The Relationship Between Maternal Metabolic Environment And Neonatal Birth Weight And Underlying Mechanism

BackgroundAdverse metabolic environmental factors that affect fetus in utero will cause abnormal birth weight and series of adult diseases, including type2diabetes, obesity, etal. In the process, it may involve methylation modification of key genes that regulate fetal growth and development. Changes in the methylation level of genes will impact gene expression, and it may increase the risk of abnormal birth weight. However, it’s still unclear that whether changes in the methylation and gene expression levels of key genes had changed or not in cord blood, and its relationship with maternal metabolic condition and neonatal birth weight remains unclear either. Therefore, through the study of methylation and expression levels of the key genes in cord blood, our study aimed to clarify its relationship with maternal metabolic condition and neonatal birth weight, which can provide theoretical basis for the early prevention of abnormal birth weight and metabolic syndrome.Objective1. To determine the relationship between maternal metabolic environment(pre-pregnancy BMI, glucose and lipid metabolism)and neonatal birth weight.2. To investigate the relationship between adipocytokines, insulin-like growth factors levels and maternal metabolic environment as well as neonatal birth weight.3. To study the methylation and expression levels of key genes that regulate fetal growth and development, and its relationship with maternal metabolic environment and neonatal birth weight.Methods1. The study enrolled late-pregnant (28-37gestational weeks) women during January lth2010to June30th2011who attended regular prenatal examinations in Yongkang, Huzhou, Cixi, Ninghai, Shaoxing, Yiwu, Dongyang and Wenling maternal and children health hospitals. We established study cohort according to the inclusion and exclusion criteria. Informed consents were obtained prior to their enrollment and the pregnant women filled in the "dietary nutrition during pregnancy and background information form" and had the blood examination. Doctors filled in the "neonatal birth information form" and extracted umbilical venous blood upon their delivery.3021pairs of mothers and newborns were included, with complete information and accorded with the inclusion and exclusion criteria. Both maternal and cord serum were measured for C-peptide, insulin, Glycosylated hemoglobin (HBA1c), triglyceride (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) levels. Based on the neonatal birth weight and gestational age, the objects were divided into three groups:SGA, AGA and LGA.302pairs of mothers and newborns were selected by stratified random sampling, who received measurement of adipocytokines, insulin-like growth factors and gene expression.49newborns were also selected by stratified random sampling for measurement of DNA methylation of key genes. Maternal metabolic environmental factors in our study were as follows:pre-pregnancy BMI, maternal glucose and lipid metabolism and blood pressure.2. Total adiponectin (Total-ADPN), high-molecular-weight adiponectin (HMW-ADPN) and insulin-like growth factors (IGFs) from302pairs of maternal and cord blood were measured by enzyme-linked immunosorbent assay (ELISA).3. The cord blood from selected302pairs of subjects were used to test the mRNA expression of candidate genes by Real-time PCR. We selected fat mass and obesity associated gene (FTO), leptin (LEP), peroxisome-proliferator-activated receptor y(PPARy) and insulin-like growth factor2(IGF2) as target genes.4. The DNA methylation levels of target genes from49newborns were tested by an bisulfite sequencing PCR(BSP)method. LEP, IGF2, PPARy and pancreatic and duodenal homeobox1(PDX1) were selected as target genes.Results1. Pre-pregnancy BMI and neonatal birth weight were positively correlated. The occurrence of LGA was significantly higher in male than in female. Maternal C-peptide, HbA1c, insulin, TG levels and cord blood C-peptide, insulin levels were SGA<AGA<LGA (p<0.05). Maternal blood HDL-C, TC levels and cord blood TG level were SGA>AGA>LGA (p<0.05). After adjustment for confounding