Molecular Mechanisms of Airway Hyperresponsiveness in a Murine Model of Steroid-Resistant Airway Inflammation

Asthma is an incurable, chronic airway disease that is characterized by pulmonary inflammation, mucus secretion, and airway hyperresponsiveness (AHR). Patients suffer from variable and reoccurring symptoms and airflow obstruction in response to Ags. Clinically characterized phenotypes have provided limited insight into the underlying pathogenic mechanisms of this heterogeneous disease (1, 2). In addition, there is increasing evidence to suggest that each of the characteristic asthma endpoints may be regulated by distinct molecular pathways (3). Therefore, studies linking pathophysiologic mechanisms to subtypes of asthma are of great importance to establish targeted and effective therapies.

Allergen-specific Th2 and/or Th17 cells are known to critically orchestrate inflammatory responses and AHR, the central feature of asthma. Specifically, IL-13 and IL-17A, produced by Th2 and Th17 cells, respectively, are effectors in the pathogenesis of disease. IL-13 contributes to disease by promoting mainly steroid-responsive, Th2-dominant disease marked by eosinophil influx into the lungs, as well as increased IgE production, mucus secretion, and AHR. In contrast, IL-17 plays a critical role in driving neutrophil influx into the airways and AHR in response to innocuous Ag. IL-17 also is implicated in mucus metaplasia and airway remodeling, which are more prominent in chronic steroid-resistant disease (4, 5). Both mRNA and protein levels of IL-13 and IL-17A are elevated in patients with asthma (6–8), and increased levels of either cytokine in the lung were shown to correlate with disease severity (9–11). Our previous work identified asthmatics with concomitant “high” Th2 and Th17 inflammatory phenotype (3), who trended to have worse lung function and increased steroid usage (3). There is no available murine model that mimics this subtype of disease present in patients with asthma, which is characterized by mixed immune pathway activation and steroid insensitivity. To better understand the mechanisms underlying this “high” Th2/Th17, steroid-insensitive disease, we developed a novel murine model that recapitulates this clinical phenotype. Because murine models are not available to recapitulate this subtype of severe asthma, a novel model was developed that mimics this clinical phenotype. Finally, gene-linkage studies showed association of polymorphisms in the genes for IL-13, the receptor for IL-13 (IL-4Rα), and IL-17 in allergic asthma, further confirming the significance of these cytokines in asthma (12–15).

To elucidate the molecular mechanisms of disease in “high” Th2/Th17 steroid-resistant asthma, we established a novel T cell–adoptive transfer mouse model. Based on the pivotal roles of IL-13 and IL-17A in disease pathogenesis, we investigated the involvement of these influential proinflammatory cytokines in inflammation, mucus production, and physiological changes in the lung. The goal of this work was to determine which hallmark phenotypes associated with allergic airway disease (inflammation, mucus metaplasia, AHR) are specifically controlled by IL-13 and/or IL-17A in a model of steroid-resistant asthma with a mixed T cell phenotype.

Nine- to eleven-week-old female BALB/cJ, BALB/c SCID, STAT6^−/− (BALB/c background), and DO11.10 TCR-transgenic mice were purchased from The Jackson Laboratory. All mice were housed in a pathogen-free environment at the Children’s Hospital of Pittsburgh and were given food and water ad libitum. All animal experiments were reviewed and approved by the University of Pittsburgh Institutional Animal Care and Use Committee.

CD4⁺CD62L⁺ naive T cells from the spleens of DO11.10 TCR-transgenic (OVA-specific) mice were enriched by negative and positive selection using Ab-coated magnetic beads (CD4⁺CD62L⁺ T cell Isolation Kit II; Miltenyi Biotec). Naive T cells were cultured for 6 d with anti-CD3/anti-CD28 mouse Dynabeads (Invitrogen) under Th2 cell– or Th17 cell–polarizing conditions, as previously reported (16). Th17-polarization medium was supplemented with 10 ng/ml IL-23, 1 ng/ml TGF-β, 20 ng/ml IL-6, 10 μg/ml anti–IL-4, and 10 μg/ml anti–IFN-γ (R&D Systems). Th2 polarization medium was supplemented with 20 U/ml IL-2, 5 ng/ml IL-4, and 10 μg/ml anti–IFN-γ (R&D Systems).

