The Structural Basis of Cyclooxygenase‐2 Inhibition by Fenamates

Abstract

Two cyclooxygenase (COX) isoforms, COX‐1 and COX‐2, catalyze the first two steps of prostaglandin biosynthesis from arachidonic acid (AA) and are the pharmacological targets of non‐steroidal anti‐inflammatory drugs (NSAIDs). Mefenamic acid is one of several NSAIDs that have shown selective inhibition of 2‐arachidonylglycerol (2‐AG) oxygenation by COX‐2 (Prusakiewicz J.J. et al, Biochemistry 2009). While the X‐ray crystal structures of most classes of NSAIDs bound to COX have been solved, the binding mode of fenamates is still unknown. Here we report the structural determination of murine COX‐2 structures complexed with several fenamates and biochemical investigation of the role of individual residues in their inhibition. The crystal structures for the first time revealed that the fenamates are bound in the cyclooxygenase channel in an inverted binding mode similar to those of propionic acid derivatives such as diclofenac and lumiracoxib. The side chains of Tyr‐385 and Ser‐530 form critical H‐bonds with the carboxylic group of the fenamates. The aniline ring substituents of mefenamic acid, tolfenamic acid and flufenamic acid exhibit a preference for the space beneath Leu‐531 as opposed to the side pocket off the active site channel. In contrast, when there is an additional chloride substituent, as in the case of meclofenamic acid, this extra substituent orients toward the side pocket. These observations provide a structural basis for the fact that meclofenamic acid is a slow, tight‐binding inhibitor of COX‐2 with a low IC50 (31 nM) whereas mefenamic acid, flufenamic acid, niflumic acid, and tolfenamic acid are weak, rapidly reversible inhibitors with IC50's greater than 10 μM for AA oxygenation. Strikingly, when 2‐AG is used as the substrate, these weak, rapidly reversible inhibitors become potent inhibitors with IC50's ranging from 20 to 50 nM. In contrast, meclofenamic acid exhibits the same potency for both substrates. Our results provide the first atomic level detail of the interaction between fenamates and COX‐2, structural insights of slow, irreversible and rapid reversible fenamates, and a biochemical and structural foundation for future design and development of substrate‐selective inhibitors of COX‐2.Support or Funding InformationThe work was supported by NIH CA089450 (to L. J. M.).