SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Objective

Attention-deficit/hyperactivity disorder (ADHD) is an etiologically complex heterogeneous behavioral disorder. Several studies have reported that ADHD subjects are more likely to be overweight/obese and that this comorbidity may be due to shared genetic factors. The objective of this study is to explore the association between ADHD and FTO, a gene strongly associated with obesity in genome-wide studies.

Design and Methods

One tag SNP (single-nucleotide polymorphism, rs8050136, risk allele A) in the FTO gene was selected and its association with ADHD was tested. Family-based association tests (FBATs) were conducted with the categorical diagnosis of ADHD as well as behavioral and cognitive phenotypes related to ADHD. Furthermore, stratified FBAT analyses based on maternal smoking during pregnancy (MSDP) status were conducted.

Results

Statistically significant associations were observed between rs8050136 and several of the traits tested in the total sample. These associations were stronger when the analysis was restricted to children who were not exposed to MSDP.

Conclusions

These exploratory results suggest the involvement of the FTO SNP rs8050136 in modulating the risk for ADHD, particularly in those children who were not exposed to MSDP. If confirmed, they may explain, at least in part, the complex links between obesity and ADHD.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent and complex childhood disorder. ADHD has a strong genetic component (mean heritability 76%), and it has been suggested that multiple genes are involved in its etiology, each accounting for a small portion of increased risk [1]. Additionally, environmental factors particularly maternal smoking during pregnancy (MSDP) is thought to play a role in the disorder [2]. Despite extensive research, results from genetic association studies have been difficult to replicate, which could be due to a high level of etiological heterogeneity of this disorder.

Recently, a link between neurobiological systems implicated in attention, motor control, appetite, and body mass regulation has been suggested [3]. Furthermore, it was reported that ADHD subjects have higher body mass index standard deviations scores (BMI-SDS) and higher percentage of body fat [4] compared with controls.

Genome-wide association studies (GWAS) have strongly implicated several genes [5] in fat mass regulation and obesity. Among these, the fat mass and obesity (FTO) gene located on chromosome 16 [6] showed the strongest association with obesity and appears to be acting through the modulation of neurobiological systems. Indeed, the level of expression of this gene is highest in the brain [6]. Furthermore, a loss-of-function mutation of the FTO was associated with microcephaly, structural/functional brain abnormalities, and severe psychomotor delay [7], whereas a duplication of this gene due to 16q trisomy produces mental retardation and ADHD [8]. In the elderly, FTO has been associated with reduced brain volume [9], increased risk for Alzheimer's disease [10], and poorer performance in executive function domains [11]. In animals, a knockout of the Fto gene resulted in a phenotype with reduced body weight and fat, decreased motor activity, and increased energy expenditure. Most interestingly, these effects were specifically related to an increased sympathetic (adrenaline and noradrenalin) tone in the central nervous system [12].

Within FTO, a number of tag single-nucleotide polymorphisms (SNPs) located in intron 1 have been associated with obesity [6]. Among these, rs8050136 (A/C) is in strong linkage disequilibrium (LD) with five other SNPs (r2 ≥ 0.88) that were all associated with early-onset obesity [13]. Interestingly, allele A of rs9939609, which is in complete LD with allele A of rs8050136, was found to be associated with lower alcohol consumption and cigarette smoking [14].

Given the link between obesity and ADHD, the association of the FTO gene with obesity and behavioral phenotypes relevant for ADHD, and the implication of the FTO gene in brain development, we primarily investigated the association of rs8050136 with ADHD and also explored the relation between rs8050136 and ADHD-related neurocognitive and behavioral phenotypes. Furthermore, given the importance of MSDP in modulating risk for ADHD and the association between smoking and rs9939609, we also explored the association between rs8050136 in children stratified with regard to MSDP.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Subjects, study procedures, and ethics

Four hundred and fifty-one ADHD children (349 boys and 102 girls), ages 6-12 years (mean = 9.05; SD = 1.86), were recruited from the Disruptive Behaviour Disorders Program and the children outpatient clinic at the Douglas Mental Health University Institute. Children were referred to these specialized care facilities by schoolteachers, community social workers, and pediatricians. The research protocol was approved by the Research Ethics Board of the Douglas Hospital. Children with ADHD and their parents were explained the study in detail, and they provided verbal assent and written consent, respectively.

