Accelerating Drug Discovery by Early Protein Drug Target Prediction Based on a Multi-Fingerprint Similarity Search

From Deep Learning to the Discovery of Promising VEGFR-2 Inhibitors

Vascular endothelial growth factor receptor 2 (VEGFR-2) stands as a prominent therapeutic target in oncology, playing a critical role in angiogenesis, tumor growth, and metastasis. FDA-approved VEGFR-2 inhibitors are associated with diverse side effects. Thus, finding novel and more effective inhibitors is of utmost importance. In this study, a deep learning (DL) classification model was first developed and then employed to select putative active VEGFR-2 inhibitors from an in-house chemical library including 187 druglike compounds. A pool of 18 promising candidates was shortlisted and screened against VEGFR-2 by using molecular docking. Finally, two compounds, RHE-334 and EA-11, were prioritized as promising VEGFR-2 inhibitors by employing PLATO, our target fishing and bioactivity prediction platform. Based on this rationale, we prepared RHE-334 and EA-11 and successfully tested their anti-proliferative potential against MCF-7 human breast cancer cells with IC50 values of 26.78±4.02 and 38.73±3.84 μM, respectively. Their toxicities were instead challenged against the WI-38. Interestingly, expression studies indicated that, in the presence of RHE-334, VEGFR-2 was equal to 0.52±0.03, thus comparable to imatinib equal to 0.63±0.03. In conclusion, this workflow based on theoretical and experimental approaches demonstrates effective in identifying VEGFR-2 inhibitors and can be easily adapted to other medicinal chemistry goals.

Inhibition of SARS-CoV-2 NSP-15 by Uridine-5'-Monophosphate Analogues Using QSAR Modelling, Molecular Dynamics Simulations, and Free Energy Landscape

SARS-CoV-2 is accountable for severe social and economic disruption around the world causing COVID-19. Non-structural protein-15 (NSP15) possesses a domain that is vital to the viral life cycle and is known as uridylate-specific endoribonuclease (EndoU). This domain binds to the uridine 5'-monophosphate (U5P) so that the protein may carry out its native activity. It is considered a vital drug target to inhibit the growth of the virus. Thus, in this current study, ML-based QSAR and virtual screening of U5P analogues targeting Nsp15 were performed to identify potential molecules against SARS-CoV-2. Screening of 816 unique U5P analogues using ML-based QSAR identified 397 compounds ranked on their predicted bioactivity (pIC50). Further, molecular docking and hydrogen bond interaction analysis resulted in the selection of the top three compounds (53309102, 57398422, and 76314921). Molecular dynamics simulation of the most promising compounds showed that two molecules 53309102 and 57398422 acted as potential binders of Nsp15. The compound was able to inhibit nsp15 activity as it was successfully bound to the active site of the nsp15 protein. This was achieved by the formation of relevant contacts with enzymatically critical amino acid residues (His235, His250, and Lys290). Principal component analysis and free energy landscape studies showed stable complex formation while MM/GBSA calculation showed lower binding energies for 53309102 (ΔGTOTAL = -29.4 kcal/mol) and 57398422 (ΔGTOTAL = -39.4 kcal/mol) compared to the control U5P (ΔGTOTAL = -18.8 kcal/mol). This study aimed to identify analogues of U5P inhibiting the NSP15 function that potentially could be used for treating COVID-19.

CIRCE: Web-Based Platform for the Prediction of Cannabinoid Receptor Ligands Using Explainable Machine Learning

The endocannabinoid system, which includes cannabinoid receptor 1 and 2 subtypes (CB1R and CB2R, respectively), is responsible for the onset of various pathologies including neurodegeneration, cancer, neuropathic and inflammatory pain, obesity, and inflammatory bowel disease. Given the high similarity of CB1R and CB2R, generating subtype-selective ligands is still an open challenge. In this work, the Cannabinoid Iterative Revaluation for Classification and Explanation (CIRCE) compound prediction platform has been generated based on explainable machine learning to support the design of selective CB1R and CB2R ligands. Multilayer classifiers were combined with Shapley value analysis to facilitate explainable predictions. In test calculations, CIRCE predictions reached ∼80% accuracy and structural features determining ligand predictions were rationalized. CIRCE was designed as a web-based prediction platform that is made freely available as a part of our study.

