UCP1-independent glucose-lowering effect of leptin in type 1 diabetes: only in conditions of hypoleptinemia.

The possibility to use leptin therapeutically for lowering glucose levels in patients with type 1 diabetes has attracted interest. However, earlier animal models of type 1 diabetes are severely catabolic with very low endogenous leptin levels, unlike most patients with diabetes. Here, we aim to test glucose-lowering effects of leptin in novel, more human-like murine models. We examined the glucose-lowering potential of leptin in diabetic models of two types: streptozotocin-treated mice and mice treated with the insulin receptor antagonist S961. To prevent hypoleptinemia, we used combinations of thermoneutral temperature and high-fat feeding. Leptin fully normalized hyperglycemia in standard chow-fed streptozotocin-treated diabetic mice. However, more humanized physiological conditions (high-fat diets or thermoneutral temperatures) that increased adiposity - and thus also leptin levels - in the diabetic mice abrogated the effects of leptin, i.e., the mice developed leptin resistance also in this respect. The glucose-lowering effect of leptin was not dependent on the presence of the uncoupling protein-1 and was not associated with alterations in plasma insulin, insulin-like growth factor 1, food intake or corticosterone but fully correlated with decreased plasma glucagon levels and gluconeogenesis. An important implication of these observations is that the therapeutic potential of leptin as an additional treatment in patients with type 1 diabetes is probably limited. This is because such patients are treated with insulin and do not display low leptin levels. Thus, the potential for a glucose-lowering effect of leptin would already have been attained with standard insulin therapy, and further effects on blood glucose level through additional leptin cannot be anticipated.