Children included in this study met DSM-IV diagnosis criteria for ADHD. A comprehensive clinical evaluation was used to establish the diagnosis of ADHD (see details in Ref. [15]). Briefly, ADHD diagnosis was based on clinical examination of the child and an interview of at least one of his or her parents by a child psychiatrist (RJ or NG). In addition, a structured clinical interview with parents using the Diagnostic Interview Schedule for Children-version IV, DISC-IV, was used to corroborate the diagnosis.

Children were excluded from this study if they had an IQ <70 on the Wechsler Intelligence Scale for Children-III/IV (WISC-III or WISC-IV), Tourette syndrome, pervasive developmental disorder, or psychosis. Among the total sample, 78.1% were male, 86.9% were of Caucasian ethnicity, and 28.9% belonged to families with an annual income of less than CAD$ 20,000. A total of 52.8% met DSM-IV criteria for the combined subtype, whereas 37.3 and 9.9% were diagnosed with the inattentive and hyperactive subtypes, respectively. A total of 38.8% were previously receiving medication for their ADHD symptoms; 2.1% were underweight, 57.9% were of normal weight, 19.7% were overweight, and 20.3% were obese as per BMI category according to WHO classification (this distribution of weight categories was not significantly different between the three genotype groups). Comorbid disorders such as oppositional defiant disorder (40.4%), conduct disorder (21.7%), anxiety disorder (44.1%), and mood disorder (8.3%) were present in proportions similar to those reported in previous studies. All the behavioral and neurocognitive assessments were completed while the children were not taking any medication. In cases, where children were on medication prior to their inclusion in the study, all clinical, behavioral, neurocognitive, and task-engagement assessments were carried out at the end of a 1-week washout period.

Clinical and behavioral evaluation

Behaviors relevant to ADHD were assessed using several assessment tools for the purpose of quantitative genetic analyses as previously described [16]. Briefly, the Child Behavior Checklist (CBCL), which assesses several behavioral dimensions of the child, was completed by the parents. Furthermore, the child's behavior in the home and classroom environment was evaluated by parents and teachers using the Conners’ Global Index for Parents and Teachers (CGI-P and CGI-T), respectively.

Neurocognitive evaluation

A neuropsychological (NP) battery of tests specially designed for children was used to study different executive function domains as described in a previous publication [15]. The Wechsler Intelligence Scale (WISC) [17] was used to evaluate the full scale (FS), verbal (V), and performance (P) IQ. The Wisconsin Card Sorting Test (WCST) [18] was used to assess cognitive flexibility and set-shifting. Similarly, the Finger Windows subtest [19] was used to assess visual-spatial working memory, and the Tower of London (TOL) [20] assessed planning, organization, and problem-solving capacity. Additionally, the Self-Ordered Pointing Task (SOPT) estimated working memory, planning, and response inhibition [21]. Finally, the Conners’ Continuous Performance Test (CPT) [22] was used to measure attention, response inhibition, and impulse control.

Task-engagement evaluation

Task-oriented behavior in children with ADHD was assessed within the clinic using the Restricted Academic Situation Scale (RASS) [23]. RASS is a specialized coding system developed for observing and recording the child's behavior when he/she is assigned a task (a set of math problems in our study), during a simulated independent academic situation within a clinical setting. It assesses the child's ability for task engagement to regular, repetitive academic work in the presence of potential distractions, with no adult supervision, and has been conducted as described [24].

MSDP evaluation

Obstetric complications, including pregnancy and delivery, and perinatal complications were systematically assessed using the Kinney Medical Gynaecological Questionnaire and scored using the McNeil-Sjöström scale [25]. During this assessment, mothers were asked questions about smoking during the three trimesters of pregnancy. If mothers smoked during at least one trimester of their pregnancy, their children were coded as “exposed” (n = 181), whereas if mothers did not smoke at all during pregnancy, their children were coded as “unexposed” (n = 230).

