Impaired CC16-mediated host responses to rhinovirus in nasal epithelial cells from patients with asthma

Abstract

Background:

Rhinovirus (RV) infections increase the risk for developing asthma and are the major trigger for disease exacerbations. CC16 levels are decreased in patients with asthma and inversely associated with inflammation and exacerbation frequency.

Objective:

We sought to determine the impact of CC16 on the RV species A, type 1B (RV-A1B) infection in airway nasal epithelial cells in the context of asthma status.

Methods:

Human nasal epithelial cells (HNECs) from participants with asthma and non-asthmatic participants in air-liquid interface culture were infected with RV-A1B, with and without recombinant CC16 (rCC16). Viral RNA and expression of host factors previously identified as associated with CC16 levels, including lysozyme, SPLUNC1, lactotransferrin, and surfactant protein D, were quantified by quantitative RT-PCR. Animal models and mouse tracheal epithelial cells (MTECs) sufficient and deficient in CC16 were infected with RV-A1B, with and without rCC16, to verify findings.

Results:

HNECs from participants with asthma (n = 7) had lower gene expression of CC16 and associated host defense factors under baseline conditions compared with HNECs from non-asthmatic participants (n = 7). Whereas RV infection increased host factors in HNECs from non-asthmatic participants, HNECs from participants with asthma failed to upregulate host factors in response to RV. rCC16 induced the expression of the host defense factors and reduced viral burden in HNECs from both non-asthmatic participants and participants with asthma, which was partly dependent on integrin α2α1, or VLA-2, interactions. CC16-deficient (CC16^−/−) mice had worse RV infection, more inflammation, and lower gene expression of host defense factors in their lungs compared with wild-type mice. Secreted protein levels were verified from apical washings of MTECs from CC16^−/− mice, which were significantly decreased compared with WT MTECs. Rescue studies in CC16^−/− mice by delivery of rCC16 during RV infection resulted in decreased inflammatory cell recruitment and infection.

Conclusions:

Results suggest that CC16 reduces RV infection in epithelial cells by mediating the upregulation of host defense factors and that this mechanism may be defective in patients with asthma who have low levels of CC16.

Capsule summary:

This research provides mechanistic insight into how CC16 regulates epithelial host defense against RV infection. This response is impaired in nasal epithelial cells from individuals with asthma but can be restored by rCC16 treatment.

GRAPHICAL ABSTRACT

Asthma exacerbations are a major cause of morbidity and contribute substantially to health care costs. Each year, nearly 10 million Americans have an asthma attack,¹ and thousands of others are hospitalized. Newer sensitive RT-PCR methods detected viruses in approximately 80% of wheezing episodes in pediatric patients with asthma and in approximately 50% of episodes in adult patients with asthma,² with rhinoviruses (RVs) being the most detected viral species.

RV is a positive-sense single-stranded RNA virus from the Picornaviridae family. Although RV manifests as a common cold in healthy individuals, it is thought to be a contributing factor to asthma development in young children as well as exacerbations in children and adults with asthma.^3,4 RVs comprise 3 species, RV-A, RV-B, and RV-C, and are characterized by more than 100 serotypes, making immunization impractical. Moreover, to date, no treatment for RV is available to our knowledge. Thus, better understanding of mechanisms driving RV-induced exacerbations, including acute signaling pathways that promote epithelial-driven antiviral host responses, could ultimately lead to novel therapeutic targets and/or specific treatment modalities to potentially prevent and/or treat the significant burden of asthma exacerbations.

CC16, a member of the secretoglobin family of disulfide-linked dimeric proteins, is encoded by the SCGB1A1 gene and secreted by club cells and nonciliated epithelial cells. Even though CC16 is robustly expressed in the uterus during early pregnancy,⁵ its primary site of constitutive expression is the pulmonary epithelium.^6–8 Along with being one of the most abundant proteins in bronchoalveolar lavage fluid (BALF), CC16 is easily detectable in the blood and can be used as a pulmonary biomarker of health and disease.^9,10 Furthermore, low levels in circulation were shown to predict mortality risk in a general population of adults with cancer-related deaths.¹¹ Findings from in vitro studies and mouse models have demonstrated that CC16 plays several critical roles in the lungs, including mediating anti-inflammatory and antioxidant responses in the lungs,^12–14 inhibiting phospholipase A2 activity,^15,16 and suppressing proinflammatory cytokine expression.^17–22 Recently, we discovered that CC16 promotes epithelial-driven host responses to pathogens,^23–25 limits airway remodeling factors,^23,25 and attenuates lung infiltration by activated leukocytes through direct interactions with integrin complexes in the airway and in circulation.^24,26