Synthesis of Novel Benzo[b][1,6]naphthyridine Derivatives and Investigation of Their Potential as Scaffolds of MAO Inhibitors

In this work, 2-alkyl-10-chloro-1,2,3,4-tetrahydrobenzo[b][1,6]naphthyridines were obtained and their reactivity was studied. Novel derivatives of the tricyclic scaffold, including 1-phenylethynyl (5), 1-indol-3-yl (8), and azocino[4,5-b]quinoline (10) derivatives, were synthesized and characterized herein for the first time. Among the newly synthesized derivatives, 5c-h proved to be MAO B inhibitors with potency in the low micromolar range. In particular, the 1-(2-(4-fluorophenyl)ethynyl) analog 5g achieved an IC₅₀ of 1.35 μM, a value close to that of the well-known MAO B inhibitor pargyline.

Artificial intelligence-driven identification of morin analogues acting as CaV1.2 channel blockers: Synthesis and biological evaluation

Morin is a vasorelaxant flavonoid, whose activity is ascribable to Ca_V1.2 channel blockade that, however, is weak as compared to that of clinically used therapeutic agents. A conventional strategy to circumvent this drawback is to synthesize new derivatives differently decorated and, in this context, morin-derivatives able to interact with Ca_V1.2 channels were found by employing the potential of PLATO in target fishing and reverse screening. Three different derivatives (5a-c) were selected as promising tools, synthesized, and investigated in in vitro functional studies using rat aorta rings and rat tail artery myocytes. 5a-c were found more effective vasorelaxant agents than the naturally occurring parent compound and antagonized both electro- and pharmaco-mechanical coupling in an endothelium-independent manner. 5a, the series' most potent, reduced also Ca²⁺ mobilization from intracellular store sites. Furthermore, 5a≈5c > 5b inhibited Ba²⁺ current through Ca_V1.2 channels. However, compound 5a caused also a concentration-dependent inhibition of K_Ca1.1 channel currents.

An Integrated Machine Learning Model To Spot Peptide Binding Pockets in 3D Protein Screening

The prediction of peptide-protein binding sites is of utmost importance to tackle the onset of severe neurodegenerative diseases and cancer. In this work, we detail a novel machine learning model based on Linear Discriminant Analysis (LDA) demonstrating to be highly predictive in detecting the putative protein binding regions of small peptides. Starting from 439 high-quality pockets derived from peptide-protein crystallographic complexes, three sets of well-established peptide-binding regions were first selected through a Partitioning Around Medoids (PAM) clustering algorithm based on morphological and energetic 3D GRID-MIF molecular descriptors. Next, the best combination between all the putative interacting peptide pockets and related GRID-MIF scores was automatically explored by using the LDA-based protocol implemented in BioGPS. This approach proved successful to recognize the actual interacting peptide regions (that is, AUC = 0.86 and partial ROC enrichment at 5% of 0.48) from all the other pockets of the protein. Validated on two external collections sets, including 445 and 347 crystallographic peptide-protein complexes, our LDA-based model could be effective to further run peptide-protein virtual screening campaigns.

Random-forest model for drug-target interaction prediction via Kullbeck-Leibler divergence

Virtual screening has significantly improved the success rate of early stage drug discovery. Recent virtual screening methods have improved owing to advances in machine learning and chemical information. Among these advances, the creative extraction of drug features is important for predicting drug–target interaction (DTI), which is a large-scale virtual screening of known drugs. Herein, we report Kullbeck–Leibler divergence (KLD) as a DTI feature and the feature-driven classification model applicable to DTI prediction. For the purpose, E3FP three-dimensional (3D) molecular fingerprints of drugs as a molecular representation allow the computation of 3D similarities between ligands within each target (Q–Q matrix) to identify the uniqueness of pharmacological targets and those between a query and a ligand (Q–L vector) in DTIs. The 3D similarity matrices are transformed into probability density functions via kernel density estimation as a nonparametric estimation. Each density model can exploit the characteristics of each pharmacological target and measure the quasi-distance between the ligands. Furthermore, we developed a random forest model from the KLD feature vectors to successfully predict DTIs for representative 17 targets (mean accuracy: 0.882, out-of-bag score estimate: 0.876, ROC AUC: 0.990). The method is applicable for 2D chemical similarity.

Benzothiazole Derivatives Endowed with Antiproliferative Activity in Paraganglioma and Pancreatic Cancer Cells: Structure–Activity Relationship Studies and Target Prediction Analysis

The antiproliferative effects played by benzothiazoles in different cancers have aroused the interest for these molecules as promising antitumor agents. In this work, a library of phenylacetamide derivatives containing the benzothiazole nucleus was synthesized and compounds were tested for their antiproliferative activity in paraganglioma and pancreatic cancer cell lines. The novel synthesized compounds induced a marked viability reduction at low micromolar concentrations both in paraganglioma and pancreatic cancer cells. Derivative 4l showed a greater antiproliferative effect and higher selectivity index against cancer cells, as compared to other compounds. Notably, combinations of derivative 4l with gemcitabine at low concentrations induced enhanced and synergistic effects on pancreatic cancer cell viability, thus supporting the relevance of compound 4l in the perspective of clinical translation. A target prediction analysis was also carried out on 4l by using multiple computational tools, identifying cannabinoid receptors and sentrin-specific proteases as putative targets contributing to the observed antiproliferative activity.

