Obesity Associated Genetic Variation in FTO
Is Associated with Diminished Satiety
Jane Wardle, Susan Carnell, Claire M. A. Haworth, I. Sadaf Farooqi, Stephen O’Rahilly,
and Robert Plomin
Health Behaviour Research Centre (J.W., S.C.), Department of Epidemiology and Public Health, University College London, London WC1E
6BT, United Kingdom; Social, Genetic, and Developmental Psychiatry Centre (C.M.A.H., R.P.), Institute of Psychiatry, King’s College
London, London SE5 8AF, United Kingdom; and University of Cambridge Metabolic Research Laboratories (I.S.F., S.O.), Institute of
Metabolic Science, Addenbrooke’s Hospital, Cambridge CB2 2QQ, United Kingdom
multiple populations. However, to date, it is not known whether the association between genetic
variation in FTO and obesity is mediated through effects on energy intake or energy expenditure.
obesity risk and measures of habitual appetitive behavior.
Methods: The intronic FTO single nucleotide polymorphism (rs9939609) was genotyped in 3337
United Kingdom children in whom measures of habitual appetitive behavior had been assessed
using two scales (Satiety Responsiveness and Enjoyment of Food) from the Child Eating Behaviour
Associations of FTO genotype with indices of adiposity and appetite were assessed by ANOVA.
Results: As expected, the A allele was associated with increased adiposity in this cohort and in an
independent case-control replication study of United Kingdom children of similar age. AA ho-
analysis indicated that the association of the AA genotype with increased adiposity was explained
in part through effects on Satiety Responsiveness.
Conclusions: We have used a unique dataset to examine the relationship between a validated
measure of children’s habitual appetitive behavior and FTO obesity risk genotype and conclude
that the commonest known risk allele for obesity is likely to exert at least some of its effects by
influencing appetite. (J Clin Endocrinol Metab 93: 3640–3643, 2008)
forms of this condition predominantly influence the intake or
expenditure of energy. Common intronic variants in the FTO
gene have recently been strongly and replicably associated with
higher body weight and adiposity (1–5), with individuals who
are homozygous for the high-risk allele (AA) of rs9939609
weighing on average 3 kg more than those with two low-risk
alleles (1). Fto mRNA is highly expressed in the hypothalamus
(6), an area that is known to be involved in the regulation of
appetite, and fto expression in the arcuate nucleus of the hypo-
uman obesity is a highly heritable trait, but it is not known
whether the specific genetic variants underlying common
thalamus in rodents is modulated by acute food deprivation (7).
morphisms and weight may be due to differences in appetitive
responses. This would be consistent with findings in monogenic
obesity disorders, which are almost all characterized by an in-
creased desire to eat (8).
We tested the hypothesis that children carrying the higher-risk
FTO alleles have altered appetite in a sample of 3337 unrelated
children aged 8–11 yr recruited as part of the Twins’ Early Devel-
previously demonstrated that adiposity is highly heritable in this
Printed in U.S.A.
Copyright © 2008 by The Endocrine Society
doi: 10.1210/jc.2008-0472 Received February 29, 2008. Accepted June 12, 2008.
First Published Online June 26, 2008
Abbreviations: BMI, Body mass index; CEBQ, Child Eating Behavior Questionnaire; CI,
TEDS, Twins’ Early Development Study.
O R I G I N A LA R T I C L E
E n d o c r i n eR e s e a r c h — B r i e fR e p o r t
jcem.endojournals.orgJ Clin Endocrinol Metab. September 2008, 93(9):3640–3643
cohort (9). To assess appetite in this large sample, we used a vali-
dated, parent-completed, psychometric measure (10–12).
Subjects and Methods
The main study population was recruited from TEDS, a population-
based twin cohort whose anthropometric characteristics have been re-
ported previously (9). Children’s height, weight, and waist circumference
measurements taken by researchers in a subsample (9). Adiposity was in-
dexed with body mass index (BMI) SD scores, and central adiposity with
rs9939609 was genotyped using a TaqMan assay that incorporates
minor groove binding probe technology for allelic discrimination. The
call rate was 98%, and the single nucleotide polymorphism was in Har-
dy-Weinberg equilibrium (P ? 0.729).
