https://orcid.org/0000-0003-1621-7178
Figures
Abstract
Drug repurposing has emerged as an important strategy and it has a great potential in identifying therapeutic applications for COVID-19. An extensive virtual screening of 4193 FDA approved drugs has been carried out against 24 proteins of SARS-CoV2 (NSP1-10 and NSP12-16, envelope, membrane, nucleoprotein, spike, ORF3a, ORF6, ORF7a, ORF8, and ORF9b). The drugs were classified into top 10 and bottom 10 drugs based on the docking scores followed by the distribution of their therapeutic indications. As a result, the top 10 drugs were found to have therapeutic indications for cancer, pain, neurological disorders, and viral and bacterial diseases. As drug resistance is one of the major challenges in antiviral drug discovery, polypharmacology and network pharmacology approaches were employed in the study to identify drugs interacting with multiple targets and drugs such as dihydroergotamine, ergotamine, bisdequalinium chloride, midostaurin, temoporfin, tirilazad, and venetoclax were identified among the multi-targeting drugs. Further, a pathway analysis of the genes related to the multi-targeting drugs was carried which provides insight into the mechanism of drugs and identifying targetable genes and biological pathways involved in SARS-CoV2.
Citation: Jamir E, Sarma H, Priyadarsinee L, Kiewhuo K, Nagamani S, Sastry GN (2023) Polypharmacology guided drug repositioning approach for SARS-CoV2. PLoS ONE 18(8): e0289890. https://doi.org/10.1371/journal.pone.0289890
Editor: Chandrabose Selvaraj, Saveetha University - Poonamallee Campus: SIMATS Deemed University, INDIA
Received: April 9, 2023; Accepted: July 27, 2023; Published: August 9, 2023
Copyright: © 2023 Jamir et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: Funding to carry out the research work was obtained by DBT Centre of Excellence in Advanced Computation and Data Sciences (No. BT/PR40188/BTIS/137/27/2021).
Competing interests: The authors declare no competing interests.
Introduction
The SARS-CoV2 (severe acute respiratory syndrome coronavirus 2) has caused a pandemic [1–3] and it has affected the lives of millions of people severely collapsing the healthcare system and economy [4]. The development of therapeutics and antiviral drugs is critical for meeting the challenges currently faced by viral infections [5]. Though therapies are available for the treatment of SARS-CoV2 infection, they deal with only the symptoms of the disease but not its underlying causes and have not been able to eradicate the disease completely [6–9]. The emergence of new mutant strains may be the primary cause of the increased transmissibility of SARS-CoV2 virus [10, 11]. So far, approximately 30 SARS-CoV2 proteins have been identified [12], and the occurrence of mutations have increased the rate of infection, replication, transcription and transmission [13–15]. SARS-CoV2 is reported to be 96.2% similar to bat CoV RaTG13 sequence and it shares 79.5% identity with SARS-CoV genome [16]. A 5’ cap and a 3’ poly-A tail is present in the positive-stranded RNA genome structure [17] where a replicase complex of ORFs i.e., ORF1a and ORF1b are present at the 5’ cap. The ORF1a and ORF1b produces polypeptide pp1a and pp1ab respectively, the pp1a comprises non-structural protein (NSP) 1 to NSP11, while pp1ab comprises of NSP12 to NSP16. These NSPs are involved in viral replication, transcription and immune invasion [18]. Different structural proteins such as S (Spike), N (Nucleoprotein), M (Membrane), and E (Envelope) proteins are encoded at the 3’ end [19]. There are nine accessory proteins (ORF3a, ORF3b, ORF6, ORF7a, ORF7b, ORF8, ORF9b, ORF9c, ORF10). These proteins are significant for virulence and host interaction [20–22]. Due to the high sequence similarity among the coronavirus, these proteins could be studied for the development of an antiviral drug. S1a, S1b Fig in S1 File displays the selected proteins of SARS-CoV2 that are involved in viral entry, RNA synthesis and viral replication.
The drug and vaccine development process for the treatment and prevention of COVID-19 has intrigued the scientific community’s interest around the world [23, 24]. Several vaccines have been developed around the world, but due to the rise of new viral strains, it is still a challenge to eradicate the virus [25]. Thus, the drug repurposing [26, 27] approach may have the possibility to identify the potential drug molecules against COVID-19 within a short period. Numerous modelling approaches have been applied for repurposing pre-existing drugs [28–30] to halt the rise of the COVID-19 pandemic. Although there are a few repurposed drugs for COVID-19 available in the market, some of the drugs are reported to have limitations. For instance, patients under remdesivir treatment were reported to be experiencing adverse events during follow-up. Similarly, few of the repurposed drugs for COVID-19 such as hydroxychloroquine and chloroquine were reported to cause serious heart problems and other safety issues [31, 32]. Several approaches such as polypharmacology with drug repurposing are reported to be effective for exploring potential leads and druggable targets for infectious diseases [33]. The polypharmacology approach involves the identification of small molecules that have activity against multiple targets. This method is widely used for identifying drugs with polypharmacological properties for targeting multiple SARS-CoV2 proteins [34]. In our previous study, a combined polypharmacology and drug repurposing approach was applied to screen FDA (Food and Drug Administration) approved drugs against seven SARS-CoV2 targets. Four drugs (i.e., venetoclax, tirilazad, acetyldigitoxin, and ledipasvir) were chosen based on the docking scores, binding pose and the interaction pattern with the SARS-CoV2 targets [35]. In arriving at the drug repurposing a consensus scoring approach has been adopted [36]. Similarly, earlier in the group, we employed an integrated approach involving structure and analog-based approaches to arrive at lead molecules [37–40]. All 30 targets were carefully analyzed and 24 targets were selected based on their functions, binding pocket size and the presence of active site residues. The 24 targets considered in this study are involved in viral entry, RNA synthesis, polyprotein cleavage and replication.
Materials and methods
Selection of target proteins and their structural details
In this study, 24 SARS-CoV2 protein targets were considered based on their role in viral entry, replication, transcription, presence of a binding pocket and the availability of protein structure with a good resolution in RCSB Protein Data Bank (PDB) (http://www.rcsb.org) [41]. Table 1 provides the PDB ID, resolutions, length, and activity of the chosen targets: NSP1, NSP2, NSP3, NSP5, NSP7, NSP8, NSP9, NSP10, NSP12, NSP13, NSP14, NSP15, NSP16, S, E, N, ORF3a, ORF7a, ORF8, and ORF9b, NSP4, NSP6, M, and ORF6. Among these, four proteins namely NSP4, NSP6, M and ORF6 were obtained from the I-TASSER server (https://zhanggroup.org/I-TASSER/) [42] due to the lack of available resolved crystal structures during this study. The 3D structures of the proteins are given in S1a, S1b Fig in S1 File.
Preparation of protein structures
The selected 24 SARS-CoV2 proteins were prepared in Autodock tools (ADT version 1.5.7rc1) [43], where the solvent molecules and ions were removed, and the partial charges and polar hydrogens were added to the structure. The protein structures were then saved in the PDBQT format. In addition, a grid box was generated based on the active site and neighbouring residues as shown in S1 Table in S1 File.
