Search

Since its inception nearly a half century ago, CHARMM has been playing a central role in computational biochemistry and biophysics. Commensurate with the developments in experimental research and advances in computer hardware, the range of methods and applicability of CHARMM have also grown. This review summarizes major developments that occurred after 2009 when the last review of CHARMM was published. They include the following: new faster simulation engines, accessible user interfaces for convenient workflows, and a vast array of simulation and analysis methods that encompass quantum mechanical, atomistic, and coarse-grained levels, as well as extensive coverage of force fields. In addition to providing the current snapshot of the CHARMM development, this review may serve as a starting point for exploring relevant theories and computational methods for tackling contemporary and emerging problems in biomolecular systems. CHARMM is freely available for academic and nonprofit research at https://academiccharmm.org/program.

Identifying druggable binding sites on proteins is an important and challenging problem, particularly for cryptic, allosteric binding sites that may not be obvious from X-ray, cryo-EM, or predicted structures. The Site-Identification by Ligand Competitive Saturation (SILCS) method accounts for the flexibility of the target protein using all-atom molecular simulations that include various small molecule solutes in aqueous solution. During the simulations, the combination of protein flexibility and comprehensive sampling of the water and solute spatial distributions can identify buried binding pockets absent in experimentally determined structures. Previously, we reported a method for leveraging the information in the SILCS sampling to identify binding sites (termed Hotspots) of small mono- or bicyclic compounds, a subset of which coincide with known binding sites of drug-like molecules. Here, we build on that physics-based approach and present a ML model for ranking the Hotspots according to the likelihood they can accommodate drug-like molecules (e.g., molecular weight >200 Da). In the independent validation set, which includes various enzymes and receptors, our model recalls 67% and 89% of experimentally validated ligand binding sites in the top 10 and 20 ranked Hotspots, respectively. Furthermore, we show that the model’s output Decision Function is a useful metric to predict binding sites and their potential druggability in new targets. Given the utility the SILCS method for ligand discovery and optimization, the tools presented represent an important advancement in the identification of orthosteric and allosteric binding sites and the discovery of drug-like molecules targeting those sites.

The chemokine receptor CXCR4 is a critical target for the treatment of several cancer types and HIV-1 infections. While orthosteric and allosteric modulators have been developed targeting its extracellular or transmembrane regions, the intramembrane region of CXCR4 may also include allosteric binding sites suitable for the development of allosteric drugs. To investigate this, we apply the Gaussian Network Model (GNM) to the monomeric and dimeric forms of CXCR4 to identify residues essential for its local and global motions located in the hinge regions of the protein. Residue interaction network (RIN) analysis suggests hub residues that participate in allosteric communication throughout the receptor. Mutual residues from the network models reside in regions with a high capacity to alter receptor dynamics upon ligand binding. We then investigate the druggability of these potential allosteric regions using the site identification by ligand competitive saturation (SILCS) approach, revealing two putative allosteric sites on the monomer and three on the homodimer. Two screening campaigns with Glide and SILCS-Monte Carlo docking using FDA-approved drugs suggest 20 putative hit compounds including antifungal drugs, anticancer agents, HIV protease inhibitors, and antimalarial drugs. In vitroassays considering mAB 12G5 and CXCL12 demonstrate both positive and negative allosteric activities of these compounds, supporting our computational approach. However, in vivofunctional assays based on the recruitment of β-arrestin to CXCR4 do not show significant agonism and antagonism at a single compound concentration. The present computational pipeline brings a new perspective to computer-aided drug design by combining conformational dynamics based on network analysis and cosolvent analysis based on the SILCS technology to identify putative allosteric binding sites using CXCR4 as a showcase.

Developing production quality CHARMM force-field (FF) parameters is a very detailed process involving a variety of calculations, many of which are specific for the molecule of interest. The first version of FFParam was developed as a standalone Python package designed for the optimization of electrostatic and bonded parameters of the CHARMM additive and polarizable Drude FFs by using quantum mechanical (QM) target data. The new version of FFParam has multiple new capabilities for FF parameter optimization and validation, with an emphasis on the ability to use condensed-phase target data in optimization. FFParam-v2 allows optimization of Lennard-Jones (LJ) parameters using potential energy scans of interactions between selected atoms in a molecule and noble gases, viz., He and Ne, and through condensed-phase calculations, from which experimental observables such as heats of vaporization and free energies of solvation may be obtained. This functionality serves as a gold standard for both optimizing parameters and validating the performance of the final parameters. A new bonded parameter optimization algorithm has been introduced to account for simultaneously optimizing multiple molecules sharing parameters. FFParam-v2 also supports the comparison of normal modes and the potential energy distribution of internal coordinates towards each normal mode obtained from QM and molecular mechanics calculations. Such comparison capability is vital to validate the balance among various bonded parameters that contribute to the complex normal modes of molecules. User interaction has been extended beyond the original graphical user interface to include command-line interface capabilities that allow for integration of FFParam in workflows, thereby facilitating the automation of parameter optimization. With these new functionalities, FFParam is a more comprehensive parameter optimization tool for both beginners and advanced users.

Molecular dynamics (MD) simulations are a commonly used method for investigating molecular behavior at the atomic level. Achieving reliable MD simulation results necessitates the use of an accurate force field. In the present work, we present a protocol to enhance the quality of group 1 monatomic ions (specifically Li+, Na+, K+, Rb+, and Cs+) with respect to their interactions with common polar model compounds in biomolecules in condensed phases in the context of the Drude polarizable force field. Instead of adjusting preexisting individual parameters for ions, model compounds, and water, we employ atom-pair specific Lennard-Jones (LJ) (known as NBFIX in CHARMM) and through-space Thole dipole screening (NBTHOLE) terms to fine-tune the balance of ion–model compound, ion–water, and model compound–water interactions. This involved establishing a protocol for the optimization of NBFIX and NBTHOLE parameters targeting the difference between molecular mechanical (MM) and quantum mechanical (QM) potential energy scans (PES). It is shown that targeting PES involving complexes that include multiple model compounds and/or ions as trimers and tetramers yields parameters that produce condensed phase properties in agreement with experimental data. Validation of this protocol involved the reproduction of experimental thermodynamic benchmarks, including solvation free energies of ions in methanol and N-methylacetamide, osmotic pressures, ionic conductivities, and diffusion coefficients within the condensed phase. These results show the importance of including more complex ion–model compound complexes beyond dimers in the QM target data to account for many-body effects during parameter fitting. The presented parameters represent a significant refinement of the Drude polarizable force field, which will lead to improved accuracy for modeling ion–biomolecular interactions.

