Discovery of Novel, Potent, and Specific Cell-Death Inducers in the Jurkat Acute Lymphoblastic Leukemia Cell Line.

The limited clinical efficacy of many cancer therapeutics has initiated intense research efforts toward the discovery of novel chemical entities in this field. In this study, 31 hit candidates were selected from nearly 800,000 database compounds in a ligand-based virtual screening campaign. In turn, three of these hits were found to have (sub)micromolar potencies in proliferation assays with the Jurkat acute lymphatic leukemic cell line. In this assay, the three hits were found to exhibit higher potency than clinically tested cell-death inducers (GDC-0152, AT-406, and birinapant). Importantly, antiproliferative activity toward non-cancer peripheral blood mononuclear cells (PBMCs) was found to be marginal. Further biological characterization demonstrated the cell-death-inducing properties of these compounds. Biological testing of hit congeners excluded a nonspecific, toxic effect of the novel structures. Altogether, these findings may have profound relevance for the development of clinical candidates in tumor therapy.

Large, chemically diverse dataset of logP measurements for benchmarking studies.

Lipophilicity is a crucial parameter in drug development since it impacts both ADME properties and target affinity of drug candidates. In early drug discovery stage, accurate tools for logP prediction are highly desired. Many calculation methods were developed to aid pharmaceutical scientists in drug research; however almost all suffer from insufficient accuracy and variation of performance in several regions of the chemical space associated with new chemical entities. The low predictive power of existing software packages can be explained by limited availability and/or variable quality of experimental logP values associated with training set used, which stem from various protocols and poorly cover chemical space. In this study, a dataset of 1000 diverse test compounds out of 4.5 million was generated; logP values of 759 purchasable compounds (46% non-ionizable, 30% basic, 17% acidic, 0.5% zwitterionic and 6.5% ampholytes) from this selected set were experimentally determined by UHPLC followed by UV detection or MS detection when necessary. Finally, a data collection of 707 validated logP values ranging from 0.30 to 7.50 is now available for benchmarking of existing and development of new approaches to predict octanol/water partition coefficients of chemical compounds.

Virtual fragment screening: discovery of histamine H3 receptor ligands using ligand-based and protein-based molecular fingerprints.

Virtual fragment screening (VFS) is a promising new method that uses computer models to identify small, fragment-like biologically active molecules as useful starting points for fragment-based drug discovery (FBDD). Training sets of true active and inactive fragment-like molecules to construct and validate target customized VFS methods are however lacking. We have for the first time explored the possibilities and challenges of VFS using molecular fingerprints derived from a unique set of fragment affinity data for the histamine H(3) receptor (H(3)R), a pharmaceutically relevant G protein-coupled receptor (GPCR). Optimized FLAP (Fingerprints of Ligands and Proteins) models containing essential molecular interaction fields that discriminate known H(3)R binders from inactive molecules were successfully used for the identification of new H(3)R ligands. Prospective virtual screening of 156,090 molecules yielded a high hit rate of 62% (18 of the 29 tested) experimentally confirmed novel fragment-like H(3)R ligands that offer new potential starting points for the design of H(3)R targeting drugs. The first construction and application of customized FLAP models for the discovery of fragment-like biologically active molecules demonstrates that VFS is an efficient way to explore protein-fragment interaction space in silico.

Ligand-, structure- and pharmacophore-based molecular fingerprints: a case study on adenosine A(1), A (2A), A (2B), and A (3) receptor antagonists.

FLAP fingerprints are applied in the ligand-, structure- and pharmacophore-based mode in a case study on antagonists of all four adenosine receptor (AR) subtypes. Structurally diverse antagonist collections with respect to the different ARs were constructed by including binding data to human species only. FLAP models well discriminate "active" (=highly potent) from "inactive" (=weakly potent) AR antagonists, as indicated by enrichment curves, numbers of false positives, and AUC values. For all FLAP modes, model predictivity slightly decreases as follows: A(2B)R > A(2A)R > A(3)R > A(1)R antagonists. General performance of FLAP modes in this study is: ligand- > structure- > pharmacophore- based mode. We also compared the FLAP performance with other common ligand- and structure-based fingerprints. Concerning the ligand-based mode, FLAP model performance is superior to ECFP4 and ROCS for all AR subtypes. Although focusing on the early first part of the A(2A), A(2B) and A(3) enrichment curves, ECFP4 and ROCS still retain a satisfactory retrieval of actives. FLAP is also superior when comparing the structure-based mode with PLANTS and GOLD. In this study we applied for the first time the novel FLAPPharm tool for pharmacophore generation. Pharmacophore hypotheses, generated with this tool, convincingly match with formerly published data. Finally, we could demonstrate the capability of FLAP models to uncover selectivity aspects although single AR subtype models were not trained for this purpose.

