Respiratory syncytial virus infection in early life disrupts epithelial barriers via IL-33/ST2/MyD88 signaling axis

Background Respiratory syncytial virus (RSV) infection at early in life might impair epithelial barrier which can increase the risk of developing asthma in adulthood. However, whether and how RSV and RSV-induced IL-33 contribute to the procedure are still unclear. Methods In vivo, 7-day-old C57BL/6 mice were infected pernasally with RSV, then viral replication, lung inflammation and barrier integrity evaluated at various time points. In vitro, human epithelial cells were infected with RSV in the presence or absence of IL-33, and changes in the expression and localization of apical junction complex proteins (AJC) evaluated by western blotting and immunofluorescence. The involvement of components of the IL-33/ST2/MyD88 axis were verified by blockade of endogenous IL-33 signaling and inhibition of MyD88. Results Exposure of neonates to RSV infection resulted in impairment of the airways epithelial barrier, as shown by reduced expression of tight junction proteins (ZO-1, Occludin) and adherent junctions (E-cadherin) in the lung tissues, effects which were significantly abrogated in St2 -/- mice compared with normal controls but exacerbated by topical application of exogenous IL-33, which also activated MyD88-mediated NF-κB signaling. In vitro, knockdown of St2 by siRNA transfection or MyD88 inhibition partially restored the expression of E-cadherin, ZO-1 and Occludin in RSV-infected epithelial cells. Conclusion RSV infection of neonatal airways induces substantial release of IL-33 by epithelial cells, which in turn results in reduced expression and localization of AJC proteins by activation of the MyD88-mediated NF-κB signaling pathway, which may represent a potential target for therapeutic intervention in RSV-mediated lung diseases.

Respiratory syncytial virus (RSV) is a filamentous, enveloped, negative-sense, single-stranded RNA virus belonging to the Orthopneumovirus genus of the Pneumoviridae family [1] . Epidemiological studies have shown that RSV infection is the leading cause of acute lower respiratory infections (ALRI) in children younger than five years as well as an important cause of death in childhood. Globally, it is estimated that approximately 33.8 million new cases of RSV-associated ALRI occur annually, resulting in 3.6 million hospitalizations and more than 100,000 deaths in these children [2,3] . Similarly, a nationwide prospective surveillance study of ALRI in China from 2009 to 2019 reported that RSV (16.8%) was one of the most frequently detected viral pathogens in both preschool and school-age children, second only to the influenza virus (28.5%)[4]. Early-life RSV infection is strongly associated with subsequent long-term respiratory sequelae, including recurrent LRTI, recurrent wheezing and childhood asthma [5,6] . Although the underlying pathogenic mechanisms of these phenomena are not well understood, there is increasing evidence that long-term impairment of epithelial barrier integrity is a key factor[7,8]. The primary structure that regulates this airway epithelial barrier is the apical junctional complex (AJC), formed by tight junctions (TJs) and adherent junctions (AJs) which contribute to the maintenance of apical-basal cellular polarity and cellular signaling[9]. Previous research has shown that, in both animal and human cell-based in vitro models, RSV infection results in airways barrier dysfunction by interrupting the assembly of various AJC proteins, but the molecular mechanisms of these effects remain poorly understood [10,11]. Interleukin-33 (IL-33), a member of the IL-1 family, is released as an epithelium-derived alarmin and activates MyD88-dependent signaling pathways in cells expressing the ST2 (IL-1RL1) receptor following tissue stress or damage[12,13]. IL-33 expression in the airway epithelium is detectable early in embryogenesis and remains elevated during the first two weeks of life[14,15]. As an “alarmin” cytokine, the expression and secretion of IL-33 may be further elevated in response to tissue injury, pathogen infection or other cellular stresses[16]. In response to RSV infection, rapid production of IL-33 by the airways epithelial cells is considered a crucial step in initiating the innate, anti-viral response and subsequent activation of immune cells[17]. Saravia et al demonstrated that, compared with adult mice, RSV infection of neonatal mice resulted in significantly elevated IL-33 expression and more severe associated immunopathogenetic features such as airways hyperresponsiveness, Th2-type inflammation and mucus hypersecretion, many of which were reduced or abolished by prior administration of IL-33 neutralizing antibodies, underlining the likely critical role of IL-33 in RSV-related airways pathology, particularly in neonates[18]. Nevertheless, there is currently very little information as to if and how IL-33 might be involved in regulating epithelial integrity during RSV infection in neonates. In order to address our hypothesis that RSV infection of neonates induces the release of IL-33 by airways epithelial cells, facilitating the recruitment of the adapter protein MyD88 and subsequent activation of NF-κB signaling through its binding to the ST2 receptor, all of which eventually contribute to epithelial barrier dysfunction, we developed a murine model of neonatal RSV infection which we used to monitor structural changes in the AJC at different time points after RSV infection. We also employed an in vitro model in which human epithelial cells were infected with RSV in the presence or absence of IL-33, and changes in the expression and localization of AJC evaluated by western blotting and immunofluorescence. The involvement of components of the IL-33/ST2/MyD88 axis were verified by blockade of endogenous IL-33 signaling and inhibition of MyD88. We believe that our findings provide new insight into the molecular signaling that governs epithelial barrier integrity and defining novel pharmacological targets for antiviral therapy.

