5. CB2 Receptors and Neuropsychiatric Disorders
5.1. Stress and Anxiety
The role of the endocannabinoid system in mediating fear, stress and anxiety has been researched extensively over the past decade (for recent reviews see [133-135]). In general, CB1 receptor activation elicits complex bi-phasic effects on stress-responding and recent evidence indicates that this may in part be due to differential activation of CB1 on forebrain glutamatergic and GABAergic neurons that elicit anxiolytic and anxiogenic effects respectively [136]. As mentioned in earlier sections, several studies have demonstrated a lack of psychoactive effects such as catalepsy, hypolocomotion and hypothermia following pharmacological modulation of the CB2 receptor [98,99,137] and, as a consequence, the role of CB2 receptors in regulating stress and anxiety has received little attention. The data reviewed above in Section 2 demonstrating expression of CB2 receptors in key brain areas involved in modulating the stress response including the amygdala, hippocampus, prefrontal cortex and hypothalamus [56,60,70,71,138], suggest that the potential role of CB2 receptors in regulation of emotional responding is at least worthy of investigation. To date, most studies examining the effect of selective CB2 receptor agonists on CNS function have examined the ability of ligands to modulate locomotor activity, with few studies examining effects in validated models of emotionality/anxiety. Unconditioned responding in stressful environments or conditioned responses to a previously learned aversive stimulus are the most commonly used means of assessing anxiety-related behaviour in animals. Onaivi and colleagues have assessed stress-induced anxiety-related behaviour in the two-compartment black and white test (also known as the light-dark test) and in the elevated plus maze following the administration of CB2 receptor ligands [60,91]. Acute systemic administration of JWH-015 (1-20 mg/kg) dose dependantly induced an anxiogenic response in the black and white box, with females slightly more sensitive than males [60,139]. In contrast, JWH-015 (20 mg/kg) attenuated stress-induced gender-specific aversion to the open arms of the elevated plus maze [139] and administration of the CB2 receptor agonist GW405833 (100 mg/kg) induced anxiolytic effects in the marble burying test [97]. However, it should be noted that the behavioural effects of CB2 receptor stimulation in these studies was accompanied by reduced locomotor activity at the doses used [60,97,139], and administration of a CB2 receptor antagonist failed to reverse either the locomotor depressant or the anxiolytic effects observed in the marble burying test [95], which may have important implications for the interpretation of the effects observed. Chronic administration of JWH-015 results in an anxiolytic behavioural profile in the black and white box [60] and reduces stereotypic behaviour in non-stressed but not stressed BALB/c mice [139], which the authors interpret as an anxiolytic profile. However, it should be noted that stereotypic behaviour represents locomotor activity and rearing and, as such, may be more accurately interpreted as a measure of general activity rather than a measure of anxiety-like behaviour.
In contrast to the effects observed following CB2 receptor stimulation, CB2 receptor antagonism using SR144528 had little or no effect in the black and white box, with the exception of a decrease in time spent in the white chamber in DBA/2 male mice at 20 mg/kg SR144528 [91], again possibly a result of reduced locomotor activity observed at this dose. Repeated (3 day) intracerebroventricular administration of CB2 antisense oligonucleotide increased the amount of time spent on the open arms
of the elevated plus maze indicative of an anxiolytic-like effect [61,139], however, the effect of this treatment on locomotor activity in the maze was not reported. Only one published study to date has examined the role of brain CB2 receptors in conditioned aversion/learning, demonstrating that infusion of JWH-015 or PEA into the CA1 region of the hippocampus does not affect novel object recognition or long-term memory retention [140]. Overall, the results obtained from behavioural studies of the role of CB2 receptors in modulating the response to aversion are far from clear, highlighting the need for further studies examining the effects of selective deletion, blockade or stimulation of brain CB2 receptors on the regulation of emotional responding.
A critical component of the stress response is the activation of the hypothalamic-pituitary-adrenal (HPA) axis and the subsequent increase in glucocorticoid levels and several studies have examined the role of the endocannabinoid system in mediating this response [141]. CB2 receptor mRNA and protein have been identified in the brain regions that modulate HPA axis activity including the hippocampus, amygdala and hypothalamus [56,71], while in comparison, CB2 receptors are not expressed in the adrenal cortex [142], indicating that should CB2 ligands modulate the neuroendocrine response to stress, this would most likely occur at the level of the brain. Our studies have indicated that endotoxin-induced increases in circulating corticosterone levels are not altered by administration of the non-selective CB1/2 receptor agonist HU-210 or the CB2 receptor antagonist/inverse agonist SR144528 [125]. Further studies will be required to determine whether CB2 receptors are involved in the endocannabinoid-mediated modulation of neuroendocrine activity under basal and stress conditions.