variables, cord blood C-peptide, TG levels, maternal age, pre-pregnancy BMI and neonatal gender were independently related to SGA. Maternal blood HbA1c, HDL-C, LDL-C, TG levels and cord blood insulin level, pre-pregnancy BMI and neonatal gender were independently associated with LGA.2. Maternal blood Total-ADPN level was SGA>AGA>LGA, and cord blood Total-ADPN, IGF-1, Leptin, IGF-2levels were SGA<AGA<LGA (p<0.05). Maternal plasma IGF-1, Leptin, IGF-2levels and cord plasma IGF-1, leptin levels increased with the increased maternal pre-pregnancy BMI. Maternal plasma HMW-ADPN level in women with pre-pregnancy≥25kg/m2group was significantly increased, and cord plasma HMW-ADPN level in women with pre-pregnancy<18kg/m2group was significantly decreased. In pregnant women with abnormal glucose tolerance and diabetes group, maternal plasma HMW-ADPN level decreased and IGF-2level increased significantly. Maternal plasma leptin level increased distinctively in the pregnant women with abnormal glucose tolerance group. After correction for confounding factors, maternal plasma Total-ADPN, Leptin, IGF-2levels and cord plasma Leptin, IGF-2levels were independently asscociated with SGA. Maternal plasma Total-ADPN and cord plasma Leptin, IGF-1, IGF-2levels were independently related with LGA.3. In maternal higher TG group, mRNA level of PPARy gene was distinctively higher than in normal TG group (p<0.05), and mRNA level of LEP gene decreased significantly in maternal high HDL-C group (p<0.05). What’s more, mRNA levels of FTO, IGF2and PPARy genes in male were significantly higher than in female (p<0.05). After correction for confounding factors, mRNA levels of IGF2, LDLR genes were still independently associated with SGA.4. Compared with female, methylation levels of PDX1-DMR1and PPARy-DMR1gene in male were significantly lower (p<0.05).Conclusions1. Maternal pre-pregnancy BMI<18.5kg/m2and maternal age≥35years are independent risk factors for the incidence of SGA; pre-pregnancy BMI≥25kg/m2and male newborns are independent risk factors for the incidence of LGA. Cord serum high C-peptide level is a protective factor for SGA, but cord serum high TG level is an independent risk factor for LGA. Maternal high HbA1c, TG levels and cord serum high insulin level are independent risk factors for the incidence of LGA, and maternal higher HDL-C and lower LDL-C levels are independent protective factors for the incidence of LGA. 2. Compared with control group, maternal plamsa HMW-ADPN level decreases significantly in GDM group, which suggest maternal low HMW-ADPN level may be involved in the insulin resistance of maternal GDM. Maternal plasma HMW-ADPN level decreases distinctively along with the increase in maternal pre-pregnancy BMI, which indicates that pregnant women with high BMI may have suffered insulin resisitance. Maternal high Total-ADPN level and cord plasma low leptin, IGF-2levels are independent risk factors for the incidence of SGA, and maternal high Total-ADPN level can predict the occurrence of SGA to some degree. Cord plasma high IGF-1, IGF-2and leptin levels are independent risk factors for the incidence of LGA, and cord plasma high IGF-1and leptin levels can predict the occurrence of LGA. Maternal high Total-ADPN level and cord plasma low leptin level are independent protective factors for the incidence of LGA.3. Gene expression of PPAR-y increases significantly in maternal high TG group, which implies that the increased mRNA expression of PPAR-ygene might be one mechanism of LGA; mRNA expression of LEP gene decreases significantly in maternal high HDL-C group, indicating that low gene expression of LEP gene may cause decreased risk of LGA. Compared with girls, mRNA expression of FTO, IGF2, PPAR-y genes increases significantly in boys, and it may be a mechanism of boys being more likely to become LGA. Low mRNA expression of LDLR gene is independently associated with SGA, which indicates that decreased LDLR gene expression might be one mechanism of SGA causing lipid metabolism disorder.4. DNA methylation of PPAR-y gene in male is significantly lower than that in female, and it might be one epigenetic regulatory mechanism of boys being more likely to develop LGA.