Drug-protein Interactions Research Articles

Relational graph convolutional networks for predicting blood-brain barrier penetration of drug molecules.

Evaluating the blood-brain barrier (BBB) permeability of drug molecules is a critical step in brain drug development. Traditional methods for the evaluation require complicated in vitro or in vivo testing. Alternatively, in silico predictions based on machine learning have proved to be a cost-efficient way to complement the in vitro and in vivo methods. However, the performance of the established models has been limited by their incapability of dealing with the interactions between drugs and proteins, which play an important role in the mechanism behind the BBB penetrating behaviors. To address this limitation, we employed the relational graph convolutional network (RGCN) to handle the drug-protein interactions as well as the properties of each individual drug. The RGCN model achieved an overall accuracy of 0.872, an area under the receiver operating characteristic (AUROC) of 0.919 and an area under the precision-recall curve (AUPRC) of 0.838 for the testing dataset with the drug-protein interactions and the Mordred descriptors as the input. Introducing drug-drug similarity to connect structurally similar drugs in the data graph further improved the testing results, giving an overall accuracy of 0.876, an AUROC of 0.926 and an AUPRC of 0.865. In particular, the RGCN model was found to greatly outperform the LightGBM base model when evaluated with the drugs whose BBB penetration was dependent on drug-protein interactions. Our model is expected to provide high-confidence predictions of BBB permeability for drug prioritization in the experimental screening of BBB-penetrating drugs. The data and the codes are freely available at https://github.com/dingyan20/BBB-Penetration-Prediction. Supplementary data are available at Bioinformatics online.

Bioinformatics and System Biology Approach to Identify the Influences of COVID-19 on Rheumatoid Arthritis.

BackgroundSevere coronavirus disease 2019 (COVID -19) has led to a rapid increase in mortality worldwide. Rheumatoid arthritis (RA) was a high-risk factor for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, whereas the molecular mechanisms underlying RA and CVOID-19 are not well understood. The objectives of this study were to analyze potential molecular mechanisms and identify potential drugs for the treatment of COVID-19 and RA using bioinformatics and a systems biology approach.MethodsTwo Differentially expressed genes (DEGs) sets extracted from GSE171110 and GSE1775544 datasets were intersected to generate common DEGs, which were used for functional enrichment, pathway analysis, and candidate drugs analysis.ResultsA total of 103 common DEGs were identified in the two datasets between RA and COVID-19. A protein-protein interaction (PPI) was constructed using various combinatorial statistical methods and bioinformatics tools. Subsequently, hub genes and essential modules were identified from the PPI network. In addition, we performed functional analysis and pathway analysis under ontological conditions and found that there was common association between RA and progression of COVID-19 infection. Finally, transcription factor-gene interactions, protein-drug interactions, and DEGs-miRNAs coregulatory networks with common DEGs were also identified in the datasets.ConclusionWe successfully identified the top 10 hub genes that could serve as novel targeted therapy for COVID-19 and screened out some potential drugs useful for COVID-19 patients with RA.

The TP53 Gene and COVID-19 Virus: A Correlation Analysis

Aim: This study aimed to discover the most effective anti-cancer medicine for cancer patients infected with SARS-CoV-2. Background: The correlation between TP53 and SARS-CoV-2 was examined using biomolecular networking analysis. Objective: Cancer patients with TP53 gene mutations are more likely to be infected with the SARSCoV- 2 virus since it is the most frequently mutated tumor suppressor gene in human cancer. The main goal of this study is to discover the most effective and efficient anti-cancer therapy for patients with SARS-CoV-2 infection. Materials and Methods: Topp gene analysis was used to prioritize candidate genes based on molecular function, biological process, and pathway analysis. Biomolecular networking was carried out using Cytoscape 2.8.2. The protein-protein interaction network was used to identify the functionally associated proteins. The protein-drug interaction network was used to observe the molecular therapeutic efficiency of drugs. The network was further analyzed using CytoHubba to find the hub nodes. The molecular docking was used to study the protein-ligand interaction, and the protein-ligand complex was further evaluated through molecular dynamic simulation to determine its stability. Results: Functionally relevant genes were prioritized through Toppgene analysis. Using Cytohabba, it was found that the genes UBE2N, BRCA1, BARD1, TP53, and DPP4 had a high degree and centrality score. The drugs 5-fluorouracil, Methotrexate, Temozolomide, Favipiravir, and Levofloxacin have a substantial association with the hub protein, according to protein-drug interaction analysis. Finally, a docking study revealed that 5-fluorouracil has the highest connection value and stability compared to Methotrexate, Favipiravir, and Levofloxacin. Conclusion: The biomolecular networking study was used to discover the link between TP53 and SARSCoV- 2, and it was found that 5-fluorouracil had a higher affinity for binding to TP53 and its related genes, such as UBE2N, BRCA1, RARD1, and SARS-CoV-2 specific DPP4. For cancer patients with TP53 gene mutations and Covid-19 infection, this treatment is determined to be the most effective.

