Receptor-based 3D QSAR analysis of estrogen receptor ligands--merging the accuracy of receptor-based alignments with the computational efficiency of ligand-based methods

Are automated molecular dynamics simulations and binding free energy calculations realistic tools in lead optimization? An evaluation of the linear interaction energy (LIE) method

An extensive evaluation of the linear interaction energy (LIE) method for the prediction of binding affinity of docked compounds has been performed, with an emphasis on its applicability in lead optimization. An automated setup is presented, which allows for the use of the method in an industrial setting. Calculations are performed for four realistic examples, retinoic acid receptor gamma, matrix metalloprotease 3, estrogen receptor alpha, and dihydrofolate reductase, focusing on different aspects of the procedure. The obtained LIE models are evaluated in terms of the root-mean-square (RMS) errors from experimental binding free energies and the ability to rank compounds appropriately. The results are compared to the best empirical scoring function, selected from a set of 10 scoring functions. In all cases, good LIE models can be obtained in terms of free-energy RMS errors, although reasonable ranking of the ligands of dihydrofolate reductase proves difficult for both the LIE method and scoring functions. For the other proteins, the LIE model results in better predictions than the best performing scoring function. These results indicate that the LIE approach, as a tool to evaluate docking results, can be a valuable asset in computational lead optimization programs.

3D QSAR Studies on Protein Tyrosine Phosphatase 1B Inhibitors: Comparison of the Quality and Predictivity among 3D QSAR Models Obtained from Different Conformer-Based Alignments

A set of 65 flexible peptidomimetic competitive inhibitors (52 in the training set and 13 in the test set) of protein tyrosine phosphatase 1B (PTP1B) has been used to compare the quality and predictive power of 3D quantitative structure-activity relationship (QSAR) comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) models for the three most commonly used conformer-based alignments, namely, cocrystallized conformer-based alignment (CCBA), docked conformer-based alignment (DCBA), and global minima energy conformer-based alignment (GMCBA). These three conformers of 5-[(2S)-2-({(2S)-2-[(tert-butoxycarbonyl)amino]-3-phenylpropanoyl}amino)3-oxo-3-pentylamino)propyl]-2-(carboxymethoxy)benzoic acid (compound number 66) were obtained from the X-ray structure of its cocrystallized complex with PTP1B (PDB ID: 1JF7), its docking studies, and its global minima by simulated annealing. Among the 3D QSAR models developed using the above three alignments, the CCBA provided the optimal predictive CoMFA model for the training set with cross-validated r2 (q2)=0.708, non-cross-validated r2=0.902, standard error of estimate (s)=0.165, and F=202.553 and the optimal CoMSIA model with q2=0.440, r2=0.799, s=0.192, and F=117.782. These models also showed the best test set prediction for the 13 compounds with predictive r2 values of 0.706 and 0.683, respectively. Though the QSAR models derived using the other two alignments also produced statistically acceptable models in the order DCBA>GMCBA in terms of the values of q2, r2, and predictive r2, they were inferior to the corresponding models derived using CCBA. Thus, the order of preference for the alignment selection for 3D QSAR model development may be CCBA>DCBA>GMCBA, and the information obtained from the CoMFA and CoMSIA contour maps may be useful in designing specific PTP1B inhibitors.

FlexNovo: structure-based searching in large fragment spaces

We present a new molecular design program, FlexNovo, for structure-based searching within large fragment spaces following a sequential growth strategy. The fragment spaces consist of several thousands of chemical fragments and a corresponding set of rules that specify how the fragments can be connected. FlexNovo is based on the FlexX molecular docking software and makes use of its incremental construction algorithm and the underlying chemical models. Interaction energies are calculated by using standard scoring functions. Several placement geometry, physicochemical property (drug-likeness), and diversity filter criteria are directly integrated into the "build-up" process. FlexNovo has been used to design potential inhibitors for four targets of pharmaceutical interest (dihydrofolate reductase, cyclin-dependant kinase 2, cyclooxygenase-2, and the estrogen receptor). We have carried out calculations using different diversity parameters for each of these targets and generated solution sets containing up to 50 molecules. The compounds obtained show that FlexNovo is able to generate a diverse set of reasonable molecules with drug-like properties. The results, including an automated similarity analysis with the Feature Tree program, indicate that FlexNovo often reproduces structural motifs as well as the corresponding binding modes seen in known active structures.

SAR and QSAR modeling of endocrine disruptors

A number of xenobiotics by mimicking natural hormones can disrupt crucial functions in wildlife and humans. These chemicals termed endocrine disruptors are able to exert adverse effects through a variety of mechanisms. Fortunately, there is a growing interest in the study of these structurally diverse chemicals mainly through research programs based on in vitro and in vivo experimentations but also by means of SAR and QSAR models. The goal of our study was to retrieve from the literature all the papers dealing with structure-activity models on endocrine disruptor xenobiotics. A critical analysis of these models was made focusing our attention on the quality of the biological data, the significance of the molecular descriptors and the validity of the statistical tools used for deriving the models. The predictive power and domain of application of these models were also discussed.

