Binding energies of tyrosine kinase inhibitors: Error assessment of computational methods for imatinib and nilotinib binding

The binding energies of imatinib and nilotinib to tyrosine kinase have been determined by quantum mechanical (QM) computations, and compared with literature binding energy studies using molecular mechanics (MM). The potential errors in the computational methods include these critical factors:

• •Errors in X-ray structures such as structural distortions and steric clashes give unrealistically high van der Waals energies, and erroneous binding energies.
• •MM optimization gives a very different configuration to the QM optimization for nilotinib, whereas the imatinib ion gives similar configurations
• •Solvation energies are a major component of the overall binding energy. The QM based solvent model (PCM/SMD) gives different values from those used in the implicit PBSA solvent MM models. A major error in inhibitor—kinase binding lies in the non-polar solvation terms.
• •Solvent transfer free energies and the required empirical solvent accessible surface area factors for nilotinib and imatinib ion to give the transfer free energies have been reverse calculated. These values differ from those used in the MM PBSA studies.
• •An intertwined desolvation—conformational binding selectivity process is a balance of thermodynamic desolvation and intramolecular conformational kinetic control.
• •The configurational entropies (TΔS) are minor error sources.

     

A displaced-solvent functional analysis of model hydrophobic enclosures

Calculation of protein-ligand binding affinities continues to be a hotbed of research. Although many techniques for computing protein-ligand binding affinities have been introduced--ranging from computationally very expensive approaches, such as free energy perturbation (FEP) theory; to more approximate techniques, such as empirically derived scoring functions, which, although computationally efficient, lack a clear theoretical basis--there remains pressing need for more robust approaches. A recently introduced technique, the displaced-solvent functional (DSF) method, was developed to bridge the gap between the high accuracy of FEP and the computational efficiency of empirically derived scoring functions. In order to develop a set of reference data to test the DSF theory for calculating absolute protein-ligand binding affinities, we have pursued FEP theory calculations of the binding free energies of a methane ligand with 13 different model hydrophobic enclosures of varying hydrophobicity. The binding free energies of the methane ligand with the various hydrophobic enclosures were then recomputed by DSF theory and compared with the FEP reference data. We find that the DSF theory, which relies on no empirically tuned parameters, shows excellent quantitative agreement with the FEP. We also explored the ability of buried solvent accessible surface area and buried molecular surface area models to describe the relevant physics, and find the buried molecular surface area model to offer superior performance over this dataset.     

Probing the conformational space available to inhibitors in the thermolysin active site using Monte Carlo/energy minimization techniques

The conformational space available to four inhibitors of the bacterial enzyme thermolysin has been searched in the enzyme binding site using a method that combines Monte Carlo type techniques with energy minimization for exploration of the conformational potential energy hypersurface. Molecular mechanics methodology using the AMBER force field was employed for computation of the molecular energetics. Solvation energies were also included in the calculations by employing a technique that estimates hydration energies based on the exposed solvent accessible surface area for each atom of the inhibitor and active site. It was found that in each case, the crystallographically observed conformation was among the low energy conformers discovered. In fact, in three of the calculations it was the lowest energy conformation. The methodology described in this article is expected to be quite useful for studies involving computer aided design and evaluation of enzyme inhibitors.     

Treating entropy and conformational changes in implicit solvent simulations of small molecules

Implicit solvent models are increasingly popular for estimating aqueous solvation (hydration) free energies in molecular simulations and other applications. In many cases, parameters for these models are derived to reproduce experimental values for small molecule hydration free energies. Often, these hydration free energies are computed for a single solute conformation, neglecting solute conformational changes upon solvation. Here, we incorporate these effects using alchemical free energy methods. We find significant errors when hydration free energies are estimated using only a single solute conformation, even for relatively small, simple, rigid solutes. For example, we find conformational entropy (TDeltaS) changes of up to 2.3 kcal/mol upon hydration. Interestingly, these changes in conformational entropy correlate poorly (R2 = 0.03) with the number of rotatable bonds. The present study illustrates that implicit solvent modeling can be improved by eliminating the approximation that solutes are rigid.     

