Duodenal fat such as oleate is known to influence gut functions by release of cholecystokinin (CCK), but the contribution of CCK endogenously released by duodenal fat or by diversion of pancreatic juice from the duodenum in the mechanism of mucosal integrity and gastroprotection has been little studied. This study was designed to compare the effect of CCK-8 and intraduodenal (i.d.) instillation of sodium oleate, or diversion of the pancreatic biliary secretions that are known to release CCK, on the gastric mucosal lesions induced by topical application of 100% ethanol or acidified aspirin (ASA) in rats with or without the pretreatment with a CCK-A receptor antagonist, loxiglumide, or with L-365,260 to block CCK-B receptors. In addition, the effect of suppression of prostaglandin (PG) biosynthesis by indomethacin (5 mg/kg i.p.), inhibition of nitric oxide (NO)-synthase by L-NAME (5 mg/kg i.v.), or blockade of sensory nerves by capsaicin (125 mg/kg s.c.) on the protective activity of sodium oleate was determined. Sodium oleate (50-200 mM i.d.), or diversion of pancreatic juice from the duodenum for 3 h that produced significant rise in plasma CCK levels, significantly reduced gastric lesions induced by 100% ethanol to an extent similar to that induced by exogenous CCK-8 (5 nmol/kg s.c.). The protective effect of oleate or diversion of pancreatic juice was accompanied by an increase in gastric blood flow (GBF). Both protection and accompanying hyperemia were completely abolished by blockade of CCK-A receptors with loxiglumide, whereas L-365,260, an antagonist of CCK-B receptors, had no effect. Oleate given i.d. significantly attenuated acidified ASA-induced gastric lesions and gastric secretion while increasing the luminal concentration of somatostatin. These effects were significantly reduced by loxiglumide but not by L-365,260. In contrast, CCK-8, which stimulated gastric acid secretion, failed to affect the lesions induced by acidified ASA and the decrease in the GBF produced by this ulcerogen. Indomethacin, which suppressed PG generation by approximately 90%, failed to influence the protective activity of oleate or CCK-8 against ethanol-induced lesions, whereas L-NAME, vagotomy, or sensory denervation significantly attenuated this protection and accompanying hyperemia. Addition to L-NAME of L-arginine, but not D-arginine, restored the protective and hyperemic effects of CCK-8 and duodenal oleate against gastric lesions induced by ethanol or acidified ASA. We conclude that endogenous CCK released by oleate or diversion of pancreatic secretion exerts a potent gastroprotective action on the stomach involving predominantly CCK-A receptors and depending on vagal activity, and hyperemia mediated by NO and sensory nerves but unrelated to acid secretory effects and endogenous PG.