Modelling Enzymatic Mechanisms with QM/MM Approaches: Current Status and Future Challenges

Our recent review about studying enzymatic mechanisms using QM/MM mythologies has just been accepted on Israel Journal of Chemistry.

Abstract

Quantum mechanics/molecular mechanics (QM/MM) methods are presently a well‐established alternative for the study of enzymatic reaction mechanisms. They enable the description of a small part of the enzyme, where reactions take place through QM, while the majority of the thousands of atoms that comprise these biomolecules are handled through MM. While different “flavors” and variations in the QM/MM field exist, this review will focus more on the application of the ONIOM methodology, presenting a fresh perspective on the application of this popular method in light of the growth in computational power and level of sophistication of the different methodologies that it can combine. In addition to a brief presentation of the basic principles behind these methods, this review will discuss different examples of applicability, common choices, practical considerations, and main problems involved, stemming from our experience in this field. Finally, a reflection on the future challenges for the next decade in the QM/MM modeling of enzymatic mechanisms is presented.

Authors: Magalhães RP, Fernandes HS, and Sousa SF

The catalytic mechanism of the Serine Hydroxymethyltransferase – a computational ONIOM QM/MM study

Henrique Silva Fernandes, Maria João Ramos, and Nuno M.F.S.A. Cerqueira

Published on 18th September 2018
Journal: ACS Catalysis

http://dx.doi.org/10.1021/acscatal.8b02321 | Download citation

Abstract

Serine Hydroxymethyltransferase (SHMT) is an important drug target to fight malaria – one of the most devastating infectious diseases that accounted in 2016 with 216 million new cases and almost 450 thousand deaths. In this paper, computational studies were carried out to unveil the catalytic mechanism of SHMT using QM/MM methodologies. This enzyme is responsible for the extraordinary cyclisation of a tetrahydrofolate (THF) into 5,10-methylene-THF. This process is catalyzed by a pyridoxal-5’-phosphate (PLP) cofactor that binds L-serine and from which one molecule of L-glycine is produced. The results show that the catalytic process takes place in eight sequential steps that involve an α-elimination, the cyclization of the 5-hydroxymethyl-THF intermediate into 5,10-methylene-THF and the protonation of the quinonoid intermediate. According to the calculated energetic profile, the rate-limiting step of the full mechanism is the elimination of the hydroxymethyl group, from which results a formaldehyde intermediate that then becomes covalently bonded to the THF cofactor. The calculated barrier (DLPNO-CCSD(T)/CBS:ff99SB) for the rate-limiting step (18.0 kcal/mol) agrees very well with the experimental kinetic results (15.7-16.2 kcal/mol). The results also highlight the key role played by Glu57 during the full catalytic process and particularly in the first step of the mechanism that requires an anionic Glu57, contrasting with some proposals available in the literature for this step. It was also concluded that the cyclisation of THF must take place in the enzyme, rather than in solution as it has been proposed also in the past. All of these results together provide new knowledge and insight on the catalytic mechanism of SHMT that now can be used to develop new inhibitors targeting SHMT and therefore new anti-malaria drugs.

molUP is cover of the Journal of Computational Chemistry 39(19)

Cover

First published: 10 June 2018 | https://doi.org/10.1002/jcc.25372

Abstract

The molUP plugin, developed by Nuno Sousa Cerqueira and colleagues and described on page 1344, overcomes some of the most common problems in computational chemistry concerning the analysis of big data. MolUP was developed for use in the analysis of quantum chemistry (QM), QM/MM (molecular mechanics), and QM/QM calculations, molecular dynamics (MD) simulations, as well as the preparation of input files. MolUP also provides new tools to analyze and visualize existing computational chemistry information in a more userfriendly way that simplifies the current complex and time‐demanding practices used in the field. (DOI: 10.1002/jcc.25189)

The Catalytic Mechanism of the Pyridoxal-5′-phosphate-Dependent Enzyme, Histidine Decarboxylase: A Computational Study

Henrique Silva Fernandes, Maria João Ramos, and Nuno M.F.S.A. Cerqueira

Published on June 14th 2017
DOI: http://dx.doi.org/10.1002/chem.201701375 | Download citation

Abstract

The catalytic mechanism of histidine decarboxylase (HDC), a pyridoxal-5′-phosphate (PLP)-dependent enzyme, was studied by using a computational QM/MM approach following the scheme M06-2X/6–311++G(3df,2pd):Amber. The reaction involves two sequential steps: the decarboxylation of l-histidine and the protonation of the generated intermediate from which results histamine. The rate-limiting step is the first one (ΔG=17.6 kcal mol−1; ΔGr=13.7 kcal mol−1) and agrees closely with the available experimental kcat (1.73 s−1), which corresponds to an activation barrier of 17.9 kcal mol−1. In contrast, the second step is very fast (ΔG=1.9 kcal mol−1) and exergonic (ΔGr=−33.2 kcal mol−1). Our results agree with the available experimental data and allow us to explain the role played by several active site residues that are considered relevant according to site-directed mutagenesis studies, namely Tyr334B, Asp273A, Lys305A, and Ser354B. These results can provide insights regarding the catalytic mechanism of other enzymes belonging to family II of PLP-dependent decarboxylases.