Three-dimensional models of non-steroidal ligands: a comparative molecular field analysis

The estrogen receptor, ER, is an important biological target whose inhibition is known to be therapeutically relevant in the treatment of postmenopausal osteoporosis. In the present study, two prediction methods (CoMFA and GRIND (Almond)) were used to describe the binding modes of a set of estrogen receptor ligands. The critical alignment step presented in CoMFA was solved by using the information of the molecular descriptors space generated by grid-independent descriptors (GRIND). Then, it was possible to build robust and high predictive models based on the alignment-independent model. Since the structure of estrogen receptor is solved, the results of the present 3D QSAR models, given by the PLS maps based on molecular interaction fields (MIF) were compared to ligand-binding ER domains and showed good agreement.

Effects of variable docking conditions and scoring functions on corresponding protein-aligned comparative molecular field analysis models constructed from diverse human protein tyrosine phosphatase 1B inhibitors

The effects of variable docking conditions and scoring functions on corresponding protein-aligned comparative molecular field analysis (CoMFA) models have been assessed. To this end, a group of diverse inhibitors were docked into the active site of human protein tyrosine phosphatase 1B (h-PTP 1B). The docked structures were utilized to construct corresponding protein-aligned CoMFA models by employing probe-based (H+, OH, CH3) energy grids and genetic partial least squares (G/PLS) statistical analysis. A total of 48 different docking configurations were evaluated, of which some succeeded in producing self-consistent and predictive CoMFA models. However, the best CoMFA model coincided with docking the un-ionized ligands into the hydrated form of the binding site via the PLP1 scoring function and restricted docking settings (r2(LOO) = 0.647, r2(PRESS) against 27 test compounds = 0.617). Interestingly, the most significant CoMFA models were orthogonal and corresponded to significantly different docked conformers/poses. To utilize the predictive potentials of the best CoMFA models collectively, it was decided to combine them in a single quantitative structure-activity relationship (QSAR) model. The combination model illustrated excellent statistical properties (r2(LOO) = 0.890, r2(PRESS) against 27 test compounds = 0.750).

A functionally orthogonal estrogen receptor-based transcription switch specifically induced by a nonsteroid synthetic ligand

It is highly desirable to design ligand-dependent transcription regulation systems based on transactivators unresponsive to endogenous ligands but induced by synthetic small molecules unable to activate endogenous receptors. Using molecular modeling and yeast selection, we identified an estrogen receptor ligand binding domain double mutant (L384M, M421G) with decreased affinity to estradiol and enhanced binding to compounds inactive on estrogen receptors. Nonresponsiveness to estrogen was achieved by additionally adding the G521R substitution while introducing an "antagonistic-type" side chain in the compound, as in 4-hydroxytamoxifen. The triple-substituted ligand binding domain is insensitive to physiological concentrations of estradiol and has nanomolar affinity for the ligand. In this binary system, both receptor and ligand are, therefore, reciprocally specific. The mutated variant in the context of a chimeric transcription factor provides tight, ligand-dependent regulation of reporter gene expression.

Design, Synthesis, Biological Evaluation, and Molecular Modeling Studies of TIBO-Like Cyclic Sulfones as Non-Nucleoside HIV-1 Reverse Transcriptase Inhibitors

TIBO- and TBO-like sulfone derivatives 1 and 2 were designed, synthesized, and tested for their ability to block the replication cycle of HIV-1 in infected cells. The anti-HIV-1 activities of sulfones 3, which were intermediates in the syntheses of 1 and 2, were also evaluated. Surprisingly, the sulfone analogues of TIBO R82913 (compounds 1) were inactive, whereas interesting results were obtained for truncated derivatives 2. Compound 2 w was the most potent among this series in cell-based assays (EC50=0.07 microM, CC50>200 microM, SI>2857). It was twofold less potent than R82913, but more selective. An X-ray crystallographic analysis was carried out to establish the absolute configuration of 2 w and its enantiomer 2 x, which were obtained by semipreparative HPLC of 2 v, one of the most potent racemates. Compounds 1-3 were proven to target HIV-1 RT. In fact, representative derivatives inhibited recombinant HIV-1 RT in vitro at concentrations similar to those active in cell-based assays. 3D QSAR studies and docking simulations were developed on TIBO- and TBO-like sulfone derivatives to rationalize their anti-HIV-1 potencies and to predict the activity of novel untested sulfone derivatives. Predictive 3D QSAR models were obtained with a receptor-based alignment by docking of TIBO- and TBO-like derivatives into the NNBS of RT.

