Protein flexibility and species specificity in structure-based drug discovery: dihydrofolate reductase as a test system

In structure-based drug discovery, researchers would like to identify all possible scaffolds for a given target. However, techniques that push the boundaries of chemical space could lead to many false positives or inhibitors that lack specificity for the target. Is it possible to broadly identify the appropriate chemical space for the inhibitors and yet maintain target specificity? To address this question, we have turned to dihydrofolate reductase (DHFR), a well-studied metabolic enzyme of pharmacological relevance. We have extended our multiple protein structure (MPS) method for receptor-based pharmacophore models to use multiple X-ray crystallographic structures. Models were created for DHFR from human and Pneumocystis carinii. These models incorporate a fair degree of protein flexibility and are highly selective for known DHFR inhibitors over drug-like non-inhibitors. Despite sharing a highly conserved active site, the pharmacophore models reflect subtle differences between the human and P. carinii forms, which identify species-specific, high-affinity inhibitors. We also use structures of DHFR from Candida albicans as a counter example. The available crystal structures show little flexibility, and the resulting models give poorer performance in identifying species-specific inhibitors. Therapeutic success for this system may depend on achieving species specificity between the related human host and these key fungal targets. The MPS technique is a promising advance for structure-based drug discovery for DHFR and other proteins of biomedical interest.     

Incorporating dynamics in E. coli dihydrofolate reductase enhances structure-based drug discovery

Escherichia coli dihydrofolate reductase (DHFR) is a long-standing target for enzyme studies. The influence of protein motion on its catalytic cycle is significant, and the conformation of the M20 loop is of particular interest. We present receptor-based pharmacophore models-an equivalent of solvent-mapping of binding hotspots-based on ensembles of protein conformations from molecular dynamics simulations of DHFR.NADPH in both the closed and open conformation of the M20 loop. The optimal models identify DHFR inhibitors over druglike non-inhibitors; furthermore, high-affinity inhibitors of E. coli DHFR are preferentially identified over general DHFR inhibitors. As expected, models resulting from simulations with DHFR in the productive conformation with a closed M20 loop have better performance than those from the open-loop simulations. Model performance improves with increased dynamic sampling, indicating that including a greater degree of protein flexibility can enhance the quest for potent inhibitors. This was compared to the limited conformational sampling seen in crystal structures, which were suboptimal for this application.     

Hydrophobic core flexibility modulates enzyme activity in HIV-1 protease

Human immunodeficiency virus Type-1 (HIV-1) protease is crucial for viral maturation and infectivity. Studies of protease dynamics suggest that the rearrangement of the hydrophobic core is essential for enzyme activity. Many mutations in the hydrophobic core are also associated with drug resistance and may modulate the core flexibility. To test the role of flexibility in protease activity, pairs of cysteines were introduced at the interfaces of flexible regions remote from the active site. Disulfide bond formation was confirmed by crystal structures and by alkylation of free cysteines and mass spectrometry. Oxidized and reduced crystal structures of these variants show the overall structure of the protease is retained. However, cross-linking the cysteines led to drastic loss in enzyme activity, which was regained upon reducing the disulfide cross-links. Molecular dynamics simulations showed that altered dynamics propagated throughout the enzyme from the engineered disulfide. Thus, altered flexibility within the hydrophobic core can modulate HIV-1 protease activity, supporting the hypothesis that drug resistant mutations distal from the active site can alter the balance between substrate turnover and inhibitor binding by modulating enzyme activity.     

WinDock: structure-based drug discovery on Windows-based PCs

In recent years, virtual database screening using high-throughput docking (HTD) has emerged as a very important tool and a well-established method for finding new lead compounds in the drug discovery process. With the advent of powerful personal computers (PCs), it is now plausible to perform HTD investigations on these inexpensive PCs. To make HTD more accessible to a broad community, we present here WinDock, an integrated application designed to help researchers perform structure-based drug discovery tasks under a uniform, user friendly graphical interface for Windows-based PCs. WinDock combines existing small molecule searchable three-dimensional (3D) libraries, homology modeling tools, and ligand-protein docking programs in a semi-automatic, interactive manner, which guides the user through the use of each integrated software component. WinDock is coded in C++.     