QSAR modeling and data mining link Torsades de Pointes risk to the interplay of extent of metabolism, active transport, and HERG liability.

We collected 1173 hERG patch clamp (PC) data (IC50) from the literature to derive twelve classification models for hERG inhibition, covering a large variety of chemical descriptors and classification algorithms. Models were generated using 545 molecules and validated through 258 external molecules tested in PC experiments. We also evaluated the suitability of the best models to predict the activity of 26 proprietary compounds tested in radioligand binding displacement (RBD). Results proved the necessity to use multiple validation sets for a true estimation of model accuracy and demonstrated that using various descriptors and algorithms improves the performance of ligand-based models. Intriguingly, one of the most accurate models uncovered an unexpected link between extent of metabolism and hERG liability. This hypothesis was fairly reinforced by using the Biopharmaceutics Drug Disposition Classification System (BDDCS) that recognized 94% of the hERG inhibitors as extensively metabolized in vivo. Data mining suggested that high Torsades de Pointes (TdP) risk results from an interplay of hERG inhibition, extent of metabolism, active transport, and possibly solubility. Overall, these new findings might improve both the decision making skills of pharmaceutical scientists to mitigate hERG liability during the drug discovery process and the TdP risk assessment during drug development.

IAP antagonists: promising candidates for cancer therapy.

A promising strategy in cancer therapy aims to promote apoptosis in cancer cells. Targeting inhibitor of apoptosis proteins (IAPs) with small-molecule inhibitors has attracted increasing interest in triggering cancer cell death. It is considered to have great potential for cancer drug discovery because IAPs block apoptosis at the core of the apoptotic machinery and are aberrantly expressed in various tumors. This review focuses on the current development of small-molecule IAP antagonists for cancer therapy.

Large-scale evaluation of log P predictors: local corrections may compensate insufficient accuracy and need of experimentally testing every other compound.

A large variety of log P calculation methods failed to produce sufficient accuracy in log P prediction for two in-house datasets of more than 96000 compounds contrary to their significantly better performances on public datasets. The minimum Root Mean Squared Error (RMSE) of 1.02 and 0.65 were calculated for the Pfizer and Nycomed datasets, respectively, in the 'out-of-box' implementation. Importantly, the use of local corrections (LC) implemented in the ALOGPS program based on experimental in-house log P data significantly reduced the RMSE to 0.59 and 0.48 for the Pfizer and Nycomed datasets, respectively, instantly without retraining the model. Moreover, more than 60% of molecules predicted with the highest confidence in each set had a mean absolute error (MAE) less than 0.33 log units that is only ca. 10% higher than the estimated variation in experimental log P measurements for the Pfizer dataset. Therefore, following this retrospective analysis, we suggest that the use of the predicted log P values with high confidence may eliminate the need of experimentally testing every other compound. This strategy could reduce the cost of measurements for pharmaceutical companies by a factor of 2, increase the confidence in prediction at the analog design stage of drug discovery programs, and could be extended to other ADMET properties.

Calculation of molecular lipophilicity: State-of-the-art and comparison of log P methods on more than 96,000 compounds.

We first review the state-of-the-art in development of log P prediction approaches falling in two major categories: substructure-based and property-based methods. Then, we compare the predictive power of representative methods for one public (N = 266) and two in house datasets from Nycomed (N = 882) and Pfizer (N = 95809). A total of 30 and 18 methods were tested for public and industrial datasets, respectively. Accuracy of models declined with the number of nonhydrogen atoms. The Arithmetic Average Model (AAM), which predicts the same value (the arithmetic mean) for all compounds, was used as a baseline model for comparison. Methods with Root Mean Squared Error (RMSE) greater than RMSE produced by the AAM were considered as unacceptable. The majority of analyzed methods produced reasonable results for the public dataset but only seven methods were successful on the both in house datasets. We proposed a simple equation based on the number of carbon atoms, NC, and the number of hetero atoms, NHET: log P = 1.46(+/-0.02) + 0.11(+/-0.001) NC-0.11(+/-0.001) NHET. This equation outperformed a large number of programs benchmarked in this study. Factors influencing the accuracy of log P predictions were elucidated and discussed.

