Predicting Reactive Cysteines with Implicit-Solvent-Based Continuous Constant pH Molecular Dynamics in Amber

Machine Learning Models to Interrogate Proteome-Wide Covalent Ligandabilities Directed at Cysteines

Machine learning (ML) identification of covalently ligandable sites may accelerate targeted covalent inhibitor design and help expand the druggable proteome space. Here, we report the rigorous development and validation of the tree-based models and convolutional neural networks (CNNs) trained on a newly curated database (LigCys3D) of over 1000 liganded cysteines in nearly 800 proteins represented by over 10,000 three-dimensional structures in the protein data bank. The unseen tests yielded 94 and 93% area under the receiver operating characteristic curves for the tree models and CNNs, respectively. Based on the AlphaFold2 predicted structures, the ML models recapitulated the newly liganded cysteines in the PDB with over 90% recall values. To assist the community of covalent drug discoveries, we report the predicted ligandable cysteines in 392 human kinases and their locations in the sequence-aligned kinase structure, including the PH and SH2 domains. Furthermore, we disseminate a searchable online database LigCys3D (https://ligcys.computchem.org/) and a web prediction server DeepCys (https://deepcys.computchem.org/), both of which will be continuously updated and improved by including newly published experimental data. The present work represents an important step toward the ML-led integration of big genome data and structure models to annotate the human proteome space for the next-generation covalent drug discoveries.

DeepKa Web Server: High-Throughput Protein pKa Prediction

DeepKa is a deep-learning-based protein pKa predictor proposed in our previous work. In this study, a web server was developed that enables online protein pKa prediction driven by DeepKa. The web server provides a user-friendly interface where a single step of entering a valid PDB code or uploading a PDB format file is required to submit a job. Two case studies have been attached in order to explain how pKa's calculated by the web server could be utilized by users. Finally, combining the web server with post processing as described in case studies, this work suggests a quick workflow of investigating the relationship between protein structure and function that are pH dependent. The web server of DeepKa is freely available at http://www.computbiophys.com/DeepKa/main.

phbuilder: A Tool for Efficiently Setting up Constant pH Molecular Dynamics Simulations in GROMACS

Constant pH molecular dynamics (MD) is a powerful technique that allows the protonation state of residues to change dynamically, thereby enabling the study of pH dependence in a manner that has not been possible before. Recently, a constant pH implementation was incorporated into the GROMACS MD package. Although this implementation provides good accuracy and performance, manual modification and the preparation of simulation input files are required, which can be complicated, tedious, and prone to errors. To simplify and automate the setup process, we present phbuilder, a tool that automatically prepares constant pH MD simulations for GROMACS by modifying the input structure and topology as well as generating the necessary parameter files. phbuilder can prepare constant pH simulations from both initial structures and existing simulation systems, and it also provides functionality for performing titrations and single-site parametrizations of new titratable group types. The tool is freely available at www.gitlab.com/gromacs-constantph. We anticipate that phbuilder will make constant pH simulations easier to set up, thereby making them more accessible to the GROMACS user community.

Machine Learning Models to Interrogate Proteomewide Covalent Ligandabilities Directed at Cysteines

Machine learning (ML) identification of covalently ligandable sites may accelerate targeted covalent inhibitor design and help expand the druggable proteome space. Here we report the rigorous development and validation of the tree-based models and convolutional neural networks (CNNs) trained on a newly curated database (LigCys3D) of over 1,000 liganded cysteines in nearly 800 proteins represented by over 10,000 three-dimensional structures in the protein data bank. The unseen tests yielded 94% and 93% AUCs (area under the receiver operating characteristic curve) for the tree models and CNNs, respectively. Based on the AlphaFold2 predicted structures, the ML models recapitulated the newly liganded cysteines in the PDB with over 90% recall values. To assist the community of covalent drug discoveries, we report the predicted ligandable cysteines in 392 human kinases and their locations in the sequence-aligned kinase structure including the PH and SH2 domains. Furthermore, we disseminate a searchable online database LigCys3D (https://ligcys.computchem.org/) and a web prediction server DeepCys (https://deepcys.computchem.org/), both of which will be continuously updated and improved by including newly published experimental data. The present work represents a first step towards the ML-led integration of big genome data and structure models to annotate the human proteome space for the next-generation covalent drug discoveries.

