Resumen de Development of a framework for the computational design and evolution of enzymes
Enzymatic design is at the heart of modern biotechnology and, increasingly so, the so-called fine chemistry and green chemistry. Designing enzymes for applications in industrial or bioremediation contexts, for example, involves having a deep knowledge of enzymatic systems to propose rational changes that improve their catalytic properties. In recent years, a large number of computational methods have been developed to design or improve new enzymes. However, achieving enzymatic predictions through these methods to reach the power of natural enzymes is still an unattained scientific challenge. In this thesis, we propose the combination of two robust methodologies to devise a computational framework for enzyme design and evolution. On the one hand, a successful protein design methodology— the Rosetta modelling environment— and on the other, an efficient method for evaluating chemical reactivity based on molecular simulations— the Empirical Valence Bond method. Both tools, working collectively, are an attractive proposition for tackling state-of-the-art challenges in the field of enzymatic design. After applying our methodology in a proof-of-concept chemical system, the catalytic reaction of Kemp eliminase, we found a series of obstacles that need to be addressed before creating a successful framework for the computational design and evolution of enzymes. This work explores these challenges in-depth and suggests new directions to improve different aspects of the proposed methodology. Specifically, on the one hand, we make a careful dissection of the interaction energies provided by Rosetta, a key aspect for a better prediction of structural scaffolds on which to build new enzymatic designs. On the other hand, we propose a new practical implementation of a structure based simulation model in the OpenMM package of molecular simulations. Both elements are a critical step in achieving an efficient and robust "toolbox" for exploring the structure-function map of designed enzymes.
