Virus-Triggered Cell Aging May Cause Lung Problems After Flu Infection

To assess the extent of IAV‐induced lung damage and the fate of senescent lung cells over time, we investigated mice exposed to a sublethal dose of H1N1p2009 until 90 days post‐infection (dpi) (for the viral load in lungs and body weight loss and recovery, see Figure S1A,B). A histochemical analysis (H&E staining) of lung sections revealed pulmonary lesions at 4 dpi that predominated at 7 and 14 dpi, including structural bronchial and alveolar lesions, inflammatory infiltrates, and airway epithelial abrasion that appeared complete in some areas (Figure 1A). On 28 dpi, the lungs appeared to have recovered partly from these structural changes, with a reduction in inflammatory infiltrates but persistence of abrasion of the airway epithelium (Figure 1A, lower panels). Interestingly, lung emphysema lesions (assessed by measurement of the mean linear intercept, Figure 1B) were observed at 28 dpi, together with the appearance of fibrotic lesions (as assessed with the modified Ashcroft score, Figure 1C). On 90 dpi, the bronchial airway epithelium had recovered partly, although the persistence of lung emphysema and fibrotic lesions was consistent with the establishment of a chronic lung disease (Figure 1A–C).

Alongside these structural changes, immunolabeling analyses revealed a massive accumulation of cells expressing the senescence markers p16 and p21 (Figure 1D,E). Cells stained for p16 or p21 were detected as early as 4 dpi in the airway epithelium, with a peak for p21‐positive cells at 7 dpi and a peak for p16‐positive cells at 14 dpi. Lung cells stained for p16 or p21 were rarer but still present at 28 dpi, indicating that senescent lung cells persisted after the virus had been cleared. Consistent with these observations, thoracic bioluminescence was observed in p16luc/+ heterozygous mice at 7–14 dpi (Figure S1C). At 4 dpi, almost all p16‐positive cells were bronchial epithelial cells that also expressed viral antigen (Figure 2A). At 7 dpi, an extension of p16 staining was observed in the lung parenchyma in cells not expressing the viral antigen, indicating the spreading of the senescence process. Beyond 14 dpi, the expression of viral proteins became poorly detectable. At 90 dpi, overexpression of p16 and p21 was no longer observed (Figure 1D and not shown).

Elevated expression of the galactosidase beta 1 (GLB1), the enzyme mediating the senescence‐associated‐β‐galactosidase activity, is observed during cellular senescence (Lee et al. 2006; Sun et al. 2022). Expression of GLB1 was enhanced in IAV‐infected lungs as assessed by Western blotting (7 dpi, Figure S1D). Histone H2A.X phosphorylated at serine 139 (gamma‐H2A.X) is a marker of the DNA damage response associated with cellular senescence. As shown in Figure 2B, elevated levels of gamma‐H2A.X (and p21) were observed in the lungs of IAV‐infected mice with a peak at 14 dpi. Immunolabeling confirmed the enhanced expression of γ‐H2A.X in lung tissue during infection (Figure 2C). In contrast, the lung levels of the DNA damage marker 53BP1 decreased during IAV infection (Figure S1E). Such opposing changes in lung gamma‐H2A.X and 53BP1 expression have been recently reported in response to a severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection (Gioia et al. 2023), supporting a common response elicited by IAV and SARS‐CoV‐2. Taken together, in agreement with other studies (Kulkarni et al. 2019; Lv et al. 2022; Schulz et al. 2023), IAV infection leads to cellular senescence in the lungs and this senescence associates with lasting damage to lung tissue.

