Efforts Towards Drugging The Undruggable: Novel Approaches To Targeting Protein Tyrosine Phosphatases (PTPs)

Abstract

There is mounting evidence in the literature that implicates the malfunction of protein tyrosine phosphatases (PTPs) in a diverse range of diseases including cancer and diabetes. Consequently PTPs have been identified as important therapeutic targets of interest and have thus been the focus of countless drug discovery investigations. A multitude of approaches have been deployed resulting in a wealth of PTP inhibitors being reported in the literature. It may come as some surprise therefore, that no PTP inhibitor to date has made it to market. This high rate of attrition can be attributed to key structural features associated with PTPs and their active site. Firstly, the high homology of PTPs conserved active site makes it hard to create inhibitors that are selective for a given PTP, which can result in toxicity. A further obstacle to PTP inhibitor development is that in order to have strong interactions at the active site and mimic the natural phosphate containing substrate, the inhibitor often has to contain a highly negatively charged or polar group, resulting in inhibitors with poor cell permeability. Despite a wide variety of approaches being implemented no PTP inhibitor has made it to market; this fact together with the structural restraints surrounding the active site has resulted in PTPs being classified as undruggable. Given the therapeutic potential of PTPs, there is a clear need for selective, cell active chemical inhibitors to probe PTPs validity as drug targets. This thesis explores the development of new approaches aimed at tackling these issues that have plagued PTP inhibitor development to date. One of the main focuses of this thesis surrounds the development of a user-friendly semi-automated pipeline for structure-based virtual screening that utilises existing tools AutoDock and OpenBabel together with software developed in-house, to create an end-to-end virtual screening workflow ranging from the preparation of receptor and ligands to the visualisation of results. An essential feature of this pipeline was the automatic generation of ligand efficiency indices (LEI). LEIs are important metrics that can aid the development of drug-like compounds, however, their application to the PTP realm to date has been limited. A further feature that was incorporated into this pipeline was the option to use a process known as blind docking, an effective tool for the identification of functional pockets, however, this had yet to be utilised in the PTP field. To validate the use of blind docking of compound libraries as a method for the identification of functionally relevant pockets in PTPs a case study was undertaken with the human protein tyrosine phosphatase PTP1B. Blind docking successfully identified several sites of known functional importance including the active site and allosteric site. Further to this, several sub-pockets were identified notably in the vicinity of the allosteric site, but these have yet to be exploited. Assessment of the LEI plots of binders associated with the active and allosteric sites identified the allosteric site to reside in a more drug-like chemical biological space. To further probe the value of application of LEI values to the PTP realm, an investigation that focused on the development of HePTP inhibitors with improved LEI values was undertaken. An 14 assessment of the druggability of different conformational states of HePTP identified a novel pocket (P2) only present in the open state of HePTP. Pockdrug and LEI plot analysis of this pocket and the active site (P1) found P2 to possess a higher degree of druggability. A series of compounds were then synthesised to facilitate a structure activity relationship analysis of inhibitors that target this P2 pocket, resulting in sub-micromolar inhibitors that had LEI values greater than that of the best current cell active HePTP inhibitors in the literature. The binding mode of these compounds was also investigated with molecular docking and NMR stud

Date of Award	31 Dec 2019
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Lydia Tabernero (Supervisor)