Introduction

Food allergy represents an adverse reaction to dietary proteins involving immunological mechanisms and includes IgE-mediated food allergy, non-IgE-mediated food allergy, and mixed IgE- and non-IgE-mediated reactions.¹

In patients with IgE-mediated food allergy, symptoms usually appear within minutes after ingestion of the culprit food and present a broad spectrum of severity, ranging from anaphylaxis to acute urticaria, vomiting, diarrhea, and rhinitis.² Non-IgE-mediated food allergy is characterized by clinical manifestations that usually develop between 6 and 72 hours after ingestion and include vomiting, diarrhea, hematochezia, constipation, food refusal, failure to thrive, chronic cough, and eczema.^3,4 A particular type of non-IgE-mediated food allergy is eosinophilic esophagitis (EoE), a chronic immune-mediated inflammatory disease that affects both children and adults.^2,5 In younger children, the clinical manifestations of this disease might be less specific, consisting in failure to thrive, vomiting, regurgitation, and feeding refusal, while adolescents usually present with food impaction and dysphagia.^6,7

Although the prevalence of food allergy in children has shown a continuous increase over the last decades, many details regarding the exact etiology and pathophysiology still need to be elucidated.^8–10 One of the key immunological mechanisms involved in food allergy is the activation of T helper type 2 (Th2) pathways, characterized by elevated levels of proinflammatory cytokines, such as interleukin (IL)-13, IL-4, and IL-5.^10,11 Being the most studied cytokine involved in the pathogenesis of EoE, IL-13 takes part in multiple pathways that lead to the development of this disease and other allergic diseases, such as atopic dermatitis, asthma, and allergic rhinitis.^5,12,13 IL-13 is produced by T cells, natural killer cells, mast cells, basophils, and eosinophils and plays a central role in the regulation of IgE synthesis.¹³ IL-13 is able to determine the migration of eosinophils, T cells, and mast cells¹⁴ and also generates disruptions of the epithelial barrier, alterations of smooth muscle contractility, and tissue remodeling by promoting collagen deposits and angiogenesis.^5,12,14,15 IL-13 also plays a crucial role in the development of atopic dermatitis, a condition frequently encountered in patients with food allergies, the relationship between these two pathologies being characterized by complex, bilateral etiopathogenic and clinical interactions.^16,17 The clinical importance of exploring the implications of IL-13 in allergic disease is also supported by the therapeutic benefits of monoclonal antibodies that interfere with IL-4/IL-13 signaling in the management of EoE and atopic dermatitis.^7,18,19

Food allergies are characterized by a strong but only partially described genetic etiology, with first-degree relatives of patients being at an increased risk of developing the disease compared to the general population.^6,20

The IL-13 gene is located on chromosome 5, in the region 5q31, and is part of the type 2 cytokine cluster along with IL-4 and IL-5 genes.¹⁷ Several studies have found significant associations between IL-13 gene polymorphisms and atopic dermatitis risk, severity, and other characteristics of the disease.^16,17,21 Polymorphisms in the IL-13 gene were also explored in studies that found associations with IgE-mediated food allergy,^22,23 food sensitization,^24–26 asthma,^27–31 respiratory allergies,³² allergic rhinitis,^33–36 high levels of total serum IgE,^{24,25,31,37–40} and elevated cord blood IgE levels.^41–44

In this study, we aimed to assess the association between two IL-13 single-nucleotide polymorphisms (SNPs), rs20541 and rs1800925, and both IgE-mediated and non-IgE-mediated food allergy. The IL-13 rs20541 SNP is a functional variant in the coding region of the IL-13 gene (within the fourth exon) and determines an exchange of arginine with glutamine at codon 144.^25,45 The IL-13 rs1800925 SNP represents a regulatory functional polymorphism that is able to enhance the IL-13 promoter activity in Th2 human lymphocytes.⁴⁶

Given the reduced specificity of symptoms and the lack of accurate diagnostic tools in the case of food allergies, diagnosis remains challenging, leading to an increased risk of both underdiagnosing and overdiagnosing these diseases.² The cornerstone in the management of food allergy consists in strict adherence to the elimination diet,² with a considerable impact on the quality of life of children and their families and a possible risk of nutritional inadequacies. All of the aspects mentioned above point to the need to identify new risk factors and diagnostic tools that could help reach a faster, more accurate diagnosis and eventually develop more efficient preventive and therapeutic strategies.