Preparation of compound structures
The drug molecules were retrieved from DrugBank [44] and DrugCentral [45] databases. The redundant molecules and molecules containing ions were removed and 4193 unique approved molecules were considered in this study. Further, the raccoon.py script of MGL tools was used for optimizing the geometry of the compounds, and the compounds were converted to 3D structures and PDBQT format.
Virtual screening
The virtual screening of the 4193 known drugs against the 24 SARS-CoV2 targets was performed using AutodockVina1.1.2. Based on the docking scores the best conformer was identified out of the five conformers generated and the selected complexes were then taken for further analysis. Further to validate the virtual screening, three other programs namely Molecular Transfer Drug Target Interaction prediction (MT-DTI), SwissDock, and iGEMDOCK was used to calculate the binding energies of the top 10 molecules for each target. The MT-DTI is a drug-target interaction predicting module using a deep learning library DeepPurpose [46]. SwissDock is a web-based program for protein-ligand docking and virtual screening and its scoring function is based on empirical force filed and statistical functions [47]. The iGEMDOCK is a software tool that predicts the binding mechanism and affinities of protein-ligand complexes and rates the screened compounds using pharmacological interactions and an energy-based scoring function [48].
Analysis of most active and least active molecules
The docking score of all the drug molecules for each target was arranged in ascending order and the frequency of the docking score was analysed. The maximum number of drugs that fall in a range of docking scores were considered as the “threshold” value and the list of drugs was classified into two different portions. The docking scores higher than the threshold were considered active molecules and lesser than the threshold was considered inactive molecules. Further, the top 10 ranked candidates from the active molecule and 10 low-ranking candidates from the inactive molecules were analysed in detail.
Distribution of therapeutic indications toward the SARS-CoV2 targets
The top 10 and bottom 10 compounds were selected to analyze the most active and least active therapeutic indications for the SARS-CoV2 targets. The therapeutic indication of each compound was retrieved from the drug databases namely DrugBank [44] and DrugCentral databases [45] and then the distribution of drugs with a category of therapeutic indications towards the SARS-CoV2 targets was carried out for the top 10 and bottom 10 compounds. In addition, the reported FDA approved drugs and drugs under clinical trials for COVID-19 disease were identified among the top-scored drugs in the study. Fig 1 depicts the schematic workflow employed in the current study.
Results
Virtual screening of drug molecules against 24 targets
Virtual screening of the known drug molecules obtained from DrugBank [44] and DrugCentral database [45] was carried out against the 24 SARS-CoV2 viral targets. The molecules were distributed based on the docking score and a bar graph was plotted for each target to identify the number of drugs falling under different docking scores (S2a, S2b Fig in S1 File). It can be observed that the average maximum docking score for each protein is around -5.0 kcal/mol to -7.0 kcal/mol and the least docking scores were found to be around -1.0 kcal/mol to -4.0 kcal/mol.
Further drug molecules were categorised into active and inactive dataset where the docking score cut-off was determined for each target (i.e., the score range that had maximum no. of drugs) and the drug molecules which scored higher than the cut-off were considered as most active molecules whereas the molecules whose docking score was less than the cut-off were considered as least active molecules. The threshold value was selected based on the highest docking score for each protein distributed across the drug. Table 2 shows the grouping of the total compounds into active and least active groups based on the cut-off scores. The top ten molecules from the most active groups and the bottom ten compounds from the least active groups were selected for further analysis. From the distribution of active and least active drugs for 24 SARS-CoV2 targets, it can be observed that among the total number of drugs, the average percentage of drugs falling under the most active drugs was found to be 53.5% while the average percentage of drugs falling under the least active drugs was found to be 46.5%. The drugs falling under the most and least active drugs were further taken for analysis of therapeutic indications. In addition, to confirm the binding affinity of the drug-target interaction, MT-DTI prediction, SwissDock, and iGEMDOCK was used to validate the top 10 molecules displaying the best docking score against each target. It has been observed that the binding scores obtained through the three programs were comparable with that of the docking energy obtained from Autodock Vina. Several studies have also reported the prediction of antivirals such as remdesivir and lopinavir for SARS-CoV2 using MT-DTI and also the use of SwissDock and iGEMDOCK for screening potential leads against SARS-CoV2 targets [49–51]. Thus, the confirmation of docking scores of top 10 FDA approved drugs across the 24 SARS-CoV2 targets with the binding affinity obtained from MT-DTI model, SwissDock and iGEMDOCK validates the screening and selection of top 10 drugs. The docking and binding scores calculated from Autodock vina, MT-DTI, SwissDock, and iGEMDOCK are listed in S2 Table in S1 File. In order to validate the docking protocol, the free energy calculation was carried out using online server FastDRH (http://cadd.zju.edu.cn/fastdrh/submit) for multi targeting drugs. This server integrates Autodock Vina, Autodok GPU docking engines, structure truncated MM/PB (GB)SA, per-residue energy decomposition analysis for multiple residues to predict the binding free energies. The analysis revealed that the binding free energies are well co-related with docking score (S3 Table in S1 File).
The total number of drugs that has docking score > threshold value has been considered as active (%) control, whereas the drugs that have docking score < threshold value has been considered as inactive (%).
Therapeutic indications of the top 10 and bottom 10 drugs
The therapeutic indication of all the drug molecules was identified through DrugBank [44] and DrugCentral [45] databases which were then analysed to identify drugs with a category of therapeutic indications that may be repurposed against 24 SARS-CoV2 targets. The top 10 drugs that showed good binding affinity towards the 24 SARS-CoV2 targets are listed in Tables 3–5 along with their existing therapeutic indication for structural, non-structural and accessory SARS-CoV2 proteins respectively. From the table, it can be observed that among a total number of 240 drugs in the top 10, around 162 drugs were unique and 39 drugs occurred multiple times across the 24 SARS-CoV2 targets.
The overall binding score distribution of the top 10 drugs across the 24 SARS-CoV2 targets is given in Fig 2. From the graph in Fig 2A, among the non-structural proteins, it can be observed that sonidegib, which is an anticancer drug, has the highest docking score of -11.8 kcal/mol with NSP16 protein. And drugs such as desloratadine, mianserin, metapramine and mecloqualone which are used as anti-depressive agents have the lowest docking score of around -5.5 kcal/mol with NSP7 protein. In the case of structural proteins shown in Fig 2B, nucleoprotein interacting with paritaprevir drug which is used for the treatment of hepatitis was found to have the highest docking score of -12.6 kcal/mol. While drugs interacting with envelope proteins which showed therapeutic indications for liver disorder, pain, diabetes, nausea, cancer, and pancreas disease appeared to have lower docking score of around -6.0 kcal/mol compared to the other docking score. Then, among the accessory proteins as shown in Fig 2C, ORF6 is observed to have the highest docking score of -10.1 kcal/mol with bisdequalinium chloride which is used for bacterial infections. While, the protein ORF7a had a lower docking score of around -5.5 kcal/mol with drugs such as hydromorphone, naproxen, phenformin, primidone which are commonly used for treating pain, haemorrhage and epilepsy.