Covalent drug design is an important component in drug discovery. Traditional drugs interact with their target in a reversible equilibrium, while irreversible covalent drugs increase the drug–target interaction duration by forming a covalent bond with targeted residues and thus may offer a more effective therapeutic approach. To facilitate the design of this class of ligands, computational methods can be used to help identify reactive nucleophilic residues, frequently cysteines, on a target protein for covalent binding, to test various warhead groups for their potential reactivities, and to predict noncovalent contributions to binding that can facilitate drug–target interactions that are important for binding specificity. To further aid covalent drug design, we extended a functional group mapping approach based on explicit solvent all-atom molecular simulations (SILCS: site identification by ligand competitive saturation) that intrinsically considers protein flexibility, functional group, and protein desolvation along with functional group–protein interactions. Through docking of a library of representative warhead fragments using SILCS-Monte Carlo (SILCS-MC), reactive cysteines can be correctly identified for proteins being tested. Furthermore, a machine learning model was trained to quantify the effectiveness of various warhead groups for proteins using metrics from SILCS-MC as well as experimental model compound warhead reactivity data. The ability to rank covalent molecular binders with similar warheads using SILCS ligand grid free energy (LGFE) ranking was also tested for several proteins. Based on these tools, an integrated SILCS-based workflow was developed, named SILCS-Covalent, which can both qualitatively and quantitatively inform covalent drug discovery.

Enveloped viruses depend on the host endoplasmic reticulum (ER) quality control (QC) machinery for proper glycoprotein folding. The endoplasmic reticulum quality control (ERQC) enzyme α-glucosidase I (α-GluI) is an attractive target for developing broad-spectrum antivirals. We synthesized 28 inhibitors designed to interact with all four subsites of the α-GluI active site. These inhibitors are derivatives of the iminosugars 1-deoxynojirimycin (1-DNJ) and valiolamine. Crystal structures of ER α-GluI bound to 25 1-DNJ and three valiolamine derivatives revealed the basis for inhibitory potency. We established the structure–activity relationship (SAR) and used the Site Identification by Ligand Competitive Saturation (SILCS) method to develop a model for predicting α-GluI inhibition. We screened the compounds against SARS-CoV-2 in vitroto identify those with greater antiviral activity than the benchmark α-glucosidase inhibitor UV-4. These host-targeting compounds are candidates for investigation in animal models of SARS-CoV-2 and for testing against other viruses that rely on ERQC for correct glycoprotein folding.

Molecular dynamic simulations are an effective tool to study complex molecular systems and are contingent upon the availability of an accurate and reliable molecular mechanics force field. The Drude polarizable force field, which allows for the explicit treatment of electronic polarization in a computationally efficient fashion, has been shown to reproduce experimental properties that were difficult or impossible to reproduce with the CHARMM additive force field, including peptide folding cooperativity, RNA hairpin structures, and DNA base flipping. Glycoproteins are essential components of glycoconjugate vaccines, antibodies, and many pharmaceutically important molecules, and an accurate polarizable force field that includes compatibility between the protein and carbohydrate aspect of the force field is essential to study these types of systems. In this work, we present an extension of the Drude polarizable force field to glycoproteins, including both N- and O-linked species. Parameter optimization focused on the dihedral terms using a reweighting protocol targeting NMR solution J-coupling data for model glycopeptides. Validation of the model include eight model glycopeptides and four glycoproteins with multiple N- and O-linked glycosylations. The new glycoprotein carbohydrate force field can be used in conjunction with the remainder of Drude polarizable force field through a variety of MD simulation programs including GROMACS, OPENMM, NAMD, and CHARMM and may be accessed through the Drude Prepper module in the CHARMM-GUI.

Generalized force fields (FFs) act as extensions to biomolecular FFs to provide a wide coverage of organic molecules. However, their precise application to an arbitrary molecule presents a separate challenge. We show that MATCH assigns different atom types and bonded and nonbonded parameters than CGenFF, and the AM1-BCC charge model, commonly used with GAFF/GAFF2, does not exactly reproduce the performance of the RESP charge model. The results indicate the need for caution when employing FFs to ensure their integrity with respect to their implementation and validation.

The Drude polarizable force field (FF) captures electronic polarization effects viaauxiliary Drude particles that are attached to non-hydrogen atoms, distinguishing it from commonly used additive FFs that rely on fixed charges. The Drude FF currently includes parameters for biomolecules such as proteins, nucleic acids, lipids, and carbohydrates and small-molecule representative of those classes of molecules as well as a range of atomic ions. Extension of the Drude FF to novel small druglike molecules is challenging as it requires the assignment of partial charges, atomic polarizabilities, and Thole scaling factors. In the present article, deep neural network (DNN) models are trained on quantum mechanical (QM)-based partial charges and atomic polarizabilities along with Thole scale factors trained to target QM molecular dipole moments and polarizabilities. Training of the DNN model used a collection of 39 421 molecules with molecular weights up to 200 Da and containing H, C, N, O, P, S, F, Cl, Br, or I atoms. The DNN model utilizes bond connectivity, including 1,2, 1,3, 1,4, and 1,5 terms and distances of Drude FF atom types as the feature vector to build the model, allowing it to capture both local and nonlocal effects in the molecules. Novel methods have been developed to determine restrained electrostatic potential (RESP) charges on atoms and external points representing lone pairs and to determine Thole scale factors, which have no QM analogue. A penalty scheme is devised as a performance predictor of the trained model. Validation studies show that these DNN models can precisely predict molecular dipole and polarizabilities of Food and Drug Administration (FDA)-approved drugs compared to reference MP2 calculations. The availability of the DNN model allowing for the rapid estimation of the Drude electrostatic parameters will facilitate its applicability to a wider range of molecular species.