PTM-Psi: A python package to facilitate the computational investigation of post-translational modification on protein structures and their impacts on dynamics and functions

Post-translational modification (PTM) of a protein occurs after it has been synthesized from its genetic template, and involves chemical modifications of the protein's specific amino acid residues. Despite of the central role played by PTM in regulating molecular interactions, particularly those driven by reversible redox reactions, it remains challenging to interpret PTMs in terms of protein dynamics and function because there are numerous combinatorially enormous means for modifying amino acids in response to changes in the protein environment. In this study, we provide a workflow that allows users to interpret how perturbations caused by PTMs affect a protein's properties, dynamics, and interactions with its binding partners based on inferred or experimentally determined protein structure. This Python-based workflow, called PTM-Psi, integrates several established open-source software packages, thereby enabling the user to infer protein structure from sequence, develop force fields for non-standard amino acids using quantum mechanics, calculate free energy perturbations through molecular dynamics simulations, and score the bound complexes via docking algorithms. Using the S-nitrosylation of several cysteines on the GAP2 protein as an example, we demonstrated the utility of PTM-Psi for interpreting sequence-structure-function relationships derived from thiol redox proteomics data. We demonstrate that the S-nitrosylated cysteine that is exposed to the solvent indirectly affects the catalytic reaction of another buried cysteine over a distance in GAP2 protein through the movement of the two ligands. Our workflow tracks the PTMs on residues that are responsive to changes in the redox environment and lays the foundation for the automation of molecular and systems biology modeling.

Understanding Cysteine Reactivity in Protein Environments with Electric Fields

The role cysteine residues play in proteins is mediated by their protonation state, whereby the thiolate form of the side chain is highly reactive while the thiol form is more inert. However, the pKa of cysteine residues is hard to predict as it can differ widely from its reference value in solution, an effect that is accentuated by local effects in the heterogeneous protein environment. Here, we present a new approach to the prediction of cysteine reactivity based on electric field calculations at the thiol/thiolate group. We validated our approach by predicting the protonation state of cysteine residues in different protein environments (in the active site, at the protein surface, and buried within the protein interior), including Cys-25 in papaya protease omega, which was proven problematic for the more traditional constant pH molecular dynamics (MD) technique. We predict pKa shifts consistent with experimental observations, and the decomposition of the electric fields into contributions from molecular fragments provides a direct handle to rationalize local pH and pKa effects in proteins without introducing parameters other than those of the force field used for MD simulations.

Quantum Descriptors for Predicting and Understanding the Structure-Activity Relationships of Michael Acceptor Warheads

Predictive modeling and understanding of chemical warhead reactivities have the potential to accelerate targeted covalent drug discovery. Recently, the carbanion formation free energies as well as other ground-state electronic properties from density functional theory (DFT) calculations have been proposed as predictors of glutathione reactivities of Michael acceptors; however, no clear consensus exists. By profiling the thiol-Michael reactions of a diverse set of singly- and doubly-activated olefins, including several model warheads related to afatinib, here we reexamined the question of whether low-cost electronic properties can be used as predictors of reaction barriers. The electronic properties related to the carbanion intermediate were found to be strong predictors, e.g., the change in the Cβ charge accompanying carbanion formation. The least expensive reactant-only properties, the electrophilicity index, and the Cβ charge also show strong rank correlations, suggesting their utility as quantum descriptors. A second objective of the work is to clarify the effect of the β-dimethylaminomethyl (DMAM) substitution, which is incorporated in the warheads of several FDA-approved covalent drugs. Our data suggest that the β-DMAM substitution is cationic at neutral pH in solution and promotes acrylamide's intrinsic reactivity by enhancing the charge accumulation at Cα upon carbanion formation. In contrast, the inductive effect of the β-trimethylaminomethyl substitution is diminished due to steric hindrance. Together, these results reconcile the current views of the intrinsic reactivities of acrylamides and contribute to large-scale predictive modeling and an understanding of the structure-activity relationships of Michael acceptors for rational TCI design.

