Role of genetic variants in ADIPOQ in human eating behavior |
| |
Authors: | Kerstin Rohde Maria Keller Annette Horstmann Xuanshi Liu Fabian Eichelmann Michael Stumvoll Arno Villringer Peter Kovacs Anke T?njes Yvonne B?ttcher |
| |
Affiliation: | .IFB Adiposity Diseases, University of Leipzig, Liebigstraße 21, 04103 Leipzig, Germany ;.Department of Neurology, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany ;.Bioinformatics Group, Department of Computer Science, University of Leipzig, Leipzig, Germany ;.Department of Medicine, University of Leipzig, Leipzig, Germany ;.Clinic of Cognitive Neurology, University of Leipzig, Leipzig, Germany |
| |
Abstract: | The beneficial effects of adiponectin and its negative correlation with BMI are well described. Adiponectin serum levels are altered in eating disorders such as anorexia nervosa, bulimia nervosa or binge eating. Here, we tested the hypothesis that (1) adiponectin serum levels correlate with human eating behavior factors and (2) that genetic variants of the ADIPOQ locus influence both serum levels and eating behavior. We analyzed 11 SNPs within ADIPOQ and in the 5′ UTR and measured serum adiponectin levels in 1,036 individuals from the German Sorbs population. The German version of the three-factor eating questionnaire (FEV) was completed by 548 Sorbs. For replication purposes, we included an independent replication cohort from Germany (N = 350). In the Sorbs, we observed positive correlations of restraint with adiponectin serum levels (P = 0.001; r = 0.148) which, however, did not withstand adjustment for covariates (P = 0.083; r = 0.077). In addition, four SNPs were nominally associated with serum adiponectin levels (all P < 0.05). Of these, two variants (rs3774261; rs1501229, all P < 0.05) were also related to disinhibition. Furthermore, three variants were exclusively associated with hunger (rs2036373, P = 0.049) and disinhibition (rs822396; rs864265, all P < 0.05). However, none of these associations withstood Bonferroni corrections for multiple testing (all P > 9.3 × 10−4). In our replication cohort, we observed similar effect directions at rs1501229 for disinhibition and hunger. A meta-analysis resulted in nominal statistical significance P = 0.036 (Z score 2.086) and P = 0.017 (Z score 2.366), respectively. Given the observed relationship of the SNPs with adiponectin levels and eating behavior, our data support a potential role of adiponectin in human eating behavior. Whether the relationship with eating behavior is mediated by the effects of circulating adiponectin warrants further investigations. |
| |
Keywords: | Adiponectin serum levels Genetics Eating behavior Human studies |
|
|