* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Innate Type 2 Immunity Is Associated with Eosinophilic Pleural
Monoclonal antibody wikipedia , lookup
Molecular mimicry wikipedia , lookup
Lymphopoiesis wikipedia , lookup
Immune system wikipedia , lookup
Hygiene hypothesis wikipedia , lookup
Polyclonal B cell response wikipedia , lookup
Adaptive immune system wikipedia , lookup
Cancer immunotherapy wikipedia , lookup
Sjögren syndrome wikipedia , lookup
Immunosuppressive drug wikipedia , lookup
Psychoneuroimmunology wikipedia , lookup
Innate Type 2 Immunity Is Associated with Eosinophilic Pleural Effusion in Primary Spontaneous Pneumothorax Bo-In Kwon1*, Seokchan Hong1*z, Kihyuk Shin1, Eun-Hye Choi1,2, Jung-Joo Hwang1,3, and Seung-Hyo Lee1 1 Graduate School of Medical Science and Engineering, Biomedical Research Center, KAIST Institute for the BioCentury, Korea Advanced Institute of Science and Technology, Daejeon, Korea; 2Eulji Medi-Bio Research Institute, Eulji University, Daejeon, Korea; and 3Department of Thoracic and Cardiovascular Surgery, Eulji University Hospital, Daejeon, Korea Rationale: Eosinophilic pleural effusion (EPE) is characterized by greater than 10% eosinophilia and is frequently associated with air and/or blood in the pleural cavity. Primary spontaneous pneumothorax (PSP), defined as the spontaneous presence of air in the pleural space, is one of the most common causes of EPE. Recent studies have shown that type 2 immune responses play important roles in eosinophilic airway inflammation resulting in pleural pathology. Objectives: To determine the predominant immune responses associated with PSP in humans, and to examine whether IL-33, thymic stromal lymphopoietin (TSLP), or type 2 innate lymphoid cell (ILC2)mediated immune responses are associated factors. Methods: Eosinophil-associated cytokines were measured in the pleural fluid of patients with PSP and control subjects. Th2 cell and ILC2 responses in the pleural cavity and peripheral blood were also evaluated by in vitro restimulation and intracellular cytokine staining of T cells and ILC2s in patients with PSP (n ¼ 62) and control subjects (n ¼ 33). IL-33–mediated IL-5 production by ILC2s was also evaluated. Measurements and Main Results: Significantly higher concentrations of IL-5 and eotaxin-3 were detected in the pleural fluid of patients with PSP, in addition to significantly higher concentrations of IL-33 and TSLP. Although IL-5 production was induced by IL-33 treatment of ILC2s, other Th2 cell–mediated immune responses were not detected. Conclusions: Our results indicate that innate immune responses characterized by the production of IL-33, TSLP, and IL-5 are associated with the development of EPE in PSP by an ILC2-dependent and Th2independent mechanism. Keywords: pneumothorax; eosinophil; IL-33; IL-5; type 2 innate lymphoid cells (Received in original form February 14, 2013; accepted in final form June 10, 2013) * These authors made equal contributions to this work. z Current Address: Division of Rheumatology, Department of Internal Medicine, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea Supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2013R1A1A2010714), the Bio & Medical Technology Development Program of the National Research Foundation funded by the Korean government (2012M3A9C7050093), and a grant from Eulji Medi-Bio Research Institute (2013EMBRIDJ0002). Author Contributions: Conception and design, J.-J.H. and S.-H.L. Analysis and interpretation, B.-I.K., S.H., K.S., and E.-H.C. Drafting the manuscript for important intellectual content, B.-I.K., S.H., J.-J.H., and S.-H.L. Correspondence and requests for reprints should be addressed to Seung-Hyo Lee, Ph.D., Cellular Immunology Laboratory, Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology, Daejeon, 305701, Korea. E-mail: [email protected]; or Jung-Joo Hwang, M.D., Ph.D., Department of Thoracic and Cardiovascular Surgery, Eulji University Hospital, Daejeon, 302-799, Korea. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Crit Care Med Vol 188, Iss. 5, pp 577–585, Sep 1, 2013 Copyright ª 2013 by the American Thoracic Society Originally Published in Press as DOI: 10.1164/rccm.201302-0295OC on July 1, 2013 Internet address: www.atsjournals.org AT A GLANCE COMMENTARY Scientific Knowledge on the Subject Primary spontaneous pneumothorax (PSP) is one of the most common causes of eosinophilic pleural effusion (EPE), but its pathogenesis and associated immune responses are unknown. What This Study Adds to the Field The present study shows that innate immune responses mediated by IL-33 and type 2 innate lymphoid cells are associated with the development of EPE in PSP. Eosinophilic pleural effusion (EPE), defined as greater than 10% eosinophilia in the pleural fluid, is often associated with the presence of air and/or blood in the pleural space (1). Although a common manifestation of various pleural diseases, the precise mechanism underlying EPE is largely unknown (1). Eosinophils are responsible for host protective immunity against parasitic helminth and fungal infections (2, 3) and the pathogenesis of allergic diseases through the production of various factors including IL-4, IL-5, chemokines, and lipid mediators (4, 5). IL-5 and eotaxin, well-known regulators of eosinophils, have been implicated in the initiation and propagation of eosinophil recruitment in EPE (6–9). Primary spontaneous pneumothorax (PSP) is one of the common causes of EPE (1); however, the mechanisms and clinical significance of eosinophil accumulation in PSP are not well understood (1, 10). One of the main issues in the management of PSP is the high rate of recurrence (up to 52%) (11, 12). Inflammatory responses associated with cigarette smoking, an independent risk factor for the recurrence of