In cultures of the differentiated clones Faza 967, Fao and HF, derived from Reuber hepatoma, physiological doses of glucocorticoid induce chenodeoxycholate 6 beta-hydroxylation, a microsomal cytochrome-P-450-mediated activity (enhanced in liver by phenobarbital and not by benzo[a]anthracene). Whereas 12-O-tetradecanoylphorbol 13-acetate (TPA) alone has no effect the tumor promoter, when added to dexamethasone, enhances this induction. This enhancement, half-maximum with 10 ng/ml TPA, is a function of the dose between 1 ng/ml and 50 ng/ml; 50 ng/ml (80 nM) increase 4-7-fold the induction rate (as measured in cultures by the amount of bile acid hydroxylated per 10(6) cells in 24 h, and in homogenates from treated cells) and 2.5-fold the maximum activity attained by the third day of induction. When added to cultures of the dedifferentiated clone H5, treated with benzo[a]anthracene, TPA does not influence benzo[a]pyrene hydroxylase induction, as shown by the total and relative amounts of the various hydrosoluble benzo[a]pyrene metabolites. TPA does not affect tyrosine aminotransferase induction in dexamethasone-treated Fao cultures. The enhancement is not suppressed by indomethacin, an inhibitor of prostaglandin synthesis. After dexamethasone removal from induced Faza 967 cultures, addition of TPA to the medium does not affect the decay rate of the chenodeoxycholate-hydroxylating activity. Retinoic acid similarly enhances the induction by dexamethasone of chenodeoxycholate hydroxylation, both in treated Faza 967 cultures and in homogenates from treated cultures. The effects of TPA and retinoic acid are additive. These results suggest a possible cooperation at the transcriptional level between transactive factors, involving TPA-mediated alterations, retinoic acid and glucocorticoid receptors. The system described might provide a convenient experimental approach in the study of its mechanism.