Endocannabinoids and the Control Of Behavior and Cardiovascular Function

National Institute On Alcohol Abuse And Alcoholism

Investigators

Linked publications, trials & patents

Abstract

The vagal gut-brain axis has been recognized as a vital regulator of metabolic, motivational, and emotional states. These include anxiety, depression, reinforcement, food and alcohol craving, and involve the endocannabinoid system. Our previous observations indicated that the brain non-penetrant CB1 receptor (CB1R) antagonist, JD5037, suppresses alcohol preference in mice by blocking CB1R in ghrelin-producing cells and hampering the permissive role of ghrelin to drink alcohol (Cell Metab 2019; 29: 1320-1333). In the previous reporting periods, we have identified subsets of sensory afferent neurons of the nodose ganglion that express CB1 receptors, the activation of which promotes voluntary ethanol drinking (VEI). These neuronal subsets selectively expressed advillin and Phox2B proteins. We then further narrowed the subtype of CB1 receptor-expressing afferent neurons relevant to VEI. Namely, nodose ganglion projections of Phos2B positive, CB1R expressing neurons have been detected in both the muscular and mucosal layers of the GI tract. These projections represent two non-overlapping populations of vagal afferents whose genetic markers are the Glp-1 receptor (Glp1r) and GPR65, respectively. To determine the contribution of these two populations of sensory neurons in CB1R-mediated alcohol drinking, we eliminated CB1R selectively from the muscular and mucosal layers of the gut by crossing Cnr1(lox/lox) mice with Glp1r(iresCre) or GPR65(iresCre) mice, respectively. The resulting mutant mice, i.e. muscular-specific CB1Rko and mucosal-specific CB1Rko and their respective littermate controls were assessed by Cre/lox based genotyping and confirmed by RNAscope. When these mice were analyzed in the two-bottle/free choice drinking paradigm, the deletion of CB1R in vagal afferents expressing GPR65 was sufficient to abolish the inhibitory effect of JD5037, whereas the drug remained fully effective after knocking out CB1R in Glp1r-expressing vagal afferents innervating the muscular layer. Thus, CB1R on vagal afferent terminals in the mucosal layer (where ghrelin-producing cells are localized) but not those in the smooth muscle layer are involved in the control of VEI. Furthermore, JD5037 inhibition of voluntary ethanol intake (VEI) was lost in mice with CB1R deletion in Phox2b+ nodose ganglia (NGA), but not in Wnt+ dorsal root ganglia. Additionally, deleting the ghrelin receptor (Ghsr) from Phox2b+ NGA neurons blocked the inhibition of VEI by either peripheral Ghsr or CB1R antagonists. Reciprocally, VEI by advillin+ NGA-CB1RKO mice were similarly unaffected by the peripherally restricted ghrelin receptor antagonist PF-5190457. These findings suggest a mutual interdependence of endocannabinoid and ghrelin signaling in controlling VEI via a gut-brain axis. Finally, VEI in mice was also inhibited by the novel, peripherally restricted biased CB1 receptor antagonist discovered in our laboratory (MRI-1891, also called INV-202 or monlunabant), which showed anti-obesity efficacy in a phase 1b clinical trial in obese people with the metabolic syndrome. Thus, it will soon be possible to test the therapeutic efficacy of peripheral CB1 receptor blockade in people with AUD. In the current review period, a manuscript summarizing these findings has been completed, submitted for publication and has just been accepted by Molecular Psychiatry.