The GK rat is a spontaneous model of NIDDM. The insulin response to 16.7 mmol/l glucose was markedly impaired in both isolated perfused pancreas and isolated islets from GK rats compared with control Wistar rats. Depolarization with 30 mmol/l KCl in the presence of 3.3 mmol/l glucose and 250 micromol/l diazoxide induced similar insulin responses in perfused pancreases of GK and control rats. In contrast, the glucose-stimulated insulin release was also severely impaired in GK pancreases in the depolarized state. Forskolin (1 micromol/l) markedly enhanced insulin release at 3.3 mmol/l glucose in GK but not control pancreases (54 +/- 15 vs. 3 +/- 1 pmol/10 min, P < 0.001). Dibutyryl cAMP (1 mmol/l) exerted effects similar to forskolin on insulin release in the perfused pancreas. In depolarized pancreases of GK but not control rats, forskolin also induced a marked insulin response at 3.3 mmol/l glucose (163 +/- 48 vs. 16 +/- 1 pmol/20 min, P < 0.03). Similarly, in studies on isolated islets from GK rats cultured in 5.5 or 16.7 mmol/l glucose for 48 h, forskolin (5 pmol/l) restored insulin release in response to 16.7 mmol/l glucose but had no effect on islet glucose utilization at 3.3 or 16.7 mmol/l glucose. Forskolin markedly stimulated insulin release at 3.3 mmol/l glucose in GK but not control rat islets cultured for 48 h in 5.5 mmol/l glucose, whereas 20 mmol/l arginine had an almost identical effect in both islet varieties. However, in islets cultured in 16.7 mmol/l glucose, forskolin stimulated insulin release similarly both in control and GK islets at 3.3 mmol/l glucose. In conclusion, this study suggests that the insulinotropic effects of glucose are coupled to a direct regulation of the exocytotic machinery in the pancreatic beta-cell. This pathway is markedly impaired in GK rats, contributing to defective insulin response to glucose. In this model, cAMP generation restores the insulin response to 16.7 mmol/l glucose and exerts a marked insulin release even at 3.3 mmol/l glucose.