Combining docking, scoring and molecular field analyses to probe influenza neuraminidase-ligand interactions

In this project, several docking conditions, scoring functions and corresponding protein-aligned molecular field analysis (CoMFA) models were evaluated for a diverse set of neuraminidase (NA) inhibitors. To this end, a group of inhibitors were docked into the active site of NA. The docked structures were utilized to construct a 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 16 different docking configurations were evaluated, of which some succeeded in producing self-consistent and predictive CoMFA models. However, the best model coincided with docking the ionized ligands into the hydrated form of the binding site via PLP1 scoring function (r2LOO=0.735, r2PRESS against 24 test compounds=0.828). The highest-ranking CoMFA models were employed to probe NA-ligand interactions. Further validation by comparison with a co-crystallized ligand-NA crystallographic structure was performed. This combination of docking/scoring/CoMFA modeling provided interesting insights into the binding of different NA inhibitors.     

3D-QSAR and molecular docking studies of 1,3,5-triazene-2,4-diamine derivatives against r-RNA: novel bacterial translation inhibitors

Aminoglycoside mimetics inhibit bacterial translation by interfering with the ribosomal decoding site. To elucidate the structural properties of these compounds important for antibacterial activity, comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were applied to a set of 56 aminoglycosides mimetics. The successful CoMFA model yielded the leave-one-out (LOO) cross-validated correlation coefficient (q(2)) of 0.708 and a non-cross-validated correlation coefficient (r(2)) of 0.967. CoMSIA model gave q(2)=0.556 and r(2)=0.935. The CoMFA and CoMSIA models were validated with 36 test set compounds and showed a good r(pred)(2) of 0.624 and 0.640, respectively. Contour maps of the two QSAR approaches show that electronic effects dominantly determine the binding affinities. These obtained results were agreed well with the experimental observations and docking studies. The results not only lead to a better understanding of structural requirements of bacterial translation inhibitors but also can help in the design of novel bacterial translation inhibitors.     

A CoMFA study of COX-2 inhibitors with receptor based alignment

A diverse set of 53 cyclooxygenase-2 (COX-2) inhibitors which were aligned in two different ways were subjected to CoMFA analysis. The first method of alignment of the molecules was based on the binding information sourced from the crystallographic study, from which CoMFA Model 1 was derived. The second mode of alignment was generated by docking the inhibitors in the binding pocket using the DOCK and AFFINITY suite of programs; this gave a second model. The CoMFA Model 2 was slightly better than Model 1 in terms of the statistical parameters r(2) and q(2). The two models could predict very well the activity of a test set of diverse molecules, with a predictive r(2) of 0.593 and 0.768, respectively. Besides the QSAR results, the docking studies give a deep insight into the H-bonding interactions between the inhibitors and residues in the active site of the enzyme, which can be exploited in designing better inhibitors. Useful ideas on activity improvement could be gleaned from these models.     

Three-dimensional quantitative structure-activity relationship (3D-QSAR) analyses of choline acetyltransferase inhibitors

As a basis for predicting structural features that may lead to the design of more potent and selective inhibitors of choline acetyltransferase (ChAT), the three-dimensional quantitative structure-activity relationship (3D-QSAR) studies were carried out on a series of trans-1-methyl-4-(1-naphthylvinyl)pyridinium (MNVP+) analogs, which are known ChAT inhibitors. 3D-QSAR studies were carried out using the comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) methods. Since these inhibitors have extremely shallow potential energy minimum energy wells and low barriers to rotation, two dihedral angles unique to these inhibitors were systematically modified to reflect the energetically preferred conformations as determined by force field calculations. An optimum alignment rule was devised based on the conformations obtained from the molecular mechanics studies, using a common substructure alignment method. The studies involve a set of 21 compounds and experimentally determined molar IC50 values were used as the dependent variable in the analysis. The 3D-QSAR models have conventional r2-values of 0.953 and 0.954 for CoMFA and CoMSIA, respectively; similarly, cross-validated coefficient q2-values of 0.755 and 0.834 for CoMFA and CoMSIA, respectively, were obtained. On the basis of these predictive r2-values the model was tested using previously determined IC50 values. CoMSIA 3D-QSAR yielded better results than CoMFA.     

Three-dimensional QSAR of HPPD inhibitors, PSA inhibitors, and anxiolytic agents: effect of tautomerism on the CoMFA models

The present study was design to examine the effect of tautomerism upon the CoMFA results. Three selected data sets involving protropic tautomerism, which are 21 p-hydroxyphenylpyruvate dioxygenase (HPPD) inhibitors, 35 inhibitors of puromycin-sensitive aminopeptidase (PSA), and 67 anxiolytic agents, were used for this purpose. Atom-by-atom alignment technique was adopted to superimpose the molecules in the data sets onto a template. The structural alignments using different tautomeric forms had no significant difference except the atoms involved in tautomerism, which ensures, to a great extent, that the differences of the CoMFA results result primarily from the tautomerism. All-orientation and all-placement search (AOS-APS) based CoMFA models, in addition to the conventional ones, were derived for each system and proved to be capable of yielding much improved statistical results. In the cases of the data sets of HPPD inhibitors and PSA inhibitors, excellent AOS-APS CoMFA models (q2>0.8 with four components for the former and q2>0.7 with seven components for the latter) were obtained, and almost no significant difference in statistical quality was observed when using different tautomeric forms to derive the models. However, it was not the case when treating the data set of anxiolytic agents. The keto tautomer, which was the active form of the PBI type inhibitors, produced measurably better results (q2=0.54 with eight components) than that the enol one (q2=0.37 with five components), indicating the importance of selecting proper tautomer in the CoMFA studies. Furthermore, there existed some substantial differences of the electrostatic field contours between the two different tautomeric forms for all of the three systems considered, whereas the differences in the steric field contour maps were limited. This implies that the resulting new potent ligands may be quite different if one utilizes the CoMFA models of different tautomeric forms for guiding further structural refinements.     