BALB/c SCID mice were challenged with 50 μg OVA (Sigma-Aldrich) via oropharyngeal aspiration on day 0. On day 1, 2 × 10⁶ Th2 and Th17 cells (1 × 10⁶ of each), 2 × 10⁶ Th2 cells, or 2 × 10⁶ Th17 cells were adoptively transferred by retro-orbital injection. Mice were challenged via oropharyngeal aspiration with 50 μg OVA daily for three consecutive days after cell transfer (days 2–4). Twenty-four hours after the last challenge, mice were euthanized (day 5), and allergic airway disease was assessed as detailed below. Control mice received all OVA challenges but received PBS retro-orbitally at the time of cell transfer. For dexamethasone (DEX)-sensitivity studies, OVA-challenged BALB/c SCID mice received 2.5 mg/kg DEX (APP Pharmaceuticals), 2 h prior to Th2/Th17 cell transfer, and OVA treatment on days 1 and 3, respectively. This dosage regimen was previously shown to inhibit Th2-mediated allergic airway disease in our model (16). To neutralize IL-13 and/or IL-17A, mice were treated with 300 μg control IgG, 100 μg anti–IL-13 Ab, and/or 300 μg anti–IL-17A Ab (all from Janssen Research & Development) via i.p. injection on days 1 and 3. To assess the role of IL-19, recombinant mouse IL-19 (4 μg/mouse via oropharyngeal aspiration; R&D Systems) was administered to Th2/Th17 cell–transferred, OVA-challenged mice on days 1 and 3. To overexpress IL-33 in the lungs, Th2/Th17 cell–transferred, OVA-challenged mice were infected with 5 × 10⁸ PFU adenovirus expressing IL-33 (AdIL-33) or adenovirus expressing control virus via oropharyngeal aspiration on day 2.

Following experimental exposures, pulmonary function was assessed by mechanical ventilation of anesthetized (90 mg/kg pentobarbital-NA, i.p.) and tracheotomized mice using a computer-controlled small-animal mechanical ventilator (flexiVent; SCIREQ) as previously described (17). Mice were mechanically ventilated at 200 breaths/min with a tidal volume of 0.25 ml and a positive end-expiratory pressure of 3 cm H₂O (mimicking spontaneous ventilation). The quasi-static mechanical properties of the lung were measured using pressure-volume curves, which involve inflating the lungs in a series of 1-s steps from atmospheric pressure up to the total lung capacity (∼30 cmH₂O pressure). Respiratory mechanics measurements were made prior to and following the administration of the drug methacholine (dose range: 0–50 mg/ml), which causes the smooth muscle surrounding the airways to constrict. Multiple linear regression was used to fit measured pressure and volume in each individual mouse to the linear model of the lung (18, 19). Model fits that resulted in a coefficient of determination < 0.8 were excluded.

Following the analyses of lung mechanics, bronchoalveolar lavage fluid (BALF) was collected from mice via the intratracheal instillation and recovery of 1 ml PBS, and total cells in BALF were counted using a hemocytometer. Cytospin preparations of BALF cells were used for differential counting of neutrophils, eosinophils, macrophages, and lymphocytes. Lung lobes were separated and flash frozen in liquid nitrogen for cytokine analysis by LINCOplex and Western blotting or gene expression analyses using qualitative real-time PCR (qRT-PCR), or they were inflation-fixed with 10% buffered formalin and embedded in paraffin for histology (20, 21).

Total RNA was isolated from lung tissue using an Absolutely RNA Miniprep Kit (Agilent Technologies, Santa Clara, CA). For qRT-PCR, cDNA was generated using an iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA). qRT-PCR was performed using Assay On Demand TaqMan primer and probe sets for target genes of interest (Life Technologies, Grand Island, NY) with the Applied Biosystems 7900 machine. The following target genes were analyzed: Muc5ac (Mm01276725_g1), Muc5b (Mm00466376_m1), Clca3 (Mm01320697_m1), Il4 (Mm00445259_m1), Il6 (Mm00446190_m1), Il13 (Mm00434204_m1), and Ifng (Mm01168134_m1). Relative gene expression was quantified using the ΔΔCt (cycles to threshold) method, with Hprt (Mm03024075_m1) as the lung endogenous control gene.

Inflammation and mucus production were assessed by H&E and periodic acid–Schiff (PAS) staining, respectively. To characterize the location of inflammation in the lung, standard H&E-stained lung sections were scored by a pathologist (M.L.M.) who was blinded to the sample group identity. The entire lung section was observed with a light microscope (original magnification ×40), and peribronchial, perivascular, and parenchymal inflammation was scored as previously described (3). Each sample was given a score based on the percentage of the specific area that contained cellular inflammation according to the following scale: 0 = no inflammation, 1 = up to 25%, 2 = 25–50%, 3 = 50–75%, and 4 = 75–100%. The inflammation score is reported as the mean of the scores for each sample group for each area.

