The neutrophil-mobilizing cytokine interleukin-26 in the airways of long-term tobacco smokers

Abstract

Long-term tobacco smokers with chronic obstructive pulmonary disease (COPD) or chronic bronchitis display an excessive accumulation of neutrophils in the airways; an inflammation that responds poorly to established therapy. Thus, there is a need to identify new molecular targets for the development of effective therapy. Here, we hypothesized that the neutrophil-mobilizing cytokine interleukin (IL)-26 (IL-26) is involved in airway inflammation amongst long-term tobacco smokers with or without COPD, chronic bronchitis or colonization by pathogenic bacteria. By analyzing bronchoalveolar lavage (BAL), bronchail wash (BW) and induced sputum (IS) samples, we found increased extracellular IL-26 protein in the airways of long-term smokers in vivo without further increase amongst those with clinically stable COPD. In human alveolar macrophages (AM) in vitro, the exposure to water-soluble tobacco smoke components (WTC) enhanced IL-26 gene and protein. In this cell model, the same exposure increased gene expression of the IL-26 receptor complex (IL10R2 and IL20R1) and nuclear factor κ B (NF-κB); a proven regulator of IL-26 production. In the same cell model, recombinant human IL-26 in vitro caused a concentration-dependent increase in the gene expression of NF-κB and several pro-inflammatory cytokines. In the long-term smokers, we also observed that extracellular IL-26 protein in BAL samples correlates with measures of lung function, tobacco load, and several markers of neutrophil accumulation. Extracellular IL-26 was further increased in long-term smokers with exacerbations of COPD (IS samples), with chronic bronchitis (BAL samples ) or with colonization by pathogenic bacteria (IS and BW samples). Thus, IL-26 in the airways emerges as a promising target for improving the understanding of the pathogenic mechanisms behind several pulmonary morbidities in long-term tobacco smokers.

Keywords: Airways; IL-26; Smokers with COPD; Th17 cytokines; host defense; inflammation.

Conflict of interest statement

The authors declare that there are no competing interests associated with the manuscript.

© 2018 The Author(s).

Figures

Figure 1. IL-26 protein in the airway lumen, airway leukocytes, and bronchial tissue biopsy samples from smokers and nonsmokers

In this figure, datasets on human samples from one cohort (COSMIC cohort) are presented and all the IL-26 protein concentrations were quantitated in cell-free fluid samples using ELISA. (A) Concentrations of IL-26 in BAL fluid samples from healthy nonsmokers (n=37), smokers without COPD (n=40), and smokers with COPD (n=33). (B) Concentrations of IL-26 in BW fluid from healthy nonsmokers (n=34), smokers without COPD (n=33), and smokers with COPD (n=33). (C) Concentrations of protein in BAL fluid from smokers with or without COPD who have chronic bronchitis (n=18) and those without chronic bronchitis (n=55). (D) Representative pictures (ICF) for cellular IL-26 protein expression in AM from healthy nonsmokers (left panel), smokers without COPD (middle panel), and smokers with COPD (right panel). Each panel shows nuclear staining (blue) in the upper left quadrant, CD68 staining (of AM, red) in the lower left quadrant, IL-26 staining (green) in the upper right quadrant and CD68 and IL-26 co-staining (red and green) in the lower right quadrant. (E) The integrated density (CTCF using ImageJ after the ICF staining) of IL-26 expression in AM, n=8 for all groups. (F) Representative pictures (IHC) for the isotype control (upper left quadrant), IL-26 staining (brown) in healthy nonsmokers (lower left quadrant), IL-26 staining (brown) in smokers without COPD (upper right quadrant), and IL-26 staining (brown) in smokers with COPD (lower right quadrant). (G) The intensity of IL-26 expression in the bronchial tissue biopsies, n=8 for all groups. The horizontal lines in (A–C, E, G) indicate medians and the P-values are according to the Mann–Whitney test. The P-values <0.05 are considered significant. Concentrations of IL-26 on the y-axis (A–C) are represented in log scale.

Figure 2. Expression of the IL-26 gene and functionally related genes in airway cells from smokers with or without COPD

In this figure, datasets on human samples from one cohort (COSMIC cohort) are presented and the mRNA expression was measured by hybridization. (A) mRNA signal intensities for inherent IL-26, RORCVAR2, IL-10R2, IL-20R1, STAT1, and STAT3 genes in unsorted BAL cells from smokers with or without COPD (n=36). (B–D) Correlation of mRNA signal intensities of IL-26 and RORCVAR2 (B), IL-26 and IL-10R2 (C) and IL-26 and STAT3 genes (D). (E–H) mRNA signal intensities of inherent IL-10R2 (E), IL-20R1 (F), STAT1 (G), and STAT3 genes (H), in BAL cells from smokers with or without COPD (n=36) compared with healthy nonsmokers (n=16). The horizontal lines in (E–H) indicate medians. The data in (B–D) as well as the P-values indicted are according to the Spearman’s correlation test and the data in (E–H) are according to the Mann–Whitney test. The P-values <0.05 are considered significant.

