 partial agonist, SL65.0155 (1 mg/kg/day) for 7 days prior to the learning trial, inhibited the amnesic effect of BAP 1–42 by increasing the latency to re-enter the dark box in the first [F(3,24)=57.52; P<0.001] and second [F(3,24)=4.089; P<0.001] retention test. Furthermore, the injection of SL65.0155 per se improved memory capacity of control mice in passive avoidance paradigm [F(3,24)=57.52; P<0.001 and F(3,24)=4.089; P<0.001].#

Intracerebroventricular injection of GAL 1–29 (3 μg/mouse) 15 min prior to the learning trial significantly shortened step-through latency of passive avoidance response in comparison to vehicle injected control mice (control vs. GAL vehicle-treated group). This effect was observed in both retention tests, made 1 [F(3,35)=46.22; P<0.001] and 7 [F(3,35)=35.96; P<0.05] days after the learning trial. The amnesic effect of GAL 1–29 was counteracted by repeated administration of SL65.0155 (1 mg/kg/day) for 7 days prior to the learning trial. This effect, represented by an increased latency to re-enter the dark box, was observed in the first [F(3,24)=57.52; P<0.001] and in the second retention test of passive avoidance paradigm. A significantly higher memory retention in the first and second retention test of a passive avoidance paradigm was also observed in vehicle-pretreated mice injected with SL65.0155 [F(3,35=46.22; P<0.001; F(3,35)=35.96; P<0.05] (Fig. 2).#

Fig. 3 shows that exposure to CO impaired memory capacity of mice tested in passive avoidance paradigm, as indicated by the decreased latency to re-enter the apparatus in the first and second retention test made 1 [F(3,35)=63.56; P<0.001] and 7 [F(3,35)=32.90; P<0.001] days after the learning trial. Treatment with SL65.0155 (1 mg/kg/day) for 7 days attenuated the decrease of latency time in both retention tests [F(3,35)=63.56; P<0.001] and 7 [F(3,35)=32.90; P<0.001] in mice exposed to CO. Furthermore, the injection of SL65.0155 per se increased memory retention of control mice in passive avoidance paradigm [F(3,35)=63.56; P<0.001and F(3,35)=32.90; P<0.001].#

The results concerning the influence of the serotonin 5-HT4 partial agonist on behavioral performance of animals with NBM lesions and tested in the active avoidance task, are shown in Table 1. Animals with NBM lesions exhibited a significant decrease in learning and memory capacity, as indicated by the reduction in the number of CARs [F(3,76)=68.30 P<0.0001] and the percent number of learners in the shuttle-box in comparison to control group (intact animals treated with vehicle). The repeated treatment (7 days) with SL65.0155 (1 mg/kg/day) increased memory retention both of animals with NBM lesions and the control group.#

Rats exposed prenatally to MAM and treated with vehicle showed a marked deficit in the acquisition of radial maze behavior compared to animals exposed to saline and treated with vehicle (Fig. 4). They exhibited, from day 2 through day 7, a slower learning of the task with an increased number of errors per trial. The cognitive impairment in prenatally MAM-exposed rats was counteracted by repeated administration of SL65.0155 (1 mg/kg/day) as demonstrated by the decreased number of errors made from day 2 through day 7 of the task. The cognitive-enhancing effect of the compound was also found both in prenatally saline-exposed group and in intact animals. They showed a better cognitive performance from day 2 through day 7 of the task in comparison to all vehicle-treated groups [F(2,216)=6.95; P<0.0001 for drug effect and F(5,216)=3.61; P<0.0001 and day effect].#

The present study shows that repeated administration of SL65.0155, a selective serotonin 5-HT4 receptor partial agonist, is followed by enhanced cognitive capacity in rodents as it improved animals' performance in different experimental models of learning and memory deficit. Consistently with these data, Moser et al. (2002) have shown that SL65.0155 reverses amnesia in animal behavioral tasks with no peripheral side effects or receptor desensitization. Because of the high density of serotonin 5-HT4 receptors in specific brain regions, several studies have focused on the role of these receptors in learning and memory processes (Medhurst et al. 2001; Waeber et al. 1993). Pharmacological evidence has been provided that serotonin 5-HT4 receptor ligands may affect learning and memory processes (Bockaert et al. 2004; Meneses 1999). For instance, in animal models of amnesia induced by scopolamine or atropine, the serotonin 5-HT4 receptor agonists BIMU-1, BIMU-8, RS17017 and RS67333 reverse working memory deficits associated to lower cholinergic activity. A specific role for serotonin 5-HT4 receptors in these studies was supported by the antagonism exerted by selective serotonin 5-HT4 antagonists, such as GR125487 and RS67532 (Fontana et al. 1987; Galeotti et al. 1998; Lelong et al. 2003; Letty et al. 1997).#