Virus preparation and Animal model of RSV infection Human respiratory syncytial virus (RSV) A2 strain was kindly provided by Prof. Yuan Jing (Department of Bacteriology, Capital Institute of Pediatrics) and propagated in monolayers of HEp-2 cells (American-type culture collection) grown in MEM (Invitrogen Life Technologies) supplemented with 5% fetal bovine serum (FBS). The viral titer was determined by plaque assay as previously described[19], then aliquoted and stored at 80°C until use. The neonatal C57BL/6 mice were infected intranasally with 1.0 × 10 6 plaque-forming units (pfu) RSV A2 in 10 μL phosphate-buffered saline (PBS) at day 7 after birth, while control animals were mock inoculated intranasally with an equal volume of PBS containing 10% sucrose. Cell culture and treatment The human bronchial epithelial cell line 16HBE14o (henceforth referred to as 16HBE cells) were used for in vitro assays, and maintained in DMEM (Life Technologies, Grand Island, NY) containing 10% FBS, 10mM glutamine, 100IU/ml penicillin, and 100μg/ml streptomycin. An in vitro model of RSV infection was set up as previously described[20]: briefly, when the cells reached 80 to 90% confluence, the monolayers were infected with RSV at a multiplicity of infection (MOI) of 0.1. Following 2 hours of viral adsorption, the cells were rinsed twice with cold PBS to discard the medium containing virus, and then cultured with fresh DMEM medium in the presence or absence of human recombinant IL-33 (rhIL-33) (10 ng/mL, R&D systems, MN, USA) for an additional 72 hours at 37°C. RSV infection was confirmed by immunofluorescence staining with a FITC-labeled monoclonal antibody to RSV (GeneTex, CA, USA). For some inhibition experiments, the 16HBE cells were preincubated with MyD88 inhibitor TJ-M2010-M (hereinafter referred to as TJ-5) for 2 h prior to RSV infection. At the indicated time points, the cells and culture supernatants were collected for further examination. Flow cytometric analysis of infiltrating inflammatory cells Single-cell suspensions were isolated from the upper lobes of the right lungs of the experimental animals as previously described [21]. Briefly, lung samples were cut into pieces and digested in DPBS medium containing DNase I (Sigma-Aldrich) (50 U/mL) and collagenase VIII (Sigma-Aldrich) (250 U/mL) at 37 °C for 30 min. Single-cell suspensions were obtained by passing the lung tissue digest through a 70-μm cell strainer. After centrifugation at 300 × g, the retrieved cells were resuspended in RPMI-1640 complete medium and counted using an automatic cell counting instrument (Bio-Rad TC20, Bio-Rad Laboratories Inc.). Surface staining was performed based on a previously published method[22], and a fixable viability dye (eBioscience) was used to exclude dead cells. The cell-surface markers to characterize inflammatory cells and the antibodies used in present study are summarized in Table 1, 2. The samples were acquired on a BD LSRFortessax-20 flow cytometer (BD Biosciences, San Jose, CA, USA) and analyzed using FlowJo software (version 10, Tree Star, Inc., Ashland, OR, USA). Real-time quantitative PCR analysis (RT-PCR) of Viral and Cytokine gene expression Total RNA from lung tissues or cell pellets was extracted using Trizol (Thermo Fisher Scientific) and isopropanol (Aladdin, Shanghai, China), and its quality assessed using a NanoDrop2000 Spectrophotometer (Thermo Fisher). cDNA was synthesized using 5 × All-In-One RT MasterMix (Applied Biological Materials, Richmond, BC, Canada) following the manufacturer’s protocol. The transcript levels of the target genes were determined by real-time PCR using the Applied Biosystems 7300 real-time PCR system (Applied Biosystems, Foster City, CA, USA) with BlasTaqTM 2 × qPCR MasterMix. (Abcam, BC, Canada). PCR specific for target genes was performed at 50 °C for 2 min and 95 °C for 10 min, followed by 40 °C two-step cycles (15 s at 95 °C, 1 min at 60 °C). Cycling conditions for cytokine RNA comprised an initial denaturation step at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 60 s. The mRNA levels were calculated using the 2 −△△Ct method and normalized to that of GAPDH. Primer sequences specific for target genes were designed with Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast) and are detailed in Table 3. Lung Histology analysis After fixation with 4% formaldehyde, the left lungs were paraffin-embedded and sliced into 4 μm thick sections. Deparaffinized sections were stained with periodic acid-Schiff (PAS, Wuhan Servicebio Technology Co. Ltd., Wuhan, China) to detect mucus production. Inflammatory infiltrates were observed using H&E staining (Beijing Solarbio Technology Co. Ltd., Beijing, China). All procedures have been described in detail previously[23]. Measurement of cytokine concentrations in lung homogenates 100 mg of right lung tissue were weighed then homogenized with 5 × volume PBS containing 1% Triton X-100 and protease inhibitor cocktail (Roche Diagnostics GmbH). After centrifugation to remove debris, aliquots of the homogenate supernatants were collected and stored at -80°C for cytokine measurement. IFN-α, IFN-γ, IL-33, IL-25, thymic stromal lymphopoietin (TSLP), IL-1β, IL-6, and TNF-α concentrations in lung homogenates were measured using commercial ELISA kits (Invitrogen, San Diego, CA, USA) according to the manufacturer’s instructions. St2 siRNA synthesis and cellular transfection St2 -specific small interfering RNA (siRNA) oligonucleotides were purchased from GenePharma (GenePharma, Suzhou, China). When growing to siRNA in Opti-MEM medium (Invitrogen, Carlsbad, CA) using lipofectamine RNAiMAX reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Scrambled (non-target) siRNA was used as a negative control. The efficiency of siRNA transfection was measured by RT-PCR or western blot analysis. The sequences of siRNA used in the experiments were as follows: St2: sense: 5′-GCG AAU GUC ACC AUA UAU ATT-3′ and anti-sense: 5′-UAUAUAUGGUGACAUUCGCTT-3′. Non-specific siRNA sequences were used as negative controls: sense: 5′-UUC UCC GAA CGU GUC ACG UTT-3′ and anti-sense: 5′-ACG UGA CAC GUU CGG AGA ATT-3′. Immunohistochemistry and immunocytochemical staining of junctional proteins Immunohistochemical staining was processed as described previously[24]. Briefly, the paraffin-embedded tissue sections were deparaffinized in xylene and ethanol, then heated in citric acid-EDTA solution to unmask antigen. After permeabilizing in PBS containing 0.1% Triton X-100 for 20 min, the slides were incubated with the primary antibodies overnight at 4°C, followed by incubation with HRP-conjugated secondary antibody. The positive cells were developed by diaminobenzidine reagent, and the nuclei stained with hematoxylin. At least 10 microscopic