Paolo Benedetti

FLAP: GRID molecular interaction fields in virtual screening. validation using the DUD data set.

The performance of FLAP (Fingerprints for Ligands and Proteins) in virtual screening is assessed using a subset of the DUD (Directory of Useful Decoys) benchmarking data set containing 13 targets each with more than 15 different chemotype classes. A variety of ligand and receptor-based virtual screening approaches are examined, using combinations of individual templates 2D structures of known actives, a cocrystallized ligand, a receptor structure, or a cocrystallized ligand-biased receptor structure. We examine several data fusion approaches to combine the results of the individual virtual screens. In doing so, we show that excellent chemotype enrichment is achieved in both single target ligand-based and receptor-based approaches, of approximately 17-fold over random on average at a false positive rate of 1%. We also show that using as much starting knowledge as possible improves chemotype enrichment, and that data fusion using Pareto ranking is an effective method to do this giving up to 50% improvement in enrichment over the single methods. Finally we show that if inactivity or decoy data is incorporated, automatically training the scoring function in FLAP improves recovery still further, with almost 2-fold improvement over the enrichments shown by the single methods. The results clearly demonstrate the utility of FLAP for virtual screening when either a limited or wide range of prior knowledge is available.

Isozyme-specific ligands for O-acetylserine sulfhydrylase, a novel antibiotic target.

The last step of cysteine biosynthesis in bacteria and plants is catalyzed by O-acetylserine sulfhydrylase. In bacteria, two isozymes, O-acetylserine sulfhydrylase-A and O-acetylserine sulfhydrylase-B, have been identified that share similar binding sites, although the respective specific functions are still debated. O-acetylserine sulfhydrylase plays a key role in the adaptation of bacteria to the host environment, in the defense mechanisms to oxidative stress and in antibiotic resistance. Because mammals synthesize cysteine from methionine and lack O-acetylserine sulfhydrylase, the enzyme is a potential target for antimicrobials. With this aim, we first identified potential inhibitors of the two isozymes via a ligand- and structure-based in silico screening of a subset of the ZINC library using FLAP. The binding affinities of the most promising candidates were measured in vitro on purified O-acetylserine sulfhydrylase-A and O-acetylserine sulfhydrylase-B from Salmonella typhimurium by a direct method that exploits the change in the cofactor fluorescence. Two molecules were identified with dissociation constants of 3.7 and 33 µM for O-acetylserine sulfhydrylase-A and O-acetylserine sulfhydrylase-B, respectively. Because GRID analysis of the two isoenzymes indicates the presence of a few common pharmacophoric features, cross binding titrations were carried out. It was found that the best binder for O-acetylserine sulfhydrylase-B exhibits a dissociation constant of 29 µM for O-acetylserine sulfhydrylase-A, thus displaying a limited selectivity, whereas the best binder for O-acetylserine sulfhydrylase-A exhibits a dissociation constant of 50 µM for O-acetylserine sulfhydrylase-B and is thus 8-fold selective towards the former isozyme. Therefore, isoform-specific and isoform-independent ligands allow to either selectively target the isozyme that predominantly supports bacteria during infection and long-term survival or to completely block bacterial cysteine biosynthesis.

Exposition and reactivity optimization to predict sites of metabolism in chemicals

Chemical modifications of drugs induced by phase I biotransformations significantly affect their pharmacokinetic properties. Because the metabolites produced can themselves have a pharmacological effect and an intrinsic toxicity, medicinal chemists need to accurately predict the sites of metabolism (SoM) of drugs as early as possible. However, site of metabolism prediction is rarely accompanied by a prediction of the relative abundance of the various metabolites. Such a prediction would be a great help in the study of drug– drug interactions and in the process of reducing the toxicity of potential drug candidates. The aim of this paper is to present recent developments in the prediction of xenobiotic metabolism and to use concrete examples to explain the computational mechanism employed.

