Background: Influenza A virus (IAV) infection is a significant risk factor for respiratory diseases, but the host defense mechanisms against IAV remain to be defined. Immune regulators such as surfactant protein A (SP-A) and Toll-interacting protein (Tollip) have been shown to be involved in IAV infection, but whether SP-A and Tollip cooperate in more effective host defense against IAV infection has not been investigated. Methods: Wild-type (WT), Tollip knockout (KO), SP-A KO, and Tollip/SP-A double KO (dKO) mice were infected with IAV for four days. Lung macrophages were isolated for bulk RNA sequencing. Precision-cut lung slices (PCLS) from WT and dKO mice were pre-treated with SP-A and then infected with IAV for 48 h. Results: Viral load was significantly increased in bronchoalveolar lavage (BAL) fluid of dKO mice compared to all other strains of mice. dKO mice had significantly less recruitment of neutrophils into the lung compared to Tollip KO mice. SP-A treatment of PCLS enhanced expression of TNF and reduced viral load in dKO mouse lung tissue. Pathway analysis of bulk RNA sequencing data suggests that macrophages from IAV-infected dKO mice reduced expression of genes involved in neutrophil recruitment, IL-17 signaling, and Toll-like receptor signaling. Conclusions: Our data suggests that both Tollip and SP-A are essential for the lung to exert more effective innate defense against IAV infection.
Keywords: Tollip; Surfactant protein A; Influenza A virus; Neutrophilic inflammation
Supplementary Information The online version contains supplementary material available at https://doi.org/10.1186/s12931-024-02820-3.
Influenza A virus (IAV), specifically H1N1, is a respiratory virus commonly known as the flu that causes seasonal epidemics, but most importantly was responsible for the 2009 pandemic [[
Toll-interacting protein (Tollip) is an innate immune regulator that is expressed in epithelial cells, macrophages, alveolar type II cells, and basal cells [[
SP-A, one of four surfactant proteins produced and secreted primarily from lung alveolar type II cells, has multiple functions in lung immunity. SP-A is a hydrophilic protein including an N-terminal triple-helical collagen region that interacts with immune cells, and a carbohydrate recognition domain (CRD) that acts as a pattern recognition receptor by recognizing proteins on the outside of microbes and allergens [[
While it has been shown that Tollip and SP-A modulate inflammation separately, their cooperation in lung defense against viral infection remains unknown. In the present study, we hypothesized that Tollip and SP-A cooperate to more effectively defend against IAV infection. To test this hypothesis, we performed an acute IAV infection experiment using our newly generated Tollip/SP-A double knockout (dKO) mice, and in precision-cut lung slice (PCLS) model which maintains the three-dimensional structure of the lung and the interactions of multiple cell types (e.g., alveolar epithelial cells, monocytes, and macrophages).
The virus used in this study was Pandemic Influenza A/California/04/2009 (CA04) virus, which was kindly provided by Dr. Kevin Harrod from University of Alabama at Birmingham. The virus was propagated in Madin-Darby canine kidney (CCL-34, MDCK, ATCC, Manassas, VA) cells, as previously published [[
Tollip knockout (KO) mice on a C57/BL6 background were obtained from Dr. Liwu Li [[
To confirm Tollip/SP-A double knockout, lung tissue from each mouse was digested for DNA extraction using the REDExtract-N-Amp reagent kit (XNAT-100RXN, MilliporeSigma, Burlington, MA) following the manufacturer's instructions. The DNA product underwent PCR using custom primers (Table 1) [[
Table 1 Custom primer sequences for Tollip/SP-A genotyping
Primer Sequence Tollip Sufficient Forward Primer 5'-GGATTTGGGATTCATCAGAGGC-3' Tollip Sufficient Reverse Primer 5'-ACAAGAGTGGGACGGAAAACTTC-3' Tollip Deficient Reverse Primer 5'-GGAGAGGCTATTCGGCTATG-3' SP-A Sufficient Forward Primer 5'-ACAGAAGTTTGTGCCGGAAG-3' SP-A Sufficient Reverse Primer 5'-ATGGTCACCCAGAAAACAGG-3' SP-A Deficient Reverse Primer 5'-GCTACTTCCATTTGTCACGTCC-3'
The PCR product was run on a 1% agarose gel, and the band size determined the genotypes. Tollip sufficient mice had a single band around 600 base pairs (bp), Tollip deficient mice had a single band approximately 1,100 bp (Supplementary Fig. 1A), which includes a neomycin cassette. SP-A sufficient mice had a single band at 167 bp, homozygous SP-A KO mice had a single band at 320 bp (Supplementary Fig. 1B), which includes a pGKneoBPA insert.