Genetic analyses

The affected child and his/her family, including the parents and the unaffected siblings, were invited to participate in the genetic component of the study. DNA was extracted for each participant and his/her associated family member, using a blood sample, a buccal swab, or a saliva sample. This study included a total of 380 nuclear families having one or more child with a DSM-IV diagnosis of ADHD. Of the 380, 184 were trios with information from both parents, 18 were trios with two affected children, 49 were trios with information from one parent and one or more unaffected sibling, 115 were duos including the proband and one parent, whereas 14 were families with two affected siblings and one parent. rs8050136 SNP was genotyped using Sequenom iPlex Gold Technology [26]; genotyping error was estimated using duplicates of two reference samples included in each plate. Furthermore, genotypes for all the samples included in the study were read with 100% accuracy and the genotype distribution (AA = 13.4%, AC = 48.7%, and CC = 38.0%) of this marker did not depart from Hardy–Weinberg equilibrium (P > 0.05).

Statistical methods

The family-based association test (FBAT) statistical package (version 2.0.3) [27] was used to examine the overtransmission of a specific allele from parent to affected offspring. Genetic association of rs8050136 with ADHD diagnosis and behavioral, cognitive, and task-engagement quantitative traits/phenotypes were examined. All the analyses were conducted under the assumption of an additive model, with the null hypothesis of no linkage and no association. At the first level, FBAT analysis was conducted with the total sample. However, given the results of earlier studies indicating that MSDP is the environmental risk factor most consistently associated with ADHD [2] and the fact that the FTO gene was associated with smoking behavior, possible modulation of the association between the FTO genotype and ADHD-related phenotypes was explored by stratifying participants into two groups based on MSDP status (±MSDP). At the time of submission of this manuscript, this was, to our knowledge, the first study exploring possible association between FTO candidate gene and ADHD as a diagnostic entity, and given the previous literature indicating that mutations in this gene may cause attention and cognitive deficits, the significance level of association with ADHD as a diagnosis was set at P = 0.05 and not corrected for multiple testing.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

In the total sample, FTO SNP rs8050136 was marginally associated with ADHD diagnosis (P = 0.05) (Table 1). More specifically, the A allele (associated with obesity in weight studies) was undertransmitted from parents to the affected child. Furthermore, exploratory quantitative FBAT analysis with the clinical, behavioral, IQ, EF, and task-engagement phenotypes revealed that the A allele of rs8050136 was associated with better performance on the Finger windows total score (P = 0.004) and the CPT reaction time (P = 0.04) (Table 2). Likewise, concerning task-oriented behaviors, the A allele of rs8050136 was positively associated with RASS total score (P = 0.01) (Table 3). However, the A allele of rs8050136 was not associated with BMI score (Table 4).

Table 1. FBAT output detailing the association between rs8050136 and clinical and behavioral dimensions
 Total sample of ADHD childrenADHD children exposed to maternal smoking during pregnancyADHD children not exposed to maternal smoking during pregnancy
TraitFamily no.Z statisticP-valueFamily no.Z statisticP-valueFamily no.Z statisticP-value
  1. For the CBCL, a lower t-score is indicative of better behavior. FTO SNP rs8050136 (A/C), allele frequency for the A allele = 0.384 and for the C allele = 0.616 in the total ADHD sample. Results in the table have been depicted for the A allele because this is considered the risk allele for obesity according to published GWAS.

  2. CBCL, Child Behaviour Checklist; Conners’ Parents, Conners’ Global Index-Parents version questionnaire; Conners’ Teachers, Conners’ Global Index-Teacher version questionnaire; DISC-IV, Diagnostic Interview Schedule for Children-version IV.

  3. a

    P ≥ 0.01 and ≤ 0.05 (trends for association); **P < 0.01.

