Skeletal muscle inducible nitric oxide synthase (NOS) protein is greatly elevated in people with type 2 diabetes mellitus, whereas endothelial NOS is at normal levels. Diabetic rat studies suggest that skeletal muscle neuronal NOS (nNOS) micro protein expression may be reduced in human insulin resistance. The aim of this study was to determine whether skeletal muscle nNOSmicro protein expression is reduced in people with impaired glucose homeostasis and whether exercise training increases nNOSmicro protein expression in these individuals because exercise training increases skeletal muscle nNOSmicro protein in rats. Seven people with type 2 diabetes mellitus or prediabetes (impaired fasting glucose and/or impaired glucose tolerance) and 7 matched (sex, age, fitness, body mass index, blood pressure, lipid profile) healthy controls aged 36 to 60 years participated in this study. Vastus lateralis muscle biopsies for nNOSmicro protein determination were obtained, aerobic fitness was measured (peak pulmonary oxygen uptake [Vo(2) peak]), and glucose tolerance and insulin homeostasis were assessed before and after 1 and 4 weeks of cycling exercise training (60% Vo(2) peak, 50 minutes x 5 d wk(-1)). Skeletal muscle nNOSmicro protein was significantly lower (by 32%) in subjects with type 2 diabetes mellitus or prediabetes compared with that in controls before training (17.7 +/- 1.2 vs 26.2 +/- 3.4 arbitrary units, P < .05). The Vo(2) peak and indicators of insulin sensitivity improved with exercise training in both groups (P < .05), but there was no effect of exercise training on skeletal muscle nNOSmicro protein in either group. In conclusion, individuals with impaired glucose homeostasis have reduced skeletal muscle nNOSmicro protein content. However, because exercise training improves insulin sensitivity without influencing skeletal muscle nNOSmicro protein expression, it seems that changes in skeletal muscle nNOSmicro protein are not central to the control of insulin sensitivity in humans and therefore may be a consequence rather than a cause of diabetes.