fields per section were counted and Image-Pro Plus 6.0 used for statistical analysis, and the immunoreactivity for each antibody was quantified by calculating the percentage of positive staining area relative to the total area of the selected region. For immunofluorescent labeling of TJ and AJ proteins, the 16HBE cells following the indicated treatments were fixed in cold methanol for 15min. The fixed cells were incubated with specific primary antibodies directed against Occludin, E-cadherin and ZO-1, followed by incubation with Alexa Fluor-labeled secondary antibodies and then stained with DAPI. The signals were analyzed using a confocal laser scanning microscope (TCS SP8 STED, Leica, Germany). Image-Pro Plus software (Media Cybernetics, Silver Spring, MD) was used to quantify the fluorescence intensity. The antibodies and working dilutions used to stain the sections are listed in Table 4. Western immunoblotting After the indicated treatments, the total protein of lung tissue and cell lysates was extracted using RIPA lysis buffer (Thermo Scientific, Waltham, MA). Equal concentrations of protein were separated by 10% SDS-PAGE. After transferring to the PVDF membranes (Millipore, USA), the membranes were blocked and incubated with specific primary antibodies overnight at 4°C. Then, the blots were incubated with corresponding goat anti-rabbit or goat anti-mouse second antibody at 37°C for 1 h. Ultimately, the blots were visualized with regular or enhanced chemiluminescence (ECL) substrates (Thermo Scientific, Waltham, MA), and immunoreactive bands imaged using ChemiDoc Imaging Systems (Bio-Rad Laboratories). Glyceraldehyde-3- phosphate dehydrogenase (GAPDH) was used as a lane-loading control. Densitometric quantification was performed using NIH Image J software. The antibodies and working dilution used in the section are listed in Table 4. Statistical Analysis Statistical analysis of experimental data was performed with GraphPad Prism 8 (La Jolla, CA). Results are expressed as mean ± SEM except where noted from at least 3 experiments. The independent Student’s t test was used to compare 2 groups, while 1- or 2-way ANOVA was used to compare the equality of means with more than 2 conditions, followed by Bonferroni correction for parametric statistical evaluation. Statistical significance was set at P < 0.05. Early-life RSV infection caused pulmonary inflammation and epithelial barrier dysfunction To obtain a dynamic evaluation of pathological changes in mice infected with RSV at neonatal stage, the mice were inoculated pernasally with RSV or vehicle control on the 7th day after birth. Lung morphology, mucus production and epithelial barrier integrity were visualized by H&E staining, PAS staining and immunohistological staining at days 0, 3, 10, 21 and 35 following viral exposure (Figure 1A). RT-PCR analysis revealed that the viral replication was first detected at day 3 post infection (3 dpi), reaching a peak at day 10 dpi then followed by a rapid decrease in viral copies from 11 to 35 dpi (Figure 1B). Although H&E stained sections showed no obvious structural alterations or differences in inflammatory cellular infiltration between RSV-infected and mock-infected mice at 3 dpi, elevated mucus production, as demonstrated by PAS staining, was observed in the RSV-infected animals at this time point. In line with the viral peak detected at 10dpi, RSV-infected mice exhibited a significantly higher inflammatory response compared with control mice, which was characterized by extensive peribronchial and perivascular inflammation, alveolar septal thickening and mucus overproduction. At 21 days post-infection (dpi), no significant differences in inflammation scores were observed between the two groups, although excessive production of mucus was still observed in the lungs of the RSV-infected mice, exhibiting a higher PAS score compared to the control animals at this time point (Figure 1D). At 35 dpi both inflammation score and PAS score were comparable between the RSV-infected mice and the controls, indicating that the acute lung injury caused by intranasal inoculation with RSV had resolved(Figure 1C,D). Next, immunohistochemical analysis was used to examine the expression and localization of different Tight Junction (TJ) (ZO-1, Occludin) and Adherent Junction (AJ) (E-cadherin) proteins in order to evaluate the integrity of the Apical Junctional Complex (AJC) in the bronchiolar epithelium of the mock- and RSV-infected animals at various time points following viral inoculation. While the mock-infected, control animals demonstrated a well-developed AJC in the stratified epithelium lining the bronchiolar lumen, RSV infection induced significant changes in the organization of the AJC featured most obviously by a reduced expression of ZO-1, Occludin, and E-cadherin in the apical epithelial layers at 10 dpi. Although the expression of these 3 proteins was observed to increase gradually from 21 dpi, quantification analysis revealed that, compared with baseline expression (0 dpi) the levels of expression of ZO-1, Occludin and E-cadherin at 35 dpi remained significantly depressed (Figure 1E). Early-life RSV exposure resulted in marked inflammatory cellular infiltration and pro-inflammatory cytokine production In order to gain insight into the inflammatory cellular process in response to RSV infection within the neonatal period, the numbers of NK cells, lymphocytes, eosinophils and neutrophils infiltrating the lung tissues at various time points following RSV exposure were analyzed and quantified by flow cytometry (Figure 2A). We observed a significant increase in the mean total numbers of CD45 + cells in the lungs of RSV-infected mice beginning from 3 dpi to 21 dpi when compared to those of the mock-infected animals. This was accompanied by an early, significant increase in the mean numbers of neutrophils and NK cells at 3 dpi, followed by an elevated mean number of eosinophils at 10 dpi. In addition, compared with the saline-challenged control animals, a significantly higher mean number of both CD4 + T cells and CD8 + T cells were observed in the RSV-challenged mice at 10 and 21 dpi. Interestingly, a persistent, significant elevation of the mean numbers of NK cells was observed at days 3, 10 and 21 following RSV exposure when compared with saline-challenged control animals (Figure 2B). ELISA analysis of lung homogenates demonstrated that pernasal infection of the animals with RSV at day 7 after birth resulted in a surge in the mean local concentrations of the pro-inflammatory cytokines IFN-α, IFN-γ, IL-6 and IL-1β, particularly from 3 dpi to 10 dpi with the mean concentration of TNF-α additionally significantly elevated at 10 dpi when compared to the mock-infected controls. Local mean