Studies of hepatitis C virus envelope proteins : interaction with host cells and as targets for the humoral response

Abstract

Hepatitis C virus (HCV) is a major cause of chronic hepatitis infection worldwide. It leads to chronic infection in over 80% in infected patients and may result in liver cirrhosis, hepatocellular carcinoma and autoimmune disorders.

The aim of this thesis was to study the functional role of human antibodies to different HCV envelope protein E2 (E2) epitopes, to characterize the interaction of die E2 protein with cell surface molecules, and to study the role of glycosylation of El, the other envelope protein of HCV, in pseudotype virus infectivity. The antibody response to hyper variable region 1 (HVR1) of the E2 protein was studied in five patients infected with the same HCV strain. The patients had different clinical outcomes: three developed chronic infection and two resolved the infection. In this study, an early antibody response to HVR1 correlated with the resolution of the infection but not the E2 antibody response. This suggests that early anti-HVR1 antibodies may play an important role in the clearance of the HCV infection.

To further study the nature of protective antibodies to HCV, human monoclonal antibodies against E2 were isolated from a combinatorial Fab library displayed on phages. The library was established from the bone marrow of a chronically HCV infected patient. This patient was infected by HCV of genotype 2b and the recombinant E2 used to select the monoclonal antibodies was of genotype la. Seven distinct antibody clones were further studied. Three clones had high affinity for E2 of genotype lb. All seven clones recognized conformation dependent epitopes, which seem conserved between different genotypes. Four clones expressed as IgG that bound to two or three different E2 epitopes, were NOB-positive for blocking the CD81-E2 interaction in vitro. This suggests the existence of at least two conserved epitopes in E2 that mediate inhibition of the E2-CD81 interaction. An assay for determination of serum antibodies to two of these epitopes was established. No correlation was found between levels of such serum antibodies and the clinical outcome of acute HCV infection.

Since there is not an efficient in vitro system for the propagation of HCV, a pseudotype virus using Vesicular Stomatitis Virus (VSV) bearing HCV glycoproteins was developed. The HCV glycoproteins, El and E2, contain ER retention signals in their transmembrane (TM) domain. We generated chimeric El and E2 glycoprotein constructs (E1-G and E2-G, respectively) by replacing their native TM with sequences encoding TM of the G protein of VSV and its cytoplasmic tail, allowing translocation of E1-G or E2-G to the cell surface.

Subsequently, a deletion of the HVR1 sequence from E2-G was performed, creating the recombinant protein E2(delta)HVR1-G. Pseudotype virus generated using E2(delta)HVR1-G had a reduced plaquing efficiency (50%) as compared to the E2-G pseudotype virus (with intact HVR1). Cells treated with pronase were not permissive for infection with E2(delta)HVR1-G or E2-G pseudotypes. However, heparinase I treatment of cells reduced only E2-G infectivity (by 40%). but not the infectivity of E2(delta)HVR1. Similarly, plaquing efficiency of E2-G pseudotype viruses could be greatly reduced by addition of heparin, while E2(delta)HVR1 could not.. Our results suggest that the HVR1 of E2 glycoprotein binds to cell surface proteoglycans, and may facilitate virushost interaction. The role of glycosylation of chimeric E1-G for intracellular transport, and infectivity of pseudotype virus was investigated. Surface expressed E1-G exhibited sensitivity to Endo-H treatment similar to full length El, suggesting that additional complex oligosaccarides were not added while E1-G was in transit from ER to cell surface. E1-G's four N-glycosylation sites were mutated separately (aspargine ---- glutamine), or in certain combinations. FACS analysis and confocal microscopy revealed a gradual decrease in cell surface expression of E1-G protein with an increasing number of mutations to the glycosylation sites. VSV pseudotype virus generated from E1-G mutants with only two of the four-glycosylation sites still intact displayed infectivity. Nglycosidase F treatment of pseudotype viruses generated from E1-G or its mutants decreased virus titer by - 35%, and the neutralization activity of patient sera did not significantly alter with N-glycosidase F treated pseudotype virus. Our results suggest that glycosylation may not play a key role for interaction of E1-G ectodomain with the host cell surface for pseudotype entry.

In conclusion, several aspects studied in this thesis may provide critical insights into HCV vaccine design and new therapeutic strategies, if any.