Common marmosets (Callithrix jacchus) are potentially useful nonhuman primate models for preclinical studies. An anti-inflammatory drug, diclofenac is reportedly metabolized mainly by human cytochrome P450 (P450) 2C9 to 4'-hydroxydiclofenac and minorly by P450 3A4 to 5-hydroxydiclofenac that leads to reactive intermediates. In this study, in vivo pharmacokinetics in six marmosets and in vitro metabolism in nine marmoset liver microsomes were analyzed using diclofenac to evaluate marmosets as preclinical drug metabolism models. In wild-type marmosets genotyped for P450 2C19 p.[(Phe7Leu; Ser254Leu; Ile469Thr)], plasma levels of 5-hydroxydiclofenac and 4'-hydroxydiclofenac were roughly similar, but the homozygotes showed approximately three-times higher plasma levels of 5-hydroxydiclofenac than those of 4'-hydroxydiclofenac after oral administrations of diclofenac (50 mg/kg). Nine marmoset liver microsomes catalyzed mainly diclofenac 5-hydroxylation with no significant effects of the the P450 2C19 genotype, and partly diclofenac 4'-hydroxylation that depended on the P450 2C19 genotype, at substrate concentrations of 10 µM and 100 µM. Both Michaels-Menten constant Km values for diclofenac 4'-hydroxylation in human and marmoset liver microsomes were ∼30 μM and those for diclofenac 5-hydroxylation were ∼120 μM. Ketoconazole significantly suppressed only diclofenac 5-hydroxylation in marmoset or human liver microsomes at 0.030 μM, indicating main contribution of P450 3A enzymes, which were found to be P450 3A5/90 using recombinant marmoset P450 3A enzymes. These results suggest that marmosets would be a functional animal model for in vivo and in vitro metabolites likely generated via arene oxide intermediates of diclofenac by P450 3A enzymes in humans, primarily because marmosets lack the ortholog of human P450 2C9.
