Mutational analysis of the hepatitis C virus E1 glycoprotein in retroviral pseudoparticles and cell-culture-derived H77/JFH1 chimeric infectious virus particles

Abstract

Cell entry by enveloped viruses is mediated by viral glycoproteins, and generally involves a short hydrophobic peptide (fusion peptide) that inserts into the cellular membrane. An internal hydrophobic domain within E1 (aa262-290) of hepatitis C virus (HCV) may function as a fusion peptide. Retrovirus-based HCV-pseudotyped viruses (HCVpp; genotype 1a) containing Ala or Pro substitutions at conserved amino acid positions within this putative fusion peptide were generated. Mutation of conserved residues significantly reduced efficiency of HCVpp entry into Huh-7 cells. The majority of amino acid substitutions appeared to disrupt necessary interactions between E1 and E2. For some mutants, reductions in HCVpp-associated E1 were associated with the incorporation of a high molecular weight, hyperglycosylated E2 that displayed decreased CD81-binding. Other entry-deficient mutants displayed normal E1E2 incorporation into pseudoparticles and normal CD81-binding, and therefore might affect viral fusion. One mutant (S283P) consistently displayed two- to threefold higher infectivity than did wild-type. Three mutations that decreased HCVpp infectivity also reduced levels of HCVcc infectious virus production. However, the S283P mutation had a different effect in the two systems as it did not increase production of infectious HCVcc. This comprehensive mutational analysis of the putative HCV fusion peptide provides insight into the role of E1 in its interaction with E2 and in HCV entry.

Fig. 1

Effects of E1 mutations on HCVpp entry. Plasmids encoding E1 mutated glycoproteins at (a) Gly or Cys, (b) Leu, (c) Arg, or (d) Ser residues, in the context of wild-type E2 were used to generate mutant HCVpp, which were then inoculated onto Huh-7 cells. Infected cells were detected by FACS analysis of GFP expression. HCVpp containing wild-type E1 and E2 typically infected 4–5% of the cultured Huh-7 cells. Infectivity of pseudoparticles generated in the presence of mutated E1 glycoproteins is expressed as a percentage of wild-type, which is arbitrarily set at 100%. Results shown represent three independent transfection/infection experiments and standard errors are indicated on the bars. Supernatants from cells transfected without E1E2-expressing plasmid (no E1E2) and untransfected cells (Untransf.) served as negative controls, which were never above 0.02% GFP-positive.

Fig. 2

Western blot analyses of (a) Gly or Cys, (b) Leu, (c) Arg, or (d) Ser mutants were performed on transfected HEK-293T cell lysates (Cells) and HCVpp pelleted by ultracentrifugation (HCVpp) under reducing conditions. MLV Gag protein was detected with a rabbit antiserum against MLV CAp30. E1 and E2 glycoproteins were detected with Monoclonal antibody A4 and a hyperimmune rabbit serum, LMF87 respectively [22]. Western blot analyses of ‘no E1E2’ controls were always non-reactive with α-E1 and α-E2 Ab, and bands were not detected in lanes representing untransfected cells.

Fig. 3

Analysis of glycans associated with HCV E2. Pelleted mutant and wild-type HCVpp were digested with endo H (H) or PGNase F (F). Undigested controls (U) were treated the same as endo H samples except enzyme was omitted. E2 glycoproteins were then analysed by Western blotting.

Fig. 4

Effects of E1 mutations of E2-CD81 binding. Pelleted HCVpp were resuspended in 250 μL of RSB and a 15 μL aliquot was analysed by Western blotting for E2 (HCVpp-associated; upper panels) as described above. The remainder was subjected to a GST-CD81 pull-down assay and CD81-bound E2 was analysed by Western blotting (CD81-bound; lower panels). Wild-type and mutant HCVpp lysates were adsorbed with GST-human CD81-LEL. As negative controls, binding assays were performed with HCVpp lacking glycoprotein (no E1E2), as well as wild-type HCVpp in the presence of a GST-CD81-LEL double mutant (F186L+E188K) which does not bind E2 (Neg Ctrl) (23). Results shown are representative from two independent experiments. Mean HCVpp infectivity data for each mutant was transposed from Fig. 1 for comparison.

Fig. 5

Effects of E1 mutations on production of extracellular infectious HCVcc. One million Huh-7.5 cells were transfected with in vitro-transcribed RNA representing HJ3/QL or each of the indicated HCVcc mutants. Extracellular infectious virus accumulated by day 4 was measured by focus-forming assays [24,25] in triplicate and means plus standard errors are plotted. ‘Negative’ represents titration of culture fluids from untransfected Huh-7.5 cells. The dotted line represents the cutoff of the assay, which was 10 ffu/mL. These data were consistent for two independent experiments and the results shown were generated from one transfection experiment preformed in triplicate.