A challenging system: Free energy prediction for factor Xa

Statistical Analysis on the Performance of Molecular Mechanics Poisson-Boltzmann Surface Area versus Absolute Binding Free Energy Calculations: Bromodomains as a Case Study

Binding free energy calculations that make use of alchemical pathways are becoming increasingly feasible thanks to advances in hardware and algorithms. Although relative binding free energy (RBFE) calculations are starting to find widespread use, absolute binding free energy (ABFE) calculations are still being explored mainly in academic settings due to the high computational requirements and still uncertain predictive value. However, in some drug design scenarios, RBFE calculations are not applicable and ABFE calculations could provide an alternative. Computationally cheaper end-point calculations in implicit solvent, such as molecular mechanics Poisson-Boltzmann surface area (MMPBSA) calculations, could too be used if one is primarily interested in a relative ranking of affinities. Here, we compare MMPBSA calculations to previously performed absolute alchemical free energy calculations in their ability to correlate with experimental binding free energies for three sets of bromodomain-inhibitor pairs. Different MMPBSA approaches have been considered, including a standard single-trajectory protocol, a protocol that includes a binding entropy estimate, and protocols that take into account the ligand hydration shell. Despite the improvements observed with the latter two MMPBSA approaches, ABFE calculations were found to be overall superior in obtaining correlation with experimental affinities for the test cases considered. A difference in weighted average Pearson ([Formula: see text]) and Spearman ([Formula: see text]) correlations of 0.25 and 0.31 was observed when using a standard single-trajectory MMPBSA setup ([Formula: see text] = 0.64 and [Formula: see text] = 0.66 for ABFE; [Formula: see text] = 0.39 and [Formula: see text] = 0.35 for MMPBSA). The best performing MMPBSA protocols returned weighted average Pearson and Spearman correlations that were about 0.1 inferior to ABFE calculations: [Formula: see text] = 0.55 and [Formula: see text] = 0.56 when including an entropy estimate, and [Formula: see text] = 0.53 and [Formula: see text] = 0.55 when including explicit water molecules. Overall, the study suggests that ABFE calculations are indeed the more accurate approach, yet there is also value in MMPBSA calculations considering the lower compute requirements, and if agreement to experimental affinities in absolute terms is not of interest. Moreover, for the specific protein-ligand systems considered in this study, we find that including an explicit ligand hydration shell or a binding entropy estimate in the MMPBSA calculations resulted in significant performance improvements at a negligible computational cost.

Localization of Millisecond Dynamics: Dihedral Entropy from Accelerated MD

Here, we demonstrate a method to capture local dynamics on a time scale 3 orders of magnitude beyond state-of-the-art simulation approaches. We apply accelerated molecular dynamics simulations for conformational sampling and extract reweighted backbone dihedral distributions. Local dynamics are characterized by torsional probabilities, resulting in residue-wise dihedral entropies. Our approach is successfully validated for three different protein systems of increasing size: alanine dipeptide, bovine pancreatic trypsin inhibitor (BPTI), and the major birch pollen allergen Bet v 1a. We demonstrate excellent agreement of flexibility profiles with both large-scale computer simulations and NMR experiments. Thus, our method provides efficient access to local protein dynamics on extended time scales of high biological relevance.

Kinetic barriers in the isomerization of substituted ureas: implications for computer-aided drug design

Urea derivatives are ubiquitously found in many chemical disciplines. N,N'-substituted ureas may show different conformational preferences depending on their substitution pattern. The high energetic barrier for isomerization of the cis and trans state poses additional challenges on computational simulation techniques aiming at a reproduction of the biological properties of urea derivatives. Herein, we investigate energetics of urea conformations and their interconversion using a broad spectrum of methodologies ranging from data mining, via quantum chemistry to molecular dynamics simulation and free energy calculations. We find that the inversion of urea conformations is inherently slow and beyond the time scale of typical simulation protocols. Therefore, extra care needs to be taken by computational chemists to work with appropriate model systems. We find that both knowledge-driven approaches as well as physics-based methods may guide molecular modelers towards accurate starting structures for expensive calculations to ensure that conformations of urea derivatives are modeled as adequately as possible.

Quantitative Correlation of Conformational Binding Enthalpy with Substrate Specificity of Serine Proteases

Members of the same protease family show different substrate specificity, even if they share identical folds, depending on the physiological processes they are part of. Here, we investigate the key factors for subpocket and global specificity of factor Xa, elastase, and granzyme B which despite all being serine proteases and sharing the chymotrypsin-fold show distinct substrate specificity profiles. We determined subpocket interaction potentials with GRID for static X-ray structures and an in silico generated ensemble of conformations. Subpocket interaction potentials determined for static X-ray structures turned out to be insufficient to explain serine protease specificity for all subpockets. Therefore, we generated conformational ensembles using molecular dynamics simulations. We identified representative binding site conformations using distance-based hierarchical agglomerative clustering and determined subpocket interaction potentials for each representative conformation of the binding site. Considering the differences in subpocket interaction potentials for these representative conformations as well as their abundance allowed us to quantitatively explain subpocket specificity for the nonprime side for all three example proteases on a molecular level. The methods to identify key regions determining subpocket specificity introduced in this study are directly applicable to other serine proteases, and the results provide starting points for new strategies in rational drug design.