The mechanisms underlying the cognitive-enhancing effect of SL65.0155 are unknown, but they could involve the localization and the expression changes of serotonin 5-HT4 receptors, and the excitability and neurotransmitter release in serotonergic neurons. The distribution of serotonin 5-HT4 receptors has been described in the brain of mouse, rat, guinea-pig, monkey and humans, particularly in limbic structures, hippocampus, basal ganglia, i.e. predominantly in areas related to cognitive processes where they are subjected to expression changes during memory formation (Bockaert et al., 2004; Manuel-Apolinar et al., 2005). Within the hippocampus, serotonin 5-HT4 receptors are located on CA1 pyramidal neurons and their activation results in an increased content of cyclic AMP (cAMP) with activation of the protein kinase A (PKA) pathway and of neuronal excitability (Roychowdhury et al. 1994; Torres et al. 1995). This is supported by electrophysiological studies demonstrating that serotonin 5-HT4 receptor agonists enhance population spike amplitude and tetanus-induced long-term potentiation (LTP), which are thought to be a form of the basic cellular mechanisms correlated to learning and memory (Matsumoto et al. 2001; Spencer et al. 2004). Interestingly, these effects are blocked either by the selective serotonin 5-HT4 receptors antagonist GR 113808 or by scopolomine, suggesting a functional interaction between the 5-HT system mediated via serotonin 5-HT4 receptors and the ACh system in cognitive processes. This possibility has been confirmed by microdialysis studies showing increased extracellular levels of ACh in the hippocampus and frontal cortex of rats after administration of serotonin 5-HT4 receptors agonists (Bockaert et al. 2004; Matsumoto et al. 2001).#

Various experimental models of amnesia can provide with insight the neurobiological bases and pathogenic mechanism of cognitive changes commonly observed in AD. Given the apparent central importance of Aβ deposition in the pathogenesis of AD, different animal models of Aβ toxicity have been developed. There is evidence that central administration of Aβ mimics some of the pathological processes of AD, as evidenced by neurodegeneration and impairment of learning and memory in rodents (Maurice et al. 1996; Mazzola et al. 2003). Furthermore, since familiar forms of AD are associated with mutations in the Aβ precursor protein (APP) and presenilin genes (PS-1) (Selkoe and Podlisny, 2002), over-expression of mutant human transgenes in mice has therefore been utilized to generate potential animal models of AD. The over-expressing transgenic mice typically show cerebral plaque-like Aβ deposits with associated cognitive impairment (Arendash et al. 2001; Howlett et al. 2004). Interestingly, serotonin 5-HT4 receptor activation stimulates the release of the soluble non-amyloidogenic peptide fragment (sAPPα) from the APP, a process which may reduce production of toxic Aβ during AD, suggesting that serotonin 5-HT4 receptor activation may have disease-modifying properties in addition to its use as a symptomatic treatment of AD (Lezoualc'h and Robert, 2003). Furthermore, Spencer et al. (2004) have shown that serotonin 5-HT4 receptors remain functional in a transgenic mouse line that over-produces Aβ. Interestingly, in line with the correlation between serotonin 5-HT4 receptors functionality and Aβ deposits, we have shown that repeated administration of SL65.0155 reversed the amnesia induced by BAP 1–42 in mice tested in a step-through passive avoidance paradigm. Probably, this effect is also linked to the increased release of ACh induced by SL65.0155.#

GAL is a 29 amino acid neuropeptide that, when administered into the lateral ventricles or hippocampus, inhibits ACh release and produces deficits in acquisition or working memory of rodents tested in several learning and memory tasks (Crawley, 1996). GAL also acts inhibiting the activation of adenylate cyclase necessary to stimulate the cAMP-dependent PKA pathway, a critical intracellular second messenger necessary for long-term memory consolidation (Kinney et al., 2003). Furthermore, it is over-expressed in the basal forebrain of patients with AD, supporting the emerging hypothesis that cognitive impairments typical of AD may be in part due to the inhibitory action of GAL (Bowser et al. 1997; Hökfelt et al. 1987; Mufson et al. 1993; Steiner et al. 2001). This assumption is in agreement with the present results showing that repeated administration of SL65.0155 is capable of reversing the amnesia induced by GAL, possibly acting on ACh release and on the PKA pathways.#