Although Onaivi and colleagues demonstrated similar effects of the CB2 receptor ligand JWH-015 on locomotor activity and stereotypic behaviour in three strains of mice, namely C57Bl6/J, DBA/2 and BABL/c, they did not compare behavioural responses between these different strains in tests of emotionality and anxiety [139]. BALB/c mice have been proposed as a model of anxiety and are regarded as more stress-sensitive than other mouse strains including C57Bl6/J and DBA/2 mice [143-145]. For example, BALB/c mice exhibit an anxiogenic profile in the black and white box, manifested as a reduced time in the white compartment of the test apparatus when compared to C57Bl6/J counterparts [145]. Therefore, due to the contribution of genetic background to stress responding, it is possible that CB2 ligands may elicit differential effects depending on the strain of mouse used. In accordance, low doses of JWH-015 (5 mg/kg) increased stereotypic behaviour in male DBA/2 but not C57Bl6/J or BALB/C mice [60]. No significant difference in whole brain CB2 receptor expression prior to or following chronic stress was observed between the three different mouse strains [139]. However, it remains to be determined if the density or function of CB2 receptors or other components of the endocannabinoid system are differentially altered between these different strains of mice in brain regions associated with stress responding. Accordingly, early life stress such as maternal deprivation induces an anxiogenic behavioural phenotype [146] which has recently been associated with enhanced CB2 receptor expression in the hippocampus [70]. Thus, greater understanding of the role of CB2 receptors in pathological anxiety states will be reached by examining the expression, distribution and functional activity of these receptors in preclinical models that exhibit ethological validity.
Pharmaceuticals 2010, 3 2528
5.2. Depression
The involvement of the endocannabinoid system in the regulation of mood and affective responding has received increasing interest in the past few years (for recent reviews see [147-149]), particularly in light of the withdrawal of rimonabant (CB1 receptor antagonist/inverse agonist) as an anti-obesity agent due to the increased risk of psychiatric side effects, including depression. The involvement of CB1 receptors in regulating mood is further supported by the upregulation of CB1 receptor expression and function in the prefrontal cortex of depressed suicide victims [150] and enhanced CB1 receptor density in the prefrontal cortex of alcoholic suicide victims compared with alcoholic controls [151]. Furthermore, allele variations in the CB1 receptor gene CNR1, plays a role in the antidepressant response in major depressed patients [152] and a cohort of elderly depressed Parkinsonian patients have demonstrated a polymorphism (AATn) of the CNR1 [153]. Preclinical evidence further supports the involvement of CB1 receptors in depressive-like behaviour [154,155] and suggests that the effects of currently used antidepressants might depend upon endocannabinoid system modification. For example, endocannabinoid-CB1 receptor signaling in the brain is altered by several interventions that elicit antidepressant activity in humans, including chronic tricyclic antidepressant treatment [156,157], repeated electroconvulsive therapy [158] and sleep deprivation [159]. In contrast, the involvement of the CB2 receptor in affective responding has not received similar attention. A recent study has, however, demonstrated an association between depression and a polymorphism in the CB2 receptor gene at position Q63R in Japanese patients [69]. Polymorphisms in the Q63R in the CB2 gene have also been linked with eating disorders, alcoholism, osteoporosis, autoimmune disease and schizophrenia [93,102,160,161], many of which often demonstrate co-morbidity with depressive illness. Although it remains to be determined if this genetic link exists across other ethnic groups, it is possible that genetic variation in the CB2 gene may be a predisposing factor in the development of depression. CB2 receptor mediated effects of cannabinoid agonists such as WIN55,212-2 and 2-AG are reduced in the presence of the Q63R polymorphism [160,162] and serum levels of 2-AG and anandamide are reduced in patients with major depression, an effect directly correlated with the duration of the depressive episode [163]. Although further studies are required, in addition to alterations in endocannabinoid levels and CB1 receptor signaling, central CB2 receptor expression or function may be altered in depressed patients.
