You are here

  • Home
  • Online First
  • Role of the IL-33/ST2 axis in cigarette smoke-induced airways remodelling in chronic obstructive pulmonary disease

Article Text

Article menu

other Versions

Chronic obstructive pulmonary disease
Original research
Role of the IL-33/ST2 axis in cigarette smoke-induced airways remodelling in chronic obstructive pulmonary disease
Free
  1. Qiong Huang1,
  2. Chen Duo Li1,
  3. Yi Ran Yang1,
  4. Xiao Feng Qin1,
  5. Jing Jing Wang2,
  6. Xin Zhang1,
  7. Xiao Nan Du1,
  8. Xia Yang3,
  9. Ying Wang4,
  10. Lun Li5,
  11. Mi Mu6,
  12. Zhe Lv1,
  13. Ye Cui1,
  14. Kewu Huang4,
  15. Chris J Corrigan7,
  16. Wei Wang1,
  17. Sun Ying1
  1. 1Department of Immunology, School of Basic Medical Sciences, Capital Medical University, Beijing, China
  2. 2Department of Laboratory Animal Sciences, Capital Medical University, Beijing, China
  3. 3Department of Respiratory and Critical Care Medicine, Tianjin Medical University General Hospital, Tianjin, China
  4. 4Department of Pulmonary and Critical Care Medicine, Beijing Chao-Yang Hospital, Capital Medical University, Beijing, China
  5. 5Department of Respiratory Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
  6. 6Department of Respiratory Medicine, the Eighth Medical Center, Chinese PLA General Hospital, Beijing, China
  7. 7Faculty of Life Sciences & Medicine, School of Immunology & Microbial Sciences, Asthma UK Centre in Allergic Mechanisms of Asthma, King's College London, London, UK
  1. Correspondence to Dr Sun Ying, School of Basic Medical Sciences, Department of Immunology, Capital Medical University, Beijing 100069, China; ying.sun{at}ccmu.edu.cn

Abstract

Background Efficient therapy and potential prophylaxis are confounded by current ignorance of the pathogenesis of airway remodelling and blockade in COPD.

Objective To explore the role of the IL-33/ST2 axis in cigarette smoke (CS) exposure-induced airways remodelling.

Methods C57BL/6, BALB/c and IL-1RL1-/- mice exposed to CS were used to establish an animal surrogate of COPD (air-exposed=5~8, CS-exposed=6~12). Hallmarks of remodelling were measured in mice. Cigarette smoke extract (CSE)-induced proliferation and protein production in vitro by fibroblasts in the presence of anti-interleukin-33 (anti-IL-33) or hST2 antibodies were measured. Expression of IL-33 and ST2 and other remodelling hallmarks were measured, respectively, in bronchoalveolar lavage fluid (BALF) (controls=20, COPD=20), serum (controls=59, COPD=90) and lung tissue sections (controls=11, COPD=7) from patients with COPD and controls.

Results Wild-type mice exposed to CS elevated expression of hallmarks of tissue remodelling in the lungs and also in the heart, spleen and kidneys, which were significantly abrogated in the IL-1RL1-/- mice. Fibroblasts exposed to CSE, compared with control, exhibited early cellular translocation of IL-33, accompanied by proliferation and elevated protein synthesis, all inhabitable by blockade of IL-33/ST2 signalling. Expression of IL-33 and ST2 and hallmarks of tissue remodelling were significantly and proportionally elevated in BALF, serum and tissue samples from patients with COPD.

Conclusions Exposure to CS induces remodelling changes in multiple organs. The data support the hypothesis that CS-induced lung collagen deposition is at least partly a result of CS-induced IL-33 translocation and release from local fibroblasts.

  • COPD àü mechanisms
  • COPD pathology
  • cytokine biology

https://doi.org/10.1136/thoraxjnl-2020-214712

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Key messages

    What is the key question?

    • Is interleukin-33 (IL-33) involved in the mechanisms by which cigarette smoke (CS) exposure causes remodelling in COPD?

    What is the bottom line?

    • IL-33 is implicated in aspects of CS-induced remodelling in COPD such as collagen deposition; blockade of the IL-33/ST2 axis abrogates this effect.

    Why read on?

    • Learn how anti-IL-33 strategies may provide a therapeutic and possibly prophylactic approach to COPD management.

    Introduction

    COPD is a heterogeneous disease characterised by persistent, irreversible airway obstruction and associated symptomatology. At present, COPD is the third leading cause of death worldwide, responsible for 3.2 million deaths annually, which is predicted to rise to 4.4 million annually by 2040.1 Recognised risk factors for the development of COPD include exposure to environmental smoke and other ambient air pollution, as well as early life factors such as chronic cough in childhood and underweight birth. Smoking cessation and reduction of environmental pollution by biomass limitation are major, effective preventive strategies.2 The main pathophysiological changes of COPD include severe airway obstruction associated with chronic airway inflammation and tissue remodelling (collagen deposition, smooth muscle thickening and angiogenesis), the later changes being irreversible and one of the principal causes of accelerated disease progression.3

    The mechanisms of airway remodelling are multifarious and complex. Many types of immune cell, such as T helper 2 (Th2) cells, M2 macrophages, Th17 cells and CD8+ T cells, as well as cytokines including transforming growth factor beta 1 (TGF-β1), interleukin-10 (IL)-10, IL-13, IL-17A and IL-9 have been implicated.4 Recently, attention has focused on the possible role of IL-33, one of the members of the IL-1 family, as a key driver of remodelling changes in the intestine,5 heart6 and lung.7 8 IL-33 is expressed principally not only by epithelial cells, but also by endothelial cells and fibroblast-like cells as well as some immune cells, including macrophages and neutrophils.9 IL-33 functions as an ‘alarmin’ cytokine, produced non-specifically in response to environmental stimuli, such as smoke exposure, causing local cellular injury which results in the release of IL-33 from nuclear stores. Thus, IL-33 and other alarmins play an important role in tissue homoeostasis in health as well as in tissue damage and restructuring in disease. It acts by binding to its unique receptors ST2 (encoded by the gene IL-1RL1) and IL-1 receptor accessory protein (IL-1RAcP).10–12 Numerous studies have implicated IL-33 in the pathogenesis of chronic airway disorders such as asthma and COPD, but the drivers of IL-33 release and the full range of its possible effects once released remain to be fully elucidated.13–15 In the present study, we address the hypothesis that airway remodelling caused by exposure to environmental cigarette smoke (CS) is at least partly mediated by IL-33 release, and that this release is not confined to the airways.

