Abstract Introduction Large-scale Genome Wide Association Studies (GWAS) have been performed for a variety of sleep traits including insomnia, excessive daytime sleepiness, sleep duration, and chronotype. These studies have provided a wealth of data identifying common genetic variants associated with sleep function and informing biological mechanisms underlying each trait. A primary impediment to interpreting the biological relevance of GWAS loci is that the majority of these variants reside in non-protein-coding regions of the genome often distal to any protein-coding gene. Our work outlines an integrative approach to pinpoint causal genes associated with each GWAS locus using human spatial and functional genomics data paired with cross-species sleep phenotyping. Methods We collected lead and correlated (r20.6) proxy variants from published GWAS summary statistics for insomnia and excessive daytime sleepiness. Cis-regulatory regions were defined using ATACseq and Promoter Capture C/Hi-C data collected from 9 different human-derived brain cell types including neural progenitors, neurons, and glia. GWAS variants were intersected with cell-specific cis-regulatory regions to identify putative causal effector genes contacted by each variant. These implicated effector genes were then screened for their effects on sleep using cell-specific GAL4-RNAi (glial or neuronal subtypes) knockdown in Drosophila. Genes showing significant effects on sleep were then validated in a vertebrate model, zebrafish, using CRISPR-mediated knockout paired with sleep phenotyping. Results We identified multiple highly conserved regulators of sleep that mapped to both the nearest gene of the GWAS association (e.g. MEIS1 and CADM2) as well as distal genes that were not previously implicated by these GWAS (e.g. PIGQ and AP3B2). Each of these significantly influenced sleep in both Drosophila and zebrafish in the same direction and showed cell-specific effects when knocked down in Drosophila. Loss of orthologs to MEIS1, which was identified from the insomnia GWAS, showed short and fragmented sleep. Additionally, loss of orthologs to CADM2 and AP3B2, which were identified from the excessive daytime sleepiness GWAS, displayed increased sleep. Conclusion Together our integrative cross-species approach defined highly conserved regulators of sleep from human GWAS associations revealing novel biological targets for intervention. Support (if any) T32 HL07953, T32 HL170968, R01 HL143790, P01 HL094307, and P01 HL160471
Zimmerman et al. (Fri,) studied this question.