Dynamics Govern Specificity of a Protein-Protein Interface: Substrate Recognition by Thrombin

Biomolecular recognition is crucial in cellular signal transduction. Signaling is mediated through molecular interactions at protein-protein interfaces. Still, specificity and promiscuity of protein-protein interfaces cannot be explained using simplistic static binding models. Our study rationalizes specificity of the prototypic protein-protein interface between thrombin and its peptide substrates relying solely on binding site dynamics derived from molecular dynamics simulations. We find conformational selection and thus dynamic contributions to be a key player in biomolecular recognition. Arising entropic contributions complement chemical intuition primarily reflecting enthalpic interaction patterns. The paradigm “dynamics govern specificity” might provide direct guidance for the identification of specific anchor points in biomolecular recognition processes and structure-based drug design.

The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities

INTRODUCTION

The molecular mechanics energies combined with the Poisson-Boltzmann or generalized Born and surface area continuum solvation (MM/PBSA and MM/GBSA) methods are popular approaches to estimate the free energy of the binding of small ligands to biological macromolecules. They are typically based on molecular dynamics simulations of the receptor-ligand complex and are therefore intermediate in both accuracy and computational effort between empirical scoring and strict alchemical perturbation methods. They have been applied to a large number of systems with varying success.

AREAS COVERED

The authors review the use of MM/PBSA and MM/GBSA methods to calculate ligand-binding affinities, with an emphasis on calibration, testing and validation, as well as attempts to improve the methods, rather than on specific applications.

EXPERT OPINION

MM/PBSA and MM/GBSA are attractive approaches owing to their modular nature and that they do not require calculations on a training set. They have been used successfully to reproduce and rationalize experimental findings and to improve the results of virtual screening and docking. However, they contain several crude and questionable approximations, for example, the lack of conformational entropy and information about the number and free energy of water molecules in the binding site. Moreover, there are many variants of the method and their performance varies strongly with the tested system. Likewise, most attempts to ameliorate the methods with more accurate approaches, for example, quantum-mechanical calculations, polarizable force fields or improved solvation have deteriorated the results.

Interface dynamics explain assembly dependency of influenza neuraminidase catalytic activity

Influenza virus neuraminidase (iNA) is a homotetrameric surface protein of the influenza virus and an established target for antiviral drugs. In contrast to neuraminidases (NAs) of other biological systems (non-iNAs), enzymatic activity of iNA is only observed in a quaternary assembly and iNA needs the tetramerization to mediate enzymatic activity. Obviously, differences on a molecular level between iNA and non-iNAs are responsible for this intriguing observation. Comparison between protein structures and multiple sequence alignment allow the identification of differences in amino acid composition in crucial regions of the enzyme, such as next to the conserved D151 and the 150-loop. These differences in amino acid sequence and protein tetramerization are likely to alter the dynamics of the system. Therefore, we performed molecular dynamics simulations to investigate differences in the molecular flexibility of monomers, dimers, and tetramers of iNAs of subtype N1 (avian 2004, pandemic 1918 and pandemic 2009 iNA) and as comparison the non-iNA monomer from Clostridium perfringens. We show that conformational transitions of iNA are crucially influenced by its assembly state. The protein-protein interface induces a complex hydrogen-bonding network between the 110-helix and the 150-loop, which consequently stabilizes the structural arrangement of the binding site. Therefore, we claim that these altered dynamics are responsible for the dependence of iNA's catalytic activity on the tetrameric assembly. Only the tetramerization-induced balance between stabilization and altered local flexibility in the binding site provides the appropriate arrangement of key residues for iNA's catalytic activity.

Explicit Ligand Hydration Shells Improve the Correlation between MM-PB/GBSA Binding Energies and Experimental Activities

Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) and Molecular Mechanics Generalized Born Surface Area (MM-GBSA) methods are widely used for drug design/discovery purposes. However, it is not clear if the correlation between predicted and experimental binding affinities can be improved by explicitly considering selected water molecules in the calculation of binding energies, since different and sometimes diverging opinions are found in the literature. In this work, we evaluated how variably populated hydration shells explicitly considered around the ligands may affect the correlation between MM-PB/GBSA computed binding energy and biological activities (IC50 and ΔGbind, depending on the available experimental data). Four different systems-namely, the DNA-topoisomerase complex, α-thrombin, penicillopepsin, and avidin-were considered and ligand hydration shells populated by 10-70 water molecules were systematically evaluated. We found that the consideration of a hydration shell populated by a number of water residues (Nwat) between 30 and 70 provided, in all of the considered examples, a positive effect on correlation between MM-PB/GBSA calculated binding affinities and experimental activities, with a negligible increment of computational cost.

Rapid and accurate prediction and scoring of water molecules in protein binding sites

Water plays a critical role in ligand-protein interactions. However, it is still challenging to predict accurately not only where water molecules prefer to bind, but also which of those water molecules might be displaceable. The latter is often seen as a route to optimizing affinity of potential drug candidates. Using a protocol we call WaterDock, we show that the freely available AutoDock Vina tool can be used to predict accurately the binding sites of water molecules. WaterDock was validated using data from X-ray crystallography, neutron diffraction and molecular dynamics simulations and correctly predicted 97% of the water molecules in the test set. In addition, we combined data-mining, heuristic and machine learning techniques to develop probabilistic water molecule classifiers. When applied to WaterDock predictions in the Astex Diverse Set of protein ligand complexes, we could identify whether a water molecule was conserved or displaced to an accuracy of 75%. A second model predicted whether water molecules were displaced by polar groups or by non-polar groups to an accuracy of 80%. These results should prove useful for anyone wishing to undertake rational design of new compounds where the displacement of water molecules is being considered as a route to improved affinity.