The amount of mucus production was quantified by measuring the amount of PAS staining in all bronchioles in the lung section. Individual fields were examined with a light microscope (original magnification ×100), and every bronchiole in the entire lung section was scored by a pathologist (M.L.M.) who was blinded to sample group identity. Scoring in each field was based on the percentage of the bronchiole that had PAS staining according to the following scale: 0 = no mucus staining, 1 = up to 25%, 2 = 25–50%, 3 = 50–75%, and 4 = 75–100%. The PAS score of bronchioles is reported as a ratio of the sum of all scores divided by the total number of fields counted for each sample.

Frozen lungs were homogenized in PBS with protease and phosphatase inhibitors (complete ULTRA and PhosSTOP mixture tablets; Roche) and processed for Western blotting. A total of 20 μg protein/lung homogenate sample was subjected to SDS-PAGE with 4–15% gradient gels (Bio-Rad). Proteins were transferred onto a nitrocellulose membrane (Bio-Rad), and the membrane was blocked in 5% BSA in TBS-T. The membrane was then incubated with a 1:1000 dilution of anti–p-STAT6 or anti–STAT-6 (Santa Cruz) in 1% BSA in TBS-T or with a 1:5000 dilution of mouse anti–β-actin (Abcam). After washing, the membrane was incubated with a 1:5000 dilution of an HRP-conjugated goat anti-rabbit IgG Ab (Cell Signaling). The membrane was developed using SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher) and visualized on a Bio-Rad imaging system. Loading control on lung homogenate blots was performed by stripping and reprobing the membrane for β-actin.

Data were analyzed using GraphPad Prism 5.0 (GraphPad, La Jolla, CA). Experiments involving two variables (AHR measurements) were analyzed by two-way ANOVA with a Bonferroni post hoc test. Data with one variable were analyzed using one-way ANOVA with the Tukey post hoc test. Data comparing two groups were analyzed using an unpaired t test. All figures show combined data from multiple studies or independent repeats (three or more throughout). Data shown are mean ± SEM. A p value <0.05 was considered statistically significant.

Our previous work established that severe asthmatics have immunologically distinct phenotypes of disease (3). Specifically, severe asthmatics were identified with concurrent “high” Th2 and Th17 inflammatory phenotypes, which may correlate with worse lung function and increased steroid usage (3), yet many existing murine models do not recapitulate this disease phenotype. To examine the molecular mechanisms underlying a mixed “high” Th2 and Th17 disease phenotype, a murine model of steroid-resistant allergic airway disease was established. Previous work characterized the pathophysiological differences between murine models of allergic airway disease established by OVA-specific Th2 and Th17 cells alone (16). Combining these individual adoptive-transfer approaches established an allergic airway disease model with pulmonary inflammation, characterized by elevated neutrophils and eosinophils and Th2 and Th17 cytokines, AHR, and mucus metaplasia representative of “high” Th2/Th17 disease (Figs. 1, 2).

Next, DEX sensitivity of this model of “high” Th2/Th17 allergic airway disease was assessed. Adoptive transfer of Th2 and Th17 cells into OVA-challenged BALB/c SCID mice resulted in inflammatory cell recruitment into the airspace that was significantly reduced by DEX treatment (Fig. 3A). Differential counting of BALF cells revealed that neutrophils, macrophages, eosinophils, and lymphocytes were elevated in mice with “high” Th2/Th17 allergic airway disease compared with control (Fig. 3B). The number of neutrophils, eosinophils, and lymphocytes increased significantly in response to DEX treatment, although not to the level observed in the control mice (Fig. 3B). Histological analyses and characterization of tissue inflammation in lungs of mice with “high” Th2/Th17 allergic airway disease, with and without DEX treatment, showed that DEX slightly reduced, but did not significantly alter, tissue inflammation (Fig. 3C, 3D). Specifically, perivascular, peribronchial, and parenchymal-associated inflammation were not affected by DEX treatment in “high” Th2/Th17 allergic airway disease in mice (Fig. 3D). Further, there were no significant differences in the levels of Th2-related (Fig. 3E) and Th17-related (Fig. 3F) cytokines and chemokines in the lungs of mice with “high” Th2/Th17 allergic airway disease, with and without DEX treatment. Pulmonary gene expression of Il13 was significantly decreased, whereas Il4, Il6, and Ifng expression did not change in the lungs following DEX treatment in mice with “high” Th2/Th17 allergic airway disease (Fig. 3G). Il13 mRNA expression decreased and protein levels tended to be lower in the lungs of mice with “high” Th2/Th17 allergic airway disease following IL-13 and/or IL-17A neutralization compared with control. The lower levels of mRNA expression observed may be predictive of a decrease in protein at a later time point than examined in this work.