Predicting clinical trial outcomes using drug bioactivities through graph database integration and machine learning

The ability to estimate the probability of a drug to receive approval in clinical trials provides natural advantages to optimizing pharmaceutical research workflows. Success rates of clinical trials have deep implications for costs, duration of development, and under pressure due to stringent regulatory approval processes. We propose a machine learning approach that can predict the outcome of the trial with reliable accuracies, using biological activities, physicochemical properties of the compounds, target-related features, and NLP-based compound representation. In the above list, biological activities have never been used as an independent variable towards the prediction of clinical trial outcomes. We have extracted the drug-disease pair from clinical trials and mapped target(s) to that pair using multiple data sources. Empirical results demonstrate that ensemble learning outperforms independently trained, small-data ML models. We report results and inferences derived from a Random forest classifier with an average accuracy of 93%, and an F1 score of 0.96 for the "Pass" class. "Pass" refers to one of the two classes (Pass/Fail) of all clinical trials, and the model performed well in predicting the "Pass" category. Through the analysis of feature contributions to predictive capability, we have demonstrated that bioactivity plays a statistically significant role in predicting clinical trial outcome. A significant effort has gone into the production of the dataset that, for the first time, integrates clinical trial information with protein targets. Cleaned, organized, integrated data and code to map these entities, created as a part of this work, are available open-source. This reproducibility and the freely available code ensure that researchers with access to deep curated and proprietary clinical trial databases (we only use open-source data in this study) can further expand the scope of the results.

PLATO: A Predictive Drug Discovery Web Platform for Efficient Target Fishing and Bioactivity Profiling of Small Molecules

PLATO (Polypharmacology pLATform predictiOn) is an easy-to-use drug discovery web platform, which has been designed with a two-fold objective: to fish putative protein drug targets and to compute bioactivity values of small molecules. Predictions are based on the similarity principle, through a reverse ligand-based screening, based on a collection of 632,119 compounds known to be experimentally active on 6004 protein targets. An efficient backend implementation allows to speed-up the process that returns results for query in less than 20 s. The graphical user interface is intuitive to give practitioners easy input and transparent output, which is available as a standard report in portable document format. PLATO has been validated on thousands of external data, with performances better than those of other parallel approaches. PLATO is available free of charge (http://plato.uniba.it/ accessed on 13 April 2022).

Bioactivity assessment of natural compounds using machine learning models trained on target similarity between drugs

Natural compounds constitute a rich resource of potential small molecule therapeutics. While experimental access to this resource is limited due to its vast diversity and difficulties in systematic purification, computational assessment of structural similarity with known therapeutic molecules offers a scalable approach. Here, we assessed functional similarity between natural compounds and approved drugs by combining multiple chemical similarity metrics and physicochemical properties using a machine-learning approach. We computed pairwise similarities between 1410 drugs for training classification models and used the drugs shared protein targets as class labels. The best performing models were random forest which gave an average area under the ROC of 0.9, Matthews correlation coefficient of 0.35, and F1 score of 0.33, suggesting that it captured the structure-activity relation well. The models were then used to predict protein targets of circa 11k natural compounds by comparing them with the drugs. This revealed therapeutic potential of several natural compounds, including those with support from previously published sources as well as those hitherto unexplored. We experimentally validated one of the predicted pair’s activities, viz., Cox-1 inhibition by 5-methoxysalicylic acid, a molecule commonly found in tea, herbs and spices. In contrast, another natural compound, 4-isopropylbenzoic acid, with the highest similarity score when considering most weighted similarity metric but not picked by our models, did not inhibit Cox-1. Our results demonstrate the utility of a machine-learning approach combining multiple chemical features for uncovering protein binding potential of natural compounds.

Bipartite network models to design combination therapies in acute myeloid leukaemia

Combination therapy is preferred over single-targeted monotherapies for cancer treatment due to its efficiency and safety. However, identifying effective drug combinations costs time and resources. We propose a method for identifying potential drug combinations by bipartite network modelling of patient-related drug response data, specifically the Beat AML dataset. The median of cell viability is used as a drug potency measurement to reconstruct a weighted bipartite network, model drug-biological sample interactions, and find the clusters of nodes inside two projected networks. Then, the clustering results are leveraged to discover effective multi-targeted drug combinations, which are also supported by more evidence using GDSC and ALMANAC databases. The potency and synergy levels of selective drug combinations are corroborated against monotherapy in three cell lines for acute myeloid leukaemia in vitro. In this study, we introduce a nominal data mining approach to improving acute myeloid leukaemia treatment through combinatorial therapy.

