(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Fibroblast growth factor-9 expression in airway epithelial cells amplifies the type I interferon response and alters influenza A virus pathogenesis [1]

['Bradley E. Hiller', 'Department Of Medicine', 'Washington University School Of Medicine', 'St. Louis', 'Missouri', 'United States Of America', 'Yongjun Yin', 'Department Of Developmental Biology', 'Unites States Of America', 'Yi-Chieh Perng']

Date: 2022-08 

Influenza A virus (IAV) preferentially infects conducting airway and alveolar epithelial cells in the lung. The outcome of these infections is impacted by the host response, including the production of various cytokines, chemokines, and growth factors. Fibroblast growth factor-9 (FGF9) is required for lung development, can display antiviral activity in vitro, and is upregulated in asymptomatic patients during early IAV infection. We therefore hypothesized that FGF9 would protect the lungs from respiratory virus infection and evaluated IAV pathogenesis in mice that overexpress FGF9 in club cells in the conducting airway epithelium (FGF9-OE mice). However, we found that FGF9-OE mice were highly susceptible to IAV and Sendai virus infection compared to control mice. FGF9-OE mice displayed elevated and persistent viral loads, increased expression of cytokines and chemokines, and increased numbers of infiltrating immune cells as early as 1 day post-infection (dpi). Gene expression analysis showed an elevated type I interferon (IFN) signature in the conducting airway epithelium and analysis of IAV tropism uncovered a dramatic shift in infection from the conducting airway epithelium to the alveolar epithelium in FGF9-OE lungs. These results demonstrate that FGF9 signaling primes the conducting airway epithelium to rapidly induce a localized IFN and proinflammatory cytokine response during viral infection. Although this response protects the airway epithelial cells from IAV infection, it allows for early and enhanced infection of the alveolar epithelium, ultimately leading to increased morbidity and mortality. Our study illuminates a novel role for FGF9 in regulating respiratory virus infection and pathogenesis.

Influenza viruses are respiratory viruses that cause significant morbidity and mortality worldwide. In the lungs, influenza A virus primarily infects epithelial cells that line the conducting airways and alveoli. Fibroblast growth factor-9 (FGF9) is a growth factor that has been shown to have antiviral activity and is upregulated during early IAV infection in asymptomatic patients, leading us to hypothesize that FGF9 would protect the lung epithelium from IAV infection. However, mice that express and secrete FGF9 from club cells in the conducting airway had more severe respiratory virus infection and a hyperactive inflammatory immune response as early as 1 day post-infection. Analysis of the FGF9-expressing airway epithelial cells found an elevated antiviral and inflammatory interferon signature, which protected these cells from severe IAV infection. However, heightened infection of alveolar cells resulted in excessive inflammation in the alveoli, resulting in more severe disease and death. Our study identifies a novel antiviral and inflammatory role for FGFs in the lung airway epithelium and confirms that early and robust IAV infection of alveolar cells results in more severe disease.

Funding: This study was supported by grants from the National Institutes of Health ( https://www.nih.gov/grants-funding ): R01 AI080172 to D.J.L., 5T32 CA009547 to Y.-C.P. and L.E.F., 5T32 AI007163 and F31 AI 14999 to M.C.L., R01 AI134862 and R01 AI137062 to I.A.C. and C.B.L., and R01 HL111190 and R01 HL154747 to D.M.O. We also acknowledge support from the Washington University Rheumatic Diseases Research Resource-based Center ( https://sites.wustl.edu/rdrrc/ ) P30 AR073752 for genome engineering and sequencing. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2022 Hiller et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

To test our hypothesis, Fgf9 was placed under doxycycline-inducible control directed by the Scgb1a1 (Cc10, Ccsp) promoter, allowing us to temporally regulate FGF9 expression and secretion from club cells [ 23 , 24 ]. Since club cells comprise the majority of the conducting airway epithelium in the mouse lung, we expected that expressing FGF9 in club cells would protect the entire conducting airway epithelium from IAV infection. However, FGF9-overexpressing mice displayed excessive lung inflammation and alveolar edema, increased cytokine and chemokine expression, accelerated infection in the alveolar space, and increased mortality. Characterization of the conducting airway epithelium revealed that FGF9 overexpression primed the epithelium for an amplified type I IFN response that protected it from IAV infection. These findings provide important insights into how FGF signaling in the lung impacts the pathogenesis of respiratory virus disease.

