Cell-cell adhesion-mediated signaling determines epithelial polarization in the liver

ABSTRACT The liver is our largest metabolic organ. It produces proteins, lipids, clotting factors and glycogen while dispensing bile and detoxifying xenobiotics. In order to transport these different substances, a sophisticated network of liver venules, capillaries and interstitial conduits has evolved. An essential feature of this network are the lumen-forming epithelia that give rise to two major liver cell populations: (1) mature hepatocytes - the main parenchymal cell type, and (2) bile duct cells. Hepatocytes form single-cell cords with a capillary-like luminal network (bile canaliculi) running between them. In contrast, bile duct cells form tubules, each with a central lumen that receives the content of the bile canaliculi that has formed next to hepatocytes. During initial liver development and bouts of regeneration, both hepatocytes and bile duct cells are derived from a common epithelial precursor. How hepatocytes and biliary epithelia obtain their unique morphological and functional phenotypes from this common precursor is poorly understood. Indeed, because bile canaliculi are not readily visible by conventional H&E tissue stain, the study of epithelial polarity in the liver has largely been neglected. The resulting gap in our knowledge has greatly hindered our ability to better understand the molecular basis of common liver diseases, which typically present with changes in lumen organization. It also severely limits our ongoing efforts to engineer hepatic tissue that can be used for transplantation, toxicology and gene therapy studies. To tackle these issues, we have developed a unique tissue culture model to examine and explain how hepatocyte and bile duct luminal phenotypes form. It strongly suggests lumen polarity is established in two distinct steps. First, cell-cell adhesion triggers initial lumen formation at cell-cell contact sites. Second, luminal stability is then modified through E-cadherin signaling. Strong E-cadherin signaling antagonizes these luminal cell-cell contacts to determine a bile duct epithelium with a single apical surface. In contrast, weak E-cadherin signaling fails to block these cell-cell contacts and as a result the default pathway ensures hepatocytic polarity proceeds where multiple bile canaliculi can now form luminal junctions running between them. This hypothesis is supported by published and preliminary data that we obtained by developing cell systems in which the polarity phenotype can be switched. We have furthermore established stem cell-derived biliary and hepatocyte primary cultures to compare signaling mechanisms in the two liver epithelial cell types directly. It is expected that these novel approaches will enable us to understand the fundamental core mechanisms that drive cell type-specification and polarity phenotypes in the liver.