Identifying effective drug combinations to treat cancer is a challenging task, either experimentally or computationally. Here, the authors develop a bipartite network modelling approach to propose drug combination strategies in acute myeloid leukaemia using patient and cell line drug screening data.

Quantitative Polypharmacology Profiling Based on a Multifingerprint Similarity Predictive Approach

We present a new quantitative ligand-based bioactivity prediction approach employing a multifingerprint similarity search algorithm, enabling the polypharmacological profiling of small molecules. Quantitative bioactivity predictions are made on the basis of the statistical distributions of multiple Tanimoto similarity θ values, calculated through 13 different molecular fingerprints, and of the variation of the measured biological activity, reported as ΔpIC₅₀, for all of the ligands sharing a given protein drug target. The application data set comprises as much as 4241 protein drug targets as well as 418 485 ligands selected from ChEMBL (release 25) by employing a set of well-defined filtering rules. Several large internal and external validation studies were carried out to demonstrate the robustness and the predictive potential of the herein proposed method. Additional comparative studies, carried out on two freely available and well-known ligand-target prediction platforms, demonstrated the reliability of our proposed approach for accurate ligand-target matching. Moreover, two applicative cases were also discussed to practically describe how to use our predictive algorithm, which is freely available as a user-friendly web platform. The user can screen single or multiple queries at a time and retrieve the output as a terse html table or as a json file including all of the information concerning the explored similarities to obtain a deeper understanding of the results. High-throughput virtual reverse screening campaigns, allowing for a given query compound the quick detection of the potential drug target from a large collection of them, can be carried out in batch on demand.

Web-Based Quantitative Structure-Activity Relationship Resources Facilitate Effective Drug Discovery

Traditional drug discovery effectively contributes to the treatment of many diseases but is limited by high costs and long cycles. Quantitative structure-activity relationship (QSAR) methods were introduced to evaluate the activity of compounds virtually, which saves the significant cost of determining the activities of the compounds experimentally. Over the past two decades, many web tools for QSAR modeling with various features have been developed to facilitate the usage of QSAR methods. These web tools significantly reduce the difficulty of using QSAR and indirectly promote drug discovery. However, there are few comprehensive summaries of these QSAR tools, and researchers may have difficulty determining which tool to use. Hence, we systematically surveyed the mainstream web tools for QSAR modeling. This work may guide researchers in choosing appropriate web tools for developing QSAR models, and may also help develop more bioinformatics tools based on these existing resources. For nonprofessionals, we also hope to make more people aware of QSAR methods and expand their use.

First-in-Class Isonipecotamide-Based Thrombin and Cholinesterase Dual Inhibitors with Potential for Alzheimer Disease

Recently, the direct thrombin (thr) inhibitor dabigatran has proven to be beneficial in animal models of Alzheimer’s disease (AD). Aiming at discovering novel multimodal agents addressing thr and AD-related targets, a selection of previously and newly synthesized potent thr and factor Xa (fXa) inhibitors were virtually screened by the Multi-fingerprint Similarity Searching aLgorithm (MuSSeL) web server. The N-phenyl-1-(pyridin-4-yl)piperidine-4-carboxamide derivative 1, which has already been experimentally shown to inhibit thr with a K_i value of 6 nM, has been flagged by a new, upcoming release of MuSSeL as a binder of cholinesterase (ChE) isoforms (acetyl- and butyrylcholinesterase, AChE and BChE), as well as thr, fXa, and other enzymes and receptors. Interestingly, the inhibition potency of 1 was predicted by the MuSSeL platform to fall within the low-to-submicromolar range and this was confirmed by experimental K_i values, which were found equal to 0.058 and 6.95 μM for eeAChE and eqBChE, respectively. Thirty analogs of 1 were then assayed as inhibitors of thr, fXa, AChE, and BChE to increase our knowledge of their structure-activity relationships, while the molecular determinants responsible for the multiple activities towards the target enzymes were rationally investigated by molecular cross-docking screening.