FGFs in the lung have diverse roles during respiratory virus infection. During IAV infection in mice, FGF7 increases the severity of IAV infection by accelerating AT2 cell infection and proliferation [ 18 ]. During IAV infection of both humans and mice, FGF2 expression is elevated in the lungs [ 19 ]. FGF2 administration in mice reduced the severity of IAV disease while knockout of Fgf2 increased IAV lung injury by impairing neutrophil recruitment and activation [ 19 ]. FGF10, however, protects the lung from severe IAV infection by driving epithelial repair [ 20 ]. During IAV infection of humans, increased FGF9 transcripts specifically in the serum of asymptomatic patients but not symptomatic patients has been observed at 1 dpi, suggesting that FGF9 may function in the early response to IAV infection [ 21 ]. In addition, a screen of 756 human secreted proteins identified members of the FGF9 subfamily (FGFs 9, 16, and 20) as inhibitors of vesicular stomatitis virus (VSV) replication [ 22 ]. Given these findings, we hypothesized that FGF9 may provide protective or antiviral functions during IAV infection in the adult lung.

Multiple secreted proteins in the lung, including cytokines and growth factors, can regulate virus infection, the inflammatory response, and post-infection repair to combat severe IAV disease. Fibroblast growth factors (FGFs) are a family of secreted proteins that regulate tissue maintenance, metabolism, regeneration, and repair of adult tissues by binding to the extracellular domain of four transmembrane tyrosine kinase FGF receptors (FGFRs) [ 5 , 6 ]. In the developing lung, FGF1, FGF2, FGF3, FGF7, FGF9, FGF10, and FGF18 all perform essential regulatory functions mediating organogenesis [ 7 – 12 ]. FGF signaling is required for epithelial and mesenchymal maintenance, and elevated FGF signaling or FGF expression can have beneficial or detrimental impacts upon lung biology. In the adult lung, increased expression of FGF2, FGF7, and FGF10 promotes repair after lung injury, and FGF10 also promotes resolution of acute lung injury and acute respiratory distress syndrome [ 13 – 15 ]. In contrast, FGF9 is elevated in tissue samples from patients with mild to severe idiopathic pulmonary fibrosis (IPF), and FGF9 and FGF18 both contribute to the development of IPF in tissue culture [ 16 , 17 ].

Influenza A virus (IAV) is a respiratory pathogen in the Orthomyxoviridae family that causes both devastating pandemics and seasonal epidemics, resulting in significant morbidity and mortality worldwide. Once inhaled, IAV primarily targets differentiated epithelial cells of the upper and lower respiratory tract, including ciliated cells, club cells, and type I and type II alveolar epithelial cells (AT1 and AT2 cells, respectively) [ 1 ]. Infection of these cells, in addition to abortive infection of immune cells, leads to virus recognition by multiple pattern recognition receptors (PRRs), resulting in the production of proinflammatory cytokines and type I interferons (IFNs) which activate hundreds of IFN-stimulated genes (ISGs) [ 2 – 4 ]. While antiviral ISGs reduce viral burden and protect neighboring cells from IAV infection, excessive inflammation from type I IFN-induced cytokines and chemokines can exacerbate lung disease. Therefore, proper resolution of IAV infection requires a coordinated respiratory epithelial and immune cell response to limit viral replication, contain inflammation, and promote repair of the lung epithelium.

