Finding Reactive Configurations: A Machine Learning Approach for Estimating Energy Barriers Applied to Sirtuin 5

Recent Advances in the Discovery of SIRT1/2 Inhibitors via Computational Methods: A Perspective

Sirtuins (SIRTs) are classified as class III histone deacetylases (HDACs), a family of enzymes that catalyze the removal of acetyl groups from the ε-N-acetyl lysine residues of histone proteins, thus counteracting the activity performed by histone acetyltransferares (HATs). Based on their involvement in different biological pathways, ranging from transcription to metabolism and genome stability, SIRT dysregulation was investigated in many diseases, such as cancer, neurodegenerative disorders, diabetes, and cardiovascular and autoimmune diseases. The elucidation of a consistent number of SIRT-ligand complexes helped to steer the identification of novel and more selective modulators. Due to the high diversity and quantity of the structural data thus far available, we reviewed some of the different ligands and structure-based methods that have recently been used to identify new promising SIRT1/2 modulators. The present review is structured into two sections: the first includes a comprehensive perspective of the successful computational approaches related to the discovery of SIRT1/2 inhibitors (SIRTIs); the second section deals with the most interesting SIRTIs that have recently appeared in the literature (from 2017). The data reported here are collected from different databases (SciFinder, Web of Science, Scopus, Google Scholar, and PubMed) using "SIRT", "sirtuin", and "sirtuin inhibitors" as keywords.

Entropy and Energy Profiles of Chemical Reactions

The description of chemical processes at the molecular level is often facilitated by the use of reaction coordinates or collective variables (CVs). The CV measures the progress of the reaction and allows the construction of profiles that track how specific properties evolve as the reaction progresses. Whereas CVs are routinely used, especially alongside enhanced sampling techniques, the links among reaction profiles, thermodynamic state functions, and reaction rate constants are not rigorously exploited. Here, we report a unified treatment of such reaction profiles. Tractable expressions are derived for the free-energy, internal-energy, and entropy profiles as functions of only the CV. We demonstrate the ability of this treatment to extract quantitative insight from the entropy and internal-energy profiles of various real-world physicochemical processes, including intramolecular organic reactions, ionic transport in superionic electrolytes, and molecular transport in nanoporous materials.

Prediction of Enzyme Catalysis by Computing Reaction Energy Barriers via Steered QM/MM Molecular Dynamics Simulations and Machine Learning

The prediction of enzyme activity is one of the main challenges in catalysis. With computer-aided methods, it is possible to simulate the reaction mechanism at the atomic level. However, these methods are usually expensive if they are to be used on a large scale, as they are needed for protein engineering campaigns. To alleviate this situation, machine learning methods can help in the generation of predictive-decision models. Herein, we test different regression algorithms for the prediction of the reaction energy barrier of the rate-limiting step of the hydrolysis of mono-(2-hydroxyethyl)terephthalic acid by the MHETase ofIdeonella sakaiensis. As a training data set, we use steered quantum mechanics/molecular mechanics (QM/MM) molecular dynamics (MD) simulation snapshots and their corresponding pulling work values. We have explored three algorithms together with three chemical representations. As an outcome, our trained models are able to predict pulling works along the steered QM/MM MD simulations with a mean absolute error below 3 kcal mol-1 and a score value above 0.90. More challenging is the prediction of the energy maximum with a single geometry. Whereas the use of the initial snapshot of the QM/MM MD trajectory as input geometry yields a very poor prediction of the reaction energy barrier, the use of an intermediate snapshot of the former trajectory brings the score value above 0.40 with a low mean absolute error (ca. 3 kcal mol-1). Altogether, we have faced in this work some initial challenges of the final goal of getting an efficient workflow for the semiautomatic prediction of enzyme-catalyzed energy barriers and catalytic efficiencies.

Benchmark of general-purpose machine learning-based quantum mechanical method AIQM1 on reaction barrier heights

Artificial intelligence-enhanced quantum mechanical method 1 (AIQM1) is a general-purpose method that was shown to achieve high accuracy for many applications with a speed close to its baseline semiempirical quantum mechanical (SQM) method ODM2*. Here, we evaluate the hitherto unknown performance of out-of-the-box AIQM1 without any refitting for reaction barrier heights on eight datasets, including a total of ∼24 thousand reactions. This evaluation shows that AIQM1's accuracy strongly depends on the type of transition state and ranges from excellent for rotation barriers to poor for, e.g., pericyclic reactions. AIQM1 clearly outperforms its baseline ODM2* method and, even more so, a popular universal potential, ANI-1ccx. Overall, however, AIQM1 accuracy largely remains similar to SQM methods (and B3LYP/6-31G* for most reaction types) suggesting that it is desirable to focus on improving AIQM1 performance for barrier heights in the future. We also show that the built-in uncertainty quantification helps in identifying confident predictions. The accuracy of confident AIQM1 predictions is approaching the level of popular density functional theory methods for most reaction types. Encouragingly, AIQM1 is rather robust for transition state optimizations, even for the type of reactions it struggles with the most. Single-point calculations with high-level methods on AIQM1-optimized geometries can be used to significantly improve barrier heights, which cannot be said for its baseline ODM2* method.

