Diminished allergic disease in patients with STAT3 mutations reveals a role for STAT3 signaling in mast cell degranulation

Abstract

Background: Severe atopic conditions associated with elevated serum IgE are heterogeneous with few known causes. Nearly every patient with autosomal-dominant hyper-IgE syndrome (AD-HIES) due to signal transducer and activator of transcription 3 (STAT3) mutations has a history of eczematous dermatitis and elevated IgE; however, clinical atopy has never been systematically studied.

Objective: Understanding of genetic determinants of allergic disease may lead to novel therapies in controlling allergic disease.

Methods: We conducted clinical evaluation of the rates of food allergies and anaphylaxis in patients with AD-HIES, a cohort of patients with no STAT3 mutation but with similar histories of elevated IgE and atopic dermatitis, and healthy volunteers with no history of atopy. Morphine skin prick testing, ImmunoCAP assays for allergen-specific IgE, and basophil activation were measured. A model of systemic anaphylaxis was studied in transgenic mice carrying an AD-HIES mutation. STAT3 was silenced in LAD2 and primary human mast cells to study the role of STAT3 in signaling and degranulation after IgE cross-linking.

Results: Food allergies and anaphylaxis were markedly diminished in patients with AD-HIES compared with a cohort of patients with no STAT3 mutation but with similar histories of elevated IgE and atopic dermatitis. Morphine skin prick testing and basophil activation were diminished in patients with AD-HIES, whereas mice carrying an AD-HIES mutation were hyporesponsive to systemic anaphylaxis models. Rapid mast cell STAT3 serine727 phosphorylation was noted after IgE cross-linking, and inhibition of STAT3 signaling in mast cells lead to impaired FcεRI-mediated proximal and distal signaling, as well as reduced degranulation.

Conclusion: This study serves as an example for how mutations in specific atopic pathways can lead to discrete allergic phenotypes, encompassing increased risk of some phenotypes but a relative protection from others.

Keywords: 2,4-Dinitrophenol; AD-HIES; Autosomal dominant hyper-IgE syndrome; BSB; Basophile stimulation buffer; DNP; ERK; Extracellular signal-regulated kinase; Job syndrome; NIH; National Institutes of Health; PE; Phycoerythrin; STAT3; Short hairpin RNA; Signal transducer and activator of transcription 3; degranulation; mast cell; shRNA; signal transducer and activator of transcription 3.

FIG. 1

STAT3-mutant (AD-HIES) patients have reduced incidence of food allergies compared with patients with similar degrees of atopic dermatitis, high IgE, and no STAT3 mutation. A, Physician-diagnosed eczema in healthy volunteer control, patients with AD-HIES (STAT3-mutant), and atopic (no STAT3 mutation) control cohorts. B, Serum IgE levels in healthy volunteer controls (N = 25), patients with AD-HIES (N = 72), and atopic controls (N = 60). Each symbol represents a single patient. Line represents the median concentration of serum IgE. C, Physician-diagnosed food allergies in healthy volunteers, patients with AD-HIES, and atopic control patients were determined by interview. D, Incidence of physician-diagnosed anaphylaxis in patents with AD-HIES and control patients with atopy. Significance for panels C and D as determined by a 2-tailed χ² test. E–G, Allergen-specific IgE in serum or plasma was measured by ImmunoCAP assay for 3 common food allergens: egg, milk, and peanut. Dotted line represents the 95% positive predictive value of each allergen as previously reported. P values between healthy volunteers and atopic controls are P = .0013 for milk, P = .0001 for egg, and P = .0001 for peanut, respectively. Significance was determined by 2-tailed Mann-Whitney test.

FIG. 2

Reduced mast cell degranulation in STAT3-mutant humans and mice. A, Healthy volunteers (N = 26) and patients with AD-HIES (N = 31) were skin prick tested for morphine reactivity with increasing doses of morphine. Scores of 0 or 1 were considered nonreactive. Error was determined by 1-tailed Mann-Whitney test. B, Histamine reactivity in healthy volunteers (N = 26) and patients with AD-HIES (N = 31). C, A basophil activation test was performed on healthy volunteers (N = 5) and patients with AD-HIES (N = 5) with increasing concentrations of anti-IgE. Activated basophils were gated on Lin1^– CD123⁺ CD203c⁺ CD63^high. Error was determined by 1- tailed Mann-Whitney test (*P = .0278, **P = .0476). D, Immunohistochemical staining of tryptase (magnification, ×20) from skin biopsies of human flank in 2 controls (top) and 2 patients with AD-HIES (bottom).

FIG. 3

Reduced mast cell degranulation in STAT3-mutant mice. A, IgE-mediated systemic anaphylaxis in mut-Stat3 mice compared with wild-type littermate controls. Data are representative of 2 independent experiments with 4 to 5 mice per group. B, Compound 48/80 was used to induce IgE-independent anaphylaxis in mut-Stat3 and wild-type mice. Data are representative of 4 mice per group. C, Dot plot of surface FcεRI expression versus CD117 expression. D, Percentage of Lin⁻IgE⁺FcεRI⁺CD117⁺ peritoneal mast cells. Data are cumulative of 2 independent experiments of age- and sex-matched animals with 4 to 6 mice per group.

FIG. 4

STAT3 contributes to human mast cell proximal and distal FcεRI signaling events. A, STAT3 phosphorylation on Serine727 and Tyrosine705 after 2 minutes of IgE cross-linking in LAD2 and primary human mast cells (HuMCs). The change in mean fluorescence intensity between stimulated (black line) and unstimulated (gray solid) cells is given in the upper left corner. As a positive control, ERK phosphorylation at 0 (solid gray), 2 (black line), and 15 (dashed red line) minutes after cross-linking was measured by flow cytometry. Data are representative of 2 independent experiments with LAD2 cells and 1 experiment in primary human mast cells. B, Western blot of lysates from the human mast cell line LAD2 transduced with 5 different shRNAs directed against STAT3. C, Mast cell degranulation was measured by FcεRI cross-linking and subsequent β-hexosaminidase (β-hex) release in LAD2 cells transduced with 5 different shRNAs against STAT3. Data are representative of 2 independent experiments. LAD2, unstimulated control; LAD+, FcεRI cross-linking. D, Mast cell degranulation was measured by FcεRI cross-linking and subsequent β-hexosaminidase release in primary human mast cells transduced with 2 different shRNAs against STAT3. HuMC, unstimulated control; HuMC+, FcεRI cross-linking. E, Silencing of STAT3 reduced proximal and distal FcεRI-induced mast cell signaling. Phosphorylation of STAT3 (S727) was evident with antigen stimulation that was much less marked with shSTAT3. Early and proximal FcεRI signaling occurs through Lyn and Syk kinases and is evident at 2 minutes (arrows) in the scramble control cells. However, silencing of STAT3 reduces phosphorylation at 2 minutes. FcεRI signaling defects were also evident with phospholipase C, γ 1 (PLCγ1) phosphorylation at 2 minutes. Defects in distal FcεRI signaling (AKT and ERK), which occur later, were also evident with STAT3 silencing. Ag, Antigen.