The association between airway eosinophilic inflammation and IL-33 in stable non-atopic COPD

Chronic Obstructive Pulmonary Disease (COPD) is one the most common chronic diseases in adults, and one of the main causes of morbidity and mortality [1]. The prevalence of COPD at Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage 2 or higher has been estimated at 10.1% overall, although the precise figure varies according to the region [2]. A growing body of evidence indicates that eosinophilic COPD is a distinct phenotype of the disease, and one that is characterized by a higher risk of exacerbations but better response to treatment with inhaled corticosteroids [3, 4]. COPD subjects have higher percentages of eosinophils in sputum and higher concentrations of eosinophil cationic protein than healthy controls, and their airway eosinophil numbers further increase during disease exacerbation [5, 6]. Data from the ECLIPSE (Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points) study showed that over three years of observation, 37.4% of COPD patients demonstrated a persistent blood eosinophil count greater than or equal to 2% [7]. Non-allergic COPD patients with blood eosinophil counts > 250 cells/μL are characterized by higher airway eosinophil numbers, higher sputum interleukin(IL)-5 and haptoglobin levels, greater CCL20 and CCL24 concentrations in bronchoalveolar lavage (BAL) and more pronounced airway remodeling [8].

IL-33 is one of the epithelial alarmins generated in response to danger signals, and is primarily produced by epithelial cells [9]. Recent evidence has demonstrated that IL-33 precedes IL-5 in regulating eosinophil commitment and is required for eosinophil homeostasis [10]. One such result of IL-33 activity is that activates and augments the maturation of hemopoietic progenitor cells (HPC). HPC are considered to be innate effector cells in allergic asthma [11, 12]: They produce an abundance of pro-inflammatory cytokines and differentiate into mature eosinophils in the process of “in situ hemopoiesis” at sites of inflammation [13].

The aim of the present study was to determine the presence of an association between IL-33 level, the expression of its receptor and the degree of eosinophilic inflammation in non-atopic COPD.

Our findings show for the first time that increased expression of IL-33 and its receptor ST2 is associated with airway eosinophilia in non-atopic COPD patients.

IL-33 has been extensively studied in allergic asthma and other allergic conditions. Some studies suggest that IL-33 may be also implicated in the pathogenesis of COPD. IL-33 and sST2 serum levels are elevated in COPD patients compared to healthy smokers [18]. Furthermore, immunofluorescence staining showed that the expression of IL-33 was increased in the lung tissues of patients with COPD, indicating that IL-33 upregulation was probably related to systemic and airway inflammation in COPD [18‐20]. It has also been reported that cigarette smoke markedly enhanced the expression of IL-33 and ST2 in the lung tissue of mice, which was accompanied by neutrophil and macrophage infiltration and elevated expression of pro-inflammatory cytokines and chemokines in the airways [19]. Interestingly, Kim et al. found blood eosinophil counts to be correlated with plasma IL-33 levels in COPD patients [20].

Recently, Gluck et al. showed that IL-33 protein is detectable at low levels in EBC, and its level is elevated in asthmatics compared with healthy controls [21]. Our results demonstrate that IL-33 is elevated in EBC collected from non-atopic COPD subjects to the same extent as in asthma. Furthermore, EBC IL-33 levels correlated positively with blood eosinophil numbers and percentages, not only in asthma but also in COPD.

To further explore these findings, the second part of our study investigated the expression of IL-33, its receptor and its soluble form in the periphery and the airways of patients with and without airway eosinophilia. Our results showed that the level of IL-33 protein is markedly increased in the serum and sputum of patients with eosinophilic COPD.

sST2 is a decoy receptor regulating the activity of IL-33. As sST2 is not detected in EBC, its levels were not assessed in the EBC obtained from COPD patients [21]. However, our findings confirm those of previous reports showing increased levels of serum sST2 in COPD [18]. These findings also suggest a trend toward overexpression of sputum sST2 in the whole COPD group. Animal studies have shown that sST2 may exert a protective function in acute lung injury and ovalbumin-induced allergic asthma [22, 23], and Hacker et al. suggest that sST2 plays a critical role in the anti-inflammatory regulatory mechanism in early stages of COPD [24]. Although our findings do not indicate significant differences in sST2 concentrations between COPD subjects with and without sputum eosinophilia, sputum sST2 levels nevertheless correlated negatively with sputum eosinophil counts, suggesting it plays a protective role against airway eosinophilic inflammation.