A Pipeline To Enhance Ligand Virtual Screening: Integrating Molecular Dynamics and Fingerprints for Ligand and Proteins

The importance of taking into account protein flexibility in drug design and virtual ligand screening (VS) has been widely debated in the literature, and molecular dynamics (MD) has been recognized as one of the most powerful tools for investigating intrinsic protein dynamics. Nevertheless, deciphering the amount of information hidden in MD simulations and recognizing a significant minimal set of states to be used in virtual screening experiments can be quite complicated. Here we present an integrated MD-FLAP (molecular dynamics-fingerprints for ligand and proteins) approach, comprising a pipeline of molecular dynamics, clustering and linear discriminant analysis, for enhancing accuracy and efficacy in VS campaigns. We first extracted a limited number of representative structures from tens of nanoseconds of MD trajectories by means of the k-medoids clustering algorithm as implemented in the BiKi Life Science Suite ( http://www.bikitech.com [accessed July 21, 2015]). Then, instead of applying arbitrary selection criteria, that is, RMSD, pharmacophore properties, or enrichment performances, we allowed the linear discriminant analysis algorithm implemented in FLAP ( http://www.moldiscovery.com [accessed July 21, 2015]) to automatically choose the best performing conformational states among medoids and X-ray structures. Retrospective virtual screenings confirmed that ensemble receptor protocols outperform single rigid receptor approaches, proved that computationally generated conformations comprise the same quantity/quality of information included in X-ray structures, and pointed to the MD-FLAP approach as a valuable tool for improving VS performances.

GBR compounds and mepyramines as cocaine abuse therapeutics: chemometric studies on selectivity using grid independent descriptors (GRIND).

Cocaine is one of the most widely abused drugs in the industrial world. Substantial evidence has accumulated that the dopamine transporter (DAT) is a key target for cocaine regarding its reinforcing effects. This work describes the application of chemometric methods to a data set of 54 N(1)-benzhydryl-oxy-alkyl-N(4)-phenyl-alk(en)yl-piperazines (GBR compounds) and chemically related mepyramines as putative candidates in cocaine abuse therapy. The aim of the study is to gain insight into the structural requirements that determine the affinity of the data set molecules to the DAT and the serotonin transporter (SERT) as well as their inhibitory potency on dopamine uptake. The compounds in the dataset are described using the recently developed GRID independent descriptors (GRIND), which allow one to obtain fast three-dimensional quantitative structure-activity relationship models without the need of aligning and superimposing the structures; the results are interpreted in a convenient pharmacophoric-like fashion. In the first part of the work, the selectivity of the database molecules for DAT binding vs dopamine reuptake inhibition is investigated. In the second part, the selectivity of the compounds for DAT binding vs SERT binding is studied. In both cases, significant models are obtained, which define the structural features responsible for the respective selectivity profiles. Moreover, the information has potential interest for the design of new derivatives with improved selectivity.

Inhibitor of ovarian cancer cells growth by virtual screening: a new thiazole derivative targeting human thymidylate synthase.

Human thymidylate synthase (hTS) was targeted through a virtual screening approach. The most optimal inhibitor identified, 2-{4-hydroxy-2-[(2-hydroxybenzylidene)hydrazono]-2,5-dihydrothiazol-5-yl}-N-(3-trifluoromethylphenyl)acetamide (5), showed a mixed-type inhibition pattern, with a K(i) of 1.3 μM and activity against ovarian cancer cell lines with the same potency as cisplatin. X-ray studies revealed that it binds the inactive enzyme conformation. This study is the first example of a nonpeptidic inhibitor that binds the inactive hTS and exhibits anticancer activity against ovarian cancer cells.