Ligand Bias of Scoring Functions in Structure-Based Virtual Screening

A total of 945 known actives and roughly 10 000 decoy compounds were docked to eight different targets, and the resulting poses were scored using 10 different scoring functions. Three different score postprocessing methods were evaluated with respect to improvement of the enrichment in virtual screening. The three procedures were (i) multiple active site correction (MASC) as has been proposed by Vigers and Rizzi, (ii) a variation of MASC where corrections terms are predicted from simple molecular descriptors through PLS, PLS MASC, and (iii) size normalization. It was found that MASC did not generally improve the enrichment factors when compared to uncorrected scoring functions. For some combinations of scoring functions and targets, the enrichment was improved, for others not. However, by excluding the standard deviation from the MASC equation and transforming the scores for each target to a mean of 0 and a standard deviation of 1 (unit variance normalization), the performance was improved as compared to the original MASC method for most combinations of targets and scoring functions. Furthermore, when the molecular descriptors were fit to the mean scores over all targets and the resulting PLS models were used to predict mean scores, the enrichment as compared to the raw score was improved more often than by straightforward MASC. A high to intermediate linear correlation between the score and the number of heavy atoms was found for all scoring functions except FlexX. There seems to be a correlation between the size dependence of a scoring function and the effectiveness of PLS MASC in increasing the enrichment for that scoring function. Finally, normalization by molecular weight or heavy atom count was sometimes successful in increasing the enrichment. Dividing by the square or cubic root of the molecular weight or heavy atom count instead was often more successful. These results taken together suggest that ligand bias in scoring functions is a source of false positives in structure-based virtual screening. The number of false positives caused by ligand bias may be decreased using, for example, the PLS MASC procedure proposed in this study.

3D-QSAR Studies on Cannabinoid CB1 Receptor Agonists: G-Protein Activation as Biological Data

G-protein activation via the CB1 receptor was determined for a group of various CB1 ligands and utilized as biological activity data in subsequent CoMFA and CoMSIA studies. Both manual techniques and automated docking at CB1 receptor models were used to obtain a common alignment of endocannabinoid and classical cannabinoid derivatives. In the final alignment models, the endocannabinoid headgroup occupies a unique region distinct from the classical cannabinoid structures, supporting the hypothesis that these structurally diverse molecules overlap only partially within the receptor binding site. Both CoMFA and CoMSIA produce statistically significant models based on the manual alignment and a docking alignment at one receptor conformer. Leave-half-out cross-validation and progressive scrambling were successfully used in assessing the predictivity of the QSAR models.

Comparing the Performance of FLUFF-BALL to SEAL-CoMFA with a Large Diverse Estrogen Data Set: From Relevant Superpositions to Solid Predictions

In this work a template-based molecular mechanistic superposition algorithm FLUFF (Flexible Ligand Unified Force Field) and an accompanying local coordinate QSAR method BALL (Boundless Adaptive Localized Ligand) are validated against the benchmark techniques SEAL (Steric and Electrostatic Alignment) and CoMFA (Comparative Molecular Field Analysis) using a large diverse set of 245 xenoestrogens extracted from the EDKB (Endocrine Disruptor Knowledge Base) maintained by NCTR (National Centre for Toxicological Research). The results indicate that FLUFF is capable of generating relevant superpositions not only for BALL but also for CoMFA, as both techniques give predictive QSAR models. When the BALL and CoMFA methods are compared, it is clear that the BALL algorithm met or even exceeded the results of the standard 3D-QSAR method CoMFA using alignments either from the tailor-made superposition technique FLUFF or the reference method SEAL. The FLUFF-BALL method can be easily automated, and it is computationally light, providing thus a good computational "sieve" capable of fast screening of large molecule libraries.

Extraction and Visualization of Potential Pharmacophore Points Using Support Vector Machines: Application to Ligand-Based Virtual Screening for COX-2 Inhibitors

Support vector machines (SVM) were trained to predict cyclooxygenase 2 (COX-2) and thrombin inhibitors. The classifiers were obtained using sets of known COX-2 and thrombin inhibitors as "positive examples" and a large collection of screening compounds as "negative examples". Molecules were encoded by topological pharmacophore-point triangles. In retrospective virtual screening, 50-90% of the known active compounds were listed within the first 0.1% of the ranked database. To check the validity of the constructed classifiers, we developed a method for feature extraction and visualization using SVM. As a result, potential pharmacophore points were weighted according to their importance for COX-2 and thrombin inhibition. Known thrombin and COX-2 pharmacophore points were correctly recognized by the machine learning system. In a prospective virtual screening study, several potential COX-2 inhibitors were predicted and tested in a cellular activity assay. A benzimidazole derivative exhibited significant inhibitory activity with an IC(50) of 0.2 microM, which is better than Celecoxib in our assay. It was demonstrated that the SVM machine-learning method can be used in virtual screening and be analyzed in a human-interpretable way that results in a set of rules for designing novel molecules.