The figure depicts the distribution of top the 10 FDA approved drugs against SARS-CoV2 proteins namely, a) non-structural proteins namely NSP1, NSP2, NSP3, NSP4, NSP5, NSP6, NSP7, NSP8, NSP 9, NSP10, NSP12, NSP13, NSP14, NSP15 and NSP16, b) structural proteins namely Envelope, Spike and Nucleoprotein, and c) accessory proteins namely ORF3a, ORF6, ORF7a, ORF8, ORF9.
The overall distribution of therapeutic indications in terms of disease area for the top 10 and bottom 10 drugs across the 24 SARS-CoV2 targets are shown in Fig 3. From the graph, it can be observed that the drugs with high binding energies were found to have maximum therapeutic indications for CNS (central nervous system) followed by immunological disorder with the least for pain management. Whereas, the drugs with low binding energies were mostly found to have therapeutic indications for immunological disorders followed by drugs for vitamin deficiencies and anesthetics.
The classification of a) top 10 and b) bottom 10 drugs based on their therapeutic areas. The figure depicts that the drugs with good docking score are mainly used for neurological disorders whereas poor docking score drugs are mainly from immune related disease.
Identification of existing approved drugs for COVID-19
According to the WHO report (WHO-2019-nCoV-therapeutics-2022) two known drugs namely remdesivir (antiviral compound) and baricitinib (anti-arthritis compound) are reported to be approved for the treatment of COVID-19. Along with these, several studies have reported a list of drugs under clinical trials and those used for emergency treatment of COVID-19. In the current study, these known drugs were identified among the drugs interacting with 24 SARS-CoV2 targets. Although the existing FDA approved drugs and drugs which are under clinical trials for COVID-19 were not found among the top 10 drugs, few of them showed good binding affinity towards the selected SARS-CoV2 targets and have been highlighted along with the top 10 drugs as shown in S3a, S3c Fig in S1 File. From the graphs it can be observed that the known drugs remdesivir, baricitinib, ritonavir, hydroxychloroquine, lopinavir, favipiravir and dexamethasone had good docking scores due to stable interaction with almost all the 24 SARS-CoV2 targets. The two reported approved drugs for COVID-19, remdesivir and baricitinib were found to be interacting with all the SARS-CoV2 targets except for envelope and ORF7a proteins. Several studies have reported remdesivir as an antiviral drug to inhibit the function of RNA-dependent RNA polymerase (NSP12) with a binding score of -5.0 kcal/mol to -10 kcal/mol [52–54]. From the study, it can be compared that remdesivir has a binding score of -7.0 kcal/mol with NSP12 which confirms the significance of the study. The highest docking score of -8.1 kcal/mol was observed with membrane protein, while the lowest docking score of -2.2 kcal/mol with NSP7 and an average binding score of -6.5 kcal/mol across the interacting targets. Similarly, baricitinib an anti-arthritis drug, is reported to interrupt the entry and intracellular assembly of SARS-CoV2 into the host cell and interfere with the immune response [55–58]. In this regard, baricitinib was observed to have the highest docking score of -7.9 kcal/mol with membrane, which is involved in humoral response during the SARS-CoV2 and host interaction. Meanwhile, the lowest score of -3.4 kcal/mol was observed for nucleoprotein with an average binding score of -5.9 kcal/mol across the interacting targets, where some of the targets are involved in viral entry and immune response. This suggests that some of the existing known drugs for COVID-19 have shown good binding energy along with the top 10 drugs, which can be of significant importance for repurposing against multiple SARS-CoV2 targets.
Privileged scaffold identification and their analysis
For the identification of the privileged scaffolds, a total of 162 unique drugs present in the top 10 drugs across the 24 SARS-CoV2 targets were taken to generate 3 levels of scaffolds. The scaffold tree displayed three levels 1,2, and 3, which consist of 71, 103 and 123 scaffolds respectively. The identified scaffolds were then compared with the scaffolds generated in each level for the existing FDA approved drugs and drugs under clinical trials for COVID-19 disease. From the analysis, four common scaffolds (6,7,8,9-Tetrahydro-5H-cyclohepta[c]pyridine, 1-Benzazepine, decalin and leucoline) in level 2 and two common scaffolds (4,5,6,7,8,8a,9,10-Octahydro-2(3H)-phenanthrenone, Gona-1,3,5(10)-trien-3-ol) in level 3 of the scaffold tree were observed and are shown in Fig 4. The compounds present in level 2 and 3 of the scaffold trees generated from the unique compounds can be considered as potential candidates for designing drugs to inhibit the effect of SARS-CoV2. In addition, 2D interaction analysis was carried out to analyse the interactions of the identified scaffold with 24 SARS-CoV2 targets. Targets namely, envelope, membrane, NSP4, NSP5, NSP14 and ORF9b were observed to have interactions with the scaffolds forming bonds such as π-alkyl, π - π stacked, conventional hydrogen bonds, π -sigma and mostly alkyl bonds (S4 Table in S1 File). The common scaffolds along with the interacting site information can also serve as starting points for the development of novel SARS-CoV2 inhibitors.
The level 2 and level 3 common scaffolds that are present among the top scored drug molecules across all the 24 SARS-CoV2 targets, existing approved drugs and drugs under clinical trials for COVID-19.
Polypharmacological property analysis
The effectivity of antivirus progressively goes down due to the increased rate of viral mutation and develops resistance toward the drug action [59]. Hence, targeting multiple targets involved in the same or different pathways through polypharmacology approach [60, 61] seems to be an effective strategy for combating a complex disease like COVID-19. In this regard, the FDA approved drugs in the top ten list were filtered based on the drugs interacting with multiple targets of more than 3 SARS-CoV2 proteins. A graph was generated representing the number of interacting targets against the drug as shown in Fig 5.
A total of 15 drugs are observed to have interaction with multiple targets (more than 3) and the maximum number of targets are found to be interacting with Dhydroergotamine, Ergotamine and Midostaurin.
From Fig 5, it can be observed that out of the total number of 162 unique drugs present in the top 10 drugs across the 24 SARS-CoV2, 15 drugs were found to have interaction with 3 or more targets. Among these drugs, dihydroergotamine and ergotamine showed a maximum number of interactions with around 8 SARS-CoV2 targets. Drugs such as bisdequalinium chloride, midostaurin, temoporfin, tirilazad and venetoclax had interaction with around 5 numbers of SARS-CoV2 targets. These drugs have therapeutic indications for migraine, bacterial infection, central nervous system disorder and most commonly cancer (Table 6). Several studies have reported that dihydroergotamine and ergotamine are having stable interactions with 3CLpro and also it has the polpypharmacology efficacy [55, 62]. In addition, the drugs tirilazad and venetoclax were reported to show stable binding affinity towards multiple targets for SARS-CoV2 [35, 56], revealing their polypharamocological property.
