Molecular mechanisms of interferon resistance mediated by NS5A and core genes of hepatitis C viruses and the effects of NS5A protein on the regulation of interferon signaling pathway

Abstract

Response or non-response of hepatitis C virus (HCV) infection to interferon- alpha (IFN-) based therapy has been documented to be associated with both host and viral factors, including the genotypes of HCV. Mutations in several genomic regions of HCV have been implicated in influencing the response to IFN treatment. The present study investigated the correlation between amino acid (aa) mutations in the core and non-structural protein 5A (NS5A) regions of HCV genotypes 1a, 1b, 3a, 3b, and 6f and clinical outcome in HCV-infected patients who were treated with pegylated IFN-α (Peg-IFN-α) and ribavirin (RBV). The full-length core and NS5A genes of HCV were amplified from patient’s sera by using nested reverse- transcription polymerase chain reaction (RT-PCR) and then the PCR products were sequenced by direct DNA sequencing method. The nucleotide sequences were translated into aa sequences and were compared with the corresponding reference sequences of each genotype. Analysis of the core protein in pretreatment sera revealed a highly conserved sequences among HCV genotypes (1a, 1b, 3a, and 3b) and no significant difference between the sequences of the viruses from responders, vi non-responders, and relapsers, with the exception of HCV-6f where the mean number of mutations in the core was significantly higher in the viruses obtained from responders compared to treatment failure group (non-responders and relapsers) with P-value of 0.023. Analysis of NS5A revealed that the number of aa mutations in full- length NS5A, C-terminus, IFN- sensitivity determining region (ISDR), variable region 3 (V3), and V3 plus flanking region of HCV-1b and RNA-activated protein kinase binding domain (PKRBD) of HCV-6f were significantly higher in responders than in the treatment failure group (P=0.010, 0.031, 0.046, 0.020, 0.006, and 0.022, respectively). Moreover, specific aa substitutions in NS5A appeared to be significantly associated with the treatment failure group were observed at positions 78 and 305 for HCV-1b (P=0.028), 64 for HCV-1a (P=0.033), and 52 for HCV-6f (P=0.045). All together, the data imply that aa mutations in NS5A protein correlate with the response to Peg-IFN-α and RBV treatment and these mutations might be a predictive value for the clinical outcome. Analysis of aa variations before and after Peg-IFN-α and RBV treatment in the core and NS5A proteins of HCV isolated from non-responders and relapsers revealed that aa sequences in the core, ISDR, PKRBD, and V3 of NS5A before treatment were highly conserved and very similar to those of the viruses after treatment. However, when comparing the sequences of the viruses from relapsers to those from non-responders, it was found that the number of mutations in full-length NS5A, particularly at N-terminus from relapsers were significantly greater than those from non-responders, P=0.046 and 0.017, respectively. Study of effects of NS5A proteins on IFN-α induced Jak-STAT signaling pathway was done by using both NS5A expressing plasmids and NS5A recombinant viruses. In order to express NS5A protein in cell cultures, a mammalian expression vector, pCAGGS-V5/His, was generated and used to construct a recombinant plasmid encoding the HCV NS5A of genotypes 1a, 1b, 3a, and 3b. Expression of NS5A was performed by transfecting the recombinant plasmid DNAs into Huh7.5.1 cells and the viral protein products were detected by Western blot analysis using both anti-V5 and anti-NS5A monoclonal antibodies. The effects of NS5A protein on the IFN signaling were investigated in the presence of IFN-α treatment. Whole cell lysates were harvested to measure relative luciferase activity and to perform Western blot analysis vii and immunoprecipitation (IP) assay, and mRNAs were collected for performing real time PCR in order to examine expression levels of interferon-stimulated genes (ISGs). The results showed that the expressed NS5A proteins of HCV genotypes 1a, 1b, 3a, and 3b significantly inhibited IFN-induced ISRE promoter activity, STAT1 phosphorylation (P-STAT1) levels, and ISG expression. In addition, NS5A proteins of genotypes 1a and 1b had stronger inhibitory effects against IFN than those of genotypes 3a and 3b. IP assay demonstrated that NS5A proteins of genotype 1 exhibited stronger binding to STAT1 than those of genotype 3. Domain mapping study revealed, for the first time, that the C-terminal NS5A conferred these inhibitory effects on IFN signaling through binding to STAT1 protein. Further experiments were performed using NS5A/JFH1 recombinant virus replication model in which complete NS5A in J6/JFH1 virus was replaced with NS5A of genotype 1a or 3a to generate infectious recombinant viruses H77 (1a NS5A-J6/JFH1) and S52 (3a NS5A- J6/JFH1), respectively. Huh7.5.1 cells were infected with recombinant viruses and the inhibitory effects of NS5A on IFN signaling were examined. The results demonstrated that H77 recombinant virus conferred more resistance to IFN treatment than S52 through stronger reduction of IFN-induced ISRE signaling, P-STAT1 levels, and ISG expression. These findings lead us to propose a model in which HCV mediates its antagonistic effects on IFN treatment through C-terminal NS5A binding to STAT1 which results in the reduction of ISRE promoter activity, P-STAT1, and ISG expression, and therefore, leads to inhibition of IFN activity. In conclusion, this study provides a clue for better understanding of the role of aa mutations in the core and NS5A proteins of HCV genotypes 1a, 1b, 3a, 3b, and 6f in responding or non-responding to Peg-IFN-α and RBV therapy and provides a new insights into the mechanisms by which NS5A protein from different genotypes influence IFN signaling. These observations may account for genotype-related differences in the response of HCV to IFN and RBV treatment