Exercise training improves insulin sensitivity in subjects with and without type 2 diabetes. However, the mechanism by which this occurs is unclear. The present study was undertaken to determine how improved insulin signaling, GLUT4 expression, and glycogen synthase activity contribute to this improvement. Euglycemic clamps with indirect calorimetry and muscle biopsies were performed before and after 8 weeks of exercise training in 16 insulin-resistant nondiabetic subjects and 6 type 2 diabetic patients. Training increased peak aerobic capacity (Vo(2peak)) in both nondiabetic (from 34 +/- 2 to 39 +/- 2 mL O(2)/kg fat-free mass [FFM]/min, 14% +/- 2%, P <.001) and diabetic (from 26 +/- 3 to 34 +/- 3 mL O(2)/kg FFM/min, 32% +/- 4%) subjects. Training also increased insulin-stimulated glucose disposal in nondiabetic (from 6.2 +/- 0.5 to 7.1 +/- 0.7 mg/kg FFM/min) and diabetic subjects (from 4.3 +/- 0.6 to 5.5 +/- 0.6 mg/kg FFM/min). Total glycogen synthase activity was increased by 46% +/- 17% and 45% +/- 12% in nondiabetic and diabetic subjects, respectively, in response to training (P <.01 v before training). Moreover, after training, glycogen synthase fractional velocity was correlated with insulin-stimulated glucose storage (r = 0.53, P <.05) and the training-induced improvement in glucose disposal was accounted for primarily by increased insulin-stimulated glucose storage. Training also increased GLUT4 protein by 38% +/- 8% and 22% +/- 10% in nondiabetic and diabetic subjects, respectively (P <.05 v. before training). Akt protein expression, which was decreased by 29% +/- 3% (P <.05) in the diabetic subjects before training (compared to the nondiabetics), increased significantly in both groups (P <.001). In contrast, exercise training did not enhance the ability of insulin to stimulate insulin receptor substrate-1 (IRS-1)-associated phosphatidylinositol 3 (PI 3)-kinase activity. The present data are consistent with a working model whereby 8 weeks of exercise training increases insulin-stimulated glucose disposal primarily by increasing GLUT4 protein expression without enhancing insulin-stimulated PI 3-kinase signaling, and that once the glucose enters the myocyte, increased glycogen synthase activity preferentially shunts it into glycogen synthesis.