Modeling the value of predictive affinity scoring in preclinical drug discovery.

Drug discovery is widely recognized to be a difficult and costly activity in large part due to the challenge of identifying chemical matter which simultaneously optimizes multiple properties, one of which is affinity for the primary biological target. Further, many of these properties are difficult to predict ahead of expensive and time-consuming compound synthesis and experimental testing. Here we highlight recent work to develop compound affinity prediction models, and extensively investigate the value such models may provide to preclinical drug discovery. We demonstrate that the ability of these models to improve the overall probability of success is crucially dependent on the shape of the error distribution, not just the root-mean-square error. In particular, while scoring more molecule ideas generally improves the probability of project success when the error distribution is Gaussian, fat-tail distributions such as a Cauchy distribution, can lead to a situation where scoring more ideas actually decreases the overall probability of success.

A novel inhibitor of active protein kinase G attenuates chronic inflammatory and osteoarthritic pain.

Activating PKG-1α induces a long-term hyperexcitability (LTH) in nociceptive neurons. Since the LTH correlates directly with chronic pain in many animal models, we tested the hypothesis that inhibiting PKG-1α would attenuate LTH-mediated pain. We first synthesized and characterized compound N46 (N-((3R,4R)-4-(4-(2-fluoro-3-methoxy-6-propoxybenzoyl)benzamido)pyrrolidin-3-yl)-1H-indazole-5-carboxamide). N46 inhibits PKG-1α with an IC50 of 7.5 nmol, was highly selective when tested against a panel of 274 kinases, and tissue distribution studies indicate that it does not enter the CNS. To evaluate its antinociceptive potential, we used 2 animal models in which the pain involves both activated PKG-1α and LTH. Injecting complete Freund's adjuvant (CFA) into the rat hind paw causes a thermal hyperalgesia that was significantly attenuated 24 hours after a single intravenous injection of N46. Next, we used a rat model of osteoarthritic knee joint pain and found that a single intra-articular injection of N46 alleviated the pain 14 days after the pain was established and the relief lasted for 7 days. Thermal hyperalgesia and osteoarthritic pain are also associated with the activation of the capsaicin-activated transient receptor protein vanilloid-1 (TRPV1) channel. We show that capsaicin activates PKG-1α in nerves and that a subcutaneous delivery of N46 attenuated the mechanical and thermal hypersensitivity elicited by exposure to capsaicin. Thus, PKG-1α appears to be downstream of the transient receptor protein vanilloid-1. Our studies provide proof of concept in animal models that a PKG-1α antagonist has a powerful antinociceptive effect on persistent, already existing inflammatory pain. They further suggest that N46 is a valid chemotype for the further development of such antagonists.

WScore: A Flexible and Accurate Treatment of Explicit Water Molecules in Ligand-Receptor Docking.

We have developed a new methodology for protein-ligand docking and scoring, WScore, incorporating a flexible description of explicit water molecules. The locations and thermodynamics of the waters are derived from a WaterMap molecular dynamics simulation. The water structure is employed to provide an atomic level description of ligand and protein desolvation. WScore also contains a detailed model for localized ligand and protein strain energy and integrates an MM-GBSA scoring component with these terms to assess delocalized strain of the complex. Ensemble docking is used to take into account induced fit effects on the receptor conformation, and protein reorganization free energies are assigned via fitting to experimental data. The performance of the method is evaluated for pose prediction, rank ordering of self-docked complexes, and enrichment in virtual screening, using a large data set of PDB complexes and compared with the Glide SP and Glide XP models; significant improvements are obtained.

Implications of Dynamic Occupancy, Binding Kinetics, and Channel Gating Kinetics for hERG Blocker Safety Assessment and Mitigation.