Morpholine-based chalcones as dual-acting monoamine oxidase-B and acetylcholinesterase inhibitors: synthesis and biochemical investigations

Nine compounds (MO1–MO9) containing the morpholine moiety were assessed for their inhibitory activities against monoamine oxidases (MAOs) and acetylcholinesterase (AChE). Most of the compounds potently inhibited MAO-B; MO1 most potently inhibited with an IC₅₀ value of 0.030 µM, followed by MO7 (0.25 µM). MO5 most potently inhibited AChE (IC₅₀ = 6.1 µM), followed by MO9 (IC₅₀ = 12.01 µM) and MO7 most potently inhibited MAO-A (IC₅₀ = 7.1 µM). MO1 was a reversible mixed-type inhibitor of MAO-B (K_i = 0.018 µM); MO5 reversibly competitively inhibited AChE (K_i = 2.52 µM); and MO9 reversibly noncompetitively inhibited AChE (K_i = 7.04 µM). MO1, MO5 and MO9 crossed the blood–brain barrier, and were non-toxic to normal VERO cells. These results show that MO1 is a selective inhibitor of MAO-B and that MO5 is a dual-acting inhibitor of AChE and MAO-B, and that both should be considered candidates for the treatment of Alzheimer’s disease.

Trimethoxylated Halogenated Chalcones as Dual Inhibitors of MAO-B and BACE-1 for the Treatment of Neurodegenerative Disorders

Six halogenated trimethoxy chalcone derivatives (CH1-CH6) were synthesized and spectrally characterized. The compounds were further evaluated for their inhibitory potential against monoamine oxidases (MAOs) and β-secretase (BACE-1). Six compounds inhibited MAO-B more effectively than MAO-A, and the 2',3',4'-methoxy moiety in CH4-CH6 was more effective for MAO-B inhibition than the 2',4',6'-methoxy moiety in CH1-CH3. Compound CH5 most potently inhibited MAO-B, with an IC₅₀ value of 0.46 µM, followed by CH4 (IC₅₀ = 0.84 µM). In 2',3',4'-methoxy derivatives (CH4-CH6), the order of inhibition was -Br in CH5 > -Cl in CH4 > -F in CH6 at the para-position in ring B of chalcone. CH4 and CH5 were selective for MAO-B, with selectivity index (SI) values of 15.1 and 31.3, respectively, over MAO-A. CH4 and CH5 moderately inhibited BACE-1 with IC₅₀ values of 13.6 and 19.8 µM, respectively. When CH4 and CH5 were assessed for their cell viability studies on the normal African Green Monkey kidney cell line (VERO) using MTT assays, it was noted that both compounds were found to be safe, and only a slightly toxic effect was observed in concentrations above 200 µg/mL. CH4 and CH5 decreased reactive oxygen species (ROS) levels of VERO cells treated with H₂O₂, indicating both compounds retained protective effects on the cells by antioxidant activities. All compounds showed high blood brain barrier permeabilities analyzed by a parallel artificial membrane permeability assay (PAMPA). Molecular docking and ADME prediction of the lead compounds provided more insights into the rationale behind the binding and the CNS drug likeness. From non-test mutagenicity and cardiotoxicity studies, CH4 and CH5 were non-mutagenic and non-/weak-cardiotoxic. These results suggest that CH4 and CH5 could be considered candidates for the cure of neurological dysfunctions.

Trimethoxy Crown Chalcones as Multifunctional Class of Monoamine Oxidase Enzyme Inhibitors

BACKGROUND

Chalcones with methoxy substituents are considered as a promising framework for the inhibition of monoamine oxidase (MAO) enzymes.

METHODS

A series of nine trimethoxy substituted chalcones (TMa-TMi) was synthesized and evaluated as a multifunctional class of MAO inhibitors. All the synthesized compounds were investigated for their in vitro MAO inhibition, kinetics, reversibility, blood-brain barrier (BBB) permeation, and cytotoxicity and antioxidant potentials.

RESULTS

In the present study, compound (2E)-3-(4-nitrophenyl)-1-(3,4,5-trimethoxyphenyl)prop-2-en-1-one (TMf) was provided with an MAO-A inhibition constant value equal to 3.47±0.09 μM and with a selectivity of 0.008. Thus, it was comparable to that of moclobemide, a well known potent hMAO-A inhibitor (SI=0.010). Compound (2E)-3-(4-bromophenyl)-1-(3,4,5-trimethoxyphenyl)prop-2-en-1-one (TMh) showed good MAO-B inhibition, with an inhibition constant of 0.46±0.009 μM. The PAMPA assay demonstrated that all the synthesized derivatives can cross the BBB successfully. The cytotoxicity studies revealed that TMf and TMh have 88.22 and 80.18 % cell viability at 25 µM. Compound TMf appeared as the most promising antioxidant molecule with IC₅₀ values, relative to DPPH and H₂O₂ radical activities, equal to 6.02±0.17 and 7.25±0.07 μM. To shed light on the molecular interactions of TMf and TMh towards MAO-A and MAO-B, molecular docking simulations and MM/GBSA calculations have been carried out.

