Computational studies on biological membranes and viral capsids

Chan, Chun Kit

Permalink

https://hdl.handle.net/2142/115327 Copy

Description

Title

Computational studies on biological membranes and viral capsids

Author(s)

Chan, Chun Kit

Issue Date

2022-02-11

Director of Research (if dissertation) or Advisor (if thesis)

Tajkhorshid, Emad

Doctoral Committee Chair(s)

Aksimentiev, Aleksei

Committee Member(s)

• Singharoy, Abhishek
• Gennis, Robert

Department of Study

Physics

Discipline

Physics

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Brownian dynamics
• Molecular dynamics
• Diffusion
• Bioenergetics
• Viral capsid

Abstract

Biological membranes are ubiquitous and often involved in complex processes. A well-known example are bioenergetic membranes, which use an array of integral membrane proteins and their complexes, small molecules, as well as soluble proteins to produce adenosine triphosphate (ATP), the molecular currency of energy in the cell for metabolism. This production of ATP requires transmembrane gradient of protons, which is generated by coupling proton translocation across the membrane to an electron transport chain. The electron transport chain employs small, lipophilic molecules and, often, small, soluble proteins to shuttle electrons between the integral membrane proteins. Disruption of electron shuttling mediated by these small molecules and/or soluble proteins frustrates the electron transport chain and subsequently prohibits the generation of ATP. Thus, elucidating the interaction between these electron shuttling agents and their redox partners is key to understanding conditions that facilitate/hinder cellular energy transduction. The diffusion of the electron shuttles is a stochastic pro- cess and therefore susceptible to environmental variations, including fluctuations in the environmental salinity, changes in the local lipid composition, and conformational changes of the redox partners. Also, the time span involved in their diffusion can range from a few hundred nanoseconds to several micro/milliseconds. Thus, computational techniques that can account for the important atomic details of electron shuttles and their redox partners, and at the same time can simulate their stochastic diffusion on relevant timescales are required for properly capture their mechanism. Molecular dynamics (MD) and Brownian dynamics (BD) are tools whose combination may satisfy these criteria. Employing MD and/or BD simulations, I have characterized the inter- actions between three commonly studied electron shuttles and their redox partners. These include cytochrome c2 (cyt. c2), quinones, and cytochrome c (cyt. c), and their redox partners cytochrome bc1 complex (bc1), cy- tochrome bo3 (cyt. bo3), and the yeast supercomplex (SC). In the first case, I illustrate the involvement of anionic lipids in cyt. c2 - bc1 interactions, and how sensitive it is to the environmental salinity whose variation can potentially enhance cyt. c2 - bc1 association when the redox states of the two proteins are complementary. In the second case, I demonstrate the role of a particular transmembrane helix of cyt. bo3 in stabilizing the protein interaction with quinones and in facilitating proton translocation from the cofactors to the bulk solu- tion. In the third case, I show the role of a flexible subunit in the SC in alleviating undesirable impacts from lipids on the diffusion of cyt. c as well as in regulating the affinity between cyt. c and individual subunits of SC during the cyt. c - mediated electron transport. Finally, I have used a combined MD/BD approach to study viral capsids. I have examined platelet factor 4 (PF4), which interacts with the capsid of ChAdOx1, a viral vector used in vaccines produced by AstraZeneca (AZ) to combat the SARS-CoV-2 virus. I illustrate that PF4 interaction with ChAdOx1 is electrostatically driven, and supported by their geometric complementarity. I have also characterized residues in ChAdOx1 that serve as hotspots for PF4 - ChAdOx1 association. With the help of experimental virologists, we use the information to devise strategies to interrupt PF4 - ChAdOx1 association and to further reduce the chance for the rare blood-clotting complication of AZ vaccines.

Graduation Semester

2022-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2022, Chun Kit Chan.

Owning Collections