key: cord-0262900-qwmbdaaf authors: Radzikowska, U.; Eljaszewicz, A.; Tan, G.; Stocker, N.; Heider, A.; Westermann, P.; Steiner, S.; Dreher, A.; Wawrzyniak, P.; Ruckert, B.; Rodriguez-Coira, J.; Zhakparov, D.; Huang, M.; Jakiela, B.; Sanak, M.; Moniuszko, M.; OMahony, L.; Kebadze, T.; Jackson, D. J.; Edwards, M. R.; Thiel, V.; Johnston, S. L.; Akdis, C.; Sokolowska, M. title: Epithelial RIG-I inflammasome activation suppresses antiviral immunity and promotes inflammatory responses in virus-induced asthma exacerbations and COVID-19 date: 2021-11-17 journal: nan DOI: 10.1101/2021.11.16.21266115 sha: 50485f6f94a029f2c3ecef543148935e0931aff0 doc_id: 262900 cord_uid: qwmbdaaf Rhinoviruses (RV) and inhaled allergens, such as house dust mite (HDM) are the major agents responsible for asthma onset, its life-threatening exacerbations and progression to severe disease. The role of severe acute respiratory syndrome coronavirus (SARS-CoV-2) in exacerbations of asthma or the influence of preexisting viral or allergic airway inflammation on the development of coronavirus disease 2019 (COVID-19) is largely unknown. To address this, we compared molecular mechanisms of HDM, RV and SARS-CoV-2 interactions in experimental RV infection in patients with asthma and healthy individuals. RV infection was sensed via retinoic acid-inducible gene I (RIG-I) helicase, but not via NLR family pyrin domain containing 3 (NLRP3), which led to subsequent apoptosis-associated speck like protein containing a caspase recruitment domain (ASC) recruitment, oligomerization and RIG-I inflammasome activation. This phenomenon was augmented in bronchial epithelium in patients with asthma, especially upon pre-exposure to HDM, which itself induced a priming step, pro-IL-1{beta} release and early inhibition of RIG-I/TANK binding kinase 1/I{kappa}B kinase {epsilon}/type I/III interferons (RIG-I/TBK1/IKK{epsilon}/IFN-I/III) responses. Excessive activation of RIG-I inflammasomes was partially responsible for the alteration and persistence of type I/III IFN responses, prolonged viral clearance and unresolved inflammation in asthma. RV/HDM-induced sustained IFN I/III responses initially restricted SARS-CoV-2 replication in epithelium of patients with asthma, but even this limited infection with SARS-CoV-2 augmented RIG-I inflammasome activation. Timely inhibition of the epithelial RIG-I inflammasome and reduction of IL-1{beta} signaling may lead to more efficient viral clearance and lower the burden of RV and SARS-CoV-2 infection. Asthma is one of the most common chronic inflammatory lung diseases affecting more than 5% of the global population 1 . Its pathogenesis and clinical presentation is complex 2 , with a common feature of susceptibility to exacerbations leading to loss of disease control, hospitalizations, and in some cases, progressive loss of lung function 3, 4 . Exacerbations of asthma are most often caused by common respiratory viruses 5, 6 , with rhinoviruses (RV) responsible for up to 80% of asthma attacks 5 . RVs that have been initially considered as benign viruses, now are also linked to the early-life development of asthma, severe bronchiolitis in infants and fatal pneumonia in elderly and immunocompromised patients [7] [8] [9] [10] . Likewise, human coronaviruses have not been strongly linked with asthma pathology 11 . However, the current pandemic of severe acute respiratory syndrome coronavirus (SARS-CoV-2) has been challenging this view, resulting in contradictory observations of asthma being considered a risk factor for SARS-CoV-2 infection and coronavirus disease 2019 (COVID-19) severity [12] [13] [14] or constituting a protection from the disease 15, 16 . Another important factor for asthma development and exacerbations is exposure to inhaled allergens. House dust mite (HDM) is the most significant source of perennial allergens worldwide. HDM sensitization is found in around 50%-85% of patients with asthma, and HDM exposure correlates with asthma severity 17, 18 . There are strong epidemiological links between RV infections, allergen exposure and sensitization on the risk of asthma development and the rates of exacerbations 9, 19 . Children with early life RV-induced wheezing and aeroallergen sensitization have an extremely high incidence of asthma in later years 9 . Combination of virus detection in the airways with the high allergen exposure markedly increases the risk of hospital admission 20 . In line with this, HDM immunotherapy significantly reduces risk of asthma exacerbations 21 . It has been also recently suggested that allergen exposure might influence SARS-CoV-2 infection patterns in the general population 22,23 . However, the underlying mechanisms of these noxious, reciprocal allergen-virus effects in asthma are incompletely understood 24 . The host response to the RV infection encompasses its RNA recognition by the endosomal toll-like receptor 3 (TLR) 3, TLR7/8 and cytoplasmic RNA helicases: retinoic acid-inducible gene I (RIG-I) and melanomadifferentiation-associated gene 5 (MDA5) 25-27 , whereas its capsid might interact with the cell surface TLR2 and initiate myeloid differentiation primary response 88 (MyD88)-dependent nuclear factor `kappa-light-chain-sufficient to clear RV infection in healthy airways 34 . We and others demonstrated several alternations in RVinduced type I/III IFN responses and other antiviral mechanisms in asthma [35] [36] [37] [38] [39] . Despite increased understanding of IFN signaling, no preventive methods or treatments targeting these pathways are available for virus-induced asthma exacerbations, suggesting greater complexity than previously anticipated [40] [41] [42] . RV infection also leads to the release of proinflammatory cytokines, chemokines and growth factors via activation of MyD88/NF-kB pathway 26, 27, 43 . This response is likely beneficial in the early infection phase and necessary for effective viral clearance. However, when it is sustained and excessive, it might lead to tissue damage and unresolved inflammation 44 . HDM also induces expression of proinflammatory mediators in the airway epithelium 24 via activation of MyD88/NF-kB and other transcriptional pathways, leading to an increase in expression of proinflammatory proteins [45] [46] [47] . Expression and release of mature IL-1b needs to be tightly regulated, by the transcriptional activation of pro-IL-1b, called priming, followed by an activation of supramolecular complexes called inflammasomes and release of mature, active forms of IL-1b and/or proinflammatory cell death called pyroptosis 48 . Inflammasomes are composed of at least a sensor protein and caspase-1 and often the adaptor protein apoptosis-associated speck-like protein containing a CARD (ASC) 49 . ASC is recruited by receptors not containing CARD domains, like NLR family pyrin domain containing 3 (NLRP3), to start oligomerization and further recruitment and activation of pro-caspase 1, via CARD-CARD interactions 49 . In the case of RIG-I (containing two CARD domains), ASC is recruited to couple and enhance caspase-1 recruitment and mature IL-1b 50 production, whereas MAVS