(CEBQ); a parent-completed, psychometric instrument that has been
over time (10–12). We used two scales that assess underlying appetitive
drivers of food intake, namely Satiety Responsiveness, a measure of the
ease with which satiety is achieved (e.g. my child cannot eat a meal if
he/she has had a snack just before), and Enjoyment of Food, a measure
of the extent to which presentation of palatable foods provokes eating
(e.g. my child loves food). Scores on these scales have been shown to be
correlated with adiposity (14).
pathways, we assessed the mediating effect of the two appetitive phe-
notypes on the association between genotype and adiposity using the
Sobel test (15, 16). We also carried out analysis of covariance including
BMI SD scores to test whether FTO was associated with appetite inde-
pendently of adiposity.
rs9939609 was also genotyped in a second United Kingdom Cauca-
sian cohort, the Severe Childhood Onset Obesity Project United King-
dom (SCOOP-UK) which comprises 1000 United Kingdom Caucasian
subjects with severe early-onset obesity of unknown etiology (536 fe-
Data were compared with published data from normal-weight United
Kingdom children of the same age in the ALSPAC study (cohort char-
acteristics summarized in Ref. 1).
The rs9939609 genotype distribution in the TEDS sample
(n ? 3337) was similar to that reported in other population-
based samples (AA ? 14.6%; AT ? 49.2%; TT ? 36.2%) (1).
As expected, we replicated the direction and magnitude of the
known association between FTO and adiposity, with each
additional copy of the A allele being associated with an in-
crease of between 0.13 and 0.18 BMI SD scores (weight dif-
ferences from 0.7–1.4 kg). We demonstrated similar effects
for waist circumference, with increases of between 0.60 and
children with the TT genotype, the odds of meeting the In-
ternational Obesity Task Force criterion for pediatric over-
weight/obesity increased from 1.39 [95% confidence interval
(CI), 1.11–1.75] for AT to 1.94 (95% CI, 1.45–2.59) for AA.
This effect was replicated in an independent case-control
study of 926 obese United Kingdom children (SCOOP-UK)
compared with 4022 normal-weight controls from the AL-
SPAC cohort (Table 2).
in relation to FTO genotype (Table 1 and Fig. 1). There was no
TABLE 1. Association between FTO genotype, anthropometric, and appetitive measures in United Kingdom children from the
Group difference and ES
F(1,3354)? 2.8; P ? 0.059;
ES ? 0.002
F(2,3256)? 16.77; P ? 0.001;
ES ? 0.010
F(2,3402)? 7.65; P ? 0.001;
ES ? 0.004
F(2,3299)? 14.68; P ? 0.001;
ES ? 0.009
?2? 20.77; P ? 0.001
P ? 0.008; ES ? 0.003
F(2,3450)? 1.85; P ? 0.157;
ES ? 0.001
TT (n ? 1209)
AT (n ? 1641)
AA (n ? 487)
BMI SD score
?0.123 (?0.194 to ?0.502)0.073 (0.012–0.134) 0.242 (0.131–0.354)
Waist61.99 (61.54–62.44)62.59 (62.19–62.98) 63.55 (62.83–64.26)
Waist SD score0.702 (0.645–0.760) 0.802 (0.752–0.851)0.997 (0.906–1.087)
CEBQ Satiety Responsiveness
score (range 1–5)
CEBQ Enjoyment of Food
2.666 (2.628–2.704)2.654 (2.621–2.687)2.551 (2.499–2.618)
4.109 (4.068–4.149) 4.113 (4.078–4.148)4.179 (4.115–4.243)
Means and 95% CI for anthropometric and appetitive measures with significance values and effect sizes (ES) for differences are shown.
TABLE 2. FTO genotype in obese children from the SCOOP-UK
cohort and normal weight control children from the ALSPAC
A total of 1000 children from the SCOOP-UK cohort were genotyped, and 926
successful genotypes were obtained. Genotype results did not deviate
significantly from Hardy-Weinberg equilibrium. We studied the association of
FTO gene variation with the risk of being obese. We found that children
homozygous for the A allele at rs9939609 were at substantially increased risk of
being obese compared with those homozygous for the low-risk T allele (P value
relates to the per A allele odds ratio).