To further identify the lung cell types that undergo senescence upon IAV infection, we performed co‐immunolabeling for p16 and specific cell markers on 14 dpi (the peak of p16 staining). Pulmonary vascular and microvascular endothelial cells, bronchial and alveolar epithelial cells, and macrophages expressed p16 protein, as shown by double‐immunofluorescence staining for p16 and CD31, Muc1, and CD68, respectively (Figure 3). A significant proportion of endothelial cells (the frequency of which being decreased on 14 dpi) and macrophages (increased frequency) expressed p16 (Figure 3A,B). Given the presence of bronchial epithelium abrasion on 14 dpi, only a few of the remaining bronchial epithelial cells were co‐stained for p16 and Muc1 at this time point (Figure 3C). In fact, most of the p16‐positive cells in the surrounding bronchial walls were positive for CD68 (Figure 3D). At 90 dpi, most of the bronchi had a normal epithelium (Figure 3E). The bronchi that did not recover contained senescent cells in the vessel wall. Figure 3E shows samples from the same mouse lung: the absence of p16‐positive cells and the complete airway epithelium repair in one bronchus (left panel) contrasts with the persistence of p16‐positive cells and only partial repair in a nearby bronchus (right panel).

To determine whether these phenomena occurred in non‐human primates, we examined the lungs of macaques infected with IAV and sacrificed on 28 dpi (Figure S2). Immunohistochemical studies of lung sections revealed a large number of p16‐positive cells primarily at sites of persistent alveolar damage and in thrombotic pulmonary blood vessels, as we have described previously following a SARS‐CoV‐2 infection (Lipskaia et al. 2022). As we observed in mice, the airway epithelium had not recovered at 28 dpi, and this time point was associated with the persistence of p16‐positive cells in the bronchial wall (Figure S2).

The epithelial cell is IAV's first target. To assess whether IAV could directly induce epithelial lung cell senescence, we analyzed publicly available datasets from human bronchial epithelial cells (GSE71766) (Kim et al. 2015) and mouse epithelial alveolar cells sorted from IAV‐infected mice (GSE57008) (Stegemann‐Koniszewski et al. 2016) on 3 dpi. Strikingly, IAV infection resulted in an enrichment in cell senescence‐related genes and aging‐related genes in both human and mouse lung epithelial cells (Figure S3).

We next assessed the effects of senescent cell elimination on post‐influenza chronic lung damage. To this end, we studied p16‐ATTAC mice expressing an inducible suicide gene (caspase 8) under the control of the p16 promoter and activated by the rapalog AP20187 (Born et al. 2023). It is noteworthy that IAV‐associated lung damage was similar in p16‐ATTAC mice and wild type mice, with the presence of lung emphysema and lung fibrosis on 28 and 90 dpi (Figure S4A,B). Compared with vehicle, the initiation of AP20187 treatment at the time of the IAV infection effectively eliminated p16‐positive cells in the lung parenchyma and the airways (Figure 4A). In particular, the CD68‐positive cells that accumulated in perivascular and peribronchial areas were eliminated upon AP20187 treatment (28 dpi, Figure 4A,B). Accordingly, recurrent administration of AP20187 was associated with markedly lower lung p16, p21, and gamma‐H2A.X protein levels on 28 dpi (Figure 4C, Figure S4C). Although not significant, treatment with AP20187 ameliorated the time course of weight loss and recovery (Figure S4D). Lung emphysema was less intense in p16‐ATTAC mice treated with AP20187 than in vehicle‐treated counterparts (Figure 4D). Remarkably, the airway epithelium completely recovered in AP20187‐treated mice by 28 dpi; this indicated marked acceleration of the epithelial repair process (Figure 4E). Lastly, pulmonary fibrosis was reduced in AP20187‐treated p16‐ATTAC mice (Figure 4F). In line with these histological findings, treatment with AP20187 was associated with lower expression of collagen 1 alpha 1, collagen 3, and phospho‐SMAD proteins—the main markers of pulmonary fibrosis (Figure 4G, Figure S4E). Hence, the genetic removal of p16‐expressing cells is associated with amelioration of chronic post‐influenza lung damage.