However, a gap in knowledge remains pertaining to mechanisms by which CC16 may impart epithelial-driven protection during viral infections. Therefore, the goal of this study was to better understand the role of CC16 during RV infection using primary human nasal cells from participants with asthma and nonasthmatic participants. In this study, we further defined mechanisms by which CC16 mediates host defense mechanisms during RV infection and discovered that those mechanisms were impaired in nasal cells from patients with asthma but could be rescued by recombinant CC16 (rCC16) treatment, paving the way for future therapeutic development. Using cells from CC16-deficient (CC16^−/−) mice, we validate gene expression findings from human studies and provide further evidence at the protein level using mass spectrometry. Finally, we demonstrate the impact of CC16 deficiency in vivo on RV burden and inflammation and the therapeutic rescue by delivery of rCC16.

METHODS

Study design

This study used a combination of mouse models and human primary cells as detailed in the graphical abstract.

Human nasal epithelial cell isolation and culturing

Human nasal epithelial cell (HNEC) brushings were obtained from nonasthmatic and asthmatic donors under University of Arizona institutional review board–approved protocols (principal investigator: Ledford) and cultured as previously described.²⁴

Experimental mice

All experiments were conducted in accordance with University of Arizona Institutional Animal Care and Use Committee–approved animal protocols. Wild-type (WT) and CC16^−/− male and female mice generated by Stripp et al²⁷ on a C57BL/6J background were approximately 6 to 8 weeks of age at the time of RV and rCC16 treatments. Mice >12 weeks of age were used for collection of mouse tracheal epithelial cells (MTECs).

RV infections in mice

RV A, type 1B (RV-A1B) was purchased from American Type Culture Collection (ATCC, VR-1645; Manassas, Va) and delivered to mice via intranasal instillation (4 × 10⁷ plaque-forming units [PFU]/50 μL) under isoflurane anesthesia. See this article’s Methods section in the Online Repository available at www.jacionline.org for additional methods.

Plaque assays with mouse lung tissue

Infected lung tissue from mice was weighed and homogenized, and lysates were added to H1 HeLa cells (ATCC). See Methods section in the Online Repository at www.jacionline.org for additional methods.

Mouse BALF collection and differential staining

At necropsy, each mouse was cannulated, and 1.5 mL of PBS (Gibco, Waltham, Mass) with 0.5 mM EDTA (Sigma, St Louis, Mo) was flushed in and out of the lungs 3 times. Recovered fluid was centrifuged, and pelleted cells were resuspended, counted, and cytospun at a density of approximately 200,000 cells/slide. Slides were stained with hematoxylin and eosin, and different cell types were assessed by light microscopy.

rCC16 treatment for MTECs and HNECs

MTECs were isolated and cultured as previously described.^23,24 See Methods section in the Online Repository at www.jacionline.org for additional methods.

MTECs and HNECs from non-asthmatic donors were treated with rCC16 (25 μg/mL) (R&D Systems, Minneapolis, Minn) for 24 hours at concentrations we have previously determined effective and in the physiologic range.²⁴ rCC16, diluted in DMEM/F12 (Gibco) (MTEC treatments) or PneumaCult ALI Medium (STEMCELL Technologies, Vancouver, BC, Canada) without supplements (HNEC treatments), was added to the apical chambers.

