The influenza virus PB2 protein evades antiviral innate immunity by inhibiting JAK1/STAT signalling

Abstract

Influenza A virus (IAV) polymerase protein PB2 has been shown to partially inhibit the host immune response by blocking the induction of interferons (IFNs). However, the IAV PB2 protein that regulates the downstream signaling pathway of IFNs is not well characterized. Here, we report that IAV PB2 protein reduces cellular sensitivity to IFNs, suppressing the activation of STAT1/STAT2 and ISGs. Furthermore, IAV PB2 protein targets mammalian JAK1 at lysine 859 and 860 for ubiquitination and degradation. Notably, the H5 subtype of highly pathogenic avian influenza virus with I283M/K526R mutations on PB2 increases the ability to degrade mammalian JAK1 and exhibits higher replicate efficiency in mammalian (but not avian) cells and mouse lung tissues, and causes greater mortality in infected mice. Altogether, these data describe a negative regulatory mechanism involving PB2-JAK1 and provide insights into an evasion strategy from host antiviral immunity employed by IAV.

Fig. 1. IAV PB2 protein inhibits IFN signaling pathway.

a, b Luciferase activity in HEK293T cells transfected with IFNβ promoter, ISRE or NF-κB luciferase reporter, Renilla luciferase plasmid, and PB2 plasmids (including NS1-PR8 and empty vector [Vec] were chosen as positive and negative controls, respectively) and treated with SeV (a) or poly(I:C) (b) (n = 3 biologically independent samples). c, d Luciferase activity in HEK293T cells transfected with ISRE or STAT1 promoter-luciferase reporter plasmid, Renilla luciferase plasmid, and PB2 plasmids and treated with IFNβ (c) or IFNα (d) (n = 3 biologically independent samples). e, f qPCR analysis of IFIT1 and ISG15 mRNA in A549 cells transfected with PB2 plasmids and treated with IFNβ (e) or IFNα (f); mRNA results are presented relative to those of untreated cells transfected with a control plasmid (n = 3 biologically independent samples). g HEK293T cells were transfected with PB2 or NS1 plasmid and infected with SeV. The supernatants were inactivated by ultraviolet radiation and collected to treat fresh HEK293T cells for 24 h, followed by infection for 12 h with VSV-GFP. The cells were observed under microscopy and then assessed by flow cytometry. Scale bars, 200 μm. h, i Immunoblot analysis of phosphorylated and total STAT1 or STAT2 in HEK293T cells transfected with PB2 plasmids and treated with IFNβ (h) or IFNα (i) (upper). Densitometry analysis of phosphorylated STAT/total STAT ratio (lower). Data are presented as the mean ± SD and are one representative of three independent experiments. Statistical significance in a–f, h, i was determined by unpaired two-tailed Student’s t test.

Fig. 2. IAV PB2 promotes the degradation of JAK1.

a Immunoblots of HEK293T cells were transfected with different concentrations of PB2 plasmids from PR8 (left) or CZ (right). b, c Immunoblots of HEK293T cells transfected with JAK1-His (b) or Flag-STAT1 (c) with different concentrations of PB2 plasmids. d qPCR analysis of JAK1 mRNA level in A549 cells transfected with different concentrations of PB2 plasmids (n = 3 biologically independent samples). Data are presented as the mean ± SD. Statistical significance was determined by unpaired two-tailed Student’s t test. ^ns P > 0.05. e Immunoblots of A549 cells infected with PR8 (left) or CZ (right) at an MOI of 1. f Immunoblots of stable PB2 knockdown HEK293T cells infected with PR8 virus (MOI = 0.1). g Immunoblots of HEK293T cells were transfected with PB2-PR8 and treated with CHX. h Immunoblots of HEK293T cells were transfected with the PB2 plasmids and treated with DMSO, MG132, NH4Cl, or chloroquine (CQ). Data are one representative of three independent experiments.

Fig. 3. IAV PB2 protein increases the K48-linked ubiquitination of JAK1.

a Ni-NTA pull-down analysis of the ubiquitination of JAK1 in HEK293T cells transfected with JAK1-His, HA-ubiquitin (HA-Ub), and PB2 plasmids and treated with MG132. b Ni-NTA pull-down analysis of the ubiquitination of JAK1 in HEK293T cells transfected with JAK1-His, HA-Ub, or its mutants K48 and K63 [K at indicated residue, and K at other residues were simultaneously mutated to arginines], and PB2 plasmids and treated with MG132. c Co-IP analysis of the ubiquitination of JAK1 in A549 cells infected with PR8 or CZ at an MOI of 0.01. d Co-ip analysis of the ubiquitination of JAK1 in stable PB2 knockdown HEK293T cells infected with PR8 virus. WCL, whole-cell lysates. Data are one representative of three independent experiments.

