To define the roles of beta- and alpha-adrenergic receptors in intense exercise, 17 lean healthy fit young males underwent 13.6 +/- 0.2 (+/-SE) min of cycle ergometer exercise: 6 at 100% maximum oxygen uptake (VO2max; MAX), 7 at their maximum possible (87 +/- 2.3%) during iv propranolol (P; 150 micrograms/kg bolus 30 min preexercise, then 80 micrograms/kg.min), and 7 (including 3 of the P subjects) at 87% VO2max (C) as controls for P. Plasma glucose increased from similar resting values to a peak in the early recovery period at 7.2 +/- 0.44 in MAX and 6.8 +/- 0.37 in P, but only 5.2 +/- 0.3 mmol/L in C. The rate of glucose appearance (Ra) rose about 8-fold in both MAX and P, but only 4-fold in C (P = 0.001). The rate of glucose disappearance (Rd) increased 4-fold in MAX, 5.5-fold in P, and 3-fold in C (P = 0.001). Plasma insulin declined during exercise (P < 0.05) in MAX and P, but not in C, whereas plasma glucagon increased modestly in all groups. The mean peak plasma norepinephrine level was 36.3 +/- 4.5 in MAX, 20.2 +/- 3.4 in P, and 15.2 +/- 2.9 nmol/L in C (P = 0.002); epinephrine reached 7141 +/- 1790 in MAX and 5605 +/- 1532 in P (P = NS), but only 1715 +/- 344 pmol/L in C (P = 0.03). Therefore, 1) an "unmasked" alpha-adrenergic effect, directly and/or via an altered glucagon/insulin ratio, probably contributed to increased Ra with P treatment; and 2) the marked facilitation of the increase in Rd with P supports a major role for beta-adrenergic restraint of Rd at this exercise intensity.

We have investigated catecholamine-glucagon-insulin interactions using three stress models: 1) hypoglycemia; 2) exercise; and 3) epinephrine infusion. Phlorizin caused mild hypoglycemia with hypoinsulinemia. Plasma glucagon increased as did hepatic glucose production. Catecholamines did not increase. Insulin caused severe hypoglycemia. Metabolic counterregulation was due mainly to the 40-fold increase in epinephrine. Glucagon played a role only in the recovery from insulin-induced hypoglycemia, which could reflect increased hepatic sensitivity to glucagon with declining plasma insulin. Glucagon suppression during exercise caused transient hypoglycemia due to an inadequate rise in glucose production. Exaggerated epinephrine release during hypoglycemic exercise prevented severe hypoglycemia by inhibiting glucose utilization and stimulating glucose production, with an associated increase in lactate and free fatty acid levels. Hypoglycemic exercise also caused increased cortisol release. Counterregulation was prevented by a euglycemic clamp. We conclude that, during exercise, glucagon is directly responsible for 80% of the increment of glucose production and controls glucose uptake by the muscle indirectly; thus glucagon spares muscle glycogen by increasing hepatic glucose production. Epinephrine infusion in normal dogs caused a transient increase in glucose production and a sustained inhibition of glucose clearance, resulting in hyperglycemia. Insulin rose transiently, followed by a relative inhibition of secretion. Glucagon suppression did not modify the metabolic effects of epinephrine. In alloxan-diabetic dogs, the glucagon response to epinephrine was augmented, whereas in depancreatized dogs, during subbasal insulin infusion, the hepatic response to glucagon was excessive. Glucagon suppression diminished hepatic responsiveness to epinephrine in both models. Stress-induced diabetic instability could relate to exaggerated glucagon release or to increased hepatic sensitivity to glucagon. Thus, during hypoglycemia, exercise, or epinephrine infusion, prevailing plasma insulin levels govern the relative metabolic roles of epinephrine and glucagon.

Stress-induced hyperglycemia can lead to significant deterioration in glycemic control in individuals with diabetes. Previously, we have shown in normal dogs that, after intracerebroventricular (ICV) administration of carbachol (a model of moderate stress), increases in both the metabolic clearance rate (MCR) of glucose and endogenous glucose production (GP) occur. Howe v e r, in hyperglycemic diabetic dogs subjected to the same stress, the MCR of glucose does not increase and glycemia therefore markedly deteriorates because of stimulation of GP. Our aims were to determine the following: 1) whether insulin-induced acute normalization of glycemia, with or without -blockade, would correct glucose clearance and prevent the hyperglycemic eff e c t of stress, and 2) whether hyperinsulinemia per se could correct these abnormalities. Stress was induced by ICV carbachol in 27 experiments in five alloxan-administered diabetic dogs subjected to the following protocols in random order: 1) basal insulin infusion (BI) to restore normoglycemia; 2) basal insulin infusion with -blockade (BI+block); 3) normoglycemic-hyperinsulinemic clamp with threefold elevation of insulin above basal (3 BI); and 4) normoglycemic-hyperinsulinemic clamp with fivefold elevation of insulin above basal ( 5 BI). The BI+block protocol fully prevented stressinduced hyperglycemia, both by increasing MCR ( MCR at peak: 0.72 ± 0.25 m l · k g -1 · min -1 vs. no change in BI, P < 0.05) and by diminishing the stressinduced increment in GP observed in BI ( GP at peak: 3.72 ± 0.09 µmol · k g -1 · min -1 for BI+block vs. 14.10 ± 0.31 µmol · k g -1 · min -1 for BI, P < 0.0001). In contrast, 3 BI and 5 BI treatments with normoglycemichyperinsulinemic clamps proportionately increased basal MCR at baseline, but paradoxically were not associated with an increase in MCR in response to stress, which induced a twofold increase in GP. Thus, in alloxan-administered diabetic dogs, stress increased GP but not MCR, despite normalization of glycemia with basal or high insulin. In contrast, -a d r e n e r g i c blockade almost completely restored the metabolic response to stress to normal and prevented marked hyperglycemia, both by limiting the rise in GP and by increasing glucose MCR. We conclude that acute normalization of glycemia with basal insulin or hyperinsulinemia does not prevent hyperglycemic effects of stress unless accompanied by -blockade, and we speculate that short-term -blockade may be a useful treatment modality under some stress conditions in patients with diabetes. Diabetes 4 9 :2 5 3-262, 2000 T he increased demand for glucose by the brain, heart, and other tissues during stress necessitates adaptive alterations in glucose metabolism. In many forms of stress, the rate of glucose production rises above the rate of glucose clearance, resulting in hyperglycemia. It is well recognized that stress induces a greater derangement in glucose regulation in patients with diabetes than in nondiabetic individuals. Consequently, in diabetes there can be an exaggerated ...