RV infections in MTECs and HNECs

MTEC or HNEC transwells were washed twice with 1× PBS to remove excess mucus. For MTEC infections, RV-A1B was diluted in DMEM/F12 (1 × 10⁷ PFU/200 μL) and added to the apical side. For HNEC infections, RV-A1B was diluted in PneumaCult ALI media without supplements (1 × 10⁶ PFU/12 mm transwell, or 3 × 10⁵ PFU/6.5 mm transwell). See Methods section in the Online Repository at www.jacionline.org for additional methods. UV-irradiated RV-A1B did not induce expression of inflammatory cytokines and host defense factors in air-liquid interface (ALI) culture of human airway epithelial cells (Fig E1 in the Online Repository at www.jacionline.org).

Statistical analysis

Statistical analysis was performed in GraphPad Prism Version 10.3.1 (GraphPad Software, Boston, Mass). Student t test was used for comparison in mouse and MTEC experiments and RV-infected asthmatic derived HNECs with and without rCC16 treatment. One-way ANOVA and Tukey multiple comparison were used for other experiments using HNECs.

RESULTS

Human nasal cells from patients with asthma have decreased host defense gene expression under baseline conditions

Patients were recruited through the Asthma and Airway Disease Research Center (Table E1 in the Online Repository at www.jacionline.org). Cells obtained from 7 patients with asthma (4 women, 3 men) and 7 non-asthmatic control participants (4 women, 3 men) were used in the studies. Mean age and body mass index were similar between the groups. Study participants with asthma had reduced FEV₁ (postbronchodilator 88.6% predicted) compared with non-asthmatic participants (postbronchodilator 100.7% predicted). Five patients with asthma reported use of short-acting β-agonist and 4 patients reported use of inhaled corticosteroid; no non-asthmatic participants reported use of these medications. Only 1 non-asthmatic participant was a current or former smoker; none of the asthmatic participants reported current or former smoking.

After 3 weeks of differentiation at an ALI, gene expression was examined for cell populations and each host defense factors of interest previously identified as significantly associated with CC16 gene expression in bronchial epithelial cells across studies of patients with severe asthma.²⁴ As shown in Fig 1, under baseline conditions without stimulation, gene expression was reduced for SCGB1A1 (CC16) (Fig 1, A), BPIFA1 (SPLUNC1) (Fig 1, B), LTF (Fig 1, C), LYZ2 (Fig 1, D), and SFTPD (Fig 1, E) from cells derived from patients with asthma compared with non-asthmatic participants. There was no difference in expression of tested host factors among cells from patients with asthma who used inhaled corticosteroids and patients who did not use inhaled corticosteroids (Fig E2 in the Online Repository at www.jacionline.org). No differences in cell populations were observed from the differentiated transwells grown at air-liquid interface (Figs E3 and E4 in the Online Repository at www.jacionline.org).

FIG 1.

Decreased SCGB1A1 and associated host defense gene expression in nasal cells from patients with asthma under baseline conditions. (A-E) Cells from patients with asthma and non-asthmatic participants were grown at an ALI for 3 weeks under identical conditions, after which gene expression for each factor was determined by quantitative RT-PCR of cell lysates. Each patient sample set had a minimum of 3 wells tested, and both groups contained a balance of cells from male and female participants. Replicates of transwells for each participant were averaged, and each human sample is represented as a single data point. Analysis is shown as fold relative to non-asthma cells for each factor examined. Data are presented as mean ± SEM. **P < .01, ****P < .0001 by t test. FC, Fold change.

Epithelial cells isolated from patients with asthma have heightened RV infectivity and fail to upregulate expression of host defense factors during infection

We sought to understand how HNECs from patients with asthma respond during RV infection compared to HNECs from non-asthmatic participants. After 3 weeks at ALI, HNECs were treated with medium (control) or RV-A1B, after which RV burden, SCGB1A1 (CC16) gene expression, and gene expression of the key host defense factors detailed earlier were measured after 24 hours. HNECs from patients with asthma had significantly increased RV burden compared with HNECs from non-asthmatic participants (Fig 2, A), and host defense factors CC16 (Fig 2, B), SPLUNC1 (Fig 2, C), lactotransferrin (Fig 2, D), and SP-D (Fig 2, F) significantly increased in HNECs from non-asthmatic participants during infection. In stark contrast, HNECs from patients with asthma failed to upregulate these host defense genes on RV infection (Fig 2, B–F). To verify that the low expression of host factors in cells from patients with asthma was not the result of an in vitro artifact, we measured expression level of another inflammatory cytokine, TNF-α, which was significantly increased in HNECs from both patients with asthma and non-asthmatic participants (Fig E5 in the Online Repository at www.jacionline.org).

