Enzyme-induction dependent bioactivation of troglitazone and troglitazone quinone in vivo

Troglitazone (TGZ), a 2,4-thiazolidinedione antidiabetic, causes hepatotoxi city in 1.9% of patients. TGZ is an inducer of, and substrate for, hepatic P450 3A. Microsomal metabolism yields a benzoquinone (TGZQ) and reactive in termediates. Kassahun et al. [Kassahun et al. (2001) Chem. Res. Toxicol. 14 , 62-70] have trapped the intermediates as thioester, thioether, and disulf ide conjugates of glutathione and found five conjugates in rat bile. The th ioether was substituted in the chromane moiety. We have investigated the ef fect of the P450 3A inducer, dexamethasone (DEX), on metabolism of TGZ and TGZQ in rats and assessed the compounds' cytotoxicity. TGZ-glucuronide and sulfonate were confirmed as principal biliary metabolites of TGZ (50 mg/kg, iv). Bile from noninduced animals also contained a TGZ-glutathione thioeth er adduct (ML3) but it was substituted in the thiazolidinedione moiety. Pre treatment with DEX (50 mg/kg/day for 3 days) resulted in a 2-5-fold increas e in the biliary concentration of ML3 and a 2-fold increase in the concentr ation of TGZQ, which was commensurate with the induction of hepatic P450 3A . Three of the known glutathione-conjugated metabolites were also found. TG ZQ (50 mg/kg, iv) was metabolized to an analogue of one of the TGZ-glutathi one thioesters and a glutathione adduct of TGZQ hydroquinone after DEX pret reatment. TGZ quinol glucuronide was a biliary metabolite of TGZ and TGZQ. Its formation would represent deactivation of TGZQ. TGZ was toxic to rat he patocytes and Hep-G2 cells at concentrations exceeding 50 and 25 muM, respe ctively, after 24 h. In contrast, TGZQ was nontoxic to rat hepatocytes and toxic to Hep G2 cells only at concentrations exceeding 100 muM. Our results show that TGZQ as well as TGZ yields reactive metabolites in vivo, and tha t bioactivation is enhanced by induction of P450 3A. However, hepatotoxicit y is unlikely to be due to either TGZQ or its metabolites.