The association between adverse life events and the development of depression is widely acknowledged, and, as a consequence, many of the preclinical models of depression or antidepressant-like activity exploit this link by examining behavioural and physiological responses to stress. One of the most widely used behavioural tests for antidepressant-like activity is the forced swim test, where rodents exposed to a confined swim arena will initially attempt to escape but after some time will assume a floating posture (immobility) that is thought to be related to a state of behavioural despair [164]. Antidepressant agents increase escape behaviour thereby reducing the duration of immobility, while in comparison, an increase in immobility is regarded as a depressive-like state. Antidepressant-like activity of cannabinoid ligands and endocannabinoid modulators such as FAAH and anandamide reuptake inhibitors, have been demonstrated in the forced swim test [131,155,165-171]. In many of the studies, the antidepressant-like effects have primarily been attributed to activity at the CB1 receptor, confirmed using receptor antagonists and/or genetic deletion. Recent evidence has indicated that CB1
Pharmaceuticals 2010, 3 2529
receptors on subpopulations of glutamatergic, but not GABAergic, neurons appear to mediate the forced swim stress-induced behavioural and neuroendocrine effects [172]. However, non-selective CB1/CB2 receptor agonists and endocannabinoid modulators may also activate CB2 receptors. Antidepressant-like activity following chronic CB1/CB2 receptor stimulation using HU210, but not CB1 receptor agonism (AM281) alone, was observed in the forced swim test [131]. In addition, intra-hippocampal administration of HU210 induced an antidepressant-like effect in the forced swim test, an effect only partially attenuated by pharmacological blockade of the CB1 receptor [169]. Although further studies are required in order to further determine the mechanism of action underpinning these effects of non-selective CB1/CB2 agonists, it is possible that activation of central CB2 receptors may co-operatively augment the effects of CB1 receptor activation on stress-induced behaviour and monoaminergic function. Presently, only one study has been published directly assessing the effect of pharmacological CB2 agonism in the forced swim test, reporting that acute administration of the CB2 agonist GW405833 did not alter time spent immobile in the forced swim test [173]. However, in the presence of neuropathic pain, CB2 receptor activation reduced nociceptive responding to a mechanical stimulus while concurrently attenuating the enhanced immobility observed in the forced swim test [173]. The present results indicate that CB2 receptor agonism may alleviate neuropathic pain and the co-morbid depressive symptoms that often accompany this disorder. In an interesting study published recently, García-Gutiérrez and colleagues employed both pharmacological and genetic approaches to investigate the role of the CB2 receptor in depressive-like behaviour [138]. Transgenic mice engineered to over-express the CB2 receptor (including over-expression in key brain regions implicated in depression) exhibited reduced depressive-like behaviour in the tail suspension test and in a novelty suppressed feeding test, compared with wildtype controls. However, acute intraperitoneal administration of the CB2 receptor antagonist AM630, at doses that had no effect on locomotor activity, had an antidepressant-like effect in the forced swim test in wildtype mice, but not in CB2 over-expressing mice [138]. Though the depression-resistant endophenotype associated with CB2 over-expression and the antidepressant-like effects of CB2 receptor blockade pose an apparent discrepancy which is difficult to reconcile, these data do at least demonstrate an important role for the CB2 receptor in regulating depressive state in mice. More studies examining the effects of direct activation/inhibition of brain CB2 receptors are required in order to determine conclusively the role this receptor plays a role in stress-induced behavioural changes.
A detailed understanding of the neurobiological underpinnings of depression requires pre-clinical models that mimic the neurological and physiological alterations that characterise this psychiatric disorder. Chronic mild or unpredictable stress (CMS) is a widely used and validated preclinical model of depression displaying several behavioural and physiological alterations that mimic those observed in the clinical setting [174,175]. CMS is associated with reduced 2-AG levels in the hippocampus, enhanced anandamide levels in limbic and cortical areas and differential expression of CB1 receptor with increased receptor expression in the prefrontal cortex and a concurrent decrease in the hippocampus, hypothalamus and ventral striatum [156,176]. One of the first studies to imply that the CB2 receptor may be implicated in depression was that of Onaivi et al., where they demonstrated that CB2 protein levels measured by western immunoblotting in whole brain extract were enhanced in mice subjected to CMS for a period of 4 weeks [61,69]. However, the anatomical region(s) associated with the enhanced CB2 protein expression or the identity of the cells that express this receptor, i.e., glia or
of the elevated plus maze indicative of an anxiolytic-like effect [61,139], however, the effect of this treatment on locomotor activity in the maze was not reported. Only one published study to date has examined the role of brain CB2 receptors in conditioned aversion/learning, demonstrating that infusion of JWH-015 or PEA into the CA1 region of the hippocampus does not affect novel object recognition or long-term memory retention [140]. Overall, the results obtained from behavioural studies of the role of CB2 receptors in modulating the response to aversion are far from clear, highlighting the need for further studies examining the effects of selective deletion, blockade or stimulation of brain CB2 receptors on the regulation of emotional responding.
A critical component of the stress response is the activation of the hypothalamic-pituitary-adrenal (HPA) axis and the subsequent increase in glucocorticoid levels and several studies have examined the role of the endocannabinoid system in mediating this response [141]. CB2 receptor mRNA and protein have been identified in the brain regions that modulate HPA axis activity including the hippocampus, amygdala and hypothalamus [56,71], while in comparison, CB2 receptors are not expressed in the adrenal cortex [142], indicating that should CB2 ligands modulate the neuroendocrine response to stress, this would most likely occur at the level of the brain. Our studies have indicated that endotoxin-induced increases in circulating corticosterone levels are not altered by administration of the non-selective CB1/2 receptor agonist HU-210 or the CB2 receptor antagonist/inverse agonist SR144528 [125]. Further studies will be required to determine whether CB2 receptors are involved in the endocannabinoid-mediated modulation of neuroendocrine activity under basal and stress conditions.
