Pigs lacking TMPRSS2 displayed fewer lung lesions and reduced inflammatory response when infected with influenza A virus

Abstract

Influenza A virus (IAV) infection is initiated by hemagglutinin (HA), a glycoprotein exposed on the virion's lipid envelope that undergoes cleavage by host cell proteases to ensure membrane fusion, entry into the host cells, and completion of the viral cycle. Transmembrane protease serine S1 member 2 (TMPRSS2) is a host transmembrane protease expressed throughout the porcine airway epithelium and is purported to play a major role in the HA cleavage process, thereby influencing viral pathogenicity and tissue tropism. Pigs are natural hosts of IAV and IAV disease causes substantial economic impact on the pork industry worldwide. Previous studies in mice demonstrated that knocking out expression of TMPRSS2 gene was safe and inhibited the spread of IAV after experimental challenge. Therefore, we hypothesized that knockout of TMPRSS2 will prevent IAV infectivity in the swine model. We investigated this hypothesis by comparing pathogenesis of an H1N1pdm09 virus challenge in wildtype (WT) control and in TMPRSS2 knockout (TMPRSS2 ^-/-) pigs. We demonstrated that TMPRSS2 was expressed in the respiratory tract in WT pigs with and without IAV infection. No differences in nasal viral shedding and lung lavage viral titers were observed between WT and TMPRSS2 ^-/- pigs. However, the TMPRSS2 ^-/- pig group had significantly less lung lesions and significant reductions in antiviral and proinflammatory cytokines in the lung. The virus titer results in our direct challenge model contradict prior studies in the murine animal model, but the reduced lung lesions and cytokine profile suggest a possible role for TMPRSS2 in the proinflammatory antiviral response. Further research is warranted to investigate the role of TMPRSS2 in swine IAV infection and disease.

Keywords: RNAscope; TMPRSS2 gene; influenza A; knockout; proinflammatory response; somatic cell nuclear transfer; swine.

FIGURE 1

RNAscope detection of TMPRSS2 transcripts in the respiratory tract of clinically healthy wildtype pigs without IAV infection. There was abundant detection in pseudostratified ciliated epithelium of the turbinate (A) and trachea (B) and circumferential detection within the respiratory epithelium lining the mainstem bronchi (C) and bronchioles (D). Transcripts were detected in cells comprising the alveolar septa (E) but rarely detected in endothelial cells (F). Original magnifications: ×600 (A,B,F); 40× (C); 400× (D,E).

FIGURE 2

RNAscope detection of TMPRSS2 transcripts in the respiratory tract of IAV infected wildtype pigs at 5 days post inoculation (dpi). Diminished transcripts were observed in areas of erosion in the turbinate (A) and trachea (B). Differential visual observation of TMPRSS2 transcripts was noted in bronchi (C,D) and bronchioles (E,F) affected by necrotizing bronchiolitis and bronchitis after IAV infection compared to non-challenged pigs. Increased transcription was observed in cells surrounding affected bronchioles (E,F) when compared to healthy tissue. Original magnifications: ×200 (D–F); 100× (C); 400× (A,B).

FIGURE 3

Nasal shedding of A/swine/Iowa/A02524480/2020 (IA/20) H1N1 by virus titration on cell culture in wildtype (WT) and TMPRSS2 ^−/− pigs. Virus titers showed as log¹⁰ transformed group mean TCID50/ml ± standard error of the mean from nasal swabs of 1–5 days post infection in negative controls (grey, N = 5), wildtype (green, N = 10), or TMPRSS2 ^−/− pigs (black, N = 7). Point of detection was 0.5. Numbers above the error bars show the number of positive pigs/total pig numbers in group and different lower-case letters (a, b, c) indicate statistical difference between group means (p ≤ 0.05) by ordinary one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons (GraphPad Prism, GraphPad Software, La Jolla, CA).

FIGURE 4

Infection of IA/20 H1N1 in WT and TMPRSS2 ^−/− pigs. (A) Bronchioalveolar lavage fluid (BALF) TCID₅₀/ml viral titers log¹⁰ transformed. (B) Weighted percentage of macroscopic lung lesion. (C) Group average microscopic lung scores. (D) Group average microscopic trachea scores. Negative controls (grey, N = 5), wildtype (green, N = 10) and in TMPRSS2 ^−/− pigs (black, N = 7). Different lower-case letters (a, b, c) indicate statistical difference between group means (p ≤ 0.05) by ANOVA with Tukey’s multiple comparisons (GraphPad Prism, GraphPad Software, La Jolla, CA).

FIGURE 5

Immunohistochemical (IHC) staining of Influenza A virus nucleoprotein in wildtype (green, N = 10) and TMPRSS2 ^−/− (black, N = 7) pig respiratory tract tissue. (A) IHC group average score with standard error of the mean from tracheal epithelium. (B) Group average with standard error of the mean from conducting airway epithelium of the lung. (C) Group average with standard error of the mean from non-airway cells including pneumocytes, macrophages and exudate of the lung. Different lower-case letters (a, b) indicate statistical difference between group means (p ≤ 0.05) by t-test (GraphPad Prism, GraphPad Software, La Jolla, CA).

FIGURE 6

Cytokine and chemokine panel performed on bronchioalveolar lavage fluid (BALF) collected from negative controls (grey, N = 5), wildtype (green, N = 10) and TMPRSS2 ^−/− (black, N = 7) pigs 5 days post infection with IA/20 H1N1 virus. Different lower-case letters (a, b, c) indicate statistical difference between group means (p ≤ 0.05) by ANOVA with Tukey’s multiple comparisons (GraphPad Prism, GraphPad Software, La Jolla, CA).