A Genomic Approach to Schizophrenia

University Of Washington, Seattle WA

Investigators

Linked publications & trials

Abstract

? DESCRIPTION (provided by applicant): Our goal is to identify genes important to schizophrenia. In the first cycle of our project, we demonstrated that damaging de novo mutations in persons with schizophrenia, from otherwise healthy families, disrupt genes that are co-expressed in the dorsolateral and ventrolateral prefrontal cortex during fetal development. Compared to their unaffected siblings, schizophrenia patients were significantly more likely to harbor such alleles. Proteins encoded by these genes functioned in neuronal migration, synaptic transmission, signaling, and transcriptional regulation. Integration of genetic data and expression data suggested possible schizophrenia-related processes and even potential targets for treatment. In the next cycle of our project, we ask whether these and other candidate genes are enriched for severe mutations in schizophrenia, by comparing mutation profiles of each gene for all cases and controls in the NIMH repository. In Aim 1, we will select ~500 candidate genes from among those with de novo damaging mutations in schizophrenia or autism, using network analytic tools to suggest those genes most likely to be causal and to identify other candidate genes. In a preliminary study, we show that an expanded network seeded by the 54 genes with damaging de novo mutations in our cases is significantly enriched for genes involved in chromatin modification and synaptic function, and for genes with de novo mutations in schizophrenia or autism from other studies. In Aim 2, we will sequence 500 candidate genes in DNA from all European American and African American schizophrenia cases and controls from the NIMH repository, ~12,000 subjects total. For each gene, we will sequence coding regions, UTRs, and non-coding potential regulatory regions. We will compare distributions of damaging alleles of cases and controls gene-by-gene and by network-defined functional groups. In a preliminary study, we identified novel truncating mutations in 3 different genes, each in one of 24 patients sequenced for 281 genes from a co-expression network. We use a method of pre-capture pooling and hybridization that yields a sequencing cost per complete gene per sample of 13 cents. In Aim 3, we will use C. elegans to characterize the roles of schizophrenia candidate genes in neurodevelopment. In candidate genes with orthologs in C. elegans, we will measure the consequences of RNAi-reduced expression on neurological phenotypes, including defects in axon guidance, dendritic branching, or neurotransmitter-specific neuron function. Then, for selected human mutations, we will create transgenic animals to examine the consequences of replacing the C. elegans gene with the human mutant vs normal gene. Together, these aims are designed to identify genes or clusters of genes important to schizophrenia.