Associations Of Exposure To Endocrine-Disrupting Chemicals During Early Pregnancy And Genetic Factors With Physical Development In Early Childhood

Endocrine-disrupting chemicals（EDCs）are ubiquitous in the daily environment and can interfere with the normal function of endocrine system of humans.Due to the wide range of exposure sources and routes,people are generally exposed to a variety of environmental EDCs.As pregnant women and developing fetuses are vulnerable to environmental pollutants,the exposure assessment of typical environmental EDCs during pregnancy is important in protecting maternal and infant health.Phthalates,bisphenols,benzophenones and triclosan（TCS）are environmental EDCs commonly used in daily products such as cosmetics,sunscreens,perfumes,detergents and food packaging materials,to which the pregnant women are frequently exposed.Previous studies have found that exposure to these typical environmental EDCs during pregnancy may affect early childhood physical development,but the results are inconsistent.Furthermore,most studies investigating these chemicals in association with child growth have been restricted to single exposure models,ignoring the potential mixed effects of exposure to multiple EDCs during pregnancy.In addition,studies have found that environmental exposures during pregnancy can interact with genetic factors,leading to adverse effects on child health.However,few studies have investigated the effects of the interaction between exposure to EDCs during pregnancy and genetic predisposition on early childhood physical development.Based on a prospective birth cohort,this study aimed to assess the concentrations of16 common typical environmental EDCs in urine samples collected during early pregnancy,and to further investigate the association between exposure to EDCs individually and in mixture during pregnancy with early childhood physical development.Meanwhile,based on large multi-ethnic GWAS summary statistics,this study constructed a polygenic score（PGS）for birth weight for assessing the genetic predisposition of early childhood physical development.Finally,the interactions between prenatal EDCs exposure and genetic predisposition were examined in relation to early childhood physical development,in order to prove an important reference for identifying health risks related to EDCs exposure during pregnancy.Part 1.Associations between exposure to endocrine-disrupting chemicals during pregnancy and physical development in early childhoodObjectives:To investigate the exposure characteristics and influencing factors of 4types of environmental EDCs in pregnant women and to investigate the associations between exposure to EDCs during pregnancy and early childhood physical development at birth and 2 years of age.Methods:This study was based on a prospective birth cohort established in Wuhan,China,with 2139 mother-infant pairs recruited between 2013 and 2015.Ultra-performance liquid chromatography-triple quadrupole mass spectrometry was used to measure the concentrations of 16 EDCs metabolites（specifically:9 phthalates,3 bisphenols,3benzophenones,and TCS）in urine samples collected during early pregnancy.Correlations between the compounds were analyzed by Spearman’s rank correlation,and factors influencing the urinary EDCs concentrations were analyzed using generalized linear regression models.Birth length and birth weight of newborns were measured,and height and weight of children at 2 years of age were measured at follow-up.Physical developmental indicators were standardized using the WHO Child Growth Standards.Physical developmental outcomes[low birth weight（LBW）,Macrosomia,small for gestational age（SGA）,large for gestational age（LGA）,and overweight/obesity（OW/OB）]were determined according to clinical definitions and WHO criteria.Associations between EDCs exposure and physical developmental indicators were analyzed using generalized linear regression models,and associations between EDCs exposure and physical developmental outcomes were analyzed using unconditional logistic regression models.Weighted quantile sum（WQS）regression models were used to analyze the association of mixed exposure to multiple EDCs with child physical developmental indicators and outcomes,and to estimate the weight of each EDCs contributing to the overall association.Results:（1）Among the 16 EDCs metabolites detected in the urine samples of this study population,the detection rates of all substances were above 75%,except