Date

2017

Document Type

Dissertation

Degree

Doctor of Philosophy

Department

Chemical Engineering

First Adviser

Jagota, Anand

Other advisers/committee members

Berger, Bryan W.; Mittal, Jeetain; Vezenov, Dimitri; Vavylonis, Dimitrios

Abstract

Molecular adhesion is the basis for many complicated biological processes. In particular the role of regulated adhesion of the vesicle to the plasma membrane in exocytosis as part of synaptic transmission is crucial. This adhesion must overcome the long-range electrostatic and hydration repulsion in a mediated fashion. Complications within this neurological process can lead to serious diseases and disorders such as schizophrenia and botulism. Little is understood about the mechanistic details of this process. Therefore, developing fundamental knowledge is invaluable to the design of treatments and therapies. Many molecular players have been identified to take part in this process, but we focus on SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptor) because that is the central conserved player. SNARE proteins are the workhorses of this adhesion and therefore developing an understanding of their role in the process is important. This is a very complex system, so in addition to experiments, modeling of the system is also required, and there are several options available. Because of the combination of long length scale (10s of nm) and time scales (1-10s of µs, to seconds), it is difficult to study this system using all-atom simulations. Therefore residue based coarse-grained (CG) Brownian dynamics simulations are used that are validated by both experimental data and detailed all-atom simulation results. The CG model of SNARE included an elastic spring-network model for intrahelical interactions and chemically specific Miyazawa and Jernigan potentials for helix-helix interactions.Using this CG model this thesis investigates two underlying questions dealing with synaptic vesicle to membrane docking/fusion that have yet to be definitively answered: (1) how many SNAREs are required in this process? and (2) how do SNAREs assemble? The force-displacement relationship for the unzippering of SNARE was determined using CG displacement control simulations. These results were combined with a continuum model of the vesicle/membrane including electrostatic and hydration repulsion to predict that 1 SNARE can bring the vesicle to within 3 nm of the membrane. This docking distance can be reduced as additional SNAREs are added. However, adding more than 4-6 SNAREs increases the minimum distance between the vesicle and plasma membrane. The vesicle was never brought closer than 2nm to the membrane, suggesting that SNAREs alone are not sufficient for vesicle to membrane fusion, that their principal role is docking.Next, the SNARE assembly process was studied using the CG model. Two models for SNARE assembly are proposed in the literature: (1) Munc18 acts as a template for SNARE assembly by holding SNARE in a semi-zippered conformation and promoting helicity, and (2) SNARE serves as a self-template for zippering and assembly. Force control unzippering simulations were performed to set up initial states for assembly simulations. Several simulations were performed mimicking the two hypotheses. We find that assembly time grows exponentially with how far the SNARE has been unzippered, and this assembly time is increased even more with the degree of unfolding. We find that helical SNAREs assemble rapidly, however it is known (from experiments) that unstructured, completely disassembled SNAREs assemble in ms to s timescales. Therefore, Munc18 or another chaperone would most likely be required to promote a half-zippered SNARE state prior to assembly. We tackled another system of importance due to its deadly and contagious characteristics: the Ebola virus (EBOV) internalization by the host cell which is an adhesion problem of comparable length and time scales. A better understanding of this system will lead to the development of possible vaccines/therapies. Again this system included two negatively charged surfaces, EBOV and the host cell membrane. We developed an analytical model to investigate the parameters required for EBOV ingestion into the host cell. We studied this system at two limits: (1) membrane bending dominates in resisting deformation and (2) membrane tension dominates in resisting deformation. From the membrane bending limit study, a dimensionless parameter representing the ratio of membrane bending stiffness to adhesion was found that determines whether EBOV will be engulfed into the host cell. From the membrane tension limit study we also extract a dimensionless parameter representing the ratio of membrane tension to adhesion that determines whether engulfment will occur. In particular, these dimensionless parameters can be used to relate single-molecule force spectroscopy measurements to the behavior at the length scale of the full virus.

Share

COinS