An obesity-associated risk allele within the FTO gene affects human brain activity for areas important for emotion, impulse control and reward in response to food images.

TL;DR: The results suggest that the two genotypes are associated with differential neural processing of food images, which may influence weight status through diminished impulse control and reward processing.

Abstract: Understanding how genetics influences obesity, brain activity and eating behaviour will add important insight for developing strategies for weight-loss treatment, as obesity may stem from different causes and as individual feeding behaviour may depend on genetic differences. To this end, we examined how an obesity risk allele for the FTO gene affects brain activity in response to food images of different caloric content via functional magnetic resonance imaging (fMRI). Thirty participants homozygous for the rs9939609 single nucleotide polymorphism were shown images of low- or high-calorie food while brain activity was measured via fMRI. In a whole-brain analysis, we found that people with the FTO risk allele genotype (AA) had increased activity compared with the non-risk (TT) genotype in the posterior cingulate, cuneus, precuneus and putamen. Moreover, higher body mass index in the AA genotype was associated with reduced activity to food images in areas important for emotion (cingulate cortex), but also in areas important for impulse control (frontal gyri and lentiform nucleus). Lastly, we corroborate our findings with behavioural scales for the behavioural inhibition and activation systems. Our results suggest that the two genotypes are associated with differential neural processing of food images, which may influence weight status through diminished impulse control and reward processing.

Summary

INTRODUCTION

• Several different FTO single nucleotide polymorphisms (SNPs) are associated with a higher body mass index (BMI) (Sällman Almén et al., 2013; Scuteri et al., 2007a) , and higher energy intake (Speakman, 2013) .
• Moreover, experiments in rodents show that changes in FTO expression levels in the hypothalamus affect feeding behavior (Frederiksen, Skakkebaek, & Andersson, 2007; Olszewski et al., 2009; Tung et al., 2010) .
• Personality scales for the Behavioral Inhibition System (BIS) and Behavioral Activation System (BAS), which measure punishment and reward sensitivity respectively, are two such tools which correlate with inactivity and poor diet (Carver & White, 1994; Dietrich, Federbusch, Grellmann, Villringer, & Horstmann, 2014; Meule, 2013; Voigt et al., 2009) .

FTO Associated Brain Activity 5

• To date, however, few fMRI studies have examined how genetic profile is associated with brain responses to food in obesity.
• A recent study found that people with the FTO risk allele for rs8050136 had reduced activity in the right prefrontal cortex while viewing food images in a postprandial state, but not while fasting (Heni et al., 2014) .
• Notably, their cohort of participants had a normal BMI with no obese participants.
• Against this background, the authors explore for the first time the association between FTO genotype, BMI, and neural responses to food images of either low-or high-calorie content.

Participants

• Prior to any experimental procedures, all participants gave written informed consent to the study which conformed to the Declaration of Helsinki and approved by the local ethics committee.
• Genotyping of the FTO single nucleotide polymorphism (SNP) rs9939609 was performed with a pre-designed Taqman single-nucleotide polymorphism genotyping assay (Applied Biosystems, Foster City, USA) and an ABI7900 genetic analyzer with SDS 2.2 software at the Uppsala Genome Center (http://www.genpat.uu.se/node462).
• Only homozygous participants were included in the study.
• Hunger ratings were also assessed on a 1-10 scale with higher numbers indicating greater feelings of hunger.

Behavioral Questionnaires

• Clinical measures for punishment sensitivity and reward-seeking behavior were acquired using the Behavioral Inhibition System (BIS) and Behavioral Activation System (BAS) questionnaires (Carver & White, 1994) .
• Each item was represented by a statement, where the participant indicated how much s/he agreed or FTO Associated Brain Activity 7 disagreed on a four-point scale.
• The BIS included only one scale, evaluating the reactions to the anticipation of punishment and anxiety, while the BAS included three subscales: Drive, which represents the pursuit of desired goals; Fun Seeking, which evaluates the desire for new rewards and impulsivity; and Reward Responsiveness, which focuses on positive reactions anticipating rewards.