Although Onaivi and colleagues demonstrated similar effects of the CB2 receptor ligand JWH-015 on locomotor activity and stereotypic behaviour in three strains of mice, namely C57Bl6/J, DBA/2 and BABL/c, they did not compare behavioural responses between these different strains in tests of emotionality and anxiety [139]. BALB/c mice have been proposed as a model of anxiety and are regarded as more stress-sensitive than other mouse strains including C57Bl6/J and DBA/2 mice [143-145]. For example, BALB/c mice exhibit an anxiogenic profile in the black and white box, manifested as a reduced time in the white compartment of the test apparatus when compared to C57Bl6/J counterparts [145]. Therefore, due to the contribution of genetic background to stress responding, it is possible that CB2 ligands may elicit differential effects depending on the strain of mouse used. In accordance, low doses of JWH-015 (5 mg/kg) increased stereotypic behaviour in male DBA/2 but not C57Bl6/J or BALB/C mice [60]. No significant difference in whole brain CB2 receptor expression prior to or following chronic stress was observed between the three different mouse strains [139]. However, it remains to be determined if the density or function of CB2 receptors or other components of the endocannabinoid system are differentially altered between these different strains of mice in brain regions associated with stress responding. Accordingly, early life stress such as maternal deprivation induces an anxiogenic behavioural phenotype [146] which has recently been associated with enhanced CB2 receptor expression in the hippocampus [70]. Thus, greater understanding of the role of CB2 receptors in pathological anxiety states will be reached by examining the expression, distribution and functional activity of these receptors in preclinical models that exhibit ethological validity.
Pharmaceuticals 2010, 3 2528
5.2. Depression
The involvement of the endocannabinoid system in the regulation of mood and affective responding has received increasing interest in the past few years (for recent reviews see [147-149]), particularly in light of the withdrawal of rimonabant (CB1 receptor antagonist/inverse agonist) as an anti-obesity agent due to the increased risk of psychiatric side effects, including depression. The involvement of CB1 receptors in regulating mood is further supported by the upregulation of CB1 receptor expression and function in the prefrontal cortex of depressed suicide victims [150] and enhanced CB1 receptor density in the prefrontal cortex of alcoholic suicide victims compared with alcoholic controls [151]. Furthermore, allele variations in the CB1 receptor gene CNR1, plays a role in the antidepressant response in major depressed patients [152] and a cohort of elderly depressed Parkinsonian patients have demonstrated a polymorphism (AATn) of the CNR1 [153]. Preclinical evidence further supports the involvement of CB1 receptors in depressive-like behaviour [154,155] and suggests that the effects of currently used antidepressants might depend upon endocannabinoid system modification. For example, endocannabinoid-CB1 receptor signaling in the brain is altered by several interventions that elicit antidepressant activity in humans, including chronic tricyclic antidepressant treatment [156,157], repeated electroconvulsive therapy [158] and sleep deprivation [159]. In contrast, the involvement of the CB2 receptor in affective responding has not received similar attention. A recent study has, however, demonstrated an association between depression and a polymorphism in the CB2 receptor gene at position Q63R in Japanese patients [69]. Polymorphisms in the Q63R in the CB2 gene have also been linked with eating disorders, alcoholism, osteoporosis, autoimmune disease and schizophrenia [93,102,160,161], many of which often demonstrate co-morbidity with depressive illness. Although it remains to be determined if this genetic link exists across other ethnic groups, it is possible that genetic variation in the CB2 gene may be a predisposing factor in the development of depression. CB2 receptor mediated effects of cannabinoid agonists such as WIN55,212-2 and 2-AG are reduced in the presence of the Q63R polymorphism [160,162] and serum levels of 2-AG and anandamide are reduced in patients with major depression, an effect directly correlated with the duration of the depressive episode [163]. Although further studies are required, in addition to alterations in endocannabinoid levels and CB1 receptor signaling, central CB2 receptor expression or function may be altered in depressed patients.