interaction with CARD9 is responsible for NF-kB activation and transcriptional priming of pro-IL-1b 50 . Activation of NLRP3 as well as RIG-I inflammasome has been demonstrated in the context of some respiratory RNA viruses, including RV 51-53 , influenza A (IAV) 52,54,55 , SARS-CoV-1 56,57 and most recently SARS-CoV-2 58,59 with major differences depending on the hematopoetic 50,51 or epithelial cellular origin 52 and their differentiation state 53 . It remains unclear whether activation of airway epithelial inflammasomes is necessary to clear infection or in contrast, whether it initiates mucosal hyperinflammation delaying virus clearance 60,61 . Additionally, an involvement of NLRP3 priming and/or inflammasome activation in HDM-models of asthma and in severe asthma in humans has been demonstrated, however data are conflicting and remain poorly understood 46,62-64 . Finally, little is known about the airway epithelial response in health or during the preexisting HDM-induced inflammation in asthma and combined infection with RV and SARS-CoV-2. Therefore, in the current study, we investigated the effects of HDM, RV and SARS-CoV-2 on differentiated primary human bronchial epithelium in vitro and in experimental in vivo RV infection in healthy subjects and in patients with asthma. We identified that RV infection and replication activated the RIG-I/ASC, but not the NLRP3, inflammasome, which was further augmented in the presence of HDM. Activation of the RIG-I inflammasome was stronger in patients with asthma and was partially responsible for disturbed RIG-I-especially in the presence of HDM exposure. NLRP3 and MDA5 inflammasomes were not involved in the responses to RV-A16 infection or HDM exposure in differentiated human primary bronchial epithelial cells. To evaluate whether the enhanced epithelial RIG-I inflammasome activation in response to RV infection might have further implications on the overall inflammatory responses at the bronchus barrier sites in asthma, we investigated inflammasome-and IL-1b-mediated immune responses. Using targeted proteomics, we found that in addition to IL-1b, also IL-18, IL-1a, tumor necrosis factor (TNF) and TNF-related activation-induced cytokine (TRANCE) were released 24h after RV-A16 infection in HBECs from both control individuals and patients with asthma (Fig. 3A ). Monocyte chemoattractant protein (MCP)-1 was significantly upregulated only in asthma, while MCP-3 was significantly upregulated only in controls, but overall, we observed rather similar changes in the expression of the analyzed proinflammatory proteins in HBECs from both patients with asthma and control individuals 24h after infection. Therefore, to evaluate if the enhanced IL-1b release and stronger RV-A16 infection in patients with asthma might influence an even broader range of responses, we analyzed publicly available next-generation sequencing (NGS) raw data of RV-A16-infected HBECs 66 . An unbiased analysis of pathways and ontologies, revealed upregulated interferon signaling and innate immune responses to RNA viral infections (Fig. 3B , Suppl. Table S1 ). We also noted a significant enrichment of inflammasomemediated immune responses, in both -control and asthma samples, but it ranked slightly higher in patients with asthma (Fig. 3B) . Accordingly, epithelium from both patients with asthma and control individuals showed increased inflammasome-mediated immune responses after RV-A16 infection (Fig. 3C, D) , but the inflammasome-related molecules, cytokines and chemokines, such as CASP1 (caspase-1), IL6, NLR family CARD domain containing 5 (NLRC5), CXCL1 and others were significantly more upregulated in asthma (Fig. 3E ). In agreement with the NGS results, we also observed that infection with RV-A16 increased expression of NLRC5 and CASP1 (caspase-1), which was significantly more enhanced in patients with asthma than control individuals (Fig. 3F) . In summary, we demonstrated here that early after RV-A16 infection, other inflammasome-and IL-1b-mediated immune responses are boosted in both groups, controls and patients with asthma, but in case of several inflammasome-related molecules, cytokines and chemokines this increase is much more pronounced in patients with asthma. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint Next, we analyzed samples from a previously reported 67 controlled, experimental RV-A16 infection study of patients with asthma and control individuals, including BAL fluid, bronchial biopsies and bronchial brushings taken two weeks before infection (d-14, baseline) and 4 days after infection (Fig. 4A ). Both groups were seronegative for anti-RV-A16 antibodies prior to infection and only individuals without recent natural respiratory infection underwent the experimental infection, as described previously 67 . Quite strikingly, we found a strong upregulation of the same transcriptome profiles of inflammasome-mediated immune responses in bronchial brushings from patients with asthma, in a sharp contrast to the downregulation of similar genes in control individuals at the same time (4 days after infection) (Fig. 4B , C, S3A, B). We further validated the expression of IL-1b, caspase-1, and RIG-I proteins in bronchial biopsies from the same patients. In line with the results demonstrated above, we found higher expression of IL-1b in the epithelial area of the bronchial biopsies of patients with asthma at baseline as compared to healthy controls (Fig. 4D ). In accordance with its gene expression in bronchial brushings, in healthy controls, we noted a decrease in IL-1b expression following RV-A16 infection (Fig. 4D) . Additionally, we assessed concentrations of mature IL-1b protein secreted into the bronchoalveolar lavage fluid. In agreement with the epithelial mRNA and protein expression of IL-1b, we found significantly decreased IL-1b protein concentration in BAL fluid from control individuals 4 days after infection, whereas in patients with asthma IL-1b protein concentrations in BAL fluid tended to be increased, though this increase was not statistically significant (Fig. 4E) . Likewise, we also noted downregulation of epithelial expression of caspase-1 in bronchial biopsies from healthy controls 4 days after infection, whereas it did not change in patients with asthma (Fig. 4F, S3C ). Here we also noticed that in vivo RIG-I protein expression in bronchial biopsies at baseline was higher in healthy controls than in patients with asthma, confirming our above-demonstrated in vitro data (Fig. S3D ). After RV-A16 infection, there was a trend towards an increase in RIG-I expression in bronchial biopsies from patients with asthma, and towards a decrease in controls, but those trends were not statistically significant. Next, we investigated the dynamics and relationship of inflammasome-mediated immune responses after in vitro, and in vivo RV-A16 infection in patients with asthma and control subjects. We observed a strong negative correlation between gene expression of inflammasome-mediated immune responses after infection at 24h in vitro and at 4 days in vivo in control individuals (Fig. 4G, left