Hepatocellular Epithelial to Fibroblastoid Conversion

Hepatocellular carcinogenesis is a multistep process emerging from the progressive de-differentiation and malignant transformation of hepatocytes. Our recent establishment of an unique ex vivo tumor model system is based on a novel finding which was so far unknown to occur: Murine hepatocytes expressing constitutive active Ha-Ras show a polarized epithelial phenotype, however, treatment with transforming growth factor (TGF)-beta1 converts these hepatocytes to a de-polarized fibroblastoid morphology. In contrast, parental normal hepatocytes undergo growth arrest under these conditions. The epithelial to fibroblastoid conversion (EFC) of hepatocytes through the cooperation of Ha-Ras and TGF-beta1 is associated with growth in polylayers, loss of cell adhesion, disruption of functional tight junctions and dramatic changes in the cytoskeletal framework. After EFC, de- differentiated fibroblastoid Ras-expressing cells secrete TGF-beta1, indicating the involvement of an autocrine regulation of TGF-beta signaling, which is sufficient to maintain a highly malignant and invasive phenotype. The current project proposal focuses on the cellular and molecular aspects of this phenotypical conversion. The project aims to identify changes in signal transduction and in regulation of gene expression which are responsible for the invasive hepatocellular phenotype. The study of this process may facilitate a better understanding of the molecular basis of hepatocellular de-differentiation upon carcinogenesis and should result in novel strategies to develop specific diagnostic and therapeutical tools.

Carcinomas arising from epithelial cells represent the most prevalent malignancies, and metastasis is the major cause for the death of carcinoma patients. The breakdown of epithelial cell homeostasis leading to aggressive cancer progression is associated with the loss of epithelial characteristics and the acquisition of an invasive phenotype. This epithelial to mesenchymal transition (EMT) is considered as a hallmark of late stage tumorigenesis. Our findings indicate that EMT of murine hepatocytes is caused by the cooperation of hyperactive Ras and transforming growth factor (TGF)-beta which leads to a dramatic increase in malignancy including local tumor infiltration and distal metastasis. As frequently observed in human hepatocellular carcinomas (HCCs), this process is accompanied by loss of the tumor suppressor E-cadherin, nuclear accumulation of beta-catenin and secretion of TGF-beta1. The regulatory events involved in establishing the invasive phenotype depend on the crosstalk of TGF-beta and MAPK signaling in the induction phase of EMT, and the cooperation of Smads with both MAPK and PI3K signals during maintenance of the metastatic phenotype. Furthermore, epithelial hepatocytes show a potential to trans- differentiate into myofibroblasts, which localize in fibrotic deposits. Together with the comparison of hepatocellular EMT and the TGF-beta driven activation of hepatic stellate established in this study, these data indicate at first a direct link between liver carcinogenesis and hepatic fibrosis. In addition to the de- and trans- differentaion of hepatocytes during the gain of malignancy, the parental p19ARF deficient hepatocytes employed in these investigations are also capable to retro-differentiate into hepatic progenitor cells during liver restoration, which finally differentiate into both normal cholangiocytes as well as hepatocytes. These investigations on hepatocellular epithelial plasticity show the vast differentiation repertoire of hepatocytes, and strongly contribute to the identification of markers and novel molecular targets which serve as a basis to develop auspicious strategies for tumor diagnosis and anti-cancer therapies.