dKO, Tollip KO, SP-A KO, and WT mice, ages 8–12 weeks (age and gender matched), were inoculated intranasally with 1 × 10
Graph: Fig. 1 Body weight measurements during four days of IAV infection. Tollip KO mice have significantly more weight loss than all other strains. Interestingly, dKO mice lost weight after IAV infection, but there was no statistical significance between dKO mice and WT and SP-A KO mice
Mice were euthanized by intraperitoneal injection of pentobarbital sodium (Fatal-Plus) in sodium chloride. Lungs were lavaged with 1 ml of sterile saline. Cell-free BAL fluid was used for viral load measurement and cytokine measurement. BAL fluid cell differential cytospin slides were stained with a Diff-Quick stain kit (23–122929, 23–122937, 23–122952, Fisher Scientific, Hampton, NH) for cell differential counts. Leukocyte differentials were determined as a percentage of 500 counted leukocytes. The left lung was used for viral load and antiviral mRNA expression, and the right lungs were used for macrophage isolation.
Right lung tissues from IAV infected WT, Tollip KO, SP-A KO, and dKO mice were digested with proteases and DNase as previously published [[
Naïve WT, Tollip KO, SP-A KO, and dKO mouse lungs were inflated with 1.5% low-melting agarose (BMA50002, Fisher Scientific, Hampton, NH) and sliced into consecutive 250 µm thick sections using a Compresstome VF-300 vibratome (Precisionary Instruments, Natick, MA). The slices were transferred into a 24-well plate containing DMEM media with antifungal agents and antibiotics and incubated overnight at 37˚C supplemented with 5% CO
To determine if SP-A inhibits viral entry or viral replication, naïve WT mouse lungs were inflated and sliced as stated above. Slices were pre-treated with 10 µg/ml of recombinant murine SP-A for 30 min, and were then infected with 3 × 10
Intracellular and released IAV was measured by RT-qPCR. Intracellular IAV was measured from homogenized lung tissue, and RNA was isolated using the TRIzol reagent method. Released IAV was measured from BAL fluid of infected mice, where equal volumes of BAL fluid were used to extract RNA using Mini Spin Columns (1940–250, Epoch Life Science Inc., Missouri City, TX) according to the manufacturer's instructions.
Total RNA was extracted from mouse PCLS as we previously described [[
Custom-made primers and probe (Integrated DNA Technologies, Coralville, IA) for the M1 gene of IAV were 5'-GACCRATCCTGTCACCTCTGAC-3' (forward), 5'-AGGGCATTYTGG-ACAAAKC-3' (reverse), and 5'-TGCAGTCCTCGCTCACTGGGCACG-3' (probe) [[
Target gene expression was normalized to the housekeeping gene 18S rRNA (ThermoFisher, Waltham, MA). The comparative threshold cycle method (∆∆Ct) was applied to determine the relative levels of target genes.
Serial dilutions of BAL fluid and half of the left lung homogenate from IAV-infected mice were plated on MDCK cells for one hour at 37˚C. Viral titer was calculated by quantitative plaque assay as described previously [[
LIX, also known as CXCL5, was measured in BAL fluid supernatants using a Mouse LIX DuoSet ELISA kit (DY443, R&D Systems, Minneapolis, MN) following manufacturer's instructions.
RNA extracted from isolated lung macrophages was used for bulk RNA sequencing as described previously [[
To determine the transcriptomic changes of the entire macrophage population, lung macrophages were isolated for bulk RNA sequencing. It has been shown that macrophages present in lung tissue are largely comprised of alveolar macrophages and interstitial lung macrophages [[
Statistical analyses were performed with GraphPad Prism 10 software, and graphical data were presented as median with interquartile range (IQR). Due to data variation in our models, all data were analyzed as nonparametric data using the Mann–Whitney U test for two group comparisons, and Kruskal–Wallis test followed by Dunn's test for multiple comparisons. A p-value < 0.05 was considered statistically significant.
