Metabolic activation of diclofenac by human cytochrome P450 3A4: Role of 5-hydroxydiclofenac

Cytochrome P450 2C11 in rats was recently found to metabolize diclofenac in to a highly reactive product that covalently bound to this enzyme before it could diffuse away and react with other proteins. To determine whether cyt ochromes P450 in human liver could catalyze a similar reaction, we have stu died the covalent binding of diclofenac in vitro to liver microsomes of 16 individuals. Only three of 16 samples were found by immunoblot analysis to activate diclofenac appreciably to form protein adducts in a NADPH-dependen t pathway. Cytochrome P450 2C9, which catalyzes the major route of oxidativ e metabolism of diclofenac to produce 4'-hydroxydiclofenac, did not appear to be responsible for the formation of the protein adducts, because sulfaph enazole, an inhibitor of this enzyme, did not affect protein adduct formati on. In contrast, troleandomycin, an inhibitor of P450 3A4, inhibited both p rotein adduct formation and Ei-hydroxylation of diclofenac. These findings were confirmed with the use of baculovirus-expressed human P450 2C9 and P45 0 3A4. One possible reactive intermediate that would be expected to bind co valently to liver proteins was the p-benzoquinone imine derivative of 5-hyd roxydiclofenac. This product was formed by an apparent metal-catalyzed oxid ation of 5-hydroxydiclofenac that was inhibited by EDTA, glutathione, and N ADPH. The p-benzoquinone imine decomposition product bound covalently to hu man liver microsomes in vitro in a reaction that was inhibited by GSH. In c ontrast, GSH did not prevent the covalent binding of diclofenac to human li ver microsomes. These results suggest that for appreciable P450-mediated bi oactivation of diclofenac to occur in vivo, an individual may have to have both high activities of P450 3A4 and perhaps low activities of other enzyme s that catalyze competing pathways of metabolism of diclofenac. Moreover, t he p-benzoquinone imine derivative of 5-hydroxydiclofenac probably has a ro le in covalent binding in the liver only under the conditions where levels of NADPH, GSH, and other reducing agents would be expected to be low.