Parallel phosphoproteomics and metabolomics map the global metabolic tyrosine phosphoproteome

Abstract

Tyrosine phosphorylation of metabolic enzymes is an evolutionarily conserved post-translational modification that facilitates rapid and reversible modulation of enzyme activity, localization or function. Despite the high abundance of tyrosine phosphorylation events detected on metabolic enzymes in high-throughput mass spectrometry-based studies, functional characterization of tyrosine phosphorylation sites has been limited to a subset of enzymes. Since tyrosine phosphorylation is dysregulated across human diseases, including cancer, understanding the consequences of metabolic enzyme tyrosine phosphorylation events is critical for informing disease biology and therapeutic interventions. To globally identify metabolic enzyme tyrosine phosphorylation events and simultaneously assign functional significance to these sites, we performed parallel phosphoproteomics and polar metabolomics in non-tumorigenic mammary epithelial cells (MCF10A) stimulated with epidermal growth factor (EGF) in the absence or presence of the epidermal growth factor receptor (EGFR) inhibitor erlotinib. We performed an integrated analysis of the phosphoproteomic and metabolomic datasets to identify tyrosine phosphorylation sites on metabolic enzymes with functional consequences. We identified two previously characterized (PKM, PGAM1) and two novel (GSTP1, GLUD1) tyrosine phosphorylation sites on metabolic enzymes with purported functions based on metabolomic analyses. We validated these hits using a doxycycline-inducible CRISPR interference (CRISPRi) system in MCF10A cells, in which target metabolic enzymes were depleted with simultaneous re-expression of wild-type, phosphomutant or phosphomimetic isoforms. Together, these data provide a framework for identification, prioritization and characterization of tyrosine phosphorylation sites on metabolic enzymes with functional significance.