Neurogenetic disorders with distinct genetic etiologies and known psychiatric phenotypes present a compelling model to study the links between genotype and phenotype. In recent years, the study of quantitative endophenotypes has become increasingly relevant in garnering a comprehensive understanding of psychiatric dysfunction. Neurofibromatosis Type 1 (NF1) and 22q11.2 Deletion Syndrome (22q11DS) are two compelling examples of neurogenetic disorders that are highly penetrant for neuropsychiatric phenotypes, and are also characterized by prefrontal cognitive dysfunction. We investigated endophenotypes of executive control in these populations, using a reverse-genetic approach.

In Chapter 2, we measured risk-taking behavior using a child-friendly gambling task in patients with NF1 and controls (N=29 NF1, 23 controls). We used functional magnetic resonance imaging (fMRI) to investigate neural activity associated with risky decision making, as well as age-associated changes in these behavioral and neural processes. Behaviorally, patients with NF1 tended to made fewer risky decisions than controls. Neuroimaging analyses revealed significant hypoactivation of multiple brain regions involved in higher-order semantic processing and motivation (i.e., anterior cingulate, paracingulate, supramarginal, and angular gyri) in patients with NF1 relative to controls, in both decision-making and outcome phases of the task. We also observed atypical age-associated changes in neural activity in patients with NF1, such that during risk taking, neural activity tended to decrease with age in controls, whereas it tended to increase with age in patients with NF1. Findings suggest that developmental trajectories of neural activity during risky decision-making may be disrupted in youth with NF1.

In Chapter 3, we reviewed recent literature on the utility of studying endophenotypes of psychiatric illness in 22q11DS. We provided an overview of neuropsychiatric findings to date, which highlight the value of this syndrome in mapping the developmental trajectory of dimensional phenotypes that traverse multiple diagnostic categories. Potential sources of genetic variability that may contribute to the disorder’s heterogeneous presentation were reviewed. We discussed the use of how animal models can readily be developed that recapitulate specific aspects of the syndrome. Future research directions involve translational models and potential for drug screenable targets in the context of this human model system.

In Chapter 4, we investigated executive function and its relationship with structural neuroanatomy in patients with 22q11DS and matched controls (N=43 22q11DS, 43 controls), along with cognitive measures that tap behavioral regulation and metacognitive aspects of executive function. Behaviorally, patients with 22q11DS were impaired on multiple executive function measures. Right orbitofrontal cortical thickness showed a differential relationship between real-world executive function in patients with 22q11DS and controls. We also observed a group difference in the relationship between behavioral regulation and metacognition measures with thickness of ventral and dorsolateral prefrontal regions, respectively. Findings suggest that executive dysfunction characteristic of 22q11DS is underscored by altered prefrontal cortical structure.

In Chapter 5, we investigated gene expression levels and behavioral correlates in patients with 22q11DS and matched controls (N=56 22q11DS, 48 controls). Using real-time quantitative polymerase chain reaction (RT-qPCR), we measured gene expression levels of three genes that are hemizygous in 22q11DS (COMT, DGCR8, and ZDHHC8). We found decreased expression of all three genes in patients with 22q11DS. We found a positive relationship between age and COMT expression in patients with 22q11DS, but not in controls. We also investigated the relationship between gene expression levels and relevant cognitive measures related to executive function, and found interaction effects between group and IQ for DGCR8 and ZDHHC8. Lastly, we found a relationship between working memory and COMT expression in patients with 22q11DS, but not in controls. Results suggest that the relationships between gene expression and cognitive function may differ between patients with 22q11DS and controls.

Overall, the collected work provides insight into the use of endophenotypes of executive control in neurogenetic disorders. In clinical psychiatry, current methods for diagnosing psychiatric illness rely on clinical symptoms. Our research is in line with the NIMH RDoC framework, which supports the notion of studying intermediate endophenotypes in order to better diagnose and treat patients with psychiatric illness. Having a comprehensive understanding of the building blocks of psychiatric disease can lead not only to better methods of diagnosis, but also to better treatment strategies.

Coordinated, large-scale efforts to study brain disorders are rare, with most neuroimaging studies collecting fewer than 100 study participants. Neuroimaging research is one of many biomedical fields that have suffered from low study power and reproducibility. Genetic variation and psychiatric disorders confer subtle effects on markers of brain structure and function, further complicating the discovery and validation of reproducible neuroimaging biomarkers.

