John W. Liebeschuetz

Potential and Limitations of Ensemble Docking

A major problem in structure-based virtual screening applications is the appropriate selection of a single or even multiple protein structures to be used in the virtual screening process. A priori it is unknown which protein structure(s) will perform best in a virtual screening experiment. We investigated the performance of ensemble docking, as a function of ensemble size, for eight targets of pharmaceutical interest. Starting from single protein structure docking results, for each ensemble size up to 500,000 combinations of protein structures were generated, and, for each ensemble, pose prediction and virtual screening results were derived. Comparison of single to multiple protein structure results suggests improvements when looking at the performance of the worst and the average over all single protein structures to the performance of the worst and average over all protein ensembles of size two or greater, respectively. We identified several key factors affecting ensemble docking performance, including the sampling accuracy of the docking algorithm, the choice of the scoring function, and the similarity of database ligands to the cocrystallized ligands of ligand-bound protein structures in an ensemble. Due to these factors, the prospective selection of optimum ensembles is a challenging task, shown by a reassessment of published ensemble selection protocols.

Pose prediction and virtual screening performance of GOLD scoring functions in a standardized test

The performance of all four GOLD scoring functions has been evaluated for pose prediction and virtual screening under the standardized conditions of the comparative docking and scoring experiment reported in this Edition. Excellent pose prediction and good virtual screening performance was demonstrated using unmodified protein models and default parameter settings. The best performing scoring function for both pose prediction and virtual screening was demonstrated to be the recently introduced scoring function ChemPLP. We conclude that existing docking programs already perform close to optimally in the cognate pose prediction experiments currently carried out and that more stringent pose prediction tests should be used in the future. These should employ cross-docking sets. Evaluation of virtual screening performance remains problematic and much remains to be done to improve the usefulness of publically available active and decoy sets for virtual screening. Finally we suggest that, for certain target/scoring function combinations, good enrichment may sometimes be a consequence of 2D property recognition rather than a modelling of the correct 3D interactions.

The good, the bad and the twisted: a survey of ligand geometry in protein crystal structures

The protein databank now contains the structures of over 11,000 ligands bound to proteins. These structures are invaluable in applied areas such as structure-based drug design, but are also the substrate for understanding the energetics of intermolecular interactions with proteins. Despite their obvious importance, the careful analysis of ligands bound to protein structures lags behind the analysis of the protein structures themselves. We present an analysis of the geometry of ligands bound to proteins and highlight the role of small molecule crystal structures in enabling molecular modellers to critically evaluate a ligand model’s quality and investigate protein-induced strain.

Systematic conformational bias in small-molecule crystal structures is rare and explicable

Analysis of the Cambridge Structural Database, together with DFT and crystal structure prediction calculations, show that the observation of higher-energy planar conformers of biphenyl (BP) and cyclobutane (CB) is possible because of improved intermolecular interactions in their crystal structures. Such intermolecular/intramolecular energy compensation almost always occurs when crystallographic and molecular symmetry elements coincide. For BP and CB, almost exclusively, a crystallographic inversion centre coincides with a centre of symmetry in a Ci-symmetric molecule. We conclude that the observation of higher energy conformers (with the compensation of conformational energies up to ≈8–10 kJ.mol.−1 above the global minimum) together with this symmetry coincidence is rare. The work shows that concerns, expressed by some drug discovery chemists and other scientists, that conformations observed in crystal structures are systematically biased due to ‘crystal packing effects’ is overstated: only 16% of BP and CB fragments are exactly planar in small-molecule crystal structures, while the remaining conformations are close to their gas-phase energy minima. Thus, crystal structure conformations are good guides to conformational preferences in other phases and in other applications, e.g. in conformer generation or in the study of protein–ligand binding.

Validating and understanding ring conformations using small molecule crystallographic data.