Empirical methods for computing molecular partition coefficients: II. Inclusion of conformational flexibility within fragment–based approaches

A novel algorithm for computing the water/1-octanol partition coefficient, log P , of conformationally flexible molecules, has been investigated using calculations upon a number of uncharged, linear dipeptides. In this method (which appears to be the first to consider explicitly the effects of the population of accessible conformational minima in both phases), the partition coefficient for each dipeptide was calculated from the overall energy change associated with moving the relevant gas-phase conformational distribution into water and into 1-octanol. These energies were computed using solvation contributions based upon the solvent accessible molecular surface area and two sets of empirical parameters. In these initial studies, gas-phase conformational minima were generated using systematic search methods. While the standard error in the computed log P values was disappointing, reasonable agreement was observed between calculated and experimental log P values for the set of model dipeptides, especially when specific hydration interactions involving polar fragments were correctly included in the empirical solvation term. These results indicate that the physical basis of many correction factors employed in the ClogP algorithm for computing log P probably arise from neglect of the redistribution of conformer populations as flexible molecules partition between water and 1-octanol.     

Binding free energy, energy and entropy calculations using simple model systems

Free energy differences are calculated for a set of two model host molecules, binding acetone and methanol. Two active sites of different characteristics were constructed based on an artificially extended C60 fullerene molecule, possibly functionalised to include polar interactions in an otherwise apolar, spherical cavity. The model host systems minimise the necessary sampling of conformational space while still capturing key aspects of ligand binding. The estimates of the free energies are split up into energetic and entropic contributions, using three different approaches investigating the convergence behaviour. For these systems, a direct calculation of the total energy and entropy is more efficient than calculating the entropy from the temperature dependence of the free energy or from a direct thermodynamic integration formulation. Furthermore, the compensating surrounding–surrounding energies and entropies are split off by calculating reduced ligand-surrounding energies and entropies. These converge much more readily and lead to properties that are more straightforwardly interpreted in terms of molecular interactions and configurations. Even though not experimentally accessible, the reduced thermodynamic properties may prove highly relevant for computational drug design, as they may give direct insights into possibilities to further optimise ligand binding while optimisation in the surrounding–surrounding energy or entropy will exactly cancel and not lead to improved affinity.     

Conformational energy penalties of protein-bound ligands

The conformational energies required for ligands to adopt their bioactive conformations were calculated for 33 ligand–protein complexes including 28 different ligands. In order to monitor the force field dependence of the results, two force fields, MM3 and AMBER, were employed for the calculations. Conformational analyses were performed in vacuo and in aqueous solution by using the generalized Born/solvent accessible surface (GB/SA) solvation model. The protein-bound conformations were relaxed by using flat-bottomed Cartesian constraints. For about 70% of the ligand–protein complexes studied, the conformational energies of the bioactive conformations were calculated to be 3 kcal/mol. It is demonstrated that the aqueous conformational ensemble for the unbound ligand must be used as a reference state in this type of calculations. The calculations for the ligand–protein complexes with conformational energy penalties of the ligand calculated to be larger than 3 kcal/mol suffer from uncertainties in the interpretation of the experimental data or limitations of the computational methods. For example, in the case of long-chain flexible ligands (e.g. fatty acids), it is demonstrated that several conformations may be found which are very similar to the conformation determined by X-ray crystallography and which display significantly lower conformational energy penalties for binding than obtained by using the experimental conformation. For strongly polar molecules, e.g. amino acids, the results indicate that further developments of the force fields and of the dielectric continuum solvation model are required for reliable calculations on the conformational properties of this type of compounds.     

Binding-affinity predictions of HSP90 in the D3R Grand Challenge 2015 with docking,MM/GBSA,QM/MM,and free-energy simulations

We have estimated the binding affinity of three sets of ligands of the heat-shock protein 90 in the D3R grand challenge blind test competition. We have employed four different methods, based on five different crystal structures: first, we docked the ligands to the proteins with induced-fit docking with the Glide software and calculated binding affinities with three energy functions. Second, the docked structures were minimised in a continuum solvent and binding affinities were calculated with the MM/GBSA method (molecular mechanics combined with generalised Born and solvent-accessible surface area solvation). Third, the docked structures were re-optimised by combined quantum mechanics and molecular mechanics (QM/MM) calculations. Then, interaction energies were calculated with quantum mechanical calculations employing 970–1160 atoms in a continuum solvent, combined with energy corrections for dispersion, zero-point energy and entropy, ligand distortion, ligand solvation, and an increase of the basis set to quadruple-zeta quality. Fourth, relative binding affinities were estimated by free-energy simulations, using the multi-state Bennett acceptance-ratio approach. Unfortunately, the results were varying and rather poor, with only one calculation giving a correlation to the experimental affinities larger than 0.7, and with no consistent difference in the quality of the predictions from the various methods. For one set of ligands, the results could be strongly improved (after experimental data were revealed) if it was recognised that one of the ligands displaced one or two water molecules. For the other two sets, the problem is probably that the ligands bind in different modes than in the crystal structures employed or that the conformation of the ligand-binding site or the whole protein changes.     

