Computer based screening of compound databases: 1. Preselection of benzamidine-based thrombin inhibitors

We present a computational protocol which uses the known three-dimensional structure of a target enzyme to identify possible ligands from databases of compounds with low molecular weight. This is accomplished by first mapping the essential interactions in the binding site with the program GRID. The resulting regions of favorable interaction between target and ligand are translated into a database query, and with UNITY a flexible 3D database search is performed. The feasibility of this approach is calibrated with thrombin as the target. Our results show that the resulting hit lists are enriched with thrombin inhibitors compared to the total database.     

Protein-ligand binding free energy estimation using molecular mechanics and continuum electrostatics. Application to HIV-1 protease inhibitors

A method is proposed for the estimation of absolute binding free energy of interaction between proteins and ligands. Conformational sampling of the protein-ligand complex is performed by molecular dynamics (MD) in vacuo and the solvent effect is calculated a posteriori by solving the Poisson or the Poisson-Boltzmann equation for selected frames of the trajectory. The binding free energy is written as a linear combination of the buried surface upon complexation, SASbur, the electrostatic interaction energy between the ligand and the protein, Eelec, and the difference of the solvation free energies of the complex and the isolated ligand and protein, deltaGsolv. The method uses the buried surface upon complexation to account for the non-polar contribution to the binding free energy because it is less sensitive to the details of the structure than the van der Waals interaction energy. The parameters of the method are developed for a training set of 16 HIV-1 protease-inhibitor complexes of known 3D structure. A correlation coefficient of 0.91 was obtained with an unsigned mean error of 0.8 kcal/mol. When applied to a set of 25 HIV-1 protease-inhibitor complexes of unknown 3D structures, the method provides a satisfactory correlation between the calculated binding free energy and the experimental pIC5o without reparametrization.     

Potent thrombin inhibitors via a 20-membered ring olefin metathesis macrocyclization

Twenty-membered ring pyrazinone derived macrocycles were prepared as a means to enhance the potency of existing thrombin inhibitors. Macrocyclization was accomplished via Grubbs olefin metathesis of a highly functionalized allyl-alloc scaffold, thus further confirming the power of such methodology.     

Comment on the validation of continuum electrostatics models

A validation based on solvation energies (vacuum to water transfer) is not sufficient to justify the use of approximated models of electrostatics to rank ligand/protein complexes. A full validation should be based on energies in solution, i.e., solvation plus vacuum Coulomb energies, because of the anticorrelation between solvation and vacuum energies. The energy in solution is the relevant quantity in simulations of biological macromolecules and complexes. ©1999 John Wiley & Sons, Inc. J Comput Chem 20: 1533–1536, 1999     

Molecular dynamics simulation and linear interaction energy study of d-Glu-based inhibitors of the MurD ligase

The biosynthetic pathway of the bacterial peptidoglycan, where MurD is an enzyme involved at the intracellular stage of its construction, represents a collection of highly selective macromolecular targets for novel antibacterial drug design. In this study as part of our investigation of the MurD bacterial target two recently discovered classes of the MurD ligase inhibitors were investigated resulting from the lead optimization phases of the N-sulfonamide d-Glu MurD inhibitors. Molecular dynamics simulations, based on novel structural data, in conjunction with the linear interaction energy (LIE) method suggested the transferability of our previously obtained LIE coefficients to further d-Glu based classes of MurD inhibitors. Analysis of the observed dynamical behavior of these compounds in the MurD active site was supported by static drug design techniques. These results complement the current knowledge of the MurD inhibitory mechanism and provide valuable support for the d-Glu paradigm of the inhibitor design.     