    Methods

    See online supplemental methods for a more detailed version.

    Subjects

    Lung tissue was obtained, with written, informed consent, from patients with COPD and non-COPD controls undergoing lung resection surgery for the removal of tumours in Chao-Yang Hospital, Capital Medical University, Beijing, China. Tissue was fixed with 10% formaldehyde and then embedded in paraffin. Serum and bronchoalveolar lavage fluid (BALF) samples were collected, with written, informed consent, from patients with COPD and controls (via health physical examination personnel) in Chao-Yang Hospital, Capital Medical University, Beijing, China, and stored at −80℃ until used. The criterion of COPD was defined as a post-bronchodilator forced expiratory volume in 1 s/forced vital capacity ratio (post-FEV1/FVC%)<70%. The clinical characteristics of the subjects involved in the study are shown in tables 1–3.

    View this table:
    Table 1

    Clinical characteristics of patients providing lung tissue

    View this table:
    Table 2

    Clinical characteristics of serum donors

    View this table:
    Table 3

    Clinical characteristic of BALF donors

    Animal surrogate

    C57BL/6, BALB/c and IL-1RL1-/- (BALB/c background) (male, 6–8 weeks old) mice were used to establish a murine surrogate of smoking-induced COPD using a Cigarette Smoke Generator (TSE systems, Germany) and Inhalation Towers (EMMS, UK). The schedule of challenge of the animals with CS is shown in figures 1A and 2A.

    Figure 1
    Figure 1

    COPD-like changes in the lung of mice exposed to cigarette smoke (CS). (A) Schedule of murine challenge. (B–D) Dot plots showing body weight, inspiratory capacity (IC) and right ventricular hypertrophy index (RVHI) of mice exposed to air (Con) or CS. (E–I) Dot plots showing the number of total cells, macrophages, neutrophils, eosinophils and lymphocytes in bronchoalveolar lavage fluid (BALF) of mice. (J–K) Representative photomicrographs showing H&E-stained sections of the lung tissue from mice, indicating inflammatory cellular infiltration and alveolar destruction (measured by calculating mean linear intercept (MLI)). Scale bars: 100 µm. The data are expressed as mean±SEM (n≥5).

    Figure 2
    Figure 2

    Blockade of IL-33/ST2 signalling inhibited cigarette smoke (CS)-induced remodelling in multiple organs. (A) Schedule of murine challenge. (B, C) Dot plots showing body weight and number of total cells in bronchoalveolar lavage fluid (BALF) of wild-type (WT) and IL1RL1-/- mice exposed to air (Con) or CS. (D–F) Representative photomicrographs showing Masson-stained sections of spleen, lung and kidney of WT and IL1RL1-/- mice exposed to air or CS. (G) Immunoreactivity for α-smooth muscle actin (α-SMA) (brown) in sections of heart of WT and IL1RL1-/- mice exposed to air or CS. (H) Immunoreactivity for CD31 (brown) in sections of lung of WT and IL1RL1-/- mice exposed to air or CS. Scale bars: 100 µm. The results are expressed as mean±SEM (n≥5).

    Statistical analysis

    Data were expressed as mean±SEM and analysed using GraphPad Prism V.8 (La Jolla, California, USA). Statistical comparison between two groups was performed using a t-test, between more than two groups using one-way analysis of variance (ANOVA) and involving two different categorical independent variables using two-way ANOVA. The data from human subjects were analysed using the Mann-Whitney U test and the sex distributions compared using Fisher’s exact test. The Pearson coefficient was used to examine the association between FEV1/FVC and expression of IL-33 and hallmarks of tissue remodelling of donors providing the lung tissue. A p value <0.05 was considered as statistically significant.

    Results

    COPD-like changes, including airway inflammation, airways remodelling and emphysematous destruction, developed in mice following exposure to CS for 5 months

    Exposure of the experimental animals to CS was associated with a significant reduction in the mean body weight (33.43±0.6711 vs 26.70±0.3390, p<0.0001) (figure 1B) accompanied by significant increases in the mean respiratory inspiratory capacity (IC) (0.9578±0.02505 vs 1.077±0.03910, p=0.0328) (figure 1C) and right ventricular hypertrophy index (24.94±1.336 vs 32.30±1.535, p=0.0030) (figure 1D) compared with control, air-exposed animals. In addition, the mean total number of cells (8.21×104±1.83×104 vs 26.29×104±3.325×104, p=0.0025), macrophages (1.738×104±0.4788×104 vs 8.449×104±0.9834×104, p=0.0025) and neutrophils (5.660×104±1.324×104 vs 16.11×104±2.834×104, p=0.0177) in the BALF of the CS-exposed animals were significantly elevated compared with the controls, whereas the mean numbers of eosinophils (0.06112×104±0.03763×104 vs 0.02180×104±0.01410×104, p=0.4697) and lymphocytes (0.7505×104±0.2623×104 vs 1.707×104±0.5557×104, p=0.3434) were not significantly altered (figure 1E–I). Analysis of paraffin-embedded lung sections stained with H&E revealed significantly elevated mean indices of inflammatory cellular infiltration into the lung tissue (1.143±0.2384 vs 1.796±0.1207, p=0.0177), alveolar destruction (42.32±1.479 vs 50.89±1.418, p=0.0011) (figure 1J,K) and airway wall thickness (7.115±1.178 vs 12.60±1.154, p=0.0015), and a significantly decreased mean index of alveolar surface area (11.30±0.3096 vs 10.44±0.1548, p=0.0132) in the CS-exposed animals compared with the control animals (online supplemental figure 1A,B).

