Recent studies from our laboratory have shown that derivatization of the carboxylate moiety in substrate analogue inhibitors, such as 5,8,11,14-eicosatetraynoic acid, and in nonsteroidal antiinflammatory drugs (NSAIDs), such as indomethacin and meclofenamic acid, results in the generation of potent and selective cyclooxygenase-2 (COX-2) inhibitors (Kalgutkar et al. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 925-930). This paper summarizes details of the structure-activity studies involved in the transformation of the arylacetic acid NSAID, indomethacin, into a COX-2-selective inhibitor. Many of the structurally diverse indomethacin esters and amides inhibited purified human COX-2 with ICo5 values in the low-nanomolar range but did not inhibit ovine COX-1 activity at concentrations as high as 66 microM. Primary and secondary amide analogues of indomethacin were more potent as COX-2 inhibitors than the corresponding tertiary amides. Replacement of the 4-chlorobenzoyl group in indomethacin esters or amides with the 4-bromobenzyl functionality or hydrogen afforded inactive compounds. Likewise, exchanging the 2-methyl group on the indole ring in the ester and amide series with a hydrogen also generated inactive compounds. Inhibition kinetics revealed that indomethacin amides behave as slow, tight-binding inhibitors of COX-2 and that selectivity is a function of the time-dependent step. Conversion of indomethacin into ester and amide derivatives provides a facile strategy for generating highly selective COX-2 inhibitors and eliminating the gastrointestinal side effects of the parent compound.