Our aim is to replicate the analyses of the PIAMA birth cohort in the ALSPAC birth cohort. We hypothesise that the association of the IL1RL1 pathway with asthma, the number of eosinophils and the wheezing phenotypes can be replicated in the ALSPAC birth cohort. Therefore, we would like to study the association of SNPs of the IR1RL1 pathway (genes IL33, IL1RL1, IL1RAcP, MYD88, TIRAP, IRAK1, IRAK4 and TRAF6) with a doctor diagnosis of asthma in childhood, with the number of eosinophils in childhood with wheezing phenotypes in childhood in the ALSPAC birth cohort. Furthermore, we would like to study 10 selected interactions of SNPs in the IL1RL1 pathway and the presence of a doctor diagnosis of asthma in childhood.

Hip fracture is a major cause of morbidity in older people, leading to loss of independence, and a huge burden economically in terms of healthcare costs. Osteoporosis, defined as reduction in bone mineral density (BMD) and bone strength, is strongly related to hip fracture risk. Several mechanisms contribute to the age-related decline in BMD and bone strength, including an age-related decline in the intensity and amount of PA, leading to a reduction in the quality and quantity of mechanical strains (defined as deformation of bone per unit length in relation to loading). For example, less than 30% of 65-74 year-olds and 15% of adults aged 75 and over report any moderate intensity PA lasting at least ten minutes in the previous four weeks (Craig R et al, 2009). These low levels of PA have numerous adverse health outcomes, reflecting the importance of physical activity for a range of systems including bone, leading to initiatives to co-ordinate existing activity promotion schemes (http://www.ageuk.org.uk).

Aims: To clarify the genetic component of white matter disease and outcome in preterm infants, by exploring genetic variation and correlation with imaging endophenotype and neurodevelopmental phenotype.

Hypothesis: Genetic variation in preterm infants correlates with white matter damage and developmental outcome at age 2.

Exposure variable: Preterm birth less than 37 weeks Gestational Age

Outcome variables:

1. Whole organism - Bayley III neurodevelopmental assessment at age 2 years

2. Endophenotype - White matter damage assessed by quantitative MRI at term equivalent age (Tract Based Spatial Statistics TBSS and Deformation Based Morphometry DBM)

Confounding variables: Sex, birth weight, IUGR, days of assisted ventilation, antenatal steroids, culture positive sepsis, chronic lung disease

Background: More than 1 in 10 babies worldwide are born prematurely, and 40-50% of children with birthweight less than 1000g have neurodevelopmental problems (Marlow et al. 2005). Rates of prematurity are increasing worldwide (WHO 2012) and the predominant type of pathology now consists of diffuse white matter injury rather than florid parenchymal lesions such as cystic periventricular leukomalacia.

Infection, inflammation, chronic lung disease, gender and intra-uterine growth restriction have all been shown to impact brain structure and outcome but this does not fully account for the range of neurological outcomes for preterm babies with similar clinical features in similar environments. Individual susceptibility to injury might have a role to play, and be modulated by genetic factors. The most severe functional end-point of the various brain parenchyma pathologies remains cerebral palsy, a multi-factorial heterogeneous phenotype with a stable incidence despite changes in obstetric and neonatal practices (Stanley et al 2000). A genetic component for cerebral palsy has been suggested by previous case-control studies (Wu et al 2009, Hollegaard et al 2012) but studies tend to report conflicting results and have to contend with small sample sizes and other methodological issues.

The ALSPAC dataset will allow us to overcome some of these hurdles by starting with a large cohort and further increasing power by focusing on biological pathways within those data rather than searching for individually significant SNPs. The documented clinical variables and outcome measures in the database provide the opportunity to account for confounders and assess the clinical significance of resulting gene candidates. Using our bioinformatics expertise we hope to apply novel pathway analysis methods to genome-wide data in order to uncover biological patterns. We aim to combine prior knowledge from three sources in order to select a final group of gene candidates and focus validation:

1. GWAS data (ALSPAC)

2. Biologically driven (neonatal mouse subplate co-expression patterns)

3. Literature evidence

MRI is a safe and non-invasive method of obtaining an intermediate measure of the brain substrate, and our perinatal neuroimaging group has extensive experience in this field. The data obtained by advanced neuroinformatics methods such as TBSS and DBM are quantitative endophenotypes that are known to relate to outcome in the preterm population (Boardman et al 2010, van Kooij et al 2012). We will use advanced bioinformatics strategies to integrate the genetic and imaging datasets, both large amounts of data that can be more informative in conjunction than separately. This will then guide validation in our cohort by correlation of gene variation and linked quantitative imaging, thus opening a window on genetic effects on brain structure and function.

