Computational Drug Repositioning and Elucidation of Mechanism of Action of Compounds against SARS-CoV-2
Francesco Napolitano, Gennaro Gambardella, Diego Carrella, Xin Gao, Diego di Bernardo
CComputational Drug Repositioning and Elucidationof Mechanism of Action of Compounds againstSARS-CoV-2
Francesco Napolitano , Gennaro Gambardella , Xin Gao , and Diego di Bernardo Computational Bioscience Research Center, King Abdullah University of Science and Technology (KAUST),Thuwal 23955-6900, Saudi Arabia Telethon Institute of Genetics and Medicine (TIGEM), Pozzuoli (NA) 80078, Italy and Department of Chemical, Materials and Industrial Production Engineering, University of Naples Federico II, 80125Naples, Italy. * [email protected] ABSTRACT
The COVID-19 crisis called for rapid reaction from all the fields of biomedical research. Traditional drug development involvestime consuming pipelines that conflict with the urgence of identifying effective therapies during a health and economicemergency. Drug repositioning, that is the discovery of new clinical applications for drugs already approved for differenttherapeutic contexts, could provide an effective shortcut to bring COVID-19 treatments to the bedside in a timely manner.Moreover, computational approaches can help accelerate the process even further. Here we present the application ofcomputational drug repositioning tools based on transcriptomics data to identify drugs that are potentially able to counteractSARS-CoV-2 infection, and also to provide insights on their mode of action. We believe that mucolytics and HDAC inhibitorswarrant further investigation. In addition, we found that the DNA Mismatch repair pathway is strongly modulated by drugs withexperimental in vitro activity against SARS-CoV-2 infection. Both full results and methods are publicly available.
Drug repositioning or drug repurposing aims to find a new clinical application for a drug already in use but for a differentpurpose. Usually, in drug repurposing, a known and potentially therapeutic target is selected and an experimental search forexisting compounds able to modulate the target activity is performed. Computational drug repurposing offers a complementaryapproach to prioritise compounds for experimental validation. Several different methods have been proposed in the literature and some have already been applied to SARS-CoV-2 . These can be broadly classified into methods searching for smallmolecules able to bind an active pocket starting from three-dimensional structure of the target protein and those searching forcompounds able to modulate the expression of the target mRNA . While most of the existing drug repositioning pipelinesfocus on the former approach, here we used the latter to identify FDA approved drugs able to modulate the expression of genesexpressed in the airway epithelium and that are known to interact with SARS-CoV-2 proteins. Next, we attempted a completelyagnostic approach to identify drugs reversing the gene expression profile induced by SARS-CoV-2 infection. Finally, weinvestigated potential effective mechanisms of action by identifying molecular pathways that are consistently dysregulated by aset of 24 drugs recently proposed as effective to reduce SARS-CoV-2 infection in treated Vero cells. By application of theproper computational tools, we were thus able to identify a set of drugs that should be further experimentally validated and thatmay have a potential beneficial role in COVID19 treatment, together with insights about mechanisms of action that could helpto identify the most effective targets against the infection. All the used computational tools share the same database of 6,100 drug induced gene expression profiles included in theConnectivity Map 2.0 (CMap) . CMap data were produced using Affymetrix Micrarrays, which we pre-processed to abtain theexpression values for 12,012 genes. Moreover, expression profiles induced by the same drug across different replicates andexperimental conditions were merged together into a single consensus profile. Details of the preprocessing are reported in .he Gene2drug and Drug Set Enrichment Analysis (DSEA) are two bioinformatic tools for drug-target prioritization anddrug mode of action elucidation, respectively (see next Sections). They both use a pathway-based version of the CMap. APathway-based Expression Profiles (PEP) is obtained from a Gene Expression Profiles (GEP) by iteratively applying Gene SetEnrichment Analysis (GSEA ) to the GEP for each pathway in a database such as the Gene Ontology or KEGG. In particular,both tools use all the pathway collections included in the MSigDB v6.1 , as previously published .The publicly available online implementation of Gene2drug (http://gene2drug.tigem.it) was used to identify drugs downreg-ulating ACE2 and TMPRSS2. The tool automatically identifies all the gene sets involving the input genes across all of thepathway databases. Similarly, for the DSEA analysis the online tool was used (http://dsea.tigem.it).Since the Gene2drug analysis applied to the SARS-CoV-2 human molecular interactors required customization, it wasperformed offline using the gep2pep Bioconductor package . In particular, all the human interactors of each for each of the27 SARS-CoV-2 proteins were defined as a gene-set. The 27 obtained gene sets were then added to each of the pathwaycollections included in the MSigDB v7 . Finally, the CMap GEPs were converted to PEPs based on these newly created geneset collections. In order to prioritize drugs according to their PEPs, Gene2drug analysis was performed for the 27 SARS-CoV-2related gene sets, using all the MSigDB gene sets as statistical background. Since the statistical background was different foreach gene set collection, the average scores were computed to obtain the final prioritization. The dataset of PEPs is publiclyavailable .Drugs predicted to reverse the SARS-CoV-2 induced signature were identified using the publicly avaliable MANTRA tool(http://mantra.tigem.it). RNA-seq raw counts for the identification of differentially expressed genes (DEGs) in response toSARS-CoV-2 were downloaded from GEO database (accession number GSE147507). Raw counts were first normalized usingthe edgeR tool while DEGs identified by using the limma package with its voom method in the R statistical environment 3.6.The obtained GEP was added to the MANTRA network using the "reverse node" feature. In this way, the GEP is reversed(most up-regulated genes appear as the most down-regulated and vice-versa) and the closest nodes in the network correspond todrugs inducing the opposite GEP as compared to that of SARS-CoV-2 infection.Two-dimensional structure images and annotations for the compounds reported in the tables were obtained from thePubChem database . It has been recently shown tha SARS-CoV-2 entry in host cells is mediated by the ACE2 receptor and requires priming by theTMPRSS2 protease . We thus sought to identify drugs that could potentially reduce the expression of both genes, although thebenefit of such an approach is being debated . To this end, we applied a computational drug repositioning approach namedGene2Drug (http://gene2drug.tigem.it) . This tool computes an Enrichment Score (ES) and the corresponding P-value for eachof 1309 small molecules of the Connectivity Map dataset (Broad Institute), including FDA approved drugs, based on how muchthey tend to down-regulate the genes of interest (ACE2 and TMPRSS2), as well as other genes involved in the same pathways.Results from the analysis identified 10 drugs shown in Table 1 ( p < . carbenoxolone has shown antiviral activity against the Dengue virus ; indomethacin is a non-steroidal anti-inflammatorydrug (NSAID) inhibiting Prostaglandin E2 synthase (PTGES2) with a demonstrated efficacy against SARS-CoV . Interestingly,in a recent study, SARS-CoV-2 viral NSP7 was found to interact with PTGES2 and thus the authors suggested indomethacinas potentially useful in treating patients. Here, however, indomethacin was selected by Gene2Drug because it appears to lowerexpression of genes in the ACE2 pathway, which does not include PTGES2 itself. Therefore indomethacin could be a highpriority molecule to be tested as it could have a double effect. Another NSAID is present among the drugs identified in Table1, nimesulide which could have a similar effect as indomethacin. Gene2Drug also identified nicarpidine , an angiotensininhibitor, an obvious candidate but found using a purely data-driven approach. Another potentially effective approach to the identification of therapeutic agents counteracting SARS-CoV-2 infection is totarget the host-virus interaction. In order to investigate cellular proteins interacting with viral proteins, we analyzed a set of332 protein interactions between 26 SARS-CoV-2 proteins (plus 1 mutant) and human proteins obtained by performed anaffinity purification-mass spectrometry analysis . Since each of these interactions could be key to hijack the host during thecourse of infection, we sought to prioritize existing drugs potentially interfering with them. In particular, we applied oncemore Gene2drug , this time exploiting its gene-set wise analysis capability. Specifically, for each viral protein, we generated acorresponding gene-set containing all the host protein interactors, thus obtaining 27 gene-sets. We then applied Gene2Drugto identify those compounds that are able to downregulate the expression of most of the