PSP, may be involved in the pathogenesis of this disease (10, 13). Therefore, it is necessary to understand the mechanisms underlying development of pleural inflammation characterized by eosinophilia in PSP. Recently it has been shown that innate immune cells, such as eosinophils and epithelial cells, play critical roles in inflammation beyond simply functioning to link innate and adaptive immunity. For example, IL-33 and thymic stromal lymphopoietin (TSLP) produced by innate immune cells have been shown to mediate airway inflammation through Th2 dependent and independent mechanisms (14, 15). IL-33, a member of IL-1 super-family, is expressed by multiple cell types including epithelial cells and macrophages (14, 16). IL-33 binds to a multimeric receptor that includes ST2 and IL-1 receptor accessory proteins and initiates many inflammatory responses (17, 18). Recent studies have reported increased IL-33 in the blood and lungs during allergic diseases and hypereosinophilia in human patients and in an animal model (19–22). Moreover, a novel cell type, type 2 578 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE innate lymphoid cells (ILC2s), which secrete IL-5 and IL-13 in response to IL-33, have been discovered (23–27). Thus, IL-33 acts on innate immune cells to mediate type 2 inflammatory responses through the production of IL-5 and IL-13. In this study, we investigated whether type 2 adaptive and innate immunity is associated with PSP with EPE. We found significantly elevated IL-33 and TSLP in pleural fluid of patients with PSP. Moreover, the concentration of IL-33 in pleural fluid was positively correlated with IL-5 and eosinophilia. Finally, we showed that ILC2s produce significant amounts of IL-5 when incubated with IL-33 or pleural fluid from patients with PSP. Together, our findings indicate that type 2 innate immunity is strongly associated with the pathogenesis of EPE in PSP. METHODS VOL 188 2013 TABLE 2. CHARACTERISTICS OF STUDY SUBJECTS IN CD4 T-CELL ASSAY Sex, M/F Age, yr Height, cm Weight, kg BMI Diagnosis Pneumothorax (n ¼ 30) Control Subjects (n ¼ 24) 28/2 18.7 6 3.2* 174.2 6 6.9* 60.7 6 8.3 20.0 6 2.3* PSP (n ¼ 30) 12/12 45.4 6 21.2 163.5 6 9.2 64.8 6 12.1 24.2 6 3.5 Tbc (n ¼ 4) Cancer (n ¼ 4) Pneumonia (n ¼ 6) Hyperhydrosis (n ¼ 10) Definition of abbreviations: BMI ¼ body mass index; PSP ¼ primary spontaneous pneumothorax; Tbc ¼ tuberculosis. Data are mean 6 SD. * P , 0.001 compared with control subjects. Patient Population The study population consisted of 62 patients with PSP and 33 control subjects whose clinical characteristics are listed in Table 1. Pleural fluid lavage was performed on 30 patients with PSP and 24 control subjects and was used to isolate CD4 T cells (Table 2). Peripheral blood mononuclear cells (PBMCs) were isolated from 21 patients with PSP and 12 control subjects (Table 3). PSP was confirmed by conventional chest radiography. Antibodies and Reagents Monoclonal antibodies specific for human Pacific blue-CD3 (clone: UCHT1), Pacific blue-CD4 (clone: RPA-T4), PE-Cy7-CD8 (clone: RPA-T8), APC-Cy7-CD45 (clone: HI30), PE-Cy7-c-Kit (clone: 104D2), PerCP-Cy5.5-CD294 (CRTH2, clone: BM16), PE-ST2 (clone: 2A5), APC-Cy7-CD11b (clone: M1/70), PE-Siglec-8 (clone: 7C9), APC-IL-5 receptor a (clone: 26,815), V450-CD45 (clone: HI30), PE-IL-4 (clone: 8D4– 8), APC-IL-5 (clone: TRFK5), and fluorescein isothiocyanate–cocktail of lineage marker (Lin: CD2 [RPA-2.10], CD3 [OKT3], CD14 [61D3], CD16 [CB16], CD19 [HIB19], CD56 [CB56], and CD235a [HIR2]) were purchased from BD Biosciences (San Diego, CA), BioLegend (San Diego, CA), eBioscience (San Diego, CA), MBL International (Woburn, MA), and R&D Systems (Minneapolis, MN). Recombinant IL-5 and IL-33 was purchased from Peprotech (Rocky Hill, NJ) and polyclonal antibody against IL-5 and IL-33 from R&D Systems. Pleural Fluid Collection and Cytokine Measurement Pleural fluids were centrifuged at 350 3 g for 10 minutes at 48 C, and cell-free supernatants were stored at 2808 C. The concentrations of IL-4 (eBioscience), IL-5 (BD Biosciences), eotaxin-3 (R&D Systems), IL-13 (BD Biosciences), IL-33 (R&D Systems), and TSLP (R&D Systems) were determined by standard ELISA. TABLE 1. CHARACTERISTICS OF STUDY SUBJECTS IN CYTOKINE ASSAY Sex, M/F Age, yr Height, cm Weight, kg BMI Diagnosis Pneumothorax (n ¼ 62) Control Subjects (n ¼ 33) 57/5 18.7 6 3.2* 173.9 6 6.3* 59.6 6 8.2† 19.7 6 2.2* PSP (n ¼ 62) 18/15 49.5 6 20.6 163.9 6 9.1 64.2 6 11.5 23.9 6 3.5 Tbc (n ¼ 7) Cancer (n ¼ 9) Pneumonia (n ¼ 7) Hyperhydrosis (n ¼ 10) Definition of abbreviations: BMI ¼ body mass index; PSP ¼ primary spontaneous pneumothorax; Tbc ¼ tuberculosis. Patients with PSP had no history, nor clinical or radiographic signs of any other lung disorder except pneumothorax. *P , 0.001 compared with control subjects. y P , 0.05 compared with control subjects. Collection of Pleural Lavage Fluid and Purification of Cells Pleural fluid lavage was performed as described with minor modification (10, 28). Briefly, 200 ml of saline was injected into the pleural space, and fluids were harvested during bleb resection or other surgical procedure. Collected fluids were filtered through 40-mm nylon mesh and washed with Hanks’ balanced salt solution. CD41 and CD1a/CD1421 cell populations were isolated using magnetic cell sorting (autoMACSpro; Miltenyi Biotec Inc., Bergisch Gladbach, Germany). Cell Cultures Cell cultures were performed as described previously (29). CD4 T cells (2 3 105) cultured with or without irradiated antigen-presenting cells (APCs; 2 3 105; 3,000 rad) were stimulated with 1 mg/ml of antiCD3/CD28 antibodies for 4 days, then supernatants were collected and cytokines assessed using Milliplex MAP kit (Millipore, Bedford, MA) and a Bioplex 200 system (Bio-Rad Laboratories, Hercules, CA). Flow Cytometry