Modeling and Design of Peptides to Control Biomineral Nucleation and Growth

Embargo until
Journal Title
Journal ISSN
Volume Title
Johns Hopkins University
Biological organisms use biomolecules in combination with compartmentalization and ion sequestering strategies to craft intricately structured hard tissues such as nacre, tooth, and bone. A molecular-level mechanistic understanding of these biomineralization processes would enable biomimetic strategies to synthesize custom nano-structured materials and would aid in developing treatments for biomineralization-related diseases. With the advent of in situ atomic force microscopy and single molecule force spectroscopy, detailed kinetic and thermodynamic information on biomineralization processes is now available. Molecular simulation can connect this information to the atomic structure of proteins adsorbed on mineral surfaces and enable the rational design of molecules to control mineral growth and nucleation. A previously developed method, RosettaSurface, served as a starting point for protein structure prediction on mineral surfaces. In this dissertation, I detail my work expanding upon the RosettaSurface framework, comparing the results of RosettaSurface with experimental measurements on peptide-biomineral systems, and rationally designing peptides to control the growth and nucleation of biominerals. First, I describe the computational details of the RosettaSurface algorithm, using the osteocalcin/hydroxyapatite system as a model. Next, I compare the results of the RosettaSurface algorithm with an experimental benchmark of kinetic and thermodynamic measurements on peptide-biomineral interactions taken from atomic force microscopy. The RosettaSurface algorithm successfully identifies which mineral face and step edges will bind peptides the strongest; however, the algorithm struggles to predict the correct rank order of binding for multiple peptides to the same face or step edge. Next, I detail my work in studying the adsorption of single amino acids on vaterite mineral surfaces and connect my computational results to experimental observations of chiral hierarchical structures in vaterite crystals. Finally, I design peptides intended to control calcium carbonate nucleation and growth by binding both flat surfaces and mineral step edges, targeting specific kinetic and thermodynamic mechanisms for growth and nucleation.
Biomineralization, peptide, mineral, crystal growth, crystal nucleation, molecular simulation, Monte Carlo, benchmark, protein design