panel) . Genes upregulated in vitro at 24h post infection, were downregulated or not changed anymore at 4 days after in vivo infection. In contrast, in patients with asthma there was a strong positive correlation between the expression of similar genes at 24h in vitro and 4 days in vivo (Fig. 4G, right panel) showing that genes already strongly upregulated in asthma at an early time point after infection, still stayed significantly upregulated in vivo 4 days after infection. Altogether, combining the data from in vivo and in vitro approaches in humans, we demonstrated that the RIG-I inflammasome is indeed activated in human bronchial epithelium after in vivo infection with RV-A16. Moreover, our data revealed that All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint RIG-I inflammasome-mediated inflammation in healthy individuals was either being actively suppressed or already resolved 4 days after infection, whereas it was still ongoing in asthma. Since the major function of RIG-I is recognition of RNA viruses 30 , we also analyzed the status of antiviral genes and proteins involved in in vivo and in vitro responses to RV-A16 infection. In line with inflammasomemediated immune responses, the majority of genes encoding antiviral pathways were still upregulated 4 days after in vivo RV-A16 infection in patients with asthma while they were either downregulated or not changed in healthy controls (Fig. 5A , B, Supplementary Table S2 ). These data suggest less effective resolution of RV-A16 infection and delayed clearance of the virus in asthma. Indeed, RV-A16 load in the bronchoalveolar lavage fluid in asthma was around 100-fold higher than in controls and the peak nasal lavage virus load was 25-fold higher in patients with asthma than in healthy controls 67 , though this difference was not statistically significant here that bronchial epithelium from healthy individuals can efficiently respond to RV-A16 infection which leads to rapid virus clearance and subsequent resolution of antiviral responses. In contrast, in asthma, the lack of resolution of antiviral responses and delayed virus clearance suggest that there is an ongoing process in epithelium, which impairs the effectiveness of RIG-I-induced antiviral mechanisms. Having demonstrated excessive RIG-I inflammasome activation and IL-1b secretion in epithelium of patients with asthma in response to RV-A16, we hypothesized that this could result in persistent, but less efficient anti-RV-A16 response in asthma. Therefore, we further studied whether RIG-I activation of MAVS/TBK1/IKKe and downstream interferon signaling inhibits RIG-I inflammasome activation and conversely if formation of RIG-I All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint inflammasome inhibits interferon signaling. First, we used BX795, a chemical inhibitor of TBK1 and IKKe. As expected, it blocked expression of IFNL2/3 (IFN-l) in HBECs of patients with asthma ( Fig. 6A ) and subsequently decreased expression of DDX58 (RIG-I) (Fig. 6B ). BX795 treatment also reduced expression of interferonstimulated chemokines: CXCL10, CXCL11 and CCL3 (Fig. 6C ). It also led to a trend to increased RV-A16 infection ( Fig. 6D) , as well as to significantly augmented inflammasome priming (Fig. 6E ) and activation (Fig. 6F ). Next, we blocked IL-1b processing by RIG-I inflammasome with the use of the caspase-1 inhibitor YVAD and investigated interferon signaling and RV-A16 infection in bronchial epithelium. Inhibition of RIG-I inflammasome activation and subsequent IL-1b signaling indeed led to trends of increasing expression of IFNB (IFN-b) (Fig. 6G ) and DDX58 (RIG-I) mRNA (Fig. 6H) and it further increased the production of CXCL10, CXCL11, CCL3, and CCL4 (Fig. 6I) . However, it did not change the infection at the same early timepoint (Fig. S1L ). In summary, these data suggest that increased RV-A16-dependent RIG-I inflammasome activation in bronchial epithelium disturbed the effectiveness of RIG-I dependent anti-RV-A16 responses in asthma. Therefore, a timely blockade of the excessive RIG-I inflammasome activation and IL-1b signaling may lead to more efficient viral clearance and lower burden of infection. Having demonstrated that HDM increased RV-A16-induced RIG-I inflammasome activation, we continued to explore the effect of HDM pre-exposure on the timing and strength of antiviral responses. We found that HDM pre-treatment decreased RV-A16-induced mRNA expression of IFNB (IFN-b) and DDX58 (RIG-I) only in HBECS from patients with asthma 24h, but not 6h after RV-A16 infection in vitro (Fig. 7A , B, S4A, B). Accordingly, when we analyzed protein expression and enriched biological pathways by targeted proteomics in the same conditions, we observed decreased cell-and IFNs-mediated antiviral responses in HBECs from control individuals and patients with asthma at 24h post infection (Fig. 7C ). In addition, HDM in the presence of RV- Table S3) . Importantly, HDM pre-stimulation had a slightly additive effect to the antiviral RIG-I pathway inhibitor (BX795) and further reduced BX795-decreased protein expression of the ISGs: CXCL10, CXCL11, CCL3 and CCL4 (Fig. 7D) . It all suggests that pre-exposure to HDM, before RV-A16 infection decreases IFN type I response in a non-specific way in patients with asthma and control individuals. However, only in patients with asthma, HDM exposure contributes further to the enhanced inflammasome-mediated and other immune responses, and associated impairment of the effectiveness of antiviral responses. Finally, facing the current pandemic and noting the contradictory results about asthma as a risk factor for COVID-19 in different populations [12] [13] [14] [15] [16] , we investigated if RV-A16-induced RIG-I inflammasome activation and HDM-mediated decrease of IFN responses may affect SARS-CoV-2 infection. We first treated primary HBECs from healthy controls and patients with asthma with or without HDM, next after 24h we infected them with RV-A16, and after a further 24h we infected them with SARS-CoV-2 for 48h (Fig. 8A) . We confirmed infection with SARS-CoV-2 by the detection of its nucleocapsid protein N (Fig. 8B ) and the increase of SARS-CoV-2 viral RNA (Fig. 8C ). In patients with asthma, but not in healthy controls, we observed lower infection with SARS-CoV-2 in samples pre-infected with RV-A16 (Fig. 8C ). Individual samples with high RV-A16 virus loads (Fig. 8D) had lower SARS-CoV-2 infection, and vice versa. This effect was diminished upon HDM pre-stimulation in samples from patients with asthma, suggesting that HDM pre-stimulation tended to increase SARS-CoV-2 infection in RV-A16+SARS-CoV-2 infected epithelium in asthma ( Table S4 ) and IL1B (IL-1b) (Fig. 8K, S5E) . Notably, HDM pre-stimulation resulted in an amplified secretion of proinflammatory proteins, such as TRAIL, IL-15, CXCL9, IL-17C and CCL8 only in epithelium of patients with asthma infected with both viruses, whereas these conditions induced IL-18 secretion both in patients with asthma and control individuals ( Table S4 ). In summary, we found here that pre-existing RV-A16 infection restricted SARS-CoV-2 replication