and Intracellular Cytokine Staining Pleural fluid cells and PBMCs were incubated with PMA (50 ng/ml; Sigma-Aldrich, St. Louis, MO) and ionomycin (500 ng/ml) for 5 hours at 378 C in the presence of brefeldin-A (10 mg/ml). Cells were then fixed using fluorescence-activated cell sorter lysing solution, permeabilized with 0.5% saponin in phosphate-buffered saline, and incubated with antibodies for surface markers or cytokines for 30 minutes at room temperature. After washing, stained cells were analyzed by flow cytometry (LSRII; BD Biosciences). For IL-33–mediated IL-5 production, pleural fluid cells and PBMCs were treated with IL-33 with or without neutralizing antibody against IL-33 for 5 hours. For IL-5/IL-33–mediated IL-4 production, pleural fluid cells were treated with IL-5/IL-33 or pleural fluid with or without neutralizing antibodies against IL-5 and/or IL-33. TABLE 3. CHARACTERISTICS OF STUDY SUBJECTS IN PBMC ASSAY Sex, M/F Age, yr Height, cm Weight, kg BMI Diagnosis Pneumothorax (n ¼ 21) Control Subjects (n ¼ 12) 19/2 18.4 6 2.7* 175.5 6 6.7* 61.0 6 7.5 19.8 6 2.0† PSP (n ¼ 21) 5/7 33.1 6 15.4 163.1 6 9.7 66.3 6 13.5 24.8 6 3.1 Cancer (n ¼ 3) Hyperhydrosis (n ¼ 9) Definition of abbreviations: BMI ¼ body mass index; PBMC ¼ peripheral blood mononuclear cell; PSP ¼ primary spontaneous pneumothorax. Data are mean 6 SD. * P , 0.01 compared with control subjects. y P , 0.001 compared with control subjects. Kwon, Hong, Shin, et al.: ILC2-Dependent IL-5 Production in Pneumothorax Statistical Analysis Comparisons were made by nonparametric Mann-Whitney U test using GraphPad Prism Software V.5.0 (Graph Pad Software, San Diego, CA). To assess correlations, Spearman correlation analysis was performed. P values (,0.05) were considered statistically significant. RESULTS IL-4, IL-5, IL-13, and Eotaxin-3 Concentrations Are Increased in the Pleural Fluid of Patients with PSP The demographic and clinical features of patients with PSP and control subjects are described in Table 1. As shown, the percentages of eosinophils were significantly increased in the pleural fluid of patients with PSP compared with control subjects (Figures 1A and 1B), consistent with the higher frequency of EPE in patients with PSP (43 [72.9%] of 59 vs. 1 [3%] of 33). Previous studies demonstrated that in pleural fluid, the levels of IL-5 and eotaxin-3 are closely correlated with the number of eosinophils suggesting these molecules may mediate the accumulation of eosinophils (8, 9, 30). In agreement with these reports, we detected a significant difference in IL-5 concentration in the pleural fluids of patients with PSP compared with control subjects (Figure 1C) (P , 0.0001). Similarly, eotaxin-3 was detected at a much higher concentration in the pleural fluid of PSP when compared with control subjects (Figure 1D) (P , 0.0001). Given that eosinophilia and IL-5 production represent some of the key features of Th2 cell-driven immune responses, these findings point to an association between Th2 immunity and EPE in PSP. To address this, we next measured the concentration of IL-4, a canonical cytokine of Th2 cells, and IL-13. Significantly higher concentrations of IL-4 and IL-13 where detected in patients with PSP compared with control subjects (Figures 1E and 1F). Therefore, in agreement with previous reports (31, 32) our 579 findings indicate that the expression of Th2 cell-associated cytokines is increased in the pleural fluid of patients with PSP. Th2 Cell–mediated Immune Responses Are Not Involved in PSP Based on our detection of Th2 cytokines and chemokines in the pleural fluid of PSP, we next sought to determine the contribution of CD4 T cells to the production of the canonical Th2 cytokines IL-4, IL5, and IL-13 in pleural fluid. Therefore, we isolated CD4 T cells from freshly obtained pleural lavage fluid using a smaller representative of cohort of PSP subjects (Table 2) as described previously (10). CD4 T cells were either directly activated with anti-CD3/CD28 antibodies or cocultured with irradiated autologous CD1a/CD141 cells as APCs in the presence of anti-CD3/CD28 antibodies. Because the basal levels of cytokine production were highly variable, we expressed data as fold differences in which the amount of cytokine produced by antibody stimulation was expressed as a fraction of cytokine levels without stimulation. However, significant differences in the production of IL-4 and IL-5 between PSP and control groups were not detected after in vitro stimulation (Figures 2A and 2B). We also could not detect significant differences in the levels of IFN-g between in vitro stimulated PSP and control CD4 T cells (Figure 2C). Similarly, no significant difference in the production of IL-4, IL-5, or IFN-g from in vitro– stimulated blood CD4 T cells was detected (Figures 2D–2F). In addition, because normalized values can be misleading and background values could affect calculated results considerably, we also presented the data as absolute values. Regardless of how the data are presented, the results between groups were not different except that cytokine production was enhanced with antibody stimulation of blood CD4 T cells, but again these effects were not different between control subjects and patients Figure 1. Eosinophils, IL-5, eotaxin-3, IL-4, and IL-13 levels in the pleural fluid from patients with primary spontaneous pneumothorax and control subjects. (A) Percentage of eosinophils in the pleural fluid of patients with PSP or control subjects. (B) Representative Giemsa-stained cytospin slides are presented. The levels of IL-5 (C), eotaxin-3 (D), IL-4 (E ), and IL-13 (F ) were quantified in pleural fluid from patients with primary spontaneous pneumothorax or control subjects. Each dot represents an individual subject. Horizontal lines represent the mean. Statistical analysis was performed using Mann-Whitney U tests. ***P , 0.001. PF ¼ pleural fluid. 