Thiazolidinediones (TZDs) improve glycemic control and insulin sensitivity in patients with type 2 diabetes. To determine whether the TZD-induced improvement in glycemic control is associated with enhanced insulin receptor signaling in skeletal muscle, 20 type 2 diabetic patients received a 75-g oral glucose tolerance test (OGTT) and euglycemic insulin (80 mU x m(-2) x min(-1)) clamp with [3-(3)H]glucose/indirect calorimetry/vastus lateralis muscle biopsies before and after 16 weeks of rosiglitazone treatment. Six age-matched nondiabetic subjects served as control subjects. RSG improved fasting plasma glucose (185 +/- 8 to 139 +/- 5 mg/dl), mean plasma glucose during the OGTT (290 +/- 9 to 225 +/- 6 mg/dl), HbA(1c) (8.5 +/- 0.3 to 7.1 +/- 0.3%), insulin-mediated total-body glucose disposal (TGD) (6.9 +/- 0.7 to 9.2 +/- 0.8 mg x kg(-1) fat-free mass x min(-1)) (all P < 0.001), and decreased fasting plasma free fatty acid (FFA) (789 +/- 59 to 656 +/- 50 micro Eq/l) and mean FFA during the OGTT (644 +/- 41 to 471 +/- 35 micro Eq/l) (both P < 0.01). Before RSG treatment, insulin infusion did not significantly increase insulin receptor tyrosine phosphorylation (0.95 +/- 0.10 to 1.08 +/- 0.13 density units; NS) but had a small stimulatory effect on insulin receptor substrate (IRS)-1 tyrosine phosphorylation (1.05 +/- 0.10 to 1.21 +/- 0.12 density units; P < 0.01) and the association of p85 with IRS-1 (0.94 +/- 0.06 to 1.08 +/- 0.06 activity units; P < 0.01). RSG therapy had no effect on basal or insulin-stimulated insulin receptor tyrosine phosphorylation but increased insulin stimulation of IRS-1 tyrosine phosphorylation (1.13 +/- 0.11 to 1.56 +/- 0.17 density units; P < 0.01 vs. prerosiglitazone) and p85 association with IRS-1 (1.00 +/- 0.06 to 1.27 +/- 0.07 activity units; P < 0.05 vs. prerosiglitazone). In control and type 2 diabetic subjects, TGD/nonoxidative glucose disposal correlated positively with the insulin-stimulated increments in IRS-1 tyrosine phosphorylation (r = 0.52/r = 0.57, P < 0.01) and inversely with the plasma FFA concentration during the insulin clamp (r = -0.55/r = -0.53, P < 0.01). However, no significant association between plasma FFA concentrations during the insulin clamp and the increment in either IRS-1 tyrosine phosphorylation or the association of p85 with IRS-1 was observed. In conclusion, in type 2 diabetic patients, rosiglitazone treatment enhances downstream insulin receptor signaling in muscle and decreases plasma FFA concentration while improving glycemic control.