The ablation of virus‐induced senescent cells has been shown to decrease the pulmonary viral load and/or inflammatory marker expression soon after a SARS‐CoV‐2 infection (Lee et al. 2021; Tsuji et al. 2022; Aguado et al. 2023). To investigate whether the protection against the above‐described long‐term effects of IAV was due to an effect during the early phases of infection, AP20187‐treated p16‐ATTAC mice were analyzed on 7 dpi (the peak of the inflammatory response). Compared with vehicle‐treated mice, AP20187‐treated mice had a slightly but significantly higher pulmonary load of IAV (Figure 5A, left panel, in a quantitative RT‐PCR). Expression of Isg15 (but not Ifnb nor Oas3) was significantly elevated upon treatment with AP20187 (Figure 5A, right panels). Furthermore, the depletion of p16‐expressing cells did not change the pulmonary expression of transcripts encoding inflammatory markers like interleukin (IL)‐6, IL‐1β, and CCL2 (Figure 5B). It is well known that influenza disrupts the epithelial barrier (Short et al. 2016). The IAV‐induced reduction in the expression of epithelial cell markers (such as zonula occludens‐1 (encoded by Tjp1) and the tight junction protein occludin (encoded by Ocln) was not modified by AP20187 treatment (Figure 5C)). Inflammatory monocytes play a key role in pulmonary pathogenesis during influenza (Schmit et al. 2022). As expected, the recruitment of inflammatory monocytes/macrophages to the lungs was intense on 7 dpi—particularly in the perivascular and peribronchial areas (Figure 5D,E). Most of these cells were p16‐negative at this time point. Accordingly, AP20187 treatment failed to significantly lower the number of inflammatory monocytes/macrophages (Figure 5D,E). We conclude that the removal of p16‐expressing cells does not affect the early, acute response to influenza.

The use of senolytic drugs is a potentially feasible clinical approach for reducing the harmful effects of senescent cells (Justice et al. 2019; Chaib et al. 2022). To determine whether the pharmacological elimination of senescent cells had much the same effects as genetic senolysis in p16‐ATTAC mice, wild type mice were treated with ABT‐263 (navitoclax). The latter drug induces senescent cell apoptosis by counteracting the anti‐apoptotic function of B cell lymphoma‐2 (Bcl‐2) family members (Zhu et al. 2016). The oral administration of ABT‐263 was initiated 24 h after the start of the IAV infection and continued until the end of the experiment (28 dpi). As assessed by Western blotting, ABT‐263 efficiently reverted the virus‐induced expression of the senescence markers p16, p21, and gamma‐H2A.X in lung extracts (Figure 6A, Figure S5A). In line with these data, immunofluorescence staining revealed a reduction (but not total abrogation) in the number of p16‐ and p21‐positive cells in the lung parenchyma and bronchi (Figure S5B,C). ABT‐263 treatment did not alter body weight loss and recovery (Figure S5D), slightly reduced the viral load (Figure 6B) but had no significant effects on the expression of transcripts encoding ISGs, inflammatory genes, and epithelial barrier markers (7 dpi, Figure S5E and not shown). Quantification of IL‐6, IL‐1β, TNF‐α and IFN‐gamma proteins in lung extracts confirmed the lack of anti‐inflammatory effect of ABT‐263 (Figure S5F).

We then analyzed the potential effect of senolysis on long‐term pulmonary sequelae. In contrast to the p16‐ATTAC system, recurrent ABT‐263 treatment did not prevent the development of lung emphysema or lung fibrosis on 28 dpi (Figure 6C,D). We then investigated whether ABT‐263 inoculation impacts the airway epithelium repair. Interestingly, ABT‐263 treatment led to complete recovery of the airway epithelium on 28 dpi, together with the elimination of p16‐positive cells in the airway walls (Figure 6E,F). We conclude that the pharmacological depletion of senescent cells by means of ABT‐263 favors airway epithelial repair during influenza.

A question of clinical importance is whether early or late removal of senescent cells can improve pulmonary sequelae post‐influenza. As senescent (p16‐positive) cells persist after the clearance of the virus, we investigated the possibility that their genetic removal at late time points could have an impact on lung sequelae. To this end, p16‐ATTAC mice were treated with AP20187 from 15 dpi to the sacrifice at 28 dpi (AP D1‐15). Another group of mice was treated at the time of the IAV infection to 15 dpi (AP D15‐28). Remarkably, both protocols led to a complete restoration of the bronchial epithelium (Figure 7A) and to a trend towards improvement of pulmonary fibrosis (Figure 7B). Protection against emphysema was only observed when treatment was initiated at the time of infection but not on 15 dpi (Figure 7C).

