Identification of a genetic network for an ecologically relevant behavioural phenotype in Drosophila melanogaster

Abstract

Pupation site choice of Drosophila third‐instar larvae is critical for the survival of individuals, as pupae are exposed to various biotic and abiotic dangers while immobilized during the 3–4 days of metamorphosis. This singular behavioural choice is sensitive to both environmental and genetic factors. Here, we developed a high‐throughput phenotyping approach to assay the variation in pupation height in Drosophila melanogaster, while controlling for possibly confounding factors. We find substantial variation of mean pupation height among sampled natural stocks and we show that the Drosophila Genetic Reference Panel (DGRP) reflects this variation. Using the DGRP stocks for genome‐wide association (GWA) mapping, 16 loci involved in determining pupation height could be resolved. The candidate genes in these loci are enriched for high expression in the larval central nervous system. A genetic network could be constructed from the candidate loci, which places scribble (scrib) at the centre, plus other genes known to be involved in nervous system development, such as Epidermal growth factor receptor (Egfr) and p53. Using gene disruption lines, we could functionally validate several of the initially identified loci, as well as additional loci predicted from network analysis. Our study shows that the combination of high‐throughput phenotyping with a genetic analysis of variation captured from the wild can be used to approach the genetic dissection of an environmentally relevant behavioural phenotype.

Testing implications of the polygenic model for the genetic analysis of loci identified through genome-wide association

“…The spread of pupal case length among the DGRP inbred lines exceeded that of the DSPR recombinant inbred stocks (ranging from 2.9 mm to 3.5 mm) (Reeves and Tautz, 2017), suggesting that DGRP lines capture indeed the existing variation of pupal case length in Drosophila melanogaster . The same conclusion applies also to the pupation site choice phenotype, as we have shown previously (Zhang et al, 2020).…”

“…Our results show a much higher contribution of autosomes to the variance of pupal case length (Table 1-figure supplement 1), with the exception of chromosome 4, mostly likely due to the limited number of genetic variants within this chromosome. Hence, pupal case length is a highly heritable phenotype and the DGRP lines include the genetic variation suitable for mapping, similarly as we have previously shown for the pupation site choice phenotype (Zhang et al, 2020).…”

“…We adapted and modified the earlier image-analysis based phenotyping pipeline (Reeves and Tautz, 2017) for the high-throughput measurement of pupal case length. The same approach allows also to score the independent phenotype of pupation site choice (pupation height) and for both we have shown that a very large number of phenotypes can be measured with high precision (Reeves and Tautz, 2017; Zhang et al, 2020).…”

“…We have chosen Drosophila pupal size and pupation site choice as quantitative phenotypes, for which we have previously developed automated phenotyping pipelines and have shown their polygenic basis (Reeves and Tautz, 2017; Zhang et al, 2020). The pupal stage is an immobile life history stage of transformation between larval and adult stages found in holometabolous insects (Jones and Reiter, 1975; Price, 1970).…”

“…Pupation site choice (or pupation height) is measured as the distance of the pupation site above the surface of the food. It reflects an ecologically relevant trait that is behaviourally determined and where a core network could be identified (Zhang et al, 2020). The phenotypes are not directly related to each other (Zhang et al, 2020).…”