We developed 3-{5-[4-(cyclopentyloxy)-2-hydroxybenzoyl]-2-[(3-hydroxy-1,2-benzisoxazol-6-yl)methoxy]phenyl} propionic acid (T-5224) as a novel inhibitor of the c-Fos/activator protein-1 for rheumatoid arthritis therapy. We predicted the metabolism of T-5224 in humans by using human liver microsomes (HLM), human intestinal microsomes (HIM), recombinant human cytochrome P450 (P450), and UDP-glucuronosyltransferases (UGTs). T-5224 was converted to its acyl O-glucuronide (G2) by UGT1A1 and UGT1A3 and to its hydroxyl O-glucuronide (G3) by several UGTs, but it was not metabolized by the P450s. A comparison of the intrinsic clearances (CL(int)) between HLM and HIM suggested that the glucuronidation of T-5224 occurs predominantly in the liver. UGT1A1 showed a higher k(cat)/K(m) value than UGT1A3 for G2 formation, but a lower k(cat)/K(m) value than UGT1A3 for G3 formation. A high correlation was observed between G2 formation activity and UGT1A1-specific activity (β-estradiol 3-glucuronidation) in seven individual HLM. A high correlation was also observed between G2 formation activity and UGT1A1 content in the HLM. These results strongly suggest that UGT1A1 is responsible for G2 formation in human liver. In contrast, no such correlation was observed with G3 formation, suggesting that multiple UGT isoforms, including UGT1A1 and UGT1A3, are involved in G3 formation. G2 is also observed in rat and monkey liver microsomes as a major metabolite of T-5224, suggesting that G2 is not a human-specific metabolite. In this study, we obtained useful information on the metabolism of T-5224 for its clinical use.