Protein-ligand interaction analysis
One of the most important steps in the drug development process is understanding how drugs interact with target proteins and identifying the important interacting residues [63]. Thus, the 15 drugs interacting with more than 3 SARS-CoV2 targets were further taken for 2D interaction analysis with their respective proteins using Discovery studio. This was carried out to identify the important interacting residues along with the non-covalent bonds contributing to the protein-ligand binding. Among the multi-targeting drugs, dihydroegotamine and ergotamine were observed to have interaction with the maximum number (8) of SARS-CoV2 proteins which is shown in Fig 6. It can be observed that both the drugs have interaction with eight proteins and shares common interacting proteins such as membrane, NSP2, NSP12, nucleoprotein, ORF3a and ORF9b protein. Further, the detailed 2D interactions of drugs with eight, five, four and three numbers of SARS-CoV2 targets are given in S4a-S4d Fig in S1 File. From the figures depicting the 2D interactions of drugs and proteins, it can be observed that amino acid residues such as ILE, LEU, ALA, ARG and VAL are some of the most commonly occurring amino acids involved in the overall interactions. These amino acids are mostly found to be hydrophobic in nature and hydrophobic interactions are reported to be the driving force between protein-ligand interactions as it contributes to the binding affinity of the complexes [64]. Among the various non-covalent bonds occurring in a drug-protein interaction, hydrogen bond plays a significant role in forming the protein-ligand complex and these interactions are reported to be important for consideration while designing inhibitors against a target [65]. In regard to this, it can be observed that polar-amino acids [66] such as SER, ASN, GLN, THR are involved majorly in hydrogen formation (S5 Table in S1 File). Other positively charged and non-polar amino acids like ARG and GLY respectively are also found to be involved in the hydrogen bond formation. Along with hydrogen bonds, other non-covalent interactions such as π-alkyl, π - π stacked, conventional hydrogen bonds, π -sigma and alkyl bonds are observed in the drug-protein interaction. The hydrophobic and hydrogen bond forming residues are also observed among the active site residues of the 24 SARS-CoV2 proteins as shown in S1a, S1b Fig in S1 File. For example, the active site residue of NSP12 was found to interact with Dihydroergotamine through the residue ARG553, which is involved in the formation of hydrogen bonds. In the case of ergotamine, the interacting residues and active site residues of the membrane were THR52, and for NSP3, it was GLY163 and GLU167, which are involved in -sigma bonds and hydrogen bonds, respectively. Similarly, among the drugs interacting with 5, 4 and 3 targets it was observed that few of the active site residues of the protein were shown to be among the interacting residues which are hydrophobic in nature and are mainly involved in hydrogen bond formation. These identified residues involved in the formation of drug-protein interactions will help in understanding the therapeutic effect of a drug against the target and also in identifying the important residues which can be targeted while designing high binding and target specific inhibitors.
Among the drugs interacting with multiple targets, (a) Dihydroergotamine and (b) ergotamine was observed to have interaction with the maximum number of SARS-CoV2 targets (8 targets). The figure represents an example of 2D interaction with membrane protein among the 8 proteins commonly for both the drugs. The drug molecule is seen to bound in the cavity of the protein through various non-covalent interactions with the amino acid residues that contributes to the protein-ligand binding affinity.
Pathway analysis for drug repurposing
Gene enrichment analysis.
Analyzing the biological pathway of a particular drug’s targets can provide valuable insight into the mechanism of a drug’s action and its impact on repurposing the drug for new indications [67]. 15 drugs interacting with multiple targets were selected and their original targets were obtained from publicly accessible databases such as Drug bank [44] and Drug Central [45]. A total of 26 targets have been obtained, and a gene enrichment analysis was performed using the network analysis tool ShinyGo (http://bioinformatics.sdstate.edu/go/). The gene enrichment analysis was performed for the genes involved in biological processes, cellular components, and molecular function pathways. A cut-off minimum 0.05 P-value was maintained for FDR (false discovery rate) and the top 20 pathways were selected for further analysis. Fig 7 depicts the constructed network and gene characteristic plot.
Gene enrichment analysis for a) Biological process, b) Cellular component and c) Molecular function of genes associated with top screened drugs interacting with multiple SARS-CoV2 targets. The network represents the interaction of nodes (genes involved in biological processes, cellular components and molecular functions) connected through edges. From the enrichment plot, it can be observed that for biological processes, the maximum number of gene is involved in regulation of biological quality, for Cellular component, the maximum number of genes are involved in cell junction and for Molecular function, the maximum number of genes are involved in molecular transducer activity.
The network in Fig 7 depicts the interacting nodes of genes involved in biological processes, cellular components, and molecular function respectively. It displays the functional categories of genes that are most highly enriched as well as how the pathways have connected the genes. From the enrichment plot, it can be observed that the majority of genes in the biological process categories were involved in the control of biological quality. This indicates that most of the genes in the biological process are found to be involved in the regulation and control of the biological quality of cellular functions such as cell cycle regulation, DNA repair, apoptosis etc [68]. Due to the short size of their genomes, viruses have a restricted ability to code and as a result, they use the host cellular mechanism and components to enable them to replicate [69]. In the process, the virus influences the host cell cycle leading to failure in safeguarding the host cell thus affecting host DNA replication and repair mechanism [69]. The pathway with the most preferable FDR score was observed to be phospholipase C-activating G protein-coupled receptor (GPCR) signalling pathway (Fig 7A). GPCR signalling pathway plays an important role in governing biological events such as cell differentiation, apoptosis and also cellular growth and is known to interact with integral membrane proteins [70]. Studies have indicated the connection between the host immune response during SARS-CoV2 infections and the G protein-coupled receptor signalling pathway, as the virus enters the host by attaching to a GPCR namely angiotensin-converting enzyme 2 (ACE2) receptor [71]. In the category of cellular components, the majority of the genes were found to be a part of the integral component of plasma membrane and this pathway was also observed to have the most preferable FDR score (Fig 7B). The plasma membrane is a highly specific permeable barrier that encompasses cell components and performs a number of functions such as sustaining the cell structure and shape, controlling the transfer of stimulants signal transmission [72]. According to a study reported by Deng et al. (2021), the human transmembrane protease serine 2 is activated when the SARS-CoV2 enters the host cell through endocytosis and directs fusing of the viral membrane with the host plasma membrane [73]. So, it is probable that the genes identified as a part of the integral component of the plasma membrane are essential for preserving cell functionality and the proper administration of membrane during SARS-CoV2 infection. Further, in the molecular function category, the maximum number of genes were involved in molecular transducer activity (Fig 7C). This includes genes that encode receptors that are involved in ion channelling and cell signalling [72]. According to several studies, the number of certain key ions present in channelling and signalling during the viral-host attachment is an important factor that influences the attachment, infection and invasion of SARS-CoV2 in the host [74]. In addition, from the enrichment plot, it was observed that the pathway associated with G protein-coupled receptor activity showed the lowest FDR score similar to the pathway in the biological process as mentioned earlier. This indicates the significance of genes involved in this pathway and the crucial role of G protein-coupled receptor activity in the immune response during the SARS-CoV2 infection [70, 71]. Thus, identifying the involvement of drugs’ genes in pathways linked to various biological processes during SARS-CoV2 infection supports the finding of the current study in determining potential drugs that may be repurposed against SARS-CoV2. From the overall analysis, it can be seen that majority of the top 10 drugs are predominantly involved in pathways related to fundamental cellular mechanisms like cell cycle regulation, cell functionality, cell signalling, transmission, ion channelling, DNA repair, apoptosis and immune response (S6 Table in S1 File) This suggests that these pathways and the genes involved in the pathways are significant targets for drug repurposing and may have therapeutic application for COVID-19 disease.
