Overweight And Obesity In Early Childhood:Its Linkage With Pregnancy-related Risk Factors And DNA Methylation Mechanism

Background:Childhood obesity is a serious public health issue.The combined effect of genetic and environmental factors may contribute to the occurrence and development of childhood obesity.The nutrition status in the first 1,000 days of life(i.e.,from conception to 2 years of age)is important for individual metabolic“programming”,which may affect weight status later in life.Pre-pregnancy overweight and obesity(OWO)and hypertensive disorders during pregnancy(HDP)are two important factors leading to offspring high weight status through altering intrauterine environment.However,it remains unclear whether these two factors have an interactive effect on offspring high weight status.Besides,breastfeeding has many benefits as an important source of nutrition for children in early childhood.However,it is still widely debated on the preventive effect of breastfeeding on offspring high weight status.The mechanism on the association of maternal factors with childhood obesity in previous epidemiological studies is unclear.Growing evidence suggests that epigenetic modification,such as DNA methylation,may play an important role.Therefore,we aim to explore the linkage of OWO in early childhood:related factors during pregnancy and DNA methylation.Part 1 The association between pre-pregnancy weight status,blood pressure status during pregnancy and offspring weight status at 12 month of ageObjective:To evaluate the association between pre-pregnancy OWO,HDP and offspring body mass index z-score(BMI-z score)in infancy and weight status at 12months of age.Methods:The study participants were recruited in Shenyang,Wuhan and Guangzhou cities in China from 2009-2010(cohort1)and in Zhuhai during2014-2016(cohort2).A total of 3,765 mother and child pairs were combined from the two cohorts for analysis(cohort1:1,969 at birth,1,698 at 3 months,1,574 at 6 months,and 1,605 at 12 months;cohort2:1,796 at birth,1,601 at 3 months,1,574 at 6 months,and 1,665 at 12 months)Demographic information was collected using questionnaires and anthropometric measurements were measured by trained staff or obtained from medical records.Pre-pregnancy OWO was defined as BMI?24kg/m² and HDP as systolic blood pressure and/or diastolic blood pressure?140/90 mm Hg.High BMI status at 12months was defined as BMI?85^th percentile according to the children's growth reference of the World Health Organization.The generalized estimating equation model was used to evaluate the association between pre-pregnancy weight status and blood pressure status during pregnancy and BMI-z score at birth,3 months,6 months,and 12 months.Excel sheet established by Andersson et al.was used to calculate the additive interactive effect(i.e.Relative excess risk of interaction,RERI and attributable proportion,AP)of pre-pregnancy OWO and HDP on offspring high BMI status in infancy.Logistic regression model was used to analyze the multiplicative interaction.RERI>0 or AP>0 was considered as a significant difference.Results:Compared to pre-pregnancy normal weight and normal gestational blood pressure,pre-pregnancy OWO and HDP could increase the offspring BMI-z score at 12 months by 0.62 and jointly increase the risk of offspring high BMI status at 12 months after adjusting for potential confounders(odds ratio,OR 3.10,95%CI1.59,6.04),with 51%of the risk attributed to additive interaction.Conclusions:Pre-pregnancy OWO and HDP can interactively increase the risk of high BMI status in infancy.Regularly monitoring pre-pregnancy weight status and blood pressure status during pregnancy is of great significance for the prevention of offspring high weight status.Part 2 The association between pre-pregnancy overweight/obesity,breastfeeding duration and overweight/obesity in preschoolersObjective: To examine the association between pre-pregnancy OWO and OWO in preschoolers and the association stratified by breastfeeding duration(<6 months and ?6 months).Methods: A total of 1,899 preschoolers were included in this study,recruited from the cross-sectional study conducted in Zhuhai city in China from 2017 to 2018.Demographic information was collected using questionnaires and anthropometric measurements were measured by trained staff.Pre-pregnancy OWO was defined as BMI?24kg/m².OWO in preschoolers was defined as BMI ? age-and sex-specific BMI24 cut-off based on the reference of Chinese children aged 2-18 years.Breastfeeding(including exclusive and mixed breastfeeding)duration was classified as <6 and ?6 months.The restricted cubic spline model was used to assess the dose-response relationship between breastfeeding duration and OWO at preschool age.Logistic regression model was used to analyze the effect of pre-pregnancy OWO and breastfeeding duration on OWO at preschool age.Results: The prevalence of OWO among preschoolers with a mean age of 3.1 years was 12.2%(14.5% for boys and 9.7 % for girls).Pre-pregnancy OWO could increase the risk of OWO at preschool age(OR 2.36,95% CI 1.41,3.94).The restricted cubic spline model showed that there was a non-linear dose-response relationship between breastfeeding duration and OWO at preschool age(P<0.001).However,the