Specific pathogen‐free C57BL/6J mice (8‐10‐week‐old, male) were purchased from Janvier (Le Genest‐St‐Isle, France). Mice expressing luciferase under the control of the p16 promoter (heterozygous p16LUC/+) mice were obtained from Dr. N. Sharpless (University of North Carolina School of Medicine, Chapel Hill, NC) (Burd et al. 2013). Mice expressing a killer gene construct driven by the p16 promoter (p16‐ATTAC mice) have been described (Born et al. 2023). Mice were maintained in a biosafety level 2 facility at the Animal Resource Center at the Institut Pasteur de Lille for at least 1 week prior to use to allow appropriate acclimatization. Mice were fed standard rodent chow (SAFE A04) (SAFE, Augy, France) and water ad libitum. Experiments in mice complied with current national and institutional regulations and ethical guidelines (Institut Pasteur de Lille/B59‐350009). Protocols were approved by the regional Animal Experimentation Ethics Committee (Comité d'Ethique en Expérimentation Animale, Hauts de France, CEEA 75) and the French Ministry of Higher Education and Research (Ministère de l'Education Nationale, de l'Enseignement Supérieur et de la Recherche) (authorization numbers: APAFIS#27417‐2020093019014739 v3). Cynomolgus macaques (Macaca fascicularis) originating from Mauritian AAALAC certified breeding centers were used in this study. All animals were housed in IDMIT infrastructure facilities (CEA, Fontenay‐aux‐roses), under BSL‐2 and BSL‐3 containment when necessary (Animal facility authorization #D92‐032‐02, Préfecture des Hauts de Seine, France) and in compliance with European Directive 2010/63/EU, the French regulations and the Standards for Human Care and Use of Laboratory Animals, of the Office for Laboratory Animal Welfare (OLAW, assurance number #A5826‐01, US). The protocols were approved by the institutional ethical committee "Comité d'Ethique en Expérimentation Animale du Commissariat à l'Energie Atomique et aux Energies Alternatives" (CEtEA #44) under statement number A18_025. The study was authorized by the "Research, Innovation and Education Ministry" under registration number APAFIS# 17836‐2018112716574241 v1.

AP20187 (HY‐13992, Cliniscience, Nanterre, France) was dissolved in ethanol, then diluted in 10% polyethylene glycol 400% and 0.2% Tween 20. ABT‐263 (Cliniscience) was dissolved in dimethyl sulfoxide, then diluted in 10% PEG400 and 60% Phosal 50 PG.

For infection, mice were anesthetized by intraperitoneal administration of 1.25 mg of ketamine and 0.25 mg of xylazine in 100 μL of phosphate buffered saline (PBS), and then intranasally (i.n.) infected with 50 μL of PBS containing 100 pfu of H1N1 A/California/04/2009 (pdm09) strain. Mice treated i.n. with PBS served as controls (non‐infected mice). The appearance of senescent lung cells was detected by bioluminescence tissue imaging using IAV‐infected heterozygous p16LUC/+ mice. Isoflurane‐anesthetized mice were injected intravenously with D‐luciferin substrate (L6882, Sigma, St Quentin Fallavier, France). The bioluminescence signal was acquired during 10 min with Photon Imager Optima (Biospace Lab, Paris, France). Results were analyzed using PhotoAcquisition software (Biospace Lab). To genetically eliminate senescent cells, p16‐ATTAC mice were treated intraperitoneally with AP20187 at 0.5 mg/kg three times a week starting the day before infection until the sacrifice. Mice were also treated between 1 and 15 dpi or between 15 and 28 dpi. To pharmacologically eliminate senescent (Bcl2‐expressing) cells, mice were administered in a daily manner by oral gavage with ABT‐263 (50 mg/kg) starting 1 day before infection until the sacrifice. Vehicle served as controls. Mice in both groups were sacrificed at different time points post‐infection. The right lung was quickly removed and immediately snap‐frozen in liquid nitrogen and then stored at −80°C until total RNA extraction for real‐time polymerase chain reaction analysis. The left lung was fixed by intratracheal infusion of 4% paraformaldehyde aqueous solution at a transpleural pressure of 30 cm H₂O for histological or immunofluorescence studies.

