In Silico Identification of Anti-SARS-CoV-2 Medicinal Plants Using Cheminformatics and Machine Learning

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative pathogen of COVID-19, is spreading rapidly and has caused hundreds of millions of infections and millions of deaths worldwide. Due to the lack of specific vaccines and effective treatments for COVID-19, there is an urgent need to identify effective drugs. Traditional Chinese medicine (TCM) is a valuable resource for identifying novel anti-SARS-CoV-2 drugs based on the important contribution of TCM and its potential benefits in COVID-19 treatment. Herein, we aimed to discover novel anti-SARS-CoV-2 compounds and medicinal plants from TCM by establishing a prediction method of anti-SARS-CoV-2 activity using machine learning methods. We first constructed a benchmark dataset from anti-SARS-CoV-2 bioactivity data collected from the ChEMBL database. Then, we established random forest (RF) and support vector machine (SVM) models that both achieved satisfactory predictive performance with AUC values of 0.90. By using this method, a total of 1011 active anti-SARS-CoV-2 compounds were predicted from the TCMSP database. Among these compounds, six compounds with highly potent activity were confirmed in the anti-SARS-CoV-2 experiments. The molecular fingerprint similarity analysis revealed that only 24 of the 1011 compounds have high similarity to the FDA-approved antiviral drugs, indicating that most of the compounds were structurally novel. Based on the predicted anti-SARS-CoV-2 compounds, we identified 74 anti-SARS-CoV-2 medicinal plants through enrichment analysis. The 74 plants are widely distributed in 68 genera and 43 families, 14 of which belong to antipyretic detoxicate plants. In summary, this study provided several medicinal plants with potential anti-SARS-CoV-2 activity, which offer an attractive starting point and a broader scope to mine for potentially novel anti-SARS-CoV-2 drugs.     

Predicting the Mechanical Properties of Polyurethane Elastomers Using Machine Learning

Bridging the gap between the computation of mechanical properties and the chemical structure of elastomers is a long-standing challenge. To fill the gap, we create a raw dataset and build predictive models for Young’s modulus, tensile strength, and elongation at break of polyurethane elastomers(PUEs). We then construct a benchmark dataset with 50.4% samples remained from the raw dataset which suffers from the intrinsic diversity problem, through a newly proposed recursive data elimination protoc...     

In Silico Evaluation and In Vitro Determination of Neuroprotective and MAO-B Inhibitory Effects of Pyrrole-Based Hydrazones: A Therapeutic Approach to Parkinson’s Disease

Parkinson’s disease is a huge burden in modern medicinal practice. A serious drawback of current antiparkinsonian therapy is its symptomatic nature. This directed our investigations in the search for new more potent derivatives, affecting not only the loss of dopaminergic neurons but also the oxidative damage of neuronal cells. Thus in vitro neurotoxicity and neuroprotective analysis on a group of N-pyrrolyl hydrazide–hydrazones were performed. The neurotoxicity of the target derivatives was determined on a subcellular level in isolated rat synaptosomes, mitochondria and microsomes determining their effect on cellular vitality, GSH depletion and MDA production. The neuroprotective effects of the evaluated hydrazones were measured in three models of induced oxidative stress: 6-OHDA, t-BuOOH and Fe²⁺/AA-induced lipid peroxidation. Molecular docking simulations along with in vitro evaluation of MAO-B inhibitory potential of the target molecules were also performed. The results identified the ethyl 5-(4-bromophenyl)-1-(3-hydrazinyl-3-oxopropyl)-2-methyl-1H-pyrrole-3-carboxylate (12) as the most promising compound with the lowest neurotoxicity and highest neuroprotection on all evaluated parameters and inhibiting the hMAOB enzyme by 50%, comparable with the activity of the reference, Selegiline. The compatibility of the in silico and in vitro evaluations is a good prerequisite for these methods to be applied in future assessment of pyrrole-based compounds as anti-Parkinson agents.     