Fig. 4. PB2-mediated JAK1 degradation at residues 859K and 860K.

a Schematic representation of the deletion mutants of JAK1 (upper) and immunoblots of HEK293T cells transfected with JAK1 or its deletion mutants and PB2 plasmids (lower). b Ubiquitination modification online prediction of JAK1 by CPLM 1.0. c Immunoblots of HEK293T cells transfected with JAK1 or its mutants and PB2 plasmids. d Intensities analysis of the bands on the immunoblots (c) from three independent experiments. Data are presented as the mean ± SD and statistical significance was determined by unpaired two-tailed Student’s t test. ^ns P > 0.05. e Immunoblots of HEK293T cells transfected with JAK1 or its mutants and PB2 plasmids. f Molecular model of the JAK1 kinase domain generated by PyMOL (PDB: 4ehz). Residues K859/860 are highlighted in pink. g Ni-NTA pull-down analysis of the ubiquitination of JAK1 in HEK293T cells transfected with JAK1 or its mutants, HA-Ub and PB2 plasmids, and treated with MG132. WCL, whole-cell lysates. Data are one representative of three independent experiments.

Fig. 5. The effects of JAK1 on IAV replication.

a, b Viral titers in HEK293T cells transfected with JAK1 plasmid and infected with PR8 virus (a) or CZ virus (b) (n = 3 biologically independent samples). c Immunoblots of HEK293T cells transfected with control shRNA (shGFP) or shRNA targeting JAK1 (shJAK1) and JAK1 plasmids. d, e Viral titers in shJAK1 HEK293T cells infected with PR8 virus (d) or CZ virus (e) (n = 3 biologically independent samples). hpi, h post-infection. f JAK1 was replaced with an shRNA off-target JAK1 mutant (ΔJAK1) with a 7-nucleotide nonsense mutation in the target sequence of the shJAK1 plasmid. g Immunoblots of stable JAK1 knockdown HEK293T cells transfected with JAK1-His, ΔJAK1-His, or ΔJAK1-K859/860R-His plasmid (upper). Intensities of the bands on the immunoblots from three independent experiments were quantified and normalized with actin (lower). h, i Viral titers in shJAK1 HEK293T cells transfected with ΔJAK1-WT or ΔJAK1-K859/860R plasmid and infected with PR8 virus (h) or CZ virus (i) (n = 5 biologically independent samples). Data are presented as the mean ± SD and are one representative of three independent experiments. Statistical significance in a, b, d, e, g–i was determined by unpaired two-tailed Student’s t test. ^ns P > 0.05.

Fig. 6. JAK1 interacts with IAV PB2.

a Co-ip and Ni-NTA pull-down analysis of the interaction of PB2 with JAK1 in HEK293T cells transfected PB2 and JAK1 plasmids. b Co-ip and Ni-NTA pull-down analysis of the interaction of PB2 with JAK1 in HEK293T cells transfected with JAK1 plasmids and infected with PR8 or CZ virus. c Colocalization of endogenous JAK1 (green) and PB2 (red) in PR8 or CZ-infected A549 cells. Nuclei were stained with DAPI (blue). Scale bars, 10 μm. Intensities of fluorescence at indicated locations were scanned by LAS X Software. d Schematic representation of the deletion mutants of PB2. e, f Ni-NTA pull-down (e) and Co-ip analysis (f) of the interaction of JAK1 with PB2 and its truncation mutants in HEK293T cells. WCL whole-cell lysates, HC Heavy Chain, LC Light Chain. Data are one representative of three independent experiments.