Figure 3. IL-26 production in AM and gene expression of functionally related genes in response to WTC in vitro

In this figure, datasets on human samples from one cohort (CYREBAC cohort) are presented. The AM were enriched from BAL cell samples harvested from healthy volunteers and stimulated with WTC. The concentrations of IL-26 in the cell-free conditioned media were measured using ELISA. Gene expression analyses on the cells were performed using RT-PCR. (A) IL-26 protein concentrations in cell-free conditioned media from AM (n=4). The rest of the graphs show gene expression for (B) IL-26, (C) IL-10R2, (D) IL-20R1, (E) NF-κB, (F) STAT1, and (G) STAT3 in AM. The P-values are according to the parametric paired t test whereby (A,B) are according to the one-tailed test and (C–G) according to the two-tailed test. The horizontal lines indicate means and the P-values <0.05 are considered significant.

Figure 4. IL-26 protein concentrations in IS of smokers with COPD, in relation to exacerbations and age

In this figure, datasets on human samples from one cohort (BALO cohort) are presented and all the IL-26 protein concentrations were quantitated in cell-free IS fluid samples using ELISA. (A) Concentrations of IL-26 in IS fluid during stable clinical conditions compared with exacerbations (n=13), (B) concentrations of IL-26 in IS fluid over time; during stable clinical conditions, prior to exacerbation and during exacerbation (n=9). (C) Concentrations of IL-26 in IS fluid from COPD-current smokers (n=10) and COPD-former smokers (n=21) during stable clinical conditions and for COPD-current smokers (n=5) and COPD-former smokers (n=11) during exacerbations. (D) Concentrations of IL-26 in IS fluid from healthy nonsmokers without COPD (n=8) and smokers with COPD (n=31) and COPD exacerbations, as assessed during stable clinical conditions (n=16). (E) Concentrations of IL-26 in IS fluid samples from smokers with COPD in relation to age during stable clinical conditions (n=31). (F) Concentrations of IL-26 in IS fluid samples from smokers with COPD, in relation to age during exacerbations (n=16). The P-values are according to Wilcoxon signed ranked test for (Figure 3A), Friedman test (Figure 3B), Mann–Whitney test (Figure 3C,D), and Spearman’s rank correlation (Figure 3E,F). The P-values <0.05 are considered significant. Concentrations of IL-26 on the y-axis are represented in log scale. Notably, the term “stable” in the figure panels signifies “stable clinical conditions” in the legend.

Figure 5. IL-26 protein concentration in relation to long-term and short-term exposure to tobacco smoke and lung function

In this figure, datasets on human samples from three cohorts (COSMIC, Smoke Expo, and CYREBAC cohort) are presented and all IL-26 protein concentrations were quantitated in cell-free fluid samples using ELISA. (A) Concentrations of IL-26 in BAL fluid samples from smokers with or without COPD (COSMIC cohort), in relation to pack years (n=73). (B) Concentrations of IL-26 in BAL fluid samples from smokers with or without COPD (COSMIC cohort) in relation to the number of cigarettes smoked per day within the last 6 months (n=63). (C) Concentrations of IL-26 in BAL fluid samples from occasional smokers without COPD after smoking ten cigarettes within 48 h (n=9) and in never-smokers without COPD (n=5) (Smoke Expo cohort). (D) Concentrations of IL-26 in BW fluid samples (COSMIC cohort) in relation to FEV1% predicted (E) and FEV1/FVC% (n=100). The P-values are according to Spearman’s rank correlation test (A,B,D,E), Mann–Whitney U test (C). The horizontal lines in (C) indicate medians. The P-values <0.05 are considered significant. Concentrations of IL-26 on the y-axis (A,B,D,E) are represented in log scale. Abbreviations: FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity.