References

Marlow N, Wolke D, Bracewell MA, Samara M, N Engl J Med, 2005;352(1):9-19

Born Too Soon: The Global Action Report on Preterm Birth, WHO 2012-08-13

Stanley F, Blair E, Alberman E. MacKeith Press; 2000

Wu D, Yan-Feng Z, Xiao-Yan X, Li Y, Gong-Chun Z, Xi-Song B, Jiu-Lai T, Dev Med Child Neurol, 2011;53: 217-225

Hollegaard MV, Skogstrand K, Thorsen P, Norgaard-Pedersen B, Hougaard DM, Grove J, Hum Mutat, 2012; Epub ahead of print

van Kooij BJ, de Vries LS, Ball G, van Haastert, Benders MJ, Groenendaal F, Counsell SJ, AJNR Am J Neuroradiol, 2012; 33(1):188-94

Boardman JP, Craven C, Valappil S, Counsell SJ, Dyet LE, Rueckert D, Aljabar P, Rutherford MA, Chew AT, Allsop JM, Cowan F, Edwards AD, Neuroimage, 2010; 52(2): 409-14

Circulating concentrations of asymmetric dimethylarginine (ADMA) strongly predict CV outcomes [1] however causal relationships have been difficult to establish. Local regulation of ADMA levels is achieved through metabolism by three enzymes dependent on the tissue and cell type: the two isoforms of dimethylarginine dimethylaminohydrolase (DDAH) and the poorly characterised enzyme alanine-glyoxylate aminotransferase-2 (AGXT2, which is mainly restricted to kidney).

Recent reports demonstrate a low-frequency coding polymorphism (V140I) in the AGXT2 gene alters urinary metabolite levels [2]. I have now demonstrated that AGXT2 plays a physiological role in ADMA metabolism by phenotyping AGXT2 null mice and quantifying the association between kidney tissue AGXT2 expression and circulating methylarginines in human transplant recipients (accounting for ~20% of ADMA variation). Furthermore I have shown disruption of the AGXT2 gene leads to hypertension in mice and, by association with V140I, diastolic blood pressure in humans [3]. The increased circulating ADMA observed with AGXT2 disruption may lead to hypertension by inhibiting production of endothelial-derived nitric oxide (NO) with consequent endothelial dysfunction.However local NO production has also been shown to inhibit renal sodium reabsorption along the nephron [4] and this is an alternative mechanism by which AGXT2 disruption might mediate hypertension.

As part of an intermediate fellowship application, which includes a number of epidemiological, human physiological and animal studies, I plan to examine the mechanisms underlying AGXT2 mediated hypertension. Specifically I hope to examine in collaboration with Professor John Deanfield the association between V140I and flow-mediated dilatation in humans to investigate if AGXT2 disruption leads to hypertension through endothelial dysfunction in humans.

Hypothesis: Functional variants in methylarginine metabolising enzymes are associated with impaired endothelial function

Exposure: Common variation in the genes encoding AGXT2 (and DDAH1) at loci previously reported to associate with an intermediate phenotype e.g. rs37369 (AGXT2 - V140I).

Outcomes: Per cent flow mediated dilatation (%FMD).

Adjustment for confunding by ethnicity (reported ethnic origin).

Power calculations: There will be 90% (alpha=0.05)power to detect an absolute difference of 0.5% in percent FMD (mean FMD: 8%; SD:4%) in those carrying the minor alle (assuming a MAF 0.1) at rs37369.

The opportunity to investigate the mechanism underlying the ADMA-CVD relationship using genetic tools will help us to establish the causal relationships in this area. Results of these studies may lead to the development of biomarkers for risk prediction as well as the potential to develop strategies to target this pathway.

1. Aucella, F., et al., Methylarginines and mortality in patients with end stage renal disease: A prospective cohort study. Atherosclerosis, 2009. 207(2): p. 541-545.

2. Suhre, K., et al., A genome-wide association study of metabolic traits in human urine. Nature Genetics, 2011. 43(6): p. 565-9.

3. Caplin, B., et al., Alaninie-Glyoxylate Aminotransferase-2 Metabolises Endogenous Methylarginines, Regulates Nitric Oxide and Controls Blood Pressure. . Arterioscler Thromb Vasc Biol, In Revision.

4. Garvin, J.L., M. Herrera, and P.A. Ortiz, Regulation of renal NaCl transport by nitric oxide, endothelin, and ATP: clinical implications. Annual review of physiology, 2011. 73: p. 359-76.

Background:

Pregnancy represents a metabolic challenge to women - in a normal pregnancy, a woman will become relatively insulin resistant, hyperlipidaemic, and have an up-regulation of coagulation factors and the inflammatory cascade (Sattar) in order to support the pregnancy. Most (Green et al.; Ness et al.; Parikh), but not all (Steenland) studies have found a positive association between parity and CVD in later life, suggesting that the more pregnancies a women has in her lifetime, the greater her risk of cardiovascular disease. However the mechanisms underlying this association are poorly understood. One possibility is that the stress test of pregnancy has long term effects and indeed there is some evidence to support this.

To our knowledge, no study has examined the association of inter-pregnancy interval and later cardiovascular health. It is possible that exposure to multiple pregnancies within a short time frame may be a risk factor for later cardiovascular health, since the woman's metabolic system would have less time to recover to its usual state between pregnancies.

Shorter inter-pregnancy interval may also be associated with greater gestational weight gain, which in ALSPAC has been shown to be associated with adverse outcomes in both the mother and offspring (Fraser 2010; Fraser 2011).

Aims:

To examine the association of inter-pregnancy interval with i) gestational weight gain, and ii) ii) cIMT, arterial stiffness and cardiovascular risk factors measured approximately 17 years postpartum in the mothers of the Avon Longitudinal Study of Parents and Children (ALSPAC).