The Enhancing Neuro Imaging Genetics through Meta Analysis Consortium (ENIGMA) has developed harmonized processing and analysis protocols to empower large-scale neuroimaging studies of genetic variation and psychiatric illness. By using standardized methods, prospective meta- and mega-analysis carried out by the ENIGMA consortium improve upon conventional retrospective meta-analyses. ENIGMA working groups include both conventional studies of psychiatric disorders such as the ENIGMA Bipolar Disorder Working Group, as well as genetics-first approaches like the ENIGMA 22q11.2 deletion working group, which study populations at high risk for psychiatric illness. These ENIGMA studies represent the largest collaborative efforts to study neuroimaging markers in their respective disorders.

The ENIGMA bipolar working group has reported altered cortical and subcortical volumes, which show significant predictive power in discriminating patient populations. The ENIGMA 22q11.2 deletion syndrome working group has reported robust cortical, subcortical and white matter differences between 22q11.2 deletion subjects and healthy controls, differences between common microdeletion subtypes, as well as associations between 22q11.2 deletion-related psychosis and idiopathic schizophrenia. Large-scale, harmonized processing and analysis efforts such as these make cross-diagnostic comparisons possible on an unprecedented scale. Given the known overlap between major psychiatric disorders, such cross-disorder studies may provide robust markers that improve diagnoses, help monitor disease progression, and provide insights into novel therapeutic targets.

Efforts to bridge the gap between genes, brain, and behavior are facilitated by the study of highly-penetrant, disease-associated copy number variations (CNV). CNVs at the 22q11.2 locus present as deletions and duplications, each associated with clinical phenotypes that implicate perturbed neurodevelopmental processes. In fact, 22q11.2 CNVs confer some of the greatest known genetic risks for psychiatric disorders, and thus represent a compelling model to yield biological insights into how gene dosage may influence downstream brain and behavior. In this body of work, I present three studies that offer a comprehensive approach tethering cortical morphometric changes, cognitive and behavioral impairments, and transcriptomic dysregulation to reveal novel insights about the molecular effects of 22q11.2 CNVs. The first two studies are published in The Journal of Neuroscience and Biological Psychiatry, respectively, while the third study will soon be submitted to Brain, Behavior, and Immunity. In Chapter 1, I briefly introduce the challenge of studying neurodevelopmental psychiatric disorders, and how the use of highly penetrant, disease-associated CNVs can help to mitigate some of these biological complexities. Then, I explain how reciprocal 22q11.2 CNVs have great potential to offer key insights into the underlying biology of neuropsychiatric disorders. In Chapter 2, I show for the first time that 22q11.2 deletions (22qDel) and duplications (22qDup) confer global opposing effects on brain morphometry by comparing 66 molecularly-confirmed 22qDel carriers, 21 22qDup carriers, and 56 demographically-matched controls. 22qDel was associated with widespread reductions in cortical surface area, with corresponding enlargement in 22qDup. Cortical thickness showed the opposite pattern, but with more localized effects. Regional patterns of thickness difference between 22qDel and 22qDup carriers were also observed in subcortical regions which, in concert with the diffuse surface area changes, imply global and early impacts of 22q11.2 CNVs on brain development. Lastly, these findings were not accounted for by a subset of the sample, but rather the entire distribution was shifted, which suggests a highly penetrant effect of gene dosage. In Chapter 3, I systematically examine neurodevelopmental phenotypes relevant to intellectual functioning, Autism Spectrum Disorder (ASD), and psychosis in 106 22qDel carriers, 38 22qDup carriers, and 82 demographically-matched controls. This study aimed to leverage dimensional phenotyping, based on the hypothesis that variation in specific quantitative traits may better reflect underlying biological variation than categorical diagnoses. I found that reciprocal 22q11.2 deletions and duplications were associated with distinct impacts on intellectual functioning and psychosis-related symptomatology, but shared global deficits in the domain of ASD-related symptomatology. However, when more fine-grained subdomain measures of ASD-related traits were assayed, subtle differences in ASD profiles were found to distinguish deletion and duplication carriers. Lastly, while controls showed an inverse relationship with cortical thickness and processing speed, this association was absent in both CNV groups. This altered relationship between normative cortical thinning and cognitive processing speed implies disrupted development of the cortical mantle may underlie impaired processing efficiency in 22q11.2 CNV carriers. In Chapter 4, I leverage high-throughput transcriptomic data as an additional lens by which to gain mechanistic insights. This study represents the first genome-wide gene expression study to characterize transcriptomic dysregulation in peripheral blood in reciprocal 22q11.2 CNV carriers. Specifically, analyzing samples from 82 22qDel carriers, 29 22qDup carriers, and 68 demographically-matched controls, I showed that the 22q11.2 deletion significantly alters gene expression across the genome, and that these differences are substantially attenuated after adjustment for cell type heterogeneity. Moreover, expression of two cytoskeletal- and ASD-relevant genes differed between 22q11.2 CNV carriers with and without a diagnosis of ASD. I also extend findings regarding cell type differences between 22q11.2 deletion carriers and controls beyond known T cell differences to include mast cell and macrophage subtypes, two cell types that have multifunctional, immune-related roles throughout the body. These findings demonstrate the challenges inherent in using peripheral blood, a common tissue of analysis in psychiatric genetics, to study brain-related diseases. Nevertheless, this work shows how blood tissue, which contains a wealth of information regarding the immune system, can still offer valuable insights regarding immune mechanisms in psychiatric disorders, provided confounds such as cell type heterogeneity are properly accounted for. In sum, the work described herein seeks to reveal fundamental biology, generate hypotheses for future targeted experiments to validate and test in animal or in vitro models, and ultimately inform novel therapeutic targets.