concentrations of the epithelium-derived alarmins IL-33 (on days 10 and 21 following infection) and TSLP (on days 3 and 10 following infection) were also significantly elevated in the RSV-infected mice compared with the controls, although we observed no significant change in the mean concentrations of IL-25 at any time point (Figure 2C). In vitro administration of IL-33 exacerbated the depression of expression of epithelial junctional proteins through activation of MyD88-mediated NF-κB signaling Immunofluorescent staining analysis revealed that exposure of human bronchial epithelial cells to RSV induced the formation of enlarged, multinucleated syncytia reflecting fusion of neighboring cells, while RT-PCR analysis revealed a significantly elevated mean number of viral copies in the cells from 24 to 72 hours following infection compared with mock-infected cells (Figure 3A,B). The mean relative expression of IL-33 mRNA and protein was significantly elevated in the infected cells commencing from 48 up to 72 hours following RSV infection, in contrast to the mock-infected controls (Figure 3C,D). To further investigate the possible involvement of IL-33 in the pathogenesis of RSV-induced epithelial damage, the epithelial cells were infected with RSV in the presence or absence of exogenous IL-33. Firstly, there was no significant difference in the viral load recovered from RSV-infected cells in the presence and absence of exogenous IL-33, suggesting a lack of influence of the exogenous IL-33 on the RSV infectivity (Supplementary Fig 1). Secondly, as shown by immunofluorescent staining, compared with control group displaying an appropriately intercellular localization of E-cadherin, ZO-1 and Occludin, the RSV-treated cells showed discontinuous and abnormal subcellular staining of these junctional proteins. Furthermore, administration of exogenous IL-33 resulted in further exacerbation of barrier dysfunction in RSV-infected cells, as evidenced by further, significantly diminished staining intensity of these junctional proteins (Figure 3E). Similarly, western blotting demonstrated that, in comparison to RSV-infected cells, exogenous IL-33 treatment resulted in a significant reduction in the expression of the ZO-1, E-cadherin, and Occludin proteins (Figure 3F). Previous studies have demonstrated that myeloid differentiation factor 88 (MyD88) serves as a crucial adaptor molecule in transducing interleukin-1 receptor (IL-1R)-mediated signaling and initiating downstream inflammation cascades, including the activation of NF-κB signaling and mitogen-activated protein kinases[25]. As shown by our present data, compared with RSV-infected cells and IL-33-treated cells both showing a modest increase in MyD88 expression, administration of exogenous IL-33 resulted in a marked, highly significant further elevation of expression of MyD88 in the RSV-infected cells, as well as a higher ratio of pNF-κB to total NF-κB (Figure 3F). Taken together, these data show that IL-33 exacerbates the impairment of epithelial integrity in the process of RSV infection, a phenomenon which might be at least partly dependent on MyD88-mediated NF-κB signaling activation. Blockade of IL-33 signaling attenuated epithelial barrier damage in response to RSV exposure Having demonstrated that IL-33 can amplify the decline in the expression of AJ and TJ proteins in response to RSV, we then addressed the question whether or not blockade of endogenous IL-33 signaling may alleviate RSV-induced epithelial impairment. To address this question, we repeated the neonatal RSV infection protocol using 7-day-old, St2 −/− mice. As demonstrated by H&E staining of lung tissues obtained from these animals 10 days following pernasal RSV infection, this infection resulted in marked inflammatory infiltration in the lung parenchyma of St2 −/− mice compared with mock-infected animals, the magnitude of which, as measured by the inflammation score, did not significantly differ from that observed in the WT mice (Figure 4A). The integrity of the epithelial barrier was assessed by immunohistochemical staining of E-cadherin, ZO-1 and Occludin proteins in lung sections. As shown in Figure 4B, the reduction in expression of all three proteins in response to RSV infection was significantly attenuated in the lung tissue of the St2 −/− mice compared with the WT mice. Correspondingly, western blotting showed significantly elevated expression of all three proteins in the lung tissue of the St2 −/− mice compared with the WT mice, as well as significant attenuation of the expression of MyD88 as well as the ratio of pNF-κB to total NF-κB (Figure 4C). We finally proceeded specifically to silence the St2 gene in 16HBE cells using siRNA transfection, a procedure which we validated by showing absence of expression of the gene at the level of both mRNA and protein (Supplementary Fig 2), and examined its effects on the expression and localization of junctional proteins using both immunofluorescent staining and western blotting analysis. As shown in Figure 5A, in comparison to scrambled siRNA-transfected 16HBE cells, knockdown of the St2 gene in 16HBE cells resulted in significantly increased intensity of staining of the Tight Junction (ZO-1 and Occludin) and Adherent Junction (E-cadherin) proteins in RSV-infected cells in the presence and absence of exogenous IL-33. Similarly, using western blotting analysis of cell lysates, we observed that the effects of IL-33 in reducing expression of E-cadherin, ZO-1 and Occludin proteins in RSV-infected cells could be abrogated by prior deletion of the St2 gene (Figure 5B). MyD88 inhibition abrogates RSV-induced epithelial damage in the presence and absence of exogenous IL-33 Since the aforementioned data demonstrate that RSV-induced epithelial barrier injury is related at least in part to MyD88-mediated activation of the NF-κB signalling pathway, we hypothesised that the observed RSV-induced epithelial barrier dysfunction and AJC disassembly can be mitigated by targeting MyD88 protein. To test our hypothesis, we investigated the effects of a new small-molecule MyD88 inhibitor, TJ-5, in preserving the integrity of the airways epithelium during RSV infection. As shown by immunofluorescent staining of 16HBE cells cultured in vitro, reduction of the expression of E-cadherin, ZO-1 and Occludin-1 at intracellular junctions in RSV-infected cells was significantly reversed in the presence of TJ-5 in the presence and absence of exogenous IL-33 (Figure 6A). Finally, western blotting analysis confirmed the protective effects of TJ-5 on the expression of junctional proteins, and its significant reversal of the elevated ratio of pNF-κB to NF-κB observed in RSV-infected cells in the presence and absence of exogenous IL-33 (Figure 6B).