Overexpression of the anti-apoptotic BCL-2 proteins is associated with the development and progression of a range of cancers. Venetoclax, an FDA-approved BCL-2 inhibitor, is fast becoming the standard-of-care for acute myeloid leukemia and chronic lymphocytic leukemia. However, the median survival offered by venetoclax is only 18 months (as part of a combination therapy regimen), and one of the primary culprits for this is the concomitant upregulation of sister anti-apoptotic proteins, in particular MCL-1 (and BCL-xL), which provides an escape route that manifests as venetoclax resistance. Since inhibition of BCL-xL leads to thrombocytopenia, we believe that a dual MCL-1/BCL-2 inhibitor may provide an enhanced therapeutic effect relative to a selective BCL-2 inhibitor. Beginning with a carboxylic acid-containing literature compound that is a potent inhibitor of MCL-1 and a moderate inhibitor of BCL-2, we herein describe our efforts to develop dual inhibitors of MCL-1 and BCL-2 by scaffold hopping from an indole core to an indazole framework. Subsequently, further elaboration of our novel N2-substituted, indazole-3-carboxylic acid lead into a family of indazole-3-acylsulfonamides resulted in improved inhibition of both MCL-1 and BCL-2, possibly through occupation of the p4 pocket, with minimal or no inhibition of BCL-xL.

Potential energy parameters for α-methyl amino acids were generated with ab initio calculations on α-methyl-N-acetylalanyl-N′-methylamide (the α-methyl “alanine dipeptide”) which served as an input to a grid-based correction to the backbone torsional potential (known as CMAP) consistent with the CHARMM36m additive protein force field. The new parameters were validated by comparison with experimentally determined helicities of the 22 residue C-terminal peptide (H10) from apolipoprotein A1 and five α-methylated variants in water and 0.3:0.7 trifluoroethanol (TFE)/water. Conventional molecular dynamics simulation totaling 30 μs for each peptide is in overall good agreement with the experiment, including the increased helicity in 30% TFE. An additional 500 ns of simulation using two-dimensional dihedral biasing (bpCMAP) replica exchange reduced left-handed conformations, increased right-handed helices, and thereby mostly decreased agreement with the experiment. Analysis of side chain–side chain salt bridges suggests that the overestimation of the helical content may be, in part, due to such interactions. The increased helicity of the peptides in 30% TFE arises from decreased hydrogen bonding of the backbone atoms to water and a concomitant increase in intramolecular backbone hydrogen bonds.

Nonstandard amino acids are both abundant in nature, where they play a key role in various cellular processes, and can be synthesized in laboratories, for example, for the manufacture of a range of pharmaceutical agents. In this work, we have extended the additive all-atom CHARMM36 and CHARMM General force field (CGenFF) to a large set of 333 nonstandard amino acids. These include both amino acids with nonstandard side chains, such as post-translationally modified and artificial amino acids, as well as amino acids with modified backbone groups, such as chromophores composed of several amino acids. Model compounds representative of the nonstandard amino acids were parametrized for protonation states that are likely at the physiological pH of 7 and, for some more common residues, in both d- and l-stereoisomers. Considering all protonation, tautomeric, and stereoisomeric forms, a total of 406 nonstandard amino acids were parametrized. Emphasis was placed on the quality of both intra- and intermolecular parameters. Partial charges were derived using quantum mechanical (QM) data on model compound dipole moments, electrostatic potentials, and interactions with water. Optimization of all intramolecular parameters, including torsion angle parameters, was performed against information from QM adiabatic potential energy surface (PES) scans. Special emphasis was put on the quality of terms corresponding to PES around rotatable dihedral angles. Validation of the force field was based on molecular dynamics simulations of 20 protein complexes containing different nonstandard amino acids. Overall, the presented parameters will allow for computational studies of a wide range of proteins containing nonstandard amino acids, including natural and artificial residues.

Fc glycosylation profoundly impacts the effector functions of antibodies and often dictates an antibody’s pro- or anti-inflammatory activities. It is well established that core fucosylation of the Fc domain N-glycans of an antibody significantly reduces its affinity for FcγRIIIa receptors and antibody-dependent cellular cytotoxicity (ADCC). Previous structural studies have suggested that the presence of a core fucose remarkably decreases the unique and favorable carbohydrate–carbohydrate interactions between the Fc and the receptor N-glycans, leading to reduced affinity. We report here that in contrast to natural core fucose, special site-specific modification on the core fucose could dramatically enhance the affinity of an antibody for FcγRIIIa. The site-selective modification was achieved through an enzymatic transfucosylation with a novel fucosidase mutant, which was shown to be able to use modified α-fucosyl fluoride as the donor substrate. We found that replacement of the core l-fucose with 6-azide- or 6-hydroxy-l-fucose (l-galactose) significantly enhanced the antibody’s affinity for FcγRIIIa receptors and substantially increased the ADCC activity. To understand the mechanism of the modified fucose-mediated affinity enhancement, we performed molecular dynamics simulations. Our data revealed that the number of glycan contacts between the Fc and the Fc receptor was increased by the selective core-fucose modifications, showing the importance of unique carbohydrate–carbohydrate interactions in achieving high FcγRIIIa affinity and ADCC activity of antibodies. Thus, the direct site-selective modification turns the adverse effect of the core fucose into a favorable force to promote the carbohydrate–carbohydrate interactions.

The development of the CHARMM lipid force field (FF) can be traced back to the early 1990s with its current version denoted CHARMM36 (C36). The parametrization of C36 utilized high-level quantum mechanical data and free energy calculations of model compounds before parameters were manually adjusted to yield agreement with experimental properties of lipid bilayers. While such manual fine-tuning of FF parameters is based on intuition and trial-and-error, automated methods can identify beneficial modifications of the parameters via their sensitivities and thereby guide the optimization process. This work introduces a semi-automated approach to reparametrize the CHARMM lipid FF with consistent inclusion of long-range dispersion through the Lennard-Jones particle-mesh Ewald (LJ-PME) approach. The optimization method is based on thermodynamic reweighting with regularization with respect to the C36 set. Two independent optimizations with different topology restrictions are presented. Targets of the optimizations are primarily liquid crystalline phase properties of lipid bilayers and the compression isotherm of monolayers. Pair correlation functions between water and lipid functional groups in aqueous solution are also included to address headgroup hydration. While the physics of the reweighting strategy itself is well-understood, applying it to heterogeneous, complex anisotropic systems poses additional challenges. These were overcome through careful selection of target properties and reweighting settings allowing for the successful incorporation of the explicit treatment of long-range dispersion, and we denote the newly optimized lipid force field as C36/LJ-PME. The current implementation of the optimization protocol will facilitate the future development of the CHARMM and related lipid force fields.