Preprocessing of fMRI data

• All preprocessing steps were performed using software package Statistical Parametric Mapping (SPM, version 8, http://www.fil.ion.ucl.ac.uk/spm/), implemented in MATLAB (version R2014a, 11 FEB 2014 .
• The images were realigned and estimated to remove movement artefacts in the data.
• EPI images were further matched with the structural image using coregistration.
• The anatomical image was segmented to strip away unnecessary tissue in the images.
• Tissue probability maps were introduced in the segmentation step to differentiate between gray matter, white matter and cerebrospinal fluid in each individual.

Statistical Analysis

• All fMRI statistical analysis was performed using the same versions of SPM and MATLAB listed in preprocessing steps.
• For all whole-brain results, a family wise error FTO Associated Brain Activity 9 (FWE) corrected significance level was set at p < 0.05 to correct for multiple testing.
• This contrast was then tested using a between-groups t-test followed by directional post-hoc comparisons as well as with a multiple regression analysis testing for interactions between genotype and BMI, BIS, or BAS individually.
• Bilateral masks of such areas were produced using the Wake Forest University Pickatlas toolbox (Maldjian, Laurienti, Kraft, & Burdette, 2003) within SPM.
• Results for the PCA were considered significant if the percentage of inertia summing from the two largest eigenvalues exceeded values listed in a significance table based on 10,000 analyses with similar numbers of individuals and independent variables (Lê et al., 2008) .

RESULTS

• The obesity-associated FTO SNP rs9939609, is associated with increased activity in response to food images.
• BOLD signals were measured as participants were shown images of low-calorie (LC) food, high-calorie (HC) food, or control images in a block design format.
• Areas included the posterior cingulate cortex (PCC), cingulate gyrus, cuneus, and precuneus (Table 2 ).
• A multiple regression analysis found an interaction between genotype and BMI, post-hoc comparisons found significant clusters for the AA genotype while BMI was decreasing in the PCC, cingulate gyrus, middle occipital gyrus, and precuneus (Supplementary Table 1 ).
• Within the t-test comparison between genotypes, a significant cluster showing greater activity in the AA genotype was FTO Associated Brain Activity 11 found in the putamen after performing a small-volumes correction using a 6 mmradius sphere over the lowest FWE-corrected p-value in the cluster (Table 2 ).

Differential patterns of behavior for each FTO genotype depending on body-mass index.

• The authors next tested if behavioral questionaires corrobarated the findings from the imaging experiments.
• The authors then performed a principle component analysis within each genotype using the BIS and the three BAS subscales (Drive, Fun Seeking, and Reward Responsiveness) as variables of interest with BMI as a quantitative supplementary variable.
• For both analyses, all the variables of interest were well projected and the first two dimensions accounted for ≈80% of the variablity (considered significant based on critera listed in methods under statistical analysis subheading, 81.4 > 80.0 for the AA group and 79.2 > 76.5 in the TT group).
• Moreover, the variables of interest projected to the same quandrants except for the Drive and Fun Seeking subscales, which were switched between the two different genotypes.
• Furthermore, the authors followed up the association between the BIS and BMI using a multiple regression analysis testing if BIS scores could be predicted by genotype, BMI, or their interaction.

DISCUSSION

• The authors examined whether an obesity-associated genotype affects the neural processing of food images with different caloric content and to what extent body-mass index (BMI) is an important factor.
• The authors found the AA genotype had increased brain activity compared to the TT genotype when viewing food images with different caloric contents, specifically in areas important for emotion (cingulate gyrus), memory, and self-image (cuneus and precuneus) and reward .
• Thus, discrimination between HC and LC foods may be handled differently for each genotype depending on BMI.
• Next, the authors corroborate their findings in the imaging study with personality questionnaires examining behavioral characteristics related to impulsivity and rewardprocessing: namely the Behavioral Inhibition System (BIS) and Behavioral Activation System (BAS) scales.
• The authors found that the BIS as well as subscales of the BAS correlated with BMI oppositely in each genotype.