Synthesis and Evaluation of (17α,20Z)-21-(4-Substituted-phenyl)-19-norpregna-1,3,5(10),20-tetraene-3,17β-diols as Ligands for the Estrogen Receptor-α Hormone Binding Domain: Comparison with 20E-Isomers

As part of our ongoing program to develop probes for the hormone binding domain of the estrogen receptor-alpha (ERalpha), we prepared and evaluated a series of 17alpha,Z-(4-substituted-phenyl)vinyl estradiol derivatives. The results indicated that the relative binding affinities (RBAs) at 25 degrees C for the new compounds were significant (RBA = 9-57) although less than that of estradiol (RBA = 100) or of the parent unsubstituted phenylvinyl estradiol (RBA = 66). All of the Z-compounds were full agonists in the uterotrophic assay, indicating that the ligands formed estrogen-like complexes with the estrogen receptor-alpha hormone binding domain (ERalpha-HBD). Comparison of corresponding Z- and E-4-substituted phenylvinyl ligands complexed with the ERalpha-HBD indicated small but significant differences in binding modes that may account for the differing trends seen in the structure-activity relationships for the two series.

Docking and 3-D QSAR Studies on Indolyl Aryl Sulfones. Binding Mode Exploration at the HIV-1 Reverse Transcriptase Non-Nucleoside Binding Site and Design of Highly Active N-(2-Hydroxyethyl)carboxamide and N-(2-Hydroxyethyl)carbohydrazide Derivatives

Three-dimensional quantitative structure-activity relationship (3-D QSAR) studies and docking simulations were developed on indolyl aryl sulfones (IASs), a class of novel HIV-1 non-nucleoside reverse transcriptase (RT) inhibitors (Silvestri, et al. J. Med. Chem. 2003, 46, 2482-2493) highly active against wild type and some clinically relevant resistant strains (Y181C, the double mutant K103N-Y181C, and the K103R-V179D-P225H strain, highly resistant to efavirenz). Predictive 3-D QSAR models using the combination of GRID and GOLPE programs were obtained using a receptor-based alignment by means of docking IASs into the non-nucleoside binding site (NNBS) of RT. The derived 3-D QSAR models showed conventional correlation (r(2)) and cross-validated (q(2)) coefficients values ranging from 0.79 to 0.93 and from 0.59 to 0.84, respectively. All described models were validated by an external test set compiled from previously reported pyrryl aryl sulfones (Artico, et al. J. Med. Chem. 1996, 39, 522-530). The most predictive 3-D QSAR model was then used to predict the activity of novel untested IASs. The synthesis of six designed derivatives (prediction set) allowed disclosure of new IASs endowed with high anti-HIV-1 activities.

Structure-based prediction of free energy changes of binding of PTP1B inhibitors

The goals were (1) to understand the driving forces in the binding of small molecule inhibitors to the active site of PTP1B and (2) to develop a molecular mechanics-based empirical free energy function for compound potency prediction. A set of compounds with known activities was docked onto the active site. The related energy components and molecular surface areas were calculated. The bridging water molecules were identified and their contributions were considered. Linear relationships were explored between the above terms and the binding free energies of compounds derived based on experimental inhibition constants. We found that minimally three terms are required to give rise to a good correlation (0.86) with predictive power in five-group cross-validation test (q2 = 0.70). The dominant terms are the electrostatic energy and non-electrostatic energy stemming from the intra- and intermolecular interactions of solutes and from those of bridging water molecules in complexes.

CoMFA and docking study of novel estrogen receptor subtype selective ligands

We present the results from a Comparative Molecular Field Analysis (CoMFA) and docking study of a diverse set of 36 estrogen receptor ligands whose relative binding affinities (RBA) with respect to 17beta-Estradiol were available in both isoforms of the nuclear estrogen receptors (ER alpha, ER beta). Initial CoMFA models exhibited a correlation between the experimental relative binding affinities and the molecular steric and electrostatic fields; ER alpha: r2 = 0.79, q2 = 0.44 ER beta: r2 = 0.93, q2 = 0.63. Addition of the solvation energy of the isolated ligand improved the predictive nature of the ER beta model initially; r2 = 0.96, q2 = 0.70 but upon rescrambling of the data-set and reselecting the training set at random, inclusion of the ligand solvation energy was found to have little effect on the predictive nature of the CoMFA models. The ligands were then docked inside the ligand binding domain (LBD) of both ER alpha and ER beta utilizing the docking program Gold, after-which the program CScore was used to rank the resulting poses. Inclusion of both the Gold and CScore scoring parameters failed to improve the predictive ability of the original CoMFA models. The subtype selectivity expressed as RBA(ER alpha/ER beta) of the test sets was predicted using the most predictive CoMFA models, as illustrated by the cross-validated r2. In each case the most selective ligands were ranked correctly illustrating the utility of this method as a prescreening tool in the development of novel estrogen receptor subtype selective ligands.