Despite the progress in modeling complex molecular systems of ever-increasing complexity, a quantitatively accurate computational treatment of ion permeation through narrow membrane channels remains challenging. An important factor to reach this goal is induced electronic polarization, which is likely to impact the permeation rate of small ions through narrow molecular pores. In this work, we extended the recently developed polarizable force field based on the classical Drude oscillators to assess the role of induced polarization effects on the energetics of sodium and potassium ion transport across the gramicidin A (gA) ion channel. The inclusion of induced polarization lowers barriers present in 1D potential of mean force (PMF) for cation permeation by ∼50% compared to those obtained with the additive force field. Conductance properties calculated with 1D PMFs from Drude simulations are in better agreement with experimental results. Polarization of single-file water molecules and protein atoms forming the narrow pore has a direct impact on the free-energy barriers and cation-specific solid-state NMR chemical shifts. Sensitivity analysis indicates that small changes to water–channel interactions can alter the free energy barrier for ion permeation. These results, illustrating polarization effects present in the complex electrostatic environment of the gA channel, have broad implications for revising proposed mechanisms of ion permeation and selectivity in a variety of ion channels.

Drug-induced cardiotoxicity is a potentially lethal and yet one of the most common side effects with the drugs in clinical use. Most of the drug-induced cardiotoxicity is associated with an off-target pharmacological blockade of K+currents carried out by the cardiac Human-Ether-a-go-go-Related (hERG1) potassium channel. There is a compulsory preclinical stage safety assessment for the hERG1 blockade for all classes of drugs, which adds substantially to the cost of drug development. The availability of a high-resolution cryogenic electron microscopy (cryo-EM) structure for the channel in its open/depolarized state solved in 2017 enabled the application of molecular modeling for rapid assessment of drug blockade by molecular docking and simulation techniques. More importantly, if successful, in silico methods may allow a path to lead-compound salvaging by mapping out key block determinants. Here, we report the blind application of the site identification by the ligand competitive saturation (SILCS) protocol to map out druggable/regulatory hotspots in the hERG1 channel available for blockers and activators. The SILCS simulations use small solutes representative of common functional groups to sample the chemical space for the entire protein and its environment using all-atom simulations. The resulting chemical maps, FragMaps, explicitly account for receptor flexibility, protein–fragment interactions, and fragment desolvation penalty allowing for rapid ranking of potential ligands as blockers or nonblockers of hERG1. To illustrate the power of the approach, SILCS was applied to a test set of 55 blockers with diverse chemical scaffolds and pIC50 values measured under uniform conditions. The original SILCS model was based on the all-atom modeling of the hERG1 channel in an explicit lipid bilayer and was further augmented with a Bayesian-optimization/machine-learning (BML) stage employing an independent literature-derived training set of 163 molecules. BML approach was used to determine weighting factors for the FragMaps contributions to the scoring function. pIC50 predictions from the combined SILCS/BML approach to the 55 blockers showed a Pearson correlation (PC) coefficient of >0.535 relative to the experimental data. SILCS/BML model was shown to yield substantially improved performance as compared to commonly used rigid and flexible molecular docking methods for a well-established cohort of hERG1 blockers, where no correlation with experimental data was recorded. SILCS/BML results also suggest that a proper weighting of protonation states of common blockers present at physiological pH is essential for accurate predictions of blocker potency. The precalculated and optimized SILCS FragMaps can now be used for the rapid screening of small molecules for their cardiotoxic potential as well as for exploring alternative binding pockets in the hERG1 channel with applications to the rational design of activators.

Protein therapeutics typically require a concentrated protein formulation, which can lead to self-association and/or high viscosity due to protein–protein interaction (PPI). Excipients are often added to improve stability, bioavailability, and manufacturability of the protein therapeutics, but the selection of excipients often relies on trial and error. Therefore, understanding the excipient–protein interaction and its effect on non-specific PPI is important for rational selection of formulation development. In this study, we validate a general workflow based on the site identification by ligand competitive saturation (SILCS) technology, termed SILCS-Biologics, that can be applied to protein therapeutics for rational excipient selection. The National Institute of Standards and Technology monoclonal antibody (NISTmAb) reference along with the CNTO607 mAb is used as model antibody proteins to examine PPIs, and NISTmAb was used to further examine excipient–protein interactions, in silico. Metrics from SILCS include the distribution and predicted affinity of excipients, buffer interactions with the NISTmAb Fab, and the relation of the interactions to predicted PPI. Comparison with a range of experimental data showed multiple SILCS metrics to be predictive. Specifically, the number of favorable sites to which an excipient binds and the number of sites to which an excipient binds that are involved in predicted PPIs correlate with the experimentally determined viscosity. In addition, a combination of the number of binding sites and the predicted binding affinity is indicated to be predictive of relative protein stability. Comparison of arginine, trehalose, and sucrose, all of which give the highest viscosity in combination with analysis of B22and kDand the SILCS metrics, indicates that higher viscosities are associated with a low number of predicted binding sites, with lower binding affinity of arginine leading to its anomalously high impact on viscosity. The present study indicates the potential for the SILCS-Biologics approach to be of utility in the rational design of excipients during biologics formulation.

