Godlewski and colleagues reported that administration of the peripherally-restricted CB1R inverse agonist, JD5037, reduced the intake of ethanol in wild-type mice; however, it was ineffective in whole-body CB1R- and GHS-R1a-null mice. Ethanol-consuming mice also had elevated levels of the eCB, anandamide, in stomach cells, and inhibiting peripheral CB1Rs with JD5037 blocked formation of the bio-active form of ghrelin, octanoyl-ghrelin. These results suggest that CB1Rs in stomach cells promote ethanol intake by a mechanism that includes controlling production of ghrelin. Next, a mouse gastric ghrelinoma cell line — which contains CB1Rs and enzymatic machinery for eCB metabolism—was used to identify mechanisms of CB1R-mediated ghrelin production. Inhibition of CB1Rs in MGN3-1 cells with JD5037 blocked formation of octanoyl-ghrelin by a mechanism that includes increased oxidative degradation of the ghrelin substrate, octanoyl-carnitine. Moreover, given expression of ghrelin receptors in the brain as well as vagal afferent neurons, the authors aimed to identify if the actions of JD5037 to reduce ethanol intake via changes in ghrelin signaling required gastric vagal afferent neurons. Both JD5037 and the CB1R inverse agonist, rimonabant, were ineffective at reducing ethanol intake in mice subjected to chemical ablation of sensory afferents by neonatal exposure to capsaicin. Interestingly, mice denervated by capsaicin displayed moderate increases in preference and intake of ethanol. Additionally, mice treated with JD5037 displayed no changes in ad-libitum intake of standard rodent chow under these specific conditions. Together, these studies provide evidence that CB1Rs in mouse stomach cells control intake and preference for ethanol by a mechanism that includes regulating production of ghrelin and indirect control of gut-brain vagal signaling. Future studies will be important to clarify if activating CB1Rs stimulates production of ghrelin by increasing conversion of octanoyl-carnitine to octanoyl-ghrelin,cannabis drying and if roles for these pathways extend beyond intake and preference for ethanol to other reinforcers, such as palatable food.
In addition, the precise impact that CB1R-mediated control of ghrelin production has on vagal neurotransmission and associated firing rates remains to be determined. Notably, ghrelin and CCK inversely affect vagal afferent neural activity, with ghrelin decreasing and CCK increasing activity. Accordingly, it is possible that eCB signaling at CB1Rs on stomach cells that produce ghrelin and on CCK-containing cells in the upper small-intestinal epithelium that inhibit CCK release results in similar reductions in vagal afferent neural activity and increases in food intake. Moreover, these pathways may coordinate vagal afferent neural activity associated with feeding status and become imbalanced in diet-induced obesity. A direct test of these hypotheses, however, remains for future investigations. In addition to indirect mechanisms, eCBs may also activate CB1Rs located on the afferent vagus nerve and directly impact gut-brain neurotransmission and food intake. Indeed, a series of studies by Burdyga and colleagues suggest that expression of CB1Rs on rat gastric vagal afferent neurons is impacted by feeding status and gutderived hormones. Immunoreactivity and mRNA for CB1Rs were identified in the rat and human nodose ganglia, and fasting for up to 48 h in rats was associated with time-dependent increases in their expression. Refeeding after a 48 h fast led to reductions in expression of mRNA for CB1Rs in nodose ganglia by 2-hrs after reintroduction of food, an effect mimicked by administration of bio-active CCK-8 in fasted rats. In addition,administration of a CCKA receptor antagonist blocked refeeding-induced reductions in expression of mRNA for CB1Rs in fasted rats, which suggests a key role for CCK in these processes. Similarly, administration of ghrelin blocked refeeding-induced reductions in expression of mRNA for CB1Rs in nodose ganglia and blocked the actions of CCK-8 administration on expression of mRNA for CB1Rs in fasted rats. These results highlight the opposing actions that gut-derived satiation and hunger signals have on expression of CB1Rs in rodent vagal afferent neurons. Several studies suggest that expression of CB1Rs in the nodose ganglia is dysregulated in rodent models of diet-induced obesity. Immunoreactivity for CB1Rs was elevated in the nodose ganglia in Zucker or Sprague Dawley rats that were maintained on high-fats diet for 8 weeks when compared to lean controls.