Viral RNA levels were measured in BAL fluid of infected mice (Fig. 2A) four days after IAV infection. While SP-A KO mice had similar levels of viral load in BAL fluid compared to WT mice, dKO mice had significantly more virus released into the BAL fluid compared to all three strains of mice. This suggests that SP-A deficiency alone is not able to worsen viral infection in Tollip sufficient mice. Viral RNA levels measured in homogenized lung tissue of infected mice (Fig. 2B) were also significantly higher in dKO mice compared to WT mice, but there was not a statistically significant difference compared to Tollip KO and SP-A KO mice. Tollip KO mice and SP-A KO mice trended to have more viral RNA in lung tissue compared to WT mice, but this was also not statistically significant. Additionally, plaque assay was performed to quantify levels of infectious viruses. Unlike the viral RNA data, IAV PFU level in the BAL fluid was not different among various groups of mice as PFU was undetectable in some mice (Fig. 2C). Our plaque assay data in BAL fluid could be explained by the fact that BAL fluid may contain antiviral components that inactivate or lessen the infectivity of IAV and other respiratory viruses [[
Graph: Fig. 2 SP-A/Tollip double knockout increased viral load but did not induce a robust inflammatory response. Wild-type (WT), Tollip KO, SP-A KO, and Tollip/SP-A double KO (dKO) mice were intranasally infected with 1 × 102 PFU/mouse of IAV or PBS for four days. After four days of IAV infection dKO mice had significantly more IAV mRNA levels present in (A) BAL fluid and (B) homogenized lung tissue. These data were recapitulated in plaque forming units of IAV found in (C) BAL fluid and (D) homogenized lung tissue. Tollip KO mice had significantly more IFN-beta mRNA expression compared to WT and dKO mice (E). dKO had significantly less neutrophils (F) and (G) LIX compared to Tollip KO mice. Each symbol represents an individual mouse (n = 4–9 mice per group) and the different colored symbols represent an individual/independent experiment where a different batch of IAV was used (n = 2 total individual/independent experiments)
Tollip KO mice infected with IAV were able to significantly increase IFN-beta mRNA expression compared to WT and dKO mice (Fig. 2E). There was no significant difference in IFN-beta levels in WT, SP-A KO, and dKO mice.
After IAV infection, there was a significant increase in neutrophil numbers in the BAL fluid of all strains of mice (Fig. 2F). Tollip KO mice infected with IAV had significantly more neutrophils than WT mice. SPA KO mice had similar number of neutrophils as compared to the WT mice. In contrast to the viral load data in BAL fluid, dKO mice had significantly lower numbers of neutrophils than the Tollip KO mice and failed to increase neutrophils as compared to the WT mice. Neutrophils have been shown to be able to limit viral infection in mice [[
LIX (CXCL5), a neutrophil chemokine, was significantly upregulated in all strains of mice after four days of IAV infection (Fig. 2G). dKO mice had significantly less LIX release compared to WT and Tollip KO mice. These data suggest that dKO mice have impaired ability to initiate a neutrophil recruitment process.
To determine if SP-A restores the antiviral function in lung tissue with both SP-A and Tollip deficiency, we treated PCLS from naïve WT and dKO mice with recombinant SP-A, followed by IAV infection for 48 h. As shown in Fig. 3A, dKO lung tissue had significantly higher tissue viral load compared to WT slices. With the addition of recombinant SP-A, dKO lung tissue showed a significant decrease in viral load. SP-A significantly reduced viral levels in supernatants of PCLS of WT mice, but not dKO mice (Fig. 3B).
Graph: Fig. 3 Recombinant SP-A treatment reduces viral load in Tollip/SP-A deficient mouse precision-cut lung slices (PCLS). PCLS from naïve WT and dKO mice were pre-treated with 10 µg/ml of recombinant mouse SP-A and then infected with 3 × 105 PFU/well of IAV for 48 h. A dKO had significantly more viral load in PCLS tissue, which was significantly decreased by SP-A. B SP-A decreased viral load released into the supernatant of WT and dKO slices. C dKO PCLS had significantly higher IFN-beta mRNA expression levels compared to WT. SP-A increased IFN-beta in WT PCLS. D IAV increased NOS2 levels in dKO slices. IAV-infected dKO slices trended to have higher NOS2 mRNA levels than WT slices. SP-A treatment reduced NOS2 in WT and dKO slices with and without infection. E IAV infection significantly increased TNF mRNA levels in both WT and dKO slices. TNF levels were similar between WT and dKO PCLS after infection, but SP-A treatment trended to increase TNF levels in dKO slices. F LIX release was significantly less in dKO slices without infection. IAV infection significantly increased LIX release in both strains, but IAV-infected dKO slices had significantly less LIX release than WT. SP-A treatment did not increase LIX levels. Each symbol represents an individual PCLS slice from n = 1 mouse and the different colored symbols in the IAV group represent an individual/independent experiment (n = 2 total individual/independent experiments)
To further understand the role of SP-A during IAV infection in PCLS, we measured anti-viral genes Interferon-beta (IFN-beta), and macrophage activation-associated cytokine TNF-α. IFN-beta RNA expression was significantly higher in dKO slices compared to WT slices (Fig. 3C). The addition of recombinant SP-A significantly increased IAV-induced IFN-beta levels in WT slices, and decreased IFN-beta levels in dKO slices to levels similar to WT treated with SP-A. As nitric oxide (NO) may be involved in IAV infection [[
The role of SP-A during pathogen infections has been attributed to its direct binding to the virus particle to aid in phagocytosis [[
Graph: Fig. 4 Recombinant SP-A treatment inhibits IAV replication in WT mouse precision-cut lung slices (PCLS). PCLS from naïve WT mice were pre-treated with 10 µg/ml of recombinant mouse SP-A and then infected with 3 × 105 PFU/well of IAV for 0, 24, and 48 h. There was a significant increase in IAV load 24 h post infection compared to 0 h of infection. SP-A treatment did not change viral load at 0 h, however, SP-A significantly decreased IAV load 24 and 48 h post infection. Each symbol represents an individual PCLS slice from n = 1 mouse done in one individual/independent experiment
To elucidate the potential mechanism whereby Tollip/SP-A deficiency impaired the antiviral function, we performed bulk RNA-sequencing in lung macrophages isolated from WT, Tollip KO, SP-A KO, and dKO mice with or without IAV infection.
