(197j) A High-Throughput Computational Framework for Investigating Protein-Polymer Bioconjugates with Molecular Dynamics Simulations

Protein-polymer bioconjugates are a class of hybrid biological molecules that strategically utilize the natural specificity of proteins and the synthetic modularity of polymers. Recently, the design and assembly of protein-polymer bioconjugates have gained much attention due to the potential applications of these bioconjugates in drug therapeutics, biomineralization, and biosensing. The conjugation of polymers to protein surfaces can enhance key properties of the biomolecule such as its structural stability and affinity for specific targets, while maintaining its natural biological functions. To optimally design and synthesize these hybrid molecules, the underlying mechanisms that influence their enhanced thermodynamic and kinetic properties must be understood at the molecular level. Computational approaches such as molecular dynamics (MD) simulations provide an excellent entry point for such studies, via the development of accurate and compatible molecular models for studying broad variations of the protein-polymer bioconjugate system.

Herein, we showcase the development of a computational framework for systematically building and testing models of protein-polymer bioconjugates that can be readily scaled for high-throughput processing. This methodological workflow encompasses various in silico tools such as Python, C++, CGENFF, Gaussian, Avogadro, ChemDraw, VMD, and GROMACS. Specifically, we built and tested this framework on bovine alpha-chymotrypsin and hen egg-white lysozyme conjugated with sulfobetaine methacrylamide through atom transfer radical polymerization (ATRP) initiators, as well as on human interleukin 12 (IL-12) conjugated with activated polyethylene glycol (PEG). The framework itself, when combined with fundamental analysis methods such as MD, will support the rational design of novel bioconjugates with improved efficacy in terms of function and stability. To this end, we conducted MD simulations of the above bioconjugates in different chemical and biological environments, namely: (1) ATRP-initiated bioconjugates interacting with silica at various temperatures, and (2) PEG-ylated bioconjugates interacting with phospholipid membranes. In the former case, we sought to elucidate structural changes in the bioconjugates that could affect their adsorption to nanoparticle surfaces in drug delivery schemes. In the latter case, we sought to identify adsorption free energy differences between the bioconjugates and their non-conjugated counterparts to inform the use of these bioconjugates in protein therapeutics applications. Finally, through the development of this computational framework, we gained new atomistic insights into advanced protein functions and an improved understanding of the molecular driving forces governing bioconjugate stability.