To determine whether inhibition of airway inflammation by DEX treatment could impact AHR and mucus metaplasia, AHR to increasing doses of methacholine, as well as quasi-static lung compliance and hysteresis, was measured in Th2/Th17 cell–transferred, OVA-challenged mice with and without DEX treatment. Airway resistance (Rn), tissue damping (G), and tissue elastance (H) in response to methacholine were unchanged by DEX treatment in mice with “high” Th2/Th17 disease (Fig. 4A–C). Further, quasi-static lung compliance and hysteresis did not change in response to DEX treatment in these mice (Fig. 4D, 4E). Clca3, Muc5ac, and Muc5b pulmonary gene expression (Fig. 4F) and PAS staining of lung tissue (Fig. 4G, 4H) from mice with “high” Th2/Th17 disease were not significantly changed by DEX treatment. Overall, although cellular inflammation in the airways was limited by DEX, adoptive transfer of Th2 and Th17 cells in OVA-treated BALB/c SCID mice produced steroid-resistant disease, characterized by DEX-insensitive tissue inflammation, AHR, and mucus metaplasia. This novel murine model mimics the high-neutrophil, high-eosinophil steroid-resistant disease identified previously in severe asthmatics (3).

IL-13 and IL-17A are mainly produced by activated Th2 and Th17 cells, respectively, and are implicated in asthma pathogenesis. To determine the influence of IL-13 and/or IL-17A in “high” Th2/Th17 allergic airway disease, the adoptive transfer model was used to establish this disease in mice (Figs. 1, 2) in conjunction with neutralizing Abs to these cytokines. Adoptive transfer of Th2 and Th17 cells into OVA-challenged BALB/c SCID mice resulted in inflammatory cell recruitment into the lungs (Fig. 5A). Differential counting of BALF cells revealed that neutrophils, macrophages, and eosinophils were elevated in mice with “high” Th2/Th17 allergic airway disease. Neutralization of IL-13 alone, IL-17A alone, or both cytokines did not significantly affect airspace inflammation in these mice (Fig. 5B). Histological analyses of the lung illustrate that perivascular, peribronchial, and parenchymal-associated inflammation are not attenuated in mice with “high” Th2/Th17 allergic airway disease, regardless of Ab treatment (Fig. 5C, 5D). Further, neutralization of IL-17A, IL-13, or both cytokines in mice with “high” Th2/Th17 allergic airway disease did not significantly alter protein levels of proinflammatory cytokines or chemokines in the lungs, with the exception of IL-13, which decreased significantly following all Ab treatments (Fig. 5E, 5F). Neutralization of IL-17A, IL-13, and both cytokines inhibited transcription of Il13 in mice with “high” Th2/Th17 allergic airway disease (Fig. 5G). In addition, anti–IL-13 treatment, alone or in combination with IL-17A neutralization, resulted in decreased transcription of Il6 in the lungs of these mice (Fig. 5G). Pulmonary gene expression of Il4 and Ifng was not affected by neutralization of IL-13 and/or IL-17A in mice with “high” Th2/Th17 allergic airway disease. These data suggest that IL-13 and/or IL-17A are only partially required for airspace and tissue inflammation in “high” Th2/Th17 allergic airway disease in mice.

Because we observed little change in inflammation, and previous studies showed that inflammation and AHR may not be mechanistically linked (3), we investigated AHR to increasing doses of methacholine, as well as quasi-static lung compliance and hysteresis in mice with “high” Th2/Th17 allergic airway disease. Adoptive transfer of Th2 and Th17 cells significantly increased Rn, G, and H in response to methacholine, and neutralization of IL-17A alone had minimal effect on Rn, G, and H in Th2/Th17–cell transferred, OVA-challenged mice (Fig. 6A–C). However, anti–IL-13 treatment alone was sufficient to significantly decrease Rn, G, and H (Fig. 6B, 6C), whereas neutralization of both IL-13 and IL-17A resulted in a further decrease in Rn in these mice (Fig. 6A). Th2 and Th17 cell transfer also significantly decreased quasi-static lung compliance (Fig. 6D) and significantly increased hysteresis (Fig. 6E) compared with control mice (Fig. 4). In our previous studies, quasi-static compliance correlated with cellular inflammation in the lung (3); therefore, as expected, we observed that neutralization of IL-13 and IL-17A had an insignificant effect on pulmonary compliance and hysteresis (Fig. 6D, 6E). These data indicate that IL-13 and, to a lesser extent, IL-17A may be necessary for the development of AHR in Th2/Th17-dominant allergic airway disease, despite little impact on lung stiffening.

