Download Hi-Res ImageDownload to MS-PowerPointCite This:J. Chem. Inf. Model. 2017, 57, 4, 897-909

Evaluation and Characterization of Trk Kinase Inhibitors for the Treatment of Pain: Reliable Binding Affinity Predictions from Theory and Computation

View Author Information

† Centre for Computational Science, Department of Chemistry, University College London, London WC1H 0AJ, United Kingdom

‡ Pfizer Global R&D, The Portway Building, Granta Park, Cambridge CB21 6GS, United Kingdom

§ Pfizer Worldwide Research and Development, 610 Main Street, Cambridge, Massachusetts 02139, United States

∥ Pfizer Worldwide Research and Development, Groton Laboratories, Groton, Connecticut 06340, United States

*E-mail: [email protected]

Cite this: J. Chem. Inf. Model. 2017, 57, 4, 897–909

Publication Date (Web):March 20, 2017

https://doi.org/10.1021/acs.jcim.6b00780

Article Views

2044

Altmetric

Citations

LEARN ABOUT THESE METRICS

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

PDF (9 MB)

Get e-Alerts

Supporting Info (1)»Supporting Information Supporting Information

SUBJECTS:

Go to Journal of Chemical Information and Modeling

Get e-Alerts

Abstract

Optimization of ligand binding affinity to the target protein of interest is a primary objective in small-molecule drug discovery. Until now, the prediction of binding affinities by computational methods has not been widely applied in the drug discovery process, mainly because of its lack of accuracy and reproducibility as well as the long turnaround times required to obtain results. Herein we report on a collaborative study that compares tropomyosin receptor kinase A (TrkA) binding affinity predictions using two recently formulated fast computational approaches, namely, Enhanced Sampling of Molecular dynamics with Approximation of Continuum Solvent (ESMACS) and Thermodynamic Integration with Enhanced Sampling (TIES), to experimentally derived TrkA binding affinities for a set of Pfizer pan-Trk compounds. ESMACS gives precise and reproducible results and is applicable to highly diverse sets of compounds. It also provides detailed chemical insight into the nature of ligand–protein binding. TIES can predict and thus optimize more subtle changes in binding affinities between compounds of similar structure. Individual binding affinities were calculated in a few hours, exhibiting good correlations with the experimental data of 0.79 and 0.88 from the ESMACS and TIES approaches, respectively. The speed, level of accuracy, and precision of the calculations are such that the affinity predictions can be used to rapidly explain the effects of compound modifications on TrkA binding affinity. The methods could therefore be used as tools to guide lead optimization efforts across multiple prospective structurally enabled programs in the drug discovery setting for a wide range of compounds and targets.

Introduction

ARTICLE SECTIONS

Jump To

The availability of computational methods that can reliably, rapidly, and accurately predict the binding affinities of ligands to a target protein of interest would greatly facilitate drug discovery programs by enabling project teams to more effectively triage design ideas and therefore synthesize only those compounds with a high probability of being pharmacologically active. The overall effect of this approach would be to reduce the number of “design–synthesis–test” cycles needed to generate compounds of sufficient bioactivity to progress to the clinic. Until recently, in silico methods of binding affinity prediction have not been regarded as reliable enough to produce such actionable results. The American Chemical Society’s Cross-Pharmaceutical Industry group has been discussing this particular challenge of conducting elaborative studies and has come out with an opinion piece recently, in which it is specifically noted that “coordinated, blinded prediction challenges offer the best opportunity to develop broad understanding and broadly applicable methods”. (1) The Coveney group has now developed a suite of computational methods that deliver rapid, accurate, precise, and reliable binding affinity predictions. We report here a prospective computational study on a then-ongoing project at Pfizer (2) to assess the effectiveness of these methods based on the use of a Binding Affinity Calculator (BAC) software tool and associated services. (3) The approach makes use of an automated workflow (4) running in a high-performance computing environment that builds models, runs large numbers of calculations, and analyzes the output data in order to place reliable error bounds on predicted ligand binding affinities. In order to assess the reliability of these methods, the predictions were performed blind at UCL and subsequently compared with the experimentally determined TrkA binding affinity data. (2) For a new method being pursued by an industrial computational chemist, these types of live blinded individual project team collaborations are the best kinds of pilot studies that can be performed.

Tropomyosin receptor kinase A (TrkA) is used as the target protein in this study. TrkA is the cognate receptor of the nerve growth factor (NGF) neuropeptide, a neurotrophic factor involved in the regulation of growth, maintenance, proliferation, and survival of certain target neurons. (5) Preclinical and clinical studies have identified a crucial role for NGF in the pathogenesis of pain; the clinical efficacy of anti-NGF monoclonal antibody (mAb) therapies against several pain end points is well-documented, and preclinically the inhibition of the TrkA kinase domain by small-molecule kinase inhibitors has been shown to reverse the effects of NGF-mediated pain transduction. (6, 7) In view of the key role of NGF in modulating pain, there is significant interest in the clinical development of small-molecule TrkA inhibitors to complement anti-NGF mAb treatment options.

The UCL group has recently introduced new approaches termed Enhanced Sampling of Molecular dynamics with Approximation of Continuum Solvent (ESMACS) and Thermodynamic Integration with Enhanced Sampling (TIES) (8) for the reliable prediction of ligand–protein free energies. The ESMACS approach centers on the molecular mechanics Poisson–Boltzmann surface area (MMPBSA) method, (9) which employs a continuum approximation for the aqueous solvent, while TIES is based on thermodynamic integration. The approaches emphasize the necessity of invoking ensemble-based sampling to reliably compute macroscopic quantities using microscopic modeling methods. (10) The BAC tool (3) is employed to automate much of the complexity of running and marshalling the required molecular dynamics simulations as well as collecting and analyzing data. The BAC tool utilized to perform the ESMACS and TIES studies constitutes a computational pipeline built from an array of software tools and services and requires access to suitable computing resources. Integration and automation are central to the reliability of the method, ensuring that the results are reproducible and can be delivered rapidly.

Computational Methods

ARTICLE SECTIONS

Jump To

The TrkA binding affinities of the compounds in Table 1 were calculated using ESMACS and TIES. These compounds were picked as representatives of the full series previously studied experimentally, (2) covering the dynamic range of the TrkA pharmacology assay and containing the structural features deemed as key determinants of TrkA activity. (A second series of Pfizer pan-Trk inhibitors, structurally dissimilar to those highlighted in Table 1, also showed good agreement between their ESMACS-predicted and experimentally determined TrkA binding affinities. Studies of those inhibitors will be published in due course.) These congeneric compounds all have the same net neutral charge. In ESMACS, the MMPBSA.py.MPI (11) module of the AMBER12 package (12) was employed for the free energy calculations of the complex (G_complex), the receptor (G_receptor), and the compound (G_ligand) (eq 1). The polar solvation free energy was calculated using the Poisson–Boltzmann equation with a grid spacing of 0.5 Å and dielectric constants of 1 and 80 for the protein and solvent, respectively. The energetic analyses, including the configurational entropy calculations, were conducted on 50 snapshots of complexes, protein, and compounds extracted evenly from each 4 ns production run in the ensemble MD simulations of complexes only (1-traj and 2-traj; see below) or from separate ensemble simulations of complexes and compounds (3-traj). In TIES, the energy derivatives ∂V/∂λ (eq 2) were recorded every 2 fs during the simulations, and a stochastic integration (10, 13) was performed using a trapezoidal method. Hydrogen-bond analyses were performed for the ensemble simulation trajectories. A hydrogen bond is considered to be formed when the distance between a hydrogen-bond acceptor and a hydrogen-bond donor is less than a defined distance cutoff and the acceptor–hydrogen–donor angle is greater than an angle cutoff. The cutoffs used in the current study are 3.0 Å for the distance and 135° for the angle. The protein and the compounds can have intramolecular and intermolecular hydrogen bonds, of which the intermolecular ones contribute significantly to the binding affinities. Some “bridging” water molecules, which are hydrogen-bonded to the protein and the compounds at the same time, also play an important role in binding of ligands to proteins, in addition to their direct interactions. In ESMACS predictions, the inter- and intramolecular interactions of the compounds and the protein are calculated explicitly, while the interactions with water, including such bridging water molecules, are taken into account implicitly.

Table 1. Ligands Considered in This Study, Numbered as Per the Order in the Files Provided by Pfizer^a

^{Table a}

All of the ligands have the same net neutral charge. The experimental TrkA inhibitory values (IC₅₀) and the binding free energies derived from them are shown. Experimental IC₅₀ measurements were conducted independently in two separate laboratories using an identical protocol; (2) Pfizer, Sandwich (U.K.) IC₅₀ values are shown in black; TCG Lifescience (India) IC₅₀ values are shown in blue.

In ESMACS, the free energy is evaluated approximately on the basis of the extended MMPBSA method, (8, 14) including configurational entropy and the free energy of association. (15) Free energy changes (ΔG_binding) are determined for the molecules in their solvated states. The binding free energy change is then calculated as

(1)where G_complex, G_receptor, and G_ligand are the free energies of the complex, the receptor, and the ligand, respectively.

The three terms on the right-hand side of eq 1 can be generated from single simulations of the complexes or from separate simulations of complexes, receptor(s), and ligand(s). The former is the so-called one-trajectory (1-traj) method, in which the trajectories of the receptor(s) and ligand(s) are extracted from those of the complexes. The latter is the so-called three-trajectory (3-traj) method when the energies are derived from independent simulations of the three components or the two-trajectory (2-traj) method when the energy is derived from independent simulations for only the receptor or the ligand. In the drug development field, binding is usually investigated for a set of ligands bound to the same protein target. The free energy of the receptor, G_receptor, is then the same for all of ligands when it is derived from an independent simulation of the receptor and hence can be treated as a constant. In the current study, we employ three approaches to calculate ΔG_binding: (a) employing only simulations of the complexes (1-traj); (b) the same as (a) but using the average of the receptor free energies ⟨G_receptor⟩ from the 1-traj method (denoted as 2-traj); and (c) the same as (b) but also invoking separate ligand simulations for the derivation of G_ligand (denoted as 3-traj).

In our more recent TIES method, (13) an “alchemical transformation” (16) for the mutated entity, either the ligand or the protein (in this study it is always the ligand), is used in both aqueous solution and within the ligand–protein complex. The relative free energy changes for the alchemical mutation processes, ΔG_aq^alch and ΔG_complex^alch, are calculated as

(2)where λ (0 ≤ λ ≤ 1) is a coupling parameter such that λ = 0 and λ = 1 correspond to the initial and final thermodynamic states, V(λ) is the potential energy of an intermediate state λ, and ⟨···⟩_λ denotes an ensemble average over configurations representative of the state λ. The relative binding free energy difference is then calculated as

(3)

Compared with ESMACS, the TIES approach, like all other alchemical-based free energy methods, is usually more accurate in comparing one congeneric ligand to another. However, the nature of these alchemical methods implies that their applications will be limited when diverse sets of compounds are of interest. A recent publication (17) in which the authors claimed that their implementation of free energy perturbation (FEP) calculations successfully predicted a number of more active compounds (a correlation coefficient of 0.71 between the predicted and experimental binding affinities was achieved) has the same limitations. ESMACS, however, has no such limitations. Our previous studies (8) have shown that ESMACS can be applied to sets of compounds that are highly diverse in terms of both the number of atoms and the net charge. In addition, ESMACS is able to make assessments of the effect that an individual ligand has on the protein to which it binds. In particular, the application of two- and three-trajectory versions of ESMACS allows one to probe the extent to which both the ligand and protein adjust their conformations on binding, broadly according to a “lock and key” or “induced fit” recognition mechanism.

In the current study, we apply the ESMACS approach for all of TrkA compounds highlighted in Table 1 and consider TIES for a selected subset. In the evaluation and characterization of TrkA binding affinity predictions from the Pfizer compound set, the ESMACS and TIES protocols were conducted as recently described. (8, 13) TrkA compounds were optimized at the Hartree–Fock level with the 6-31G* basis (HF/6-31G*) in Gaussian 03 (18) and parametrized using Antechamber and RESP in AmberTools 12 with the general AMBER force field (GAFF). (19) The Amber ff99SBildn force field (20) was used for the protein. In ESMACS, we used 25 replicas in an ensemble calculation for each ligand. In TIES, five replicas were used for each selected pair of ligands. The MD package NAMD2.9 (21) was used throughout the equilibration and production simulations with periodic boundary conditions. We used a protocol established in our previous publications (13, 14) in which 2 ns of equilibration and 4 ns of production were conducted for each replica. Energy analyses showed that a 4 ns production run was sufficient to have convergence results for the current molecular systems (see Figure S1 in the Supporting Information). In ESMACS, the MMPBSA.py.MPI (11) module of the Amber package (12) was used to extract the free energies of the complexes, the protein, and the compounds (eq 1), including configurational entropy from normal mode calculations.

ESMACS was performed on SuperMUC, two separate high-end supercomputers with a total of ca. 245 000 cores in a combination of various processor technologies (https://www.lrz.de/services/compute/supermuc/systemdescription/), at the Leibniz Rechenzentrum (Leibniz Supercomputing Centre) in Garching, Germany. We were able to run our calculations on them as if it were a single supercomputer. (22) TIES calculations were performed on ARCHER, a Cray XC30 supercomputer (equipped with ca. 110 000 cores), the U.K.’s National High Performance Computing Service located in Edinburgh. A single ESMACS study of one ligand–protein interaction and a TIES study of the binding free energy difference for two congeneric ligands can both be completed in about 8 wall-clock hours on 1200 and 3120 CPU cores, respectively, on ARCHER (see details in our previous publication (8, 13)). (It is possible to further reduce the time required by use of GPU accelerators.) The ESMACS and TIES approaches are both scalable, allowing a large number of calculations to be performed concurrently, depending only on the computing resources available.

Results

ARTICLE SECTIONS

Jump To

We report a collaborative study performed between academic computational chemists and a pharmaceutical company on a data set of 16 pan-Trk ligands (Table 1). (2) Of the TrkA cocrystal structures provided by Pfizer, (2) we elected to utilize the cocrystal structure of TrkA and 1 (Figure 1, PDB code 5JFV) as it contained the highest number of crystallographically defined TrkA residues. The missing residues were built by ModLoop, (23) while the initial structures of the complexes were constructed by a docking method employing UCSF DOCK. (24) Nine of the 16 compounds were successfully docked into the binding site of the TrkA protein with orientations agreeing with those from available X-ray structures. Compounds that failed to generate suitable cocomplexes via docking were manually positioned on the basis of X-ray and/or modeled structures of similar compounds.

Figure 1. Crystal structure of 1 bound to TrkA, viewed from the N-lobe to the C-lobe of the kinase. Hydrogen bonds are displayed by dashed lines. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. For clarity, the N-lobe is not shown.

Comparison of Experimental TrkA Binding Free Energies with ESMACS Predictions

The comparison of the TrkA binding free energies obtained from ESMACS predictions and those determined experimentally is shown in Figure 2. The predicted binding free energies from the 1-traj approach exhibit a moderate Pearson correlation with the experimental data (with a coefficient of 0.42; Figure 2a). The 2-traj approach, in which the same free energy of the receptor averaged from the 1-traj method is used, significantly improves the correlation between the predictions and experimental data (with a coefficient of 0.76; Figure 2b). Incorporating the free energies of the ligands from their individual simulations in water (the 3-traj approach) generates a correlation similar to that from the 2-traj approach (with a coefficient of 0.79; Figure 2c). The improvements are achieved by including the adaptation energies (8) of the receptor in the current study, and possibly of the ligands in general cases, which quantify the free energy changes of a molecule between its bound and free states. The 3-traj ESMACS results are in good agreement with the experimental measurements. It should be noted that the ESMACS method is capable of comparing the binding affinities of the entire set of ligands, notwithstanding the relatively significant structural differences. It should also be noted that the experimental TrkA binding free energies were approximated from half-maximal inhibitory concentration (IC₅₀) values, which provide only semiquantitative estimates. In addition, experimental IC₅₀ values for individual compounds vary between intralab measurements, and their average IC₅₀ values vary between 1.6- and 4.2-fold when measured independently across the two laboratories (Table 1 and Table S1 in the Supporting Information). A root-mean-square (RMS) error of ∼24% has recently been inferred from interlaboratory variations of reported binding affinities by Chodera and Mobley. (25)

Figure 2. Comparisons of the TrkA experimental data and the calculated binding free energies from (a) 1-traj, (b) 2-traj, and (c) 3-traj ESMACS approaches. The equation in each panel indicates the calculations used in each case. The term G_com/rec/lig^com/lig represents the free energy of the component (subscript: complex, receptor, or ligand) obtained from simulation of the same or different component (superscript: complex or free ligand in aqueous solution). The calculated binding free energies are associated with standard errors of ca. 0.6 kcal/mol in the 1-traj approach and ca. 3 kcal/mol in the 2- and 3-traj approaches (see Table S2), which are not shown in the figures for reasons of clarity. The experimental data from two sites (Pfizer, Sandwich and TCG Lifescience) are displayed in black and red, respectively. The correlation coefficients shown in the figure were calculated using the averages of the calculated binding free energies and the experimental data from Pfizer, Sandwich (black circles) and TCG Lifescience (red circles) where the former are not available. Large uncertainties are associated with the correlation coefficients because of the large error bars of the calculated and experimental binding affinities. Further analyses with bootstrapping resampling (see details in the Supporting Information) generate correlation coefficients of 0.39 ± 0.26, 0.60 ± 0.26, and 0.62 ± 0.27 for the 1-, 2-, and 3-traj approaches, respectively. It is evident that the 2- and 3-traj methods improve the ranking provided by the 1-traj method.

Improvement of Docking Predictions by the ESMACS Approach

For the nine compounds that were successfully docked into the TrkA protein, the binding free energies were also calculated using the structures after docking and energy minimization. The minimization process optimizes the geometry of a collection of atoms so that the net interatomic force on each atom is acceptably close to zero. The remaining seven compounds were not included in these calculations because the manually constructed structures exhibit some close contacts between the protein and the compounds, to which the binding free energy estimations are very sensitive. When the compounds could be docked into a single protein structure, the ranking of their binding affinities was reasonably predicted from the docked structures (Figure 3a). However, the results from the docking structures overestimate the free energy differences among the compounds, as shown in Figure 3a, in which the large slope of the regression line (3.33) indicates that the predicted energy differences are on average more than 3 times larger than the experimentally measured ones. The energy minimization significantly decreases the overestimations (Figure 3b, with a regression slope of 1.81). The ESMACS study further decreases the overestimation and makes the points on the scatter plot more concentrated around the regression line (Figure 3c). The docking method ranks the binding affinities reasonably well only for the compounds with closely related structures. Its results are also strongly dependent on the quality and choice of the structure(s) of the target protein. (26) The ESMACS approach, however, is not sensitive to the initial structure and achieves a similar correlation coefficient between the calculated and experimental binding affinities for all of the compounds irrespective of whether they can be successfully docked (Figure 2c).

Figure 3. Calculated binding free energies for 1, 3, 4, 6, 7, 8, 13, 16, and 22 (a) from single structures after docking, (b) from 25 structures for each compound after 11 000-step minimization with a sophisticated conjugate-gradient method, and (c) from ensemble averages from three-trajectory ESMACS studies. Compounds 1, 3, 4, 6, 7, 8, 13, 16, and 22 are highly structurally similar (see Table 1).

ESMACS Binding Affinities: Contributions from Energy Components

The components of the free energy calculations (see the Supporting Information for more details) may provide insight into the mechanism of compound binding. (27) In this study, no apparent correlations could be found for the individual energy components of the internal van der Waals, electrostatic, or electrostatic solvation energies. The total bonded energies, including bond, angle, and dihedral interactions, manifest a moderate correlation with the experimental binding free energies in the 3-traj approach, with a Pearson coefficient of 0.54 (Figure 4a). The nonbonded interactions, including the internal electrostatic, solvation electrostatic, and van der Waals interactions, have a slightly weaker correlation (Pearson coefficient of 0.43) than the bonded energies. The combination of bonded and nonbonded interactions significantly improves the correlation, with a Pearson coefficient of 0.75 (Figure 4c). The bonded and nonbonded energies are the two components of the total adaptation energy (see below), which is associated with the conformational changes upon compound binding. It is therefore not surprising that both of them show correlations with the binding affinities. The configurational entropy components do not exhibit a significant correlation with the experimental binding affinity (Figure S2). They vary in a range of 4.6 kcal/mol; the MMPBSA energies vary in a range of 14.41 kcal/mol by comparison. Inclusion of the contribution from configurational entropy into the MMPBSA energy does not improve the Pearson correlation coefficients significantly (see Figure 2c, which includes configurational entropy, and Figure 4c, which does not).

Figure 4. Correlations of free energy components and the experimental data from the 3-traj approach. Both (a) bonded and (b) nonbonded energy terms contribute to the ranking of binding affinities, with similar correlation coefficients between the calculations and experimental data. Their combination, the MMPBSA energy (c), exhibits better correlations with the experimental data than the components themselves.