in asthma, but not in controls, which was in line with the sustained type I/III IFNs at this time-point, induced by RV-A16 in asthma. SARS-CoV-2 infection alone did not induce DDX58 (RIG-I), IFIH1 (MDA5), IFNB (IFN-b), IFNL1 (IFN-l1) at 48h after infection. However, especially in the presence of HDM, we observed enhanced All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint proinflammatory responses in patients with asthma after SARS-CoV-2 co-infection with RV-A16, in spite of the reduced viral load of SARS-CoV-2. In the current study, we investigated mechanisms, interactions and dynamics of inflammasomes, antiviral and proinflammatory responses in the airways of healthy controls and patients with asthma upon combined exposures with house dust mite, rhinovirus and SARS-CoV-2. We analyzed the direct in vivo responses in humans, paired with the experiments in patients` primary cells. We showed, that recognition of replicating RV in bronchial epithelial cells by RIG-I helicase led to ASC recruitment, ASC oligomerization, activation of caspase-1, processing and release of mature IL-1b, independently of NLRP3 inflammasome activation. The same mechanisms occurred both in healthy individuals and in patients with asthma, but they were significantly more pronounced in asthma, especially when preceded with HDM exposure. Overactivation of epithelial RIG-I inflammasome in asthma compromised the dynamics of RIG-I-dependent type I/III IFNs and ISG responses, leading to less effective virus clearance, as well as to sustained inflammasome-, and IFN-dependent airway inflammation. In addition to enhancing RV-dependent RIG-I inflammasome activation, HDM also partly inhibited early type I/III IFN responses. Interestingly, this sustained RV-induced antiviral response in HBECS from patients with asthma led to restrained SARS-CoV-2 replication, but augmented inflammasome activation and proinflammatory responses. There are four main inflammasomes described to date to be involved in innate antiviral immunity against RNA viruses -the NLRP3, RIG-I and in some cases MDA5 and absent in melanoma 2 (AIM2) inflammasomes 50,55,78,79 . They are activated by several stimuli involved in the viral infection, such as viral nucleic acids, viroporins, RNAmodulating proteins, reactive oxygen species (ROS) and others 61 . Here, we found, that in vivo in humans and in fully differentiated primary human bronchial epithelium, infection with RV, a single stranded RNA virus, leads to increased priming of pro-IL-1b in a replication independent and dependent manner, and to assembly of RIG-I/ASC inflammasome, in a replication dependent manner. We did not see involvement of NLRP3 inflammasome upon RV infection or HDM exposure or even significant expression of NLRP3 in airway epithelium at baseline in any of our in vivo, in vitro or data mining approaches 66,80 . We also did not see MDA5 forming inflammasome. Other groups observed that infection of human peripheral blood mononuclear cells, human macrophages and mouse bone-marrow derived cells with other single stranded RNA viruses, vesicular stomatitis virus (VSV) and IAV, activates RIG-I/MAVS-dependent pro-IL-1b transcription and RIG-I/ASCdependent, but NLRP3-independent, inflammasome activation and mature IL-1b and IL-18 production 50,81 . In contrast, in undifferentiated, submerged cultures of primary human airway epithelial cells infection with IAV All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint revealed RIG-I and NLRP3-inflammasome-dependent mature IL-1b release 52 , while infection with RV in a similar model led to activation of NLRP3/NLRC5/ASC complexes and mature IL-1b release 53 . These differences might come from i) the undifferentiated state of the cells in the previous studies, as in non-differentiated epithelium, lacking ciliated cells, viral infection might engage different pathways than in human airways in vivo 82,83 ; ii) from different expression of inflammasome components in undifferentiated and differentiated mature epithelium lining human airways, as we showed previously 84 ; as well as iii) from the differences in the virus strains and serotypes 85 . Importantly, our in vitro and in vivo data consistently showed the same results that RIG-I is engaged as inflammasome in bronchial epithelium upon RV infection, which constitutes an important early time-point event, triggering subsequent airway inflammation. It is also possible, that in vivo in humans both inflammasomes are engaged in different cellular compartments: RIG-I/ASC in airway epithelium and NLRP3 in the infiltrating inflammatory cells in the airways. Indeed, RV infection in mice leads to partly macrophage-derived, NLRP3 inflammasome-dependent airway inflammation 51 . However, neither depletion of macrophages nor NLRP3 knockout leads to the complete blockade of mature caspase-1 and IL-1b processing upon RV infection, underlining that RV might induce other inflammasomes in airway bronchial epithelium 51 . An appropriate balance between activation of RIG-I epithelial inflammasome and subsequent IL-1b/IL-1 receptor (IL1R) signaling with RIG-I-dependent type I/III IFN responses should lead to the limitation of viral replication, efficient virus clearance and timely resolution of airway inflammation 52 . Indeed, we observed here that in the bronchial epithelium of healthy subjects at early time points during RV infection there was an activation of RIG-I inflammasome and inflammasome-mediated immune responses, together with efficient type I/III IFN and ISG-responses. Importantly all of these responses were actively inhibited or went back to the pre-infection state, already 4 days after in vivo infection. In contrast, in epithelium of patients with asthma, there was enhanced RIG-I inflammasome activation accompanied by augmented inflammasome/IL1Rmediated proinflammatory responses starting early after infection and still non-resolved in vivo 4 days after infection. Overactivation of epithelial RIG-I inflammasome and subsequent increases in mature IL-1b release might be at least partially responsible for the delayed and sustained type I/III IFN/ISG responses. We demonstrated this here by blocking caspase-1 with YVAD which led to an increase in IFN-b (IFNB) and RIG-I (DDX58) mRNA together with IFN-responsive chemokines such as CXCL10, CXCL11, CCL3, and CCL4. Our findings are in line with early observations showing that IL-1β is able to attenuate transcription and translation of type I IFNs and excessive IFNα/β-induced effects via proteasome-dependent mechanisms or by induction of prostaglandin E2 86-88 . Interestingly, we also noted here that this cross-talk between IL-1b and type I IFNs is reciprocal, as blocking phosphorylation of TBK1 and IKKe by their inhibitor BX795 and thus reducing RIG-Iinduced type I interferons, significantly increased pro-IL-1b transcription and its processing by RIG-I All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint inflammasome in airway epithelium. This was previously elegantly demonstrated in the case of NLRP3 inflammasome 89 and in transcriptional IL-1b regulation in macrophages and other