We next determined the impact of IL-13 and/or IL-17A neutralization on mucus metaplasia. Th2/Th17 cell adoptive transfer resulted in increased expression of Clca3, Muc5ac, and Muc5b in the lungs compared with control mice (Fig. 7A–C). Although neutralization of IL-17A significantly reduced Clca3 gene expression in the lungs of Th2/Th17 cell–transferred, OVA-challenged mice, blocking IL-17A significantly elevated Muc5ac and did not alter pulmonary Muc5b expression compared with IgG-treated Th2/Th17 cell–transferred, OVA-challenged mice. Notably, anti–IL-13 treatment alone or anti–IL-13/anti–IL-17A treatment significantly reduced gene expression of Clca3, Muc5ac, and Muc5b compared with IgG-treated Th2/Th17 cell–transferred, OVA-challenged mice (Fig. 7A–C). Similar results were observed when mucus production was assessed by histological scoring of PAS⁺ bronchioles (Fig. 7D, 7E). These data suggest that IL-13 critically regulates mucus metaplasia in this “high” Th2/Th17 model of asthma.

Having observed that the neutralization of IL-13 could alter AHR and not classical Th2 and Th17 pulmonary inflammation, we investigated additional immunoregulatory cytokines that are known to influence Th2 and Th17 immunity. Specifically, we measured pulmonary expression of IL-19, IL-25, and IL-33, which are known to promote Th2 immune responses, as well as IL-10, which is known to limit Th1 and Th2 immune responses (22–25). Expression of IL-19, IL-25, IL-33, and IL-10 was attenuated by anti–IL-13 and anti–IL-13/anti–IL-17A treatment in Th2/Th17 cell–transferred, OVA-challenged mice compared with IgG-treated Th2/Th17 cell transferred, OVA-challenged mice (Fig. 8A–D). Further, anti–IL-17A treatment alone did not significantly alter gene expression of these cytokines in the lung (Fig. 8A–D). Overall, these results show that neutralizing IL-13 may alter the expression of cytokines known to be influential in the regulation of Th cell immune responses in mice with “high” Th2/Th17 allergic airway disease.

To determine whether these Th2-promoting cytokines themselves could promote Th2/Th17-dominant allergic airway disease downstream of IL-13 and/or IL-17A, IL-19 and IL-33 were overexpressed in the lungs of Th2/Th17 cell–transferred, OVA-challenged mice following IL-13 and IL-17A neutralization. Both IL-19 and IL-33 were unable to worsen inflammation or AHR in these mice (Figs. 9, 10). Further, we assessed STAT expression to determine downstream effectors of the signaling pathway mediated by these Th2-promoting cytokines. STAT3 was not affected by anti–IL-13 or anti–IL-13/anti–IL-17A treatment in mice with “high” Th2/Th17 allergic airway disease (Fig. 8E). Therefore, the reduction in AHR and mucus metaplasia observed after IL-13 and/or IL-17A neutralization in Th2/Th17 cell–transferred, OVA-challenged mice is likely not due to decreased Th2-promoting cytokines or STAT3 expression.

AHR and mucus hypersecretion in Th2-dominant asthma models are known to be STAT6 dependent (26). Because STAT6 is stimulated by IL-4 and IL-13 binding to IL-4R (27), it is possible that limiting IL-13 would alter STAT6 activation in the lung, a potential mechanism by which IL-13 neutralization limits AHR and mucus metaplasia in Th2/Th17-dominant allergic airway disease. Western blot analyses for p-STAT6 revealed that there was less STAT6 activation following neutralization of IL-13 alone and in combination with IL-17A in mice with “high” Th2/Th17 allergic airway disease (Fig. 11B, 11C). Because IL-13 and IL-17A may regulate cytokine levels by impacting translation and secretion, we also examined the effect of anti–IL-13 and anti–IL-13/anti–IL-17A treatment on Stat6 mRNA levels in the lungs and found that neutralization of IL-13 alone or in combination with IL-17A inhibited Stat6 transcription in the lungs of mice with “high” Th2/Th17 allergic airway disease (Fig. 11A). Overall, these results show that IL-13 is necessary for Stat6 gene expression and phosphorylation in “high” Th2/Th17-mediated disease.