genes across the 27 gene-sets at thesame time, which appears a particularly effective strategy given the currently limited understanding of the specific infectionmechanisms. The top 20 hits are reported in Table 2. Interestingly, one of the hits is niclosamide , an antihelminthic drug, that as found to have antiviral efficacy against SARS-CoV-2 in a recent screening using VERO cells . Other notable candidatesfound by our analysis are: fenoterol , which was also identified as potentially effective in a recent study on computational drugrepositioning ; alexidine , an antibiotic and a selective inhibitor of the mitochondrial phosphatase Ptpmt1, that was found toinhibit replication of the CytoMegaloVirus (CMV) infection ; oxytetracycline , another antibiotic that has shown antiviralactivity against Dengue virus and the ability to reduce HIV-1 RNA transcripts within extracellular vescicles ; clofibrate ,a peroxisome-proliferator activated receptor-alpha (PPAR-) agonist previously used as a cholesterol-lowering agent, whichwas found to have antiviral activity against MDV, an alphaherpesvirus that infects chickens and causes a deadly lymphoma ; flupentixol , a neuroleptic agent that was found to inhibit hepatitis C virus entry and coxsackievirus replication ; gossypol , anatural phenol derived from the cotton plant known to inhibit spermatogenesis, that was shown to have also antiviral activity ; ms-275 and trichostatin A , two histone deacetylases inhibitors (HDACi), the latter found to inhibit expression of herpessimplex virus genes ; interestingly valproic acid, another HDACi, was proposed for repositioning against SARS-CoV-2, as theviral protein NSP5 was found to interact with HDAC2 . A data-driven approach to drug repositioning is to identify compounds that are able to revert a disease-related signature . Wedeveloped the MANTRA tool exploiting the cMap dataset (Broad Institute) to generate and explore drug networks for drugrepositioning . In drug networks, nodes represent drug-induced transcriptional profiles, and edges represent similaritiesbetween them. We sought to use this approach to prioritize drugs reverting SARS-CoV-2 induced gene expression at the whole-genome level without assuming any prior knowledge of molecular mechanisms involved in the disease progression. Recently,gene expression profiles of primary human bronchial epithelial (NHBE) infected with SARS-CoV-2 have been made publiclyavailable (GSE147507 from the GEO database). We obtained the corresponding raw data and computed the differentialexpression of 15,330 genes between infected and non infected cells. Out of these, only 37 genes were significantly differentiallyexpressed (adjusted p < . at the top of the list and those most up-regulated at the bottom . We thenqueried the MANTRA drug network with this profile and looked for drugs inducing a similar transcriptional profile, whichtherefore may potentially induce a profile opposite to that of the infection. We thus found 11 significant drugs (transcriptionaldistance < .
8) listed in Table 3. One of the 11 drugs is sirolimus, also known as rapamycin, an inhibitor of mechanistic TargetOf Rapamycin Complex 1 (mTORC1), and used clinically as an immunosuppressive agent. This is an obvious candidate forthis kind of drug repositioning, where the aim is to find a drug inducing an opposite repsonse to that of the virus, as most ofthe genes overexpressed during viral infection are immune-related. Of course, this does not mean that it is clinically relevantas suppressing the immune response could be detrimental to the cells. Nevertheless, Rapamycin was also suggested in theSARS-CoV-2 interactors’ study based on the nuclecapsid interaction with the mTOR translational repressors LARP1 andon rapamycin reported in vitro activity against MERS . Other interesting candidates found by MANTRA are: ambroxol (ametabolite of bromhexine), a mucolytic agent with reported antiviral activity against influenza-virus among others andwith inhibition activity against TMPRSS2 ; corticosterone , a glucocorticoid; idoxuridine , a nucleoside analogue used toinhibit replication of DNA viruses; naltrexone , a mu-opiod receptor antagonist used for opioid and alcohol dependence, withreported antiinflammatory effect ; nordihydroguaiaretic acid an antioxidant compound found in the creosote bush (Larreatridentata) with reported antiviral activity against West Nile and Zika viruses . in vitro antiviral action against SARS-CoV-2 infec-tion. Understanding the biological mechanisms underlying a pathological condition at the molecular level can greatly help theidentification of potentially effective drugs. One way of obtaining such knowledge is the analysis of existing small moleculesthat