Recent studies have suggested that the canalicular bile salt transport system of rat liver corresponds to a 100-kDa membrane glycoprotein. In the present study we attempted to functionally reconstitute the 100-kDa protein into artificial proteoliposomes. Canalicular membrane proteins were solubilized with octyl glucoside in the presence ofasolectin phospholipids. The extracts were treated with preimmune serum or the 100-kDa protein selectively immunoprecipitated with a polyclonal antiserum. Proteins remaining in the supernatant were then incorporated into proteoliposomes by gel-filtration chromatography. Canalicular proteoliposomes containing the 100-kDa protein exhibited t ulatable taurocholate uptake that could be inhibited by 4,4'-diisothiocyanato-2,2'-stilbenedisulfonic acid (DmDS). In contrast, no DIDS-sensitive t imulatable taurocholate uptake was found in 100-kDa protein-free canalicular proteoliposomes. However, when the immunoprecipitated 100-kDa protein was dissociated from the antibodies and exclusively incorporated into liposomes, reconstitution of DIDSsensitive transstimulatable and electrogenic taurocholate anion transport was again positive. Although incorporation of solubilized basolateral membrane proteins into liposomes also resulted in a prompt reconstitution of Na gradient-driven taurocholate uptake, the anti-100-kDa antibodies had no effects on the reconstituted transport activity of basolateral proteins. Thus, the findings establish that the previously characterized canalicularspecific 100-kDa protein is directly involved in the transcanalicular secretion of bile salts.

The mechanisms and driving forces for hepatic uptake of sulfate were investigated in basolateral (sinusoidal) rat liver plasma membrane vesicles. A transmembrane pH difference (pH 8.0 inside, 6.0 outside) stimulated sulfate uptake above equilibrium ("overshoot"). This pH gradient-stimulated sulfate uptake was saturable with increasing concentrations of sulfate and could be inhibited by probenecid, 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS), carbonyl cyanide p-(trifluoromethoxy)-phenylhydrazone, and nigericin. At low buffer concentrations and pH 6.0 an inwardly directed sodium gradient also stimulated sulfate uptake. This sodium-dependent sulfate uptake could be inhibited by amiloride and DIDS, indicating indirect coupling of sodium and sulfate flux through concomitant sodium-proton and sulfate-hydroxyl exchange. Cisinhibition of initial pH gradient-stimulated sulfate uptake, as well as transstimulation of sulfate uptake under pH-equilibrated conditions (pH 7.5 inside and outside), were observed with sulfate, thiosulfate, oxalate, and succinate, but not with chloride, bicarbonate, acetate, lactate, pyruvate, p-aminohippurate, citrate, glutamate, aspartate, and taurocholate. Furthermore, cholate and sulfobromophthalein exhibited competitive inhibition of pH gradient-stimulated sulfate uptake. In addition, an inside-to-outside hydroxyl gradient also stimulated uptake of cholate and this pH gradient-sensitive portion of cholate uptake was inhibited by extravesicular sulfate. In contrast to basolateral membranes, no evidence for multispecific sulfate-hydroxyl exchange was found in canalicular plasma membrane vesicles.

To identify the sinusoidal bile acid uptake system(s) of skate liver, photoaffinity labeling and kinetic transport studies were performed in isolated plasma membranes as well as intact hepatocytes. In both preparations photoaffinity labeling with the photolabile bile salt derivative (7,7-azo-3 alpha, 12 alpha-dihydroxy-5 beta-[3 beta-3H]cholan-24-oyl)-2-aminoethanesulfonate revealed the presence of a predominant bile salt binding polypeptide with an apparent molecular weight of 54,000. The labeling of this polypeptide was inhibited by taurocholate and cholate in a concentration-dependent manner and was virtually abolished by 1 mM of the anion transport inhibitor 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid. Kinetic studies of hepatic uptake with taurocholate, cholate, and the photoreactive bile salt derivative indicated the involvement of a single transport system, and all three substrates mutually competed with the uptake of each other. Finally, irreversible inhibition of the bile salt uptake system by photoaffinity labeling of hepatocytes with high concentrations (250 microM) of photolabile derivative reduced the Vmax but not the Km of taurocholate uptake. These findings strongly indicate that a single polypeptide with an apparent molecular weight of 54,000 is involved in sinusoidal bile salt uptake into skate hepatocytes. These findings contrast with similar studies in rat liver that implicate both a 54,000- and 48,000-K polypeptide in bile salt uptake and are consistent with a single Na+-independent transport mechanism for hepatic bile salt uptake in this primitive vertebrate.

