Biotransformation and mechanism of cell death of thiazolidinedione (TZD) ring containing compounds

The glitazones were developed for the treatment of type II diabetes. A common structural feature in these drugs is a 2,4-thiazolidinedione (TZD) ring. Liver injury has been reported in the clinic following use of these drugs. The mechanism by which the glitazones exert toxicity is controversial and unpredicted. In contrast, 3-(3,5- dichlorophenyl)-2,4-thiazolidinedione (DCPT) is hepatotoxic in rats and cytotoxic in human HepG2 cells. DCPT is converted by HepG2 cells into [[[(3,5- dichlorophenyl)amino]carbonyl]thio]glycolic acid (DCTA), which can potentially degrade into 3,5-dichlorophenyl isocyanate (DPI) and mercaptoacetic acid (MAA). DCTA, MAA and DPI could conceivably contribute to DCPT toxicity. Our hypothesis is that DCPT undergoes biotransformation in the TZD ring to yield a metabolite that is involved in cell death. We propose to test this hypothesis through the following specific aims: (1) characterizing the metabolic profile of DCPT and (2) investigating the mechanism of cytotoxicity of DCPT and its metabolites. Rat and pooled human liver microsomes were used to investigate DCPT metabolism. Incubations contained the microsomes, buffer, glutathione and DCPT. Following pre-incubation the reactions were started by the addition of NADPH or NADH. Control incubations were performed in the absence of the cofactors. Additional experiments were conducted by boiling the microsomes before use. All incubations were assayed by reversed phase HPLC. The concentrations of DCTA and DPI in the incubations increased with time. DPI formation was completely blocked in the minus NADPH incubations or when boiled microsomes were used. DPI was detected in both the plus/minus NADH incubations but not until the iv 30 min time point and to a lesser degree when compared to plus NADPH incubations. In comparison, there was an apparent increase in production of DCTA when NADPH was omitted compared to complete incubations. To investigate whether there was a difference in DCPT metabolism between genders, female rat liver microsomes (FRLM) were used. In FRLM, significantly more DCTA was formed over time in comparison to MRLM. DPI formation was NADPH dependent in FRLM similar to what was seen in MRLM. Thus, metabolism of DCPT in vitro appears to be both cofactor-dependent and hydrolytic in HLM and MRLM/FRLM. Further, our results suggest a role for CYP-mediated metabolism in male rat and pooled human liver microsomes.;To address the mechanism of cytotoxicity, HepG2 cells were treated with 200 muM DCPT, DCTA, DPI, MAA or a combination of DPI plus MAA. Cell viability and proliferation were assessed over a 24 hr period. DCTA and the combination treatment (DPI plus MAA) were significantly more cytotoxic than DCPT, DPI or MAA alone, and both produced about a 65% loss of cell viability at 24 hr. In contrast, at the same time point, cell viability was reduced ca. 47% or 27% by DCPT or DPI alone, respectively. There was no significant loss in cell viability over 24 hr in cells treated only with MAA. To further study the cytotoxicity of the combination treatment (DPI plus MAA), HepG2 cells were pretreated (4 hr) with either 200 muM DPI or MAA, and then challenged with the alternative compound. The pretreatment groups profoundly reduced cell viability, as shown by the MTS assay, compared to the combination (DPI plus MAA) treatment. Our results suggest that the presence of both hydrolytic products of DCTA may contribute to DCPT cytotoxicity in HepG2 cells. Furthermore, there appears to be an additive/synergistic cytotoxic effect between DPI and MAA. Additional experiments are needed to understand the metabolism and mechanism of cell death of DCPT. (Abstract shortened by UMI.).