Previous CNP Funded Pilots:
Bearden – A ‘Human Knockout’ Model for Cross-Diagnostic Investigation of Memory and Response Inhibition Phenotypes
Groman/Brown – Assessment of Dopamine D1 and D2 Receptors in Brain and Their Association with Behavioral Phenotypes
Molnar-Szakacs – Web-Based Cognitive Phenotyping: Reliability and Validity Studies in the LA2K
Mumford – New methods for functional connectivity phenotype identification
Congdon - Components of Impulsivity and Risky Decision Making
Karlsgodt - Translational Assessment of Complex Phenotypes: Bridging the Gap Between Behavior and Brain Function
James/Seu - Developing a Model of Risk-Taking Behavior for Whole Genome Studies
Jasinska - Identification of gene expression traits involved in stress response
Fears - Functional Brain Phenotypes and Behavioral Correlations in BXD Mice
Groman - Mechanistic Studies Related to D2-Receptor Signaling in Impulsivity
Kaufman & Ghahremani - Human Reversal Learning Task Development
Funded Project Abstracts
The 22q11.2 deletion syndrome (Velocardiofacial/DiGeorge syndrome; 22q11del) is a compelling model for the study of cross-diagnostic phenotypes, as it represents one of the greatest known risk factors for the development of psychotic illness, and is also associated with high rates of bipolar disorder and attentional disturbance. Because of the variety of psychiatric disorders associated with this syndrome, which span a range of DSM-IV diagnostic categories, as well as the presence of several CNS-relevant genes in the deleted region, including a gene involved in prefrontal dopamine metabolism (COMT)2, this disorder presents a fascinating ‘human knockout’ model allowing for investigation across multiple levels of analysis within the same subjects. Here we propose to investigate memory and response inhibition (RI) phenotypes in 22q11del, using identical neural, cognitive and clinical assays to those utilized in the CNP.
A full understanding of what predisposes to risky behaviors common across psychiatric disorders will require a characterization of how different processes interact with subfactors of impulsivity, and while much is known about response inhibition, incentive motivation, and reinforcement learning separately, their relationships remain poorly understood. Furthermore, there is a significant lack of research on these relationships in healthy, high impulsive individuals and this is critical as they represent the link between normal and pathological variation in impulsivity and component processes of decision making. In order to address this critical gap in knowledge about the predisposing mechanism to risky behaviors, we propose a pilot study which takes advantage of the CNP infrastructure, resources, and recruitment efforts, and which offers future opportunities for translational research. In particular, we propose to invite participants, pre-selected based on trait impulsivity scores (assessed in LA2K), to participate in a third scan session, in which they will perform additional tasks related to response inhibition, incentive motivation, and reinforcement learning.
The proposed pilot study aims to test relationships between quantitative behavioral and neurochemical measures with phenotypes in order to clarify the potential contributions of D1-like vs. D2-like dopamine (DA) receptor systems to variation in behavioral measures of impulsive temperament, response inhibition, and memory. This pilot uses PET scanning to determine in vivo DA D1-like and D2-like receptor availability in striatal and extra-striatal brain regions of subjects and to relate these measures to indices of impulsivity, response inhibition, and memory.
While the application of imaging measures to animal models is growing in popularity, there is a paucity of data on how to use the resulting measures as meaningful phenotypes, and how to interpret their translation to human investigations. Rodent models are powerful tools for assessment of disease- related phenotypes due to the ability to manipulate individual genes of interest. This project aims to more closely link the available levels of analysis in human subjects and genetic mouse models of schizophrenia using two methodologies well suited to cross-species research: long term memory and neuroimaging. We will test the viability of a novel imaging based phenotype of cellular function in mice and further assess whether the proposed imaging measures relate to behavioral indices. Specifically, we will use a combination of behavioral testing, manganese enhanced MRI (MEMRI), and cerebral blood volume (CBV) in dysbindin mutant mice to try to build a bridge between imaging measures used in mice and those use in humans, as well as to understand the behavioral implications of changes in these measures.This approach has potential utility for explorations of brain-behavior relationships in models of a variety of major mental disorders, including schizophrenia.
By implementing a novel behavioral measure of risk-taking, the rodent-appropriate version of the Balloon Analogue Risk Task (BART), which was designed with the specific goal of cross-species translatability, the proposed study will establish a construct valid method of phenotyping decision- making under risk in a genetically-tractable model organism. Furthermore, it will initiate studies that will reveal candidate genomic regions which control variation in this psychiatric disorder-associated phenotype; these findings will directly inform LA2K genome-wide association efforts to detect genes relevant to risk proneness in humans.
Identification of genetic and environmental determinants of cognitive, behavioral and neurostructural traits is a challenging task due to the complexity of relations between these traits at different levels, their uncertain genetic architecture, and the difficulty of quantifying or controlling environmental exposures. The proposed project intends to take advantage of existing extensive genomic and phenomic resources. It will use gene transcript levels as intermediate phenotypes to explore the effects of exposure to an acute environmental stressor and map stress related expression Quantitative Trait Loci (eQTL); we hypothesize this study will reveal the genetic basis of stress response. Anticipated identification of genes related to relocation stress (RS) and genetic differences among individuals affecting their response to the stressor, at a molecular level, will help to elucidate molecular mechanisms underlying complex behavioral and cognitive traits modulated by stress. Such traits are considered important endophenotypes involved in psychiatric disorders, for example memory processes in schizophrenia and response inhibition in attention deficit/hyperactivity disorder, and therefore are intensively investigated by CNP.
Psychiatry research has converged on the idea that advancing understanding of complex neuropsychiatric syndromes, like schizophrenia, requires the identification of intermediate phenotypes that exist between gene and disease diagnosis. Yet for this dynamic approach to have the most impact, it will need to a valid high-throughput approach in order to acquire large enough sample sizes in the amount of time needed. We propose to validate the current web-based data-collection platform used by CNP investigators (BrainTest.org) with respect to traditional lab-based procedures implemented by the LA2K, and assess possible ascertainment biases that may affect web-assessment in the context of the CNP.
Neuroimaging phenotypes are almost always defined in terms of localized activity in specific brain regions, but an important challenge for neuroimaging is to characterize how the function of these individual brain regions is integrated to achieve complex mental function. Unfortunately, there are no widely accepted methods for the analysis of functional integration or connectivity. The goal of this pilot proposal is to develop flexible new approaches for studying functional integration and characterizing the phenotypes that comprise the functional connectome. We will develop extensions of recently developed graph- theoretical approaches to gene expression mapping, and will apply these methods to identify voxel coactivation networks in the brain. By adapting these methods to fMRI data, we will provide new and interesting ways to compare brain activation patterns across genotypes, possibly revealing new phenotypic measures of mental illness that may be more powerful than localization-based phenotypes.