Modulated Antimicrobial Activity and Drug-Protein Interaction Ability of Zinc Oxide and Cadmium Sulfide Nanoparticles: Effect of Doping with Few First-Row Transition Metals

Modulated Antimicrobial Activity and Drug-Protein Interaction Ability of Zinc Oxide and Cadmium Sulfide Nanoparticles: Effect of Doping with Few First-Row Transition Metals

Investigation of Pazopanib and Human Serum Albumin Interaction Using Spectroscopic and Molecular Docking Approaches

Pazopanib (PAZ), a tyrosine kinase inhibitor, is used to treat advanced renal cell carcinoma (RCC) and advanced soft tissue sarcoma (STS). The FDA approved PAZ for RCC in 2009 and for STS in 2012. The antitumor activity of pazopanib, according to the degree of inhibition, shows different results depending on the dose. Renal cell carcinoma is the most sensitive carcinoma to pazopanib, with 77% inhibition at the 10 mg/kg dose. Clinical studies have shown 53% to 65% inhibition in carcinomas such as breast carcinoma, prostate carcinoma, and melanoma. Plasma proteins such as human serum albumin (HSA) have a critical role in transporting and storing bioactive components. This feature of HSA is very important for the development of cancer therapy. Here, we investigated the interaction between PAZ and HSA to evaluate their binding strength, binding types, and conformational change in HSA. We used spectroscopic methods to assess the drug–protein interaction. Fluorescence measurements revealed that the interaction of PAZ with HSA occurred via the static quenching mechanism. The calculated binding number and binding constants were 1.041 and 1.436 × 106 M−1, respectively, at 298.15 K based on fluorescence screening. The high binding constant and calculated Gibbs free energy at different temperatures showed spontaneous and strong binding. Circular dichroism measurements showed that the α-helix structure of HSA was retained as the secondary structure, with a slight reduction in its percentage after adding PAZ. Furthermore, molecular modeling studies suggested that the docking score of PAZ is higher than those of bicalutamide and ibuprofen, the drugs that were chosen as model competitors against PAZ. Accordingly, PAZ was found to replace bicalutamide and ibuprofen on the HSA binding site, which was also confirmed by UV absorption spectroscopy.

BridgeDPI: a novel Graph Neural Network for predicting drug-protein interactions.

Exploring drug-protein interactions (DPIs) provides a rapid and precise approach to assist in laboratory experiments for discovering new drugs. Network-based methods usually utilize a drug-protein association network and predict DPIs by the information of its associated proteins or drugs, called 'guilt-by-association' principle. However, the 'guilt-by-association' principle is not always true because sometimes similar proteins cannot interact with similar drugs. Recently, learning-based methods learn molecule properties underlying DPIs by utilizing existing databases of characterized interactions but neglect the network-level information. We propose a novel method, namely BridgeDPI. We devise a class of virtual nodes to bridge the gap between drugs and proteins and construct a learnable drug-protein association network. The network is optimized based on the supervised signals from the downstream task-the DPI prediction. Through information passing on this drug-protein association network, a Graph Neural Network can capture the network-level information among diverse drugs and proteins. By combining the network-level information and the learning-based method, BridgeDPI achieves significant improvement in three real-world DPI datasets. Moreover, the case study further verifies the effectiveness and reliability of BridgeDPI. The source code of BridgeDPI can be accessed at https://github.com/SenseTime-Knowledge-Mining/BridgeDPI. The source data used in this study is available on the https://github.com/IBM/InterpretableDTIP (for the BindingDB dataset), https://github.com/masashitsubaki/CPI_prediction (for the C.ELEGANS and HUMAN) datasets, http://dude.docking.org/ (for the DUD-E dataset), repectively.