for bisphenol AF（BPAF）and mono-methyl phthalate（MMP）,which were below 50%.Among them,mono-n-butyl phthalate（Mn BP）had the highest geometric mean concentration（59.70μg/L）,followed by mono-iso-butyl phthalate（Mi BP）（17.62μg/L）.The internal exposure levels of dibutyl phthalates（DBP）characterized by the sum molar concentrations of Mn BP and Mi BP were high with a urine specific gravity（SG）corrected geometric mean concentration of 0.39μmol/L.The correlation analysis showed that there were significant positive correlations among phthalates（correlation coefficient:0.30-0.98,P<0.01）.In addition,benzophenones also showed modest positive correlations with triclosan（TCS）,and low molecular weight phthalates[mono-ethyl phthalate（MEP）,Mn BP,and Mi BP]（correlation coefficients:0.20-0.32,P<0.05）.Maternal age,pre-pregnancy BMI,education level and household income level had significantly positive correlation with the concentrations of multiple EDCs in urine.Urinary EDCs concentrations during early pregnancy showed significantly seasonal differences,showing generally higher concentration in summer and lower in winter.（2）Single-pollutant model analysis revealed that the levels of urinary MEP and DBP in early pregnancy were significantly and positively associated with BMIz scores at 2years of age,and each unit increase in natural log-transformed urinary MEP and DBP concentrations was associated with 3.51%（95%CI:0.02%,7.00%）and 7.54%（95%CI:2.96%,12.13%）increase in BMIz scores,respectively;multi-pollutant mixed exposure model analysis found that mixed exposure to multiple EDCs during early pregnancy showed a significantly positive correlation with BMIz score at 2 years of age（%Change=10.02%,95%CI:3.78%,16.26%）,with DBP contributing the greatest weight（49%）,followed by MEP（28%）.In addition,Single-pollutant model analysis found that urinary DBP concentration levels in early pregnancy were significantly associated with increased risk of OW/OB in children at age 2 years（OR=1.24,95%CI:1.09,1.41）;multi-pollutant mixed exposure model analysis found that exposure to EDCs mixture during early pregnancy was significantly positively associated with increased risk of childhood OW/OB（OR=1.31,95%CI:1.08,1.57）,with DBP contributing the greatest weight（46%）,followed by MEP（16%）.Conclusions:The pregnant women in Wuhan are widely exposed to multiple environmental EDCs,with DBP having the highest internal exposure levels in urine.The correlation between multiple EDCs was significantly positive,and there may be a common source of exposure.Concentrations of EDCs in urine samples from pregnant women were influenced by various factors including age,pre-pregnancy BMI,education level,and sampling season.Multiple EDCs exposure in mixture during pregnancy are significantly associated with increased risk of overweight and obesity in early childhood,with DBP contributing the greatest weight.Part 2.The interactions between genetic factors and exposure to endocrine-disrupting chemicals during pregnancy on early childhood physical developmentObjectives:To construct a PGS for birth weight to assess the effects of genetic factors on early childhood physical development and to investigate the interactions between prenatal EDCs exposure and the PGS on early childhood physical development.Methods:（1）Based on the prospective birth cohort,2803 children complete a2-year-old physical examination at the hospital until December 2019 and had available cord blood sample were selected for the study.The Illumina Infinium Asian Screening Array-24 v1.0（ASA）gene chip was used to measure cord blood DNA samples to genotype the study subjects.Based on the summary statistics reported in a large multi-ethnic birth weight GWAS,the PGS for birth weight was constructed in our study population using the PRSice2 and LDpred2 algorithms,respectively,and the best PGS was selected for subsequent analysis by comparing the strength of relationship between the constructed PGS and birth weight under different algorithms and parameter conditions.The association between PGS and physical developmental indicators was analyzed using linear regression models,and the association between PGS and physical developmental outcomes was analyzed using unconditional logistic regression.The dose-response relationship between PGS and physical developmental indicators and outcomes was analyzed using Restricted cubic spline（RCS）models.