Zebrafish screen of high-confidence targets at insomnia GWAS loci implicates MEIS1, SKIV2L, and ARFGAP2 as conserved regulators of sleep-wake behaviors

Study objectives Insomnia is a pervasive sleep disorder with heritability estimates ranging from 22-59 percent. Human genome-wide association studies (GWAS) have identified more than 200 genome-wide significant loci associated with the pathogenesis of insomnia; however, few of the implicated effector genes at these loci have undergone functional characterization, and little is known of their relevance to sleep dysfunction in insomnia. High-throughput phenotyping in model organisms, especially for conserved traits like sleep, provides an opportunity for enhanced gene prioritization to improve mechanistic understanding of this complex polygenic disorder and to yield strong genetic candidates for therapeutic intervention. Methods Leveraging existing datasets, we performed a functional screen of candidate insomnia-associated genes in zebrafish. Our previous 3D-genomics approach, which integrated ATAC-seq RNA-seq and high-resolution promoter-focused Capture C in neural progenitor cells, revealed putatively causal variants and corresponding effector genes at insomnia GWAS loci. We went on to report first-pass neuron-specific RNA interference screening in Drosophila to implicate high-confidence effector genes at these genetic signals. This current study builds on those initial observations by leveraging CRISPR/Cas9 in a diurnal vertebrate model, namely larval zebrafish (5-7 days post fertilization), to perform functional validation of six top candidates to determine conserved regulators of sleep behaviors relevant to insomnia. Results CRISPR mutation of three high-confidence insomnia-implicated effector genes ( MEIS1 , SKIV2L , and ARFGAP2 ) resulted in robust changes to sleep and activity patterns in larval zebrafish. Specifically, we observed phenotypes that reflect insomnia-like behaviors, including altered sleep duration, poor sleep consolidation, and impaired latency to sleep onset after lights off. Moreover, mutation of ARFGAP2 also impacted development resulting in a significant motor phenotype. Conclusions Our results reveal the utility of CRISPR/Cas9 in F0 larval zebrafish for high-throughput screening of GWAS-implicated candidate effector genes. We validated three of the six strongest candidate effector genes identified in our previous combined 3D Genomic and D rosophila screen, two of which were 3D-mapped to different genes rather than the nearest gene reported in the original insomnia GWAS study. Together, we demonstrate the necessity for functional validation of GWAS-implicated effector genes, and we provide evidence for two novel genes producing sleep-wake phenotypes not previously suggested through positional mapping alone. ### Competing Interest Statement The authors have declared no competing interest.