To determine whether STAT6 signaling mediates airway inflammation in Th2/Th17-induced allergic airway disease, Th2 and Th17 cells were adoptively transferred into OVA-challenged wild-type BALB/cJ (WT) or STAT6^−/− mice. Adoptive transfer of Th2 and Th17 cells into OVA-challenged WT and STAT6^−/− mice increased inflammatory cell recruitment into the lungs equally (Fig. 12A). Differential counting of BALF cells revealed that Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice and WT mice had similar levels of neutrophils, macrophages, and lymphocytes in the airspaces (Fig. 12B). However, Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice had significantly fewer eosinophils in their BALF compared with their WT counterparts (Fig. 12B). Histological analyses of the lung also confirmed that the peribronchial and parenchymal-associated inflammation were similar in Th2/Th17 cell–transferred, OVA-challenged WT and STAT6^−/− mice, but perivascular inflammation was significantly reduced in the STAT6^−/− mice (Fig. 12C, 12D). WT and STAT6^−/− mice with “high” Th2/Th17 allergic airway disease had similar protein levels of IL-13, IL-6, and CXCL1 in the lungs (Fig. 12E, 12F). Protein levels of IL-4 and IL-5 were significantly decreased, whereas G-CSF levels were significantly increased, in the lungs of Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice compared with WT mice (Fig. 12E, 12F). Il13 and Il4 expression was lower in the lungs of Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice compared with wild-type mice, whereas Il6 and Ifng levels were similar between the groups (Fig. 12G). Similar to Fig. 3, the difference in the mRNA and protein levels of IL-13 may be explained by the timing of the measurement. It is likely that mRNA expression precedes protein expression changes; thus, the decrease in Il13 expression observed may be predictive of a decrease in protein at a later time points than assessed in this study. It is also possible that differential posttranscriptional regulation of IL-13 expression occurs in our model. Overall, Th2-induced eosinophilia was reduced, but it had little effect on overall airspace and tissue inflammation in “high” Th2/Th17-induced allergic airway disease in mice.

To determine whether STAT6 is necessary for the development of AHR, Rn, G, and H in response to methacholine were measured in Th2/Th17 cell–transferred, OVA-challenged WT and STAT6^−/− mice (Fig. 13A–C). STAT6^−/− mice had significantly less G and H in response to methacholine compared with WT mice (Fig. 13B, 13C). However, deletion of STAT6 was not sufficient to alter Rn (Fig. 13A). Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice had similar quasi-static lung compliance (Fig. 13D) and significantly decreased hysteresis (Fig. 13E) compared with their WT counterparts. Together, these results show that STAT6 is required for the development of peripheral lung AHR in Th2/Th17-induced allergic airway disease.

To determine whether STAT6 is necessary for the development of mucus metaplasia in Th2/Th17-dominant allergic airway disease, Clca3, Muc5ac, and Muc5b gene expression in the lungs was measured. Th2/Th17 cell adoptive transfer in OVA-challenged STAT6^−/− mice resulted in significantly decreased pulmonary expression of Clca3, Muc5ac, and Muc5b compared with their WT counterparts (Fig. 13F). Similar results were obtained using histological analyses of mucus production from PAS-stained lung tissue (Fig. 13G, 13H). Overall, because these STAT6^−/− mice received WT in vitro–differentiated Th cells, these results further indicate that STAT6 expression, specifically in the lung, is necessary for the development of AHR and mucus metaplasia in “high” Th2/Th17-induced allergic airway disease.

Asthma is a chronic inflammatory disease of the airways that represents an increasingly significant health burden worldwide. Although the heterogeneity of this condition is well-established, mechanistic studies focused on linking underlying disease pathogenesis, steroid responsiveness, and disease phenotypes are limited. Elucidating the underlying pathophysiologic mechanisms of disease and identifying and characterizing specific asthma endotypes are essential to the development of targeted and, ultimately, more effective therapeutic approaches. The current study elucidates which hallmark phenotypes associated with allergic airway disease (inflammation, mucus hypersecretion, and AHR) are specifically controlled by IL-13 and/or IL-17A in a model of steroid-resistant asthma with a mixed T cell phenotype. Ultimately, we determined that AHR and mucus metaplasia are controlled by IL-13 in a STAT6-dependent fashion in steroid-resistant, Th2/Th17-dominant allergic airway disease, which significantly adds to the current understanding and establishment of this asthma endotype.