Blockade of the hERG potassium channel prolongs the ventricular action potential (AP) and QT interval, and triggers early after depolarizations (EADs) and torsade de pointes (TdP) arrhythmia. Opinions differ as to the causal relationship between hERG blockade and TdP, the relative weighting of other contributing factors, definitive metrics of preclinical proarrhythmicity, and the true safety margin in humans. Here, we have used in silico techniques to characterize the effects of channel gating and binding kinetics on hERG occupancy, and of blockade on the human ventricular AP. Gating effects differ for compounds that are sterically compatible with closed channels (becoming trapped in deactivated channels) versus those that are incompatible with the closed/closing state, and expelled during deactivation. Occupancies of trappable blockers build to equilibrium levels, whereas those of non-trappable blockers build and decay during each AP cycle. Occupancies of ~83% (non-trappable) versus ~63% (trappable) of open/inactive channels caused EADs in our AP simulations. Overall, we conclude that hERG occupancy at therapeutic exposure levels may be tolerated for nontrappable, but not trappable blockers capable of building to the proarrhythmic occupancy level. Furthermore, the widely used Redfern safety index may be biased toward trappable blockers, overestimating the exposure-IC50 separation in nontrappable cases.

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.

Designing tight-binding ligands is a primary objective of small-molecule drug discovery. Over the past few decades, free-energy calculations 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 commercial application of free-energy simulations has been limited due to the lack of large-scale validation coupled with the technical challenges traditionally associated with running these types of calculations. 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 chemical perturbations, many of which involve significant changes in ligand chemical structures. In addition, we have applied the method in prospective drug discovery projects and found a significant improvement in the quality of the compounds synthesized that have been predicted to be potent. Compounds predicted to be potent by this approach have a substantial reduction in false positives relative to compounds synthesized on the basis of other computational or medicinal chemistry 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.

Docking covalent inhibitors: a parameter free approach to pose prediction and scoring.

Although many popular docking programs include a facility to account for covalent ligands, large-scale systematic docking validation studies of covalent inhibitors have been sparse. In this paper, we present the development and validation of a novel approach for docking and scoring covalent inhibitors, which consists of conventional noncovalent docking, heuristic formation of the covalent attachment point, and structural refinement of the protein-ligand complex. This approach combines the strengths of the docking program Glide and the protein structure modeling program Prime and does not require any parameter fitting for the study of additional covalent reaction types. We first test this method by predicting the native binding geometry of 38 covalently bound complexes. The average RMSD of the predicted poses is 1.52 Å, and 76% of test set inhibitors have an RMSD of less than 2.0 Å. In addition, the apparent affinity score constructed herein is tested on a virtual screening study and the characterization of the SAR properties of two different series of congeneric compounds with satisfactory success.

The translocation kinetics of antibiotics through porin OmpC: insights from structure-based solvation mapping using WaterMap.

Poor permeability of the lipopolysaccharide-based outer membrane of Gram-negative bacteria is compensated by the existence of protein channels (porins) that selectively admit low molecular weight substrates, including many antibiotics. Improved understanding of the translocation mechanisms of porin substrates could help guide the design of antibiotics capable of achieving high intracellular exposure. Energy barriers to channel entry and exit govern antibiotic fluxes through porins. We have previously reported a hypothesis that the costs of transferring protein solvation to and from bulk medium underlie the barriers to protein-ligand association and dissociation, respectively, concomitant with the gain and loss of protein-ligand interactions during those processes. We have now applied this hypothesis to explain the published rates of entry (association) and exit (dissociation) of six antibiotics to/from reconstituted E. coli porin OmpC. WaterMap was used to estimate the total water transfer energies resulting from transient occupation by each antibiotic. Our results suggest that solvation within the porin cavity is highly energetically favorable, and the observed moderately fast entry rates of the antibiotics are consistent with replacement of protein-water H-bonds. The observed ultrafast exit kinetics is consistent with the lack of intrachannel solvation sites that convey unfavorable resolvation during antibiotic dissociation. These results are aligned with known general relationships between antibiotic efficacy and physicochemical properties, namely unusually low logP, reflecting an abundance of H-bond partners. We conclude that antibiotics figuratively "melt" their way through porin solvation at a rate determined by the cost of exchanging protein-solvent for protein-antibiotic H-bonds.

Significant reduction in errors associated with nonbonded contacts in protein crystal structures: automated all-atom refinement with PrimeX.