Results

Generation and characterization of FGF9-overexpressing transgenic mice To express Fgf9 from club cells in the conducting airway epithelium with temporal specificity, TRE-Fgf9-IRES-eGfp transgenic mice [10] were mated to transgenic mice with a reverse tetracycline transactivator (rtTA) driven under the Scgb1a1 promoter (Scgb1a1-rtTA). This cross, hereafter referred to as FGF9-overexpressing (FGF9-OE) mice, results in the induction of the rtTA element primarily in club cells after replacing the standard mouse diet with doxycycline-containing chow (DOX) (Fig 1A). Elevated levels of Fgf9 RNA were detected by real-time quantitative PCR (RT-qPCR) from whole lungs of FGF9-OE mice upon DOX administration for 24, 48, and 72 hours compared to uninduced FGF9-OE mice (S1A Fig). To further evaluate the induction of the TRE-Fgf9-IRES-eGfp cassette, lung sections isolated from FGF9-OE mice and mice containing only the TRE-Fgf9-IRES-eGfp or Scgb1a1-rtTA transgene (hereafter referred to as littermate “control” mice) were analyzed after 3 days of DOX administration. Lung sections from DOX-treated control mice or from uninduced FGF9-OE mice displayed no detectable eGFP activity in the airway epithelium (Figs 1B and S2). In contrast, DOX-treated FGF9-OE mice showed eGFP fluorescence co-localized within the club cells as denoted by SCGB1A1 staining (Figs 1B and S2). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. FGF9 overexpression from club cells after 3 days of DOX administration does not alter lung histology. (A) Schematic of the doxycycline chow (DOX)-inducible double-transgenic mouse breeding resulting in single-transgenic control mice (Scgb1a1-rtTA or TRE-Fgf9-IRES-eGfp) or double-transgenic FGF9-OE mice (Scgb1a1-rtTA and TRE-Fgf9-IRES-eGfp). (B-C) Representative images of control and FGF9-OE mouse lung sections after 3 days of DOX administration; (B) stained with DAPI (blue), anti-SCGB1A1 (red), and eGFP (green), scale bars = 50 μm, (C) stained with H&E, scale bars = 500 μm. (D) FGF9-OE and control mice were given DOX for 3 days, lungs were harvested, and single cell suspensions were analyzed for epithelial cells by flow cytometry as described in Materials and Methods. Data are represented as mean ± SEM and analyzed by unpaired student’s t test. https://doi.org/10.1371/journal.ppat.1010228.g001 To confirm that FGF9 is expressed in the FGF9-OE airway epithelium, lungs sections from control and FGF9-OE mice after 3 days of DOX treatment were stained for FGF9 by immunohistochemistry (S1B Fig). While FGF9 protein was indeed observed in the FGF9-OE airway epithelium, we only detected a modest FGF9 signal in the FGF9-OE airway compared to the 105−106 fold increase in Fgf9 transcripts we detected in FGF9-OE lungs by RT-qPCR (S1A Fig). Together, these data suggest that while FGF9 RNA is highly upregulated in the FGF9-OE lungs, DOX treatment results in only moderate expression of FGF9 protein from the airway epithelium. Histological analysis of the lung sections revealed no gross differences in the conducting airway or alveolar epithelium or evidence of cellular infiltrates in the lungs among DOX-treated FGF9-OE mice, DOX-treated control mice, or FGF9-OE mice without DOX (Fig 1C). To confirm whether 3 days of FGF9 overexpression altered epithelial cell number in the lungs, FGF9-OE and control mice were administered DOX for 3 days, and whole right lung lobes were harvested and analyzed by flow cytometry (S3A Fig). Consistent with our histological analysis of lung sections, we detected no differences in the number of total epithelial cells (EpCAM+), conducting airway epithelial cells (EpCAM+CD24+) or alveolar epithelial cells (EpCAM+CD24−) between control and FGF9-OE mice (Fig 1D).

