Investigating Genetically Engineered Virus-Like Particles Using Molecular Dynamics Simulations for Enhanced Vaccine Development

Virus-like particles (VLPs) mimic the structure of viruses without containing viral genetic material, making them safe and highly immunogenic platforms for vaccine development. Genetic modifications can expand the use of VLPs, such as the insertion of antigens to target various diseases. However, designing stable VLP-based vaccines through genetic engineering is challenging and unpredictable due to the complex nature of protein interactions involved in VLP assembly and the unknown effects of genetic modifications on VLP scaffolds. These challenges have limited the development of VLP-based vaccines. This project aims to overcome these limitations by leveraging computational tools to investigate and predict the overall stability of genetically modified VLP-based vaccines. This project began by constructing and establishing purification pathways for chimeric VLPs incorporating hydrophobic insertions using Hepatitis B core (HBc) and Human heavy chains ferritin (HFn). Initial investigations revealed that models predicted by AlphaFold2 were stable as initial structures, providing a basis for accurate simulation studies. Subsequently, the self-assembly properties of eight HBc derivatives were investigated through experimental validation and molecular dynamics (MD) simulations from a thermodynamic perspective. It was found that HBc derivative assembly failures could result in different morphologies as dissociative dimers or large multi-subunit polymers. Instability in dimer formations or improper (reduced) intradimer distances were identified as possible explanations of assembly-incompetent dimers. Polar solvation energies were found to play a critical role in causing assembly-incompetent dimers. MD simulations on dimers can offer preliminary predictions regarding the assembly properties of HBc derivatives, enhancing vaccine design efficiency by reducing the risk of self-assembly failures in engineered proteins. To simulate the surface physical properties of VLP assemblies more accurately, large partial VLP models, each consisting of 17 chains, were constructed for the HBc chimeric model vaccines using the wild-type (wt) HBc assembly template. This approach allowed for a comparison of the effects of inserted epitopes and insertion strategies on HBc modifications. Simulation results showed that the insertion site significantly affects the conformational stability of the VLP and secondary structures of inserted epitopes. Separate insertions at different sites on HBc, i.e. at the N-terminus and major immunodominant region (MIR), caused instability, while serial insertions at the MIR hindered VLP assembly. However, inserting a different epitope of the same length at the MIR led to successful assembly, indicating that the amino acid sequence is more critical than length. These insertions also modulate surface hydrophobicity and overall stability, which could be accurately predicted by MD simulations. The methodology was further applied to the Murine polyomavirus (MPV) capsid protein VP1. Study results showed that binding energies calculated from MD simulations correlate well with the thermal stability of genetically engineered VP1 capsomeres. Larger foreign insertions lead to less compact backbones and increased flexible regions, resulting in reduced thermal stability in VP1 capsomeres. Van der Waals interactions were identified as key stabilizing factors of the capsomere structures. These findings suggest that the influence of inserted antigens on the viral structural protein is complicated (depending on both size and sequence), and therefore using small protein tags and minimizing structural disruption is a more effective strategy for antigen replacement on VLP platforms. The integration of computational modeling and simulation with research validation in this thesis represents a significant advancement in the field of VLP vaccine development. The findings not only improve our understanding of genetically modified VLP derivatives but also provide a methodological framework for reducing development costs and accelerating the response to emerging infectious diseases. Future research should continue to explore the integration of computational methods with traditional vaccine development processes to establish a new paradigm that enhances the efficiency and precision of vaccine design.

Author dc:creator

• Luo, Hong

Advisors dc:contributor.advisor

• Bi, Jingxiu
• Zhang, Songping (Chinese Academy of Science, Institution of Process Engineering)

• Virus-like particles (VLPs)
• Molecular dynamics (MD) simulations
• AlphaFold
• Genetic engineering
• Heptitis B core (HBc) protein
• Murine polyomavirus virus (MPV)
• Protein engineering
• Vaccine development

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/2440/144833

OAI identifier oai:identifier

oai:digital.library.adelaide.edu.au:2440/144833