Obesity rates are rising rapidly in the United States, reaching epidemic proportions. Insights into which genes predispose individuals to develop obesity are a necessity. If people at risk for obesity can be identified, individualized treatment programs can be designed based on the individuals' genetic and epigenetic predisposition to help decrease the rate of obesity and obesity-related diseases and deaths. This study will be focusing on the genes FTO, MAOA, SH2B1, CCKAR, NEGR1, LEPR, DNMT3B, and BDNF that have been previously associated with obesity risk and obesity-related phenotypes.

Transcript levels of FTO and MAOA were analyzed using quantitative real-time RTPCR, promoter methylation was examined utilizing methylation-sensitive restriction enzyme digestion assays designed for each of the eight gene promoters, and the genotype at eight SNPs, previously associated with obesity, were examined. These data were compared to data gathered on body composition, eating behavior, and temperament. The goals of this project were to replicate results from previous research suggesting associations between certain genetic variants to body composition measures, to identify novel associations between genetic and epigenetic variations and body composition, eating behavior, and temperament, and to provide evidence that the genes previously correlated to obesity in adults is also correlated to measures of obesity and obesity-related phenotypes in children.

Decreased levels of methylation in the promoter of BDNF were associated with different eating behaviors including, decreased food fussiness and decreased satiety response. These results were statistically significant after Bonferroni correction for multiple testing. Genotype analysis at the SNP, rs4923461, in BDNF identified an association between the G allele and increased emotional under-eating in males. This association also remained significant after Bonferroni correction. These data gathered for BDNF may suggest a novel role for BDNF in the regulation of energy balance and obesity.

The data analysis for all expression, methylation, and genotype data identified associations with 16 different obesity-related phenotypes. These phenotypes included; three measures of body composition, seven eating behaviors, two measures of food intake, one measure of self-regulation, and three measures of temperament. These associations were held to a lower statistical standard and are considered suggestive pending replication in a larger sample. This research was able to provide novel insight into genetic and epigenetic alterations that modify obesity-related phenotypes in African American children.

A cumulative genetic and epigenetic "obesity risk factor" score was derived using all significant and suggestive associations to obesity-related phenotypes. The score was derived from the methylation analysis from all eight gene promoters, SNPs from LEPR, DNMT3B, and _BDNF, and expression data for MAOA and FTO. The "obesity-risk factor" score was significantly higher in obese compared to non-obese individuals, suggesting the combined genetic and epigenetic approach has value in the prediction of childhood obesity in African Americans.