In WT macrophages, IAV infection as compared to PBS control, up-regulated 1,824 genes while it down-regulated 1,789 genes (Fig. 5A). As expected, IAV infection, as compared to the PBS control in WT macrophages increased genes associated with pathways such as IL-17 signaling, TNF signaling, RIG-I-like receptor signaling, Toll-like receptor signaling, and NF-kappa B signaling (Supplementary Tables 1 and 2), and down-regulated pathways such as focal adhesion and tight junctions.
Graph: Fig. 5 Volcano plots of differentially regulated genes in isolated lung macrophages from IAV-infected mice. A Genes up- and down-regulated in IAV-infected WT macrophages compared to PBS treated (wild-type, WT) macrophages. B Genes up- and down- regulated in IAV-infected Tollip KO (vs. WT) macrophages. C SP-A KO and D dKO macrophages compared with WT macrophages infected with IAV. E Genes up- and down- regulated in IAV-infected dKO macrophages compared to IAV infected Tollip KO macrophages and F IAV infected SP-A KO macrophages
IAV infected Tollip KO macrophages, as compared to IAV infected WT macrophages, up-regulated 535 genes and down-regulated 646 genes (Fig. 5B). Pathways associated with the up-regulated genes were natural killer cell mediated cytotoxicity and cell adhesion molecules. The down-regulated genes were associated with pathways such as HIF-1 signaling and MAPK signaling (Tables 2 and 3).
Table 2 Pathways of genes altered by Tollip deficiency (vs. Tollip/SP-A sufficiency) in mouse lung macrophages infected with IAV
Signaling Pathways Gene Count % Up- or Down- Regulated Natural killer cell mediated cytotoxicity 10 6.5 Up 1.90E-04 Cell adhesion molecules (CAMs) 8 5.2 Up 1.76E-02 HIF-1 signaling pathway 11 6.1 Down 9.59E-05 MAPK signaling pathway 13 7.2 Down 4.76E-02
Table 3 List of genes in selected pathways altered by Tollip deficiency (vs. Tollip/SP-A sufficiency) in mouse lung macrophages infected with IAV
Genes Log2 fold change Genes Log2 fold change Natural killer cell mediated cytotoxicity Cell adhesion molecules (CAMs) Klrc1 1.18 H2-M2 3.69 Gzmb 1.17 Ctla4 1.25 Fasl 1.35 Tigit 0.93 Klrc2 1.21 Icos 1.24 Prf1 0.77 Cd28 1.03 Klra4 0.78 Pdcd1 0.89 Ncr1 1.18 Lrrc4b 6.34 Fcgr4 0.87 Itga8 1.56 Klrb1c 0.68 Ulbp1 0.66 Genes Log2 fold change Genes Log2 fold change HIF-1 signaling pathway MAPK signaling pathway Egln1 -1.51 Met -1.16 Ldha -0.78 Map2k6 -2.15 Pgk1 -1.27 Dusp9 -1.78 Vegfa -0.83 Arrb1 -1.48 Eno1b -1.16 Vegfa -0.83 Mtor -0.89 Map3k13 -2.27 Eno1 -1.00 Hspb1 -0.97 Pfkl -1.03 Ereg -0.75 Hmox1 -0.61 Cacnb3 -0.64 Serpine1 -0.74 Ngf -0.73 Aldoa -0.73 Hspa1a -0.48 Hspa1b -0.46 Map4k4 -0.45
SP-A KO macrophages infected with IAV up-regulated 552 genes and down-regulated 257 genes compared to IAV infected WT macrophages (Fig. 5C). Pathway analysis did not show any significant pathways associated with antiviral defense, but instead showed significant association with pathways such as focal adhesion, cell adhesion molecules, and leukocyte transendothelial migration. Down-regulated genes were associated with the cytokine-cytokine receptor interaction pathway (Tables 4 and 5).