FIG 2.

Nasal cells from patients with asthma fail to upregulate key host defense genes during RV infection and have a higher infection rate compared with non-asthma cells. Normal and asthmatic HNECs were grown at ALI for 3 weeks and treated with media or RV-A1B for 4 hours at multiplicity of infection of 1 after which unbound virus was removed. (A-F) Assessment of RV burden by RT-PCR in cell lysates (A) and host factors in cell lysates 24 hours later (B-F). A minimum of 3 transwells per patient were examined from both male and female participants, and replicates were averaged so that each human sample is represented as a single data point. Analysis is shown as fold relative to non-asthma cells for each factor examined. Data are presented as mean ± SEM. *P < .05, **P < .01, ***P < .001, ****P < .0001 by 1-way ANOVA Tukey multiple comparisons test. Ctrl, Control.

Host defense factors are negatively correlated to RV burden only in HNECs from non-asthmatic participants

Using the expression data from Fig 2, we performed correlation analyses to determine how expression of the key host defense factors correlated to RV burden within HNECs from nonasthmatic participants and patients with asthma when analyzed separately. We found that RV burden was in general negatively correlated to the host defense factors (Fig 3, A–E) within HNECs from non-asthmatic participants, indicating that in samples from non-asthmatic individuals, the higher the host defense gene expression, the lower the RV burden. However, these same host defense factors from HNECs from patients with asthma had no associations with RV infectivity.

FIG 3.

Host defense factors are negatively correlated to RV burden in normal HNECS, but not in asthmatic HNECs. (A-E) Biological replicates from each transwell measured in Fig 2 are shown for host defense gene expression and RV burden. We performed correlation analyses using a Pearson correlation coefficient model separating 6 asthma samples from 7 non-asthma samples as shown: non-asthma: r = −0.72, P = .0002; asthma: r = 0.26, P = .26 (A); non-asthma: r = −0.57, P = .006; asthma: r = −0.22, P = .36 (B); non-asthma: r = −0.74, P < .0001; asthma: r = 0.079, P = .74 (C); non-asthma: r = −0.64, P = .001; asthma: r = −0.23, P = .33 (D); non-asthma: r = −0.52, P = .012; asthma: r = −0.20, P = .40 (E). FC, Fold change.

rCC16 delivered exogenously to HNECs upregulates gene expression of host defense factors

We next examined if rCC16 delivered exogenously to HNECs would result in an upregulation of the host response factors in question. For these studies, rCC16 or medium (control) was added to the apical surface of transwells of HNECs from non-asthmatic participants and patients with asthma, and gene expression was examined 24 hours later. Expression of SCGB1A1 (CC16), BPIFA1 (SPLUNC1), LTF, LYZ2, and SFTPD was significantly increased by rCC16 in cells from both patients with asthma and non-asthmatic participants at this time point, compared medium only (Fig 4, A and B).

FIG 4.

rCC16 increases host defense factors in HNECs from patients with asthma and non-asthma participants. (A and B) HNECs from non-asthma participants (A) and patients with asthma (B) were differentiated at an ALI for 3 weeks, after which they were treated apically with either medium (vehicle control) or rCC16 (2.5 mg/mL) for 24 hours. SCGB1A1, BPIFA1, LTF, LYZ2, and SFTPD expression in cell lysates were measured by RT-PCR; GAPDH was used as a housekeeping control. A minimum of 3 transwells per patient were examined from both male and female participants, and replicates were averaged so that each human sample is represented as a single data point. Each gene is shown in comparison with their respective nontreated control. Data are presented as mean ± SEM. *P < .05, **P < .01, by t test. FC, Fold change.