Exposure variable:

Inter-pregnancy interval (analysis restricted to women who have 2 offspring in order to remove confounding by parity)

Outcome variables:

1. Gestational weight gain - as modelled using linear spline multilevel models by Kate Tilling and used in previous publications (Fraser 2010; Fraser 2011). Since these data are only available for the index child of ALSPAC, these analysis will be restricted to women where the ALSPAC index child is their second child, and they have no further children. The association between inter-pregnancy interval and gestational weight gain will be assessed by fitting an interaction between a categorised indicator of inter-pregnancy interval (short, medium (reference), long) and gestational age in the multilevel model of gestational weight gain

2. ii) cIMT, arterial stiffness and Cardiovascular risk factors measured at the 'focus on mothers' clinic 18 years postpartum. Detailed measures to be examined are the calculated 10-year risk of CVD based on the Framingham risk score, cIMT, arterial stiffness, BMI, waist circumference, systolic and diastolic blood pressure, glucose, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, insulin, pro-insulin, triglycerides, and C-reactive protein. Linear or logistic regression will be used to assess the association between interpregnancy interval and each outcome.

Confounding variables:

Maternal age at delivery and parity were obtained from obstetric records. Information on further pregnancies was obtained from various questionnaires. Information on prepregnancy weight and height, maternal smoking in pregnancy, maternal education, and household social class was based on questionnaire responses. Maternal education was categorized as below or above university level. The highest parental occupation was used to allocate the children to family social class groups (classes I [professional/managerial] to V [unskilled manual workers], according to the 1991 British Office of Population and Census Statistics classification). Maternal smoking in pregnancy was categorized as follows: never smoked; smoked before pregnancy or in the first trimester and then stopped; and smoked throughout pregnancy.

Additional relevant variables:

Information on diabetes mellitus and CVD diagnosed during follow-up was collected by a questionnaire completed 18 years after the index pregnancy. Women reported having been told they had a heart attack, heart failure, angina, and/or stroke.

References:

Fraser A, Tilling K, Macdonald-Wallis C, Sattar N, Brion MJ, Benfield L, Ness A, Deanfield J, Hingorani A, Nelson SM, Smith GD, Lawlor DA (2010). Association of maternal weight gain in pregnancy with offspring obesity and metabolic and vascular traits in childhood. Circulation;121:2557-64

Fraser A, Tilling K, Macdonald-Wallis C, Hughes R, Sattar N, Nelson SM, Lawlor DA. (2011) Associations of gestational weight gain with maternal body mass index, waist circumference, and blood pressure measured 16 y after pregnancy: the Avon Longitudinal Study of Parents and Children (ALSPAC); Am J Clin Nutr. 93: 1285-92

Green A, Beral V, Moser K (1988) Mortality in women in relation to their childbearing history. BMJ; 297: 391-395.

Ness RB, Harris T, Cobb J, Flegal KM, Kelsey JL, et al. (1993) Number of pregnancies and the subsequent risk of cardiovascular disease. N Engl J Med; 328: 1528-1533.

Parikh NI, Cnattingius S, Dickman PW, Mittleman MA, Ludvigsson JF, et al. (2010) Parity and risk of later-life maternal cardiovascular disease. American Heart Journal; 159: 215-221

Sattar, N. (2004) Do pregnancy complications and CVD share common antecedents? Atherosclerosis Supplements; 5: 3-7

Steenland K, Lally C, Thun M (1996) Parity and coronary heart disease among women in the American Cancer Society CPS II population. Epidemiology; 7: 641-643.

Aims: Few studies have investigated the aetiology of chronic disabling fatigue in children. Our aim is to investigate whether chronic disabling fatigue in 16 year-old children in the ALSPAC cohort is predicted by a range of factors, including maternal psychopathology, childhood psychological problems, exercise, and obesity.

Hypotheses: Maternal and childhood psychological, environmental and lifestyle factors may be associated with chronic disabling fatigue at age 16-17 years.

Exposure variables: episodes of maternal anxiety (Crown-Crisp Experiential Index anxiety subscale score greater than 8) and depression (Edinburgh Depression Scale score greater than 12) occurring at up to multiple time points since pregnancy; childhood psychological problems (derived from the 'Development and Well-Being Assessment' (DAWBA) questionnaire) at age 15 years; body mass index; and physical activity (questionnaire and accelerometer data); chronic fatigue at at 13-14 years (defined as fatigue that had prevented school attendance or participation in activities and that was not due to sport, possible sleep apnoea or probable depression [Crawley E, Hughes R, Northstone K, Tilling K, Emond A, Sterne JA. Pediatrics. 2012;130(1):e71-9]).

Outcome variables: We will identify ALSPAC 16 year-olds with high levels of fatigue using an appropriate threshold on the Chalder Fatigue Scale. This questionnaire was included in TF3. The threshold will be defined by looking at Chalder Fatigue scores in children diagnosed with CFS/ME according to National Insititute for Health & Clinical Excellence (NICE) diagnostic criteria using data from a cohort of paediatric patients which is maintained in the Centre for Child & Adolescent Health.

