Andrey O. Demkiv: Computational Modelling of Protein Dynamics, Specificity and Evolution

Abstract

Proteins are the foundational molecules driving nearly all biochemical processes essential for life. Their ability to catalyse reactions with high specificity and efficiency is key to biological function and holds significant potential for applications in drug discovery, disease treatment, and green chemistry. However, understanding the intricate mechanisms underlying enzyme catalysis and ligand binding requires not only structural insights but also a deep understanding of protein dynamics, which can be challenging to capture experimentally.

This thesis leverages various computational methods to explore the dynamic behaviour of proteins, providing critical insights that complement experimental approaches. We developed an implementation of a replica exchange enhanced sampling technique to model enzymatic reactions with the EVB approach - Q-RepEx, which allows us to study the reaction within the context of larger protein motions. EVB and MD simulations enabled us to uncover the catalytic promiscuity of the lactonase GcL, revealing its ability to utilise multiple pathways depending on the substrate—a feature that could be exploited for designing selective quorum quenchers. Additionally, our MD studies on disembodied P-loop peptides provided new perspectives on their role as potential evolutionary precursors of the P-loop NTPase family, challenging existing hypotheses on their minimal functional ancestor. Overall, this work underscores the role of computational methods in advancing our understanding of protein dynamics and function, offering valuable insights that are essential for both fundamental biology and the development of novel biotechnological applications.