（2）A total of 940 mother-infant pairs both completed a 2-year-old physical examination at the hospital and had birthweight PGS data in the follow up of study population in Part 1 were selected to investigate the effect of the interaction between prenatal EDCs exposure and PGS on physical development in early childhood.The 940infants were divided into low PGS group（N=314）,intermediate PGS group（N=312）and high PGS group（N=314）according to tertiles of PGS.The association between exposure to EDCs during pregnancy and physical developmental outcomes was analyzed among the total population and each PGS group using unconditional logistic regression,and the association between exposure to EDCs and physical developmental indicators was analyzed using generalized linear models.The interaction effects were estimated by including an interaction term of EDCs and PGS in the models.The population were further divided into high and low exposure groups according to the higher tertiles of each EDCs,and the population was then divided into six EDCs and PGS combination groups.The low EDCs exposure combined intermediate PGS group was used as the reference group to analyze the joint effects of prenatal EDCs exposure and PGS in the association with physical developmental indicators and outcomes in early childhood.Results:（1）PGS constructed by LDpred2 performed better than PRSice2.The best PGS was significantly and positively associated with birth weight.For each standard deviation increase in PGS,birth weight increased by 79.51 g(P=1.67×10^-29)after adjusting for gestational age and sex.Birthweight PGS was significantly correlated with z-scores of physical developmental indicators at birth and 2 years of age（b:0.08-0.17;P<0.01）,and RCS analysis showed a significant linear dose-response relationship between PGS and physical developmental indicators(all P_linear<0.01).For each standard deviation increase in PGS,adjusted odds ratios（ORs）were 1.58（95%CI:1.33,1.89）for macrosomia,1.52（95%CI:1.36,1.70）for LGA,1.09（95%CI:1.00,1.20）for OW/OB,0.72（95%CI:0.53,0.96）for LBW and 0.62（95%CI:0.54,0.70）for SGA,respectively.The results of the RCS analysis showed that PGS was associated with risk of these outcomes in a linear dose-response manner(all P_linear<0.05).Infants with a PGS in the top 10%were at significantly higher risk of macrosomia and LGA（P<0.01）,and infants with a PGS in the bottom 10%were at significantly higher risk of SGA（P<0.01）.（2）Among 940 infants,we categorized population to low,intermediate and high PGS group according to PGS tertiles.The adjusted ORs were 2.49（95%CI:1.52,4.18）for LGA and 1.40（95%CI:0.95,2.07）for OW/OB,respectively,in high PGS group compared with the intermediate PGS group,and the OR was 1.76（95%CI:1.08,2.90）for SGA in low PGS group compared with the intermediate PGS group.Stratified analysis by PGS revealed that there was a significant interaction between prenatal urinary Mi BP and PGS with the risk of LGA(P_interaction<0.05),with a significant risk effect observed only in the high PGS group（OR=1.42,95%CI:1.05,2.08）.The positive association of prenatal DBP exposure with the risk of LGA and OW/OB were only significant in the high PGS group（P<0.05）.Joint effects analysis found that there were significant joint effects of high DBP exposure and high PGS in the association with the risk of LGA and OW/OB,with significantly highest ORs in high DBP and high PGS group compared to low DBP and intermediate PGS group with the risk of LGA（OR=3.51,95%CI:1.64,7.69）and OW/OB（OR=2.41,95%CI:1.33,4.69）.In addition,compared to low exposure and intermediate PGS group,there was a significant joint effect between high butylbenzyl phthalate（BBz P）exposure and high PGS on the risk of OW/OB（P<0.05）,and there were significant joint effects of TCS,BBz P and di-2-ethylhexyl phthalate（DEHP）exposure with high PGS on the risk of LGA（P<0.05）,and there were significant joint effects of 2,4-Dihydroxybenzophenone（BP-1）,Bisphenol A（BPA）and Bisphenol S（BPS）with low PGS on the risk of SGA（P<0.05）.Conclusions:There were significantly positive associations between birthweight PGS and early childhood physical developmental indicators,and high PGS indicated higher genetic risks of macrosomia,LGA and childhood OW/OB,whereas low PGS indicated higher genetic risks of LBW and SGA.DBP exposure during pregnancy was associated with increased risk of LGA and childhood OW/OB,and the association was more pronounced among infants with high PGS.Prenatal exposure to EDCs may interact with genetic factors to affect the physical development in early childhood.