QM/MM Energy Decomposition Using the Interacting Quantum Atoms Approach

A novel cascade catalysis for one-pot enzymatically modified isoquercitrin (EMIQ) conversion from rutin and sucrose using rationally designed gradient temperature control

Enzymatically modified isoquercitrin (EMIQ) is a glyco-chemically modified flavonoid exhibiting notably high biological activity, such as antioxidant, anti-inflammatory and anti-allergic properties. However, the utilization of expensive substrates such as isoquercitrin and cyclodextrin in the conventional approach has hindered the industrial-scale production of EMIQ due to high cost and low yields. Hence, the development of a cost-effective and efficient method is crucial for the biological synthesis of EMIQ. In this study, a natural cascade catalytic reaction system was constructed with α-L-rhamnosidase and amylosucrase using the inexpensive substrates rutin and sucrose. Additionally, a novel approach integrating gradient temperature regulation into biological cascade reactions was implemented. Under the optimal conditions, the rutin conversion reached a remarkable 95.39% at 24 h. Meanwhile, the productivity of quercetin-3-O-tetraglucoside with the best bioavailability reached an impressive 41.69%. This study presents promising prospects for future mass production of EMIQ directly prepared from rutin and sucrose.

Exploring the Reaction Mechanism of Polyethylene Terephthalate Biodegradation through QM/MM Approach

The enzyme PETase fromIdeonella sakaiensis (IsPETase) strain 201-F6 can catalyze the hydrolysis of polyethylene terephthalate (PET), mainly converting it into mono(2-hydroxyethyl) terephthalic acid (MHET). In this study, we used quantum mechanics/molecular mechanics (QM/MM) simulations to explore the molecular details of the catalytic reaction mechanism of IsPETase in the formation of MHET. The QM region was described with AM1d/PhoT and M06-2X/6-31+G(d,p) potential. QM/MM simulations unveil the complete enzymatic PET hydrolysis mechanism and identify two possible reaction pathways for acylation and deacylation steps. The barrier obtained at M06-2X/6-31+G(d,p)/MM potential for the deacylation step corresponds to 20.4 kcal/mol, aligning with the experimental value of 18 kcal/mol. Our findings indicate that deacylation is the rate-limiting step of the process. Furthermore, per-residue interaction energy contributions revealed unfavorable contributions to the transition state of amino acids located at positions 200-230, suggesting potential sites for targeted mutations. These results can contribute to the development of more active and selective enzymes for PET depolymerization.

Discovering trends in the Lewis acidity of beryllium and magnesium hydrides and fluorides with increasing clusters size

We have reported in the last years the strong effect that Be- and Mg-containing Lewis acids have on the intrinsic properties of typical bases, which become acids upon complexation. In an effort to investigate these changes when the Be and Mg derivatives form clusters of increasing size, we have examined the behavior of the (MX2)n (M = Be, Mg; X = H, F; n = 1, 2, 3) clusters when they interact with ammonia, methanimine, hydrogen cyanide and pyridine, and with their corresponding deprotonated forms. The complexes obtained at the M06-2X/aug-cc-pVTZ level were analyzed using the MBIE energy decomposition formalism, in parallel with QTAIM, ELF, NCIPLOT and AdNDP analyses of their electron density. For n = 1 the interaction enthalpy for the different families of monomers, Be (Mg) hydrides and Be (Mg) fluorides, follows the same trend as the intrinsic basicity of the base that interacts with them. This interaction is greatly reinforced after the deprotonation of the base, resulting in a significant enhancement of the intrinsic acidity of the corresponding MX2-Base complex. For (MX2)2 clusters a further reinforcement of the interaction with the base is observed, this reinforcement being again larger for the deprotonated complexes. However, the concomitant increase of their intrinsic acidity is one order of magnitude larger for hydrides than for fluorides. Unexpectedly, the cyclic conformers (MX2)3, which are more unstable than the linear ones, become the global minima after association with the base and the same is true for the deprotonated complex. Accordingly, a further increase of the intrinsic acidity of the (MX2)3-Base complexes with respect to the (MX2)2-Base ones is observed. This effect is maximum for (MgF2)3 clusters, to the point that the (MgF2)3-Base complexes become more acidic than nitric acid, the extreme case being the cluster (MgF2)3-NCH, whose acidity is higher than that of perchloric acid.

