In intact rats 16 h of fasting reduced the plasma insulin response to i.v. stimulation with either glucose, tolbutamide or glibenelamide by 50--80 %, without affecting the extractable insulin content of the pantreas. In subsequent studies with the isolated perfused rat panereas two distinct patterns of insulin release could be discerned during the secondary phase. In the fed state, glucose 1.5 mg/ml induced a more or less constant elevation of the insulin secretion rate over 30 rain (type I). At glucose concentrations of 2 and 3 mg/ml the release pattern was characterized by progressively increasing secretion rates (type II pattern). Infusion of tolbutamide (0.2 mg/ml) lowered the threshold for glueose stimulation and induced both patterns of secretion at lower glucose concentrations. Fasting for 24~ h caused a 70--80 % in-hibition of insulin secretion per 30 rain a~ glucose levels of 1.5 and. 2 mg/ml. Decreased glucose sensitivity was indicated by a shift to the right of the entire dose-response curve for glucose and by reduced inhibition (30 %) at a glucose level of 3 mg/ml. The effect of tolbutamide was also strongly diminished. The percent inhibition of the response to tolbutamide at different levels of glucose showed a pattern of inhibition similar to that observed with glucose alone. These findings suggest that the glucose-dependent release mechanism is highly sensitive to relatively short periods of fasting.

The release of insulin from ligated rat pancreas in vitro was measured after pre-incubation of the tissue at either high or low glucose concentrations. Insulin release was estimated by determining the glucose uptake of rat hemidiaphragms after joint incubation with the pancreatic tissue. Addition of bovine growth hormone (GH) to the pre-incubation medium in a concentration of 0.1 mg/ml significantly increased the response to glucose stimulation. Pancreatic tissue of hypophysectomized rats failed to respond to GH in vitro, although previous hypophysectomy was shown to reduce the response to glucose stimulation. Pretreatment of hypophysectomized rats with GH in doses which restored the growth rate to normal, caused the reappearance of the effect of GH in vitro. The glucose sensitivity of pancreatic tissue was also increased by pretreatment with GH in vivo alone. It is concluded from these findings that GH may increase the sensitivity of the β-cells to glucose by direct action. Since hypophysectomy exerts an opposite effect, it is suggested that physiological levels of endogenous GH may be involved in the regulation of insulin release.