Expression of key host defense factors during RV infection is dependent on CC16 signaling through VLA-2

Our group has previously determined that CC16 induces the expression of SPLUNC1 by signaling through VLA-2 within the respiratory epithelium.²⁴ Therefore, to understand the mechanism by which CC16 induces the expression of these other key host defense factors during RV infection, we decided to also look at the relevance of VLA-2 signaling. Addition of rCC16 to HNECs from non-asthmatic participants resulted in significantly increased expression of LYZ2, SFPTD, and LTF compared with media only (Fig 5, A–C). The addition of the VLA-2 inhibitor (BTT3033) during rCC16 treatment resulted in significantly decreased expression of BPIFA1 (not shown but previously published²⁴), SFTPD, and LTF, but not LYZ2, compared with rCC16 treatment alone. As CC16 induces the expression of BPIFA1, SFPTD, and LTF through VLA-2-dependent mechanisms, we also examined RV burden in HNECs treated with a VLA-2 inhibitor during RV infection. As expected, addition of rCC16 during RV infection resulted in significantly decreased RV burden by almost 2 logs compared with RV only. However, this response was negated when VLA-2 was inhibited during rCC16 treatment (Fig 5, D), with those cells showing similarly high levels of infection as the RV-only group.

FIG 5.

CC16 signaling through VLA-2 induces the expression of several host defense factors and attenuates RV burden in HNECs. Non-asthmatic HNECs were differentiated on transwells at an ALI for 3 weeks after which they were treated with medium, rCC16, or combined rCC16 and BTT3033 for 24 hours, all apically. (A-C) Gene expression was assessed for LTF (A), SFTPD (B), and LYZ2 (C) by RT-PCR. BPIFA1 was examined in a previous publication and found to be VLA-2 dependent. (D) An additional set of transwells was infected with RV-A1B (multiplicity of infection = 1) for 4 hours, after which unbound virus was removed and medium with or without rCC16 was added in the presence or absence of BTT3033, and RV burden was assessed by RT-PCR 24 hours later. A minimum of triplicate wells from 2 participants is shown in these mechanistic studies. Data are presented as mean ± SEM. *P < .05, **P < .01, ****P < .0001 by 1-way ANOVA Tukey multiple comparisons test. FC, Fold change; ND, not detected.

Exogenous rCC16 rescues host defense gene expression during RV infection in nasal cells from patients with asthma

Nasal cells from patients with asthma failed to upregulate CC16 and associated host defense genes on RV infection, but when rCC16 was given under naive conditions, those factors were increased in cells from patients with asthma as well as non-asthmatic participants. We next examined whether rCC16 could also upregulate host defense genes during RV infection and therefore reduce RV burden in the asthmatic nasal cells, which were impaired in host responses compared with nonasthmatic cells, as shown in Figs 2 and 3. RV burden was decreased in asthmatic HNECS in which rCC16 was given during infection compared with vehicle controls (Fig 6, A). Similar to naive conditions, when rCC16 was given exogenously to RV-infected HNECs from patients with asthma, several host defense genes were significantly upregulated, including SCGB1A1, BPIFA1, LTF, and LYZ2 (Fig 6, B–E), whereas SFTPD gene expression (Fig 6, F) was increased when rCC16 was administered during RV infection, but did not achieve statistical significance.

FIG 6.

Exogenous rCC16 increases expression of host defense factors and reduces RV infection in HNECs from patients with asthma. Asthmatic HNECs were differentiated at ALI for 3 weeks and infected with RV-A1B (multiplicity of infection = 1) for 4 hours, after which unbound virus was removed. Indicated transwells had rCC16 added apically for 24 hours compared with medium controls. (A) RV burden in cell lysates and (B-F) gene expression of SCGB1A1 (B), BPIFA1 (C), LTF (D), LYZ2 (E), and SFTPD (F) by RT-PCR, fold change normalized to GAPDH. A minimum of 3 transwells per patient were examined from both male and female participants; due to high variability among RV-infected wells, all replicates are displayed from 6 patient samples for RV infection and 4 patient samples for RV combined with rCC16. Data are presented as mean ± SEM. *P < .05, **P < .01, ***P < .001, ****P < .0001 by t test. FC, Fold change; ns, not significant.