AHR is a defining clinical feature of asthma; however, the mechanisms underlying its development are still not fully understood. Research has focused on understanding the effects of TNF-α, IL-13, IL-17A, and other cytokines on mediating immune responses in asthma (28–30). Because IL-13 and IL-17A are recognized as important effectors in asthma, we hypothesized that limiting IL-13 and IL-17A in a Th2 and Th17–dominant model of allergic airway disease would alter lung responses to allergen. Surprisingly, blocking IL-13, IL-17A, or both cytokines did not alter airspace or tissue inflammation in “high” Th2/Th17 allergic airway disease, but it did attenuate AHR and mucus metaplasia. Specifically, we observed that quasi-static lung compliance is associated with airspace inflammation (Fig. 3A), similar to our previous findings in a Th17-induced, steroid-resistant allergic airway disease model (3). However, we discovered that the relationship between tissue inflammation and AHR is more complex in “high” Th2/Th17 allergic airway disease than in a solely Th17-induced disease. AHR was attenuated by anti–IL-13 and/or anti–IL-17A treatment in mice with “high” Th2/Th17 allergic airway disease (Fig. 4A–C), whereas soluble factors in the lung, as well as tissue inflammation, remained unchanged (Fig. 3). Our findings indicate that IL-13 may be influential for the development of tissue-related AHR, whereas both IL-13 and IL-17 may contribute to Rn in “high” Th2/Th17 disease.

Clinically, IL-13 and IL-17A have been targeted for treatment of asthma with differing efficacy, depending on disease phenotype. Although murine studies suggest that IL-4Rα antagonists have therapeutic potential (31), pitrakinra, an rIL-4 variant that competitively inhibits IL-4Rα to interfere with IL-4/IL-13 signaling, was only successful in reducing exacerbations and improving asthma symptoms in patients with eosinophilia and not in a broad population of asthmatics (32). More recently, another study by Wenzel et al. (33, 34) showed that dupilumab, a human mAb to IL-4Rα, was associated with fewer exacerbations in patients with persistent, moderate-to-severe asthma and elevated eosinophil levels who were taking inhaled corticosteroids. However, dupilumab improved lung function and reduced levels of Th2-associated inflammatory markers when inhaled glucocorticoids and long-acting β-agonists were withdrawn in these patients. Although the therapies are well tolerated, anti–IL-13 therapy did not improve asthma control, pulmonary function, or exacerbations in patients with severe, steroid-insensitive asthma (35). Another study by Corren et al. (36) showed that the anti–IL-13 therapy lebrikizumab was associated with improved lung function in asthmatics with high serum levels of periostin. Taken together, these studies suggest that inhibition of IL-4/IL-13 signaling may be beneficial, but only for certain asthmatics. Similarly to IL-13–related therapies, there was no effect of brodalumab, a human anti–IL-17RA mAb, on primary or secondary disease endpoints in subjects with inadequately controlled moderate to severe asthma taking inhaled corticosteroids (37). These clinical results are not surprising because mAb therapies focus on blocking a specific mechanism of disease, and asthma is a heterogeneous disease with various clinical phenotypes that may respond to only specific biological therapies. Therefore, these studies also support the need for targeting of interventions to distinct patient populations with disease most likely to respond to therapy.

Although pulmonary inflammation may not necessarily contribute to AHR, effector T cell differentiation is directed by the cytokine milieu. Additionally, STAT proteins are known to be engaged in cytokine signaling transduction and critically regulate the differentiation of Th1, Th2, and Th17 cell subsets (38). Specifically, STAT6 is stimulated by IL-4 and IL-13 binding to IL-4R, which then stimulates the development of Th2 cells (27, 39). Our results show that blocking IL-13, alone or in combination with IL-17A, in a Th2/Th17-dominant allergic airway disease model resulted in less AHR, mucus metaplasia, and STAT6 activation. In addition, although Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice had less mucus metaplasia, Rn was not rescued compared with WT controls. Overall, this study suggests that IL-13 blockade may be beneficial for improving AHR in a mouse model of Th2/Th17-dominant allergic airway disease but is not sufficient to limit airspace or tissue inflammation in the lungs.

