Genetic Basis of Reward Learning and Symptom, Smoking, and Neural Signature Correlates

Mclean Hospital, Belmont MA

Investigators

Linked publications, trials & patents

Abstract

Project Summary Schizophrenia [SZ] and bipolar disorder [BPD] are two the most costly and debilitating psychiatric diseases worldwide, and are associated with many impairments. One of these is a deficit in reward-based learning. However, the underlying genetic basis and neurobiological mechanisms remain largely unknown. Such incomplete understanding partially arises from the fact that current classification systems lack validity and encompass a heterogeneous set of disorders. Cross-disorder genome-wide association studies (GWAS) have shown overlapping polygenic risk among SZ and BPD, and major depression, indicating pleiotropic effects of some risk variants across DSM classification. Neuroscience research has identified that dysfunction of reward learning cuts across diagnostic boundaries. To date, virtually all prior studies designed to characterize brain pathophysiology and its underlying biological mechanisms associated with reward learning in disease states have relied on DSM-based diagnoses, rather than on a dimensional approach. To address this issue, the NIMH created an opportunity (PAR-14-008) for secondary analyses of existing clinical research data. Our proposal responds to this opportunity by combining two well characterized datasets and using them to: i) investigate the ?Reward Learning? construct of the RDoC Positive Valence System domain in healthy and clinical samples, ii) define reward learning in relation to clinical symptoms (anhedonia), a specific environmental risk factor (smoking), and to brain electrophysiology, and iii) dissect the genetic basis of reward learning. In line with the RDoC effort, we take a dimensional approach to examine data across self-report, behavioral, physiological, and genetic units of analyses. We will make use of two unique resources, consisting of patients with diagnoses of affective or psychotic disorders, and normal individuals (total N=855) with measures of reward learning, clinical symptoms, medication, substance use, and genomewide genotyping data, as well as a subgroup with neurophysiological measures. We will conduct the first and cutting-edge polygenic risk analyses to dissect the genetic basis of reward learning. The polygenic analysis is designed to capture greater phenotypic variance compared with single marker analyses. In addition, we will genotype the risk SNP markers identified by the large scale GWAS analyses and perform SNP-based association analyses to examine the effects of risk variants for psychiatric disorders on reward learning. Furthermore, we will use the combined phenotypic data to dissect the relationships between reward learning, clinical symptoms, anhedonia, nicotine use, and medication. Finally, we will examine the link between reward learning and brain electrophysiology and the role of nicotine as a potential modulator. Upon completing this project, we expect to provide the field with important insights of genetic underpinning of reward learning, and its associations with clinical symptoms, smoking, and brain neurophysiology across diagnostic classifications.