Similarly, mRNA for CB1Rs was elevated in nodose ganglia in mice fed a high-fat diet for 12 weeks. In addition, refeeding after a fast or administration of CCK-8 in Wistar rats maintained on a high-fat diet both failed to reduce levels of immunoreactivity for CB1Rs in nodose ganglia. Moreover, levels of mRNA for CB1Rs were elevated in the nucleus of the solitary tract in rats maintained on a high-fat and sugar diet. Collectively, these studies suggest that CB1R expression in rodent vagal afferent neurons is controlled by feeding status, and their meal-related expression is dysregulated by chronic exposure to high-fat diets. Roles in food intake for CB1Rs expressed in vagal afferent neurons are unclear; however, Elmquist and colleagues reported that genetic deletion of CB1Rs in the afferent and efferent vagus nerve had no impact on food intake, body weight, or energy expenditure in mice maintained on standard rodent chow or a high-fat diet. These results suggest that CB1Rs on vagal afferent neurons may be sufficient to promote food intake but are not required in these processes. With regards to food intake, these findings are also in line with the transient nature of feeding suppression in rodents administered CB1R antagonists, which suggests that CB1Rs may not be required for the long-term maintenance of food intake. Nonetheless, a series of important studies investigated the impact of activating CB1Rs on the neurochemical phenotype of associated neurons and the function of gastric vagal afferent neurons in mice. Similar to CB1Rs, fasting was associated with time-dependent increases in expression of melanin-concentrating hormone 1 receptor in the nodose ganglia of rats, albeit at later time-points when compared to CB1Rs. In contrast,growers equipment fasting was associated with time-dependent reductions in expression of neuropeptide Y receptor type 2. Administration of CCK-8 reversed the effects of fasting by decreasing expression of CB1Rs and MCH1Rs, and increasing expression of Y2Rs. Notably, administration of the eCB, anandamide, dose-dependently reversed the effects of CCK-8 on expression of CB1Rs, MCH1Rs, and Y2Rs. Moreover, administration of a CB1R inverse agonist reduced expression of CB1Rs and increased expression of Y2Rs with no effect on expression of MCH1Rs in fasted rats. Together, these studies reveal distinct changes in the neurochemical composition of vagal afferents neurons in response to CB1R activation and inactivation, and suggest that CB1Rs may directly modulate activity of vagal afferent neurons in response to food-related signals released from the gut. Elegant studies conducted by Christie and colleagues suggest that CB1Rs control mechanosensitivity of gastric vagal afferent neurons, which may be dysregulated in dietinduced obesity. For these studies, a mouse in vitro electrophysiological preparation was utilized that consists of isolated stomach and esophagus with attached vagal fibers and measurement of vagal afferent neural function and mechanosensitivity. Application of methanandamide—a stable analog of anandamide —led to a biphasic effect on activity of vagal fibers in response to stretch, with low doses reducing responses to stretch and high doses increasing responses. These effects were found only in tension sensitive fibers, but not those innervating gastric mucosa. In contrast to mice maintained on standard rodent chow, mice maintained for 12 weeks on a high-fat diet were only responsive to the inhibitory effects of methanandamide on gastric vagal afferent neural activity.
