OBJECTIVE: Endogenous low-molecular-weight glycans containing pinitol (3-O-methyl-D-chiro-inositol) and D-chiro-inositol are thought to mediate certain actions of insulin. We tested the hypothesis that oral administration of soybean-derived pinitol would improve insulin sensitivity in obese subjects (BMI = 36.6 kg/m2) with diet-treated type 2 diabetes or glucose intolerance (HbA1c = 6.8%). RESEARCH DESIGN AND METHODS: There were 22 subjects randomized to receive either pinitol 20 mg x kg(-1) x day(-1) (n = 12) or placebo (n = 10) in a 28-day double-blinded trial. RESULTS: No toxicity due to the pinitol was observed during the study The sensitivity of glucose and lipid metabolism to insulin were assessed by measurement of whole-body glucose, palmitate, and glycerol kinetics during basal conditions and a hyperinsulinemic-euglycemic clamp. Metabolic measurements were made at baseline and again at the end of the study Final plasma levels of pinitol were 48-fold (1.06 +/- 0.15 vs. 0.02 +/- 0.01 micromol/l, P < 0.0001) and D-chiro-inositol levels 14-fold (0.56 +/- 0.08 vs. 0.04 +/- 0.02 micromol/l, P < 0.0001) greater in the pinitol group compared with placebo. CONCLUSIONS: Four weeks of pinitol treatment did not alter baseline glucose production, insulin-mediated glucose disposal, or rates of appearance of free fatty acids and glycerol in plasma. We conclude that plasma levels of both pinitol and D-chiro-inositol are very responsive to pinitol ingestion, but insulin sensitivity does not increase after pinitol treatment in individuals with obesity and mild type 2 diabetes.