The airways epithelium is considered the first line of defense against harmful environmental stimuli including microbes, pollutants and allergens. The intercellular contacts between epithelial cells lining the airways are established and maintained by the apical junctional complexes (AJCs). AJC consist of tight junctions (TJs) and adherent junctions (AJs), which play a key role in maintaining cellular polarity, facilitating the transport of nutrients and waste products, and enabling the formation of distinct tissue compartments with unique chemical environments[26]. Emerging evidence has shown that airway barrier dysfunction and increased barrier permeability are common characteristics of infected cells during respiratory bacterial and viral infection[27-29]. As a result, pathogens and other external molecules such as allergens are better able to penetrate into the airways sub epithelial tissues [30,31]. In the present study, our data add significantly to previous findings suggesting that RSV infection at the neonatal stage of development may exert long-term effects on the airways epithelial barrier at least partly by regulating the expression and structural localization of epithelial junctional proteins, which is in turn dependent on the IL-33/ST2/MyD88 signaling axis. RSV, a common respiratory pathogen detected in infants and young children, is responsible for a large burden of severe lower respiratory tract infection illness such as bronchiolitis and pneumonia. In the clinic, most infants with RSV infection develop a mild, self-limiting illness, which is usually managed in the outpatient setting, with recovery in the ensuing 1-2 weeks. There remains, however, a substantial proportion of patients (25-40%) who develop severe RSV disease which may require hospitalization[32]. Severe RSV infection in infants induces bronchiolitis, interstitial pneumonitis and sometimes alveolitis. The peribronchiolar infiltrates include primarily macrophages/monocytes, lymphocytes, neutrophils and occasionally eosinophils 28 . In the present study we have developed a murine surrogate of neonatal RSV infection in which the peak of viral replication in the airways was detected 10 days post-infection (dpi), with viral clearance achieved within 3 weeks after inoculation. Evidence from our lung histological and inflammatory infiltration analysis indicates that, in this surrogate, peribronchial and perivascular inflammation may be observed as early as 3dpi and peaks at 10 dpi. Following RSV infection, a rapid recruitment of NK cells and neutrophils to the airways was detectable as early as 3 dpi, later followed by CD4 + and CD8 + T cells and limited numbers of eosinophils within 10 dpi. In addition, local production of a significant amount of both IFN-γ and IFN-α was observed in the early stages of infection (3 dpi), followed by the release of other pro-inflammatory cytokines such as TNF-α, IL-1β and IL-6, which are the main effectors responsible not only for facilitating viral clearance, but also inducing host tissue damage and pathology. It is worthy of note that the murine surrogate of neonatal RSV infection we have established in the present study recapitulates many of the pathologies associated with RSV infection in infants. Recently, Carrie-Anne Malinczak and colleagues in a study of the long-term effects of pulmonary viral infections in mice developing a decline in lung function first manifested a defect in alveolar development approximately 5 weeks after neonatal RSV infection[27]. In the present study we noted a similar time course of structural alterations of the epithelial barrier first evident from 10 and persisting up to 35 days following RSV infection. The associated loss of expression of epithelial junctional proteins such as E-cadherin, ZO-1 and Occludin, although partially restored during the later phase of infection, remained significantly reduced. Based on these findings, it has been suggested that neonatal RSV infection may result in long-term airway barrier dysfunction through regulation of the expression of junctional proteins. Rezaee et al. in contrast proposed that RSV-related epithelial barrier injury may result from rearrangements and internalization of the tight junctions rather than the expression and total quantity of AJC proteins[33]. Furthermore, other studies have conversely reported induction of tight junction molecules such as claudin-4 and Occludin along with elevated local production of pro-inflammatory cytokines such as IL-8 and TNF-α in human airways nasal epithelial cells following RSV infection[34,35]. Such discrepancies suggest that the expression of junctional proteins may be dependent on multiple variables which remain to be defined. Previous studies have confirmed abundant, local release of IL-33 into the airways of infants in association with severe RSV infection[36,37]. Our present data firmly support the hypothesis that neutralization of the local effects of IL-33 on the airways epithelium in neonates in association with RSV infection may exert long-term effects on the responsiveness of the epithelium to subsequent exposure to environmental pathogens, and other environmental insults. Similarly, other studies suggest that early-life RSV infection may result in persistent upregulation of IL-33 expression secondary to epigenetic changes in the promoter of the IL-33 gene in the alveolar epithelial cells[38]. Similarly, Vittoria Palmieri’s work demonstrated that IL-33 is involved in promoting the long-term enhancement of gut mucosal permeability[39]. In the present study we observed that IL-33 production was induced both in RSV-infected mice in vivo and in bronchial epithelial cell supernatants after exposure to RSV in vitro . In vitro RSV infection in the presence of IL-33 resulted in a reduction in the expression of TJ and AJ proteins, which was accompanied by the overexpression of MyD88 and NF-κB signaling activation. Blockage of endogenous IL-33 signaling (via deletion of the St2 gene) or pharmacological inhibition of MyD88 prior to RSV exposure attenuated the RSV-induced epithelial barrier damage, consistent with a critical role of the IL-33/ST2/Myd88 axis signaling in regulating the integrity of epithelial barrier. In addition to its role as an extracellular cytokine transducing signaling by binding to its receptor ST2, IL-33 can also function as a nuclear transcription factor, increasing NF-κB p65 mRNA expression by binding to the NF-κB p65 promoter in endothelial cells[40]. A role for NF-kB in regulating alveolar barrier function was demonstrated by Lee and colleagues, who showed that the overall expression of claudin-18 was reduced in alveolar epithelial cells following stimulation with multiple pro-inflammatory cytokines such as TNF-α, IL-1β and IFN-γ, an effect partially abrogated by prior treatment with the IkB Kinase inhibitor BMS-345541. However, an excessive and inappropriate inhibition of NF-kB pathway may result ultimately to pulmonary edema, even in an otherwise unperturbed lung. Thus, the authors proposed that the maintenance of constitutive NF-kB signaling pathway functions at baseline is crucial in promoting formation of alveolar tight junction[41]. Likewise, Nenci and colleagues reported that complete inhibition of classical NF-kB signaling in NEMO (NF-κB essential modulator) knockout mice caused chronic intestinal inflammation, suggesting a constitutive role for NF-kB in promoting a healthy gut epithelium[42]. Taken together, these findings strongly suggest that NF-κB may not function as a simple on/off switch in regulating the expression of junctional proteins. Although we did observe massive release of IL-33 and MyD88-mediated NF-κB signaling activation in the RSV-related pathological process, the question whether or not IL-33 enters the nucleus and exerts a direct role in regulating genetic expression of junctional protein in the course of RSV infection remains unresolved, necessitating further investigation in future studies. In conclusion, our findings clearly demonstrate dynamic variation in the airway epithelial integrity in neonatal RSV-infected mice, and that such infection induces long-term alteration of the structure and composition of airway epithelial junctions, associated with increased activation of IL-33/St2/MyD88 signaling. We believe that our findings will lay the groundwork for future investigations to design interventional or therapeutic approaches to treat the acute and chronic sequelae of RSV infection. Data availability Data generated from the current study are available from the corresponding author upon reasonable request.

