Author: Stephanie Cynthia Contreras

Publisher:

Published: 2019

Total Pages:

ISBN-13: 9781392859971

Protein engineering has two prevailing strategies: rational design and directed evolution. The method of directed evolution was pioneered by Frances Arnold in the 1990’s. It involves the construction and screening of variant libraries for the identification of mutations, generated randomly, for the improvement of enzymes through an evolutionary strategy (1). Rational design proposes mutations on the basis of structure and biochemical properties for the improvement of enzymes (2). With recent developments in computing power and accessibility of computation, computational protein design has become a popular method in protein engineering. In silico prediction and design of a protein structure presents the challenge of obtaining an accurate energy state of the protein by modeling and sampling a large number of different conformations of atoms (3). Stephen Mayo’s group was the first to successfully overcome this challenge with the de novo protein design of the second zinc finger module of the DNA binding protein Zif268. The successful redesign of this protein was done through the use of their algorithm that implemented the dead-end elimination theorem and use of physical chemistry principles (4). From this pioneering work other programs were developed to address the in silico challenge and resulted in another achievement in the field from David Baker’s group with the development of the biomolecular modeling suite Rosetta. The Baker group showed success with Rosetta by creating a new protein structure of a globular protein whereas previous methods sought to only redesign naturally occurring proteins (5). Rosetta uses a score function based on physical principles and a Monte Carlo search protocol for structure prediction, protein design, and other applications. Since the publication of this designed protein structure, Rosetta has reached many milestones when it comes to structure prediction and design (6). Advancements of computational and experimental methods within the protein engineering field have provided scientists with the tools needed to engineer proteins for use in real-world applications, such as in medicine, biofuels, and the food industry. With the continuous progress and efforts made to improve protein design and structure prediction we can get closer to obtaining more accurate models of engineered and naturally occurring proteins. The research being presented here builds on this progress of both the improvement of structure prediction and the design of novel enzymes.