Treatment with small doses of topical capsaicin protects the gastric mucosa from the damage by strong irritants but functional ablation of sensory nerves by pretreatment with larger dose of parenteral capsaicin augments the formation of gastric lesions via unknown mechanism. This study was designed to determine the role of gastric acid secretion, mucosal blood flow (GBF) and prostaglandins (PG) generation in the gastroprotection induced by small doses of topical or parenteral capsaicin in rats with intact or capsaicin-deactivated sensory nerves. Gastric lesions were produced in rats with intact sensory nerves (series A) or capsaicin-deactivated nerves (series B) using intragastric (i.g.) application of 100% ethanol, acidified aspirin (ASA) or water immersion and restraint stress (WRS). Pretreatment with i.g. capsaicin (0.12-1.0 mg/kg) in rats with intact sensory nerves (series A) reduced dose-dependently the mucosal damage caused by ethanol, ASA or WRS, the dose inhibiting the lesion area by 50% (ID50) being 0.3, 0.5 and 0.7 mg/kg, respectively. This protection was accompanied by a significant rise in gastric mucosal blood flow (GBF). Parenteral application of capsaicin (1.2-10 mg/kg s.c.) that in intact rats dose-dependently increased GBF, also dose-dependently reduced gastric damage induced by ASA or WRS (but not by ethanol), the ID50 being 5 and 3 mg/kg, respectively. The reduction by i.g. capsaicin of ethanol-or WRS-induced mucosal lesions was accompanied by a rise in GBF and this effect was reversed by indomethacin at a dose that suppressed endogenous PG biosynthesis by about 90%, indicating that PG are involved in the protective activities of topical capsaicin. Furthermore, topical and to a lesser extent parenteral capsaicin given to rats with intact or deactivated sensory nerves inhibited gastric acid and pepsin outputs, suggesting that this inhibition could contribute to the capsaicin-induced gastroprotection against acid-dependent mucosal lesions (ASA or WRS). Capsaicin deactivation of sensory nerves aggravated mucosal lesions induced by all three ulcerogens and this effects was accompanied by a marked decrease in GBF. In such capsaicin-deactivated rats, topical capsaicin also reduced ethanol-, ASA- or WRS-induced lesions, while parenteral capsaicin was effective only in the protection against the damage induced by acidified ASA and WRS but not by ethanol. The protection against WRS lesions and accompanying rise in GBF by parenteral capsaicin were also reversed by the pretreatment with indomethacin applied in a dose suppressing the generation of PG. We conclude that capsaicin is capable of protecting gastric mucosa in rats with both intact and capsaicin-deactivated rats and that this protective activity depends, at least in part, upon its hyperemic and antisecretory effects that may be mediated, at least in part, by endogenous release of PG.