Influenza virus NS1- C/EBPβ gene regulatory complex inhibits RIG-I transcription

Abstract

Influenza virus non-structural protein 1 (NS1) counteracts host antiviral innate immune responses by inhibiting Retinoic acid inducible gene-I (RIG-I) activation. However, whether NS1 also specifically regulates RIG-I transcription is unknown. Here, we identify a CCAAT/Enhancer Binding Protein beta (C/EBPβ) binding site in the RIG-I promoter as a repressor element, and show that NS1 promotes C/EBPβ phosphorylation and its recruitment to the RIG-I promoter as a C/EBPβ/NS1 complex. C/EBPβ overexpression and siRNA knockdown in human lung epithelial cells resulted in suppression and activation of RIG-I expression respectively, implying a negative regulatory role of C/EBPβ. Further, C/EBPβ phosphorylation, its interaction with NS1 and occupancy at the RIG-I promoter was associated with RIG-I transcriptional inhibition. These findings provide an important insight into the molecular mechanism by which influenza NS1 commandeers RIG-I transcriptional regulation and suppresses host antiviral responses.

Keywords: C/EBPβ; Influenza; NS1; RIG-I; Transcriptional regulation.

Fig. 1.

hC/EBPβ binds to predicted binding site on RIG-I promoter. (A) RIG-I promoter region sequence with potential regulatory elements (underlined) identified by TFSEARCH. (B) Sequence of probes used in EMSA. (C) EMSA showing hC/EBPβ binding to hC/EBPβ sequence in RIG-I promoter. Lanes 1, free Cy5 hC/EBPβ probe without incubation with A549 nuclear extracts (NE); lane 2, Cy5 hC/EBPβ probe incubated with uninfected A549 NE; Lane 3, Cy5 hC/EBPβ probe incubated with PR8-infected A549 NE; lane 4, PR8-infected A549 NE pre-incubated with hC/EBPβ cold competitor (100 molar excess) followed by Cy5 probe incubation; lane 5, PR8-infected A549 NE pre-incubated with consensus cold competitor (100 molar excess) followed by Cy5 probe incubation; lane 6, PR8-infected A549 NE pre-incubated with nonspecific (N.S.) cold competitor 1 (100 molar excess) followed by Cy5 probe incubation; lane 7, PR8-infected A549 NE pre-incubated with nonspecific (N.S.) cold competitor 2 (100 molar excess) followed by Cy5 probe incubation. Probes were separated on a 4-12% gradient non-denaturing polyacrylamide gel and detected using Fluorescent Imaging Odyssey as described.

Fig. 2.

hC/EBPβ negatively regulates RIG-I and IFNβ activation. A549 cells were untransfected or co-transfected with vector or hC/EBPβ expression vector/ control siRNA or hC/EBPβ siRNA and RIG-I or mutant RIG-I promoter luciferase reporter plasmids for 24 h. Cells were then infected with PR8 (1.0 MOI) for indicated time periods and harvested for various analysis. A549 cells were mock infected or infected with PR8 (1.0 MOI) infection at indicated time period and analysed for (A) RIG-I and NS1 protein expression by western blot and (B) NPvRNA, (C) RIG-I mRNA and (D) IFNβ mRNA expression by qRT-PCR. A549 cells transfected with vector or hC/EBPβ expression vector were mock infected or infected with PR8 (1.0 MOI) infection for 6 h and analysed for (E) RIG-I, C/EBPβ and NS1 protein expression by western blot and (F) NPvRNA, (G) RIG-I mRNA and (H) IFNβ mRNA expression by qRT-PCR. A549 cells transfected with control siRNA or hC/EBPβ siRNA were mock infected or infected with PR8 (1.0 MOI) infection for 6 h and analysed for (I) RIG-I, C/EBPβ and NS1 protein expression by western blot and (J) NPvRNA, (K) RIG-I mRNA and (L) IFNβ mRNA expression by qRT-PCR. (M, N, O and P) Luciferase reporter activity (RLU) in A549 cells untransfected or co-transfected with vector or hC/EBPβ expression vector and RIG-I promoter luciferase reporter plasmids for 24 h, followed by PR8 (1.0 MOI) infection for 6 h. Data shown are means ± S.D. from three independent experiments. *p<0.05 compared to respective controls.

Fig. 3.

