Chapter 18 Computationally Assisted Protein Design

Abstract

This chapter discusses the different aspects of a computationally assisted protein design. Protein design refers to the identification of amino acid sequences that fold to predetermine structures. In some cases, structure itself may also be an element of design. De novo designed proteins have the potential to serve as novel therapeutics, catalysts, biomaterials, and molecular scaffolds. In addition, protein design tests the understanding of folding and structure–function relationships. The choice of the target structure is a key decision made during computational protein design (CPD). Most redesign efforts start with a high-resolution structure obtained through X-ray crystallography or solution nuclear magnetic resonance sometimes with modifications to reflect specific design goals. Two types of degrees of freedom that must be simultaneously optimized during CPD are the amino acid identities and their side chain conformations. The protein design algorithm DEZYMER has been used to introduce targeted mutations in thioredoxin, thereby evolving a tetrahedral tetrathiolate iron center. The designed protein, which forms a 1:1 monomeric complex with Fe(III), supports multiple cycles of oxidation and reduction.