Association between fetal growth and ADHD. Low birth weight (less than 2500 grams) is a well-established risk factor for ADHD (Nigg et al., 2010). Across prospective case-control studies LBW cases are at approximately 2-times increased risk to developing ADHD or clinically significant ADHD symptoms (e.g., Breslau et al., 1996). Other fetal growth phenotypes such as small for gestational age, ponderal index, and head circumference at birth are also associated with ADHD symptomtatology (Indredavik et al., 2004; Lahti et al., 2006; Heinonen et al., 2011). The relationship between reduced fetal growth and prematurity exists after controlling for a variety of confounding child factors including: child sex, season of birth or duration of breast feeding (Elgen, Sommerfelt, & Markestad, 2003; Horwood, Mogridge, & Darlow, 1998); or by parental factors including: marital status, age, maternal education, paternal education, maternal stress, parental psychopathology, substance abuse, maternal smoking, or parental nurturance (Breslau et al., 1996; Elgen et al., 2003; Horwood et al., 1998; Indredavik et al., 2004; Linnet et al., 2006; Zubrick et al., 2000). Furthermore, evidence suggests that there is a dose-response relationship between birth weight and child behavioral problems (van Os et al., 2001; Wichers et al., 2002; see Wilens et al., 2006 for null findings), including ADHD (Boulet et al., 2009). Given the association between fetal growth and ADHD has been examined primarily in prospective case-control studies, it is necessary to utilize population birth cohorts to further understand the nature of the relationship across the full spectrum of fetal growth and ADHD symptom severity.

Fetal growth and ADHD neurodevelopmental risk. Growth restricted infants have reductions in overall white and gray matter compared to normally grown infants (Brown et al., 2009; Larroque et al., 2003; Tolsa et al., 2004). Such reductions relate to poorer performance on early measures of attention, negative neurodevelopmental outcomes (Peterson et al., 2003; Tolsa et al., 2004) and are consistent with findings in ADHD samples (Filipek et al., 1997; Kates et al., 2002; Mostofsky et al., 2002; Overmeyer et al., 2001). In addition to between group findings, neuroanatomical abnormalities, such as parenchymal lesions or ventricular enlargement, predict increased risk for ADHD within a LBW cohort (Whitaker et al., 1997; Whitaker et al., 2011). Together, these findings suggest that youth who have experienced restricted fetal growth, tend to display neurodevelopmental abnormalities which are functionally related to ADHD symptomatology and consistent with findings in ADHD samples.

Restricted fetal growth as a proxy for an ADHD environmental pathogen(s). Evidence from twin studies suggest that environmental factors that underlie fetal growth, increase vulnerability for ADHD and that this relationship is not confounded by genetic factors (Groen-Blokhuis et al., 2011). This suggests that restricted fetal growth is a proxy for a constellation of prenatal environmental risk factors which restrict fetal growth and are implicated in the complex pathophysiology of ADHD. Genetic factors, however, cannot be discounted and likely play an important role in modifying the relationship between fetal growth and vulnerability for ADHD.

Prenatal ischemia-hypoxia as a possible common pathway linking prenatal environmental risk with ADHD. Suboptimal maternal-placental-fetal nutrient transport is a main determinant of restricted fetal growth (Ghidini, 1996). Thus, prenatal ischemia-hypoxia is believed to underlie restricted fetal growth in the majority of cases, especially within well-nourished populations (Henriksen & Clausen, 2002). This suggests that prenatal ischemia-hypoxia may represent a common prenatal pathway which alters neurodevelopment (Schimdt-Kastner et al., 2012) and subsequent vulnerability for ADHD. For example, both prospective and retrospective case-control studies have shown that ADHD is associated with ischemia-hypoxia related obstetric complications (Pineda et al., 2007; Getahun et al., 2013). Furthermore, low neonatal cerebral blood flow is associated with greater dopamine receptor availability and increased ADHD-related executive functioning deficits (Lou et al., 2004). These findings suggest that prenatal ischemia-hypoxia and related neurodevelopmental and cerebral vascular sequelae may represent an early developmental pathway to ADHD.