risk of OWO for preschoolers with mothers having pre-pregnancy OWO was not increased when they were breastfed for more than 6 months(OR 1.70,95% CI 0.79,3.63).Conclusions: Pre-pregnancy OWO might increase the risk of offspring OWO at preschool age.It is critical to guarantee breastfeeding duration for 6 months to reduce the risk of OWO in early childhood and partly offset the increased risk of OWO in early childhood caused by pre-pregnancy OWO.Part 3 The cord blood DNA methylation mechanism of overweight/obesity in preschoolersObjective: To test the association between and genome-wide DNA methylation of maternal blood/cord blood and pre-pregnancy BMI and BMI-z score at 3 years of age;To explore the genome-wide differentially methylated regions(DMRs)and differentially methylated probes(DMPs)in cord blood between OWO and normal-weight preschoolers and further validated the selected Cp G sites.Methods: The maternal and cord blood sample were obtained from the cohort2 2014-2016 and followed-up until 3 years of age.A total of 30 maternal blood samples in early pregnancy and 30 cord blood samples collected from 2014 to 2016 with data of pre-pregnancy weight and height,weight and height of preschoolers,were tested using Illumina Infinium Methylation EPIC Bead Chip(850k)to assess the association between genome-wide DNA methylation and pre-pregnancy BMI and BMI-z score at preschool age.The cord blood of OWO preschoolers(n=4)and age-and sex-matched controls(n=12)were used to identify the genome-wide DMPs.Related Cytosine polyguanine(CpG)sites and CpG sites in nearby susceptible area(100-700bp)were selected for validation based on the genome-wide analysis of DNA methylation in cord blood with BMI-z score(N=30)and OWO(4 cases vs.12 control)at preschool age.The cord blood of OWO preschoolers(n=10)and age-and sex-matched controls(n=10)were used to validate the selected Cp G sites.Demographic information was collected using questionnaires and anthropometric measurements were measured by trained staff.OWO in preschoolers was defined as BMI ? age-and sex-specific BMI24 cut-off based on the reference of Chinese children aged 2-18 years.The linear regression model in CHAMP Package was used to assess the association between genome-wide DNA methylation of maternal blood/cord blood and pre-pregnancy BMI or BMI-z score at preschool age.Bumphunter algorithm in CHAMP Package was used to analyze genome-wide DMRs in cord blood between OWO and normal-weight preschoolers.A linear model for categorical variables in CHAMP Package was used to compare the genome-wide DMPs in cord blood between OWO and normal-weight preschoolers.Gaussian model of generalized linear regression was used to validate the selected Cp G sites between OWO and normal-weight preschoolers and analyze the association between validated Cp G sites in cord blood and pre-pregnancy BMI or BMI-z score at preschool age.Results: In the genome-wide analysis of DNA methylation,no significant CpG sites in maternal blood/cord blood(30 mother and child pairs)were associated with pre-pregnancy BMI or BMI-z score at preschool age after FDR correction.However,the CpG site with small P values(near 1×10^-6)might be associated with pre-pregnancy BMI or BMI-z score at preschool age.In the genome-wide analysis of DNA methylation in cord blood between 4 OWO and 12 normal-weight preschoolers,a total of 59 DMRs and 371 DMPs were found.The DMRs were mainly concentrated on metabolic and adipocytokines signaling pathways,glucagon signaling pathway,and HIF-1 signaling pathway and DMPs concentrated on regulation of actin cytoskeleton and cell adhesion molecules pathways.CpG sites in FOXN3(cg23501836),ZNF264(cg27437574)and SKIV2L(cg13015485)(P values ranked 1^st,2^nd,and 4^th,respectively)based on the genome-wide analysis of DNA methylation in 30 cord blood with BMI-z score at preschool age,DMPs in AHRR(cg26487191),ARNT2(cg16622471),SORCS2(cg26686068),and SMAD3(cg17726760)in cord blood between 4 OWO and 12 normal-weight preschoolers,and Cp G sites in the nearby susceptible region were selected to validate.There was a significant difference of the cord-blood Cp G sites in AHRR(chr5: 355067-355068)and FOXN3(chr14: 89630264-89630272 and chr14: 89630387-89630388)between 10 OWO and 10 normal-weight preschoolers(P<0.05).In addition,Cp G sites in FOXN3(chr14: 89630264-89630272 and chr14: 89630387-89630388)were associated with pre-pregnancy BMI(The P-value of Cp G site in the FOXN3 gene ranked 62 in the genome-wide analysis of DNA methylation in cord blood with pre-pregnancy BMI).Cp G sites of chr14: 89630264-89630272 and chr14: 89630295-89630296 in FOXN3 and chr19: 57703104-57703107 and chr19: 57703301-57703307 in ZNF264 were associated with BMI-z score at preschool age.Conclusions: The epigenome-wide association study and validated study indicated that cord blood DNA methylation in FOXN3 and ZNF264 might be associated with BMI-z score in early childhood and DNA methylation in FOXN3 and AHRR was associated with OWO in early childhood.In addition,DNA methylation in FOXN3 might partly mediate the association between pre-pregnancy BMI and BMI-z score or OWO in early childhood,suggesting an important role of DNA methylation for the development of obesity in early childhood.