Benchmarking In Silico Tools for Cysteine pKa Prediction

Accurate estimation of the pK_a's of cysteine residues in proteins could inform targeted approaches in hit discovery. The pK_a of a targetable cysteine residue in a disease-related protein is an important physiochemical parameter in covalent drug discovery, as it influences the fraction of nucleophilic thiolate amenable to chemical protein modification. Traditional structure-based in silico tools are limited in their predictive accuracy of cysteine pK_a's relative to other titratable residues. Additionally, there are limited comprehensive benchmark assessments for cysteine pK_a predictive tools. This raises the need for extensive assessment and evaluation of methods for cysteine pK_a prediction. Here, we report the performance of several computational pK_a methods, including single-structure and ensemble-based approaches, on a diverse test set of experimental cysteine pK_a's retrieved from the PKAD database. The dataset consisted of 16 wildtype and 10 mutant proteins with experimentally measured cysteine pK_a values. Our results highlight that these methods are varied in their overall predictive accuracies. Among the test set of wildtype proteins evaluated, the best method (MOE) yielded a mean absolute error of 2.3 pK units, highlighting the need for improvement of existing pK_a methods for accurate cysteine pK_a estimation. Given the limited accuracy of these methods, further development is needed before these approaches can be routinely employed to drive design decisions in early drug discovery efforts.

GPU-Accelerated All-Atom Particle-Mesh Ewald Continuous Constant pH Molecular Dynamics in Amber

Constant pH molecular dynamics (MD) simulations sample protonation states on the fly according to the conformational environment and user specified pH conditions; however, the current accuracy is limited due to the use of implicit-solvent models or a hybrid solvent scheme. Here, we report the first GPU-accelerated implementation, parametrization, and validation of the all-atom continuous constant pH MD (CpHMD) method with particle-mesh Ewald (PME) electrostatics in the Amber22 pmemd.cuda engine. The titration parameters for Asp, Glu, His, Cys, and Lys were derived for the CHARMM c22 and Amber ff14sb and ff19sb force fields. We then evaluated the PME-CpHMD method using the asynchronous pH replica-exchange titration simulations with the c22 force field for six benchmark proteins, including BBL, hen egg white lysozyme (HEWL), staphylococcal nuclease (SNase), thioredoxin, ribonuclease A (RNaseA), and human muscle creatine kinase (HMCK). The root-mean-square deviation from the experimental pK_a's of Asp, Glu, His, and Cys is 0.76 pH units, and the Pearson's correlation coefficient for the pK_a shifts with respect to model values is 0.80. We demonstrated that a finite-size correction or much enlarged simulation box size can remove a systematic error of the calculated pK_a's and improve agreement with experiment. Importantly, the simulations captured the relevant biology in several challenging cases, e.g., the titration order of the catalytic dyad Glu35/Asp52 in HEWL and the coupled residues Asp19/Asp21 in SNase, the large pK_a upshift of the deeply buried catalytic Asp26 in thioredoxin, and the large pK_a downshift of the deeply buried catalytic Cys283 in HMCK. We anticipate that PME-CpHMD will offer proper pH control to improve the accuracies of MD simulations and enable mechanistic studies of proton-coupled dynamical processes that are ubiquitous in biology but remain poorly understood due to the lack of experimental tools and limitation of current MD simulations.

Cysteine Oxidation in Proteins: Structure, Biophysics, and Simulation

Cysteine side chains can exist in distinct oxidation states depending on the pH and redox potential of the environment, and cysteine oxidation plays important yet complex regulatory roles. Compared with the effects of post-translational modifications such as phosphorylation, the effects of oxidation of cysteine to sulfenic, sulfinic, and sulfonic acid on protein structure and function remain relatively poorly characterized. We present an analysis of the role of cysteine reactivity as a regulatory factor in proteins, emphasizing the interplay between electrostatics and redox potential as key determinants of the resulting oxidation state. A review of current computational approaches suggests underdeveloped areas of research for studying cysteine reactivity through molecular simulations.

A Guide to the Continuous Constant pH Molecular Dynamics Methods in Amber and CHARMM [Article v1.0]

Like temperature and pressure, solution pH is an important environmental variable in biomolecular simulations. Virtually all proteins depend on pH to maintain their structure and function. In conventional molecular dynamics (MD) simulations of proteins, pH is implicitly accounted for by assigning and fixing protonation states of titratable sidechains. This is a significant limitation, as the assigned protonation states may be wrong and they may change during dynamics. In this tutorial, we guide the reader in learning and using the various continuous constant pH MD methods in Amber and CHARMM packages, which have been applied to predict pK a values and elucidate proton-coupled conformational dynamics of a variety of proteins including enzymes and membrane transporters.

