Asthma and chronic obstructive pulmonary disease (COPD) are common chronic airway diseases that frequently coexist, and together they are referred to specifically as asthma–COPD overlap (ACO). Because asthma and COPD are heterogeneous diseases, ACO pathogenesis is more complex and its immune mechanisms and optimal treatments remain largely unknown. Interleukin 33 (IL-33) is an epithelium-derived pro-type 2 cytokine that functions as an alarmin. We previously reported that the sputum of patients with COPD or ACO contains high levels of IL-33.¹ Alternaria alternata, an airborne allergen, triggers IL-33-induced asthma-like type 2 airway inflammation by activating group 2 innate lymphocyte (ILC2).² We hypothesized that IL-33-derived inflammation is involved in ACO pathogenesis. We established a type-2 airway inflammation and pulmonary emphysema overlap model by combining Alternaria-derived type 2 innate airway inflammation and porcine pancreatic elastase (PPE)-induced pulmonary emphysema to test this hypothesis. We also cultured normal human bronchial epithelial (NHBE) cells to assess IL-33 expression and release after exposure to neutrophil elastase (NE).

We first confirmed that the emphysema model developed pulmonary emphysema (Supporting Information S1: Figure S1A–C). IL-33 was upregulated in the lungs from Days 1 to 3 (Supporting Information S1: Figure S1D). Most of the IL-33 was intracellular full-length and the remainder was extracellularly processed (Supporting Information S1: Figure S1E, S1F). We also assessed immune cell counts and cytokine levels in the bronchoalveolar lavage (BAL) fluid and lung lysate (Supporting Information S1: Figure S2A, S2B). In culture experiments, human bronchial epithelial cells (Table E1) exhibited stable intracellular IL-33 expression (Supporting Information S1: Figure S1G). NE promoted both the expression and release of IL-33 (Supporting Information S1: Figure S1H). These findings indicate that NE upregulate intracellular IL-33 in the lungs, and this is partially released in the airspaces.

We next hypothesized that the emphysema model would release more IL-33 in the alveoli after Alternaria alternata inhalation. We compared inflammation patterns in the type-2 airway inflammation (A. alternata), emphysema (PPE), and overlap models (both) (Figure 1A and Supporting Information S1: Figures S3A, S3B). The emphysema and overlap models developed pulmonary emphysema (Figure 1B, C). Lung IL-33 concentrations were higher in the emphysema model than in the control model from hour 1 to 7 (total IL-33, Days 1–3; full-length IL-33, Days 3–7; processed IL-33, Hour 1 to Day 3, Supporting Information S1: Figure S1D–F), but this was no longer the case on day 10 (Figure 1D). Lung IL-33 levels (total, full-length, and processed) were higher in the overlap model than in the others (Figure 1D). Paradoxically, the overlap model showed lower IL-4, IL-5, and IL-13 levels in the BAL fluid than the type-2 airway inflammation model (Figure 1E). We then suspected that ILC2-antagonizing cytokines² were upregulated in the overlap model lung. As expected, interferon-γ (IFNγ), but not IL-12 or IL-27, was upregulated in the overlap model lung (Figure 1F). This indicates that IL-33 and IFNγ may positively and negatively regulate, respectively, type 2 airway inflammation in our overlap model.

A previous study reported that the papain-induced ACO model develops allergic asthma.³ Another study demonstrated that the PPE and ovalbumin-induced ACO model also develops allergic asthma.⁴ Our overlap model used the previously reported A. alternata administration schedule.² This administration schedule increases the number of ILC2 and CD4⁺ T cells containing IL-5 and -13 intracellularly in BAL fluid.² Thus, our ACO model may have innate and allergic asthma aspects.

More importantly, our overlap model showed that type 2 airway inflammation is stronger than in the emphysema model and weaker than in the type-2 airway inflammation model. These results are consistent with clinical observations that ACO patients exhibit higher fractional exhaled nitric oxide (FeNO) levels than COPD patients and lower FeNO levels than asthmatics^5-7, because FeNO is a biomarker of type-2 inflammation⁸ and IL-13 derives inducible nitric oxide synthase production in human bronchial epithelial cells.⁹ Our overlap model had upregulated IL-33 and IFNγ levels in the lungs, and these positively and negatively regulate ILC2, respectively.² These results suggest that IL-33 and IFNγ may mediate FeNO levels in ACO patients.

Our overlap and innate type-2 airway inflammation models exhibited upregulated IL-33 release in the alveoli. Monoclonal anti-IL-33 antibodies reduce exacerbation in patients with moderate-to-severe COPD¹⁰ or moderate-to-severe asthma.¹¹ Thus, IL-33-targeting therapies may reduce exacerbation in patients with ACO (i.e., COPD plus innate asthma), although further investigations are required.

This study was subject to some limitations. PPE does not trigger all the physiological events associated with cigarette smoke.¹² Thus, our PPE-induced emphysema model was not fully relevant to human COPD. We did not assess airway hypersensitivity in our mouse model after A. alternata administration. Thus, this was not an asthma model but an asthma-like type 2 airway inflammation model.

In conclusion, our overlap model shows airway inflammation relevant to clinical ACO, and we found that IL-33 is a possible therapeutic target in patients with ACO.

AUTHOR CONTRIBUTIONS

Chika Miyaoka: Conceptualization (equal); data curation (equal); formal analysis (equal); investigation (equal); resources (equal); validation (equal); visualization (equal); writing—original draft preparation (equal). Masato Watanabe: Conceptualization (equal); data curation (supporting); funding acquisition (lead); investigation (equal); methodology (lead); project administration (lead); supervision (lead); writing—original draft preparation (equal); writing—review & editing (lead). Keitaro Nakamoto: Conceptualization (supporting); formal analysis (equal); resources (equal); validation (equal). Yuki Yoshida: Data curation (equal); formal analysis (equal); resources (supporting). Aya Hirata: Methodology (supporting); resources (supporting). Jumpei Aso: Data curation (equal); validation (supporting). Hiroki Nunokawa: Validation (supporting). Manabu Ishida: Formal analysis (supporting); investigation (supporting). Kojiro Honda: Formal analysis (supporting). Saori Takata: Investigation (supporting). Takeshi Saraya: Investigation (supporting); writing—review & editing (supporting). Haruyuki Ishii: Writing—review & editing (supporting).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

ETHICS STATEMENT

The Experimental Animal Ethics Committee of Kyorin University (No. 236).