Troglitazone, a thiazolidinedione derivative, is a widely used antidiabetic drug that binds and activates peroxisome proliferator-activated receptor γ (PPARγ) and enhances insulin sensitivity. It induces differentiation of adipocytes, which highly express PPARγ. We report that human prostate cancer cells expressed PPARγ at prominent levels and normal prostate tissues had very low expression. Dose-response clonogenic assays of the PC-3 prostate cancer cell line with troglitazone showed an antiproliferative effect (ED50, 3 × 10-7 m) and other PPARγ ligands (BRL49653: ED50, 8 × 10-8 m; 15-deoxy-Δ12,14-prostaglandin J2: ED50, 2 × 10-6 m; ciglitizone: ED50, not reached; indomethacin: ED50, not reached) showed similar effects. Combinations of troglitazone and a ligand specific for either retinoid X receptor or retinoic acid receptor did not show a synergistic effect. Pulse-exposure to troglitazone (10-5 m) for different durations showed that 4 days of pulse-exposure to the agent irreversibly inhibited 50% clonal growth of PC-3 cells. Interestingly, PC-3 cells cultured with troglitazone (10-5 m) showed dramatic morphological changes both by light and electron microscopy, suggesting that the cells became less malignant. Nevertheless, troglitazone did not affect either the cell cycle or several markers of differentiation. LNCaP cells constitutively produced prostate-specific antigen, and levels were markedly enhanced by all-trans-retinoic acid. Troglitazone (10-5 m, 4 days) decreased by 50% the levels of prostate-specific antigen produced by these cells. In vivo treatment of PC-3 tumors growing in male BNX triple immunodeficient mice with oral troglitazone (500 mg/kg/day) produced significant inhibition of tumor growth (P = 0.01). The only objective side effect of troglitazone in mice was the elevation of serum transaminases. Short-term culture of four surgically obtained human prostate cancer tumors with troglitazone (10-5 m, 4 days) produced marked and selective necrosis of the cancer cells (about 60%) but not the adjacent normal prostate cells. Taken together, these results suggest that troglitazone may be a useful therapeutic agent for the treatment of prostate cancer, especially in the setting of low disease burden.
This work was supported by NIH and United States Army grants and also in part by the CaP CURE Foundation and Parker Hughes Fund.