The role of the peripheral endocannabinoid system in insulin resistance and obesity
There is increasing appreciation that the endocannabinoid system (ECS) influences multiple physiological processes, including those important for metabolic and energy homeostasis, via both central and peripheral actions. Major components of this system include the cannabinoid [CB1 and CB2] G-protein coupled receptors and their endogenous lipid-derived ligands, anandamide (AEA) and 2-arachidonoylglycerol (2-AG), which are synthesised and released ‘on-demand’ in response to increased intracellular calcium. Although CB1 is abundantly expressed in the brain it is also present in significant amounts in peripheral tissues such as liver, pancreas, adipose tissue and skeletal muscle of rodents and humans. CB2, by contrast, is primarily expressed in cells and tissues involved with immune function (i.e. spleen and leukocytes). An increase in ECS tone within peripheral tissues has been implicated in the control of numerous processes as highlighted in the figure below.
The importance of the ECS as a regulator of metabolism and energy balance is underscored by the observation that mice deficient in CB1 are resistant to diet-induced obesity, and that chronic CB1 blockade using rimonabant induces weight loss independently of its anorectic action. Indeed, there is now a significant evidential base showing that the ECS can regulate lipid and glucose metabolism by direct effects on peripheral tissues. For example, in liver, HU210, a mixed CB1/CB2 agonist, induces hepatic fatty acid synthesis – an effect not observed in CB1 deficient mice that are also protected against diet-induced hepatic steatosis. In adipose tissue, treatment with AEA or agonist-induced CB1 stimulation induces lipogenesis and represses mitochondrial biogenesis. In skeletal muscle, CB1 activation impairs insulin-dependent MAPK signalling, whereas rimonabant not only antagonises this effect, but stimulates glucose uptake independently of insulin by upregulating PI3-kinase signalling. Importantly, it is noteworthy that CB1 blockade using peripherally acting neutral antagonists that exhibit poor brain penetrance also promote significant weight reduction and improve glucose tolerance in obese rodents thus paving the path for testing their efficacy as anorectic agents in human subjects.
The finding that CB1 blockade exerts “insulin-like” effects implies that increased peripheral ECS activity would normally be insulin-desensitising. Indeed, increased levels of AEA and 2-AG have been correlated with increased adiposity and elevated CB1 expression has been documented in liver, pancreas and adipose tissue of obese individuals. In line with such findings, we have discovered that increased peripheral endocannabinoid tone may also be a feature of aging and one contributing to age-related insulin resistance. Analysis of PKB/Akt-directed insulin signalling in skeletal muscle, liver and adipose tissue of young (4 month old) and aged (16 month) mice reveals a significant reduction in insulin signalling capacity in aged animals, which is associated with elevated tissue CB1 expression. Strikingly, we find that two weeks of rimonabant administration induces significant gains in tissue insulin signalling in aged, but not young mice. Whilst the drug causes a modest anorectic effect as reported in previous studies, it suppresses expression of proadipogenic genes (FAS, SREBP-1, PPARγ2), whilst inducing that of the lipolytic enzyme, ATGL (adipose triglyceride lipase) in adipose tissue of older mice. These changes coincide with significantly reduced body fat mass in aged but not young mice in response to CB1 blockade. In addition, rimonabant conveys an anti-inflammatory effect based on increased adipocyte abundance of IkBα (a key repressor of NFkB dependent transcription) and reduced IL-6 mRNA expression, a proinflammatory cytokine whose expression is regulated by NFkB.
Precisely how over-activation of CB1 contributes to the decline in insulin sensitivity and metabolic capacity in obese and aged animals is poorly understood, but mounting evidence suggests that ECS-induced disturbances in mitochondrial function and integrity promote dysregulation of fuel and energy homeostasis. Intriguingly, whilst the precise signalling mechanisms by which CB1 blockade brings about these beneficial metabolic changes remain unclear, recent and emerging data suggests that reducing CB1 activity enhances AMPK-directed signalling, mitochondrial mass and respiratory function and that these may help counter the development of insulin resistance as seen during aging and obesity.
In addition to our work on CB1 and CB2, we are also interested in the role played by GPR55, a G-protein coupled receptor that was initially deorphanised as a cannabinoid receptor in the regulation of tissue inflammation, energy homeostasis and metabolism in skeletal muscle. The receptor, however, shares little homology with CB1 or CB2 and was subsequently shown to bind lysophophatidylinositol (LPI), as its physiological ligand that intriguingly does not interact with CB1 or CB2. In addition to LPI, the receptor can also be activated by oleoylethanolamide (OEA) and palmitoyl-ethanolamine (PEA) suggesting that it may function as a novel lipid sensing receptor. We (and others) have found that binding of these ligands, as well as the synthetic GPR55 agonist, O-1602, induces changes in intracellular Ca2+ as well as signalling via Rho kinase and the classical extracellular signal-regulated kinase 1/2 (ERK) pathway in cell based studies. These events are likely to be triggered by the activated receptor coupling to Gα12/13 and Gαq proteins and whilst these signalling responses may be cell type dependent they are repressed by the presence of synthetic GPR55 antagonists such as CID16020046 or ML193. Our studies currently evaluating what role GPR55 plays in the control cellular energy balance as well as cell/tissue inflammation.
For more information on our work in this area click on the links to following Journals.