Data-driven enzyme engineering to identify function-enhancing enzymes

Identifying function-enhancing enzyme variants is a 'holy grail' challenge in protein science because it will allow researchers to expand the biocatalytic toolbox for late-stage functionalization of drug-like molecules, environmental degradation of plastics and other pollutants, and medical treatment of food allergies. Data-driven strategies, including statistical modeling, machine learning, and deep learning, have largely advanced the understanding of the sequence-structure-function relationships for enzymes. They have also enhanced the capability of predicting and designing new enzymes and enzyme variants for catalyzing the transformation of new-to-nature reactions. Here, we reviewed the recent progresses of data-driven models that were applied in identifying efficiency-enhancing mutants for catalytic reactions. We also discussed existing challenges and obstacles faced by the community. Although the review is by no means comprehensive, we hope that the discussion can inform the readers about the state-of-the-art in data-driven enzyme engineering, inspiring more joint experimental-computational efforts to develop and apply data-driven modeling to innovate biocatalysts for synthetic and pharmaceutical applications.

Mechanistic Insights into Enzyme Catalysis from Explaining Machine-Learned Quantum Mechanical and Molecular Mechanical Minimum Energy Pathways

With the increasing popularity of machine learning (ML) applications, the demand for explainable artificial intelligence techniques to explain ML models developed for computational chemistry has also emerged. In this study, we present the development of the Boltzmann-weighted cumulative integrated gradients (BCIG) approach for effective explanation of mechanistic insights into ML models trained on high-level quantum mechanical and molecular mechanical (QM/MM) minimum energy pathways. Using the acylation reactions of the Toho-1 β-lactamase and two antibiotics (ampicillin and cefalexin) as the model systems, we show that the BCIG approach could quantitatively attribute the energetic contribution in one system and the relative reactivity of individual steps across different systems to specific chemical processes such as the bond making/breaking and proton transfers. The proposed BCIG contribution attribution method quantifies chemistry-interpretable insights in terms of contributions from each elementary chemical process, which is in agreement with the validating QM/MM calculations and our intuitive mechanistic understandings of the model reactions.

Exponential averaging versus umbrella sampling for computing the QM/MM free energy barrier of the initial step of the desuccinylation reaction catalyzed by sirtuin 5

The computational characterization of enzymatic reactions poses a great challenge which arises from the high dimensional and often rough potential energy surfaces commonly explored by static QM/MM methods such as adiabatic mapping (AM). The present study highlights the difficulties in estimating free energy barriers via exponential averaging over AM pathways. Based on our previous study [von der Esch et al., J. Chem. Theory Comput., 2019, 15, 6660-6667], where we analyzed the first reaction step of the desuccinylation reaction catalyzed by human sirtuin 5 (SIRT5) by means of QM/MM adiabatic mapping and machine learning, we use, here, umbrella sampling to compute the free energy profile of the initial reaction step. The computational investigations show that the initial step of the desuccinylation reaction proceeds via an S_N2-type reaction mechanism in SIRT5, suggesting that the first step of the deacylation reactions catalyzed by sirtuins is highly conserved. In addition, the direct comparison of the extrapolated free energy barrier from minimal energy paths and the computed free energy path from umbrella sampling further underlines the importance of extensive sampling.

Machine Learning to Predict Diels-Alder Reaction Barriers from the Reactant State Electron Density

Reaction barriers are key to our understanding of chemical reactivity and catalysis. Certain reactions are so seminal in chemistry that countless variants, with or without catalysts, have been studied, and their barriers have been computed or measured experimentally. This wealth of data represents a perfect opportunity to leverage machine learning models, which could quickly predict barriers without explicit calculations or measurement. Here, we show that the topological descriptors of the quantum mechanical charge density in the reactant state constitute a set that is both rigorous and continuous and can be used effectively for the prediction of reaction barrier energies to a high degree of accuracy. We demonstrate this on the Diels-Alder reaction, highly important in biology and medicinal chemistry, and as such, studied extensively. This reaction exhibits a range of barriers as large as 270 kJ/mol. While we trained our single-objective supervised (labeled) regression algorithms on simpler Diels-Alder reactions in solution, they predict reaction barriers also in significantly more complicated contexts, such a Diels-Alder reaction catalyzed by an artificial enzyme and its evolved variants, in agreement with experimental changes in k_cat. We expect this tool to apply broadly to a variety of reactions in solution or in the presence of a catalyst, for screening and circumventing heavily involved computations or experiments.

