Role of RANK-L as a potential inducer of ILC2-mediated type 2 inflammation in chronic rhinosinusitis with nasal polyps

Abstract

Chronic rhinosinusitis with nasal polyps (CRSwNP) is characterized by type 2 inflammation with accumulation of activated group 2 innate lymphoid cells (ILC2s) and elevation of thymic stromal lymphopoietin (TSLP). A member of the TNF superfamily (TNFSF), TNFSF15, is known to induce the production of type 2 cytokines in ILC2s. Although ILC2s have been implicated in CRSwNP, the presence and role of TNFSFs in ILC2-mediated type 2 inflammation in CRSwNP has not been elucidated. Here, we investigate the involvement of TNFSFs in ILC2-mediated type 2 inflammation in CRSwNP. We found that receptor activator of NF-κB (RANK) ligand (RANK-L (TNFSF11)) was significantly elevated in nasal polyps (NPs), and that the receptor of RANK-L, RANK, was expressed on ILC2s in human peripheral blood and NPs. An agonistic antibody against RANK induced production of type 2 cytokines in human ILC2s, and TSLP significantly enhanced this reaction. The membrane-bound RANK-L was detected mainly on CD45 + immune cells, including T_H2 cells in NPs. The co-culture of NP-derived ILC2s and T_H2 cells significantly enhanced production of type 2 cytokines, and anti-RANK-L monoclonal antibody suppressed this enhancement. In conclusion, RANK-L, together with TSLP, may play an inductive role in the ILC2-mediated type 2 inflammation in CRSwNP.

Fig. 1

Elevation of RANK-L in NPs and expression of RANK on human ILC2s. a, e Total RNA was extracted from whole control UT (Con, n = 21) and NP tissue (n = 39) from patients with CRSwNP. Expression of mRNAs for RANK-L and OPG was analyzed using qPCR. Gene expression was normalized to a housekeeping gene, β-glucuronidase (GUSB), and expression levels were shown as % expression of GUSB. b Representative histograms of flow cytometric plots for RANK on ILC2s in blood and NPs (n = 6) are shown. Levels of cell surface expression of RANK on ILC2s are shown by geometric mean fluorescence intensity (gMFI). c A comparison of the gMFI ratio of RANK to isotype IgG1 between blood ILC2s and NP ILC2s is shown. d Protein extracts were generated from control UT (Con, n = 13), NPs from CRSwNP patients who did not have NSAID sensitivity (n = 69) and who did have NSAID sensitivity (n = 12). Expression of RANK-L protein in tissue homogenates was determined by Luminex. RANK-L protein concentrations were normalized to the concentration of total protein.*p < 0.05, **p < 0.01 and ***p < 0.001 were calculated by Mann-Whitney test (a, e), 1-way ANOVA Kruskal-Wallis test (d), paired-t test (b) and Wilcoxon test (c).

Fig. 2

RANK-mediated production of type 2 cytokines in ILC2s. a Sorted blood ILC2s (n = 14) and b NP ILC2s (n = 8) were cultured in the presence or absence of 10 μg/ml agonistic anti-RANK antibody (a-RANK) for 4 days. c-e Sorted blood ILC2s were cultured with 10 ng/ml TSLP, 10 μg/ml a-RANK and their combination (c, n = 12) in the presence or absence of 0.001% dimethyl sulfoxide (DMSO) (vehicle control), 100 nM dexamethasone (Dex) (d, n = 7) and 10 μM IMD0354 (IMD) (e, n = 3) for 4 days. The concentrations of IL-5 and IL-13 were measured by Luminex assay. Not significant (NS), *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 were calculated by Wilcoxon test (a, b) and 1-way ANOVA Kruskal-Wallis test (c-e).