have been observed to elicit varying levels of therapeutic efficacy through drug screening. Although the mode of action ofpositive hits from large-scale drug screening can be difficult to identify, it is likely that such hidden mechanism is shared amongmost of the hits. In order to investigate and exploit this assumption, we previously developed the Drug Set Enrichment Analysis(DSEA , http://dsea.tigem.it). DSEA is able to highlight molecular pathways that are consistently and specifically dysregulatedby most drugs in a set by analyzing drug-induced expression profiles in the CMap dataset.In order to identify compounds that are able to counteract SARS-CoV-2 infection, Jeon et al recently screened a collectionof 35 small molecules previously observed as potentially effective against SARS-CoV infection, plus 15 drugs suggestedby infectious diseases specialists. Among the 50 screened drugs, 24 showed reduction of the SARS-CoV-2 N protein levelsin infected VERO cells . Of the 24 drugs, 10 were present in the cMap dataset used by DSEA (amodiaquine, ciclosporin,desoxycortone, digoxin, loperamide, mefloquine, niclosamide, ouabain, proscillaridin, tetrandrine). We thus sought to analyze hese 10 drugs through the DSEA tool in order to understand the common mechanism of action underlying their antiviral effect.These drugs were found by DSEA to modulate targets of the transcription factors (TFs) shown in Table 4, including: NF-kB , amaster regulator of stress response;
ATF6 , a TF involved in the integrated stress response and with a known role in enhancingherpesvirus gene expression ; STAT5B , a member of the STAT family activated in response to cytokines and growth factor.DSEA also highlighted several biological processes as modulated by these drugs (Table 4), among which those related to DNAmistmatch repair pathway (MMR). Interestingly, DNA MMR pathway activation has been recently reported to be required forthe replication of the coronavirus infection bronchitis virus (IBV) . Computational methods for drug repositioning could narrow down the search for effective drugs to counteract novel epidemics.Here we showed how a number of existing tools can be used towards this aim, covering different approaches such as targetingof host proteins interacting with viral proteins, reversion of disease-induced transcriptional profiles, and mode of actioninvestigation for drugs with demonstrated in vitro activity against SARC-CoV-2 infection. Overall, we found 39 compoundsthat could be tested experimentally among which NSAIDs, antihistamines, antibiotics and antihelmintics, antipsychotics,corticosteroids, mucolytics and HDAC inhibitors (HDACi). Bromhexine, of which ambroxol is a metabolite, has been recentlyproposed as a prophylactic treatment for SARS-COV-2 patients based on its reported activity against TMPRSS2 and itsover-the-counter availability and safety levels. In our study, we find that ambroxol also induces a transcriptional profile oppositeto that induced by SARS-CoV-2 infection in vitro, adding to the evidence of a possible role of this drug in counteracting thevirus. Of interest, HDAC2 was found to interact with SARS-CoV-2 proteins and the screening on inhibitors of infection onVERO cells identified several members of the cardiac glycosides as effective . Cardiac glycosides have been recently found tohave strong HDACi activity . It will be therefore interesting to verify experimentally whether HDACi have beneficial role inreducing viral replication in host cells. Finally, by interrogating the possible mode of action of positive hits of a drug screeningin VERO cells , we identified the DNA MMR pathway as possibly involved in the mode of action of these drugs, and thereforehinting that this pathway could be involved in the host response to the viral infection. Interestingly, HDAC proteins are involvedin the modulation of the DNA MMR pathway, thus hinting that HDAC inhibtors, if proven to be effective, may act through thispathway. All the cited methods, the full lists of prioritized drugs and pathways, and all the necessary data to reproduce the results, arepublicly available . The research reported in this publication was supported by the King Abdullah University of Science and Technology (KAUST)Office of Sponsored Research (OSR) under Award No. FCC/1/1976-04, FCC/1/1976-06, FCC/1/1976-17, FCC/1/1976-18,FCC/1/1976-23, FCC/1/1976-25, FCC/1/1976-26, URF/1/3450-01, URF/1/4098-01-01, and REI/1/0018-01-01 to XG and byFondazione Telethon grant to DdB.
We thank Diego Carrella of the TIGEM Bioinformatics Core for support with the MANTRA webtool.
References Iorio, F., Saez-Rodriguez, J. & di Bernardo, D. Network based elucidation of drug response: from modulators to targets.