Confounding variables: family adversity

Aims:

To use GWAS data to identify genetic variation that influences maternal iodine status, as measured by urinary iodine-to-creatinine ratio. We would then aim to extrapolate this to evaluate if there is an interaction between iodine status, SNPs in relevant pathways and child outcome data.

We have hypothesised that our observed association between maternal iodine status and child cognition is driven by maternal thyroid hormones, which are required for fetal brain and neurological development, and a sufficient iodine supply is required for thyroid hormone production. However, we did not have maternal thyroid measures in the pilot study and therefore wish to confirm the relationships between iodine and thyroid function.

We wish to investigate the relationships between maternal iodine and selenium status. We will then include selenium in the model with iodine status and child cognition to evaluate if maternal selenium influences the obvserved relationships in the pilot study.

We also aim to use ALSPAC dietary data and iodine status measures in pregnancy to develop a screening tool that could be used in a clinical setting to identify women who are at risk of iodine deficiency. The dietary data in ALSPAC would be analysed to establish dietary patterns that protect women from iodine deficiency and thus provide the data for a risk screening tool. This screening tool will be developed in combination with data from another UK study (SPRINT - a selenium intervention study that has baseline measures of iodine status and thyroid function), allowing the tool to be tested in different groups and populations.

Regulation of gene expression is a highly heritable, critical component of a variety of biological processes. Studies integrating expression data and genotypic data have successfully identified common-variant expression quantitative trait loci (eQTLs) in a range of cell and tissues samples. As the majority of common genetic variants associated with complex traits map to non-coding regions of the genome, genome-wide studies of the genetic regulation of gene expression are key to interpreting the wealth of data generated in association studies and to understanding the genetic architecture of gene regulation, and by extension, the genetic architecture of complex traits.

We propose to fine-map common variant and identify rare variant eQTLs by integrating UK10K whole genome sequences from the TwinsUK and ALSPAC cohorts with gene expression data generated from the same individuals as part of the MuTHER study (TwinsUK) and ALSPAC expression study. Gene expression data is available from three tissues in the MuTHER individuals - lymphoblastoid cell lines (LCLs), adipose and skin, and LCLs in the ALSPAC individuals. In each case roughly half of the individuals have whole genome sequences generated as part of the cohorts arm of UK10K and the remaining will have imputed genotypes based on the UK10K-reference panel. Both the ALSPAC and TwinsUK expression profiles were generated on the Illumina HT-12 array in the same facility, making them ideally matched for joint analyses. We are applying separately to the UK10K management committee for access to the ALSPAC genome sequences. The proposed research will generate a novel resource of fine-scale regulatory data and inform our understanding of the genetic regulation of gene expression.

We will carry out global single point (MAF greater than 1%) and rare-variants burden scans to identify cis-eQTLs. We will condition on significant peaks to find secondary (and potentially tertiary signals). We will use the discovered cis-eQTLs to answer the following questions: A) How much more of the heritability of expression of a transcript is explained by rare variant or incompletely tagged common variants? B) Where do the cis-eQTLs map and can we use functional annotation to improve the sensitivity of our scans? C) Where we find a secondary eQTL signal under a primary one already linked to a GWAS signal is there evidence that the secondary eQTL also affects the GWAS trait? (We will focus on traits measured in the UK10K analysis of the TwinsUK/ALSPAC full cohort) In addition, we will fine map trans-eQTLs previously discovered with common variant genotypic data in these datasets.

The benefits of physical activity (PA) and cardio-respiratory fitness (CRF) for children's and adolescents' physical [1, 2] and psychological health [3] are well established. However, only 33% of British boys and 21% of British girls aged 4-15 years meet PA guidelines when PA is objectively assessed [4]. Meanwhile, British children spend on average between 3.4 - 4.1 hours in sedentary pursuits outside of their school time and the amount of time spent sedentary increases with age [4]. Sedentary behaviours such as TV viewing and passive travel negatively predict CRF [5, 6] and are associated with increased adiposity [6]. Data also indicates that British youth have become alarmingly unfit in the course of the last decade [7]. Inactive lifestyles not only pose a health concern during adolescent years, but also track relatively well into adulthood, thus increasing adult risk of morbidity and mortality [8].

The emerging body of evidence suggests that low levels of PA, CRF and fatness are not only associated with adverse physical health outcomes but negatively affect brain and cognitive development, as well as academic achievement in youth [9-12]. Conversely, CRF, short bouts of exercise and exercise interventions have been associated with superior cognitive performance [10, 13-15] and larger volumes in brain structures sub-serving memory and learning [13], as well as with higher academic achievement in youth and children [10, 16]. Investigating the relationships among PA, CRF, cognitive function and academic outcomes in school-aged children and youth is important, as these relationships can contribute to long term positive educational sequel [17].