Dielectric relaxation in proteins: a continuum electrostatics model incorporating dielectric heterogeneity of the protein and time‐dependent charges

A boundary element formulation of continuum electrostatics is used to examine time‐independent dielectric relaxation and screening in two proteins, and time‐dependent relaxation in two simpler solutes. Cytochrome c oxidation is modeled by inserting partial charges on the heme, using one to three dielectric regions in the protein. It was suggested recently that for charge insertion on a protein‐bound ligand, all or part of the ligand should be treated as a cavity within the protein medium. Here, the effect of an internal cavity surrounding the central heme atoms is examined, considering separately the static and relaxation (or reorganization) free energies. The former is the free energy to remove the redox electron while maintaining the rest of the structure and charge distribution fixed; the latter is the free energy associated with the relaxation into the product state after the corresponding constraints are released. The effect of the cavity is found to be small for the static free energy, while for the relaxation free energy it is large, as polarization of groups immediately around the heme dominates the relaxation. If the protein surface groups are treated as a distinct medium with a dielectric of 25 (as suggested by recent molecular dynamics simulations), the relaxation free energy decreases significantly (from −37.0 to −43.9 kcal/mol), compared to a model where the whole protein has a dielectric constant of two. Therefore, with this model, although polarization of groups immediately around the heme still dominates the relaxation, polar groups near the protein surface also contribute significantly, and solvent negligibly. The screening of an applied field within myoglobin is calculated, with the protein surrounded by either a low‐dielectric or a high‐dielectric glass. In the vicinity of the CO ligand, the screening is approximately isotropic with a low‐dielectric glass. It is anisotropic with a high‐dielectric glass, but the applied and local fields are still approximately parallel. This has implications for experiments that probe dielectric screening in proteins with the newly developed technique of vibrational Stark spectroscopy: with a high‐dielectric glass, a single, rotationally averaged screening factor can be used, the local field being about 1.65 times the applied field. Finally, we calculate the time‐dependent relaxation in response to instantaneous charge insertion within a spherical cavity in a Debye solvent, and to photoexcitation of a tryptophan solute, illustrating the extension of the boundary element formulation to time‐dependent problems. © 2001 John Wiley & Sons, Inc. J Comput Chem 22: 290–305, 2001     

Multipole electrostatics in hydration free energy calculations

Hydration free energy (HFE) is generally used for evaluating molecular solubility, which is an important property for pharmaceutical and chemical engineering processes. Accurately predicting HFE is also recognized as one fundamental capability of molecular mechanics force field. Here, we present a systematic investigation on HFE calculations with AMOEBA polarizable force field at various parameterization and simulation conditions. The HFEs of seven small organic molecules have been obtained alchemically using the Bennett Acceptance Ratio method. We have compared two approaches to derive the atomic multipoles from quantum mechanical calculations: one directly from the new distributed multipole analysis and the other involving fitting to the electrostatic potential around the molecules. Wave functions solved at the MP2 level with four basis sets (6-311G*, 6-311++G(2d,2p), cc-pVTZ, and aug-cc-pVTZ) are used to derive the atomic multipoles. HFEs from all four basis sets show a reasonable agreement with experimental data (root mean square error 0.63 kcal/mol for aug-cc-pVTZ). We conclude that aug-cc-pVTZ gives the best performance when used with AMOEBA, and 6-311++G(2d,2p) is comparable but more efficient for larger systems. The results suggest that the inclusion of diffuse basis functions is important for capturing intermolecular interactions. The effect of long-range correction to van der Waals interaction on the hydration free energies is about 0.1 kcal/mol when the cutoff is 12?, and increases linearly with the number of atoms in the solute/ligand. In addition, we also discussed the results from a hybrid approach that combines polarizable solute with fixed-charge water in the HFE calculation.     

Application of the linear interaction energy method for rational design of artemisinin analogues as haeme polymerisation inhibitors

The anti-malarial activity of artemisinin-derived drugs appears to be mediated by an interaction of the drug's endoperoxide bridge with intra-parasitic haeme. The binding affinity of artemisinin analogues with haeme were computed using linear interaction energy with a surface generalised Born (LIE-SGB) continuum solvation model. Low levels of root mean square error (0.348 and 0.415 kcal/mol) as well as significant correlation coefficients (r ² = 0.868 and 0.892) between the experimental and predicted free energy of binding (FEB) based on molecular dynamics and hybrid Monte Carlo sampling techniques establish the SGB-LIE method as an efficient tool for generating more potent inhibitors of haeme polymerisation inhibition.     

