3D QSAR in modern drug design

Develop a High-Throughput Screening Method to Identify C-P4H1 (Collagen Prolyl 4-Hydroxylase 1) Inhibitors from FDA-Approved Chemicals

Collagen prolyl 4-hydroxylase 1 (C-P4H1) is an α-ketoglutarate (α-KG)-dependent dioxygenase that catalyzes 4-hydroxylation of proline on collagen. C-P4H1-induced prolyl hydroxylation is required for proper collagen deposition and cancer metastasis. Therefore, targeting C-P4H1 is considered a potential therapeutic strategy for collagen-related cancer progression and metastasis. However, no C-P4H1 inhibitors are available for clinical testing, and the high content assay is currently not available for C-P4H1 inhibitor screening. In the present study, we developed a high-throughput screening assay by quantifying succinate, a byproduct of C-P4H-catalyzed hydroxylation. C-P4H1 is the major isoform of collagen prolyl 4-hydroxylases (CP4Hs) that contributes the majority prolyl 4-hydroxylase activity. Using C-P4H1 tetramer purified from the eukaryotic expression system, we showed that the Succinate-GloTM Hydroxylase assay was more sensitive for measuring C-P4H1 activity compared with the hydroxyproline colorimetric assay. Next, we performed high-throughput screening with the FDA-approved drug library and identified several new C-P4H1 inhibitors, including Silodosin and Ticlopidine. Silodosin and Ticlopidine inhibited C-P4H1 activity in a dose-dependent manner and suppressed collagen secretion and tumor invasion in 3D tissue culture. These C-P4H1 inhibitors provide new agents to test clinical potential of targeting C-P4H1 in suppressing cancer progression and metastasis.

Teaching and Learning Computational Drug Design: Student Investigations of 3D Quantitative Structure-Activity Relationships through Web Applications

The increasing use of information technology in the discovery of new molecular entities encourages the use of modern molecular-modeling tools to help teach important concepts of drug design to chemistry and pharmacy undergraduate students. In particular, statistical models such as quantitative structure-activity relationships (QSAR)-often as its 3D QSAR variant-are commonly used in the development and optimization of a leading compound. We describe how these drug discovery methods can be taught and learned by means of free and open-source web applications, specifically the online platform www.3d-qsar.com. This new suite of web applications has been integrated into a drug design teaching course, one that provides both theoretical and practical perspectives. We include the teaching protocol by which pharmaceutical biotechnology master students at Pharmacy Faculty of Sapienza Rome University are introduced to drug design. Starting with a choice among recent articles describing the potencies of a series of molecules tested against a biological target, each student is expected to build a 3D QSAR ligand-based model from their chosen publication, proceeding as follows: creating the initial data set (Py-MolEdit); generating the global minimum conformations (Py-ConfSearch); proposing a promising mutual alignment (Py-Align); and finally, building, and optimizing a robust 3D QSAR models (Py-CoMFA). These student activities also help validate these new molecular modeling tools, especially for their usability by inexperienced hands. To more fully demonstrate the effectiveness of this protocol and its tools, we include the work performed by four of these students (four of the coauthors), detailing the satisfactory 3D QSAR models they obtained. Such scientifically complete experiences by undergraduates, made possible by the efficiency of the 3D QSAR methodology, provide exposure to computational tools in the same spirit as traditional laboratory exercises. With the obsolescence of the classic Comparative Molecular Field Analysis Sybyl host, the 3dqsar web portal offers one of the few available means of performing this well-established 3D QSAR method.

Structure Based Thermostability Prediction Models for Protein Single Point Mutations with Machine Learning Tools

Thermostability issue of protein point mutations is a common occurrence in protein engineering. An application which predicts the thermostability of mutants can be helpful for guiding decision making process in protein design via mutagenesis. An in silico point mutation scanning method is frequently used to find “hot spots” in proteins for focused mutagenesis. ProTherm (http://gibk26.bio.kyutech.ac.jp/jouhou/Protherm/protherm.html) is a public database that consists of thousands of protein mutants’ experimentally measured thermostability. Two data sets based on two differently measured thermostability properties of protein single point mutations, namely the unfolding free energy change (ddG) and melting temperature change (dTm) were obtained from this database. Folding free energy change calculation from Rosetta, structural information of the point mutations as well as amino acid physical properties were obtained for building thermostability prediction models with informatics modeling tools. Five supervised machine learning methods (support vector machine, random forests, artificial neural network, naïve Bayes classifier, K nearest neighbor) and partial least squares regression are used for building the prediction models. Binary and ternary classifications as well as regression models were built and evaluated. Data set redundancy and balancing, the reverse mutations technique, feature selection, and comparison to other published methods were discussed. Rosetta calculated folding free energy change ranked as the most influential features in all prediction models. Other descriptors also made significant contributions to increasing the accuracy of the prediction models.

Modeling the excitation wavelengths (lambda(ex)) of boronic acids

The quantitative structure-property relationship (QSPR) method was used to model the fluorescence excitation wavelengths (lambda(ex)) of 42 boronic acid-based fluorescent biosensors (30 in the training set and 12 in the test set). In this QSPR study, unsupervised forward selection (UFS), stepwise multiple linear regression (SMLR), partial least squares regression (PLS) and associative neural networks (ASNN) were employed to simulate linear and nonlinear models. All models were validated by a test set and Tropsha's validation model. The resulting ASNN nonlinear model demonstrates significant improvement on the predictive ability of the neural network compared to the SMLR and PLS linear models. The descriptors used in the models are discussed in detail. These QSPR models are useful tools for the prediction of fluorescence excitation wavelengths of arylboronic acids.

High-throughput ligand screening via preclustering and evolved neural networks

The pathway for novel lead drug discovery has many major deficiencies, the most significant of which is the immense size of small molecule diversity space. Methods that increase the search efficiency and/or reduce the size of the search space, increase the rate at which useful lead compounds are identified. Artificial neural networks optimized via evolutionary computation provide a cost and time-effective solution to this problem. Here, we present results that suggest preclustering of small molecules prior to neural network optimization is useful for generating models of quantitative structure-activity relationships for a set of HIV inhibitors. Using these methods, it is possible to prescreen compounds to separate active from inactive compounds or even actives and mildly active compounds from inactive compounds with high predictive accuracy while simultaneously reducing the feature space. It is also possible to identify "human interpretable" features from the best models that can be used for proposal and synthesis of new compounds in order to optimize potency and specificity.