FGF9-OE mice are more susceptible to respiratory virus infection than control mice Given the ability of FGF9 to inhibit recombinant VSVs [22], we hypothesized that FGF9 may inhibit IAV infection and thereby protect the lungs from severe disease. To test this hypothesis, FGF9-OE and littermate control mice were continuously administered DOX beginning 3 days prior to infection (d-3), infected intranasally (i.n.) (d0) with IAV A/WSN/33 (WSN), and then monitored for weight loss and survival. Although we expected the FGF9-OE mice to be protected from IAV infection, all FGF9-OE mice lost weight and succumbed to infection between 5–9 dpi (Fig 2A and 2B). In contrast, while the control mice initially lost weight at the same rate as the FGF9-OE mice, 100% of the control mice recovered and survived WSN infection. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. FGF9-OE mice are more susceptible to respiratory virus infection than control mice. (A-B) FGF9-OE and control mice were administered DOX beginning 3 days prior to infection (d-3), inoculated intranasally (i.n.) on d0 with 6x104 PFU WSN, and monitored for (A) weight loss (control n = 12, FGF9-OE n = 8) and (B) survival (control n = 31, FGF9-OE n = 28). (C-D) DOX-induced FGF9-OE and control mice were inoculated i.n. on d0 with 5 PFU PR8 (control n = 10, FGF9-OE n = 8) (C) or 1x104 PFU SeV-52 (control n = 23, FGF9-OE n = 7) (D) and were monitored for survival. (E) WT FVB/NJ mice were treated i.n. with PBS (n = 7) or 5 μg rFGF9 (n = 7) on d-3 and again concurrently with WSN infection on d0 and were monitored for survival. (F) Survival curve of WSN-infected FGF9-OE mice treated with DOX 1 day pre-infection (d-1, n = 9) or 1 dpi (d+1, n = 9) compared to historic d-3 DOX administration from panel B. For all experiments, weight loss and survival were monitored until 14 dpi. Data were pooled from 2 or more separate experiments. Data are represented as mean ± SEM and were analyzed by Mantel-Cox test (*, p < 0.05; **, p < 0.01; ****, p < 0.0001). https://doi.org/10.1371/journal.ppat.1010228.g002 To determine whether FGF9-OE mice were similarly more susceptible to other respiratory viruses, we evaluated survival of DOX-treated mice infected with IA A/PR/8/34 (PR8) or Sendai virus (SeV-52). While the majority of control mice recovered from PR8 infection (Fig 2C) and SeV-52 infection (Fig 2D), the FGF9-OE mice were more susceptible to both respiratory viruses compared to control mice. Together, these results demonstrate that overexpression of FGF9 from club cells leads to increased susceptibility to respiratory virus infection. To characterize this FGF9-driven increase in disease, we focused on elucidating the mechanisms of heightened WSN pathogenesis in FGF9-OE mice since the difference in survival between FGF9-OE and control mice was most striking during WSN infection. Because club cells in the wild-type adult mouse lung do not specifically express high levels of FGF9, we wanted to dispel the possibility that the FGF9-OE mice are more susceptible to WSN infection due to unintended effects of the double-transgenic mouse model unrelated to FGF9 expression. Therefore, we next determined whether rFGF9 similarly promoted susceptibility to WSN infection. WT FVB/NJ mice were treated with 5 μg rFGF9 or PBS i.n. on d-3 and again concurrently with WSN infection on d0, and survival was monitored. While all PBS-treated mice recovered from WSN infection, nearly all rFGF9-treated mice succumbed to the infection (Fig 2E), confirming that increased FGF9 in the respiratory tract enhanced influenza severity. Next, to determine when FGF9 overexpression was required for increased susceptibility to IAV infection, we induced FGF9 overexpression at varying time points. FGF9-OE mice were given DOX 1 day prior to infection (d-1) or 1 day post infection (d+1), were infected with WSN at d0, and were monitored for survival. FGF9-OE mice given DOX on d-1 displayed intermediate lethality compared to what we observed when treatment was initiated on d-3, with approximately 45% of the mice recovering from infection (Fig 2F). However, if DOX administration was started on d+1, then nearly all the mice survived WSN infection (Fig 2F). Taken altogether, these results demonstrate that forced overexpression of FGF9 in club cells increases susceptibility to different respiratory viruses, but only if overexpression is initiated prior to infection.