    Long-term CS exposure resulted in remodelling of multiple organs associated with IL-33 expression

    Histological (Masson) and immunohistochemical staining of tissue sections revealed that long-term, regular exposure of animals to CS compared with air control was associated with significant elevation of the mean global collagen deposition in the spleen (0.6230±0.09402 vs 1.348±0.1939, p=0.0177), lung (1.774±0.1041 vs 2.426±0.2178, p=0.0101) and kidney (1.099±0.2140 vs 2.455±0.3259, p=0.0025) (figure 3A), α-smooth muscle actin (α-SMA) deposition in the heart (0.2530±0.04677 vs 0.6152±0.06062, p=0.0025) (figure 3B) and expression of CD31, a marker of angiogenesis in the spleen (5.930±0.6986 vs 14.69±1.423, p=0.0025), lung (3.464±0.1181 vs 5.021±0.2040, p=0.0013) and kidney (2.183±0.1868 vs 3.232±0.2348, p=0.0088) (figure 4A). The mean concentrations of the cytokines IL-33 (11487±1564 vs 17230±1878, p=0.048) and IL-13 (52.40±3.469 vs 78.13±9.287, p=0.0051) were significantly elevated in the lung homogenates of the CS-exposed mice compared with the air exposed controls (figure 4E,F), while the mean concentrations of TGF-β1 (1665±950.6 vs 722.9±547.2, p=0.6098), IL-10 (4031±276.1 vs 3879±317.7, p=0.3434), IL-9 (6803±704.0 vs 6002±636.6, p=0.3434) and IL-17A (146.2±21.96 vs 167.5±38.46, p=0.8763) in the lung homogenates (figure 4B–D,G) as well as IL-33 in the BALF (6.959±0.9755 vs 8.574±1.447, p=0.4556) (online supplemental figure 2A) were not significantly altered. In addition, the mean number of fibroblasts expressing IL-33 (α-SMA+IL-33+ cells) (8.671±1.022 vs 13.82±0.4321, p<0.0001) in the lung tissue of the CS-exposed animals was significantly elevated compared with the controls, whereas the mean number of proliferating fibroblasts (α-SMA+PCNA+ cells) (1.823±0.8303 vs 2.853±0.3720, p=0.0617) was not significantly altered (online supplemental figure 3A,B).

    Figure 3
    Figure 3

    Long-term cigarette smoke (CS) exposure caused collagen deposition and smooth muscle hypertrophy in multiple organs. (A) Representative photomicrographs showing Masson-stained sections of the heart, liver, spleen, lung and kidney of mice exposed to air (Con) or CS. (B) Immunoreactivity for α-smooth muscle actin (α-SMA) (brown) in sections of the heart, liver, spleen, lung and kidney of mice exposed to air (Con) or CS. Scale bars: 100 µm. The results are expressed as mean±SEM (n≥5).

    Figure 4
    Figure 4

    Long-term cigarette smoke (CS) exposure was associated with elevated expression of CD31, a marker of angiogenesis, in the spleen, lung and kidney, and interleukin-33 (IL-33) and IL-13 in lung tissue homogenates. (A) Immunoreactivity for CD31 (brown) in sections of the heart, liver, spleen, lung and kidney of mice exposed to air (Con) or CS. Scale bars: 100 µm. (B–G) Concentrations of TGF-β1, IL-10, IL-9, IL-13, IL-33 and IL-17A in lung homogenates of mice exposed to air (Con) or CS measured by ELISA. The results are expressed as mean±SEM (n≥5).

    Blockade of IL-33/ST2 signalling alleviated CS-induced inflammation in the lung and abolished remodelling in multiple organs

    Exposure of the IL-1RL1-/- mice to CS did not significantly alter their mean body weight compared with air-exposed IL-1RL1-/- mice (27.1±0.7969 vs 26±0.9399, p=0.6536) (figure 2B). The mean total number of cells in the BALF of the CS-exposed IL-1RL1-/- mice was significantly reduced compared with the WT mice (17.33×104±2.394×104 vs 28×104±3.690×104, p=0.0104) (figure 2C). Compared with WT mice, exposure of IL-1RL1-/- mice to CS significantly abrogated collagen deposition in the spleen (2.377±0.2637 vs 0.9787±0.1511, p=0.0002) and lung (1.457±0.2536 vs 0.5040±0.09428, p=0.0015), and the expression of immunoreactivity for α-SMA in the heart (0.2914±0.01162 vs 0.1970±0.02568, p=0.0037) and CD31 in the lung tissues (14.13±0.6100 vs 11.75±0.4506, p=0.0394). For all of these outcomes, the effects of exposure of the IL-1RL1-/- mice to CS did not significantly differ from those observed in the air-exposed IL-1RL1-/- mice (figure 2D–H).

    Exposure to cigarette smoke extract modified proliferation, production of fibrosis-related proteins and IL-33 translocation in normal fibroblasts in vitro

    In vitro, exposure of both murine (MLg) and human (MRC5) lung fibroblast cell lines to high dilutions of cigarette smoke extract (CSE) significantly increased their mean proliferation, as measured by the CCK8 assay, compared with diluent control (MLg: 0.2% CSE 1.474±0.02686 vs 1.722±0.04153, p=0.0086 and 1% CSE 1.474±0.02686 vs 1.986±0.05532, p=0.0028; MRC5: 1% CSE 1.166±0.008161 vs 1.287±0.02652, p=0.0033) (figures 5A and 6A). Western blotting showed that exposure of MLg and MRC5 fibroblasts to 1% CSE was also associated with a significantly elevated mean production, compared with exposure to diluent control of collagen I (MLg: 0.6900±0.08638 vs 0.9397±0.08054, p=0.0195; MRC5: 0.8059±0.05683 vs 1.310±0.1449, p=0.0156), fibronectin (MLg: 0.6757±0.05414 vs 0.7442±0.04672, p=0.0322; MRC5: 0.7939±0.07247 vs 0.9825±0.1226, p=0.0469) and ST2 (MLg: 0.6609±0.06729 vs 0.8095±0.06286, p=0.0155; MRC5: 0.8317±0.03106 vs 1.111±0.09212, p=0.0263) (figures 5B and 6B). Laser scanning confocal microscopic analysis revealed that, at rest, both MLg and MRC5 cells expressed IL-33 predominantly in the nucleus, whereas as early as 4 hours after exposure to 1% CSE, immunoreactivity for IL-33 was apparent in the cytoplasm (figures 5C and 6C). Western blotting analysis confirmed a significant elevation in the mean IL-33/GAPDH ratio in the cytoplasm (MLg: 0.7963±0.1139 vs 0.8765±0.1028, p=0.0180; MRC5: 0.7504±0.05114 vs 0.8590±0.08286, p=0.0313) of both cell lines 12 hours after exposure to 1% CSE, with a trend for a corresponding reduction in the ratio in the nuclei (MLg: 0.5611±0.07377 vs 0.5236±0.1319, p=0.7490; MRC5: 0.5819±0.07219 vs 0.5053±0.07406, p=0.2188) (figures 5C,D and 6C,D), which had normalised within 48 hours after initiation of CSE exposure (MLg-cytoplasm: 0.8077±0.1172 vs 0.7493±0.08742, p=0.3243; MRC5-cytoplasm: 0.9215±0.02645 vs 1.031±0.04175, p=0.125) (MLg-nuclei: 0.5366±0.1891 vs 0.4806±0.1588, p=0.3230; MRC5-nuclei: 0.6748±0.1373 vs 0.6719±0.1523, p>0.9999) (figures 5C,E and 6C,E). In addition, the mean concentrations of IL-33 in the supernatants of the cultured MLg (4.609±1.215 vs 5.684±1.432, p=0.0469) and MRC5 (0.3294±0.08192 vs 0.6627±0.06207, p=0.0013) cells treated with 1% CSE were significantly elevated compared with diluent control (figure 7A,D).