FTO Associated Brain Activity 14

• In between-groups comparisons, as well as multiple regression analysis, the authors found significant clusters of brain activity when testing a contrast for caloric discrimination (HC food images opposed to LC food images).
• Specifically, the authors found increased neural activation in the AA genotype compared to the TT genotype within the posterior cingulate cortex (PCC), cingulate gyrus, cuneus and precuneus.
• The PCC is a well-connected and multifunctional brain area associated with emotional processing, and a central node in the default mode network (DMN): involved in arousal/awareness, balancing external and internal thought, and emotion (Leech & Sharp, 2014) .
• Thus, the AA genotype in their cohort confirms previous reports equating impulsivity with obesity/overeating (Meule, 2013) specifically in one study which also found a negative correlation between BIS and BMI in males (Dietrich et al., 2014) .
• In conclusion, their findings offer insight into the relationship between FTO, obesity, and brain activity; and suggest that overweight/obese populations have different attitudes and functional processing for food images depending on genetic background.

B)

• A region of interest analysis found a significant cluster within the putamen after a smallvolumes correction.
• The BIS evaluates inhibitory behavior in the anticipation of punishment and anxiety, while the BAS included three subscales: Drive, which represents the pursuit of desired goals; Fun Seeking, which evaluates the desire for new rewards and impulsivity; and Reward Responsiveness, which focuses on positive reactions anticipating rewards.
• The behavioral variables were all well projected in each group.
• A) variables factor map for the AA genotype.
• BMI was positively correlated with BIS and negatively correlated with the BAS Fun Seeking subscale.

Figures

Figure 1. Participants homozygous for the FTO risk-allele (rs9939609) have increased 541 brain activity in response to high- and low-calorie food images. 13 participants 542 homozygous for the FTO risk allele (A) and 17 for the non-risk allele (T) were shown images 543 of either low-calorie (LC), high-calorie (HC) food, or control while brain activations were 544 measured via fMRI. A whole-brain, between-groups comparison tested the neural response to 545 foods of different caloric content (HC versus LC contrast) using total brain volume (TBV) 546 and body-mass index (BMI) as covariates of no interest. Greater activity was found in four 547 clusters for participants homozygous for the risk allele (A) compared to the normal allele (T). 548 A) Axial brain slices showing significant clusters. Exact MNI z-coordinate given below 549 slices in white numbers, and t-statistics indicated by color-coded scale bar (critical t = 3.5). 550 B) A region of interest analysis found a significant cluster within the putamen after a small-551 volumes correction. 552

Full text

Wiemerslage, L, Nilsson, EK, Solstrand Dahlberg, L, Ence-Eriksson, F, Castillo,
S, Larsen, AL, Bylund, SBA, Hogenkamp, PS, Olivo, G, Bandstein, M, Titova,
OE, Larsson, E-M, Benedict, C, Brooks, SJ and Schiöth, HB
An obesity-associated risk allele within the FTO gene affects human brain
activity for areas important for emotion, impulse control and reward in
response to food images.
http://researchonline.ljmu.ac.uk/id/eprint/9287/
Article
LJMU has developed LJMU Research Online for users to access the research output of the
University more effectively. Copyright © and Moral Rights for the papers on this site are retained by
the individual authors and/or other copyright owners. Users may download and/or print one copy of
any article(s) in LJMU Research Online to facilitate their private study or for non-commercial research.
You may not engage in further distribution of the material or use it for any profit-making activities or
any commercial gain.
The version presented here may differ from the published version or from the version of the record.
Please see the repository URL above for details on accessing the published version and note that
access may require a subscription.
For more information please contact researchonline@ljmu.ac.uk
http://researchonline.ljmu.ac.uk/
Citation (please note it is advisable to refer to the publisher’s version if you
intend to cite from this work)
Wiemerslage, L, Nilsson, EK, Solstrand Dahlberg, L, Ence-Eriksson, F,
Castillo, S, Larsen, AL, Bylund, SBA, Hogenkamp, PS, Olivo, G, Bandstein,
M, Titova, OE, Larsson, E-M, Benedict, C, Brooks, SJ and Schiöth, HB (2016)
An obesity-associated risk allele within the FTO gene affects human brain
LJMU Research Online

