The present study was designed to investigate the mechanisms by which insulin regulates the disposal of an intravenous glucose load in man. A combined tracer-hepatic vein catheter technique was used to quantitate directly the components of net splanchnic glucose balance (NSGB), i.e., splanchnic glucose uptake and hepatic glucose output, and peripheral (extrasplanchnic) glucose uptake. Four different protocols were performed: (a) intravenous infusion of glucose alone (6.5 mg kg(-1) min(-1)) for 90 min (control group); (b) glucose plus somatostatin (0.6 mg/h) and glucagon (0.8 ng kg(-1) min(-1); (c) glucose plus somatostatin, glucagon, and insulin (0.15 mU kg(-1) min(-1)); and (d) glucose plus somatostatin, glucagon, and insulin (0.4 m U kg(-1) min(-1)). In groups 2-4, arterial blood glucose was raised to comparable levels to those of controls ( approximately 170 mg/dl) by a variable glucose infusion. In the control group, plasma insulin levels reached 40 muU/ml at 90 min. NSGB switched from a net output of 1.71+/-0.13 to a net uptake of 1.5-1.6 mg kg(-1) min(-1) due to a 90-95% suppression of hepatic glucose output (P < 0.01) and a 105-130% elevation of splanchnic glucose uptake (from 0.78+/-0.13 to 1.6-1.8 mg kg(-1) min(-1); P < 0.01). Peripheral glucose uptake rose by 150-160% (P < 0.01). In group 2, plasma insulin fell to <5 muU/ml. Net splanchnic glucose output initially rose twofold but later returned to basal values. This response was entirely accounted for by similar changes in hepatic glucose output since splanchnic glucose uptake remained totally unchanged in spite of hyperglycemia. In contrast, peripheral glucose uptake rose consistently by 100% (P < 0.01) despite insulin deficiency. In an additional group of experiments, glucose metabolism by the forearm muscle tissue was quantitated during identical conditions to those of group 2 (hyperglycemia plus insulin deficiency). Both the arterial-deep venous blood glucose difference and forearm glucose uptake increased markedly by 300-400% (P < 0.05 - <0.01). In group 3, plasma insulin was maintained at near-basal, peripheral levels (12-14 muU/ml). Hepatic glucose output decreased slightly by 35-40% (P < 0.05) while splanchnic glucose uptake remained unchanged. Consequently, the net glucose overproduction seen in group 2 was totally prevented although NSGB still remained as a net output. In group 4, peripheral insulin levels were similar to those of the control group (35-40 muU/ml). The suppression of hepatic glucose output was more pronounced (60-65%) and splanchnic glucose uptake rose consistently by 65% (P < 0.01). Consequently, NSGB did not remain as a net output but eventually switched to a small uptake (0.3 mg kg(-1) min(-1)). Peripheral glucose uptake rose to the same extent as in controls. IT IS CONCLUDED THAT: (a) the suppressive effect of hyperglycemia on hepatic glucose output is strictly dependent on the degree of hepatic insulinization; (b) insulin plays an essential role in promoting splanchnic glucose uptake after an intravenous glucose load whereas hyperglycemia per se is totally unable to activate this process; (c) peripheral glucose uptake is markedly stimulated by hyperglycemia even in the face of insulin deficiency. Direct evidence also demonstrates that the skeletal muscle is involved in this response. Our data, thus, indicate that insulin rather than hyperglycemia regulates splanchnic glucose disposal in man. On the other hand, hyperglycemia per se appears to be an important regulator of glucose disposal by peripheral tissues.