NS1 promotes hC/EBPβ phosphorylation and interacts with hC/EBPβ. A549 cells, untransfected or co-transfected with vector, wild type hC/EBPβ or mutants threonine 235 to alanine (T235A), tyrosine 274 to phenylalanine (Y274F) or leucine 320 to arginine (L320R) or myc-NS1 expression vector and/or RIG-I promoter reporter plasmids for 24 h. Cells were then infected or mock infected with PR8 or PR8ΔNS1 (1.0 MOI) for 6 h in the presence or absence of U0126 (10 μM). (A) Immunoblot analysis of RIG-I, phC/EBPβ, hC/EBPβ, pERK, ERK and NS1 in A549 cells mock infected or infected with PR8 (1.0 MOI) for 6 h in the presence or absence of U0126 (10 μM). (B) RIG-I promoter activity (RLU) in A549 cells transfected with RIG-I promoter reporter plasmids for 24 h followed by mock or PR8 or PR8ΔNS1 (1.0 MOI) infection for 6 h in the presence or absence of U0126 (10 μM). (C) RIG-I promoter activity (RLU) in A549 cells, transfected with vector or wild type hC/EBPβ or hC/EBPβ mutants threonine 235 to alanine (T235A), tyrosine 274 to phenylalanine (Y274F) or leucine 320 to arginine (L320R) and/or RIG-I promoter reporter plasmids for 24 h followed by mock or PR8 or PR8ΔNS1 (1.0 MOI) infection for 6 h. (D) Total cell lysate from PR8-infected A549 cells were subjected to immunoprecipitation using anti-C/EBPβ or anti-NS1 antibodies. Immunoprecipitates were analysed for presence of NS1 and hC/EBPβ by immunoblotting. (E) Immunoblot analysis of phospho hC/EBPβ, hC/EBPβ, pERK, ERK and NS1 in A549 cells transfected with vector or myc-NS1 expression vector. (F) RIG-I promoter activity (RLU) in A549 cells co-transfected with RIG-I promoter reporter plasmids and vector or myc-NS1 expression vector, followed by mock or PR8ΔNS1 (1.0 MOI) infection. (G) IFNβ promoter activity (RLU) in A549 cells co-transfected with RIG-I promoter reporter plasmids and vector or myc-NS1 expression vector, followed by mock or PR8ΔNS1 (1.0 MOI) infection. (H) Total cell lysate from vector or myc-NS1 expression vector transfected A549 cells were subjected to immunoprecipitation using anti-myc, anti-C/EBPβ or control antibodies. Immunoprecipitates were analysed for presence of NS1 and hC/EBPβ by immunoblotting. Data shown are means ± S.D. from three independent experiments. *p<0.05 compared to respective controls.

Fig. 4.

NS1 co-operatively enhances hC/EBPβ recruitment at the RIG-I promoter region. ChIP assay using antibodies against phospho-hC/EBPβ and NS1 coupled with real-time PCR were performed on A549 cells infected with PR8 (1.0 MOI) or NS1ΔPR8 (1.0 MOI) in the presence or absence of U0126 at the indicated times. The precipitated DNA was analysed for the presence of RIG-I or CXCL10 promoter region by quantitative real-time PCR. (A) p-hC/EBPβ occupancy at RIG-I promoter in A549 cells infected with infected with PR8 or NS1ΔPR8 in the presence or absence of U0126. (B) NS1 occupancy at RIG-I promoter in A549 cells infected with PR8 or NS1ΔPR8 in the presence or absence of U0126. (C) p-hC/EBPβ occupancy at RIG-I promoter in A549 cells transfected with vector, wt hC/EBPβ or mutants T235A, Y274F or L320R followed by PR8 or NS1ΔPR8 infection. (D) p-hC/EBPβ occupancy at CXCL10 promoter in A549 cells infected with infected with PR8 or NS1ΔPR8 in the presence or absence of U0126. (E) NS1 occupancy at CXCL10 promoter in A549 cells infected with infected with PR8 or NS1ΔPR8 in the presence or absence of U0126. (F) p-hC/EBPβ occupancy at CXCL10 promoter in A549 cells transfected with vector, wt hC/EBPβ or mutants T235A, Y274F or L320R followed by PR8 or NS1ΔPR8 infection. All data were normalized to the amount of chromatin added to each precipitation reaction and expressed as fold-increase over irrelevant (TCR) antibody controls. Data shown are means ± S.D. from three independent experiments. *p<0.05 compared to respective controls.

Fig. 5.

Heat map of differentially expressed genes based on PCR Array analysis. NHBE cells were transfected with vector, hC/EBPβ, control siRNA or hC/EBPβ siRNA for 24h. Cells were then mock infected or infected with PR8 (1.0 MOI) for 6 h. Gene expression were analysed by real-time RT-PCR using PCR array kit as per manufacturer protocol. Data shown are average from two independent experiments. The scale shows the level of gene expression where red corresponds to upregulation (log2 fold).