Computational Design And Modeling Of Helical Peptide Bundles In One-Dimensional Nanomaterials

Filamentous materials have a range of applications in biology and materials science. Control of their assembly is desired to develop materials for novel applications. Peptides are a desirable building block for one-dimensional nanomaterials because of their ability to self-assemble into defined structures. Intricate noncovalent interactions dictate the structures of peptide assemblies and can be made even more powerful by the addition of covalent links between peptides. Previous work has generated rigid rod nanomaterials with long persistence lengths (tens of microns) based on the self- assembly of non-natural peptides that form helical coiled coil bundles. These peptide bundle building blocks were computationally designed de novo and functionalized with complementary functional groups to covalently link the building blocks. However, the atomic structure of the bundle-bundle interface and the determinants of rigidity are not well understood. In this work, we seek to develop an atomistic understanding of these existing one-dimensional peptide assemblies and suggest alterations to the bundles that should modulate the properties of the nanostructures. We also develop greater versatility in these types of peptide nanostructures and building block units. We use probabilistic computational protein design to identify peptide sequences likely to fold into target structures. A series of peptides of different lengths are introduced which can be used to change the periodicity of peptide building blocks in peptide nanorods. In addition, a series of peptide bundles are designed to provide differential display of reactive groups at the termini of a building block. These computational studies and their experimental characterization aim to improve our understanding of peptide-peptide interactions and designed self-assembly of biomolecular nanomaterials.