We examined the gastric ulcerogenic property of selective COX-1 and/or COX-2 inhibitors in rats, and investigated whether COX-1 inhibition is by itself sufficient for induction of gastric damage. Animals fasted for 18 h were given various COX inhibitors p.o., either alone or in combination, and they were killed 8 h later. The nonselective COX inhibitors such as indomethacin, naproxen and dicrofenac inhibited PG production, increased gastric motility, and provoked severe gastric lesions. In contrast, the selective COX-2 inhibitor rofecoxib did not induce any damage in the stomach, with no effect on the mucosal PGE(2) contents and gastric motility. Likewise, the selective COX-1 inhibitor SC-560 also did not cause gastric damage, despite causing a significant decrease in PGE(2) contents. The combined administration of SC-560 and rofecoxib, however, provoked gross damage in the gastric mucosa, in a dose-dependent manner. SC-560 also caused a marked gastric hypermotility, whereas rofecoxib had no effect on basal gastric motor activity. On the other hand, the COX-2 mRNA was expressed in the stomach after administration of SC-560, while the normal gastric mucosa expressed only COX-1 mRNA but not COX-2 mRNA. These results suggest that the gastric ulcerogenic property of conventional NSAIDs is not accounted for solely by COX-1 inhibition and requires the inhibition of both COX-1 and COX-2. The inhibition of COX-1 up-regulates the COX-2 expression, and this may counteract the deleterious influences, such as gastric hypermotility and the subsequent events, due to a PG deficiency caused by COX-1 inhibition.