Hepatitis C virus NS3-4A inhibits the peroxisomal MAVS-dependent antiviral signalling response

Abstract

Hepatitis C virus (HCV) is the cause of one of the most prevalent viral infections worldwide. Upon infection, the HCV genome activates the RIG-I-MAVS signalling pathway leading to the production of direct antiviral effectors which prevent important steps in viral propagation. MAVS localizes at peroxisomes and mitochondria and coordinate the activation of an effective antiviral response: peroxisomal MAVS is responsible for a rapid but short-termed antiviral response, while the mitochondrial MAVS is associated with the activation of a stable response with delayed kinetics. The HCV NS3-4A protease was shown to specifically cleave the mitochondrial MAVS, inhibiting the downstream response. In this study, we have analysed whether HCV NS3-4A is also able to cleave the peroxisomal MAVS and whether this would have any effect on the cellular antiviral response. We show that NS3-4A is indeed able to specifically cleave this protein and release it into the cytosol, a mechanism that seems to occur at a similar kinetic rate as the cleavage of the mitochondrial MAVS. Under these conditions, RIG-I-like receptor (RLR) signalling from peroxisomes is blocked and antiviral gene expression is inhibited. Our results also show that NS3-4A is able to localize at peroxisomes in the absence of MAVS. However, mutation studies have shown that this localization pattern is preferred in the presence of a fully cleavable MAVS. These findings present evidence of a viral evasion strategy that disrupts RLR signalling on peroxisomes and provide an excellent example of how a single viral evasion strategy can block innate immune signalling from different organelles.

Keywords: Hepatitis C virus; MAVS; NS3-4A; peroxisomes.

Figure 1

Localization pattern of the different peroxisomal MAVS used in this study. (A) Schematic representation of MAVS‐WT and mutant MAVS constructs used in this study. The cleavage site is represented by a scissors. (B) (a–c) MAVS‐PEX intracellular localization in Mefs MAVS‐KO cells (a) Myc‐MAVS‐PEX, (b) PMP70, (c) merge image of a and b. (d–f) MAVS500‐PEX intracellular localization in Mefs MAVS‐KO cells (d) Myc‐MAVS500‐PEX, (e) PMP70, (f) merge image of d and e. Arrows indicate co‐localization loci. Bars represent 10 μm. Representative images of three independent experiments.

Figure 2

NS3‐4A cleaves the peroxisomal MAVS with similar kinetics as the mitochondrial MAVS. (A) (a–d) MAVS‐PEX is cleaved by NS3‐4A in Mefs MAVS‐KO cells. (a) Myc‐MAVS‐PEX, (b) PMP70, (c) GFP‐NS3‐4A, (d) merge image of a, b and c. (e‐h) MAVS500‐PEX is not cleaved by NS3‐4A (e) Myc‐MAVS500‐PEX, (f) PMP70, (g) GFP‐NS3‐4A, (h) merge image of e, f and g. Arrows indicate co‐localization loci between MAVS‐PEX or MAVS500‐PEX with peroxisomes and NS3‐4A. Full‐head arrows indicate co‐localization loci between peroxisomes and NS3‐4A. Bars represent 10 μm. (B) Western blot analysis of NS3‐4A cleavage of WT and mutant MAVS in Mefs MAVS‐KO cells. Arrow indicates the cleavage product of MAVS. (C) Time course of Myc‐MAVS‐PEX and Flag‐MAVS‐MITO cleavage by GFP‐NS3‐4A in Mefs MAVS‐KO cells. Arrow indicates the cleavage product of MAVS. (D) Pull‐down analysis of the Myc‐MAVS‐PEX and Flag‐MAVS‐MITO in Mefs MAVS‐KO cells. Immunoprecipitation was performed with antibodies against Myc and Flag. Arrow indicates the cleavage product of MAVS. Representative images of three independent experiments.

Figure 3

Cleavage of the peroxisomal MAVS by NS3‐4A inhibits the peroxisomal‐dependent production of antiviral compounds. RT‐qPCR analysis of IRF1 and viperin mRNA expression in Mefs MAVS‐KO cells expressing Myc‐MAVS‐PEX and stimulated with GFP‐RIG‐I‐CARD in the presence or absence of GFP‐NS3‐4A. GADPH was used as control. Data represent the means ± S.E.M. of three independent experiments. Error bars represent S.E.M.. *P < 0.05 in one‐way anova, with Bonferroni's post‐test, conditions were compared with the control Myc‐MAVS‐PEX condition.

Figure 4

NS3‐4A intracellular localization analysis in Mefs MAVS‐KO cells. (A) (a–c) NS3‐4A intracellular localization in the absence of MAVS‐PEX (a) GFP‐NS3‐4A, (b) PMP70, (c) merge image of a and b. (d–f) NS3‐4A intracellular localization in the presence of MAVS‐PEX, (d) GFP‐NS3‐4A, (e) PMP70, (f) merge image of d and e. Arrows indicate co‐localization loci. Bars represent 10 μm. Representative images of three independent experiments. (B) Co‐localization between NS3‐4A and peroxisomes was analysed using Manders' coefficient. Data represent the means ± S.E.M. of three independent experiments, 45 cells were analysed for each condition. Error bars represent S.E.M.. **P < 0.01 in one‐way anova, with Bonferroni's post‐test.

Figure 5

Model of organelle‐specific MAVS antiviral defence and HCV NS3‐4A effect. (A) During infection, viral RNA is released into the cytosol where it is recognized by RLR receptors. RLR receptors activate MAVS present at peroxisomes, mitochondria and MAM. Peroxisomal MAVS induces an early antiviral response, which is complemented by mitochondrial MAVS activation that mediates the expression of ISG through the secretion of type I IFN, promoting a delayed but sustained response. (B) HCV produces NS3‐4A, a non‐structural protein that, among other functions, allows evasion of the cellular antiviral defences. HCV NS3‐4A cleaves the adaptor protein MAVS at peroxisomes, mitochondria and MAM. MAVS cleavage leads to its release into the cytosol impairing the downstream signalling from peroxisomal and mitochondrial MAVS.