http://researchonline.ljmu.ac.uk/

FTO Associated Brain Activity 1
Title 1
An obesity-associated risk allele within the FTO gene affects brain activity for areas 2
important for emotion, impulse control, and reward in response to food images. 3
4
Running Title 5
FTO Associated Brain Activity 6
7
Author names and affiliation 8
Lyle Wiemerslage*
§
, Emil K Nilsson
§
, Linda Solstrand Dahlberg
§
, Fia Ence-Eriksson
§
, 9
Sandra Castillo
§
, Anna L Larsen
§
, Simon BA Bylund
§
, Pleunie S Hogenkamp
§
, Gaia 10
Olivo
§
, Marcus Bandstein
§
, Olga E Titova
§
, Elna-Marie Larsson
┼
, Christian Benedict
§
, 11
Samantha J Brooks
╪
, Helgi B Schiöth
§
12
13
Uppsala University § 14
Department of Neuroscience, Functional Pharmacology 15
Biomedicinska Centrum (BMC) 16
Husargatan 3, Box 593 17
751 24 Uppsala, Sweden 18
19
Section of Neuroradiology ┼ 20
Department of Radiology, Uppsala University 21
Akademiska Sjukhuset 22
751 85 Uppsala, Sweden 23
24
University of Cape Town ╪ 25
Department of Psychiatry 26
Old Groote Schuur Hospital 27
J2 Building 28
Anzio Road 29
Observatory, Cape Town, South Africa. 30
31
Corresponding author 32
Uppsala University * 33
Department of Neuroscience, Functional Pharmacology 34
Biomedicinska Centrum (BMC) 35

FTO Associated Brain Activity 2
Husargatan 3, Box 593 36
751 24 Uppsala, Sweden 37
lyle.wiemerslage@neuro.uu.se 38
39
Number of: 40
• Figures = 3 41
• Tables = 1 42
• Words: 43
o Abstract = 197 44
o Introduction = 565 45
o Entire Manuscript (excluding references and figure legends) = 3,972 46
47
Keywords: 48
FTO, fMRI, SNP, obesity, food 49
50
Conflict of Interest: 51
The authors declare no conflicts of interest. 52

FTO Associated Brain Activity 3
ABSTRACT 53
Understanding how genetics influences obesity, brain activity, and eating behavior will 54
add important insight for developing strategies for weight-loss treatment, as obesity may stem 55
from different causes and as individual feeding behavior may depend on genetic differences. 56
To this end, we examined how an obesity risk-allele for the FTO gene affects brain activity in 57
response to food images of different caloric content via fMRI. 30 participants homozygous 58
for the rs9939609 single nucleotide polymorphism were shown images of low- or high-calorie 59
food while brain activity was measured via fMRI. In a whole-brain analysis, we found that 60
people with the FTO risk-allele genotype (AA) had increased activity than the non-risk (TT) 61
genotype in the posterior cingulate, cuneus, precuneus, and putamen. Moreover, higher BMI 62
in the AA genotype was associated with reduced activity to food images in areas important for 63
emotion (cingulate cortex), but also in areas important for impulse control (frontal gyri and 64
lentiform nucleus). Lastly, we corroborate our findings with behavioral scales for the 65
behavioral inhibition and activation systems (BIS/BAS). Our results suggest that the two 66
genotypes are associated with differential neural processing of food images, which may 67
influence weight status through diminished impulse control and reward processing. 68
69
Keywords: 70
fMRI, FTO, SNP, BMI, food images, obesity 71

Frequently asked questions

Q1. What are the contributions in this paper?

56 To this end, the authors examined how an obesity risk-allele for the FTO gene affects brain activity in 57 response to food images of different caloric content via fMRI. Their results suggest that the two 66 genotypes are associated with differential neural processing of food images, which may 67 influence weight status through diminished impulse control and reward processing.