Comparative IP-MS Reveals HSPA5 and HSPA8 Interacting with Hemagglutinin Protein to Promote the Replication of Influenza A Virus

Abstract

The influenza A viruses (IAV) are the principal pathogens for annual (seasonal) influenza, which cause world-wide outbreaks in poultry and pose a persistent threat to public health. The Hemagglutinin protein (HA) of IAV promotes virus infection by binding the host membrane receptor and mediating virus-host membrane fusion. Immunoprecipitation-mass spectrometry (IP-MS) provides global insights into IAV HA-host protein interactions. However, various experimental conditions might affect the identification of interactions. Here, we performed a serial IP-MS to compare interactors of IAV HA in accidental host human, chicken and reservoir host duck cells. We found that the positive ratio of interactors identified by the IP-MS was improved when the transfected HA plasmid had a similar expression level to HA proteins found in IAV virus infection. Comparing interactors in human, chicken and duck cells, we found that HA-interacting host factors might play a role in the susceptibility of accidental hosts (human and chicken) to IAV infection compared to reservoir hosts (duck). We then focused on the function of two heat shock proteins (HSPA5 and HSPA8), which interacted with IAV HA proteins in all three species (human, chicken and duck). We found that both HSPA5 and HSPA8 promoted the IAV replication by enhancing the viral attachment and internalization. These findings extend our knowledge about the mechanisms of IAV entry to host cells and provide target genes to create chickens resistant to avian influenza.

Keywords: HSPA5 and HSPA8; Hemagglutinin (HA); immunoprecipitation–mass spectrometry (IP-MS); influenza A virus (IAV).

Figure 1

Screening strategy for comparing host factors interacting with HA from IAV and plasmids. (A) IP-MS experimental workflow for identifying interactions between HA and host proteins from human HEK293T, chicken CEF and duck DEF cells. HEK293T cells were transfected with H1N1 HA plasmid or infected with H1N1; CEF and DEF cells were transfected with H5N1 HA plasmid or infected with H5N1. (B,D,F) HA expression was detected by RT-qPCR and Western Blotting (C,E,G) in HEK293T, CEF and DEF cells were transfected with plasmid or infected with IAV. The red boxes are the selected samples (plasmid, high and virus) from each species.

Figure 2

Identification of the host factors interacting with HA protein of IAV and plasmids. (A–C) HA protein enrichment was detected by silver staining assay in the selected samples (high, plasmid and virus) from HEK293T, CEF and DEF cells, with IgG antibodies as the control. The red arrow refers to the enriched HA protein. (D) The number of HA interaction factors screened in different categories (high, plasmid and virus) of human, chicken and duck species was counted in the bar chart. (E) Interactions between HA and indicated cellular binders were examined by immunoprecipitation (IP)-Western Blotting. HA and one of the Flag-tagged cellular proteins were co-expressed in HEK293T cells. Cells were lysed 48 h post transfection. Cellular proteins were pulled down with Flag beads and detected for co-eluted HA using an antibody against HA.

Figure 3

Functional analysis of HA-interacting host proteins. (A) Venn diagram showing the overlap of HA-interacting host proteins identified in samples (high, plasmid and virus) of human, chicken and duck species. (B) GO enrichment analysis of genes that were identified to be interacting with HA in samples (high, plasmid and virus) of human, chicken and duck species. The database for Annotation, Visualization and Integrated Discovery (DAVID) was applied for this analysis. (C) GO enrichment analysis of HA-interacting host proteins in species of human, chicken and duck. The database for Annotation, Visualization and Integrated Discovery (DAVID) was applied for this analysis.

Figure 4

HSPA5 or HSPA8 interacts with IAV HA protein. (A) Venn diagram showing the overlap of HA-interacting host proteins identified in human, chicken and duck. (B) Interactions between HSPA5 or HSPA8 and H5N1 HA in DF1 cells were examined by immunoprecipitation at 48 h after co-transfection. (C) Colocalization of H5N1 HA and HSPA5 or HSPA8 in DF1 cells was monitored at 48 h after co-transfection. DAPI (blue) for nuclei visualization. Colocalization (yellow) between HSPA5 or HSPA8 (red) and HA (green) was observed under a Nikon confocal microscope. (D) The Pearson’s correlation and overlap coefficient are shown, deduced from three independent experiments. n = 3.

Figure 5

HSPA5 or HSPA8 positively regulate the IAV replication. (A) The knockdown of HSPA5 or HSPA8 in DF1 cells transfected with specific or scrambled shRNA for 12 h, 24 h, 36 h and 48 h was detected by RT-qPCR. n = 3, *, p < 0.1; **, p < 0.01; ***, p < 0.001. (B) Cell viability was determined using a CCK8 assay, with WT (wild-type) and Mock (Lipofectamine -only) DF1 cells as controls. n = 3, *, p < 0.1; **, p < 0.01. HSPA5 shRNA- (C), HSPA8 shRNA- (D) or scrambled shRNA-transfected DF1 cells and Mock (Lipofectamine only) DF1 cells were infected with H3N8 (MOI = 0.1), H9N2 (MOI = 0.1) or H5N1 (MOI = 0.1) virus. Virus titers at 12 h, 24 h, 36 h and 48 h were determined by means of TCID₅₀ on MDCK cells. n = 3, *, p < 0.1; **, p < 0.01.

Figure 6

HSPA8 or HSPA5 is involved in the early stage of the IAV replication cycle. HSPA5 shRNA-, HSPA8 shRNA- or scrambled shRNA-transfected DF1 cells were infected with CK/0513 virus at an MOI of 1. Localization of NP in DF1 cells was monitored at 4 h (A), 6 h (B) and 8 h (C) by means of Immunofluorescence assay. (D) Quantitative analysis of NP localization in virus-infected cells; for each bar graph 100 cells were taken for analysis.

Figure 7

HSPA8 or HSPA5 promotes IAV attachment and internalization. HSPA5 shRNA-, HSPA8 shRNA- or scrambled shRNA-transfected DF1 cells were infected with CK/0513 virus (MOI = 1) for 2 h at 4 °C or transferred to 37 °C for 30 min after 2 h. (A) Virion attachment and endocytosis were observed by electron microscope. (B) NP copies were detected by RT-qPCR. n = 3, **, p < 0.01; ***, p < 0.001. (C) The colocalization of H5N1 HA and HSPA5 or HSPA8 in DF1 cells was monitored at 48 h after co-transfection without permeabilization. Colocalization (yellow) between HSPA5 or HSPA8 (red) and HA (green) was observed under a Nikon confocal microscope. (D) The Pearson’s correlation and overlap coefficient from three independent experiments were analyzed. n = 3.