Adaptation Energy: A Measure of Conformational Change upon Binding

The improvement obtained with the 3-traj approach compared with the 1-traj approach is achieved by relaxing the assumption that the receptor and the compounds sample similar conformations in both the free and bound states. Unfavorable adaptation energies (8) are usually induced when the conformations within the free state are significantly perturbed upon compound binding. The adaptation energies from the 3-traj approach indeed provide more insights into the mechanism of compound binding. They are related to the structural modifications made to the compounds and the protein and shed important light on the optimization. As the ranking of the ligand binding is the main concern of the study and the relative adaptation energies of the protein are energetically as informative as the absolute ones in the binding affinity comparison, no attempt has been made here to compute accurate absolute adaptation energies for the protein. All of the adaptation energies for the protein are relative ones in the current paper. Non-negligible adaptation energies are associated with conformational changes upon compound binding, meaning that the binding involves “induced fit” recognition. Compound 8, for example, introduces the largest adaptation energy within the protein, while 16 has the largest adaptation energy within the compound (Figure 5). The binding of 8 induces a conformational change within the protein because there is a reluctance to accommodate the methoxy group in the linker region of 8 into the binding site; the arrangement of the amide group in 16 forces the compound to adopt a high-energy conformation in the protein’s binding site (see details below).

Figure 5. Adaptation free energies of (a) the receptor and (b) the compounds showing the binding free energy changes between the 1- and 3-traj approaches. The terms G_rec/lig^com/lig are the free energies of receptor or ligands (subscript) calculated from simulations performed for the complex or the ligands (superscript).

Comparison of Experimental TrkA Binding Free Energies with TIES Predictions

The TIES method provides a more accurate approach to estimate relative binding affinities, but the scope for its use is rather tightly circumscribed, as it is applicable only for pairs of compounds that do not present significant structural differences. Our previous studies have demonstrated that TIES offers more quantitative accuracy in its predictions than ESMACS. (13, 28) Here we apply TIES to the TrkA data set to highlight the accuracy of the approach. In TIES, studies are performed for 14 pairs of TrkA compounds (Figure 6). The calculated binding free energy differences correlate well with the experimental data, with a Pearson correlation of 0.88 (Figure 6), compared with 0.79 from the 3-traj ESMACS study (Figure 2c). A directional agreement is achieved if a prediction has the same sign as the experimental observation; otherwise, the result is deemed to be directional disagreement. For 10 out of the 14 pairs studied, TIES successfully achieves directional agreements, meaning that the results predict the direction of the change in the binding affinity correctly (Figure 6). As shown in the figure, all but one of the TIES predictions that directionally disagree lie on the border of quadrants II and III within their calculated error bars. When the error bars of the experimental ΔΔG are also taken into account, a better agreement might be achieved than that reported here.

Figure 6. Correlation between TIES-predicted relative binding affinities and experimental data. The black line is the correlation line, while the dotted lines (x = 0 and y = 0) create four quadrants. Ten out of the 14 data points are in quadrants I (x > 0 and y > 0) and III (x < 0 and y < 0), meaning that the calculated binding free energy differences have the same sign as those from the experimental data.

Discussion

ARTICLE SECTIONS

Jump To

The calculations, especially the ESMACS ones, can provide further insights for binding affinity variation between structurally similar ligands. In order to analyze the results, it is necessary to consider the differences between the one-, two-, and three-trajectory methods. The 1-traj simulation calculates ΔG_binding assuming there is no energetic penalty for the adaptation of the protein or ligand to the binding conformation; the 2-traj method takes the change in protein energy into consideration, and the 3-traj method accounts for the changes in both protein and ligand energies. The simulations enable the exploration of key differences in binding modes for structurally related ligands, as we now discuss.

Hinge Binding Group Modifications: Compounds 1, 4, and 22

Initial compounds in the pyrrolopyrimidine series, such as 1, contain an amino group at the 4-position of the hinge binding group (Table 2). Compounds such these, while TrkA-active, required optimization of their kinase selectivity profile. (29) It is known within the scientific literature that minimizing the number of hydrogen-bonding interactions made between a kinase inhibitor and the kinase hinge binding region can lead to an enhanced kinome selectivity profile. (30, 31) In order to design compounds with an optimal kinase selectivity profile, an analysis of the crystal structure of 1 bound to TrkA and the nonliganded TrkA apo structure was undertaken. (29) The aminopyrrolopyrimidine motif of 1 makes a two-point hydrogen-bonding interaction with hinge residues E590 and M592. When 1 is superpositioned onto the TrkA apo crystal structure, the −NH₂ motif of the aminopyrrolopyrimidine group overlays directly with a conserved water molecule observed in the apo structure. The −NH₂ group was therefore removed to generate compounds such as 4, in which a bridging water molecule forms hydrogen bonds with the backbone carbonyl oxygen at E590 and the ketone oxygen of the TrkA ligand. The hydrogen bond between the ligand and the backbone N–H of hinge residue M592 is maintained. The deletion of the −NH₂ group had only a minimal effect on the TrkA activity, as 1 and 4 had broadly similar IC₅₀ values in the TrkA pharmacology assay.

Table 2. Structures and Properties of 1, 4, and 22^a

^{Table a}

TrkA free energy activities derived from experimental IC₅₀ values are denoted as ΔG_exp. The relative binding free energies from experiment and ESMACS and TIES calculations are denoted as ΔΔG_exp, ΔΔG_ESMACS, and ΔΔG_TIES, respectively. Experimental data from Pfizer, Sandwich (U.K.) are shown in black, and those from TCG Lifescience (India) are shown in blue.

ESMACS calculations show that compounds without the −NH₂ group at the 4-position (all compounds except 1, 3, 6, and 22; Table 2) have a bridging water molecule between the ketone oxygen of the ligand and the backbone carbonyl oxygen of hinge residue E590, with an average occupancy of 41 ± 9%. The occupancy indeed underscores the presence of a water molecule at the location as the molecule dynamically moves into and out of the cutoff distance and angle that designate the occurrence of a bridging water molecule. Simulations of compounds containing the −NH₂ group at the 4-position (1, 3, and 6; Table 2) show the same bridging water to be present but at a far lower frequency (occupancy values of just 2 ± 0% for 1, 3, and 6) (Figure 7).

Figure 7. ESMACS simulations of (a) 1, (b) 22, and (c) 4 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 frames. Key residues E590 and M592 (for clarity, side-chain atoms are not shown) at the TrkA hinge region are highlighted. Hydrogen bonds are shown as dashed lines. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. For reasons of clarity, the N-lobe of the protein has been removed.

ESMACS confirms that the bridging water molecule highlighted in Figure 7 helps mediate and thus satisfy key protein–compound interactions, rendering the overall binding free energies of compounds such as 4 less affected by the absence of the H-bond when −NH₂ is not present. This capability is reflected in the close similarity between the calculated TrkA free energy values for 1 and 4 and those derived from TrkA pharmacology experiments (Table 2). Compound 22 was synthesized as part of an effort to characterize putative compound metabolites. As highlighted in Table 2, 22 is only weakly active in the TrkA pharmacology (IC₅₀ = 1756 nM). ESMACS simulations show a low average occupancy of the aforementioned bridging water molecule (average occupancy 10 ± 6%) due to the presence of the −OH group at the 4-position (Figure 7). The calculations also show that the −OH group of 22 is more likely to form an intramolecular H-bond with the carbonyl group of the ligand than to form an intermolecular H-bond with hinge residue E590 of the protein. Within the 25-member ensemble, only one ESMACS replica has an intermolecular H-bond forming with an occupancy of 84%, compared with <5% for all of the others. The −OH group occupies the space of the bridging water molecule and does not have a direct hydrogen-bonding interaction with E590. The intramolecular H-bond of the −OH group places the −OH oxygen in a position where an electrostatic repulsion arises with the backbone carbonyl oxygen of E590 (Figure 7b). The result of having a nonsatisfied hydrogen-bonding group (−OH) in 22 is a reduction in overall ligand binding affinity. This is reflected not only in the measured activity data (ΔG_exp = −7.1 kcal/mol) but also in the relative binding affinities calculated by ESMACS and TIES (Table 2).

Hinge Binding Group Modifications: Compounds 12, 17, 23, and 24

The TrkA cocrystal structure of compounds such as 12 indicated that a polar atom capable of making an additional hydrogen-bonding interaction with M592 might be accommodated at the 2-position of the hinge binding group (Table 3). Compound 17 was synthesized initially to test this theory and proved to be more active in the TrkA pharmacology model than the parent molecule 12. Compound 23 was then synthesized to see whether adding a methyl group to the amine would further boost the potency through the introduction of hydrophobic interactions with the lipophilic side chain of Y591. Compound 23 was potent in the TrkA assay, albeit slightly less so than 12. Compound 24 was synthesized as a putative metabolite of 12 and was found to be >15-fold less active at TrkA.

Table 3. Structures and Properties of 12, 17, 23, and 24^a

^{Table a}

ESMACS shows that addition of polar groups at the 2-position increases the total occupancy of hydrogen bonds with M592 from 23 ± 4% (for 12) to 85 ± 4%, 80 ± 4%, and 91 ± 19% (for 17, 23, and 24, respectively). This includes the occupancy of the hydrogen bond made between the pyrimidyl nitrogen and the −NH group of M592, which is common to all of the compounds. The increased occupancies of 17, 23, and 24 result in the additional hydrogen bonds between the polar groups at the 2-position of the compounds and the C═O group of M592. The simulations also show that addition of an −NH₂ group (17) does not introduce any unfavorable steric interaction within the protein or the ligand (see Figure 8). The slightly improved binding affinity of 17 versus 12 stems from the favorable electrostatic interaction (including the solvation electrostatic component from the Poisson–Boltzmann calculation) between the compound and the protein. The addition of the methyl group on the amine (23), however, induces an unfavorable adaptation energy within the protein (see the comparison of relative adaptation energies for 12/17 and 23 in Figure 5a) due to steric hindrance between the protein and 23 (Figure 8c). The presence of an −NH₂ group at the 2-position (17) does not induce an unfavorable adaptation energy. The introduction of an −OH group (24), however, introduces a significant adaptation energy in the protein because the −OH group forms the strongest hydrogen bond with M592 among the subgroup of compounds 12, 17, 23, and 24, which induces steric hindrance within the complex and reduces the overall binding affinity (Figure 8).

Figure 8. ESMACS simulations of (a) 12, (b) 17, (c) 23, and (d) 24 bound to TrkA. Representative conformations are displayed using the final conformations of 4 ns production runs of one replica from each 25-member ensemble. The protein is shown by the surface representation and the ligand by the ball-and-stick representation with hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. The surfaces of the R2-position groups are shown in the wireframe representation.

Linker Group Modifications: Compounds 1, 4, and 22

The TrkA cocrystal structure of compounds such as 1 highlighted a conserved water molecule that formed hydrogen-bonding interactions with the N–H group of the amide linker and protein residues K544 and E560 (Table 4 and Figure 9). A methyl group was added to the amide nitrogen to assess the effect of displacing the conserved water molecule (and the potentially altered amide conformation relative to the pyridyl ring) on the TrkA activity. The effect of transposing the amide N–H to the other side of the carbonyl group was also investigated, as this, if the activity were retained, could open up new parallel chemistry opportunities within the chemotype. Compound 3, an example of the N-methylamide, remained active at TrkA, although with reduced affinity versus its desmethyl congener 1. Compound 6, an example of the transposed N–H amide, also remained active at TrkA with almost identical affinity as 1.

Table 4. Structures and Properties of 1, 3, and 6^a

^{Table a}

Figure 9. ESMACS simulations of (a) 1, (b) 3, and (c) 6 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 frames. The key TrkA protein residues D668, K544, and E560 are highlighted. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. Hydrogen bonds between the linker group carbonyl oxygen of 1, 3, and 6 and the backbone −NH group of D668 are shown as dashed lines.

ESMACS shows that the modifications highlighted in Table 4 do not alter the key hydrogen-bonding interaction between the carbonyl oxygen of the amide group of 1, 3, and 6 and the backbone N–H of D668, with occupancies of 56 ± 7%, 54 ± 3%, and 51 ± 6%, respectively (Figure 9). The altered position of the amide group in 1 and 6 slightly affects the appearance of the bridging water molecules between the ligands and residue E560, with occupancies of 15 ± 6% and 20 ± 5% for 1 and 6, respectively. Minimal direct hydrogen-bonding interactions are predicted between the amide −NH group of 1 and 6 and residue K544 or E560. The replacement of the amide −NH group of 1 by the more hydrophobic N-methyl group is predicted to displace the adjacent bridging water molecule (Figure 9a,b) and overall make binding to TrkA energetically less favorable. There is a good correlation between the ESMACS-predicted and experimentally derived free energies of TrkA binding for 1, 3, and 6 (Table 4).

Linker Group Modifications: Compounds 7 and 8

Although the linker group of the pyrrolopyrimidine/pyrrolopyridine series binds to a relatively narrow region of the ligand binding site, between the two adjacent ATP and DFG pockets, the tolerance of the TrkA protein toward bulkier linker groups was briefly investigated as part of the TrkA program. An example from this work is 8, in which a methoxy group has been added to the linker-group carbonyl. As highlighted in Table 5, a significant reduction in TrkA activity is observed for 8 versus 7.

Table 5. Structures and Properties of 7 and 8^a

^{Table a}

TrkA free energy activities derived from experimental IC₅₀ values are denoted as ΔG_exp. The relative binding free energies from experiment and ESMACS and TIES calculations are denoted as ΔΔG_exp, ΔΔG_ESMACS, and ΔΔG_TIES, respectively. Experimental data were obtained from Pfizer, Sandwich (U.K.) for these two compounds.

ESMACS calculations indicate that the oxygen atom of the methoxy group of 8 is located proximal (3.47 ± 0.31 Å) to the carbonyl group of residue D668. The occupancy of a bridging water molecule between the methoxy group and the carbonyl is only 5 ± 2%, indicating that the orientation and the space between these groups preclude a bridging water molecule being available to quench any electrostatic repulsion (Table 5 and Figure 10). The chirality of the methoxy group also prevents the polar oxygen from having any direct or favorable water-bridged interactions with E560 or D668. The binding of 8 induces the largest adaptation energy (Figure 5a) in the TrkA protein of all compounds assessed. Energetic analyses show that although the bonded energy and the van der Waals interactions are favorable for 8 compared with 7 (by 3.70 and 2.29 kcal/mol, respectively), a significantly unfavorable electrostatic energy (10.59 kcal/mol), mainly from the interactions between the polar oxygen of the methoxy group and the polar/charged groups of the protein (carbonyl group of D668 and carboxylate group of E560; Figure 10), is introduced by the inclusion of the (S)-OMe group.

Figure 10. ESMACS simulations of (a) 7 and (b) 8 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 frames. The key TrkA protein residues D668, E560, and L564 are highlighted. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. The unfavorable electrostatic interactions between the methoxy group of 8 and the carbonyl group of D668 and between the methoxy group and the carboxylate group of E560 are highlighted by pink arrows.

The calculated free energies of TrkA binding for 7 and 8 correlate well with those obtained experimentally. Hence, utilization of ESMACS to predict the TrkA binding affinity of 8 during the Pfizer TrkA project might well have influenced the medicinal chemistry team to deprioritize this compound and redirect synthetic efforts in other directions.

Linker Group Modifications: Compounds 4 and 16

The hydrogen bond between the linker-group carbonyl and the backbone N–H of D668 is a key interaction across the pyrrolopyrimidine/pyrrolopyridine series. During the Pfizer TrkA program, reversing the amide group was modeled to assess whether the usual ligand binding mode could be adapted to maintain this interaction in the new amide arrangement. Docking of 16 into TrkA using a Pfizer docking protocol (data not shown) suggested that the reverse amide would likely not be able to adapt a conformation in which this interaction would be retained. However, 16 was synthesized as an example of a reverse amide system to challenge and/or verify the docking result. As can be seen in Table 6, reversing the amide group led to a significant reduction in TrkA activity.

Table 6. Structures and Properties of 4 and 16^a

^{Table a}

TrkA free energy activities derived from experimental IC₅₀ values are denoted as ΔG_exp. The relative binding free energies from experiment and ESMACS and TIES calculations are denoted as ΔΔG_exp, ΔΔG_ESMACS, and ΔΔG_TIES, respectively. Experimental data were obtained from Pfizer, Sandwich (U.K.) for these two compounds.

ESMACS-based component binding analysis shows that 16 has the least favorable bonding and nonbonding interactions of the compounds in Table 1. The reversed positions of the −NH and C═O groups prevent formation of the hydrogen bond between the ligand and residue D668 (Figure 11), with an H-bond occupancy of <1% compared with occupancies of 30–56% for the rest of the compound set (except for 26, whose occupancy is 14% as a result of the competition from an extra hydrogen bond between the polar oxygen at R4 and residue K544) (Table 1). Indeed, ESMACS calculations reveal that the −NH and C═O groups of 16 align in a noncomplementary fashion with the −NH and C═O groups of residue D668, introducing a large unfavorable electrostatic interaction (Figure 11). Compound 16 has the largest ligand adaptation energy, indicating that the switching of N–H and C═O does not enable an alternate binding mode to be adopted, and therefore, 16 must adopt a high-energy conformation to complex with the protein. As is clear in Table 6, ESMACS correctly predicts the reduced TrkA potency of 16 compared with 4.

Figure 11. ESMACS simulations of (a) 4 and (b) 16 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 conformations. The key TrkA protein residue D668 is highlighted. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. The hydrogen bond between the linker-group carbonyl oxygen of 4 and the backbone −NH group of D668 is shown as a dashed line. As can be seen in (b), switching of the N–H and C═O groups prevents the formation of a hydrogen bond between the carbonyl oxygen of 16 and the backbone −NH group of D668.

Conclusions

ARTICLE SECTIONS

Jump To

Methodologies that can predict the binding affinities of molecules ahead of synthesis represent a key area of interest in the pharmaceutical industry. The ability to effectively prioritize prospective compounds on the basis of their likely activity in a key pharmacology assay offers the potential to rapidly improve lead optimization timelines in drug discovery programs by reducing the number of “design–synthesis–screen” cycles needed to identify candidate-quality molecules suitable for clinical testing. (32)

Despite the relative diversity of the compounds in this study, the 3-traj version of the ESMACS approach provides good agreement between the theoretically predicted binding free energies and those derived from experimentally measured activities for the entire data set. TIES generates even better agreement between the calculated and experimental binding free energy differences for a selected subgroup of ligands. This is manifest in the better correlation coefficient obtained in TIES (0.88) than in ESMACS (0.79) and accurate binding free energy differences from the former. This is the case for all of molecular systems we have studied. (13, 28, 33) ESMACS also provides structural and energetic insight into the binding of individual ligand–protein systems: some compounds bind by a “lock and key” recognition mechanism, for which no significant conformational adjustment is required by the protein and the ligands; others bind according to an “induced fit” recognition mechanism involving significant conformational changes associated with non-negligible adaptation energies.

The Binding Affinity Calculator (BAC) used in this study automates the workflows for ESMACS and TIES calculations, making the use of these approaches easy and user-friendly. With powerful computing resources now widely available, these robust approaches should become more routine for industrial groups in the field of structure-based drug design. The results described herein, based on an analysis of TrkA ligands synthesized as part of the TrkA program at Pfizer, suggest that the binding affinity calculations have the potential to be successfully applied in real-time prospective ligand design.

Supporting Information

ARTICLE SECTIONS

Jump To

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jcim.6b00780.

Detailed description of the energy decomposition, the energy convergence, and the error analyses, experimental measurements of TrkA inhibitory activity, and the predicted binding free energies (PDF)

ci6b00780_si_001.pdf (300.28 kb)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

ARTICLE SECTIONS

Jump To

Corresponding Author
- Peter V. Coveney - Centre for Computational Science, Department of Chemistry, University College London, London WC1H 0AJ, United Kingdom; http://orcid.org/0000-0002-8787-7256; Email: [email protected]
Authors
- Shunzhou Wan - Centre for Computational Science, Department of Chemistry, University College London, London WC1H 0AJ, United Kingdom
- Agastya P. Bhati - Centre for Computational Science, Department of Chemistry, University College London, London WC1H 0AJ, United Kingdom
- Sarah Skerratt - Pfizer Global R&D, The Portway Building, Granta Park, Cambridge CB21 6GS, United Kingdom
- Kiyoyuki Omoto - Pfizer Worldwide Research and Development, 610 Main Street, Cambridge, Massachusetts 02139, United States
- Veerabahu Shanmugasundaram - Pfizer Worldwide Research and Development, Groton Laboratories, Groton, Connecticut 06340, United States
- Sharan K. Bagal - Pfizer Global R&D, The Portway Building, Granta Park, Cambridge CB21 6GS, United Kingdom
Notes
The authors declare no competing financial interest.

Acknowledgment

ARTICLE SECTIONS

Jump To

The authors acknowledge the support from the EPSRC via the 2020 Science Programme (http://www.2020science.net/, EP/I017909/1), the EU H2020 Projects ComPat (http://www.compat-project.eu/, 671564) and CompBioMed (http://www.compbiomed.eu/, 675451), the Qatar National Research Fund (7-1083-1-191), the MRC Medical Bioinformatics Project (MR/L016311/1), and the UCL Provost. We are grateful to the Hartree Centre for access to the BlueWonder2 computer and for the help of its scientific support staff. We acknowledge the Leibniz Supercomputing Centre for providing access to SuperMUC and the assistance of its scientific support staff. We also made use of ARCHER, the U.K.’s National High Performance Computing Service, funded by the Office of Science and Technology through the EPSRC’s High-End Computing Programme. Access to ARCHER was provided through the 2020 Science Programme. This research was also partially supported by the PLGrid Infrastructure, through which access to Prometheus, the Polish supercomputer run by ACK Cyfronet AGH in Krakow, was provided. A.P.B. is supported by an Overseas Research Scholarship from UCL and an Inlaks Scholarship from the Inlaks Shivdasani Foundation.

