With the recent cloning of a second gene coding for the prostaglandin endoperoxide (PGH) synthase (cyclooxygenase), it has become obvious that mammalian cells contain two related, but unique, isozymes of PGH synthase. Both of these isozymes catalyze the conversion of arachidonic acid to PGH2, leading to production of biologically active prostaglandins. Although the first of these isozymes, PGH synthase-1 (PGHS-1), has long been thought to be the primary and sole site of action of nonsteroidal antiinflammatory drugs (NSAIDs), it is now known that the second isozyme, PGH synthase-2 (PGHS-2), is also sensitive to NSAIDs. Cloning of complementary DNAs for murine PGHS-1 and PGHS-2 has permitted individual expression of these two isozymes in the cos-1 cell system and comparison of their relative inhibition by several common NSAIDs in vitro. These studies have demonstrated that the two mouse isozymes, PGHS-1 and PGHS-2, are pharmacologically distinct. PGHS-1 is a constitutively expressed enzyme that early observations indicate is the principal enzyme involved in producing prostaglandins that regulate cellular housekeeping functions, such as gastric cytoprotection, vascular homeostasis, and kidney function. In contrast, PGHS-2 appears only to be expressed in inflamed tissue or following exposure to growth factors, lymphokines, or other mediators of inflammation. Expression of PGHS-2 is inhibited by antiinflammatory glucocorticoids, lending further support to the hypothesis that this enzyme produces prostaglandins involved in inflammation. We have identified NSAIDs that preferentially inhibit murine PGHS-1 or PGHS-2 or inhibit both isozymes equally. The finding that the two isozymes can be differentially inhibited provides a possible mechanism for identifying safer, more effective NSAIDs. Screening for drugs that preferentially inhibit PGHS-2 may allow identification of NSAIDs that reduce inflammation, but spare renal and gastric prostaglandin synthesis, thus reducing the untoward side effects commonly associated with most NSAIDs. Thus far, nabumetone is the only NSAID identified that preferentially inhibits murine PGHS-2.