Glycyrrhizin (Glycyrrhizic Acid) HMGB1 (high mobility group box 1) inhibitor upregulate mitochondrial function in adipocyte, cell viability and in-silico study

BackgroundMitochondrial plays a vital role in regulating obesity and related comorbidity. Targeting mitochondrial function could be a potent therapeutic approach to inhibit metabolic-related diseases like obesity, liver disease. Prolonged use of existing drug moieties demonstrated severe adverse effects. MethodsWe apply Ucp1-A-GFP immortalized reporter cell lines and HEK293T cell lines to evaluate cell viability, mitochondrial ATP production, and the in-silico model. ResultsWe found Glycyrrhizin, an HMGB1 (high mobility group box 1) inhibitor, plays a significant role in modulating mitochondrial function against obesity. At the cellular level, the adipocytes treated with Glycyrrhizin have increased mitochondrial function. Further analysis shows that compared with the control group, the cells in the treatment group contain more mitochondria. Glycyrrhizin demonstrated a nontoxic effect on the HEK293T cell line, upregulating mitochondrial DNA and reducing mitochondrial ATP production levels. In-silico study exhibited drug-protein interaction and binding side with UCP1. ConclusionGlycyrrhizin improves mitochondrial function that would be an effective drug candidate to treat metabolic diseases and obesity-related diseases. Further investigation will require both the human and animal models to reveal new insight into the mechanism against obesity, metabolic diseases or mitochondrial dysfunction-related diseases.

Retraction of "Predictions of Drug-Protein Interactions and Study of Magnetically Assisted Release Dynamics of 5-Fluorouracil from Soya Protein-Coated Iron Oxide Core-Shell Nanoparticles".

Retraction of "Predictions of Drug-Protein Interactions and Study of Magnetically Assisted Release Dynamics of 5-Fluorouracil from Soya Protein-Coated Iron Oxide Core-Shell Nanoparticles".

Elucidation and Prediction of Molecular Mechanisms and Associated Pathways of the Prostaglandin-endoperoxide Synthase-2 as a Common Key Gene for Diabetes and Cachexia: Bioinformatics Strategies

Cachexia and diabetes both feature severe weight loss as a phenotype, and both diseases share several gene and protein abnormalities. This study investigates the PTGS2 gene as a common key gene for diabetes and cachexia. A variety of biological databases and enrichment approaches were used to explore this gene. Probable genes and substances that could be used to address disease symptoms were identified using STICH, a protein-protein, and protein-drug interactions. It was also established that PTGS2 is strongly linked to the pharmacology of the ibuprofen route of action. These findings and annotations will aid in further research into the therapeutic option of targeting the PTGS2 gene in the treatment of both disorders.

Structure-Aware Multimodal Deep Learning for Drug–Protein Interaction Prediction

Identifying drug-protein interactions (DPIs) is crucial in drug discovery, and a number of machine learning methods have been developed to predict DPIs. Existing methods usually use unrealistic data sets with hidden bias, which will limit the accuracy of virtual screening methods. Meanwhile, most DPI prediction methods pay more attention to molecular representation but lack effective research on protein representation and high-level associations between different instances. To this end, we present the novel structure-aware multimodal deep DPI prediction model, STAMP-DPI, which was trained on a curated industry-scale benchmark data set. We built a high-quality benchmark data set named GalaxyDB for DPI prediction. This industry-scale data set along with an unbiased training procedure resulted in a more robust benchmark study. For informative protein representation, we constructed a structure-aware graph neural network method from the protein sequence by combining predicted contact maps and graph neural networks. Through further integration of structure-based representation and high-level pretrained embeddings for molecules and proteins, our model effectively captures the feature representation of the interactions between them. As a result, STAMP-DPI outperformed state-of-the-art DPI prediction methods by decreasing 7.00% mean square error (MSE) in the Davis data set and improving 8.89% area under the curve (AUC) in the GalaxyDB data set. Moreover, our model is an interpretable model with the transformer-based interaction mechanism, which can accurately reveal the binding sites between molecules and proteins.

In Vitro and In Vivo Pharmacological Characterization of a Novel TRPM8 Inhibitor Chemotype Identified by Small-Scale Preclinical Screening.

