Search

Fas (Apo1/CD95) is a member of the tumour necrosis factor/nerve growth factor receptor superfamily and mediates apoptosis in various cell types (for review sec [1]). Although this apoptotic activity has been clearly related to homeostasis in the immune system and pathological situations in non-lymphoid organs, the Fas signaling pathway remains mostly elusive. We and others previously showed that Fas-induced apoptosis of primary culture hepatocytes requires either an inhibitor of translation or a protein kinase inhibitor, suggesting that two distinct pathways of Fas signaling exist in hepatocytes. We report here that activation of ICE-like and CPP32-like cysteine proteases are required for Fas-mediated apoptosis, but that these pathways involve different subclasses of serine proteases and are selectively modulated by inhibitors of protein tyrosine kinases. These results confirm that distinct pathways can lead to Fas-induced apoptosis in hepatocytes. Further understanding of these pathways could facilitate the rational design of anti-apoptotic drugs in liver diseases associated with massive Fas-mediated hepatocyte apoptosis, including fulminant hepatitis.

In liver, apoptosis is a physiological process involved in the clearance of injured cells and in homeostatic control [1]. However, in patients with viral fulminant hepatitis or with nonacute liver diseases [2], dramatic liver failure or secondary cirrhosis results from the death of hepatocytes, which express the cell-surface receptor Fas, by apoptosis. To date, treatment of fulminant hepatitis relies mainly on orthotopic liver transplantation, which is limited by immunological complications and graft availability. Unravelling the molecular mechanisms that underlie acute liver failure could allow the design of an appropriate therapy. Ligand-bound Fas and tumour necrosis factor alpha (TNF-alpha) induce hepatic apoptosis in mice [3-6]. In various cell types, Fas- or TNF-alpha-induced apoptosis is blocked by viral proteins (such as p35 and CrmA) as well as by a decoy peptide (YVADcmk) [7-11], suggesting that these mechanisms of apoptosis involve ICE (interleukin-1 beta converting enzyme)-like proteases. Here, we report that, in vivo, pre-treatment of mice with YVADcmk protects them from the lethal effect of anti-Fas antibody and from liver failure induced by injection of TNF-alpha. Remarkably, YVADcmk administration is also highly effective in rescuing mice that have been pretreated with anti-Fas antibody from rapid death, despite extensive hepatic apoptosis. This dramatic curative effect could be of clinical benefit for the treatment of viral and inflammatory liver diseases.

Fas is an apoptosis-signalling cell surface antigen that has been shown to trigger cell death upon specific ligand or antibody binding. Treatment of mice with an anti-Fas antibody causes fulminant hepatic failure due to massive apoptosis. To test a putative protective effect of the anti-apoptotic Bcl-2 protein, transgenic mice were generated to express the human bcl-2 gene product in hepatocytes. Early onset of massive hepatic apoptosis leading to death was observed in all nontransgenic mice treated with an anti-Fas antibody. By contrast, hepatic apoptosis was delayed and dramatically reduced in transgenic animals, yielding a 93% survival rate. These results demonstrate that Bcl-2 is able to protect from in vivo Fas-mediated cytotoxicity, and could be of significance for preventing fulminant hepatic failure due to viral hepatitis in humans.

A transgenic mouse model for hepatocarcinoma has been previously produced by targeting SV40 T-antigen expression to the liver. To evaluate the perturbation of cell death occurring during hepatocarcinogenesis, we examined the Fas-induced apoptosis on hepatocytes expressing T-antigen. Whereas anti-Fas antibody induced apoptosis in primary cultured normal hepatocytes, they imparted a weak cytotoxicity on primary cultured hepatocytes expressing T-antigen. This resistance of hepatic Fas-mediated apoptosis appears to result in an enhancement of a protective mechanism involving the protein kinase C signaling pathway rather than in a down-regulation of Fas-antigen expression. We further demonstrated that anti-Fas antibody does not have as efficient a lethal effect in T-antigen transgenic mice as in wild-type mice. The livers of transgenic mice injected with anti-Fas mAbs showed large intact regions with a few scattered apoptotic bodies: these regions strictly corresponded with carcinoma nodules, expressing high level of T-antigen. Our results describe a novel function for SV40 T-antigen which could contribute to viral pathogenesis by protecting infected cells against the host apoptotic defense mechanism.