FGF9-OE mice have delayed viral clearance and exacerbated inflammation during IAV infection Increased susceptibility to WSN required FGF9 overexpression prior to infection. However, lethality in FGF9-OE mice did not occur until 5–6 dpi, likely due to increased viral burden, altered viral clearance, or an excessive inflammatory response. First, to evaluate the impact of FGF9 overexpression on viral burden, the lungs of DOX-treated, WSN-infected FGF9-OE and control mice were harvested at 1, 3, and 6 dpi to measure infectious virus by plaque assay (Fig 3A). At 1 dpi, the levels of replicating virus did not differ between the control and FGF9-OE mice. By 3 dpi, the FGF9-OE mice sustained viral titers of 105 PFU/ml while the viral titers in control mice were beginning to decline, with an approximate 1-log reduction in viral burden. At 6 dpi, nearly all the control mice had cleared replicating virus from their lungs, while the FGF9-OE mice failed to clear infectious virus and still displayed titers as high as 104 PFU/ml. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. FGF9-OE mice sustain higher viral titers and have increased lung inflammation. (A-C) FGF9-OE and control mice were administered DOX beginning on d-3 and infected on d0 with 6x104 PFU WSN i.n. Lungs were harvested at 1, 3, and 6 dpi. At each time point, n = 6–13 control or FGF9-OE mice were analyzed. (A) Infectious virus was quantified by plaque assay; dotted line represents 50 PFU/ml (limit of detection). (B) H&E-stained slides at 1 dpi (control n = 3, FGF9-OE n = 5) and 6 dpi (control n = 1, FGF9-OE n = 3) were scored for peribronchiolitis, squamous epithelial metaplasia, airway epithelial denudation, and alveolitis as described in the Materials and Methods. (C) Representative images of H&E-stained control and FGF9-OE whole lung sections at 1 and 6 dpi; (bottom) corresponding magnified insets (scale bars = 200 μm). Data are represented as mean ± SEM and were analyzed within each time point by unpaired student’s t test (*, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001). https://doi.org/10.1371/journal.ppat.1010228.g003 We next scored histopathology in tissue sections collected from DOX-treated FGF9-OE and control mice at early (1 dpi) and late (6 dpi) infection time points. At 1 dpi, we observed thickening of the FGF9-OE airways driven by significantly increased peribronchiolitis and a higher prevalence of squamous epithelial cell metaplasia, compared to relatively low levels of peribronchiolitis and a lack of squamous epithelial cell metaplasia in control lung sections (Fig 3B and 3C). By 6 dpi, control mouse lung sections displayed increased peribronchiolitis and alveolitis characterized by a moderate number of infiltrating leukocytes in the alveolar space. In contrast, we observed significantly more severe alveolitis in FGF9-OE lung sections at 6 dpi, characterized by a high number of infiltrating leukocytes, alveolar edema, and alveolar consolidation. We additionally observed significant denuding of the FGF9-OE airway epithelium at 6 dpi with severe peribronchiolitis around the remaining conducting airways (Fig 3B and 3C). Altogether, these results demonstrate that club cell-driven FGF9 overexpression led to elevated and sustained viral titers in the lung tissue. We also observed pervasive inflammation of the airway epithelium and, most significantly, the lung parenchyma during IAV infection, correlating to when the FGF9-OE mice began to succumb to infection.