References

ARTICLE SECTIONS

Jump To

This article references 33 other publications.

1
Sherborne, B.; Shanmugasundaram, V.; Cheng, A. C.; Christ, C. D.; DesJarlais, R. L.; Duca, J. S.; Lewis, R. A.; Loughney, D. A.; Manas, E. S.; McGaughey, G. B.; Peishoff, C. E.; van Vlijmen, H. Collaborating to Improve the Use of Free-Energy and Other Quantitative Methods in Drug Discovery J. Comput.-Aided Mol. Des. 2016, 30, 1139– 1141 DOI: 10.1007/s10822-016-9996-y
Google Scholar
1
Collaborating to improve the use of free-energy and other quantitative methods in drug discovery
Sherborne, Bradley; Shanmugasundaram, Veerabahu; Cheng, Alan C.; Christ, Clara D.; DesJarlais, Renee L.; Duca, Jose S.; Lewis, Richard A.; Loughney, Deborah A.; Manas, Eric S.; McGaughey, Georgia B.; Peishoff, Catherine E.; van Vlijmen, Herman
Journal of Computer-Aided Molecular Design (2016), 30 (12), 1139-1141CODEN: JCADEQ; ISSN:0920-654X. (Springer)
In May and August, 2016, several pharmaceutical companies convened to discuss and compare experiences with Free Energy Perturbation (FEP). This unusual synchronization of interest was prompted by Schrodinger's FEP+ implementation and offered the opportunity to share fresh studies with FEP and enable broader discussions on the topic. This article summarizes key conclusions of the meetings, including a path forward of actions for this group to aid the accelerated evaluation, application and development of free energy and related quant., structure-based design methods.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XitFClsLfN&md5=c00386f775bd826c6b845bbbef4a975d
2
Skerratt, S. E.; Andrews, M.; Bagal, S. K.; Bilsland, J.; Brown, D.; Bungay, P. J.; Cole, S.; Gibson, K. R.; Jones, R.; Morao, I.; Nedderman, A.; Omoto, K.; Robinson, C.; Ryckmans, T.; Skinner, K.; Stupple, P.; Waldron, G. The Discovery of a Potent, Selective, and Peripherally Restricted Pan-Trk Inhibitor (PF-06273340) for the Treatment of Pain J. Med. Chem. 2016, 59, 10084– 10099 DOI: 10.1021/acs.jmedchem.6b00850
Google Scholar
2
The Discovery of a Potent, Selective, and Peripherally Restricted Pan-Trk Inhibitor (PF-06273340) for the Treatment of Pain
Skerratt, Sarah E.; Andrews, Mark; Bagal, Sharan K.; Bilsland, James; Brown, David; Bungay, Peter J.; Cole, Susan; Gibson, Karl R.; Jones, Russell; Morao, Inaki; Nedderman, Angus; Omoto, Kiyoyuki; Robinson, Colin; Ryckmans, Thomas; Skinner, Kimberly; Stupple, Paul; Waldron, Gareth
Journal of Medicinal Chemistry (2016), 59 (22), 10084-10099CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
The neurotrophin family of growth factors, comprised of nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin 3 (NT3) and neurotrophin 4 (NT4), is implicated in the physiol. of chronic pain. Given the clin. efficacy of anti-NGF monoclonal antibody (mAb) therapies, there is significant interest in the development of small mol. modulators of neurotrophin activity. Neurotrophins signal through the tropomyosin related kinase (Trk) family of tyrosine kinase receptors, hence Trk kinase inhibition represents a potentially "druggable" point of intervention. To deliver the safety profile required for chronic, non-life threatening pain indications, highly kinase-selective Trk inhibitors with minimal brain availability are sought. Herein the authors describe how the use of SBDD, 2D QSAR models and Matched Mol. Pair data in compd. design enabled the delivery of the highly potent, kinase-selective and peripherally restricted clin. candidate PF-06273340.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhslWjsbfL&md5=f8c904a92c1716062c217fefe2374ce6
3
Sadiq, S. K.; Wright, D.; Watson, S. J.; Zasada, S. J.; Stoica, I.; Coveney, P. V. Automated Molecular Simulation Based Binding Affinity Calculator for Ligand-Bound HIV-1 Proteases J. Chem. Inf. Model. 2008, 48, 1909– 1919 DOI: 10.1021/ci8000937
Google Scholar
3
Automated Molecular Simulation Based Binding Affinity Calculator for Ligand-Bound HIV-1 Proteases
Sadiq, S. Kashif; Wright, David; Watson, Simon J.; Zasada, Stefan J.; Stoica, Ileana; Coveney, Peter V.
Journal of Chemical Information and Modeling (2008), 48 (9), 1909-1919CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
The successful application of high throughput mol. simulations to det. biochem. properties would be of great importance to the biomedical community if such simulations could be turned around in a clin. relevant timescale. An important example is the detn. of antiretroviral inhibitor efficacy against varying strains of HIV through calcn. of drug-protein binding affinities. We describe the Binding Affinity Calculator (BAC), a tool for the automated calcn. of HIV-1 protease-ligand binding affinities. The tool employs fully atomistic mol. simulations alongside the well established mol. mechanics Poisson-Boltzmann solvent accessible surface area (MMPBSA) free energy methodol. to enable the calcn. of the binding free energy of several ligand-protease complexes, including all nine FDA approved inhibitors of HIV-1 protease and seven of the natural substrates cleaved by the protease. This enables the efficacy of these inhibitors to be ranked across several mutant strains of the protease relative to the wildtype. BAC is a tool that utilizes the power provided by a computational grid to automate all of the stages required to compute free energies of binding: model prepn., equilibration, simulation, postprocessing, and data-marshaling around the generally widely distributed compute resources utilized. Such automation enables the mol. dynamics methodol. to be used in a high throughput manner not achievable by manual methods. This paper describes the architecture and workflow management of BAC and the function of each of its components. Given adequate compute resources, BAC can yield quant. information regarding drug resistance at the mol. level within 96 h. Such a timescale is of direct clin. relevance and can assist in decision support for the assessment of patient-specific optimal drug treatment and the subsequent response to therapy for any given genotype.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXhtVShtLjN&md5=c616ac9330dc1f4cb9808d68fabc7e4c
4
Groen, D.; Bhati, A. P.; Suter, J.; Hetherington, J.; Zasada, S. J.; Coveney, P. V. FabSim: Facilitating Computational Research through Automation on Large-Scale and Distributed E-Infrastructures Comput. Phys. Commun. 2016, 207, 375– 385 DOI: 10.1016/j.cpc.2016.05.020
Google Scholar
4
FabSim: Facilitating computational research through automation on large-scale and distributed e-infrastructures
Groen, Derek; Bhati, Agastya P.; Suter, James; Hetherington, James; Zasada, Stefan J.; Coveney, Peter V.
Computer Physics Communications (2016), 207 (), 375-385CODEN: CPHCBZ; ISSN:0010-4655. (Elsevier B.V.)
We present FabSim, a toolkit developed to simplify a range of computational tasks for researchers in diverse disciplines. FabSim is flexible, adaptable, and allows users to perform a wide range of tasks with ease. It also provides a systematic way to automate the use of resources, including HPC and distributed machines, and to make tasks easier to repeat by recording contextual information. To demonstrate this, we present three use cases where FabSim has enhanced our research productivity. These include simulating cerebrovascular bloodflow, modeling clay-polymer nanocomposites across multiple scales, and calcg. ligand-protein binding affinities.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28Xptl2nu7s%253D&md5=4b7e50658ec8307f7fa5fcd16747cab6
5
Norman, B. H.; McDermott, J. S. Targeting the Nerve Growth Factor (NGF) Pathway in Drug Discovery. Potential Applications to New Therapies for Chronic Pain J. Med. Chem. 2017, 60, 66– 88 DOI: 10.1021/acs.jmedchem.6b00964
Google Scholar
5
Targeting the Nerve Growth Factor (NGF) Pathway in Drug Discovery. Potential Applications to New Therapies for Chronic Pain
Norman, Bryan H.; McDermott, Jeff S.
Journal of Medicinal Chemistry (2017), 60 (1), 66-88CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
A review. The neurotrophin nerve growth factor (NGF) has been implicated as a key mediator of chronic pain. NGF binds the tropomyosin receptor kinase A (TrkA) and p75, resulting in the activation of downstream signaling pathways that have been linked to pro-nociception. While anti-NGF antibodies have demonstrated analgesia both preclinically and in patients, the mechanism of action of these agents remains unclear. The authors describe ligands targeting NGF, its receptors and downstream/related targets. This perspective highlights large and small mol. approaches to targeting the NGF-TrkA pathway both extra- and intracellularly. In addn., the authors present a strategic framework for future drug discovery efforts in this pathway beyond the targeting of NGF or its receptors. While existing tools have greatly informed NGF-mediated signaling, ongoing and future pathway research may help focus new drug discovery efforts on key novel targets and mechanisms. This may result in highly differentiated therapeutics with greater efficacy and/or improved safety profiles.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhslSqtLjF&md5=606097686779a102db9e6e54f8252a40
6
Hefti, F. F.; Rosenthal, A.; Walicke, P. A.; Wyatt, S.; Vergara, G.; Shelton, D. L.; Davies, A. M. Novel Class of Pain Drugs Based on Antagonism of NGF Trends Pharmacol. Sci. 2006, 27, 85– 91 DOI: 10.1016/j.tips.2005.12.001
Google Scholar
6
Novel class of pain drugs based on antagonism of NGF
Hefti, Franz F.; Rosenthal, Arnon; Walicke, Patricia A.; Wyatt, Sean; Vergara, German; Shelton, David L.; Davies, Alun M.
Trends in Pharmacological Sciences (2006), 27 (2), 85-91CODEN: TPHSDY; ISSN:0165-6147. (Elsevier Ltd.)
A review. Nerve growth factor (NGF) was identified originally as a survival factor for sensory and sympathetic neurons in the developing nervous system. In adults, NGF is not required for survival but it has a crucial role in the generation of pain and hyperalgesia in several acute and chronic pain states. The expression of NGF is high in injured and inflamed tissues, and activation of the NGF receptor tyrosine kinase trkA on nociceptive neurons triggers and potentiates pain signaling by multiple mechanisms. Inhibition of NGF function and signaling blocks pain sensation as effectively as cyclooxygenase inhibitors and opiates in rodent models of pain. Several pharmaceutical companies have active drug-discovery and development programs that are based on a variety of approaches to antagonize NGF, including NGF capture', blocking the binding of NGF to trkA and inhibiting trkA signaling. NGF antagonism is expected to be a highly effective therapeutic approach in many pain states, and to be free of the adverse effects of traditional analgesic drugs.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD28XhtV2lsLo%253D&md5=3cfe9e06069c3b6b9aa9d96b7411d199
7
Mantyh, P. W.; Koltzenburg, M.; Mendell, L. M.; Tive, L.; Shelton, D. L. Antagonism of Nerve Growth Factor-TrkA Signaling and the Relief of Pain Anesthesiology 2011, 115, 189– 204 DOI: 10.1097/ALN.0b013e31821b1ac5
Google Scholar
There is no corresponding record for this reference.
8
Wan, S.; Knapp, B.; Wright, D. W.; Deane, C. M.; Coveney, P. V. Rapid, Precise, and Reproducible Prediction of Peptide-MHC Binding Affinities from Molecular Dynamics That Correlate Well with Experiment J. Chem. Theory Comput. 2015, 11, 3346– 3356 DOI: 10.1021/acs.jctc.5b00179
Google Scholar
8
Rapid, precise, and reproducible prediction of peptide-MHC binding affinities from molecular dynamics that correlate well with experiment
Wan, Shunzhou; Knapp, Bernhard; Wright, David W.; Deane, Charlotte M.; Coveney, Peter V.
Journal of Chemical Theory and Computation (2015), 11 (7), 3346-3356CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
The presentation of potentially pathogenic peptides by major histocompatibility complex (MHC) mols. is one of the most important processes in adaptive immune defense. Prediction of peptide-MHC (pMHC) binding affinities is therefore a principal objective of theor. immunol. Machine learning techniques achieve good results if substantial exptl. training data are available. Approaches based on structural information become necessary if sufficiently similar training data are unavailable for a specific MHC allele, although they have often been deemed to lack accuracy. In this study, we use a free energy method to rank the binding affinities of 12 diverse peptides bound by a class I MHC mol. HLA-A*02:01. The method is based on enhanced sampling of mol. dynamics calcns. in combination with a continuum solvent approxn. and includes ests. of the configurational entropy based on either a one or a three trajectory protocol. It produces precise and reproducible free energy ests. which correlate well with exptl. measurements. If the results are combined with an amino acid hydrophobicity scale, then an extremely good ranking of peptide binding affinities emerges. Our approach is rapid, robust, and applicable to a wide range of ligand-receptor interactions without further adjustment.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhtVaksLjN&md5=860aa35a6e8e4013e2b0c7f0ae99170f
9
Kollman, P. A.; Massova, I.; Reyes, C.; Kuhn, B.; Huo, S.; Chong, L.; Lee, M.; Lee, T.; Duan, Y.; Wang, W.; Donini, O.; Cieplak, P.; Srinivasan, J.; Case, D. A.; Cheatham, T. E., III Calculating Structures and Free Energies of Complex Molecules: Combining Molecular Mechanics and Continuum Models Acc. Chem. Res. 2000, 33, 889– 897 DOI: 10.1021/ar000033j
Google Scholar
9
Calculating Structures and Free Energies of Complex Molecules: Combining Molecular Mechanics and Continuum Models
Kollman, Peter A.; Massova, Irina; Reyes, Carolina; Kuhn, Bernd; Huo, Shuanghong; Chong, Lillian; Lee, Matthew; Lee, Taisung; Duan, Yong; Wang, Wei; Donini, Oreola; Cieplak, Piotr; Srinivasan, Jaysharee; Case, David A.; Cheatham, Thomas E., III
Accounts of Chemical Research (2000), 33 (12), 889-897CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
A review, with 63 refs. A historical perspective on the application of mol. dynamics (MD) to biol. macromols. is presented. Recent developments combining state-of-the-art force fields with continuum solvation calcns. have allowed us to reach the fourth era of MD applications in which one can often derive both accurate structure and accurate relative free energies from mol. dynamics trajectories. We illustrate such applications on nucleic acid duplexes, RNA hairpins, protein folding trajectories, and protein-ligand, protein-protein, and protein-nucleic acid interactions.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3cXmvFGiu7g%253D&md5=8436ee610ae145894428db1a1deff73c
10
Coveney, P. V.; Wan, S. On the Calculation of Equilibrium Thermodynamic Properties from Molecular Dynamics Phys. Chem. Chem. Phys. 2016, 18, 30236– 30240 DOI: 10.1039/C6CP02349E
Google Scholar
10
On the calculation of equilibrium thermodynamic properties from molecular dynamics
Coveney, Peter V.; Wan, Shunzhou
Physical Chemistry Chemical Physics (2016), 18 (44), 30236-30240CODEN: PPCPFQ; ISSN:1463-9076. (Royal Society of Chemistry)
The purpose of statistical mechanics is to provide a route to the calcn. of macroscopic properties of matter from their constituent microscopic components. It is well known that the macrostates emerge as ensemble avs. of microstates. However, this is more often stated than implemented in computer simulation studies. Here we consider foundational aspects of statistical mechanics which are overlooked in most textbooks and research articles that purport to compute macroscopic behavior from microscopic descriptions based on classical mechanics and show how due attention to these issues leads in directions which have not been widely appreciated in the field of mol. dynamics simulation.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XntVyisr8%253D&md5=f7950af8255cf494a117d46bc1b5191e
11
Miller, B. R., 3rd; McGee, T. D., Jr.; Swails, J. M.; Homeyer, N.; Gohlke, H.; Roitberg, A. E. Mmpbsa.Py: An Efficient Program for End-State Free Energy Calculations J. Chem. Theory Comput. 2012, 8, 3314– 3321 DOI: 10.1021/ct300418h
Google Scholar
11
MMPBSA.py: An Efficient Program for End-State Free Energy Calculations
Miller, Bill R., III; McGee, T. Dwight, Jr.; Swails, Jason M.; Homeyer, Nadine; Gohlke, Holger; Roitberg, Adrian E.
Journal of Chemical Theory and Computation (2012), 8 (9), 3314-3321CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
MM-PBSA is a post-processing end-state method to calc. free energies of mols. in soln. MMPBSA.py is a program written in Python for streamlining end-state free energy calcns. using ensembles derived from mol. dynamics (MD) or Monte Carlo (MC) simulations. Several implicit solvation models are available with MMPBSA.py, including the Poisson-Boltzmann Model, the Generalized Born Model, and the Ref. Interaction Site Model. Vibrational frequencies may be calcd. using normal mode or quasi-harmonic anal. to approx. the solute entropy. Specific interactions can also be dissected using free energy decompn. or alanine scanning. A parallel implementation significantly speeds up the calcn. by dividing frames evenly across available processors. MMPBSA.py is an efficient, user-friendly program with the flexibility to accommodate the needs of users performing end-state free energy calcns. The source code can be downloaded at http://ambermd.org/ with AmberTools, released under the GNU General Public License.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhtV2gtrzP&md5=cc4148bd8f70c7cad94fd3ec6f580e52
12
Case, D. A.; Cheatham, T. E., III; Darden, T.; Gohlke, H.; Luo, R.; Merz, K. M., Jr.; Onufriev, A.; Simmerling, C.; Wang, B.; Woods, R. J. The Amber Biomolecular Simulation Programs J. Comput. Chem. 2005, 26, 1668– 1688 DOI: 10.1002/jcc.20290
Google Scholar
12
The amber biomolecular simulation programs
Case, David A.; Cheatham, Thomas E., III; Darden, Tom; Gohlke, Holger; Luo, Ray; Merz, Kenneth M., Jr.; Onufriev, Alexey; Simmerling, Carlos; Wang, Bing; Woods, Robert J.
Journal of Computational Chemistry (2005), 26 (16), 1668-1688CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
The authors describe the development, current features, and some directions for future development of the Amber package of computer programs. This package evolved from a program that was constructed in the late 1970s to do Assisted Model Building with Energy Refinement, and now contains a group of programs embodying a no. of powerful tools of modern computational chem., focused on mol. dynamics and free energy calcns. of proteins, nucleic acids, and carbohydrates.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXht1SlsbbM&md5=93be29ff894bab96c783d24e9886c7d0
13
Bhati, A. P.; Wan, S.; Wright, D. W.; Coveney, P. V. Rapid, Accurate, Precise, and Reliable Relative Free Energy Prediction Using Ensemble Based Thermodynamic Integration J. Chem. Theory Comput. 2017, 13, 210– 222 DOI: 10.1021/acs.jctc.6b00979
Google Scholar
13
Rapid, Accurate, Precise, and Reliable Relative Free Energy Prediction Using Ensemble Based Thermodynamic Integration
Bhati, Agastya P.; Wan, Shunzhou; Wright, David W.; Coveney, Peter V.
Journal of Chemical Theory and Computation (2017), 13 (1), 210-222CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
The accurate prediction of the binding affinities of ligands to proteins is a major goal in drug discovery and personalized medicine. The time taken to make such predictions is of similar importance to their accuracy, precision and reliability. In the last few years, an ensemble based mol. dynamics approach has been proposed that provides a route to reliable predictions of free energies based on the mol. mechanics Poisson-Boltzmann surface area method which meets the requirements of accuracy, precision and reliability. Here, we describe an equiv. methodol. based on thermodn. integration to substantially improve the accuracy, precision and reliability of calcd. relative binding free energies. We report the performance of the method when applied to a diverse set of protein targets and ligands. The results are in very good agreement with exptl. data (90% of calcns. agree to within 1 kcal/mol) while the method is reproducible by construction. Statistical uncertainties of the order of 0.5 kcal/mol or less are achieved. We present a systematic account of how the uncertainty in the predictions may be estd.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XitVGntrrJ&md5=510b70188112a3578030e291ce19b127
14
Wright, D. W.; Hall, B. A.; Kenway, O. A.; Jha, S.; Coveney, P. V. Computing Clinically Relevant Binding Free Energies of HIV-1 Protease Inhibitors J. Chem. Theory Comput. 2014, 10, 1228– 1241 DOI: 10.1021/ct4007037
Google Scholar
14
Computing Clinically Relevant Binding Free Energies of HIV-1 Protease Inhibitors
Wright, David W.; Hall, Benjamin A.; Kenway, Owain A.; Jha, Shantenu; Coveney, Peter V.
Journal of Chemical Theory and Computation (2014), 10 (3), 1228-1241CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
The use of mol. simulation to est. the strength of macromol. binding free energies is becoming increasingly widespread, with goals ranging from lead optimization and enrichment in drug discovery to personalizing or stratifying treatment regimes. To realize the potential of such approaches to predict new results, not merely to explain previous exptl. findings, it is necessary that the methods used are reliable and accurate, and that their limitations are thoroughly understood. However, the computational cost of atomistic simulation techniques such as mol. dynamics (MD) has meant that until recently little work has focused on validating and verifying the available free energy methodologies, with the consequence that many of the results published in the literature are not reproducible. Here, we present a detailed anal. of two of the most popular approx. methods for calcg. binding free energies from mol. simulations, mol. mechanics Poisson-Boltzmann surface area (MMPBSA) and mol. mechanics generalized Born surface area (MMGBSA), applied to the nine FDA-approved HIV-1 protease inhibitors. Our results show that the values obtained from replica simulations of the same protease-drug complex, differing only in initially assigned atom velocities, can vary by as much as 10 kcal mol-1, which is greater than the difference between the best and worst binding inhibitors under investigation. Despite this, anal. of ensembles of simulations producing 50 trajectories of 4 ns duration leads to well converged free energy ests. For seven inhibitors, we find that with correctly converged normal mode ests. of the configurational entropy, we can correctly distinguish inhibitors in agreement with exptl. data for both the MMPBSA and MMGBSA methods and thus have the ability to rank the efficacy of binding of this selection of drugs to the protease (no account is made for free energy penalties assocd. with protein distortion leading to the over estn. of the binding strength of the two largest inhibitors ritonavir and atazanavir). We obtain improved rankings and ests. of the relative binding strengths of the drugs by using a novel combination of MMPBSA/MMGBSA with normal mode entropy ests. and the free energy of assocn. calcd. directly from simulation trajectories. Our work provides a thorough assessment of what is required to produce converged and hence reliable free energies for protein-ligand binding.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXht12ltrk%253D&md5=56bc7e7c8f6bbdd694bfc76c20dcec63
15
Swanson, J. M.; Henchman, R. H.; McCammon, J. A. Revisiting Free Energy Calculations: A Theoretical Connection to MM/PBSA and Direct Calculation of the Association Free Energy Biophys. J. 2004, 86, 67– 74 DOI: 10.1016/S0006-3495(04)74084-9
Google Scholar
15
Revisiting free energy calculations: A theoretical connection to MM/PBSA and direct calculation of the association free energy
Swanson, Jessica M. J.; Henchman, Richard H.; McCammon, J. Andrew
Biophysical Journal (2004), 86 (1, Pt. 1), 67-74CODEN: BIOJAU; ISSN:0006-3495. (Biophysical Society)
The prediction of abs. ligand-receptor binding affinities is essential in a wide range of biophys. queries, from the study of protein-protein interactions to structure-based drug design. End-point free energy methods, such as the Mol. Mechanics Poisson-Boltzmann Surface Area (MM/PBSA) model, have received much attention and widespread application in recent literature. These methods benefit from computational efficiency as only the initial and final states of the system are evaluated, yet there remains a need for strengthening their theor. foundation. Here a clear connection between statistical thermodn. and end-point free energy models is presented. The importance of the assocn. free energy, arising from one mol.'s loss of translational and rotational freedom from the std. state concn., is addressed. A novel method for calcg. this quantity directly from a mol. dynamics simulation is described. The challenges of accounting for changes in the protein conformation and its fluctuations from sep. simulations are discussed. A simple first-order approxn. of the configuration integral is presented to lay the groundwork for future efforts. This model has been applied to FKBP12, a small immunophilin that has been widely studied in the drug industry for its potential immunosuppressive and neuroregenerative effects.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXlsV2hug%253D%253D&md5=720b2ec6ef0462edcda34f7be768bc44
16
Chodera, J. D.; Mobley, D. L.; Shirts, M. R.; Dixon, R. W.; Branson, K.; Pande, V. S. Alchemical Free Energy Methods for Drug Discovery: Progress and Challenges Curr. Opin. Struct. Biol. 2011, 21, 150– 160 DOI: 10.1016/j.sbi.2011.01.011
Google Scholar
16
Alchemical free energy methods for drug discovery: Progress and challenges
Chodera, John D.; Mobley, David L.; Shirts, Michael R.; Dixon, Richard W.; Branson, Kim; Pande, Vijay S.
Current Opinion in Structural Biology (2011), 21 (2), 150-160CODEN: COSBEF; ISSN:0959-440X. (Elsevier Ltd.)
A review. Improved rational drug design methods are needed to lower the cost and increase the success rate of drug discovery and development. Alchem. binding free energy calcns., one potential tool for rational design, have progressed rapidly over the past decade, but still fall short of providing robust tools for pharmaceutical engineering. Recent studies, esp. on model receptor systems, have clarified many of the challenges that must be overcome for robust predictions of binding affinity to be useful in rational design. In this review, inspired by a recent joint academic/industry meeting organized by the authors, we discuss these challenges and suggest a no. of promising approaches for overcoming them.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXkt1Kisrk%253D&md5=fa17c0982d921cd1c32cd095fd34420e
17
Wang, L.; Wu, Y.; Deng, Y.; Kim, B.; Pierce, L.; Krilov, G.; Lupyan, D.; Robinson, S.; Dahlgren, M. K.; Greenwood, J.; Romero, D. L.; Masse, C.; Knight, J. L.; Steinbrecher, T.; Beuming, T.; Damm, W.; Harder, E.; Sherman, W.; Brewer, M.; Wester, R.; Murcko, M.; Frye, L.; Farid, R.; Lin, T.; Mobley, D. L.; Jorgensen, W. L.; Berne, B. J.; Friesner, R. A.; Abel, R. Accurate and Reliable Prediction of Relative Ligand Binding Potency in Prospective Drug Discovery by Way of a Modern Free-Energy Calculation Protocol and Force Field J. Am. Chem. Soc. 2015, 137, 2695– 2703 DOI: 10.1021/ja512751q
Google Scholar
17
Accurate and Reliable Prediction of Relative Ligand Binding Potency in Prospective Drug Discovery by Way of a Modern Free-Energy Calculation Protocol and Force Field
Wang, Lingle; Wu, Yujie; Deng, Yuqing; Kim, Byungchan; Pierce, Levi; Krilov, Goran; Lupyan, Dmitry; Robinson, Shaughnessy; Dahlgren, Markus K.; Greenwood, Jeremy; Romero, Donna L.; Masse, Craig; Knight, Jennifer L.; Steinbrecher, Thomas; Beuming, Thijs; Damm, Wolfgang; Harder, Ed; Sherman, Woody; Brewer, Mark; Wester, Ron; Murcko, Mark; Frye, Leah; Farid, Ramy; Lin, Teng; Mobley, David L.; Jorgensen, William L.; Berne, Bruce J.; Friesner, Richard A.; Abel, Robert
Journal of the American Chemical Society (2015), 137 (7), 2695-2703CODEN: JACSAT; ISSN:0002-7863. (American Chemical Society)
Designing tight-binding ligands is a primary objective of small-mol. drug discovery. Over the past few decades, free-energy calcns. have benefited from improved force fields and sampling algorithms, as well as the advent of low-cost parallel computing. However, it has proven to be challenging to reliably achieve the level of accuracy that would be needed to guide lead optimization (∼5× in binding affinity) for a wide range of ligands and protein targets. Not surprisingly, widespread com. application of free-energy simulations has been limited due to the lack of large-scale validation coupled with the tech. challenges traditionally assocd. with running these types of calcns. Here, we report an approach that achieves an unprecedented level of accuracy across a broad range of target classes and ligands, with retrospective results encompassing 200 ligands and a wide variety of chem. perturbations, many of which involve significant changes in ligand chem. structures. In addn., we have applied the method in prospective drug discovery projects and found a significant improvement in the quality of the compds. synthesized that have been predicted to be potent. Compds. predicted to be potent by this approach have a substantial redn. in false positives relative to compds. synthesized on the basis of other computational or medicinal chem. approaches. Furthermore, the results are consistent with those obtained from our retrospective studies, demonstrating the robustness and broad range of applicability of this approach, which can be used to drive decisions in lead optimization.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhsF2iuro%253D&md5=37a4f4a6c085f47ed531342643b6c33b
