Study On Susceptibility Genes And Gene-Environment Interactions Of Preeclampsia Based On Whole Exome Sequencing

Objectives:(1)Based on a case-control study with small sample size,wholeexome sequencing was conducted,and bioinformatics analysis and association analysis were used to screen candidate gene loci related to the occurrence of preeclampsia(PE).(2)Based on a case-control study with an expanded sample size,genotyping technology was applied to detect the PE-related candidate gene loci which were screened by whole exome sequencing,and the association analysis was used to verify the association between the above candidate gene loci and the occurrence of PE.(3)Based on the epidemiological data of the above two case-control studies,combined with the candidate gene loci screened by whole exome sequencing,this study explores the independent environmental factors related to PE and evaluates the association of gene-gene and geneenvironment interactions with the development of PE.Methods:A case-control study design was used.The diagnosis of PE refers to the Guidelines for the Diagnosis and Treatment of Hypertensive Disorders in Pregnancy(2015)formulated by the Hypertensive Diseases Group in Pregnancy of the Chinese Society of Obstetrics and Gynecology.Pregnant women with PE who visited the Hunan Maternal and Child Health Hospital and the Third Xiangya Hospital of Central South University from October2019 to October 2022 were recruited as the case group.Healthy pregnant women who received pregnancy care at the obstetrics clinic of the above hospitals during the same time were recruited as the control group.After signing the informed consent form,demographic characteristics,periconceptional epidemiological data,and blood samples were obtained,and all participants were followed up until 12 weeks after delivery.The research contents and methods were as follows:(1)Based on a case-control study with a small sample size,wholeexome sequencing was conducted.Firstly,bioinformatics analysis including combined GO/KEGG enrichment analysis,protein interaction network analysis,and biological distance evaluation with known PE genes was applied to prioritize candidate genes.Secondly,combined the results of candidate gene prioritization and the expression levels of candidate genes in normal placental tissue to screen the low-frequency variants with high pathogenic effects.Furthermore,a logistic regression model based on the additive model was used to explore the association between common variants and the occurrence of PE to identify candidate gene loci that are most likely associated with the occurrence of PE.(2)Based on a case-control study with an expanded sample size,the Mass Array time-of-flight mass spectrometry was used to detect SNP genotyping of the screened candidate gene loci.Both univariate and multivariate logistic regression models were used to evaluate the association between the polymorphism of candidate gene loci and the distribution of its genetic model and the development of PE.(3)Based on the epidemiological data of the above case-control study and selected candidate gene loci,univariate and multivariate logistic regression analyses were used to explore the environmental factors independently related to the occurrence of PE.Additionally,the crossgeneration analysis,generalized multifactor dimension reduction,and logistic regression model were used to evaluate the association of genegene and gene-environment interactions with the occurrence of PE.Results:(1)A total of 235,012 variants were detected by whole exome sequencing in 50:50 case-controls after preliminary filtering,of which108,515 variants were located in the exon region or splicing region.Through bioinformatics analysis and association analysis,26 lowfrequency variation loci(0.01 ≤ MAF < 0.05)in 11 genes and 15 common variation loci(MAF ≥ 0.05)in 14 genes were preliminarily identified as new candidate gene loci for PE.(2)A 150:350 case-control study was conducted to validate the preliminarily identified candidate gene loci.After controlling the suspected confounding factors and multiple corrections,it was found that(1)the polymorphism of rs13306364 locus in the FN1 gene was associated with the occurrence of PE(dominant model: OR=4.57,95%CI: 1.89-11.06).In addition,the cumulative effect analysis was performed using the FN1rs17449032 A allele,rs6728999 T allele,and rs13306364 G allele as risk alleles,and it was found that for each additional risk allele carried by pregnant women,the risk of PE increased by 1.97 times(OR=1.97,95%CI:1.26-3.08).(2)The polymorphism of rs35713889 locus in the LAMB2 gene was associated with the occurrence of PE(dominant model: OR=5.30,95%CI: 1.91-14.69).The cumulative effect analysis was performed using the LAMB2 rs148818522 T allele,rs79677861 T allele,and rs35713889 T allele as risk alleles,and it was found that for each additional risk allele carried by pregnant women,the risk of developing PE increased by 2.00times(OR=2.00,95%CI: 1.18-3.36).(3)The mutations of the KDR gene may be associated with the occurrence of PE.The cumulative effect analysis was conducted using the KDR rs34231037 G allele and rs35636987 T allele as risk alleles,and it was found that for each additional risk allele carried by pregnant women,the risk of developing PE increased by 2.15 times(OR=2.15,95%CI: 1.15-4.01).