myeloid cells 90 , but it has not been demonstrated for RIG-I inflammasome or considered to be important at the epithelial barrier. There is a long-lasting discussion about the underlying origins and mechanisms of the frequent viral infections and exacerbations in patients with asthma, which gained a special weight during the current SARS-CoV-2 pandemic 16, 91 . An impairment in epithelial type I/III IFN responses has been demonstrated in patients with uncontrolled and/or severe asthma, while it is less or not detectable in mild or well-controlled asthma 92-94 . More recently, it was suggested that this impairment might result from the unfitting grade (too high or too low), incorrect timing of initiation (delayed) 95 or resolution of the antiviral and inflammatory response (sustained) in bronchi of patients with asthma 96,97 . Treatment with nebulized IFN-b initiated early after development of common cold symptoms, did not provide improvement in comparison to placebo in the whole studied group including a majority of patients with mild asthma 42 , however in those with moderate/severe asthma, improvements in symptoms and lung function were observed 42,98 . Indeed, adding INF-b in vitro to monocyte-derived macrophages and epithelial cells prior to, but not after infection with IAV drastically reduced the number of nucleoprotein-1 positive cells 97 . These data combined suggest that type I/III IFNs might be extremely important in the antiviral response in the very early phase of infection, whereas they might not be relevant or could be even detrimental in the latter stages. Interestingly, the phenomenon of the frequent and severe exacerbations appears in both atopic, type-2 asthma, as well as in non-atopic, type-2 low asthma, suggesting that there might be a common mechanistic link to it 44,99 . Here, we showed that RV infection in asthma leads to the simultaneous overactivation of RIG-I inflammasome in their airway epithelium and impaired early RIG-I-dependent type I/III IFNs, which overall led to sustained inflammation, but insufficient viral clearance as compared to healthy controls. In patients with asthma, partially due to the higher pro-IL-1b expression, but lower RIG-I expression at baseline, RV infection led to the strong stimulation of RIG-I transcription and translation, and overactivation of both RIG-I-induced pathways: inflammasome and type I IFN responses. Since RIG-I inflammasome is also induced in healthy individuals in the early phase of RV infection, it might provide a physiological mechanism to fight an infection, as shown in ferrets, where the N protein of H5N1 influenza inhibiting this mechanism, resulted in higher mortality 52 . However, its overactivation in patients with asthma, especially at the very early stage of the infection, might contribute rather to the damage of the infected ciliated epithelium than to the effective antiviral response. Such damage might occur via pyroptosis 100 and lead to the release of active IL-1b and RIG-I/ASC/caspase-1 inflammasome complexes to the extracellular space and subsequent barrier impairment 2,101 , recruitment of macrophages and neutrophils 101 , and as we also showed here, to the inhibition of a timely and functional antiviral response and delayed in vivo viral clearance. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in Presence of other airway barrier-damaging and/or activating factors, such as exposure to allergens in addition to the viral infection, worsens the clinical outcomes, leads to a more severe exacerbation, hospitalization or respiratory failure 11 . Mechanistically, it might be connected with the multiplication of pathways activated in airway epithelium and in cells infiltrating the airways, and/or with the additive effects of different triggers on the same pathway 24,91 . HDM activates airway epithelium via, among others, TLR2/4, C-type lectins, and PARs in an allergen-non-specific way to initiate allergen sensitization, but also to perpetuate already developed allergic and probably non-allergic airway inflammation in the absence of the sufficiently developed inhibitory signals 45-47 . It is known that one of the strongest effects of HDM-specific immunotherapy and antiimmunoglobulin E (IgE) treatment is the reduction of the rate of asthma exacerbations. It might be at least partially connected to the fact that in sensitized individuals, allergen binding to specific IgE, captured by the Co-infections with two or more respiratory viruses occur often and likely acts as additional factors increasing airway epithelial damage. Patients with asthma are at a greater risk of developing respiratory failure as shown in the case of H1N1 influenza infection 104 . Thus, it has been somewhat surprising that so far during the current SARS-CoV-2 pandemic, epidemiological cohorts of COVID-19 patients from different geographical locations have resulted in partially contradictory observations that asthma is (USA, United Kingdom, Australia) or is not (Europe, China) a risk factor for SARS-CoV-2 infection and/or severity of COVID-19 15, 105, 106 . We and others demonstrated that the expression of angiotensin-converting enzyme 2 (ACE2), the main SARS-CoV-2 receptor All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint and other plausible points of entry, are not changed in patients with asthma, even though different types of airway inflammation or inhaled steroids might modulate their expression 107,108 and as such it seems unlikely that it would be the main reason for observed discrepancies. Different geographical locations might represent variable levels and quality of environmental exposures such as viruses or allergens, which may interfere with the rate of SARS-CoV-2 infections 23,109 and COVID-19 severity. SARS-CoV-2, an enveloped, positive-sense, ssRNA virus 110 has been shown to be sensed, depending on the cell type, by MDA5 111 , RIG-I 111 and NLRP3 112,113 . However, due to several evasion properties and encoding by non-structural (Nsp) and accessory proteins, such as Nsp1, 6,12,13 114 , various open read frames (ORFs) 115 , protein M 116 , protein N 117 and others, which antagonize interferon pathways on many levels 118 , induction of IFNs by SARS-CoV-2 itself is reduced or delayed 114,119 with the augmented proinflammatory mediator release. Several in vitro and in vivo animal studies revealed that SARS-CoV-2, similarly to SARS-CoV-1, is very sensitive to pre-treatment with type I/III IFNs 120-122 , which inhibit its replication. Patients with disrupted IFN gene expression and production or patients with autoantibodies against type I IFNs have been shown to be at a greater risk of severe COVID-19. In accordance with these studies, we observed here that infection of epithelium from patients with asthma at the moment of heightened and sustained IFNs type I/III response induced by RV, results in reduced SARS-CoV-2 viral load, in contrast to the healthy epithelium, where the RV-induced IFN-I/III response has been already actively resolved. However, even if restricted, SARS-CoV-2 infection in combination with RV and HDM led to an increase in activation