Materials and Methods

Study population

We performed a study that included two groups of pediatric patients, one with non-IgE-mediated food allergy (NFA, N=80) and one with IgE-mediated food allergy (FA, N=85), in the evidence of Marie Curie Emergency Children’s Hospital in Bucharest. All patients were unrelated Caucasians of Romanian origin. In 74 cases, the diagnosis of non-IgE-mediated food allergy was made based on the presence of suggestive symptoms (chronic diarrhea, hematochezia, vomiting, failure to thrive, chronic abdominal pain, anemia, etc.), a positive response to the elimination diet, and the exclusion of other diagnoses. Six of the patients were diagnosed with EoE after performing upper gastrointestinal endoscopy with biopsies that revealed a peak eosinophil count of at least 15 eosinophils/high power field. For the group of patients with IgE-mediated food allergies, the diagnosis was made based on a relevant patient history of specific symptoms developed shortly after ingestion of the culprit foods and elevated levels of specific serum IgE. The control group included 115 unrelated healthy Romanian subjects with no history of allergic diseases.

The study obtained approval from the Ethics Committee of Marie Curie Emergency Children’s Hospital. All parents of patients and all controls provided consent for inclusion in the study after being informed about the tests to be performed.

Genotyping and statistical analysis

Peripheral venous blood samples were collected from all patients and controls for DNA extraction. The QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany) was used according to the manufacturer’s protocol to obtain genomic DNA. Two SNPs in the IL-13 gene were selected for analysis: rs1800925 and rs20541. Genotyping of these polymorphisms was performed using real-time polymerase chain reaction with TaqMan Allelic Discrimination Assays C_8932051_10 and C_2259921_20 (Real-time PCR System, Applied Biosystems by Thermo Fisher Scientific, Foster City, CA, USA).

OpenEpi software and PLINK v1.07 were used to assess the statistical significance of the associations. A two-sided p-value of 0.05 was considered significant. The groups of patients were described based on general and demographic characteristics, which were reported as absolute and relative frequencies for categorical variables and medians and interquartile ranges (IQR) for non-normally distributed continuous variables. Normality was assessed using Kolmogorov–Smirnov tests. The Hardy–Weinberg equilibrium (HWE) was assessed for both SNPs using the chi-squared test. Mid-P exact tests were performed in order to compare the frequencies of genotypes and alleles between NFA, FA, and the control group. Moreover, the NFA group of patients was further divided based on the presence of atopic dermatitis, a positive family history of atopy, and elevated total IgE levels, and differences in genotype and allele distribution were evaluated separately for each case. The odds ratio was calculated, and a 95% confidence interval (CI) was used. Haplotype frequencies were also analyzed, and association tests were performed.

Results

Demographic characteristics for patients are presented in Table 1. We noticed a significant difference in sex ratio between the two patient groups: NFA (38.7% males) and FA (64.7% males, p=9x10^-4, OR 0.34, 95% CI = 0.18–0.64).

For the NFA group, data were available on the presence of atopic dermatitis (28.7%), a positive family history of allergic diseases (40%), and elevated total IgE levels (32.5%).

Patient and control groups were in Hardy-Weinberg equilibrium for both investigated SNPs. The genotyping success rate was 100% for rs1800925 and 99.2% for rs20541.