Table 4 Pathways of genes altered by SP-A deficiency (vs. Tollip/SP-A sufficiency) in mouse lung macrophages infected with IAV
Signaling Pathways Gene Count % Up- or Down- Regulated ECM-receptor interaction 11 5.7 Up 2.16E-05 Focal adhesion 17 8.8 Up 5.94E-05 Cell adhesion molecules (CAMs) 11 5.7 Up 3.48E-03 Leukocyte transendothelial migration 7 3.6 Up 3.52E-02 Cytokine-cytokine receptor interaction 8 4.0 Down 2.39E-02
Table 5 List of genes in selected pathways altered by SP-A deficiency (vs. Tollip/SP-A sufficiency) in mouse lung macrophages infected with IAV
Genes Log2 fold change Genes Log2 fold change Focal adhesion Cell adhesion molecules (CAMs) Itga8 2.79 Icam2 1.50 Cav1 1.20 Itga8 2.79 Mylk 1.48 Esam 1.84 Lama5 1.45 Cldn5 1.95 Col4a2 1.40 Pecam1 1.07 Col4a1 1.29 Cdh5 1.71 Itga1 1.13 Ptprm 1.85 Prkcg 1.64 Ctla4 0.93 Col6a2 1.28 Vcam1 0.72 Pdgfra 0.91 Lrrc4b 5.83 Vwf 0.96 Cadm1 1.18 Pdgfd 1.73 Lamc3 1.69 Leukocyte transendothelial migration Col6a1 1.22 Esam 1.84 Vegfd 1.46 Cldn5 1.95 Col6a3 1.01 Pecam1 1.07 Mylk3 5.65 Rapgef3 1.51 Cdh5 1.71 Prkcg 1.64 Vcam1 0.72 Cytokine-cytokine receptor interaction Genes Log2 fold change Genes Log2 fold change Ccr9 -2.29 Tnfrsf11b -1.77 Cxcl5 -2.10 Il12rb2 -1.08 Tnfrsf9 -1.81 Il1a -0.83 Il2rb -0.94 Inhba -0.87
After IAV infection, dKO macrophages up-regulated 241 genes and down-regulated 447 genes (Fig. 5D) compared to IAV infected WT macrophages. Cytokine-cytokine receptor interactions was one of the few significant pathways associated with the up-regulated genes. Importantly, the chemokine signaling pathway was significantly associated with the down-regulated genes (Tables 6 and 7).
Table 6 Pathways of genes altered by Tollip/SP-A deficiency (vs. Tollip/SP-A sufficiency) in mouse lung macrophages infected with IAV
Cytokine-cytokine receptor interaction 6 0.3 Up 1.93E-02 Proteoglycans in cancer 5 0.3 Up 2.53E-02 African trypanosomiasis 2 0.1 Up 3.14E-02 Tyrosine metabolism 2 0.1 Up 3.32E-02 Graft-versus-host disease 3 0.1 Up 8.03E-03 Autoimmune thyroid disease 3 0.1 Up 1.22E-02 Chemokine signaling pathway 10 6.6 Down 8.64E-03 Cell adhesion molecules (CAMs) 8 5.3 Down 1.93E-02 ABC transporters 4 2.6 Down 1.75E-02 Citrate cycle (TCA cycle) 3 2.0 Down 3.50E-02 Biosynthesis of amino acids 11 7.2 Down 8.17E-07 Glycolysis / Gluconeogenesis 10 6.6 Down 1.38E-06 Carbon metabolism 10 6.6 Down 2.87E-04 Pyruvate metabolism 5 3.3 Down 1.38E-03 HIF-1 signaling pathway 8 5.3 Down 2.40E-03
Table 7 List of genes in selected pathways altered by Tollip/SP-A deficiency (vs. Tollip/SP-A sufficiency) in mouse lung macrophages infected with IAV
Cytokine-cytokine receptor interaction Genes Log2 fold change Genes Log2 fold change Il18rap 0.99 Fasl 1.27 Ccl21a 1.91 Il5ra 1.96 Gm10591 1.47 Il18 1.58 Chemokine signaling pathway Genes Log2 fold change Genes Log2 fold change Cxcl15 -1.88 Grk1 -6.56 Ccl24 -2.27 Ccl9 -1.07 Cxcl3 -1.52 Cxcl14 -1.44 Cxcl5 -1.95 Arrb1 -1.56 Ccl22 -1.31 Ccr9 -1.10
dKO (vs. Tollip KO) macrophages had 644 up-regulated genes, and 689 down-regulated genes (Fig. 5E). Influenza A and asthma pathways were significantly up-regulated in dKO macrophages as compared to Tollip KO macrophages (Tables 8 and 9). It is important to note, that of the down-regulated genes, those associated with the chemokine signaling pathway (CXCL5, Ccl9, Ccl12) and the IL-17 signaling pathway (IL-17) were significantly reduced in dKO macrophages compared to Tollip KO macrophages.
