There is renewed controversy regarding safe exposure levels for dioxin. At the heart of this controversy is the hypothesis that toxic effects of dioxin are receptor-mediated and therefore a "threshold" should exist below which no toxic effects can occur. Our research focuses on the ability of dioxin to alter protein levels in rodent livers. Established effects of exposure to dioxin are the induction of cytochrome P450-1A1 and P450-1A2 and a reduction in the maximal binding of the epidermal growth factor receptor in rat livers. An initiation-promotion protocol was used to study the effects of dioxin in female Sprague-Dawley rats. Animals were administered a single initiating dose of diethylnitrosamine followed by 16 biweekly gavage doses of 2,3,7,8-TCDD. Steady-state pharmacodynamic models were fit to these data assuming a combination of Hill kinetics and Michaelis-Menten kinetics. Two classes of models were developed which postulate two different mechanisms for the constitutive expression and TCDD-induced alterations in the levels of these proteins. The results are consistent with models which follow proportionate response in the low-dose region (no threshold) and with models which allow for a low-dose threshold. In all cases studied, the best fitting model exhibited no "threshold" for the effects of TCDD on the modulation of these proteins. The finding is consistent with the knowledge that for some receptor-mediated responses, there is a proportional relationship between receptor occupancy and biological response, even at low ligand concentrations.(ABSTRACT TRUNCATED AT 250 WORDS)