Using molecular dynamics to probe protein conformational transition mechanisms at the membrane surface

Far from being a passive demarcation between the cell and the environment, the cellular membrane is a vibrant and dynamic ecosystem. Membrane proteins constitute an integral piece of this ecosystem, allowing for communication between the cell and its environment that is necessary for adaptation and robustness of the cell. Often, membrane proteins are expressed in a default inactive state requiring a signal from either the extracellular milieu or from the cell itself to be activated, undergoing a significant structural shift in order to perform its function. Due to the transient nature of these activated states, however, it has been difficult to characterize the structural transitions needed to activate peripheral and integral membrane proteins. The work presented herein leverages molecular dynamics (MD) simulations as a method to explore and characterize the dynamics of both peripheral and integral membrane proteins in the context of the membrane. While MD simulations have unrivaled spatiotemporal resolution, the timescales accessible to MD simulations often prevents characterization of spontaneous membrane insertion of peripheral proteins. Many biasing protocols have been developed to observe the binding of peripheral membrane proteins on the timescales routinely accessible in MD simulations. These protocols, however, inherently bias the final structure derived from the study. A novel membrane mimetic model, termed the highly mobile membrane mimetic (HMMM), is presented which shows expedited lipid dynamics and decreased time to insertion for peripheral proteins. The physicochemical properties of the model are fully characterized and compared to conventional membranes, showing good agreement with key elastic, electric, and energetic properties of the membrane. Moreover, the HMMM model is utilized to simulate the unbiased binding of a peripheral protein, the talin F2F3 subdomain. In addition to characterizing the membrane-bound form of this protein, which has remained elusive to multiple crystallization studies, the simulations describe a large, interdomain conformational change that reconciles previous discrepancies between biochemical and structural studies. In addition to the studies on peripheral proteins, MD simulations were utilized to probe the dynamics of a ligand-gated ion channel (LGIC) in response to anesthetics, which antagonize channel function. There is significant debate over the proper anesthetic binding site in LGICs, with extracellular, transmembrane, and ion conduction pore mechanisms proposed. A variety of molecular dynamics techniques, starting from parameterization of the anesthetics to free energy calculations, have been utilized and show that anesthetics bind to a transmembrane site identified in crystal structures. An additional and previously undescribed binding site was discovered in the simulations that stabilizes the anesthetic in the previously unidentified site as shown by free energy and mutation studies. Moreover, conformational changes in the transmembrane accompanied anesthetic binding that immensely increase the energy barrier to sodium conduction.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Biophysics & Computnl Biology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Arcario, Mark

Contributors dc:contributor

• Tajkhorshid, Emad
• Morrissey, James H.
• Sligar, Stephen G.
• Aksimentiev, Aleksei

• Molecular Dynamics
• Membrane Protein
• Coagulation
• Talin
• Integrin
• Anesthetics
• Gloeobacter violaceus ligand-gated ion channel (GLIC)
• Ion Channel

Statement dc:rights

• Copyright 2014 Mark Arcario

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/72776

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/72776