FGF9-OE airway epithelial cells have a strong IFN signature at 1 dpi Secreted FGFs tightly associate with heparan sulfate proteoglycans on the surface of cells, which limits their diffusion and allows for robust autocrine and paracrine signaling on nearby neighboring cells [29–32]. In our FGF9-OE mouse model, FGF9 is secreted from club cells, which comprise the majority of the conducting airway epithelium in the mouse lung [8]. Therefore, we reason that the club cells, ciliated cells, and other epithelial cell types of the conducting airway are the cells that receive the most FGF9 stimulus and become transcriptionally altered. To investigate this further, we sorted airway epithelial cells (CD45−CD326+CD24+) (S3A Fig) from lung digests of DOX-induced FGF9-OE and control mice at 1 dpi and performed RNA-seq analysis on extracted RNA. Gene set enrichment analysis (GSEA) of the Gene Ontology Biological Processes database revealed that the top 10 most significant positively-enriched pathways in the FGF9-OE airway epithelial cells compared to control airway epithelial cells were all related to the innate immune response and IFN signaling, including “Cellular response to IFNβ,” “Defense response to virus,” and “Inflammatory response” (Figs 6A and S4 and S1 Table). Ifnb1, multiple Ifna subtypes (e.g. Ifna11 and Ifna14), and Ifnl2 were all upregulated by 1–3 log fold change in the FGF9-OE airway epithelial cells at 1 dpi, as well as multiple known antiviral ISGs (e.g. Isg15, Oas3, and Rsad2) (Fig 6B). Interestingly, several cytokines and chemokines found highly expressed in whole lung lysates at 1 dpi (Fig 4) were also upregulated in FGF9-OE airway epithelial cells (Il6, Ccl2, Ccl4, and Cxcl9) whereas G-CSF (Csf3) and many other highly-expressed cytokines and chemokines are not differentially expressed in FGF9-OE airway epithelial cells, suggesting that the FGF9-OE airway epithelial cells may be the source of only a subset of the cytokines and chemokines overexpressed in FGF9-OE lungs at 1 dpi. Importantly, FGF9 was highly upregulated in the FGF9-OE airway epithelial cells but was expressed at very low levels in the control airway epithelial cells. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. FGF9-OE airway epithelial cells display an elevated IFN response at 1 dpi. (A-D) FGF9-OE and control mice were administered DOX beginning on d-3 and infected on d0 with 6x104 PFU WSN i.n. (A-C) RNA extracted from sorted airway epithelial cells (CD45− CD326+ CD24+) at 1 dpi was analyzed by bulk RNA sequencing as described in Materials and Methods. (A) Top 10 most significant positively-enriched pathways, normalized enrichment scores, and adjusted p-values from GSEA analysis of Gene Ontology Biological Processes terms comparing control and FGF9-OE airway epithelial cells at 1 dpi. (B) Selected representative genes and log fold change from GSEA analysis in FGF9-OE airway epithelial cells at 1 dpi. (C) RT-qPCR validation of Ifnb1, Isg15, and Rsad2 expression in sorted FGF9-OE or control airway epithelial cells at 0 and 1 dpi. Gene expression was normalized to Gapdh and graphed as fold change over control (0 dpi) using the 2−ΔΔCt method. Data are represented as mean ± SEM and analyzed using unpaired student’s t test (**, p < 0.01; ***, p < 0.001). (D) Representative images of lung sections from control or FGF9-OE mice at 0 and 1 dpi stained with anti-ISG15 polyclonal sera (yellow) and analyzed by immunofluorescence microscopy (blue = DAPI, scale bars = 100 μm). https://doi.org/10.1371/journal.ppat.1010228.g006 RT-qPCR analysis of RNA extracted from sorted airway epithelial cells at 1 dpi validated the upregulation of several of these genes, including Ifnb1, Isg15, and Rsad2 (viperin) in FGF9-OE airway epithelial cells as compared to cells isolated from control mice (Fig 6C). We also assessed ISG15 protein expression by immunofluorescence microscopy. At 1 dpi we saw a modest upregulation of ISG15 protein expression in the control airway epithelium, but in the FGF9-OE lungs, ISG15 expression was dramatically upregulated in the airway epithelial cells (Fig 6D). Finally, to determine if FGF9 signaling alone resulted in this increased IFN signature in the conducting airway epithelium, we analyzed expression of several genes in FGF9-OE mice that were treated with DOX for 3 days but were not infected (0 dpi). RT-qPCR analysis of sorted airway epithelial cells from FGF9-OE and control mice at 0 dpi revealed no significant differences in Ifnb1, Isg15, and Rsad2 expression (Fig 6C). We also observed no ISG15 protein expression by immunofluorescence microscopy in the lungs of FGF9-OE mice following treatment with DOX but without viral infection (Fig 6D). Together, these data demonstrate that 3 days of FGF9 overexpression alone does not significantly alter IFN signaling in the conducting airway epithelium. Instead, high FGF9 signaling sensitizes the airway epithelial cells to induce a rapid, dramatic IFN signature, especially type I IFN, in response to infection in conducting airway epithelial cells.

[END]
---
[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1010228

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/