Abstract
Phenotypic complexity results from the contributions of environmental factors and multiple genetic loci, interacting or acting independently. Studies of yeast and Arabidopsis found that the majority of natural variation across phenotypes is attributable to independent additive quantitative trait loci. Most of the loci were identified, facilitated by the power to detect almost all heritable components underlying trait variation in these model organisms. By contrast, many heritable components underlying phenotypic variation in metazoan models remain undetected. Before the relative impacts of additive and interactive variance components on metazoan phenotypic variation can be dissected, high replication and precise phenotypic measurements are required to obtain sufficient statistical power to detect loci contributing to this missing heritability. Here, we used a panel of 265 recombinant inbred advanced intercross lines of Caenorhabditis elegans and a high-throughput fitness assay to detect loci underlying responses to 16 different toxins, including heavy metals, chemotherapeutic drugs, pesticides, and neuropharmaceuticals. Using linkage mapping, we found many genomic regions that underlie variation in responses to these toxins and predicted the relative contributions of additive loci and genetic interactions across various growth parameters. Additionally, we identified three genomic regions that impact responses to multiple classes of toxins. We generated near-isogenic lines and chromosome-substitution strains to experimentally validate these QTL hotspots and implicated interacting loci that underlie toxin-response variation. The discovery of these QTL hotspots might be indicative of pleiotropic loci that control responses to multiple conditions and could underlie the means by which large regions of the genome were swept across the C. elegans species.