(4)The polymorphism of rs6738031 locus in the SCN7 A gene was associated with the occurrence of PE(additive model: OR=1.64,95%CI: 1.18-2.28).(5)The polymorphism of rs2956114 locus in the APIP gene was associated with the occurrence of PE(dominant model: OR=2.90,95%CI: 1.85-4.55;additive model: OR=2.26,95%CI: 1.57-3.25).(6)The polymorphism of rs2276360 locus in the NADSYN1 gene was associated with a decreased risk of PE(dominant model: OR=0.51,95%CI: 0.32-0.80;additive model: OR=0.60,95%CI:0.43-0.83).(3)The multifactorial logistic regression analysis of the association between environmental factors and the risk of PE showed that gestational age of 35 years and older(OR=1.94,95%CI: 1.13-3.32),living in rural areas(OR=4.04,95%CI: 1.94-8.43),being overweight or obese before pregnancy(OR=4.14,95%CI: 2.44-7.04),having periodontitis during early pregnancy(OR=2.99,95%CI: 1.24-7.22),using hormone medication in early pregnancy(OR=2.37,95%CI: 1.28-4.39),taking iron supplements in early pregnancy(OR=1.66,95%CI: 1.06-2.61),and having gestational diabetes mellitus(OR=5.15,95%CI: 2.64-10.05)were all independently associated with an increased risk of developing PE in pregnant women.In addition,pregnant women with a bachelor’s degree or above were associated with a lower risk of developing PE(OR=0.29,95%CI: 0.10-0.89)compared with those with junior middle school and below.Furthermore,low pre-pregnancy BMI(OR=0.39,95%CI: 0.18-0.82)and second-hand smoke exposure in peri-conception(OR=0.62,95%CI: 0.40-0.98)were also associated with a decreased risk of PE in pregnant women.(4)Gene-gene interaction revealed a first-order interaction effect of antagonistic type between APIP gene rs2956114 locus and NADSYN1 gene rs2276360 locus(RERI=-0.71,95%CI:-2.81--0.29).Analysis of the generalized multifactor dimensionality reduction model showed that,after controlling for suspected confounding factors,the four-factor,three-order interaction model consisting of the SCN7 A gene rs6738031 locus,the NARS2 gene rs10501429 locus,the APIP gene rs2956114 locus,and the NADSYN1 gene rs2276360 locus was the best gene-gene high-order interaction model with maximum cross-validation consistency(10/10)and relatively high accuracy of training samples and test samples(0.715 and0.633)in this study.Logistic regression analysis showed that SCN7 A gene rs6738031,APIP gene rs2956114,NADSYN1 gene rs2276360,and NARS2 gene rs10501429 had a multiplication interaction on the occurrence of PE(P=0.017).(5)Gene-environment interaction revealed a first-order interaction between SCN7 A gene rs6738031 and gestational diabetes mellitus(RERI=11.79,95%CI: 0.10-33.32),APIP gene rs2956114 and gestational age(RERI=4.56,95%CI: 0.62-12.78),pre-pregnancy overweight or obesity(RERI=9.85,95%CI: 1.89-25.86)and gestational diabetes mellitus(RERI=24.6,95%CI: 5.38-84.16).There was an antagonistic interaction between NADSYN1 gene rs2276360 and iron supplementation in early pregnancy(RERI=-1.18,95%CI:-3.38--0.18).GDMR model analysis showed that the three-factor interaction model consisting of pre-pregnancy BMI level,SCN7 A gene rs6738031,and APIP gene rs2956114 was the best gene-gene interaction model in this study.This model had the highest cross-validation consistency(10/10),and the highest accuracy in test samples(0.710).Logistic regression analysis showed that pre-pregnancy BMI level,SCN7 A gene rs6738031,and APIP gene rs2956114 had a multiplicative interaction with PE(P=0.026).Conclusions:(1)The genetic variants at FN1 gene rs13306364,LAMB2 gene rs35713889,SCN7 A gene rs6738031,APIP gene rs2956114,and NADSYN1 gene rs2276360 loci may be genetic susceptibility variants associated with the onset of PE.(2)The risk alleles of FN1 rs17449032 A,rs6728999 T,and rs13306364 G,as well as LAMB2 rs148818522 T,rs79677861 T,and rs35713889 T,and KDR rs34231037 G and rs35636987 T,all have a significant cumulative effect on the development risk of PE.The risk of PE increases with the number of risk alleles carried by each gene.(3)Gestational age,educational level,residence,pre-pregnancy BMI level,periodontitis in early pregnancy,hormone drugs used in early pregnancy,second-hand smoke exposure in peri-conception pregnancy,iron supplementation in early pregnancy,and gestational diabetes are independently associated with the development of PE.(4)There was antagonistic interaction between APIP gene rs2956114 and NADSYN1 gene rs2276360 on the risk of PE.A high-order gene-gene interaction between SCN7 A gene rs6738031,NARS2 gene rs10501429,APIP gene rs2956114,and NADSYN1 gene rs2276360 may be associated with the development of PE in pregnant women.(5)There were synergistic interactions between SCN7 A gene rs6738031 and gestational diabetes mellitus,APIP gene rs2956114 and gestational age,pre-pregnancy overweight or obesity,and gestational diabetes mellitus on the risk of PE.There was an antagonistic interaction between NADSYN1 gene rs2276360 and iron supplementation in early pregnancy on the risk of PE.A high-order gene-environment interaction between pre-pregnancy BMI level,SCN7 A gene rs6738031 and APIP gene rs2956114 may be associated with the development of PE in pregnant women.