Abstract
Context. A number of studies have reported replicable associations between common genetic loci and obesity indices. One of these loci is the fat mass and obesity associated locus (FTO). We aimed to assess whether breastfeeding mediated the known association between FTO and indices of body fatness. Methods. This study includes three independent pediatric cohorts, two of Greek origin (the Gene-Diet Attica Investigation: GENDAI, n = 1 138 and the "Growth, Exercise and Nutrition Epidemiological Study In preschoolers": the GENESIS study, n = 2 374) and one British (the Avon Longitudinal Study of Parents and Children: ALSPAC, n = 4 325). Among other information, breastfeeding history was recorded. A DNA sample was ascertained by either blood or saliva. Genotyping for FTO variants was performed in GENDAI and ALSPAC for the rs9939609, while in GENESIS, for the rs17817449 variant. Results. In all cohorts, multivariate analysis showed that the association between FTO : rs9939609 and measures of obesity was consistent across newly presented cohorts (GENDAI: Body mass index [BMI], beta = 0.43, p = 0.009; Waist Circumference, beta = 1.067, p = 0.019; triceps skinfold, beta = 0.972, p = 0.003; subscapular skinfold, beta = 0.593, p = 0.023; GENESIS: Waist Circumference, beta = 0.473, p = 0.008 and subscapular skinfold, beta = 0.227, p = 0.014). Inclusion of one month of breastfeeding as an interaction term effectively removed these associations with indices of obesity (BMI, Waist-Hip-Ratio and subscapular skinfold). No evidence of such interaction was observed for the independent cohort of British children. Conclusions. Our findings indicate that in two moderately sized Greek samples, breastfeeding may exert a modifying effect on the relationship between variants at the FTO locus and indices of adiposity. These findings were not replicated in a larger British collection.