580 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 188 2013 Figure 2. Th1 (IFN-g) and Th2 (IL-4, IL-5) responses of CD4 T cells from patients with primary spontaneous pneumothorax and control subjects. After stimulation of pleural fluid CD4 T cells from patients with primary spontaneous pneumothorax or control subjects with anti-CD3 and CD28 antibodies, IL-4 (A), IL-5 (B), and IFN-g (C) production was measured by ELISA. Stimulation of CD4 T cells from blood was similarly performed and IL4 (D), IL-5 (E), and IFN-g (F) production was measured. Each dot represents an individual subject. Horizontal lines represent the mean. Statistical analysis was performed using Mann-Whitney U tests. with PSP (see Figure E1 in the online supplement). Because CD4 T-cell stimulation in the presence of APCs may affect overall results in which cytokine production by activated CD4 T cells can be regulated by APCs, we also stimulated CD4 T cells with antibodies without APCs and still did not detect significant differences between groups (see Figure E2). Furthermore, we performed intracellular cytokine staining to address the question of whether the secretion of cytokines in pleural fluid was derived from CD4 T cells. Because expression of CD4 is down-regulated after phorbol myristate acetate and ionomycin stimulation, surface staining for CD4 was not possible (33). We thus used CD82/CD31 cells to select for CD4 T cells as previously described (34) (Figure 3A). The frequency of IL- 4–producing CD4 T cells in pleural fluid was not significantly increased in the PSP group (Figure 3B). Because a large number of CD8 T cells were also seen in pleural fluid of PSP and control subjects, we compared the frequency of CD8 T cells between groups and determined whether these cells produced IL-4. Similar to the results from CD4 T cells, neither the numbers of CD8 T cells nor the frequency of IL-4–producing CD8 T cells in pleural fluid differed between control and PSP groups (Figure 3C). In addition, we observed no IL-5 production from CD4 or CD8 T cells from both subjects (data not shown). Finally, we measured Th1- and Th17-associated cytokines, such as IFN-g and IL-17, and found that they were also not increased (data not shown). Collectively, these data indicate that Figure 3. Flow cytometric detection of intracellular cytokine production in CD4 and CD8 T cells obtained from pleural fluid. (A) Gating method for the identification of T cells (left and center) and representative dot plots of the staining strategy for IL-4– and IL-5–producing CD4 (right bottom) and CD8 (right top) T cells in pleural fluid. Percentages of IL4–producing CD4 (B) and CD8 (C) T cells from pleural fluid after restimulation with phorbol myristate acetate and ionomycin are shown. Each dot represents an individual subject. Horizontal lines represent the mean. Statistical analysis was performed using Mann-Whitney U tests. Kwon, Hong, Shin, et al.: ILC2-Dependent IL-5 Production in Pneumothorax 581 neither classical Th2 immune responses nor CD8 T cell–mediated responses are involved in the pathogenesis of PSP. with the frequency of eosinophils and the concentrations of IL-5 and eotaxin-3. Increased IL-33 and TSLP Levels in Pleural Fluid of Patients with PSP Induction of IL-5 in ILC2s by Treatment with IL-33 or Pleural Fluid of Patients with PSP Our findings so far indicated that adaptive immune cells including CD4 and CD8 T cells in the pleural fluid are not a significant cellular source of Th2 cytokines and chemokines, which are implicated in eosinophil recruitment in PSP. Recent evidence suggests that innate immune cells, such as macrophages, eosinophils, and epithelial cells, have diverse and critical roles in the development and progression of inflammatory diseases (35, 36). These cells can participate in allergic inflammation by producing several cytokines, chemokines, and other effector molecules in response to various stimuli including antigens, infection, and mechanical stress, and they mediate complex immune responses not confined to classical Th2 immunity (35, 36). Moreover, it was discovered that IL-33 secretion could be elicited by mechanical stress in the absence of cellular necrosis (37). Thus, we hypothesized that IL-33 and TSLP, two recently discovered cytokines produced by epithelial cells, could be involved in the activation and recruitment of eosinophils in PSP. To test this, we first measured the levels of IL-33 and TSLP in the pleural fluid of patients with PSP and found both were significantly higher than in control subjects (Figures 4A and 4B). Next, we examined the correlation between the levels of IL-33 and the frequency of eosinophils, and the levels of IL-5 and eotaxin-3 in pleural fluid to assess the potential role of these cytokines in eosinophil recruitment. As shown in Figure 4C, there was a positive correlation between the frequency of eosinophils and the concentration of IL-33 and a strong positive correlation between the concentrations of IL-5 and IL-33, and eotaxin-3 and IL-33 in pleural fluid of patients with PSP (Figures 4D and 4E). Thus, IL-33 and TSLP are induced in the pleural cavity of patients with PSP, and IL-33 levels are positively correlated Because it has been shown that IL-33 induces eosinophilia by stimulating IL-5 secretion from several different cells including CD41 Th2 cells and ILC2s (21, 25, 38), we investigated whether IL-33 detected in the pleural fluid of patients with PSP could promote the production of IL-5 from ILC2s. Immune cells were isolated from pleural fluid and the peripheral blood of patients with PSP and treated with either recombinant IL-33 or pleural fluid obtained from subjects with elevated concentrations of IL-33. First, we