The pre-clinical assessment of rapamycin-eluting, durable polymer-free stent coating concepts.

All four currently FDA-approved drug-eluting stents (DESs) contain a durable polymeric coating which can negatively impact vascular healing processes and eventually lead to adverse cardiac events. Aim of this study was the pre-clinical assessment of two novel rapamycin-eluting stent (RES) coating technologies that abstain from use of a durable polymer. Two distinctive RES coating technologies were evaluated in vitro and in the porcine coronary artery stent model. The R-poly(S) stent platform elutes rapamycin from a biodegradable polymer that is top coated with the resin shellac to minimize the amount of polymer. The R-pro(S) stent platform allows dual drug release of rapamycin and probucol, blended by shellac. HPLC-based determination of pharmacokinetics indicated drug release for more than 28 days. At 30 days, neointimal formation was found to be significantly decreased for both DESs compared to bare-metal stents. Assessment of vascular healing revealed absence of increased inflammation in both DESs, which is commonly observed in DES with non-erodible polymeric coating. In conclusion, the pre-clinical assessment of RESs with resin-based or dual drug coating indicated an adequate efficacy profile as well as a beneficial effect for vascular healing processes. These results encourage the transfer of these technologies to clinical evaluation.

Novel TOPP descriptors in 3D-QSAR analysis of apoptosis inducing 4-aryl-4H-chromenes: comparison versus other 2D- and 3D-descriptors.

Novel 3D-descriptors using Triplets Of Pharmacophoric Points (TOPP) were evaluated in QSAR-studies on 80 apoptosis-inducing 4-aryl-4H-chromenes. A predictive QSAR model was obtained using PLS, confirmed by means of internal and external validations. Performance of the TOPP approach was compared with that of other 2D- and 3D-descriptors; statistical analysis indicates that TOPP descriptors perform best. A ranking of TOPP>GRIND>BCI 4096=ECFP>FCFP>GRID-GOLPE>>DRAGON>>>MDL 166 was achieved. Finally, in a 'consensus' analysis predictions obtained using the single methods were compared with an average approach using six out of eight methods. The use of the average is statistically superior to the single methods. Beyond it, the use of several methods can help to easily investigate the presence/absence of outliers according to the 'consensus' of the predicted values: agreement among all the methods indicates a precise prediction, whereas large differences between predicted values (for the same compounds by different methods) would demand caution when using such predictions.

Virtual screening for novel openers of pancreatic K(ATP) channels.

Ligand-based virtual screening approaches were applied to search for new chemotype KCOs activating Kir6.2/SUR1 KATP channels. A total of 65 208 commercially available compounds, extracted from the ZINC archive, served as database for screening. In a first step, pharmacokinetic filtering via VolSurf reduced the initial database to 1913 compounds. Afterward, six molecules were selected as templates for similarity searches: similarity scores, obtained toward these templates, were calculated with the GRIND, FLAP, and TOPP approaches, which differently encode structural information into potential pharmacophores. In this way, we obtained 32 hit candidates, 16 via GRIND and eight each via FLAP and TOPP. For biological testing of the hit candidates, their effects on membrane potentials in HEK 293 cells expressing Kir6.2/SUR1 were studied. GRIND, FLAP, and TOPP all yielded hits, but no method top-ranked all the actives. Thus, parallel application of different approaches probably improves hit detection.

Structure-activity relationships of K(ATP) channel openers.

Given their many physiological functions, K(ATP) channels represent promising drug targets. Sulfonylureas like glibenclamide block K(ATP) channels; they are used in the therapy of type 2 diabetes. Openers of K(ATP) channels (KCOs) e.g. relax smooth muscle and induce hypotension. KCOs are chemically heterogeneous and include as different classes as the benzopyrans, cyanoguanidines, thioformamides, thiadiazines and pyridyl nitrates. Examples for new chemical entities more recently developed as KCOs include cyclobutenediones, dihydropyridine related structures, and tertiary carbinols. Structure-activity relationships of the main chemical classes of KCOs are discussed.

Second-generation K(ATP) channel openers.