Electronic polarization effects have been suggested to play an important role in proton binding to titratable residues in proteins. In this work, we describe a new computational method for pKacalculations, using Monte Carlo (MC) simulations to sample protein protonation states with the Drude polarizable force field and Poisson–Boltzmann (PB) continuum electrostatic solvent model. While the most populated protonation states at the selected pH, corresponding to residues that are half-protonated at that pH, are sampled using the exact relative free energies computed with Drude particles optimized in the field of the PB implicit solvation model, we introduce an approximation for the protein polarization of low-populated protonation states to reduce the computational cost. The highly populated protonation states used to compute the polarization and pKa’s are then iteratively improved until convergence. It is shown that for lysozyme, when considering 9 of the 18 titratable residues, the new method converged within two iterations with computed pKa’s differing only by 0.02 pH units from pKa’s estimated with the exact approach. Application of the method to predict pKa’s of 94 titratable side chains in 8 proteins shows the Drude-PB model to produce physically more correct results as compared to the additive CHARMM36 (C36) force field (FF). With a dielectric constant of two assigned to the protein interior the Root Mean Square (RMS) deviation between computed and experimental pKa’s is 2.07 and 3.19 pH units with the Drude and C36 models, respectively, and the RMS deviation using the Drude-PB model is relatively insensitive to the choice of the internal dielectric constant in contrast to the additive C36 model. At the higher internal dielectric constant of 20, pKa’s computed with the additive C36 model converge to the results obtained with the Drude polarizable force field, indicating the need to artificially overestimate electrostatic screening in a nonphysical way with the additive FF. In addition, inclusion of both synand antiorientations of the proton in the neutral state of acidic groups is shown to yield improved agreement with experiment. The present work, which is the first example of the use of a polarizable model for the prediction of pKa’s in proteins, shows that the use of a polarizable model represents a more physically correct model for the treatment of electrostatic contributions to pKashifts in proteins.

The CHARMM Drude-2013 polarizable force field (FF) was developed to include the explicit treatment of induced electronic polarizability, resulting in a more accurate description of the electrostatic interactions in molecular dynamics (MD) simulations. While the Drude-2013 protein FF has shown success in improving the folding properties of α-helical peptides and to reproduce experimental observables in simulations up to 1 μs, some limitations were noted regarding the stability of β-sheet structures in simulations longer than 100 ns as well as larger deviations from crystal structures in simulations of a number of proteins compared to the additive CHARMM36 protein FF. The origin of the instability has been identified and appears to be primarily due to overestimated atomic polarizabilities and induced dipole–dipole interactions on the Cβ, Cγ, and Cδ side chain atoms. To resolve this and other issues, a number of aspects of the model were revisited, resulting in Drude-2019 protein FF. Backbone parameters were optimized targeting the conformational properties of the (Ala)5peptide in solution along with gas phase properties of the alanine dipeptide. Dipeptides that contain N-acetylated and N′-methylamidated termini, excluding Gly, Pro, and Ala, were used as models to optimize the atomic polarizabilities and Thole screening factors on selected Cβ, Cγ, and Cδ carbons by targeting quantum mechanical (QM) dipole moments and molecular polarizabilities. In addition, to obtain better conformational properties, side chain χ1and χ2dihedral parameters were optimized targeting QM data for the respective side chain dipeptide conformations as well as Protein Data Bank survey data based on the χ1, χ2sampling from Hamiltonian replica-exchange MD simulations of (Ala)4-X-(Ala)4in solution, where X is the amino acid of interest. Further improvements include optimizing nonbonded interactions between charged residues to reproduce QM interaction energies of the charged-protein model compounds and experimental osmotic pressures. Validation of the optimized Drude protein FF includes MD simulations of a collection of peptides and proteins including β-sheet structures, as well as transmembrane ion channels. Results showed that the updated Drude-2019 protein FF yields smaller overall root-mean-square differences of proteins as compared to the additive CHARMM36m and Drude-2013 FFs as well as similar or improved agreement with experimental NMR properties, allowing for long time scale simulation studies of proteins and more complex biomolecular systems in conjunction with the remainder of the Drude polarizable FF.

RNA molecules perform a variety of biological functions for which the correct three-dimensional structure is essential, including as ribozymes where they catalyze chemical reactions. Metal ions, especially Mg2+, neutralize these negatively charged nucleic acids and specifically stabilize RNA tertiary structures as well as impact the folding landscape of RNAs as they assume their tertiary structures. Specific binding sites of Mg2+in folded conformations of RNA have been studied extensively; however, the full range of interactions of the ion with compact intermediates and unfolded states of RNA is challenging to investigate, and the atomic details of the mechanism by which the ion facilitates tertiary structure formation is not fully known. Here, umbrella sampling combined with oscillating chemical potential Grand Canonical Monte Carlo/molecular dynamics simulations are used to capture the energetics and atomic-level details of Mg2+-RNA interactions that occur along an unfolding pathway of the Twister ribozyme. The free energy profiles reveal stabilization of partially unfolded states by Mg2+, as observed in unfolding experiments, with this stabilization being due to increased sampling of simultaneous interactions of Mg2+with two or more nonsequential phosphate groups. Notably, these results indicate a push-pull mechanism in which the Mg2+-RNA interactions actually lead to destabilization of specific nonsequential phosphate-phosphate interactions (i.e., pushed apart), whereas other interactions are stabilized (i.e., pulled together), a balance that stabilizes unfolded states and facilitates the folding of Twister, including the formation of hydrogen bonds associated with the tertiary structure. This study establishes a better understanding of how Mg2+-ion interactions contribute to RNA structural properties and stability.

Sodium laureth sulfate (SLES) and fatty acids are common ingredients in many cosmetic products. Understanding how neutral and charged fatty acid compounds partition between micellar and water phases is crucial to achieve the optimal design of the product formulation. In this paper, we first study the formation of mixed SLES and fatty acid micelles using molecular dynamics (MD) simulations. Micelle/water partition coefficients of neutral and charged fatty acids are then calculated using COSMOmic as well as a MD approach based on the potential of mean force (PMF) calculations performed using umbrella sampling (US). The combined US/PMF approach was performed with both the additive, non-polarizable CHARMM general force field (CGenFF) and the classical Drude polarizable force field. The partition coefficients for the neutral solutes are shown to be accurately calculated with the COSMOmic and additive CGenFF US/PMF approaches, while only the US/PMF approach with the Drude polarizable force field accurately calculated the experimental partition coefficient of the charged solute. These results indicate the utility of the Drude polarizable force field as a tool for the rational development of mixed micelles.

