Activation of STAT3 in the COPD airway epithelium

Abstract

Rationale

The mechanisms driving epithelial pathology in COPD are steadily being unveiled, confirming a role for inflammation in the disease. This study explored the interleukin (IL)-6/STAT3 axis, previously reported to link inflammation and epithelial-to-mesenchymal transition, two features of COPD at the airway epithelium level.

Methods

Bronchoalveolar lavage fluid (BALF) and surgical lung tissue were obtained from nonsmoker controls, smokers and COPD patients. The activation of STAT3 and IL-6 levels were measured in these samples. Primary air–liquid interface (ALI) cultures were carried out from nonsmokers, smokers and COPD patients, and IL-6 release and STAT3 mRNA levels were assessed. BEAS-2B cell cultures were exposed to sputum supernatants from COPD patients versus nonsmokers, with and without a pan-gp130 blocking monoclonal antibody. Finally, primary ALI cultures from nonsmokers were exposed to IL-6 versus vehicle and assessed for epithelial-to-mesenchymal transition and cell differentiation.

Results

IL-6 and Tyr705-phospho-STAT3 levels were increased in samples from COPD patients compared to controls, both in BALF and in the airway epithelium, as well as in ALI cultures. BEAS-2B cells exposed to COPD sputum supernatants displayed STAT3 activation that was inhibited by the pan-gp130 blocking monoclonal antibody. In addition, stimulation of ALI cultures with IL-6 induced increased vimentin expression and fibronectin release and reduced the expression of apical junctional complexes proteins, indicating epithelial-to-mesenchymal transition. Finally, no impact on airway cell differentiation was observed.

Conclusions

The IL-6/STAT3 axis is activated in the COPD airway epithelium, presumably contributing to epithelial-to-mesenchymal transition.

Shareable abstract

The STAT3 pathway is activated in the COPD airway epithelium and promotes epithelial-to-mesenchymal transition in an IL-6-dependent manner https://bit.ly/46RKYpk

Introduction

COPD currently represents the third cause of death worldwide [1], with limited therapeutic options, none of which are able to reverse the course of the disease. COPD is mainly due to noxious airborne stimuli, most importantly cigarette smoke (CS), while genetic mechanisms and impaired lung growth also play a role in its pathogenesis [2, 3]. COPD is characterised by the destruction of the lung parenchyma (i.e., emphysema), the irreversible narrowing and disappearance of the small airways [4] and remodelling of the epithelium lining the airways [5].

As the first barrier encountered by inhaled toxics, the airway epithelium is particularly prone to display structural and functional alterations in COPD, including altered cell differentiation [6, 7], barrier dysfunction [8], reduced local mucosal immunity [9] and aberrant repair mechanisms, including sustained epithelial-to-mesenchymal transition (EMT) [7, 10]. EMT is a dynamic process during which epithelial cells lose their polarity and adhesiveness and acquire migratory abilities as well as mesenchymal features. This shift is marked by the downregulation of epithelial markers such as E-cadherin, and the upregulation of mesenchymal markers including vimentin and fibronectin. EMT participates in embryogenesis (type I EMT), tissue repair (type II) and cancer metastasis (type III). EMT programming has been shown to occur in both small and large airways in COPD, where it is thought to contribute to airway remodelling and promote subepithelial fibrosis as well as carcinogenesis and metastasis potential [11–15]. In addition, airway epithelial cells from COPD patients, cultured at the air–liquid interface (ALI), replicate EMT and spontaneously display mesenchymal features, allowing the in vitro evaluation of this process [7, 16].

EMT is regulated by an intricate network of pathways, including the Janus kinase (JAK)-signal transducers and activators of transcription (STAT) pathways [17]. Their relevance to COPD pathophysiology is currently unknown, with only a few descriptive studies suggesting that their activity could be modified in COPD, with no further exploration of the role of such dysregulation [18, 19]. Yet, interleukin (IL-)6, whose signalling requires signal transduction by JAK family members JAK1 and/or Tyk2, and activation of STAT transcription factors [20–22], is thought to participate in COPD pathogenesis. Indeed, IL-6 levels are increased in serum, bronchoalveolar lavage fluid (BALF) and exhaled breath condensate from COPD patients [23–26], while IL6_-174G/C polymorphism has been associated with accelerated lung function decline and susceptibility to COPD [27]. Whether IL-6 plays a role in exacerbations is still debated [28, 29], but cumulative evidence suggests that IL-6-associated inflammation plays an important role in COPD pathogenesis at the epithelial level. Whether IL-6 contributes to the promotion of EMT in COPD remains unclear.

