Soluble ST2 enhances IL‐33–induced neutrophilic and pro‐type 2 inflammation in the lungs

To the Editor,

Soluble suppression of tumorigenicity 2 (sST2) is a decoy receptor for interleukin 33 (IL‐33). We previously reported that high serum levels of sST2 predict unfavorable outcomes in patients with pneumonia ¹ or asthma, ² although its function in the lungs remains poorly understood. We hypothesized that sST2 modifies IL‐33–induced neutrophilic inflammation in the lungs. We measured levels of sST2, IL‐33, and pro‐neutrophilic or pro‐Th17 chemokines in the sputum of patients with asthma, chronic obstructive pulmonary disease (COPD), or asthma–COPD overlap (ACO), and administered a recombinant soluble form of ST2 (ST2‐Fc) and/or IL‐33 to mice.

Both sST2 and IL‐33 were high in patients with ACO, whereas only sST2 was high in patients with asthma and only IL‐33 was high in patients with COPD (Figure 1A,B; Table S1). There was no correlation between sST2 and IL‐33 levels (Figure 1C). IL‐33 levels were positively correlated with chemokine (CXC motif) ligand 8 (CXCL8) and chemokine (CC motif) ligand 20 (CCL20) but were not correlated with CXCL1 levels (Figure 1D). We expected sST2 to negatively regulate IL‐33–induced pro‐neutrophilic inflammation, but this was not the case (Figure 1E). Conversely, elevated levels of both sST2 and IL‐33 (characteristic of ACO patients) were associated with elevation of CXCL1 (as reported previously ³ ) and CCL20 levels (Figure 1F). Cell counts were not conducted at the time of collection, so they could not be analyzed. These results do not support the hypothesis that sST2 works as a decoy receptor for IL‐33 in human airways.

FIGURE 1.

Levels of sST2, IL‐33, and chemokines in the airways of patients with asthma, COPD, or ACO. (A, B) Levels of sST2 (A) or IL‐33 (B) in sputum from patients with COPD, ACO, or asthma. Data are expressed as median and interquartile range. (C–E) Scatter plots of sST2 and IL‐33 (C), IL‐33 and chemokine (D), and sST2 and chemokine (E) levels in sputum from the same patients. (F) Levels of CXCL8, CCL20, and CXCL1 in sputum from the same patients. p‐values were calculated using Spearman's rank correlations. ACO, asthma–COPD overlap; COPD, chronic obstructive pulmonary disease

Based on the clinical data, we suspected that sST2 may be a carrier protein. To test this idea, we administered mice with mouse (m) IL‐33 and/or mST2‐Fc at 2.5 and 25 μg protein, respectively (molar ratio, 1:2.8), because the concentration of sST2 in the sputum of patients with ACO was approximately 10 times that of IL‐33 (medians of 927 and 74.8 pg/m, respectively; Table S1). Surprisingly, mST2‐Fc augmented the mIL‐33–induced influx of neutrophils and dendritic cells into the alveolar spaces, as well as bronchial inflammation and secretion of CXCL1, matrix metalloproteinase 9 (MMP9), dsDNA, and CCL17 into the airspaces (Figure 2A–C). Also, mIL‐33 promoted an influx of eosinophils (Figure 2A) and secretion of IL‐5 (data not shown) into the airspaces, but ST2‐Fc did not augment this process. Furthermore, we detected ST2‐Fc–IL‐33 complexes and enhanced recovery of mIL‐33 and mST2‐Fc in BAL fluid (Figure 2D). Finally, we confirmed that ST2‐Fc–IL‐33 complex formation did occur between mST2‐Fc and mIL‐33 in vitro (Figure 2E). Therefore, mST2‐Fc augments mIL‐33–induced airway neutrophilia and pro‐type 2 inflammation via complex formation.

FIGURE 2.

Soluble form of ST2 augments IL‐33–induced inflammation in mouse lungs. Recombinant mIL‐33 (2.5 μg), mST2‐Fc (25 μg), both mIL‐33 and mST2‐Fc (2.5 and 25 μg, respectively), or PBS (50 μl) were administrated to female BALB/c mice via intratracheal spray. Differential cell counts (A), HE stains of lung tissue (B), secretion of CXCL1, MMP9 (with gelatin zymography of MMP9, upper panel), dsDNA, and CCL17 (C), and recovery of mST2‐Fc–mIL‐33 complexes, IL‐33, and ST2 (D) in BAL fluid were assessed 24 h after administration. (E) Presence of mST2‐Fc–mIL‐33 complexes in a solution of mST2‐Fc (4 μg) and mIL‐33 (0.4 μg) in PBS (50 μl) after incubation for 1 h at 37°C. The complex (mIL‐33 bound to mST2‐Fc) was isolated via immunoprecipitation using protein A beads and detected via immunoblot assay. A mixture that was not immunoprecipitated (containing free mIL‐33, free mST2‐Fc, and mST2‐Fc–mIL‐33 complexes) was also assessed (input). Data are pooled from three experiments (n = 6, each group) and are expressed as mean ± SEM. ****p < .0001, ***p < .001, **p < .01, *p < .05. p‐values were calculated using one‐way ANOVA and post hoc Holm–Sidak tests (A, C, and D). BAL, bronchoalveolar lavage; PBS, Phosphate‐buffered saline