CC16 deficiency results in increased infectivity and mixed granulocytic responses to acute RV infection in vivo

WT and CC16^−/− mice were infected with RV-A1B, which has known infectivity in mouse models.²⁸ After 24 hours, BALF was assessed for cellular inflammation, and lung tissue was assessed for RV burden. Compared with RV-infected WT mice, we found heightened neutrophils in the infected CC16^−/− mice (Fig 7, B). Macrophages remained consistent and not different between WT and CC16^−/− mice (Fig 7, A). RT-PCR and plaque assays of mouse lung tissue both demonstrated that CC16^−/− mice had significantly increased RV burden compared with the infected WT mice (Fig 7, C and D). Additionally, rescue studies were performed with treatment of CC16^−/− mice with rCC16 during RV infection, which resulted in significantly decreased neutrophils (Fig 7, F) and RV burden (Fig 7, G and H) compared with saline treatment. Macrophage numbers were not impacted by rCC16 treatment (Fig 7, E).

FIG 7.

CC16^−/− mice have increased inflammation and RV infection, which can be rescued by rCC16 delivery. (A-D) WT and CC16^−/− mice were infected intranasally with RV-A1B (4 × 10⁷ PFU/50 μL) for 24 hours, and inflammatory cells from BALF were examined by differential cell counts. (C and G) RV burden was measured in lung tissue by RT-PCR; Gapdh was used as a housekeeping control. (D and H) Plaque assays were performed using lung tissue lysates. (E-H) For rescue studies, CC16^−/− mice were treated with either rCC16 (16.4 μg/mouse) or saline (vehicle controls) at the time of RV infection and assessed 24 hours later. *P < .05, **P < .001 by unpaired t test. ns, Not significant.

MTECs lacking CC16 have heightened RV infectivity

To support the HNEC studies, we sought to determine the impact of CC16 during viral infection using differentiated MTEC cultures from WT and CC16^−/− mice at an ALI challenged with RV-A1B. For these studies, MTECs were infected for 4 hours, after which any nonadherent virus is washed from the transwells. Then, cells were allowed to incubate for an additional 24 hours, after which RT-PCR was performed for RV infectivity. As shown in Fig 8, A, cells lacking CC16 had significantly enhanced infection with RV compared with WT cells, suggesting an impairment in antiviral host responses as well as an impairment in antimicrobial responses we have previously observed with Mycoplasma pneumoniae.²³

FIG 8.

Loss of CC16 results in increased infectivity and decreased expression of host defense factors in MTECs and mouse lungs. (A) WT and CC16^−/− MTECS were grown at ALI for 2 weeks and infected with RV-A1B at 2 × 10⁶ PFU (multiplicity of infection = 1) for 4 hours after which unbound virus was removed. RV burden was determined by RT-PCR of cell lysates 24 hours later. (B-E) Gene expression of host defense factors (lysosome, SPLUNC1, lactotransferrin, and surfactant protein D) in naive WT and CC16^−/− mouse lung tissue by RT-PCR. (F) Apical secretion of these key antimicrobial host defense proteins by naive MTECs using mass spectrometry and quantitative proteomics. The heat map illustrates expression of each of these proteins by CC16^−/− MTECs, relative to WT MTECs. Data are presented as mean ± SEM. *P < .05, **P < .001 by unpaired t test, ****P < .0001 by t test.

Validation of host defense factors in mouse lungs

We sought to validate the key host defense factors from HNECs (detailed above), which were identified from our previously published mass spectrometry data.²⁴ As shown in Fig 8, lysozyme (Lyz2) (Fig 8, B), SPLUNC1 (Bpifa1) (Fig 8, C), lactotransferrin (Ltf) (Fig 8, D), and SP-D (Sftpd) (Fig 8, E) all had significantly lower gene expression in CC16^−/− lung lysates compared with WT controls. Additionally, we validated these factors at the protein level using mass spectrometry, which demonstrated that naive CC16^−/− MTECs have significantly decreased apical protein secretion of these 4 host response factors, as well as CC16, compared with the WT MTECs (Fig 8, F), similar to our previously published findings.²⁴