Our study aims were to assess the activation of the IL-6/STAT3 pathway at the COPD epithelium level and to explore whether this promotes EMT.

Methods

Study population

Lung tissue from nonsmoker and smoker controls as well as mild–moderate (Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages I–II) COPD patients was collected during surgical resection for suspected cancer, at maximal distance from the lesion. Samples were collected in Belgium (UCLouvain, Brussels) and in the UK (Manchester). Lung explant tissue from severe and very severe (GOLD stage III and IV, respectively) COPD patients was also collected in Belgium. Samples collected in Belgium were used to generate ALI epithelial cell culture, while tissue collected in the UK was used for immunohistochemical analysis. The Belgian cohort comprised 51 tissue samples from nonsmoker controls (n=10), smoker controls (n=15) and mild–moderate COPD (n=13) as well as severe and very severe COPD (explants, n=13). The UK cohort was composed of 43 samples from nonsmoker controls (n=9), smoker controls (n=10) and COPD patients (n=24). All COPD patients had a smoking history of >10 pack-years. They included a mix of current and former smokers, except for GOLD stage III and IV COPD patients from the Belgian cohort, all of whom were ex-smokers who had quit smoking for at least 6 months.

Bronchoalveolar lavage (n=34) and sputum samples (n=22) were also collected in Manchester, for cytokine assessments and epithelial pSTAT3 induction. COPD was diagnosed based on the 2017 GOLD recommendations [30]. Controls originated from smokers without airflow limitation, and nonsmokers (history <1 pack-year). Subjects with a history of any other respiratory disease, such as asthma, interstitial lung disease and cystic fibrosis, were excluded from the study. This research was approved by local research ethics committees (Belgium lung tissue: 2007/19MARS/58; UK lung tissue: 03/SM/396; UK bronchoalveolar lavage: 06/Q1403/156; and UK sputum: 05/Q1402/41). All subjects provided written informed consent. The clinical characteristics of the subjects are summarised in supplementary tables S1–S4.

Immunohistochemistry

Tissue preparation and processing for immunohistochemistry are described in the supplementary material. For quantification of STAT3 staining, small airway epithelial expression was graded as: Grade 0: no pSTAT3 immunoreactivity, Grade 1: pSTAT3⁺ basal cells with scattered positive columnar/cuboidal cells, Grade 2: pSTAT3⁺ basal cells with foci of positive columnar/cuboidal cells, Grade 3: pSTAT3⁺ epithelium throughout airway. Illustrative pictures are presented in figure 1a. The number of pSTAT3⁺ cells·per mm² of subepithelium was calculated using the manual tag facility of the ImagePro Plus 6.0 image analysis software (Media Cybernetics, Rockville, MD, USA).

FIGURE 1.

Increased pSTAT3 levels in the COPD airway epithelium and COPD-derived air–liquid interface (ALI) cultures. Immunohistochemistry was performed on lower airway tissue samples from nonsmokers (NS; n=9), smokers (Smo; n=10) and COPD (n=24) patients to identify pSTAT3 expression (UK cohort). Tissue samples were graded 0–3 depending on the level of pSTAT3 expression (a) with levels compared between groups (b). ALI-reconstituted airway epithelium (Belgium cohort) issued from nonsmokers (n=9), smoker controls (n=15) and COPD patients (n=26) were assessed at 4 weeks (4w) for STAT3 gene expression by RT-PCR (c) as well as total STAT3 and pSTAT3 protein expression by Western blot, with normalisation to GAPDH (d: illustrative blots and e: comparison between groups). Comparisons between groups was assessed by Kruskal–Wallis test, with Dunn's multiple comparison test. HKG: housekeeping gene. **: p<0.01; ***: p<0.001; ****: p<0.0001.

In vitro reconstitution of primary human bronchial epithelium on ALI culture

Primary cultures of human bronchoepithelial cells (HBEC) from nonsmokers (n=10), smokers (n=15) and mild-to-moderate COPD patients (n=13), as well as severe and very severe COPD patients (n=13) were maintained at the ALI for up to 4 weeks, as previously reported [31]. Detailed information on the management of cultures is provided in the supplementary material.