selected ILC2s by examining their specific cell surface molecule expression pattern identified as CD451c-Kit1CRTH21Lin2 (Figure 5A, top), as described previously (23–25). Furthermore, ILC2s express ST2, a receptor for IL-33 (Figure 5A) (14). On the contrary, the number of ILC2s in control subjects was too low to detect and significantly lower than that of ILC2 in the pleural fluid of patients with PSP (Figures 5A, bottom, and 5B). In agreement with previous reports (19, 24), we detected the induction of IL-5 by pleural fluid ILC2s from patients with PSP after IL-33 treatment, and we found that IL-5 induction was ablated with IL-33 neutralizing antibody treatment (Figures 5C and 5D). Furthermore, pleural fluid ILC2s showed enhanced production of IL-5 with pleural fluid treatment and this effect was also inhibited by the addition of anti–IL-33 antibody (Figures 5C and 5D). Because not only ILC2 but also other cells, such as CD41 Th2 cells, could produce IL-5 in response to IL-33, we examined IL-5 production by intracellular cytokine staining and showed that only CRTH2 1 ILC2 produced IL-5 in the patients with PSP (Figure 5E). Similar analyses from control subjects were effectively impossible because of the lack of ILC2 (Figures 5A, 5B and 5F). We also quantified IL-33–stimulated IL-5 and IL-13 secretion by pleural fluid cells because ILC2s produce IL-5 and Figure 4. Levels of IL-33 and thymic stromal lymphopoietin (TSLP) in patients with primary spontaneous pneumothorax (PSP) or control subjects, and the correlation of IL-33 levels with eosinophils frequency, and IL-5 and eotaxin-3 levels. The levels of IL-33 (A) and TSLP (B) in pleural fluid collected from patients with PSP and control subjects were analyzed by ELISA. (C) Correlation between IL-33 levels and the percentage of eosinophils in the pleural fluid from patients with PSP. Correlations of IL-33 with IL-5 (D) and eotaxin-3 (E). Each value represents a sample from an individual subject. Statistical analysis was performed using Mann-Whitney U tests to compare the cytokine levels among groups. ***P , 0.001. Spearman correlation analysis was performed to assess for significant correlations between quantitative variables. 582 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 188 2013 Figure 5. Induction of IL-5 by IL-33, and pleural fluid (fluid) treatment in type 2 innate lymphoid cells (ILC2s) of patients with primary spontaneous pneumothorax (PSP). (A) Gating method to identify ILC2s that express CD45, c-kit, CRTH2, and ST2 without lineage markers. (B) The frequency of ILC2s in CD451 lineage2 cells between patients with PSP and control subjects was calculated. Representative flow analysis to quantify IL-5 production induced by IL-33 and pleural fluid is presented (C) and cumulative data for IL-5 induction by IL-33 and pleural fluid treatment (D). Twodimensional scatter plot analysis for intracellular IL-5 production by ILC2 in response to IL-33 and pleural fluid (E). Aggregate IL-5 (F) and IL-13 (G) secretion data as quantified by ELISA from pleural fluid cells treated with and without IL33. Statistical analysis was performed using Mann-Whitney U tests. *P , 0.05, **P , 0.01, ***P , 0.001. IL-13 in response to IL-33 (23–25), and found significantly higher amounts of both cytokines from patients with PSP relative to control subjects (Figures 5F and 5G). Because we showed the increased levels of IL-4 in pleural fluid of the patients with PSP (Figure 1E) and neither CD4 T cells nor ILC2s produced IL-4 in response to IL-33, we tested a possibility whether eosinophils produce IL-4 by a treatment with IL-33. To so do, first we identified CD451CD11b1Siglec-81IL-5Ra (CD125)1 eosinophils with fluorescence-activated cell sorter analysis (Figure 6A) (39). After treatment with IL-33/IL-5 or pleural fluid, eosinophils in the pleural fluid of patients with PSP produced significant amounts of IL-4; however, we could not detect IL-4 production in control subjects probably because of the absence of eosinophils (Figures 6B and 6C). Lastly, we investigated a possibility in which CD41 Th2 cells might secrete IL-5 after stimulation with IL-33 because it has been shown that IL-33–stimulated Th2 cells could be activated and produce effector cytokines including IL-5 (38). Similar to anti-CD3/ CD28 antibody stimulation, we could not detect IL-5 production in CD4 T cells in the pleural fluid of patients with PSP treated with either IL-33 or pleural fluid (see Figure E3). Together, these results demonstrate that the levels of IL-33 were increased in the pleural cavity of patients with PSP and that this cytokine directly promoted the production of IL-5 from ILC2s, and specifically not Th2 cells, ultimately resulting in EPE. DISCUSSION In this study, we addressed the question of whether innate and/or adaptive immune responses are involved in the process of EPE in Kwon, Hong, Shin, et al.: ILC2-Dependent IL-5 Production in Pneumothorax 583 Figure 6. Eosinophils of patients with primary spontaneous pneumothorax (PSP) produce IL-4. (A) Gating method to identify eosinophils that express CD45, CD11b, Siglec-8, and IL-5 receptor a chain (CD125) without lineage markers. (B) Representative flow histogram analysis to quantify IL-4 induced by IL-5/IL-33 and cumulative results are presented. The IL-4 production levels were decreased by treatment with anti–IL-5/IL-33 antibodies. (C) Representative flow analysis and cumulative data for IL-4 induction by pleural fluid (fluid) treatment under the indicated conditions. Statistical analysis was performed using MannWhitney U tests. *P , 0.05, ***P , 0.001. PSP. We have shown that eosinophil-associated cytokines and chemokines, IL-5, eotaxin-3, IL-4, and IL-13, are significantly elevated in pleural fluid from patients with PSP compared with that of