Unravelling the expression of interleukin-9 in chronic rhinosinusitis: A possible role for Staphylococcus aureus

Abstract

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a Th2 biased inflammation, associated with nasal colonization of Staphylococcus (S.) aureus. Interleukin (IL)-9 is a pro-inflammatory Th2 cytokine with a pivotal role in asthma, allergy and chronic obstructive pulmonary disease (COPD), but is less studied in CRSwNP. We aimed to characterize the expression and cellular source of IL-9 and examined S. aureus as potential local trigger in CRSwNP. We showed increased numbers of interleukin-9 producing neutrophils and mononuclear cells in the tissue of CRSwNP patients. This interleukin-9 production was stimulated by S. aureus and its enterotoxin B in vitro. These findings underline the contribution of S. aureus and define IL-9 as another relevant cytokine in type 2 CRSwNP.

To the Editor

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a chronic Th2 biased inflammation of the sinonasal mucosa, with patients suffering facial pressure, nasal obstruction and reduced smell caused by the presence of nasal polyps and excessive sticky mucus production [1]. CRSwNP is associated with nasal colonization of Staphylococcus (S.) aureus in 67% of the patients [2]. S. aureus can infiltrate the nasal mucosa and modulate the CRSwNP pathogenesis via secretion of enterotoxins (Toxic shock syndrome toxin-1 (TSST-1), S. aureus enterotoxin A and B (SEA and SEB)), which causes an excessive stimulation of T-cells through their superantigen function. More specifically, S. aureus contributes to the polarization of type 2 immune responses in CRSwNP, characterized by increased levels of Th2 cytokines as interleukin (IL)-4, IL-5 and IL-13 [3]. IL-9 is another pro-inflammatory Th2 cytokine that plays a pivotal role in multiple chronic airway inflammations as asthma, chronic obstructive pulmonary disease (COPD) and allergy, but is less studied in CRSwNP [4]. In asthmatic animal models and in vitro experiments, IL-9 causes bronchial hyperresponsiveness, lung eosinophilia, elevated levels of IgE, increased mucin secretion and subepithelial accumulation of collagen – all major characteristics of the CRSwNP pathogenesis [5,6,7]. In addition, concentrations of IL-9 in bronchoalveolar lavage correlate with clinical parameters of bronchoconstriction in asthmatic patients [6]. Despite the clear effects of IL-9 on other respiratory diseases, studies on IL-9 in CRSwNP are rather limited, and cellular sources, triggers of IL-9 production and possible functions in CRSwNP are still unidentified. Therefore, we aimed to characterize the expression of IL-9 and IL-9R, and made use of S. aureus as local trigger of IL-9 production in the tissue of CRSwNP patients. Details of methods and materials used in this study are shown in Additional file 1: Materials and methods. Patient characterization of the subjects used in this study is summarized in Additional file 2: Table S1.

Both lL9 and IL9R gene expression were found significantly increased in CRSwNP (n = 40) compared to control (n = 20) tissues (p < 0.05 and p < 0.001; Fig. 1 a, b). As IL-9 protein levels in the tissue were undetectable via ELISA tests, further confirmation of increased IL-9 in CRSwNP was provided by quantification of IL-9 + cells via immunohistochemistry (IHC) staining. Indeed, numbers of IL-9⁺ cells were found significantly increased in CRSwNP (n = 13) compared to controls (n = 5) (p < 0.01; Fig. 1c). These findings are in line with an earlier study in Caucasian patients showing the increased presence of IL-9⁺ cells in CRSwNP, while to the best of our knowledge, this is the first report describing increased IL-9R expression in Caucasian CRSwNP [8]. In addition, we found significantly elevated numbers of IL-9⁺ cells in CRSwNP patients with comorbid asthma (n = 8), compared to those without comorbid asthma (n = 5) (p < 0.05; Additional file 3: Fig. S1). No association was observed between local IL-9 production and the patients’ allergy status (data not shown). Related to these observations, it was shown that numbers of IL-9-expressing cells are increased in asthmatic airways compared to healthy controls and in bronchial biopsies of asthmatic patients with nasal polyps compared to those without nasal polyps [6].

