Genome-wide association studies (GWAS) provide robust statistical evidence linking individual sequence variants to increased risk of disease. However, the majority of GWAS-identified risk variants identified to date, including those for psychiatric disorders, do not affect protein-coding sequence but instead map to noncoding DNA. GWAS alone yield little direct insight into the mechanisms by which these variants confer increased risk. It is widely assumed that many affect the function of distant-acting transcriptional enhancers, but the historically poor annotation of noncoding functional sequences in the human genome has rendered their interpretation challenging. Over the past decade, our lab and others made substantial progress toward identifying enhancers in the genome at scale. For example, our group has used ChIP-seq from mouse and human brain tissues to identify initial collections of developmentally active brain enhancers. Furthermore, as members of the ENCODE consortium, we generated the first high-resolution time series mapping the brain chromatin landscape throughout mouse prenatal development. These resources, along with complementary data from NIH-funded consortia, are now available to aid in the interpretation of GWAS results. Here we propose to bridge the current gap between noncoding GWAS findings and mechanistic understanding of psychiatric disorder etiology. We will couple extensive pre-existing epigenomic resources to cutting-edge mouse engineering and single cell-resolution transcriptome analyses in order to understand how enhancer variants contribute to psychiatric disease risk. Specifically we will: 1) Perform an integrative analysis of psychiatric disorder GWAS results and epigenomically predicted brain enhancers to identify regulatory sequences that harbor disease-associated variants and prioritize them for functional validation, 2) Use high-throughput mouse transgenic assays to validate predicted enhancers in vivo and determine the exact brain regions in which they are active, 3) Use single cell RNA-sequencing of transgenic mice to determine the exact cell type(s) in which a brain enhancer functions, and 4) Use these in vivo functional genomic methods to uncover whether and how disease-associated sequence variants impact the cell type-specific activity of each enhancer. As whole genome sequencing of psychiatric disease cohorts continues to progress, we will also assess variants from those studies. In combination, our work will uncover how noncoding disease-associated variants alter enhancer function in vivo, link noncoding GWAS findings to specific cell types, and provide a unique panel of well-characterized enhancers that can be used to label discrete neuronal cell populations for further downstream characterization. The results will substantially improve our mechanistic understanding of how noncoding sequence changes contribute to mental illness and provide entry points for potential downstream therapies.
Human genetics studies routinely implicate noncoding sequence variants, particularly those that reside in tissue-specific transcriptional enhancers, in psychiatric disorders. However, it remains unclear how these noncoding sequence changes are mechanistically linked to disease and even which specific cell types they are acting through. This project will use large-scale in vivo testing of predicted human brain enhancers harboring psychiatric disorder-associated variants to understand how noncoding sequence differences alter enhancer function and identify the cell types whose dysregulation may underlie psychiatric phenotypes.