In a subset of experiments, ALI-HBECs were exposed to IL-6 at 5 ng·mL⁻¹ (product code: 130–093-929; Miltenyi Biotec, Germany) in the basolateral compartment for 48 h. This concentration was determined from a preliminary titration experiment (10, 5 and 2.5 ng·mL⁻¹), where no cytotoxicity (release of lactate dehydrogenase <5%) was observed at 5 ng·mL⁻¹.

Western blots

Western blots for Tyr705-phospho-(p-)STAT3 (product code: CRL-3588, ATCC, US) were performed on BEAS-2B cell line cultures stimulated with sputum supernatants (n=25), and blots for Tyr705-phospho-(p-)STAT3 (product code: CRL-3588, ATCC, US) and vimentin (product code: M0725, Dako, Carpinteria, CA) on controls versus COPD primary HBEC ALI cultures (n=51) and on IL-6-stimulated HBEC ALI cultures (n=6) as reported previously [31, 32]. Detailed information is provided in the supplementary material.

Reverse transcriptase quantitative polymerase chain reaction

RNA extraction, reverse transcription and quantification were performed as previously reported [7]. Detailed information is provided in the supplementary material.

BEAS-2B cell culture

The epithelial cell lines BEAS-2B were maintained in RPMI-1640 (product code: R0882, Sigma-Aldrich, Poole, UK) supplemented with 10% v/v fetal calf serum (FCS; product code: A5256801, Invitrogen, Paisely, UK), 2 mM L-Glutamine (product code: 25030081; Invitrogen), 100 U·mL⁻¹ penicillin and 100 µg·mL⁻¹ streptomycin (product code: P3032 and S9137, respectively, Sigma-Aldrich). All cultures were grown in a humidified atmosphere (5% CO₂ in air) at 37°C. Cells received fresh serum-free media 24 h prior to the experiments. Confluent cells were treated with sputum supernatant (diluted 1:2 in supplemented RPMI media) from healthy nonsmokers or severe COPD patients for 1 h, in the presence or absence of the pan gp130 monoclonal antibody, B-R3 (dilution: 1:400; product code: ARG23320; 2BScientific, Upper Heyford, UK). Cell cultures received fresh serum-free media 24 h prior to the experiments.

Sputum supernatant collection

Sputum was collected from nonsmoker controls (n=11) and COPD patients (n=11). Sputum was induced and processed using a two-step method with a Dulbecco's phosphate-buffered saline (PBS, product code: D8537; Sigma-Aldrich) wash step followed by a dithiothreitol (DTT, product code: D2650, Sigma-Aldrich) step as published previously [33]. PBS processed sputum supernatants were prepared and stored directly at −80°C.

Enzyme-linked immunosorbent assays (ELISA)

Levels of IL-6 in BALF and in culture supernatants were determined using Quantikine high-sensitivity ELISA (product code: HS600C; R&D systems, Abingdon, UK). The lower limit of quantification of IL-6 was 0.2 pg·mL⁻¹. Oncostatin M (OSM), IL-11 and IL-27 levels in sputum supernatants were assessed by ELISA (product codes: DY295, DY218 and DY2526, respectively; R&D systems). The lower limits of quantification were: OSM 31.2 pg·mL⁻¹; IL-11 31.2 pg·mL⁻¹; and IL-27 15.6 pg·mL⁻¹. All assays were performed as per the manufacturer's instructions.

Fibronectin release at the basolateral side of ALI cultures was assessed by using direct ELISA (product code: 14-9869-82; Thermo-Fisher), as described in the supplementary material.

BALF collection

BALF was retrieved from healthy nonsmokers (n=10), smoker controls (n=6), mild-to-moderate COPD patients (n=12) and severe COPD patients (n=6), as previously reported [34]. IL-6 levels were determined as described above.

Tyramide signal amplification (TSA) immunofluorescence staining of ALI-reconstituted airway epithelium

ALI-reconstituted epithelium from control patients, submitted to control conditions versus IL-6 stimulation, were stained for p63, FOXJ1, MUC5AC, vimentin and α-SMA, as well as for E-cadherin, occludin, α-SMA and vimentin, using TSA immunofluorescence. This technique, previously described [7, 31], is detailed in the supplementary material.

Statistical analysis

Data are expressed as medians and interquartile ranges unless otherwise stated, and were analysed with JMP Pro, Version 14 (SAS Institute Inc., Cary, NC, USA) and GraphPad Prism version 8.0.2 for Windows (GraphPad Software, La Jolla, CA, USA). p-values <0.05 were considered statistically significant.

