Proximity Labeling-assisted Identification of Endogenous Kinase Substrates

Abstract

Mass spectrometry-based phosphoproteomics can identify more than 10,000 phosphorylated sites in a single experiment. But, despite the fact that enormous phosphosite information has been accumulated in public repositories, protein kinase-substrate relationships remain largely unknown. Here, we describe a method to identify endogenous substrates of kinases by means of proximity labeling. We used a proximity-dependent biotin identification approach, called BioID, in combination with kinase-perturbed phosphoproteomics profiling and phosphorylation sequence motifs derived from in vitro kinase assay to find molecules that interact with a target kinase, that show altered phosphorylation in response to kinase perturbation, and that are directly phosphorylated by the kinase in vitro; i.e., endogenous kinase substrates. Application of this methodology to casein kinase 2 (CK2) and protein kinase A (PKA) identified 33 and 52 putative substrates, respectively. We also show that known cancer-associated missense mutations near phosphosites of substrates affect phosphorylation by CK2 or PKA, and thus might alter downstream signaling in cancer cells bearing these mutations. This study extends our knowledge of kinase-substrate networks by proposing a new large-scale approach to identify endogenous substrates of kinases.