A number of studies using neurophysiological and behavioural measures of cognition indicate that pre-adolescent children with higher levels of CRF perform significantly better on cognitive tasks requiring executive control (mental processes regulating goal directed behaviour, which include working memory, inhibition, and cognitive flexibility), attention and relational memory [13, 18-20]. Superior cognitive performance was linked to greater volumes in brain structures supporting learning and memory: (i) basal ganglia [19, 21, 22], and (ii) hippocampus [18], thus indicating structural physiological underpinnings for the effects of CRF on children's cognition. Working memory and inhibition in particular are thought to be important for academic achievement and are linked to achievement in mathematics and reading [23-25]. Likewise teachers' ratings of pupils' attention show a strong association with academic achievement in mathematics and reading [26-28]. However, studies employing objective cognitive measures of attention in relation to academic achievement are sparse [29, 30]. Likewise, only a paucity of studies investigated the effects of PA (exercise intervention) or CRF on children's attention and these results are promising [20, 31]. Thus the relationship between PA, CRF, attention and academic achievement warrants further investigation.

Evidence is emerging linking childhood and adolescent overweight and obesity to the declines in cognitive function and academic achievement [11, 12]. Research also suggests that exercise can positively affect this relationship and improve executive function and academic performance in overweight and sedentary children [32-34]. Thus it is possible that PA, CRF and adiposity interact in producing different cognitive and academic outcomes in youth. Further research investigating complex associations between PA, CRF, adiposity and cognition as well as academic achievement is warranted.

Studies to date focused on the effects of CRF [10] and broadly defined PA, including time spent in physical education (PE) [10, 35, 36], self-reported PA, or the effects of a particular exercise intervention as a proxy for PA [10, 15, 35]. Such methodological approaches taken together with the heterogeneity of PA measures and definitions do not allow for any inferences about what patterns of habitual PA (intensity, frequency and duration, as well as the number of PA bouts) are most strongly associated with positive cognitive and academic outcomes in youth and children. Evidence linking objectively measured habitual PA to these outcomes is lacking [13, 14]. Furthermore, none of the studies to date assessed the unique effects of sedentary time on cognitive function and academic achievement in children or adolescents. Although few studies reported positive effects of short bouts of exercise during class on children's concentration and engagement with learning tasks [37, 38], these data do not inform our knowledge of how sedentary time independent of moderate to vigorous PA uniquely contribute to gains in cognitive function and, possibly, academic achievement. The knowledge of how patterns of habitual PA and sedentary time uniquely contribute to various aspects of adolescents' cognitive function in comparison to their effects on academic achievement is important. Such knowledge will afford mapping of complex pathways through which habitual PA can affect processes supporting learning as well as learning outcomes.

Present study proposes to address these gaps in research and to investigate:

1) the mediated effects of cognition (working memory, inhibition, selective attention, divided attention, attentional control) on the relationship between habitual PA (and sedentary behaviour) on academic achievement at age 11 years;

Research question (RQ) 1: Is the relationship between the total and accumulated bouts (e.g. greater than 10 minutes) of PA (and time sedentary) and academic achievement (in reading, spelling, English, mathematics and science) at age 11 mediated by cognitive performance on the tests of working memory and inhibition at age 10 years, and divided attention, selective attention and attentional control at age 11 years?

2) the longitudinal effects of habitual PA (and sedentary behaviour) at age 11 years on youth's cognition at age 15.5 years (inhibition), and academic achievement at 13 years (Key Stage 3), after controlling for academic achievement at age 11 years and habitual PA (and sedentary behaviour) at age 13 and 15.5 years;

RQ 2: What is the effect of the total and accumulated bouts (e.g. greater than 10 minutes) of PA (and time sedentary) at age 11 years on inhibition at age 15.5 years?

RQ3: What is the effect of the total and accumulated bouts (e.g. greater than 10 minutes) of PA (and time sedentary) at age 11 years on academic achievement (in reading, spelling, English, mathematics and science) at age 13 years?

3) and to test complex effects models to elucidate the mediating and moderating effects of CRF and adiposity, on the proposed relationships between habitual PA (and sedentary behaviour) and cognition (selective attention, divided attention, attentional control, working memory and inhibition).

RQ4: Does cardio-respiratory fitness at age 15.5 mediate or moderate the relationship between the total and accumulated bouts (e.g. greater than 10 minutes) of PA (and time sedentary) and cognition (inhibition) at age 15.5 years?

RQ5: Does adiposity at age 11 and 15.5 years mediate or moderate the relationship between the total and accumulated bouts (e.g. greater than 10 minutes) of PA (and time sedentary) and cognitive performance at age 10 years (working memory, inhibition), 11 years (selective attention, divided attention and attentional control) and 15.5 years (inhibition)?

PA and sedentary behaviour will be objectively measured by the Actigraph accelerometer (please refer to Appendix 2A for further details of the proposed use of the accelerometer data). The objective measures of components of executive function most consistently associated with academic outcomes (working memory, inhibition), as well as the measures of attention (selective attention, divided attention and attentional control) will be taken using validated computer and pen and paper tasks. Inhibition will be measured with Stop Signal Task [39] and working memory with a computerised Counting Span Task [40]. The Test of Everyday Attention for Children (TEACh) (adapted from the adult version by Robertson [41]) will be employed to test various aspects of attention. Scores from the Standard Assessment Tests (SATS) of the National Curriculum at Key Stages 2 and 3 (after controlling for child's results at Key Stage 1) in mathematics, reading, spelling, English and science will be used as outcome measures of academic achievement. Heart rate taken during an exercise bike session will be used as a proxy measure of cardio-respiratory fitness. Percent body fat will be assessed with the whole Dual X-Ray Absorptiometry (DXA) and BMI. BMI will be used to classify children into normal weight, overweight and obese categories in accordance with the criteria recommended by the Centre for Disease Control (CDC) [42] to test moderating effects model. Biological maturation has been related to body composition [43-47] as well as academic achievement [48, 49] in adolescent boys and girls, and its sequel affects maturation of brain structures involved in cognitive function [50]. Therefore biological maturation will be controlled for in all the analyses. Measures of sexual maturation (stage of pubic hair development in boys and girls, breast development and the onset of menses in girls, and genital development in boys), and somatic maturation (predicted age at peak height velocity (APHV)) will be used as maturity indicators.

