Systematic identification of changes in the yeast protein interaction network in response to environmental, chemical, and genetic perturbation [interactome data]

To understand the principles underlying protein-protein interaction (PPI) complex changes in response to external perturbations, we created a highly multiplexed version of the murine dihydrofolate reductase protein complementation assay (mDHFR PCA) in Saccharomyces cerevisiae, allowing quantitative PPI complex profiling in vivo. We investigated the effects of 14 different conditions (including small molecules, abiotic stress factors, and nutrient composition) on a total of 1383 PPIs. More than half of PPIs (758) were found to be variable, and their Gene Ontology (GO) annotations were found to be informative of both the nature of the perturbation within each condition, as well as the overall variability of the interactions across conditions. Many perturbations triggered network changes characterized by large connected modules centered around highly connected proteins ('hubs'), suggesting that cellular control of a few proteins (e.g., by mRNA levels) can induce widespread PPI remodeling. Under a diauxic shift from glucose to ethanol as the main carbon source, we found a striking relationship between PPI changes measured by our assay and those predicted by mRNA expression under a simple law of mass action based model. In the murine Dihydrofolate Reductase-based Protein-fragment Complementation Assay (mDHFR PCA), the interaction of two proteins of interest allows a yeast strain to grow in the presence of the DHFR inhibitor methotrexate (MTX). We have created a pool of mDHFR-PCA strains each containing unique 20-bp barcode sequences flanked by common priming sites, allowing the relative abundance of individual strains to be quantitatively monitored within a pool of competitively-grown strains. To identify protein-protein interactions that change in response to environmental perturbation, we measured the fitness of 1383 mDHFR-PCA strains in parallel in minimal media (i.e., yeast nitrogen base and ammonium sulfate) in the presence or absence of methotrexate and in the presence of an environmental perturbation. Each condition was tested in duplicate. To test the impact of single protein deletion on the network, we implemented a Tet-inducible CRISPRi system to repress the transcription of a target gene in a controlled manner. In this system, the expression of a guide RNA is activated by addition of anhydrotetracycline (ATC) into the media, which then recruits a Cas9-repressor fusion protein to a locus of interest. We targeted the RBD2 gene in this manner.