Carolina Ortega-Azorín 1 , 2 , Jose V Sorlí 1 , 2 , Eva M Asensio 1 , 2 , Oscar Coltell 2 , 3 , 4 , Miguel Ángel Martínez-González 5 , Jordi Salas-Salvadó 2 , 6 , Maria-Isabel Covas 2 , 7 , Fernando Arós 8 , José Lapetra 2 , 9 , Lluís Serra-Majem 10 , Enrique Gómez-Gracia 11 , Miquel Fiol 2 , 12 , Guillermo Sáez-Tormo 13 , Xavier Pintó 14 , Miguel Angel Muñoz 15 , Emilio Ros 2 , 16 , Jose M Ordovás 4 , 17 , 18 , Ramon Estruch 2 , 19 , Dolores Corella , 1 , 2 , 4 , 20
6 November 2012
Although the Fat Mass and Obesity ( FTO) and Melanocortin-4 Receptor ( MC4R) genes have been consistently associated with obesity risk, the association between the obesity-risk alleles with type 2 diabetes is still controversial. In some recent meta-analyses in which significant results have been reported, the associations disappeared after adjustment for body mass index (BMI). However gene-diet interactions with dietary patterns have not been investigated. Our main aim was to analyze whether these associations are modulated by the level of adherence to the Mediterranean Diet (MedDiet).
Case-control study in 7,052 high cardiovascular risk subjects (3,430 type 2 diabetes cases and 3,622 non-diabetic subjects) with no differences in BMI. Diet was assessed by validated questionnaires. FTO-rs9939609 and MC4R-rs17782313 were determined. An aggregate genetic score was calculated to test additive effects. Gene-diet interactions were analyzed.
Neither of the polymorphisms was associated with type 2 diabetes in the whole population. However, we found consistent gene-diet interactions with adherence to the MedDiet both for the FTO-rs9939609 (P-interaction=0.039), the MC4R-rs17782313 (P-interaction=0.009) and for their aggregate score (P-interaction=0.006). When adherence to the MedDiet was low, carriers of the variant alleles had higher type 2 diabetes risk (OR=1.21, 95%CI: 1.03-1.40; P=0.019 for FTO-rs9939609 and OR=1.17, 95%CI:1.01-1.36; P=0.035 for MC4R-rs17782313) than wild-type subjects. However, when adherence to the MedDiet was high, these associations disappeared (OR=0.97, 95%CI: 0.85-1.16; P=0.673 for FTO-rs9939609 and OR=0.89, 95%CI:0.78-1.02; P=0.097 for MC4R-rs17782313). These gene-diet interactions remained significant even after adjustment for BMI. As MedDiet is rich in folate, we also specifically examined folate intake and detected statistically significant interaction effects on fasting plasma glucose concentrations in non-diabetic subjects. However these findings should be interpreted with caution because folate intake may simply reflect a healthy dietary pattern.