Studies of enzyme design using novel electronic structure methods
Title: Studies of enzyme design using novel electronic structure methods
DNr: SNIC 2016/1-15
Project Type: SNIC Medium Compute
Principal Investigator: Tore Brinck <tore@kth.se>
Affiliation: Kungliga Tekniska högskolan
Duration: 2016-01-22 – 2017-02-01
Classification: 10407 10402 10405
Homepage: https://www.kth.se/en/che/divisions/tfk/staff/seniors/brinck-1.80061
Keywords:

Abstract

Enzymes are highly effective catalysts and offer important benefits from a green chemistry perspective. The possibility of studying the enzyme active site in detail on a quantum mechanical level is limited due to the system size and associated high computational cost. The common approach is to consider a limited region containing the active site quantum mechanically and the surrounding environment semi-empirically with the so-called Q\ M/MM method. This has proven a powerful method when it comes to modeling chemical reactions of these extended systems. When designing enzymes, understanding the importance of different residues to the general activity of the enzyme for a given reaction. With a QM/MM approach, calculation of, for instance, an interaction energy between a residue and the active site would require two calculations of the system. One with and one without the residue of interest present in the system. To investigate many the impact of many different residues leads to numerous calculations which is computationally costly. With a novel method called the Fragment Molecular Orbital (FMO) method, which is implemented in the quantum chemistry software GAMESS, the enzyme system is divided into fragments which are explicitly calculated together with their dimers at a given level of theory in the Coulomb field of the whole system. This allows for obtaining interaction energies between all residues by means of one single calculation. This project aims at using the FMO method on important enzymatic reactions in order to identify key residues in the reactions of interest and comparing the result to similar, but computationally more demanding methods such as the differential transition state stabilization method. In addition, comparison and to some extent benchmarking will be made to and against molecular orbital, or standard-type density functional theory, calculations, which, if applied to systems of the mentioned size will be much more costly.