REPLICATION AND PATHOGENESIS
OF HEPATITIS B VIRUSES

CHRISTOPH SEEGER, Ph.D., Senior Member

This laboratory investigates the biology of hepatitis B viruses (HBVs). We focus our research program on the regulation of woodchuck (WHV) and duck HBV (DHBV) replication during transient and chronic infections. Aims include investigations on host proteins that control morphogenesis and disassembly of virus particles in infected hepatocytes and on the mechanism(s) by which the immune system inhibits viral replication during recovery from transient infections. We anticipate that the information gained from our studies will shed light on host-virus interactions and, thus, can be exploited for the design of effective antiviral therapies necessary to cure chronic hepadnavirus infections.

A previous study on acute hepadnavirus infection in woodchucks revealed that viral clearance from the liver requires less than four weeks even if every hepatocyte is infected. Based on results obtained with cytological and serological assays, it has been proposed that replacement of infected hepatocytes through liver regeneration can contribute to clearance, but cannot solely account for it (Kajino et al., J. Virol. 68:5792, 1994). Thus, hepatocytes must have a means to accelerate the turnover of all viral components in order to prevent the intracellular accumulation of newly synthesized virus particles. Our goal is to understand the molecular mechanism(s) that account(s) for the sudden disappearance of viral proteins and DNA from infected livers. Our hypothesis is that immune mediated secretion of certain cytokines and/or chemokines causes changes in hepatocytes that lead to inhibition of virus replication. This hypothesis is supported by studies with HBV transgenic mice suggesting that such pathways could be activated by interferon gamma (IFN-g) and tumor necrosis factor alpha (TNF-a) (Guidotti et al., Immunity 4:25, 1996).

To understand how a transient infection of HBV resolves, we have characterized the immune response in the liver of 15 WHV-infected woodchucks with woodchuck specific reagents including markers for Tlymphocytes (CD3, CD4, CD8), cytokines (IFN-g, TNF-a), and 2'5'-oligoadenylate synthetase (2'5'-OAS) as an indicator of an IFN-a response. Our results showed that infections, whether transient or chronic, lead to an influx into the liver of T cells with CD4 and CD8 markers, and that this influx was associated with the expression of IFN-g and TNF-a. Similarly, we observed a transient increase of 2',5'-oligoadenylate synthetase (OAS) expression during the recovery phase. All transiently infected animals exhibited significantly higher levels of CD3- and CD8-positive T cells in the liver in comparison to chronically infected animals. Interestingly, the CD4 levels varied among two cohorts of transiently infected woodchucks. In cohort I, CD4 levels were increased approximately 25-fold as compared to uninfected animals whereas in cohort II the increase in CD4 was only 3-fold, similar to chronically infected woodchucks. The observed levels of CD4 correlated well with the observed levels of the cytokines INF-g and TNF-a. Notably, the viremic phase in woodchucks of cohort I lasted several weeks longer than in cohort II. Histologic analysis of liver samples from a woodchuck biopsied just at the time of recovery revealed an approximate 5-fold elevation in apoptosis of hepatocytes associated with increased levels of cells expressing proliferating cell nuclear antigen (PCNA), a marker for liver regeneration.

From the data generated thus far, we draw the following preliminary conclusions from our results: 1) Recovery from an infection can be accompanied by increased expression of certain cytokines, an observation that is consistent with a model invoking cytokines as instigators of an antiviral cellular response. 2) Clearance from a transient infection appears also to occur under conditions where cytokine expression does not significantly differ from expression observed with chronically infected animals. 3) Recovery appears to always be associated with an increase of CD8-positive cells, but is not necessarily associated with a comparable increase in CD4-positive T cells.

Hepadnavirus infections can cause many pathological changes that vary in severity among infected patients and animals. The most obvious changes observed during acute and chronic hepatitis include infiltration of lymphocytes into the parenchyma of the liver, spotty to piecemeal necroses, fibrosis leading to cirrhosis, and finally the development of primary hepatocellular carcinoma. The molecular events responsible for these pathological changes remain largely elusive. Identification of the genes whose expression is altered during hepadnaviral infections may provide insight into the disease mechanism and may lead to the development of diagnostic tools for the prediction of sequelae associated with chronic liver disease.