    Figure 5
    Figure 5

    Cigarette smoke extract (CSE) stimulation increased proliferation, fibrosis-related proteins and ST2 production and affected interleukin-33 (IL-33) translocation in MLg cells. (A) Proliferation of MLg cells treated with CSE at various dilutions or diluent control, quantified with CCK8. (B) Western blotting showing expression of collagen I, fibronectin and ST2 immunoreactivity in MLg cells treated with medium alone (Con) or 1% CSE. (C) Representative confocal images show nucleus (blue) and immunoreactivity for IL-33 (green) in MLg cells treated with 1% CSE for 0, 4, 12, 24 and 48 hours (separate cultures). (D, E) Western blotting showing expression of IL-33 immunoreactivity in cytoplasmic and nuclear extracts of MLg cells treated with 1% CSE for 12 or 48 hours. Scale bars: 50 µm. The results are expressed as mean±SEM (n≥4).

    Figure 6
    Figure 6

    Cigarette smoke extract (CSE) stimulation increased proliferation, fibrosis-related proteins and ST2 production and affected interleukin-33 (IL-33) translocation in MRC5 cells. (A) Proliferation of MRC5 cells treated with CSE at various dilutions or diluent control, quantified with CCK8. (B) Western blotting showing expression of collagen I, fibronectin and ST2 immunoreactivity in MRC5 cells treated with medium alone (Con) or 1% CSE. (C) Representative confocal images show nucleus (blue) and immunoreactivity for IL-33 (red) in MRC5 cells treated with 1% CSE for 0, 4, 12, 24 and 48 hours (separate cultures). (D, E) Western blotting showing the expression of IL-33 immunoreactivity in cytoplasmic and nuclear extracts of MRC5 cells treated with 1% CSE for 12 or 48 hours. Scale bars: 75 µm. The results are expressed as mean±SEM (n≥4).

    Figure 7
    Figure 7

    Blockade of IL-33/ST2 signalling inhibited cigarette smoke extract (CSE)-induced proliferation and production of fibrosis-related proteins by fibroblasts. (A) Concentration of interleukin-33 (IL-33) in the supernatants of MLg cells treated with medium (Con) or 1% CSE, detected by ELISA. (B) Proliferation of MLg cells treated with medium (Con) or 1% CSE in the presence/absence of anti-IL-33 antibody, measured with CCK8. (C) Production of collagen Ⅰ and fibronectin by MLg cells treated with medium (Con) or 1% CSE in the presence/absence of anti-IL-33 antibody (1 ng/mL). (D) Concentration of IL-33 in the supernatants of MRC5 cells treated with medium (Con) or 1% CSE, detected by ELISA. (E) Proliferation of MRC5 cells treated with medium (Con) or 1% CSE in the presence/absence of anti-human ST2 antibody (hST2), measured with CCK8. (F) Production of collagen Ⅰ and fibronectin by MRC5 cells treated with medium (Con) or 1% CSE in the presence/absence of anti-hST2 antibody (3 µg/mL). The results are expressed as mean±SEM (n≥6).

    Blockade of IL-33/ST2 signalling abrogated CSE-induced proliferation of, and fibrosis-related protein production by fibroblasts

    Neutralisation of IL-33 and ST2 using specific blocking antibodies inhibited CSE-induced proliferation of MLg (1% CSE: 117.7±4.504 vs 1% CSE+1 ng/mL anti-IL-33: 108.8±4.887, p=0.0384) and MRC5 (1% CSE: 114.5±4.690 vs 1% CSE+1 µg/mL anti-hST2: 101.1±2.037, p=0.0441; 1% CSE: 114.5±4.690 vs 1% CSE+3 µg/mL anti-hST2 99.14±2.233, p=0.0441) cells (figure 7B,E). The mean CSE-induced production of collagen I (MLg-1% CSE: 1.036±0.05017 vs MLg-1% CSE anti-IL-33: 0.8827±0.07293, p=0.0453; MRC5-1% CSE: 0.7967±0.03815 vs MRC5-1% CSE+anti-hST2: 0.5921±0.02783, p=0.0060) and fibronectin (MLg-1% CSE: 0.9920±0.07168 vs MLg-1% CSE+anti-IL-33: 0.8288±0.1028, p=0.0500; MRC5-1% CSE: 0.8855±0.1294 vs MRC5-1% CSE+anti-hST2: 0.6850±0.1251, p=0.0323) by MLg and MRC5 cells was also significantly attenuated in the presence of anti-IL-33 or hST2 antibody (figure 7C,F).

    Increased expression of IL-33 and remodelling-related proteins in resected lung tissue from patients with COPD and controls, and association with lung function

    Immunohistochemistry and Masson staining were performed on sections of paraffin-embedded lung tissues obtained from patients with COPD and control patients. The data showed that the thickness of the smooth muscle surrounding the airways, as detected by mean global immunoreactivity for α-smooth muscle actin (α-SMA) (0.1255±0.02716 vs 0.7744±0.1756, p=0.0047) (figure 8A) and the degree of angiogenesis in the lung parenchyma, as detected by mean, global immunoreactivity for von Willebrand Factor (vWF)(15.23±1.881 vs 24.69±3.228, p=0.0225) (figure 8B) were both increased significantly in lung tissue sections of patients with COPD compared with the controls. In addition, the mean, global immunoreactivity for fibronectin (1.774±0.4584 vs 5.145±1.464, p=0.0204) and global Masson staining, reflecting collagen deposition (3.567±1.268 vs 8.735±2.839, p=0.0358), were also clearly and significantly elevated in the lung sections from patients with COPD compared with the controls (figure 8C,D). Finally, expression of mean global immunoreactivity for IL-33 (19.12±3.582 vs 68.33±9.153, p=0.0002), detectable both in the airway mucosa and the lung parenchyma, was significantly elevated in the lung tissue sections from patients with COPD compared with the controls (figure 8E).