Native folded and compact intermediate states of RNA typically involve tertiary structures in the presence of divalent ions such as Mg2+in a background of monovalent ions. In a recent study, we have shown how the presence of Mg2+impacts the transition from partially unfolded to folded states through a “push-pull” mechanism where the ion both favors and disfavors the sampling of specific phosphate-phosphate interactions. To further understand the ion atmosphere of RNA in folded and partially folded states results from atomistic umbrella sampling and oscillating chemical potential grand canonical Monte Carlo/molecular dynamics (GCMC/MD) simulations are used to obtain atomic-level details of the distributions of Mg2+and K+ions around Twister RNA. Results show the presence of 100 mM Mg2+to lead to increased charge neutralization over that predicted by counterion condensation theory. Upon going from partially unfolded to folded states, overall charge neutralization increases at all studied ion concentrations that, while associated with an increase in the number of direct ion-phosphate interactions, is fully accounted for by the monovalent K+ions. Furthermore, K+preferentially interacts with purine N7 atoms of helical regions in partially unfolded states, thereby potentially stabilizing the helical regions. Thus, both secondary helical structures and formation of tertiary structures leads to increased counterion condensation, thereby stabilizing those structural features of Twister. Notably, it is shown that K+can act as a surrogate for Mg2+by participating in specific interactions with nonsequential phosphate pairs that occur in the folded state, explaining the ability of Twister to self-cleave at submillimolar Mg2+concentrations.

In this work, we report the development of Drude polarizable force field parameters for the carboxylate and N-acetyl amine derivatives, extending the functionality of the existing Drude polarizable carbohydrate force field. The force field parameters have been developed in a hierarchical manner, reproducing the quantum mechanical gas-phase properties of small model compounds representing the key functional group in the carbohydrate derivatives, including optimization of the electrostatic and bonded parameters. The optimized parameters were then used to generate the models for carboxylate and N-acetyl amine carbohydrate derivatives. The transferred parameters were further tested and optimized to reproduce crystal geometries and J-coupling data from nuclear magnetic resonance experiments. The parameter development resulted in the incorporation of d-glucuronate, l-iduronate, N-acetyl-d-glucosamine (GlcNAc), and N-acetyl-d-galactosamine (GalNAc) sugars into the Drude polarizable force field. The parameters developed in this study were then applied to study the conformational properties of glycosaminoglycan polymer hyaluronan, composed of d-glucuronate and N-acetyl-d-glucosamine, in aqueous solution. Upon comparing the results from the additive and polarizable simulations, it was found that the inclusion of polarization improved the description of the electrostatic interactions observed in hyaluronan, resulting in enhanced conformational flexibility. The developed Drude polarizable force field parameters in conjunction with the remainder of the Drude polarizable force field parameters can be used for future studies involving carbohydrates and their conjugates in complex, heterogeneous systems.

In the domain of computer-aided drug design, achieving precise and accurate estimates of ligand–protein binding is paramount in the context of screening extensive drug libraries and performing ligand optimization. A fundamental aspect of the SILCS (site identification by ligand competitive saturation) methodology lies in the generation of comprehensive 3D free-energy functional group affinity maps (FragMaps), encompassing the entirety of the target molecule structure. These FragMaps offer an intricate landscape of functional group affinities across the protein, bilayer, or RNA, acting as the basis for subsequent SILCS-Monte Carlo (MC) simulations wherein ligands are docked to the target molecule. To augment the efficiency and breadth of ligand sampling capabilities, we implemented an improved SILCS-MC methodology. By harnessing the parallel computing capability of GPUs, our approach facilitates concurrent calculations over multiple ligands and binding sites, markedly enhancing the computational efficiency. Moreover, the integration of a genetic algorithm (GA) with MC allows us to employ an evolutionary approach to perform ligand sampling, assuring enhanced convergence characteristics. In addition, the potential utility of parallel tempering (PT) to improve sampling was investigated. Implementation of SILCS-MC on GPU architecture is shown to accelerate the speed of SILCS-MC calculations by over 2-orders of magnitude. Use of GA and PT yield improvements over Markov-chain MC, increasing the precision of the resultant docked orientations and binding free energies, though the extent of improvements is relatively small. Accordingly, significant improvements in speed are obtained through the GPU implementation with minor improvements in the precision of the docking obtained via the tested GA and PT algorithms.

Additive force fields are designed to account for induced electronic polarization in a mean-field average way, using effective empirical fixed charges. The limitation of this approximation is cause for serious concerns, particularly in the case of lipid membranes, where the molecular environment undergoes dramatic variations over microscopic length scales. A polarizable force field based on the classical Drude oscillator offers a practical and computationally efficient framework for an improved representation of electrostatic interactions in molecular simulations. Building on the first-generation Drude polarizable force field for the dipalmitoylphosphatidylcholine 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) molecule, the present effort was undertaken to improve this initial model and expand the force field to a wider range of phospholipid molecules. New lipids parametrized include dimyristoylphosphatidylcholine (DMPC), dilauroylphosphatidylcholine (DLPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), dipalmitoylphosphatidylethanolamine (DPPE), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), and 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE). The iterative optimization protocol employed in this effort led to lipid models that achieve a good balance between reproducing quantum mechanical data on model compound representative of phospholipids and reproducing a range of experimental condensed phase properties of bilayers. A parametrization strategy based on a restrained ensemble–maximum entropy methodology was used to help accurately match the experimental NMR order parameters in the polar headgroup region. All the parameters were developed to be compatible with the remainder of the Drude polarizable force field, which includes water, ions, proteins, DNA, and selected carbohydrates.

Development of accurate force field parameters for molecular ions in the context of a polarizable energy function based on the classical Drude oscillator is a crucial step toward an accurate polarizable model for modeling and simulations of biological macromolecules. Toward this goal we have undertaken a hierarchical approach in which force field parameter optimization is initially performed for small molecules for which experimental data exists that serve as building blocks of macromolecular systems. Small molecules representative of the ionic moieties of biological macromolecules include the cationic ammonium and methyl substituted ammonium derivatives, imidazolium, guanidinium and methylguanidinium, and the anionic acetate, phenolate, and alkanethiolates. In the present work, parameters for molecular ions in the context of the Drude polarizable force field are optimized and compared to results from the nonpolarizable additive CHARMM general force field (CGenFF). Electrostatic and Lennard-Jones parameters for the model compounds are developed in the context of the polarizable SWM4-NDP water model, with emphasis on assuring that the hydration free energies are consistent with previously reported parameters for atomic ions. The final parameters are shown to be in good agreement with the selected quantum mechanical (QM) and experimental target data. Analysis of the structure of water around the ions reveals substantial differences between the Drude and additive force fields indicating the important role of polarization in dictating the molecular details of aqueous solvation. The presented parameters represent the foundation for the charged functionalities in future generations of the Drude polarizable force field for biological macromolecules as well as for drug-like molecules.