Comparing the Quality and Predictiveness between 3D QSAR Models Obtained from Manual and Automated Alignment

A set of 113 flexible cyclic urea inhibitors of human immunodeficiency virus protease (HIV-1 PR) was used to compare the quality and predictive power of CoMFA and CoMSIA models for manually or automatically aligned inhibitor set. Inhibitors that were aligned automatically with molecular docking were in agreement with information obtained from existing X-ray structures. Both alignment methods produced statistically significant CoMFA and CoMSIA models, with the best q(2) value being 0.649 and the best predictive r(2) being 0.754. The manual alignment gave statistically higher values, whereas the automated alignment gave more robust models for predicting the activities of an external inhibitor set. Both models utilized similar amino acids in the HIV-1 PR active site, supporting the idea that hydrogen bonds form between an inhibitor and the backbone carbonyl oxygens of Gly48 and Gly48' and also the backbone NH group of Asp30, Gly48, Asp29', and Gly48' of the enzyme. These results suggest that an automated inhibitor alignment can yield predictive 3D QSAR models that are well comparable to manual methods. Thus, an automated alignment method in creating 3D QSAR models is encouragable when a well-characterized structure of the target protein is available.

Improving structure-based virtual screening by multivariate analysis of scoring data

Three different multivariate statistical methods, PLS discriminant analysis, rule-based methods, and Bayesian classification, have been applied to multidimensional scoring data from four different target proteins: estrogen receptor alpha (ERalpha), matrix metalloprotease 3 (MMP3), factor Xa (fXa), and acetylcholine esterase (AChE). The purpose was to build classifiers able to discriminate between active and inactive compounds, given a structure-based virtual screen. Seven different scoring functions were used to generate the scoring matrices. The classifiers were compared to classical consensus scoring and single scoring functions. The classifiers show a superior performance, with rule-based methods being most effective. The precision of correctly predicting an active compound is about 90% for three of the targets and about 25% for acetylcholine esterase. On the basis of these results, a new two-stage approach is suggested for structure-based virtual screening where limited activity information is available.

Spectroscopic QSAR Methods and Self-Organizing Molecular Field Analysis for Relating Molecular Structure and Estrogenic Activity

The performance of three "spectroscopic" quantitative structure-activity relationship (QSAR) methods (eigenvalue (EVA), electronic eigenvalue (EEVA), and comparative spectra analysis (CoSA)) for relating molecular structure and estrogenic activity are critically evaluated. The methods were tested with respect to the relative binding affinities (RBA) of a diverse set of 36 estrogens previously examined in detail by the comparative molecular field analysis method. The CoSA method with (13)C chemical shifts appears to provide a predictive QSAR model for this data set. EEVA (i.e., molecular orbital energy in this context) is a borderline case, whereas the performances of EVA (i.e., vibrational normal mode) and CoSA with (1)H shifts are substandard and only semiquantitative. The CoSA method with (13)C chemical shifts provides an alternative and supplement to conventional 3D QSAR methods for rationalizing and predicting the estrogenic activity of molecules. If CoSA is to be applied to large data sets, however, it is desirable that the chemical shifts are available from common databases or, alternatively, that they can be estimated with sufficient accuracy using fast prediction schemes. Calculations of NMR chemical shifts by quantum mechanical methods, as in this case study, seem to be too time-consuming at this moment, but the situation is changing rapidly. An inherent shortcoming common to all spectroscopic QSAR methods is that they cannot take the chirality of molecules into account, at least as formulated at present. Moreover, the symmetry of molecules may cause additional problems. There are three pairs of enantiomers and nine symmetric (C(2) or C(2)(v)) molecules present in the data set, so that the predictive ability of full 3D QSAR methods is expected to be better than that of spectroscopic methods. This is demonstrated with SOMFA (self-organizing molecular field analysis). In general, the use of external test sets with randomized data is encouraged as a validation tool in QSAR studies.