of epithelial inflammasome and release of higher amounts of IL-18 and other proinflammatory cytokines. In context of timing and possible clinical relevance it may mean that patients with asthma with pre-existing RV-infection might have a slightly restricted SARS-CoV-2 infection at first, but due to excessive inflammasome-related damage and proinflammatory signaling, together with SARS-CoV-2-induced inhibition of type I/III IFNs, they may in fact succumb eventually to more severe COVID-19. Importantly, in the presence of HDM, these potentially adverse effects are further heightened, meaning that HDM reduces RVinduced IFN response, which leads to higher SARS-CoV-2 replication and it enhances RIG-I inflammasome activation, inflammation and tissue damage. All in all, we showed here in vivo and in vitro that the lack of balance between activation of RIG-I inflammasome and the RIG-I-IFNs-axis in response to a common respiratory virus, is an important driving factor of epithelial damage, lack of viral clearance and sustained airway inflammation in patients with asthma. Timely targeting of this abnormal response by the yet-to-be developed early therapeutics or even prophylactic approaches might provide in the future a beneficial strategy to prevent RV-induced exacerbations of asthma and potentially severe COVID-19. All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint to generate passage 1 working stocks, which were used for all experiments. Titers were determined using standard plaque assay, as described previously 123 . Experimental in vivo rhinovirus infection in 11 control individuals and 28 patients with asthma was performed as reported previously 67 . Briefly, non-smoking, non-atopic control individuals, and non-smoking mild/moderate patients with asthma without any recent viral illness and without serum neutralizing antibodies towards RV-A16, who passed inclusion criteria, underwent infection on day 0 with RV-A16 at the dose of 100 TCID50. Bronchial brushings, bronchial biopsies and bronchoalveolar lavage (BAL) fluid were collected around 2 weeks before and at 4 days after RV-A16 infection. Additionally, nasal lavage (NL) samples at the peak of RV-A16 infection were collected to assess RV-A16 infection rates. Only subjects who had sufficient remaining samples to be analyzed in this study and/or subjects who had successful infection in the lungs, as assessed by viral RNA copies by qPCR, were included in the BAL, NL, and biopsies analyses (n=9 healthy control, n=19 patients with asthma), and bronchial brushing microarray analysis (n=7 healthy controls, n=17 patients with asthma). The study received ethical approval from the St. Mary`s Hospital Research Ethics Committee (09/H0712/59). All participants gave written, informed consent. The clinical characteristics of the 9 control and 19 asthma study participants who had sufficient remaining samples to be analyzed in this study is presented in Supplementary Table S7 . Clinical characteristics of the study participants is presented in the Supplementary Table S7. Primary Human Bronchial Epithelial cells (HBECs) were obtained from the above-listed cohorts or from the doctor-diagnosed asthma and control individuals from two independent commercial sources: Lonza (Basel, Switzerland), and Epithelix (Plan-les-Ouates, Switzerland). Characteristics of the HBECs used in the manuscript are presented in Supplementary Table S8 . All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint HBECs from the control subjects and patients with asthma were cultured and differentiated in the air-liquid interface (ALI) conditions as described previously, with minor alterations of the previous protocol 126 . Briefly, cells from passage 2 were grown in 20 mL of bronchial epithelial basal medium (Lonza, Basel, Switzerland) supplemented with the SingleQuot Kit (Lonza, Basel, Switzerland) placed in 150cm 2 T-flask in humidified incubator at 37°C with 5% CO2 for maximum 10 days, or until 80%-90% confluency. Next, cells were trypsinized (ThermoFisher Scientific, Waltham, USA) and seeded at a density of 1.5x10 5 Cells were grown submerged for 3-5 days in the apical medium and were in contact with the basolateral medium. After they obtained a full confluence, the apical medium was removed and cells were kept in the airliquid interface (ALI) cultures for at least 21 days. BEGM/DMEM/ATRA medium was maintained only basolaterally to differentiate the HBECs. During the cell culture process, medium was exchanged every 2-3 days and, periodically, excess of produced mucus was removed from the wells. All experiments were performed on the fully differentiated HBECs from the same passage, between 21 and 28 days of ALI culture ( Fig. S1A ). House dust mite (HDM) stimulation, followed by rhinovirus A16 (RV-A16) infection experiments were performed in the OptiMEM medium (LifeTechnologies, ThermoFisher Scientific, Waltham, USA). ALIdifferentiated HBECs from control individuals and patients with asthma were treated apically with the HDM extract (Allergopharma, Reinbek, Germany) at a dose of 200 µg/mL of the total protein in 200 µl OptiMEM on the apical side, and 600 µl of clear OptiMEM on the basolateral side ( Fig. S1A ), in the humidified incubator at 37°C with 5% CO2. After 24h of HDM stimulation cells, were apically infected with RV-A16 at the MOI of 0.1 or as otherwise specified, or stimulated with UV-RV-A16 at the same MOI, and cultured in the humidified incubator at 34.5°C with 5% CO2 for the next 24h (Fig. S1A ). Next, cell supernatants (apical and basolateral), RNA, and protein cellular lysates were collected and stored in -80°C. Some cells were fixed with 4% PFA (Fluka/Sigma Aldrich Buch, Switzerland) and were stored wet at 4°C for 1-2 weeks before the subsequent confocal analyses. All doses and time-points used for the final experiments were based on the preliminary dose-dependent and time-course experiments. Briefly, two different HDM extracts: main HDM extract used All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The ALI-differentiated MucilAir cultures (Epithelix, Plan-les-Ouates, Switzerland) from primary human bronchial epithelium obtained from 4 control individuals and 5 patients with asthma (Supplementary Table S8 ) were cultured for 7 days in the MucilAir Medium (Epithelix, Plan-les-Ouates, Switzerland) in ALI conditions, with basolateral medium changed every other day. At the day of the experiment, performed in a biosafety level 3 (BSL3) laboratory, cells were washed with warm PBS to remove an excess of mucus. The experiment was performed in the OptiMEM medium (LifeTechnologies, ThermoFisher Scientific, Waltham, USA) in the volume of 250 µl on the apical, and 600 µl on the basolateral side. Through the whole experiment cells were kept in the humidified incubator at 37°C with 5% CO2. First, HBECs were stimulated apically with 200 µg/mL of protein content of HDM extract or vehicle. 