DISCUSSION

A major finding in this study is that at baseline under naive conditions HNECs from patients with asthma grown at an ALI had overall reduced expression of host defense genes compared with HNECs from non-asthmatic participants. We also discovered that RV infection alone upregulates gene expression of host defense factors in non-asthmatic HNECs—a response that was surprisingly absent in HNECs from patients with asthma. We also observed that rCC16 was able to upregulate SCGB1A1 gene expression, which is the gene encoding CC16 protein. Although previous studies have hinted at an autoregulatory role of CC16, this area warrants further investigation.²⁹ In line with previous publications, we also observed that HNECs from patients with asthma had significantly increased RV burden compared with non-asthmatic HNECs.^{2–4,30–33} However, to our knowledge, we are the first to show that during RV infection, asthmatic HNECs have significantly decreased CC16 expression, as well as a failure to respond to infection by upregulating gene expression of the host defense factors examined, compared with the non-asthmatic control participants. Previous publications have shown that CC16 expression is significantly decreased within the circulation, BALF, and bronchial airway cells of patients with asthma^14,34–36 and that patients have significantly increased incidence of pulmonary infection^3,30,37,38; therefore, our data tie these concepts together by shedding light on novel mechanisms driving the dysfunction of the pulmonary epithelium during respiratory infection likely due to decreased CC16 levels.

In addition to being a human pathogen,³⁹ RV-A1B, a minor group of RV, could infect mice because the virus uses the low-density lipoprotein receptor to enter cells. Even though the minor group and the major group RVs attach to a different receptor, they have been shown to induce similar responses in both BEAS2B and primary human airway epithelial cells.^40,41 Interestingly, the observation that NECs from asthma donors have higher RV-A1B burden is consistent with our earlier finding for the type C RV-C15,⁴² which uses CDHR3 for cell entry. This highlights the possibility that CC16-influencing host defense deficiency in asthma is not receptor specific.

Although CC16 is predominantly produced by nonciliated club cells in the distal airways, the nasal epithelium also contains a subset of CC16-producing club cells,^43,44 which were evident in our cultures as well. Nasal cells from both groups demonstrated differentiation into ciliated, goblet, basal, and club cells, without notable differences between non-asthmatic and asthmatic patient samples under naive conditions (Figs E3 and E4).

The HNEC studies were performed only from cells that had been expanded and differentiated at an ALI for 3 weeks, and therefore we do not know if any of these factors would be different in fresh nasal brushings from patients with asthma. We acknowledge that the differences observed in response to RV in our HNECs may have been impacted by the expansion and growth process. However, in our previous study, we found that these same host defense factors were significantly associated with epithelial CC16 expression from freshly collected bronchial cells from patients with asthma by RNA sequencing.³⁶ Future studies are needed to determine if these factors are also reduced in fresh nasal brushings from patients with asthma compared with control participants.

Mechanistically, our study adds to the previous knowledge that CC16 induces SPLUNC1 production and BPIFA1 gene expression through VLA-2-dependent signaling,²⁴ by demonstrating that CC16 mediated induction of lactotransferrin and SP-D is also VLA-2 dependent, whereas lysozyme appears to be VLA-2 independent. Interestingly, the host defense factors examined all have AP-1 binding sites in their promoters and are under the control of an AP-1 transcription factor under naive or stimulated conditions.^45–47 It is likely that CC16 mediates these host defense factors through pathways that activate AP-1 transcription factor binding; however, validation of this is still ongoing and beyond the scope of these studies.

VLA-2 inhibition during rCC16 treatment resulted in significantly increased RV burden compared with rCC16 treatment alone, which is likely due in part to the decreased expression of SPLUNC1, SP-D, and lactotransferrin, which have known antiviral activities. However, we cannot exclude the possibility that additional host defense factors, such as SP-A, may also be at play during acute RV infection.

We selected sex-matched HNECs from non-asthmatic individuals and individuals with asthma to ensure that sex differences were considered. We did not see any differences based on sex of the derived HNECs regarding RV infection and CC16 activity. However, we did not examine any cells from pediatric patients and recognize that asthma susceptibility and viral infection may be impacted differently in a pediatric population. Lastly, our work has focused on 4 key host defense factors that were identified in our previous mass spectrometry screening²⁴; however, CC16 possibly influences the expression of many other factors that could also impact host defense. Likewise, focusing our studies on a single time point (24 hours) is a limitation and provides only a snapshot of the epithelial host response to infection. We also acknowledge that our patient sample size was small (7 patients with asthma and non-asthmatic participants), all were recruited from the Tucson, Arizona, area, and all patients had mild asthma. Future studies will be needed in patients with moderate to severe asthma, which may be challenging due to disease heterogeneity in these patients and increased medications used.