All-atom models are essential for many applications in molecular modeling and computational chemistry. Nonbonded atomic contacts much closer than the sum of the van der Waals radii of the two atoms (clashes) are commonly observed in such models derived from protein crystal structures. A set of 94 recently deposited protein structures in the resolution range 1.5-2.8 Å were analyzed for clashes by the addition of all H atoms to the models followed by optimization and energy minimization of the positions of just these H atoms. The results were compared with the same set of structures after automated all-atom refinement with PrimeX and with nonbonded contacts in protein crystal structures at a resolution equal to or better than 0.9 Å. The additional PrimeX refinement produced structures with reasonable summary geometric statistics and similar R(free) values to the original structures. The frequency of clashes at less than 0.8 times the sum of van der Waals radii was reduced over fourfold compared with that found in the original structures, to a level approaching that found in the ultrahigh-resolution structures. Moreover, severe clashes at less than or equal to 0.7 times the sum of atomic radii were reduced 15-fold. All-atom refinement with PrimeX produced improved crystal structure models with respect to nonbonded contacts and yielded changes in structural details that dramatically impacted on the interpretation of some protein-ligand interactions.

Diversity-oriented synthesis of a library of substituted tetrahydropyrones using oxidative carbon-hydrogen bond activation and click chemistry.

Eighteen (2RS,6RS)-2-(4-methoxyphenyl)-6-(substituted ethyl)dihydro-2H-pyran-4(3H)ones were synthesized via a DDQ-mediated oxidative carbon-hydrogen bond activation reaction. Fourteen of these tetrahydropyrans were substituted with triazoles readily assembled via azide-alkyne click-chemistry reactions. Examples of a linked benzotriazole and pyrazole motif were also prepared. To complement the structural diversity, the alcohol substrates were obtained from stereoselective reductions of the tetrahydropyrone. This library provides rapid access to structurally diverse non-natural compounds to be screened against a variety of biological targets.

Contribution of explicit solvent effects to the binding affinity of small-molecule inhibitors in blood coagulation factor serine proteases.

The prevention of blood coagulation is important in treating thromboembolic disorders, and several serine proteases involved in the coagulation cascade have been classified as pharmaceutically relevant. Whereas structure-based drug design has contributed to the development of some serine protease inhibitors, traditional computational methods have not been able to fully describe structure-activity relationships (SAR). Here, we study the SAR for a number of serine proteases by using a method that calculates the thermodynamic properties (enthalpy and entropy) of the water that solvates the active site. We show that the displacement of water from specific subpockets (such as S1-4 and the ester binding pocket) of the active site by the ligand can govern potency, especially for cases in which small chemical changes (i.e., a methyl group or halogen) result in a substantial increase in potency. Furthermore, we describe how relative binding free energies can be estimated by combining the water displacement energy with complementary terms from an implicit solvent molecular mechanics description binding.

IDSite: An accurate approach to predict P450-mediated drug metabolism.

Accurate prediction of drug metabolism is crucial for drug design. Since a large majority of drugs metabolism involves P450 enzymes, we herein describe a computational approach, IDSite, to predict P450-mediated drug metabolism. To model induced-fit effects, IDSite samples the conformational space with flexible docking in Glide followed by two refinement stages using the Protein Local Optimization Program (PLOP). Sites of metabolism (SOMs) are predicted according to a physical-based score that evaluates the potential of atoms to react with the catalytic iron center. As a preliminary test, we present in this paper the prediction of hydroxylation and O-dealkylation sites mediated by CYP2D6 using two different models: a physical-based simulation model, and a modification of this model in which a small number of parameters are fit to a training set. Without fitting any parameters to experimental data, the Physical IDSite scoring recovers 83% of the experimental observations for 56 compounds with a very low false positive rate. With only 4 fitted parameters, the Fitted IDSite was trained with the subset of 36 compounds and successfully applied to the other 20 compounds, recovering 94% of the experimental observations with high sensitivity and specificity for both sets.

Understanding kinase selectivity through energetic analysis of binding site waters.

Kinases remain an important drug target class within the pharmaceutical industry; however, the rational design of kinase inhibitors is plagued by the complexity of gaining selectivity for a small number of proteins within a family of more than 500 related enzymes. Herein we show how a computational method for identifying the location and thermodynamic properties of water molecules within a protein binding site can yield insight into previously inexplicable selectivity and structure-activity relationships. Four kinase systems (Src family, Abl/c-Kit, Syk/ZAP-70, and CDK2/4) were investigated, and differences in predicted water molecule locations and energetics were able to explain the experimentally observed binding selectivity profiles. The successful predictions across the range of kinases studied here suggest that this methodology could be generally applicable for predicting selectivity profiles in related targets.