Single-marker analysis revealed that the genotype distribution for IL-13 rs1800925 was significantly different between controls and NFA patients (Table 2). Considering the largely used dominant model, we found that carriers of the minor allele were more frequent in the NFA group compared to controls (50% vs. 35.6%, p=0.04, OR 1.80). Based on the over-dominant model that assumes the heterozygote has the strongest impact, we observed that subjects with the CT genotype presented a higher risk for the disease compared to CC and TT homozygotes (p=0.01, OR 2.13). The minor allele T for this polymorphism was more frequent in patients, but the difference was not statistically significant. No association with the risk of NFA was observed for rs20541.

For IgE-mediated food allergy (FA) patients, the minor allele frequencies for the two SNPs were also higher compared with controls, but not significantly. No significant differences in the allele or genotype distributions were obtained when the NFA group of patients was compared with the FA group.

A low level of linkage disequilibrium was obtained for the two polymorphisms of the IL-13 gene (r²=0.2). The analysis of haplotypes showed that all four haplotype combinations had frequencies over 5% (Table 3). Although we observed several differences in haplotype frequencies between patient groups and controls, none were statistically significant.

The analysis of the two IL-13 gene polymorphisms in relation to clinical characteristics of the patients (Tables 4, 5) revealed that in the case of children with NFA who associated AD, the minor allele T of rs1800925 was more frequent compared to children without AD (32.6% vs. 24.5%), but the difference was not statistically significant (p=0.3).

Children with NFA and a positive family history of atopic disease were more frequently carriers of the minor allele T of rs20541 (41.9% vs. 34%), and the minor allele T was more frequent in this subgroup of patients (27.4% vs. 19.1%), but statistical significance was not reached (p=0.49 and p=0.23, respectively).

The minor allele T of rs20541 showed a higher frequency in children with elevated total serum IgE levels (30.7% vs. 18.2%), but the difference was not statistically significant (p=0.08). Moreover, the TT and CT genotypes were more frequent in these patients (50.0% vs. 30.7%), but the difference did not reach statistical significance (p=0.1).

Discussions

The present study investigates, for the first time, IL-13 gene variability in non-IgE-mediated food allergy (NFA). Despite extensive research, we could not identify in the literature any association study specifically regarding IL-13 gene polymorphisms and the broad category of non-IgE-mediated food allergies. The only partially related finding was a phenome-wide association study that identified a cluster of associations between the IL-5–IL-13 region and eosinophilic esophagitis (EoE), a particular type of NFA.⁴⁷

We found a significant association between IL-13 rs1800925 and the risk of NFA. Carriers of the minor allele were at high risk for the disease. We also found that the heterozygote genotype is a risk factor for NFA compared to homozygote genotypes. The single-nucleotide polymorphism rs1800925 (-1112C/T, also known as -1055 or -1111) located in the promoter of the IL-13 gene is a functional SNP that affects the level of gene expression.⁴⁶ The 1112*T allele, which, according to comparative analysis of the IL-13 promoter, is the ancestral allele, increases gene transcription depending on various factors such as cell type, cell differentiation, and surrounding polymorphisms.⁴⁶ High levels of mRNA for IL-13 and IL-4 and low levels of mRNA for IL-5 and IL-10 were found in children with NFA compared with healthy children.⁴⁸ Another study addressing cytokine gene expression reported significantly higher levels of mRNA for IL-4, IL-10, and IL-13 in the rectum compared to the duodenum in NFA children.⁴⁹ All these data suggest a dysregulation of gene expression involving the Th2 cluster in non-IgE-mediated food allergy that requires more research efforts.

Regarding IgE-mediated food allergy, we did not find any associations for rs20541 or rs1800925 SNPs with the risk of the disease. Data on this pathology and the IL-13 gene are also sparse, but they have been reported in a couple of studies presented below.