Synthesis and evaluation of 17alpha-20E-21-(4-substituted phenyl)-19-norpregna-1,3,5(10),20-tetraene-3,17beta-diols as probes for the estrogen receptor alpha hormone binding domain

As part of our program to develop probes for the hormone binding domain of the estrogen receptor alpha (ERalpha), we prepared a series of 4-para-substituted phenylvinyl estradiol derivatives using a combination of solution and solid-phase Pd(0)-catalyzed methods. The compounds 5a-j were evaluated for their binding affinity using the ERalpha hormone binding domain (HDB) isolated from transfected BL21 cells. The results indicated that although the new compounds were somewhat lower in relative binding affinity (RBA at 25 degrees C is 1-60%) than estradiol (100%), most had higher affinity than the unsubstituted parent phenylvinyl estradiol (RBA = 9%). Because the substituents did not generate a structure-activity relationship directly based on physicochemical properties, the series was evaluated using molecular modeling and molecular dynamics to determine key interactions between the ligand, especially the para substituent, and the protein. The results suggest that the observed relative binding affinities are directly related to the calculated binding energies and that amino acids juxtaposed to the para position play a significant but not dominant role in binding. In conclusion, we have identified the 17alpha-E-(4-substituted phenyl)vinyl estradiols as a class of ligands that retain significant affinity for the ERalpha-HBD. In particular, 4-substitution tends to increase receptor affinity compared to the unsubstituted analogue, as exemplified by 5e (4-COCH(3)), which had the highest RBA value (60%) of the series. Palladium(0)-catalyzed coupling reactions on solid support or in solution using suitably substituted iodo arenes and 17alpha-E-tributylstannylvinyl estradiols offer a flexible approach to their preparation. Molecular modeling studies of the receptor suggest that there exists additional ligand accessible regions within the ERalpha-HBD to generate interactions that may enhance receptor affinity or modify efficacy in developing new therapeutic agents. Studies to undertake modification in the properties and/or position of the aryl substituents in subsequent series to further define that role are in progress.

Structure-activity relationship study of a diverse set of estrogen receptor ligands (I) using MultiCASE expert system

The MultiCASE expert system was used to construct a quantitative structure-activity relationship model to screen chemicals with estrogen receptor (ER) binding potential. Structures and ER binding data of 313 chemicals were used as inputs to train the expert system. The training data set covers inactive, weak as well as very powerful ER binders and represents a variety of chemical compounds. Substructural features associated with ER binding activity (biophores) and features that prevent receptor binding (biophobes) were identified. Although a single phenolic hydroxyl group was found to be the most important biophore responsible for the estrogenic activity of most of the chemicals, MultiCASE also identified other biophores and structural features that modulate the activity of the chemicals. Furthermore, the findings supported our previous hypothesis that a 6 A distant descriptor may describe a ligand-binding site on an ER. Quantitative structure-activity relationship models for the chemicals associated with each biophore were constructed as part of the expert system and can be used to predict the activity of new chemicals. The model was cross validated via 10 x 10%-off tests, giving an average concordance of 84.04%.

Study on molecular mechanism and 3D-QSAR of influenza neuraminidase inhibitors

Neuraminidase (NA) is a critical enzyme of the influenza virus and many inhibitors targeting to this enzyme are quite efficient and encouraging as anti-influenza agents. In this paper the binding model of five series of inhibitors to NA was examined using molecular simulation method. The resulted conformation and orientation of the compounds were directly put into CoMSIA study. The most significant amino acid residues at binding sites and the requirement for features of substituents were applied to direct design of new inhibitors. The robust QSAR model and its three-dimensional contour map provided guidelines to building novel compounds with new scaffold and for structural optimization of current molecules.

Binding affinity prediction of novel estrogen receptor ligands using receptor-based 3-D QSAR methods

We have recently reported the development of a 3-D QSAR model for estrogen receptor ligands showing a significant correlation between calculated molecular interaction fields and experimentally measured binding affinity. The ligand alignment obtained from docking simulations was taken as basis for a comparative field analysis applying the GRID/GOLPE program. Using the interaction field derived with a water probe and applying the smart region definition (SRD) variable selection procedure, a significant and robust model was obtained (q(2)(LOO)=0.921, SDEP=0.345). To further analyze the robustness and the predictivity of the established model several recently developed estrogen receptor ligands were selected as external test set. An excellent agreement between predicted and experimental binding data was obtained indicated by an external SDEP of 0.531. Two other traditionally used prediction techniques were applied in order to check the performance of the receptor-based 3-D QSAR procedure. The interaction energies calculated on the basis of receptor-ligand complexes were correlated with experimentally observed affinities. Also ligand-based 3-D QSAR models were generated using program FlexS. The interaction energy-based model, as well as the ligand-based 3-D QSAR models yielded models with lower predictivity. The comparison with the interaction energy-based model and with the ligand-based 3-D QSAR models, respectively, indicates that the combination of receptor-based and 3-D QSAR methods is able to improve the quality of prediction.