Fig. 7. PB2s of H5 subtype AIVs degrade mammalian JAK1 differently.

a Immunoblots of HEK293T cells transfected with JAK1 and PB2-CZ, PB2M-CZ (M283I-R526K), PB2-JY, or PB2M-JY (I283M-K526R) plasmids (upper). The intensities of the bands on the immunoblots from three independent experiments were quantified and normalized with actin (lower). Data are presented as the mean ± SD. b Immunoblots of HEK293T cells transfected with JAK1 and PB2-CZ, PB2-R526K-CZ, PB2-M283I-CZ, or PB2M-CZ plasmids (upper). The intensities of the bands on the immunoblots from three independent experiments were quantified and normalized with actin (lower). Data are presented as the mean ± SD. c Co-ip analysis of the interaction of PB2 or its mutants with JAK1 in HEK293T cells. d Colocalization of endogenous JAK1 (green) and PB2 (red) in rJY, rCZM (M283I-R526K), or rJYM (I283M-K526R) infected A549 cells. Nuclei were stained with DAPI (blue). Scale bars, 10 μm. Intensities of fluorescence at indicated locations were scanned by LAS X Software. e Ni-NTA pull-down analysis of the ubiquitination of JAK1 in HEK293T cells transfected with PB2 or its mutant plasmids and treated with MG132. WCL, whole-cell lysates. Statistical significance in a, b was determined by unpaired two-tailed Student’s t test. ^ns P > 0.05. Data are one representative of three independent experiments.

Fig. 8. AIV PB2 inhibits IFN-mediated signaling by JAK1 degradation.

a, b qPCR analysis of IFIT1, ISG15, and TAP1 mRNA in A549 cells transfected with PB2 or its mutant plasmids and treated with IFNβ (a) or IFNα (b) (n = 3 biologically independent samples). c, d Luciferase activity in HEK293T cells transfected with ISRE or STAT1 promoter-luciferase reporter, Renilla luciferase plasmid, and PB2 or its mutants plasmids and treated with IFNβ (c) or IFNα (d) (n = 3 biologically independent samples). e, f Immunoblot analysis of phosphorylated and total STAT in HEK293T cells transfected with PB2 or its mutant plasmids and treated with IFNβ (e) or IFNα (f) (upper). Densitometry analysis of the ratio of phospho-STAT/total STAT on the immunoblots (lower). g qPCR analysis of IFIT1, ISG15, and TAP1 mRNA in shJAK1 HEK293T cells transfected with ΔJAK1-WT or ΔJAK1-K859/860R plasmid and treated with IFNβ. Data are presented as the mean ± SD and are one representative of three independent experiments. Statistical significance was determined by unpaired two-tailed Student’s t test. ^ns P > 0.05.

Fig. 9. JAK1 degradation mediated by AIV PB2 affects virus replication in mammalian cells.

a, b Immunoblots of A549 cells infected with rCZ, rCZM, rJY, or rJYM virus (left). The intensities of the bands on the immunoblots from three independent experiments were quantified and normalized with actin (right). c, d Growth curve of viruses in A549 (c) or DF-1 cells (d) infected with rCZ, rCZM, rJY, or rJYM at an MOI = 0.01 (n = 3 biologically independent samples). hpi, h post-infection. e The plaque morphology of MDCK cells infected with rCZ, rCZM, rJY, or rJYM virus. Data are presented as the mean ± SD. Statistical significance was determined by unpaired two-tailed Student’s t test. ^ns P > 0.05.

Fig. 10. JAK1 degradation mediated by AIV PB2 significantly increases the virulence of AIV in mice.

a, b The mice (n = 10 per group) were intranasally inoculated (10⁶ EID₅₀ per mouse) with rJY, rJYM, rCZ, or rCZM virus and monitored for 14 d for body weight loss (a) and survival (b). c Gloss lesions of lungs from the infected mice on 2 and 5 dpi. Severe pneumonia with diffuse consolidation was outlined in yellow. The images are from representative one of five mice. d Viral titers in the lungs (n = 5 per group) were determined on 2 and 5 dpi by plaque assay. e–h Hematoxylin/eosin (HE) staining (e) and scoring (f), immunohistochemistry (IHC) staining (g), and scoring of lung sections (h). Scale bar, 200 μm. The images are from representative one of five mice. The scores were calculated from five mice. i Immunoblots of lung tissue from the infected mice. j Quantification of JAK1 expression on the immunoblots (i) and normalized with actin (n  =  3). k, l qPCR analysis of ISG15 (k) and IFIT1 (l) mRNA in lung from the infected mice on 2 dpi (n = 3). dpi, days post-infection. Data are presented as the mean ± SD. Statistical significance was determined by unpaired two-tailed Student’s t test in a, d, f, h, j–l or log-rank test in b.