We compared the effects of oral vanadyl sulfate (100 mg/day) in moderately obese NIDDM and nondiabetic subjects. Three-hour euglycemic-hyperinsulinemic (insulin infusion 30 mU / m / min) clamps were performed after 2 weeks of placebo and 3 weeks of vanadyl sulfate treatment in six nondiabetic control subjects (age 37 +/- 3 years; BMI 29.5 +/- 2.4 kg/m2 ) and seven NIDDM subjects (age 53 +/- 2 years; BMI 28.7 +/-1.8 kg/m2). Glucose turnover ([3-3 H]glucose), glycolysis from plasma glucose, glycogen synthesis, and whole-body carbohydrate and lipid oxidation were evaluated. Decreases in fasting plasma glucose (by approximately 1.7 mmol/l) and HbAlc (both P < 0.05) were observed in NIDDM subjects during treatment; plasma glucose was unchanged in control subjects. In the latter, the glucose infusion rate (GIR) required to maintain euglycemia (40.1 +/- 5.7 and 38.1 +/- 4.8 micromol / kg fat-free mass FFM / min) and glucose disposal (Rd) (41.7 +/- 5.7 and 38.9 +/-4.7 micromol / kg FFM / min were similar during placebo and vanadyl sulfate administration, respectively. Hepatic glucose output (HGO) was completely suppressed in both studies. In contrast, in NIDDM subjects, vanadyl sulfate increased GIR approximately 82% (17.3 +/- 4.7 to 30.9 +/- 2.7 micromol / kg FFM / min, P < 0.05); this improvement in insulin sensitivity was due to both augmented stimulation of Rd (26.0 +/-4.0 vs. 33.6 +/- 2.22 micromol / kg FFM / min, P < 0.05) and enhanced suppression of HGO (7.7 +/- 3.1 vs. 1.3 +/- 0.9 micromol / kg FFM / min, P < 0.05). Increased insulin-stimulated glycogen synthesis accounted for >80% of the increased Rd with vanadyl sulfate (P < 0.005), but plasma glucose flux via glycolysis was unchanged. In NIDDM subjects, vanadyl sulfate was also associated with greater suppression of plasma free fatty acids (FFAs) (P < 0.01) and lipid oxidation (P < 0.05) during clamps. The reduction in HGO and increase in Rd were both highly correlated with the decline in plasma FFA concentrations during the clamp period (P < 0.001). In conclusion, small oral doses of vanadyl sulfate do not alter insulin sensitivity in nondiabetic subjects, but it does improve both hepatic and skeletal muscle insulin sensitivity in NIDDM subjects in part by enhancing insulin's inhibitory effect on lipolysis. These data suggest that vanadyl sulfate may improve a defect in insulin signaling specific to NIDDM.