Exp Clin Endocrinol Diabetes 2008; 116 - N3
DOI: 10.1055/s-0028-1096330

The growth hormone – IGF1 axis as a mediator for association between FTO variants and obesity

D Rosskopf 1, C Schwahn 2, F Neumann 1, A Bornhorst 1, M Mischke 1, S Wolf 1, I Geissler 1, T Kocher 2, HJ Grabe 3, M Nauck 4, J Hebebrand 5, HK Kroemer 1, N Friedrich 6, 7, H Völzke 6, H Wallaschofski 7
  • 1Department of Pharmacology, Center for Pharmacology and Experimental Therapeutics, Ernst-Moritz-Arndt University, Greifswald, Germany
  • 2Community Medicine, Dentistry Clinic, Ernst-Moritz-Arndt University, Greifswald, Germany
  • 3Department of Psychiatry and Psychotherapy, Ernst-Moritz-Arndt University, Stralsund, Germany
  • 4Institute for Clinical Chemistry and Laboratory Medicine, Ernst-Moritz-Arndt University, Greifswald, Germany
  • 5Department of Child and Adolescent Psychiatry, University of Duisburg-Essen, Germany
  • 6Institute for Community Medicine, Ernst-Moritz-Arndt University, Greifswald, Germany
  • 7Department of Gastroenterology, Endocrinology and Nutrition, Ernst-Moritz-Arndt University, Greifswald, Germany

Evidence from numerous epidemiological studies in more than 50,000 individuals indicate that risk alleles of the fat mass- and obesity-associated gene (FTO) are distinctly associated with increased BMI values and an enhanced risk for obesity and type 2 diabetes. The molecular mechanisms that link variants in FTO to increased fat mass remain obscure. We re-analyzed the association of rs9926289, one of the tightly linked FTO risk variants, with BMI in the Study of Health in Pomerania (SHIP), a population-based cross-sectional health survey in the north-eastern part of Germany. SHIP comprised 4310 individuals recruited in each five years age strata from 20 to 79 years of age. FTO genotypes of 4068 individuals were obtained, the distribution of which mirrored that of other studies in Caucasian populations. The entire SHIP population was invited to participate in a five years follow-up survey, termed SHIP1, and 3300 individuals (76.6% of the SHIP cohort) were successfully re-examined. FTO genotypes were available for 3267 subjects of SHIP1, which allowed for a longitudinal analysis how FTO risk alleles affected BMI and weight changes within a five years period. For the entire SHIP population we observed a significant association of the risk allele with obesity as indicated by an odds ratio (OR) of 1.27 (95% confidence intervals (CI) 1.06–1.53; p=0.0096; recessive model, logistic regression). We observed a distinct age dependency for the association of FTO risk alleles with BMI. For individuals younger than 56 years of age mean (n=2340) BMI for homozygous carriers of the risk allele amounted to 30.0±0.2kg/m2 compared to 26.2±0.2 and 26.2±0.1kg/m2 for homozygous carriers of the low-risk allele and heterozygotes, respectively (p=0.018; linear model, adjusted for age and gender). OR for obesity in homozygous carriers of the risk allele amounted to 1.64 (CI 1.27–2.10; recessive model; logistic regression adjusted for age, gender). For the SHIP population older than 56 years of age an association of the FTO risk allele with BMI or obesity was no longer observable. The levels of plasma insulin-like growth factor 1 (IGF1) also follow a strong age dependency. Homozygous carriers of the FTO risk allele exhibited significantly lower plasma IGF1 levels (crude models and after adjustment for age, gender, BMI and IGF1 binding protein 3 levels). We propose that FTO, which is predominantly expressed in the hypothalamus, affects the growth hormone – IGF1 axis which may in turn affect the body composition and fat mass.