Transient receptor potential melastatin type 8 (TRPM8) is a target for the treatment of different physio-pathological processes. While TRPM8 antagonists are reported as potential drugs for pain, cancer, and inflammation, to date only a limited number of chemotypes have been investigated and thus a limited number of compounds have reached clinical trials. Hence there is high value in searching for new TRPM8 antagonistic to broaden clues to structure-activity relationships, improve pharmacological properties and explore underlying molecular mechanisms. To address this, the EDASA Scientific in-house molecular library has been screened in silico, leading to identifying twenty-one potentially antagonist compounds of TRPM8. Calcium fluorometric assays were used to validate the in-silico hypothesis and assess compound selectivity. Four compounds were identified as selective TRPM8 antagonists, of which two were dual-acting TRPM8/TRPV1 modulators. The most potent TRPM8 antagonists (BB 0322703 and BB 0322720) underwent molecular modelling studies to highlight key structural features responsible for drug–protein interaction. The two compounds were also investigated by patch-clamp assays, confirming low micromolar potencies. The most potent compound (BB 0322703, IC50 1.25 ± 0.26 μM) was then profiled in vivo in a cold allodinya model, showing pharmacological efficacy at 30 μM dose. The new chemotypes identified showed remarkable pharmacological properties paving the way to further investigations for drug discovery and pharmacological purposes.

Predicting the Activities of Drug Excipients on Biological Targets using One-Shot Learning.

Excipients are major components of drugs and are used to improve drug attributes such as stability and appearance. Excipients approved by the U.S. Food and Drug Administration (FDA) are regarded as safe for humans in allowed concentrations, but their potential interactions with drug targets have not been investigated systematically, which might influence a drug's efficacy. Deep learning models have been used for the identification of ligands that could bind to the drug targets. However, due to the limited available data, it is challenging to reliably estimate the likelihood of a ligand-protein interaction. One-shot learning techniques provide a potential approach to address this low data problem as these techniques require only one or a few examples to classify the new data. In this study, we apply one-shot learning models to data sets that include ligands binding to G-protein-coupled receptors (GPCRs) and kinases. The predicted results suggest that one-shot learning could be used for predicting ligand-protein interactions, and the models attain better performance when protein targets contain conserved binding pockets. The trained models are also used to predict interactions between excipients and drug targets, which provides a potential efficient strategy to explore the activities of drug excipients. We find that a large number of drug excipients could interact with biological targets and influence their function. The results demonstrate how one-shot learning can be used to make accurate predictions for excipient-protein interactions, and these methods could be used for selecting excipients with limited drug-protein interactions.

In vitro and in silico interactions of antiulcer, glucocorticoids and urological drugs on human carbonic anhydrase I and II isozymes.

Carbonic anhydrases (CAs, Enzyme Commission 4.2.1.1) convert carbon dioxide to bicarbonate in metabolism and use Zn2+ ions as a cofactor for their catalytic activity. The activators or inhibitors of CA-I and CA-II, which are the most abundant CA isozymes in erythrocytes, have pharmacological applications in medicine. So, investigation of drug-protein interaction of these isozymes is significant. On this basis, the objective of this study was to clarify the primer effects of widely used drugs on the activity of human CA-I and CA-II enzymes and elucidate the inhibition mechanism through molecular docking studies. For this aim isozymes were purified from human erythrocytes by affinity chromatography technique. Then inhibition profiles of antiulcer, glucocorticoids, and urological drugs were investigated. As a result, while budesonide had the highest inhibitory potency on hydratase activity of hCA-I with the IC50 of 0.08mM, levofloxacin showed the highest inhibition effect on hCA-II with the IC50 of 0.886mM. The most effective inhibitor on the esterase activity of isozymes was found as fluticasone propionate with the Ki values of 0.0365±0.016mM and 0.054±0.018mM respectively. However, by molecular docking study, it was estimated that budesonide showed maximum inhibition potency for both isozymes with the free binding energy of -7.58 and -6.97kcal/mol, respectively. Consequently, it was observed that some of the drugs studied did not show any inhibitory effect. Drug-enzyme interactions were also estimated by molecular docking. This study could contribute to the discovery of new drug candidates and as well as target proteins.

ALDPI: adaptively learning importance of multi-scale topologies and multi-modality similarities for drug-protein interaction prediction.