    Figure 8
    Figure 8

    Global immunoreactivity for interleukin-33 (IL-33) and remodelling-related proteins increased in patients with COPD compared with controls. (A–C) Immunoreactivity for α-smooth muscle actin (α-SMA), von Willebrand Factor (vWF) and fibronectin (brown) in sections of resected lung tissues of patients with COPD and controls. (D) Representative photomicrographs of Masson-stained sections of resected lung tissues of patients with COPD and controls. (E) Immunoreactivity for IL-33 (brown) in sections of resected lung tissues of patients with COPD and controls. Scale bars: 100 µm. The results are expressed as mean±SEM.

    Correlation analysis revealed significantly negative correlations between FEV1/FVC, the standard index of airways obstruction in COPD and global immunoreactivity for vWF (r=−0.5625, 95% CI −0.8275 to −0.09264, p=0.0233), α-SMA (r=−0.7824, 95% CI −0.9209 to −0.4683, p=0.0003) and IL-33 (r=−0.6990, 95% CI −0.8919 to −0.2909, p=0.0037), but not fibronectin (r=−0.2980, 95% CI −0.6811 to 0.2132, p=0.2453) or collagen (r=−0.2733, 95% CI −0.6664 to 0.2387, p=0.2886) in sections of lung tissues of subjects with or without COPD (figure 9A–E). Furthermore, we observed clear, positive correlations between the mean expression of global immunoreactivity for IL-33 and that of fibronectin (r=0.6841, 95% CI 0.2851 to 0.8811, p=0.0035), vWF (r=0.5769, 95% CI 0.06673 to 0.8479, p=0.0308) and α-SMA (r=0.7768, 95% CI 0.4189 to 0.9258, p=0.0011) as well as global collagen staining (r=0.6638, 95%CI 0.2506 to 0.8724, p=0.0050) (figure 9F–I).

    Figure 9
    Figure 9

    Correlations between FEV1/FVC and mean, global expression of immunoreactivity/staining for remodelling-related proteins and interleukin-33 (IL-33). (A) Correlation between FEV1/FVC and global collagen deposition. (B–E) Correlations between FEV1/FVC and global immunoreactivity for fibronectin, von Willebrand Factor (vWF), α-smooth muscle actin (α-SMA) and IL-33. (F) Correlation between global immunoreactivity for IL-33 and collagen deposition. (G–I) Correlations between global immunoreactivity for IL-33 and fibronectin, vWF and α-SMA.

    Increased concentrations of IL-33 in BALF and ST2 in serum from patients with COPD

    Using ELISA, we measured the concentrations of IL-33 in the BALF and ST2 in the serum of two different groups of patients with COPD and controls undergoing health physical examinations. The mean BALF concentration of IL-33 (3.150±0.8418 vs 7.893±1.585, p=0.0061) as well as the mean serum concentration of ST2 (12178±687.3 vs 32006±4794, p=0.0013) were significantly elevated in patients with COPD as compared with the control subjects (figure 10A,B).

    Figure 10
    Figure 10

    Concentrations of (A) interleukin (IL)-33 in bronchoalveolar lavage fluid (BALF) from health physical examination personnel (n=20) and patients with COPD (n=20) and (B) ST2 in sera from health physical examination personnel (n=59) and patients with COPD (n=90) by ELISA.

    Discussion

    Airway remodelling, one of the major features of COPD, plays a major role in symptomatology and morbidity, but the precise mechanisms of this remodelling remain poorly characterised. In the present study, we explored the potential mechanisms of environmental CS-induced airway remodelling using an animal surrogate and in vivo and in vitro experiments.

    First, we observed the initiation of hallmarks of tissue remodelling not only in the lung tissue of our experimental animals exposed regularly to CS but also in other organs including the heart, liver, spleen and kidney. So far as we are aware, this is a previously undocumented phenomenon, although some other studies have reported the presence of altered microRNA species in livers, kidneys, spleen and heart in addition to the lungs in other animal surrogates exposed to CS16 and that mitochondrial function and redox homeostasis are compromised in these organs.17 Our data suggest that environmental exposures responsible for the pathogenesis of airway remodelling in COPD might initiate similar processes in remote organs, which could of course contribute to overall morbidity, most obviously the known, associated cardiovascular morbidity.18 19 Our data show that cardiac smooth muscle thickness and right ventricular hypertrophy were significantly elevated in our experimental animals following prolonged CS exposure, changes typical of pulmonary hypertension.20 In addition, collagen deposition and angiogenesis were markedly increased in the lungs, kidneys and spleen in the experimental mice, which might be associated with other clinically reported phenomena such as the relatively high incidence of acute kidney injury in patients with COPD, especially during exacerbation.21 Although some previous studies have reported collagen deposition in the kidneys of mice exposed to CS,22–24 we believe ours is the first to demonstrate evidence of collagen deposition in the spleen. Interestingly, we also observed brownish-black sediments in sections of both lung and spleen tissue of the experimental mice (photomicrographs shown in online supplemental figure 4), as commonly observed in smokers’ lungs. Further studies would be required to determine how these sediments form in the spleen and what, if any, are the functional consequences.

    It has been shown that fibroblasts cocultured with CSE-treated human bronchial epithelial cells or exosomes from these cells can differentiate into myofibroblasts.25–27 Our data show that, in vitro, CSE, at what one might consider a low concentration, being only a fraction (1%) of the potential stimulus delivered by smoking a single cigarette, accelerated the proliferation and production of collagen I and fibronectin by both murine and human lung fibroblasts. Furthermore, in vivo, the expression of IL-33 and IL-13 was significantly elevated in the lung homogenates of the experimental animals after CS exposure. We have previously shown that IL-33 may contribute directly to angiogenesis and collagen deposition in the lung, at least partly by acting on vascular endothelial cells and fibroblasts,7 8 while others have implicated IL-33 in the pathological changes of COPD.28–30 In the present study, we have further investigated the potential role and mechanisms of action of IL-33 in airway remodelling induced by an archetypal stimulus for COPD, namely, exposure to environmental CS. Our data with the IL-1RL1-/- mice clearly demonstrate that blockade of IL-33/ST2 signalling inhibited the remodelling-related changes induced by CS exposure in multiple organs, including collagen deposition in the spleen, lungs and kidneys, cardiac smooth muscle hypertrophy and angiogenesis in the airways of the experimental animals.