Table 8 Pathways of genes altered by Tollip/SP-A deficiency (vs. Tollip deficiency) in mouse lung macrophages infected with IAV
Signaling Pathways Gene Count % Up- or Down- Regulated Notch signaling pathway 7 3.0 Up 4.31E-04 Longevity regulating pathway – multiple species 7 3.0 Up 1.80E-03 FoxO signaling pathway 11 4.7 Up 6.32E-03 Asthma 4 1.7 Up 7.45E-03 mTOR signaling pathway 12 5.1 Up 7.47E-03 TGF-beta signaling pathway 8 3.4 Up 7.49E-03 ECM-receptor interaction 8 3.4 Up 7.70E-03 Influenza A 12 5.1 Up 9.59E-03 Antigen processing and presentation 7 3.0 Up 1.56E-02 Wnt signaling pathway 10 4.2 Up 2.81E-02 Cell cycle 9 3.8 Up 3.11E-02 Oxidative phosphorylation 9 3.8 Up 4.26E-02 Cell adhesion molecules (CAMs) 21 7.7 Down 3.96E-07 Cytokine-cytokine receptor interaction 26 9.6 Down 1.69E-05 Hematopoietic cell lineage 12 4.4 Down 2.40E-04 Complement and coagulation cascades 10 3.7 Down 8.34E-04 IL-17 signaling pathway 11 4.1 Down 1.02E-03 T cell receptor signaling pathway 10 3.7 Down 8.07E-03 Chemokine signaling pathway 15 5.5 Down 8.11E-03 Primary immunodeficiency 5 1.8 Down 1.27E-02 ECM-receptor interaction 8 3.0 Down 2.06E-02 Adipocytokine signaling pathway 7 2.6 Down 2.44E-02 Th1 and Th2 cell differentiation 8 3.0 Down 2.67E-02 Nitrogen metabolism 3 1.1 Down 2.70E-02
Table 9 List of genes in selected pathways altered by Tollip/SP-A deficiency (vs. Tollip deficiency) in mouse lung macrophages infected with IAV
Genes Log2 fold change Genes Log2 fold change Influenza A Asthma Hspa1a 0.61 Fcer1g 0.76 Hspa1b 0.62 Il10 0.61 Hspa2 0.60 H2-Eb1 0.37 H2-Eb1 0.37 H2-DMb2 0.49 Ifna1 5.46 Ivns1abp 0.29 H2-DMb2 0.49 Ep300 0.37 Nup98 0.29 Eif2ak4 0.45 Nxf1 0.30 Crebbp 0.32 Genes Log2 fold change Genes Log2 fold change Cytokine-cytokine receptor interaction IL-17 signaling pathway Ccl9 -0.98 Lcn2 -0.93 Ccl12 -0.94 Ccl12 -0.94 Il1Rl1 -1.00 Csf2 -0.87 Ccr8 -1.28 Il17f -3.31 Csf2 -0.87 Mapk10 -4.39 Il17f -3.31 Ikbke -0.50 Cxcl15 -2.00 Mapk8 -0.54 Gdf3 -1.59 Cxcl5 -0.61 Ccl1 -1.40 Ccl2 -0.34 Inhba -0.46 Mmp13 -0.78 Inhbb -1.13 Il5 -1.03 Il1a -0.33 Il34 -2.59 Chemokine signaling pathway Cxcl9 -0.46 Ccl9 -0.98 Cx3cl1 -0.59 Ccl12 -0.94 Cxcr6 -0.61 Ccr8 -1.28 Cxcl5 -0.61 Cxcl15 -2.00 Ccr4 -0.67 Ccl1 -1.40 Il22ra1 -5.20 Cxcl9 -0.46 Ccl2 -0.34 Cx3cl1 -0.59 Il5 -1.03 Cxcr6 -0.61 Ccr5 -0.32 Cxcl5 -0.61 Ccr1 -0.34 Ccr4 -0.67 Bmp6 -0.61 Ccl2 -0.34 Ccl5 -0.32 Ccr5 -0.32 Acvrl1 -0.27 Ccr1 -0.34 Ccl5 -0.32 Gnb4 -0.51
dKO (vs. SP-A KO) macrophages had 698 up-regulated genes, and 1050 down-regulated genes (Fig. 5F). Influenza A and cytokine-cytokine receptor interaction pathways were significantly higher in dKO macrophages as compared to SP-A KO macrophages (Supplementary Tables 3 and 4). Of the down-regulated genes, those associated with focal adhesion and cell adhesion molecules were significantly reduced in dKO macrophages compared to SP-A KO macrophages.
These data suggest SP-A deficiency in the absence of Tollip further inhibits the appropriate antiviral inflammatory response involved in neutrophil chemokine signaling.