18
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. J.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.Gaussian 03; Gaussian, Inc.: Wallingford, CT, 2004.
Google Scholar
There is no corresponding record for this reference.
19
Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A. Development and Testing of a General Amber Force Field J. Comput. Chem. 2004, 25, 1157– 1174 DOI: 10.1002/jcc.20035
Google Scholar
19
Development and testing of a general Amber force field
Wang, Junmei; Wolf, Romain M.; Caldwell, James W.; Kollman, Peter A.; Case, David A.
Journal of Computational Chemistry (2004), 25 (9), 1157-1174CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
We describe here a general Amber force field (GAFF) for org. mols. GAFF is designed to be compatible with existing Amber force fields for proteins and nucleic acids, and has parameters for most org. and pharmaceutical mols. that are composed of H, C, N, O, S, P, and halogens. It uses a simple functional form and a limited no. of atom types, but incorporates both empirical and heuristic models to est. force consts. and partial at. charges. The performance of GAFF in test cases is encouraging. In test I, 74 crystallog. structures were compared to GAFF minimized structures, with a root-mean-square displacement of 0.26 Å, which is comparable to that of the Tripos 5.2 force field (0.25 Å) and better than those of MMFF 94 and CHARMm (0.47 and 0.44 Å, resp.). In test II, gas phase minimizations were performed on 22 nucleic acid base pairs, and the minimized structures and intermol. energies were compared to MP2/6-31G* results. The RMS of displacements and relative energies were 0.25 Å and 1.2 kcal/mol, resp. These data are comparable to results from Parm99/RESP (0.16 Å and 1.18 kcal/mol, resp.), which were parameterized to these base pairs. Test III looked at the relative energies of 71 conformational pairs that were used in development of the Parm99 force field. The RMS error in relative energies (compared to expt.) is about 0.5 kcal/mol. GAFF can be applied to wide range of mols. in an automatic fashion, making it suitable for rational drug design and database searching.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXksFakurc%253D&md5=2992017a8cf51f89290ae2562403b115
20
Lindorff-Larsen, K.; Piana, S.; Palmo, K.; Maragakis, P.; Klepeis, J. L.; Dror, R. O.; Shaw, D. E. Improved Side-Chain Torsion Potentials for the Amber FF99SB Protein Force Field Proteins: Struct., Funct., Bioinf. 2010, 78, 1950– 1958 DOI: 10.1002/prot.22711
Google Scholar
20
Improved side-chain torsion potentials for the Amber ff99SB protein force field
Lindorff-Larsen, Kresten; Piana, Stefano; Palmo, Kim; Maragakis, Paul; Klepeis, John L.; Dror, Ron O.; Shaw, David E.
Proteins: Structure, Function, and Bioinformatics (2010), 78 (8), 1950-1958CODEN: PSFBAF ISSN:. (Wiley-Liss, Inc.)
Recent advances in hardware and software have enabled increasingly long mol. dynamics (MD) simulations of biomols., exposing certain limitations in the accuracy of the force fields used for such simulations and spurring efforts to refine these force fields. Recent modifications to the Amber and CHARMM protein force fields, for example, have improved the backbone torsion potentials, remedying deficiencies in earlier versions. Here, the authors further advance simulation accuracy by improving the amino acid side-chain torsion potentials of the Amber ff99SB force field. First, the authors used simulations of model alpha-helical systems to identify the four residue types whose rotamer distribution differed the most from expectations based on Protein Data Bank statistics. Second, the authors optimized the side-chain torsion potentials of these residues to match new, high-level quantum-mech. calcns. Finally, the authors used microsecond-timescale MD simulations in explicit solvent to validate the resulting force field against a large set of exptl. NMR measurements that directly probe side-chain conformations. The new force field, which the authors have termed Amber ff99SB-ILDN, exhibits considerably better agreement with the NMR data. Proteins 2010. © 2010 Wiley-Liss, Inc.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXkvFegtLo%253D&md5=447a9004026e2b93f0f7beff165daa09
21
Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable Molecular Dynamics with NAMD J. Comput. Chem. 2005, 26, 1781– 1802 DOI: 10.1002/jcc.20289
Google Scholar
21
Scalable molecular dynamics with NAMD
Phillips, James C.; Braun, Rosemary; Wang, Wei; Gumbart, James; Tajkhorshid, Emad; Villa, Elizabeth; Chipot, Christophe; Skeel, Robert D.; Kale, Laxmikant; Schulten, Klaus
Journal of Computational Chemistry (2005), 26 (16), 1781-1802CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
NAMD is a parallel mol. dynamics code designed for high-performance simulation of large biomol. systems. NAMD scales to hundreds of processors on high-end parallel platforms, as well as tens of processors on low-cost commodity clusters, and also runs on individual desktop and laptop computers. NAMD works with AMBER and CHARMM potential functions, parameters, and file formats. This article, directed to novices as well as experts, first introduces concepts and methods used in the NAMD program, describing the classical mol. dynamics force field, equations of motion, and integration methods along with the efficient electrostatics evaluation algorithms employed and temp. and pressure controls used. Features for steering the simulation across barriers and for calcg. both alchem. and conformational free energy differences are presented. The motivations for and a roadmap to the internal design of NAMD, implemented in C++ and based on Charm++ parallel objects, are outlined. The factors affecting the serial and parallel performance of a simulation are discussed. Finally, typical NAMD use is illustrated with representative applications to a small, a medium, and a large biomol. system, highlighting particular features of NAMD, for example, the Tcl scripting language. The article also provides a list of the key features of NAMD and discusses the benefits of combining NAMD with the mol. graphics/sequence anal. software VMD and the grid computing/collab. software BioCoRE. NAMD is distributed free of charge with source code at www.ks.uiuc.edu.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXht1SlsbbJ&md5=189051128443b547f4300a1b8fb0e034
22
Highfield, R. Supercomputer bid to create the first truly personalised medicine. https://blog.sciencemuseum.org.uk/supercomputer-bid-to-create-the-first-truly-personalised-medicine/ (accessed Feb 22, 2017).
Google Scholar
There is no corresponding record for this reference.
23
Fiser, A.; Sali, A. Modloop: Automated Modeling of Loops in Protein Structures Bioinformatics 2003, 19, 2500– 2501 DOI: 10.1093/bioinformatics/btg362
Google Scholar
23
ModLoop: automated modeling of loops in protein structures
Fiser, Andras; Sali, Andrej
Bioinformatics (2003), 19 (18), 2500-2501CODEN: BOINFP; ISSN:1367-4803. (Oxford University Press)
Summary: ModLoop is a web server for automated modeling of loops in protein structures. The input is the at. coordinates of the protein structure in the Protein Data Bank format, and the specification of the starting and ending residues of one or more segments to be modeled, contg. no more than 20 residues in total. The output is the coordinates of the non-hydrogen atoms in the modeled segments. A user provides the input to the server via a simple web interface, and receives the output by e-mail. The server relies on the loop modeling routine in MODELLER that predicts the loop conformations by satisfaction of spatial restraints, without relying on a database of known protein structures. For a rapid response, ModLoop runs on a cluster of Linux PC computers.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXpvVSisLk%253D&md5=4b0fdaca0a412715682496883ab2019c
24
Allen, W. J.; Balius, T. E.; Mukherjee, S.; Brozell, S. R.; Moustakas, D. T.; Lang, P. T.; Case, D. A.; Kuntz, I. D.; Rizzo, R. C. Dock 6: Impact of New Features and Current Docking Performance J. Comput. Chem. 2015, 36, 1132– 1156 DOI: 10.1002/jcc.23905
Google Scholar
24
DOCK 6: Impact of new features and current docking performance
Allen, William J.; Balius, Trent E.; Mukherjee, Sudipto; Brozell, Scott R.; Moustakas, Demetri T.; Lang, P. Therese; Case, David A.; Kuntz, Irwin D.; Rizzo, Robert C.
Journal of Computational Chemistry (2015), 36 (15), 1132-1156CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
This manuscript presents the latest algorithmic and methodol. developments to the structure-based design program DOCK 6.7 focused on an updated internal energy function, new anchor selection control, enhanced minimization options, a footprint similarity scoring function, a symmetry-cor. root-mean-square deviation algorithm, a database filter, and docking forensic tools. An important strategy during development involved use of three orthogonal metrics for assessment and validation: pose reprodn. over a large database of 1043 protein-ligand complexes (SB2012 test set), cross-docking to 24 drug-target protein families, and database enrichment using large active and decoy datasets (Directory of Useful Decoys [DUD]-E test set) for five important proteins including HIV protease and IGF-1R. Relative to earlier versions, a key outcome of the work is a significant increase in pose reprodn. success in going from DOCK 4.0.2 (51.4%) → 5.4 (65.2%) → 6.7 (73.3%) as a result of significant decreases in failure arising from both sampling 24.1% → 13.6% → 9.1% and scoring 24.4% → 21.1% → 17.5%. Companion cross-docking and enrichment studies with the new version highlight other strengths and remaining areas for improvement, esp. for systems contg. metal ions. The source code for DOCK 6.7 is available for download and free for academic users at. © 2015 Wiley Periodicals, Inc.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXnt1Smtbs%253D&md5=8b5fc6cc7533f975e1e740acb3688807
25
Chodera, J. D.; Mobley, D. L. Entropy-Enthalpy Compensation: Role and Ramifications in Biomolecular Ligand Recognition and Design Annu. Rev. Biophys. 2013, 42, 121– 142 DOI: 10.1146/annurev-biophys-083012-130318
Google Scholar
25
Entropy-enthalpy compensation: role and ramifications in biomolecular ligand recognition and design
Chodera, John D.; Mobley, David L.
Annual Review of Biophysics (2013), 42 (), 121-142CODEN: ARBNCV; ISSN:1936-122X. (Annual Reviews)
A review. Recent calorimetric studies of interactions between small mols. and biomol. targets have generated renewed interest in the phenomenon of entropy-enthalpy compensation. In these studies, entropic and enthalpic contributions to binding are obsd. to vary substantially and in an opposing manner as the ligand or protein is modified, whereas the binding free energy varies little. In severe examples, engineered enthalpic gains can lead to completely compensating entropic penalties, frustrating ligand design. Here, we examine the evidence for compensation, as well as its potential origins, prevalence, severity, and ramifications for ligand engineering. We find the evidence for severe compensation to be weak in light of the large magnitude of and correlation between errors in exptl. measurements of entropic and enthalpic contributions to binding, though a limited form of compensation may be common. Given the difficulty of predicting or measuring entropic and enthalpic changes to useful precision, or using this information in design, we recommend ligand engineering efforts instead focus on computational and exptl. methodologies to directly assess changes in binding free energy.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3sXhtFGrs7bP&md5=52a30a1d0f4f9ae49128e87d2a23e3fa
26
Bissantz, C.; Folkers, G.; Rognan, D. Protein-Based Virtual Screening of Chemical Databases. 1. Evaluation of Different Docking/Scoring Combinations J. Med. Chem. 2000, 43, 4759– 4767 DOI: 10.1021/jm001044l
Google Scholar
26
Protein-Based Virtual Screening of Chemical Databases. 1. Evaluation of Different Docking/Scoring Combinations
Bissantz, Caterina; Folkers, Gerd; Rognan, Didier
Journal of Medicinal Chemistry (2000), 43 (25), 4759-4767CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
Three different database docking programs (Dock, FlexX, Gold) have been used in combination with seven scoring functions (Chemscore, Dock, FlexX, Fresno, Gold, Pmf, Score) to assess the accuracy of virtual screening methods against two protein targets (thymidine kinase, estrogen receptor) of known three-dimensional structure. For both targets, it was generally possible to discriminate about 7 out of 10 true hits from a random database of 990 ligands. The use of consensus lists common to two or three scoring functions clearly enhances hit rates among the top 5% scorers from 10% (single scoring) to 25-40% (double scoring) and up to 65-70% (triple scoring). However, in all tested cases, no clear relationships could be found between docking and ranking accuracies. Moreover, predicting the abs. binding free energy of true hits was not possible whatever docking accuracy was achieved and scoring function used. As the best docking/consensus scoring combination varies with the selected target and the physicochem. of target-ligand interactions, we propose a two-step protocol for screening large databases: (i) screening of a reduced dataset contg. a few known ligands for deriving the optimal docking/consensus scoring scheme, (ii) applying the latter parameters to the screening of the entire database.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3cXotFymurY%253D&md5=ee74ac9a99e55759c2df27fd1db58010
27
Gohlke, H.; Kiel, C.; Case, D. A. Insights into Protein-Protein Binding by Binding Free Energy Calculation and Free Energy Decomposition for the Ras-Raf and Ras-RaiGDS Complexes J. Mol. Biol. 2003, 330, 891– 913 DOI: 10.1016/S0022-2836(03)00610-7
Google Scholar
27
Insights into protein-protein binding by binding free energy calculation and free energy decomposition for the Ras-Raf and Ras-RalGDS complexes
Gohlke, Holger; Kiel, Christina; Case, David A.
Journal of Molecular Biology (2003), 330 (4), 891-913CODEN: JMOBAK; ISSN:0022-2836. (Elsevier Science Ltd.)
Abs. binding free energy calcns. and free energy decompns. are presented for the protein-protein complexes H-Ras/C-Raf1 and H-Ras/RalGDS. Ras is a central switch in the regulation of cell proliferation and differentiation. In our study, we investigate the capability of the mol. mechanics (MM)-generalized Born surface area (GBSA) approach to est. abs. binding free energies for the protein-protein complexes. Averaging gas-phase energies, solvation free energies, and entropic contributions over snapshots extd. from trajectories of the unbound proteins and the complexes, calcd. binding free energies (Ras-Raf: -15.0(±6.3) kcal mol-1; Ras-RalGDS: -19.5(±5.9) kcal mol-1) are in fair agreement with exptl. detd. values (-9.6 kcal mol-1; -8.4 kcal mol-1), if appropriate ionic strength is taken into account. Structural determinants of the binding affinity of Ras-Raf and Ras-RalGDS are identified by means of free energy decompn. For the first time, computationally inexpensive generalized Born (GB) calcns. are applied in this context to partition solvation free energies along with gas-phase energies between residues of both binding partners. For selected residues, in addn., entropic contributions are estd. by classical statistical mechanics. Comparison of the decompn. results with exptl. detd. binding free energy differences for alanine mutants of interface residues yielded correlations with r2=0.55 and 0.46 for Ras-Raf and Ras-RalGDS, resp. Extension of the decompn. reveals residues as far apart as 25 A from the binding epitope that can contribute significantly to binding free energy. These "hotspots" are found to show large at. fluctuations in the unbound proteins, indicating that they reside in structurally less stable regions. Furthermore, hotspot residues experience a significantly larger-than-av. decrease in local fluctuations upon complex formation. Finally, by calcg. a pair-wise decompn. of interactions, interaction pathways originating in the binding epitope of Raf are found that protrude through the protein structure towards the loop L1. This explains the finding of a conformational change in this region upon complex formation with Ras, and it may trigger a larger structural change in Raf, which is considered to be necessary for activation of the effector by Ras.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXlt1aqtbs%253D&md5=0a8acf22e053534f34365f264ed4192a
28
Bunney, T. D.; Wan, S.; Thiyagarajan, N.; Sutto, L.; Williams, S. V.; Ashford, P.; Koss, H.; Knowles, M. A.; Gervasio, F. L.; Coveney, P. V.; Katan, M. The Effect of Mutations on Drug Sensitivity and Kinase Activity of Fibroblast Growth Factor Receptors: A Combined Experimental and Theoretical Study EBioMedicine 2015, 2, 194– 204 DOI: 10.1016/j.ebiom.2015.02.009
Google Scholar
28
The Effect of Mutations on Drug Sensitivity and Kinase Activity of Fibroblast Growth Factor Receptors: A Combined Experimental and Theoretical Study
Bunney Tom D; Thiyagarajan Nethaji; Ashford Paul; Katan Matilda; Wan Shunzhou; Coveney Peter V; Sutto Ludovico; Gervasio Francesco L; Williams Sarah V; Knowles Margaret A; Koss Hans
EBioMedicine (2015), 2 (3), 194-204 ISSN:.
Fibroblast growth factor receptors (FGFRs) are recognized therapeutic targets in cancer. We here describe insights underpinning the impact of mutations on FGFR1 and FGFR3 kinase activity and drug efficacy, using a combination of computational calculations and experimental approaches including cellular studies, X-ray crystallography and biophysical and biochemical measurements. Our findings reveal that some of the tested compounds, in particular TKI258, could provide therapeutic opportunity not only for patients with primary alterations in FGFR but also for acquired resistance due to the gatekeeper mutation. The accuracy of the computational methodologies applied here shows a potential for their wider application in studies of drug binding and in assessments of functional and mechanistic impacts of mutations, thus assisting efforts in precision medicine.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A280%3ADC%252BC2srotlKmtA%253D%253D&md5=8414f3530d24bbba0ea8cc98cb784b96
29
Andrews, M. D.; Bagal, S. K.; Gibson, K. R.; Omoto, K.; Ryckmans, T.; Skerratt, S. E.; Stupple, P. A. Pyrrolo[2,3-d]pyrimidine Derivatives as Inhibitors of Tropomyosin-Related Kinases and Their Preparation and Use in the Treatment of Pain. WO2012137089A1, 2012.
Google Scholar
There is no corresponding record for this reference.
30
Xing, L.; Klug-Mcleod, J.; Rai, B.; Lunney, E. A. Kinase Hinge Binding Scaffolds and Their Hydrogen Bond Patterns Bioorg. Med. Chem. 2015, 23, 6520– 6527 DOI: 10.1016/j.bmc.2015.08.006
Google Scholar
30
Kinase hinge binding scaffolds and their hydrogen bond patterns
Xing, Li; Klug-Mcleod, Jacquelyn; Rai, Brajesh; Lunney, Elizabeth A.
Bioorganic & Medicinal Chemistry (2015), 23 (19), 6520-6527CODEN: BMECEP; ISSN:0968-0896. (Elsevier B.V.)
Protein kinases constitute a major class of intracellular signaling mols., and describe some of the most prominent drug targets. Kinase inhibitors commonly employ small chem. scaffolds that form hydrogen bonds with the kinase hinge residues connecting the N- and C-terminal lobes of the catalytic domain. In general the satisfied hydrogen bonds are required for potent inhibition, therefore constituting a conserved feature in the majority of inhibitor-kinase interactions. From systematically analyzing the kinase scaffolds extd. from Pfizer crystal structure database (CSDb) the authors recognize that large no. of kinase inhibitors of diverse chem. structures are derived from a relatively small no. of common scaffolds. Depending on specific substitution patterns, scaffolds may demonstrate versatile binding capacities to interact with kinase hinge. Afforded by thousands of ligand-protein binary complexes, the hinge hydrogen bond patterns were analyzed with a focus on their three-dimensional configurations. Most of the compds. engage H6 NH for hinge recognition. Dual hydrogen bonds are commonly obsd. with addnl. recruitment of H4 CO upstream and/or H6 CO downstream. Triple hydrogen bonds accounts for small no. of binary complexes. An unusual hydrogen bond with a non-canonical H5 conformation is obsd., requiring a peptide bond flip by a glycine residue at the H6 position. Addnl. hydrogen bonds to kinase hinge do not necessarily correlate with an increase in potency; conversely they appear to compromise kinase selectivity. Such learnings could enhance the prospect of successful therapy design.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhsVentr3L&md5=048b4b99628c6e1ae273b00a47ab2377
31
Morphy, R. Selectively Nonselective Kinase Inhibition: Striking the Right Balance J. Med. Chem. 2010, 53, 1413– 1437 DOI: 10.1021/jm901132v
Google Scholar
31
Selectively Nonselective Kinase Inhibition: Striking the Right Balance
Morphy, Richard
Journal of Medicinal Chemistry (2010), 53 (4), 1413-1437CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
A review. Marketed kinase inhibitors (MKIs) can deliver superior efficacy compared to inhibitors with high specificity for a single kinase, and the recent introduction of several MKIs to the market opens the door to a new era of safer and effective anticancer therapy. The key to combining high efficacy with acceptable safety is to inhibit multiple targets in a selectively nonselective fashion. Strategies for intentionally designing MKIs are emerging, but the field is still in its infancy and we are as medicinal chemists currently on the steepest part of the learning curve. MTDD can be time-consuming and expensive, and we need to become more proficient first at identifying disease-relevant target combinations and second at discovering MKIs that combine optimal physicochem. and biol. properties. Bold and innovative medicinal chem. strategies are required to tackle "difficult combinations" where the disease rationale is compelling but where it is a struggle to combine all the desired attributes of an oral MKI drug into a single mol. At present it is unclear to what extent MKIs with highly tuned selectivity profiles can be rationally designed, particularly for targets that are unrelated by sequence. In addn. to the well-known selectivity challenge, the physicochem. property profiles of AT P-competitive MKIs can be inherently challenging and limited scope for patentability can also be a serious hindrance. On the plus side, the amt. of kinase-specific structural information is growing very rapidly, and ultimately this may reveal distinct features and design rules that enable a medicinal chemist to rationally modify and refine the profile of MKIs. In addn., increasing SAR knowledge is emerging from large scale panel screening with the binding profiles starting to reveal to medicinal chemists how chem. structure affects cross-reactivity across large parts of the kinome. The merit of MKIs compared with single kinase inhibitors is a subject of controversy in drug discovery that is unlikely to be resolved in the near future. At the start of a new MTDD project, a rigorous debate needs to take place as to whether it makes more sense to seek a combination of highly selective agents or a DML. Many factors need to be taken into account in this decision such as the no., similarity, and promiscuity of the targets in the profile and the disease area. Conformational plasticity and the occurrence of multiple binding modes complicate the in silico prediction of kinase polypharmacol. based solely upon protein structure. The use of ligand-based similarity to assess the feasibility of a given combination can add real value. Currently, serendipity plays a significant role in MKI discovery and many, if not most, MKIs have been discovered by chance during the search for selective inhibitors. Medicinal chemists need to be alert to the possibilities when a surprising combination is found by chance. To exploit such serendipity, you need a good appreciation of when you have a sufficiently high quality starting compd. and then you need to be able to make and test sufficient analogs to explore your new disease-based hypothesis. MKIs are costly to develop and are consequently priced at a premium level, so they will need to show clear improvements in order to get reimbursement. There have already been problems with reimbursement for some MKIs in some markets due to concerns from funding bodies over insufficient efficacy. The true value of MKIs relative to other anticancer drugs still has to be established, and the results from recent clin. trials have been mixed. Despite the broad activity profile of many MKIs, the patient response can be inconsistent and unpredictable. The identification of predictive biomarkers of response or resistance is a crit. step to ascertain which specific combination of targets produces a significant clin. benefit with respect to specific tumor types. More clin. feedback is needed to facilitate the design of the next generation of inhibitors with more precisely defined profiles. Although it might seem immeasurably distant at the present time, the ultimate goal should be to derive the prerequisite knowledge and tools so that MTDD becomes a rational endeavor rather than a black box approach that relies upon serendipity. This will help banish claims that MKIs are merely dirty, nonspecific drugs with insufficient specificity for treating a wider range of human diseases.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXhtlWmtbfF&md5=3c915b5debf9d005993c8ad31c68bc2e
32
Ou-Yang, S. S.; Lu, J. Y.; Kong, X. Q.; Liang, Z. J.; Luo, C.; Jiang, H. Computational Drug Discovery Acta Pharmacol. Sin. 2012, 33, 1131– 1140 DOI: 10.1038/aps.2012.109
Google Scholar
32
Computational drug discovery
Ou-Yang, Si-sheng; Lu, Jun-yan; Kong, Xiang-qian; Liang, Zhong-jie; Luo, Cheng; Jiang, Hualiang
Acta Pharmacologica Sinica (2012), 33 (9), 1131-1140CODEN: APSCG5; ISSN:1671-4083. (Nature Publishing Group)
A review. Computational drug discovery is an effective strategy for accelerating and economizing drug discovery and development process. Because of the dramatic increase in the availability of biol. macromol. and small mol. information, the applicability of computational drug discovery has been extended and broadly applied to nearly every stage in the drug discovery and development workflow, including target identification and validation, lead discovery and optimization and preclin. tests. Over the past decades, computational drug discovery methods such as mol. docking, pharmacophore modeling and mapping, de novo design, mol. similarity calcn. and sequence-based virtual screening have been greatly improved. In this review, we present an overview of these important computational methods, platforms and successful applications in this field.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38Xht12lt7vL&md5=2a0bff666d5881d1109124049480b1a2
33
Wan, S.; Bhati, A. P.; Zasada, S. J.; Wall, I.; Green, D.; Bamborough, P.; Coveney, P. V. Rapid and Reliable Binding Affinity Prediction of Bromodomain Inhibitors: A Computational Study J. Chem. Theory Comput. 2017, 13, 784– 795 DOI: 10.1021/acs.jctc.6b00794
Google Scholar
33
Rapid and Reliable Binding Affinity Prediction of Bromodomain Inhibitors: A Computational Study
Wan, Shunzhou; Bhati, Agastya P.; Zasada, Stefan J.; Wall, Ian; Green, Darren; Bamborough, Paul; Coveney, Peter V.
Journal of Chemical Theory and Computation (2017), 13 (2), 784-795CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
Binding free energies of bromodomain inhibitors are calcd. with recently formulated approaches, namely ESMACS (enhanced sampling of mol. dynamics with approxn. of continuum solvent) and TIES (thermodn. integration with enhanced sampling). A set of compds. is provided by GlaxoSmithKline, which represents a range of chem. functionality and binding affinities. The predicted binding free energies exhibit a good Spearman correlation of 0.78 with the exptl. data from the 3-trajectory ESMACS, and an excellent correlation of 0.92 from the TIES approach where applicable. Given access to suitable high end computing resources and a high degree of automation, the authors can compute individual binding affinities in a few hours with precisions no greater than 0.2 kcal/mol for TIES, and no larger than 0.34 kcal/mol and 1.71 kcal/mol for the 1- and 3-trajectory ESMACS approaches.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XitFektLnF&md5=713436084662420482684fb50db0832e