Specificity rule discovery in HIV-1 protease cleavage site analysis

Several machine learning algorithms have recently been applied to modeling the specificity of HIV-1 protease. The problem is challenging because of the three issues as follows: (1) datasets with high dimensionality and small number of samples could misguide classification modeling and its interpretation; (2) symbolic interpretation is desirable because it provides us insight to the specificity in the form of human-understandable rules, and thus helps us to design effective HIV inhibitors; (3) the interpretation should take into account complexity or dependency between positions in sequences. Therefore, it is necessary to investigate multivariate and feature-selective methods to model the specificity and to extract rules from the model. We have tested extensively various machine learning methods, and we have found that the combination of neural networks and decompositional approach can generate a set of effective rules. By validation to experimental results for the HIV-1 protease, the specificity rules outperform the ones generated by frequency-based, univariate or black-box methods.     

Development of a protease production platform for structure-based drug design

Structure-based drug design (SBDD) has played an integral role in the development of highly specific, potent protease inhibitors resulting in a number of drugs in clinical trials and on the market. Possessing biochemical assays and structural information on both the target protease and homologous family members helps ensure compound selectivity. We have redesigned the path from clone to protein eliminating many of the traditional bottlenecks associated with protein production to ensure a constant supply to feed many diverse protease drug discovery programs. The process was initiated with the design of a multi-system vector, capable of expression in both eukaryotic and prokaryotic hosts; this vector also facilitated high-throughput cloning, expression and purification. When combined into an expression screen, supplemented with salvage screens for detergent extraction and refolding, a route for protein production was established rapidly. Using this process-orientated approach we have successfully expressed and purified all mechanistic classes of active human and viral proteases for enzymatic assays and crystallization studies. While exploiting recent developments in high-throughput biochemistry, we still employ classical biophysical techniques such as light-scattering and analytical ultracentrifugation, to ensure the highest quality protein enters crystallization trials. We have drawn on examples from our own research programs to illustrate how these strategies have been successfully used in the production of proteases for SBDD.     

HIV-1 gp120 V3 loop for structure-based drug design

HIV-1 cell entry is mediated by sequential interactions of the envelope protein gp120 with the receptor CD4 and a coreceptor, usually CCR5 or CXCR4, depending on the individual virion. Considerable efforts on exploiting the HIV coreceptors as drug targets have led to the new class of coreceptor antagonists. While these antiretroviral drugs aim at preventing virus/coreceptor interaction by binding to host proteins, neutralizing antibodies directed against the coreceptor-binding sites on gp120 have attracted attention as possible vaccine candidates. However, both approaches are complicated by the multiple protective mechanisms of gp120 which allow for rapid escape from selective pressures exerted by drugs or antibodies. Thus, advances in rational drug and vaccine design rely heavily on improved insights into the relation between genotype and phenotype, the evolution of coreceptor usage, and, ultimately the structural biology of coreceptor usage and inhibition. The third variable (V3) loop of gp120, crucially involved in all these aspects, will be a major focus of this review.     

Hit diffusion: limitations to drug discovery and structure-based design

Journal of Computer-Aided Molecular Design - Modern drug discovery employs a ‘screening funnel’ to pick compounds worthy of advancing to the clinic, a multi-step process linking a...     

Impact of computational structure-based predictive toxicology in drug discovery

Computational tools for predicting toxicity have been envisioned to have the potential to broadly impact up on the attrition rate of compounds in pre-clinical drug discovery and development. An integrated approach of computer-assisted, predictive, and physico-chemical properties of a compound, along with its in vitro and in vivo analysis, needs to be routinely exercised in the lead identification and lead optimization processes. Starting with a good lead can save a lot of money and it can significantly reduce the entire drug discovery process. The journey towards triple R's- reduce, replace and refine, further proves to be successful in predicting adverse drug reactions in patients (or animals) enrolled in clinical trials. However, the impact of predictive toxicity analysis was modest and relatively narrow in scope, due to the limited domain knowledge in this field. It is important to note that advances within medical science and newer approaches in drug development will require predictive toxicology applications to be viable. The field of computational toxicology has been heading in a direction more relevant to human diseases by reducing the adverse drug reactions. Therefore, efforts must be directed to integrating these tools relevant to the goal of preventing undesired toxicity in pre-clinical trials followed by different phases of clinical trials.     

Combating susceptibility to drug resistance: lessons from HIV-1 protease

Drug resistance is a major obstacle in modern medicine. However, resistance is rarely considered in drug development and may inadvertently be facilitated, as many designed inhibitors contact residues that can mutate to confer resistance, without significantly impairing function. Contemporary drug design often ignores the detailed atomic basis for function and primarily focuses on disrupting the target's activity, which is necessary but not sufficient for developing a robust drug. In this study, we examine the impact of drug-resistant mutations in HIV-1 protease on substrate recognition and demonstrate that most primary active site mutations do not extensively contact substrates, but are critical to inhibitor binding. We propose a general, structure-based strategy to reduce the probability of drug resistance by designing inhibitors that interact only with those residues that are essential for function.     