Explainable chemical artificial intelligence from accurate machine learning of real-space chemical descriptors

Machine-learned computational chemistry has led to a paradoxical situation in which molecular properties can be accurately predicted, but they are difficult to interpret. Explainable AI (XAI) tools can be used to analyze complex models, but they are highly dependent on the AI technique and the origin of the reference data. Alternatively, interpretable real-space tools can be employed directly, but they are often expensive to compute. To address this dilemma between explainability and accuracy, we developed SchNet4AIM, a SchNet-based architecture capable of dealing with local one-body (atomic) and two-body (interatomic) descriptors. The performance of SchNet4AIM is tested by predicting a wide collection of real-space quantities ranging from atomic charges and delocalization indices to pairwise interaction energies. The accuracy and speed of SchNet4AIM breaks the bottleneck that has prevented the use of real-space chemical descriptors in complex systems. We show that the group delocalization indices, arising from our physically rigorous atomistic predictions, provide reliable indicators of supramolecular binding events, thus contributing to the development of Explainable Chemical Artificial Intelligence (XCAI) models.

QM/MM-Based Energy Decomposition Analysis Method for Large Systems

In this work, a QM/MM-based EDA method, called GKS-EDA(QM/MM), is proposed. As an extension of GKS-EDA, this scheme divides the total interaction energy into electrostatic, exchange-repulsion, polarization, and correlation/dispersion terms. GKS-EDA(QM/MM) can be applied to describe the interactions of large-scale systems combined with various QM/MM platforms. By using the examples of a hydrated hydronium ion complex in water solution, the barnase-barstar complex, and MMP-13-pyrimidinetrione in a metalloprotein, the capability of GKS-EDA(QM/MM) for various interactions in large systems is validated.

Effect of the QM Size, Basis Set, and Polarization on QM/MM Interaction Energy Decomposition Analysis

Herein, an Energy Decomposition Analysis (EDA) scheme extended to the framework of QM/MM calculations in the context of electrostatic embeddings (QM/MM-EDA) including atomic charges and dipoles is applied to assess the effect of the QM region size on the convergence of the different interaction energy components, namely, electrostatic, Pauli, and polarization, for cationic, anionic, and neutral systems interacting with a strong polar environment (water). Significant improvements are found when the bulk solvent environment is described by a MM potential in the EDA scheme as compared to pure QM calculations that neglect bulk solvation. The predominant electrostatic interaction requires sizable QM regions. The results reported here show that it is necessary to include a surprisingly large number of water molecules in the QM region to obtain converged values for this energy term, contrary to most cluster models often employed in the literature. Both the improvement of the QM wave function by means of a larger basis set and the introduction of polarization into the MM region through a polarizable force field do not translate to a faster convergence with the QM region size, but they lead to better results for the different interaction energy components. The results obtained in this work provide insight into the effect of each energy component on the convergence of the solute-solvent interaction energy with the QM region size. This information can be used to improve the MM FFs and embedding schemes employed in QM/MM calculations of solvated systems.

The IQA Energy Partition in a Drug Design Setting: A Hepatitis C Virus RNA-Dependent RNA Polymerase (NS5B) Case Study

The interaction of the thumb site II of the NS5B protein of hepatitis C virus and a pair of drug candidates was studied using a topological energy decomposition method called interacting quantum atoms (IQA). The atomic energies were then processed by the relative energy gradient (REG) method, which extracts chemical insight by computation based on minimal assumptions. REG reveals the most important IQA energy contributions, by atom and energy type (electrostatics, sterics, and exchange–correlation), that are responsible for the behaviour of the whole system, systematically from a short-range ligand–pocket interaction until a distance of approximately 22 Å. The degree of covalency in various key interatomic interactions can be quantified. No exchange–correlation contribution is responsible for the changes in the energy profile of both pocket–ligand systems investigated in the ligand–pocket distances equal to or greater than that of the global minimum. Regarding the hydrogen bonds in the system, a “neighbour effect” was observed thanks to the REG method, which states that a carbon atom would rather not have its covalent neighbour oxygen form a hydrogen bond. The combination of IQA and REG enables the automatic identification of the pharmacophore in the ligands. The coarser Interacting Quantum Fragments (IQF) enables the determination of which amino acids of the pocket contribute most to the binding and the type of energy of said binding. This work is an example of the contribution topological energy decomposition methods can make to fragment-based drug design.

A Molecular Mechanics Energy Partitioning Software for Biomolecular Systems

The partitioning of the molecular mechanics (MM) energy in calculations involving biomolecular systems is important to identify the source of major stabilizing interactions, e.g., in ligand–protein interactions, or to identify residues with considerable contributions in hybrid multiscale calculations, i.e., quantum mechanics/molecular mechanics (QM/MM). Here, we describe Energy Split, a software program to calculate MM energy partitioning considering the AMBER Hamiltonian and parameters. Energy Split includes a graphical interface plugin for VMD to facilitate the selection of atoms and molecules belonging to each part of the system. Energy Split is freely available at or can be easily installed through the VMD Store.