KEGG pathway analysis.
KEGG pathway analysis was carried out to study the functional role of genes and identify specific pathways that are differentially regulated depending on the disease condition or drug being used. The 26 genes linked to the 15 drugs were taken for the KEGG pathway analysis using the online server Metascape (https://metascape.org) and the parameters for calculation was maintained as default [75]. From the study, enriched ontology clusters, a protein-protein interaction network and an enrichment heatmap was generated (Fig 8) which were analysed for identifying particular biochemical interactions and response networks involved in a biological process.
a) Enriched ontology clusters represent the clusters of genes which are involved in the different pathway and the nodes of the colour represent the genes belonging to the same clusters. The nodes are linked by edges where the thickness of the edges represents the similarity score of the genes. b) Protein-protein interaction network of the genes involved in different pathways. The proteins which are most densely connected have been clustered through MCODE algorithm to identify the protein neighbourhoods. c) Enrichment analysis of genes-disease association represents the genes which are involved in different diseases. The images were generated using Metascape [71].
In Fig 8, the enriched gene cluster (Fig 8C) represent clusters of genes in colours which are involved in a particular pathway and size of the cluster represents the number of genes involved. Each cluster is connected by edges and the thickness of the edges indicates the level of similarity among the clusters of pathways. From the enriched gene cluster it can be observed that majority of the genes are involved in specific role such as G protein-coupled receptor activity, positive regulation of mitogen-activated protein kinase (MAPK) cascade, signalling pathways, cellular response and regulation. These pathways involved in G protein-coupled receptor activity, cellular response and regulation and cell signalling have also been identified in the gene enrichment analysis and linked to SARS-CoV2 infection [70–72]. From the protein-protein interaction network (Fig 8B), it can be observed that two main clusters of protein that are densely connected to each other have been identified among the various other clusters of proteins obtained through Molecular Complex Detection (MCODE). The densely connected proteins represents highly interconnected clusters which are associated to a particular biological process and can be analysed to identify potential targets for therapeutics [76]. From the identified clusters, it has been observed that 12 genes (ADRA1A, HTR2A, ADRA1D, P2RY2, ADRA1B, AVPR1A, AGTR1, OPRM1, HTR1B, OPRD1, OPRK1, HTR1D) are highly interconnected. These genes were mainly found to be involved in pathways such as G protein-coupled receptor activity, cellular response and regulation, cell signalling, immune response and apoptosis in both the gene enrichment and KEGG pathway analysis. The presence of these genes and pathways in both the analysis indicates the important role of these genes and the pathways involved in various biological processes particularly in SARS-CoV2 infection. In addition, the enrichment analysis of genes-disease association (Fig 8C) has shown the involvement of genes mostly in diseases related to neurological disorder, pain and some in cancer. This supports the identification of multi targeting drugs among the top 10 drugs with therapeutic indication for migraine, bacterial infection, neurological disorder and cancer.
Discussion
The study involved screening of FDA approved drugs against the 24 SARS-CoV2 targets, followed by drug categorization into most active and least active based on high and low docking score respectively across the targets. The docking scores of the top 10 FDA approved drugs against the 24 SARS-CoV2 targets obtained from Autodock Vina were validated using MT-DTI, SwissDock, and iGEMDOCK and it was observed that similar docking scores were obtained through all the three methods. Upon categorization, the therapeutic indications of the drugs were allocated to identify the categories of drugs with significant therapeutic indications that may have the potential for repurposing against SARS-CoV2. It was found that the majority of the compounds with high docking scores had the most therapeutic potential for neurological disorders, followed by immunological disorders, cancer, pain, viral, and bacterial diseases. Several studies have also reported drugs with antiviral, anticancer, and anti-inflammatory properties for targeting SARS-CoV2 proteins [77]. This was then followed by scaffold analysis of the 162 drugs present in the top 10 drugs across the 24 SARS-CoV2 targets. Four common scaffolds (6,7,8,9-Tetrahydro-5H-cyclohepta[c]pyridine, 1-Benzazepine, decalin and leucoline) in level 2 and two common scaffolds (4,5,6,7,8,8a,9,10-Octahydro-2(3H)-phenanthrenone, Gona-1,3,5(10)-trien-3-ol) in level 3 of the scaffold tree were identified. The identified scaffolds along with the interacting residues of proteins that are involved in the interaction can be used to develop novel inhibitors with higher binding affinity. Since it is of paramount importance to find drugs that can bind to multiple targets in order to counteract viral mutational modifications, a polypharmacological property analysis was carried out where the top ten FDA approved drugs were filtered based on their interactions with more than three SARS-CoV2 target proteins. A total of 15 drugs were found to interact with multiple SARS-CoV2 targets, with dihydroergotamine and ergotamine leading the list by binding with eight targets. Then five number SAR-CoV2 targets were shown to interact with bisdequalinium chloride, midostaurin, temoporfin, tirilazad, and venetoclax. The original therapeutic indications of the identified drugs are quite varied, ranging from migraine, bacterial infection, neurological disorder and cancer. These multi-target drugs and their bound complex structure with the corresponding proteins resulted in identifying the key interacting residues involved in drug-target binding. ILE, LEU, ALA, ARG, and VAL are some of the most frequently occurring amino acid residues in the overall interactions, which are mostly hydrophobic in nature and contribute to the complexes’ binding affinity. Among non-covalent bonds, hydrogen bonds play an important role in maintaining the stability of drug-target interactions, and according to the study, polar amino acids such as SER, ASN, GLN, and THR are majorly found to be involved in hydrogen formation. Targeting these residues will not only aid in comprehending the therapeutic effect of the drug on the target but will also contribute to developing high binding and target specific inhibitors. In the context of drug repurposing, pathway analysis can offer valuable insights into the mechanism of action of a drug. Therefore, in order to identify the functional category of key biological pathways implicated among the targets of the 15 drugs, a gene enrichment analysis of targets related to the 15 drugs that interact with multiple targets was conducted. Majority of the top 10 drugs mostly influence pathways which are related to basic biological functions such as cell cycle regulation, cell functionality, cell signalling, transmission, ion channelling, DNA repair, apoptosis and immune response. Furthermore, studies indicating that the identified pathways are linked to various biological process during SARS-CoV2 infections supports that these pathways and involved genes are important drug repurposing targets and could be used for the development of drugs for COVID-19. In addition, the KEGG pathway analysis have led to the identification of important genes involved in a specific pathway and how the expression of gene is influenced under different condition particularly in relation to the SARS-CoV2 infection. The gene enrichment analysis and the KEGG pathways of genes connected to multi-target drugs reveals details about several pathways that could help in understanding the mechanisms of action underlying these drugs and how they could be repurposed against SARS-CoV2 targets. Drugs obtained through virtual screening with long term efficacy can be challenging in drug discovery process as virus undergo rapid mutation effecting the binding affinity and efficacy of drug. However, several studies have reported using a structure-based virtual screening strategy to find suitable alternative drugs for those that have developed resistance to SARS-CoV2 variants [78, 79]. One such strategy is the multi-target approach which is employed in the current study to identify approved drugs with polypharmacological properties. In the current study, FDA approved drugs were screened through a unique combination of docking study, therapeutic indication analysis along with polypharmacology and network pharmacology approach. This approach has led to identification of drugs targeting multiple proteins and pathways which is an essential component to enhance the efficacy of drug against viral variants. The traditional approaches of one target for one drug may become inefficient when the targets undergo frequent mutations. Besides considering the large number of pathways in viral entry, progression, replication, translocation, etc., a drug which can simultaneously inhibit multiple targets appear to be an optimal choice. Thus, the current study adopts polypharmacology assisted by network pharmacology.