In silico fragment-mapping method: a new tool for fragment-based/structure-based drug discovery

Here, we propose an in silico fragment-mapping method as a potential tool for fragment-based/structure-based drug discovery (FBDD/SBDD). For this method, we created a database named Canonical Subsite–Fragment DataBase (CSFDB) and developed a knowledge-based fragment-mapping program, Fsubsite. CSFDB consists of various pairs of subsite–fragments derived from X-ray crystal structures of known protein–ligand complexes. Using three-dimensional similarity-matching between subsites on one protein and another, Fsubsite compares the surface of a target protein with all subsites in CSFDB. When a local topography similar to the subsite is found on the surface, Fsubsite places a fragment combined with the subsite in CSFDB on the target protein. For validation purposes, we applied the method to the apo-structure of cyclin-dependent kinase 2 (CDK2) and identified four compounds containing three mapped fragments that existed in the list of known inhibitors of CDK2. Next, the utility of our fragment-mapping method for fragment-growing was examined on the complex structure of tRNA-guanine transglycosylase with a small ligand. Fsubsite mapped appropriate fragments on the same position as the binding ligand or in the vicinity of the ligand. Finally, a 3D-pharmacophore model was constructed from the fragments mapped on the apo-structure of heat shock protein 90-α (HSP90α). Then, 3D pharmacophore-based virtual screening was carried out using a commercially available compound database. The resultant hit compounds were very similar to a known ligand of HSP90α. As a result of these findings, this in silico fragment-mapping method seems to be a useful tool for computational FBDD and SBDD.     

Novel, customizable scoring functions, parameterized using N-PLS, for structure-based drug discovery

The ability to accurately predict biological affinity on the basis of in silico docking to a protein target remains a challenging goal in the CADD arena. Typically, "standard" scoring functions have been employed that use the calculated docking result and a set of empirical parameters to calculate a predicted binding affinity. To improve on this, we are exploring novel strategies for rapidly developing and tuning "customized" scoring functions tailored to a specific need. In the present work, three such customized scoring functions were developed using a set of 129 high-resolution protein-ligand crystal structures with measured Ki values. The functions were parametrized using N-PLS (N-way partial least squares), a multivariate technique well-known in the 3D quantitative structure-activity relationship field. A modest correlation between observed and calculated pKi values using a standard scoring function (r2 = 0.5) could be improved to 0.8 when a customized scoring function was applied. To mimic a more realistic scenario, a second scoring function was developed, not based on crystal structures but exclusively on several binding poses generated with the Flo+ docking program. Finally, a validation study was conducted by generating a third scoring function with 99 randomly selected complexes from the 129 as a training set and predicting pKi values for a test set that comprised the remaining 30 complexes. Training and test set r2 values were 0.77 and 0.78, respectively. These results indicate that, even without direct structural information, predictive customized scoring functions can be developed using N-PLS, and this approach holds significant potential as a general procedure for predicting binding affinity on the basis of in silico docking.     

Interpretation of protein/ligand crystal structure using QM/MM calculations: case of HIV-1 protease/metallacarborane complex

Deltahedral metallacarborane compounds have recently been discovered as potent, specific, stable, and nontoxic inhibitors of HIV-1 protease (PR), the major target for AIDS therapy. The 2.15 A-resolution X-ray structure has exhibited a nonsymmetrical binding of the parental compound [Co(3+)-(C2B9H11)2](-) (GB-18) into PR dimer and a symmetrical arrangement in the crystal of two PR dimer complexes into a tetramer. In order to explore structural and energetic details of the inhibitor binding, quantum mechanics coupled with molecular mechanics approach was utilized. Realizing the close positioning of anionic inhibitors in the active site cavity, the possibility of an exchange of structural water molecules Wat50 and Wat128 by Na+ counterions was studied. The energy profiles for the rotation of the GB-18 molecules along their longitudinal axes in complex with PR were calculated. The results show that two Na+ counterions are present in the active site cavity and provide energetically favorable and unfavorable positions for carbon atoms within the carborane cages. Eighty-one rotamer combinations of four molecules of GB-18 bound to PR out of 4 x 10(5) are predicted to be highly populated. These results lay ground for further calculations of interaction energies between GB-18 and amino acids of PR active site and will make it possible to interpret computationally the binding of similar metallacarborane molecules to PR as well as to resistant PR variants. Moreover, this computational tool will allow the design of new, more potent metallacarborane-based HIV-1 protease inhibitors.     

