A comparison of methods for pharmacophore generation with the catalyst software and their use for 3D-QSAR: application to a set of 4-aminopyridine thrombin inhibitors

In silico identification of novel lead compounds with AT1 receptor antagonist activity: successful application of chemical database screening protocol

Background

AT₁ receptor antagonists are clinically effective drugs for the treatment of hypertension, cardiovascular, and related disorders. In an attempt to identify new AT₁ receptor antagonists, a pharmacophore-based virtual screening protocol was applied. The pharmacophore models were generated from 30 training set compounds. The best model was chosen on the basis of squared correlation coefficient of training set and internal test set. The validity of the developed model was also ensured using catScramble validation method and external test set prediction.

Results

The final model highlighted the importance of hydrogen bond acceptor, hydrophobic aliphatic, hydrophobic, and ring aromatic features. The model satisfied all the statistical criteria such as cost function analysis and correlation coefficient. The result of estimated activity for internal and external test set compounds reveals that the generated model has high prediction capability. The validated pharmacophore model was further used for mining of 56000 compound database (MiniMaybridge). Total 141 hits were obtained and all the hits were checked for druggability, this led to the identification of two active druggable AT₁ receptor antagonists with diverse structure.

Conclusion

A highly validated pharmacophore model generated in this study identified two novel druggable AT₁ receptor antagonists. The developed model can also be further used for mining of other virtual database.

In silico pharmacophore model for tabun-inhibited acetylcholinesterase reactivators: a study of their stereoelectronic properties

Organophosphorus (OP) nerve agents that inhibit acetylcholinesterase (AChE; EC 3.1.1.7) function in the nervous system, causing acute intoxication. If untreated, death can result. Inhibited AChE can be reactivated by oximes, antidotes for OP exposure. However, OP intoxication caused by the nerve agent tabun (GA) is particularly resistant to oximes, which poorly reactivate GA-inhibited AChE. In an attempt to develop a rational strategy for the discovery and design of novel reactivators with lower toxicity and increased efficacy in reactivating GA-inhibited AChE, we developed the first in silico pharmacophore model for binding affinity of GA-inhibited AChE from a set of 11 oximes. Oximes were analyzed for stereoelectronic profiles and three-dimensional quantitative structure-activity relationship pharmacophores using ab initio quantum chemical and pharmacophore generation methods. Quantum chemical methods were sequentially used from semiempirical AM1 to hierarchical ab initio calculations to determine the stereoelectronic properties of nine oximes exhibiting affinity for binding to GA-inhibited AChE in vivo. The calculated stereoelectronic properties led us to develop the in silico pharmacophore model using CATALYST methodology. Specific stereoelectronic profiles including the distance between bisquarternary nitrogen atoms of the pyridinium ring in the oximes, hydrophilicity, surface area, nucleophilicity of the oxime oxygen, and location of the molecular orbitals on the isosurfaces have important roles for potencies for reactivating GA-inhibited AChE. The in silico pharmacophore model of oxime affinity for binding to GA-inhibited AChE was found to require a hydrogen bond acceptor, a hydrogen bond donor at the two terminal regions, and an aromatic ring in the central region of the oximes. The model was found to be well-correlated (R = 0.9) with experimental oxime affinity for binding to GA-inhibited AChE. Additional stereoelectronic features relating activity with the location of molecular orbitals and weak electrostatic potential field over the aromatic rings were found to be consistent with the pharmacophore model. These results provided the first predictive pharmacophore model of oxime affinity for binding toward GA-inhibited AChE. The model may be useful for virtual screening of compound libraries to discover and/or custom synthesize more efficacious and less toxic reactivators that may be useful for GA intoxication.

A Novel Approach Using Pharmacophore Ensemble/Support Vector Machine (PhE/SVM) for Prediction of hERG Liability

A novel approach by using a panel of plausible pharmacophore hypothesis candidates to constitute the pharmacophore ensemble (PhE) and subject them to regression by support vector machine (SVM) has been developed for predicting the liability of human ether-a-go-go-related gene (hERG). This PhE/SVM scheme takes into account the protein conformational flexibility while interacting with structurally diverse ligands, which is crucial yet often neglected by most of the analogue-based modeling methods. Thirty-nine molecules were carefully selected and cross-examined from the literature data for this study, of which 26 and 13 molecules were deliberately treated as the training set and the test set to generate the model and to validate the generated model, respectively. The final PhE/SVM model gave rise to an r(2) value of 0.97 for observed vs predicted pIC(50) values for the training set, a q(2) value of 0.89 by the 10-fold cross-validation and an r(2) value of 0.94 for the test set. Thus, this PhE/SVM model provides a fast and accurate tool for predicting liability of hERG and can be utilized to guide medicinal chemistry to avoid molecules with an inhibition potential of this potassium channel.

Ligand-Based Molecular Modeling Study on a Chemically Diverse Series of Cholecystokinin-B/Gastrin Receptor Antagonists: Generation of Predictive Model

Pharmacophore hypotheses were developed for six structurally diverse series of cholecystokinin-B/gastrin receptor (CCK-BR) antagonists. A training set consisting of 33 compounds was carefully selected. The activity spread of the training set molecules was from 0.1 to 2100 nM. The most predictive pharmacophore model (hypothesis 1), consisting of four features, namely, two hydrogen bond donors, one hydrophobic aliphatic, and one hydrophobic aromatic feature, had a correlation (r) of 0.884 and a root-mean-square deviation of 1.1526, and the cost difference between null cost and fixed cost was 81.5 bits. The model was validated on a test set consisting of six different series of 27 structurally diverse compounds and performed well in classifying active and inactive molecules correctly. This validation approach provides confidence in the utility of the predictive pharmacophore model developed in this work as a 3D query tool in the virtual screening of drug-like molecules to retrieve new chemical entities as potent CCK-BR antagonists. The model can also be used to predict the biological activities of compounds prior to their costly and time-consuming synthesis.