ADHD166−1.9470.051*600.4980.618103−2.630.008a
Total number of DISC ADHD items171−1.3140.188610.4160.677105−1.7490.080
Conners’ parents168−0.2720.785600.9740.329104−1.0640.287
Conners’ teachers164−1.0090.313590.6770.498101−1.670.094
CBCL total score175−0.7690.441620.6260.531108−1.4920.135
CBCL internalizing behavior173−0.530.596620.7110.477107−1.2510.210
CBCL externalizing behavior172−0.570.568610.7470.455106−1.3230.185
CBCL withdrawn145−1.060.289580.7240.46982−2.2690.023a
CBCL somatic complaints138−0.8260.408500.2540.79982−1.0980.272
CBCL anxious/depressed161−0.8070.419590.8230.41097−1.5480.121
CBCL social problems166−0.4660.641611.3920.16499−1.50.133
CBCL thought problems154−1.0560.290540.4710.63794−1.7520.079
CBCL attention problems173−1.430.152620.5020.615106−2.0110.044a
CBCL delinquent behavior360.5670.570190.7980.42417−0.1740.861
CBCL aggressive behavior165−0.7480.454610.8140.41599−1.6930.090
Table 2. FBAT output detailing the association between rs8050136 and cognitive endophenotypes
 Total sample of ADHD childrenADHD children exposed to maternal smoking during pregnancyADHD children not exposed to maternal smoking during pregnancy
TraitFamily no.Z statisticP-valueFamily no.Z statisticP-valueFamily no.Z statisticP-value
  1. For the WCST, FW, and TOL, a higher standard score is indicative of better performance; for the SOPT, a lower score is indicative of better performance; for the CPT, the t-scores are standard scores that use a mean of 50, where a high t-score (≥60) indicates slow response speed. FTO SNP rs8050136 (A/C), allele frequency for the A allele = 0.384 and for the C allele = 0.616 in the total ADHD sample. Results in the table have been depicted for the A allele because this is considered the risk allele for obesity according to published GWAS.

  2. CPT, Conners’ Continuous Performance Test; FW, Finger Windows (visual-spatial working memory); SOPT, Self-Ordered Pointing Task; TOL, Tower of London test (planning, organization, and problem-solving capacity); WCST, Wisconsin Card Sorting Test (measure of cognitive flexibility and set-shifting); WISC, Wechsler Intelligence Scale-version III or IV.

  3. a

    P ≥ 0.01 and ≤ 0.05 (trends for association);

  4. b

    P < 0.01.

WISC IQ1551.4440.148541.1380.255970.5410.588
WISC verbal IQ1500.6760.49954−0.6630.507910.9190.357
WISC performance IQ1480.840.401531.3680.17190−0.3390.734
WCST total errors standard score1520.0250.98055−0.4510.65193−0.2310.817
WCST perseverative responses1500.830.406520.3110.755940.1590.873
WCST perseverative errors1520.4230.672540.3590.71994−0.4550.649
WCST nonperseverative errors151−0.3860.69954−0.7550.450930.1530.878
FW score108−2.830.004b32−0.3190.74973−2.9940.002b
SOPT score168−1.5370.124610.5970.550102−2.3420.019a
TOL score139−1.0770.281520.1120.91084−1.6640.096
CPT         
Omission errors142−0.5560.577460.1290.89792−0.0030.997
Commission errors170−0.980.32761−0.7260.467105−0.7360.461
Attention score170−1.0110.31261−1.2860.198105−0.0120.990
Hit reaction time169−1.9630.049a60−1.5590.118105−1.0620.288
Perseveration score142−1.3690.170460.0940.92492−2.2140.026a
Overall index84−0.0680.945390.250.802400.220.825
Table 3. FBAT output detailing the association between rs8050136 and task-engagement endophenotypes
 Total sample of ADHD childrenADHD children exposed to maternal smoking during pregnancyADHD children not exposed to maternal smoking during pregnancy
TraitFamily no.Z statisticP-valueFamily no.Z statisticP-valueFamily no.Z statisticP-value
  1. FTO SNP rs8050136 (A/C), allele frequency for the A allele = 0.384 and for the C allele = 0.616 in the total ADHD sample. Results in the table have been depicted for the A allele because this is considered the risk allele for obesity according to published GWAS.

  2. RASS, Restricted Academic situation scale.

  3. a

    P ≥ 0.01 and ≤ 0.05 (trends for association);

  4. b

    P < 0.01.

RASS         
Total score172−2.4720.013a62−0.0010.999105−3.1280.001b
Vocalization104−1.2970.194400.1110.91162−1.7230.084
Fidgeting167−2.3950.016a59−0.0520.958103−2.8330.004b
Off-task156−2.4510.014a55−0.0580.95396−3.1580.001b
Plays with object150−2.5250.011a55−0.3070.75990−3.0530.002b
Out of seat132−0.9080.363480.5350.59281−1.4020.161
Table 4. FBAT output detailing the association between rs8050136 and BMI in ADHD children
 Total sample of ADHD childrenADHD children exposed to maternal smoking during pregnancyADHD children not exposed to maternal smoking during pregnancy
TraitFamily no.Z statisticP-valueFamily no.Z statisticP-valueFamily no.Z statisticP-value
  1. FTO SNP rs8050136 (A/C), allele frequency for the A allele = 0.384 and for the C allele = 0.616 in the total ADHD sample. Results in the table have been depicted for the A allele because this is considered the risk allele for obesity according to published GWAS.

