PgmNr P2096: Genetic dissection of variation in sleep using the Drosophila Synthetic Population Resource.

Authors:
S. J. Macdonald; B. R. Smith


Institutes
Department of Molecular Biosciences, University of Kansas, Lawrence, KS.


Abstract:

In humans there is considerable genetic variation for the duration and quality of sleep, and studies have implicated a lack of sleep as a risk factor for a range of health problems. Drosophila, in addition to being widely employed as a model genetic system to understand fundamental aspects of the control of complex trait variation, is recognized as an important translational model for the study of human health and disease. Flies exhibit a sleep-like state, and similar to mammals show increased rest following a period of sleep deprivation, an age-related decline in the duration of sleep, and reduced rest following exposure to caffeine. Here, we use Drosophila to dissect natural variation in sleep, identify loci for future functional testing, and facilitate experimental exploration of the mechanistic basis of variation in sleep. As with all complex, polygenic traits identifying the molecular pathways and causative genes responsible for phenotypic variation is challenging. Thus, we employed a multi-tiered approach, encompassing high-resolution QTL mapping, expression QTL (eQTL) data, and functional validation with RNAi. We initially measured a battery of sleep traits in multiple individuals from each of 600 heterozygous genotypes derived from the Drosophila Synthetic Population Resource (DSPR). The DSPR is a large set of multiparental advanced generation intercross lines that facilitates powerful mapping of QTL to small genomic regions. We observed extensive genetic variation in sleep traits in the population, and successfully mapped a number of QTL that collectively explain significant fractions of variation in sleep. Under the assumption that some fraction of phenotypic variation is due to changes in gene regulation, true causative genes implicated by QTL may often additionally segregate for cis-eQTL. Merging sleep QTL data with a large Drosophila head transcriptome eQTL mapping dataset from the same population allowed us to refine the list of plausible candidate causative sleep loci. This set includes genes with previously characterized effects on sleep (e.g., timeless), in addition to novel candidates. Subsequently, we employed adult nervous system specific RNAi to functionally test the effects of several loci, identifying significant effects on sleep following knockdown of the Dopa decarboxylase and dyschronic genes. The genes we identify are highly likely to harbor causative, regulatory variation contributing to variation in sleep-like phenotypes.