Search

To elucidate how actin filaments participate in bile formation, polarity of actin filaments in the pericanalicular cytoplasm was determined with myosin subfragment 1 by transmission electron microscopy of ultrathin sections and deep-etching replicas. Densely concentrated actin filaments were identified around the bile canaliculi in the forms of microvillous core filaments, pericanalicular web filaments, and filaments on the functional complex. They bound subfragment 1 to form double-helical strands on the deep-etching replica or typical arrowheads on the ultrathin section. All microvillous core filaments showed their arrowheads pointing basally, suggesting the molecular growth occurring at their apical ends. In contrast, filaments of the pericanalicular web, running in parallel to the cell surface, showed unfixed polarities as indicated by their arrowheads. Furthermore, neighboring filament pairs often showed opposite polarities, an alignment necessary for filament sliding. The junctional complex had filaments with arrowheads pointed mostly at the cell center with a small number in opposite direction. In addition, a group of sporadic filaments appeared to be installed to link to both the canalicular membrane and coated vesicles. Such regionally specialized actin filaments are considered inclusively to form a cytoskeletal system that is in charge of (a) maintenance of length of the microvilli, (b) contraction of the canalicular walls, and (c) translocation of coated vesicles in the pericanalicular cytoplasm.

It was recently shown that intrahepatic bile duct epithelial cells in situ or after isolation from rat liver have coated pits and vesicles, suggesting that they participate in receptor-mediated endocytosis. Therefore, using a morphologic approach and epidermal growth factor coupled to horseradish peroxidase or colloidal gold as probes, we studied freshly isolated or short-term cultured intrahepatic bile duct epithelial cells prepared from normal rat liver to determine if they participate in receptor-mediated endocytosis. Immunoelectron microscopy using a monoclonal antibody against the epidermal growth factor receptor was also used to examine for the presence of the growth factor receptor on the cells. Immediately after isolation, the cells did not internalize either epidermal growth factor-horseradish peroxidase or epidermal growth factor-colloidal gold; no growth factor receptor could be shown on these cells by immunocytochemistry, either. In contrast, cells cultured for 24 h bound and internalized both epidermal growth factor-horseradish peroxidase and epidermal growth factor-colloidal gold at 37 degrees C and showed growth factor receptors diffusely distributed on the plasma membrane. When cultured cells exposed to epidermal growth factor-colloidal gold were fixed with glutaraldehyde containing saponin and tannic acid, colloidal gold particles were observed in coated pits and in coated and uncoated vesicles. Preincubation of cultured cells with native epidermal growth factor completely blocked the internalization of both epidermal growth factor-horseradish peroxidase and epidermal growth factor-colloidal gold. When rat liver was stained in situ for epidermal growth factor receptor, reaction product was observed by immunoelectron microscopy exclusively on the basal surface of the plasma membrane of the intrahepatic bile duct epithelial cells. These results indicate that bile duct epithelial cells internalize epidermal growth factor by endocytosis via coated pits containing receptors localized in situ exclusively to the basal domain of their plasma membranes. The data demonstrate for the first time that intrahepatic bile duct epithelial cells participate in receptor-mediated endocytosis and raise the possibility that they are a target for epidermal growth factor.

The mechanism of bile duct destruction observed in primary biliary cirrhosis (PBC) is still not clear. Recent study regarding cell death suggests the contribution of perforinor Fas-related apoptosis in the pathogenesis of immune-mediated cholangiopathy. The measurement of Fas-ligand (Fas-L) has been available recently and serum Fas-L level is reportedly related to the various inflammatory processes. However, its importance in immunemediated cholangiopathy is not known. Also, the usefulness of measuring Fas-L for evaluation of the clinical stage or activity is not known. Thus the Aims of this study are; (1) to measure the Fas-L in the serum from patients with PBC and compare their Fas-L levels to clinical parameters, (2) to compare Fas-L levels in PBC patients with those in other autoimmune liver disease patients (autoimmune hepatitis (AIH), primary sclerosing cholangitis (PSC)), and (3) to examine the relationship between serum Fas-L levels and histological Fas-L expression of the liver-infiltrating lymphocytes. [Methods] Histologically diagnosed 43 PBC cases, 5 AIH cases, and 2 PSC cases were enrolled in this study. The clinical parameters analyzed included serum ALT, ALP, IgG, IgM levels and histological staging of PBC by Scheuer. Serum Fas-L level was measured by ELISA. The Fas-L expression in liver specimens from PBC patients was evaluated by indirect immunohistochemistry using anti-Fas-L antibody. The density of Fas-L positive cells in the portal tract were calculated using NIH image, and evaluated in accordance with the degree of cholangitis and fibrosis in routine HE PBC 0.55 1.01, AIH 0.84 + 0.62, PSC 2.20--. 2.54, and the difference was not significant. (2) There was no statistical corelation between serum Fas-L levels and other clinical parameters (serum ALT, ALP, IgG, IgM levels). (3) In PBC, scattered Fas-L-positive cells were observed. The number of Fas-L positive cells per portal tract (mean_+ SD) was 2.79 +-1.21 (range 1.29 to 5.14). However, there was no statistical corelation of those positivities of histological Fas-L expression with serum Fas-L levels or degree of fiborsis/cholangitis. [Conclusions] Measurement of serum Fas-L levels were possible in human cholangiopathies including PBC. However, its relationship to other clinical factors and histological Fas-L expression was not proved. This research was funded by the grant from the Ministry of Education of Japan #07670555 and #09770339.

To study directly the functions of the cells that line the bile ducts inside the liver, we developed a new technique for isolating intrahepatic bile duct epithelial cells (IBDECs) from normal rat liver. Parenchymal and nonparenchymal cells were separated from whole liver by enzymatic digestion and mechanical disruption; subpopulations of individual nonparenchymal cells then were isolated by serial counter-flow elutriation, isopycnic centrifugation, and immunoaffinity separation with a specific monoclonal antibody against an antigen on the plasma membrane of IBDECs. Using this approach, we isolated 1.2 +/- 0.2 x 10(6) (mean +/- SE) viable (greater than 95% trypan blue exclusion) cells, greater than 95% of which were identified as IBDECs by morphologic appearance and specific cytochemical markers. The IBDECs averaged 7.4 +/- 0.16 microM in diameter and retained their in situ appearance, including morphologic polarity. They appeared as single cells or as cell doublets attached by tight junctions that excluded ruthenium red. Microvilli were abundant and were restricted to the apical (i.e., luminal) domain of the plasma membrane. Coated pits were observed on both apical and basolateral cell surfaces. Internally, IBDECs contained a well-developed system of organelles, including mitochondria, Golgi, and discrete types of vesicles, such as coated vesicles, multivesicular bodies, and lysosomes. These results indicate that a highly purified suspension of viable, morphologically intact, and polar IBDECs can be prepared from normal rat liver using a novel approach that separates liver cells on the basis of size, density, and specific membrane components. The availability of such a model will allow experimental studies to be performed directly on IBDECs, an approach that has not previously been possible.