Mitochondrial protein functions revealed by global protein‐lipid‐metabolite profiles

Abstract

A central challenge in experimental biology and clinical medicine is to rapidly quantify phenotype changes with enough breadth to capture diagnostic features. Here, using systematic mass spectrometry-based profiling, we quantify 3,250 biochemical phenotypes in 1,000 single gene deletion yeast cultures to define functions for uncharacterized genes and fingerprints of mitochondrial dysfunction. By profiling diverse mitochondrial disease gene homologs, we identify a general respiration deficiency response, define gene-specific perturbations, and predict new disease gene candidates. These include previously unrecognized roles for Hfd1p—a conserved dehydrogenase linked to human disease—in the production of 4-hydroxybenzoic acid for ubiquinone biosynthesis. Integration of proteome, lipidome, and metabolome perturbations reveals functional networks that include membership in protein complexes and unexpected inter-ome correlations between perturbation profiles of functionally related genes, which can predict function. Collectively, our results provide molecular insight into mitochondrial biology and, more broadly, establish a high-throughput approach for defining gene function and quantifying diagnostic phenotypes. Support or Funding Information This work was supported by a Searle Scholars Award, and by National Institutes of Health grants U01GM94622 and R01GM112057 (to D.J.P.), NIH Ruth L. Kirschstein National Research Service Award F30AG043282 (to J.A.S.), and National Institutes of Health (R01 GM080148) and the National Science Foundation (0701846) (to J.J.C.). Figure 1Open in figure viewerPowerPoint Experimental design for our “yeast 3,000 (Y3K)” mass spectrometry-based protein functional annotation strategy