The mechanism(s) and driving forces for biliary excretion of sulfate were investigated in canalicular rat liver plasma membrane vesicles (cLPM). Incubation of cLPM vesicles in the presence of an inside-to-outside (in, out) bicarbonate gradient (50 mM in, 5 mM out, pH 8.0 in and out), but not pH (pH 8.0 in, 6.0 out) or out-to-in sodium gradients, stimulated sulfate uptake 10-fold compared with the absence of bicarbonate and approximately 2-fold above sulfate equilibrium ("overshoot"). Initial rates of this bicarbonate gradient-driven sulfate uptake were saturable with increasing concentrations of sulfate (apparent Km, approximately 0.3 mM) and could be inhibited by probenecid, N-(4-azido-2-nitrophenyl)-2-aminoethylsulfonate, acetazolamide, furosemide,4-acetamido-4'-isothiocyanostilbene-2,2'-disulfonic acid, and 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (IC50, approximately 40 microM). Cisinhibition of initial bicarbonate gradient-stimulated sulfate uptake and transstimulation of sulfate uptake in the absence of bicarbonate were observed with sulfate, thiosulfate, and oxalate but not with chloride, nitrate, phosphate, acetate, lactate, glutamate, aspartate, cholate, taurocholate, dehydrocholate, taurodehydrocholate, and reduced or oxidized glutathione. These findings indicate the presence of a sulfate (oxalate)-bicarbonate anion exchange system in canalicular rat liver plasma membranes. In conjunction with the previously reported chloride-bicarbonate exchanger (J. Clin. Invest. 75: 1256-1263, 1985), these findings support the concept that bicarbonate-sensitive transport system might play an important role in bile acid-independent canalicular bile formation.

The inner membrane of rat liver mitochondria contains a highly active phospholipase Az which has alkaline pH optimum and requires Ca2+ in the micromolar range. The phospholipase is particularly active on the endogenous phosphatidylethanolamine and release relatively high amounts of docosahexanoic acid. The phospholipase A2 of mitochondria or mitoplasts is not dependent on calmodulin. Using fluorescamine-labelled mitoplasts there are indications that the enzyme is localized on both sides of the inner membrane.Mitochondria are known to possess a Ca2+-dependent phospholipase A2 which, beside being involved in the turnover of membrane phospholipids, is responsible for the membrane damage observed in aged mitochondria [l -31 and might be involved in the process of Ca2+-release from mitochondria [4,5]. The mitochondrial phospholipase A2 has an alkaline pH optimum and it prefers phosphatidylethanolamine as endogenous substrate [6]. The enzyme seems to be primarly located on the outer membrane of the mitochondrion [7], however some activity has also been found associated with the inner membrane [7,8]. Presumably the outer membrane enzyme is mainly involved in the turnover of membrane phospholipids whereas the inner membrane enzyme might be more involved in determining the permeability properties and the functional integrity of the inner membrane. It is still unclear how the activity of the mitochondrial and the other intracellular phospholipase A2 is regulated [9]. Since very little has been done to characterize the inner membrane phospholipase A2 we have decided to carry out a detailed study of its functional properties. For this study we have used both mitochondria completely devoid of outer membrane (mitoplasts) and inner membrane vesicles. The results obtained are presented in this article and are discussed in the general context of the metabolic regulation and function of the mitochondrial phospholipase Az. MATERIALS A N D METHODS Preparation of Mitochondria and MitoplastsLiver mitochondria were prepared from Wistar rats by the standard sucrose procedure of Schneider [lo]. The homogenization medium was buffered with 20 mM Tris/Cl pH 7.4 and contained 0.2 mM EDTA. The mitochondrial pellets were washed three times with medium without EDTA to reduce microsomal contamination. The mitochondria which were directly used for phospholipase A2 assay were treated with 30 pg digitonin/mg of protein to eliminate most of the contaminant lysosomes. Mitoplasts were prepared from intact mitochondria with a procedure similar to that described by Abbreviations. PtdEtn, phosphatidylethanolamine; IysoPtdEtn, lysophosphatidylethanolamine; PtdCho, phosphatidylcholine; Hcpes, 4-(2-hydroxyethy1)-l -piperazineethane sulfonic acid.Chan et al. [I I ] using 130 pg digitonin/mg of mitochondrial protein. To obtain complete removal of the outer membrane, the mitoplasts were suspended in 250 mM sucrose, 10 mM Tris/Cl pH 7.5, 0.2 mM EDTA at a concentration of 5 mg/ml. Aliquots of 3.5 ml were then sonicated for 5 s at 0 "C using a Branson sonifier (...