Invariant natural killer T-cell neutralization is a possible novel therapy for human eosinophilic esophagitis

Abstract

Eosinophilic esophagitis (EoE) is a recently recognized inflammatory disorder that needs a potential therapeutic strategy. We earlier showed that iNKT cell-deficient mice are protected from allergen-induced EoE. Therefore, we now tested the hypothesis that iNKT cells are induced in the human EoE and is a novel possible target for the treatment of human EoE. Accordingly, we examine number of iNKT cells and eosinophils and expression of iNKT-associated cell surface receptors and chemokines by performing immunofluorescence, qPCR and ELISA in the esophageal biopsies and blood samples of normal subjects (comparison control) and EoE patients. Herein, we show that iNKT cell number, their receptor subcomponents Vα24 and Vβ11 expression, and associated chemokine CXCL16 levels (or expression) are induced significantly in EoE patients compared with normal individuals. In addition, we show that CXCL16 levels (or expression) correlate with the mRNA levels of Vα24 receptor but not well with esophageal eosinophilia in human EoE. Of note, we show that in vivo activation of iNKT cells is sufficient to induce EoE in mice. Furthermore, we show that anti-mCD1d- and anti-hVα24Jα18-neutralizing antibody treatment protects allergen-induced experimental EoE. Taken together, we have shown first time that iNKT cells have a critical pathogenic role in human and experimental EoE. iNKT cell neutralization by humanized anti-CD1d and anti-Vα24Jα18 antibodies might be a novel and potential therapy for human EoE.

Keywords: CXCL16; PBS57; anti-CD1d; anti-Vα24Jα18; iNKT cells.

Figure 1

Detection of iNKT cells in the esophageal biopsies of normal individuals and EoE patients. Immunofluorescence staining of esophageal biopsies was performed by using an iNKT cell-specific receptor anti-hVα24Jα18 antibody (eBioscience) and DAPI mounting material of normal (comparison control) individuals and EoE patients. A representative photomicrograph of an EoE patient biopsy detected a number of anti-hVα24Jα18-positive cells in the esophageal biopsies of EoE patients ((A), original magnification × 400 photomicrograph (b)). The DAPI-merged photomicrograph clearly shows that the Vα24Jα18 receptors are expressed on a number of cells in the epithelial mucosa ((A), original magnification × 400 photomicrograph (c)). Esophageal biopsies of normal (comparison control) subjects had non-detectable anti-hVα24Jα18-positive cells as shown following merging with a DAPI-stained photomicrograph ((B), original magnification × 400 (e, f)). The esophageal biopsies of isotype-matched IgG (eBioscience) antibody did not show any positive staining in the esophageal epithelial mucosa of EoE patients ((C), original magnification × 400 (h); n=6 normal (comparison control) individuals and n=7 EoE patients). The quantitation of iNKT cells in normal (comparison control) individuals and EoE patients are shown in (D). The level of significance was calculated by using the Mann–Whitney test. P<0.0001, n=6–7. A high magnification photomicrograph shows cell surface staining (red) by an anti-hVα24Jα18 antibody of DAPI stained nucleus ((E), original magnification × 1000 (i)).

Figure 2

Analysis of number of iNKT cells in the blood of normal individuals and EoE patients. Flow cytometry analysis was performed to examine iNKT cells in the total blood cells using an APC-conjugated human anti-hVα24Jα18 antibody. A representative dot-plot analysis of normal (comparison control) subjects and EoE patients is shown (a, b). Total blood cells from normal subjects (a), or EoE patients (b) were analyzed for iNKT cells by gating CD3+ cells for Vα24Jα18+ utilizing isotype-matched control IgG or loaded APC-conjugated anti-hVα24Jα18 antibody. The absolute number as well as percent of iNKT cells is shown in normal (comparison control) subjects, active EoE patients and non-EoE or chronic esophagitis (CE) patients, n=8/subject (c, d). The levels of mRNA expression of iNKT cell-specific receptors Vβ11 and Vα24 normalized with GADPH in normal individuals and EoE patients (e, f). The level of significance was calculated by using the Mann–Whitney test.

