Effects of gastric inhibitory polypeptide on leucine- and arginine-stimulated insulin release.

Insulin release stimulated by single amino acids, with and without gastric inhibitory polypeptide (GIP), was studied in vivo in anesthetized rats and in vitro in collagenase-digested isolated rat pancreatic islets. Insulin release in vivo during a 15-min intravenous infusion of leucine (0.97 mM) or arginine (0.97 mM) with GIP (17 ng/min) was greater than with infusion of either amino acid or GIP alone. Serum immunoreactive GIP was in the physiological range, and serum glucose showed no significant change in these studies. In contrast, insulin release did not occur with valine infused alone or with GIP. Insulin release in vitro by isolated islets was greater during a 45-min incubation period with leucine (5-20 mM) or arginine (20 mM) plus GIP (10 ng to 10 micrograms/ml) than with either amino acid alone. This effect that occurred in the absence of glucose could not be demonstrated with low concentrations of leucine (1.5 mM) or with 20 mM valine. In vitro insulin release during paired perfusion studies of isolated islets was greater with leucine (20 mM) plus GIP (50 ng/min) than with leucine alone, and augmentation of insulin release by GIP was demonstrable in both the early and late phases of insulin release. From these results it is concluded that insulin release is greater, both in vivo and in vitro, after leucine or arginine plus GIP than with either amino acid or GIP alone. This effect appears to be specific for the insulinotropic amino acids tested, does not appear to be glucose dependent, and can be demonstrated in both the early and late phases of insulin release.