This study evaluated the contribution of supraspinal opioid receptors to gastric mucosal protection in the rat. Intracerebroventricular (i.c.v.) and intracisternal (i.c.) injections of selective delta- [[D-Ala(2),D-Leu(5)]-enkephalin (DADLE), [D-Pen(2),D-Pen(5)]-enkephalin (DPDPE), deltorphin II], selective mu- [[D-Ala(2),Phe(4),Gly(5)-ol]-enkephalin (DAGO)] opioid receptor agonists and beta-endorphin (ligand of both receptor types) produced a dose-dependent inhibition of acidified ethanol-induced gastric mucosal damage. The ED(50) values for beta-endorphin, DAGO, DADLE, deltorphin II, and DPDPE were 3.5, 6.8, 75, 120, and 1100 pmol/rat, respectively, following i.c.v. and 0.8, 9.0, 45, 0.25, and 7 pmol/rat following i.c. injection. The gastroprotective effect of DADLE, deltorphin II, and DPDPE, but not that of DAGO, was inhibited by naltrindole, the selective delta-receptor antagonist. Since the delta(2)-receptor agonist deltorphin II was more potent than the delta(1)-receptor agonist DPDPE, the dominant role of central delta(2)-receptors in gastroprotection might be raised. The site of action for delta-receptor agonists is likely to be the brain stem since the peptides were more potent following i.c. than following i.c.v. administration. The gastroprotective effect was reduced following acute bilateral cervical vagotomy. Moreover, both the nitric-oxide synthase inhibitor N(G)-nitro-L-arginine (3 mg/kg i.v.) and the prostaglandin synthesis inhibitor indomethacin (20 mg/kg p.o.) decreased the protective effect of opioid peptides. The results indicate that 1) activation of supraspinal delta- and mu-opioid receptors induces gastric mucosal protection, 2) integrity of vagal nerve is necessary for the gastroprotective action of opioids, and 3) mucosal nitric oxide and prostaglandins may be involved in the opioid-induced gastroprotection.