This review discusses structural aspects of second-generation K(ATP) channel openers (KCOs), which exhibit improved tissue-selectivity. Their therapeutic profile is debated with main focus on cardiac ischemia, asthma, and urinary incontinence.

Binding studies and GRIND/ALMOND-based 3D QSAR analysis of benzothiazine type K(ATP)-channel openers.

For seventeen 1,4-benzothiazine potassium channel openers, we performed binding studies in rat aortic smooth muscle cells and cardiomyocytes, compared their binding affinities with published relaxation data, and derived 3D-QSAR models using GRIND/ALMOND descriptors. Binding affinities in smooth muscle cells range from a pK(D) of 4.76 for compound 3e to 9.10 for compound 4c. Comparison of data for smooth muscle relaxation and binding shows preferentially higher pEC(50)s for the former. In cardiomyocytes, pK(D) values range from 4.21 for 3e to 8.16 for 4c. 3D-QSAR analysis resulted in PLS models of two latent variables for all three activities with determination coefficients of 0.97 (smooth muscle relaxation) and 0.94 (smooth muscle cells- and cardiomyocytes-binding). Internal validation yielded q(2) values of 0.69, 0.66, and 0.64. The carbonyl on the N-4 substituent, the hydrogen bond acceptor at C-6, the five-membered ring at N-4, and the gem-dimethyls mainly guide strong binding and strong smooth muscle relaxation.

Pharmacophore, drug metabolism, and pharmacokinetics models on non-peptide AT1, AT2, and AT1/AT2 angiotensin II receptor antagonists.

About 20 non-peptide angiotensin II receptor antagonists are in various stages of clinical development. Different modeling approaches were used to predict the pharmacophoric requirements for AT(1) (angiotensin II receptor subtype 1) affinity. However, to our knowledge, none was used to predict both the selectivity toward AT(1) and AT(2) (angiotensin II receptor subtype 2) receptor subtypes. In this paper, partial least squares discriminant analysis is applied to derive the chemical features guiding AT(1) and AT(2) selectivity or mixed AT(1)/AT(2) receptor binding. The method can be used to modulate AT(1) versus AT(2) selectivity. Concerns that unopposed stimulation of the AT(2) receptor might produce adverse effects initiated a search for new balanced antagonists. Moreover, it can serve as a fast filtering procedure in database searches. Finally, some relevant pharmacokinetics and metabolic properties of the database of 53 compounds are calculated using the VolSurf and MetaSite software to allow the simultaneous characterization of pharmacodynamic and pharmacokinetics properties of the chemical space of angiotensin II receptor antagonists.

log P estimation of 1,2-dithiole-3-thiones and 1,2-dithiole-3-ones: a comparison of experimental and calculative approaches.

PURPOSE: To estimate experimental log P values of formerly described 5-formyl- and 5-acyl-dithiole-3-thiones (DTT) and -dithiole-3-ones (DTO) and to check the validity of five log P calculation programs via experimental log P for a database of 68 DTT and DTO. METHODS: Experimental log P values were measured by means of octanol/water partitioning; for determining solute concentrations in water, RP-HPLC with spectrophotometric detection was used. For calculating log P, the fragmental methods ACD/log P, CLOGP, and KOWWIN, the atom-based approach XLOGP, and the whole-molecule approach QLOGP were applied. RESULTS: Quality of calculations significantly differs depending on the subset under consideration. For database compounds 01-48, comprising alkyl and aryl substitution in 4- and 5-position, the fragmental methods ACD/log P, CLOGP, and KOWWIN perform significantly better than the atom-based approach XLOGP and the whole-molecule method QLOGP. For database compounds 49-68, comprising formyl and acyl substitution in 4- and 5-position, superiority of the whole-molecule method QLOGP over the substructure-based approaches is observed. The strong underestimation of log P for compounds 49-68 probably indicates hidden physicochemical phenomena resulting from the juxtaposition of the acyl and dithiole moieties. CONCLUSIONS: All calculation methods included in this study need a thorough refinement to adequately cope with particular solvation behavior suspected to prevail in formyl- or acyl-DTT and DTO, which represent a chemical class of high pharmacological interest.

Comparison of ligand-based and structure-based 3D-QSAR approaches: a case study on (aryl-)bridged 2-aminobenzonitriles inhibiting HIV-1 reverse transcriptase.

