Sodium-dependent bile acid transport is a well-established function of the sinusoidal segment of the hepatocyte plasma membrane. Evidence has been provided previously by the authors for the existence of a putative sinusoidal plasma membrane sodium-dependent bile acid transporter with a mass of 49 kD. This protein has been partially characterized with a monoclonal antibody and by reconstitution in proteoliposomes. Further characterization is provided in the paper under discussion. The transporter was isolated from plasma membranes and from endoplasmic reticulum by immunoprecipitation with a monoclonal antibody. The protein was separated from antibody by means of HPLC and SDS-PAGE with electroelution. Amino acid analysis and NH2-terminal amino acid sequencing were performed. For both preparations, it was shown that the NH2-terminal amino acid sequence was identical, and the amino acid composition was similar to results obtained for the microsomal enzyme epoxide hydrolase. Purified mEH was immunoprecipitated with the monoclonal antibody which precipitated the plasma membrane protein. Isoelectric focusing established that the isoelectric points for epoxide hydrolase and the proteins isolated from the endoplasmic reticulum and plasma membranes were identical. The peptide maps prepared by subjecting the three proteins to enzymatic fragmentation with subtilisin were essentially identical. Because of the known contamination of plasma membrane fractions with endoplasmic reticulum and because of the high specific activity of epoxide hydrolase in endoplasmic reticulum, special attention was devoted to the question of the localization of the enzyme/transporter in plasma membrane. It was possible to show that virtually all of the immunoprecipitated protein obtained from intact cells derived from plasma membrane and that only a small fraction could be explained as having originated in endoplasmic reticulum. Finally, a hydrophobic plot of the amino acid sequence of epoxide hydrolase suggested that the protein contained four potential transmembrane domains. It could be concluded that a/the 49-kD sinusoidal plasma membrane sodium-dependent bile acid transporter protein was closely similar or identical to epoxide hydrolase. In view of the relatively small mass of the protein and its limited number of potential transmembrane domains, it appeared probable that the physiological membrane transporter consists of an assembly of 49-kD monomers or an assembly of heterologous proteins including the 49-kD protein.