Chemical fragment cosolvent sampling techniques have become a versatile tool in ligand-protein binding prediction. Site-identification by ligand competitive saturation (SILCS) is one such method that maps the distribution of chemical fragments on a protein as free energy fields called FragMaps. Ligands are then simulated via Monte Carlo techniques in the field of the FragMaps (SILCS-MC) to predict their binding conformations and relative affinities for the target protein. Application of SILCS-MC using a number of different scoring schemes and MC sampling protocols against multiple protein targets was undertaken to evaluate and optimize the predictive capability of the method. Seven protein targets and 551 ligands with broad chemical variability were used to evaluate and optimize the model to maximize Pearson’s correlation coefficient, Pearlman’s predictive index, correct relative binding affinity, and root-mean-square error versus the absolute experimental binding affinities. Across the protein–ligand sets, the relative affinities of the ligands were predicted correctly an average of 69% of the time for the highest overall SILCS protocol. Training the FragMap weighting factors using a Bayesian machine learning (ML) algorithm led to an increase to an average 75% relative correct affinity predictions. Furthermore, once the optimal protocol is identified for a specific protein–ligand system average predictabilities of 76% are achieved. The ML algorithm is successful with small training sets of data (30 or more compounds) due to the use of physically correct FragMap weights as priors. Notably, the 76% correct relative prediction rate is similar to or better than free energy perturbation methods that are significantly computationally more expensive than SILCS. The results further support the utility of SILCS as a powerful and computationally accessible tool to support lead optimization and development in drug discovery.

The derivation of atomic polarizabilities for polarizable force field development has been a long-standing problem. Atomic polarizabilities were often refined manually starting from tabulated values, rendering an automated assignment of parameters difficult and hampering reproducibility and transferability of the obtained values. To overcome this, we trained both a linear increment scheme and a multilayer perceptron neural network on a large number of high-quality quantum mechanical atomic polarizabilities and partial atomic charges, where only the type of each atom and its connectivity were used as input. The predicted atomic polarizabilities and charges had average errors of 0.023 Å3and 0.019 e using the neural net and 0.063 Å3and 0.069 e using the simple increment scheme. As the algorithm relies only on the connectivities of the atoms within a molecule, thus omitting dependencies on the three-dimensional conformation, the approach naturally assigns like charges and polarizabilities to symmetrical groups. Accordingly, a convenient utility is presented for generating the partial atomic charges and atomic polarizabilities for organic molecules as needed in polarizable force field development.

NSD2, a histone methyltransferase specific for methylation of histone 3 lysine 36 (H3K36), exhibits a glutamic acid to lysine mutation at residue 1099 (E1099K) in childhood acute lymphocytic leukemia (ALL), and cells harboring this mutation can become the predominant clone in relapsing disease. We studied the effects of this mutant enzyme in silico, in vitro, and in vivo using gene edited cell lines. The E1099K mutation altered enzyme/substrate binding and enhanced the rate of H3K36 methylation. As a result, cell lines harboring E1099K exhibit increased H3K36 dimethylation and reduced H3K27 trimethylation, particularly on nucleosomes containing histone H3.1. Mutant NSD2 cells exhibit reduced apoptosis and enhanced proliferation, clonogenicity, adhesion, and migration. In mouse xenografts, mutant NSD2 cells are more lethal and brain invasive than wildtype cells. Transcriptional profiling demonstrates that mutant NSD2 aberrantly activates factors commonly associated with neural and stromal lineages in addition to signaling and adhesion genes. Identification of these pathways provides new avenues for therapeutic interventions in NSD2 dysregulated malignancies.

Halogenated ligands can participate in nonbonding interactions with proteins via halogen bond (XB) or halogen–hydrogen bond donor (X-HBD) interactions. In the context of molecular dynamics (MD) simulations, the accuracy of the simulations depends strongly on the force field (FF) used. To ensure good reproduction of XB and X-HBD interactions with proteins, we optimized the previously developed additive CHARMM36/CHARMM General force field (CGenFF) and Drude polarizable force field by including atom pair-specific Lennard-Jones parameters for aromatic halogen–protein interactions. The optimization targeted quantum mechanical interaction energy surfaces with the developed parameters then examined for their ability to reproduce experimental halogen-containing ligand–protein interactions in MD simulations. The calculated halogenated ligand interaction geometries were in good overall agreement with the experimental crystal data for both the polarizable and additive FFs, showing that these models can accurately treat both XB and X-HBD interactions. Analysis of the ligand–protein interactions shows significant contributions of polarizability to binding occurring in the Drude FF, with self-polarization energy making both favorable and unfavorable contributions to binding. Further analysis of the dipole moments from aqueous solution to protein indicates the polarizable FF accounts for subtle changes of the environment of the ligands that can impact binding. The present work demonstrates the utility of the updated additive CHARMM36/CGenFF and polarizable Drude FFs for the study of halogenated ligand–protein interactions in computer-aided drug design.

Cation-π interactions between tryptophan and choline or trimethylated lysines are vital for many biological processes. The performance of the additive CHARMM36 force field against target quantum mechanical data is shown to reproduce QM equilibrium geometries but required modified Lennard-Jones potentials to accurately reproduce the QM interaction energies. The modified parameter set allows accurate modeling, including free energies, of cation-π indole-choline and indole-trimethylated lysines interactions relevant for protein–ligand, protein–membrane, and protein–protein interfaces.

Lipid membrane permeation of drug molecules was investigated with Heterogeneous Dielectric Generalized Born (HDGB)-based models using solubility-diffusion theory and machine learning. Free energy profiles were obtained for neutral molecules by the standard HDGB and Dynamic HDGB (DHDGB) to account for the membrane deformation upon insertion of drugs. We also obtained hybrid free energy profiles where the neutralization of charged molecules was taken into account upon membrane insertion. The evaluation of the predictions was done against experimental permeability coefficients from Parallel Artificial Membrane Permeability Assays (PAMPA), and effects of partial charge sets, CGenFF, AM1-BCC, and OPLS, on the performance of the predictions were discussed. (D)HDGB-based models improved the predictions over the two-state implicit membrane models, and partial charge sets seemed to have a strong impact on the predictions. Machine learning increased the accuracy of the predictions, although it could not outperform the physics-based approach in terms of correlations.