Results

The clinical characteristics of enrolled subjects in the UK and Belgium are summarised in supplementary tables S1–S4. COPD patients who donated lung tissue were age-matched to nonsmoker and smoker tissue donors. Healthy nonsmokers and smokers who donated sputum and bronchoalveolar lavage were significantly younger than the corresponding patients with COPD. All other differences between groups were as expected.

Increased expression of phosphorylated STAT3 in airway epithelium of COPD patients compared to controls

Immunohistochemistry was used to assess Tyr705-pSTAT3 in the airway epithelium of resected lung samples (UK cohort). Epithelial pSTAT3 expression was significantly higher in COPD patients compared to nonsmokers (p=0.0038) and smoker controls (p<0.0001) (figure 1a,b). This result was supported by the observation that primary airway epithelial cells from COPD patients (Belgium cohort), grown on ALI cultures for 4 weeks, also expressed higher levels of pSTAT3 compared to cells from nonsmokers (p=0.0001) and smoker controls (p=0.0096) (figure 1d). In these same cultures, STAT3 mRNA levels were similar in cells from COPD patients and (non)-smoker controls (figure 1c). Taken together, these results indicate increased STAT3 activation in the COPD epithelium, independently from its transcription.

IL-6 triggers STAT3 activation in the COPD epithelium

To assess what stimuli is responsible for elevated pSTAT3 in airway epithelial cells from COPD patients, airway epithelial cells (BEAS-2B cell line) were treated with sputum supernatants either from healthy nonsmokers or from COPD patients (UK cohort). Sputum supernatant was chosen as it is a physiologically relevant matrix containing COPD-specific markers [35]. Supernatants from COPD patients induced higher levels of pSTAT3 compared to supernatants from healthy nonsmokers (p=0.046) (figure 2a). Pre-treating the epithelial cells with neutralising antibodies against the glycoprotein (gp) 130 receptor protein, prior to COPD sputum exposure, suppressed pSTAT3 induction, demonstrating that pSTAT3 stimulation was induced by an IL-6 superfamily cytokine (figure 2b). Levels of IL-6 are known to be higher in sputum supernatants from COPD patients compared to controls [26, 36]. Attempts were made to measure the gp130-dependent IL-6 superfamily cytokines OSM, IL-11 and IL-27 in sputum supernatants from COPD patients and controls, but all were undetectable in all samples tested (data not shown), suggesting that IL-6 is the primary candidate in sputum supernatant for activating STAT3 in airway epithelial cells.

FIGURE 2.

STAT3 activation in COPD is gp130-dependent. a) Phosphorylated STAT3 levels in BEAS-2B epithelial cells following 1 h exposition to sputum supernatants from nonsmokers (NS) and COPD patients (UK cohort) were assessed by Western blot, with normalisation to β-actin levels (bar illustrates mean value). Comparisons between nonsmoker and COPD sputum responses were assessed by t-test; *: p<0.05. b) To access the mechanism of STAT3 activation by sputum supernatants, BEAS-2B cells were treated with (+) or without (−) a pan gp130 receptor blocking antibody (0.05 µg–50 µg) prior to sputum exposure.

Increased IL-6 levels in BALF from COPD patients

Sputum is typically classed as an upper airway sample and may not represent the cytokine profile that the lower airway epithelial cells, which were assessed for pSTAT3 in the resected lung samples, are exposed to. Therefore, IL-6 levels were also assessed in BALF from COPD patients and controls (UK cohort). Higher IL-6 levels were observed in BALF from both mild-to-moderate COPD patients (median and interquartile range: 3.34 pg·mL⁻¹ (0.93–2.87), p=0.003 versus nonsmokers) and severe and very severe COPD patients (8.85 pg·mL⁻¹ (4.93–15.35), p=0.0004 versus nonsmokers) as compared to healthy nonsmokers (0.786 pg·mL⁻¹ (0.46–1.14)) (figure 3a). A non-significant upward trend was also observed versus smoker controls (1.64 pg·mL⁻¹ (0.93–2.87)). IL-6 levels were also higher in ALI cultures (Belgium cohort) from smokers (303.6 pg·mL⁻¹ (112.7–1470), p=0.0028), mild-to-moderate COPD patients (219.4 pg·mL⁻¹ (73.8–570.3), p=0.022) and severe COPD patients (355.5 pg·mL⁻¹ (153.2–654.7), p=0.0044) compared to nonsmoker controls (13.24 pg·mL⁻¹ (0.16–176.0), figure 3b). No differences between COPD and (non-)smoker controls were observed in Il6 mRNA levels in the same ALI cultures (figure 3c).