Fig. 1

IL9–IL9R gene expression and numbers of IL-9⁺ cells are increased in CRSwNP compared to controls. a IL9 and b IL9R gene expression were significantly (resp. p < .05 and p < .0001) increased in tissue of CRSwNP (n = 40) compared to controls (n = 20). c Significantly (p < .01) increased numbers of IL-9⁺ cells were observed in tissue of CRSwNP (n = 13) compared to controls (n = 5). Gene expression is expressed as normalized relative quantity (NRQ). d IL-9⁺ neutrophil (arrow) and IL-9⁻ neutrophil (arrowhead) located near each other in the same patient. Both IL-9⁺ neutrophils (e, g, i) and IL-9⁻ neutrophils (f, h, j) were observed in patients with CRSwNP. Levels of statistical significance are expressed as *p < .05, **p < .01, ***p < .001 and ****p < .0001

Morphologic analysis of the IHC stainings further showed that, similar to asthmatic models, the main producers of IL-9 in CRSwNP were mononuclear cells and neutrophils [4]. Interestingly, only a fraction (80.1 ± 14.0%) of the neutrophils were IL-9 positive (Fig. 1d–j), which might indicate that IL-9 production in neutrophils is subtype, micro-environment or activation dependent. Besides classic triggering of the T-cell receptor, CD4( +) T-cells require an additional stimulus to produce and secrete IL-9 [7]. CRSwNP pathology is affected by S. aureus and its enterotoxins as described above, but the effect of S. aureus and SEB on IL-9 production is still unknown. We therefore stimulated peripheral blood derived mononuclear cells (PBMCs) with vehicle (NS), lipopolysaccharide (LPS), S. epidermidis, S. aureus and SEB for 24 h and analyzed the IL9 gene expression in 7 independent experiments. Significant increases in IL9 gene expression were observed in PBMCs upon stimulation with S. aureus (p < 0.001) and SEB (p < 0.01), but not with S. epidermidis and LPS (Fig. 2). Moreover, the increase in IL9 gene expression in PBMCs was concentration dependent for S. aureus (Additional file 4: Fig. S2). Interestingly, S. aureus is a very specific trigger for IL-9 production as no increase was observed upon stimulation with the same concentration of S. epidermidis. These observations are in line with earlier findings from our group showing that S. aureus colonization is associated with type 2 immune responses and that SEB can induce the production of several type 2 cytokines as IL-4, IL-5 and IL-13 in CRSwNP [2, 9]. To confirm our findings, CRSwNP tissue cubes were stimulated with S. aureus and SEB, but due to low RNA quality after stimulation, and the absence of a SEB-compatible ELISA kit, we were unable to measure IL-9 production ex vivo. Altogether, we showed here that the increased levels of IL-9 in CRSwNP are produced by a fraction of neutrophils and mononuclear cells and that the IL-9 production in mononuclear cells is stimulated by S. aureus and its enterotoxin B. The increased levels of IL-9 in CRSwNP could play a vital role in the chronicity and severity of CRSwNP, but further research will be necessary to show the exact relevance of IL-9 in CRSwNP. These findings underline the contribution of S. aureus to the CRSwNP pathogenesis, and define IL-9 as another relevant cytokine, upregulated with type 2 cytokines such as IL-4, IL-5, IL-13 and others, which deserves further study.

Fig. 2

S. aureus induced IL9 gene expression in PBMCs. Levels of IL9 gene expression in PBMCs after 24 h stimulation with culture medium (NS), LPS (1 µg/ml), S. epidermidis (10⁵ CFUs/ml), S. aureus (10⁵ CFUs/ml) and SEB (1 µg/ml) in n = 7 independent experiments. IL9 gene expression in significantly increased upon 24 h stimulation with S. aureus (p < .001) and SEB (p < .01), but not upon stimulation with S. epidermidis and LPS. Levels of gene expression are expressed as fold induction (FI) versus non stimulated conditions, divided by 10³. Levels of statistical significance are expressed as *p < .05, **p < .01, ***p < .001 and ****p < .0001