An Investigation Of Six SNPs Interaction To Confer Susceptibility To Atopic Dermatitis In Chinese Han Population

Background Atopic dermatitis(AD) is a common, chronic, inflammatory and itchy skin disease. The factors such as environment, gene together with infection play fatal roles in the pathogenesis of AD. Of all the factors that concerning the pathogenesis of AD, genetics is considered to be a fatal factor in the development of AD. Gene-gene/environment interactions are strongly considered to contribute to the genetic risk of common diseases, such as psoriasis, SLE and so on. Lots of gene-environment interactions of atopic dermatitis were performed, but there are few comprehensive investigations on the gene-gene(or genetic variants) interactions performed for atopic dermatitis. The association model of 6 single nucleotide polymorphisms(SNPs) which taged the most significant(P<10E-05) in our previous genome wide association study(GWAS) for atopic dermatitis were searched and the possible genetic variant interactions based on the previous GWAS data was investigated by using Generalized Multifactor Dimensionality Reduction(GMDR), Multifactor Dimensionality Reduction(MDR) and Plink 1.07Objective To investigate the genetic model for the 6 single nucleotide polymorphisms(SNPs) which taged the most significant(P<10E-05) in our previous genome wide association study(GWAS) for atopic dermatitis, and search for the possible genetic variant interactions based on the previous GWAS dataMethod In total, 4,636 AD cases(2,766 men and 1,870 women, with the mean age of5.45 years) and 13,559 unrelated healthy controls(7,159 men and 6,400 women, with the mean age of 28.01 years) were recruited in this study. All the samples used in this study were from Chinese Han population from multiple hospitals of China and the patients were diagnosed by at least two experienced dermatologists. The AD cases meeting the Hanifin-Rajka diagnostic criteria were diagnosed by at least two experienced dermatologists and the clinical data was collected from the affected individuals through a full clinical checkup. Additional demographic information was collected from the cases and controls through a structured questionnaire. All controls were healthy individuals without atopic dermatitis, other atopic diseases, systemic disorders or family history of atopic dermatitis(including first-, second- and third-degree relatives). 6 most significant SNPs rs3126085( FLG/1q21.3), rs12085366(ATP6V1G3/1q31.3), rs7701890( TMEM232- SLC25A46/5q22.1), rs17173197( PRKAG2/7q36.1, rs2393903( ZNF365/10.21.2), and rs6010620(TNFRSF6B- ZGPAT/20q13.33), all SNPs with P<10E-05) from our previous GWAS study that showed supportive association evidence in multiple stages for atopic dermatitis were conducted in an interaction study and the possible genetic variant interactions for atopic dermatitis were investigated. Comparisions of genotypes frequencies were performed using the Pearson’s Chi-square test. Fisher’s exact test was used when the expected count was <5. The genetic model(additive, dominant, or recessive model) of the single variant association for each of the SNPs was assumed based on the strongest association significance(determined through comparing, the p-value). Multifactor dimensionality reduction(MDR) algorithm were used to evaluate genetic variant interactions and models were evaluated based on the testing balanced accuracy(TBA) statistic, the cross-validation consistency(CVC) and the statistical significance of the model. The TBA measures how often individuals are correctly classified with respect to their case/control status, and the CVC evaluates the consistency with which individuals are classified. A satisfactory TBA score is above 0.50, and P is under 10E-08, CVC takes the maximum.Results The dominant model was best to describe the association of 3 SNPs(rs3126085, rs12085366, rs7701890,the statistic significance is 3.21E-12, 5.27E-04 and 9.23E-08respectively), the most significant associated evidence was observed under recessive model for SNP rs17173197(P=4.39E-20)and additive model for 2 SNPs(rs2393903 and rs6010620,the statistic significance is 1.10E-09 and 2.24E-07 respectively). We observed three pair-wise interactions including PRKAG2 and FLG SNPs(rs17173197 × rs3126085, Pcombined=1.11E-15), PRKAG2 and TMEM232-SLC25A46 SNPs(rs17173197 × rs7701890, Pcombined = 2.22E-15), PRKAG2 and TNFRSF6B-ZGPAT SNPs(rs17173197 × rs6010620, Pcombined = 6.66E-16), and a three-way interaction among PRKAG2 × TMEM232-SLC25A46 × TNFRSF6B-ZGPAT SNPs(rs17173197 × rs7701890 × rs6010620, Pcombined = 5.99E-15) were associated with atopic dermatitis.Conclusion This study indicates that three pair-wise interactions including PRKAG2 and FLG, PRKAG2 and TMEM232-SLC25A46, PRKAG2 and TNFRSF6B-ZGPAT, and a three-way interaction among PRKAG2 × TMEM232-SLC25A46 × TNFRSF6B-ZGPAT confered susceptibility to AD in our Chinese Han Population.