The skin barrier function gene SPINK5 is associated with challenge‐proven IgE‐mediated food allergy in infants

SE Ashley, HTT Tan, P Vuillermin, SC Dharmage… - Allergy, 2017 - Wiley Online Library
SE Ashley, HTT Tan, P Vuillermin, SC Dharmage, MLK Tang, J Koplin, LC Gurrin, A Lowe
Allergy, 2017Wiley Online Library
Background A defective skin barrier is hypothesized to be an important route of sensitization
to dietary antigens and may lead to food allergy in some children. Missense mutations in the
serine peptidase inhibitor Kazal type 5 (SPINK 5) skin barrier gene have previously been
associated with allergic conditions. Objective To determine whether genetic variants in and
around SPINK 5 are associated with IgE‐mediated food allergy. Method We genotyped 71
“tag” single nucleotide polymorphisms (tag‐SNP s) within a region spanning~ 263 kb …
Background
A defective skin barrier is hypothesized to be an important route of sensitization to dietary antigens and may lead to food allergy in some children. Missense mutations in the serine peptidase inhibitor Kazal type 5 (SPINK5) skin barrier gene have previously been associated with allergic conditions.
Objective
To determine whether genetic variants in and around SPINK5 are associated with IgE‐mediated food allergy.
Method
We genotyped 71 “tag” single nucleotide polymorphisms (tag‐SNPs) within a region spanning ~263 kb including SPINK5 (~61 kb) in n=722 (n=367 food‐allergic, n=199 food‐sensitized‐tolerant and n=156 non‐food‐allergic controls) 12‐month‐old infants (discovery sample) phenotyped for food allergy with the gold standard oral food challenge. Transepidermal water loss (TEWL) measures were collected at 12 months from a subset (n=150) of these individuals. SNPs were tested for association with food allergy using the Cochran‐Mantel‐Haenszel test adjusting for ancestry strata. Association analyses were replicated in an independent sample group derived from four paediatric cohorts, total n=533 (n=203 food‐allergic, n=330 non‐food‐allergic), mean age 2.5 years, with food allergy defined by either clinical history of reactivity, 95% positive predictive value (PPV) or challenge, corrected for ancestry by principal components.
Results
SPINK5 variant rs9325071 (A⟶G) was associated with challenge‐proven food allergy in the discovery sample (P=.001, OR=2.95, CI=1.49‐5.83). This association was further supported by replication (P=.007, OR=1.58, CI=1.13‐2.20) and by meta‐analysis (P=.0004, OR=1.65). Variant rs9325071 is associated with decreased SPINK5 gene expression in the skin in publicly available genotype‐tissue expression data, and we generated preliminary evidence for association of this SNP with elevated TEWL also.
Conclusions
We report, for the first time, association between SPINK5 variant rs9325071 and challenge‐proven IgE‐mediated food allergy.
Wiley Online Library