Figure 3

Analysis of iNKT cell-specific genes in human esophageal biopsies. The mRNA levels of iNKT cell surface molecule, TCR and T-cell components like CD1d (a), Vα24 (b), Vβ11 (c), Jα18, (d), CXCR6 (e) and chemokine CXCL16 (f) were examined by performing quantitative real-time PCR analysis. Each data point represents an individual patient (n=12–15 normal and 24–28 in EoE). The mRNA expression was normalized to GAPDH and expressed as relative expression. Statistical significance was calculated using the Mann–Whitney test. P-values for each experiment are provided in the figure (a–f).

Figure 4

Analysis of CXCL16 expression in human esophageal biopsies. A representative photomicrograph of baseline esophageal CXCL16 expression in normal (comparison control) subjects (a, b) and EoE patients (c, d) stained with anti-CXCL16 antibody (a–d) or isotype-matched IgG (e, f) and overlapped with a DAPI-stained mounted reagent. Correlations between the peak eosinophil number/hpf vs CXCL16 mRNA expression and Vα24 vs CXCL16 mRNA expression normalized with GADPH in human EoE (g, h). The r-value was calculated using the Spearman correlation test (n=24–28). Statistical significance was calculated using both the Mann–Whitney and Kruskal–Wallis test.

Figure 5

In vivo iNKT cell activation by PBS57 and esophageal eosinophil analysis. Mice (BALB/c) were exposed to repeated exposure of intravenous (i.v.) or intranasal (i.n.) PBS57  as per the protocol shown in (a). After 20–24 h of the last PBS57 or vehicle challenge, mice were killed, and eosinophil levels were evaluated in the esophagus (b) and BALF (c). Representative anti-MBP antibody-immunostained high-resolution photomicrographs of intraepithelial eosinophils in the esophagus after vehicle (d, original magnification × 100) or intranasal PBS57 challenge (e, original magnification × 100 and f, original magnification × 400). The absolute numbers of eosinophils in the esophagus of CD1d-null mice and wild-type mice after intranasal PBS57 challenge of mice (g). Data are expressed as mean±s.d., n=12 mice/group. EP, epithelium; LP, lamina propria; MS, muscularis mucosa.

Figure 6

iNKT neutralization protects mice from food- and aeroallergen-induced experimental EoE. The Balb/c mice were subjected to iNKT cell neutralization by injecting the anti-CD1d monoclonal antibody or isotype-matched IgG in an allergen-induced mouse model of EoE as per the protocol (a, b). The number of eosinophils in the esophagus was analyzed in the food allergen-sensitized mice treated with anti-CD1d or IgG and challenged with saline (−) or peanut (+) is shown following 24 h after last saline or peanut challenge (c). Data are expressed as mean±s.d., n=9–12 mice/group. The anti-CD1d or IgG treated and intranasally challenged with saline (−) or aeroallergen (Aspergillus fumigatus; +) were also examined for esophageal eosinophilia and are shown as eosinophils/mm² following 20 h after last Aspergillus or saline challenge (d). Data are expressed as mean+s.d., n=12 mice/group. The eosinophil numbers in bronchoalveolar lavage fluid (BALF) were counted in food allergen-sensitized mice treated with anti-CD1d or IgG and challenged with saline (−) or peanut (+) and CD1d-treated and challenged with saline or Aspergillus (e and f). Statistical significance was calculated using both the Mann–Whitney and Kruskal–Wallis test, and the significance levels are shown in respective figures.

Figure 7

Diagrammatic representation of proposed pathway of iNKT cell-induced EoE. Allergen taken by antigen-presenting cells (APCs) is offered to the conventional (CD4+ T cells) and non-conventional T cells (iNKT cells). On the basis of our previous and current data in the manuscript, we propose that both conventional and non-conventional T cells produce eosinophil active Th2 cytokines that are regulated by signal transducer and activator of transcription (STAT) family of molecules. The induction and activation of iNKT cells along with eosinophil active cytokines, and eotaxin-3 in esophageal epithelial mucosa accumulates eosinophils into the esophagus. This cartoon model summarizes the mechanism that STAT5 regulates iNKT cell-induced eosinophilic esophageal inflammation in EoE; whereas STAT6 regulates other allergic Th2 cytokines-associated eosinophilic disorders.