Effect of hyperinsulinemia on plasma leptin concentrations and food intake in rats.

We investigated the dose- and time-dependent effect of insulin infusion on peripheral and portal plasma leptin concentrations in normal rats. Three groups were studied: group I: euglycemic (6 mmol/l) insulin (6 mU . kg-1 . min-1) clamps for 0, 2, 4, 12, and 24 h; group II: euglycemic insulin (18 mU . kg-1 . min-1) clamp for 2 h; and group III: euglycemic insulin (3 mU . kg-1 . min-1) clamp for 7 days. In group III, food intake was quantified during days 1-7. After an overnight fast, peripheral and portal plasma leptin levels were identical (1.5 +/- 0.2 and 1.6 +/- 0.2 ng/ml). Insulin infusion (6 mU . kg-1 . min-1) for 2 h had no effect on plasma leptin levels (1.5 +/- 0.2 ng/ml). After 4 h (2.0 +/- 0.2 ng/ml), 12 h (2.2 +/- 0. 4 ng/ml), and 24 h (2.7 +/- 0.6 ng/ml; all P < 0.05) of insulin infusion, a progressive time-related increase in plasma leptin concentration was observed; portal vein leptin levels rose in parallel and were similar to peripheral levels. When insulin (18 mU . kg-1 . min-1) was infused for 2 h, plasma leptin levels increased to 3.0 +/- 0.3 ng/ml (P < 0.01). Seven days of constant insulin infusion (3 mU . kg-1 . min-1) resulted in a progressive increase in fasting plasma leptin and a parallel decrease in food intake. A mean increase in plasma leptin concentration of 1 ng/ml during the 7-day insulin infusion period was associated with a mean decrease in food intake of 2.5 g/day (multivariate ANOVA, P < 0.05). We conclude that the insulin-induced rise in peripheral and portal vein leptin levels is similar and both dose and time dependent. The inverse relationship between plasma leptin concentration and food intake during prolonged hyperinsulinemia, but not during short-term hyperinsulinemia, supports the role of leptin in long-term food consumption.