It was established that IL-13 and IL-4 have distinct roles and can contribute individually to experimental asthma in mice (40, 41). Although IL-4 is present, STAT6 appears to be IL-13 dependent in this Th2/Th17-dominant murine model of allergic airway disease. Indeed, other studies linked IL-17A and IL-13 to STAT6 activation and showed that STAT6 signaling is necessary for the development of allergic asthma. Early studies showed that mice deficient in STAT6 had significantly less pulmonary eosinophilia, mucus metaplasia, and AHR compared with wild-type mice following OVA sensitization and challenge (42, 43). Newcomb et al. (44) showed that STAT6^−/− mice adoptively transferred with OVA-specific Th17 cells and treated with OVA had increased IL-17A and IL-13 protein compared with control-treated mice. Another study showed that eosinophilia, AHR, and mucus hypersecretion were also STAT6 dependent in Th2-dominant allergic airway disease (26). Kuperman et al. (45, 46) also determined that STAT6 signaling is necessary and sufficient for mucus metaplasia and AHR in response to IL-13. Specifically, increased Muc5ac expression in the lungs was identified as a central feature of Ag-induced mucus metaplasia, whereas Muc5b is required for mucociliary clearance and maintaining immune defenses in the lung (47, 48). In the current study, we observed that blocking IL-13, alone or concurrently with IL-17A, significantly minimized Muc5ac expression in the lungs of mice with “high” Th2/Th17 disease, whereas blocking IL-17A alone increased Muc5ac expression. In this murine model, it is possible that blocking IL-17A shifts the environment in the lungs to a more Th2 phenotype during “high” Th2/Th17 disease either by altering the cytokine milieu and/or by directly mediating T cell plasticity in vivo. Because previous work showed that BALB/c SCID mice adoptively transferred with Th2 cells have more mucus metaplasia than those with Th17 cell transfer (16), this may be one mechanism by which we observed increased Muc5ac in the lungs of mice with “high” Th2/Th17 disease following IL-17 neutralization. Together, this previous work and the current findings suggest that blocking IL-13 to prevent STAT6 activation may be one mechanism by which AHR and mucus metaplasia are reduced in Th2/Th17-dominant allergic airway disease. These findings further call into question the relative roles of inflammation in mediating altered lung physiology in asthma.

Localization of STAT6 may affect its regulatory role in allergic airway disease. Specifically, our results show that pulmonary expression of STAT6 is necessary for the development of eosinophilia, Th2 cytokine expression, peripheral airway mechanics, and mucus production, but not neutrophilia, Th17 cytokine expression, and Rn in “high” Th2/Th17 allergic airway disease. Previous work also showed that goblet cell metaplasia and mucus production in an allergic asthma in mice are dependent upon expression of STAT6 in the lung tissue (49).

It is possible that there are other contributing factors to Rn responses seen in our Th2/Th17 cell–transferred, OVA-challenged STAT6^−/− mice. STAT6-expressing inflammatory dendritic cells in the lungs play an important role in the induction of Rn, independent of cellular inflammation (50). Another group found that STAT6^−/− mice had significantly higher numbers of regulatory T cells compared with wild-type mice (49) and that this STAT6-mediated suppression of regulatory T cells may contribute to the development of allergic airway inflammation (51). Type 2 innate lymphoid cells were recently reported to contribute to allergic asthma pathogenesis, and studies showed that the function of ILCs in Th2-dominant disease is independent of STAT6 (52, 53). Our data suggest that eosinophilia, mucus metaplasia, and tissue-related AHR are dependent on STAT6. Using transgenic mice with selective expression of IL-4Rα on smooth muscle, Perkins et al. (54) demonstrated that activation of smooth muscle by IL-4, IL-13, or allergen is sufficient, but not necessary, to induce AHR. Further, they demonstrated that IL-13 could promote expression of genes important for smooth muscle cell migration, proliferation, and contractility (54). Ultimately, there are likely multiple mechanisms that control AHR.

In our study, we used a mixed Th2 and Th17 model of allergic airway disease, whereas the majority of published studies focused on allergen sensitization and challenge models and disease driven exclusively by one distinct population of Th cell. Recent work is supportive of our findings in this murine model of mixed immune phenotype, but these studies did not directly assess AHR or investigate the molecular pathways underlying “high” Th2/Th17 disease (55). To our knowledge, STAT6-dependent mechanisms of asthma have never been investigated in a mixed “high” Th2/Th17, steroid-resistant allergic airway disease model. Furthermore, most studies of allergic airway disease in mice do not compare or report responses for central versus peripheral Rn. Our results show that the development of AHR is cellular and tissue inflammation independent and that IL-13, via STAT6 signaling, is a dominant driving force of AHR and mucus metaplasia in “high” Th2/Th17 allergic airway disease in mice. This work significantly contributes to the understanding of how AHR is differentially regulated and further clarifies the underlying molecular mechanism of disease that contribute to asthma pathogenesis. Overall, identification of a patient’s asthma endotype is essential for the implementation of successful biological therapies.

We thank the laboratory of Dr. Jay K. Kolls for providing the Western blotting equipment used for this work and Waleed Elsegeiny for technical assistance with the equipment.

The IL-13– and IL-17A–neutralizing Abs used in these studies were provided by Janssen Research & Development, LLC to J.F.A. under a research funding agreement with the University of Pittsburgh. P.L.D. and M.M.E. are employees of Janssen Research & Development, LLC.