Targeting Cystalysin, a Virulence Factor of Treponema denticola‐Supported Periodontitis

Cystalysin from Treponema denticola is a pyridoxal 5′‐phosphate dependent lyase that catalyzes the formation of pyruvate, ammonia, and sulfide from cysteine. It is a virulence factor in adult periodontitis because its reaction contributes to hemolysis, which sustains the pathogen. Therefore, it was proposed as a potential antimicrobial target. To identify specific inhibitors by structure‐based in silico methods, we first validated the crystal structure of cystalysin as a reliable starting point for the design of ligands. By using single‐crystal absorption microspectrophotometry, we found that the enzyme in the crystalline state, with respect to that in solution, exhibits: 1) the same absorption spectra for the catalytic intermediates, 2) a close pKa value for the residue controlling the keto enamine ionization, and 3) similar reactivity with glycine, L‐serine, L‐methionine, and the nonspecific irreversible inhibitor aminoethoxyvinylglycine. Next, we screened in silico a library of 9357 compounds with the Fingerprints for Ligands and Proteins (FLAP) software, by using the three‐dimensional structure of cystalysin as a template. From the library, 17 compounds were selected and experimentally evaluated by enzyme assays and spectroscopic methods. Two compounds were found to competitively inhibit recombinant T. denticola cystalysin, with inhibition constant (Ki) values of 25 and 37 μM. One of them exhibited a minimum inhibitory concentration (MIC) value of 64 μg mL−1 on Moraxella catarrhalis ATCC 23246, which proves its ability to cross bacterial membranes.

Determination of free fatty acids in beer wort

The importance of free fatty acids (FFAs) in wort has been known for a long time because of their influence on beer quality and yeast metabolism. Lipids have a beneficial effect on yeast growth during fermentation as well as negative effects on beer quality. Lipids content of beer affects the ability to form a stable head of foam and plays an important role in beer staling. Moreover, the ratio of unsaturated and saturated fatty acids seems to be related to gushing problems. A novel, simple, and reliable procedure for quantitative analysis of FFAs in wort was developed and validated. The determination of FFAs in wort was achieved via liquid-liquid cartridge extraction, purification of FFA fraction by solid phase extraction, boron trifluoride in methanol methylation, and injection into GC-FID system. The proposed method has high accuracy (<0.3%, expressed as the bias), high precision (<1.2%, RSD), and recoveries ranging from 74% to 98%. The method was tested on two different wort samples (9° and 12° Plato).

Distant collaboration in drug discovery: The LINK3D project

The work describes the development of novel software supporting synchronous distant collaboration between scientists involved in drug discovery and development projects. The program allows to visualize and share data as well as to interact in real time using standard intranets and Internet resources. Direct visualization of 2D and 3D molecular structures is supported and original tools for facilitating remote discussion have been integrated. The software is multiplatform (MS-Windows, SGI-IRIX, Linux), allowing for a seamless integration of heterogeneous working environments. The project aims to support collaboration both within and between academic and industrial institutions. Since confidentiality is very important in some scenarios, special attention has been paid to security aspects. The article presents the research carried out to gather the requirements of collaborative software in the field of drug discovery and development and describes the features of the first fully functional prototype obtained. Real-world testing activities carried out on this prototype in order to guarantee its adequacy in diverse environments are also described and discussed.

From Experiments to a Fast Easy-to-Use Computational Methodology to Predict Human Aldehyde Oxidase Selectivity and Metabolic Reactions

Aldehyde oxidase (AOX) is a molibdo-flavoenzyme that has raised great interest in recent years, since its contribution in xenobiotic metabolism has not always been identified before clinical trials, with consequent negative effects on the fate of new potential drugs. The fundamental role of AOX in metabolizing xenobiotics is also due to the attempt of medicinal chemists to stabilize candidates toward cytochrome P450 activity, which increases the risk for new compounds to be susceptible to AOX nucleophile attack. Therefore, novel strategies to predict the potential liability of new entities toward the AOX enzyme are urgently needed to increase effectiveness, reduce costs, and prioritize experimental studies. In the present work, we present the most up-to-date computational method to predict liability toward human AOX (hAOX), for applications in drug design and pharmacokinetic optimization. The method was developed using a large data set of homogeneous experimental data, which is also disclosed as Supporting Information .