Binding of aminoglycoside antibiotics to the small ribosomal subunit: a continuum electrostatics investigation

The binding of paromomycin and similar antibiotics to the small (30S) ribosomal subunit has been studied using continuum electrostatics methods. Crystallographic information from a complex of paromomycin with the 30S subunit was used as a framework to develop structures of similar antibiotics in the same ribosomal binding site. Total binding energies were calculated from electrostatic properties obtained by solution of the Poisson-Boltzmann equation combined with a surface area-dependent apolar term. These computed results showed good correlation with experimental data. Additionally, calculation of the ribosomal electrostatic potential in the paromomycin binding site provided insight into the electrostatic mechanisms for aminoglycoside binding and clues for the rational design of more effective antibiotics.     

Continuum solvation models in the linear interaction energy method

The linear interaction energy (LIE) method in combination with two different continuum solvent models has been applied to calculate protein-ligand binding free energies for a set of inhibitors against the malarial aspartic protease plasmepsin II. Ligand-water interaction energies are calculated from both Poisson-Boltzmann (PB) and Generalized Born (GB) continuum models using snapshots from explicit solvent simulations of the ligand and protein-ligand complex. These are compared to explicit solvent calculations, and we find close agreement between the explicit water and PB solvation models. The GB model overestimates the change in solvation energy, and this is caused by consistent underestimation of the effective Born radii in the protein-ligand complex. The explicit solvent LIE calculations and LIE-PB, with our standard parametrization, reproduce absolute experimental binding free energies with an average unsigned error of 0.5 and 0.7 kcal/mol, respectively. The LIE-GB method, however, requires a constant offset to approach the same level of accuracy.     

In silico binding free energy predictability by using the linear interaction energy (LIE) method: bromobenzimidazole CK2 inhibitors as a case study

Protein kinase CK2 is essential for cell viability, and its control regards a broad series of cellular events such as gene expression, RNA, and protein synthesis. Evidence of its involvement in tumor development and viral replication indicates CK2 as a potential target of antineoplastic and antiviral drugs. In this study the Linear Interaction Energy (LIE) Method with the Surface Generalized Born (SGB) continuum solvation model was used to study several bromobenzimidazole CK2 inhibitors. This methodology, developed by Aqvist, finds a plausible compromise between accuracy and computational speed in evaluating binding free energy (DeltaGbind) values. In this study, two different free binding energy models, named "CK2scoreA" and "CK2scoreB", were developed using 22 inhibitors as the training set in a stepwise approach useful to appropriately select both the tautomeric form and the starting binding position of each inhibitor. Both models are statistically acceptable. Indeed, the better one is characterized by a correlation coefficient (r2) of 0.81, and the predictive accuracy was 0.65 kcal/mol. The corresponding validation, using an external test set of 16 analogs, showed a correlation coefficient (q2) of 0.68 and a prediction root-mean-square error of 0.78 kcal/mol. In this case, the LIE approach has been proved to be an efficient methodology to rationalize the difference of activity, the key interactions, and the different possible binding modes of this specific class of potent CK2 inhibitors.     

Molecular dynamics and continuum electrostatics studies of inactivation in the HERG potassium channel

Fast inactivation of the HERG potassium channel plays a critical role in normal cardiac function. Malfunction of these channels due to either genetic mutations or blockade by drugs leads to cardiac arrhythmias. An unusually long S5-P linker in the outer mouth of HERG is implicated in the fast inactivation mechanism. To examine the role of the S5-P linker in this inactivation mechanism, we study the permeation properties of the open and inactive states of a recent homology model of HERG. This model was constructed using the KcsA potassium channel as a template and contains specific conformations of the S5-P linker in the open and inactive states. We perform molecular dynamics simulations on the HERG model, followed by free energy, structural, and continuum electrostatics calculations. Our free energy calculations lead to selectivity results of the model channel (K+ over Na+) that are different in some respects from those of other potassium channels but consistent with experimental observations. Our structural results show that, in the inactive state, the S5-P linkers move closer to the channel axis, possibly causing a steric hindrance to permeating K+ ions. Our electrostatics calculations reveal, in the inactive state, an electrostatic potential energy barrier of approximately 14 kT at the extracellular pore entrance, again sufficient to stop K+ ion permeation through the pore. These results suggest that a steric and/or electrostatic plug mechanism contributes to inactivation in the HERG homology model.     