An Australian study demonstrated an association between IL-13 gene polymorphism rs1295686 (found in complete linkage disequilibrium with the functional variant rs20541) and challenge-proven IgE-mediated food allergy, while no association was found in the case of the other eight assessed polymorphisms, including rs1800925.²² In the same study, rs1295686 and rs1800925 were found to be associated with increased total IgE levels.²² These two SNPs (rs1295686 and rs20541) were reported to be significantly associated with total IgE concentrations in a genome-wide association study on the Framingham Heart Study cohort.⁵⁰

Another study that described the association of rs20541 and rs1800925 SNPs with food allergy included Indian patients with challenge-proven shrimp allergy, one of the most frequent food allergies and a common cause of food-induced anaphylaxis.²³ Significant associations were found between the homozygous genotypes IL-13 rs20541 AA (OR=3.60, 95% CI=1.30–9.95, p=0.01) and IL-13 rs1800925 TT (OR=5.48, 95% CI=1.16–26.01, p=0.03) and the presence of challenge-proven shrimp allergy.²³ Moreover, these genotypes were associated with higher levels of shrimp-specific IgE and a higher risk of shrimp allergy in younger people compared to older people.²³ In the haplotype analysis, the combination of rs1800925 T and rs20541 A alleles was associated with both challenge-proven shrimp allergy and shrimp sensitivity.²³

The two IL-13 polymorphisms selected for our research were also investigated in four studies, including European children with food sensitization. In German children, Liu et al. found that the IL-13 rs1800925 SNP was associated with a higher risk of food sensitization in the pediatric population (children with increased levels of specific IgE antibodies to hen’s egg, cow’s milk, soy, and wheat, OR=3.49, 95% CI=1.52–8.02), while no association was found for rs20541.²⁶ In a previous study performed by the same author and including the same study population, both rs20541 and rs1800925 were associated with elevated total serum IgE levels (p<0.0001 and p=0.0002, respectively).³⁷ Bottema et al. found that the rs1800925 SNP was associated with egg sensitization at the ages of 1 and 2 years old in a study including three Dutch birth cohorts; furthermore, IL-13 rs20541 and rs1800925 SNPs were associated with increased total serum IgE levels.²⁴ The association between rs1800925 and egg sensitization was confirmed by a study conducted by Zitnik et al. that included children with atopic dermatitis (p=0.0001).²⁵ Moreover, they found associations between rs20541 and higher IgE levels (p=0.013), sensitization against egg (p=0.0001), milk (p=0.0006), and peanut (0.04).²⁵ An interesting finding of this research is that IgE levels did not differ significantly between individuals with TT and CC genotypes for rs1800925, but heterozygotes presented slightly higher levels of total IgE.²⁵

In the literature, there are multiple reports of associations between IL-13 gene SNPs rs20541 and rs1800925 and atopic dermatitis, elevated levels of total serum IgE, or asthma, while there are also studies with contradictory results.

A meta-analysis and systematic review of candidate gene associations in atopic dermatitis analyzed the IL-13 gene SNP rs1800925. The authors found a significant association between this polymorphism and the risk of developing atopic dermatitis in patients with European ancestry (based on four studies including 931 patients), but no significant associations with atopic dermatitis in patients with Asian ancestry (two studies, N=1803).¹⁷ One of the studies included in this meta-analysis also found associations between the rs1800925 SNP and increased IgE levels and asthma in the Polish population, while no associations were found with allergic rhinitis and early onset of atopic dermatitis.²¹ Lesiak et al. described no correlation between IL-13 levels and a specific genotype of the rs1800925 SNP.⁵¹ In a previous study by the same author, no associations were found between any of the analyzed polymorphisms in the IL-13, IL-4, and IL-10 genes and the severity of atopic dermatitis, clinical manifestations, or early onset of asthma.⁵² On the other hand, in a whole-population birth cohort from 1989 from the Isle of Wight (N=1456), no significant associations were found between IL-13 rs20541 and rs1800925 genotypes and the prevalence of eczema, the disease being assessed at different ages from birth to 10 years old.⁵⁵

