Characterisation of the Hepatitis C Virus Genotype 3 Glycoproteins

Shaw, Megan Louise (2001) Characterisation of the Hepatitis C Virus Genotype 3 Glycoproteins. PhD thesis, University of Glasgow.

Abstract

Hepatitis C virus (HCV) can be classified into six genotypes (1-6) which show 30% nucleotide sequence variability throughout the genome. HCV genotypes 1, 2 and 3 have a world-wide distribution but their prevalence differs from one geographical area to another. In Scotland there is an approximate 50/50 split between individuals infected with HCV genotype 1 and genotype 3. There is little evidence that disease progression or severity differs between the genotypes. On the other hand, one difference which has been consistently demonstrated is the better response of patients infected with genotypes 2 and 3 to interferon treatment than those infected with genotype 1. HCV encodes three structural proteins, core, E1 and E2. The two glycoproteins, E1 and E2, are bE1ieved to be the envE1ope proteins. The lack of a cE1l culture system for the production of HCV virions has meant that the proteins within the envE1ope of the HCV virion remain uncharacterised. However, recombinant forms of the E1 and E2 proteins have been shown to localise to the endoplasmic reticulum (ER) when expressed in cultured cE1ls and to interact with one another to form a complex (Ralston et al., 1993). E2 has also recently been shown to bind to the cE1l-surface molecule, CD81, which, as a result, has been proposed to be an HCV receptor (Pileri et al., 1998). These characteristics of the HCV glycoproteins have been described using constructs derived from genotype 1 only. Comparison of the properties of E1 and E2 with those of other genotypes would identify both conserved features, which are possibly essential to the virus life cycle, and genotype-specific features. In this study, the E1 and E2 proteins of genotype 3 were compared with those of genotype 1 with respect to antibody recognition, subcE1lular localisation, complex formation, glycosylation status and CD81-binding. The role of E2 in mediating resistance to interferon (IFN) through a proposed interaction with PKR (Taylor et al., 1999) was also investigated. The genotype 3 structural genes were amplified by reverse-transcription polymerase chain reaction (RT-PCR) from the serum of an HCV genotype 3a-infected patient. This isolate was named HCV 3a-Gla (Gla-3a) and was used as the genotype 3 representative, whereas genotype 1 was represented by the H77 isolate. Sequence comparison of Gla-3a with four published genotype 3a isolates and the genotype 1 isolate revealed that the genotype 3 E2 protein was 6 amino acids longer and had one fewer cysteine residue than that of genotype 1. In terms of length and the number of cysteine residues, the E1 proteins of genotypes 1 and 3 were identical. E2 of Gla-3a (and three other genotype 3 isolates) had one fewer predicted glycosylation site compared to genotype 1, whereas the predicted glycosylation sites in E1 were identical between genotypes 3 and 1. Expression of the core, E1 and E2 proteins was achieved using either the Semliki Forest Virus (SFV) expression system for production of large amounts of recombinant protein or the pcDNA system, in which recombinant gene expression is under control of a cytomegalovirus (CMV) promoter. None of the genotype 1 anti-E1 antibodies were able to recognise genotype 3 E1. Three genotype 1 anti-E2 monoclonal antibodies recognised genotype 3 E2 and they all recognised epitopes in a highly conserved sixteen amino acid region. The genotype 3 glycoproteins localised to the ER and showed evidence of both aggregate and native complex formation as described for those of genotype 1. The rE1ative mobilities of untreated and glycosidase-treated proteins confirmed the predicted glycosylation status of the E2 proteins of genotypes 1 and 3 but the genotype 3 E1 protein appeared to have one additional glycan compared to genotype 1 E1. The deglycosylated genotype 1 E1 protein also seemed smaller than that of genotype 3. Neither of these observations was predicted from the sequence comparison. In contrast to genotype 1 E2, the recombinant genotype 3 E2 protein did not bind to human CD81, although interestingly, prE1iminary results on the ability of virions in human sera to bind CD81 suggested that genotype 3 virions did bind but that genotype 1 virions did not. The ability of the PKR-eIF2alpha phosphorylation homology domain (PePHD) in the E2 protein to predict response to IFN treatment was not validated for genotype 3 and the proposed interaction of genotype 1 E2 with PKR was not confirmed, either by co-localisation or immunoprecipitation. Therefore the role of E2 in mediating IFN resistance remains unclear. In conclusion, the glycoproteins of both genotypes 1 and 3 behave similarly with respect to ER localisation and complex formation in this experimental system. Glycosylation status of the glycoproteins varies between the genotypes and further investigation is required to determine the true glycosylation status of genotype 3 E1. The lack of CD81 binding of the genotype 3 E2 protein is intriguing. This will have to be confirmed with additional genotype 3 isolates but it does highlight the need to include representatives of other genotypes when analysing interactions between viral and cE1lular proteins.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Adviser: Liz McCruden
• Keywords: Virology
• Date of Award: 2001
• Depositing User: Enlighten Team
• Unique ID: glathesis:2001-75741
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 19 Nov 2019 18:17
• Last Modified: 19 Nov 2019 18:17
• URI: https://theses.gla.ac.uk/id/eprint/75741