Impact Of Genetic Susceptibility And Interaction With Lifestyle Factors On Childhood Obesity In Beijing, China

[Background]Obesity has been becoming a severe public health problem. With the change of living evironment (e.g. excessive energy intake, low level of physical activity and more sedentary lifestyle), the prevalence of obesity increases greatly. As is known, obesity could cause many related chronic diseases, such as type 2 diabetes, cardiovascular diseases, dyslipoproteinemia, hypertension, obstructive sleep apnea and some cancers. In addition, obesity leads to enormous social and economic burden.The pathogenesis of obesity may have a genetic and environmental basis. Family and twin studies showed that genetic factors accounted for 40-70% in the pathogenesis of obesity. Genome wide association (GWA) studies suggested that rs9939609, in the intron 1 of FTO gene, was significantly associated with BMI and fat mass percentage (FMP) and inceased the risk of obesity in one Caucasian population (including adult and children). Subsequently, the association was replicated and confirmed in other Caucasian populations. However, the results were controversial in East adult Asians. Obesity is a complex disease, so one gene independently could not explain its genetic etiology. The widely accepted viewpoint is that the gene-gene, genes-environment interaction could induce the disease. Based on the results of GWA studies, the further studies on the gene-environment interaction add further valuable insight and contribute to our understanding of the pathogensis of complex diseases such as obesity.[Objective]1. Using the data derived from the Beijing Child and Adolescent Metabolic Syndrome (BCAMS) study, we aim to investigate the association of seven genes' polymorphisms (FTO rs9939609, FAIM2 rs7138803, NPC1 rs 1805081, FTO rs6499640, MC4R rs17782313, BDNF rs6265, GNPDA2 rs10938397) with BMI and obesity;2. To investigate the effect of gene-gene interaction on obesity in Chinese children;3. To investigate the effect of genes-environment interaction on obesity in Chinese children. [Methods]Subjects were recruited from the BCAMS study, a cross-sectional population-based survey carried out in 2004. The survey included a questionnaire, medical examination, anthropometric measurement, and finger capillary blood tests among a representative sample (n= 19593, boys accounted for 50% of the sample) of Beijing school-age children aged 6 to 18 years. Anthropometric measurement included weight, height, waist circumference, and FMP by bioimpedance analysis. Birth weight was collected based on a self-report questionnaire, which was completed by parents or guardians. Within this large group of children,1229 obese,655 overweight and 1619 normal weight children were recruited and diagnosed by the Chinese age-and sex-specific BMI cutoffs. Venipuncture blood samples were collected after a 12-hour overnight fast. Genomic DNA was isolated from peripheral blood white cells using the salt fractionation method. SNPs were genotyped by ABI 5700 real time PCR (FTO rs9939609) and TaqMan Allelic Discrimination Assays with the GeneAmp 7900 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) (FAIM2 rs7138803, NPC1 rs1805081, FTO rs6499640, MC4R rs 17782313, BDNF rs6265, GNPDA2 rs 10938397). Access 2003 was used for data entry and SPSS, version 13.0 (SPSS Inc., Chicago, Illinois) was used for data analysis. The statistical methods included ANCOVA, multiple linear regression, multiple logistic regression and multifactor dimensionality reduction.[Results]1. To correct for multiple comparisons, false discovery rate (FDR) approach was used. After adjusted for sex and age, SNP rs9939609, rs7138803, rs 17782313, rs6265 and rs 10938397 were significantly associated with BMI (P=0.0000076,0.0031, 0.0000010,0.014 and 0.000092). However, there were no statistical significant association between rs 1805081, rs6499640 and BMI (P=0.104 and 0.049). BMI increased by 0.210 to 0.786 kg/m2 for per risk allele increase. The results of multinomial logistic regression analysis showed that six SNPs, except for rs 1805081 were significantly associated with obesity risk in Chinese children. Odds ratios (ORs) for obesity were 1.40 (1.19-1.65,P=0.000055),1.15 (1.02-1.29,P=0.024),1.12 (0.98-1.28, P=0.083),1.19 (1.03-1.37, P=0.020),1.37 (1.20-1.56, P=0.0000017),1.16 (1.04-1.30, P=0.0062),1.24 (1.11-1.40, P=0.00022), respectively.2. The obesity risk increased greatly with the increase of genetic risk scores (GRSs). Compared with children whose GRSs<2, subjects with GRSs equal to 3,4,5,6, 7 and>8 had higher risk for obesity, ORs were 1.44(0.97-2.12),1.63(1.14-2.34),2.16 (1.51-3.10),2.31 (1.60-3.33),2.99 (2.00-4.48) and 3.37 (2.14-5.31), respectively. Overall, the risk for children obesity increased by 21% for per risk allele change.3. The interaction of MC4R and GNPDA2 genes increased the risk for children obesity. After adjusted for sex and age, compared with children carrying MC4R rs 17782313 CC and GNPDA2 rs 10938397 GG genotypes, a significantly increased risk of obesity was observed for children carring the genotypes of rs]7782313 TT and rs10938397 GG (OR=2.16,95%CI:1.25-3.74,P=0.006), genotypes of rs17782313 TT and rs10938397 GA (OR=2.72,95%CI:1.60-4.59, P＜0.001), and genotypes of rs17782313 TT and rs10938397 AA have the highest risk (OR=4.47,95%CI: 1.67-11.90,P=0.003).4. Among all lifestyle factors collected, five common factors including protein, vegetables & fruits, tobacco & alcohol, sedentary behavior and leisure time physical exercise were extracted with factor analysis. Logistic regression analysis showed that the interaction of FTO rs9939609 with protein foods,vegetables & fruits, sedentary behavior and leisure time physical exercise accounted for 19.25%,5.67%,11.66% and 17.56%, respectively. The interaction of MC4R rs17782313 with vegetables & fruits, sedentary behavior and leisure time physical exercise accounted for 2.35%,36.89% and 14.20%, respectively.[Conclusions]1. To our knowledge, this is the first study replicating and confirming the association of five SNP (rs9939609, rs7138803, rs17782313, rs6265and rs10938397) with both BMI and obesity in Chinese children.2. Using the genetic risk scores with the known SNPs, we may identify the high risk individuals who carry more risk alleles, which is beneficial to reduce the occurance obesity in obesity susceptible children through lifestyle modification.3. The interaction of MC4R and GNPDA2 genes, and the interactions of genes and lifestyle factors increase the risk of obesity in Chinese children, which contributes to our understanding of the etiology of obesity.