The association between adverse life events and the development of depression is widely acknowledged, and, as a consequence, many of the preclinical models of depression or antidepressant-like activity exploit this link by examining behavioural and physiological responses to stress. One of the most widely used behavioural tests for antidepressant-like activity is the forced swim test, where rodents exposed to a confined swim arena will initially attempt to escape but after some time will assume a floating posture (immobility) that is thought to be related to a state of behavioural despair [164]. Antidepressant agents increase escape behaviour thereby reducing the duration of immobility, while in comparison, an increase in immobility is regarded as a depressive-like state. Antidepressant-like activity of cannabinoid ligands and endocannabinoid modulators such as FAAH and anandamide reuptake inhibitors, have been demonstrated in the forced swim test [131,155,165-171]. In many of the studies, the antidepressant-like effects have primarily been attributed to activity at the CB1 receptor, confirmed using receptor antagonists and/or genetic deletion. Recent evidence has indicated that CB1
Pharmaceuticals 2010, 3 2529
receptors on subpopulations of glutamatergic, but not GABAergic, neurons appear to mediate the forced swim stress-induced behavioural and neuroendocrine effects [172]. However, non-selective CB1/CB2 receptor agonists and endocannabinoid modulators may also activate CB2 receptors. Antidepressant-like activity following chronic CB1/CB2 receptor stimulation using HU210, but not CB1 receptor agonism (AM281) alone, was observed in the forced swim test [131]. In addition, intra-hippocampal administration of HU210 induced an antidepressant-like effect in the forced swim test, an effect only partially attenuated by pharmacological blockade of the CB1 receptor [169]. Although further studies are required in order to further determine the mechanism of action underpinning these effects of non-selective CB1/CB2 agonists, it is possible that activation of central CB2 receptors may co-operatively augment the effects of CB1 receptor activation on stress-induced behaviour and monoaminergic function. Presently, only one study has been published directly assessing the effect of pharmacological CB2 agonism in the forced swim test, reporting that acute administration of the CB2 agonist GW405833 did not alter time spent immobile in the forced swim test [173]. However, in the presence of neuropathic pain, CB2 receptor activation reduced nociceptive responding to a mechanical stimulus while concurrently attenuating the enhanced immobility observed in the forced swim test [173]. The present results indicate that CB2 receptor agonism may alleviate neuropathic pain and the co-morbid depressive symptoms that often accompany this disorder. In an interesting study published recently, García-Gutiérrez and colleagues employed both pharmacological and genetic approaches to investigate the role of the CB2 receptor in depressive-like behaviour [138]. Transgenic mice engineered to over-express the CB2 receptor (including over-expression in key brain regions implicated in depression) exhibited reduced depressive-like behaviour in the tail suspension test and in a novelty suppressed feeding test, compared with wildtype controls. However, acute intraperitoneal administration of the CB2 receptor antagonist AM630, at doses that had no effect on locomotor activity, had an antidepressant-like effect in the forced swim test in wildtype mice, but not in CB2 over-expressing mice [138]. Though the depression-resistant endophenotype associated with CB2 over-expression and the antidepressant-like effects of CB2 receptor blockade pose an apparent discrepancy which is difficult to reconcile, these data do at least demonstrate an important role for the CB2 receptor in regulating depressive state in mice. More studies examining the effects of direct activation/inhibition of brain CB2 receptors are required in order to determine conclusively the role this receptor plays a role in stress-induced behavioural changes.
A detailed understanding of the neurobiological underpinnings of depression requires pre-clinical models that mimic the neurological and physiological alterations that characterise this psychiatric disorder. Chronic mild or unpredictable stress (CMS) is a widely used and validated preclinical model of depression displaying several behavioural and physiological alterations that mimic those observed in the clinical setting [174,175]. CMS is associated with reduced 2-AG levels in the hippocampus, enhanced anandamide levels in limbic and cortical areas and differential expression of CB1 receptor with increased receptor expression in the prefrontal cortex and a concurrent decrease in the hippocampus, hypothalamus and ventral striatum [156,176]. One of the first studies to imply that the CB2 receptor may be implicated in depression was that of Onaivi et al., where they demonstrated that CB2 protein levels measured by western immunoblotting in whole brain extract were enhanced in mice subjected to CMS for a period of 4 weeks [61,69]. However, the anatomical region(s) associated with the enhanced CB2 protein expression or the identity of the cells that express this receptor, i.e., glia or