    Analysis by immunofluorescence and western blotting revealed that, in the ‘resting’ state, immunoreactivity for IL-33 was located principally in the nuclei of both the murine and the human lung fibroblast cell lines MLg and MRC5 in vitro, a finding consistent with previous reports.10 31 32 After exposure to 1% CSE, the principal location of the IL-33 immunoreactivity changed progressively over the ensuing 12 hours period from nucleus to cytoplasm and then back again within 48 hours, consistent with a process of transient, intracellular translocation of IL-33 from the cell nucleus to the cytoplasm, with subsequent reversal and resolution. Kunisch and colleagues have similarly reported that TNF-α induces translocation of IL-33 to the nucleus in synovial fibroblasts derived from patients with rheumatoid arthritis at 24 hours after stimulation.33 In contrast, Nile and colleagues have presented data suggesting that IL-33 protein is located principally in the cytoplasm of monocytes and becomes sequestered in the nucleus when the cells start to undergo apoptosis, and is not released into the extracellular milieu until cells undergo necrosis.34 Taken together, these data suggest that the mechanisms of translocation and release of IL-33 into the extracellular milieu differ according to the type of stimulus and the type of cell of origin. Nevertheless, we believe that this is the first report that exposure to CSE directs time-dependent translocation of IL-33 in pulmonary fibroblasts.

    We additionally observed that CSE exposure significantly increased the expression of the IL-33 receptor ST2 by the cultured pulmonary fibroblasts in vitro, as well as the mean concentration of IL-33 in the culture supernatants. Furthermore, blockade of IL-33 or ST2 using anti-IL-33 or hST2 antibodies at least attenuated proliferation and production of collagen I and fibronectin by MLg and MRC5 fibroblasts. Our data suggest that IL-33 promotes these effects by acting on its specific receptor ST2 expressed by the fibroblasts, potentially in an autocrine or paracrine fashion.

    In harmony with our animal and in vitro data, we noted marked upregulation of expression of immunoreactivity for IL-33 in the sections of lung tissue resected from patients with COPD compared with controls, to a degree which correlated inversely with the degree of airways obstruction (FEV1/FVC), but positively with the expression of a range of hallmarks of tissue remodelling, including collagen deposition and expression of immunoreactivity for fibronectin, vWF and α-SMA. It is also of interest that some of the control subjects in this study had been chronic smokers, suggesting that there exists interindividual susceptibility to CS-induced IL-33 release in the airways and progression to COPD through mechanisms yet to be identified. If our hypothesis that IL-33 is directly involved with remodelling changes in the airways in COPD is correct, one would expect its activity to correlate closely with the laydown of principal remodelling proteins such as collagen and fibronectin. Global airway obstruction, as reflected by the FEV1/FVC ratio at spirometry, is, on the other hand, likely to be influenced not only by remodelling but also airways mucosal narrowing from inflammation and mucus, and possible gas trapping from emphysematous changes, which is contributed to by many additional factors other than the remodelling changes investigated in the present study, and so would not be expected to correlate so clearly with IL-33 expression. Finally, we found that the mean concentrations of IL-33 in BALF and ST2 in serum were also clearly elevated in patients with COPD compared with healthy controls. Although for operational reasons we were unable to match the age ranges of our serum and BALF donors, and the sample size for the lung tissue is relatively small, we are aware of no precedent in the literature that the difference in age range of the serum and BALF sample donors might have confounded our data.

    In conclusion, we have uncovered that CS exposure in an animal surrogate induces tissue remodelling changes in the airways characteristic of COPD and additionally in other organs, and provided evidence that this is at least in part a result of IL-33 translocation in local fibroblasts and release into the extracellular milieu, which, in turn, stimulates autocrine proliferation and remodelling protein production by these cells. Consequently, targeting of the IL-33/ST2 axis may be a useful approach to retarding and even preventing airway remodelling in COPD and in addition reducing comorbidity in other organs.

    Acknowledgments

    We are most grateful to Professor Andrew N.J. McKenzie (the Medical Research Council Laboratory of Molecular Biology, Cambridge, UK) for kindly providing the IL-1RL1-/- mice.