control subjects. Furthermore, the levels of IL-33 and TSLP, two recently discovered cytokines that mediate various inflammatory diseases through the recruitment and activation of eosinophils and other immune cells, are also significantly higher in the pleural fluid of patients with PSP. Although we were unable to detect any significant difference in Th2 immune responses between pleural fluid cells of patients with PSP and control subjects, we did find significant induction of IL-5 by ILC2s treated with IL-33. Therefore, our study suggests that innate immune responses characterized by the production of IL5, IL-33, and TSLP participate in the development of EPE in PSP through a Th2-independent and ILC2-dependent mechanism. Our results are consistent with previous reports that IL-5 and eotaxin-3 are closely correlated with EPE in post–coronary artery bypass graft pleural effusions, suggesting that these cytokines play a critical role in the development of eosinophilia in pleura (8, 9). Another study also demonstrated that pleural fluid in EPE after PSP has higher levels of several inflammatory cytokines including IL-5, IL-6, and tumor necrosis factor-a. IL-4, which is the canonical cytokine of Th2 immunity, is a potent inducer of eotaxin expression, and eosinophilic inflammation is one of the main features of Th2-mediated inflammation (2, 35). In contrast to our findings, IL-4 was not detectable in the pleural fluid of patients with EPE associated with coronary artery bypass graft, malignancy, or trauma (31, 40). Taken together, these findings strongly suggest that eosinophilic immune responses play an important role in the pathogenesis of EPE, but classical Th2-mediated responses may not be involved. We were unable to detect a significant increase in CD41 Th2associated immune responses in pleural fluid of patients with PSP. For example, there was no significant increase of IL-4– producing CD4 T cells in the pleural fluid of patients with PSP in two independent experiments. Thus, these data suggest that Th2-mediated adaptive immunity is not implicated in the eosinophilic inflammation associated with pneumothorax. Indeed, this is not surprising considering that adaptive immunity is characterized by delayed response through antigen recognition by specific T-cell receptors, but PSP occurs rapidly, before adaptive immunity is likely to become fully activated. In addition, a study involving an animal model of pneumothorax provides evidence that conventional Th2-associated immunity may not be involved in pleural eosinophilia (41). This same study also demonstrated that pneumothorax-associated EPE in mice is abolished in IL-5–deficient, but not in IL-13 knock-out mice (41). Furthermore, recent studies have suggested that innate immune cells, including airway epithelial cells and ILC2s, play a role in the initiation and propagation of airway inflammation and their mechanism of action is not confined to the classical Th2 immune response (36). Thus, our present findings indicate that innate eosinophilic immune responses, but not Th2 immunity, are significantly involved in the pathogenesis of PSP. Indeed, our data clearly show that significantly increased numbers of ILC2s exist in the pleural fluid of patients with PSP. Furthermore, ILC2s have been shown to respond to IL-33 with the expression of its receptor, ST2, and the production of IL-5 and IL-13. However, we could not find significantly increased 584 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE production of IL-5 from blood ILC2s (data now shown). Therefore, it is reasonable to hypothesize that mechanical stress may cause the release of IL-33, and ILC2s responding to IL-33 secrete IL-5 during eosinophilic inflammation over the course of PSP. IL-33, produced by hematopoietic and nonhematopoietic cells, such as dendritic cells, macrophages, epithelial cells, and endothelial cells, was originally discovered as a ligand for Th2 cell-associated receptor, ST2. It has been shown that IL-33/ ST2 signaling participates in the pathogenesis of various inflammatory diseases, such as allergic asthma, rheumatoid arthritis, and inflammatory bowel disease (18). IL-33 can activate eosinophils to produce CXCL8 and superoxide, and it enhances adhesion and survival of these cells (42, 43). In addition, IL-33 can activate bone marrow–derived mast cells and basophils to produce proinflammatory cytokines and chemokines including IL-6; IL-13; regulated upon activation, normal; and monocyte chemoattractant protein-1 (44, 45). Recent evidence suggests that IL-33 has a role in the development of airway inflammation by a T cell–independent mechanism (46). This study showed that exposure of house dust mite allergen can directly induce production of IL-33 from airway epithelium through a Toll-like receptor-mediated signaling pathway (46). We detected significantly higher levels of IL-33 and TSLP in the pleural fluid from patients with PSP, which is to our knowledge the first report describing this phenomenon. Because T cell– mediated adaptive immune responses were not apparent, we conclude that ILC2s play an important role in the course of EPE in PSP through immediate secretion of inflammatory cytokines including IL-5 and/or IL-13, resulting in inflammation, such as tissue eosinophilia. The cells responsible for the production of these cytokines were not defined; however, given that the epithelial cell lining of airways play critical roles in the initiation and progression of diverse inflammatory diseases, it is likely that mesothelial cells, which express markers for epithelial cells, and lining of the pleura contribute to eosinophilic inflammation in EPE. By definition, PSP occurs in subjects without underlying lung disease, and subpleural bullae are found in 76–100% of patients, as compared with 