Hepatitis A virus suppresses RIG-I-mediated IRF-3 activation to block induction of beta interferon

Abstract

Hepatitis A virus (HAV) antagonizes the innate immune response by inhibition of double-stranded RNA (dsRNA)-induced beta interferon (IFN-beta) gene expression. In this report, we show that this is due to an interaction of HAV with the intracellular dsRNA-induced retinoic acid-inducible gene I (RIG-I)-mediated signaling pathway upstream of the kinases responsible for interferon regulatory factor 3 (IRF-3) phosphorylation (TBK1 and IKKepsilon). In consequence, IRF-3 is not activated for nuclear translocation and gene induction. In addition, we found that HAV reduces TRIF (TIR domain-containing adaptor inducing IFN-beta)-mediated IRF-3 activation, which is part of the Toll-like receptor 3 signaling pathway. As IRF-3 is necessary for IFN-beta transcription, inhibition of this factor results in efficient suppression of IFN-beta synthesis. This ability of HAV seems to be of considerable importance for HAV replication, as HAV is not resistant to IFN-beta, and it may allow the virus to establish infection and preserve the sites of virus production in later stages of the infection.

FIG. 1.

HAV inhibits dsRNA-induced activation of IRF-3 but not of NF-κB. FRhK-4 cells stably transfected with PRDII-CAT (A) or PRDIII-I-CAT (B) reporter plasmids were infected with HAV/7 at an MOI of 1 and transfected with poly(I-C) by DEAE-dextran (DD) 9 days postinfection; 16 h later, cell extracts were analyzed for NF-κB-dependent (A) or IRF-3-dependent (B) CAT expression. Noninfectedcells and cells treated with medium or DEAE-dextran alone were used as controls. Data represent means of two replicates, and experiments were carried out twice. (C) Nuclear translocation of NF-κB was examined by immunofluorescence. MRC-5 cells infected with HAV-GBM were transfected with poly(I-C) by DEAE-dextran (DD); after 3 h, cells were fixed and permeabilized with 4% paraformaldehyde/methanol, and immunostained for NF-κB p65 (fluorescein isothiocyanate) and HAV (Texas Red). Nontransfected and/or noninfected cells were used as controls. Results are representative of two independent experiments. Magnification, ×400 (Axioskop II, Zeiss).

FIG. 2.

HAV inhibits IRF-3 nuclear translocation and phosphorylation. (A to D) FRhK-4/GFP-IRF-3 cells infected with HAV/7 were infected with NDV (MOI, 10); 4 h postinfection, cells were fixed with 4% paraformaldehyde and 0.2% Triton X-100 and immunostained for HAV (Texas Red) (D). Noninfected cells (A) and cells infected only with HAV or NDV (B and C) were used as controls. Magnification, ×400 (Axioskop II, Zeiss). (E) IRF-3 C-terminal phosphorylation was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting. FRhK-4/IRF-3 cells noninfected or infected with HAV/7 were inoculated with NDV at an MOI of 10 for 2 h. At different times postinfection, whole-cell extracts were prepared. Phosphorylated IRF-3 was detected as an additional, slower-migrating form. All results shown are representative of a series of three independent experiments.

FIG. 3.

NDV replication kinetics under one-step growth conditions in FRhK-4 cells infected or not infected with HAV/7. The kinetics show the total NDV titers in the course of 24 h. Titers were determined in FRhK-4 cells by cytopathic effect 6 days after inoculation. Each data point is an average obtained from duplicate titrations of two separate experiments. Error bars indicate standard deviations of the mean. HAV infection of 100% of the cells was proved by indirect immunofluorescence prior to infection with NDV (MOI of 10) (lower half).

FIG. 4.

HAV does not inhibit IFN-β induction and IRF-3 activation by IKKɛ or TBK1. (A) In order to analyze IKKɛ-dependent IFN-β induction (IFN-β-CAT) or IRF-3 activity (PRDIII-I-CAT), noninfected or HAV/7-infected FRhK-4 cells were cotransfected with CAT reporter gene plasmids and an IKKɛ expression vector. NDV infection (MOI of 0.1) was used as a control for functional HAV infection in cells cotransfected with the empty vector. (B) For analysis of TBK1-dependent IFN-β induction (IFN-β-Luc) or IRF-3 activity (PRDIII-I-Luc), noninfected or HAV/7-infected FRhK-4 cells were cotransfected with luciferase reporter gene plasmids and a TBK1 expression vector. NDV infection (MOI of 0.1) was used as a control for functional HAV infection in cells cotransfected with the empty vector. The right panel includes TBK1-transfected plus NDV-infected controls to demonstrate the integrity of the inhibitory mechanism of HAV independently of TBK1 overexpression. In all experiments, NDV infection was performed 24 h after transfection, and reporter gene expression was analyzed 42 h after transfection. Data represent means of at least two replicates, and experiments were carried out at least twice.

FIG. 5.

HAV inhibits IFN-β induction and IRF-3 activation by RIG-I. In order to analyze RIG-I-dependent IFN-β induction (IFN-β-CAT) (A) or IRF-3 activity (PRDIII-I-CAT) (B), noninfected or HAV/7-infected FRhK-4 cells were cotransfected with CAT reporter gene plasmids and a RIG-I expression vector (upper panels). Empty vector cotransfection was used as a control. CAT expression was analyzed 42 h after transfection. Flag-RIG-I overexpression was verified by immunoblotting (lower panels). Data represent means of at least two replicates, and experiments were carried out at least twice.

FIG. 6.

RIG-I N-terminal deletion mutant (RIG-IC) suppresses IFN-β enhancer activation induced by poly(I-C) transfection or NDV infection. FRhK-4 cells were cotransfected with IFN-β-CAT and a RIG-IC expression vector encoding the RIG-I dsRNA-binding helicase domain; 24 h later, cells were transfected with poly(I-C) via DEAE-dextran (DD) (A) or infected with NDV (MOI of 0.1) (B) and 18 h later, cell extracts were analyzed for CAT expression. Data represent means of two replicates, and experiments were carried out twice.

FIG. 7.

HAV negatively affects IFN-β induction and IRF-3 activation by TRIF. TRIF-dependent IFN-β induction (IFN-β-CAT) (A) or IRF-3 activity (PRDIII-I-CAT) (B) was examined in noninfected or HAV/7-infected FRhK-4 cells after cotransfection with CAT reporter gene plasmids and a TRIF expression vector. Empty vector cotransfection was used as a control. CAT expression was analyzed 42 h after transfection. Data represent means of at least two replicates, and experiments were carried out at least twice.

FIG. 8.

Model of HAV interaction with the dsRNA-induced IRF-3 activation pathway. Intracellular dsRNA binds to RIG-I, which regulates downstream activation of IKKɛ or TBK1, which form a complex with adapters, such as TANK (9), by still unknown intermediate components. Activated IKKɛ/TBK1 then phosphorylate IRF-3, leading to its dimerization, nuclear translocation, and transactivity. Extracellular dsRNA binds to TLR3 associated with intracellular vesicles, inducing recruitment of TBK1 and IRF-3 via its adapter TRIF, resulting in IRF-3 phosphorylation and activation. Hepatitis A virus suppresses intracellular dsRNA-induced RIG-I-mediated IRF-3 phosphorylation by a mechanism acting upstream of IKKɛ/TBK1. Additionally, HAV negatively affects TRIF-induced IRF-3 activation, probably inhibiting TLR3 signaling as well.