Ligand- (GRIND) and structure-based (GLUE/GRIND) 3D-QSAR approaches were compared for 55 (aryl-)bridged 2-aminobenzonitriles inhibiting HIV-1 reverse transcriptase (HIV-1 RT). The ligand-based model was built from conformers selected by in vacuo minimization. The available X-ray structure of 3v in complex with HIV-1 RT allowed comparative structure-based calculations using the new docking software GLUE for conformer selection. Both models were validated via statistics and via virtual receptor sites (VRS) considering pharmacophoric regions and mutual distances, which were also compared with experimental evidence. The statistics show slight superiority of the structure-based approach in terms of fitting and prediction. By encoding relevant molecular interaction fields (MIF) into pharmacophoric regions, 10 such regions were derived from both models; they all fit the real receptor except HBD2. Also mutual distances highly agree between the real site and both VRS. Although distances from the structure-based approach are closer to the real receptor, present data prove the validity of the ligand-based GRIND approach.

The impact of lipophilicity in drug research: a case report on beta-blockers.

The key importance of lipophilicity in bio-studies is discussed for beta-blockers. Examples of their lipophilicity-dependent pharmacological properties including pharmacokinetic, pharmacodynamic and clinical aspects are reviewed. Comprehensive lipophilicity compilations of beta-blockers are lacking so far. LogP calculations with 10 programs for 30 clinically relevant beta-blockers are presented for the first time in this review.

GRIND/ALMOND investigations on CysLT1 receptor antagonists of the quinolinyl(bridged)aryl type.

One of the current routes in developing antiasthmatics is CysLT(1) receptor antagonism. For a training set of 54 CysLT(1) receptor antagonists of the quinolinyl(bridged)aryl type we developed chemometric QSAR models applying GRID independent descriptors (=GRIND). PLS analysis resulted in a two-component model explaining 67% of the variance for CysLT(1) receptor binding (r2=0.67, SDEC = 0.47, q2=0.54). GRIND variables 11-50 and 22-55 are responsible for high-affinity binding; variable 11-62 is detrimental. The predictivity of the above chemometric model is tested with a set of 69 CysLT(1) receptor antagonists, exhibiting varying chemical similarity to the training set. Nearly 50% of the test set are quite well predicted. The quality of prediction coincides in part with chemical subclassification: phenylene bridged compounds are quite well predicted; for structures with bridging heterocycles predictions are rather poor. For explaining the outlier behavior, a PLS discriminant analysis including the training set and the strongest outliers of the test set was performed. The scores plot of discriminant PLS shows an almost complete separation between the two subsets. A PLS coefficients plot explains which GRIND variables are important for the discrimination between the training set and the outliers of the test set.

Chemometric studies on the bactericidal activity of quinolones via an extended VolSurf approach.

An extended VolSurf approach, that additionally includes SHAPE descriptors, was applied to a dataset of 55 quinolones. Bactericidal activity was measured at Bayer AG, Germany, for Gram-negative (Escherichia coli and Pseudomonas aeruginosa) and Gram-positive bacteria (Staphylococcus aureus and Enterococcus faecalis). Chemometric analysis was first approached via a classical VolSurf approach. The following descriptors were found most important: bactericidal activity particularly increases with high values of the best volume (BV11(OH2)) and the minimum energy (Emin1(OH2)) of the water probe, high values of the integy moment (ID(DRY)) of the lipophilic probe, and high values of the hydrophilic region (W(O)) of the hydrogen bond acceptor probe. Best volume (BV31(OH2)) of the water probe and best volume (BV12(DRY)) and lipophilic regions (D(DRY)) of the lipophilic probe as well as H-bonding capacity derived with the CO probe (HB(O)) are inversely related to activity. PLS analysis yields a five-component model with an r(2) of 0.83 and a q(2) of 0.43 after variable selection via fractional factorial design (FFD). Chemometric modeling could be improved by including newly derived SHAPE descriptors, which were merged with the VolSurf descriptors and subjected to PLS analysis. The global model of this extended VolSurf approach is optimal with two components and exhibits a significantly improved statistical quality; a marginally reduced r(2) (0.75 versus 0.83) is more than compensated by a highly improved predictivity with a q(2) of 0.63 versus 0.43. To prove model quality, external prediction of seven test set quinolones was performed. The precise prediction of all test set molecules nicely demonstrates the robustness and statistical significance of the obtained chemometric model using the extended VolSurf approach.