Conclusions
In the current study, virtual screening of FDA approved drugs was carried out against 24 SARS-CoV2 targets along with categorization of top docking scored drugs based on their therapeutic indications through a combination of polypharmacological and network pharmacology approaches. The common scaffolds generated by molecules with high docking scores are helpful in designing drugs for disrupting the virulence of SARS-CoV2, due to their simultaneous involvement in inhibiting multiple pathways. Further, analysing the gene enrichment and KEGG pathway analysis of genes related to multi target drugs, has led to the identification of various genes and biological pathways that can be targetable for drug repurposing. The repurposed drugs with original indications against neurological disorders (tirilazad), pain (dihydroergotamine, ergotamine), cancer (midostaurin, venetoclax, temoporfin) and bacterial infections (bisdequalinium chloride) may emerge as effective anti SARS-CoV2 drugs, owing to their multi-targeting nature.
Acknowledgments
EJ thanks DST for Inspire Fellowship, KK thanks UGC for SRF Fellowship and HS thanks DBT Centre of Excellence for the fellowship.
References
- 1.
Coronavirus Disease Situation Reports (WHO) [Accessed in February, 2022].
- 2. Zhu N, Zhang D, Wang W, Li X, Yang B, Song J, et al. China Novel Coronavirus Investigating and Research Team. A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med. 2020; 382:727–733.
- 3. Pedersen SF, Ho YC. SARS-CoV-2: a storm is raging. J Clin Invest. 2020; 130:2202–2205. pmid:32217834
- 4. Reid AH, Taubenberger JK. The origin of the 1918 pandemic influenza virus: a continuing enigma. J Gen Virol. 2003; 84:2285–2292. pmid:12917448
- 5. Lotfi M, Hamblin MR, Rezaei N. COVID-19: Transmission, prevention, and potential therapeutic opportunities. Clin Chim Acta. 2020; 508:254–266. pmid:32474009
- 6. Menni C, Klaser K, May A, Polidori L, Capdevila J, Louca P, et al. Vaccine side-effects and SARS-CoV-2 infection after vaccination in users of the COVID Symptom Study app in the UK: a prospective observational study. Lancet Infect Dis. 2021; 21:939–949. pmid:33930320
- 7. Ho P, Zheng JQ, Wu CC, Hou YC, Liu WC, Lu CL, et al. Perspective Adjunctive Therapies for COVID-19: Beyond antiviral therapy. Int J Med Sci. 2021; 18:314–324. pmid:33390800
- 8. Borba MGS, Val FFA, Sampaio VS, Alexandre MAA, Melo GC, Brito M, et al. Effect of high vs low doses of chloroquine diphosphate as adjunctive therapy for patients hospitalized with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Infection: A randomized clinical trial. JAMA Netw Open. 2020;3: e208857. pmid:32330277
- 9. Chorin E, Dai M, Shulman E, Wadhwani L, Bar-Cohen R, Barbhaiya C, et al. The QT interval in patients with COVID-19 treated with hydroxychloroquine and azithromycin. Nat Med. 2020; 26:808–809. pmid:32488217
- 10. Singh J, Samal J, Kumar V, Sharma J, Agrawal U, Ehtesham NZ, et al. Structure-function analyses of new SARS-CoV-2 variants B.1.1.7, B.1.351 and B.1.1.28.1: Clinical, diagnostic, therapeutic and public health implications. Viruses. 2021; 13:439. pmid:33803400
- 11. Wang R, Chen J, Gao K, Hozumi Y, Yin C, Wei GW. Analysis of SARS-CoV-2 mutations in the United States suggests presence of four substrains and novel variants. Commun Biol. 2021; 4:228. pmid:33589648
- 12. Arya R, Kumari S, Pandey B, Mistry H, Bihani SC, Das A, et al. Structural insights into SARS-CoV-2 proteins. J Mol Biol. 2021; 433:166725. pmid:33245961
- 13. Hossain MU, Bhattacharjee A, Emon MTH, Chowdhury ZM, Ahammad I, Mosaib MG, et al. Novel mutations in NSP-1 and PLPro of SARS-CoV-2 NIB-1 genome mount for effective therapeutics. J Genet Eng Biotechnol. 2021; 19:52. pmid:33797663
- 14. Banoun H. Evolution of SARS-CoV-2: Review of mutations, role of the host immune system. Nephron. 2021;145:392–403. pmid:33910211
- 15. Fehr AR, Perlman S. Coronaviruses: an overview of their replication and pathogenesis. Methods Mol Biol. 2015; 1282:1–23. pmid:25720466
- 16. Wang C, Liu Z, Chen Z, Huang X, Xu M, He T, et al. The establishment of reference sequence for SARS-CoV-2 and variation analysis. J Med Virol. 2020; 92:667–674. pmid:32167180
- 17. Kim D, Lee JY, Yang JS, Kim JW, Kim VN, Chang H. The architecture of SARS-CoV-2 transcriptome. Cell. 2020; 181:914–921. pmid:32330414
- 18. Subissi L, Imbert I, Ferron F, Collet A, Coutard B, Decroly E, et al. SARS-CoV ORF1b-encoded nonstructural proteins 12–16: replicative enzymes as antiviral targets. Antiviral Res. 2014; 101:122–130. pmid:24269475
- 19. Schoeman D, Fielding BC. Coronavirus envelope protein: current knowledge. Virol J. 2019; 16:69. pmid:31133031
- 20. Redondo N, Zaldívar-López S, Garrido JJ, Montoya M. SARS-CoV-2 Accessory Proteins in Viral Pathogenesis: Knowns and Unknowns. Front Immunol. 2021; 12:708264. pmid:34305949
- 21. Jin Z, Du X, Xu Y, Deng Y, Liu M, Zhao Y, et al. Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors. Nature. 2020; 582:289–293. pmid:32272481
- 22. Osipiuk J, Azizi SA, Dvorkin S, Endres M, Jedrzejczak R, Jones KA, et al. Structure of papain-like protease from SARS-CoV2 and its complexes with non-covalent inhibitors. Nat. Commun. 2021; 12:1–9.