20% of control subjects (12). The mechanisms of bulla formation remains uncertain but a respectable explanation is that degradation of elastic fibers in lung occurs, induced by influx of inflammatory cells. Eosinophils have long been related to fibrosis because they secrete such products as inflammatory and fibrogenic mediators including lipid mediators, chemokines, and cytokines (47). After bullae have formed, inflammation-induced obstruction of the small airway increases alveolar pressure, resulting in an air leak, consequently causing pneumothorax. Furthermore, the relatively high recurrence rate of pneumothorax is a major problem in the management of PSP, and several reports suggest that PSP is associated with poor lung function (11, 12). EPE in PSP can cause continuous inflammation through their products and these could be related to the high recurrence rate of pneumothorax. Cigarette smoking is a risk factor for the development and recurrence of PSP, and some studies suggest that smoking can evoke eosinophilic inflammation in the lung (48, 49). Indeed, one report clearly demonstrates that cigarette smoke is a major etiologic factor for the development of acute eosinophilic pneumonia, using the provocation test (49). Further studies are needed to address the effects of smoking on the development of EPE in PSP. In conclusion, we have found that levels of IL-5, eotaxin-3, IL-33, and TSLP are significantly increased in patients with PSP. However, enhancement of Th2-mediated adaptive immune responses was not detected in CD4 T cells from the pleural fluid of the patients with PSP. On the contrary, our results suggest that VOL 188 2013 innate immune responses including IL-33 and TSLP secretion are likely to play an important role in the pathogenesis of EPE in PSP by a Th2-independent mechanism. More importantly, our results show that IL-33–treated ILC2s can produce IL-5, and this production is inhibited by treatment with a neutralizing antibody against IL-33. Therefore, our study suggests that innate immune responses characterized by the production of IL-5, eotaxin-3, IL33, and TSLP, participate in the development of EPE in PSP through an ILC2-dependent mechanism. Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors thank Drs. David Corry and Farrah Kheradmand and all the members of the Hye Hwa Forum for helpful comments on the manuscript. References 1. Kalomenidis I, Light RW. Eosinophilic pleural effusions. Curr Opin Pulm Med 2003;9:254–260. 2. Rothenberg ME, Hogan SP. The eosinophil. Annu Rev Immunol 2006; 24:147–174. 3. Porter P, Polikepahad S, Qian Y, Knight JM, Lu W, Tai WM, Roberts L, Ongeri V, Yang T, Seryshev A, et al. Respiratory tract allergic disease and atopy: experimental evidence for a fungal infectious etiology. Med Mycol 2011;49:S158–S163. 4. Gleich GJ. Mechanisms of eosinophil-associated inflammation. J Allergy Clin Immunol 2000;105:651–663. 5. Walsh ER, August A. Eosinophils and allergic airway disease: there is more to the story. Trends Immunol 2010;31:39–44. 6. Antony VB. Immunological mechanisms in pleural disease. Eur Respir J 2003;21:539–544. 7. Katayama H, Yokoyama A, Kohno N, Sakai K, Hiwada K, Yamada H, Hirai K. Production of eosinophilic chemokines by normal pleural mesothelial cells. Am J Respir Cell Mol Biol 2002;26:398–403. 8. Mohamed KH, Abdelhamid AI, Lee YC, Lane KB, Conner B, Hawthorne M, Light RW. Pleural fluid levels of interleukin-5 and eosinophils are closely correlated. Chest 2002;122:576–580. 9. Kalomenidis I, Stathopoulos GT, Barnette R, Guo Y, Peebles RS, Blackwell TS, Light RW. Eotaxin-3 and interleukin-5 pleural fluid levels are associated with pleural fluid eosinophilia in post-coronary artery bypass grafting pleural effusions. Chest 2005;127:2094–2100. 10. De Smedt A, Vanderlinden E, Demanet C, De Waele M, Goossens A, Noppen M. Characterisation of pleural inflammation occurring after primary spontaneous pneumothorax. Eur Respir J 2004;23:896–900. 11. Schramel FM, Postmus PE, Vanderschueren RG. Current aspects of spontaneous pneumothorax. Eur Respir J 1997;10:1372–1379. 12. Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J Med 2000; 342:868–874. 13. Goven D, Boutten A, Lecon-Malas V, Marchal-Somme J, Soler P, Boczkowski J, Bonay M. Induction of heme oxygenase-1, biliverdin reductase and H-ferritin in lung macrophage in smokers with primary spontaneous pneumothorax: role of HIF-1alpha. PLoS ONE 2010;5: e10886. 14. Schmitz J, Owyang A, Oldham E, Song Y, Murphy E, McClanahan TK, Zurawski G, Moshrefi M, Qin J, Li X, et al. IL-33, an interleukin1-like cytokine that signals via the IL-1 receptor-related protein ST2 and induces T helper type 2-associated cytokines. Immunity 2005;23: 479–490. 15. Allakhverdi Z, Comeau MR, Jessup HK, Yoon BR, Brewer A, Chartier S, Paquette N, Ziegler SF, Sarfati M, Delespesse G. Thymic stromal lymphopoietin is released by human epithelial cells in response to microbes, trauma, or inflammation and potently activates mast cells. J Exp Med 2007;204:253–258. 16. Mirchandani AS, Salmond RJ, Liew FY. Interleukin-33 and the function of innate lymphoid cells. Trends Immunol 2012;33:389–396. 17. Kakkar R, Lee RT. The IL-33/ST2 pathway: therapeutic target and novel biomarker. Nat Rev Drug Discov 2008;7:827–840. 18. Liew FY, Pitman NI, McInnes IB. Disease-associated functions of IL-33: the new kid in the IL-1 family. Nat Rev Immunol 2010;10: 103–110. Kwon, Hong, Shin, et al.: ILC2-Dependent IL-5 Production in Pneumothorax 19. Chang YJ, Kim HY, Albacker LA, Baumgarth N, McKenzie AN, Smith DE, Dekruyff RH, Umetsu DT. Innate lymphoid cells mediate influenza-induced airway hyper-reactivity independently of adaptive immunity. Nat Immunol 2011;12:631–638. 