Reply

Abstract

We thank Dr. Zhang for the comment on our article.(1) First, we would like to address the question regarding the role of the immune system in the human liver chimeric mouse model. We chose this model because it is the only small animal model allowing the investigation of hepatitis B virus (HBV)/hepatocyte interaction during a robust infection in vivo. Because of the immunodeficiency of the model and the absence of engrafted human immune cells, this model enables to study only HBV-hepatocyte interactions in the absence of HBV-specific immune responses. As stated in our article, the objective of our in vivo studies was not to determine which pathways are activated during the early steps of infection, but to analyze the impact of long-term infection on the expression of innate immune-related genes in infected hepatocytes. Using this model, we observed a decrease in cyclic GMPAMP synthase (cGAS), but also partial interferon (IFN)-stimulated gene expression, such as IFNB, after 16 weeks of persistent HBV infection, suggesting that these alterations are independent of antiviral immune responses or human immune nonparenchymal cells (see Fig. 7 of our article(1)). Furthermore, we had addressed Dr. Zhang’s comment already in the discussion of our article by stating that it would be of interest to study the role of Kupffer cells in HBV-cGAS interaction.(1) However, this question was out of the scope of our study aiming to study direct HBV-hepatocyte interactions.(1) Furthermore, this question cannot be conclusively addressed in classical human liver chimeric mice engrafted with human hepatocytes, but would require other animal models with a functional immune system modeling antiviral immune responses. Finally, we would like to comment on the technical questions of Dr. Zhang regarding the analysis of the liver function in the human liver chimeric mouse model: The detection of human albumin by a human albumin-specific enzyme-linked immunosorbent assay is a standard method to address cell viability of the engrafted hepatocytes in chimeric mouse models, as shown by many laboratories in the field including ours.(2-4) Serum human albumin directly correlates with the viability and number of engrafted hepatocytes. Thus, as stated in the Results section of our article, we used albumin analyses to exclude that expression levels were confounded by differences in engraftment or viability. In contrast, alanine aminotransferase (ALT) levels are not generally used to assess the injury or viability of the engrafted human hepatocytes: (1) The enzymatic assays to quantify ALT do not easily discriminate mouse ALT from human ALT; (2) unspecific alterations of ALT levels, because of the mouse liver injury of the model, do not allow to make arresting conclusions regarding HBVinduced elevations of ALT levels; and (3) because of absent HBV-specific immune and inflammatory responses in immunodeficient uPA-SCID mice, HBV-induced elevation of ALT is not expected to occur.(5) Given these limitations of the mouse model, we did not assess ALT levels.