Converging free energy estimates: MM-PB(GB)SA studies on the protein-protein complex Ras-Raf

Estimating protein-protein interaction energies is a very challenging task for current simulation protocols. Here, absolute binding free energies are reported for the complex H-Ras/C-Raf1 using the MM-PB(GB)SA approach, testing the internal consistency and model dependence of the results. Averaging gas-phase energies (MM), solvation free energies as determined by Generalized Born models (GB/SA), and entropic contributions calculated by normal mode analysis for snapshots obtained from 10 ns explicit-solvent molecular dynamics in general results in an overestimation of the binding affinity when a solvent-accessible surface area-dependent model is used to estimate the nonpolar solvation contribution. Applying the sum of a cavity solvation free energy and explicitly modeled solute-solvent van der Waals interaction energies instead provides less negative estimates for the nonpolar solvation contribution. When the polar contribution to the solvation free energy is determined by solving the Poisson-Boltzmann equation (PB) instead, the calculated binding affinity strongly depends on the atomic radii set chosen. For three GB models investigated, different absolute deviations from PB energies were found for the unbound proteins and the complex. As an alternative to normal-mode calculations, quasiharmonic analyses have been performed to estimate entropic contributions due to changes of solute flexibility upon binding. However, such entropy estimates do not converge after 10 ns of simulation time, indicating that sampling issues may limit the applicability of this approach. Finally, binding free energies estimated from snapshots of the unbound proteins extracted from the complex trajectory result in an underestimate of binding affinity. This points to the need to exercise caution in applying the computationally cheaper "one-trajectory-alternative" to systems where there may be significant changes in flexibility and structure due to binding. The best estimate for the binding free energy of Ras-Raf obtained in this study of -8.3 kcal mol(-1) is in good agreement with the experimental result of -9.6 kcal mol(-1), however, further probing the transferability of the applied protocol that led to this result is necessary.     

Modeling hydrophobic solvation of nonspherical systems: Comparison of use of molecular surface area with accessible surface area

It is shown that the molecular surface and the accessible surface lead to exactly the same results when calculating solvation free energies and transfer free energies, from methods using the surface tension as a parameter if the exact geometric curvature is used with the accessible surface. However, the use of the exact curvature is not necessarily the best approach chemically. Other modifications, including an approximate curvature improves the approach. Such modifications are difficult to include in methods in which the molecular surface rather than the accessible surface is used to calculate solvent effects. A modification of a Gaussian curvature term is necessary if dissociation is to be accounted for properly. The inclusion of a Gaussian curvature term, in addition to the usual mean curvature term, reconciles the difference in magnitude of the microscopic and macroscopic surface tension in the case of the accessible surface area. © 1997 by John Wiley & Sons, Inc.     

Oxoanion binding by guanidiniocarbonylpyrrole cations in water: a combined DFT and MD investigation

Structures and properties of nonbonding interactions involving guanidinium-functionalized hosts and carboxylate substrates were investigated by a combination of ab initio and molecular dynamics approaches. The systems under study are on one hand intended to be a model of the arginine-anion bond, so often observed in proteins and nucleic acids, and on the other to provide an opportunity to investigate the influence of molecular structure on the formation of supramolecular complexes in detail. Use of DFT calculations, including extended basis sets and implicit water treatment, allowed us to determine minimum-energy structures and binding enthalpies that compared well with experimental data. Intermolecular forces were found to be mostly due to electrostatic interactions through three hydrogen bonds, one of which is bifurcate, and are sufficiently strong to induce a conformational change in the ligand consisting of a rotation of about 180 degrees around the guanidiniocarbonylpyrrole axis. Free binding energies of the complexes were evaluated through MD simulations performed in the presence of explicit water molecules by use of the molecular mechanics Poisson-Boltzmann solvent accessible surface area (MM-PBSA) and linear interaction energy (LIE) approaches. LIE energies were in quantitative agreement with experimental data. A detailed analysis of the MD simulations revealed that the complexes cannot be described in terms of a single binding structure, but that they are characterized by a significant internal mobility responsible for several low-energy metastable structures.     