Development of biologically active compounds by combining 3D QSAR and structure-based design methods

One of the major challenges in computational approaches to drug design is the accurate prediction of the binding affinity of novel biomolecules. In the present study an automated procedure which combines docking and 3D-QSAR methods was applied to several drug targets. The developed receptor-based 3D-QSAR methodology was tested on several sets of ligands for which the three-dimensional structure of the target protein has been solved--namely estrogen receptor, acetylcholine esterase and protein-tyrosine-phosphatase 1B. The molecular alignments of the studied ligands were determined using the docking program AutoDock and were compared with the X-ray structures of the corresponding protein-ligand complexes. The automatically generated protein-based ligand alignment obtained was subsequently taken as basis for a comparative field analysis applying the GRID/GOLPE approach. Using GRID interaction fields and applying variable selection procedures, highly predictive models were obtained. It is expected that concepts from receptor-based 3D QSAR will be valuable tools for the analysis of high-throughput screening as well as virtual screening data.

A model binding site for testing scoring functions in molecular docking

Prediction of interaction energies between ligands and their receptors remains a major challenge for structure-based inhibitor discovery. Much effort has been devoted to developing scoring schemes that can successfully rank the affinities of a diverse set of possible ligands to a binding site for which the structure is known. To test these scoring functions, well-characterized experimental systems can be very useful. Here, mutation-created binding sites in T4 lysozyme were used to investigate how the quality of atomic charges and solvation energies affects molecular docking. Atomic charges and solvation energies were calculated for 172,118 molecules in the Available Chemicals Directory using a semi-empirical quantum mechanical approach by the program AMSOL. The database was first screened against the apolar cavity site created by the mutation Leu99Ala (L99A). Compared to the electronegativity-based charges that are widely used, the new charges and desolvation energies improved ranking of known apolar ligands, and better distinguished them from more polar isosteres that are not observed to bind. To investigate whether the new charges had predictive value, the non-polar residue Met102, which forms part of the binding site, was changed to the polar residue glutamine. The structure of the resulting Leu99Ala and Met102Gln double mutant of T4 lysozyme (L99A/M102Q) was determined and the docking calculation was repeated for the new site. Seven representative polar molecules that preferentially docked to the polar versus the apolar binding site were tested experimentally. All seven bind to the polar cavity (L99A/M102Q) but do not detectably bind to the apolar cavity (L99A). Five ligand-bound structures of L99A/M102Q were determined by X-ray crystallography. Docking predictions corresponded to the crystallographic results to within 0.4A RMSD. Improved treatment of partial atomic charges and desolvation energies in database docking appears feasible and leads to better distinction of true ligands. Simple model binding sites, such as L99A and its more polar variants, may find broad use in the development and testing of docking algorithms.

A validation study on the practical use of automated de novo design

The de novo design program Skelgen has been used to design inhibitor structures for four targets of pharmaceutical interest. The designed structures are compared to modeled binding modes of known inhibitors (i) visually and (ii) by means of a novel similarity measure considering the size and spatial proximity of the maximum common substructure of two small molecules. It is shown that the Skelgen algorithm generates representatives of many inhibitor classes within a very short time and that the new similarity measure is useful for comparing and clustering designed structures. The results demonstrate the necessity of properly defining search constraints in practical applications of de novo design.

Molecular docking of competitive phosphodiesterase inhibitors

Mammalian phosphodiesterases types 3 and 4 (PDE3 and PDE4) hydrolyze cAMP and are essential for the regulation of this intracellular second messenger. These enzymes share structural and biochemical similarities, but each can be distinguished by its sensitivity to isoenzyme-specific, substrate-competitive inhibitors. We present a model configuration for the PDE4 substrate (cAMP) and a PDE4-specific inhibitor (rolipram) within the active site of the enzyme. The docked models were also used to examine the structural consequences of mutations that confer resistance to rolipram and other PDE4-specific inhibitors. The proposed rolipram-binding configuration is consistent with the substrate-competitive nature of inhibition and also provides a structural basis for the observed specificity of binding to the R- versus S-enantiomer. For mutations that render the enzyme rolipram-insensitive, there was generally an inverse relationship between the magnitude of the drug resistance and the distance of the altered residue from the predicted binding site. We observed a direct correlation between the net loss of protein residue interactions (van der Waals contacts and hydrogen bond interactions) and the degree of rolipram resistance. The positions of several drug sensitivity-determinant residues define a surface leading to the substrate- and drug-binding sites, suggesting a possible approach channel leading to the enzyme active site. The binding of other PDE4 inhibitors (high- and low-affinity) was also modeled and used to predict the involvement of residues that were not previously implicated in pharmacological interactions.