- 23. Liu C, Zhou Q, Li Y, Garner LV, Watkins SP, Carter LJ, et al. Research and development on therapeutic agents and vaccines for COVID-19 and related human coronavirus diseases. ACS Cent Sci. 2020; 6:315–331. pmid:32226821
- 24. Guo D. Old weapon for new enemy: Drug repurposing for treatment of newly emerging viral diseases. Virol Sin.2020;35:253–255. pmid:32048130
- 25. Moore JP, Offit PA. SARS-CoV2 vaccines and the growing threat of viral variants. Jama.2021;325: 821–822. pmid:33507218
- 26. Bohari MH, Sastry GN. FDA approved drugs complexed to their targets: evaluating pose prediction accuracy of docking protocols. J Mol Model. 2012; 18:4263–4274. pmid:22562231
- 27. Janardhan S, John L, Prasanthi M, Poroikov V, Sastry GN. A QSAR and molecular modelling study towards new lead finding: polypharmacological approach to Mycobacterium tuberculosis. SAR QSAR Environ. Res. 2017;28: 815–832. pmid:29183232
- 28. Kumar N, Sarma H, Sastry GN. Repurposing of approved drug molecules for viral infectious diseases: a molecular modelling approach. J Biomol. Struct. 2021;1–17. pmid:33810775
- 29. Choudhary MI, Shaikh M, Tul-Wahab A, Ur-Rahman A. In silico identification of potential inhibitors of key SARS-CoV-2 3CL hydrolase (Mpro) via molecular docking, MMGBSA predictive binding energy calculations, and molecular dynamics simulation. PLoS One. 2020;15(7):e0235030sss. pmid:32706783
- 30. Ippolito M, Cortegiani A. Length of remdesivir treatment in patients with severe COVID-19. Breathe (Sheff). 2021; 17:200276. pmid:34295393
- 31. Kumari P, Pradhan B, Koromina M, Patrinos GP, Steen KV. Discovery of new drug indications for COVID-19: A drug repurposing approach. PLoS One. 2022;17(5):e0267095. pmid:35609015
- 32. Gupta AK, Parker BM, Priyadarshi V, Parker J. Cardiac adverse events with remdesivir in COVID-19 infection. Cureus.2020;12:e11132. pmid:33240723
- 33. Madugula S.S, Nagamani S, Jamir E, Priyadarsinee L, Sastry G.N. Drug repositioning for anti-tuberculosis drugs: an in silico polypharmacology approach. Mol Diversity. 2021;26:1675–1695.
- 34. Artigas L, Coma M, Matos-Filipe P, Aguirre-Plans J, Farrés J, Valls R, et al. In-silico drug repurposing study predicts the combination of pirfenidone and melatonin as a promising candidate therapy to reduce SARS-CoV-2 infection progression and respiratory distress caused by cytokine storm. PLoS One. 2020;15(10):e0240149. pmid:33006999
- 35. Jamir E, Sarma H, Priyadarsinee L, Nagamani S, Kiewhuo K, Gaur AS, et al. Applying polypharmacology approach for drug repurposing for SARS-CoV2. J Chem Sci. 2022;134:157. pmid:35498548
- 36. Choudhury C, Priyakumar UD, Sastry GN. Dynamics based pharmacophore models for screening potential inhibitors of mycobacterial cyclopropane synthase. J Chem Inf Model. 2015;55:848–860. pmid:25751016
- 37. Reddy AS, Pati SP, Kumar PP, Pradeep HN, Sastry GN. Virtual screening in drug discovery—a computational perspective. Curr Protein Pept Sci. 2007;4:329–351. pmid:17696867
- 38. Badrinarayan P, Sastry GN. Virtual high throughput screening in new lead identification. Comb Chem High Throughput Screen. 2011;14:840–860. pmid:21843146
- 39. Ravindra GK, Achaiah G, Sastry GN. Molecular modeling studies of phenoxypyrimidinyl imidazoles as p38 kinase inhibitors using QSAR and docking. Eur J Med Chem. 2008;43:830–838. pmid:17706839
- 40. Srivani P, Srinivas E, Raghu R, Sastry GN. Molecular modeling studies of pyridopurinone derivatives—potential phosphodiesterase 5 inhibitors. J Mol Graph Model. 2007;26:378–390. pmid:17307372
- 41. Berman H, Henrick K, Nakamura H Announcing the worldwide protein data bank. Nat Struct Mol Biol. 2003;10:980–980.
- 42. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res 2018;46:W296–W303. pmid:29788355
- 43. Trott O, Olson AJ AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem. 2010;31:455–461. pmid:19499576
- 44. Wishart DS, Feunang YD, Guo AC, Lo EJ, Marcu A, Grant JR, et al. DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res. 2018; 46(D1):D1074–D1082. pmid:29126136
- 45. Ursu O, Holmes J, Knocke J, Bologa CG, Yang JJ, Mathias SL, et al. DrugCentral: online drug compendium. Nucleic Acids Res. 2017;45: D932–D939. pmid:27789690
- 46. Huang K, Fu T, Glass LM, Zitnik M, Xiao C, Sun J. DeepPurpose: a deep learning library for drug-target interaction prediction. Bioinformatics 2021;36:5545–5547. pmid:33275143
- 47. Grosdidier A, Zoete V, Michielin O. SwissDock, a protein-small molecule docking web service based on EADock DSS. Nucleic Acids Res. 2011;39(Web Server issue):W270–7. pmid:21624888
- 48. Hsu KC, Chen YF, Lin SR, Yang JM. iGEMDOCK: a graphical environment of enhancing GEMDOCK using pharmacological interactions and post-screening analysis. BMC Bioinform. 2011;12:S33. pmid:21342564
- 49. Choi Y, Shin B, Kang K, Park S, Beck BR. Target-Centered Drug Repurposing Predictions of Human Angiotensin-Converting Enzyme 2 (ACE2) and Transmembrane Protease Serine Subtype 2 (TMPRSS2) Interacting Approved Drugs for Coronavirus Disease 2019 (COVID-19) Treatment through a Drug-Target Interaction Deep Learning Model. Viruses. 202018;12(11):1325. pmid:33218024
- 50. Kumar S, Kashyap P, Chowdhury S, Kumar S, Panwar A, Kumar A. Identification of phytochemicals as potential therapeutic agents that binds to Nsp15 protein target of coronavirus (SARS-CoV-2) that are capable of inhibiting virus replication. Phytomedicine. 2021;85:153317. pmid:32943302
- 51. Vincent S, Arokiyaraj S, Saravanan M, Dhanraj M. Molecular Docking Studies on the Anti-viral Effects of Compounds From Kabasura Kudineer on SARS-CoV-2 3CLpro. Front Mol Biosci. 2020;7:613401. pmid:33425994