Screening for cell migration inhibitors via automated microscopy reveals a Rho-kinase inhibitor

Small-molecule kinase inhibitors are predominantly discovered in pure protein assays. We have discovered an inhibitor of Rho-kinase (ROCK) through an image-based, high-throughput screen of cell monolayer wound healing. Using automated microscopy, we screened a library of approximately 16,000 compounds finding many that affected cell migration or cell morphology as well as compounds that blocked mitotic progression. We tested approximately 200 compounds in a series of subassays and chose one, 3-(4-pyridyl)indole (Rockout), for more detailed characterization. Rockout inhibits blebbing and causes dissolution of actin stress fibers, phenocopying Rho-kinase inhibitors. Testing Rho-kinase activity in vitro, Rockout inhibits with an IC50 of 25 microM ( approximately 5-fold less potent than Y-27632) but has a similar specificity profile. We also profile the wound healing assay with a library of compounds with known bioactivities, revealing multiple pathways involved in the biology.     

A molecule on the surface or inside a spherical nanoparticle: the computation of the interaction energy in terms of the dielectric continuum model

The electrostatic interaction energy between a charged or polar molecule and a spherical polarizable nanoparticle is studied within the advanced dielectric continuum model proposed previously. The molecule can be located either inside or outside the nanoparticle or in the vicinity of its boundary surface. The nanoparticle and its environment are considered as a polarizable medium and described in terms of a nonuniform dielectric continuum approximation with a position-dependent dielectric permittivity function e(r) \varepsilon (r) , where r is the position vector. A special construction of this function accounts for the proper treatment of sophisticated boundary effects. Test computations are performed for a number of sample molecules.     

On the transferability of hydration-parametrized continuum electrostatics models to solvated binding calculations

Using molecular mechanics force field partial atomic charges, we show the nonuniqueness of the parametrization of continuum electrostatics models with respect to solute atomic radii and interior dielectric constant based on hydration (vacuum-to-water transfer) free energy data available for small molecules. Moreover, parameter sets that are optimal and equivalent for hydration free energy calculations lead to large variations of calculated absolute and relative electrostatic binding free energies. Hence, parametrization of solvation effects based on hydration data, although a necessary condition, is not sufficient to guarantee its transferability to the calculation of binding free energies in solution.     

Solution conformations of structured peptides: continuum electrostatics versus distance-dependent dielectric functions

To compare different implicit solvent potentials, the folding thermodynamics of the helical peptide RN24 and the β-hairpin peptide BH8 are studied by molecular dynamics simulation with adaptive umbrella sampling. As the potential energy functions, the analytical continuum solvent (ACS) potential and three simplified variants, termed EPSR₁, EPSR₄, and EPSR₁₀, are used. The ACS potential is a combination of the standard CHARMM force field for the internal energy (bonds, angles, dihedrals) and the van der Waals energy with the analytical continuum electrostatic (ACE) potential and a non-polar solvation potential. The EPSR potentials differ from the ACS potential by the use of Coulomb's law with a distance-dependent dielectric function to calculate the electrostatic energy. With the ACS potential, quantitative agreement with experiment is obtained for the helix propensity (RN24: 62% calculated vs 50–60% experiment) and the β-hairpin propensity (BH8: 33% calculated vs 19–37% experiment) of the peptides. During the simulations with the EPSR potentials, no significant formation of secondary structure is observed. It is shown that the preference for coil conformations over conformations with secondary structure by the EPSR potentials is due to an overestimation of the energy of salt bridge formation, independent of the magnitude of the Coulomb energy relative to the other energy terms. Possible improvements of the distance-dependent dielectric functions which may permit their application to the simulation of peptide folding, are discussed. Received: 11 July 1998 / Accepted: 22 September 1998 / Published online: 17 December 1998     

Parameterization of a coarse-grained model of cholesterol with point-dipole electrostatics