Neurodevelopmental syndromes such as schizophrenia and autism are driven by a confluence of genetic and environmental risk factors leading to differences in brain development. Rare copy number variants (CNVs) involving deletions or duplications of multiple genes can incur high risk for neuropsychiatric illness and represent a ‘genetic-first’ model for understanding important biological pathways contributing to risk. Investigation of brain phenotypes in populations with behaviorally defined risk factors for mental illness represents a complementary ‘top down’ approach for linking clinical symptoms to brain systems and underlying biology. In this dissertation, three separate studies apply magnetic resonance imaging (MRI) in cohorts with genetic and clinical risk factors to better understand the developmental differences in brain structure and function that contribute to neuropsychiatric risk. In the first study, we assessed longitudinal development of functional connectivity between the thalamus and cortex in individuals with 22q11.2 deletion syndrome (22qDel) who are at high risk for schizophrenia and autism compared to neurotypical controls. We found altered developmental trajectories in 22qDel, with children displaying a pattern of thalamocortical hyperconnectivity in somatomotor regions and hypoconnectivity in frontoparietal regions, which had been previously observed in studies of schizophrenia, clinical high risk (CHR) for psychosis, and a cross-sectional analysis of 22qDel. In the second study, we investigated gene dosage effects on maturation of subcortical brain volumes across 22qDel, neurotypical controls, and carriers of 22q11.2 duplications (22qDup) who are at high risk for autism but not schizophrenia. We found gene dosage effects on multiple subcortical structures and a wide range of developmental alterations across the two CNV groups, illuminating the role of the 22q11.2 locus in subcortical development. In the final study we compared resting-state functional MRI case-control phenotypes in 22qDel versus CHR for psychosis to better understand similarities between these two high-risk populations. We found broadly dissimilar effects on long-range connectivity, local connectivity, and brain signal variability in these two cohorts, and related 22qDel case-control differences to multiple established brain maps, most notably positron emission tomography maps of hemodynamics and metabolism from neurotypical adults. These three studies provide important new insights into brain development and neuropsychiatric risk.

The representation and utilization of temporal information is a basic human ability that permeates many aspects of daily life, such as estimating the duration of an event or predicting the duration of a behavioral response. The ability to discriminate temporal durations develops in infancy; however, the precision (i.e., consistency) of timing abilities improves from early childhood to adolescence. In recent years, there has been a surge of interest in the neurobiological basis of temporal processing, and the frontal-striatal systems have been implicated in second-range timing functions. Frontal-striatal abnormalities are well documented in attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorders (ASD), and the genetic syndrome resulting from a deletion at 22q11.2 (22q11DS); therefore, temporal processing is a candidate endophenotype that may serve as a clinical indicator of aberrant frontal-striatal function. The purpose of this study was to: (1) characterize time reproduction accuracy and variability in three different neurodevelopmental disorders, (2) model age-related changes in time reproduction accuracy and consistency, and (3) assess the relative contributions of symptom dimensions (inattention, hyperactivity, and autistic traits) to time reproduction performance. Data were collected over a three year period as an adjunct to ongoing research studies in ADHD, ASD, and 22q11DS. The time reproduction task was a previously validated measure of interval timing with target durations of 4, 8, 12, 16, and 20 seconds repeated four times each in random order. Time reproduction accuracy and consistency were analyzed separately in each of the three samples using repeated measures mixed effects regression. Across all samples, younger age was the most consistent predictor of time reproduction variability. High levels of inattention in the ADHD group were also associated with increased variability. Both the ASD and 22q11DS groups showed evidence of increased response variability relative to typically developing comparison groups; however, in the ASD group the effect of diagnosis was moderated by working memory. The results have implications for empirical investigations of temporal processing across multiple dimensions of psychopathology and highlight the importance of considering both response variability and developmental factors (e.g., maturation of frontal-striatal circuits) in the formation of new theories linking neurobiological substrates to the emergence of symptom constellations.

Objective

Method

Results

Conclusions