Cited By

ARTICLE SECTIONS

Jump To

This article is cited by 29 publications.

Alexander D. Wade, Agastya P. Bhati, Shunzhou Wan, Peter V. Coveney. Alchemical Free Energy Estimators and Molecular Dynamics Engines: Accuracy, Precision, and Reproducibility. Journal of Chemical Theory and Computation 2022, 18 (6) , 3972-3987. https://doi.org/10.1021/acs.jctc.2c00114
Shunzhou Wan, Agastya P. Bhati, David W. Wright, Ian D. Wall, Alan P. Graves, Darren Green, Peter V. Coveney. Ensemble Simulations and Experimental Free Energy Distributions: Evaluation and Characterization of Isoxazole Amides as SMYD3 Inhibitors. Journal of Chemical Information and Modeling 2022, 62 (10) , 2561-2570. https://doi.org/10.1021/acs.jcim.2c00255
Agastya P. Bhati, Peter V. Coveney. Large Scale Study of Ligand–Protein Relative Binding Free Energy Calculations: Actionable Predictions from Statistically Robust Protocols. Journal of Chemical Theory and Computation 2022, 18 (4) , 2687-2702. https://doi.org/10.1021/acs.jctc.1c01288
Agastya P. Bhati, Shunzhou Wan, Peter V. Coveney. Ensemble-Based Replica Exchange Alchemical Free Energy Methods: The Effect of Protein Mutations on Inhibitor Binding. Journal of Chemical Theory and Computation 2019, 15 (2) , 1265-1277. https://doi.org/10.1021/acs.jctc.8b01118
Sharan K. Bagal, Mark Andrews, Bruce M. Bechle, Jianwei Bian, James Bilsland, David C. Blakemore, John F. Braganza, Peter J. Bungay, Matthew S. Corbett, Ciaran N. Cronin, Jingrong Jean Cui, Rebecca Dias, Neil J. Flanagan, Samantha E. Greasley, Rachel Grimley, Kim James, Eric Johnson, Linda Kitching, Michelle L. Kraus, Indrawan McAlpine, Asako Nagata, Sacha Ninkovic, Kiyoyuki Omoto, Stephanie Scales, Sarah E. Skerratt, Jianmin Sun, Michelle Tran-Dubé, Gareth J. Waldron, Fen Wang, Joseph S. Warmus. Discovery of Potent, Selective, and Peripherally Restricted Pan-Trk Kinase Inhibitors for the Treatment of Pain. Journal of Medicinal Chemistry 2018, 61 (15) , 6779-6800. https://doi.org/10.1021/acs.jmedchem.8b00633
Agastya P. Bhati, Shunzhou Wan, Yuan Hu, Brad Sherborne, Peter V. Coveney. Uncertainty Quantification in Alchemical Free Energy Methods. Journal of Chemical Theory and Computation 2018, 14 (6) , 2867-2880. https://doi.org/10.1021/acs.jctc.7b01143
José Jiménez, Miha Škalič, Gerard Martínez-Rosell, and Gianni De Fabritiis . KDEEP: Protein–Ligand Absolute Binding Affinity Prediction via 3D-Convolutional Neural Networks. Journal of Chemical Information and Modeling 2018, 58 (2) , 287-296. https://doi.org/10.1021/acs.jcim.7b00650
Nojood A. Altwaijry, Michael Baron, David W. Wright, Peter V. Coveney, and Andrea Townsend-Nicholson . An Ensemble-Based Protocol for the Computational Prediction of Helix–Helix Interactions in G Protein-Coupled Receptors using Coarse-Grained Molecular Dynamics. Journal of Chemical Theory and Computation 2017, 13 (5) , 2254-2270. https://doi.org/10.1021/acs.jctc.6b01246
Alexander Heifetz. Accelerating COVID-19 Drug Discovery with High-Performance Computing. 2024, 405-411. https://doi.org/10.1007/978-1-0716-3449-3_19
Siyabonga Melamane, Tavonga T. Mandava, Arthur Manda, Nonhlanhla Luphade, Sandile M.M. Khamanga, Pedzisai A. Makoni, Patrick H. Demana, Scott K. Matafwali, Bwalya A. Witika. Artificial neural network–based inference of drug–target interactions. 2023, 35-62. https://doi.org/10.1016/B978-0-323-91763-6.00015-1
Shunzhou Wan, Agastya P. Bhati, David W. Wright, Alexander D. Wade, Gary Tresadern, Herman van Vlijmen, Peter V. Coveney. The performance of ensemble-based free energy protocols in computing binding affinities to ROS1 kinase. Scientific Reports 2022, 12 (1) https://doi.org/10.1038/s41598-022-13319-6
Shunzhou Wan, Agastya P. Bhati, Alexander D. Wade, Dario Alfè, Peter V. Coveney. Thermodynamic and structural insights into the repurposing of drugs that bind to SARS-CoV-2 main protease. Molecular Systems Design & Engineering 2022, 7 (2) , 123-131. https://doi.org/10.1039/D1ME00124H
Agastya P. Bhati, Shunzhou Wan, Dario Alfè, Austin R. Clyde, Mathis Bode, Li Tan, Mikhail Titov, Andre Merzky, Matteo Turilli, Shantenu Jha, Roger R. Highfield, Walter Rocchia, Nicola Scafuri, Sauro Succi, Dieter Kranzlmüller, Gerald Mathias, David Wifling, Yann Donon, Alberto Di Meglio, Sofia Vallecorsa, Heng Ma, Anda Trifan, Arvind Ramanathan, Tom Brettin, Alexander Partin, Fangfang Xia, Xiaotan Duan, Rick Stevens, Peter V. Coveney. Pandemic drugs at pandemic speed: infrastructure for accelerating COVID-19 drug discovery with hybrid machine learning- and physics-based simulations on high-performance computers. Interface Focus 2021, 11 (6) https://doi.org/10.1098/rsfs.2021.0018
Shunzhou Wan, Deepak Kumar, Valentin Ilyin, Ussama Al Homsi, Gulab Sher, Alexander Knuth, Peter V. Coveney. The effect of protein mutations on drug binding suggests ensuing personalised drug selection. Scientific Reports 2021, 11 (1) https://doi.org/10.1038/s41598-021-92785-w
Shunzhou Wan, Robert C. Sinclair, Peter V. Coveney. Uncertainty quantification in classical molecular dynamics. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 2021, 379 (2197) https://doi.org/10.1098/rsta.2020.0082
Francis Giraud, Elisabeth Pereira, Fabrice Anizon, Pascale Moreau. Recent Advances in Pain Management: Relevant Protein Kinases and Their Inhibitors. Molecules 2021, 26 (9) , 2696. https://doi.org/10.3390/molecules26092696
Shunzhou Wan, Andrew Potterton, Fouad S. Husseini, David W. Wright, Alexander Heifetz, Maciej Malawski, Andrea Townsend-Nicholson, Peter V. Coveney. Hit-to-lead and lead optimization binding free energy calculations for G protein-coupled receptors. Interface Focus 2020, 10 (6) , 20190128. https://doi.org/10.1098/rsfs.2019.0128
S. J. Zasada, D. W. Wright, P. V. Coveney. Large-scale binding affinity calculations on commodity compute clouds. Interface Focus 2020, 10 (6) , 20190133. https://doi.org/10.1098/rsfs.2019.0133
Shunzhou Wan, Agastya P. Bhati, Stefan J. Zasada, Peter V. Coveney. Rapid, accurate, precise and reproducible ligand–protein binding free energy prediction. Interface Focus 2020, 10 (6) , 20200007. https://doi.org/10.1098/rsfs.2020.0007
Anita de Ruiter, Chris Oostenbrink. Advances in the calculation of binding free energies. Current Opinion in Structural Biology 2020, 61 , 207-212. https://doi.org/10.1016/j.sbi.2020.01.016
David W. Wright, Fouad Husseini, Shunzhou Wan, Christophe Meyer, Herman van Vlijmen, Gary Tresadern, Peter V. Coveney. Application of the ESMACS Binding Free Energy Protocol to a Multi‐Binding Site Lactate Dehydogenase A Ligand Dataset. Advanced Theory and Simulations 2020, 3 (1) https://doi.org/10.1002/adts.201900194
Shunzhou Wan, Gary Tresadern, Laura Pérez‐Benito, Herman van Vlijmen, Peter V. Coveney. Accuracy and Precision of Alchemical Relative Free Energy Predictions with and without Replica‐Exchange. Advanced Theory and Simulations 2020, 3 (1) https://doi.org/10.1002/adts.201900195
José Jiménez-Luna, Laura Pérez-Benito, Gerard Martínez-Rosell, Simone Sciabola, Rubben Torella, Gary Tresadern, Gianni De Fabritiis. DeltaDelta neural networks for lead optimization of small molecule potency. Chemical Science 2019, 10 (47) , 10911-10918. https://doi.org/10.1039/C9SC04606B
David W. Wright, Shunzhou Wan, Christophe Meyer, Herman van Vlijmen, Gary Tresadern, Peter V. Coveney. Application of ESMACS binding free energy protocols to diverse datasets: Bromodomain-containing protein 4. Scientific Reports 2019, 9 (1) https://doi.org/10.1038/s41598-019-41758-1
Robert Roskoski. Targeting ERK1/2 protein-serine/threonine kinases in human cancers. Pharmacological Research 2019, 142 , 151-168. https://doi.org/10.1016/j.phrs.2019.01.039
Jumana Dakka, Matteo Turilli, David W. Wright, Stefan J. Zasada, Vivek Balasubramanian, Shunzhou Wan, Peter V. Coveney, Shantenu Jha. High-throughput binding affinity calculations at extreme scales. BMC Bioinformatics 2018, 19 (S18) https://doi.org/10.1186/s12859-018-2506-6
Robert Abel, Eric S Manas, Richard A Friesner, Ramy S Farid, Lingle Wang. Modeling the value of predictive affinity scoring in preclinical drug discovery. Current Opinion in Structural Biology 2018, 52 , 103-110. https://doi.org/10.1016/j.sbi.2018.09.002
Govindan Subramanian, Gennady Poda. In silico ligand‐based modeling of h BACE ‐1 inhibitors. Chemical Biology & Drug Design 2018, 91 (3) , 817-827. https://doi.org/10.1111/cbdd.13147
R. Charlotte Eccleston, Shunzhou Wan, Neil Dalchau, Peter V. Coveney. The Role of Multiscale Protein Dynamics in Antigen Presentation and T Lymphocyte Recognition. Frontiers in Immunology 2017, 8 https://doi.org/10.3389/fimmu.2017.00797