24h after HDM stimulation, cells were apically infected with/without RV-A16 at the MOI of 0.1. After next 24h, HBECs were apically infected with/without SARS-CoV-2 at the MOI of 0.1. Finally, 48h after SAR-CoV-2 infection experiment was harvested (Fig. 8A ). In order to inactivate SARS-CoV-2, all collected supernatants were treated with 65°C for 30 min. Cells were fixed in 4% PFA for at least 20 min. Inactivated supernatants were frozen in -80°C until further analyses. For RNA analyses, insert with the fixed cells were preserved in RNAlater (Qiagen, Hilden, Germany), left overnight in 4°C, and stored in -20°C in the new, dry tube. In order to perform confocal staining, inserts with the fixed cells were snap frozen in the Clear Frozen Section Compound (FSC22, Leica, Wetzlar, Germany). All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in Non-essential Amino Acid Solution and heat-inactivated FCS (cRPMI medium) in the 75cm 2 T-flask, and cultured for 1 day in the humidified incubator at 37°C with 5% CO2. In the following day, cells were counted, checked for viability (98%) and transferred to the 12-well cell cultures plate (0.5 mio cells/well in 1 mL of cRPMI medium). Next day, cells were stimulated with LPS (100 ng/mL, Invivogen, San Diego, USA) or vehicle for 4h followed by 2 mM ATP or vehicle (Invivogen, San Diego, USA) for 20 min. Cytospins (250xg, 3 mins, Shandon Cytospin 2, Marshall Scientific, Hampton, USA) were prepared, and cells were immediately fixed with 4% PFA (Fluka/Sigma Aldrich, Buchs, Switzerland), and stored in wet chamber before the confocal staining. Western Blotting experiments from the cell lysates and the apical supernatants were performed as previously described 126,127 . Briefly, cells were lysed in RIPA Lysis and Extraction buffer (ThermoFisher Scientific, Waltham, All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint For co-immunoprecipitation cells were lysed with the Lysing Buffer (1µM DTT + 10% Triton X100 in ddH20 Samples preserved in the RNAlater, as described above, were immersed in the increasing concentrations of ethanol (30% up to 100%, increasing every 10%). After this initial step, RNA was isolated with use of RecoverAll All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in Rhinovirus infection in control individuals and patients with asthma after experimental RV-A16 infection in vivo was performed in the nasal lavages (peak of infection) and BAL fluid (4 days post infection) with use of qPCR, as previously described. 67 Results are presented as log10 of viral RNA copies in 1 mL. Cells were fixed on the inserts with 4% paraformaldehyde (Fluka/Sigma Aldrich, Buch, Switzerland) for 10 minutes, permeabilized with detergent (PBS + 0.1% TritonX100 + 0.02% SDS) for 5 minutes and blocked with 10% goat serum (Dako, Agilent, Santa Clara, USA) in 1% BSA/PBS for 60 min at room temperature (RT). All antibodies were diluted in 4% goat serum + 1% BSA/PBS, and cells were stained from apical and basolateral sides with 100 µl of antibodies working solution. Samples were stained for ASC (2µg/mL, mouse anti-ASC, Santa Cruz Biotechnology, Santa Cruz, USA), IL-1b (10µg/mL, mouse anti-IL-1b, Abcam, Cambridge, UK), and RIG-I (2µg/mL, mouse anti-RIG-I, Santa Cruz Biotechnology, Santa Cruz, USA) for 60 minutes at RT. Proper mouse isotype controls, in the corresponding concentrations were used to control for unspecific binding. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint and Zen Software (Zeiss, Oberkochen, Germany). All pictures were taken at the 40x magnification and are presented as maximal projection (orthogonal projection) from z-stacks (3-22 Bronchial biopsies were collected from the control individuals and patients with asthma at baseline and 4 days after in vivo RV-A16 infection. Biopsies were fixed and embedded in the paraffin blocks, sections were cut, and placed on the glass slides as described previously 96 . Prepared slides were baked 30 minutes in 65°C, followed by the deparaffinization with xylol (2x10 min), graded isopropanol (2x3 mins 100%, 2x3 mins 96% and 3 mins 70%), and rehydration (2x5 mins in H202). Samples were boiled in the sodium citrate buffer (10mM sodium citrate with 0.05% Tween 20 in PBS at pH6) in the pressure cooker for 4 mins, as described previously 129 . Samples were permeabilized and blocked with the Perm/Block Buffer (1%BSA+0.2%TritonX100+10% goat serum in PBS) for 25 minutes in RT. All antibodies were diluted in 4% goat All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint serum + 0.05% Tween20 in PBS, and 50 µl of antibodies per sample were used. Primary antibodies for IL-1b All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint Distribution (normality) of the data was assessed with Shapiro-Wilk test. One-way ANOVA (Kruskal-Wallis test), RM one-way ANOVA (Friedman test) or mixed-effects model tests were performed for more than three groups comparisons depending on the data relation (paired/not-paired) and distribution (normal/notnormal). Two-tailed paired/not-paired t-test or Wilcoxon/U-Mann-Whitney tests were performed for two groups comparisons depending on the data relation and distribution. The data are presented as the mean ± Table S10 ) and ii) protein interactions and pathways analysis prepared using the STRING (version 11.0) 132 , and further processed with the Cytoscape software (version 3.8.2) 133 (Supplementary Table S3 ). The list of all proteins available for PEA measurements at the moment of the current analysis, that were used as a background reference for STRING analyses for targeted proteomics data is presented in Supplementary Table S11. Transcriptome data from bronchial brushings from control individuals and patients with asthma experimentally infected with RV-A16 has been submitted to the NCBI GEO: GSE185658 and will be publicly available at the time of publication. All other data are included in the Online Supplement or are available from the corresponding author upon request. The codes for transcriptome data analysis are available here: https://github.com/uzh/ezRun (NGS), https://github.com/ge11232002/p1688-Ula (microarray). Code for Proximity Extension Assay data analysis is available from the corresponding author upon request. All codes will be publicly available at the time of publication. The Authors are grateful to all patients, clinicians and research staff involved in securing and processing patients samples. We would like to thank David Mirer for rhinovirus plaque assay analyses. HDM extract was a gift from Allergopharma AG. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in Th17-driven cytokines Asthma (response to infection) ratio ratio **** **** **** **** **** **** **** **** **** **** * **** **** **** *** ** **** **** **** * *** * **** ** **** **** * * * ** * * * ** * ** * * * * DDX58 CCL5 IFIH1 NLRC5 IL1RN CASP1 NFKBIA IRF1 BIRC3 MYD88 IL6 TNF IL1A IRAK2 IL1B PELI1 CFLAR CXCL2 RIPK2 TXNIP DDX58 CCL5 IFIH1 NLRC5 IL1RN CASP1 NFKBIA IRF1 BIRC3 MYD88 IL6 TNF IL1A IRAK2 IL1B PELI1 CFLAR CXCL2 RIPK2 TXNIP CXCL1 MAP3K8 CASP8 IRAK3 PANX1 IRF2 CARD6 BIRC2 XIAP PEA15 NFKB1 RELA NOD2 IRAK4 BCL2L1 PELI2 SUGT1 CROC1B TAB3 MAP3K3 CASP4 Asthma -4 0 4 8 before after in vitro RV-A16 infection **** **** **** **** **** **** **** **** **** **** **** **** **** **** **** **** **** **** **** **** *** **** **** **** **** **** **** **** *** *** *** **** **** ** ** ** ** *** perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint IFIH1 IL1RN CCL2 CASP1 AIM2 CASP5 IRF1 IL1B IFNG BIRC3 NLRC5 IRAK3 IL6 RIPK2 CASP4 MYD88 NFKBIA CXCL1 TXN PYCARD CIITA CTSB CASP8 MAP3K3 MAPK9 APP MAPK1 IL33 IKBKB CARD18 MAPK11 TNFSF11 NLRP9 NLRP7 NLRP5 MAP2K4 RELA MAPK8 CHUK UBE2N NOD2 NFKB1 HSP90AB1 TAB2 IL18 PEA15 TRAF6 XIAP TNF Asthma DDX58 IFIH1 IL1RN CCL2 CASP1 AIM2 CASP5 IRF1 IL1B IFNG BIRC3 NLRC5 IRAK3 IL6 RIPK2 CASP4 MYD88 NFKBIA CXCL1 TXN PYCARD CIITA CTSB CASP8 MAP3K3 MAPK9 APP MAPK1 IL33 IKBKB CARD18 MAPK11 TNFSF11 NLRP9 NLRP7 NLRP5 MAP2K4 RELA MAPK8 CHUK UBE2N NOD2 NFKB1 HSP90AB1 TAB2 IL18 PEA15 TRAF6 XIAP perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint significant difference of indicated condition as compared to the vehicle from the same group. (&) represents a significant difference upon HDM treatment when compared to the respective condition without HDM. ($) represents a significant difference between RV-A16 and UV-RV-A16 condition. (+) represents a significant difference between HDM+RV-A16 and HDM condition. Bar graph data show mean ± SEM analysed with oneway ANOVA (Kruskal-Wallis test), RM one-way ANOVA (Friedman test) or mixed-effects model, as appropriate, depending on the data relation (paired or unpaired) and distribution (if not mentioned differently), *p-value≤0.05, **p-value≤0.01, ***p-value≤0.001, ****p-value≤0.0001. Figure 1A All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in TNFSF11 NLRP9 NLRP7 IL12B NLRP5 PELI3 TAB3 RBX1 HSP90AA1 CFLAR MAP2K4 RELA MAPK8 CHUK UBE2N NOD2 NFKB1 HSP90AB1 TAB2 IL18 PEA15 TRAF6 XIAP TNF IRAK4 DDX58 IFIH1 IL1RN CCL2 CASP1 AIM2 CASP5 IRF1 IL1B IFNG BIRC3 NLRC5 IRAK3 IL6 RIPK2 CASP4 MYD88 NFKBIA CXCL1 TXN PYCARD CIITA CTSB MAP3K8 TNFSF14 CASP8 IRF2 TNIP2 SKP1 MAP3K3 MAPK9 APP Asthma DDX58 IFIH1 IL1RN CCL2 CASP1 AIM2 CASP5 IRF1 IL1B IFNG BIRC3 NLRC5 IRAK3 IL6 RIPK2 CASP4 MYD88 NFKBIA CXCL1 TXN PYCARD CIITA CTSB MAP3K8 TNFSF14 CASP8 IRF2 TNIP2 SKP1 MAP3K3 MAPK9 APP MAPK1 IL33 IKBKB CARD18 MAPK11 TNFSF11 NLRP9 NLRP7 IL12B NLRP5 PELI3 TAB3 RBX1 HSP90AA1 CFLAR MAP2K4 RELA MAPK8 CHUK UBE2N NOD2 NFKB1 HSP90AB1 TAB2 IL18 PEA15 TRAF6 XIAP TNF perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; Supplementary Table S3 . KEGG pathways enriched in significantly changed proteins secreted from the human bronchial epithelial cells of control subjects and patients with asthma after house dust mite stimulation and rhinovirus A16 (RV-A16) infection, as compared to RV-A16 infection alone. KEGG Pathways marked in red demonstrate proteins included in cytokine-mediated signaling pathway on the Figure 7C Supplementary preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in IL1B CTCTTCGAGGCACAAGGCA GGCTGCTTCAGACACTTGAG RV-A16 POSITIVE STRAND CGGGACTGCAAACACTACCT CACCACGTGTGTCCCTAACA DDX58 TGATTGCCACCTCAGTTGCT TCCTCTGCCTCTGGTTTGGA NLRC5 GCTGGAGGAGGTCAGTTTGC TGTTTCGGCTCAGGTCAAGT IFIH1 AGATGCAACCAGAGAAGATCCA TGGCCCATTGTTCATAGGGT CASP1 GCCCACCACTGAAAGAGTGA TTCACTTCCTGCCCACAGAC IFNL2/3 CTGGGAGACAGCCCAGTTCA AGAAGCGACTCTTCTAAGGCATCTT IFNB1 perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted November 17, 2021. ; https://doi.org/10.1101/2021.11.16.21266115 doi: medRxiv preprint Bialystok, Poland. JRC was supported by an FPI-CEU predoctoral fellowship and the Swiss-European The increase in the prevalence of asthma and allergy: food for thought Does the epithelial barrier hypothesis explain the increase in allergy, autoimmunity and other chronic conditions? Global Initiative for Asthma. Global Strategy for Asthma Management and Prevention International ERS/ATS guidelines on definition, evaluation and treatment of severe asthma Viruses and bacteria in acute asthma exacerbations--a GA(2) LEN-DARE systematic review Viral Infections and Associated Factors That Promote Acute Exacerbations of Asthma Fatal respiratory infections associated with rhinovirus outbreak, Vietnam. Emerging infectious diseases Molecular characterization of a variant rhinovirus from an outbreak associated with uncommonly high mortality Early life rhinovirus wheezing, allergic sensitization, and asthma risk at adolescence Human rhinovirus species and season of infection determine illness severity Role of viruses in asthma Risk of COVID-19-related death among patients with chronic obstructive pulmonary disease or asthma prescribed inhaled corticosteroids: an observational cohort study using the OpenSAFELY platform Factors associated with COVID-19-related death using OpenSAFELY Risk of adverse outcomes in patients with underlying respiratory conditions admitted to hospital with COVID-19: a national, multicentre prospective cohort study using the ISARIC WHO Clinical Characterisation Protocol UK. The Lancet SARS-CoV-2 infection and COVID-19 in asthmatics: a complex relationship Asthma-associated risk for COVID-19 development IFNE, IFNG, IFNGR1, IFNGR2, IFNK, IFNW1, IKBKB, IKBKE, IKBKG, IL10RB, IL12A, IL12B, IL23A, IL28A, IL28B, IL28RA, IL29, IL33, IL6, ILF3, IRAK3, IRF1, IRF3, IRF5, IRF7, IRF9, ISG15, ISG20, ITCH, IVNS1ABP ** *** Supplementary Table S1 . Enrichment analysis of the ten most significant process networks in the top one hundred upregulated genes in human bronchial epithelial cells (HBECs) from control individuals and patients with asthma after rhinovirus A16 infection. Analyzed from GSE61141 66 . Total p-value In data Network objects from active data 1 Inflammation_Interferon signaling 110 3. 409E-37 27 IL29, PKR, CCL5, IFI17, IRF1, IL28A, MxB, ISG20, TAP1 (PSF1), IFI44, GBP1, I-TAC, Apo-2L(TNFSF10), STAT1/STAT2, SOCS1, IDO1, ISG54, STAT1, IRF7, MxA, IFP 35, IFI56, ISG15, IL28B, TLR3 1 Inflammation_Interferon signaling 110 1.008E-32 25 IL29, CCL5, IFI17, IL28A, MxB, ISG20, TAP1 (PSF1), IFI44, GBP1, I-TAC, Caspase-1, Apo-2L(TNFSF10), MIG, STAT1/STAT2, IDO1, ISG54, STAT1, IRF7, MxA, IFP 35, IFI56, IL28B, TLR3, PML, STAT2 2 Immune response_Innate immune response to RNA viral infection 83 1.227E-14 13 MDA-5, RIG-I, IP10, IKK-epsilon, I-TAC, WARS, 2'-5'-oligoadenylate synthetase, IDO1, STAT1, IRF7, MxA, TLR3, STAT2 3 Inflammation_Jak preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in Male ICS 70 870 11 3 Asthma mild well 21-25 Male SABA 91 121 14 4 Asthma moderate well 16-20 Female ICS 92 64 16 0 Asthma moderate well 46-50 Female ICS 88 57 8 0 Asthma mild partial 31-35 Male SABA 79 806 13 5 Asthma moderate partial 31-35 Female ICS 65 507 8 0 Asthma moderate poor 31-35 Male ICS 84 146 12 3 Asthma moderate partial 46-50 Male SABA 73 1204 7 4 Asthma mild partial 31-35 Female SABA 78 119 6 6 Asthma moderate poor 26-30 Male SABA 73 19 11 4 Asthma moderate poor 51-55 Male ICS 83 157 26 5 Asthma moderate partial 31-35 Female ICS 77 67 7 0 Asthma moderate poor 41-45 Female ICS 68 1593 3 4 Asthma mild well 21-25 Female SABA 115 213 10 4 Asthma moderate poor 41-45 Male ICS 74 228 17 0 Asthma moderate well 46-50 Male ICS 81 87 9 3 Asthma moderate partial 46-50 Female ICS 82 2106 Supplementary Table S11. All proteins available for PEA measurements used as statistical background for STRING analyses of targeted proteomics data. UniprotID