We are grateful for financial support from the National Natural Science Foundation of China (82471813, 82350710228, 82071805, 81971510), and the Beijing Natural Science Foundation (7232055, IS23094), R&D Program of Beijing Municipal Education Commission (KZ20231002541), Chinese Institutes for Medical Research, Beijing (CX24PY02). Table 1 Cell surface markers used to define inflammatory cells | CD4 + Ths | CD45 + CD3 + CD4 + | | CD8 + Ths | CD45 + CD3 + CD8 + | | NK | CD45 + CD3 - NK1.1 + | | Neutrophils | CD45 + CD3 - CD11b + Ly6G + Siglec-F − | | Eosinophils | CD45 + CD3 - CD11b + Ly6G - Siglec-F + | Table 2 Antibodies employed in the present study | CD45 | BV650 | 103151 | Biolegend | | NK1.1 | Super Bright 600 | 63-5941-82 | Invitrogen | | Ly6G | Alexa Fluor™ 700 | 56-9668-82 | Invitrogen | | CD3 | APC | 555335 | BD Biosciences | | CD8 | PE | 553032 | BD Biosciences | | CD11b | Percp-Cy5.5 | 561114 | BD Biosciences | | Siglec-F | BV421 | 565934 | BD Biosciences | Table 3 List of primer sequences used in this study | rsv-l | GAACTCAGTGTAGGTAGAATGTTTGCA | TTCAGCTATCATTTTCTCTGCCAAT | | Il33 | GGTGTGGATGGGAAGAAGCTG | GAGGACTTTTTGTGAAGGAC G | | tslp | CTGAGAGAAATGACGGTACTCAGG | GGAGATTGCATGAAGGAATACCAC | | Il25 | ACAGGGACTTGAATCGGGTC | TGGTAAAGTGGGACGGAGTTG | | gapdh | ACGGCCGCATCTTCTTGTGCA | AATGGCAGCCCTGGTGACCA | Table 4List of antibodies and dilutions used for immunocytochemistry and immunofluorescence analysis | E-cadherin | Cell signaling technology | #3195 | 1:300 (IF) 1:1000 (WB) | | ZO-1 | Proteintech | 21773-1-AP | 1:100 (IF) 1:1000 (WB) | | Occludin | Proteintech | 66378-1-Ig | 1:300 (IF) 1:2000 (WB) | | Myd88 | Proteintech | 67969-1-Ig | 1:1000 (WB) | | ST2 | Abcam | ab25877 | 1:2000 (WB) | | GAPDH | Sigma-Aldrich | G9545 | 1:10000 (WB) | | Phospho-NF-κB p65 (Ser536) | Cell signaling technology | #3033 | 1:1000 (WB) | | NF-κB p65 | Cell signaling technology | #8242 | 1:1000 (WB) | | Goat anti-Rabbit IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 488 | Thermo Fisher | A-11008 | 1:500 (IF) | | Goat anti-Mouse IgG (H+L) Secondary Antibody, HRP | Thermo Fisher | 31430 | 1:1000 (IHC) 1:10000 (WB) | | Goat anti-Rabbit IgG (H+L) Secondary Antibody, HRP | Thermo Fisher | 31460 | 1:1000 (IHC) 1:10000 (WB) | | (IF: immunofluorescence staining; IHC: Immunohistochemistry staining; WB: Western blot) |

1. Battles MB, McLellan JS. Respiratory syncytial virus entry and how to block it. Nature reviews Microbiology. 2019;17(4):233-245.2. Li Y, Wang X, Blau DM, et al. Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in children younger than 5 years in 2019: a systematic analysis. Lancet (London, England). 2022;399(10340):2047-2064.3. Li M, Cong B, Wei X, et al. Characterising the changes in RSV epidemiology in Beijing, China during 2015-2023: results from a prospective, multi-centre, hospital-based surveillance and serology study. The Lancet regional health Western Pacific. 2024;45:101050.4. Li ZJ, Zhang HY, Ren LL, et al. Etiological and epidemiological features of acute respiratory infections in China. Nature communications. 2021;12(1):5026.5. Rosas-Salazar C, Chirkova T, Gebretsadik T, et al. Respiratory syncytial virus infection during infancy and asthma during childhood in the USA (INSPIRE): a population-based, prospective birth cohort study. Lancet (London, England). 2023;401(10389):1669-1680.6. Griffiths C, Drews SJ, Marchant DJ. Respiratory Syncytial Virus: Infection, Detection, and New Options for Prevention and Treatment. Clinical microbiology reviews. 2017;30(1):277-319.7. Celebi Sözener Z, Cevhertas L, Nadeau K, Akdis M, Akdis CA. Environmental factors in epithelial barrier dysfunction. Journal of Allergy and Clinical Immunology. 2020;145(6):1517-1528.8. Veazey JM, Eliseeva SI, Hillman SE, et al. Inhibiting Protein Kinase D Promotes Airway Epithelial Barrier Integrity in Mouse Models of Influenza A Virus Infection. Frontiers in immunology. 2020;11:580401.9. Rezaee F, Georas SN. Breaking barriers. New insights into airway epithelial barrier function in health and disease. Am J Respir Cell Mol Biol. 2014;50(5):857-869.10. Smallcombe CC, Linfield DT, Harford TJ, et al. Disruption of the airway epithelial barrier in a murine model of respiratory syncytial virus infection. American journal of physiology Lung cellular and molecular physiology. 2019;316(2):L358-l368.11. Kast JI, McFarlane AJ, Głobińska A, et al. Respiratory syncytial virus infection influences tight junction integrity. Clin Exp Immunol. 2017;190(3):351-359.12. Li W, Liu M, Chu M. Strategies targeting IL-33/ST2 axis in the treatment of allergic diseases. Biochemical pharmacology. 2023;218:115911.13. Cayrol C, Girard JP. Interleukin-33 (IL-33): A critical review of its biology and the mechanisms involved in its release as a potent extracellular cytokine. Cytokine. 2022;156:155891.14. Saluzzo S, Gorki AD, Rana BMJ, et al. First-Breath-Induced Type 2 Pathways Shape the Lung Immune Environment. Cell reports. 2017;18(8):1893-1905.15. Iijima K, Kobayashi T, Matsumoto K, Ohara K, Kita H, Drake LY. Transient IL-33 upregulation in neonatal mouse lung promotes acute but not chronic type 2 immune responses induced by allergen later in life. PloS one. 2021;16(5):e0252199.16. Liu H, Wu M, Wang Q, et al. IL-33 released during challenge phase regulates allergic asthma in an age-dependent way. Cellular & molecular immunology. 2024.17. Qi F, Wang D, Liu J, et al. Respiratory macrophages and dendritic cells mediate respiratory syncytial virus-induced IL-33 production in TLR3- or TLR7-dependent manner. International immunopharmacology. 2015;29(2):408-415.18. Saravia J, You D, Shrestha B, et al. Respiratory Syncytial Virus Disease Is Mediated by Age-Variable IL-33. PLoS Pathog. 2015;11(10):e1005217.19. McKimm-Breschkin JL. A simplified plaque assay for respiratory syncytial virus–direct visualization of plaques without immunostaining. Journal of virological methods. 