A three-dimensional in silico pharmacophore model for inhibition of Plasmodium falciparum cyclin-dependent kinases and discovery of different classes of novel Pfmrk specific inhibitors

The cell division cycle is regulated by a family of cyclin-dependent protein kinases (CDKs) that are functionally conserved among many eukaryotic species. The characterization of plasmodial CDKs has identified them as a leading antimalarial drug target in our laboratory. We have developed a three-dimensional QSAR pharmacophore model for inhibition of a Plasmodium falciparum CDK, known as Pfmrk, from a set of fifteen structurally diverse kinase inhibitors with a wide range of activity. The model was found to contain two hydrogen bond acceptor functions and two hydrophobic sites including one aromatic-ring hydrophobic site. Although the model was not developed from X-ray structural analysis of the known CDK2 structure, it is consistent with the structure-functional requirements for binding of the CDK inhibitors in the ATP binding pocket. Using the model as a template, a search of the in-house three-dimensional multiconformer database resulted in the discovery of sixteen potent Pfmrk inhibitors. The predicted inhibitory activities of some of these Pfmrk inhibitors from the molecular model agree exceptionally well with the experimental inhibitory values from the in vitro CDK assay.

Generation of ligand conformations in continuum solvent consistent with protein active site topology: application to thrombin

Using the crystal structure of an inhibitor complexed with the serine protease thrombin (PDB code ) and the functional group definitions contained within the Catalyst software, a representation of the enzyme's active site was produced (structure-based pharmacophore model). A training set of 16 homologous non-peptide inhibitors whose conformations had been generated in continuum solvent (MacroModel) and clustered into conformational families (XCluster) was regressed against this pharmacophore so as to obtain a 3D-QSAR model. To test the robustness of the resulting QSAR model, the synthesis of a series of non-peptide thrombin inhibitors based on arylsuphonyl derivatives of an aminophenol ring linked to a pyridyl-based S1 binding group was undertaken. These compounds served as a test set (20-24). The crystal structure for the novel symmetrical disulfonyl compound 24, in complex with thrombin, has been solved. Its calculated binding mode is in general agreement with the crystallographically observed one, and the predicted K(i) value is in close accord with the experimental value.

Non-peptide angiotensin II receptor antagonists: chemical feature based pharmacophore identification

Chemical feature based pharmacophore models were elaborated for angiotensin II receptor subtype 1 (AT(1)) antagonists using both a quantitative and a qualitative approach (Catalyst HypoGen and HipHop algorithms, respectively). The training sets for quantitative model generation consisted of 25 selective AT(1) antagonists exhibiting IC(50) values ranging from 1.3 nM to 150 microM. Additionally, a qualitative pharmacophore hypothesis was derived from multiconformational structure models of the two highly active AT(1) antagonists 4u (IC(50) = 0.2 nM) and 3k (IC(50) = 0.7 nM). In the case of the quantitative model, the best pharmacophore hypothesis consisted of a five-features model (Hypo1: seven points, one hydrophobic aromatic, one hydrophobic aliphatic, a hydrogen bond acceptor, a negative ionizable function, and an aromatic plane function). The best qualitative model consisted of seven features (Hypo2: 11 points, two aromatic rings, two hydrogen bond acceptors, a negative ionizable function, and two hydrophobic functions). The obtained pharmacophore models were validated on a wide set of test molecules. They were shown to be able to identify a range of highly potent AT(1) antagonists, among those a number of recently launched drugs and some candidates presently undergoing clinical tests and/or development phases. The results of our study provide confidence for the utility of the selected chemical feature based pharmacophore models to retrieve structurally diverse compounds with desired biological activity by virtual screening.

Pharmacophore mapping of a series of 2,4-diamino-5-deazapteridine inhibitors of Mycobacterium avium complex dihydrofolate reductase

Pharmacophore hypotheses were developed for a series of 2,4-diamino-5-deazapteridine inhibitors of Mycobacterium avium complex (MAC) and human dihydrofolate reductase (hDHFR). Training sets consisting of 20 inhibitors were selected in each case on the basis of the information content of the structures and activity data as required by the HypoGen program in the Catalyst software. In the case of MAC DHFR inhibitors, the best pharmacophore in terms of statistics and predictive value consisted of four features: two hydrogen bond acceptors (HA), one hydrophobic (HY) feature, and one ring aromatic (RA) feature. The selected pharmacophore hypothesis yielded an rms deviation of 0.730 and a correlation coefficient of 0.967 with a cost difference (null cost minus total cost) of approximately 52. The pharmacophore was validated on a large set of test inhibitors. For the test series, a classification scheme was used to distinguish highly active from moderately active and inactive compounds on the basis of activity ranges. This classification scheme is more practical than actual estimated values because these values have no meaning for compounds yet to be tested except that they indicate whether the compounds will be active or inactive in a biological assay. For the training set, the success rate for predicting active and inactive compounds was 100%. For the test set, the success rate in predicting active compounds was greater than 92% while about 7% of the inactive compounds were predicted to be active. This successful prediction was further validated on three structurally diverse compounds active against MAC DHFR. Two compounds mapped well onto three of the four features of the pharmacophore. The third compound was mapped to all four features of the pharmacophore. This validation study provided confidence for the usefulness of the selected pharmacophore model to identify compounds with diverse structures from a database search. Comparison of pharmacophores for inhibitors of human and MAC DHFR is expected to reveal fundamental differences between these two pharmacophores that may be effectively exploited to identify and design compounds with high selectivity for MAC DHFR.