On the contrary, sputum cells obtained from COPD patients with airway eosinophilia overexpressed ST2 mRNA. ST2 is expressed by many types of cells that may be present in the airways, including eosinophils. The activation of eosinophils by IL-33 through ST2 leads to the production of superoxide and IL-8 and increases eosinophil survival [25]. It is important to note that IL-33 activates eosinophils at least to the same degree as IL-5 [25]. The significant correlations between IL-33 and ST2 and sputum eosinophil counts found in our study support the hypothesis that IL-33 may be involved in the development and maintenance of eosinophilic airway inflammation in non-atopic COPD subjects.

Finally, the HPC count as well as ST2 and intracellular IL-5 expression by these cells were measured in the patients with sputum eosinophilia and in those without. HPC are known to act as pro-inflammatory effector cells of allergic inflammation [11]. HPC express ST2 and IL-33 as potent activators of HPC, leading to the release of a number of cytokines and chemokines, including IL-5. HPC can differentiate into mature eosinophils at the site of inflammation in a process called “in situ eosinophilopoiesis”. IL-33 accelerates the maturation of HPC and modulates their migration into airways in allergic asthma [26, 27].

Little is known of the role of HPC in COPD. Some studies have reported reduced numbers of circulating HPC in COPD [28], while others report similar circulating and sputum HPC numbers between COPD and healthy non-atopic subjects [29]. Our present findings do not suggest any differences in circulating HPC numbers between COPD patients with sputum eosinophilia and those without. However, a striking difference was observed in the number of sputum HPC between the two groups of COPD patients, with significantly elevated HPC numbers found in those with sputum eosinophils > 3%. This was accompanied by overexpression of intracellular IL-5 and ST2 by sputum HPC indicating increased activation of these cells in eosinophilic COPD, analogously to allergic asthma.

As IL-33 modulates the trafficking of HPC, it is possible that increased IL-33 levels may be at least partially responsible for the augmented influx of HPC into airways observed in COPD patients with eosinophilic inflammation. In addition, increased numbers of ST2 + IL-5 + HPC were seen in the sputum of patients with airway eosinophilia. This finding suggests that IL-33 activates HPC in eosinophilic COPD. Therefore, in those subjects, HPC may act as effector cells in an analogous way to allergic asthma, by fostering the development of a local IL-5 rich environment independent of the IgE pathway.

There are several limitations to our study. First, the IL-33 protein levels were low in a significant number of exhaled breath and sputum specimens. This could be due to the rapid neutralization of IL-33 following its release from activated cells. Measuring IL-33 protein content is challenging and previous studies give varying results for serum and sputum [30, 31]. Nonetheless, our findings on ST2 expression confirm the IL-33 measurements and support the association between IL-33 and eosinophilic phenotype of COPD. The best way to determine IL-33 expression would be to measure it directly in the main source of the cytokine, i.e. the airway epithelium; however, studies comparing IL-33 expression in eosinophilic COPD involving invasive methods are warranted. In addition, the results may have been affected by the fact that our group of COPD subjects was older than those of the other two groups. However, as no correlation has been found between IL-33 and ST2 expression and the age of participant, it is unlikely that this may be the case.

In conclusion, our results suggest that increased IL-33 is associated with airway eosinophilia in non-atopic COPD. It is tempting to speculate that IL-33 is involved in the recruitment and activation of HPC into the airways. This may result in the creation of a local, IL-5 rich inflammatory state similar to that observed in allergic asthma. Thus, IL-33 may be a potential therapeutic target in the subgroup of COPD patients characterized by eosinophilic inflammation.