Concordance of two multiple analytical approaches demonstrate that interaction between BMI and ADIPOQ haplotypes is a determinant of LDL cholesterol in a general French population

Genetic and environmental factors are involved in insulin resistance (IR). IR and dyslipidemia associate with increased risk of cardiovascular diseases. Plasma low-density lipoprotein cholesterol (LDL-C) level is a marker of cardiovascular risk. In a Caucasian general population we aimed at determining the multifactorial components of LDL-C levels using 10 genes and 3 phenotypes. In the PPARG, UCP3, ADIPOQ, TNF, LIPC, CARTPT, PCSK9, SCAP, SCARB1 and ENPP1 genes known to be associated with IR or dyslipidemia we genotyped 19 single nucleotide polymorphisms (SNPs) in 846 subjects. When several SNPs were genotyped for a given gene we constructed haplotypes. Including genetic and environmental variables (gender, body mass index (BMI) and adiponectin level) we used (1) the multifactor dimensionality reduction method to explain clusters of high and low LDL-C, and (2) the restricted partition method to explain LDL-C levels. Both methods showed that BMI and haplotypes at the ADIPOQ adiponectin encoding gene but not adiponectin level itself, were discriminant regarding to LDL-C. Subjects bearing an at-risk combination of BMI and ADIPOQ genotypes were prone to have a higher LDL-C (OR=3.13, 95% CI=2.20-4.46, P<0.0001). Our results suggest that in interaction with BMI, ADIPOQ haplotypes capture genetic variation(s) from neighboring gene(s) that would modulate LDL-C level.