To identify receptor signaling pathways mediating these effects, methanandamide was co-incubated with a CB1R inverse agonist, a transient receptorpotential vanilloid-1 channel antagonist, a growth hormone secretagogue receptor antagonist, or several inhibitors of distinct second messenger pathways. The biphasic effects of methanandamide on mechanosensitivity were both inhibited by CB1R and TRPV1 blockade in mice maintained on standard rodent chow. Furthermore, the excitatory effects of methanandamide may occur via a CB1R-mediated PKC-TRPV1 second messenger pathway, whereas the inhibitory effects may occur via CB1R-mediated release of ghrelin from the stomach and its actions on GHSRs on vagal afferent neurons. Together, these studies suggest that endocannabinoids differentially control afferent vagal activity depending upon dose by mechanisms that include distinct interactions between CB1Rs, TRPV1, and GHSR signaling pathways, which may become dysregulated in diet-induced obesity. Future studies will be important to identify physiological roles in food intake and obesity for CB1R signaling in distinct populations of gastric vagal afferent neurons. Moreover, it will be important to delineate how CB1Rs on sensory vagal terminals in the gut, nodose ganglia, and at terminals in the NTS may participate in distinct or common aspects of vagal afferent neurotransmission. Fasting is associated with elevated levels of eCBs in the upper small-intestinal epithelium of rodents, and recent studies suggest that the efferent vagus nerve is required for these processes. The efferent arm of the vagus nerve communicates parasympathetic neurotransmission from the brain to peripheral organs—including the gut—via cholinergic signaling pathways, and participates in a variety of motor functions and possibly food intake. For example, c-Fos expression in the myenteric and sub-mucosal plexus in the rat proximal small intestine was induced by vagal nerve stimulation, and pharmacological blockade of peripheral muscarinic acetylcholine receptors with atropine methyl nitrate inhibited both refeeding after a fast and sham feeding of liquid diets in rats.Recent investigations, however, suggest that cholinergic neurotransmission carried by the efferent vagus nerve controls biosynthesis of the eCB, 2-AG, in the proximal small-intestinal epithelium during a fast and participates in refeeding after a fast. For these studies, rats were fasted for up to 24 h, then levels of 2-AG and its precursor, 1, stearoyl, 2-arachidonoyl-sn-glycerol , were quantified in a variety of peripheral organs by liquid chromatography/tandem mass spectrometry. Levels of 2-AG and SAG were elevated in the upper small-intestinal epithelium by 24 h after fasting; however, no changes were found in stomach, ileum, colon, liver, pancreas, or spleen. This effect was specific for 2-AG, because levels of other common monoacylglycerols in the upper small-intestinal epithelium were unaffected. Moreover, levels of 2-AG were rapidly normalized by refeeding to levels similar to those in free-feeding animals, an effect mimicked by intra-duodenal infusions of equicaloric quantities of lipid, sucrose, or protein. These results highlight that production of intestinal 2-AG in fasting rats can be rapidly reduced upon refeeding in a manner that is not dependent on macronutrient content.We next aimed to identify if efferent cholinergic activity is required for production of 2-AG in the intestinal epithelium. Rats were given full diaphragmatic vagotomy and fasted for 24 h. When compared to control rats receiving a sham surgery, levels of 2-AG failed to increase in the small-intestinal epithelium after a 24 h fast, which suggests that the efferent vagus may be required for production of 2-AG. We next aimed to identify specific cholinergic receptors involved in vagal-mediated 2-AG production during a fast. The principal neurotransmitter released from the efferent vagus nerve is acetylcholine, which activates a variety of cholinergic receptor subtypes in the periphery, including mAChRs in the intestine. Activation of mAChRs in the brain, including subtype- 3 mAChRs, enhances eCB production that, in turn, participates in the control of synaptic plasticity via CB1Rs. In addition, 2-AG in the brain is formed by a mechanism that includes phospholipase-C-dependent production of SAG, then conversion of SAG to 2-AG by diacylglycerol lipase. Notably, m3 mAChRs are Gq-protein coupled receptors that share overlapping downstream pathways as those responsible for biosynthesis of 2-AG, including activation of phospholipase-C and diacylglycerol lipase. Thus, we examined if mAChRs are required for fasting-induced production of 2-AG in the small-intestinal lining. Similar to brain, m3 mAChRs were expressed in the small intestinal epithelium, and diacylglycerol lipase activity was required for biosynthesis of 2-AG. Systemic administration of a general mAChR antagonist or intra-duodenal administration of a selective m3 mAChR antagonist both blocked fasting-induced rises in 2-AG in the small intestine. Moreover, pharmacological inhibition of peripheral CB1Rs and m3 mAChRs in the intestine equally reduced refeeding after a 24 h fast, with no additive effects when both inhibitors were co-administered. Collectively, these studies suggest that the efferent vagus nerve is recruited during a fast and participates in refeeding after a fast by activating m3 mAChRs in the intestine which, in turn, drives production of 2-AG and activation of local CB1Rs. In addition to interactions between efferent parasympathetic neurotransmission and the eCB system, studies also suggest that efferent sympathetic neurotransmission is controlled by CB1Rs, which may impact food intake through a mechanism that requires the afferent vagus nerve.