Physiological insulin action is opposed by β-adrenergic mechanisms in dogs

To investigate the possible role of adrenergic mechanisms in modulating glucose homeostasis during physiological insulin changes, we studied the effects of α-, β-, or combined α- and β-adrenergic blockade on glucose production (R(a)) and utilization (R(d)) via isotope ([3-³H]glucose) dilution during nonstressful, nonhypoglycemic conditions in response to physiological insulin changes in conscious dogs. Without adrenergic blockade, infusion of insulin at 0.275 mU · kg^-1 · min^-1 (control) caused glucose to fall from 92 ± 4 to 82 ± 4 mg/dl over 30 min, because of transient fall in R(a) from 2.8 ± 0.4 to 2.3 ± 0.3 mg · kg^-1 · min^-1, which recovered to base line by 30 min. There was a later rise in R(d) to 3.9 ± 0.4 mg · kg^-1 · min^-1 at 45 min, but no counter-regulatory hormonal changes (glucagon, cortisol, epinephrine, and norepinephrine) to account for these findings in glucose kinetics. α-Blockade alone led to an initial rise in base-line insulin and consequent fall in glucose, associated with a transient fall in R(a) but no change in R(d); infusion of insulin led to a further small fall in glucose, with no change in R(a), but with a rise at 30 min in R(d) similar to controls. β-Blockade alone led to an initial fall in insulin and modest rise in glucose; insulin infusion led to a greater rate of fall in glucose than in controls (from 112 ± 6 to 78 ± 7 mg/dl over 30 min). This greater fall in glucose was due to a transient fall in R(a), as in controls, but an earlier and greater rise in R(d) than controls (to 5.0 ± 0.9 mg · kg^-1 · min^-1 by 30 min). During combined α- and β-blockade, insulin infusion resulted in a fall in glucose over 30 min similar to controls. However, this fall in glucose was entirely due to a greater and earlier rise in R(d) (to 4.8 mg · kg^-1 · min^-1), whereas R(a) did not change. Other than adrenergic blockade-induced rises in catecholamines, there were no counter-regulatory hormone responses to account for these findings. We conclude that at physiological, nonstress levels of hyperinsulinemia in dogs 1) insulin-enhanced R(d) is opposed by β-adrenergic mechanisms since R(d) increased during β- or combined α- and β-blockade; and 2) the recovery of R(a) is not dependent on adrenergic mechanisms, since R(a) returned to base line regardless of adrenergic blockade.