Trichloroacetate (TCA) and dichloroacetate (DCA) are hepatocarcinogenic metabolites of the environmental pollutant trichloroethylene (TCE) and are common water contaminants. Induction of peroxisome proliferation via activation of the peroxisome proliferator-activated receptor alpha (PPARalpha) has been proposed as a mechanism for their hepatocarcinogenic action. However, it is unclear whether these compounds are direct ligands of PPARalpha or whether activation occurs by a ligand-independent process. The present studies were undertaken to determine whether a primary rat hepatocyte model system could be used to examine structure-activity relationships of haloacetates for the induction of peroxisomal palmitoyl-CoA oxidation. The haloacetates tested differed in both type (iodo, bromo, chloro and fluoro) and extent (mono, di and tri) substitution. Significant differences were observed in both potency and efficacy. Potency varied over about two orders of magnitude, in the order of mono > di = tri. Within the monohalo-substituted series, the order of potency was iodo > bromo > chloro, with the fluoro analog being essentially inactive. The monoiodo- and monobromo-derivatives showed significant induction at 50 and 100 microM, respectively, but cytotoxicity precluded obtaining full concentration-response curves. The dihalo- and trihalo-acetates had generally similar potency, and, with the exception of the diflouro- and dibromoacetates, showed a maximal induction of two- to three-fold. Difluoroacetate and dibromoacetate induced palmitoyl-CoA oxidation by nine- and six-fold, respectively, approaching the effectiveness of Wy-14,643 (50 microM) in this system. Of interest, the slopes of the concentration-dependence lines of the difluoro- and dibromo-acetates were markedly dissimilar from the other di- and tri-haloacetates, suggesting either a marked difference in the way they activate the PPARalpha receptor or a substantial difference in the way they are metabolized or transported by the hepatocytes.