Phospho-dependent molecular switches in cancer signalling

Protein phosphorylation plays a critical role in cellular signalling and is frequently dysregulated in human diseases such as cancer. There are thousands of sites of protein phosphorylation across the human proteome, mediated by over 500 protein kinases. Many protein kinases modify proteins that have a clear consensus sequence. Sites can also be readily identified at a whole proteome level using mass spectrometry. While the methods used to identify phosphorylation sites are efficient, exploring the functional consequences of each protein phosphorylation site is slow and laborious. Furthermore, we can predict the consequences of phosphorylation in only a limited number of cases. We are developing new approaches to improve functional annotation of phosphorylation sites so that we can focus on those most likely to have a biological function, such as in the regulation of protein-protein interactions.
Some phosphorylated proteins are recognized by “reader” modules such as BRCT and FHA. These interactions drive the formation of signalling complexes. Our recent work has extended this paradigm because we show that the phosphate can be recognized indirectly through a phosphorylation-induced helix that could be read by a wide range of proteins. This mechanism of phosopho-recognition, in which the sequence motif that governs the stabilization of the helix is distinct from that used to form the interface with the reader, provides greater flexibility to tune the molecular details of the interaction than canonical recognition motifs which are dominated by phosphate binding. Based on the structural details of the mechanism, we predict that it could be a general mechanism for the regulation of signalling events by protein kinases.
The aim of this project is to determine how phospho-dependent cryptic helix switches contribute to signalling pathways mediated by the Aurora family of protein kinases that regulate cell division. Phospho-proteomics studies have identified many potential substrates of Aurora kinases, but the structural and functional consequences of almost all of these sites are unknown. The first part of the project is to identify phosphorylation sites of Aurora that lie within a sequence that is predicted to have helical propensity, and that would be predicted to form a helix when phosphorylated. This will be done using databases of Aurora substrates, computational analysis of sequences, and molecular dynamics modelling of the peptides in solution. These predictions will be verified experimentally using Circular Dichroism and Nuclear Magnetic Resonance spectroscopy. These studies will identify Aurora substrates that undergo a disorder-to-helix transition upon phosphorylation. The final part of the project will be to explore how this structural change regulates the function of the substrate protein. Protein binding partners of the substrate in phosphorylated and unphosphorylated forms will be identified and the regulation of these interactions in cells will be probed using proximity ligation assays.
The project will be supervised by Prof Bayliss and Dr Paci, experienced scientists in structural, cellular and computational biology methods. The project exploits our state-of-the-art NMR spectrometers, part of the £17M Astbury Biostructure Laboratory (ABSL) investment in structural biology. Recent PhD graduates from the Bayliss group have published first author papers in high quality journals and progressed to research positions in academic or industrial laboratories.