Dietary Emulsifiers Exacerbate Food Allergy and Colonic Type 2 Immune Response through Microbiota Modulation

Abstract

The significant increase in food allergy incidence is correlated with dietary changes in modernized countries. Here, we investigated the impact of dietary emulsifiers on food allergy by employing an experimental murine model. Mice were exposed to drinking water containing 1.0% carboxymethylcellulose (CMC) or Polysorbate-80 (P80) for 12 weeks, a treatment that was previously demonstrated to induce significant alterations in microbiota composition and function leading to chronic intestinal inflammation and metabolic abnormalities. Subsequently, the ovalbumin food allergy model was applied and characterized. As a result, we observed that dietary emulsifiers, especially P80, significantly exacerbated food allergy symptoms, with increased OVA-specific IgE induction and accelerated type 2 cytokine expressions, such as IL-4, IL-5, and IL-13, in the colon. Administration of an antibiotic regimen completely reversed the emulsifier-induced exacerbated susceptibility to food allergy, suggesting a critical role played by the intestinal microbiota in food allergy and type 2 immune responses.

Keywords: dietary emulsifiers; food allergy; microbiota.

Figure 1

Dietary emulsifiers exacerbated food allergy-induced parameters of anaphylaxis. (A) Experimental flow chart: three-week-old BALB/c mice were treated with Water, CMC, or P80 for three months. After six weeks of exposure to Water, CMC, or P80, mice were sensitized with intraperitoneal injection with 20 mg of OVA and Alum, followed by another sensitization in two weeks. After another two weeks, the mice were orally administered 50 mg of ovalbumin five times. Mice were euthanized at the end of experiments and used for subsequent analysis. (B) Time course of hypothermia. Rectal temperature was evaluated just after the final oral challenge with OVA (n = 5, Control; n = 10, Water; n = 9, CMC; n = 10, P80). (C) Diarrhea scores. The extent of diarrhea was examined just after the final oral challenge with OVA (n = 17, Water; n = 18, CMC; n = 19, P80). (D) OVA-specific IgE in plasma collected from food allergy-induced mice was measured by ELISA (n = 10 per group). One-way ANOVA and Tukey’s Multiple Comparison Test were used to determine significance. Data are the means ± SEM. * p < 0.05, ** p < 0.01; ns, not significant.

Figure 2

(A–C) The mRNA expression for IL-4, IL-5, and IL-13 in duodenum, jejunum, ileum, and colon by using quantitative PCR. The white bar represents the control group without food allergy induction; the blue bar represents the food allergy-induced group treated with water (no dietary emulsifier treatment); orange bar represents CMC treated group; the purple bar represents P80 treated group. (n = 5, Control; n = 10, Water; n = 9, CMC; n = 9, P80). One-way ANOVA and Tukey’s Multiple Comparison Test were used to determine significance. Data are the means ± SEM. ns, not significant.

Figure 3

(A) Periodic acid-Schiff (PAS) staining for colonic tissue in food allergy-induced mice (Water, CMC, P80). (B) Goblet Cell Score of PAS-stained colonic tissue as of A. (n = 18 per group) One-way ANOVA and Tukey’s Multiple Comparison Test were used to determine significance. Data are the means ± SEM. ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 4

RNA-seq analysis of colonic CD45+ immune cells derived from the mice with food allergy. (A) Volcano plots presenting log2 fold-change of gene expression (X axis), and log10 fold-change of gene expression (Y axis) in CMC-treated mice as compared with water-treated mice. Blue dots correspond to the genes with a p value < 0.05 between CMC-treated and water-treated mice. Green dots correspond to the genes with at least a 2-fold-decreased or 2-fold-increased abundance in CMC-treated mice compared to water-treated mice. Red dots correspond to the genes with at least a 2-fold-decreased or 2-fold-increased abundance in CMC-treated mice compared to water-treated mice and a p value < 0.05. (B) Volcano plots presenting log2 fold-change of gene expression (X axis), and log10 fold-change of gene expression (Y axis) in P80-treated mice as compared with water-treated mice. Blue dots correspond to the genes with a p value < 0.05 between P80-treated and water-treated mice. Green dots correspond to the genes with at least a 2-fold-decreased or 2-fold-increased abundance in P80-treated mice compared to water-treated mice. Red dots correspond to the genes with at least a 2-fold-decreased or 2-fold-increased abundance in P80-treated mice compared to water-treated mice and a p value < 0.05. (C) Heatmaps showing the gene expressions related to mast cell activation and type 2 immune responses.

Figure 5

Alteration of microbiota in dietary emulsifiers treated mice with food allergy are expressed by Principal coordinates analysis (PCoA) plots to assess the variation among water, CMC, and P80 treated group at week 0 (A) and week 12 (B).

Figure 6

Treatment with an antibiotics regimen completely reversed aggravated parameters of food allergy. (A) Experimental flow chart. (B) Time course of hypothermia. Rectal temperature was evaluated just after the final oral challenge with OVA (n = 7, Water; n = 6, CMC; n = 6, P80). (C) Diarrhea scores. The extent of diarrhea was examined just after the final oral challenge with OVA (n = 7, Water; n = 6, CMC; n = 6, P80). (D) PAS staining for colonic tissue in food allergy-induced mice treated with an antibiotic regimen plus, water, CMC, or P80. (E) Goblet Cell Score of PAS-stained colonic tissue as of D. (n = 12 per group). One-way ANOVA and Tukey’s Multiple Comparison Test were used to determine significance. Data are the means ± SEM. * p < 0.05; ns, not significant.

Figure 6

Treatment with an antibiotics regimen completely reversed aggravated parameters of food allergy. (A) Experimental flow chart. (B) Time course of hypothermia. Rectal temperature was evaluated just after the final oral challenge with OVA (n = 7, Water; n = 6, CMC; n = 6, P80). (C) Diarrhea scores. The extent of diarrhea was examined just after the final oral challenge with OVA (n = 7, Water; n = 6, CMC; n = 6, P80). (D) PAS staining for colonic tissue in food allergy-induced mice treated with an antibiotic regimen plus, water, CMC, or P80. (E) Goblet Cell Score of PAS-stained colonic tissue as of D. (n = 12 per group). One-way ANOVA and Tukey’s Multiple Comparison Test were used to determine significance. Data are the means ± SEM. * p < 0.05; ns, not significant.