Role of Gs-alpha in regulation of adipose tissue growth and function

Weinstein, Lee

Abstract

We generated mice with Gs-alpha deficiency in adipose tissue (FGsKO mice) by repeated matings of aP2-cre recombinase transgenic mice with floxed Gs-alpha mice which have loxP recombination sites surrounding Gs-alpha exon 1. FGsKO mice had poor survival and decreased linear growth, particularly those mice in which Gs-alpha expression in adipose tissue was severely reduced. The cause of these effects are unclear, although insulin-like growth factor 1 (IGF1) levels were reduced by 50% in FGsKO mice. In this subset of mice, white adipose tissue (WAT) pads were almost absent. Fibroblasts from FGsKO embryos had significantly reduced adipogenic conversion in the presence of agents known to induce adipogenesis. These in vivo and in vitro results confirm that Gs-alpha is critical for normal adipogenesis. FGsKO mice which survived had also had reduced relative fat mass with smaller WAT pads and smaller adipocytes with less lipid content per cell. In contrast, the interscapular brown adipose tissue (BAT) pads were larger than normal and the cells had increased lipid stores with a more unilocular distribution. This histological pattern in BAT is consistent with reduced metabolic activation presumably due to an inability of the sympathetic nervous system to stimulate lipolysis via beta-adrenergic/Gs-alpha pathways. Consistent with this, the expression of PGC-1, uncoupling protein 1 (UCP1) and other genes associated with lipid metabolism were markedly reduced in BAT from FGsKO mice. FGsKO mice were hypoglycemic and hypoinsulinemic relative to controls and had improved glucose tolerance and insulin sensitivity on both regular and high-fat diets, consistent with their lean phenotype. Two forms of adaptive thermogenesis, cold- and diet-induced thermogenesis, are mediated by increased sympathetic nervous system activity. FGsKO mice are cold intolerant and cold-induced thermogenesis is proabably markedly impaired as FGsKO mice placed in a cold environment do not maintain their body temperature or raise their expression of UCP1 in BAT. This is consistent with the known role for BAT in cold-induced thermogenesis and the results in FGsKO mice that BAT fails to be activated by sympathetic stimulation despite the fact that their sympathetic activity, as determined by urine catecholamine levels, was markedly increased. In contrast, diet-induced thermogenesis is maintained and in fact greater than normal in FGsKO mice based upon the observations that FGsKO mice fail to gain weight on a high-fat diet and have markedly increase their energy expenditure on a high-fat diet. These results suggest that cold- and diet-induced thermogenesis can occur in separate tissues and we propose that muscle is the main site for diet-induced thermogenesis in these mice. To further examine this we looked at sympathetic activity and PGC-1alpha induction after either acute cold or high fat diet in control mice. In response to high fact diet sympathetic activity only increased in skeletal muscle, with no change in heart, liver, or brown fat. In contrast sympathetic activity increased in all tissues after acute cold exposure. PGC-1alpha was markedly induced in BAT after cold exposure and much less so after high fat diet. Finally, these results as well as results in heterozygous FGsKO mice suggest that adipose tissue is not the site whereby germline Gs-alpha mutations on the maternal allele lead to severe obesity and insulin resistance. More recently we knocked out Gs-alpha using adiponectin-cre and although these mice show evidence of brown adipose tissue dysfunction with cold intolerance and lack of formation of beige fat in white adipose tissue stores, the mice had relatively normal body weight and fat mass. This is also in spite of the fact that these mice had impaired metabolic responses to beta 3-adrenergic receptor agonists. This model also suggests that rates of lipolysis and lipogenesis may be coupled to each other in adipose tissue. Despite the lack of effect on adiposity, these mice did have altered insulin sensitivity and peripheral glucose metabolism, showing that adipose tissue can affect glucose metabolism in the absence of a change in adiposity, which may be the result of difference in expression of the adipokine fatty acid binding protein 4 (FABP4). The differences between these models may be that in the latter Gs-alpha is deleted at a later point in adipose cell differentiation, allowing the formation of mature white adipose tissue. Studies are also ongoing examining the role of different G protein pathways in the central nervous system on white adipose tissue browning and BAT thermogenesis. In addition studies are ongoing examining the effects of knocking out Gs-alpha in both skeletal muscle and adipose tissue on energy balance and cold-induced thermogenesis.