CONCLUSION

The lead molecules TMf and TMh with multi-functional nature can be further employed for the treatment of various neurodegenerative disorders and depressive states.

Homobivalent Lamellarin-Like Schiff Bases: In Vitro Evaluation of Their Cancer Cell Cytotoxicity and Multitargeting Anti-Alzheimer's Disease Potential

Marine alkaloids belonging to the lamellarins family, which incorporate a 5,6-dihydro-1-phenylpyrrolo[2,1-a]isoquinoline (DHPPIQ) moiety, possess various biological activities, spanning from antiviral and antibiotic activities to cytotoxicity against tumor cells and the reversal of multidrug resistance. Expanding a series of previously reported imino adducts of DHPPIQ 2-carbaldehyde, novel aliphatic and aromatic Schiff bases were synthesized and evaluated herein for their cytotoxicity in five diverse tumor cell lines. Most of the newly synthesized compounds were found noncytotoxic in the low micromolar range (<30 μM). Based on a Multi-fingerprint Similarity Search aLgorithm (MuSSeL), mainly conceived for making protein drug target prediction, some DHPPIQ derivatives, especially bis-DHPPIQ Schiff bases linked by a phenylene bridge, were prioritized as potential hits addressing Alzheimer's disease-related target proteins, such as cholinesterases (ChEs) and monoamine oxidases (MAOs). In agreement with MuSSeL predictions, homobivalent para-phenylene DHPPIQ Schiff base 14 exhibited a noncompetitive/mixed inhibition of human acetylcholinesterase (AChE) with K_i in the low micromolar range (4.69 μM). Interestingly, besides a certain inhibition of MAO A (50% inhibition of the cell population growth (IC₅₀) = 12 μM), the bis-DHPPIQ 14 showed a good inhibitory activity on self-induced β-amyloid (Aβ)_1-40 aggregation (IC₅₀ = 13 μM), which resulted 3.5-fold stronger than the respective mono-DHPPIQ Schiff base 9.

Design, synthesis and biological evaluation of imidazole and triazole-based carbamates as novel aromatase inhibitors

In the search for novel aromatase inhibitors, a series of triazole and imidazole-based carbamate derivatives were designed and synthesized. Final compounds were thus evaluated against human aromatase by in vitro kinetic experiments in a fluorimetric assay in comparison with letrozole. The effect of most active derivatives 13a and 15c was then evaluated in vitro on the human breast cancer cell line MCF7 by MTT assay, cytotoxicity assay (LDH release) and cell cycle analysis, revealing a dose-dependent inhibition profile of cell viability and low micromolar IC₅₀ values. In addition, docking simulations were also carried out to elucidate at a molecular level of detail the binding modes adopted to target human aromatase.

(Hetero-)(arylidene)arylhydrazides as Multitarget-Directed Monoamine Oxidase Inhibitors

Fourteen (hetero-)(arylidene)arylhydrazide derivatives (ABH1-ABH14) were synthesized, and their inhibitory activities against monoamine oxidases (MAOs) and acetylcholinesterase (AChE) were evaluated. Compound ABH5 most potently inhibited MAO-B with an IC₅₀ value of 0.025 ± 0.0019 μM; ABH2 and ABH3 exhibited high IC₅₀ values as well. Most of the compounds weakly inhibited MAO-A, except ABH5 (IC₅₀ = 3.31 ± 0.41 μM). Among the active compounds, ABH2 showed the highest selectivity index (SI) of 174 for MAO-B, followed by ABH5 (SI = 132). ABH3 and ABH5 effectively inhibited AChE with IC₅₀ values of 15.7 ± 6.52 and 16.5 ± 7.29 μM, respectively, whereas the other compounds were weak inhibitors of AChE. ABH5 was shown to be a reversible competitive inhibitor for MAO-A and MAO-B with K_i values of 0.96 ± 0.19 and 0.024 ± 0.0077 μM, respectively, suggesting that this molecule can be considered as an interesting candidate for further development as a multitarget inhibitor relating to neurodegenerative disorders.

Ethyl Acetohydroxamate Incorporated Chalcones: Unveiling a Novel Class of Chalcones for Multitarget Monoamine Oxidase-B Inhibitors Against Alzheimer's Disease

BACKGROUND

Chalcones are considered as the selective scaffold for the inhibition of MAO-B.

