The role of skin inflammation, barrier dysfunction, and oral tolerance in skin sensitization to gluten-derived hydrolysates in a rat model

Abstract

Background: Adverse reactions to wheat-containing skin care products have been linked to food allergy development.

Objectives: To determine the role of skin barrier dysfunction and inflammation in sensitization to gluten-derived hydrolysates via the skin in Brown Norway rats with and without oral tolerance to wheat.

Methods: Skin barrier defect was induced by mechanical disruption, and skin inflammation was induced by topical application of SLS or MC903. Unmodified, enzyme hydrolyzed, or acid hydrolyzed gluten products were applied to the skin three times per week for 5 weeks. Subsequently, rats were orally gavaged with unmodified gluten.

Results: Wheat-naïve rats were readily sensitized to gluten hydrolysates via the skin. Skin barrier defect and skin inflammation had little effect on the skin sensitization and hydrolysate-specific IgE levels. Oral administration of unmodified gluten promoted the production of unmodified gluten-specific IgE in rats sensitized via the skin. Sensitization through intact skin, disrupted skin barrier, or inflamed skin was unable to break tolerance to unmodified gluten in rats on a wheat-containing diet.

Conclusions: Mechanical skin barrier disruption and skin inflammation play a limited role in experimental skin sensitization to gluten-derived hydrolysates.

Keywords: animal model; food allergy; gluten; oral tolerance; skin barrier dysfunction; skin inflammation; skin sensitization; wheat.

FIGURE 1

Histology of intact skin, barrier disrupted skin, and SLS‐ or MC903‐treated skin from the pilot study in Brown Norway rats. Damaged skin induced by light scratching with sandpaper once a week for 2 weeks (Days 0 and 7). Skin inflammation was induced by SLS or MC903 application to skin of Brown Norway rats 5 times/week for 2 weeks (Day 0–4 and 7–11). Induction of inflammation was evaluated by histology using haematoxylin and eosin (HE)‐stained skin sections collected after sacrifice. The bar shows the length of 50 μm. Examples of fibrin deposition, epidermal thickening with enhanced keratinization, and infiltration of mononuclear cells are indicated using black, blue, and red arrows, respectively.

FIGURE 2

Animal experimental design. Barrier disrupted skin (damaged) was induced by slightly scratching using sandpaper once a week for 5 weeks (Days 7, 14, 21, 28, and 35) on shaven skin in Brown Norway rats. Skin inflammation was induced by topical application of sodium lauryl sulphate (SLS) or the vitamin D analogue MC903 5 times/week for 6 weeks (Days 0–4, 7–11, 14–18, 21–25, 28–32, and 35–39) on shaven skin. Intact skin was only shaven. Gluten‐derived products (unmodified gluten (Un Glu), enzyme hydrolyzed gluten (E Glu), or acid hydrolyzed gluten (Ac Glu 1–3) or PBS alone were applied to the rats with different skin conditions 3 times/week for 5 weeks (Days 7–9, 14–16, 21–23, 28–30, and 35–37). Oral gavages with Un Glu were administrated on Days 42 and 49. Blood samples were collected before and after the oral gavages (Days 42 and 57). An ear swelling test (EST) to Un Glu was performed on Day 56, and rats were sacrificed Day 57. Pictures were provided by BioRender.com.

FIGURE 3

Transepidermal water loss across intact skin, barrier disrupted skin, or inflamed skin induced by SLS or MC903 in wheat‐naïve Brown Norway rats. Change in transepidermal water loss (TEWL) at the time of the first skin application of gluten‐derived products compared with baseline (Day 0 vs. 7 [intact, and SLS‐ or MC903‐treated skin], and before vs. after sandpaper treatment [barrier disrupted defect, damaged]). Data show mean (±SEM) change in TEWL (n = 12–48 per group). The level of statistically significant differences between indicated groups are shown using asterisks: *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 4

Product‐specific IgG1 and IgE levels following application of gluten‐derived products to intact skin, barrier disrupted skin, or inflamed skin induced by SLS or MC903 in wheat‐naïve rats. Product‐specific (A) IgG1 and (B) IgE levels following 5 weeks of application (Day 42) of unmodified gluten (Un Glu), enzyme hydrolyzed gluten (E Glu), or acid hydrolyzed gluten (Ac Glu 1–3) products to intact skin, barrier disrupted skin (damaged), or inflamed skin induced by SLS or MC903 in wheat‐naïve Brown Norway rats. Each symbol represents a single rat and horizontal lines indicate median values (n = 7–8 per group). The level of statistically significant differences between indicated groups are shown using asterisks: *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 5

Cross‐reactivity and clinical reactivity to unmodified gluten following application of gluten‐derived products to intact skin, barrier disrupted skin, or inflamed skin induced by SLS or MC903 in wheat‐naïve rats. Unmodified gluten (Un Glu)‐specific IgE levels (A) before (Day 42), and (B) before (Day 42) versus after (Day 57) two oral gavages of Un Glu following 5 weeks application of Un Glu, enzyme hydrolyzed gluten (E Glu), or acid hydrolyzed gluten (Ac Glu 1–3) products to intact skin, barrier disrupted skin (damaged), or inflamed skin induced by SLS or MC903 in wheat‐naïve Brown Norway rats. (C) Ear swelling in response (Day 56) to intradermal Un Glu injections following the two oral gavages of Un Glu. Each symbol represents a single rat and horizontal lines indicate median values (n = 7–8 per group). The level of statistically significant differences between before and after oral postimmunization with Un Glu (B) and between indicated groups (C) are shown using asterisk: *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 6

Product‐specific IgG1 and IgE levels following application of gluten‐derived products to intact skin, barrier disrupted skin, or inflamed skin induced by SLS or MC903 in wheat‐tolerant rats. Product‐specific (A) IgG1 and (B) IgE levels following 5 weeks of application (Day 42) of PBS, unmodified gluten (Un Glu), or acid hydrolyzed gluten (Ac Glu 2) products to intact skin, barrier disrupted skin (damaged), or inflamed skin induced by SLS or MC903 in wheat‐tolerant Brown Norway rats. Each symbol represents a single rat and horizontal lines indicate median values (n = 6–8 per group). The level of statistically significant differences between control PBS group and treatment groups are shown using asterisks: *p < 0.05; **p < 0.01; ***p < 0.001.