Identification and Evaluation of Traditional Chinese Medicine Natural Compounds as Potential Myostatin Inhibitors: An In Silico Approach

Myostatin (MSTN), a negative regulator of muscle mass, is reported to be increased in conditions linked with muscle atrophy, sarcopenia, and other muscle-related diseases. Most pharmacologic approaches that treat muscle disorders are ineffective, emphasizing the emergence of MSTN inhibition. In this study, we used computational screening to uncover natural small bioactive inhibitors from the Traditional Chinese Medicine database (~38,000 compounds) for the MSTN protein. Potential ligands were screened, based on binding affinity (150), physicochemical (53) and ADMET properties (17). We found two hits (ZINC85592908 and ZINC85511481) with high binding affinity and specificity, and their binding patterns with MSTN protein. In addition, molecular dynamic simulations were run on each complex to better understand the interaction mechanism of MSTN with the control (curcumin) and the hit compounds (ZINC85592908 and ZINC85511481). We determined that the hits bind to the active pocket site (Helix region) and trigger conformational changes in the MSTN protein. Since the stability of the ZINC85592908 compound was greater than the MSTN control, we believe that ZINC85592908 has therapeutic potential against the MSTN protein and may hinder downstream singling by inhibiting the MSTN protein and increasing myogenesis in the skeletal muscle tissues.     

Synthetic Activators of Cell Migration Designed by Constructive Machine Learning

Constructive machine learning aims to create examples from its learned domain which are likely to exhibit similar properties. Here, a recurrent neural network was trained with the chemical structures of known cell-migration modulators. This machine learning model was used to generate new molecules that mimic the training compounds. Two top-scoring designs were synthesized, and tested for functional activity in a phenotypic spheroid cell migration assay. These computationally generated small molecules significantly increased the migration of medulloblastoma cells. The results further corroborate the applicability of constructive machine learning to the de novo design of druglike molecules with desired properties.     

Targeting Cytotoxin-Associated Antigen A,a Virulent Factor of Helicobacter pylori-Associated Gastric Cancer: Structure-Based In Silico Screening of Natural Compounds

Gastric cancer is the fifth most frequent cancer and the third major cause of mortality worldwide. Helicobacter pylori, a bacterial infection linked with GC, injects the cytotoxin-associated antigen A (CagA; an oncoprotein) into host cells. When the phosphorylated CagA protein enters the cell, it attaches to other cellular components, interfering with normal cellular signaling pathways. CagA plays an important role in the progression of GC by interacting with phosphatidylserine of the host cell membrane. Therefore, disrupting the CagA–phosphatidylserine connection using small molecules appears to be a promising therapeutic approach. In this report, we screened the natural compounds from ZINC database against the CagA protein using the bioinformatics tools. Hits were initially chosen based on their physicochemical, absorption, distribution, metabolism, excretion, and toxicity (ADMET) characteristics, as well as other drug-like characteristics. To locate safe and effective hits, the PAINS filter, binding affinities estimation, and interaction analysis were used. Three compounds with high binding affinity and specificity for the CagA binding pocket were discovered. The final hits, ZINC153731, ZINC69482055, and ZINC164387, were found to bind strongly with CagA protein, with binding energies of −11.53, −10.67, and −9.21 kcal/mol, respectively, which were higher than that of the control compound (−7.25 kcal/mol). Further, based on binding affinity and interaction pattern, two leads (ZINC153731, ZINC69482055) were chosen for molecular dynamics (MD) simulation analysis. MD results showed that they displayed stability in their vicinity at 100 ns. This study suggested that these compounds could be used as possible inhibitors of CagA protein in the fight against GC. However, additional benchwork tests are required to validate them as CagA protein inhibitors.     

Spectroscopic Properties of Organic Azo Compounds

Azo compounds are characterized by a low-energy ¹(n,π^*) electronic state, which gives rise to a weak band at long wavelengths in the absorption spectrum. Coupling of this state with other low-lying molecular states is responsible for the high intensity of the “azo band” compared to other forbidden transitions, for its lack of vibrational structure, and for the inability of nearly all organic azo compounds to fluoresce.     

