Type 2 and Type 17 Invariant Natural Killer T Cells Contribute to Local Eosinophilic and Neutrophilic Inflammation and Their Function Is Regulated by Mucosal Microenvironment in Nasal Polyps

Abstract

Chronic rhinosinusitis with nasal polyps (CRSwNP) is characterized by heterogeneous inflammatory endotypes of unknown etiology. Invariant natural killer T (iNKT) cells are multifunctional innate T cells that exhibit Th1-, Th2-, and Th17-like characteristics. We investigated functional relationships between iNKT cells and inflammatory subtypes of CRSwNP. Eighty patients with CRSwNP and thirty-two control subjects were recruited in this study. Flow cytometry was used to analyze the frequencies and functions of iNKT cells and their subsets in peripheral blood mononuclear cells (PBMCs) and tissues. Polyp tissue homogenates were used to study the multifunctionality of iNKT cells. iNKT cells were significantly increased in polyps (0.41%) than in control mucosa (0.12%). iNKT cells were determined in the paucigranunlocytic (n=20), eosinophilic (n=22), neutrophilic (n=23), and mixed granulocytic (n=13) phenotypes of CRSwNP. The percentages of iNKT cells and HLA-DR⁺PD-1⁺ subsets were lower in eosinophilic or mixed granulocytic polyps than those of other phenotypes. iNKT cells and subsets were enriched in polyp tissues than in matched PBMCs. The evaluation of surface markers, transcription factors, and signature cytokines indicated that the frequencies of iNKT2 and iNKT17 subsets were significantly increased in eosinophilic and neutrophilic polyps, respectively, than in the paucigranulocytic group. Moreover, the production of type 2 (partially dependent on IL-7) and type 17 (partially dependent on IL-23) iNKT cells could be stimulated by eosinophilic and neutrophilic homogenates, respectively. Our study revealed that type 2 and type 17 iNKT cells were involved in eosinophilic and neutrophilic inflammation, respectively, in CRSwNP, while different inflammatory microenvironments could modulate the functions of iNKT cells, suggesting a role of iNKT cells in feedback mechanisms and local inflammation.

Keywords: chronic rhinosinusitis with nasal polyps (CRSwNP); eosinophilia; functional subsets; invariant natural killer T cells; neutrophilia.

Figure 1

The percentages of iNKT cells and their subsets in PBMCs and tissues from CRSwNP and control (or healthy) subjects. (A) Representative flow cytometric plots showed the frequencies of iNKT cells, CD4⁺ iNKT cells, HLA-DR⁺ iNKT cells, PD-1⁺ iNKT cells, and HLA-DR⁺PD-1⁺ iNKT cells in PBMCs from healthy controls and patients. FACS analysis for stepwise iNKT cell enumeration is based on subgating T cells for CD3 and TCR Va24. (B) The percentages of T cells, iNKT cells and iNKT subsets in PBMCs were compared between healthy donors (n = 15) and CRSwNP patients (33 samples used for total T-cell and iNKT-cell analysis, 28 samples used for iNKT subset analysis). (C) The proportions of HLA-DR⁺PD-1^-, HLA-DR⁺PD-1⁺, and HLA-DR^-PD-1^- iNKT subsets in total iNKT cells from PBMCs (n = 15) of healthy and patient subjects (n = 28). (D) Representative flow cytometric plots showed the frequencies of iNKT cells, CD4+ iNKT cells, HLA-DR⁺ iNKT cells, PD-1⁺ iNKT cells, and HLA-DR⁺PD-1⁺ iNKT cells in nasal mucosa from control and patient subjects. The gating strategy is the same as stated in (A). (E) The percentages of T cells, iNKT cells and iNKT subsets in tissue samples were compared between control subjects (32 samples used for total T-cell and iNKT-cell analysis, 23 samples used for iNKT subset analysis) and patients with CRSwNP (78 samples used for total T-cell and iNKT-cell analysis, 54 samples used for iNKT subset analysis). (F) The proportions of HLA-DR⁺PD-1^-, HLA-DR⁺PD-1⁺, and HLA-DR^-PD-1^- iNKT subsets in total iNKT cells from tissues of control (n = 23) and patient (n = 54) subjects. The Mann–Whitney test was used in comparison analysis in (B, D); data presented in (B, D) as median with interquartile range; **p<0.01, ***p<0.001. ns, not significant.