Fig. 3

The main source of RANK-L was immune cells in NPs. The presence of RANK-L positive cells was assessed by immunohistochemistry. a Negative control antibody staining in NP is shown. b, c Representative immunostaining for RANK-L in NP (b) and control UT (c) are shown. d High magnification image of (b) is shown. e The number of RANK-L positive cells in control UT (n = 8) and NPs (n = 11) was counted. f, left Representative flow cytometric plots for RANK-L+ cells in NPs are shown. We gated on single, live, RANK-L positive (RANK-L+) cells compared to isotype control IgG2b. f, right RANK-L+ cells were further separated into CD45+ and CD45− populations and the frequency of RANK-L+ cells in the two populations was calculated (n = 12). *p < 0.05 and ****p < 0.0001 were calculated by Mann-Whitney test (e) and pair-t test (f).

Fig. 4

Identification of RANK-L expressing cells in NPs. a The frequency of granulocytes (Gran; SSC high and CD45+ cells), CD19+ B cells (CD45+CD19+CD3− cells), CD3+ T cells (CD45+CD3+ cells) and CD3-CD19− cells (SSC mid-low, CD45+CD3-CD19^mid-low cells) in RANK-L+CD45+ cells in NPs was calculated (n = 7). b The frequency of CD11c− cells, CD11c+ cells in RANK-L+CD45+ cells is shown (n = 7). c The frequency of mDC1 (CD1c^highCD141-CD11c+ cells), mDC2 (CD141+CD1c-CD11c+ cells) and CD11c+ non-mDCs (CD1c^mid-lowCD141− cells) in RANK-L+CD45+ cells in NPs is shown (n = 3). d, f Representative histograms of flow cytometric plots and bar graphs for RANK-L on T_H2 cells, CRTH2-CD4+ T cells (n = 9), CXCL16-HLADR+CD11c+ cells and CXCL16+HLA-DR+CD11c+ cells (n = 6), in NPs are shown. e, g Comparisons of the gMFI ratio of RANK-L to isotype IgG2b between T_H2 cells and CRTH2CD4+ T cells and between CXCL16+ cells-, CXCL16− cells and ILC2s are shown. Not significant (NS), *p < 0.05, **p < 0.01 and ****p < 0.0001 were calculated by 1-way ANOVA Kruskal-Wallis test (a, g), Holm-Sidak’s multiple comparisons test (c), paired-t test (d, f) and Wilcoxon test (b, e).

Fig. 5

The expression of CXCR6 and CCR4 on ILC2s and the elevation of CCL17 and CCL22 in NPs. a, b Representative histograms of flow cytometric plots and bar graphs for CXCR6 (a, n = 6) in NP ILC2s and CCR4 (b, n = 5) in NP ILC2s and NP T_H2 cells are shown. Total RNA was extracted from whole control UT tissue (Con, n = 21) and whole NP tissue (n = 39). c Expression of mRNAs for CCL17 and CCL22 was analyzed using qPCR. *p < 0.05, **p < 0.01 and ****p < 0.0001 were calculated by paired-t test (a, b) and Mann-Whitney test (c).

Fig. 6

Co-culture of ILC2s with RANK-L expressing cells enhanced the production of type 2 cytokines. a The schema of our co-culture study is shown. b, c, e We performed a (1:1) cell number co-culture using sorted NP ILC2s with NP CRTH2-CD4+ T cells (n = 10, b), NP ILC2s with NP T_H2 cells (n = 12, c) and NP ILC2s with NP CXCL16+ cells (n = 4, e) for 4 days. We also cultured equal numbers of NP ILC2s, NP CRTH2-CD4+ T cells, NP T_H2 and NP CXCL16+ cells separately (Individual culture) for 4 days. d, f The ratio of IL-5 and IL-13 production in the co-culture of NP ILC2s with T_H2 cells or ILC2s with CXCL16+ cells to the sum of individual culture is shown (d. n = 12, f. n = 4). g NP ILC2s were co-cultured with NP T_H2 cells in the presence or absence of 10 μg/ml denosumab and isotype human control IgG2 (n = 8) for 4 days. The concentrations of IL-5 and IL-13 were measured by Luminex assay. Cytokine amounts were shown in pg from 10,000 ILC2s plus 10,000 T cells or 10,000 CXCL16+ cells. Not significant (NS), *p < 0.05, **p < 0.01 and ***p < 0.001 were calculated by Wilcoxon test (b, c), ratio paired t-test (e) and Friedman test (g).