Download PDF

Abstract
High Resolution Image
Download MS PowerPoint Slide
Figure 1
Figure 1. Crystal structure of 1 bound to TrkA, viewed from the N-lobe to the C-lobe of the kinase. Hydrogen bonds are displayed by dashed lines. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. For clarity, the N-lobe is not shown.
High Resolution Image
Download MS PowerPoint Slide
Figure 2
Figure 2. Comparisons of the TrkA experimental data and the calculated binding free energies from (a) 1-traj, (b) 2-traj, and (c) 3-traj ESMACS approaches. The equation in each panel indicates the calculations used in each case. The term G_com/rec/lig^com/lig represents the free energy of the component (subscript: complex, receptor, or ligand) obtained from simulation of the same or different component (superscript: complex or free ligand in aqueous solution). The calculated binding free energies are associated with standard errors of ca. 0.6 kcal/mol in the 1-traj approach and ca. 3 kcal/mol in the 2- and 3-traj approaches (see Table S2), which are not shown in the figures for reasons of clarity. The experimental data from two sites (Pfizer, Sandwich and TCG Lifescience) are displayed in black and red, respectively. The correlation coefficients shown in the figure were calculated using the averages of the calculated binding free energies and the experimental data from Pfizer, Sandwich (black circles) and TCG Lifescience (red circles) where the former are not available. Large uncertainties are associated with the correlation coefficients because of the large error bars of the calculated and experimental binding affinities. Further analyses with bootstrapping resampling (see details in the Supporting Information) generate correlation coefficients of 0.39 ± 0.26, 0.60 ± 0.26, and 0.62 ± 0.27 for the 1-, 2-, and 3-traj approaches, respectively. It is evident that the 2- and 3-traj methods improve the ranking provided by the 1-traj method.
High Resolution Image
Download MS PowerPoint Slide
Figure 3
Figure 3. Calculated binding free energies for 1, 3, 4, 6, 7, 8, 13, 16, and 22 (a) from single structures after docking, (b) from 25 structures for each compound after 11 000-step minimization with a sophisticated conjugate-gradient method, and (c) from ensemble averages from three-trajectory ESMACS studies. Compounds 1, 3, 4, 6, 7, 8, 13, 16, and 22 are highly structurally similar (see Table 1).
High Resolution Image
Download MS PowerPoint Slide
Figure 4
Figure 4. Correlations of free energy components and the experimental data from the 3-traj approach. Both (a) bonded and (b) nonbonded energy terms contribute to the ranking of binding affinities, with similar correlation coefficients between the calculations and experimental data. Their combination, the MMPBSA energy (c), exhibits better correlations with the experimental data than the components themselves.
High Resolution Image
Download MS PowerPoint Slide
Figure 5
Figure 5. Adaptation free energies of (a) the receptor and (b) the compounds showing the binding free energy changes between the 1- and 3-traj approaches. The terms G_rec/lig^com/lig are the free energies of receptor or ligands (subscript) calculated from simulations performed for the complex or the ligands (superscript).
High Resolution Image
Download MS PowerPoint Slide
Figure 6
Figure 6. Correlation between TIES-predicted relative binding affinities and experimental data. The black line is the correlation line, while the dotted lines (x = 0 and y = 0) create four quadrants. Ten out of the 14 data points are in quadrants I (x > 0 and y > 0) and III (x < 0 and y < 0), meaning that the calculated binding free energy differences have the same sign as those from the experimental data.
High Resolution Image
Download MS PowerPoint Slide
Figure 7
Figure 7. ESMACS simulations of (a) 1, (b) 22, and (c) 4 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 frames. Key residues E590 and M592 (for clarity, side-chain atoms are not shown) at the TrkA hinge region are highlighted. Hydrogen bonds are shown as dashed lines. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. For reasons of clarity, the N-lobe of the protein has been removed.
High Resolution Image
Download MS PowerPoint Slide
Figure 8
Figure 8. ESMACS simulations of (a) 12, (b) 17, (c) 23, and (d) 24 bound to TrkA. Representative conformations are displayed using the final conformations of 4 ns production runs of one replica from each 25-member ensemble. The protein is shown by the surface representation and the ligand by the ball-and-stick representation with hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. The surfaces of the R2-position groups are shown in the wireframe representation.
High Resolution Image
Download MS PowerPoint Slide
Figure 9
Figure 9. ESMACS simulations of (a) 1, (b) 3, and (c) 6 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 frames. The key TrkA protein residues D668, K544, and E560 are highlighted. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. Hydrogen bonds between the linker group carbonyl oxygen of 1, 3, and 6 and the backbone −NH group of D668 are shown as dashed lines.
High Resolution Image
Download MS PowerPoint Slide
Figure 10
Figure 10. ESMACS simulations of (a) 7 and (b) 8 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 frames. The key TrkA protein residues D668, E560, and L564 are highlighted. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. The unfavorable electrostatic interactions between the methoxy group of 8 and the carbonyl group of D668 and between the methoxy group and the carboxylate group of E560 are highlighted by pink arrows.
High Resolution Image
Download MS PowerPoint Slide
Figure 11
Figure 11. ESMACS simulations of (a) 4 and (b) 16 bound to TrkA. Final conformations of 4 ns production runs from all 25 replicas are overlapped and smoothed by averaging over 10 conformations. The key TrkA protein residue D668 is highlighted. The protein is shown in cyan cartoon, and ligand atoms are colored by element: hydrogen in white, carbon in cyan, oxygen in red, and nitrogen in blue. The hydrogen bond between the linker-group carbonyl oxygen of 4 and the backbone −NH group of D668 is shown as a dashed line. As can be seen in (b), switching of the N–H and C═O groups prevents the formation of a hydrogen bond between the carbonyl oxygen of 16 and the backbone −NH group of D668.
High Resolution Image
Download MS PowerPoint Slide
References
ARTICLE SECTIONS
Jump To
This article references 33 other publications.
1. 1
  Sherborne, B.; Shanmugasundaram, V.; Cheng, A. C.; Christ, C. D.; DesJarlais, R. L.; Duca, J. S.; Lewis, R. A.; Loughney, D. A.; Manas, E. S.; McGaughey, G. B.; Peishoff, C. E.; van Vlijmen, H. Collaborating to Improve the Use of Free-Energy and Other Quantitative Methods in Drug Discovery J. Comput.-Aided Mol. Des. 2016, 30, 1139– 1141 DOI: 10.1007/s10822-016-9996-y
  Google Scholar
  1
  Collaborating to improve the use of free-energy and other quantitative methods in drug discovery
  Sherborne, Bradley; Shanmugasundaram, Veerabahu; Cheng, Alan C.; Christ, Clara D.; DesJarlais, Renee L.; Duca, Jose S.; Lewis, Richard A.; Loughney, Deborah A.; Manas, Eric S.; McGaughey, Georgia B.; Peishoff, Catherine E.; van Vlijmen, Herman
  Journal of Computer-Aided Molecular Design (2016), 30 (12), 1139-1141CODEN: JCADEQ; ISSN:0920-654X. (Springer)
  In May and August, 2016, several pharmaceutical companies convened to discuss and compare experiences with Free Energy Perturbation (FEP). This unusual synchronization of interest was prompted by Schrodinger's FEP+ implementation and offered the opportunity to share fresh studies with FEP and enable broader discussions on the topic. This article summarizes key conclusions of the meetings, including a path forward of actions for this group to aid the accelerated evaluation, application and development of free energy and related quant., structure-based design methods.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XitFClsLfN&md5=c00386f775bd826c6b845bbbef4a975d
2. 2
  Skerratt, S. E.; Andrews, M.; Bagal, S. K.; Bilsland, J.; Brown, D.; Bungay, P. J.; Cole, S.; Gibson, K. R.; Jones, R.; Morao, I.; Nedderman, A.; Omoto, K.; Robinson, C.; Ryckmans, T.; Skinner, K.; Stupple, P.; Waldron, G. The Discovery of a Potent, Selective, and Peripherally Restricted Pan-Trk Inhibitor (PF-06273340) for the Treatment of Pain J. Med. Chem. 2016, 59, 10084– 10099 DOI: 10.1021/acs.jmedchem.6b00850
  Google Scholar
  2
  The Discovery of a Potent, Selective, and Peripherally Restricted Pan-Trk Inhibitor (PF-06273340) for the Treatment of Pain
  Skerratt, Sarah E.; Andrews, Mark; Bagal, Sharan K.; Bilsland, James; Brown, David; Bungay, Peter J.; Cole, Susan; Gibson, Karl R.; Jones, Russell; Morao, Inaki; Nedderman, Angus; Omoto, Kiyoyuki; Robinson, Colin; Ryckmans, Thomas; Skinner, Kimberly; Stupple, Paul; Waldron, Gareth
  Journal of Medicinal Chemistry (2016), 59 (22), 10084-10099CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  The neurotrophin family of growth factors, comprised of nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin 3 (NT3) and neurotrophin 4 (NT4), is implicated in the physiol. of chronic pain. Given the clin. efficacy of anti-NGF monoclonal antibody (mAb) therapies, there is significant interest in the development of small mol. modulators of neurotrophin activity. Neurotrophins signal through the tropomyosin related kinase (Trk) family of tyrosine kinase receptors, hence Trk kinase inhibition represents a potentially "druggable" point of intervention. To deliver the safety profile required for chronic, non-life threatening pain indications, highly kinase-selective Trk inhibitors with minimal brain availability are sought. Herein the authors describe how the use of SBDD, 2D QSAR models and Matched Mol. Pair data in compd. design enabled the delivery of the highly potent, kinase-selective and peripherally restricted clin. candidate PF-06273340.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhslWjsbfL&md5=f8c904a92c1716062c217fefe2374ce6
3. 3
  Sadiq, S. K.; Wright, D.; Watson, S. J.; Zasada, S. J.; Stoica, I.; Coveney, P. V. Automated Molecular Simulation Based Binding Affinity Calculator for Ligand-Bound HIV-1 Proteases J. Chem. Inf. Model. 2008, 48, 1909– 1919 DOI: 10.1021/ci8000937
  Google Scholar
  3
  Automated Molecular Simulation Based Binding Affinity Calculator for Ligand-Bound HIV-1 Proteases
  Sadiq, S. Kashif; Wright, David; Watson, Simon J.; Zasada, Stefan J.; Stoica, Ileana; Coveney, Peter V.
  Journal of Chemical Information and Modeling (2008), 48 (9), 1909-1919CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  The successful application of high throughput mol. simulations to det. biochem. properties would be of great importance to the biomedical community if such simulations could be turned around in a clin. relevant timescale. An important example is the detn. of antiretroviral inhibitor efficacy against varying strains of HIV through calcn. of drug-protein binding affinities. We describe the Binding Affinity Calculator (BAC), a tool for the automated calcn. of HIV-1 protease-ligand binding affinities. The tool employs fully atomistic mol. simulations alongside the well established mol. mechanics Poisson-Boltzmann solvent accessible surface area (MMPBSA) free energy methodol. to enable the calcn. of the binding free energy of several ligand-protease complexes, including all nine FDA approved inhibitors of HIV-1 protease and seven of the natural substrates cleaved by the protease. This enables the efficacy of these inhibitors to be ranked across several mutant strains of the protease relative to the wildtype. BAC is a tool that utilizes the power provided by a computational grid to automate all of the stages required to compute free energies of binding: model prepn., equilibration, simulation, postprocessing, and data-marshaling around the generally widely distributed compute resources utilized. Such automation enables the mol. dynamics methodol. to be used in a high throughput manner not achievable by manual methods. This paper describes the architecture and workflow management of BAC and the function of each of its components. Given adequate compute resources, BAC can yield quant. information regarding drug resistance at the mol. level within 96 h. Such a timescale is of direct clin. relevance and can assist in decision support for the assessment of patient-specific optimal drug treatment and the subsequent response to therapy for any given genotype.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXhtVShtLjN&md5=c616ac9330dc1f4cb9808d68fabc7e4c
4. 4
  Groen, D.; Bhati, A. P.; Suter, J.; Hetherington, J.; Zasada, S. J.; Coveney, P. V. FabSim: Facilitating Computational Research through Automation on Large-Scale and Distributed E-Infrastructures Comput. Phys. Commun. 2016, 207, 375– 385 DOI: 10.1016/j.cpc.2016.05.020
  Google Scholar
  4
  FabSim: Facilitating computational research through automation on large-scale and distributed e-infrastructures
  Groen, Derek; Bhati, Agastya P.; Suter, James; Hetherington, James; Zasada, Stefan J.; Coveney, Peter V.
  Computer Physics Communications (2016), 207 (), 375-385CODEN: CPHCBZ; ISSN:0010-4655. (Elsevier B.V.)
  We present FabSim, a toolkit developed to simplify a range of computational tasks for researchers in diverse disciplines. FabSim is flexible, adaptable, and allows users to perform a wide range of tasks with ease. It also provides a systematic way to automate the use of resources, including HPC and distributed machines, and to make tasks easier to repeat by recording contextual information. To demonstrate this, we present three use cases where FabSim has enhanced our research productivity. These include simulating cerebrovascular bloodflow, modeling clay-polymer nanocomposites across multiple scales, and calcg. ligand-protein binding affinities.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28Xptl2nu7s%253D&md5=4b7e50658ec8307f7fa5fcd16747cab6
5. 5
  Norman, B. H.; McDermott, J. S. Targeting the Nerve Growth Factor (NGF) Pathway in Drug Discovery. Potential Applications to New Therapies for Chronic Pain J. Med. Chem. 2017, 60, 66– 88 DOI: 10.1021/acs.jmedchem.6b00964
  Google Scholar
  5
  Targeting the Nerve Growth Factor (NGF) Pathway in Drug Discovery. Potential Applications to New Therapies for Chronic Pain
  Norman, Bryan H.; McDermott, Jeff S.
  Journal of Medicinal Chemistry (2017), 60 (1), 66-88CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  A review. The neurotrophin nerve growth factor (NGF) has been implicated as a key mediator of chronic pain. NGF binds the tropomyosin receptor kinase A (TrkA) and p75, resulting in the activation of downstream signaling pathways that have been linked to pro-nociception. While anti-NGF antibodies have demonstrated analgesia both preclinically and in patients, the mechanism of action of these agents remains unclear. The authors describe ligands targeting NGF, its receptors and downstream/related targets. This perspective highlights large and small mol. approaches to targeting the NGF-TrkA pathway both extra- and intracellularly. In addn., the authors present a strategic framework for future drug discovery efforts in this pathway beyond the targeting of NGF or its receptors. While existing tools have greatly informed NGF-mediated signaling, ongoing and future pathway research may help focus new drug discovery efforts on key novel targets and mechanisms. This may result in highly differentiated therapeutics with greater efficacy and/or improved safety profiles.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhslSqtLjF&md5=606097686779a102db9e6e54f8252a40
6. 6
  Hefti, F. F.; Rosenthal, A.; Walicke, P. A.; Wyatt, S.; Vergara, G.; Shelton, D. L.; Davies, A. M. Novel Class of Pain Drugs Based on Antagonism of NGF Trends Pharmacol. Sci. 2006, 27, 85– 91 DOI: 10.1016/j.tips.2005.12.001
  Google Scholar
  6
  Novel class of pain drugs based on antagonism of NGF
  Hefti, Franz F.; Rosenthal, Arnon; Walicke, Patricia A.; Wyatt, Sean; Vergara, German; Shelton, David L.; Davies, Alun M.
  Trends in Pharmacological Sciences (2006), 27 (2), 85-91CODEN: TPHSDY; ISSN:0165-6147. (Elsevier Ltd.)
  A review. Nerve growth factor (NGF) was identified originally as a survival factor for sensory and sympathetic neurons in the developing nervous system. In adults, NGF is not required for survival but it has a crucial role in the generation of pain and hyperalgesia in several acute and chronic pain states. The expression of NGF is high in injured and inflamed tissues, and activation of the NGF receptor tyrosine kinase trkA on nociceptive neurons triggers and potentiates pain signaling by multiple mechanisms. Inhibition of NGF function and signaling blocks pain sensation as effectively as cyclooxygenase inhibitors and opiates in rodent models of pain. Several pharmaceutical companies have active drug-discovery and development programs that are based on a variety of approaches to antagonize NGF, including NGF capture', blocking the binding of NGF to trkA and inhibiting trkA signaling. NGF antagonism is expected to be a highly effective therapeutic approach in many pain states, and to be free of the adverse effects of traditional analgesic drugs.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD28XhtV2lsLo%253D&md5=3cfe9e06069c3b6b9aa9d96b7411d199
7. 7
  Mantyh, P. W.; Koltzenburg, M.; Mendell, L. M.; Tive, L.; Shelton, D. L. Antagonism of Nerve Growth Factor-TrkA Signaling and the Relief of Pain Anesthesiology 2011, 115, 189– 204 DOI: 10.1097/ALN.0b013e31821b1ac5
  Google Scholar
  There is no corresponding record for this reference.
8. 8
  Wan, S.; Knapp, B.; Wright, D. W.; Deane, C. M.; Coveney, P. V. Rapid, Precise, and Reproducible Prediction of Peptide-MHC Binding Affinities from Molecular Dynamics That Correlate Well with Experiment J. Chem. Theory Comput. 2015, 11, 3346– 3356 DOI: 10.1021/acs.jctc.5b00179
  Google Scholar
  8
  Rapid, precise, and reproducible prediction of peptide-MHC binding affinities from molecular dynamics that correlate well with experiment
  Wan, Shunzhou; Knapp, Bernhard; Wright, David W.; Deane, Charlotte M.; Coveney, Peter V.
  Journal of Chemical Theory and Computation (2015), 11 (7), 3346-3356CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  The presentation of potentially pathogenic peptides by major histocompatibility complex (MHC) mols. is one of the most important processes in adaptive immune defense. Prediction of peptide-MHC (pMHC) binding affinities is therefore a principal objective of theor. immunol. Machine learning techniques achieve good results if substantial exptl. training data are available. Approaches based on structural information become necessary if sufficiently similar training data are unavailable for a specific MHC allele, although they have often been deemed to lack accuracy. In this study, we use a free energy method to rank the binding affinities of 12 diverse peptides bound by a class I MHC mol. HLA-A*02:01. The method is based on enhanced sampling of mol. dynamics calcns. in combination with a continuum solvent approxn. and includes ests. of the configurational entropy based on either a one or a three trajectory protocol. It produces precise and reproducible free energy ests. which correlate well with exptl. measurements. If the results are combined with an amino acid hydrophobicity scale, then an extremely good ranking of peptide binding affinities emerges. Our approach is rapid, robust, and applicable to a wide range of ligand-receptor interactions without further adjustment.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhtVaksLjN&md5=860aa35a6e8e4013e2b0c7f0ae99170f
9. 9
  Kollman, P. A.; Massova, I.; Reyes, C.; Kuhn, B.; Huo, S.; Chong, L.; Lee, M.; Lee, T.; Duan, Y.; Wang, W.; Donini, O.; Cieplak, P.; Srinivasan, J.; Case, D. A.; Cheatham, T. E., III Calculating Structures and Free Energies of Complex Molecules: Combining Molecular Mechanics and Continuum Models Acc. Chem. Res. 2000, 33, 889– 897 DOI: 10.1021/ar000033j
  Google Scholar
  9
  Calculating Structures and Free Energies of Complex Molecules: Combining Molecular Mechanics and Continuum Models
  Kollman, Peter A.; Massova, Irina; Reyes, Carolina; Kuhn, Bernd; Huo, Shuanghong; Chong, Lillian; Lee, Matthew; Lee, Taisung; Duan, Yong; Wang, Wei; Donini, Oreola; Cieplak, Piotr; Srinivasan, Jaysharee; Case, David A.; Cheatham, Thomas E., III
  Accounts of Chemical Research (2000), 33 (12), 889-897CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
  A review, with 63 refs. A historical perspective on the application of mol. dynamics (MD) to biol. macromols. is presented. Recent developments combining state-of-the-art force fields with continuum solvation calcns. have allowed us to reach the fourth era of MD applications in which one can often derive both accurate structure and accurate relative free energies from mol. dynamics trajectories. We illustrate such applications on nucleic acid duplexes, RNA hairpins, protein folding trajectories, and protein-ligand, protein-protein, and protein-nucleic acid interactions.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3cXmvFGiu7g%253D&md5=8436ee610ae145894428db1a1deff73c
10. 10
  Coveney, P. V.; Wan, S. On the Calculation of Equilibrium Thermodynamic Properties from Molecular Dynamics Phys. Chem. Chem. Phys. 2016, 18, 30236– 30240 DOI: 10.1039/C6CP02349E
  Google Scholar
  10
  On the calculation of equilibrium thermodynamic properties from molecular dynamics
  Coveney, Peter V.; Wan, Shunzhou
  Physical Chemistry Chemical Physics (2016), 18 (44), 30236-30240CODEN: PPCPFQ; ISSN:1463-9076. (Royal Society of Chemistry)
  The purpose of statistical mechanics is to provide a route to the calcn. of macroscopic properties of matter from their constituent microscopic components. It is well known that the macrostates emerge as ensemble avs. of microstates. However, this is more often stated than implemented in computer simulation studies. Here we consider foundational aspects of statistical mechanics which are overlooked in most textbooks and research articles that purport to compute macroscopic behavior from microscopic descriptions based on classical mechanics and show how due attention to these issues leads in directions which have not been widely appreciated in the field of mol. dynamics simulation.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XntVyisr8%253D&md5=f7950af8255cf494a117d46bc1b5191e
11. 11
  Miller, B. R., 3rd; McGee, T. D., Jr.; Swails, J. M.; Homeyer, N.; Gohlke, H.; Roitberg, A. E. Mmpbsa.Py: An Efficient Program for End-State Free Energy Calculations J. Chem. Theory Comput. 2012, 8, 3314– 3321 DOI: 10.1021/ct300418h
  Google Scholar
  11
  MMPBSA.py: An Efficient Program for End-State Free Energy Calculations
  Miller, Bill R., III; McGee, T. Dwight, Jr.; Swails, Jason M.; Homeyer, Nadine; Gohlke, Holger; Roitberg, Adrian E.
  Journal of Chemical Theory and Computation (2012), 8 (9), 3314-3321CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  MM-PBSA is a post-processing end-state method to calc. free energies of mols. in soln. MMPBSA.py is a program written in Python for streamlining end-state free energy calcns. using ensembles derived from mol. dynamics (MD) or Monte Carlo (MC) simulations. Several implicit solvation models are available with MMPBSA.py, including the Poisson-Boltzmann Model, the Generalized Born Model, and the Ref. Interaction Site Model. Vibrational frequencies may be calcd. using normal mode or quasi-harmonic anal. to approx. the solute entropy. Specific interactions can also be dissected using free energy decompn. or alanine scanning. A parallel implementation significantly speeds up the calcn. by dividing frames evenly across available processors. MMPBSA.py is an efficient, user-friendly program with the flexibility to accommodate the needs of users performing end-state free energy calcns. The source code can be downloaded at http://ambermd.org/ with AmberTools, released under the GNU General Public License.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhtV2gtrzP&md5=cc4148bd8f70c7cad94fd3ec6f580e52
12. 12
  Case, D. A.; Cheatham, T. E., III; Darden, T.; Gohlke, H.; Luo, R.; Merz, K. M., Jr.; Onufriev, A.; Simmerling, C.; Wang, B.; Woods, R. J. The Amber Biomolecular Simulation Programs J. Comput. Chem. 2005, 26, 1668– 1688 DOI: 10.1002/jcc.20290
  Google Scholar
  12
  The amber biomolecular simulation programs
  Case, David A.; Cheatham, Thomas E., III; Darden, Tom; Gohlke, Holger; Luo, Ray; Merz, Kenneth M., Jr.; Onufriev, Alexey; Simmerling, Carlos; Wang, Bing; Woods, Robert J.
  Journal of Computational Chemistry (2005), 26 (16), 1668-1688CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
  The authors describe the development, current features, and some directions for future development of the Amber package of computer programs. This package evolved from a program that was constructed in the late 1970s to do Assisted Model Building with Energy Refinement, and now contains a group of programs embodying a no. of powerful tools of modern computational chem., focused on mol. dynamics and free energy calcns. of proteins, nucleic acids, and carbohydrates.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXht1SlsbbM&md5=93be29ff894bab96c783d24e9886c7d0
13. 13
  Bhati, A. P.; Wan, S.; Wright, D. W.; Coveney, P. V. Rapid, Accurate, Precise, and Reliable Relative Free Energy Prediction Using Ensemble Based Thermodynamic Integration J. Chem. Theory Comput. 2017, 13, 210– 222 DOI: 10.1021/acs.jctc.6b00979
  Google Scholar
  13
  Rapid, Accurate, Precise, and Reliable Relative Free Energy Prediction Using Ensemble Based Thermodynamic Integration
  Bhati, Agastya P.; Wan, Shunzhou; Wright, David W.; Coveney, Peter V.
  Journal of Chemical Theory and Computation (2017), 13 (1), 210-222CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  The accurate prediction of the binding affinities of ligands to proteins is a major goal in drug discovery and personalized medicine. The time taken to make such predictions is of similar importance to their accuracy, precision and reliability. In the last few years, an ensemble based mol. dynamics approach has been proposed that provides a route to reliable predictions of free energies based on the mol. mechanics Poisson-Boltzmann surface area method which meets the requirements of accuracy, precision and reliability. Here, we describe an equiv. methodol. based on thermodn. integration to substantially improve the accuracy, precision and reliability of calcd. relative binding free energies. We report the performance of the method when applied to a diverse set of protein targets and ligands. The results are in very good agreement with exptl. data (90% of calcns. agree to within 1 kcal/mol) while the method is reproducible by construction. Statistical uncertainties of the order of 0.5 kcal/mol or less are achieved. We present a systematic account of how the uncertainty in the predictions may be estd.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XitVGntrrJ&md5=510b70188112a3578030e291ce19b127
14. 14
  Wright, D. W.; Hall, B. A.; Kenway, O. A.; Jha, S.; Coveney, P. V. Computing Clinically Relevant Binding Free Energies of HIV-1 Protease Inhibitors J. Chem. Theory Comput. 2014, 10, 1228– 1241 DOI: 10.1021/ct4007037
  Google Scholar
  14
  Computing Clinically Relevant Binding Free Energies of HIV-1 Protease Inhibitors
  Wright, David W.; Hall, Benjamin A.; Kenway, Owain A.; Jha, Shantenu; Coveney, Peter V.
  Journal of Chemical Theory and Computation (2014), 10 (3), 1228-1241CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  The use of mol. simulation to est. the strength of macromol. binding free energies is becoming increasingly widespread, with goals ranging from lead optimization and enrichment in drug discovery to personalizing or stratifying treatment regimes. To realize the potential of such approaches to predict new results, not merely to explain previous exptl. findings, it is necessary that the methods used are reliable and accurate, and that their limitations are thoroughly understood. However, the computational cost of atomistic simulation techniques such as mol. dynamics (MD) has meant that until recently little work has focused on validating and verifying the available free energy methodologies, with the consequence that many of the results published in the literature are not reproducible. Here, we present a detailed anal. of two of the most popular approx. methods for calcg. binding free energies from mol. simulations, mol. mechanics Poisson-Boltzmann surface area (MMPBSA) and mol. mechanics generalized Born surface area (MMGBSA), applied to the nine FDA-approved HIV-1 protease inhibitors. Our results show that the values obtained from replica simulations of the same protease-drug complex, differing only in initially assigned atom velocities, can vary by as much as 10 kcal mol-1, which is greater than the difference between the best and worst binding inhibitors under investigation. Despite this, anal. of ensembles of simulations producing 50 trajectories of 4 ns duration leads to well converged free energy ests. For seven inhibitors, we find that with correctly converged normal mode ests. of the configurational entropy, we can correctly distinguish inhibitors in agreement with exptl. data for both the MMPBSA and MMGBSA methods and thus have the ability to rank the efficacy of binding of this selection of drugs to the protease (no account is made for free energy penalties assocd. with protein distortion leading to the over estn. of the binding strength of the two largest inhibitors ritonavir and atazanavir). We obtain improved rankings and ests. of the relative binding strengths of the drugs by using a novel combination of MMPBSA/MMGBSA with normal mode entropy ests. and the free energy of assocn. calcd. directly from simulation trajectories. Our work provides a thorough assessment of what is required to produce converged and hence reliable free energies for protein-ligand binding.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXht12ltrk%253D&md5=56bc7e7c8f6bbdd694bfc76c20dcec63
15. 15
  Swanson, J. M.; Henchman, R. H.; McCammon, J. A. Revisiting Free Energy Calculations: A Theoretical Connection to MM/PBSA and Direct Calculation of the Association Free Energy Biophys. J. 2004, 86, 67– 74 DOI: 10.1016/S0006-3495(04)74084-9
  Google Scholar
  15
  Revisiting free energy calculations: A theoretical connection to MM/PBSA and direct calculation of the association free energy
  Swanson, Jessica M. J.; Henchman, Richard H.; McCammon, J. Andrew
  Biophysical Journal (2004), 86 (1, Pt. 1), 67-74CODEN: BIOJAU; ISSN:0006-3495. (Biophysical Society)
  The prediction of abs. ligand-receptor binding affinities is essential in a wide range of biophys. queries, from the study of protein-protein interactions to structure-based drug design. End-point free energy methods, such as the Mol. Mechanics Poisson-Boltzmann Surface Area (MM/PBSA) model, have received much attention and widespread application in recent literature. These methods benefit from computational efficiency as only the initial and final states of the system are evaluated, yet there remains a need for strengthening their theor. foundation. Here a clear connection between statistical thermodn. and end-point free energy models is presented. The importance of the assocn. free energy, arising from one mol.'s loss of translational and rotational freedom from the std. state concn., is addressed. A novel method for calcg. this quantity directly from a mol. dynamics simulation is described. The challenges of accounting for changes in the protein conformation and its fluctuations from sep. simulations are discussed. A simple first-order approxn. of the configuration integral is presented to lay the groundwork for future efforts. This model has been applied to FKBP12, a small immunophilin that has been widely studied in the drug industry for its potential immunosuppressive and neuroregenerative effects.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXlsV2hug%253D%253D&md5=720b2ec6ef0462edcda34f7be768bc44
16. 16
  Chodera, J. D.; Mobley, D. L.; Shirts, M. R.; Dixon, R. W.; Branson, K.; Pande, V. S. Alchemical Free Energy Methods for Drug Discovery: Progress and Challenges Curr. Opin. Struct. Biol. 2011, 21, 150– 160 DOI: 10.1016/j.sbi.2011.01.011
  Google Scholar
  16
  Alchemical free energy methods for drug discovery: Progress and challenges
  Chodera, John D.; Mobley, David L.; Shirts, Michael R.; Dixon, Richard W.; Branson, Kim; Pande, Vijay S.
  Current Opinion in Structural Biology (2011), 21 (2), 150-160CODEN: COSBEF; ISSN:0959-440X. (Elsevier Ltd.)