neurons, were not identified. CB2 receptor mRNA was detected in the striatum, midbrain and hippocampus of control and CMS exposed mice, however, no significant difference in expression was observed between the groups [61]. More recently, García-Gutiérrez and colleagues demonstrated that CMS in mice for 7-8 weeks resulted in reduced levels of CB2 mRNA in the hippocampus of mice, compared with non-stressed controls. Moreover, this stress-induced reduction was prevented by chronic administration of the CB2 receptor antagonist AM630 [138]. Behaviourally, acute treatment with JWH-015 reduced stereotypic behaviour in stressed but not non-stressed mice whereas the converse was observed following chronic treatment [139]. This group have also reported that reduced spontaneous locomotor activity of mice subjected to CMS was enhanced by acute and chronic treatment with JWH-015. In addition, CMS-induced anxiogenic behaviour in the elevated plus maze was attenuated by acute administration of JWH-015 [60,69]. CMS is associated with reduced intake of palatable solutions such as sucrose, used as a measure of anhedonia, a hallmark of depressive-illness. Daily administration of either the CB2 receptor agonist JWH-015 or the CB2 receptor antagonist AM630 did not alter chronic stress-induced decreases in sucrose consumption [69,91]. Administration of JWH-015 for a period of 2 weeks increased sucrose consumption in control but not stressed animals, an effect not observed at later time points. Although the authors report no effect of AM630 on sucrose consumption, examination of the data indicates that sucrose consumption was reduced in control, but not stressed mice, from week 2 of treatment with this CB2 receptor antagonist (3 mg/kg) [69]. Based on these findings it would appear that stress blocks CB2 receptor modulation of hedonic responses. Similarly, although CB2 receptor ligands elicit no effect on alcohol consumption in control mice, enhanced alcohol consumption following CMS is augmented by JWH-015 and slightly attenuated by AM630 [93]. In addition to providing further evidence for a differential role of CB2 receptors in modulating behaviour in stressed versus non-stressed animals, the results of this latter study also raise the possibility of a role for CB2 receptors in the co-morbidity of depression and alcohol abuse. Recently it has been shown that transgenic mice engineered to over-express the CB2 receptor are resistant to CMS-induced reductions in sucrose consumption and increases in tail suspension test immobility time [138]. The CB2 over-expressing mice were also resistant to CMS-induced reductions in brain derived neurotrophic factor (BDNF) levels in the hippocampus. While these results suggest that the CB2 over-expression results in a depression-resistant endophenotype, the same study also reported that pharmacological blockade of CB2, with chronic administration of the CB2 receptor antagonist AM630 for 4 weeks, prevented the effects of CMS on tail suspension test, sucrose intake, CB2 receptor gene, BDNF gene and protein expression in wildtype mice. The authors speculate that the chronic antagonist treatment may have lead to increased expression of CB2 in key brain regions, thereby mimicking the phenotype of the CB2 over-expressing transgenic mice. Taken together, these results indicate that CB2 receptors may play an important role in mediating behavioural and molecular effects associated with CMS.
Early life stress has been linked with a predisposition to psychiatric disorders in later life, resulting in the development of several preclinical models based on this association. One such model is the maternal deprivation (MD) model which involves separation of neonatal rats from the dam for a single 24 hour episode resulting in long-lasting behavioural, neurochemical and immune changes and has been proposed as a model of several neuropsychiatric disorders including depression. The depressive-like phenotype associated with MD includes decreased latency to immobility in the forced swim test
Pharmaceuticals 2010, 3 2531
[177,178], reduced locomotor activity and social investigatory behaviour [178] and enhanced impulsivity [179]. Alterations in the endocannabinoid system have been demonstrated in this model where MD is associated with enhanced hippocampal 2-AG levels [180] and reduced CB1 and increased CB2 receptor expression in the hippocampus [70]. Although both male and female MD rats exhibited a comparable increase in hippocampal CB2 receptor expression, CB1 receptor expression demonstrated sexual dimorphism, with a greater MD-related decrease observed in males when compared to females [70]. Examination of gender-specific effects is of particularly importance due to the enhanced prevalence of neurodevelopmental and psychiatric disorders in women. Detailed examination of the location of these receptors demonstrates that hippocampal CB2 receptors are located on dendritic terminals and not microglia [70]. This is consistent with the finding that CB2 receptors are located post-synaptically in the hippocampus [55] in comparison to the CB1 receptor that is primarily located on pre-synaptic GABAergic and glutamatergic terminals [181,182]. It remains to be determined if similar alterations in the expression of cannabinoid receptors occurs in other brain regions in this model. Thus, MD-induced changes in CB2 receptor expression, and other components of the endocannabinoid system may underlie some of the behavioural, cognitive and neuroendocrine changes observed in this model and in the neuropsychiatric disorders it models.
Figure 1. The CB1/CB2 receptor agonist HU210 (100 μg/kg i.p.) and CB2 receptor antagonist SR144528 (3mg/kg i.p.) attenuated LPS (100 μg/kg i.p)-induced increases in IL-1β and TNFα levels in the prefrontal cortex in rats.
Rimonabant (SR141716A; 3mg/kg i.p.) partially blocked the HU210-induced attenuation of IL-1β following LPS administration. Cytokines were measured 2 hours post saline or LPS administration and rats received drugs or vehicle administration 30 minutes prior to saline or LPS. Data analysed by ANOVA followed by Student-Newman-Keuls post hoc test. N = 6-8 per group. ** P < 0.01 *P < 0.05 vs. Vehicle-saline. ++ P < 0.01 + P < 0.05 vs. vehicle-LPS. Reproduced with permission from Roche et al., 2006 [125].