  2. BMI, body mass index.

  3. *P ≥ 0.01 and ≤ 0.05 (trends for association); **P < 0.01.

BMI84−1.3220.186230.1310.89557−1.9150.055*

Additionally, stratified exploratory analysis based on MSDP status indicated that the genetic associations of FTO rs8050136 were significant only in the subgroup of patients who were not exposed to MSDP. For example, the association of FTO rs8050136 improved by one order of magnitude with both ADHD diagnosis (P = 0.008, Table 1) and task-engagement RASS total score (P = 0.001, Table 4) in the subgroup of children who were not exposed to MSDP. More specifically, the A allele of rs8050136 was significantly undertransmitted to ADHD children who were not exposed to MSDP and to those children with less severe behavioral and cognitive traits relevant for ADHD. FBAT–e option provided similar findings. Finally, stratified exploratory analysis based on ADHD subtype and medication naivety did not show any association between FTO SNP rs8050136 and ADHD (Supplementary Table 1 & 2).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

The FTO gene has been consistently associated with the modulation of body mass [28]. Although it may exert its effects on fat tissues through its effects on the adipocytes, [29], there is strong evidence indicating that its effects are due to the modulation of brain pathways implicated in energy metabolism and food intake [30]. Furthermore, there is evidence that its effects may depend on its differential implication in various brain regions and at different developmental stages [31]. Indeed, it has been shown that mice lacking the Fto gene show hyperphagia, decreased locomotor activity, increased sympathetic tone (adrenaline and noradrenaline), and postnatal growth retardation [12]. Interestingly, in humans, one case report suggests that overexpression of FTO (16q11.2-16q13 duplication) results in a phenotype of ADHD associated with obesity and mental retardation [8]. Consistent with this result, it has been shown that experimental duplication or triplication of the Fto gene in mice results in an overexpression of this gene and proportional increase in weight [32]. However, the literature is still unclear with regard to the effect of the risk alleles located in the first intron of the FTO gene on gene expression.

In this study, we investigated the role of the tag SNP rs8050136 within FTO in children with ADHD. Allele A of this polymorphism has been consistently associated with obesity. Additionally, we also explored the association between this gene and ADHD while stratifying children according to their exposure to MSDP. This latter analysis was motivated by a study suggesting that the obesity risk alleles within the FTO gene are associated with lower risk for cigarette smoking [14]. The stratification on the basis of MSDP was also motivated by data from our laboratory strongly suggesting that the groups of children exposed or not to MSDP may be different from a clinical [33] and etiological points of view [34].

The main result of the study showed that the A allele of FTO rs8050136 polymorphism is undertransmitted to children with ADHD from their parents. Furthermore, exploratory quantitative trait analysis showed that rs8050136 A allele is associated with lower severity of ADHD and better functioning on tests measuring executive function and task-engagement traits, but not associated with BMI. These results were stronger on children who were not exposed to MSDP. Although this article was under review, Velders et al. reported a study investigating the association between FTO and ADHD traits in a large cohort (n = 1,718) of normally developing children. In line with our results, they observed that there was no association between FTO rs9939609 and BMI in young children. More importantly, children with the A allele of FTO rs9939609 were less likely to have symptoms of ADHD (OR = 0.74, P = 0.01) and showed more emotional control (OR = 0.64, P = 0.01) compared with children without the A allele [35]. Given that rs8050136 is in perfect LD with rs9939609 (r2 = 1 and D′ = 1), this study and the study reported recently by Velders et al. are highly convergent as they replicate the same results with regard to obesity and behavioral phenotypes in ADHD and in typically developing children.