Effective computational methods to predict drug-protein interactions (DPIs) are vital for drug discovery in reducing the time and cost of drug development. Recent DPI prediction methods mainly exploit graph data composed of multiple kinds of connections among drugs and proteins. Each node in the graph usually has topological structures with multiple scales formed by its first-order neighbors and multi-order neighbors. However, most of the previous methods do not consider the topological structures of multi-order neighbors. In addition, deep integration of the multi-modality similarities of drugs and proteins is also a challenging task. We propose a model called ALDPI to adaptively learn the multi-scale topologies and multi-modality similarities with various significance levels. We first construct a drug-protein heterogeneous graph, which is composed of the interactions and the similarities with multiple modalities among drugs and proteins. An adaptive graph learning module is then designed to learn important kinds of connections in heterogeneous graph and generate new topology graphs. A module based on graph convolutional autoencoders is established to learn multiple representations, which imply the node attributes and multiple-scale topologies composed of one-order and multi-order neighbors, respectively. We also design an attention mechanism at neighbor topology level to distinguish the importance of these representations. Finally, since each similarity modality has its specific features, we construct a multi-layer convolutional neural network-based module to learn and fuse multi-modality features to obtain the attribute representation of each drug-protein node pair. Comprehensive experimental results show ALDPI's superior performance over six state-of-the-art methods. The results of recall rates of top-ranked candidates and case studies on five drugs further demonstrate the ability of ALDPI to discover potential drug-related protein candidates. zhang@hlju.edu.cn.

Large-scale Prediction of Drug-Protein Interactions Based on Network Information.

The prediction of drug-protein interaction (DPI) plays an important role in drug discovery and repositioning. Unfortunately, traditional experimental validation of DPIs is expensive and time-consuming. Therefore, it is necessary to develop in silico methods for the identification of potential DPIs. In this work, the identification of DPIs was performed by the generated recommendation of the unexplored interaction of the drug-protein bipartite graph. Three kinds of recommenders were proposed to predict the potential DPIs. The simulation results showed that the proposed models obtained good performance in crossvalidation and independent test. Our recommendation strategy based on collaborative filtering can effectively improve the DPI identification performance, especially for certain DPIs lacking chemical structure similarity or genomic sequence similarity.

Detection of molecular signatures and pathways shared by Alzheimer's disease and type 2 diabetes.

Alzheimer's disease (AD) and type 2 diabetes (T2D) are common in the general elderly population, conferring heavy individual, social, and economic stresses on families and society. Accumulating evidence indicates T2D to be a risk factor for AD. However, the underlying mechanisms for this association are largely unknown. This study aimed to identify the shared molecular signatures between AD and T2D through integrated analysis of temporal cortex gene expression data. Gene Ontology (GO) and pathway enrichment analysis, protein over-representation analysis, protein-protein interaction, DEG-transcription factor interactions, DEG-microRNA interactions, protein-drug interactions, gene-disease association analysis, and protein subcellular localization analysis of the common DEGs were performed. We identified 16 common DEGs between the two datasets, which were mainly enriched in the biological processes of apoptosis, autophagy, inflammation, and hemostasis. We also identified five hub proteins encoded by the DEGs, five central regulatory transcription factors, and six microRNAs. Protein-drug interaction analysis showed C1QB to be associated with different drugs. Gene-disease association analysis revealed that hub genes, SFN and ITGB2, were actively engaged in other diseases. Collectively, these findings provide new insights into shared molecular mechanisms between AD and T2D and provide novel candidate targets for therapeutic intervention.

An optical microfluidic calorimeter for high-throughput drug screening

High-throughput screening (HTS) is crucial to the drug discovery process enabling rapid development of drug leads and reducing costs associated with optimization of drug activity. Calorimetry is a label-free, biophysical technique used for determination of important parameters associated with drug-protein interactions, such as binding affinity and thermodynamics. However, conventional isothermal titration calorimetry (ITC) is limited by low throughput, high sample consumption (milliliters), and long measurement time (tens of minutes).