预测金属有机骨架甲烷和氢气输送能力的迁移学习建模

基于深度神经网络(DNN)和迁移学习(TL),使用少量数据建立TL模型,精准预测了金属有机骨架(MOFs)的甲烷和氢气输送性能.首先,使用8414个MOFs在298 K/65 bar～298 K/5.8 bar(1 bar=0.1 MPa)条件下的甲烷输送数据训练一个决定系数(R²)为0.973的DNN[源任务(ST)模型].随后,将ST模型的部分参数冻结,使用100个MOFs在233 K/65 bar～358 K/5.8 bar条件下的甲烷输送数据和100个MOFs在198 K/100 bar～298 K/5 bar条件下的氢气输送数据分别微调ST模型,进行TL建模.结果表明,两个TL模型的R²分别为0.968和0.945,均高于其它5个传统的ML模型.所开发的TL模型在预测小数据集时具有高精度与高稳定性.最后,使用排列特征重要度方法来计算描述符重要度,明确了模型之间的“知识”共享情况,并在此基础上探讨了重要描述符和输送能力之间的关系.     

In Silico and In Vitro Analysis of Major Cannabis-Derived Compounds as Fatty Acid Amide Hydrolase Inhibitors

Accumulated evidence suggests that enhancing the endocannabinoid (eCB) tone, in particular of anandamide (N-arachidonoylethanolamine, AEA), has therapeutic potential in many human diseases. Fatty acid amide hydrolase (FAAH) is a membrane-bound enzyme principally responsible for the degradation of AEA, and thus it represents a relevant target to increase signaling thereof. In recent years, different synthetic and natural compounds have been developed and tested on rat FAAH, but little is known of their effect on the human enzyme. Here, we sought to investigate six major cannabis-derived compounds to compare their action on rat and human FAAHs. To this aim, we combined an in silico analysis of their binding mode and affinity, with in vitro assays of their effect on enzyme activity. This integrated approach allowed to disclose differences in efficacy towards rat and human FAAHs, and to highlight the role of key residues involved in the inhibition of both enzymes. This study suggests that the therapeutic efficacy of compounds targeted towards FAAH should be always tested in vitro on both rat and human enzymes.     

Phytochemical Compounds and Pharmacological Properties of Larrea tridentata

For centuries, traditional medicine from plants (phytotherapy) was the only treatment for infectious and non-infectious diseases. Although it is still practiced in several countries with excellent therapeutic results, it is frequently underestimated because, unlike Western medicine, it is not based on an empirical scientific foundation. However, interest in the search for plant-based therapeutic resources has been stimulated by disciplines such as phytochemistry and the side effects of conventional pharmacological therapies. For example, Larrea tridentata is a perennial shrub used in traditional medicine in northern Mexico and the southern United States to treat infertility, rheumatism, arthritis, colds, diarrhea, skin problems, pain, inflammation and excess body weight. Scientific research has revealed its beneficial effects—antioxidant, antitumor, neuroprotective, regenerative, antibacterial, antiviral, antifungal, anthelmintic, antiprotozoal and insecticidal—although reports indicate that some compounds in Larrea tridentata may be hepatotoxic and nephrotoxic. Therefore, the aim of this review was to highlight the updates regarding phytochemical compounds and the pharmacological properties of Larrea tridentata.     

Modeling Interfacial Tension of n-Alkane/Water-Salt System Using Artificial Neural Networks

In this paper, a feed forward neural network is built and trained using experimental data reported in the literature to model interfacial tension of n-alkane/water-salt systems. Temperature, pressure, molecular weight of n-alkane, and ionic strength of electrolyte solution are used as input to the neural network. The model succeeded to predict interfacial tension of liquid n-alkane/water system with or without the presence of electrolyte and yielded average absolute deviation of 0.58% over all data points. The performance of the model is analyzed and compared against the performance of the other alternative models. It was found out that the proposed model outperforms the other alternatives.     

