Macromolecular machines in motion: Insights from long-timescale molecular dynamics and time-resolved cryo-EM

Thesis

Macromolecular machines undergo dynamic conformational changes across multiple timescales to execute their biological functions, yet conventional structural biology techniques typically capture only static snapshots. This thesis presents a multi-faceted approach to overcome these fundamental limitations by combining cutting-edge computational and experimental methodologies to investigate protein dynamics from microseconds to minutes. First, I present the most comprehensive molecular dynamics study of the human mitochondrial LONP1 protease to date, extending the timescales of large-complex molecular dynamics simulations and revealing transformative insights into the dynamic interplay between functional domains that governs enzymatic activity. Next, I expand experimental access to rapid conformational changes through the development of “Mix-it-up” (MIU), a cost-effective time-resolved cryo-EM platform that enables millisecond timescale studies of biochemical reactions. Finally, I employ time-resolved cryo-EM to elucidate the mechanistic details of early DNA binding by Polymerase θ, successfully capturing previously inaccessible early conformational ensembles that function in DNA repair. Together, these studies bridge the temporal gap between computational and experimental approaches, advancing our understanding of how macromolecular machines coordinate their functional cycles through exploration of diverse conformational landscapes. These mechanistic insights lay the groundwork for understanding how disruptions in protein dynamics drive disease and for guiding the design of targeted therapeutic interventions.

Authors: Alexandrescu, L

• DOI: 10.5287/ora-4rxd9p1j4
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Subjects: Biophysics