    References

    1. ↵
      1. Celli BR,
      2. Wedzicha JA
      . Update on clinical aspects of chronic obstructive pulmonary disease. N Engl J Med 2019;381:1257–66.doi:10.1056/NEJMra1900500pmid:http://www.ncbi.nlm.nih.gov/pubmed/31553837
      OpenUrlPubMed
    2. ↵
      1. Wang C,
      2. Xu J,
      3. Yang L, et al
      . Prevalence and risk factors of chronic obstructive pulmonary disease in China (the China Pulmonary Health [CPH] study): a national cross-sectional study. Lancet 2018;391:1706–17.doi:10.1016/S0140-6736(18)30841-9pmid:http://www.ncbi.nlm.nih.gov/pubmed/29650248
      OpenUrlCrossRefPubMed
    3. ↵
      1. Agustí A,
      2. Hogg JC
      . Update on the pathogenesis of chronic obstructive pulmonary disease. N Engl J Med 2019;381:1248–56.doi:10.1056/NEJMra1900475pmid:http://www.ncbi.nlm.nih.gov/pubmed/31553836
      OpenUrlPubMed
    4. ↵
      1. Distler JHW,
      2. Györfi A-H,
      3. Ramanujam M, et al
      . Shared and distinct mechanisms of fibrosis. Nat Rev Rheumatol 2019;15:705–30.doi:10.1038/s41584-019-0322-7pmid:http://www.ncbi.nlm.nih.gov/pubmed/31712723
      OpenUrlCrossRefPubMed
    5. ↵
      1. Imai J,
      2. Kitamoto S,
      3. Sugihara K, et al
      . Flagellin-mediated activation of IL-33-ST2 signaling by a pathobiont promotes intestinal fibrosis. Mucosal Immunol 2019;12:632–43.doi:10.1038/s41385-019-0138-4pmid:http://www.ncbi.nlm.nih.gov/pubmed/30742042
      OpenUrlPubMed
    6. ↵
      1. Garbern JC,
      2. Williams J,
      3. Kristl AC, et al
      . Dysregulation of IL-33/ST2 signaling and myocardial periarteriolar fibrosis. J Mol Cell Cardiol 2019;128:179–86.doi:10.1016/j.yjmcc.2019.01.018pmid:http://www.ncbi.nlm.nih.gov/pubmed/30763587
      OpenUrlPubMed
    7. ↵
      1. Shan S,
      2. Li Y,
      3. Wang J, et al
      . Nasal administration of interleukin-33 induces airways angiogenesis and expression of multiple angiogenic factors in a murine asthma surrogate. Immunology 2016;148:83–91.doi:10.1111/imm.12589pmid:http://www.ncbi.nlm.nih.gov/pubmed/27035894
      OpenUrlPubMed
    8. ↵
      1. An G,
      2. Zhang X,
      3. Wang W, et al
      . The effects of interleukin-33 on airways collagen deposition and matrix metalloproteinase expression in a murine surrogate of asthma. Immunology 2018;154:637–50.doi:10.1111/imm.12911
      OpenUrl
    9. ↵
      1. Lee JH,
      2. Hailey KL,
      3. Vitorino SA, et al
      . Cigarette smoke triggers IL-33-associated inflammation in a model of late-stage chronic obstructive pulmonary disease. Am J Respir Cell Mol Biol 2019;61:567–74.doi:10.1165/rcmb.2018-0402OCpmid:http://www.ncbi.nlm.nih.gov/pubmed/30973786
      OpenUrlPubMed
    10. ↵
      1. Liew FY,
      2. Girard J-P,
      3. Turnquist HR
      . Interleukin-33 in health and disease. Nat Rev Immunol 2016;16:676–89.doi:10.1038/nri.2016.95pmid:http://www.ncbi.nlm.nih.gov/pubmed/27640624
      OpenUrlCrossRefPubMed
    11. ↵
      1. Gabryelska A,
      2. Kuna P,
      3. Antczak A, et al
      . Il-33 mediated inflammation in chronic respiratory Diseases—Understanding the role of the member of IL-1 superfamily. Front Immunol 2019;10:1–9.doi:10.3389/fimmu.2019.00692
      OpenUrlCrossRef
    12. ↵
      1. Mantovani A,
      2. Dinarello CA,
      3. Molgora M, et al
      . Interleukin-1 and related cytokines in the regulation of inflammation and immunity. Immunity 2019;50:778–95.doi:10.1016/j.immuni.2019.03.012pmid:http://www.ncbi.nlm.nih.gov/pubmed/30995499
      OpenUrlCrossRefPubMed
    13. ↵
      1. An G,
      2. Wang W,
      3. Zhang X, et al
      . Combined blockade of IL-25, IL-33 and TSLP mediates amplified inhibition of airway inflammation and remodelling in a murine model of asthma. Respirology 2020;25:603–12.doi:10.1111/resp.13711pmid:http://www.ncbi.nlm.nih.gov/pubmed/31610614
      OpenUrlPubMed
    14. ↵
      1. Allinne J,
      2. Scott G,
      3. Lim WK, et al
      . Il-33 blockade affects mediators of persistence and exacerbation in a model of chronic airway inflammation. J Allergy Clin Immunol 2019;144:e10:1624–37. doi:10.1016/j.jaci.2019.08.039pmid:http://www.ncbi.nlm.nih.gov/pubmed/31562870
      OpenUrlPubMed
    15. ↵
      1. Kearley J,
      2. Silver JS,
      3. Sanden C, et al
      . Cigarette smoke silences innate lymphoid cell function and facilitates an exacerbated type I interleukin-33-dependent response to infection. Immunity 2015;42:566–79.doi:10.1016/j.immuni.2015.02.011pmid:http://www.ncbi.nlm.nih.gov/pubmed/25786179
      OpenUrlCrossRefPubMed
    16. ↵
      1. Izzotti A,
      2. Longobardi M,
      3. La Maestra S, et al
      . Release of microRNAs into body fluids from ten organs of mice exposed to cigarette smoke. Theranostics 2018;8:2147–60.doi:10.7150/thno.22726pmid:http://www.ncbi.nlm.nih.gov/pubmed/29721069
      OpenUrlPubMed
    17. ↵
      1. Raza H,
      2. John A,
      3. Nemmar A
      . Short-Term effects of nose-only cigarette smoke exposure on glutathione redox homeostasis, cytochrome P450 1A1/2 and respiratory enzyme activities in mice tissues. Cell Physiol Biochem 2013;31:683–92.doi:10.1159/000350087pmid:http://www.ncbi.nlm.nih.gov/pubmed/23711494
      OpenUrlPubMed
    18. ↵
      1. Bhatnagar A,
      2. Maziak W,
      3. Eissenberg T, et al
      . Water pipe (Hookah) smoking and cardiovascular disease risk: a scientific statement from the American heart association. Circulation 2019;139:E917–36.doi:10.1161/CIR.0000000000000671pmid:http://www.ncbi.nlm.nih.gov/pubmed/30845826
      OpenUrlPubMed
    19. ↵
      1. Kemény Ágnes,
      2. Csekő K,
      3. Szitter I, et al
      . Integrative characterization of chronic cigarette smoke-induced cardiopulmonary comorbidities in a mouse model. Environ Pollut 2017;229:746–59.doi:10.1016/j.envpol.2017.04.098pmid:http://www.ncbi.nlm.nih.gov/pubmed/28648837
      OpenUrlPubMed
    20. ↵
      1. Thenappan T,
      2. Ormiston ML,
      3. Ryan JJ, et al
      . Pulmonary arterial hypertension: pathogenesis and clinical management. BMJ 2018;360:j5492.doi:10.1136/bmj.j5492pmid:http://www.ncbi.nlm.nih.gov/pubmed/29540357
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Barakat MF,
      2. McDonald HI,
      3. Collier TJ, et al
      . Acute kidney injury in stable COPD and at exacerbation. Int J Chron Obstruct Pulmon Dis 2015;10:2067–77.doi:10.2147/COPD.S88759pmid:http://www.ncbi.nlm.nih.gov/pubmed/26451102
      OpenUrlPubMed
    22. ↵
      1. Pabón MA,
      2. Patino E,
      3. Bhatia D, et al
      . Beclin-1 regulates cigarette smoke–induced kidney injury in a murine model of chronic obstructive pulmonary disease. JCI Insight 2018;3.doi:10.1172/jci.insight.99592
    23. ↵
      1. Mayyas F,
      2. Alzoubi KH
      . Impact of cigarette smoking on kidney inflammation and fibrosis in diabetic rats. Inhal Toxicol 2019;31:45–51.doi:10.1080/08958378.2019.1597219pmid:http://www.ncbi.nlm.nih.gov/pubmed/30947565
      OpenUrlPubMed
    24. ↵
      1. Liang KV,
      2. Greene EL,
      3. Oei LS, et al
      . Nodular glomerulosclerosis: renal lesions in chronic smokers mimic chronic thrombotic microangiopathy and hypertensive lesions. Am J Kidney Dis 2007;49:552–9.doi:10.1053/j.ajkd.2007.01.014pmid:http://www.ncbi.nlm.nih.gov/pubmed/17386324
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Xu H,
      2. Ling M,
      3. Xue J, et al
      . Exosomal microRNA-21 derived from bronchial epithelial cells is involved in aberrant epithelium-fibroblast cross-talk in COPD induced by cigarette smoking. Theranostics 2018;8:5419–33.doi:10.7150/thno.27876pmid:http://www.ncbi.nlm.nih.gov/pubmed/30555555
      OpenUrlPubMed
    26. ↵
      1. Song M,
      2. Peng H,
      3. Guo W, et al
      . Cigarette smoke extract promotes human lung myofibroblast differentiation by the induction of endoplasmic reticulum stress. Respiration 2019;98:347–56.doi:10.1159/000502099pmid:http://www.ncbi.nlm.nih.gov/pubmed/31416082
      OpenUrlPubMed
    27. ↵
      1. Zhou L-L,
      2. Wang M,
      3. Liu F, et al
      . Cigarette smoking aggravates bleomycin-induced experimental pulmonary fibrosis. Toxicol Lett 2019;303:1–8.doi:10.1016/j.toxlet.2018.12.008pmid:http://www.ncbi.nlm.nih.gov/pubmed/30572104
      OpenUrlPubMed
    28. ↵
      1. Wu H,
      2. Yang S,
      3. Wu X, et al
      . Interleukin-33/ST2 signaling promotes production of interleukin-6 and interleukin-8 in systemic inflammation in cigarette smoke-induced chronic obstructive pulmonary disease mice. Biochem Biophys Res Commun 2014;450:110–6.doi:10.1016/j.bbrc.2014.05.073pmid:http://www.ncbi.nlm.nih.gov/pubmed/24866242
      OpenUrlCrossRefPubMed
    29. ↵
      1. Qiu C,
      2. Li Y,
      3. Li M, et al
      . Anti-interleukin-33 inhibits cigarette smoke-induced lung inflammation in mice. Immunology 2013;138:76–82.doi:10.1111/imm.12020pmid:http://www.ncbi.nlm.nih.gov/pubmed/23078031
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Byers DE,
      2. Alexander-Brett J,
      3. Patel AC, et al
      . Long-Term IL-33-producing epithelial progenitor cells in chronic obstructive lung disease. J Clin Invest 2013;123:5410doi:10.1172/JCI74125pmid:http://www.ncbi.nlm.nih.gov/pubmed/23945235
      OpenUrlPubMed
    31. ↵
      1. Smith DE
      . Il-33 meets allergens at the gate. Nat Immunol 2018;19:318–20.doi:10.1038/s41590-018-0069-3pmid:http://www.ncbi.nlm.nih.gov/pubmed/29563627
      OpenUrlCrossRefPubMed
    32. ↵
      1. Clerman A,
      2. Noor Z,
      3. Fishelevich R, et al
      . The full-length interleukin-33 (FLIL33)-importin-5 interaction does not regulate nuclear localization of FLIL33 but controls its intracellular degradation. J Biol Chem 2017;292:21653–61.doi:10.1074/jbc.M117.807636pmid:http://www.ncbi.nlm.nih.gov/pubmed/29127199
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Kunisch E,
      2. Chakilam S,
      3. Gandesiri M, et al
      . Il-33 regulates TNF-α dependent effects in synovial fibroblasts. Int J Mol Med 2012;29:530–40.doi:10.3892/ijmm.2012.883pmid:http://www.ncbi.nlm.nih.gov/pubmed/22246057
      OpenUrlPubMed
    34. ↵
      1. Nile CJ,
      2. Barksby E,
      3. Jitprasertwong P, et al
      . Expression and regulation of interleukin-33 in human monocytes. Immunology 2010;130:172–80.doi:10.1111/j.1365-2567.2009.03221.xpmid:http://www.ncbi.nlm.nih.gov/pubmed/20070408
      OpenUrlCrossRefPubMedWeb of Science