  A review. Improved rational drug design methods are needed to lower the cost and increase the success rate of drug discovery and development. Alchem. binding free energy calcns., one potential tool for rational design, have progressed rapidly over the past decade, but still fall short of providing robust tools for pharmaceutical engineering. Recent studies, esp. on model receptor systems, have clarified many of the challenges that must be overcome for robust predictions of binding affinity to be useful in rational design. In this review, inspired by a recent joint academic/industry meeting organized by the authors, we discuss these challenges and suggest a no. of promising approaches for overcoming them.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXkt1Kisrk%253D&md5=fa17c0982d921cd1c32cd095fd34420e
17. 17
  Wang, L.; Wu, Y.; Deng, Y.; Kim, B.; Pierce, L.; Krilov, G.; Lupyan, D.; Robinson, S.; Dahlgren, M. K.; Greenwood, J.; Romero, D. L.; Masse, C.; Knight, J. L.; Steinbrecher, T.; Beuming, T.; Damm, W.; Harder, E.; Sherman, W.; Brewer, M.; Wester, R.; Murcko, M.; Frye, L.; Farid, R.; Lin, T.; Mobley, D. L.; Jorgensen, W. L.; Berne, B. J.; Friesner, R. A.; Abel, R. Accurate and Reliable Prediction of Relative Ligand Binding Potency in Prospective Drug Discovery by Way of a Modern Free-Energy Calculation Protocol and Force Field J. Am. Chem. Soc. 2015, 137, 2695– 2703 DOI: 10.1021/ja512751q
  Google Scholar
  17
  Accurate and Reliable Prediction of Relative Ligand Binding Potency in Prospective Drug Discovery by Way of a Modern Free-Energy Calculation Protocol and Force Field
  Wang, Lingle; Wu, Yujie; Deng, Yuqing; Kim, Byungchan; Pierce, Levi; Krilov, Goran; Lupyan, Dmitry; Robinson, Shaughnessy; Dahlgren, Markus K.; Greenwood, Jeremy; Romero, Donna L.; Masse, Craig; Knight, Jennifer L.; Steinbrecher, Thomas; Beuming, Thijs; Damm, Wolfgang; Harder, Ed; Sherman, Woody; Brewer, Mark; Wester, Ron; Murcko, Mark; Frye, Leah; Farid, Ramy; Lin, Teng; Mobley, David L.; Jorgensen, William L.; Berne, Bruce J.; Friesner, Richard A.; Abel, Robert
  Journal of the American Chemical Society (2015), 137 (7), 2695-2703CODEN: JACSAT; ISSN:0002-7863. (American Chemical Society)
  Designing tight-binding ligands is a primary objective of small-mol. drug discovery. Over the past few decades, free-energy calcns. have benefited from improved force fields and sampling algorithms, as well as the advent of low-cost parallel computing. However, it has proven to be challenging to reliably achieve the level of accuracy that would be needed to guide lead optimization (∼5× in binding affinity) for a wide range of ligands and protein targets. Not surprisingly, widespread com. application of free-energy simulations has been limited due to the lack of large-scale validation coupled with the tech. challenges traditionally assocd. with running these types of calcns. Here, we report an approach that achieves an unprecedented level of accuracy across a broad range of target classes and ligands, with retrospective results encompassing 200 ligands and a wide variety of chem. perturbations, many of which involve significant changes in ligand chem. structures. In addn., we have applied the method in prospective drug discovery projects and found a significant improvement in the quality of the compds. synthesized that have been predicted to be potent. Compds. predicted to be potent by this approach have a substantial redn. in false positives relative to compds. synthesized on the basis of other computational or medicinal chem. approaches. Furthermore, the results are consistent with those obtained from our retrospective studies, demonstrating the robustness and broad range of applicability of this approach, which can be used to drive decisions in lead optimization.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhsF2iuro%253D&md5=37a4f4a6c085f47ed531342643b6c33b
18. 18
  Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. J.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.Gaussian 03; Gaussian, Inc.: Wallingford, CT, 2004.
  Google Scholar
  There is no corresponding record for this reference.
19. 19
  Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A. Development and Testing of a General Amber Force Field J. Comput. Chem. 2004, 25, 1157– 1174 DOI: 10.1002/jcc.20035
  Google Scholar
  19
  Development and testing of a general Amber force field
  Wang, Junmei; Wolf, Romain M.; Caldwell, James W.; Kollman, Peter A.; Case, David A.
  Journal of Computational Chemistry (2004), 25 (9), 1157-1174CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
  We describe here a general Amber force field (GAFF) for org. mols. GAFF is designed to be compatible with existing Amber force fields for proteins and nucleic acids, and has parameters for most org. and pharmaceutical mols. that are composed of H, C, N, O, S, P, and halogens. It uses a simple functional form and a limited no. of atom types, but incorporates both empirical and heuristic models to est. force consts. and partial at. charges. The performance of GAFF in test cases is encouraging. In test I, 74 crystallog. structures were compared to GAFF minimized structures, with a root-mean-square displacement of 0.26 Å, which is comparable to that of the Tripos 5.2 force field (0.25 Å) and better than those of MMFF 94 and CHARMm (0.47 and 0.44 Å, resp.). In test II, gas phase minimizations were performed on 22 nucleic acid base pairs, and the minimized structures and intermol. energies were compared to MP2/6-31G* results. The RMS of displacements and relative energies were 0.25 Å and 1.2 kcal/mol, resp. These data are comparable to results from Parm99/RESP (0.16 Å and 1.18 kcal/mol, resp.), which were parameterized to these base pairs. Test III looked at the relative energies of 71 conformational pairs that were used in development of the Parm99 force field. The RMS error in relative energies (compared to expt.) is about 0.5 kcal/mol. GAFF can be applied to wide range of mols. in an automatic fashion, making it suitable for rational drug design and database searching.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXksFakurc%253D&md5=2992017a8cf51f89290ae2562403b115
20. 20
  Lindorff-Larsen, K.; Piana, S.; Palmo, K.; Maragakis, P.; Klepeis, J. L.; Dror, R. O.; Shaw, D. E. Improved Side-Chain Torsion Potentials for the Amber FF99SB Protein Force Field Proteins: Struct., Funct., Bioinf. 2010, 78, 1950– 1958 DOI: 10.1002/prot.22711
  Google Scholar
  20
  Improved side-chain torsion potentials for the Amber ff99SB protein force field
  Lindorff-Larsen, Kresten; Piana, Stefano; Palmo, Kim; Maragakis, Paul; Klepeis, John L.; Dror, Ron O.; Shaw, David E.
  Proteins: Structure, Function, and Bioinformatics (2010), 78 (8), 1950-1958CODEN: PSFBAF ISSN:. (Wiley-Liss, Inc.)
  Recent advances in hardware and software have enabled increasingly long mol. dynamics (MD) simulations of biomols., exposing certain limitations in the accuracy of the force fields used for such simulations and spurring efforts to refine these force fields. Recent modifications to the Amber and CHARMM protein force fields, for example, have improved the backbone torsion potentials, remedying deficiencies in earlier versions. Here, the authors further advance simulation accuracy by improving the amino acid side-chain torsion potentials of the Amber ff99SB force field. First, the authors used simulations of model alpha-helical systems to identify the four residue types whose rotamer distribution differed the most from expectations based on Protein Data Bank statistics. Second, the authors optimized the side-chain torsion potentials of these residues to match new, high-level quantum-mech. calcns. Finally, the authors used microsecond-timescale MD simulations in explicit solvent to validate the resulting force field against a large set of exptl. NMR measurements that directly probe side-chain conformations. The new force field, which the authors have termed Amber ff99SB-ILDN, exhibits considerably better agreement with the NMR data. Proteins 2010. © 2010 Wiley-Liss, Inc.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXkvFegtLo%253D&md5=447a9004026e2b93f0f7beff165daa09
21. 21
  Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable Molecular Dynamics with NAMD J. Comput. Chem. 2005, 26, 1781– 1802 DOI: 10.1002/jcc.20289
  Google Scholar
  21
  Scalable molecular dynamics with NAMD
  Phillips, James C.; Braun, Rosemary; Wang, Wei; Gumbart, James; Tajkhorshid, Emad; Villa, Elizabeth; Chipot, Christophe; Skeel, Robert D.; Kale, Laxmikant; Schulten, Klaus
  Journal of Computational Chemistry (2005), 26 (16), 1781-1802CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
  NAMD is a parallel mol. dynamics code designed for high-performance simulation of large biomol. systems. NAMD scales to hundreds of processors on high-end parallel platforms, as well as tens of processors on low-cost commodity clusters, and also runs on individual desktop and laptop computers. NAMD works with AMBER and CHARMM potential functions, parameters, and file formats. This article, directed to novices as well as experts, first introduces concepts and methods used in the NAMD program, describing the classical mol. dynamics force field, equations of motion, and integration methods along with the efficient electrostatics evaluation algorithms employed and temp. and pressure controls used. Features for steering the simulation across barriers and for calcg. both alchem. and conformational free energy differences are presented. The motivations for and a roadmap to the internal design of NAMD, implemented in C++ and based on Charm++ parallel objects, are outlined. The factors affecting the serial and parallel performance of a simulation are discussed. Finally, typical NAMD use is illustrated with representative applications to a small, a medium, and a large biomol. system, highlighting particular features of NAMD, for example, the Tcl scripting language. The article also provides a list of the key features of NAMD and discusses the benefits of combining NAMD with the mol. graphics/sequence anal. software VMD and the grid computing/collab. software BioCoRE. NAMD is distributed free of charge with source code at www.ks.uiuc.edu.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXht1SlsbbJ&md5=189051128443b547f4300a1b8fb0e034
22. 22
  Highfield, R. Supercomputer bid to create the first truly personalised medicine. https://blog.sciencemuseum.org.uk/supercomputer-bid-to-create-the-first-truly-personalised-medicine/ (accessed Feb 22, 2017).
  Google Scholar
  There is no corresponding record for this reference.
23. 23
  Fiser, A.; Sali, A. Modloop: Automated Modeling of Loops in Protein Structures Bioinformatics 2003, 19, 2500– 2501 DOI: 10.1093/bioinformatics/btg362
  Google Scholar
  23
  ModLoop: automated modeling of loops in protein structures
  Fiser, Andras; Sali, Andrej
  Bioinformatics (2003), 19 (18), 2500-2501CODEN: BOINFP; ISSN:1367-4803. (Oxford University Press)
  Summary: ModLoop is a web server for automated modeling of loops in protein structures. The input is the at. coordinates of the protein structure in the Protein Data Bank format, and the specification of the starting and ending residues of one or more segments to be modeled, contg. no more than 20 residues in total. The output is the coordinates of the non-hydrogen atoms in the modeled segments. A user provides the input to the server via a simple web interface, and receives the output by e-mail. The server relies on the loop modeling routine in MODELLER that predicts the loop conformations by satisfaction of spatial restraints, without relying on a database of known protein structures. For a rapid response, ModLoop runs on a cluster of Linux PC computers.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXpvVSisLk%253D&md5=4b0fdaca0a412715682496883ab2019c
24. 24
  Allen, W. J.; Balius, T. E.; Mukherjee, S.; Brozell, S. R.; Moustakas, D. T.; Lang, P. T.; Case, D. A.; Kuntz, I. D.; Rizzo, R. C. Dock 6: Impact of New Features and Current Docking Performance J. Comput. Chem. 2015, 36, 1132– 1156 DOI: 10.1002/jcc.23905
  Google Scholar
  24
  DOCK 6: Impact of new features and current docking performance
  Allen, William J.; Balius, Trent E.; Mukherjee, Sudipto; Brozell, Scott R.; Moustakas, Demetri T.; Lang, P. Therese; Case, David A.; Kuntz, Irwin D.; Rizzo, Robert C.
  Journal of Computational Chemistry (2015), 36 (15), 1132-1156CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
  This manuscript presents the latest algorithmic and methodol. developments to the structure-based design program DOCK 6.7 focused on an updated internal energy function, new anchor selection control, enhanced minimization options, a footprint similarity scoring function, a symmetry-cor. root-mean-square deviation algorithm, a database filter, and docking forensic tools. An important strategy during development involved use of three orthogonal metrics for assessment and validation: pose reprodn. over a large database of 1043 protein-ligand complexes (SB2012 test set), cross-docking to 24 drug-target protein families, and database enrichment using large active and decoy datasets (Directory of Useful Decoys [DUD]-E test set) for five important proteins including HIV protease and IGF-1R. Relative to earlier versions, a key outcome of the work is a significant increase in pose reprodn. success in going from DOCK 4.0.2 (51.4%) → 5.4 (65.2%) → 6.7 (73.3%) as a result of significant decreases in failure arising from both sampling 24.1% → 13.6% → 9.1% and scoring 24.4% → 21.1% → 17.5%. Companion cross-docking and enrichment studies with the new version highlight other strengths and remaining areas for improvement, esp. for systems contg. metal ions. The source code for DOCK 6.7 is available for download and free for academic users at. © 2015 Wiley Periodicals, Inc.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXnt1Smtbs%253D&md5=8b5fc6cc7533f975e1e740acb3688807
25. 25
  Chodera, J. D.; Mobley, D. L. Entropy-Enthalpy Compensation: Role and Ramifications in Biomolecular Ligand Recognition and Design Annu. Rev. Biophys. 2013, 42, 121– 142 DOI: 10.1146/annurev-biophys-083012-130318
  Google Scholar
  25
  Entropy-enthalpy compensation: role and ramifications in biomolecular ligand recognition and design
  Chodera, John D.; Mobley, David L.
  Annual Review of Biophysics (2013), 42 (), 121-142CODEN: ARBNCV; ISSN:1936-122X. (Annual Reviews)
  A review. Recent calorimetric studies of interactions between small mols. and biomol. targets have generated renewed interest in the phenomenon of entropy-enthalpy compensation. In these studies, entropic and enthalpic contributions to binding are obsd. to vary substantially and in an opposing manner as the ligand or protein is modified, whereas the binding free energy varies little. In severe examples, engineered enthalpic gains can lead to completely compensating entropic penalties, frustrating ligand design. Here, we examine the evidence for compensation, as well as its potential origins, prevalence, severity, and ramifications for ligand engineering. We find the evidence for severe compensation to be weak in light of the large magnitude of and correlation between errors in exptl. measurements of entropic and enthalpic contributions to binding, though a limited form of compensation may be common. Given the difficulty of predicting or measuring entropic and enthalpic changes to useful precision, or using this information in design, we recommend ligand engineering efforts instead focus on computational and exptl. methodologies to directly assess changes in binding free energy.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3sXhtFGrs7bP&md5=52a30a1d0f4f9ae49128e87d2a23e3fa
26. 26
  Bissantz, C.; Folkers, G.; Rognan, D. Protein-Based Virtual Screening of Chemical Databases. 1. Evaluation of Different Docking/Scoring Combinations J. Med. Chem. 2000, 43, 4759– 4767 DOI: 10.1021/jm001044l
  Google Scholar
  26
  Protein-Based Virtual Screening of Chemical Databases. 1. Evaluation of Different Docking/Scoring Combinations
  Bissantz, Caterina; Folkers, Gerd; Rognan, Didier
  Journal of Medicinal Chemistry (2000), 43 (25), 4759-4767CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  Three different database docking programs (Dock, FlexX, Gold) have been used in combination with seven scoring functions (Chemscore, Dock, FlexX, Fresno, Gold, Pmf, Score) to assess the accuracy of virtual screening methods against two protein targets (thymidine kinase, estrogen receptor) of known three-dimensional structure. For both targets, it was generally possible to discriminate about 7 out of 10 true hits from a random database of 990 ligands. The use of consensus lists common to two or three scoring functions clearly enhances hit rates among the top 5% scorers from 10% (single scoring) to 25-40% (double scoring) and up to 65-70% (triple scoring). However, in all tested cases, no clear relationships could be found between docking and ranking accuracies. Moreover, predicting the abs. binding free energy of true hits was not possible whatever docking accuracy was achieved and scoring function used. As the best docking/consensus scoring combination varies with the selected target and the physicochem. of target-ligand interactions, we propose a two-step protocol for screening large databases: (i) screening of a reduced dataset contg. a few known ligands for deriving the optimal docking/consensus scoring scheme, (ii) applying the latter parameters to the screening of the entire database.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3cXotFymurY%253D&md5=ee74ac9a99e55759c2df27fd1db58010
27. 27
  Gohlke, H.; Kiel, C.; Case, D. A. Insights into Protein-Protein Binding by Binding Free Energy Calculation and Free Energy Decomposition for the Ras-Raf and Ras-RaiGDS Complexes J. Mol. Biol. 2003, 330, 891– 913 DOI: 10.1016/S0022-2836(03)00610-7
  Google Scholar
  27
  Insights into protein-protein binding by binding free energy calculation and free energy decomposition for the Ras-Raf and Ras-RalGDS complexes
  Gohlke, Holger; Kiel, Christina; Case, David A.
  Journal of Molecular Biology (2003), 330 (4), 891-913CODEN: JMOBAK; ISSN:0022-2836. (Elsevier Science Ltd.)
  Abs. binding free energy calcns. and free energy decompns. are presented for the protein-protein complexes H-Ras/C-Raf1 and H-Ras/RalGDS. Ras is a central switch in the regulation of cell proliferation and differentiation. In our study, we investigate the capability of the mol. mechanics (MM)-generalized Born surface area (GBSA) approach to est. abs. binding free energies for the protein-protein complexes. Averaging gas-phase energies, solvation free energies, and entropic contributions over snapshots extd. from trajectories of the unbound proteins and the complexes, calcd. binding free energies (Ras-Raf: -15.0(±6.3) kcal mol-1; Ras-RalGDS: -19.5(±5.9) kcal mol-1) are in fair agreement with exptl. detd. values (-9.6 kcal mol-1; -8.4 kcal mol-1), if appropriate ionic strength is taken into account. Structural determinants of the binding affinity of Ras-Raf and Ras-RalGDS are identified by means of free energy decompn. For the first time, computationally inexpensive generalized Born (GB) calcns. are applied in this context to partition solvation free energies along with gas-phase energies between residues of both binding partners. For selected residues, in addn., entropic contributions are estd. by classical statistical mechanics. Comparison of the decompn. results with exptl. detd. binding free energy differences for alanine mutants of interface residues yielded correlations with r2=0.55 and 0.46 for Ras-Raf and Ras-RalGDS, resp. Extension of the decompn. reveals residues as far apart as 25 A from the binding epitope that can contribute significantly to binding free energy. These "hotspots" are found to show large at. fluctuations in the unbound proteins, indicating that they reside in structurally less stable regions. Furthermore, hotspot residues experience a significantly larger-than-av. decrease in local fluctuations upon complex formation. Finally, by calcg. a pair-wise decompn. of interactions, interaction pathways originating in the binding epitope of Raf are found that protrude through the protein structure towards the loop L1. This explains the finding of a conformational change in this region upon complex formation with Ras, and it may trigger a larger structural change in Raf, which is considered to be necessary for activation of the effector by Ras.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXlt1aqtbs%253D&md5=0a8acf22e053534f34365f264ed4192a
28. 28
  Bunney, T. D.; Wan, S.; Thiyagarajan, N.; Sutto, L.; Williams, S. V.; Ashford, P.; Koss, H.; Knowles, M. A.; Gervasio, F. L.; Coveney, P. V.; Katan, M. The Effect of Mutations on Drug Sensitivity and Kinase Activity of Fibroblast Growth Factor Receptors: A Combined Experimental and Theoretical Study EBioMedicine 2015, 2, 194– 204 DOI: 10.1016/j.ebiom.2015.02.009
  Google Scholar
  28
  The Effect of Mutations on Drug Sensitivity and Kinase Activity of Fibroblast Growth Factor Receptors: A Combined Experimental and Theoretical Study
  Bunney Tom D; Thiyagarajan Nethaji; Ashford Paul; Katan Matilda; Wan Shunzhou; Coveney Peter V; Sutto Ludovico; Gervasio Francesco L; Williams Sarah V; Knowles Margaret A; Koss Hans
  EBioMedicine (2015), 2 (3), 194-204 ISSN:.
  Fibroblast growth factor receptors (FGFRs) are recognized therapeutic targets in cancer. We here describe insights underpinning the impact of mutations on FGFR1 and FGFR3 kinase activity and drug efficacy, using a combination of computational calculations and experimental approaches including cellular studies, X-ray crystallography and biophysical and biochemical measurements. Our findings reveal that some of the tested compounds, in particular TKI258, could provide therapeutic opportunity not only for patients with primary alterations in FGFR but also for acquired resistance due to the gatekeeper mutation. The accuracy of the computational methodologies applied here shows a potential for their wider application in studies of drug binding and in assessments of functional and mechanistic impacts of mutations, thus assisting efforts in precision medicine.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A280%3ADC%252BC2srotlKmtA%253D%253D&md5=8414f3530d24bbba0ea8cc98cb784b96
29. 29
  Andrews, M. D.; Bagal, S. K.; Gibson, K. R.; Omoto, K.; Ryckmans, T.; Skerratt, S. E.; Stupple, P. A. Pyrrolo[2,3-d]pyrimidine Derivatives as Inhibitors of Tropomyosin-Related Kinases and Their Preparation and Use in the Treatment of Pain. WO2012137089A1, 2012.
  Google Scholar
  There is no corresponding record for this reference.
30. 30
  Xing, L.; Klug-Mcleod, J.; Rai, B.; Lunney, E. A. Kinase Hinge Binding Scaffolds and Their Hydrogen Bond Patterns Bioorg. Med. Chem. 2015, 23, 6520– 6527 DOI: 10.1016/j.bmc.2015.08.006
  Google Scholar
  30
  Kinase hinge binding scaffolds and their hydrogen bond patterns
  Xing, Li; Klug-Mcleod, Jacquelyn; Rai, Brajesh; Lunney, Elizabeth A.
  Bioorganic & Medicinal Chemistry (2015), 23 (19), 6520-6527CODEN: BMECEP; ISSN:0968-0896. (Elsevier B.V.)
  Protein kinases constitute a major class of intracellular signaling mols., and describe some of the most prominent drug targets. Kinase inhibitors commonly employ small chem. scaffolds that form hydrogen bonds with the kinase hinge residues connecting the N- and C-terminal lobes of the catalytic domain. In general the satisfied hydrogen bonds are required for potent inhibition, therefore constituting a conserved feature in the majority of inhibitor-kinase interactions. From systematically analyzing the kinase scaffolds extd. from Pfizer crystal structure database (CSDb) the authors recognize that large no. of kinase inhibitors of diverse chem. structures are derived from a relatively small no. of common scaffolds. Depending on specific substitution patterns, scaffolds may demonstrate versatile binding capacities to interact with kinase hinge. Afforded by thousands of ligand-protein binary complexes, the hinge hydrogen bond patterns were analyzed with a focus on their three-dimensional configurations. Most of the compds. engage H6 NH for hinge recognition. Dual hydrogen bonds are commonly obsd. with addnl. recruitment of H4 CO upstream and/or H6 CO downstream. Triple hydrogen bonds accounts for small no. of binary complexes. An unusual hydrogen bond with a non-canonical H5 conformation is obsd., requiring a peptide bond flip by a glycine residue at the H6 position. Addnl. hydrogen bonds to kinase hinge do not necessarily correlate with an increase in potency; conversely they appear to compromise kinase selectivity. Such learnings could enhance the prospect of successful therapy design.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhsVentr3L&md5=048b4b99628c6e1ae273b00a47ab2377
31. 31
  Morphy, R. Selectively Nonselective Kinase Inhibition: Striking the Right Balance J. Med. Chem. 2010, 53, 1413– 1437 DOI: 10.1021/jm901132v
  Google Scholar
  31
  Selectively Nonselective Kinase Inhibition: Striking the Right Balance
  Morphy, Richard
  Journal of Medicinal Chemistry (2010), 53 (4), 1413-1437CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  A review. Marketed kinase inhibitors (MKIs) can deliver superior efficacy compared to inhibitors with high specificity for a single kinase, and the recent introduction of several MKIs to the market opens the door to a new era of safer and effective anticancer therapy. The key to combining high efficacy with acceptable safety is to inhibit multiple targets in a selectively nonselective fashion. Strategies for intentionally designing MKIs are emerging, but the field is still in its infancy and we are as medicinal chemists currently on the steepest part of the learning curve. MTDD can be time-consuming and expensive, and we need to become more proficient first at identifying disease-relevant target combinations and second at discovering MKIs that combine optimal physicochem. and biol. properties. Bold and innovative medicinal chem. strategies are required to tackle "difficult combinations" where the disease rationale is compelling but where it is a struggle to combine all the desired attributes of an oral MKI drug into a single mol. At present it is unclear to what extent MKIs with highly tuned selectivity profiles can be rationally designed, particularly for targets that are unrelated by sequence. In addn. to the well-known selectivity challenge, the physicochem. property profiles of AT P-competitive MKIs can be inherently challenging and limited scope for patentability can also be a serious hindrance. On the plus side, the amt. of kinase-specific structural information is growing very rapidly, and ultimately this may reveal distinct features and design rules that enable a medicinal chemist to rationally modify and refine the profile of MKIs. In addn., increasing SAR knowledge is emerging from large scale panel screening with the binding profiles starting to reveal to medicinal chemists how chem. structure affects cross-reactivity across large parts of the kinome. The merit of MKIs compared with single kinase inhibitors is a subject of controversy in drug discovery that is unlikely to be resolved in the near future. At the start of a new MTDD project, a rigorous debate needs to take place as to whether it makes more sense to seek a combination of highly selective agents or a DML. Many factors need to be taken into account in this decision such as the no., similarity, and promiscuity of the targets in the profile and the disease area. Conformational plasticity and the occurrence of multiple binding modes complicate the in silico prediction of kinase polypharmacol. based solely upon protein structure. The use of ligand-based similarity to assess the feasibility of a given combination can add real value. Currently, serendipity plays a significant role in MKI discovery and many, if not most, MKIs have been discovered by chance during the search for selective inhibitors. Medicinal chemists need to be alert to the possibilities when a surprising combination is found by chance. To exploit such serendipity, you need a good appreciation of when you have a sufficiently high quality starting compd. and then you need to be able to make and test sufficient analogs to explore your new disease-based hypothesis. MKIs are costly to develop and are consequently priced at a premium level, so they will need to show clear improvements in order to get reimbursement. There have already been problems with reimbursement for some MKIs in some markets due to concerns from funding bodies over insufficient efficacy. The true value of MKIs relative to other anticancer drugs still has to be established, and the results from recent clin. trials have been mixed. Despite the broad activity profile of many MKIs, the patient response can be inconsistent and unpredictable. The identification of predictive biomarkers of response or resistance is a crit. step to ascertain which specific combination of targets produces a significant clin. benefit with respect to specific tumor types. More clin. feedback is needed to facilitate the design of the next generation of inhibitors with more precisely defined profiles. Although it might seem immeasurably distant at the present time, the ultimate goal should be to derive the prerequisite knowledge and tools so that MTDD becomes a rational endeavor rather than a black box approach that relies upon serendipity. This will help banish claims that MKIs are merely dirty, nonspecific drugs with insufficient specificity for treating a wider range of human diseases.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXhtlWmtbfF&md5=3c915b5debf9d005993c8ad31c68bc2e
32. 32
  Ou-Yang, S. S.; Lu, J. Y.; Kong, X. Q.; Liang, Z. J.; Luo, C.; Jiang, H. Computational Drug Discovery Acta Pharmacol. Sin. 2012, 33, 1131– 1140 DOI: 10.1038/aps.2012.109
  Google Scholar
  32
  Computational drug discovery
  Ou-Yang, Si-sheng; Lu, Jun-yan; Kong, Xiang-qian; Liang, Zhong-jie; Luo, Cheng; Jiang, Hualiang
  Acta Pharmacologica Sinica (2012), 33 (9), 1131-1140CODEN: APSCG5; ISSN:1671-4083. (Nature Publishing Group)
  A review. Computational drug discovery is an effective strategy for accelerating and economizing drug discovery and development process. Because of the dramatic increase in the availability of biol. macromol. and small mol. information, the applicability of computational drug discovery has been extended and broadly applied to nearly every stage in the drug discovery and development workflow, including target identification and validation, lead discovery and optimization and preclin. tests. Over the past decades, computational drug discovery methods such as mol. docking, pharmacophore modeling and mapping, de novo design, mol. similarity calcn. and sequence-based virtual screening have been greatly improved. In this review, we present an overview of these important computational methods, platforms and successful applications in this field.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38Xht12lt7vL&md5=2a0bff666d5881d1109124049480b1a2
33. 33
  Wan, S.; Bhati, A. P.; Zasada, S. J.; Wall, I.; Green, D.; Bamborough, P.; Coveney, P. V. Rapid and Reliable Binding Affinity Prediction of Bromodomain Inhibitors: A Computational Study J. Chem. Theory Comput. 2017, 13, 784– 795 DOI: 10.1021/acs.jctc.6b00794
  Google Scholar
  33
  Rapid and Reliable Binding Affinity Prediction of Bromodomain Inhibitors: A Computational Study
  Wan, Shunzhou; Bhati, Agastya P.; Zasada, Stefan J.; Wall, Ian; Green, Darren; Bamborough, Paul; Coveney, Peter V.
  Journal of Chemical Theory and Computation (2017), 13 (2), 784-795CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  Binding free energies of bromodomain inhibitors are calcd. with recently formulated approaches, namely ESMACS (enhanced sampling of mol. dynamics with approxn. of continuum solvent) and TIES (thermodn. integration with enhanced sampling). A set of compds. is provided by GlaxoSmithKline, which represents a range of chem. functionality and binding affinities. The predicted binding free energies exhibit a good Spearman correlation of 0.78 with the exptl. data from the 3-trajectory ESMACS, and an excellent correlation of 0.92 from the TIES approach where applicable. Given access to suitable high end computing resources and a high degree of automation, the authors can compute individual binding affinities in a few hours with precisions no greater than 0.2 kcal/mol for TIES, and no larger than 0.34 kcal/mol and 1.71 kcal/mol for the 1- and 3-trajectory ESMACS approaches.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XitFektLnF&md5=713436084662420482684fb50db0832e
- PDB: 5JFV
Supporting Information
Supporting Information
ARTICLE SECTIONS
Jump To
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jcim.6b00780.
- Detailed description of the energy decomposition, the energy convergence, and the error analyses, experimental measurements of TrkA inhibitory activity, and the predicted binding free energies (PDF)
- ci6b00780_si_001.pdf (300.28 kb)
Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

