Regulation of Hepatitis B Virus (HBV) Transcription and Biosynthesis

Introduction

HBV infection is a worldwide health problem and is endemic in many regions of Asia and Africa. The clinical consequences of HBV infection can be acute or chronic and range from the subclinical to fatal forms of the disease, including fulminant hepatitis where the patient dies shortly after infection. Although HBV infection can generally be prevented by vaccination with hepatitis B surface antigen (HBsAg), chronic HBV infection remains a major clinical problem. It is estimated that there are 200 to 500 million HBV chronic carriers in the world for whom, to date, there is no reliable treatment. The consequences of chronic HBV infection can include debilitating chronic active hepatitis, and liver cirrhosis, which is a major cause of mortality. Chronic carrier status has additional clinical implications. The estimated relative risk of primary hepatocellular carcinoma (PHC) in chronic HBV carriers is approximately 100-times greater than in uninfected individuals. As a consequence of HBV infection, it is estimated that there are approximately one million deaths annually attributable to this viral infection.  Therefore, effective treatments for chronic HBV infection are required. Understanding the viral life cycle in detail may reveal potential targets for antiviral therapy. HBV replicates by reverse transcription of the viral pregenomic RNA encoded by the HBV genome. Consequently, transcription of the viral genome and the regulation of pregenomic RNA synthesis, in particular, are essential steps in virus replication.
​

Transcriptional regulation of HBV by nuclear hormone receptors is a critical determinant of viral tropism
​

We have developed a new viral replication system using non-hepatoma cells where the effect of one or more liver-enriched transcription factors on HBV RNA synthesis and replication can be examined. Using this approach, it has been possible to demonstrate that the nuclear hormone receptors, hepatocyte nuclear factor 4 (HNF4), retinoid X receptor α (RXRα), peroxisome proliferator-activated receptor (PPAR), farnesoid X receptor (FXR), liver receptor homolog 1 (LRH1) and estrogen related receptor (ERR), are the liver-enriched transcription factors that are essential for pregenomic RNA synthesis and viral replication.  Surprisingly, nuclear hormone receptors are the only essential liver-enriched transcription factors critical to pregenomic RNA synthesis and viral replication, indicating a previously unknown importance of these factors in the HBV life cycle and tissue-specific tropism of the virus. In addition, this analysis suggests nuclear hormone receptors may represent suitable targets of the development of antiviral therapies.
​

Nuclear covalently closed circular (CCC) viral genomic DNA in the liver of hepatocyte nuclear factor 1α-null hepatitis B virus transgenic mice
​

The role of hepatocyte nuclear factor 1α (HNF1α) in regulating viral transcription and replication was examined in an HBV transgenic mouse model system. In transient transfection analysis, it has been demonstrated previously that HNF1α regulates the level of transcription from the large surface antigen promoter. This observation predicts that the loss of HNF1α might be associated with a reduction in the level of the 2.4-kb HBV RNA and the large surface antigen polypeptide it encodes. As the large surface antigen polypeptide is essential for viral biosynthesis, the loss of HNF1α might be expected to limit viral biosynthesis and lead to an increased abundance of mature capsids in the cytoplasm of the cell. In turn, these mature capsids may deliver their viral genomes to the nucleus to amplify the pool of CCC DNA, as is observed in duck hepatitis B virus infection. To examine whether this occurs in vivo, the viral replication intermediates present in the liver of HNF1α-null HBV transgenic mice were examined. The levels of the HBV RNAs, including the 2.4-kb viral transcript, in the HNF1α-null HBV transgenic mice were similar to the levels present in HBV transgenic mice expressing HNF1α. However, nuclear HBV CCC DNA was present in the hepatocytes of the HNF1α-null HBV transgenic mice. This suggests that subtle alterations in the levels of the HBV RNAs resulting from the absence of HNF1α may have resulted in the translocation of HBV genomic DNA into the nucleus of the hepatocytes. Alternatively, the absence of HNF1α may alter the physiological properties of the hepatocytes in a manner that favors the translocation of HBV genomic DNA into the nucleus. In either case, it is apparent that cycling of encapsidated HBV DNA from the cytoplasm into the nucleus can occur in the HNF1α-null HBV transgenic mouse model and represents a system where the molecular events regulating this aspect of the HBV life cycle can be analyzed in detail. In addition, the presence of nuclear HBV CCC DNA in these mice permits their role in HBV-mediated hepatocellular carcinoma to be examined.