20. Kim HY, Chang YJ, Subramanian S, Lee HH, Albacker LA, Matangkasombut P, Savage PB, McKenzie AN, Smith DE, Rottman JB, et al. Innate lymphoid cells responding to IL-33 mediate airway hyperreactivity independently of adaptive immunity. J Allergy Clin Immunol 2012; 129:216–227. 21. Bartemes KR, Iijima K, Kobayashi T, Kephart GM, McKenzie AN, Kita H. IL-33-responsive lineage- CD251 CD44(hi) lymphoid cells mediate innate type 2 immunity and allergic inflammation in the lungs. J Immunol 2012;188:1503–1513. 22. Barlow JL, Bellosi A, Hardman CS, Drynan LF, Wong SH, Cruickshank JP, McKenzie AN. Innate IL-13-producing nuocytes arise during allergic lung inflammation and contribute to airways hyperreactivity. J Allergy Clin Immunol 2012;129:191–198. 23. Neill DR, Wong SH, Bellosi A, Flynn RJ, Daly M, Langford TK, Bucks C, Kane CM, Fallon PG, Pannell R, et al. Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature 2010; 464:1367–1370. 24. Moro K, Yamada T, Tanabe M, Takeuchi T, Ikawa T, Kawamoto H, Furusawa J, Ohtani M, Fujii H, Koyasu S. Innate production of T(H)2 cytokines by adipose tissue-associated c-Kit(1)Sca-1(1) lymphoid cells. Nature 2010;463:540–544. 25. Mjösberg JM, Trifari S, Crellin NK, Peters CP, van Drunen CM, Piet B, Fokkens WJ, Cupedo T, Spits H. Human IL-25- and IL-33-responsive type 2 innate lymphoid cells are defined by expression of CRTH2 and CD161. Nat Immunol 2011;12:1055–1062. 26. Spits H, Cupedo T. Innate lymphoid cells: emerging insights in development, lineage relationships, and function. Annu Rev Immunol 2012;30:647–675. 27. Spits H, Artis D, Colonna M, Diefenbach A, Di Santo JP, Eberl G, Koyasu S, Locksley RM, McKenzie AN, Mebius RE, et al. Innate lymphoid cells: a proposal for uniform nomenclature. Nat Rev Immunol 2013;13:145–149. 28. Noppen M, De Waele M, Li R, Gucht KV, D’Haese J, Gerlo E, Vincken W. Volume and cellular content of normal pleural fluid in humans examined by pleural lavage. Am J Respir Crit Care Med 2000;162: 1023–1026. 29. Lee SH, Goswami S, Grudo A, Song LZ, Bandi V, Goodnight-White S, Green L, Hacken-Bitar J, Huh J, Bakaeen F, et al. Antielastin autoimmunity in tobacco smoking-induced emphysema. Nat Med 2007;13:567–569. 30. Smit HJ, van den Heuvel MM, Barbierato SB, Beelen RJ, Postmus PE. Analysis of pleural fluid in idiopathic spontaneous pneumothorax: correlation of eosinophil percentage with the duration of air in the pleural space. Respir Med 1999;93:262–267. 31. Schandene L, Namias B, Crusiaux A, Lybin M, Devos R, Velu T, Capel P, Bellens R, Goldman M. IL-5 in post-traumatic eosinophilic pleural effusion. Clin Exp Immunol 1993;93:115–119. 32. Yokoyama A, Kohno N, Ito M, Abe M, Hiwada K, Yamada H, Matsushima K, Hirai K. Eotaxin levels in pleural effusions: comparison with monocyte chemoattractant protein-1 and IL-8. Intern Med 2000;39: 547–552. 585 33. Pala P, Hussell T, Openshaw PJ. Flow cytometric measurement of intracellular cytokines. J Immunol Methods 2000;243:107–124. 34. Grumelli S, Corry DB, Song LZ, Song L, Green L, Huh J, Hacken J, Espada R, Bag R, Lewis DE, et al. An immune basis for lung parenchymal destruction in chronic obstructive pulmonary disease and emphysema. PLoS Med 2004;1:e8. 35. Locksley RM. Asthma and allergic inflammation. Cell 2010;140:777–783. 36. Barrett NA, Austen KF. Innate cells and T helper 2 cell immunity in airway inflammation. Immunity 2009;31:425–437. 37. Kakkar R, Hei H, Dobner S, Lee RT. Interleukin 33 as a mechanically responsive cytokine secreted by living cells. J Biol Chem 2012;287: 6941–6948. 38. Kurowska-Stolarska M, Kewin P, Murphy G, Russo RC, Stolarski B, Garcia CC, Komai-Koma M, Pitman N, Li Y, Niedbala W, et al. IL-33 induces antigen-specific IL-51 T cells and promotes allergicinduced airway inflammation independent of IL-4. J Immunol 2008; 181:4780–4790. 39. Lee JJ, Jacobsen EA, Ochkur SI, McGarry MP, Condjella RM, Doyle AD, Luo H, Zellner KR, Protheroe CA, Willetts L, et al. Human versus mouse eosinophils: that which we call an eosinophil, by any other name would stain as red. J Allergy Clin Immunol 2012;130:572– 584. 40. Kalomenidis I, Mohamed KH, Lane KB, Peebles RS, Barnette R, Rodriguez RM, Light RW. Pleural fluid levels of vascular cell adhesion molecule-1 are elevated in eosinophilic pleural effusions. Chest 2003;124:159–166. 41. Kalomenidis I, Guo Y, Peebles RS, Lane KB, Papiris S, Elias J, Light RW. Pneumothorax-associated pleural eosinophilia in mice is interleukin-5 but not interleukin-13 dependent. Chest 2005;128: 2978–2983. 42. Cherry WB, Yoon J, Bartemes KR, Iijima K, Kita H. A novel IL-1 family cytokine, IL-33, potently activates human eosinophils. J Allergy Clin Immunol 2008;121:1484–1490. 43. Suzukawa M, Koketsu R, Iikura M, Nakae S, Matsumoto K, Nagase H, Saito H, Matsushima K, Ohta K, Yamamoto K, et al. Interleukin-33 enhances adhesion, CD11b expression and survival in human eosinophils. Lab Invest 2008;88:1245–1253. 44. Ali S, Huber M, Kollewe C, Bischoff SC, Falk W, Martin MU. IL-1 receptor accessory protein is essential for IL-33-induced activation of T lymphocytes and mast cells. Proc Natl Acad Sci USA 2007;104: 18660–18665. 45. Kroeger KM, Sullivan BM, Locksley RM. IL-18 and IL-33 elicit Th2 cytokines from basophils via a MyD88- and p38alpha-dependent pathway. J Leukoc Biol 2009;86:769–778. 46. Hammad H, Chieppa M, Perros F, Willart MA, Germain RN, Lambrecht BN. House dust mite allergen induces asthma via Tolllike receptor 4 triggering of airway structural cells. Nat Med 2009;15: 410–416. 47. Rothenberg ME. Eosinophilia. N Engl J Med 1998;338:1592–1600. 48. Janz DR, O’Neal HR Jr, Ely EW. Acute eosinophilic pneumonia: a case report and review of the literature. Crit Care Med 2009;37:1470–1474. 49. Uchiyama H, Suda T, Nakamura Y, Shirai M, Gemma H, Shirai T, Toyoshima M, Imokawa S, Yasuda K, Ida M, et al. Alterations in smoking habits are associated with acute eosinophilic pneumonia. Chest 2008;133:1174–1180.