2004;120(1):113-117.20. Wang JH, Sun LH, Huang SK, Chen AH. Effect of co-culturing human primary basic fibroblasts with respiratory syncytial virus-infected 16-HBE cells. Genetics and molecular research : GMR. 2016;15(1).21. Du X, Li Y, Xu Y, et al. Airways epithelial exposure to Streptococcus pneumoniae in the presence of the alarmin IL-33 induces a novel subset of pro-inflammatory ILC2s promoting a mixed inflammatory response. Inflammation research : official journal of the European Histamine Research Society [et al]. 2024;73(7):1239-1252.22. Van Hoecke L, Job ER, Saelens X, Roose K. Bronchoalveolar Lavage of Murine Lungs to Analyze Inflammatory Cell Infiltration. Journal of visualized experiments : JoVE. 2017(123).23. Peng D, Shi Y, Pang J, et al. Early-life infection of the airways with Streptococcus pneumoniae exacerbates HDM-induced asthma in a murine model. Cell Immunol. 2022;376:104536.24. Cui L, Qin X, Fu T, et al. Attenuated airways inflammation and remodeling in IL-37a and IL-37b transgenic mice with an ovalbumin-induced chronic asthma. Cell Immunol. 2023;391-392:104759.25. Deguine J, Barton GM. MyD88: a central player in innate immune signaling. F1000prime reports. 2014;6:97.26. Gao N, Rezaee F. Airway Epithelial Cell Junctions as Targets for Pathogens and Antimicrobial Therapy. Pharmaceutics. 2022;14(12).27. Malinczak CA, Fonseca W, Hrycaj SM, et al. Early-life pulmonary viral infection leads to long-term functional and lower airway structural changes in the lungs. American journal of physiology Lung cellular and molecular physiology. 2024;326(3):L280-l291.28. Chiu CC, Shi YF, Yang JJ, Hsiao YC, Tzang BS, Hsu TC. Effects of human Parvovirus B19 and Bocavirus VP1 unique region on tight junction of human airway epithelial A549 cells. PloS one. 2014;9(9):e107970.29. Takeichi M. Dynamic contacts: rearranging adherens junctions to drive epithelial remodelling. Nature reviews Molecular cell biology. 2014;15(6):397-410.30. Hakansson AP, Orihuela CJ, Bogaert D. Bacterial-Host Interactions: Physiology and Pathophysiology of Respiratory Infection. Physiological reviews. 2018;98(2):781-811.31. Eisele NA, Anderson DM. Host Defense and the Airway Epithelium: Frontline Responses That Protect against Bacterial Invasion and Pneumonia. Journal of pathogens. 2011;2011:249802.32. Piedimonte G, Perez MK. Respiratory syncytial virus infection and bronchiolitis. Pediatrics in review. 2014;35(12):519-530.33. Rezaee F, Harford TJ, Linfield DT, et al. cAMP-dependent activation of protein kinase A attenuates respiratory syncytial virus-induced human airway epithelial barrier disruption. PloS one. 2017;12(7):e0181876.34. Colpitts CC, Baumert TF. Claudins in viral infection: from entry to spread. Pflugers Archiv : European journal of physiology. 2017;469(1):27-34.35. Masaki T, Kojima T, Okabayashi T, et al. A nuclear factor-κB signaling pathway via protein kinase C δ regulates replication of respiratory syncytial virus in polarized normal human nasal epithelial cells. Mol Biol Cell. 2011;22(13):2144-2156.36. García-García ML, Calvo C, Moreira A, et al. Thymic stromal lymphopoietin, IL-33, and periostin in hospitalized infants with viral bronchiolitis. Medicine. 2017;96(18):e6787.37. Vu LD, Siefker D, Jones TL, et al. Elevated Levels of Type 2 Respiratory Innate Lymphoid Cells in Human Infants with Severe Respiratory Syncytial Virus Bronchiolitis. American journal of respiratory and critical care medicine. 2019;200(11):1414-1423.38. Morris SB, Ocadiz-Ruiz R, Asai N, et al. Long-term alterations in lung epithelial cells after EL-RSV infection exacerbate allergic responses through IL-1β-induced pathways. Mucosal immunology. 2024;17(5):1072-1088.39. Palmieri V, Ebel JF, Ngo Thi Phuong N, et al. Interleukin-33 signaling exacerbates experimental infectious colitis by enhancing gut permeability and inhibiting protective Th17 immunity. Mucosal immunology. 2021;14(4):923-936.40. Choi YS, Park JA, Kim J, et al. Nuclear IL-33 is a transcriptional regulator of NF-κB p65 and induces endothelial cell activation. Biochemical and biophysical research communications. 2012;421(2):305-311.41. Ward C, Schlingmann B, Stecenko AA, Guidot DM, Koval M. NF-κB inhibitors impair lung epithelial tight junctions in the absence of inflammation. Tissue Barriers. 2015;3(1-2).42. Nenci A, Becker C, Wullaert A, et al. Epithelial NEMO links innate immunity to chronic intestinal inflammation. Nature. 2007;446(7135):557-561. Figure Legends Figure 1. Dynamic changes in viral load and pathological changes in neonatal mice following pernasal infection with RSV (A) The protocol to establish a murine model of neonatal RSV infection and the time points of specimen sampling. (B) Viral copy numbers in lung homogenates of control and RSV-infected animals at indicated time points after viral inoculation as determined by quantitative RT-PCR. (C) Left: Representative photomicrographs of H&E stained lung tissue sections of RSV-infected animals at indicated time points. Right: semi-quantitative scoring of the severity of inflammatory infiltration of the airway. (D) Left: Representative photomicrographs of PAS stained sections of mouse airways. Right: Mucus score based on PAS staining. (E) Immunoreactivity for TJ (ZO-1 and Occludin) and AJ (E-cadherin) proteins in the sections of lung tissues at various time points following RSV exposure. Left: Representative photomicrographs showing immunoreactivity for E-cadherin, ZO-1 and Occludin (brown) in sections of the lung tissue. Right: Quantitative analysis of positively stained areas of the lung tissue sections. Data are presented as the mean ± SEM (n=6 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001 shown comparing the mock-infected with the RSV-infected groups. Figure 2. Pulmonary cellular influx and pro-inflammatory cytokine concentrations in the lung lysates following pernasal infection with RSV (A) Gating strategy