J Clin Endocrinol Metab, September 2008, 93(9):3640–3643jcem.endojournals.org
sponsiveness was significantly lower in homozygotes for the A
allele (2.55 vs. 2.65 in AT heterozygotes and 2.67 in TT ho-
mozygotes; P ? 0.008). AA homozygotes also had the highest
scores for Enjoyment of Food, but this did not reach statistical
age, family socioeconomic status, and BMI SD score, the associ-
ation between FTO and satiety responsiveness remained signif-
icant [F(2,3105)? 3.17; P ? 0.04; effect size ? 0.002], but as
expected, the effect size was reduced due to controlling for the
known association of appetite and BMI.
on BMI was significantly partially mediated by Satiety Respon-
be attenuated by error of measurement in the measure.
genotype of FTO rs9939609 in two cohorts of United Kingdom
(1–3). Possession of one copy of the A allele is sufficient to in-
crease body weight by 1.5 kg in adults (1), and we were able to
equally strong effect for waist circumference. Together these
studies provide robust support for the assertion that FTO rep-
In the present study, we were uniquely able to examine the
relationship between FTO genotype and measures of appet-
itive behavior in 3337 children recruited as part of the TEDS
cohort. We showed that the obesity-linked FTO intronic sin-
gle-nucleotide polymorphism rs9939609 was associated with
impaired satiety responsiveness. Mediation analysis indicated
that a proportion of the observed association between FTO
genotype and BMI could be explained by effects on satiety
responsiveness. The association between FTO genotype and
satiety responsiveness remained significant after controlling
for BMI SD score, consistent with the idea that FTO may have
a direct effect on appetite, which in turn influences adiposity.
However, given the cross-sectional nature of the phenotypic
data, these analyses can only be indicative of the causal
cific aspects of appetite to obesity. Schachter (17) first demon-
strated that obese adults overeat compared with normal-weight
controls under conditions of satiety but show no differences in
less sensitive to satiety cues. The same effect has been observed
notypes, including aspects of appetite, have been shown to be
heritable (19), and we have shown that Satiety Responsiveness
and Enjoyment of Food are highly heritable (20). The results are
also consistent with evidence that fto expression in the arcuate
nucleus of the hypothalamus in rodents is modulated by acute
food deprivation (7).
Despite the increased risk of obesity associated with het-
of heterozygosity on appetite in this study. This may suggest
that other obesogenic effects of the FTO A allele are operating
in the heterozygotes or, more likely in our view, that the psy-
chometric tool we used is insufficiently sensitive to detect the
very small effects on cumulative energy intake that would be
needed to result in the small increase in adiposity conveyed by
heterozygosity for the risk allele.
These results need to be replicated in samples of other ages
to determine whether the effect is also found in adults. FTO
might of course influence other facets of energy balance, al-
though recent results from the Quebec Family Study found no
differences in resting energy expenditure between genotypes
at the FTO locus (21).
The finding that individuals with two FTO A alleles have
lower responsiveness to satiety cues, and that there is a sig-
nificant indirect path between FTO genotype and BMI
through satiety responsiveness, supports the hypothesis that
the FTO association with BMI involves effects on appetite.
Inter-individual differences in susceptibility to obesity may be
determined in part by genetic variants impacting on satiety
responsiveness that in turn influence the likelihood of over-
eating in environments where large portion sizes and multiple
eating opportunities are the norm.
Enjoyment of food
FIG. 1. FTO association results: SD scores for Satiety Responsiveness and
Enjoyment of Food at age 8–11 yr. Error bars represent ?1 SE.
Wardle et al.FTO and Eating BehaviorJ Clin Endocrinol Metab, September 2008, 93(9):3640–3643
Acknowledgments Download full-text
We gratefully acknowledge the ongoing contribution of the parents and
children in the Twins’ Early Development Study (TEDS).
Health Behavior Research Centre, University College London, Gower
Street, London WC1E 6BT, United Kingdom. E-mail: email@example.com.
Research Council, and the work on obesity in TEDS is supported by a
grant from the Biotechnology and Biological Sciences Research Council
(31/D19086). J.W. is supported by Cancer Research UK, R.P. and C.H.
by the Medical Research Council, and I.S.F. and S.O by the Medical
Research Council UK, Wellcome Trust, and NIHR Cambridge Biomed-
ical Research Centre.