There is strong evidence to suggest that very pre-term (<= 33 weeks gestation) and/or very low birth weight (VLBW; <= 1500 g) children have academic difficulties and more subtle cognitive impairments [51]. Recent meta-analysis of neurobehavioural outcomes in very preterm and VLBW children suggests that these effects are especially prominent for academic achievement in mathematics (d= -0.60) and spelling (d= - .76), with medium effect sizes reported for executive function (effect sizes range from

- 0.36 to - 0.57), and attention (effect sizes range from -0.43 to -0.59) [51]. Thus it is important to control for gestational age and weight at birth in the analyses relating to both adolescents' cognition and their academic performance.

Likewise, there is epidemiological evidence linking attention-deficit/hyperactivity disorder (AD/HD) and depression to the declines in cognitive [52, 53] and academic performance in youth [54-57]. Therefore, AD/HD and depression will be controlled for in the analyses based on the scores from the Development and Well-being Assessment (DAWBA, [58]), and, at time points when these are not available, by the hyperactivity score from Strength and Difficulties Questionnaire (SDQ) [59, 60] and a computerised depression questionnaire. Cases with IQ below 85, special educational needs, learning disabilities and physical disabilities will be excluded from the analyses. Please refer to appendix 2C for the exhaustive list of exposure, outcome, and confounding variables, as well as for mediators/moderators and exclusion criteria.

Statistical analyses:

Statistical analyses will include descriptive statistics, Pearson product moment correlations, and the assessments of mutlicollinearity. Exploratory multiple regression models including all the confounders (age at test, sex, SES, IQ, handedness, gestational age, birth weight, biological maturation, AD/HD, depression, accelerometer wear, adiposity and CRF) and predictors (frequency and duration of light, moderate, moderate to vigorous and vigorous PA, as well as the number of PA bouts, duration and frequency of sedentary time and sedentary bouts) will be first assessed to ensure that the direct relationships exist among PA, cognitive and academic outcomes. To test the mediating effects model (RQ1) the subtracting method as outlined by Judd and Kenny [61] will be employed. This approach allows for testing the significance of the indirect effect as opposed to Barron and Kenny's [62] method. The test of statistical significance for the indirect effect will be based on the method designed by McKinnon et al. [63] and outlined in [64]. To answer research questions 2 and 3, mixed effects models for longitudinal data analysis will be employed [65]. To address research questions 4 and 5, the complex moderating and mediating effects models including CRF and adiposity will be assessed by pathway analyses with Structural Equation Modelling [53].

Cited literature

1. Gutin, B. and S. Owens, The influence of physical activity on cardiometabolic biomarkers in youths: A review. Pediatric Exercise Science, 2011. 23(2): p. 169-185.

2. Janssen, I. and A.G. LeBlanc, Systematic review of the health benefits of physical activity and fitness in school-aged children and youth. International Journal of Behavioral Nutrition and Physical Activity, 2010. 7: p. 40.

3. Biddle, S.J.H. and M. Asare, Physical activity and mental health in children and adolescents: a review of reviews. British Journal of Sports Medicine, 2011. 45(11): p. 886-895.

4. Aresu, M., et al., Health Survey for England 2008: Physical activity and fitness, R. Craig, J. Mindell, and V. Hirani, Editors. 2009, Joint Health Surveys Unit, National Centre for Social Research, Department of Epidemiology and Public Health, UCL Medical School: London.

5. Sandercock, G.R.H. and A.A. Ogunleye, Screen time and passive school travel as independent predictors of cardiorespiratory fitness in youth. Preventive Medicine, 2012. 54(5): p. 319-322.

6. Tremblay, M., et al., Systematic review of sedentary behaviour and health indicators in school-aged children and youth. International Journal of Behavioral Nutrition and Physical Activity, 2011. 8(1): p. 98.

7. Sandercock, G., et al., Ten year secular declines in the cardiorespiratory fitness of affluent English children are largely independent of changes in body mass index. Archives of Disease in Childhood, 2010. 95(1): p. 46-47.

8. Hallal, P.C., et al., Adolescent physical activity and health: A systematic review. Sports Medicine, 2006. 36(12): p. 1019-1030.

9. Davis, C.L. and S. Cooper, Fitness, fatness, cognition, behavior, and academic achievement among overweight children: Do cross-sectional associations correspond to exercise trial outcomes? Preventive Medicine, 2011. 52, Supplement(0): p. S65-S69.