OBJECTIVES

A previously synthesized ethyl acetohydroxamate-chalcones (L1-L22) were studied for their inhibitory activities against human recombinant monoamine oxidase A and B (hMAO-A and hMAO-B, respectively) and acetylcholinesterase (AChE) as multi-target directed ligands for the treatment of Alzheimer's Disease (AD).

METHODS

Enzyme inhibition studies of MAO-A, MAO-B and AChE is carried out. Computational studies such as Molecular docking, Molecular Mechanics/Generalized Born Surface Area calculations, ADMET prediction, and protein target prediction are also performed.

RESULTS

Among the screened compounds, compound L3 has most potent hMAO-B inhibition with an IC50 value of 0.028 ± 0.0016 µM, and other compounds, L1, L2, L4, L8, L12, and L21 showed significant potent hMAO-B inhibition with IC50 values of 0.051 ± 0.0014, 0.086 ± 0.0035, 0.036 ± 0.0011, 0.096 ± 0.0061, 0.083 ± 0.0016, and 0.038 ± 0.0021 µM, respectively. On the other hand, among the tested compounds, compound L13 showed highest hMAO-A inhibition with an IC50 value of 0.51± 0.051 µM and L9 has a significant value of 1.85 ± 0.045 µM. However, the compounds L3 and L4 only showed high selectivities for hMAO-B with Selectivity Index (SI) values of 621.4 and 416.7, respectively. Among the substituents in ring A of ethyl acetohydroxamate-chalcones (L1-L9), F atom at p-position (L3) showed highest inhibitory effect against hMAO-B. This result supports the uniqness and bizarre behavior of fluorine. Moreover, chalcones L3, L4, L9, L11, and L12 showed potential AChE inhibitory effect with IC50 values of 0.67, 0.85, 0.39, 0.30, and 0.45 µM, respectively. Inhibitions of hMAO-B by L3 or L4 were recovered to the level of the reversible reference (lazabemide), and were competitive with Ki values of 0.0030 ± 0.0002 and 0.0046 ± 0.0005 µM, respectively. Inhibitions of AChE by L3 and L11 were of the competitive and mixed types with Ki values of 0.30 ± 0.044 and 0.14 ± 0.0054 µM, respectively.

CONCLUSION

The studies indicated that L3 and L4 are considered to be promising multitarget drug molecules with potent, selective, and reversible competitive inhibitors of hMAO-B and with highly potent AChE inhibitory effect.

De Novo Drug Design of Targeted Chemical Libraries Based on Artificial Intelligence and Pair-Based Multiobjective Optimization

Artificial intelligence and multiobjective optimization represent promising solutions to bridge chemical and biological landscapes by addressing the automated de novo design of compounds as a result of a humanlike creative process. In the present study, we conceived a novel pair-based multiobjective approach implemented in an adapted SMILES generative algorithm based on recurrent neural networks for the automated de novo design of new molecules whose overall features are optimized by finding the best trade-offs among relevant physicochemical properties (MW, logP, HBA, HBD) and additional similarity-based constraints biasing specific biological targets. In this respect, we carried out the de novo design of chemical libraries targeting neuraminidase, acetylcholinesterase, and the main protease of severe acute respiratory syndrome coronavirus 2. Several quality metrics were employed to assess drug-likeness, chemical feasibility, diversity content, and validity. Molecular docking was finally carried out to better evaluate the scoring and posing of the de novo generated molecules with respect to X-ray cognate ligands of the corresponding molecular counterparts. Our results indicate that artificial intelligence and multiobjective optimization allow us to capture the latent links joining chemical and biological aspects, thus providing easy-to-use options for customizable design strategies, which are especially effective for both lead generation and lead optimization. The algorithm is freely downloadable at https://github.com/alberdom88/moo-denovo and all of the data are available as Supporting Information.

Comparing a Query Compound with Drug Target Classes Using 3D-Chemical Similarity

3D similarity is useful in predicting the profiles of unprecedented molecular frameworks that are 2D dissimilar to known compounds. When comparing pairs of compounds, 3D similarity of the pairs depends on conformational sampling, the alignment method, the chosen descriptors, and the similarity coefficients. In addition to these four factors, 3D chemocentric target prediction of an unknown compound requires compound-target associations, which replace compound-to-compound comparisons with compound-to-target comparisons. In this study, quantitative comparison of query compounds to target classes (one-to-group) was achieved via two types of 3D similarity distributions for the respective target class with parameter optimization for the fitting models: (1) maximum likelihood (ML) estimation of queries, and (2) the Gaussian mixture model (GMM) of target classes. While Jaccard-Tanimoto similarity of query-to-ligand pairs with 3D structures (sampled multi-conformers) can be transformed into query distribution using ML estimation, the ligand pair similarity within each target class can be transformed into a representative distribution of a target class through GMM, which is hyperparameterized via the expectation-maximization (EM) algorithm. To quantify the discriminativeness of a query ligand against target classes, the Kullback-Leibler (K-L) divergence of each query was calculated and compared between targets. 3D similarity-based K-L divergence together with the probability and the feasibility index, (F_m), showed discriminative power with regard to some query-class associations. The K-L divergence of 3D similarity distributions can be an additional method for (1) the rank of the 3D similarity score or (2) the p-value of one 3D similarity distribution to predict the target of unprecedented drug scaffolds.

