The adipocyte hormone leptin reduces food intake in normal animals. During uncontrolled type 1 diabetes, plasma leptin levels fall, whereas food intake increases. To test the hypothesis that low leptin levels contribute to diabetic hyperphagia, we investigated the effect on food intake of replacement of leptin at basal plasma concentrations for 7 days in Long-Evans rats with uncontrolled diabetes induced by streptozotocin (STZ). One group of STZ diabetic rats received saline (STZ + Sal) (n = 11), while the other group (STZ + Lep) (n = 15) received a subcutaneous infusion of recombinant rat leptin (100 microg x kg(-1) x day(-1)) via osmotic minipumps. A nondiabetic control group (Con) (n = 11) received saline only. In the STZ + Sal group, plasma leptin levels decreased by 75% (P < 0.05) from 2.4+/-0.5 on the day before STZ/citrate buffer vehicle (Veh) injection (day 0) to 0.6+/-0.2 ng/ml on day 7. In contrast, plasma leptin levels on days 3-7 were comparable to pretreatment values in both the STZ + Lep group (day 0: 2.6+/-0.4 vs. day 7: 2.5+/-0.3 ng/ml, NS) and the Con group (day 0: 3.8+/-0.4 vs. day 7: 2.9+/-1.0 ng/ml, NS). In the STZ + Sal group, daily food intake increased gradually to values 43% above basal by day 7 (day 0: 24+/-2 to day 7: 33+/-3 g, P < 0.05), whereas food intake did not increase in either the STZ + Lep group (day 0: 24+/-1 vs. day 7: 21+/-2 g, NS), or the Con group (day 0: 23+/-1 vs. day 7: 23+/-2 g). Plasma glucose levels exceeded nondiabetic control values (7.7+/-0.2 mmol/l) in both diabetic groups, but were lower in the STZ + Lep group (17.2+/-1.8 mmol/l) than in the STZ + Sal group (24.3+/-1.1 mmol/l, P < 0.05). To determine if sensitivity to leptin-induced anorexia was affected by STZ treatment, a second experiment was performed in which the effect of intracerebroventricular leptin injection (at doses of 0.35, 1.0, or 3.5 microg) on food intake was measured 10 days after STZ or Veh treatment. Leptin suppressed both 4- and 24-h food intake in the two groups to an equal extent at every dose (by 15, 22, and 35%, respectively). These findings support the hypothesis that the effect of uncontrolled diabetes to lower leptin levels contributes to diabetic hyperphagia and that this effect is not due to altered leptin sensitivity.