Figure 2

Relationship between percentages of iNKT cells and their subsets (CD4⁺, HLA-DR⁺, PD-1⁺, and HLA-DR⁺PD-1⁺ iNKT cells) and the infiltration patterns of inflammatory cells in CRSwNP. (A) Comparison of the percentages of iNKT cells and their subsets in paucigranulocytic, eosinophilic, neutrophilic, and mixed granulocytic polyps. (B) Comparisons of the percentages of iNKT cells and their subsets in tissues and matched PBMCs from patients with CRSwNP (n = 24). (C) Comparison of the tissue eosinophil counts in patients (n = 24) with high versus low local enrichment of total iNKT cells and their subsets. (D) Comparison of the tissue neutrophil counts in patients (n = 24) with high versus low local enrichment of total iNKT cells and their subsets. The Mann–Whitney test was used in comparison analysis in (A, C, D); the Wilcoxon matched-pairs signed rank test was used in comparison analysis in (B); data presented in (A, C, D) as median with interquartile range; *p<0.05, **p<0.01, ***p<0.001. ns, not significant.

Figure 3

Functional subsets of iNKT cells in different inflammatory endotypes of CRSwNP. (A) Representative flow cytometric plots showed the frequencies of iNKT functional subsets based on characterization of surface markers, i.e., iNKT1 (CD3⁺ TCRVα24⁺ CXCR3⁺ CCR4^-), iNKT2 (CD3⁺ TCRVα24⁺ CXCR3^- CCR4⁺), iNKT17 (CD3⁺ Vα24TCR⁺ CXCR3^- CCR4^- CCR6⁺). iNKT cells were identified through a subsequent gating strategy by lmyphocytes, CD3⁺ T cells, and TCR Va24 iNKT cells; the iNKT cells were further defined into iNKT subsets by using markers CXCR3, CCR4, and CCR6. Gates were set based on negative cell population, i.e., non-CD3⁺ T cells. (B) Comparisons of the percentages of iNKT1, iNKT2, and iNKT17 and the ratios of iNKT2/iNKT1 and iNKT17/iNKT1 in paucigranulocytic (n = 7), eosinophilic (n = 12), neutrophilic (n = 12), and mixed granulocytic (n = 6) polyp samples. (C) Comparisons of the ratios of GATA3/T-bet mRNA levels and RORc/T-bet mRNA levels in paucigranulocytic (n = 7), eosinophilic (n = 7), neutrophilic (n = 7), and mixed granulocytic (n = 6) polyp samples. The Mann–Whitney test was used in comparison analysis in (B, C); data presented in (B, C) as median with interquartile range; *p<0.05, **p<0.01, ***p<0.001.

Figure 4

Cytokine production features of iNKT cells in different inflammatory endotypes of CRSwNP. (A, B) Representative flow cytometric pictures demonstrated the production of IFN-γ, IL-5, IL-13, and IL-17A in iNKT cells and their comparisons were analyzed in paucigranulocytic (n = 4), eosinophilic (n = 4), neutrophilic (n = 4), and mixed granulocytic (n = 4) polyp samples. (C) Boolean gating for cytokine analysis. Boolean gating was analysed in Flowjo through excluding and combine gating strategy. Pie charts showed the poly-functionality of iNKT cells in four different inflammatory subtypes of CRSwNP. (D) iNKT cells producing 1, 2, and 3 types of cytokines were compared in different inflammatory groups of CRSwNP. The unpaired t test was used in comparison analysis in (B); data presented in (B, D) as mean with SD; *p<0.05, **p<0.01, ***p<0.001.

Figure 5

Functional properties of iNKT cells in PBMCs stimulated by different inflammatory endotypes of CRSwNP. (A) Representative flow cytometric pictures showed the production of IFN-γ, IL-5, IL-13, and IL-17A in iNKT cells from PBMCs which were stimulated by different inflammatory endotypes of polyp tissues. (B) Comparisons of IFN-γ⁺, IL-5⁺, IL-13⁺, and IL-17A⁺ iNKT cells from PBMCs stimulated by tissue homogenates from paucigranulocytic (n = 6), eosinophilic (n = 6), neutrophilic (n = 5) and mixed granulocytic (n = 6) polyps. The values in (B) were subtracted by the cytokine level measured in the cells incubated with IL-2 plus αGalCer. The Mann–Whitney test was used to analyze the statistical difference in (B); data presented in (B) as median with interquartile range; *p<0.05, **p<0.01.

Figure 6

Effects of neutralizing antibodies on cytokine production features of iNKT cells stimulated by tissue homogenates. (A) Percentages of IFN-γ⁺, IL-5⁺, IL-13⁺, and IL-17A⁺ iNKT cells were compared in iNKT cells treated with homogenates from eosinophilic polyps (n=5) and anti-IL-7 antibody at concentrations of 1, 10, and 100 ng/ml. (B) Percentages of IFN-γ⁺, IL-5⁺, IL-13⁺, and IL-17A⁺ iNKT cells were compared in iNKT cells treated with homogenates from neutrophilic polyps (n = 5) and anti-IL-23 antibody at concentrations of 1, 10, and 100 ng/ml. Paired t test was used in the above comparison analysis; data presented in (A, B) as mean; ^#p<0.05 versus homogenate control; **p<0.01, ***p<0.001.