Unique immune and other responses of human nasal epithelial cells infected with H5N1 avian influenza virus compared to seasonal human influenza A and B viruses

Abstract

Highly pathogenic avian influenza (HPAI) virus (e.g. H5N1) infects the lower airway to cause severe infections, and constitute a prime candidate for the emergence of disease X. The nasal epithelium is the primary portal of entry for respiratory pathogens, serving as the airway's physical and immune barrier. While HPAI virus predominantly infects the lower airway, not much is known about its interactions with the nasal epithelium. Hence, we sought to elucidate and compare the differential responses of the nasal epithelium against HPAI infection that may contribute to its pathology, and to identify critical response markers. We infected human nasal epithelial cells (hNECs) cultured at the air-liquid interface from multiple healthy donors with clinical isolates of major human seasonal influenza viruses (H1N1, H3N2, influenza B) and HPAI H5N1. The infected cells were subjected to virologic, transcriptomic and secretory protein analyses. While less adapted to infecting the nasal epithelium, HPAI H5N1 elicited unique host responses unlike seasonal influenza. Interestingly, H5N1 infection of hNECs induced responses indicative of subdued antiviral activity (e.g. reduced expression of IFNβ, and inflammasome mediators, IL-1α and IL-1β); decreased wound healing; suppressed re-epithelialization; compromised epithelial barrier integrity; diminished responses to oxidative stress; and increased transmembrane solute and ion carrier gene expression. These unique molecular changes in response to H5N1 infection may represent potential targets for enhancing diagnostic and therapeutic strategies for better surveillance and management of HPAI infection in humans.

Keywords: H5N1; Influenza; highly pathogenic avian influenza; host pathogen interactions; human nasal epithelial cells; molecular responses; upper airway.

Figure 1.

H5N1 Avian Influenza Infects the Human Nasal Epithelium with Lower Viral Replication and Elicits Comparable Host Responses. (A) Viral titre kinetics in hNECs infected with seasonal influenza viruses (at MOI of 0.1) and H5N1 virus (at MOI of 0.1 and 1.0). (B) Immunofluorescence staining with anti-H5 HA (green) in infected hNECs depicting active H5N1 infection of hNECs. (C) Secreted CXCL10 levels at 48 hpi in all influenza strains and control in hNECs infection. Statistical analysis was performed using 1-Way ANOVA with Tukey's correction for multiple comparison. *p < 0.05; **p < 0.01; ***p < 0.001. (D) Number of significantly differentially expressed genes (DEGs) in all seasonal versus avian influenza infections in hNECs. Black arrows indicate peak DEG numbers among all the time points tested (8, 24 and 48hpi).

Figure 2.

Gene Ontology Pathways of Infected in hNECs Showed Differential Upper Airway Responses Between Seasonal and Avian Influenza Infections. The top immune and non-immune pathways from up-regulated DEGs and top pathways from down-regulated DEGs at (A) 8hpi, (B) 24 hpi and (C) 48 hpi. Top up-regulated immune and non-immune pathways were determined by their functions in which immune pathways comprise antiviral immune response functions while non-immune pathways do not have direct function leading to initiation of antiviral immune response. The x-axis shows the number of significant DEGs that are present in the top pathways. Significant GO pathways were determined by hypergeometric analysis. Pathways with an adjusted p-value of <0.05 were considered significant.

Figure 3.

Human Seasonal Influenza Infections Exhibit Highly Similar Host Responses, While H5N1 Influenza Infection of hNECs Induce Unique Host Responses. (A) Venn diagram comparison of DEGs during human seasonal influenza infections of hNECs. The numbers indicate are significant DEGs that were up-regulated (red) or down-regulated (green). Influenza B induces the most DEGs among human seasonal influenza. (B) Comparison of TNFα and FGF2 protein levels in response to human seasonal influenza infection at 48hpi. Statistical analyses were performed using 1-Way ANOVA with Tukey's correction for multiple comparison. *p < 0.05; **p < 0.01; ***p < 0.001. (C) Overlapping DEGs in human influenza across all three strains were compared with DEGs from H5N1 avian influenza at 48hpi. Significant pathways of up or down-regulated DEGs (up-regulated – red; down-regulated – green) in the respective group (common, human only and avian only) were obtained via hypergeometric analysis. Pathways with an adjusted p-value of <0.05 were considered significant. (D) Venn diagrams for comparison of overlapping DEGs in between human seasonal influenza viruses and H5N1 avian influenza at 8 and 24hpi.

Figure 4.

H5N1 Avian Influenza Infection of hNECs Induces Relatively Weaker Responses to Viruses. (A) Heatmaps of selected pathway common between human and avian influenza at 48hpi, mostly being response pathway to influenza and viruses. Gene expression value is presented in Log2FC. (B) qPCR validation of selected genes from significant pathway shared between human seasonal and avian influenza infection at 48hpi. Statistical analyses were conducted using 2-Way ANOVA with Tukey's correction for multiple comparison. (C) LUMINEX validation of apically secreted cytokines in significant pathways shared between human seasonal and avian influenza infection. Statistical analyses were performed using 1-Way ANOVA with Tukey's correction for multiple comparison. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 5.

H5N1 Avian Influenza Infection of hNECs Elicit Subdued Inflammasome Activation Compared to Human Seasonal Influenza. (A) Heatmaps of selected pathway significantly up-regulated in human seasonal influenza at 48hpi, including inflammasome, NFκB and hypoxic responses. Gene expression value is presented in Log2FC. (B) LUMINEX validation of apically secreted cytokines from significant pathways only found in human seasonal influenza infection at 48hpi. Statistical analyses were performed using 1-Way ANOVA with Tukey's correction for multiple comparison. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 6.

H5N1 Avian Influenza Infection of hNECs Differentially Downregulates Epithelial Development and Wound Healing Genes. (A) Heatmaps of selected pathways significantly up-regulated in human influenza but down-regulated in avian influenza at 48hpi, including regulation of epithelial development, wound healing and SLIT/ROBO pathways. Gene expression value is presented in Log2FC. (B) qPCR validation of selected genes from significant pathways differentially up-regulated in human seasonal influenza, while down-regulated in avian influenza infection at 48hpi. Statistical analyses were conducted using 2-Way ANOVA with Tukey's correction for multiple comparison. (C) LUMINEX validation of apically secreted FGF2 identified from significant pathway up-regulated in human seasonal influenza infection at 48hpi. (D) Heatmaps of stress related responses at 48hpi showing differences between human seasonal and avian influenza infections. Gene expression value is presented in Log2FC. Statistical analyses were performed using 1-Way ANOVA with Tukey's correction for multiple comparison *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 7.

Avian (H5N1) Infection Resulted in Cardiac and Neuronal Pathway Enrichment. (A) Heatmaps of selected pathways significantly up-regulated in avian influenza at 48hpi, including cardiac conduction and neuronal processes pathways which are enriched with ion transport genes. Gene expression value is presented in Log2FC. (B) The hNECs and/or hBECs infected with H3N2 and H5N1 were subjected to treatment with peruvoside, a modulator of cellular calcium ions, at concentrations of 20, 60 and 100nM. At 100nM, peruvoside significantly reduced the H3N2 viral titre hNECs, and of H5N1 viral titre infected hBECs. Statistical analyses were conducted using 1-Way ANOVA with Dunnett's correction for multiple comparison. *p < 0.05.