To validate and better understand the impact of the loss of these CC16-mediated responses in vivo, we infected WT and CC16^−/− mice with RV for 24 hours, after which we measured RV burden and inflammatory cell recruitment. We also examined expression of host response factors under naive conditions that we previously identified by mass spectrometry of the secretome from WT and CC16^−/− MTECs,²⁴ giving us validation at the protein level. Lysozyme is an antibacterial protein that plays a role in innate immune defense, as well as providing protection against bacteria, viruses, and fungi.^48,49 SPLUNC1 is a glycoprotein that is highly expressed within the pulmonary epithelium and responsible for inducing innate immune responses and killing activity against multiple pulmonary bacterial and viral pathogens.^50–55 Lactotransferrin is an innate defense, iron-binding protein that is secreted by bronchial epithelial cells and responsible for antibacterial, antifungal, and antiviral properties.^56–58 SP-D is an innate immune defense protein expressed by the pulmonary epithelium and responsible for exerting antimicrobial effects through direct microbial interactions⁵⁹ and antiviral effects through viral neutralization and induction of phagocytosis.^60–62

CC16^−/− MTECs had increased RV burden, as well as decreased expression of each of these host response factors, compared with WT MTECs. The impact of this dysfunction in host responses was evident in CC16^−/− mice. Despite the finding that more neutrophils were recruited in RV-infected CC16^−/− mice, they had higher RV burden in the lungs. This is in alignment with our previous studies with M pneumoniae infection, suggesting an epithelial-driven impairment in responses to pathogens.²⁶ Increased leukocyte extravasation to the lungs of RV-infected CC16^−/− mice is likely due to the absence of CC16 binding to VLA-4 on leukocytes and preventing their transmigration into the airway.^23–26 However, delivery of rescue rCC16 to CC16^−/− mice resulted in fewer neutrophils present in the BALF as well as decreased RV infection.

Overall, we provide evidence in human primary nasal cells from individuals with asthma and non-asthmatic individuals that CC16 induces the expression of the key host defense factors through VLA-2- and non–VLA-2-dependent mechanisms. Further, RV burden is significantly and negatively correlated to the expression of these key host defense factors in HNECs from non-asthmatic individuals, whereas HNECs from individuals with asthma fail to mount an appropriate host defense response to RV, which we believe is linked to low CC16. Our work was further strengthened by findings that were replicated in mice deficient in CC16. The rescue studies delivering rCC16 in RV-infected CC16^−/− mice provide additional support for our findings. As several chronic respiratory diseases have been associated with low CC16 levels, including asthma,^{14,35,36,63–66} chronic obstructive pulmonary disease,^6,9,67–72 and cystic fibrosis,⁷³ and affected individuals are at an increased risk for respiratory infections, this new discovery that decreased CC16 levels may shape a nasal environment with decreased production of several key host defense proteins is of critical importance. Translationally, asthmatic HNECs with impaired host defense responses to RV were able to adequately mount a response with the addition of rCC16, supporting the potential of CC16 augmentation as a therapeutic approach during respiratory infections in vulnerable populations.

Key messages.

• Nasal epithelial cells from individuals with asthma have reduced baseline and RV-induced host defense responses, which may be driven by deficits in CC16.

• rCC16 restores protection in nasal cells from individuals with asthma by upregulating host defense factors and lowers viral burden via integrin α2β1 pathway.

• CC16 deficiency worsens infection and inflammation in mice, whereas supplementation with rCC16 reverses these effects.

Abbreviations used

ALI

Air-liquid interface

BALF

Bronchoalveolar lavage fluid

HNEC

Human nasal epithelial cell

MTEC

Mouse tracheal epithelial cell

PFU

Plaque-forming units

rCC16

Recombinant CC16

RV

Rhinovirus

RV-A1B

RV A, Type 1B

WT

Wild-type