PRO_SELECT: Combining structure-based drug design and combinatorial chemistry for rapid lead discovery. 1. Technology

This paper describes a novel methodology, PRO_SELECT, which combines elements of structure-based drug design and combinatorial chemistry to create a new paradigm for accelerated lead discovery. Starting with a synthetically accessible template positioned in the active site of the target of interest, PRO_SELECT employs database searching to generate lists of potential substituents for each substituent position on the template. These substituents are selected on the basis of their being able to couple to the template using known synthetic routes and their possession of the correct functionality to interact with specified residues in the active site. The lists of potential substituents are then screened computationally against the active site using rapid algorithms. An empirical scoring function, correlated to binding free energy, is used to rank the substituents at each position. The highest scoring substituents at each position can then be examined using a variety of techniques and a final selection is made. Combinatorial enumeration of the final lists generates a library of synthetically accessible molecules, which may then be prioritised for synthesis and assay. The results obtained using PRO_SELECT to design thrombin inhibitors are briefly discussed.     

Potent HIV-1 protease inhibitors incorporating meso-bicyclic urethanes as P2-ligands: structure-based design, synthesis, biological evaluation and protein-ligand X-ray studies

Recently, we designed a series of novel HIV-1 protease inhibitors incorporating a stereochemically defined bicyclic fused cyclopentyl (Cp-THF) urethane as the high affinity P2-ligand. Inhibitor with this P2-ligand has shown very impressive potency against multi-drug-resistant clinical isolates. Based upon the -bound HIV-1 protease X-ray structure, we have now designed and synthesized a number of meso-bicyclic ligands which can conceivably interact similarly to the Cp-THF ligand. The design of meso-ligands is quite attractive as they do not contain any stereocenters. Inhibitors incorporating urethanes of bicyclic-1,3-dioxolane and bicyclic-1,4-dioxane have shown potent enzyme inhibitory and antiviral activities. Inhibitor (K(i) = 0.11 nM; IC(50) = 3.8 nM) displayed very potent antiviral activity in this series. While inhibitor showed comparable enzyme inhibitory activity (K(i) = 0.18 nM) its antiviral activity (IC(50) = 170 nM) was significantly weaker than inhibitor . Inhibitor maintained an antiviral potency against a series of multi-drug resistant clinical isolates comparable to amprenavir. A protein-ligand X-ray structure of -bound HIV-1 protease revealed a number of key hydrogen bonding interactions at the S2-subsite. We have created an active model of inhibitor based upon this X-ray structure.     

Energetic basis for drug resistance of HIV-1 protease mutants against amprenavir

Amprenavir (APV) is a high affinity (0.15 nM) HIV-1 protease (PR) inhibitor. However, the affinities of the drug resistant protease variants V32I, I50V, I54V, I54M, I84V and L90M to amprenavir are decreased 3 to 30-fold compared to the wild-type. In this work, the popular molecular mechanics Poisson-Boltzmann surface area method has been used to investigate the effectiveness of amprenavir against the wild-type and these mutated protease variants. Our results reveal that the protonation state of Asp25/Asp25′ strongly affects the dynamics, the overall affinity and the interactions of the inhibitor with individual residues. We emphasize that, in contrast to what is often assumed, the protonation state may not be inferred from the affinities but requires pK_a calculations. At neutral pH, Asp25 and Asp25′ are ionized or protonated, respectively, as suggested from pK_a calculations. This protonation state was thus mainly considered in our study. Mutation induced changes in binding affinities are in agreement with the experimental findings. The decomposition of the binding free energy reveals the mechanisms underlying binding and drug resistance. Drug resistance arises from an increase in the energetic contribution from the van der Waals interactions between APV and PR (V32I, I50V, and I84V mutant) or a rise in the energetic contribution from the electrostatic interactions between the inhibitor and its target (I54M and I54V mutant). For the V32I mutant, also an increased free energy for the polar solvation contributes to the drug resistance. For the L90M mutant, a rise in the van der Waals energy for APV-PR interactions is compensated by a decrease in the polar solvation free energy such that the net binding affinity remains unchanged. Detailed understanding of the molecular forces governing binding and drug resistance might assist in the design of new inhibitors against HIV-1 PR variants that are resistant against current drugs.