Proteome-Wide Profiling of the Covalent-Druggable Cysteines with a Structure-Based Deep Graph Learning Network

Covalent ligands have attracted increasing attention due to their unique advantages, such as long residence time, high selectivity, and strong binding affinity. They also show promise for targets where previous efforts to identify noncovalent small molecule inhibitors have failed. However, our limited knowledge of covalent binding sites has hindered the discovery of novel ligands. Therefore, developing in silico methods to identify covalent binding sites is highly desirable. Here, we propose DeepCoSI, the first structure-based deep graph learning model to identify ligandable covalent sites in the protein. By integrating the characterization of the binding pocket and the interactions between each cysteine and the surrounding environment, DeepCoSI achieves state-of-the-art predictive performances. The validation on two external test sets which mimic the real application scenarios shows that DeepCoSI has strong ability to distinguish ligandable sites from the others. Finally, we profiled the entire set of protein structures in the RCSB Protein Data Bank (PDB) with DeepCoSI to evaluate the ligandability of each cysteine for covalent ligand design, and made the predicted data publicly available on website.

Profiling MAP kinase cysteines for targeted covalent inhibitor design

Mitogen-activated protein kinases (MAPK) are important therapeutic targets, and yet no inhibitors have advanced to the market. Here we applied the GPU-accelerated continuous constant pH molecular dynamics (CpHMD) to calculate the pK a's and profile the cysteine reactivities of all 14 MAPKs for assisting the targeted covalent inhibitor design. The simulations not only recapitulated but also rationalized the reactive cysteines in the front pocket of JNK1/2/3 and the extended front pocket of p38α. Interestingly, the DFG - 1 cysteine in the DFG-in conformation of ERK1/ERK2 was found somewhat reactive or unreactive; however, simulations of MKK7 showed that switching to the DFG-out conformation makes the DFG - 1 cysteine reactive, suggesting the advantage of type II covalent inhibitors. Additionally, the simulations prospectively predicted several druggable cysteine and lysine sites, including the αH head cysteine in JNK1/3 and DFG + 6 cysteine in JNK2, corroborating the chemical proteomic screening data. Given the low cost and the ability to offer physics-based rationales, we envision CpHMD simulations to complement the chemo-proteomic platform for systematic profiling cysteine reactivities for targeted covalent drug discovery.

Impact of different protonation states on virtual screening performance against cruzain

The cysteine protease cruzain is a Chagas disease target, exploited in computational studies. However, there is no consensus on the protonation states of the active site residues Cys25, His162, and Glu208 at the enzyme's active pH range. We evaluated the impact of different protonation states of these residues on docking calculations. Through a retrospective study with cruzain inhibitors and decoys, we compared the performance of virtual screening using four grids, varying protonation states of Cys25, His162, and Glu208. Based on enrichment factors and ROC plots, docking with the four grids affected compound ranking and the overall charge of top-ranking compounds. Different grids can be complementary and synergistic, increasing the odds of finding different ligands with diverse chemical properties.

Protein Modifications: From Chemoselective Probes to Novel Biocatalysts

Chemical reactions can be performed to covalently modify specific residues in proteins. When applied to native enzymes, these chemical modifications can greatly expand the available set of building blocks for the development of biocatalysts. Nucleophilic canonical amino acid sidechains are the most readily accessible targets for such endeavors. A rich history of attempts to design enhanced or novel enzymes, from various protein scaffolds, has paved the way for a rapidly developing field with growing scientific, industrial, and biomedical applications. A major challenge is to devise reactions that are compatible with native proteins and can selectively modify specific residues. Cysteine, lysine, N-terminus, and carboxylate residues comprise the most widespread naturally occurring targets for enzyme modifications. In this review, chemical methods for selective modification of enzymes will be discussed, alongside with examples of reported applications. We aim to highlight the potential of such strategies to enhance enzyme function and create novel semisynthetic biocatalysts, as well as provide a perspective in a fast-evolving topic.

