Probing protein-protein interactions by HDX-MS and computational approaches

  • Nathan Gittens

Student thesis: Doctoral Thesis

Abstract

In Chapter 1, an introduction to the field of hydrogen-deuterium exchange mass spectrometry (HDX-MS) is provided, which provides the foundation of much of the content within the thesis. Sections on supervised learning and molecular dynamics are also provided. In Chapter 2, a framework for the discrimination of BRD4 inhibitor binding modes is shown using hydrogen-deuterium exchange mass spectrometry and supervised learning. Issues with BRD4 digestion were tackled by the exploration of multiple different proteases. In particular, the formation of an interface between BRD4 bromodomains was shown is a key factor in being able to separate the bivalent class of inhibitors. The experimental challenges of screening large numbers of compounds is also addressed. In Chapter 3, HDX-MS and molecular dynamics simulations were used to investigate the PPIs induced by the binding of a VHL. Interestingly, HDX-Ms showed significant protection of the proteolysis-targeting chimera (PROTAC) bound to Bcl-xL and α2-helix on formation of the ternary complex, revealing changes in dynamics not previously observed form available structural data. HDX protection factors were also modelled from molecular dynamics data in an attempt to relate the experimental and computational data. Finally, in Chapter 4, the modelling of PROTAC-mediated ternary complexes is considered by protein-protein docking experiments. PROTAC-mediated ternary complexes may require the PROTAC molecule to adopt energetically disfavourable conformations to facilitate protein-protein interactions. This poses issues for modelling where linker conformations are biased towards the lowest energy structures. This chapter demonstrates the importance of building in the PROTAC linker into protein-protein complexes generated by computational docking. Data from HDX-MS experiments and molecular dynamics simulations may then be used to help to identify those ternary complex models that are consistent with the available experimental data.
Date of Award2 Jun 2022
Original languageEnglish
Awarding Institution
  • University Of Strathclyde
SponsorsUniversity of Strathclyde
SupervisorDavid Palmer (Supervisor) & Craig Jamieson (Supervisor)

Cite this

'