10. Fedewa, A.L. and S. Ahn, The effects of physical activity and physical fitness on children's achievement and cognitive outcomes: A meta-analysis. Research Quarterly for Exercise and Sport, 2011. 82(3): p. 521-535.

11. Li, Y., et al., Overweight is associated with decreased cognitive functioning among school-age children and adolescents. Obesity, 2008. 16(8): p. 1809-1815.

12. Shore, S.M., et al., Decreased scholastic achievement in overweight middle school students. Obesity, 2008. 16(7): p. 1535-1538.

13. Chaddock, L., et al., A review of the relation of aerobic fitness and physical activity to brain structure and function in hildren. Journal of the International Neuropsychological Society, 2011. 17(6): p. 975-985.

14. Hillman, C.H., K. Kamijo, and M. Scudder, A review of chronic and acute physical activity participation on neuroelectric measures of brain health and cognition during childhood. Preventive Medicine, 2011. 52: p. S21-S28.

15. Tomporowski, P.D., K. Lambourne, and M.S. Okumura, Physical activity interventions and children's mental function: An introduction and overview. Preventive Medicine, 2011. 52, Supplement(0): p. S3-S9.

16. Singh, A.P., et al., Physical activity and performance at school: A systematic review of the literature including a methodological quality assessment. Archives of Pediatrics & Adolescent Medicine, 2012. 166(1): p. 49-55.

17. Koivusilta, L.K., H. Nupponen, and A.H. Rimpela, Adolescent physical activity predicts high education and socio-economic position in adulthood. The European Journal of Public Health, 2011.

18. Chaddock, L., et al., A neuroimaging investigation of the association between aerobic fitness, hippocampal volume, and memory performance in preadolescent children. Brain Research, 2010. 1358(0): p. 172-183.

19. Chaddock, L., et al., Basal ganglia volume is associated with aerobic fitness in preadolescent children. Developmental Neuroscience, 2010. 32(3): p. 249-256.

20. Hillman, C.H., D.M. Castelli, and S.M. Buck, Aerobic fitness and neurocognitive function in healthy preadolescent children. Medicine and Science in Sports and Exercise, 2005. 37(11): p. 1967-1974.

21. Chaddock, L., et al., A functional MRI investigation of the association between childhood aerobic fitness and neurocognitive control. Biological Psychology, 2012. 89(1): p. 260-268.

22. Chaddock, L., et al., Childhood aerobic fitness predicts cognitive performance one year later. Journal of Sports Sciences, 2012. 30(5): p. 421-430.

23. Bull, R. and G. Scerif, Executive functioning as a predictor of children's mathematics ability: Inhibition, switching, and working memory. Developmental Neuropsychology, 2001. 19(3): p. 273-293.

24. Latzman, R.D., et al., The contribution of executive functioning to academic achievement among male adolescents. Journal of Clinical and Experimental Neuropsychology, 2010. 32(5): p. 455-462.

25. St Clair-Thompson, H.L. and S.E. Gathercole, Executive functions and achievements in school: Shifting, updating, inhibition, and working memory. The Quarterly Journal of Experimental Psychology, 2006. 59(4): p. 745-759.

26. Breslau, N., et al., Change in teachers' ratings of attention problems and subsequent change in academic achievement: a prospective analysis. Psychological Medicine, 2010. 40(01): p. 159-166.

27. Polderman, T.J., et al., A genetic study on attention problems and academic skills: Results of a longitudinal study in twins. Journal of the Canadian Academy of Child and Adolescent Psychiatry / Journal de l'Academie canadienne de psychiatrie de l'enfant et de l'adolescent, 2011. 20(1): p. 22-34.

28. Rabiner, D.L., et al., An exploration of the relationship between ethnicity, attention problems, and academic achievement. School Psychology Review, 2004. 33(4): p. 498-509.

29. Mayes, S.D., et al., IQ and neuropsychological predictors of academic achievement. Learning and Individual Differences, 2009. 19(2): p. 238-241.

30. Trautmann, M. and F.D. Zepf, Attentional performance, age and scholastic achievement in healthy children. PLoS ONE, 2012. 7(3).

31. Hill, L., et al., Exercising attention within the classroom. Developmental Medicine & Child Neurology, 2010. 52(10): p. 929-934.

32. Davis, C.L., et al., Effects of aerobic exercise on overweight children's cognitive functioning: A randomized controlled trial. Research Quarterly for Exercise and Sport, 2007. 78(5): p. 510-519.

33. Davis, C.L., et al., Exercise improves executive function and achievement and alters brain activation in overweight children: A randomized, controlled trial. Health Psychology, 2011. 30(1): p. 91-98.

34. O'Malley, G., Aerobic exercise enhances executive function and academic achievement in sedentary, overweight children aged 7-11 years. Journal of Physiotherapy, 2011. 57(4): p. 255.

35. Martin, K., Brain boost: Sport and physical activity enhance children's learning, 2010, The University of Western Australia. p. 16.

36. Rasberry, C.N., et al., The association between school-based physical activity, including physical education, and academic performance: A systematic review of the literature. Preventive Medicine, 2011. 52, Supplement(0): p. S10-S20.