FIGURE 3.

Interleukin (IL)-6 levels are increased in bronchoalveolar lavage fluid (BALF) and air–liquid interface (ALI) cultures derived from COPD patients. a) IL-6 levels in BALF (UK cohort) from nonsmokers (NS) (n=10), smokers (Smo) (n=6), Global Initiative for Chronic Obstructive Lung Disease (GOLD) 1–2 (n=12) and GOLD 3–4 (n=6) COPD patients were measured by ELISA. In 4 weeks (4w) ALI epithelial cell cultures (Belgian cohort) derived from nonsmoker (n=10), smoker (n=15), GOLD 1–2 (n=13) and GOLD 3–4 (n=13) COPD patient's lung tissue, IL-6 levels were assessed by ELISA in basolateral supernatants (b), while IL6 gene expression was assessed by PCR and normalised to housekeeping genes (c). Comparisons between groups were assessed by Kruskal–Wallis test, with Dunn's multiple comparison test. HKG: housekeeping gene. *: p<0.05; **: p<0.01; ***: p<0.001. Bars illustrate median with interquartile range values.

IL-6/STAT3 axis activation promotes EMT in COPD

To assess the effect of increased STAT3 activity on the airway epithelium, markers of EMT were assessed in ALI cultures from healthy nonsmokers stimulated with IL-6 (Belgium cohort). Stimulation with IL-6 (5 ng·mL⁻¹) resulted in increased STAT3 Tyr705-phosphorylation (p=0.0043), increased expression of vimentin (p=0.041) and increased fibronectin release at the basolateral surface of the ALI cultures (p=0.026) (figure 4a).

FIGURE 4.

Interleukin (IL)-6 exposure induces epithelial-to-mesenchymal transition (EMT)-related modifications in airway epithelial cells. Air–liquid interface (ALI)-reconstituted airway epithelial cells (Belgian cohort) were treated with recombinant IL-6 (5 ng·mL⁻¹) and pSTAT3 levels, as well as markers of EMT, were assessed by qPCR and Western blot. a) IL-6 stimulation led to increased protein expression of pSTAT3 and vimentin, along with enhanced fibronectin release at the basolateral pole. While mRNA levels of SNAI1, SNAI2 and TWIST1 remained unchanged (b), IL-6 stimulation significantly increased SNAI1 and SNAI2 protein expression (c). d) Representative blots for SNAI1 and SNAI2. Comparisons between IL-6-treated and vehicle control (DMSO) were assessed by Mann–Whitney test; *: p<0.05; **: p<0.01. Bars illustrate median levels and interquartile range. 4w: 4 weeks; NS: nonsignificant; HKG: housekeeping gene.

We further investigated three transcription factors known to promote EMT (SNAI1, SNAI2 and TWIST1) at both the mRNA and protein levels. While no significant changes were observed at the mRNA level (figure 4b), IL-6 stimulation led to a significant increase in SNAI1 and SNAI2 protein levels (p=0.0087 and p=0.0043, respectively), with TWIST1 showing a non-significant upward trend (figures 4c and d).

Finally, EMT was assessed by evaluating the expression of vimentin and α-SMA, as well as apical junction complex proteins. Airway epithelial cells cultured under control conditions exhibited normal intercellular expression of E-cadherin and occludin, with minimal vimentin expression (figure 5a, upper panel). In contrast, IL-6 stimulation led to increased vimentin expression and a marked reduction in both E-cadherin and occludin protein expression (figure 5a, lower panel). E-cadherin expression was significantly reduced under IL-6 stimulation, both at the protein (p=0.0312) and mRNA levels (CDH1, p=0.0312) (figure 5b). Consistently, TJP1 expression was also significantly decreased in IL-6-stimulated conditions (p=0.0312).

FIGURE 5.