Grand canonical Monte Carlo (GCMC) simulations of ionic solutions with explicit solvent models are known to be challenging. One challenge arises from the treatment of long-range electrostatics and finite-box size in Monte Carlo simulations when periodic boundary condition and Ewald summation methods are used. Another challenge is that constant excess chemical potential GCMC simulations for charged solutes suffer from inadequate insertion and deletion acceptance ratios. In this work, we address those problems by implementing an oscillating excess chemical potential GCMC algorithm with smooth particle mesh Ewald and finite-box-size corrections to treat the long-range electrostatics. The developed GCMC simulation program was combined with GROMACS to perform GCMC/MD simulations of ionic solutions individually containing Li+, Na+, K+, Rb+, Cs+, F–, Cl–, Br–, I–, Ca2+, and Mg2+, respectively. Our simulation results show that the combined GCMC/MD approach can approximate the ionic hydration free energies with proper treatment of long-range electrostatics. Our developed simulation approach can open up new avenues for simulating complex chemical and biomolecular systems and for drug discovery.

Protein–protein interactions (PPI) between the transcriptional repressor B-cell lymphoma 6 (BCL6) BTB domain (BCL6BTB) and its corepressors have emerged as a promising target for anticancer therapeutics. However, identification of potent, drug-like inhibitors of BCL6BTBhas remained challenging. Using NMR-based screening of a library of fragment-like small molecules, we have identified a thiourea compound (7CC5) that binds to BCL6BTB. From this hit, the application of computer-aided drug design (CADD), medicinal chemistry, NMR spectroscopy, and X-ray crystallography has yielded an inhibitor, 15f, that demonstrated over 100-fold improved potency for BCL6BTB. This gain in potency was achieved by a unique binding mode that mimics the binding mode of the corepressor SMRT in the aromatic and the HDCH sites. The structure–activity relationship based on these new inhibitors will have a significant impact on the rational design of novel BCL6 inhibitors, facilitating the identification of therapeutics for the treatment of BCL6-dependent tumors.

Phosphate groups are essential components of nucleic acids and proteins, whose interactions with solvent, metal ions, and ionic side chains help control folding and binding. Methyl phosphate (MP) represents a simple analog of phosphate moieties that are post-translation modifications in proteins and present at the termini of nucleic acids, among other environments. In the present study, we optimized parameters for use in polarizable molecular dynamics simulations of MP in its mono- and dianionic forms, MP–≡ CH3HPO4–and MP2–≡ CH3PO42–, along with Pi2–≡ HPO42–, in the context of the classical Drude oscillator model. Parameter optimization was done in a manner consistent with the remainder of the Drude molecular mechanics force field, choosing atomic charges and polarizabilities to reproduce molecular properties from quantum mechanics as well as experimental hydration free energies. Optimized parameters were similar to existing dimethyl phosphate parameters, with a few significant differences. The developed parameters were then used to compute magnesium binding affinities in aqueous solution, using alchemical molecular dynamics free energy simulations. Good agreement with experiment was obtained, and outer sphere binding was shown to be predominant for MP–and MP2–.

We present an extension of the CHARMM Drude polarizable force field to enable modeling of polysaccharides containing pyranose and furanose monosaccharides. The new force field parameters encompass 1↔2, 1→3, 1→4, and 1→6 pyranose–furanose linkages, 2→1 and 2→6 furanose–furanose linkages, 2→2, 2→3, and 2→4 furanose–pyranose, and 1↔1, 1→2, 1→3, 1→4, and 1→6 pyranose–pyranose linkages. For the glycosidic linkages, both simple model compounds and the full disaccharides with methylation at the reducing end were used for parameter optimization. The model compounds were chosen to be monomers or glycosidic-linked dimers of tetrahydropyran (THP) and tetrahydrofuran (THF). Target data for optimization included one- and two-dimensional potential energy scans of ω and the Φ/Ψ glycosidic dihedral angles in the model compounds and full disaccharides computed by quantum mechanical (QM) RIMP2/cc-pVQZ single point energies on MP2/6-31G(d) optimized structures. Also included in the target data are extensive sets of QM gas phase monohydrate water–saccharide interactions, dipole moments, and molecular polarizabilities for both model compounds and full disaccharides. The resulting polarizable model is shown to be in good agreement with a range of QM data, offering a significant improvement over the additive CHARMM36 carbohydrate force field, as well as experimental data including crystal structures and conformational properties of disaccharides and a trisaccharide in aqueous solution.

Meaningful efforts in computer-aided drug design (CADD) require accurate molecular mechanical force fields to quantitatively characterize protein–ligand interactions, ligand hydration free energies, and other ligand physical properties. Atomic models of new compounds are commonly generated by analogy from the predefined tabulated parameters of a given force field. Two widely used approaches following this strategy are the General Amber Force Field (GAFF) and the CHARMM General Force Field (CGenFF). An important limitation of using pretabulated parameter values is that they may be inadequate in the context of a specific molecule. To resolve this issue, we previously introduced the General Automated Atomic Model Parameterization (GAAMP) for automatically generating the parameters of atomic models of small molecules, using the results from ab initio quantum mechanical (QM) calculations as target data. The GAAMP protocol uses QM data to optimize the bond, valence angle, and dihedral angle internal parameters, and atomic partial charges. However, since the treatment of van der Waals interactions based on QM is challenging and may often be unreliable, the Lennard-Jones 6–12 parameters are kept unchanged from the initial atom types assignments (GAFF or CGenFF), which limits the accuracy that can be achieved by these models. To address this issue, a new set of Lennard-Jones 6–12 parameters was systematically optimized to reproduce experimental neat liquid densities and enthalpies of vaporization for a large set of 430 compounds, covering a wide range of chemical functionalities. Calculations of the hydration free energy indicate that optimal accuracy for these models is achieved when the molecule–water van der Waals dispersion is rescaled by a factor of 1.115. The final optimized model yields an average unsigned error of 0.79 kcal/mol in the hydration free energies.