Ontwikkeling van een fysiologisch relevant model voor hepatotrope virussen met behulp van hepatocyten afgeleid van humane pluripotente stamcellen

The pharmaceutical industry is in high need of efficient and physiologically relevant in vitro and in vivo models to study hepatotropic viruses, screen for potential antiviral drugs and test for liver toxicity that might be caused by these and other drugs. Current in vitro models rely on primary human hepatocytes (PHHs) and hepatoma cell lines, which have major drawbacks. As an alternative hepatocyte source, the use of human pluripotent stem cell derived hepatocyte-like cells (hPSC-HLCs) to study hepatotropic viruses has been proposed. Previous studies have shown that these hPSC-HLCs are indeed relevant models to study hepatotropic viruses as they are susceptible for hepatitis C, hepatitis B, hepatitis E and dengue virus. In this thesis, I examined whether hPSC-HLCs could also be a relevant model to study zika virus (ZIKV) infection. ZIKV has recently been linked to microcephaly in newborns. However, ZIKV-infected mouse models and non-human primates do not only have a high viral load in the brain, but also in other organs such as the liver. Additionally, as two reports described the association between ZIKV infection and liver injury and as ZIKV belongs to the same viral family of known hepatotropic viruses, such as hepatitis C and dengue virus, I examined whether human hepatocytes are susceptible to ZIKV infection. I demonstrated that hPSC-HLCs and the hepatoma cell line, Huh7, supports the complete ZIKV life cycle, including the entry, viral replication and the production of novel infectious ZIKV virions. Furthermore, treatment of ZIKV-infected hPSC-HLCs and Huh7 cells with 7-deaza-2'-C-methyladenosine, a known viral RNA-dependent RNA polymerase inhibitor, decreased ZIKV replication significantly in a dose-dependent manner, while 2'-C-methylcytidine and 6-fluoro-3-hydroxypyrazine-2-carboxamide only decreased ZIKV replication in ZIKV-infected Huh7 cells. Finally, ZIKV-infected hPSC-HLCs but not Huh7 cells induced an innate immune and NFκβ response, which may explain the more extensive cytopathic effect observed in ZIKV-infected Huh7 cells. These significant differences between hPSC-HLCs and hepatoma cells in the innate immune response against ZIKV and antiviral drug sensitivity highlight the need to assess ZIKV infection as well as antiviral activity not only in cell line models, but also in more physiologically relevant systems. As hPSC-HLCs used to demonstrate the susceptibility of hepatocytes to ZIKV, resemble fetal hepatocytes rather than adult primary hepatocytes and do not have drug metabolizing activity, I tested if metabolically improved hPSC-HLCs created by the Verfaillie lab at the Stem Cell Institute Leuven, consisting of inducible overexpression of 3 transcription factors (TFs) HNF1α, PROX1 and FOXA3 (termed HC3x-HLCs) and cultured in metabolically optimized differentiation culture medium, would be a suitable model to study hepatitis B virus (HBV) and to test new anti-HBV antivirals that might need to undergo biotransformation to be active. The reason to study HBV infection is that despite an adequate anti-HBV vaccine, 260 million people remain chronically infected with HBV worldwide and available treatments do not cure patients as the covalently closed circular DNA (cccDNA) is retained in the hepatocyte nucleus. Hence, there is an urgent need for novel therapeutics that can eliminate HBV infection in chronically infected patients. I here demonstrated that metabolically improved HC3x-HLCs can efficiently be infected with HBV by immunofluorescence staining of HBV core antigen and surface antigens (HBsAg). Furthermore, I detected the release of HBsAg and HBV e antigen in the supernatant, which increased over time, via ELISA, as well as high titers of infectious virus when back-titrated on HepG2-NTCP cells. This data suggested that functional cccDNA was formed by the HBV-infected HC3x-HLCs. Additionally, I validated the model for use in antiviral drug studies using various known HBV inhibitors. Finally, as the pharmaceutical industry also requires suitable in vivo models to study hepatotropic viruses and test drugs for liver toxicity, I attempted to transplant the metabolically improved HC3x-HLCs and HC6x-HLCs (that overexpress not only HNF1α, PROX1, FOXA3 but also PGC1α, SIRT1 and AMPK) into urokinase plasminogen activator - severe combined immunodeficiency (uPA-SCID) mice to generate a humanized liver mouse model. Six months after transplantation, human cells were detected in the livers of mice transplanted with HC3x-HLCs, but not HC6x-HLCs. However, engraftment efficiencies were low, despite the enhanced maturity of HC3x-HLCs compared to many other HLC populations. Studies will be required to determine if the failure to robustly repopulate murine livers is due to the specific mouse model used or due to failure to home and establish initial engraftment, and/or subsequent expansion of the commonly small numbers of initially engrafted cells.

Jaar van publicatie:2020

Toegankelijkheid:Open