Associations between BMI and the FTO gene are age dependent: results from the GINI and LISA birth cohort studies up to age 6 years.
ABSTRACT The association between polymorphisms in intron 1 of the fat mass and obesity associated gene (FTO) and obesity-related traits is one of the most robust associations reported for complex traits and is established both in adults and children. However, little is known about the longitudinal dynamics of these polymorphisms on body mass index (BMI), overweight, and obesity.
This study is based on the 2,732 full-term neonates of the German GINI-plus and LISA-plus birth cohorts, for whom genotyping data on the FTO variants rs1558902 (T>A) or rs9935401 (G>A) were available. Children were followed from birth up to age 6 years. Up to 9 anthropometric measurements of BMI were obtained. Fractional-Polynomial-Generalized-Estimation-Equation modeling was used to assess developmental trends and their potential dependence on genotype status.
We observed no evidence for BMI differences between genotypes of both variants for the first 3 years of life. However, from age 3 years onwards, we noted a higher BMI for the homozygous minor alleles carriers in comparison to the other two genotype groups. However, evidence for statistical significance was reached from the age of 4 years onwards.
This is one of the first studies investigating in detail the development of BMI depending on FTO genotype between birth and the age of 6 years in a birth cohort not selected for the phenotype studied. We observed that the association between BMI and FTO genotype evolves gradually and becomes descriptively detectable from the age of 3 years onwards.
- SourceAvailable from: Kristine Færch[show abstract] [hide abstract]
ABSTRACT: Several obesity risk alleles affecting adult adiposity have been identified by the recent wave of genome wide association studies. We aimed to examine the potential effect of these variants on fetal body composition by investigating the variants in relation to birth weight and ponderal index of the newborn. Midwife records from the Danish State Archives provided information on mother's age, parity, as well as birth weight, birth length and prematurity of the newborn in 4,744 individuals of the population-based Inter99 study. Twenty-four risk alleles showing genome-wide associations with adult BMI and/or waist circumference were genotyped. None of the 24 risk variants tested showed an association with birth weight or ponderal index after correction for multiple testing. Birth weight was divided into three categories low (≤10(th) percentile), normal (10(th)-90(th) percentile) and high birth weight (≥90th percentile) to allow for non-linear associations. There was no difference in the number of risk alleles between the groups (p = 0.57). No interactions between each risk allele and birth weight in the prediction of adult BMI were observed. An obesity risk score was created by summing up risk alleles. The risk score did not associate with fetal body composition. Moreover there was no interaction between the risk score and birth weight/ponderal index in the prediction of adult BMI. 24 common variants associated with adult adiposity did not affect or interact with birth weight among Danes suggesting that the effects of these variants predominantly arise in the post-natal life.PLoS ONE 01/2010; 5(12):e14190. · 3.73 Impact Factor