We present a new coarse-grained (CG) model of cholesterol (CHOL) for the electrostatic-based ELBA force field. A distinguishing feature of our CHOL model is that the electrostatics is modeled by an explicit point dipole which interacts through an ideal vacuum permittivity. The CHOL model parameters were optimized in a systematic fashion, reproducing the electrostatic and nonpolar partitioning free energies of CHOL in lipid/water mixtures predicted by full-detailed atomistic molecular dynamics simulations. The CHOL model has been validated by comparison to structural, dynamic and thermodynamic properties with experimental and atomistic simulation reference data. The simulation of binary DPPC/cholesterol mixtures covering the relevant biological content of CHOL in mammalian membranes is shown to correctly predict the main lipid behavior as observed experimentally.     

Intramolecular electronic energy transfer via exchange interaction in bichromophoric molecules

A theory is presented for intramolecular electronic energy transfer in bichromophoric molecules. Expressions are given for the donor moiety fluorescence (phosphorescence) decay and for its fluorescence (phosphorescence) quantum yield in terms of the average distance between the donor and acceptor moieties and the donor—acceptor bridge flexibility. Comparison with available experimental data supports the predictions of the analysis.     

Coarse-graining in interaction space: a systematic approach for replacing long-range electrostatics with short-range potentials

A short-range effective potential for long-range electrostatic interactions in homogeneously disordered condensed phase systems has been determined with a novel approach to coarse-graining in interaction space. As opposed to coarse-graining the system resolution, this approach "coarsens" the system's interactions by mapping multiple configurations of an accurate long-range atomistic potential onto a more efficient, short-range effective potential with a force-matching (FM) method. Developing an empirical potential in this manner is fundamentally different from existing strategies because it utilizes condensed-phase (as opposed to gas-phase) atomistic interactions to determine general pair potentials defined on distance meshes (as opposed to fitting predetermined functional forms). The resulting short-range ( approximately 10 A) effective potential reproduces structural, dynamical, and many thermodynamic properties of liquid water, ions in water, and hydrophobes in water, with unprecedented accuracy. The effective potential is also shown to be transferable to a nonaqueous molten salt system. With continued development, such effective potentials may provide an accurate and highly efficient alternative to Ewald-based long-range electrostatics methods.     

FDS: flexible ligand and receptor docking with a continuum solvent model and soft-core energy function

The docking of flexible small molecule ligands to large flexible protein targets is addressed in this article using a two-stage simulation-based method. The methodology presented is a hybrid approach where the first component is a dock of the ligand to the protein binding site, based on deriving sets of simultaneously satisfied intermolecular hydrogen bonds using graph theory and a recursive distance geometry algorithm. The output structures are reduced in number by cluster analysis based on distance similarities. These structures are submitted to a modified Monte Carlo algorithm using the AMBER-AA molecular mechanics force field with the Generalized Born/Surface Area (GB/SA) continuum model. This solvent model is not only less expensive than an explicit representation, but also yields increased sampling. Sampling is also increased using a rotamer library to direct some of the protein side-chain movements along with large dihedral moves. Finally, a softening function for the nonbonded force field terms is used, enabling the potential energy function to be slowly turned on throughout the course of the simulation. The docking procedure is optimized, and the results are presented for a single complex of the arabinose binding protein. It was found that for a rigid receptor model, the X-ray binding geometry was reproduced and uniquely identified based on the associated potential energy. However, when side-chain flexibility was included, although the X-ray structure was identified, it was one of three possible binding geometries that were energetically indistinguishable. These results suggest that on relaxing the constraint on receptor flexibility, the docking energy hypersurface changes from being funnel-like to rugged. A further 14 complexes were then examined using the optimized protocol. For each complex the docking methodology was tested for a fully flexible ligand, both with and without protein side-chain flexibility. For the rigid protein docking, 13 out of the 15 test cases were able to find the experimental binding mode; this number was reduced to 11 for the flexible protein docking. However, of these 11, in the majority of cases the experimental binding mode was not uniquely identified, but was present in a cluster of low energy structures that were energetically indistinguishable. These results not only support the presence of a rugged docking energy hypersurface, but also suggest that it may be necessary to consider the possibility of more than one binding conformation during ligand optimization.