CoMFA based de novo design of pyridazine analogs as PTP1B inhibitors

PTP1B plays an important role as a negative regulator in insulin and leptin signaling pathways. Potent and orally active PTP1B inhibitors can act as potential agents for the treatment of Type 2 diabetes and obesity. CoMFA (Comparative Molecular Field Analysis) and de novo ligand design using LeapFrog (LF) studies were performed on pyridazine analogs, reported to be selective and non-competitive inhibitors of PTP1B. A robust model was developed which produced statistically significant results with cross-validated and conventional correlation coefficients of 0.619 and 0.990, respectively. Further, the robustness of the model was verified by bootstrapping analysis. LeapFrog (LF) program is a de novo drug discovery tool, which uses CoMFA maps to generate hypothetical cavity and ligands. As the crystal structure of PTP1B-pyridazine complex is not yet known, the contours of CoMFA model was used to serve as a pharmacophoric model to generate hypothetical cavity for LeapFrog calculations. Ligands were optimized using this concept.     

3D QSAR studies on a series of potent and high selective inhibitors for three kinases of RTK family

For targets belonging to the same family of receptors, inhibitors often act at more than one biological target and produce a synergistic effect. Separate CoMFA and CoMSIA models were developed from our data set for the KDR, cKit and Tie-2 inhibitors. These models showed excellent internal predictability and consistency, and validation using test-set compounds yielded a good predictive power for the pIC(50) value. The field contour maps (CoMFA and CoMSIA) corresponding to the KDR, cKit and Tie-2 kinase subtypes reflected the characteristic similarities and differences between these types. These maps provided valuable information to facilitate structural modifications of the inhibitor to increase selectivity for the KDR over cKit and Tie-2.     

光系统II抑制剂的三维构效关系和相关性研究

光系统II(PSII)抑制剂要导致PSII反应中心的电子传递受阻，被认为是重要的除草剂。近年来发现许多化合物都有希尔反应抑制活性。本文利用比较分子场分析方法，在药效团模型基础上，对不同系列的光系统II抑制剂的结构和活性关系进行了分析和讨论。     

Structural investigation of PAP derivatives by CoMFA and CoMSIA reveals novel insight towards inhibition of Bcr-Abl oncoprotein

Molecular modeling by 3D-QSAR comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were employed on a series of phenylaminopyrimidine-based (PAP) Bcr-Abl inhibitors. The chemical structures of 63 PAP analogues were aligned using a template extracted from the crystal structure of STI571 bound to Abl kinase. Subsequently, the structures built were divided into training and test sets that include 53 and 10 compounds, respectively. Statistical results showed that the 3D-QSAR models generated from CoMSIA were superior to CoMFA (CoMSIA; q2=0.66, r2=0.94, N=3, F=139.09, r2pred=0.64 while CoMFA; q2=0.53, r2=0.73, N=3, F=43.53, r2pred=0.61). Based on the contour interpretation, the attachment of hydrophobic and bulky groups to the phenyl and pyrrolidine (D- and E-ring of NS-187, respectively) along with highly electronegative groups around the D-ring are important structural features for the design of second-generation Bcr-Abl inhibitors. The generated models are predictive based on reproducible values of the predicted compared with experimental activities in the test set. Further, the complementary analysis of contour maps to the Bcr-Abl binding site suggested the anchor points for binding affinity.     

3D-QSAR and molecular docking studies on pyrazolopyrimidine derivatives as glycogen synthase kinase-3beta inhibitors

Glycogen synthase kinase-3 (GSK-3), a serine/threonine kinase, is a fascinating enzyme with diverse biological actions in intracellular signaling systems, making it an emerging target for diseases such as diabetes mellitus, cancer, chronic inflammation, bipolar disorders and Alzheimer's disease. It is important to inhibit GSK-3 selectively and the net effect of the GSK-3 inhibitors thus should be target specific, over other phylogenetically related kinases such as CDK-2. In the present work, we have carried out three-dimensional quantitative structure activity relationship (3D-QSAR) studies on novel class of pyrazolopyrimidine derivatives as GSK-3 inhibitors reported to have improved cellular activity. Docked conformation of the most active molecule in the series, which shows desirable interactions in the receptor, was taken as template for alignment of the molecules. Statistically significant CoMFA and CoMSIA models were generated using 49 molecules in training set. By applying leave-one-out (LOO) cross-validation study, r(cv)2 values of 0.53 and 0.48 for CoMFA and CoMSIA, respectively and non-cross-validated (r(ncv)2) values of 0.98 and 0.92 were obtained for CoMFA and CoMSIA models, respectively. The predictive ability of CoMFA and CoMSIA models was determined using a test set of 12 molecules which gave predictive correlation coefficients (r(pred)2) of 0.47 and 0.48, respectively, indicating good predictive power. Based upon the information derived from CoMFA and CoMSIA contour maps, we have identified some key features that explain the observed variance in the activity and have been used to design new pyrazolopyrimidine derivatives. The designed molecules showed better binding affinity in terms of estimated docking scores with respect to the already reported systems; hence suggesting that newly designed molecules can be more potent and selective towards GSK-3beta inhibition.