Appropriate social behaviors are essential to humans and animals for reproduction and survival. Defects in social interaction and communication in humans can result in autism spectrum disorder (ASD). Little is understood about the basic neurogenetics controlling complex ASD-like behaviors in humans. Here we use an unbiased quantitative trait locus (QTL) approach in Drosophila melanogaster to identify genes and gene interactions that influence 1) social communication during mating (receptive and expressive) and 2) social space interactions in flies.
The Drosophila Genetic Reference Panel (DGRP) has been used to analyze a range of naturally occurring behaviors including sleep, appetite, olfaction, and aggression. The DGRP represent >4,800,00 single nucleotide polymorphisms (SNPs) and >2,900,00 non-SNP variants mirroring the variation for each of 205 individuals in a wild type population. Genetic variation within each line is fixed, while variation across lines represents the naturally occurring variation in the original population. The DGRP is thus the most powerful system available for behavioral genetics among model organisms. Using online genome wide association tools for the DGRP (http://dgrp.gnets.ncsu.edu), we identified genes associated with ASD-like behaviors across the DGRP lines.
Social communication during mating was used as a proxy for expressive and receptive social communication and social space measured using a social space assay triangle to measure average resting distance among flies. The measures used for genome wide association were time to latency (expressive/receptive communication) and average distance between flies (social space).
We have now identified at least one gene, sfl (sulfateless), associated with both social communication and social spacing behavior. sfl encodes a heparan sulfate N-deacetylase known to be essential for synaptic vesicle endocytosis. We have also identified QTLs in two known ASD gene homologs affecting receptive (neuroligin2) and expressive (Nrx-IV) communication. We have identified QTLs in 130 genes affecting ASD-like behaviors but have only tested ~20% of the fully sequenced DGRP lines. Optimizing resolution and statistical power of these associations will require assessment of all 205 lines for all ASD-like behaviors. Our end goal is to identify new gene-gene relationships or even variants in known ASD genes that are associated with these behaviors in humans.