Previous studies have significantly associated the A allele of FTO rs8050136 polymorphism with obesity [13]; thus, we were expecting an overtransmission of this allele to children affected with ADHD. Contrary to our expectation, this allele was undertransmitted to affected children in this study. Nonetheless, given the fact that allele A is an intronic variant, it may be implicated in regulating the level of expression of the FTO gene, with an effect on gene expression that remains to be determined. In case this variant is associated with a lower expression of the FTO gene, its effects may partially mimic the knockout of the Fto gene in mice. Under such scenario, children with ADHD who inherit this allele from their parents will be expected to have less severe psychopathology given that the knockout model of the Fto associates reduced motor activity and increased brain catecholamines [12]. Indeed, as ADHD has been conceived as a disorder with lower brain catecholamines (including noradrenaline) [36], it is expected that higher level of catecholamines due to a partial loss of function associated with the A allele would explain a milder form of ADHD and better executive function. Also, symptoms of ADHD are improved by the use of medication acting as selective norepinephrine-reuptake inhibitor, such as atomoxetine.

This study also shows a stronger association of the A allele of FTO rs8050136 polymorphism with ADHD-relevant traits in the subgroup of ADHD children whose mothers did not smoke during pregnancy, and this association was completely absent in those subject who were exposed to MSDP. These observations are consistent with a previous study showing that homozygosity for the A allele of rs9939609 (another FTO SNP in high LD with rs8050136) is associated with lower tobacco smoking [14]. Given the fact that ADHD is highly comorbid with smoking, drug/substance use, and alcohol abuse (reviewed in Ref. [37]), it is possible that the A allele of rs8050136 is indexing a group of children with ADHD whose mothers are less likely to smoke during pregnancy, and who have a milder form of ADHD. Interestingly, this is consistent with previous literature showing that children whose mother did not smoke during pregnancy have a milder form of psychopathology and better performances on executive function [38].

The finding that the A allele of FTO rs8050136 polymorphism is not associated with BMI is in contrast with results of previous GWAS [39]. This result may be attributed to the relatively smaller number of subjects and the resulting lack of statistical power in our study to detect an effect on weight, although a recent study with a larger sample size reported results similar to our study [35]. In case the association with ADHD and cognitive traits reported here is true, this may suggest that the effect of the FTO gene on behavioral phenotypes is stronger than on the body-weight phenotype during childhood, although this needs further confirmation in a larger sample of patients.

The primary outcome result of this study (overall association with the disorder) was not corrected for multiple testing, because mutations in the FTO candidate gene have been previously associated with cognitive deficits [11], and neurocognitive deficits are believed to result in behavioral symptoms displayed by ADHD children [40]. However, given that the primary association investigated was with rs8050136 and ADHD as a diagnostic entity, even if the Bonferroni correction was to be applied (1 SNP × three exposure strata, P = 0.05/3 = 0.016), the association results in the group where the mothers did not smoke during pregnancy would remain significant. However, the remaining results that identify associations between FTO rs8050136 polymorphism and quantitative phenotypes were not corrected for multiple testing as they are considered exploratory. Nonetheless, given the recent replication of our finding in an independent study [35], these preliminary findings gain further validity and may help inform future genetic studies of the disorder. It is critical to have independent replication before definitive conclusions can be reached.

Finally, this is the first study investigating the association between FTO and ADHD, uses a stringent family-based association design, and explores several behavioral dimensions measured in different environment (home, school, and clinic) by different raters. The convergence of positive association along many of these dimensions and the robustness of the family-based association study increases confidence in these results. Also, the selection of the genetic variants that were strongly associated with somatic phenotypes (obesity) implies that these variants may be functional. Nonetheless, further studies with bigger sample sizes are needed to confirm or negate these results. This is particularly true for the stratified analysis by MSDP. Indeed, the subgroup of children whose mother smoked during pregnancy is relatively small.

In conclusion, these results implicate FTO in the modulation of ADHD phenotype and suggest that this effect is more prominent in those children who were not exposed to MSDP. If replicated in independent large samples they shed light on the physiopathology of ADHD and its possible link with obesity.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

The authors thank Johanne Bellingham, Jacqueline Richard, Sandra Robinson, Phuong-Thao Nguyen, Rosherrie DeGuzman, Marina TerStepanian, Anna Poloskia, Matthew Lebaron, Nicole Pawliuk, and Sharon Hill for technical and clinical assistance. A special thanks to the families who participated in this research.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
oby20444-sup-0001-SuppInfo.doc29KSupporting Information
oby20444-sup-0002-SuppInfo.doc29KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.