Laundry detergent promotes allergic skin inflammation and esophageal eosinophilia in mice

Abstract

The prevalence of allergic diseases is on the rise, yet the environmental factors that contribute to this increase are still being elucidated. Laundry detergent (LD) that contains cytotoxic ingredients including microbial enzymes continuously comes into contact with the skin starting in infancy. An impaired skin barrier has been suggested as a route of allergic sensitization. We hypothesized that exposure of skin to LD damages the skin barrier resulting in systemic sensitization to allergens that enter through the impaired skin barrier. Mouse skin samples exposed in vitro to microbial proteases or LD exhibited physical damage, which was more pronounced in neonatal skin as compared to adult skin. Exposure of the skin to microbial proteases in vitro resulted in an increase in the levels of interleukin (IL)-33 and thymic stromal lymphopoietin (TSLP). BALB/c wild type mice epicutaneously exposed to LD and ovalbumin (OVA) showed an increase in levels of transepidermal water loss, serum OVA-specific immunoglobulin (Ig) G1 and IgE antibodies, and a local increase of Il33, Tslp, Il4 and Il13 compared with LD or OVA alone. Following intranasal challenge with OVA, mice epicutaneously exposed to LD showed an increase in allergen-induced esophageal eosinophilia compared with LD or OVA alone. Collectively, these results suggest that LD may be an important factor that impairs the skin barrier and leads to allergen sensitization in early life, and therefore may have a role in the increase in allergic disease.

Fig 1. LD or microbial proteases cause skin damage in mouse skin in vitro.

A-E. Representative H&E stained sections (A) and quantification of damage (B) of skin biopsies from 8-week-old adult mice. Representative H&E stained sections (C) and quantification of damage (D) of skin biopsies from 9 day-old neonatal mice, and comparison of damage between age groups (E) 24 hours after incubation with LD or proteases. Magnification, 100X. Scale bars: 100 μm. Data in E are plotted from the same experiments shown in B and D. F. Concentrations of IL-33 and TSLP in supernatant of adult and neonatal skin biopsies 1 hour after incubation with Savinase. Data are mean±SEMs. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.001 obtained by nonparametric one-way ANOVA (B, D, F) or nonparametric Mann-Whitney U test within each exposure condition (E). ns, not significant.

Fig 2. Epicutaneous sensitization with LD and allergen predisposes to the development of allergic skin inflammation.

A. Experimental plan. B. Gross appearance. C. TEWL. D-G. Representative H&E stained skin sections (D), epidermal and dermal thickness (E), numbers of dermal eosinophils per HPF (F), mRNA expression of cytokines as fold induction relative to saline controls (G). Arrows in D indicate examples of eosinophils. Magnification, 100X. Scale bars: 100 μm. n = 4–5 mice per group in B-D. Data are mean±SEMs.*p<0.05, **p<0.01, and ***p<0.001 by two-way ANOVA (C) or nonparametric one-way ANOVA (E-G). ns, not significant.

Fig 3. Epicutaneous sensitization with LD and allergen predisposes to the development of esophageal eosinophilia.

A. Experimental plan. B. Representative H&E stained sections of the esophagus. Arrows in B indicate examples of eosinophils. Magnification, 200X. Scale bars: 50 μm. C. Number of eosinophils per HPF. D-F. Representative flow cytometry analysis (D), frequencies (E), and numbers (F) of eosinophils in the esophagus. G. Serum levels of OVA-specific Igs. H. Splenocyte cytokine secretion in response to in vitro OVA stimulation. Percentages of eosinophils within live, CD45⁺lin^- cells are shown in D. Data are mean±SEMs.*p<0.05, **p<0.01, and ***p<0.001 obtained by nonparametric one-way ANOVA (C, E-H). ns, not significant.