Modeling of poly(ADP-ribose)polymerase (PARP) inhibitors. Docking of ligands and quantitative structure-activity relationship analysis

Poly(ADP-ribose)polymerase-1 (PARP-1) is a nuclear enzyme that has recently emerged as an important player in the mechanisms leading to postischemic neuronal death, and PARP inhibitors have been proposed as potential neuroprotective agents. With the aim of clarifying the structural basis responsible for PARP inhibition, we carried out a computational study on 46 inhibitors available through the literature. Our computational approach is composed of three parts. In the first one, representative PARP inhibitors have been docked into the crystallographic structure of the catalytic domain of PARP by using the Autodock 2.4 program. The docking studies thus carried out have provided an alignment scheme that has been instrumental for superimposing all the remaining inhibitors. Upon the basis of this alignment scheme, a quantitative structure-activity relationship (QSAR) analysis has been carried out after electrostatic and steric interaction energies have been computed with the RECEPTOR program. The QSAR analysis yielded a predictive model able to explain much of the variance of the 46-compound data set. The inspection of the QSAR coefficients revealed that the major driving force for potent inhibition is given by the extension of the contact surface between enzyme and inhibitors while electrostatic energy and hydrogen bonding capability play a minor role. Finally, the projection of the QSAR coefficients back onto the X-ray structure of the catalytic domain of PARP provides insights into the role played by specific amino acid residues. This information will be useful to address the design of new selective and potent PARP inhibitors.

Investigations of new lead structures for the design of selective estrogen receptor modulators

Heterocyclic derivatives of (R,S)/(S,R)-1-(2-chloro-4-hydroxyphenyl)-2-(2,6-dichloro-4-hydroxyphenyl)ethylenediamine (L1) were synthesized and tested for estrogen receptor binding. The selection of the heterocycles was based on theoretical consideration. (2R,3S)/(2S,3R)-2-(2-Chloro-4-hydroxyphenyl)-3-(2,6-dichloro-4-hydroxyphenyl)piperazine 2, (4R,5S)/(4S,5R)-4-(2-chloro-4-hydroxyphenyl)-5-(2,6-dichloro-4-hydroxyphenyl)-2-imidazoline 3, and 4-(2-chloro-4-hydroxyphenyl)-5-(2,6-dichloro-4-hydroxyphenyl)imidazole 4 possess a spatial structure with neighboring aromatic rings as is realized in hormonally active [1,2-diphenylethylenediamine]platinum(II) complexes. The 1,2-diphenylethane pharmacophor, however, cannot adapt an antiperiplanar conformation to interact with the estrogen receptor (ER) comparable to synthetic (e.g., diethylstilbestrol (DES)) or steroidal (e.g., estradiol (E2)) estrogens. Due to the different spatial structures, the heterocycles cause only a marginal displacement of E2 from its binding site (relative binding affinity (RBA) < 0.1%). Nevertheless, unequivocally ER mediated gene activation was verified on the MCF-7-2a cell line. Imidazoline 3 as the most active compound reached the maximum effect of E2 (100% activation) in a concentration of 5 x 10(-7) M, while piperazine 2 and imidazole 4 activate luciferase expression only in a small but significant amount of 20% and 27%, respectively. We therefore assigned these heterocyclic compounds to a second class of hormones (type-II-estrogens), which are attached at the ER at different amino acids than DES or E2 (type-I-estrogens).

Structure-based 3D QSAR and design of novel acetylcholinesterase inhibitors

The paper describes the construction, validation and application of a structure-based 3D QSAR model of novel acetylcholinesterase (AChE) inhibitors. Initial use was made of four X-ray structures of AChE complexed with small, non-specific inhibitors to create a model of the binding of recently developed aminopyridazine derivatives. Combined automated and manual docking methods were applied to dock the co-crystallized inhibitors into the binding pocket. Validation of the modelling process was achieved by comparing the predicted enzyme-bound conformation with the known conformation in the X-ray structure. The successful prediction of the binding conformation of the known inhibitors gave confidence that we could use our model to evaluate the binding conformation of the aminopyridazine compounds. The alignment of 42 aminopyridazine compounds derived by the docking procedure was taken as the basis for a 3D QSAR analysis applying the GRID/GOLPE method. A model of high quality was obtained using the GRID water probe, as confirmed by the cross-validation method (q2LOO = 0.937, q2L50%O = 0.910). The validated model, together with the information obtained from the calculated AChE-inhibitor complexes, were considered for the design of novel compounds. Seven designed inhibitors which were synthesized and tested were shown to be highly active. After performing our modelling study the X-ray structure of AChE complexed with donepezil, an inhibitor structurally related to the developed aminopyirdazines, has been made available. The good agreement found between the predicted binding conformation of the aminopyridazines and the one observed for donepezil in the crystal structure further supports our developed model.