AKT controls adipocyte function and systemic metabolism

Adipose tissue is a key regulator of energy homeostasis. Diseases with an increase or decrease in adiposity result in perturbations of systemic metabolism. The insulin signaling and insulin-like growth factor 1 (IGF-1) cascades are vital to the function of many tissues during development and in the mature organism. AKT, a Ser/Thr kinase, is a central node in the insulin and IGF-1 pathways. In the liver, much is known about the consequences when insulin-AKT signaling is lost, but adipose tissue has presented a unique challenge to study in vivo. The extant evidence suggests that manipulating the insulin-signaling cascade in adipose causes systemic insulin resistance (IR), but often these models exhibit a lipodystrophy, implicating insulin/IGF-1 signaling as required for adipocyte differentiation or maintenance. In vitro work has identified the insulin pathway as necessary for adipocyte differentiation. The mechanism through which adipocyte IR causes systemic IR has remained elusive. We used the AdipoQ-Cre to delete Akt1 and Akt2 simultaneously (Adipo-AKT KO) and individually (Adipo-AKT1 KO, Adipo-AKT2 KO) to understand how loss of AKT signaling affects the mature adipocyte. Adipo-AKT KO mice have a severe lipodystrophy with hyperinsulinemia, hepatosteatosis, and perturbed glucose homeostasis. Adipo-AKT1 KO mice have no obvious defects, but Adipo-AKT2 KO mice have systemic IR, with hyperinsulinemia, elevated free fatty acids (FFAs), and blunted insulin-stimulated glucose uptake in adipocytes. This defect has been proposed to account for the observed IR in models with loss of adipocyte insulin signaling, supported by work using the aP2-Cre to delete glucose transporter 4 (aP2-Glut4 KO), the insulin-responsive glucose transporter, but subsequent work has shown the aP2-Cre is not adipocyte-specific. We tested this mechanism by generating mice with AdipoQ-Cre deletion of Glut4 (Adipo-Glut4 KO). These mice do not have systemic IR, unlike the aP2-Glut4 KO. We conclude that loss of adipocyte glucose uptake cannot account for the IR in the Adipo-AKT2 KO, and that the most likely candidate for the propagation of IR from the adipocyte to the liver is elevated FFAs, which promote hepatic glucose production. We propose that loss of AKT2 causes elevated FFAs because of blunted suppression of lipolysis in response to insulin