37. Mahar, M.T., Impact of short bouts of physical activity on attention-to-task in elementary school children. Preventive Medicine, 2011. 52, Supplement(0): p. S60-S64.

38. Mahar, M.T., et al., Effects of a classroom-based program on physical activity and on-task behavior. Medicine & Science in Sports & Exercise, 2006. 38(12): p. 2086-2094.

39. Logan, G.D., W.B. Cowan, and K.A. Davis, On the ability to inhibit simple and choice reaction time responses: A model and a method. Journal of Experimental Psychology: Human Perception & Performance, 1984. 10(2): p. 276-291.

40. Case, R., D.M. Kurland, and J. Goldberg, Operational efficiency and the growth of short-term memory span. Journal of Experimental Child Psychology, 1982. 33(3): p. 386-404.

41. Robertson, I.H., et al., The structure of normal human attention: The Test of Everyday Attention. Journal of the International Neuropsychological Society, 1996. 2(06): p. 525-534.

42. Barlow, S.E. and a.t.E. Committee, Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: Summary report. Pediatrics, 2007. 120(Supplement 4): p. S164-S192.

43. Beunen, G.P., et al., Adiposity and biological maturity in girls 6-16 years of age. International Journal of Obesity and Related Metabolic Disorders, 1994. 18(8): p. 542-546.

44. Bratberg, G., et al., Early sexual maturation, central adiposity and subsequent overweight in late adolescence. A four-year follow-up of 1605 adolescent Norwegian boys and girls: the Young HUNT study. BMC Public Health, 2007. 7(1): p. 54.

45. Demerath, E.W., et al., Fifty-year trends in serial body mass index during adolescence in girls: the Fels Longitudinal Study. The American Journal of Clinical Nutrition, 2004. 80(2): p. 441-446.

46. Malina, R.M., S. Koziel, and T. Bielicki, Variation in subcutaneous adipose tissue distribution associated with age, sex, and maturation. American Journal of Human Biology, 1999. 11(2): p. 189-200.

47. Beunen, G., et al., Fatness and skeletal maturity of Belgian boys 12 through 17 years of age. American Journal of Physical Anthropology, 1982. 59(4): p. 387-392.

48. Mendle, J. and J. Ferrero, Detrimental psychological outcomes associated with pubertal timing in adolescent boys. Developmental Review, 2012. 32(1): p. 49-66.

49. Cavanagh, S.E., C. Riegle-Crumb, and R. Crosnoe, Puberty and the education of girls. Social Psychology Quarterly, 2007. 70(2): p. 186-198.

50. Blakemore, S.J. and S. Choudhury, Development of the adolescent brain: implications for executive function and social cognition. Journal of Child Psychology and Psychiatry, 2006. 47(3-4): p. 296-312.

51. Aarnoudse-Moens, C.S.H., et al., Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children. Pediatrics, 2009. 124(2): p. 717-728.

52. Han, G., et al., Selective neurocognitive impairments in adolescents with major depressive disorder. Journal of Adolescence, 2012. 35(1): p. 11-20.

53. Nigg, J.T., Neuropsychologic theory and findings in attention-deficit/hyperactivity disorder: The state of the field and salient challenges for the coming decade. Biological Psychiatry, 2005. 57(11): p. 1424-1435.

54. Jaycox, L.H., et al., Impact of teen depression on academic, social, and physical functioning. Pediatrics, 2009. 124(4): p. e596-e605.

55. Nelson, J.R.G.J.L.K.B.W., Academic achievement of K-12 students with emotional and behavioral disorders. Exceptional Children, 2004. 71(1): p. 59-73.

56. Rapport, M.D., et al., Internalizing behavior problems and scholastic achievement in children: cognitive and behavioral pathways as mediators of outcome. Journal Of Clinical Child Psychology, 2001. 30(4): p. 536-551.

57. Rothon, C., et al., The impact of psychological distress on the educational achievement of adolescents at the end of compulsory education. Social Psychiatry And Psychiatric Epidemiology, 2009. 44(5): p. 421-427.

58. Goodman, R., et al., The development and well-being assessment: Description and initial validation of an integrated assessment of child and adolescent psychopathology. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 2000. 41(05): p. 645-655.

59. Goodman, R., The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry, 1997. 38(5): p. 581-586.

60. Goodman, R., The extended version of the strengths and difficulties questionnaire as a guide to child psychiatric caseness and consequent burden. Journal of Child Psychology and Psychiatry, 1999. 40(5): p. 791-799.

61. Judd, C.M. and D.A. Kenny, Process analysis. Evaluation Review, 1981. 5(5): p. 602-619.

62. Baron, R.M. and D.A. Kenny, The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality & Social Psychology, 1986. 51(6): p. 1173-1182.

63. MacKinnon, D.P., C. Lockwood, and J. Hoffman, A new method to test for mediation. In: The annual meeting of the Society for Prevention Research1998: Park City, UT.

64. MacKinnon, D.P., et al., A Comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 2002. 7(1): p. 83-104.

65. Thorpe, J.I., Joint mixed-effects models for longitudinal data analysis: An application for the metabolic syndrome. In: Biostatistics 2009, Virginia Commonwealth University: Richmond.