The functional significance of the alterations in cannabinoid receptor expression in animal models of depression remains to be determined however, as previously highlighted, CB1 and CB2 receptors
Pharmaceuticals 2010, 3 2532
modulate neural stem cell proliferation in culture [26,183]) and/or in adult mice [25,129,184,185]. Impaired hippocampal neurogenesis has been proposed to underlie the pathophysiology of depression (for reviews see [186-188]). Neurogenesis relies on several factors including neurotrophins such as brain derived neurotrophic factor (BDNF), a reduction in which has been observed in depressed patients [189-191] and in preclinical models including CMS [192] and MD [193]. Antidepressants, as a class, reverse pathological stress-induced reductions in adult hippocampal neurogenesis [194], an effect at least partially mediated by enhancing BDNF, TrkB receptor signaling and activation of MAPK/ERK pathways [195-198]. Similarly, chronic, but not acute, treatment of rats with the CB1/CB2 agonist HU210, but not the selective CB1 agonist AM281, enhanced hippocampal neurogenesis and elicited antidepressant-like behavioural effects [131]. The inability of AM281 to modulate neurogenesis or immobility behaviour may indicate a role for CB2 receptors in the effects of HU210, promoting neuronal survival and differentiation in the hippocampus. In addition, WIN55,212-2 induced activation of both CB1 and CB2 receptors enhances neurogenesis in the hippocampus of aged rats [132]. As highlighted earlier, immunological mediators also modulate neurogenesis, with high levels of pro-inflammatory cytokines considered detrimental to neuronal viability. It has been proposed that the reduced neurogenesis observed in depression is due, at least in part, to the enhanced immune activation and elevated cytokine levels that are a feature of this psychiatric disorder [199], an effect also demonstrated in several animal models including CMS and MD [178,200,201]. Furthermore, immune stimuli such as endotoxins activate microglia and enhance inflammatory cytokine levels in the brain resulting in reduced BDNF [202], reduced neurogenesis [203,204] and depressive-like symptoms [205]. We have previously shown that HU210 attenuates endotoxin-induced increases in IL-1β and TNFα in rat brain, effects partially mediated by CB1, but not CB2, receptors [125]. However, blockade of either CB1 or CB2 receptors, using rimonabant or SR144528 respectively, also attenuated lipopolysaccaride (LPS)-induced cytokine levels in the brain [125] (Figure 1). In addition, administration of the endotoxin LPS enhances endocannabinoid levels [206-208] and increases CB2 receptor protein expression in the brain (detected using western immunoblotting) [86]. Based on the clinical and preclinical evidence, we hypothesise that depression is associated with altered endocannabinoid function including that of the brain CB2 receptors, activation of which would reduce inflammatory responses, enhance neurogenesis and result in antidepressant activity (Figure 2).
5.3. Schizophrenia
The role of the endocannabinoid system in schizophrenia has received considerable attention and has be covered in detail in several recent reviews [209,210] including those within this special issue [211,212]. Therefore, this section will concentrate primarily on the putative role of the CB2 receptor in this disorder.
Ishiguro and colleagues have recently demonstrated that Japanese schizophrenic patients exhibit an increase in the frequency of two single nucleotide polymorphisms (SNPs) in the CB2 receptor gene, namely rs12744386 and Q63R, which confer lower functioning of the CB2 receptor [160]. Low levels of CB2 receptor mRNA and protein in the brain and lympohocytes were associated with the C allele of rs1274486 gene, a genotype commonly observed in schizophrenic patients.
Figure 2. Putative mechanisms underpinning potential antidepressant effect of CB2 receptor stimulation.
Depression is associated with alterations in various neurotransmitter systems including the endocannabinoid system, enhanced immune activity and impaired neurogenesis. Stimulation of CB2 receptors, either by endocannabinoids or exogenously administered CB2 receptor agonists, may reduce neuroinflammatory responses, enhance neurogenesis and alter levels of neurotransmitters and neuropeptides, in a manner which is consistent with antidepressant activity. See main text for references and discussion pertaining to the limited number of published studies that have informed the preparation of this schematic. The model is necessarily speculative at this point and further studies are required before strong, conclusive links can be made between CB2 receptor-mediated modulation of these systems and processes and putative antidepressant activity.
As mentioned earlier, the other SNP (rs2501432) is a mismatch Q63R, the presence of which leads to poor response to CB2 ligands [160]. Thus, the presence of both SNPs may synergistically confer enhanced susceptibility to schizophrenia. In line with these observations, schizophrenia is associated with enhanced cerebrospinal fluid levels of anandamide and PEA, and remission from which is associated with reduced anandamide levels and peripheral blood mononuclear cell CB2 receptor mRNA expression [213-215]. It is unknown if similar changes in CB2 receptor expression are observed in the brain, however, neuroleptic treatment reduces G-protein functioning in schizophrenic patients [216], and as such may induce similar effects on G-protein coupled cannabinoid receptors.
