In vivo glucose uptake can occur via two mechanisms, namely, insulin-mediated glucose uptake (IMGU) and non-insulin-mediated glucose uptake (NIMGU). Although the principal tissue sites for IMGU are skeletal muscle, the tissue sites for NIMGU at a given serum glucose concentration are not known. To examine this issue, rates of whole body glucose uptake (Rd) were measured at basal and during glucose clamp studies performed at euglycemia (approximately 90 mg/dl) and hyperglycemia (approximately 220 mg/dl) in six lean healthy men. Studies were performed during hyperinsulinemia (approximately 70 microU/ml) and during somatostatin-induced insulinopenia to measure IMGU and NIMGU, respectively. During each study, leg glucose balance (arteriovenous catheter technique) was also measured. With this approach, rates of whole body skeletal muscle IMGU and NIMGU can be estimated, and the difference between overall Rd and skeletal muscle glucose uptake represents non-skeletal muscle Rd. The results indicate that approximately 20% of basal Rd is into skeletal muscle. During insulinopenia approximately 86% of body NIMGU occurs in non-skeletal muscle tissues at euglycemia. When hyperglycemia was created, whole body NIMGU increased from 128 +/- 6 to 213 +/- 18 mg/min (P less than 0.01); NIMGU into non-skeletal muscle tissues was 134 +/- 11 and 111 +/- 6 mg/min at hyperglycemia and euglycemia, respectively, P = NS. Therefore, virtually all the hyperglycemia induced increment in NIMGU occurred in skeletal muscle. During hyperinsulinemia, IMGU in skeletal muscle represented 75 and 95% of body Rd, at euglycemia and hyperglycemia, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)