Phenolic Characterization and Neuroprotective Properties of Grape Pomace Extracts

Vitis vinifera (grape) contains various compounds with acknowledged phytochemical and pharmacological properties. Among the different parts of the plant, pomace is of particular interest as a winemaking industry by-product. A characterization of the water extract from grape pomace from Montepulciano d’Abruzzo variety (Villamagna doc) was conducted, and the bioactive phenolic compounds were quantified through HPLC-DAD-MS analysis. HypoE22, a hypothalamic cell line, was challenged with an oxidative stimulus and exposed to different concentrations (1 µg/mL⁻¹ mg/mL) of the pomace extract for 24, 48, and 72 h. In the same conditions, cells were exposed to the sole catechin, in a concentration range (5–500 ng/mL) consistent with the catechin level in the extract. Cell proliferation was investigated by MTT assay, dopamine release through HPLC-EC method, PGE₂ amount by an ELISA kit, and expressions of neurotrophin brain-derived neurotrophic factor (BDNF) and of cyclooxygenase-2 (COX-2) by RT-PCR. The extract reverted the cytotoxicity exerted by the oxidative stimulus at all the experimental times in a dose-dependent manner, whereas the catechin was able to revert the oxidative stress-induced depletion of dopamine 48 h and 72 h after the stimulus. The extract and the catechin were also effective in preventing the downregulation of BDNF and the concomitant upregulation of COX-2 gene expression. In accordance, PGE₂ release was augmented by the oxidative stress conditions and reverted by the administration of the water extract from grace pomace and catechin, which were equally effective. These results suggest that the neuroprotection induced by the extract could be ascribed, albeit partially, to its catechin content.     

Insight into Gentisic Acid Antidiabetic Potential Using In Vitro and In Silico Approaches

Numerous scientific studies have confirmed the beneficial therapeutic effects of phenolic acids. Among them gentisic acid (GA), a phenolic acid extensively found in many fruit and vegetables has been associated with an enormous confirmed health benefit. The present study aims to evaluate the antidiabetic potential of gentisic acid and highlight its mechanisms of action following in silico and in vitro approaches. The in silico study was intended to predict the interaction of GA with eight different receptors highly involved in the management and complications of diabetes (dipeptidyl-peptidase 4 (DPP4), protein tyrosine phosphatase 1B (PTP1B), free fatty acid receptor 1 (FFAR1), aldose reductase (AldR), glycogen phosphorylase (GP), α-amylase, peroxisome proliferator-activated receptor gamma (PPAR-γ) and α-glucosidase), while the in vitro study studied the potential inhibitory effect of GA against α-amylase and α-glucosidase. The results indicate that GA interacted moderately with most of the receptors and had a moderate inhibitory activity during the in vitro tests. The study therefore encourages further in vivo studies to confirm the given results.     

Medicinal Properties of Anchusa strigosa and Its Active Compounds

Anchusa strigosa is a widespread weed in Greece, Syria, Turkey, Lebanon, Israel, Jordan, and Iran. The purpose of this study was to identify the phytochemicals of Anchusa strigose and estimate the pro-wound healing (pro-WH) and antimicrobial activities of its active compounds. An identification of volatile compounds was performed by GC/MS analysis; HPLC, LC-ESI-MS, and MALDI-TOF-MS were also applied. Our results demonstrate that two specific combinations of compounds from A. strigosa extract significantly enhanced WH (p < 0.001). Several flavonoids of the plant extract, including quercetin 3-O-rutinoside, kaempferol, kaempferol 3-O-β-rhamnopyranosyl(1→6)-β-glucopyranoside, and kaempferol 3-O-α-rhamnopyranosyl(1→6)-β-galactopyranoside, were effective against drug-resistant microorganisms. In addition, all the above-mentioned compounds had antibiofilm activity against Escherichia coli and Salmonella enteritidis.     

Molecular Modeling of Bioorganometallic Compounds: Thermodynamic Properties of Molybdocene–Glutathione Complexes and Mechanism of Peptide Hydrolysis

The computational study of bioinorganic complexes between transition metals and flexible ligands is still challenging, given that, besides requiring extensive conformational searches, the treatment of metal–ligand bonds demands the application of quantum chemical methods. Herein, the adducts formed between molybdocene, which exhibits antitumor activity and reacts with thiol groups to give stable water‐soluble complexes, and the tripeptide glutathione, which is a major source of biological thiols, are studied. Conformational searches are performed using the semiempirical PM6 method followed by geometry optimizations and single‐point calculations using density functional theory methods. In addition, molecular dynamics simulations of the molybdocene–glutathione complex involved in the regioselective hydrolysis of the Cys–Gly linkage are performed in explicit solvent. The reactive process is also studied theoretically on cluster models of both the molybdocene‐bound and the free peptide.