Reactivities of the Front Pocket N-Terminal Cap Cysteines in Human Kinases

The front pocket (FP) N-terminal cap (Ncap) cysteine is the most popular site of covalent modification in kinases. A long-standing hypothesis associates the Ncap position with cysteine hyper-reactivity; however, traditional computational predictions suggest that the FP Ncap cysteines are predominantly unreactive. Here we applied the state-of-the-art continuous constant pH molecular dynamics (CpHMD) to test the Ncap hypothesis. Simulations found that the Ncap cysteines of BTK/BMX/TEC/ITK/TXK, JAK3, and MKK7 are reactive to varying degrees; however, those of BLK and EGFR/ERBB2/ERBB4 possessing a Ncap+3 aspartate are unreactive. Analysis suggested that hydrogen bonding and electrostatic interactions drive the reactivity, and their absence renders the Ncap cysteine unreactive. To further test the Ncap hypothesis, we examined the FP Ncap+2 cysteines in JNK1/JNK2/JNK3 and CASK. Our work offers a systematic understanding of the cysteine structure-reactivity relationship and illustrates the use of CpHMD to differentiate cysteines toward the design of targeted covalent inhibitors with reduced chemical reactivities.

Titr-DMD-A Rapid, Coarse-Grained Quasi-All-Atom Constant pH Molecular Dynamics Framework

The pH-dependence of enzyme fold stability and catalytic activity is a fundamentally dynamic, structural property which is difficult to study. The challenges and expense of investigating dynamic, atomic scale behavior experimentally means that computational methods, particularly constant pH molecular dynamics (CpHMD), are well situated tools for this. However, these methods often struggle with affordable sampling of sufficiently long time scales while also obtaining accurate pKa prediction and verifying the structures they generate. We introduce Titr-DMD, an affordable CpHMD method that combines the quasi-all-atom coarse-grained discrete molecular dynamics (DMD) method for conformational sampling with Propka for pKa prediction, to circumvent these issues. The combination enables rapid sampling on limited computational resources, while simulations are still performed on the atomic scale. We benchmark the method on a set of proteins with experimentally attested pKa and on the pH triggered conformational change in a staphylococcal nuclease mutant, a rare experimental study of such behavior. Our results show Titr-DMD to be an effective and inexpensive method to study pH-coupled protein dynamics.

Proton-Coupled Conformational Activation of SARS Coronavirus Main Proteases and Opportunity for Designing Small-Molecule Broad-Spectrum Targeted Covalent Inhibitors

The SARS coronavirus 2 (SARS-CoV-2) main protease (Mpro) is an attractive broad-spectrum antiviral drug target. Despite the enormous progress in structure elucidation, the Mpro's structure-function relationship remains poorly understood. Recently, a peptidomimetic inhibitor has entered clinical trial; however, small-molecule orally available antiviral drugs have yet to be developed. Intrigued by a long-standing controversy regarding the existence of an inactive state, we explored the proton-coupled dynamics of the Mpros of SARS-CoV-2 and the closely related SARS-CoV using a newly developed continuous constant pH molecular dynamics (MD) method and microsecond fixed-charge all-atom MD simulations. Our data supports a general base mechanism for Mpro's proteolytic function. The simulations revealed that protonation of His172 alters a conserved interaction network that upholds the oxyanion loop, leading to a partial collapse of the conserved S1 pocket, consistent with the first and controversial crystal structure of SARS-CoV Mpro determined at pH 6. Interestingly, a natural flavonoid binds SARS-CoV-2 Mpro in the close proximity to a conserved cysteine (Cys44), which is hyper-reactive according to the CpHMD titration. This finding offers an exciting new opportunity for small-molecule targeted covalent inhibitor design. Our work represents a first step toward the mechanistic understanding of the proton-coupled structure-dynamics-function relationship of CoV Mpros; the proposed strategy of designing small-molecule covalent inhibitors may help accelerate the development of orally available broad-spectrum antiviral drugs to stop the current pandemic and prevent future outbreaks.

