A platform for the systematic interrogation of genetic interactions in human cells: 26 gene by genome-wide double knockout screens as a proof of concept

Abstract

Genetic interactions are typically studied by looking at the phenotype that results from disruption of pairs of genes, as well as from higher order combinations of perturbations. Systematically interrogating all pairwise combinations provides insights into how genes are organized into pathways and complexes to sustain cellular homeostasis and how interacting genes respond to stressors and external signals. Genetic interactions have been studied extensively in yeast, due, in part, to the availability of a systematic collection of gene knockouts, and the development of Synthetic Genetic Array (SGA) technology. In contrast, such approaches are more challenging in human cells and therefore comparable data for human cells is scarce. This study introduces an innovative approach to functionally characterize genetic interactions in human cells through CRISPR/Cas9 screens using a pooled genome-wide knockout library in NALM6 cells. By combining a single guide RNA (sgRNA) targeting the gene of interest (aka the query) in cells already infected with an inducible genome-wide sgRNA pool, it is possible to achieve near saturation of genome-wide double knockouts. We conducted 26 of these screens, which we term "gene by genome-wide" knockout screens. This approach can be rapidly performed, in part, because it bypasses the need to generate genotyped isogenic knockout clones. Data from these screens identified both expected and novel synthetic lethal and synthetic rescue interactions, demonstrating that this strategy is effective for large-scale genetic research in human cells. Additionally, we show that these GBGW screens can be combined with chemical perturbation to reveal new synthetic interactions that are not apparent without drug treatment. Finally, we show that cDNA overexpression can be incorporated with genome-wide knockouts to systematically explore gain-of-function scenarios. The complete dataset is accessible on the ChemoGenix website (URL: https://chemogenix.iric.ca).