Combining the polarizable Drude force field with a continuum electrostatic Poisson–Boltzmann implicit solvation model

In this work, we have combined the polarizable force field based on the classical Drude oscillator with a continuum Poisson–Boltzmann/solvent‐accessible surface area (PB/SASA) model. In practice, the positions of the Drude particles experiencing the solvent reaction field arising from the fixed charges and induced polarization of the solute must be optimized in a self‐consistent manner. Here, we parameterized the model to reproduce experimental solvation free energies of a set of small molecules. The model reproduces well‐experimental solvation free energies of 70 molecules, yielding a root mean square difference of 0.8 kcal/mol versus 2.5 kcal/mol for the CHARMM36 additive force field. The polarization work associated with the solute transfer from the gas‐phase to the polar solvent, a term neglected in the framework of additive force fields, was found to make a large contribution to the total solvation free energy, comparable to the polar solute–solvent solvation contribution. The Drude PB/SASA also reproduces well the electronic polarization from the explicit solvent simulations of a small protein, BPTI. Model validation was based on comparisons with the experimental relative binding free energies of 371 single alanine mutations. With the Drude PB/SASA model the root mean square deviation between the predicted and experimental relative binding free energies is 3.35 kcal/mol, lower than 5.11 kcal/mol computed with the CHARMM36 additive force field. Overall, the results indicate that the main limitation of the Drude PB/SASA model is the inability of the SASA term to accurately capture non‐polar solvation effects. © 2018 Wiley Periodicals, Inc.     

Energy–entropy method using multiscale cell correlation to calculate binding free energies in the SAMPL8 host–guest challenge

Free energy drives a wide range of molecular processes such as solvation, binding, chemical reactions and conformational change. Given the central importance of binding, a wide range of methods exist to calculate it, whether based on scoring functions, machine-learning, classical or electronic structure methods, alchemy, or explicit evaluation of energy and entropy. Here we present a new energy–entropy (EE) method to calculate the host–guest binding free energy directly from molecular dynamics (MD) simulation. Entropy is evaluated using Multiscale Cell Correlation (MCC) which uses force and torque covariance and contacts at two different length scales. The method is tested on a series of seven host–guest complexes in the SAMPL8 (Statistical Assessment of the Modeling of Proteins and Ligands) “Drugs of Abuse” Blind Challenge. The EE-MCC binding free energies are found to agree with experiment with an average error of 0.9 kcal mol^?1. MCC makes clear the origin of the entropy changes, showing that the large loss of positional, orientational, and to a lesser extent conformational entropy of each binding guest is compensated for by a gain in orientational entropy of water released to bulk, combined with smaller decreases in vibrational entropy of the host, guest and contacting water.

     

The SGB/NP hydration free energy model based on the surface generalized born solvent reaction field and novel nonpolar hydration free energy estimators

The development and parameterization of a solvent potential of mean force designed to reproduce the hydration thermodynamics of small molecules and macromolecules aimed toward applications in conformation prediction and ligand binding free energy prediction is presented. The model, named SGB/NP, is based on a parameterization of the Surface Generalized Born continuum dielectric electrostatic model using explicit solvent free energy perturbation calculations and a newly developed nonpolar hydration free energy estimator motivated by the results of explicit solvent simulations of the thermodynamics of hydration of hydrocarbons. The nonpolar model contains, in addition to the more commonly used solvent accessible surface area term, a component corresponding to the attractive solute-solvent interactions. This term is found to be important to improve the accuracy of the model, particularly for cyclic and hydrogen bonding compounds. The model is parameterized against the experimental hydration free energies of a set of small organic molecules. The model reproduces the experimental hydration free energies of small organic molecules with an accuracy comparable or superior to similar models employing more computationally demanding estimators and/or a more extensive set of parameters.     

Analyzing the robustness of the MM/PBSA free energy calculation method: application to DNA conformational transitions

The ability to predict and characterize free energy differences associated with conformational equilibria or the binding of biomolecules is vital to understanding the molecular basis of many important biological functions. As biological studies focus on larger molecular complexes and properties of the genome, proteome, and interactome, the development and characterization of efficient methods for calculating free energy becomes increasingly essential. The aim of this study is to examine the robustness of the end-point free energy method termed the molecular mechanics Poisson-Boltzmann solvent accessible surface area (MM/PBSA) method. Specifically, applications of MM/PBSA to the conformational equilibria of nucleic acid (NA) systems are explored. This is achieved by comparing A to B form DNA conformational free energy differences calculated using MM/PBSA with corresponding free energy differences determined with a more rigorous and time-consuming umbrella sampling algorithm. In addition, the robustness of NA MM/PBSA calculations is also evaluated in terms of the sensitivity towards the choice of force field and the choice of solvent model used during conformational sampling. MM/PBSA calculations of the free energy difference between A-form and B-form DNA are shown to be in very close agreement with the PMF result determined using an umbrella sampling approach. Further, it is found that the MM/PBSA conformational free energy differences were also in agreement using either the CHARMM or AMBER force field. The influence of ionic strength on conformational stability was particularly insensitive to the choice of force field. Finally, it is also shown that the use of a generalized Born implicit solvent during conformational sampling results in free energy estimates that deviate slightly from those obtained using explicitly solvated MD simulations in these NA systems.     

