Coiled coil peptides as monomers for targeted self-assembly and crosslinking of nanostructured materials

Abstract

Proteins make excellent building blocks for materials creation due to their hierarchical self-assembly, sequence programmability, tunability, and diverse functionality. The sequence-defined precision of proteins enables programmable structure-function relationships, affording high levels of control of intra- and inter-molecular interactions. The ability to design, control, and tune the supramolecular self-assembly of peptides, which are short-length proteins, makes them well-suited for the design of nanostructured materials via a bottom-up approach. In this dissertation, I explore three fundamentals for the future design of peptide-based materials using coiled coil peptide ‘bundlemers’ as monomers: controllable covalent crosslinking, tunable and controllable physical assembly, and a scalable approach to peptide synthesis. The methods developed herein provide a platform for the bottom-up design of ordered materials made from peptide building blocks. ☐ Coiled coil peptide ‘bundlemers’ are computationally designed 29 amino acid peptides that self-assemble into tetrameric coiled coils in aqueous solution. These self-assembled structures resemble a cylinder-like particle with approximate dimensions of 2 nm in diameter and 4 nm in length. The particles are highly stable due to their well-designed hydrophobic core, and site-specific modifications can be easily made to the surface of the particles to allow for tunable interactions and controllable conjugation. ☐ First, bundlemer sequences were selectively modified with allyloxycarbonyl (alloc) protected lysines and cysteines for use in a thiol-ene photo click chemistry reaction to form crosslinked peptide networks. Results demonstrated that the degree of network crosslinking could be finely tuned by manipulating the number of crosslinkable sites and the position of those sites, providing control over intra- and interbundle crosslinking via the protein-like display of chemistry on the periphery of the particles. Adding the alloc protecting groups without cysteines resulted in unexpected particle self-assembly into highly ordered porous lattice structures. These lattices were studied experimentally using transmission electron microscopy (TEM) and small-angle x-ray scattering (SAXS), and lattice packing was proposed via machine learning optimization techniques. The proposed packing highlighted the importance of side-to-side interactions in the lattice self-assembly, which are driven via hydrophobic interactions between alloc side chains. ☐ The unexpected self-assembly of alloc-modified bundlemer sequences into intricate lattices inspired a new study utilizing new natural and non-natural hydrophobic amino acid side chains to drive self-assembly into lattice particles via similar interactions. I hypothesized that placing different hydrophobic side chains in identical positions along the peptide backbone would induce self-assembly into porous lattice particles, but with subtle differences in nanostructure. The same parent sequence was selectively modified with five new non-natural and natural hydrophobic amino acid side chains, and all modified sequences were observed to form porous lattice particles in solution that were distinctly different than the previously reported structures. New sequences were studied experimentally using TEM and SAXS and computationally via simulations and modeling to provide insight into the intra- and inter-particle interactions that drive lattice formation. These results highlighted the versatility of bundlemer design in forming precision nanoporous structures via hydrophobic interactions, the importance of spatial display of chemical functionality, and the role of molecular interactions that drive the formation of these nanostructures. ☐ Hydrophobic self-assembly of bundlemers into ordered lattices has proven tunable and robust for mixed charge peptide sequences. I hypothesized that making identical modifications to any bundlemer parent sequence should enable self-assembly into lattice particles via hydrophobic interactions. Due to their different amino acid composition, single charge bundlemer sequences were chosen for modification, which are unique because they exhibit liquid crystalline behavior at high concentrations in aqueous solution, driven by electrostatic interactions. Modifying single charge sequences with alloc protected lysines enabled the functional design of a polymorphic peptide particle capable of forming hexagonal columnar liquid crystals in aqueous suspension and crystal lattice particles in concentrated salt solutions. Uniquely, despite having a very different amino acid composition, one of the new modified sequences had nearly identical bundlemer packing to the previously studied mixed charge alloc lattices, as observed via structure factor in SAXS. The observation of identical structures across two very different peptide sequences highlights the importance of the display of the alloc groups on the periphery of the bundlemer particles and the specificity of their interactions in forming ordered lattices. ☐ One overarching challenge of my research was the limitation in the amount of material for bulk materials crosslinking and characterization. To alleviate this, I turned to recombinant expression in E. coli to synthesize large quantities of peptide. A protocol for synthesizing and purifying coiled coil peptide multimers has been developed for scalable synthesis, focusing on the tagless expression and column-less purification of these multimeric proteins. Without chromatography, solution conditions such as pH, temperature, and salt concentration are to manipulate solubility and drive separation via controllable phase behavior. This dissertation includes an in-depth discussion of the design, synthesis, purification, and characterization of the coiled coil multimers and future directions for recombinant expression.