Engineering highly ordered two-dimensional arrays and cages mediated by noncovalent protein-protein interfaces

Thesis (Ph.D.)--University of Washington, 2015-12Natural proteins often assemble into higher order structures by symmetric assembly of many copies of the same protein subunit through weak, non-covalent interaction to perform their tasks. Some examples of higher order structures include cages used for exocytosis, fibers used for structural stability and channels used for the flow of ions and water in and out of the cell. Two-dimensional (2D) protein assemblies also occur in nature, either assembled in or on lipid membranes. These 2D assemblies are usually made for large-scale transport of ions or water or as a cellular barrier against antagonists. Assemblies in 2D have been a challenge to engineer through design in the past due to the complexity of the proteins and their propensity to misfold. Previous successful attempts used different avenues to create 2D arrays, for example using metal-mediated assembly or by using fused proteins but there are no examples of 2D assemblies being designed to have non-covalent interaction in their assembly as similarly seen in nature. We aimed to design such protein assemblies in order to allow for new avenues in biosensing, atomic- scale repeat patterning, structure determination and drug delivery. My thesis describes the first successful design of 2D assemblies using non-covalent interactions in different types of 2D crystal space groups (layer groups) as well as of similarly designed tetrahedral and icosahedral protein cages