Harder, better, faster, stronger: Large-scale QM and QM/MM for predictive modeling in enzymes and proteins

Computational prediction of enzyme mechanism and protein function requires accurate physics-based models and suitable sampling. We discuss recent advances in large-scale quantum mechanical (QM) modeling of biochemical systems that have reduced the cost of high-accuracy models. Tradeoffs between sampling and accuracy have motivated modeling with molecular mechanics (MM) in a multiscale QM/MM or iterative approach. Limitations to both conventional density-functional theory and classical MM force fields remain for describing noncovalent interactions in comparison to experiment or wavefunction theory. Because predictions of enzyme action (i.e. electrostatics), free energy barriers, and mechanisms are sensitive to the protocol and embedding method in QM/MM, convergence tests and systematic methods for quantifying QM-level interactions are a needed, active area of development.

Accounting for the instantaneous disorder in the enzyme-substrate Michaelis complex to calculate the Gibbs free energy barrier of an enzyme reaction

Many enzyme reactions present instantaneous disorder. These dynamic fluctuations in the enzyme-substrate Michaelis complexes generate a wide range of energy barriers that cannot be experimentally observed, but that determine the measured kinetics of the reaction. These individual energy barriers can be calculated using QM/MM methods, but then the problem is how to deal with this dispersion of energy barriers to provide kinetic information. So far, the most usual procedure has implied the so-called exponential average of the energy barriers. In this paper, we discuss the foundations of this method, and we use the free energy perturbation theory to derive an alternative equation to get the Gibbs free energy barrier of the enzyme reaction. In addition, we propose a practical way to implement it. We have chosen four enzyme reactions as examples. In particular, we have studied the hydrolysis of a glycosidic bond catalyzed by the enzyme Thermus thermophilus β-glycosidase, and the mutant Y284P Ttb-gly, and the hydrogen abstraction reactions from C13 and C7 of arachidonic acid catalyzed by the enzyme rabbit 15-lipoxygenase-1.

Machine learning models for classification tasks related to drug safety

In this review, we outline the current trends in the field of machine learning-driven classification studies related to ADME (absorption, distribution, metabolism and excretion) and toxicity endpoints from the past six years (2015-2021). The study focuses only on classification models with large datasets (i.e. more than a thousand compounds). A comprehensive literature search and meta-analysis was carried out for nine different targets: hERG-mediated cardiotoxicity, blood-brain barrier penetration, permeability glycoprotein (P-gp) substrate/inhibitor, cytochrome P450 enzyme family, acute oral toxicity, mutagenicity, carcinogenicity, respiratory toxicity and irritation/corrosion. The comparison of the best classification models was targeted to reveal the differences between machine learning algorithms and modeling types, endpoint-specific performances, dataset sizes and the different validation protocols. Based on the evaluation of the data, we can say that tree-based algorithms are (still) dominating the field, with consensus modeling being an increasing trend in drug safety predictions. Although one can already find classification models with great performances to hERG-mediated cardiotoxicity and the isoenzymes of the cytochrome P450 enzyme family, these targets are still central to ADMET-related research efforts.

Identification of the subtype-selective Sirt5 inhibitor balsalazide through systematic SAR analysis and rationalization via theoretical investigations

We report here an extensive structure-activity relationship study of balsalazide, which was previously identified in a high-throughput screening as an inhibitor of Sirt5. To get a closer understanding why this compound is able to inhibit Sirt5, we initially performed docking experiments comparing the binding mode of a succinylated peptide as the natural substrate and balsalazide with Sirt5 in the presence of NAD⁺. Based on the evidence gathered here, we designed and synthesized 13 analogues of balsalazide, in which single functional groups were either deleted or slightly altered to investigate which of them are mandatory for high inhibitory activity. Our study confirms that balsalazide with all its given functional groups is an inhibitor of Sirt5 in the low micromolar concentration range and structural modifications presented in this study did not increase potency. While changes on the N-aroyl-β-alanine side chain eliminated potency, the introduction of a truncated salicylic acid part minimally altered potency. Calculations of the associated reaction paths showed that the inhibition potency is very likely dominated by the stability of the inhibitor-enzyme complex and not the type of inhibition (covalent vs. non-covalent). Further in-vitro characterization in a trypsin coupled assay determined that the tested inhibitors showed no competition towards NAD⁺ or the synthetic substrate analogue ZKsA. In addition, investigations for subtype selectivity revealed that balsalazide is a subtype-selective Sirt5 inhibitor, and our initial SAR and docking studies pave the way for further optimization.

A protocol to obtain multidimensional quantum tunneling corrections derived from QM(DFT)/MM calculations for an enzyme reaction

The multidimensional small-curvature tunneling (SCT) method with Electrostatic Embedding calculations is a compromise between an accessible computational cost and the attainment of an accurate enough estimation of tunneling for an enzyme reaction.