Hierarchical design of metal-mediated peptide materials

Abstract

Metal–organic frameworks (MOFs) are a subset of coordination polymers often made up of an organic linker, usually a rigid small molecule, and either a metal ion or metal cluster. Despite key advances in the development and functionalization of MOFs, their prediction and rational design has remained a challenge. New methodologies have allowed for limited computational prediction of MOFs made from highly ordered metal clusters and short, very rigid linkers, however, the computational design of MOFs with more flexible linkers like peptides has not yet been realized. Furthermore, few MOFs discovered that use these more flexible but highly functional peptide components exist, making the de novo design of such structures a new and exciting endeavor.To help address this challenge, I worked towards designing MOF-inspired materials from de novo cyclic peptides. In the second chapter of this thesis, I will describe the design of symmetric cyclic peptides for use as structured modular precursors. Here, I designed and characterized 94 peptides in a variety of Cn symmetries, ranging from 6-24 residues in length. In the third chapter of this thesis, I report our efforts towards developing two different methods for the design and characterization of larger crystalline structures from these ordered cyclic peptides. The first was a bottom-up approach, developing computational methods for the design of new peptides that when coordinated to metal in a defined fashion can be geometrically compatible with four cubic space groups. The second was a top-down experimental approach, developing new ways to screen and test crystal growth assembly conditions in a high-throughput manner. Here, we selected 42 peptides for crystallization and solved the structures of three crystals that incorporated metal ions into their structure. Although these three observed crystals deviate from our computational design models, they improved our understanding of different features of the crystallization processes, including local metal coordination geometries, side chain configurations, and crystal packing density. All of these lessons will be incorporated into our computational methodology for improved crystal prediction and design. Overall, although the development of computational methods to design MOF-inspired peptide materials is still in its infancy, the methods outlined here provide a basic framework upon which future efforts can build and expand.