Virtual screening web servers: designing chemical probes and drug candidates in the cyberspace

The interplay between life sciences and advancing technology drives a continuous cycle of chemical data growth; these data are most often stored in open or partially open databases. In parallel, many different types of algorithms are being developed to manipulate these chemical objects and associated bioactivity data. Virtual screening methods are among the most popular computational approaches in pharmaceutical research. Today, user-friendly web-based tools are available to help scientists perform virtual screening experiments. This article provides an overview of internet resources enabling and supporting chemical biology and early drug discovery with a main emphasis on web servers dedicated to virtual ligand screening and small-molecule docking. This survey first introduces some key concepts and then presents recent and easily accessible virtual screening and related target-fishing tools as well as briefly discusses case studies enabled by some of these web services. Notwithstanding further improvements, already available web-based tools not only contribute to the design of bioactive molecules and assist drug repositioning but also help to generate new ideas and explore different hypotheses in a timely fashion while contributing to teaching in the field of drug development.

Structure-Based Identification and Design of Angiotensin Converting Enzyme-Inhibitory Peptides from Whey Proteins

Besides their nutritional value, whey protein (WP) peptides are food components retaining important pharmacological properties for controlling hypertension. We herein report how the use of complementary experimental and theoretical investigations allowed the identification of novel angiotensin converting enzyme inhibitory (ACEI) peptides obtained from a WP hydrolysate and addressed the rational design of even shorter sequences based on molecular pruning. Thus, after bromelain digestion followed by a 5 kDa cutoff ultrafiltration, WP hydrolysate with ACEI activity was fractioned by RP-HPLC; 2 out of 23 collected fractions retained ACEI activity and were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). In the face of 128 identified peptides, molecular docking was carried out to prioritize peptides and to rationally guide the design of novel shorter and bioactive sequences. Therefore, 11 peptides, consisting of 3-6 amino acids and with molecular weights in the range from 399 to 674 Da, were rationally designed and then purchased to determine the IC₅₀ value. This approach allowed the identification of two novel peptides: MHI and IAEK with IC₅₀ ACEI values equal to 11.59 and 25.08 μM, respectively. Interestingly, we also confirmed the well-known ACEI IPAVF with an IC₅₀ equal to 9.09 μM. In light of these results, this integrated approach could pave the way for high-throughput screening and identification of new peptides in dairy products. In addition, the herein proposed ACEI peptides could be exploited for novel applications both for food production and pharmaceuticals.

Design, synthesis and biological evaluation of oxygenated chalcones as potent and selective MAO-B inhibitors

The present study documents the synthesis of oxygenated chalcone (O1-O26) derivatives and their abilities to inhibit monoamine oxidases. All 26 derivatives examined showed potent inhibitory activity against MAO-B. Compound O23 showed the greatest inhibitory activity against MAO-B with an IC₅₀ value of 0.0021 µM, followed by compounds O10 and O17 (IC₅₀ = 0.0030 and 0.0034 µM, respectively). In addition, most of the derivatives potently inhibited MAO-A and O6 was the most potent inhibitor with an IC₅₀ value of 0.029 µM, followed by O3, O4, O9, and O2 (IC₅₀ = 0.035, 0.053, 0.072, and 0.082 µM, respectively). O23 had a high selectivity index (SI) value for MAO-B of 138.1, and O20 (IC₅₀ value for MAO-B = 0.010 µM) had an extremely high SI of >4000. In dialysis experiments, inhibitions of MAO-A and MAO-B by O6 and O23, respectively, were recovered to their respective reversible reference levels, demonstrating both are reversible inhibitors. Kinetic studies revealed that O6 and O23 competitively inhibited MAO-A and MAO-B, respectively, with respective K_i values of 0.016 ± 0.0007 and 0.00050 ± 0.00003 µM. Lead compound are also non-toxic at 200 µg/mL in normal rat spleen cells. Molecular docking simulations and subsequent Molecular Mechanics/Generalized Born Surface Area calculations provided a rationale that explained experimental data.