As an opening to such an analysis, we prepared cDNA libraries that were enriched for genes that are differentially induced in WHV infected livers. More than 1000 cDNA clones were arrayed on nylon membranes and hybridized with cDNA probes made from mRNA of normal, acute or chronic WHV-infected woodchuck livers. So far, we have identified genes that are overexpressed more than 5-fold in more than 50% of WHV infected woodchucks. Among these are two genes encoding proteins regulating lipid metabolism, stearoyl-CoA-desaturase (SCD1) and fatty acid binding protein (FABP). Expression of both genes was elevated in all four chronically infected animals examined. In five of six transiently infected animals, expression levels of both genes increased during the viremic phase and declined following viral clearance. Interestingly, the expression of SCD1 and FABP are not changed in tissue cultured cells that replicate WHV, suggesting that induction of those genes depends on the environmental changes in the liver as they occur during a natural infection.

An apparent paradox of virus replication is that infected cells must be permissive for both uncoating and assembly of virus particles. Since both pathways cannot be active at the same time, an important question concerns the mechanisms responsible for the switch from the "uncoating" to the "assembly" mode. In hepadnavirus, one possibility to keep pathways leading to virus destruction and production separated from each other is that DNA containing particles are inherently unstable in the cytoplasm of infected cells and that "escape" into the lumen of the endoplasmic reticulum shelters them from uncoating. Such a model could explain why capsids enter the uncoating pathway during the early phase of an infection when escape into the endoplasmic reticulum is not possible due to the lack of sufficient envelope components that trigger this step. Regurgitation of cores leads to the amplification of viral DNA in nuclei of infected hepatocytes and is believed to be responsible for the persistence of virus infection, a hallmark of the hepadnavirus life cycle.

In looking for possible targets involved in determining the fate of viral capsid, we found a consensus recognition sequence for the protein kinase cdc2 in the core polypeptide of DHBV. This motif is located at the end of a proposed alpha helix thought to mediate the assembly of the 120 core dimers into core particles. Notably, this site is also exposed on the surface of core shells. Our model predicts that uncoating occurs as a consequence of protein phosphorylation at this site, a mechanism similar to that controlling nuclear envelope breakdown; nuclear envelope breakdown depends on the depolymerization of nuclear lamins induced through the phosphorylation of specific serine residues by cdc2 kinase. To test our hypothesis, we mutated the threonine residue present in the cdc2 consensus motif of the DHBV core polypeptide to alanine or aspartate to mimic unphosphorylated and phosphorylated threonine residues, respectively. Consistent with our hypothesis, we found that the threonine to aspartate change leads to the degradation of core protein and prevents viral DNA replication. Conversely, we found that replacement with alanine does not affect virus production in transfected cells. However, this mutant virus is not infectious for primary hepatocyte cultures. While this observation is consistent with our model invoking phosphorylation as a means for uncoating of viral capsids, we found that the alanine variant can amplify viral DNA in transfected cell cultures. Efforts are in progress to explain the apparent discordance among results obtained with transfected cells and primary hepatocyte cultures.

SEEGER, C., MASON, W.S. Chronic hepadnavirus infections of the woodchuck and duck. In Persistent Viral Infections, edited by R. Ahmed and I. Chen. John Wiley and Sons, Ltd. pp. 607-622, 1999.

HU, J., SEEGER, C. RNA signals that control DNA replication in hepadnaviruses. Semin. Virol. 8:5502-5511, 1998.

SEEGER, C., MASON, W. Replication of the hepatitis B virus genome. In Hepatitis B virus: Molecular Mechanism in Disease and Novel Strategies for Therapy, edited by R. Koshy and W.H. Caselmann. Imperial College Press, London, pp. 1-20, 1998.

^§   Fox Chase researcher

^a   G. Chen: Present address--National Institutes of Health, Bethesda, MD 20892

^b   H. Zhou: Present address--CIGNA, Philadelphia, PA 19103

Illustrations or unpublished data in these reports should not be used without permission of the author.

Fox Chase Cancer Center

Scientific Report 1998