Obesity is becoming one the leading causes of mortality in the western world. Although
environment is a factor in its development, it is highly heritable and despite a number of genes
that have been found to be associated to the disease its genetics are still poorly understood.
Discovery of genetic pathways that influence obesity risk can provide a better understanding of
the pathophysiology of the disease and identify possible pharmacological targets for its
treatment.
This project was designed to investigate possible genetic associations between five candidate
genes and severe obesity in both adult and child French Caucasians (n=2,822). Tag SNPs were
chosen along with a selection of common SNPs not in the HapMap database and genotyped
using Sequenom iPLEX assays. Putative associations were discovered to obesity in three
genes, SIRT1 (corrected p-values: 0.034, 0.019), APLN (corrected p-value: 0.017) and IL11
(corrected p-value: 0.016), although associations do not withstand genome-wide correction.
SIRT1 and IL11 SNPs were subsequently genotyped within a family cohort for which
transcription data in adipose tissue was available. In this cohort, SIRT1 genotypes were
nominally associated with BMI (corrected p-values: 0.014, 0.019, 0.014) and a significant
difference in expression levels of the gene was observed between lean and obese individuals
(p=1.6x10-35) providing suggestive evidence of a role of this gene in the development of obesity.
Expression and genotypes of another gene, IRS1 were also analysed and although no
significant associations to obesity were found, an association between SNP variation and gene
expression was discovered (corrected p-value: 1.0x10-5). Another aim of this project was to investigate the possibility that DNA methylation influences
obesity risk. Firstly, a method for the measurement of the quantitative trait of DNA methylation
status at individual CpG sites was developed using direct sequencing. Next, methylation in the
leptin gene CpG island was measured in a subset of 184 case-control samples and a nominal
association was discovered between the quantitative measurement of methylation at a single
CpG site and obesity (p=0.013).
In summary, putative associations to obesity have been discovered with genetic variants as well
as transcription levels and CpG methylation. Replication in other populations is required in order
to confirm these associations.