Interleukin (IL)-6 exposure reduces tight junction proteins expression and enhances vimentin expression in air–liquid interface (ALI)-reconstituted airway epithelial cells. a) ALI-reconstituted airway epithelial cells (Belgian cohort) were exposed to either control conditions (DMSO, upper panels) or recombinant IL-6 (lower panels). Epithelial-to-mesenchymal transition (EMT) and tight junction integrity were assessed using TSA-enhanced immunofluorescence staining for vimentin (AlexaFluor 594, red), E-cadherin (AlexaFluor 647, orange) and occludin (AlexaFluor 555, yellow). α-SMA was also assessed (AlexaFluor 754, purple) but no expression was detected. Under control conditions, vimentin expression was minimal, while occludin and E-cadherin displayed classical apical intercellular staining. Following IL-6 stimulation, vimentin expression increased markedly, whereas both E-cadherin and occludin expression were strongly reduced. Scale bar: 10 µm. b) E-cadherin expression was reduced under IL-6 stimulation, both at the mRNA and protein levels. c) No differences were observed in the mRNA levels of TJP1. NS: nonsignificant; HKG: housekeeping gene. *p<0.05.

IL-6/STAT3 axis activation does not alter cell differentiation

To assess the effect of increased STAT3 activity on the airway epithelium, markers of airway cell differentiation were evaluated in ALI cultures from healthy nonsmokers stimulated with IL-6 (Belgian cohort). Stimulation with IL-6 (5 ng·mL⁻¹) did not result in observable changes in the numbers of p63⁺ basal cells, MU5AC⁺ goblet cells or FOXJ1⁺ ciliated cells (figure 6a). Similarly, IL-6 stimulation did not alter the mRNA expression levels of FOXJ1 and SPDEF, two transcription factors known to promote differentiation towards ciliated and goblet cells, respectively; or P63, a transcription factor associated with basal cells (figure 6b).

FIGURE 6.

Impact of interleukin (IL)-6 stimulation on cell differentiation in ALI-reconstituted airway epithelial cells. a) ALI-reconstituted airway epithelial cells (Belgian cohort) were exposed to either control conditions (DMSO, upper panels) or recombinant IL-6 (lower panels). Cell differentiation and epithelial-to-mesenchymal transition (EMT) were assessed using TSA-enhanced immunofluorescence staining for FOXJ1 (ciliated cells, AlexaFluor 488, green), p63 (basal cells, AlexaFluor 647, orange), MUC5AC (goblet cells, AlexaFluor 555, yellow) and vimentin (AlexaFluor 594, red). α-SMA was also assessed (AlexaFluor 754, purple) but no expression was detected. Under control conditions, vimentin expression was minimal, while some cells expressed FOXJ1 and/or MUC5AC. The epithelial layer was lined by p63⁺ basal cells. Upon IL-6 stimulation, few changes were observed, apart from increased vimentin expression, as seen in figure 5. Scale bar: 10 µm. b) At the mRNA level, IL-6 stimulation did not alter the expression of SPDEF, FOXJ1 or P63. HKG: housekeeping gene.

Discussion

In this study, we provide evidence that the IL-6/STAT3 axis is activated in the airway epithelium of COPD patients, and that this activation induces markers of EMT. First, elevated levels of IL-6 and Tyr705-phospho-STAT3 were observed in both patient-derived samples and ALI-cultured bronchoepithelial cells. Second, the use of a pan-gp130 blocking monoclonal antibody effectively inhibited STAT3 activation, revealing an upstream trigger by IL-6 superfamily cytokine. Finally, primary airway cell cultures exposed to IL-6 displayed increased levels of vimentin and fibronectin, as well as SNAI1 and SNAI2 transcription factors. In parallel, the expression of several components of the apical junctional complexes was reduced, supporting the role of the IL-6/STAT3 pathway in promoting EMT, a central feature of COPD pathogenesis at the epithelial level.

Regarding STAT3, data in COPD are limited. At the parenchymal level, increased pSTAT1 and pSTAT3 levels were observed in samples from COPD patients [18]. Additionally, specific polymorphisms on the STAT3 gene associated with increased STAT3 mRNA levels have been linked with the development and progression of COPD [37]. A recent conference paper also reported that inhibiting STAT3 in a CS-exposition murine model of COPD alleviated COPD features [38]. Beyond its role in inducing EMT, increased STAT3 in COPD may also contribute to disease progression by promoting senescence [39], a well-documented feature of COPD [40].