For the first time we have shown that the deficiency of both Tollip and SP-A during an IAV infection contributed to increased viral load associated with lack of robust lung immune response such as neutrophil recruitment. We have also shown that SP-A may have therapeutic implication in restoring antiviral function in lung tissues with both Tollip and SP-A deficiency.
Our group and others have shown that Tollip regulates virus inflammation by directly interacting with other proteins such as STING, or by indirectly enhancing anti-inflammatory protein expression [[
One of the important questions unsolved in this study is why SP-A deficiency alone did not impair innate immunity. Our findings showed SP-A KO mice had similar lung tissue IAV RNA levels to the dKO mice. However, the plaque assay data showed that SP-A KO mice had significantly less IAV PFU/ml than dKO mice. This suggests that the infectivity of IAV from SP-A KO mice is less than that from dKO mice. How SP-A affects viral infection has not been well understood. Previous work regarding the role of SP-A during viral infection has focused primarily on its ability to bind to virus resulting in either increased phagocytosis or neutralization of the virus particle. One study found that incubation of macrophages with SP-A before, during, or after IAV infection suppressed the infection [[
Having shown the dependency of SP-A function on Tollip during IAV infection, we explored if SP-A may have the therapeutic implication in restoring antiviral function in lungs with both SP-A and Tollip deficiency. By leveraging the PCLS model where interactions of various cell types are maintained, we tested if exogenous SP-A treatment could inhibit IAV infection. One of the interesting findings in our study is that while the PCLS tissue viral data (Fig. 3A) in dKO vs. WT mice was similar to the in vivo data, the dKO PCLS supernatant data showed a slight decrease in viral release compared to WT slices, whereas in vivo, dKO mice had more viral load released into their BAL fluid compared to WT mice. One explanation for such differences could be the lack of mucociliary mechanisms in PCLS. In vivo, mucociliary mechanism exists, which may be compromised in dKO mice. In PCLS, dKO PCLS increased IFN-β likely due to increased lung tissue viral load. The increased IFN-β may in turn reduce the release of intracellular viruses into the extracellular space. Additionally, recombinant SP-A decreased viral load in PCLS of dKO mice. Our data was in line with a study conducted by Al-Qahtani et al. who showed that full-length SP-A added to A594 cells infected with IAV decreased viral load as well as pro-inflammatory cytokines [[
There are several limitations to this study. First, our IAV model does not include allergen challenges to determine the role of Tollip/SP-A deficiency in asthma exacerbations. Second, this study only focused on a very acute IAV infection. It remains unclear if Tollip/SP-A deficiency prolongs IAV infection or delays clearance at a later time, therefore additional time points should be considered. Third, our data shows that the reduction in viral load seen in SP-A treated PCLS is likely due to SP-A-mediated inhibition of viral replication, however, the underlying mechanism needs to be further investigated. Fourth, whether the decrease in genes associated with chemokine signaling seen in dKO macrophages leads to impaired neutrophil recruitment and viral clearance could be explored. Fifth, the exact mechanism of how Tollip/SP-A double deficiency inhibited chemokine transcription and the ensuing neutrophil recruitment has yet to be elucidated. Lastly, how these two immune modulators work together, whether through direct interaction or indirectly, to reduce inflammation needs to be further determined.
By utilizing our novel dKO mouse strain for both in vivo IAV infections as well as PCLS cultures, we have demonstrated that Tollip/SP-A deficiency enhances IAV infection via an impaired innate immune response (i.e., neutrophilic inflammation, Fig. 6), and that SP-A's role in viral infection is dependent on the presence of Tollip. Supplementation of SP-A or enhancement of Tollip expression or function in hosts with SP-A deficiency may provide a new approach to attenuate the severity of IAV infection.
Graph: Fig. 6 Proposed mechanisms underlying Tollip and SP-A cooperation to modulate IAV infection and pro-inflammatory response. In Tollip- and SP-A-sufficient lung (e.g., epithelial cells and macrophages), IAV infection increases interferon (e.g., type I interferon) expression and chemokines via viral RNA signaling to effectively clear the viruses. In the absence of both Tollip and SP-A, the lung fails to clear the viruses partly due to impaired generation of antiviral mediators (e.g., type I interferons and chemokines) and lack appropriate levels of neutrophil recruitment. RIG-I = retinoic acid-inducible gene I. NLR = NOD like receptor
Not applicable.
