In the POUNDS LOST Trial, a 2-year, randomized weight-loss intervention, we found that dietary protein intake significantly modified the effect of an FTO variant on changes in body composition and fat distribution. Carriers of the risk allele (A allele) of the rs1558902 genotype had a greater loss of weight and regional fat in response to a high-protein diet compared with noncarriers, whereas an opposite genetic effect was observed regarding changes in fat distribution in response to a low-protein diet. Our data indicate that the modification effects of dietary treatment were more evident with prolonged intervention. We did not observe significant modification of dietary fat intake on the genotype effects.
The rs1558902 genotype was reported to show the strongest association with obesity in the European (25
) and other ethnic populations (28
), and it has strong linkage disequilibrium with other obesity-associated FTO
variants such as the rs9939609 genotype. In this study, the MAF of the polymorphism in all participants was similar to those in the HapMap CEU population (0.45). At baseline, no significant difference was observed in anthropometrics and metabolic estimates, body composition, or fat distribution across genotypes. The lack of association with baseline BMI is probably largely due to the fact that the participants were all overweight or obese, so that the groups had relatively smaller variances in BMI than the general population.
Several cross-sectional studies showed that diets might modify the effect of the FTO
variant on obesity, but the data from randomized diet intervention trials are conflicting and limited by small sample size or short term of follow-up (Supplementary Table 3
). Two lifestyle intervention studies with follow-up periods of 9 and 12 months did not find significant influence of FTO
polymorphisms (rs8050136 and rs9939609) on changes in body weight or fat distribution related to diet among 200 overweight and obese individuals treated by diets with reduced fat and increased fiber or reduced fat and sugar (13
). In another 10-week, hypo-energetic diet intervention with either low fat or high fat content, the FTO
variant had an effect on only changes in resting energy expenditure, insulin release, and sensitivity, not on weight loss (15
). Similarly, in the Finnish Diabetes Prevention Study, the FTO
variant did not modify weight change by individualized diet intervention with reduced fat and increased fiber during the 4-year follow-up of 255 individuals with impaired glucose tolerance (16
). In our study, when macronutrient components of diets were not considered, we found no main effects of the FTO
variant on changes in weight and body composition during the intervention.
Grau et al. (15
) reported that dietary fat/carbohydrate content interacted with some genetic variants including the FTO
variant on weight reduction or change in obesity-related phenotypes. In our study, we also found significant gene-diet interactions on changes in body composition and fat distribution. However, our data indicate that it is the dietary protein component, rather than dietary fat, that might drive the observed interactions. In previous studies, high-protein intervention has been found to result in a greater weight loss and abdominal fat mass (30
). Our results suggest that individuals with a certain genetic background may benefit more in weight loss by following a high-protein diet. The mechanism of how protein intake interacts with FTO
genotype is unclear.
Our data indicate that the genetic effects on certain fat compositions or depots may be more evident than the effects on overall adiposity. Functional studies have shown that the loss or overexpression of FTO
in mice led to different changes in fat distribution at different dissected sites (19
mRNA expression was fat depot–specific and was found to differ significantly in subcutaneous fat and in visceral fat (33
). Epidemiological studies also have shown that FTO
variants are significantly associated with distribution of fat depots (13
). Taken together, these data suggest that genetic effects of FTO
on the change of fat mass at various sites may be different, and changes in anthropometrics may not adequately reflect the effects of an FTO
The genetic effect in our study seemed to be more evident at 2 years than that at 6 months. The results were in line with a recent study by Razquin et al. (18
) in which it was found that FTO
risk allele carriers had the highest weight reduction after 3 years of intervention with a Mediterranean diet compared with several short-term diet interventions (less than 1 year) in which no influence of an FTO
variant on weight loss or change in fat distribution was found (13
). Of note, between 6 months and 2 years of intervention in our trial, the participants regained weight. Therefore, it seems that the genetic variant might affect both the reduction and regain of the adiposity measures. These data suggest that the modification effects of diet treatment on an FTO
genotype effect are more likely to be identified in long-term interventions.
Several limitations need to be considered when interpreting our findings. Even though our study is thus far the largest and longest diet intervention weight-loss trial, the relatively small sample size of the subgroups may limit the power to detect very moderate genetic effects or interactions. We did not adjust for multiple testing according to the recommendation by Rothman (38
) and Lai et al. (39
) because outcomes and the repeated measurements at 6 months and 2 years were highly correlated in our study. Overadjustment for multiple comparisons may increase the type II error and reduce power to detect significant differences. In addition, the majority of the total participants were white, and further studies are needed to determine whether our findings are generalizable to other ethnic groups. Even though the randomized clinical trial is thought to be the best model to test gene-environment interactions, we acknowledge that replication in diverse populations is needed to verify our findings.
In summary, we found that dietary protein intake might modify the FTO variant’s effect on changes in body composition and fat distribution. A high-protein diet may be beneficial for weight loss in individuals with the risk allele of an FTO variant. Further studies are warranted to verify our findings and explore the potential mechanisms.