for the identification of inflammatory cell populations in airways and lung tissue. (B) Total and differential inflammatory cell counts in the lung tissue at indicated time points following RSV infection. (C) Concentrations of IFN-α, IFN-γ, TNF-α, IL-6, IL-1β, IL-33, IL-25 and TSLP per mg of lung tissue homogenates at indicated time points following RSV infection, as determined by ELISA. Values are expressed as mean ± SEM (n=6 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001 shown comparing the mock-infected with the RSV-infected groups. Figure 3. In-vitro administration of IL-33 results in disruption of airways integrity (A) The RSV fusion protein (green) was detected in the intracellular space between 16HBE cells by immunofluorescence staining. Nuclei were counterstained with DAPI. (B) Changes in the viral RNA expression in 16HBE cells at the indicated time points following mock infection or RSV exposure were determined a RT-PCR assay. (C-D) Changes in expression of IL-33, IL-25 and TSLP in 16HBE cells at the indicated time points following RSV or mock infection examined at the levels of both mRNA and protein. (E) Left: The structure of the AJC was examined by immunostaining with antibodies against the E-cadherin, ZO-1 and Occludin proteins (red). Right: Statistical analysis of mean fluorescent intensity of E-cadherin, ZO-1 and Occludin staining following the indicated exposures. (F) Western blot and statistical analysis of fold changes in E-cadherin, ZO-1 and Occludin expression in human epithelial cells exposed to RSV in the presence or absence of IL-33. Data are presented as the mean ± SEM (n=5–6 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001. Figure 4. Deletion of the St2 gene attenuated RSV-induced epithelial damage (A) Comparison of the pulmonary inflammatory responses in wild type and St2 −/− mice at day 10 following RSV or mock infection. Left: Representative photomicrographs of H&E staining of lung tissues from wild type and St2 −/− mice. Right: semi-quantitative scoring of the severity of inflammatory infiltrates in the airways. (B) Left: Representative photomicrographs showing immunoreactivity for junctional proteins (ZO-1, Occludin and E-cadherin) (brown) in the sections of lung tissues from wild type and St2 −/− mice at day 10 following RSV or mock infection. Right: Quantitative analysis of positive staining area in lung tissue. (C) Left: Representative immunoblots of ZO-1, Occludin, E-cadherin, MyD88, p-NF-κB p65 and total NF-κB proteins in lung homogenates from RSV- or mock-infected wild type and St2 −/− mice. Right: Statistical analysis of relative expression of these proteins normalized to that of GAPDH. Data are presented as the mean ± SEM (n=5 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001. Figure 5. Silencing of the St2 gene attenuated RSV-induced epithelial damage through the IL-33/ST2/MyD88 signaling axis (A) Left: Representative confocal staining of AJC proteins (ZO-1 and Occludin and RSV) (red) in RSV-infected 16HBE cells treated with exogenous IL-33 following transfection with siRNA encoding St2 or control. Right: Statistical analysis of mean fluorescent intensity of the stained proteins in the 4 groups. (B) Left: Western blot analysis of ZO-1, Occludin, E-cadherin, MyD88, p-NF-κB p65 and total NF-κB protein expression in RSV-infected 16HBE cells treated with exogenous IL-33 following transfection with siRNA encoding St2 or control. Right: Statistical analysis of the relative expression of these proteins normalized to that of GAPDH. Data are presented as the mean ± SEM (n=5 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001. Figure 6. Prior treatment with MyD88 inhibitor attenuated RSV-induced epithelial damage (A) 16HBEs were pre-incubated with a MyD88 inhibitor (TJ-5) or DMSO diluent control for 2 h, then infected with RSV in the presence or absence of exogenous IL-33 for a further 72 h. Left: Representative confocal images of E-cadherin, ZO-1 and Occludin staining (red) in the different groups. Right: statistical analysis of the mean intensity of fluorescent staining of these proteins. (B) Effects of TJ-5 treatment on the expression of E-cadherin, ZO-1 and Occludin in the RSV-infected 16HBE cells in the presence or absence of IL-33 were examined by western blotting analysis. Right: Densitometric quantification of at least 5 independent experiments. Data are presented as the mean ± SEM (n=5 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001. Supplementary Figure 1. Comparisons of viral load recovered from RSV-infected 16HBE in the presence and absence of IL-33 The viral copy numbers in Mock-infected, RSV-infected 16HBE, RSV-infected 16HBE in combination with exogenous IL-33 treatment at indicated time points were measured by quantitative RT-PCR. Data are presented as the mean ± SEM (n=6 in each group). * P < 0.05, ** P < 0.01, *** P < 0.001 shown comparing the mock-infected with the RSV-infected groups. Supplementary Figure 2. Validation of the knockdown efficiency of St2 (A) The mRNA expression of St2 in 16HBE transfection with NC-siRNA or St2 -siRNA was determined by RT-PCR. (B) Left: Western blot analysis confirming the knockdown of ST2 in St2 -siRNA-16HBE cells. Right: Statistical analysis of the relative expression of these proteins normalized to that of GAPDH. Data are presented as the mean ± SEM (n=5 in each group). ** P < 0.01. Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License. Collection

Authors Metrics & Citations Metrics Article Usage 453views 209downloads Citations Download citation Qing Miao, Rui Yu, Fanyu Shi, et al. Respiratory syncytial virus infection in early life disrupts epithelial barriers via IL-33/ST2/MyD88 signaling axis. Authorea. 28 January 2025. DOI: https://doi.org/10.22541/au.173808583.37593598/v1 DOI: https://doi.org/10.22541/au.173808583.37593598/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.