A role for a specific cholesterol interaction in stabilizing the Apo configuration of the human A(2A) adenosine receptor.

The function of G-protein-coupled receptors is tightly modulated by the lipid environment. Long-timescale molecular dynamics simulations (totaling approximately 3 mus) of the A(2A) receptor in cholesterol-free bilayers, with and without the antagonist ZM241385 bound, demonstrate the instability of helix II in the apo receptor in cholesterol-poor membrane regions. We directly observe that the effect of cholesterol binding is to stabilize helix II against a buckling-type deformation, perhaps rationalizing the observation that the A(2A) receptor couples to G protein only in the presence of cholesterol (Zezula and Freissmuth, 2008). The results suggest a mechanism by which the A(2A) receptor may function as a coincidence detector, activating only in the presence of both cholesterol and agonist. We also observed a previously hypothesized conformation of the tryptophan "rotameric switch" on helix VI in which a phenylalanine on helix V positions the tryptophan out of the ligand binding pocket.

High-energy water sites determine peptide binding affinity and specificity of PDZ domains.

PDZ domains have well known binding preferences for distinct C-terminal peptide motifs. For most PDZ domains, these motifs are of the form [S/T]-W-[I/L/V]. Although the preference for S/T has been explained by a specific hydrogen bond interaction with a histidine in the PDZ domain and the (I/L/V) is buried in a hydrophobic pocket, the mechanism for Trp specificity at the second to last position has thus far remained unknown. Here, we apply a method to compute the free energies of explicit water molecules and predict that potency gained by Trp binding is due to a favorable release of high-energy water molecules into bulk. The affinities of a series of peptides for both wild-type and mutant forms of the PDZ domain of Erbin correlate very well with the computed free energy of binding of displaced waters, suggesting a direct relationship between water displacement and peptide affinity. Finally, we show a correlation between the magnitude of the displaced water free energy and the degree of Trp-sensitivity among subtypes of the HTRA PDZ family, indicating a water-mediated mechanism for specificity of peptide binding.

Bonded exciplex formation: electronic and stereoelectronic effects.

As recently proposed, the singlet-excited states of several cyanoaromatics react with pyridine via bonded-exciplex formation, a novel concept in photochemical charge transfer reactions. Presented here are electronic and steric effects on the quenching rate constants, which provide valuable support for the model. Additionally, excited-state quenching in poly(vinylpyridine) is strongly inhibited both relative to that in neat pyridine and also to conventional exciplex formation in polymers, consistent with a restrictive orientational requirement for the formation of bonded exciplexes. Examples of competing reactions to form both conventional and bonded exciplexes are presented, which illustrate the delicate balance between these two processes when their reaction energetics are similar. Experimental and computational evidence is provided for the formation of a bonded exciplex in the reaction of the singlet excited state of 2,6,9,10-tetracyanoanthracene (TCA) with an oxygen-substituted donor, dioxane, thus expanding the scope of bonded exciplexes.

Improving database enrichment through ensemble docking.

While it may seem intuitive that using an ensemble of multiple conformations of a receptor in structure-based virtual screening experiments would necessarily yield improved enrichment of actives relative to using just a single receptor, it turns out that at least in the p38 MAP kinase model system studied here, a very large majority of all possible ensembles do not yield improved enrichment of actives. However, there are combinations of receptor structures that do lead to improved enrichment results. We present here a method to select the ensembles that produce the best enrichments that does not rely on knowledge of active compounds or sophisticated analyses of the 3D receptor structures. In the system studied here, the small fraction of ensembles of up to 3 receptors that do yield good enrichments of actives were identified by selecting ensembles that have the best mean GlideScore for the top 1% of the docked ligands in a database screen of actives and drug-like "decoy" ligands. Ensembles of two receptors identified using this mean GlideScore metric generally outperform single receptors, while ensembles of three receptors identified using this metric consistently give optimal enrichment factors in which, for example, 40% of the known actives outrank all the other ligands in the database.