Assessment of proton-coupled conformational dynamics of SARS and MERS coronavirus papain-like proteases: Implication for designing broad-spectrum antiviral inhibitors

Broad-spectrum antiviral drugs are urgently needed to stop the Coronavirus Disease 2019 pandemic and prevent future ones. The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is related to the SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV), which have caused the previous outbreaks. The papain-like protease (PLpro) is an attractive drug target due to its essential roles in the viral life cycle. As a cysteine protease, PLpro is rich in cysteines and histidines, and their protonation/deprotonation modulates catalysis and conformational plasticity. Here, we report the pK_a calculations and assessment of the proton-coupled conformational dynamics of SARS-CoV-2 in comparison to SARS-CoV and MERS-CoV PLpros using the recently developed graphical processing unit (GPU)-accelerated implicit-solvent continuous constant pH molecular dynamics method with a new asynchronous replica-exchange scheme, which allows computation on a single GPU card. The calculated pK_a's support the catalytic roles of the Cys-His-Asp triad. We also found that several residues can switch protonation states at physiological pH among which is C270/271 located on the flexible blocking loop 2 (BL2) of SARS-CoV-2/CoV PLpro. Simulations revealed that the BL2 can open and close depending on the protonation state of C271/270, consistent with the most recent crystal structure evidence. Interestingly, despite the lack of an analogous cysteine, BL2 in MERS-CoV PLpro is also very flexible, challenging a current hypothesis. These findings are supported by the all-atom fixed-charge simulations and provide a starting point for more detailed studies to assist the structure-based design of broad-spectrum inhibitors against CoV PLpros.

Assessment of Proton-Coupled Conformational Dynamics of SARS and MERS Coronavirus Papain-like Proteases: Implication for Designing Broad-Spectrum Antiviral Inhibitors

Broad-spectrum antiviral drugs are urgently needed to stop the COVID-19 pandemic and prevent future ones. The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is related to SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV), which have caused the previous outbreaks. The papain-like protease (PLpro) is an attractive drug target due to its essential roles in the viral life cycle. As a cysteine protease, PLpro is rich in cysteines and histidines and their protonation/deprotonation modulates catalysis and conformational plasticity. Here we report the pKa calculations and assessment of the proton-coupled conformational dynamics of SARS-CoV-2 in comparison to SARS-CoV and MERS-CoV PLpros using a newly developed GPU-accelerated implicit-solvent continuous constant pH molecular dynamics method with an asynchronous replica-exchange scheme. The calculated pKa's support the catalytic roles of the Cys-His-Asp triad. We also found that several residues can switch protonation states at physiological pH, among which is C270/271 located on the flexible blocking loop 2 (BL2) of SARS-CoV-2/CoV PLpro. Simulations revealed that the BL2 conformational dynamics is coupled to the titration of C271/270, in agreement with the crystal structures of SARS-CoV-2 PLpro. Simulations also revealed that BL2 in MERS-CoV PLpro is very flexible, sampling both open and closed states despite the lack of an analogous cysteine. Our work provides a starting point for more detailed mechanistic studies to assist structure-based design of broad-spectrum inhibitors against CoV PLpros.

Catalytic Site pKa Values of Aspartic, Cysteine, and Serine Proteases: Constant pH MD Simulations

Enzymatic function and activity of proteases is closely controlled by the pH value. The protonation states of titratable residues in the active site react to changes in the pH value, according to their pK_a, and thereby determine the functionality of the enzyme. Knowledge of the titration behavior of these residues is crucial for the development of drugs targeting the active site residues. However, experimental pK_a data are scarce, since the systems’ size and complexity make determination of these pK_a values inherently difficult. In this study, we use single pH constant pH MD simulations as a fast and robust tool to estimate the active site pK_a values of a set of aspartic, cysteine, and serine proteases. We capture characteristic pK_a shifts of the active site residues, which dictate the experimentally determined activity profiles of the respective protease family. We find clear differences of active site pK_a values within the respective families, which closely match the experimentally determined pH preferences of the respective proteases. These shifts are caused by a distinct network of electrostatic interactions characteristic for each protease family. While we find convincing agreement with experimental data for serine and aspartic proteases, we observe clear deficiencies in the description of the titration behavior of cysteines within the constant pH MD framework and highlight opportunities for improvement. Consequently, with this work, we provide a concise set of active site pK_a values of aspartic and serine proteases, which could serve as reference for future theoretical as well as experimental studies.