Abstract
Targeted covalent inhibitors (TCI) are compounds that are designed to exert their therapeutic effect through the formation of a covalent bond with a biological target. In spite of several perceived benefits of pursuing a covalent mechanism of action, concerns about off-target reactivity and selectivity remain. The most efficient way to accurately characterise covalent reactivity by computational approaches can be challenging, and the mechanism of chemical inhibition of high profile drug targets for some protein kinases remains an unanswered question. Computational modelling of covalent reactivity provides an attractive approach for investigating the determinants of reactivity between a covalent inhibitor and its target and can aid in the design of safer and more selective covalent drugs.Simple ligand only reactivity metrics of covalent reactivity including proton affinity (PA) and reaction energy calculations were investigated with quantum mechanical (QM) methods. However, limitations in the predictive power of these methods for pharmaceutical lead-like compounds were identified, which led to a focus on reactivity assessments ‘in situ’. A benchmarking study of the most appropriate semi-empirical and density functional QM methods confirmed that specific methods should be used to accurately model thiol reactivity.
A comprehensive reactivity study was carried out for the covalent inhibitor ibrutinib that targets a non-catalytic cysteine residue in Bruton’s tyrosine kinase. Constant pH molecular dynamics simulations were used to calculate the pKa of Cys481 in BTK and identified the neutral thiol group to be the most likely protonation state at physiological pH. Combined quantum mechanics/molecular mechanics (QM/MM) calculations in combination with umbrella sampling simulations were used to assess chemical inhibition pathways in BTK and led to the identification of a novel mechanistic pathway in BTK that is distinct from other protein kinases.
| Date of Award | 23 Mar 2021 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Adrian J Mulholland (Supervisor) |