Using model systems, we explored a potential function of hepatic phosphatidylcholine transfer protein to extract biliary-type phosphatidylcholines from intracellular membranes (e.g., smooth endoplasmic reticulum) and deliver them to canalicular plasma membranes where biliary secretion occurs. We measured transfer rates of parinaroyl phosphatidylcholine, a naturally fluorescent phospholipid, from small unilamellar vesicles composed of sn-1 palmitoyl, sn-2 parinaroyl phosphatidylcholine, and egg yolk phosphatidylcholine (molar ratio 75:25) wherein the fluorophore is self-quenched to small unilamellar vesicles composed of phosphatidylethanolamine, sphingomyelin, phosphatidylserine, phosphatidylinositol, and cholesterol (molar ratios 22:22:10:8:38) representing model microsomal and canalicular plasma membranes, respectively. Following addition of phosphatidylcholine transfer protein (purified from bovine liver), fluorescence intensity increased exponentially indicating net phosphatidylcholine transfer from donor to acceptor vesicles. Submicellar concentrations of a wide hydrophobicity range of common and uncommon taurine and glycine conjugated bile salts species (anionic steroid detergent-like molecules), sodium taurofusidate (a conjugated fungal bile salt analog), and sodium dodecyl sulfate and octylglucoside, anionic and nonionic straight chain detergents, respectively, markedly stimulated phosphatidylcholine transfer protein activity. This 40-115-fold effect was most pronounced for the common bile salts and correlated positively with bile salt hydrophobicity. Thermodynamic analysis of net transfer revealed that the rate-limiting step was extraction of phosphatidylcholine molecules from donor vesicles and that bile salts facilitated their capture by enhancing both phosphatidylcholine transfer protein binding as well as perturbing phospholipid packing in vesicle bilayers. These data, taken together with a kinetic analysis showing that bile salt-bound vesicles acted as both donors and noncompetitive inhibitors of phosphatidylcholine movement, suggest that protein-dependent phosphatidylcholine net transfer may function vectorially in vivo to deliver phosphatidylcholine molecules from endoplasmic reticulum to canalicular plasma membranes. Since we found a highly significant correlation between protein-mediated phosphatidylcholine transfer rates in vitro and reported phosphatidylcholine secretion rates in vivo as functions of secreted bile salt hydrophobicity [bile fistula hamsters [Gurantz, D., and Hofmann, A. F. (1984) Am, J. Physiol. 247, G736-G748] and prairie dogs [Cohen, D. E., Leighton, L. S., & Carey, M. C. (1992) Am. J. Physiol. 263, C386-G395]], our results are consistent with the hypothesis that the specific phosphatidylcholine transfer protein of liver may play a physiological role in bile formation.