Interferon a therapy is a standard treatment for chronic HBV infection. Using the HNF1a-null HBV transgenic mouse model, the stability of HBV CCC DNA in the hepatocytes of these mice has been examined in response to the induction of an interferon a/b response. Induction of an interferon a/b response in HBV transgenic mice by polyinosinic-polycytidylic acid treatment dramatically reduces the level of cytoplasmic HBV replication intermediates.  Therefore it was of interest to determine if induction of an interferon a/b response in HNF1a-null HBV transgenic mice might reduce the level of cytoplasmic and nuclear viral replication intermediates to a similar or different extent.  From this analysis, it is apparent that cytoplasmic replication intermediates are much more sensitive to elimination by interferon a/b induction than nuclear HBV CCC DNA.  These observations support the contention that elimination of nuclear HBV CCC DNA is the major problem in resolving chronic HBV infection.  In addition, these findings help to explain why interferon a therapy may reduce patients’ viral load without necessarily preventing the recurrence of viral biosynthesis after the completion of therapy.
​

Characterization of the role of FoxA/HNF3 transcription factor deficiency in regulating HBV transcription and replication in vivo in HBV transgenic mice
​

Based upon the observations that FoxA/HNF3 modulates HBV biosynthesis in cell culture, it was of interest to investigate the effect of the loss of FoxA/HNF3 on HBV biosynthesis in vivo by examining the properties of FoxA/HNF3-deficient HBV transgenic mice.  Liver-specific FoxA/HNF3-deficient HBV transgenic mice have been produced and partially characterized.  This analysis demonstrated that the loss of the majority of FoxA expression around birth results in the complete loss of HBV transcription and replication indicating a critical role for FoxA/HNF3 in HBV biosynthesis.  In addition, immunohistochemical analysis of the liver from these mice supports the contention that viral biosynthesis has been lost from the vast majority of hepatocytes (and hence these mice might be considered “cured”).  However these mice do display extensive fibrosis associated with biliary epithelial cell (cholangiocyte) proliferation.  Despite these histological observations, hepatocyte metabolic function is generally only slightly altered under normal physiological conditions.  Of note, our developmental studies indicate that at one to two weeks of age there are no major differences in extracellular matrix deposition (i.e. fibrosis), cytokine levels or biliary epithelial cell proliferation between wild type and FoxA/HNF3-deficient mice.  Therefore it is apparent that it is the FoxA/HNF3 deficiency that is responsible for the loss of viral biosynthesis in this system.

The reason(s) higher levels of FoxA expression are essential for HBV biosynthesis than for appropriate liver-specific gene expression in hepatocytes is unclear but may relate to the observation that there is a very high density of FoxA sites within the viral genome (seven FoxA sites in 3.2kb DNA).  Interestingly, FoxA proteins are considered to be “pioneer” transcription factors which developmentally regulate chromatin structure in a manner marking genes for expression at later developmental stages.  This possibility led us to investigate the methylation status of the viral genome in the FoxA/HNF3-defiecient HBV transgenic mice.  Bisulfite sequencing analysis indicated that transcription was occurring from the unmethylated HBV genomes present in the FoxA3/HNF3-expressing wild-type HBV transgenic mice but not from the methylated HBV genomes present in the FoxA3/HNF3-deficient HBV transgenic mice.  This observation suggests that normal FoxA expression, but not limiting levels of FoxA, prevents the methylation of specific sites within the HBV genome leading to normal viral transcription and replication.  From these studies, it is apparent that the correct developmental expression of FoxA in the liver is required to permit HBV biosynthesis in hepatocytes.  Therefore targeted disruption of FoxA developmental expression represents a potential mode of irreversible inhibition of HBV transcription, and hence replication, which might be exploited to prevent functional transmission of virus from mother to child at birth, the most common route of HBV transmission.

Oxidative stress, hepatocellular carcinoma (HCC) and hepatitis B virus biosynthesis

The HBV transgenic mouse model of chronic HBV infection is the only convenient small animal model available for the study of this human pathogen.  We have used this model to demonstrate that HBV replication, but not transcription, is inhibited in liver tumors induced by the activation of the PI3K/PDK1/AKT signal transduction pathway in liver-specific PTEN-null HBV transgenic mice.  This observation is important because approximately 50% of primary liver tumors are associated with mutations in the PI3K/PTEN/PDK1/AKT/GSK3b/b-catenin signal transduction pathway and tumors associated with HBV infection often fail to support viral biosynthesis.  To understand this observation, “normal” PTEN-null liver tissue and PTEN-null HCC tissue transcripts plus aged matched PTEN-positive liver tissue transcripts from HBV transgenic mice were subjected to RNA-seq analysis.  The differentially regulated genes in the HCC tumor tissues relative to the adjacent “normal” PTEN-null liver tissues and the control liver tissues derived from littermate HBV transgenic mice expressing PTEN were evaluated by gene set enrichment analysis, gene ontogeny and pathway analysis plus transcription factor enrichment profiling.  In combination, the statistical analysis of the RNA-seq data suggest that the PTEN-null HCC tissue displays features at the transcriptional level that are similar to those induced by hypoxia, altered redox regulation/oxidative stress and alterations in endobiotic/xenobiotic gene regulation.  Basically, it appears that this genomic expression analysis points toward three major transcriptome networks governed by Hif, Nfe2l2 and Car being responsible for a major part of the RNA changes seen as a result of HCC tumor development.  Further analysis will be required to determine the relative contribution of these networks to the observed loss of viral biosynthesis during HCC progression.

PhD, University of Aberdeen (Scotland)