Conceptualization, N.S.; and H.W.C.; methodology, N.S.; D.C.; M.N.; and H.W.C.; software, N.S.; and H.W.C.; validation, N.S.; D.C.; and H.W.C.; formal analysis, N.S.; and H.W.C.; investigation, N.S.; and H.W.C.; resources, M.N.; J.G.L.; M.K.; and H.W.C.; data curation, N.S.; D.C.; and T.N.; writing—original draft preparation, N.S.; writing—review and editing, D.C.; T.N.; M.N.; J.G.L.; M.K.; and H.W.C.; visualization, J.G.L.; M.K.; and H.W.C.; supervision, H.W.C.; project administration, H.W.C.; funding acquisition, M.K. and H.W.C.; All authors have read and agreed to the published version of the manuscript.
This work was supported by the NIH grant: U19AI125357. This funding source was not involved in the preparation of data or the manuscript.
No datasets were generated or analysed during the current study.
The animal study protocol was approved by the Institutional animal care and use committee (IACUC) of National Jewish Health (AS2792-03-26).
Not applicable.
Drs. Ledford and Kraft are co-founders of RaeSedo, LLC, a company to develop novel peptidomimetic based therapeutics derived from an active area of SP-A. No peptidomimetic studies are reported in this manuscript. All other authors declare no conflict of interest.
Graph: Additional file 1. WT IAV vs WT PBS – downregulated genes. A table of genes downregulated by IAV in WT mouse macrophages, with the log2 fold change, and p -values.
Graph: Additional file 2. WT IAV vs WT PBS – upregulated genes. A table of genes upregulated by IAV in WT mouse macrophages, with the log2 fold change, and p -values.
Graph: Additional file 3. SP-A KO IAV vs WT IAV – downregulated genes. A table of genes that are downregulated in IAV infected SP-A KO mouse macrophages compared to IAV infected WT macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 4. SP-A KO IAV vs WT IAV – upregulated genes. A table of genes that are upregulated in IAV infected SP-A KO macrophages compared to IAV infected WT macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 5. Tollip KO IAV vs WT IAV – downregulated genes. A table of genes that are downregulated in IAV infected Tollip KO macrophages compared to IAV infected WT macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 6. Tollip KO IAV vs WT IAV – upregulated genes. A table of genes that are upregulated in IAV infected Tollip KO macrophages compared to IAV infected WT macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 7. dKO IAV vs WT IAV – downregulated genes. A table of genes that are downregulated in IAV infected dKO macrophages compared to IAV infected WT macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 8. dKO IAV vs WT IAV – upregulated genes. A table of genes that are upregulated in IAV infected dKO macrophages compared to IAV infected WT macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 9. dKO IAV vs Tollip KO IAV – downregulated genes. A table of genes that are downregulated in IAV infected dKO macrophages compared to IAV infected Tollip KO macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 10. dKO IAV vs Tollip KO IAV – upregulated genes. A table of genes that are upregulated in IAV infected dKO macrophages compared to IAV infected Tollip KO macrophages. This includes the log2 fold change and p -values.
Graph: Additional file 11: Supplementary Figure 1. Tollip/SP-A genotyping confirmation. (A) Tollip sufficient mice have a single band around 600 base pairs, and Tollip deficient mice have a single band around 1,100 bp. (B) SP-A sufficient mice have a single band around 167 bp, ad SP-A deficient mice have a single band around 320 bp.
Graph: Additional file 12: Supplementary Table 1. Pathways of genes altered by IAV infection (vs. PBS control) in wild-type (Tollip/SP-A sufficient) mouse lung macrophages. A table of signaling pathways, the gene count, the percent up- or down- regulated, and the p -value of important pathways found to be associated in mouse lung macrophages.
Graph: Additional file 13: Supplementary Table 2. List of genes in selected pathways altered by IAV infection (vs. PBS control) in wild-type (Tollip/SP-A sufficient) mouse lung macrophages. A table of genes and the log2 fold change associated with the pathways listed in Supplementary table 1.
Graph: Additional file 14: Supplementary Table 3. Pathways of genes altered by Tollip/SP-A deficiency (vs. SP-A deficiency) in mouse lung macrophages infected with IAV. A table of signaling pathways, the gene count, the percent up- or down- regulated, and the p -value of important pathways found to be associated in mouse lung macrophages.
Graph: Additional file 15: Supplementary Table 4. List of genes in selected pathways altered by Tollip/SP-A deficiency (vs. SP-A deficiency) in mouse lung macrophages infected with IAV. A table of genes and the log2 fold change associated with the pathways listed in Supplementary table 3.
• IAV
- Influenza A Virus
• Tollip
- Toll-interacting protein
• SP-A
- Surfactant Protein A
• CRD
- Carbohydrate recognition domain
• KO
- Knockout
• WT
- Wild-type
• dKO
- Double knockout
• PCLS
- Precision cut lung slices
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
By Niccolette Schaunaman; Diana Cervantes; Taylor Nichols; Mari Numata; Julie G. Ledford; Monica Kraft and Hong Wei Chu
Reported by Author; Author; Author; Author; Author; Author; Author