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors QH participated in the design of the study, data acquisition, statistical analysis and drafting of the manuscript. CDL, YRY, XFQ, XZ and XND participated in acquisition of the data. JJW bred the IL-1RL1-/- mice. XY, YW, LL, MM and KH participated in collecting clinic samples from healthy controls and patients with COPD. ZL and YC participated in the design of the study. CJC, WW and SY participated in the design of the study, reviewed and edited the manuscript. All authors read and approved the final manuscript.

    • Funding We acknowledge financial support from the National Natural Science Foundation of China (81770049, 81700026, 81971510, 81974050, 82090013), the Natural Science Foundation of Beijing Municipality (7192023) and the Support Project of High-level Teachers in Beijing Municipal Universities in the Period of 13th Five-year Plan (IDHT20190510).

    • Competing interests None declared.

    • Patient consent for publication Not required.

    • Ethics approval The study was monitored and approved by the Institutional Animal Care and Use Committee of the Capital Medical University.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data availability statement Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information. All data used to support the findings of the current study are available from the corresponding authors upon request. Corresponding author, Professor Sun YingE-mail: ying.sun@ccmu.edu.cn; Phone (Office): (+86)10-83911743.

    Linked Articles

    • Airwaves
      Highlights from this issue
      The Triumvirate
      Thorax 2021; 76 743-743 Published Online First: 19 Jul 2021. doi: 10.1136/thoraxjnl-2021-217852