The production of hydroperoxides is rapidly increased and remains at 200-280% of the control 1-24 h after the second daily application of 17 nmol of 12-O-tetradecanoylphorbol-13-acetate (TPA) to mouse skin in vivo. The levels of hydroperoxides are increased 1.63-, 2.64-, 4.07-, and 4.31-fold 18 h after one, two, three, or four applications of TPA at 24-h intervals, respectively. The hydroperoxide response to TPA observed in whole skin reflects almost entirely the increased hydroperoxide-producing activity of the epidermis. Such hydroperoxide responses are triggered to various degrees by the anthrone derivatives and the phorbol esters and diterpene with complete and/or stage 2 tumor-promoting activities but not by the agents with only inflammatory, hyperplastic or stage 1 tumor-promoting activities. However, the Ca2+ ionophores A23187 and ionomycin are potent inducers of hydroperoxide formation. Several discrepancies are observed between the hydroperoxide response to TPA and the known effects of the tumor promoter on ornithine decarboxylase (ODC) induction. In contrast to the refractory state against ODC induction caused by TPA treatments repeated at intervals of less than 48 h, the time interval required for recovery of the hydroperoxide response to TPA in TPA-pretreated skins is only 5 h. The stimulatory effects of A23187, ionomycin and various diacylglycerols (DAGs) on hydroperoxide production do not correlate with their ODC-inducing activities. The increasing susceptibilities of C57BL/6, CF-1, and SEN-CAR mice to skin tumor promotion correlate with their hydroperoxide responses but not with their ODC responses to TPA. alpha-Difluoromethylornithine (DFMO) and other inhibitors of TPA-induced ODC activity fail to alter hydroperoxide production whereas the compounds that inhibit the hydroperoxide response to TPA, such as fluocinolone acetonide, have no or only minimal inhibitory activity against ODC induction. This would suggest that the hydroperoxide response to TPA does not require ODC induction and may not be essential for ODC induction. The hydroperoxide response to TPA is mimicked, but to a lesser degree, by the activator of protein kinase C, 1,2-dioctanoyl-sn-glycerol, and inhibited by verapamil, trifluoperazine, and palmitoylcarnitine. Populations of TPA-treated keratinocytes, therefore, may be responsible not only for ODC activation but also for hydroperoxide production. However, these two responses, which involve, at least in part, Ca2+ mobilization and protein kinase C activation and play important roles in the mechanism of skin tumor promotion, do not appear to be correlated.