Spectroscopic and in silico approach to probe the binding interactions of irbesartan and human serum albumin

ObjectiveThe free and active concentration of drugs and thereby their pharmacokinetic properties are controlled by their binding to human serum albumin (HSA) protein. Irbesartan (IRB), an antihypertensive drug was aimed to be investigated in terms of its binding interactions with different sites of HSA using in silico molecular docking technique along with the commonly employed spectroscopic techniques. MethodsUsing FT-IR spectroscopy, the spectral shifting and intensity variations before and after complexation with IRB were studied for amide A, amide-I as well as amide-II of HSA. The absorbance of HSA with and without increasing concentrations of IRB was studied at 280 nm and the binding constant was determined using UV-spectroscopy. Molecular docking study was performed, and the types of interactions were predicted. ResultsThe IR spectra of IRB-HSA complex showed reductions in the intensities of amide-I and II bands as well as marked reduction in the α-helix content of HSA. The absorbance of HSA protein increased with increasing concentrations of drug. A binding constant value of 5.64 × 104 M−1 was calculated indicating good interaction. Molecular docking studies showed that IRB interacts more effectively with site-I of HSA through greater number of hydrogen bonds and strong π–charge (electrostatic) interactions than with site-II. ConclusionsThe spectroscopic and molecular docking techniques proved to be effective tools to study the drug-protein interaction which provided accurate results as evident from these studies. Studying drug-albumin interaction is of utmost importance as it directly influences the overall pharmacokinetics of the drugs including its distribution, metabolism and therefore the duration of action.

Detection of polypharmacy side effects by integrating multiple data sources and convolutional neural networks.

The consumption of drug combinations, named polypharmacy, is commonly used for treating patients with several diseases or those with complex conditions. However, the main drawback of polypharmacy is the increased probability of harmful side effects. The polypharmacy side effects are caused by an interaction between two medications. It means that the drug-drug interaction causes changes in their activities due to interfering in each other's performance. Therefore, discovering these side effects is one of the most challenging and important aspects of drug production and consumption as it is associated with human health. In this paper, a method has been introduced for predicting the polypharmacy side effects, called PSECNN. It is a multi-label multi-class deep learning method that combines various basic features of drugs to predict the polypharmacy side effects. Firstly, PSECNN collects five basic features of drugs, such as individual drug's side effects, drug-protein interactions, chemical substructures, targets, and enzymes in order to create a novel combination of drug features. A feature extraction module creates five feature vectors with the same dimension for each drug based on the Jaccard similarity index. Based on the feature vectors, a unique representative is then created for each drug. These representative vectors are given in pairs as input to the deep neural network to predict the occurrence probability of side effects. According to the experimental evaluations, PSECNN could outperform the state-of-the-art polypharmacy side effects prediction methods up to 74%. It has been found that PSECNN has better performance with polypharmacy side effects with a cause of molecular basis due to the novel combination of basic drug features.

Site-selective covalently immobilized alpha 1A adrenergic receptor for thermodynamic and extra-thermodynamic study of four ligands binding to the receptor by chromatographic methods

Immobilized G protein-coupled receptor is a versatile tool to study ligand-receptor interactions. In this work, we synthesized the immobilized alpha 1A adrenergic receptor (α1A-AR), a GPCR subtype mediating smooth muscle contraction, through a site-selective covalent method that relies on the reaction between haloalkane dehalogenase tagged α1A-AR and macroporous silica gel coated with 6-chlorohexanoic acid. To investigate thermodynamic and extra-thermodynamic parameters for ligand binding, we utilized the covalently immobilized receptor as stationary phase to perform frontal analysis and injection-amount dependent analysis as well as compared with the random immobilization method. Terazosin gave the association constant of 1.48 × 105 M−1 to α1A-AR, indicating that the oriented immobilization of α1A-AR enhances the ligand-binding activity by one order of magnitude in comparison with the random immobilization method (7.9 × 104 M−1). The binding of phentolamine and tamsulosin to the receptor was accompanied by a large absolute heat capacity (ΔCp) of 1.28 ± 0.23 kJ mol−1, demonstrating that the binding enthalpy and entropy appear to compensate for one another. These results indicated that the covalent immobilization of the receptor onto solid support has a profound impact on the ligand-binding activity of the receptor and the determination of ligand-receptor binding parameters. The receptor immobilized through the site-selective method will act as a benchmark for chromatographic determination of binding parameters in ligand-receptor interactions and can be used as an effective approach for rapid analysis of drug-protein interactions with high accuracy.