PDF [ 9MB]

All Types

Evaluation and Characterization of Trk Kinase Inhibitors for the Treatment of Pain: Reliable Binding Affinity Predictions from Theory and Computation

Article Views

Altmetric

Citations

Abstract

Introduction

Computational Methods

Results

Figure 1

Comparison of Experimental TrkA Binding Free Energies with ESMACS Predictions

Figure 2

Improvement of Docking Predictions by the ESMACS Approach

Figure 3

ESMACS Binding Affinities: Contributions from Energy Components

Figure 4

Adaptation Energy: A Measure of Conformational Change upon Binding

Figure 5

Comparison of Experimental TrkA Binding Free Energies with TIES Predictions

Figure 6

Discussion

Hinge Binding Group Modifications: Compounds 1, 4, and 22

Figure 7

Hinge Binding Group Modifications: Compounds 12, 17, 23, and 24

Figure 8

Linker Group Modifications: Compounds 1, 4, and 22

Figure 9

Linker Group Modifications: Compounds 7 and 8

Figure 10

Linker Group Modifications: Compounds 4 and 16

Figure 11

Conclusions

Supporting Information

Terms & Conditions

Author Information

Acknowledgment

References

Cited By

Abstract

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

References

Supporting Information

Supporting Information

Terms & Conditions

STEP 1:

STEP 2: