Endocannabinoids And Energy Homeostasis

National Institute On Alcohol Abuse And Alcoholism

Investigators

Linked publications & trials

Abstract

We have analyzed the endocannabinoid (EC)/cannabinoid receptor-1 (CB1R) system expressed in hepatic progenitor cells (HPC) and its role in the control of HPC differentiation and mitochondrial metabolism. The proliferation and differentiation of HPCs drive the homeostatic renewal of the liver under diverse conditions. Liver regeneration is associated with an increase in Axin2+Cnr1+ HPCs, along with a marked increase in the levels of the EC anandamide (AEA). However, the molecular mechanism linking AEA signaling to HPC proliferation and/or differentiation has not been explored. In a study just published, we show that in vitro exposure of HPCs to AEA triggers both cell cycling and differentiation along with increased expression of Cnr1, Krt19, and Axin2. Mechanistically, we found that AEA promotes the nuclear localization of the transcription factor beta-catenin, with subsequent induction of its downstream targets, including Axin2. Systemic analyses of cells after CRISPR-mediated knockout of the beta-catenin-regulated transcriptome revealed that AEA modulates beta-catenin-dependent cell cycling and differentiation, as well as interleukin pathways. Further, we found that AEA promotes OXPHOS in HPCs when amino acids and glucose are provided as substrates, whereas AEA inhibits it when the cells rely primarily on fatty acid oxidation. Thus, the EC/CB1R system promotes hepatocyte renewal and maturation by stimulating the proliferation of Axin2+Cnr1+ HPCs via the beta-catenin pathways while modulating the metabolic activity of their precursor cells. This work has been published in Cell Death Discovery (2023). Appetite regulation involves mutually antagonistic interactions between anorexic leptin and orexigenic ECs. We have earlier shown that leptin decreases hypothalamic levels of ECs (Nature, 2001), but the mechanisms by which ECs limit leptin action are not well understood. In a recently completed study we found that hypothalamic STAT3 signaling in mice, initiated by physiological elevations of leptin, is diminished by agonists of the CB1R. Measurement of STAT3 activation by semi-automated confocal microscopy in cultured neurons revealed that this CB1R-mediated inhibition requires both T cell protein tyrosine phosphatase (TC-PTP) and beta-arrestin1 but is independent of changes in cyclic AMP. Moreover, beta-arrestin1 translocates to the nucleus upon CB1R activation and binds both STAT3 and TC-PTP. Consistently, CB1R activation failed to suppress leptin signaling in beta-arrestin1 knock-out mice in vivo, and in neural cells deficient in CB1R, beta-arrestin1 or TC-PTP. Altogether, CB1R activation engages beta-arrestin1 to co-ordinate the TC-PTP-mediated inhibition of the leptin-evoked neural STAT3 response. This mechanism may restrict the anorexigenic effects of leptin when hypothalamic EC levels rise, as during fasting or in diet-induced obesity. This work has been published in iScience (2023). In other projects we are analyzing the role of peripheral CB1R in hepatocytes vs. adipocytes in obesity-related fatty liver disease. We earlier reported that the steatosis of mice with high-fat diet-induced obesity (DIO) is nearly completely reversed by treatment with a peripheral CB1R antagonist even though mutant mice that express CB1R in hepatocytes only on a global CB1R knockout background accumulate very little hepatic triglycerides (TGs) when fed the same high-fat diet (JCI 120:2953, 2010). This suggested that diet-induced steatosis may be promoted by extrahepatic peripheral CB1R, such as those in adipocytes, a known source of fatty acids (FAs) released and taken up by the liver for the synthesis of hepatic TGs. Indeed, mice with conditional knockout of CB1R in adipocytes are resistant to DIO and the associated steatosis (JCI 127:4148, 2017). Therefore, we decided to test whether adipocyte CB1R are not only necessary but also sufficient for the development of diet-induced steatosis. For this, we are generating a rescue model, i.e. mutant mice with transgenic re-expression of CB1R in adipocytes on a global CB1R-KO background. To achieve this, we generated global ko mice by inserting a stop codon flanked by two loxP sites upstream from the Cnr1 coding sequence. The resulting global CB1ko mice are then crossed with adipoqCre mice, resulting in the selective cleavage of the stop codon in adipocytes only, with the resulting re-expression of normal levels of CB1R on a global CB1ko background. This breeding program is under way. As mentioned above, hepatic CB1R have minimal role in the development of diet-induced steatosis. However, the removal of excess TG from the liver is not a simple reversal of the process of TG accumulation, and the most likely target of CB1R blockade for reversing steatosis is CB1R expressed by hepatocytes. In a project nearing completion, we found that wild-type (wt) and hepatocyte-specific CB1Rko (LCB1ko) mice became similarly obese and steatotic on a high-fat diet (HFD). JD5037 treatment reversed the weight gain similarly in the two strains, but only reversed the steatosis in wt and not in LCB1ko mice, suggesting the involvement of hepatocyte CB1R in the latter process. Targeted gene expression analyses with a focus on lipolytic pathway yielded a single gene, CD36, whose expression in wt mice was strongly induced by HFD, and the induction was fully reversed by JD5037 treatment, whereas in LCB1ko mice HFD caused only a minor, statistically insignificant increase in CD36 expression which remained unaffected by JD treatment. The FA translocase CD36 plays a key role not only in the uptake and transport of FAs into hepatocytes but also in the regulation of AMP kinase (AMPK) activity and its downstream effect in promoting FA oxidation, processes involved in the removal of hepatic TGs. We therefore further analyzed the role of hepatocyte CB1R and its effects on CD36/AMPK signaling in the removal of hepatic TG by the peripheral CB1R antagonist JD5037. rtPCR analyses confirmed the findings of the Nanostring analyses and a similar pattern of changes were found at the protein levels of CD36, analyzed by Western blotting. CD36 is known to inhibit the activation of AMPK. Consistently, we found that AMPK activation, as deduced by its phosphorylation on its threonine-178 residue, was reduced by HFD and the reduction reversed by JD5037 treatment, changes opposite to those of CD36 expression. Interestingly, we found that in lean mice on standard diet, which have low circulating FA levels and their primary energy source is carbohydrates, JD5037 had effects opposite to those in DIO mice, i.e it increased CD36 gene and protein expression and decreased AMPK phosphorylation in wt mice. These effects were again much smaller or non-existent in LCB1ko mice on standard diet. These findings confirm the role of hepatic CB1R in the control of liver TG content by regulating the CD36/AMPK signaling pathway and reveal, for the first time, the substrate-dependent opposite regulation of hepatic lipid metabolism in vivo by hepatocytes. These latter findings may account for the 'paradoxical' observation that chronic cannabis use by non-obese individuals protects from the development of fatty liver disease. A manuscript is being prepared summarizing the above findings.