Developmental Origins of Health and Disease Hypothesis. The Developmental Origins of Health and Disease hypothesis (DOHaD; Gluckman & Hanson, 2004) provides a framework to conceptualize how genetic and early environmental factors interact to confer vulnerability for ADHD. For example, neurodevelopmental delays or neurodevelopmental disruptions, which covary with restricted fetal growth, may result from a limited supply of nutrients and oxygen in utero. Within individuals exposed to prenatal ischemia-hypoxia vulnerability for ADHD may then be moderated by fetal and maternal genotype. Aberrant epigenetic regulation of gene expression is believed to mediate the relationship between gene x environment interaction with an observed phenotype. Therefore, to further understand developmental mechanisms linking prenatal ischemia-hypoxia with increased vulnerability for ADHD, it is essential to examine variation in both the epigenome and trascriptome.

In our previous research, we have showed that fetal growth moderates the relationship between genetic variants in ischemia-hypoxia regulated (IHR) neurotrophic and kynurenine genes and ADHD symptom severity (Smith et al., In preparation). This study included a limited number of candidate genes that are regulated by prenatal ischemia-hypoxia (Schmidt-Kastner et al., 2012). Therefore, there is a need to more comprehensively examine how IHR genes alter vulnerability for ADHD within a population birth cohort. Therefore, this project aims to leverage the size and scope of the ALSPAC database, along with an IHR gene database curated by Dr. Schmidt-Kastner (see Schmidt-Kastner, 2012) to examine the interplay between fetal growth, IHR genetic variation, and vulnerability for ADHD. The IHR gene database will include 388 genes that: 1) are up- or down-regulated in response to prenatal ischemia-hypoxia in microarray studies of the rodent brain; and 2) are either regulated by hypoxia inducible factor 1 or 2 (HIF-1; HIF-2 ) which regulate the cellular response to hypoxia or have a vascular function.

Summary. Although restricted fetal growth is a well-established risk factor for ADHD (Nigg et al., 2010), little is known about biological mechanisms that underlie the relationship between restricted fetal growth and ADHD. Therefore, the current proposal aims to explore the role of ischemia-hypoxia pathway (as indexed by IHR genetic and epigenetic variation) as an underlying mechanism linking restricted fetal growth and vulnerability for ADHD.

Aim 1. Estimate theproportion of genetic covariance between fetal growth and ADHD symptom severity that can be accounted for by genome-wide SNPs and separately within IHR SNPs using genome-wide complex trait analysis (GCTA; Lee et al., 2012).

Aim1.1. Examine the relationships betweenmeasured fetal and maternal SNPs within IHR genes on ADHD symptom severity, as well as pleiotropic relationships between IHR SNPs on both fetal growth and ADHD symptom severity (Hartley et al., 2012).

Maternal and fetal SNPs that have pleiotropic associations with ADHD symptom severity and fetal will inform gene-environment interaction models in Aim 2.

Aim 2. Model the interactive effects of IHR genetic variants (identified in Aim 1.1) and fetal growth on child ADHD symptoms.

Aim 3. Characterize the role of DNA methylation of CpG sites at birth within child IHR genes (identified in Aim 1.1) in mediating the relationship between fetal growth and ADHD symptom severity.

Aim 4. Examine the interrelationships between fetal growth, methylation profiles of IHR genes, ischemia-hypoxia gene expression and ADHD symptom severity.

Overall Hypothesis: These aims will examine the overall hypothesis that IHR genetic and epigenetic variability will partially explain the relationship between restricted fetal growth and increased ADHD symptom severity.

Exposure Variables: Fetal growth based on gestational age (birthweight, head circumference, crown-heel length) and ischemia-hypoxia related obstetric complications may be examined on a secondary basis (e.g., preeclampsia).

Outcome Variables: ADHD symptom severity is the primary outcome measure. From the ALSPAC site, ADHD (via parent report) is assessed multiple times with the DAWBA (ages 7, 10, 14) and SDQ (ages4, 7, 8, 10, 12, 13, and 17) providing a longitudinal measure of the primary outcome.

Genetic Determinants: Maternal and Child genome-wide data to be used in GCTA analysis. Child DNA Methylation Data at birth and 7 or 9 years of age. SNPs and CpG islands within IHR genes will be utilized in Aims 2 and 3. RNA expression data of IHR genes will be used in Aim 4. NOTE: IHR gene database (in consultation with Dr. Schmidt-Kastner) and results from Aim 1 analysis will be used to guide analyses in Aim 2, 3, and 4.

Confounding Variables: Placental weight, maternal blood pressure, hypertension during pregnancy, maternal smoking during pregnancy, prenatal alcohol use, socioeconomic status and pregnancy, delivery and neonatal health/complications.