Profiling tyrosine kinase substrate recognition using bacterial peptide display and deep sequencing.

Tyrosine kinases control a wide range of cell signaling pathways that are central to human physiology, and they are dysregulated in a variety of human diseases, most notably cancers. Our understanding of tyrosine kinase biology hinges upon a clear delineation of their protein substrates. Thus, much effort has been invested into defining the substrate specificities of tyrosine kinases, which is partly driven by recognition of the amino acid sequences surrounding the phospho-acceptor tyrosine residues. Numerous methods have been developed to profile tyrosine kinase sequence recognition, and these approaches have collectively demonstrated that different tyrosine kinases have distinct substrate sequence preferences. Here, we describe one such method that combines bacterial peptide display and deep sequencing to study tyrosine kinase substrate preferences. Our approach enables rapid measurement of relative phosphorylation efficiencies for thousands of peptides simultaneously. The genetically-encoded nature of the peptide libraries used with this method allows for facile and cheap construction of libraries tailored to answer a variety questions. Notably, our approach is compatible with genetic code expansion via Amber codon suppression, which allows for the construction and screening of libraries containing non-canonical amino acids. Importantly, results from this assay correlate strongly with quantitative measurements of enzyme kinetics, they corroborate previously reported tyrosine kinase substrate preferences, and they can reveal new insights into tyrosine kinase substrate specificity.