In contrast to STAT3, increased IL-6 levels have been widely reported in COPD. For example, a pooled analysis of the COPDGene and SPIROMICS cohorts found elevated IL-6 levels in serum of COPD patients compared to never-smokers, with these levels being associated with progressive airflow obstruction over 5 years [41]. Higher IL-6 levels in COPD patients have also been linked to increased exacerbation rates [28]. In the ECLIPSE study, elevated serum IL-6 levels were associated with poor physical outcomes and increased mortality after 3 years [42], though this association was not observed at 9 years [43]. Additionally, IL6 gene polymorphisms have been linked to a rapid decline in forced expiratory volume in 1 second and increased susceptibility to COPD in smokers [27]. Finally, elevated IL-6 levels have been observed in basal supernatants from COPD-derived ALI cultures [7, 44].

IL-6-induced STAT3 phosphorylation drives the transcription of numerous genes, including SNAIL1, SNAIL2 (also known as SLUG) and TWIST1, three transcription factors well known for promoting EMT [45]. EMT is a dynamic process in which epithelial cells lose polarity and adhesiveness while acquiring migratory abilities and mesenchymal characteristics. EMT programming is crucial to tissue repair following tissue injury (type II EMT) [17]. Growing evidence indicates that EMT occurs in the COPD airway epithelium, affecting both small [11] and large airways [12, 13, 16], contributing to subepithelial airway fibrosis and impaired epithelial repair [46], as well as carcinogenesis [14, 15]. Beyond its role in promoting EMT, IL-6 has also been associated with mucus hypersecretion, a COPD trait associated to increased morbidity and mortality [47]. Notably, IL-6 inhibition in cell cultures has been shown to reduce this effect [48]. The impact of IL-6 on airway epithelial cell differentiation has not been thoroughly explored to date. In the present study, IL-6 stimulation did not appear to alter airway epithelial cell composition; however, this may be attributed to the timing of the stimulation or analysis, which may not have captured potential delayed effects on cell differentiation.

Our observations of increased IL-6/STAT3 expression in COPD raise the question of whether inhibiting this pathway could improve COPD outcomes. JAK inhibition currently represents an attractive therapeutic target for several diseases, leading to the development and approval of selective drugs over the past 10 years [49]. These drugs have significantly improved outcomes in various inflammatory diseases, including atopic dermatitis, rheumatoid arthritis and inflammatory bowel diseases. In COPD, in vitro studies have provided promising results. In lipopolysaccharide- and CS-stimulated monocyte-derived macrophages, JAK1/2 inhibition with ruxolitinib strongly attenuated IL-6, CXCL-8 and CCL-2 release. Similarly, pan-JAK inhibition using PF1367550 reduced CXCL9, CXCL10 and CXCL11 release in interferon-γ- and tumour necrosis factor-α-stimulated primary airway epithelial cells from both control subjects and COPD patients [50]. However, the translation into clinical studies remains slow, with only early-phase clinical trials currently ongoing and results still pending. In COPD, the risk of increasing susceptibility to infections by dampening immune responses in patients already prone to respiratory infections represents a significant challenge [51].

While our study provides significant insights into IL-6/STAT3 signalling and its role in EMT in COPD, it has several limitations. Firstly, while this study used ex vivo airway samples from patients and healthy subjects to link elevated IL-6 signalling and EMT in COPD, in vivo animals may be needed to provide further supporting evidence. Secondly, COPD patient samples were assessed as a homogeneous group, which does not capture the clinical reality of the disease. Further investigations would be useful to understand if IL-6/STAT/EMT is specific to a particular subtype/endotype. Thirdly, STAT3 phosphorylation may be triggered by other cytokines and growth factors, such as epithelial growth factor, insulin-like growth factor 1, granulocyte colony-stimulating factor and fibroblast growth factors [52]. As some of these factors are relevant to COPD pathophysiology, further studies should explore their relative contribution to increased STAT3 signalling. Finally, although we attempted pharmacological inhibition of STAT3 using S3I-201, we were unable to demonstrate a clear reduction in STAT3 phosphorylation, likely due to limitations inherent to the ALI culture model. Despite these limitations, our findings lay a strong foundation for future research into targeted therapies that address key inflammatory and remodelling processes in COPD.

In conclusion, by demonstrating that STAT3 is selectively activated in the airway epithelium in COPD and that this activation promotes EMT, this study enhances our understanding of the disease and contributes to the effort for defining innovative approaches and improved care of COPD patients.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material

00497-2025.SUPPLEMENT