- 52. Shannon A, Le NT, Selisko B, Eydoux C, Alvarez K, Guillemot JC, et al. Remdesivir and SARS-CoV-2: Structural requirements at both nsp12 RdRp and nsp14 Exonuclease active-sites. Antiviral Res. 2020;178:104793. pmid:32283108
- 53. Yu R, Chen L, Lan R, Shen R, Li P. Computational screening of antagonists against the SARS-CoV-2 (COVID-19) coronavirus by molecular docking. Int J Antimicrob Agents. 2020;56:106012. pmid:32389723
- 54. Nguyen HL, Thai NQ, Truong DT, Li MS. Remdesivir strongly binds to both RNA-dependent RNA polymerase and main protease of SARS-CoV-2: Evidence from molecular simulations. J Phys Chem B. 2020;124(50):11337–1134. pmid:33264025
- 55. Gul S, Ozcan O, Asar S, Okyar A, Barıs I, Kavakli IH. In silico identification of widely used and well-tolerated drugs as potential SARS-CoV-2 3C-like protease and viral RNA-dependent RNA polymerase inhibitors for direct use in clinical trials. J Biomol Struct Dyn. 2021;39:6772–6791. pmid:32752938
- 56. Ghosh S, Bhattacherjee D, Satpati P, Bhabak KP. Venetoclax: a promising repurposed drug against SARS-CoV-2 main protease. J Biomol Struct Dyn. 2021;23:1–12. pmid:34424151
- 57. Zhang X, Zhang Y, Qiao W, Zhang J, Qi Z. Baricitinib, a drug with potential effect to prevent SARS-COV-2 from entering target cells and control cytokine storm induced by COVID-19. Int Immunopharmacol. 2020;86:106749. pmid:32645632
- 58. Richardson PJ, Corbellino M, Stebbing J. Baricitinib for COVID-19: a suitable treatment?—Authors’ reply. Lancet Infect Dis. 2020;20:1013–1014. pmid:32251639
- 59. Agarwal M, Panda PK. Use of Polypharmacy and Emergence of COVID-19 Variants—Are they Co-Related? Curr Drug Saf. 2023;18(1):2–4. pmid:35346002
- 60. Anighoro A, Bajorath J, Rastelli G. Polypharmacology: challenges and opportunities in drug discovery. J Med Chem. 2014;57(19):7874–87. pmid:24946140
- 61. Jamir E, Kiewhuo K, Priyadarsinee L, Sarma H, Nagamani S, Sastry GN. Structure-function relationships among selected human coronaviruses. Indian J Biochem Biophys. 2022;59:536–551.
- 62. Molavi Z, Razi S, Mirmotalebisohi SA, Adibi A, Sameni M, Karami F, et al. Identification of FDA approved drugs against SARS-CoV-2 RNA dependent RNA polymerase (RdRp) and 3-chymotrypsin-like protease (3CLpro), drug repurposing approach. Biomed Pharmacother. 2021;138:111544. pmid:34311539
- 63. Anusuya S, Kesherwani M, Priya KV, Vimala A, Shanmugam G, Velmurugan D, et al. Drug-Target Interactions: Prediction Methods and Applications. Curr Protein Pept Sci. 2018;19(6):537–561. pmid:27829350
- 64. Ferreira de Freitas R, Schapira M. A systematic analysis of atomic protein-ligand interactions in the PDB. Medchemcomm. 2017;8(10):1970–1981. pmid:29308120
- 65. Mahadevi AS, Sastry GN. Cooperativity in Noncovalent Interactions. Chem Rev. 2016;116(5):2775–825. pmid:26840650
- 66. Scheiner S, Kar T, Pattanayak J. Comparison of various types of hydrogen bonds involving aromatic amino acids. J Am Chem Soc. 2002;124(44):13257–64. pmid:12405854
- 67. Pan Y, Cheng T, Wang Y, Bryant SH. Pathway analysis for drug repositioning based on public database mining. J Chem Inf Model. 2014;54(2):407–18. pmid:24460210
- 68. Myhre S, Tveit H, Mollestad T, Laegreid A. Additional gene ontology structure for improved biological reasoning. Bioinformatics. 2006;22(16):2020–7. pmid:16787968
- 69. Mekawy AS, Alaswad Z, Ibrahim AA, Mohamed AA, AlOkda A, Elserafy M. The consequences of viral infection on host DNA damage response: a focus on SARS-CoVs. J Genet Eng Biotechnol. 2022;20(1):104. pmid:35829826
- 70. Kotliar IB, Lorenzen E, Schwenk JM, Hay DL, Sakmar TP. Elucidating the Interactome of G Protein-Coupled Receptors and Receptor Activity-Modifying Proteins. Pharmacol Rev. 2023;75(1):1–34. pmid:36757898
- 71. Abdel Hameid R, Cormet-Boyaka E, Kuebler WM, Uddin M, Berdiev BK. SARS-CoV-2 may hijack GPCR signaling pathways to dysregulate lung ion and fluid transport. Am J Physiol Lung Cell Mol Physiol. 2021;320(3):L430–L435. pmid:33434105
- 72. Goldberg DM, Riordan JR. Role of membranes in disease. Clin Physiol Biochem. 1986;4(5):305–36. pmid:3022980
- 73. Deng Y, Angelova A. Coronavirus-Induced Host Cubic Membranes and Lipid-Related Antiviral Therapies: A Focus on Bioactive Plasmalogens. Front Cell Dev Biol. 2021;9:630242. pmid:33791293
- 74. Dakal TC. SARS-CoV-2 attachment to host cells is possibly mediated via RGD-integrin interaction in a calcium-dependent manner and suggests pulmonary EDTA chelation therapy as a novel treatment for COVID 19. Immunobiology. 2021;226(1):152021. pmid:33232865
- 75. Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. pmid:30944313
- 76. Osterman TJ, Terry M, Miller RS. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clin Cancer Inform. 2020;4:993–1001. pmid:33136433
- 77. Aldea M, Michot JM, Danlos FX, Ribas A, Soria JC. Repurposing of anticancer drugs expands possibilities for antiviral and anti-Inflammatory discovery in COVID-19. Cancer Discov. 2021; 11:1336–1344. pmid:33846172
- 78. Kumar D T, Shaikh N, Kumar S U, Doss C GP, Zayed H. Structure-Based Virtual Screening to Identify Novel Potential Compound as an Alternative to Remdesivir to Overcome the RdRp Protein Mutations in SARS-CoV-2. Front Mol Biosci. 2021; 9 (8):645216. pmid:33898520
- 79. Wakchaure PD, Ghosh S, Ganguly B. Revealing the Inhibition Mechanism of RNA-Dependent RNA Polymerase (RdRp) of SARS-CoV-2 by Remdesivir and Nucleotide Analogues: A Molecular Dynamics Simulation Study. J Phys Chem B. 2020; 124(47):10641–10652. pmid:33190493