Understanding the conformational preferences of ring structures is fundamental to structure-based drug design. Although the Cambridge Structural Database (CSD) provides information on the preferred conformations of small molecules, analyzing this data can be very time-consuming. In order to overcome this hurdle, tools have been developed for quickly extracting geometrical preferences from the CSD. Here we describe how the program Mogul has been extended to analyze and compare ring conformations, using a library derived from over 900 000 ring fragments in the CSD. We illustrate how these can be used to understand the conformational preferences of molecules in a crystal lattice and bound to proteins.

Computational design of a lipase for catalysis of the Diels-Alder reaction

AbstractCombined molecular docking, molecular dynamics (MD) and density functional theory (DFT) studies have been employed to study catalysis of the Diels-Alder reaction by a modified lipase. Six variants of the versatile enzyme Candida Antarctica lipase B (CALB) have been rationally engineered in silico based on the specific characteristics of the pericyclic addition. A kinetic analysis reveals that hydrogen bond stabilization of the transition state and substrate binding are key components of the catalytic process. In the case of substrate binding, which has the greater potential for optimization, both binding strength and positioning of the substrates are important for catalytic efficiency. The binding strength is determined by hydrophobic interactions and can be tuned by careful selection of solvent and substrates. The MD simulations show that substrate positioning is sensitive to cavity shape and size, and can be controlled by a few rational mutations. The well-documented S105A mutation is essential to enable sufficient space in the vicinity of the oxyanion hole. Moreover, bulky residues on the edge of the active site hinders the formation of a sandwich-like nearattack conformer (NAC), and the I189A mutation is needed to obtain enough space above the face of the α,β-double bond on the dienophile. The double mutant S105A/I189A performs quite well for two of three dienophiles. Based on binding constants and NAC energies obtained from MD simulations combined with activation energies from DFT computations, relative catalytic rates (vcat/vuncat) of up to 103 are predicted. FigureUsing a combination of molecular dynamics simulations and quantum chemical calculations, it is demonstrated that a few rational mutations can improve the catalytic activity of a lipase towards the Diels-Alder reaction.

Structural and docking studies of potent ethionamide boosters.

Tuberculosis remains the second only to HIV as the leading cause of death by infectious disease worldwide, and was responsible for 1.4 million deaths globally in 2011. One of the essential drugs of the second-line antitubercular regimen is the prodrug ethionamide, introduced in the 1960s. Ethionamide is primarily used in cases of multi-drug resistant (MDR) and extensively drug resistant (XDR) TB due to severe adverse side effects. As a prodrug, ethionamide is bioactivated by EthA, a mono-oxygenase whose activity is repressed by EthR, a member of the TetR family of regulators. Previous studies have established that inhibition of EthR improves ethionamide potency. We report here the crystal structures of three EthR inhibitors at 0.8 Å resolution (3-oxo-3-{4-[3-(thiophen-2-yl)-1,2,4-oxadiazol-5-yl]piperidin-1-yl}propanenitrile (BDM31343), 4,4,4-trifluoro-1-{4-[3-(6-methoxy-1,3-benzothiazol-2-yl)-1,2,4-oxadiazol-5-yl]piperidin-1-yl}butanone (BDM41325) and 5,5,5-trifluoro-1-{4-[3-(4-methanesulfonylphenyl)-1,2,4-oxadiazol-5-yl]piperidin-1-yl}pentanone (BDM41907)), and the docking studies undertaken to investigate possible binding modes. The results revealed two distinct orientations of the three compounds in the binding channel, a direct consequence of the promiscuous nature of the largely lipophilic binding site.

Ensemble docking revisited

In recent years, the importance of considering induced fit effects in molecular docking calculations has been widely recognised in the molecular modelling community. While small-scale protein side-chain movements are now accounted for in many state-of-the-art docking strategies, the explicit modelling of large-scale protein motions such as loop movements in kinase domains is still a challenging task. For this reason ensemble-based methods have been introduced taking into account several discrete protein conformations in the conformational sampling step. Our protein-ligand docking approach GOLD [1,2] has been extended to search such conformational ensembles time-efficiently. The performance of the approach has been assessed on several protein targets using different scoring functions. A detailed analysis of pose prediction and virtual screening results in dependence of the number of protein structures considered in the conformational ensemble will be presented and limitations of the approach will be highlighted.