Calculations of pH-dependent binding of proteins to biological membranes

Binding of proteins to membranes is often accompanied by titration of ionizable residues and is, therefore, dependent on pH. We present a theoretical treatment and computational approach for predicting absolute, pH-dependent membrane binding free energies. The standard free energy of binding, DeltaG, is defined as -RTln(P(b)/P(f)), where P(b) and P(f) are the amounts of bound and free protein. The apparent pK(a) of binding is the pH value at which P(b) and P(f) are equal. Proteins bind to the membrane in the pH range where DeltaG is negative. The components of the binding free energy are (a) the free energy cost of ionization state changes (DeltaG(ion)), (b) the effective energy of transfer from solvent to the membrane surface, (c) the translational/rotational entropy cost of binding, and (d) an ideal entropy term that depends on the relative volume of the bound and free state and therefore depends on lipid concentration. Calculation of the first term requires determination of pK(a) values in solvent and on the membrane surface. All energies required by the method are obtained from molecular dynamics trajectories on an implicit membrane (IMM1-GC). The method is tested on pentalysine and the helical peptide VEEKS, derived from the membrane-binding domain of phosphocholine cytidylyltransferase. The agreement between the measured and the calculated free energies of binding of pentalysine is good. The extent of membrane binding of VEEKS is, however, underestimated compared to experiment. Calculations of the interaction energy between two VEEKS helices on the membrane suggest that the discrepancy is mainly due to the neglect of protein-protein interactions on the membrane surface.     

β-环糊精和甾类化合物的分子动力学模拟

运用分子动力学(molecular dynamics, MD)和MM-PBSA (molecular mechanics/Poisson Boltzmann surface area)相结合的方法预测了β-环糊精(cyclodextrin, CD)和甾类客体分子包结模式. 通过重原子均方根偏差(root mean square deviation, RMSD)分析可得, 两种包结模式下客体分子都可以和β-CD形成稳定的包结. 在MD轨迹采样基础上, 采用高效MM-PBSA方法计算了两种包结模式下的包结自由能. 计算结果显示, β-CD和三个甾类客体分子包结的主要驱动力为范德华相互作用, 而溶剂化能和熵变则不利于体系的包结. 进一步分析平均构象和包结自由能发现, 对于波尼松龙, D-up (D-ring up orientation)取向为优势包结模式; 而乙炔雌二醇和雌三醇的优势包结模式均为A-up (A-ring up orientation)取向. 通过比较β-CD和三个客体分子的理论包结自由能, 预测包结稳定性的次序为乙炔雌二醇＞雌三醇＞波尼松龙, 和实验结果相一致.     

Universal solvation model based on conductor‐like screening model

Atomic surface tensions are parameterized for use with solvation models in which the electrostatic part of the calculation is based on the conductor‐like screening model (COSMO) and the semiempirical molecular orbital methods AM1, PM3, and MNDO/d. The convergence of the calculated polarization free energies with respect to the numerical parameters of the electrostatic calculations is first examined. The accuracy and precision of the calculated values are improved significantly by adjusting two parameters that control the segmentation of the solvent‐accessible surface that is used for the calculations. The accuracy of COSMO calculations is further improved by adopting an optimized set of empirical electrostatic atomic radii. Finally, the electrostatic calculation is combined with SM5‐type atomic surface tension functionals that are used to compute the nonelectrostatic portions of the solvation free energy. All parameterizations are carried out using rigid (R) gas‐phase geometries; this combination (SM5‐type surface tensions, COSMO electrostatics, and rigid geometries) is called SM5CR. Six air–water and 76 water–solvent partition coefficients are added to the training set of air–solvent data points previously used to parameterize the SM5 suite of solvation models, thereby bringing the total number of data points in the training set to 2266. The model yields free energies of solvation and transfer with mean unsigned errors of 0.63, 0.59, and 0.61 kcal/mol for AM1, PM3, and MNDO/d, respectively, over all 2217 data points for neutral solutes in the training set and mean unsigned errors of 3.0, 2.7, and 3.1 kcal/mol, respectively, for 49 data points for the ions. © 2000 John Wiley & Sons, Inc. J Comput Chem 21: 340–366, 2000