Figure 6. IL-26 protein concentrations in BW samples in relation to growth of pathogenic bacteria

In this figure, datasets on human samples from one cohort (COSMIC cohort) are presented and all IL-26 protein concentrations in the cell-free BW fluid were quantitated using ELISA. (A) Concentrations of IL-26 in relation to growth of pathogenic (n=7) compared with commensal (n=54) bacterial species in samples from smokers with or without COPD. (B) Concentrations of IL-26 in relation to growth pathogenic bacterial species (n=10) compared with no growth of bacterial species (n=11) in samples from smokers with or without COPD, plus healthy nonsmokers. (C) Concentrations of IL-26 in relation to growth of pathogenic (n=10) compared with commensal (n=78) bacterial species in samples from smokers with or without COPD plus healthy nonsmokers. (D) Concentrations of IL-26 in samples from smokers without COPD, without chronic bronchitis and without growth of pathogenic bacterial species (n=24), compared with samples from healthy nonsmokers with no growth of pathogenic bacterial species (n=31). The horizontal lines indicate medians and the p-values are according to the Mann-Whitney test. The p-values < 0.05 are considered significant. Concentrations of IL-26 in the y-axis are represented in log scale.

Figure 7. IL-26 protein concentrations in IS samples in relation to growth of pathogenic bacteria

In this figure, datasets on human samples from one cohort (BALO cohort) are presented. The IL-26 protein concentrations were quantitated in cell-free fluid samples using ELISA, and bacterial growth in IS fluid samples from smokers with COPD were determined. (A) Concentrations of IL-26 in relation to growth of pathogenic bacterial species (n=16) compared with no growth of pathogenic bacterial species (n=15) during stable clinical conditions, as well as in relation to growth of pathogenic bacterial species (n=12) compared with no growth of pathogenic bacterial species (n=4) during exacerbations. (B) Concentrations of IL-26 in relation to chronic growth of pathogenic bacterial species (n=12) compared with no chronic growth of pathogenic bacterial species (n=19) during stable clinical conditions. (C) Concentrations of IL-26 in relation to growth of H. influenzae (n=11) compared with no growth of H. influenzae growth) (n=20) at any visit during stable clinical conditions, as well as in relation to growth of H. influenzae (n=8) compared with no growth of H. influenzae growth (n=8) during exacerbations. (D) Concentrations of IL-26 in relation to chronic growth of H. influenzae (+) (n=9) compared with no chronic growth of H. influenzae (–) (n=22) during stable clinical conditions. (E) Concentrations of IL-26 within the same subjects when there is growth of pathogenic bacterial species and when there is no growth of pathogenic bacterial species (n=7)during stable clinical conditions. (F) Correlation between IL-26 concentrations and purulence score during stable clinical conditions (n=31) and (G) during exacerbations (n=16). The horizontal lines indicate medians and the P-values are according to Mann–Whitney test (Figure 6A–H), Wilcoxon signed ranked test (Figure 6I), Spearman’s rank correlation (Figure 6J,K). The P-values <0.05 are considered significant. Concentrations of IL-26 on the y-axis are represented in log scale. Notably, the term “stable” in the figure panels signifies “stable clinical conditions” in the legend.

Figure 8. Correlation of IL-26 concentrations with markers of neutrophil accumulation in IS samples

In this figure, datasets from two human cohorts (COSMIC and BALO cohorts) are presented and IL-26 protein concentrations in the cell-free fluid samples were quantitated using ELISA. (A) Correlation between concentrations of IL-26 and neutrophils in BAL samples from smokers with or without COPD (COSMIC cohort: n=73). (B) Correlation between concentrations of IL-26 and IL-8 (ELISA) during stable clinical conditions (n=31) and (C) during exacerbations (n=16) in IS fluid samples from smokers with COPD (BALO cohort). (D) Correlation between concentrations of IL-26 and LTB4 (ELISA) during stable clinical conditions (n=31) and (E) during exacerbations (n=16) in IS fluid samples from smokers with COPD (BALO cohort). (F) Correlation between concentrations of IL-26 and MPO activity (ELISA) during stable clinical conditions (n=31) and (G) during exacerbations (n=16) in IS fluid samples from smokers with COPD (BALO cohort). The data and P-values indicated are according to the Spearman’s correlation test. The P-values <0.05 are considered significant. Concentrations of IL-26 on the y-axis are represented in log scale.

Figure 9. Effects of rhIL-26 on pro-inflammatory genes in human AM

In this figure, datasets on human samples from one cohort (CYREBAC cohort) are presented. AM were enriched from BAL cells harvested during bronchoscopy in healthy volunteers and these AM were stimulated with different concentrations of rhIL-26. Gene expression analyses were performed using RT-PCR. Graphs show gene expression for (A) NF-κB (n=3), (B) IL-1β (n=4), (C) IL-6 (n=4), (D) IL-8 (n=4), (E) TNF-α (n=4), (F) STAT1 (n=4), and (G) STAT3 (n=4). Data are presented as mean and S.E.M., P-values are according to the linear regression analyses and the P-values <0.05 are considered significant.