Enhancing Near Native Protein Visualization Using Liquid Phase Electron Microscopy

Restricted (Penn State Only)
- Author:
- Berry, Samantha
- Graduate Program:
- Biomedical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 24, 2024
- Committee Members:
- James Adair, Outside Field Member
Joyce Jose, Outside Unit Member
William Hancock, Chair & Dissertation Advisor
Justin Pritchard, Major Field Member
Daniel Hayes, Program Head/Chair - Keywords:
- electron microscopy
structural biology
SARS-CoV-2
liquid phase electron microscopy - Abstract:
- Structural biology has made tremendous strides in revealing the structure-function relationship of proteins. High-resolution techniques such as nuclear magnetic resonance spectroscopy, X-ray crystallography, and cryogenic electron microscopy (cryo-EM) are capable of revealing detailed protein structure at the nanoscale. Cryo-EM boasts near-native specimen preparation, meaning that proteins are as close as possible to the native state without chemical fixation or staining. This technique has helped reveal intricacies of human health and disease, from neurodegenerative diseases such as Alzheimer’s to infectious diseases such as SARS-CoV-2. However, one major limitation of cryo-EM lies in the fact that specimens are prepared through vitrification, resulting in frozen, static state. Proteins are flexible and can undergo conformational changes dependent on environmental conditions, including temperature that may be affected by cryogenic temperatures. Therefore, dynamic studies on protein interactions and assembly are highly desired in the structural biology community. Liquid phase electron microscopy (LPEM) is capable of dynamic sample visualization in liquid, and it is widely used in physical science studies. LPEM involves the encapsulation of a liquid samples in hermetically sealed enclosures for EM imaging. Life science studies using the technique are more limited, as biological samples are inherently more sensitive to the electron beam and experiments using LPEM are challenging for these soft materials. The goal of this dissertation is to determine the feasibility of LPEM in elucidating native protein structures. This goal is accomplished by carrying out liquid imaging studies on patient-derived samples and applying single particle analysis (SPA) to evaluate 3D reconstructions obtained from the data. The development of LPEM for biological studies is in its early stages, and barriers such as accessibility and resolution limitations must be overcome to unlock its full potential. Sample grid preparation is a major component of EM experiments, and current liquid enclosures have their own unique challenges. For example, commercial liquid holders are often very costly and are difficult for novice microscopists to effectively use. Graphene liquid cells are another popular enclosure type, but while they are relatively cost-effective to make, the manufacturing process can be very tedious. Achieving high-resolution information can also be a challenge for LPEM since the particles are freely diffusing in a liquid layer, as opposed to stabilized in ice with cryo-EM. This dissertation introduces a novel method of imaging viral proteins using a microchip sandwich assembly for LPEM. The method is broken down into explicit steps, as well as the suggested microscope settings and SPA model parameters. To demonstrate capabilities of LPEM, multiple samples are examined, including adeno-associated virus, SARS-CoV-2, and rotavirus. By establishing a user-friendly, reproducible protocol, this work improves the accessibility of LPEM for other researchers studying biological specimens. This dissertation focuses on the use of LPEM to visualize components of SARS-CoV-2, the virus responsible for the COVID-19 global pandemic. Specifically, spike protein has a significant role in viral infectivity, and is the model protein for mRNA vaccine design. The following procedure was used. First, SARS-CoV-2 PCR positive patient serum was enriched for potential presence of native spike protein. Second, functionalized affinity capture monolayers were applied to EM grids and incubated with histidine-tagged ACE2 receptors for spike protein capture. Third, samples were imaged in liquid phase. Finally, EM images were processed using single particle analysis to generate 3-dimensional reconstructions. Overall, this dissertation explores the development and design of liquid phase electron microscopy for biological specimens while introducing methods for native protein visualization. By improving LPEM capabilities, protein dynamics at high spatial resolution in near native state may be observed in the future, contributing to discoveries in human health.