The effect of CB2 receptor antagonists on MK-801- or methamphetamine-induced disturbances in pre-pulse inhibition (PPI) have also been examined. PPI is a widely used behavioural test of sensorimotor gating, deficits in which are commonly observed in schizophrenic patients [217,218]. Essentially, PPI refers to the ability of a weak pre-stimulus, a pre-pulse, to inhibit the startle reflex elicited by a subsequent intense stimulus. Administration of the CB2 receptor antagonist, AM630, alone, did not alter PPI, however AM630 did augment the MK-801- and metamphetamine-induced reduction in PPI and increase in locomotor activity [160]. It has been proposed that that a decrease in CB2 receptor functioning alone does not lead to the development of schizophrenia but that in the
Pharmaceuticals 2010, 3 2534
presence of other risk factors, reduced CB2 receptor functioning may confer enhanced susceptibility to the development of this disorder.
Disorganised stereotypic behaviour is common in psychotic individuals, mediated by hyperdopaminergic functioning and alleviated by antipsychotic treatment. CB1 receptor desensitization due to chronic cannabinoid treatment or CB1 receptor antagonism exacerbates dopamine receptor-induced stereotypic behaviours [219,220]. Although the role of CB2 receptors in directly mediating or modulating dopamine-induced effects has not been investigated, CB2 receptor agonists reduce stereotypic behaviour in a dose-dependent and gender specific manner [60] as outlined in previous sections. In addition, CB2 receptors modulate key neurotransmitter systems involved in schizophrenia such as dopaminergic and glutamatergic function, possibly via microglial inhibition. For example, CB2 receptor agonists prevent 6-hydroxydopamine induced dopamine depletion [221] and glutamate receptor (AMPA, kainate and NMDA) mediated excitotoxicity [25,79,122,222]. Alterations in cannabinoid receptor expression may induce profound alterations in neurotransmission (dopaminergic and/or glutamatergic) and/or, modulation of hippocampal axonal growth and plasticity, which may confer a predisposition to the development of schizophrenia. MD and/or exposure to cannabinoids during critical neurodevelopmental periods have been proposed to induce such neurochemical alterations which may underlie the psychotic-like behavioural alterations. In addition to the depressive-like phenotype exhibited following MD, this model also results in long-term behavioural alterations that resemble symptoms observed in schizophrenia, including deficits in PPI, latent inhibition and auditory sensory gating [223,224]. The behavioural alterations associated with MD, combined with the neuronal, endocrine and immune alterations observed, support its usefulness and relevance as a model based on the neurodevelopmental hypothesis of schizophrenia. As previously mentioned MD is associated with alterations in the endocannabinoid signalling system including reduced CB1 expression and enhanced CB2 receptor expression in the hippocampus [70,146,180]. Although an increase in CB2 receptor expression in the hippocampus of MD rats may seem at odds with data suggesting that schizophrenia is associated with reduced CB2 receptor functioning, it should be noted that although the density of CB2 receptors is enhanced, significant impairment in function may exist. In addition, CB2 receptor expression was assessed in pre-pubertal MD rats and the density and distribution pattern of this receptor may be different in adults.
Schizophrenia is associated with altered neuroimmune functioning, primarily an imbalance between type-1 and type-2 immune responses, which is thought to underlie altered neuronal function including neurotransmitter alterations and reduced neurogenesis (for review see [225-228]). In addition, epidemiological data have demonstrated an association between prenatal infection, enhanced pro-inflammatory cytokine levels and increased risk of psychiatric disorders in later life, including schizophrenia [229], an association exploited in order to develop more ethologically valid models of schizophrenia. This imbalance in immune function is also observed in neurodevelopmental models such as MD [178,200]. Antipsychotic medication and anti-inflammatory agents such as COX-2 inhibitors alleviate psychotic symptoms, correct the imbalance in type-1 and type-2 responses and reduce pro-inflammatory cytokine release [225,226]. It remains to be determined whether the anti-inflammatory effects of CB2 receptor agonism might confer protection and enhancement of neuronal function of sufficient magnitude to alleviate psychotic symptoms in schizophrenia.
Pharmaceuticals 2010, 3 2535
The enhanced prevalence of schizophrenia and other psychiatric disorders such as anxiety and depression in women highlights the need for more studies examining gender differences in the development of these disorders. Gonadal hormones have been demonstrated to alter expression of cannabinoid receptors [230] and females in general are more sensitive than males to the behavioural effects of cannabinoids [231]. As highlighted in earlier sections, few studies have examined the effect of CB2 ligands in animal models of neuropsychiatric disorder and for the most part these studies have been confined to examining effects in male animals. However, it has been demonstrated that the CB2 receptor agonist JWH-015 reduces stereotypic behaviour in female mice at a dose (10 mg/kg) that is ineffective in male counterparts [60]. In addition, sexually dimorphic effects on hippocampal CB1 receptor expression have been demonstrated in the MD neurodevelopmental animal model, with a more marked MD-related decrease observed in male rats compared with female rats [70]. In comparison, comparable increases in CB2 receptor expression were observed in male and female rats. As such further studies are required in order to determine if gender-specific alterations in endocannabinoid function may underlie the development of, and differential susceptibility to, neuropsychiatric disorders.