Adults, female cynomolgus macaques (macaca fascicularis) aged 2–3 years were exposed to a total dose of 2.1 × 10⁶ pfu of H1N1/2009/Cal07 via the combination of intranasal, oropharyngeal, and intratracheal routes, using atropine (0.04 mg kg⁻¹) for pre‐medication and ketamine (5 mg kg⁻¹) with medetomidine (0.05 mg kg⁻¹) for anesthesia. Animals were observed daily and clinical exams were performed at baseline and twice weekly on anesthetized animals using ketamine (5 mg kg⁻¹) and medetomidine (0.05 mg kg⁻¹). Body weight was recorded daily Animals were euthanized on 28 dpi. PBS‐treated macaques served as controls. Lung tissues were fixed in 4% PBS buffered formaldehyde for 2 days, rinsed in PBS, transferred to ethanol (70%) and then processed into paraffin‐embedded tissue blocks.

For transcriptomic analyses of datasets from IAV‐infected human bronchial epithelial cells and mouse lungs, data were extracted from the GEO dataset public database, and Gene Set Enrichment Analysis (GSEA) was performed using GSEA v2.0.13 software with default parameters (www.broadinstitute.org/gsea/↗). GSEA analyses were done on the datasets corresponding to human bronchial epithelial cells (BEAS‐2B) (GSE71766) and sorted type II alveolar epithelial cells after mouse infection (GSE57008), 3 days after IAV infection in both cases.

Paraffin‐embedded sections of lung were deparaffinized using xylene and a graded series of ethanol di‐lutions then processed for epitope retrieval using Antigen retrieval buffer (100X Tris‐EDTA buffer, pH 9.0, ab93684, Abcam) according to the manufacturer's instructions. For nuclear immunolabeling, tissues were permeabilized with 0.1% Triton X‐100 in PBS for 10 min. Saturation was achieved using Dako antibody diluents with 10% goat serum. For immunolabeling with primary antibody produced in mouse we used the M.O.M. (mouse on mouse) immunodetection kit, basic (ref. BMK‐2202, Vector Lab) according to manufacturer instructions. For double staining, first and second primary antibodies were diluted in Dako antibody diluents with 3% goat serum then incubated for 1 h at 37°C in a humidified chamber. After PBS washes, the sections were covered with secondary antibody (Dako antibody diluents with 3% of goat serum mixed with mouse or rabbit Alexa Fluor 555 or Alexa Fluor 660 [Thermofisher]) for 40 min at 37°C in a humidified chamber. The sections were secured with fluorescent mounting medium containing DAPI and protected with coverslips. Double‐immunolabeling for p16 and H1N1 HA was performed by using Tyramide SuperBoost Kits with Alexa Fluor Tyramides (B400923 and B40916, Invitrogen) according to manufacturer instructions. Fluorescence was recorded using an Axio Imager M2 imaging microscope (Zeiss, Oberkochen, Germany) and analyzed on digital photographs using Image J software (imagej.nih.gov/ij↗ /). The following primary antibodies were used for immunofluorescence: anti‐CDKN2A/p16INK4a (1:200 for common immunolabeling and 1:1000 for labelling using tyramide, ab54210, Abcam); anti‐CD31 (PECAM [platelet endothelial cell adhesion molecule]; 1:1000, ab182981, Abcam); anti‐CD68 (1:100, ab 283,654, Abcam); anti‐MUC1(1:200, ab109185, Abcam); anti‐H1N1 HA (1:1000, PA5‐34929, Invitrogen). To quantify p16, CD31, Muc1 and CD68‐positive cells, the entire tissue sections were scanned using Axioscan Z1 (Zeiss, Germany) and then subjected to digital image analysis using QuPath open software for bioimage analysis (https://qupath.github.io↗). Cells were identified and counted by DAPI nuclear staining. Cells were classified as either p16 positive or negative based on the average intensity of nuclear Alexa 630 staining, then classified as either CD31, Muc1, CD68 positive or negative based on the average intensity of cytosolic Alexa 555 staining. Double‐labeled cells were identified by nuclear 630 and cytosolic 555 average intensity.

