Roswall, Nina3; Ängquist, Lars3; Ahluwalia, Tarun Veer Singh4; Romaguera, Dora3; Larsen, Sofus C3; Østergaard, Jane N3; Halkjaer, Jytte3; Vimaleswaran, Karani S3; Wareham, Nicolas J3; Bendinelli, Benedetta3; Palli, Domenico3; Boer, Jolanda M A3; van der A, Daphne L3; Boeing, Heiner3; Loos, Ruth J F3; Sørensen, Thorkild I A4; Tjønneland, Anne3
1 Section for Metabolic Genetics, The Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, Københavns Universitet2 Department of Public Health, Department of Public Health, Faculty of Health and Medical Sciences, Københavns Universitet3 unknown4 Section for Metabolic Genetics, The Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, Københavns Universitet
influence of <i>FTO</i> and <i>TCF<sub>7</sub>L2</i> loci
BACKGROUND: Several studies have shown that adherence to the Mediterranean Diet measured by using the Mediterranean diet score (MDS) is associated with lower obesity risk. The newly proposed Nordic Diet could hold similar beneficial effects. Because of the increasing focus on the interaction between diet and genetic predisposition to adiposity, studies should consider both diet and genetics. OBJECTIVE: We investigated whether FTO rs9939609 and TCF7L2 rs7903146 modified the association between the MDS and Nordic diet score (NDS) and changes in weight (Δweight), waist circumference (ΔWC), and waist circumference adjusted for body mass index (BMI) (ΔWCBMI). DESIGN: We conducted a case-cohort study with a median follow-up of 6.8 y that included 11,048 participants from 5 European countries; 5552 of these subjects were cases defined as individuals with the greatest degree of unexplained weight gain during follow-up. A randomly selected subcohort included 6548 participants, including 5496 noncases. Cases and noncases were compared in analyses by using logistic regression. Continuous traits (ie, Δweight, ΔWC, and ΔWCBMI) were analyzed by using linear regression models in the random subcohort. Interactions were tested by including interaction terms in models. RESULTS: A higher MDS was significantly inversely associated with case status (OR: 0.98; 95% CI: 0.96, 1.00), ΔWC (β = -0.010 cm/y; 95% CI: -0.020, -0.001 cm/y), and ΔWCBMI (β = -0.008; 95% CI:-0.015, -0.001) per 1-point increment but not Δweight (P = 0.53). The NDS was not significantly associated with any outcome. There was a borderline significant interaction between the MDS and TCF7L2 rs7903146 on weight gain (P = 0.05), which suggested a beneficial effect of the MDS only in subjects who carried 1 or 2 risk alleles. FTO did not modify observed associations. CONCLUSIONS: A high MDS is associated with a lower ΔWC and ΔWCBMI, regardless of FTO and TCF7L2 risk alleles. For Δweight, findings were less clear, but the effect may depend on the TCF7L2 rs7903146 variant. The NDS was not associated with anthropometric changes during follow-up.
American Journal of Clinical Nutrition, 2014, Vol 100, Issue 4, p. 1188-1197
Adiposity; Adult; Alleles; Body Mass Index; Diet; Diet, Mediterranean; Europe; Female; Follow-Up Studies; Genetic Loci; Genetic Predisposition to Disease; Humans; Linear Models; Linkage Disequilibrium; Male; Middle Aged; Obesity; Polymorphism, Single Nucleotide; Prospective Studies; Proteins; Transcription Factor 7-Like 2 Protein; Waist Circumference; Weight Gain