The response of serum leptin to short (4 days) and prolonged (28 days) energy restriction (50% reduction in energy intake) was determined in 18 (9 male, 9 female) moderately obese humans (body mass index 32.0 +/- 0.6 kg/m2 mean +/- SEM), 9 of whom had mild non-insulin-dependent diabetes mellitus (NIDDM). Body composition was assessed before and at the end of the energy restriction using DEXA. The subjects lost a measured 2.6 +/- 0.4 kg of body fat after 28 days and an estimated 0.3 kg at 4 days. Serum leptin fell to 64 +/- 3% of baseline levels at day 4 and further to 46 +/- 4% at day 28. In a multiple correlation analysis, the change in leptin concentration at day 4 was significantly related to the change in dietary carbohydrate intake (partial r = 0.68, p < 0.005) but not to changes in fat (r = 0.12) or protein (r = 0.02) intakes. There was a 1:1 relationship between the changes in leptin and dietary carbohydrate (regression slope = 1.0 +/- 0.3). Gender, or the presence of NIDDM had no effects on these responses. This pronounced fall in serum leptin in association with reduced carbohydrate intake before substantial loss of body fat suggests a role for leptin in defending the body's carbohydrate stores and implicates leptin in the satiating effects of carbohydrate. Dietary or other interventions which maintain leptin levels during weight reduction may lead to improvements in weight loss.