Associations between the rs20541 SNP and higher total IgE levels were reported by several studies in both Caucasian and Asian populations.^31,38–40 A study that included fine mapping of the IL-13 rs20541 SNP also identified significant associations with atopic dermatitis and increased IgE levels.⁵⁴ However, in a Polish study assessing the IL-13 rs20541 SNP, no significant associations were found with atopic dermatitis, asthma, allergic rhino-conjunctivitis, and total IgE levels; the only significant correlation for the rs20541 SNP was with positive results of skin prick tests for aeroallergens.⁵⁵ No significant association between the IL-13 rs20541 SNP and elevated levels of total IgE was also found in a study involving two Caucasian prospective cohorts.⁵⁶

An interesting finding from a birth cohort study is the interaction between the IL-13 polymorphisms and birth order, as only in the case of firstborn children was there a significant association between rs1800925 and rs20541 and increased levels of total IgE, positive skin prick tests, and positive levels of serum inhalant-specific IgE.⁵⁷

With regard to gene–gene interactions, it was demonstrated that the interaction between IL-13 (rs20541) and STAT6 (rs1059513) led to an increased risk of developing eczema.⁵⁸ Interactions were also described between the IL-13 rs20541 SNP and the type of birth and prenatal antibiotic exposure with respect to the increased risk of atopic dermatitis.⁵⁹ Children with certain IL-13 rs20541 SNP genotypes also presented a higher risk of developing early-persistent atopic dermatitis after receiving antibiotic treatment.⁶⁰

Children with food allergies present an increased risk of developing asthma,⁶¹ another allergic disease in which the implications of the IL-13 rs20541 SNP and rs1800925 SNP have been largely evaluated. A meta-analysis that included 45 studies on rs20541 and 31 studies on rs1800925 concluded that both polymorphisms are significantly associated with the risk of developing asthma in both European and Asian populations.²⁹ The same conclusion was drawn by Cui et al. in another meta-analysis that assessed both polymorphisms.³⁰ Another study demonstrated the association between rs1800925 and elevated levels of total serum IgE in patients with asthma.⁶²

In a large study on a longitudinal birth cohort (N =2918) from 1946, IL-13 rs20541 and rs1800925 were significantly associated with a self-reported positive history of asthma and allergy, while no confounding factor was identified regarding any of the assessed environmental factors (breastfeeding duration, region of birth, parental smoking, atmosphere pollution during childhood, most recent occupation, etc.).⁶³ However, in models that evaluated the association of both polymorphisms together, the rs20541 SNP reduced the association of the rs1800925 SNP with asthma, as well as with allergy, to the extent that there was no longer a statistically significant independent effect on the outcome for rs1800925.⁶³

The interplay between genetic and environmental factors in the development of atopic diseases begins in the prenatal stage, the risk of allergy later in life being correlated with the level of cord blood IgE levels.⁴² Both IL-13 polymorphisms, rs1800925 and rs20541, were associated with elevated levels of IgE in cord blood samples, and interactions with SNPs found in other genes and environmental factors, such as maternal atopy or exposure to smoking, were also described.^41–44

Considering the fact that associations between IL-13 gene SNPs and diverse allergic pathologies were demonstrated in several studies, but some of these associations were not found to be statistically significant in other studies, more research in this area is still necessary. The main limitation of the present study is the reduced number of patients available for each type of food allergy. Additionally, the results need confirmation in cohorts from other populations.

Conclusion

The main finding of our study was the significant association between IL-13 rs1800925 and the risk of non-IgE-mediated food allergy. Individuals carrying the minor allele T, as well as CT heterozygotes, presented an increased risk of the disease. No associations were found for rs20541 or in the case of children with IgE-mediated food allergy. Additionally, in patients with non-IgE-mediated food allergy, rs1800925 and rs20541 were not associated with the risk of presenting atopic dermatitis, a positive family history of allergy, or elevated total IgE levels. These findings could represent a step toward identifying the potential of IL-13 to be part of the future generation of diagnostic tools used in food allergy, but validation is needed in larger groups of patients.