Role of the active-site solvent in the thermodynamics of factor Xa ligand binding.

Understanding the underlying physics of the binding of small-molecule ligands to protein active sites is a key objective of computational chemistry and biology. It is widely believed that displacement of water molecules from the active site by the ligand is a principal (if not the dominant) source of binding free energy. Although continuum theories of hydration are routinely used to describe the contributions of the solvent to the binding affinity of the complex, it is still an unsettled question as to whether or not these continuum solvation theories describe the underlying molecular physics with sufficient accuracy to reliably rank the binding affinities of a set of ligands for a given protein. Here we develop a novel, computationally efficient descriptor of the contribution of the solvent to the binding free energy of a small molecule and its associated receptor that captures the effects of the ligand displacing the solvent from the protein active site with atomic detail. This descriptor quantitatively predicts (R(2) = 0.81) the binding free energy differences between congeneric ligand pairs for the test system factor Xa, elucidates physical properties of the active-site solvent that appear to be missing in most continuum theories of hydration, and identifies several features of the hydration of the factor Xa active site relevant to the structure-activity relationship of its inhibitors.

Bonded exciplexes. A new concept in photochemical reactions.

Charge-transfer quenching of the singlet excited states of cyanoaromatic electron acceptors by pyridine is characterized by a driving force dependence that resembles those of conventional electron-transfer reactions, except that a plot of the log of the quenching rate constants versus the free energy of electron transfer is displaced toward the endothermic region by 0.5-0.8 eV. Specifically, the reactions with pyridine display rapid quenching when conventional electron transfer is highly endothermic. As an example, the rate constant for quenching of the excited dicyanoanthracene is 3.5 x 10(9) M(-1)s(-1), even though formation of a conventional radical ion pair, A*-D*+, is endothermic by approximately 0.6 eV. No long-lived radical ions or exciplex intermediates can be detected on the picosecond to microsecond time scale. Instead, the reactions are proposed to proceed via formation of a previously undescribed, short-lived charge-transfer intermediate we call a "bonded exciplex", A- -D+. The bonded exciplex can be formally thought of as resulting from bond formation between the unpaired electrons of the radical ions A*- and D*+. The covalent bonding interaction significantly lowers the energy of the charge-transfer state. As a result of this interaction, the energy decreases with decreasing separation distance, and near van der Waals contact, the A- -D+ bonded state mixes with the repulsive excited state of the acceptor, allowing efficient reaction to form A- -D+ even when formation of a radical ion pair A*-D*+ is thermodynamically forbidden. Evidence for the bonded exciplex intermediate comes from studies of steric and Coulombic effects on the quenching rate constants and from extensive DFT computations that clearly show a curve crossing between the ground state and the low-energy bonded exciplex state.

Use of an induced fit receptor structure in virtual screening.

Structured-based drug design has traditionally relied on a single receptor structure as a target for docking and screening studies. However, it has become increasingly clear that in many cases where protein flexibility is an issue, it is critical to accurately model ligand-induced receptor movement in order to obtain high enrichment factors. We present a novel protein-ligand docking method that accounts for both ligand and receptor flexibility and accurately predicts the conformation of protein-ligand binding complexes. This method can generate viable receptor ensembles that can be used in virtual database screens.

Novel procedure for modeling ligand/receptor induced fit effects.

We present a novel protein-ligand docking method that accurately accounts for both ligand and receptor flexibility by iteratively combining rigid receptor docking (Glide) with protein structure prediction (Prime) techniques. While traditional rigid-receptor docking methods are useful when the receptor structure does not change substantially upon ligand binding, success is limited when the protein must be "induced" into the correct binding conformation for a given ligand. We provide an in-depth description of our novel methodology and present results for 21 pharmaceutically relevant examples. Traditional rigid-receptor docking for these 21 cases yields an average RMSD of 5.5 A. The average ligand RMSD for docking to a flexible receptor for the 21 pairs is 1.4 A; the RMSD is < or =1.8 A for 18 of the cases. For the three cases with RMSDs greater than 1.8 A, the core of the ligand is properly docked and all key protein/ligand interactions are captured.