A simulation approach to cryptic pocket discovery in viral envelopes

Abstract

Hotspots are protein surfaces that preferentially bind drug-like molecules and their detection is crucial for rational drug design. Molecular dynamics (MD) simulation-based approaches can be used for hotspot detection, particularly for sites that are usually hidden in static protein structures (known as cryptic pockets). One such group of methods includes mixed solvent probe-mapping, where small drug-like molecules or probes are added to the simulation system and preferentially bind to protein hotspots. The dynamic aspect of simulations allows for transient opening and closing of cryptic sites and can accommodate binding of probes into the pocket cavity. However, membrane-bound proteins have long been excluded from probe-mapping efforts as small probes tend to sequester into the membrane associated portion of the protein. In this work, I develop a benzene-mapping method for cryptic pocket discovery in membrane-bound proteins and demonstrate its effectiveness by applying the methodology to viral envelope systems. Benzene-mapping on SARS-CoV-2 spike (S) proteins successfully recapitulates existing pockets and presents a novel pocket located underneath a functionally relevant loop which is involved in modulating the stability of the S protein trimers as well as in formation of S protein multimers. Benzene-mapping is also applied to the envelopes of flaviviruses (dengue, yellow fever, and Zika virus) which, despite similarities in sequence, demonstrate strain-specific dynamic behaviour. I elaborate on the characteristics of an established detergent binding site and reveal a previously undiscovered cryptic pocket located on domain interfaces that contains a conserved cluster of charges implicated in the pH-dependent conformational change crucial for the viral life cycle. Collectively, this thesis presents the development and applications of the hotspot mapping method and provides a unique link between conserved cryptic pockets and their functional roles.

Date of Award	14 Dec 2021
Original language	English
Awarding Institution	The University of Manchester
Supervisor	James Warwicker (Supervisor) & Sam Hay (Supervisor)