Briefly, paraffin‐embedded lung sections were deparaffinized using the PT Module (Epredia) and Dewax and HIER Buffer H, pH 9.0 (TA‐999‐DHBH, Epredia) according to the manufacturer's instructions. Immunohistochemical labelling was performed on the Autostainer 360‐2D (Epredia). After PBS‐Tween 20 buffer (PBT010, SkyTek Laboratories) washed, slides were treated with Peroxide Block solution (ACA500, SkyTek Laboratories). The nuclei were then permeabilized with 0.1% Triton X‐100 in PBS for 10 min. SkyTek Laboratories' EZ Block solution (EZB125) was used to reach saturation. For immunolabeling, sections were incubated with the primary antibody diluted in Normal Antibody Diluent (ABB 500, EZB 125, SkyTek Laboratories). The following primary antibodies were used: rabbit anti‐p21 polyclonal antibody (28248‐1‐AP, Proteintech, 1:200), rabbit anti‐53BP1 polyclonal antibody (NB100‐304, Novusbio, 1:200) and rabbit monoclonal anti‐phospho‐Histone H2A.X antibody (20E3, #9718, 1:500, Cell signaling Technology). The sections were then washed with PBST buffer and incubated with the anti‐rabbit HRP‐labeled polymer (K4003, Dako). The sections were then washed, incubated with DAB working solution (ImmPACT DAB EqV, Peroxidase substrate, SK‐4103, Vector Laboratories), and rinsed twice with distilled water. The nuclei were stained using hematoxylin (HAQ500, SkyTek Laboratories). The sections were then rinsed twice with water and a bluing solution (0.1%) of sodium bicarbonate. The slides were then dehydrated, cleared, secured with a permanent, non‐aqueous medium, and covered with coverslips for protection. To quantify p21‐ and 53BP1‐positive cells, the entire tissue sections were scanned using Axioscan Z1 (Zeiss, Germany) and then subjected to digital image analysis using QuPath open software for bioimage analysis (https://qupath.github.io↗). Cells were classified as either positive or negative based on the average intensity of nuclear DAB staining.

For morphometry studies, 5 μm‐thick sagittal sections along the greatest axis of the left lung were cut in a systematic manner. Lung emphysema was measured using mean linear intercept methods on hematoxylin–eosin (H&E) coloration. Briefly, 20 fields/animal light microscope fields at an overall magnification of 500 were overlapped with a 42‐point, 21‐line eyepiece according to a systematic sampling method from a random starting point. MLI measurement was performed by an independent operator using semi‐automated measurement of MLI. Both methods (manual and semi‐automated) performed by two independent observers provided the same results. To measure bronchial epithelial thickness, the entire tissue sections were scanned using Axioscan Z1 (Zeiss, Germany) and then epithelial thickness was measured using ZEN 3.0 software (ZEISS, Rueil Malmaisson, France) for bioimage analysis. The samples were examined microscopically and morphometrically by two investigators who were unaware of the origin of the material. For each animal, at least 10 transversally cut terminal bronchioles with an adequate cross‐sectional profile (less than 10% of variation in maximal and minimal diameter) were measured, and only airways with a visible full perimeter were analyzed. The average epithelium thickness was determined by measuring the distance between the basal membrane limit and the apical membrane limit in at least 10 regions of bronchia. Values measured for each of the 10 airways were averaged to provide a single data point for each animal. Lung fibrosis was quantified on lung sections stained with Sirius Red (Picro Sirius Red Stain Kit, ab150681, Abcam, Cambridge, UK) using the modified Ashcroft scale. Briefly, the lungs were scanned microscopically with a 20‐fold objective, which allowed the evaluation of fine structures while also providing a sufficiently broad view. The entire section was examined by inspecting each field in a raster pattern. Areas with dominating tracheal or bronchial tissue were omitted. The grades were summarized and divided by the number of fields to obtain a fibrotic index for the lung.