We thus demonstrated that ST2‐Fc enhances IL‐33 activity in vivo. Similar forms of augmentation, caused by several types of cytokines and their corresponding anti‐cytokine monoclonal antibodies, have been reported. For instance, anti‐IL‐4 and anti‐granulocyte‐colony stimulating factor (G‐CSF) monoclonal antibodies enhance IL‐4 and G‐CSF, respectively, in mice, via the formation of immune complexes in the circulation system. ⁴ These findings support our discovery that ST2‐Fc augments IL‐33 activity.

Cell‐surface IL‐33 receptors are composed of cell‐surface ST2 (ST2L) and IL‐1RAcP, and a fusion protein of ST2 and IL‐1RAcP, IL‐33trap, binds to IL‐33 thirty times more strongly than sST2. ⁵ These findings support our claim that sST2 carries IL‐33 to cell‐surface IL‐33 receptors. Together, ST2‐Fc and IL‐33 induced greater neutrophilic inflammation with the release of dsDNA than IL‐33 alone. Asthmatics with high levels of neutrophil‐derived extracellular DNA in their sputum experience more asthma exacerbation and poorer lung function than those without. ⁶ Thus, sST2–IL‐33 complexes may contribute to the pathogenesis of neutrophilic asthma. In conclusion, sST2 may augment IL‐33 activity via the formation of sST2–IL‐33 complexes.

FUNDING INFORMATION

M. Watanabe reports grants from Grant‐in‐Aid for Scientific Research, grants from Environmental Restoration and Conservation Agency, grants and personal fees from Novartis, grants from GSK, grants from Pfizer, personal fees from Kyorin Pharmaceutical, personal fees from AstraZeneca, during the conduct of the study; personal fees from ThermoFisher, personal fees from Abbott, outside the submitted work. K. Nakamoto, C. Miyaoka, Y. Yoshida, J. Aso, H. Nunokawa, K. Honda, M. Nakamura, M. Tamura, A. Hirata, M. Oda, T. Saraya, and H. Ishii report grants from Grant‐in‐Aid for Scientific Research, grants from Environmental Restoration and Conservation Agency, grants and personal fees from Novartis, grants from GSK, grants from Pfizer, personal fees from Kyorin Pharmaceutical, personal fees from AstraZeneca, during the conduct of the study. K. Chibana reports personal fees from GSK, personal fees from Kyorin Pharmaceutical, personal fees from Torii, personal fees from Taiho, personal fees from Novartis, personal fees from Boehringer Ingelheim, personal fees from AstraZeneca, during the conduct of the study. S. Takata reports grants from Grant‐in‐Aid for Scientific Research, grants from Environmental Restoration and Conservation Agency, grants and personal fees from Novartis, grants from GSK, grants from Pfizer, personal fees from Kyorin Pharmaceutical, personal fees from AstraZeneca, during the conduct of the study; personal fees from Lilly, personal fees from Chugai Pharmaceutical, outside the submitted work. D. Kurai reports grants from Grant‐in‐Aid for Scientific Research, grants from Environmental Restoration and Conservation Agency, grants and personal fees from Novartis, grants from GSK, grants from Pfizer, personal fees from Kyorin Pharmaceutical, personal fees from AstraZeneca, during the conduct of the study; grants from Grant‐in‐Aid for Scientific Research, personal fees from Janssen pharmaceutical, personal fees from MSD, personal fees from Meiji seika pharma, personal fees from Pfizer, personal fees from GSK, personal fees from Kyorin Pharmaceutical, personal fees from Sumitomo Dainippon Pharma, personal fees from Astellas Pharma, outside the submitted work. H. Takizawa reports grants from Grant‐in‐Aid for Scientific Research, grants from Environmental Restoration and Conservation Agency, grants from Novartis, grants from GSK, grants from Pfizer, grants from Kyorin Pharmaceutical, grants from AstraZeneca, during the conduct of the study. T. Inui and M. Sada have nothing to disclose.

CONFLICT OF INTEREST

None of the authors have any conflicts of interest to declare.

CONSENT FOR PUBLICATION

Not applicable.

Supporting information

Appendix S1