I.S.F. and S.O. conceived, designed, and funded the study in obese chil-
dren. C.M.A.H. assisted with the statistical analyses. J.W., I.S.F., and
S.O. drafted the manuscript. All authors contributed to the final critical
revision of the manuscript.
Disclosure Statement: J.W., S.C., C.M.A.H., I.S.F., S.O., and R.P.
have nothing to declare.
1. Frayling TM, Timpson NJ, Weedon MN, Zeggini E, Freathy RM, Lindgren
L, Loos RJ, Barroso I, Wareham NJ, Karpe F, Owen KR, Cardon LR, Walker
McCarthy MI 2007 A common variant in the FTO gene is associated with
body mass index and predisposes to childhood and adult obesity. Science
2. Dina C, Meyre D, Gallina S, Durand E, Korner A, Jacobson P, Carlsson LM,
Kiess W, Vatin V, Lecoeur C, Delplanque J, Vaillant E, Pattou F, Ruiz J, Weill
P, Kovacs P, Marre M, Balkau B, Cauchi S, Chevre JC, Froguel P 2007 Vari-
ation in FTO contributes to childhood obesity and severe adult obesity. Nat
Orru M, Usala G, Dei M, Lai S, Maschio A, Busonero F, Mulas A, Ehret GB,
Adams TD 2008 Association of the FTO gene with BMI. Obesity (Silver
5. Hinney A, Nguyen TT, Scherag A, Friedel S, Bronner G, Muller TD, Grallert
association (GWA) study for early onset extreme obesity supports the role of
fat mass and obesity associated gene (FTO) variants. PLoS ONE 2:e1361
6. Stratigopoulos G, Padilla S, Leduc CA, Watson E, Hattersley AT, McCarthy
MI, Zeltser LM, Chung WK, Leibel RL 2008 Regulation of Fto/Ftm gene
expression in mice and humans. Am J Physiol Regul Integr Comp Physiol
P, Prieur X, Coll AP, Ma M, Jovanovic Z, Farooqi IS, Sedgwick B, Barroso I,
Lindahl T, Ponting CP, Ashcroft FM, O’Rahilly S, Schofield CJ 2007 The
demethylase. Science 318:1469–1472
8. Farooqi IS, O’Rahilly S 2005 Monogenic obesity in humans. Annu Rev Med
influence on childhood adiposity despite the force of the obesogenic environ-
ment. Am J Clin Nutr 87:398–404
10. Wardle J, Guthrie CA, Sanderson S, Rapoport L 2001 Development of the
11. Carnell S, Wardle J 2007 Measuring behavioural susceptibility to obesity:
12. Ashcroft J, Semmler C, Carnell S, van Jaarsveld CH, Wardle J 8 August 2007
Continuity and stability of eating behaviour traits in children. Eur J Clin Nutr
13. Cole TJ, Freeman JV, Preece MA 1995 Body mass index reference curves for
the UK, 1990. Arch Dis Child 73:25–29
14. Carnell S, Wardle J Appetite and adiposity in children: evidence for a behav-
ioral susceptibility theory of obesity. Am J Clin Nutr 88:22–29
15. Preacher KJ, Hayes AF 2004 SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behav Res Methods Instrum Comput 36:
16. MacKinnon DP, Fairchild AJ, Fritz MS 2007 Mediation analysis. Annu Rev
17. Schachter S 1968 Obesity and eating. Internal and external cues differentially
affect the eating behavior of obese and normal subjects. Science 161:751–756
A 2003 Overweight children overeat after exposure to food cues. Eat Behav
19. Rankinen T, Bouchard C 2006 Genetics of food intake and eating behavior
phenotypes in humans. Annu Rev Nutr 26:413–434
20. Carnell S, Haworth CM, Plomin R, Wardle J 2008 Genetic influence on ap-
petite in children. Int J Obes, in press
21. Do R, Bailey SD, Desbiens K, Belisle A, Montpetit A, Bouchard C, Pe ´russe L,
J Clin Endocrinol Metab, September 2008, 93(9):3640–3643jcem.endojournals.org