Lung extracts were prepared as described (Delval et al. 2023). Proteins were then separated using 12 or 15% SDS‐PAGE and then transferred from the gel to a PVDF membrane. The antibodies used are as follows: anti‐β‐actin (A1978, Sigma Aldrich); anti‐γ‐H2AX (Ser139; #9718, Cell Signaling Technology); anti‐CDKN2A/p16INK4a (#PA5‐20379, Thermo Fisher Scientific); anti‐CDKN1A/p21 (#ab 8224, Abcam); anti‐collagen I (Col1A1) (#ab260043, Abcam); anti‐collagen III (#ab184993, Abcam); anti‐phospho‐Smad3 (Ser423/425) (#9520, Cell Signaling); and anti‐Smad3 (#9523, Cell Signaling). The detection was made by using the appropriate horseradish peroxidase‐conjugated secondary antibody (Sigma Merck). The signal was detected using an enhanced chemiluminescence detection system (GE Healthcare). The unsaturated images were acquired using a ChemiDoc MP imaging system (Bio‐Rad), and the signal densities were quantified using Image Studio Lite ver 5.2 (LI‐COR). To normalize, an antibody directed against β actin (#4970, Cell Signaling) was used.

Viral load in the lungs was determined by quantifying viral RNA encoding the M1 protein (segment 7) (Heumel et al. 2024). RNA was reverse‐transcribed with SuperScript II Reverse Transcriptase (Invitrogen, Waltham, MA) using primers specific for M1 (5′‐TCTAACCGAGGTCGAAACGTA‐3′). Quantitative PCR was performed using TaqMan Universal PCR Master Mix (Applied Biosystems, Waltham, MA), using primers for M1 (Table 1) and M1‐specific TaqMan probe (FAM) 5′‐TTTGTGTTCACGCTCACCGTGCC‐3′ (TAMRA). Amplification was performed using the QuantStudio 5 K Flex Real‐Time PCR System (Applied Biosystems). Assessment of gene expression was performed by quantitative RT‐PCR using standard procedures. RNA was reverse‐transcribed using a High‐Capacity cDNA Archive Kit (Life Technologies, Carlsbad, CA, USA). The resulting cDNA was amplified using SYBR Green‐based real‐time PCR and the QuantStudio 5 K Flex Real‐Time PCR System (Applied Biosystems). The primers used are listed in Table 1. Relative mRNA levels were determined according to the 2^−ΔΔCt (cycle thresholds) method by comparing (i) the PCR Ct for the gene of interest and the housekeeping gene (ΔCt) and (ii) the ΔCt values for the treated and control groups (ΔΔCt). Data were normalized against the expression of the Gapdh gene and expressed as fold‐change over the mean gene expression level in mock‐treated mice.

Lung extracts were lysed in cold RIPA buffer (50 mM Tris–HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X‐100), 0.1% sodium deoxycholate supplemented with 1 mM PMSF and protease inhibitors (Roche Diagnostics, Basel, Switzerland), and centrifuged at 10000g for 5 min at 4°C to remove debris. The protein concentrations of clarified lysates were quantified using the PierceTM BCA protein assay kit (ThermoFisher Scientific). Samples were normalized to a single protein concentration (1 mg/mL) using the same lysis buffer. Cytokines were quantified as the Olink^R target 48 mouse cytokine kit exactly as described by the manufacturer (Olink Proteomics AB, Upssala, Sweden).

Statistical analyses were performed using GraphPad Prism 9 software (San Diego, CA). One‐way or two‐way ANOVA followed by Bonferroni post hoc test were used to compare the means of more than two independent groups. Comparison between two groups was performed by using the Mann–Whitney U test. Unless otherwise stated, data are expressed as individual values and mean ± SEM or ±SD.