Nitric oxide (NO) is known io be a potent messenger in the intracellular signal transduction system in many tissues. In pancreatic beta cells, NO has been reported to be formed from L-arginine through NO synthase. To elucidate the effect of NO on insulin secretion and to investigate the intracellular mechanism of its effect, we have used sodium nitroprusside (SNP) as a NO donor. SNP inhibited glucose-induced insulin secretion in a dose-dependent manner, and its effect was reversed by hemoglobin, a known NO scavenger. However, glyceraldehyde-induced insulin secretion was not affected by SNP. Since the closure of ATP-sensitive K+ channels (K-ATP channel) has been established as a key step in glucose-induced insulin secretion, we have directly assessed the effect of SNP on K-ATP channel activity using the patch damp technique. The K-ATP channel activity reduced by glucose was found to be reversibly activated by the addition of SNP, and this activation was able to be similarly reproduced by applying S-Nitroso-N-acetyl-DL-penicillamine (SNAP), another NO generator. Furthermore, these activating effects were completely eliminated by hemoglobin, in accordance with the reversibility in inhibition of glucose-induced insulin release. However, SNP could not affect the K-ATP channel suppression by ATP applied to the inside of the plasma membrane. The activation of the K-ATP channel by NO, therefore, seems to be due to the decreased ATP production attributable to impairment of glucose metabolism in beta cells. Since SNP exhibited no effect on glyceraldehyde-induced K-ATP channel inhibition, NO may disturb a glycolytic step before glyceraldehyde-3-phosphate. The K-ATP channel activation by 2-deoxyglucose through presumable ATP consumption due to its phosphorylation by glucokinase was, however, not affected even in the presence of SNP. But in the permeabilized beta cells made by exposure to a low concentration (0.02 U/ml) of streptolysin O (open cell-attached configuration), SNP reopens K-ATP channels which have been eliminated by fructose-6-phosphate, while this effect was not observed in the K-ATP channels inhibited by fructose-1,6-bisphosphate. On the other hand, in rat ventricular myocyte K-ATP channels were not activated by SNP even under a low concentration of glucose. From these observations, the inhibition of phosphofructokinase activity is probably the site responsible for the impairment of glucose metabolism induced by NO in pancreatic beta cells. NO, therefore, seems to be a factor in the deterioration of glucose-induced insulin secretion from pancreatic beta cells through a unique intracellular mechanism.