SL65.0155 was also effective against the cognitive deficits induced in rats by CO exposure, by pretreatment with MAM or by lesions of the NBM. It has been reported that CO exposure produces a rapid impairment of learning associated to a marked dysfunction of cholinergic neurons in the frontal cortex, striatum and hippocampus; providing a good model for the investigation of memory deterioration (Nabeshima et al. 1991; Yang and Zhou 2001). MAM is a neuro-specific antimitotic agent that, injected at appropriate gestational days, impairs spatial and working memory with a reduction of the hippocampal levels of nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) (Cattabeni and Di Luca 1997; Fiore et al. 1999 2000; Le Pen et al. 2000; Wood et al. 2003). Since NGF and BDNF are known to regulate the choline acetyltransferase (ChAT) activity in brain neurons of the septum, nucleus of Meynert and striatum, prenatal disruption of ACh neuron development could be a risk factor for AD (Hellweg et al. 1998; Rylett and Williams 1994). The ACh neuronal system which projects to the fronto-parietal cortex from NBM, an ACh neuronal cell body-rich region, as well as the ACh neurons projecting to the hippocampus from the medial septal area, are functionally and neuropathologically damaged in AD (Bartus et al. 1982; Coyle et al. 1983). Experimental lesions of the basal forebrain cholinergic systems also produce memory deficits in animals that appear to reflect the symptoms of AD (Dekker et al. 1992; Mandel et al. 1989).#

The loss of ACh neurons is one of the major neurochemical deficits in AD, correlated with a loss of cognitive function and the current therapeutic approach is focused to stimulate the ACh system. Since cholinergic agents, such as the inhibitors of acetyl-cholinesterase (i.e. donepezil, rivastigmine and galantamine) have shown only limited clinical efficacy in AD due to the multifactorial nature of the disease and poorly understood etiology of it, novel therapeutic approaches are currently under intense investigation. Besides the ACh system, loss of other neurotransmitter systems including NA, DA and 5-HT, has been reported in AD, suggesting alterative therapeutic intervention either alone or in combination with compounds acting on the cholinergic system (Palmer, 1996). It is well-known that the 5-HT system plays a role in the learning and memory processes. Pharmacological studies on multiple 5-HT receptor subtypes have provided evidence of the link between these receptors and learning and memory processes, in line with the presence of 5-HT neurons in brain areas involved in these processes (Buhot et al. 2000; Meneses 1999). Various pre- and postsynaptic 5-HT markers are known to be altered in patients with AD: they include marked loss of dorsal and median raphe neurons (Chen et al., 2000), reduced concentrations of 5-HT and metabolites (Cross et al., 1983), altered binding to 5-HT re-uptake sites (Chen et al. 1996; Meneses 1999) and decreased density of serotonin 5-HT4 receptors (Lai et al., 2003). Thus, cognitive impairment in AD may be related to a general dysfunction of the serotonergic system. In fact, behavioral disturbances such as depression, apathy and anxiety coexist with cognitive symptoms in AD and selective serotonin reuptake inhibitors (SSRI) are effective in the treatment of both behavioral symptoms (e.g. depression) and cognitive dysfunctions in AD patients (Meltzer et al. 1998; Meneses 1999). Furthermore, preclinical studies have shown that SSRI, acting on multiple 5-HT receptors, enhance learning consolidation or memory formation in learning tasks such as avoidance and they are also able to reverse the impairment induced by scopolamine (Altman and Normile 1987; Flood and Cherkin 1987; Meneses and Hong 1995).#

In conclusion, the present findings provide further support to the cognitive-enhancing activity induced by drugs acting on serotonin 5-HT4 receptors, such as SL65.0155 across a range of tasks in rodents, underlying the involvement of serotonin 5-HT4 receptors in learning and memory processes. The cognitive-enhancing activity of SL65.0155 was observed in several tests assessing different areas of cognitive performance and against a variety of deficits induced by different models of amnesia, used to mimic the cognitive deficits characterizing AD. For instance, rats prenatally exposed to MAM demonstrated a marked deficit of working memory, which was reversed by SL65.0155. Thus, this compound may be active on working memory deficit, that is considered as a prominent neuropsychological feature of AD. In active and passive avoidance tests, the compound has counteracted the amnesia of rodents subjected to different experimental models, showing to improve a variety of cognitive deficits. Furthermore, the synergic action of SL65.0155 with rivastigmine (Moser et al., 2002) suggests the possibility of enhancing the modest improvements in cognitive function reported with this acetyl-cholinesterase inhibitors. Thus, the present data support the therapeutic potential of SL65.0155 in the treatment of cognitive disturbances in AD.#