BMC Syst. Biol. , 139, DOI: 10.1186/1752-0509-7-139 (2013). Sirci, F., Napolitano, F. & di Bernardo, D. Computational Drug Networks: a computational approach to elucidate drugmode of action and to facilitate drug repositioning for neurodegenerative diseases.
Drug Discov. Today: Dis. Model. ,11–17, DOI: 10.1016/j.ddmod.2017.04.004 (2016). Mottini, C., Napolitano, F., Li, Z., Gao, X. & Cardone, L. Computer-aided drug repurposing for cancer therapy: Approachesand opportunities to challenge anticancer targets.
Semin. Cancer Biol.
DOI: 10.1016/j.semcancer.2019.09.023 (2019). Zhou, Y. et al.
Network-based drug repurposing for novel coronavirus 2019-nCoV/SARS-CoV-2.
Cell Discov. , DOI:10.1038/s41421-020-0153-3 (2020). Lavecchia, A. & Di Giovanni, C. Virtual Screening Strategies in Drug Discovery: A Critical Review. . . gene2drug: a computational tool for pathway-based rational drug repositioning. Bioinformatics , 1498–1505, DOI:10.1093/bioinformatics/btx800 (2018). Keenan, A. B. et al.
The Library of Integrated Network-Based Cellular Signatures NIH Program: System-Level Catalogingof Human Cells Response to Perturbations.
Cell Syst. , 13–24, DOI: 10.1016/J.CELS.2017.11.001 (2018). Lamb, J. et al.
The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease.
Sci. (New York, N.Y.) , 1929–35, DOI: 10.1126/science.1132939 (2006). Napolitano, F., Sirci, F., Carrella, D. & Di Bernardo, D. Drug-Set Enrichment Analysis: A Novel Tool to Investigate DrugMode of Action.
Bioinformatics , 235–241, DOI: 10.1093/bioinformatics/btv536 (2016). Subramanian, A. et al.
Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wideexpression profiles.
Proc. Natl. Acad. Sci. , 15545–15550 (2005).
Napolitano, F., Carrella, D., Gao, X. & Bernardo, D. gep2pep: a Bioconductor package for the creation and analysis ofpathway-based expression profiles.
Bioinformatics (2019).
Gordon, D. E. et al.
A SARS-CoV-2-Human Protein-Protein Interaction Map Reveals Drug Targets and PotentialDrug-Repurposing. bioRxiv
DOI: 10.1101/2020.03.22.002386 (2020).
Napolitano, F., Gambardella, G., Carrella, D., Gao, X. & di Bernardo, D. Computational Drug Repositioning andElucidation of Mechanism of Action of Compounds against SARS- CoV-2 - Supplementary data, DOI: 10.5281/zenodo.3756887 (2020).
Kim, S. et al.
PubChem 2019 update: Improved access to chemical data.
Nucleic Acids Res. , D1102–D1109, DOI:10.1093/nar/gky1033 (2019). Mcmurray, J. J. V., Pfeffer, M. A., Ph, D. & Solomon, S. D. Renin – Angiotensin – Aldosterone System Inhibitors inPatients with Covid-19.
The New Engl. J. Medicine
Pu, J. et al.
Antiviral activity of Carbenoxolone disodium against dengue virus infection.
J. Med. Virol. , 571–581, DOI:10.1002/jmv.24571 (2017). Amici, C. et al.
Indomethacin has a potent antiviral activity against SARS coronavirus.
Antivir. Ther. , 1021–1030(2006). Jeon, S. et al.
Identification of antiviral drug candidates against SARS-CoV-2 from FDA-approved drugs. bioRxiv
DOI:10.1101/2020.03.20.999730 (2020).
Wu, C. et al.
Analysis of therapeutic targets for SARS-CoV-2 and discovery of potential drugs by computational methods.
Acta Pharm. Sinica B
DOI: 10.1016/j.apsb.2020.02.008 (2020).
Drug Repurposing Approach Identifies Inhibitors of the Prototypic Viral Transcription Factor IE2 that Block HumanCytomegalovirus Replication.
Cell Chem. Biol. , 340–351, DOI: 10.1016/j.chembiol.2015.12.012 (2016). Rothan, H. A. et al.
Study the antiviral activity of some derivatives of tetracycline and non-steroid anti inflammatory drugstowards dengue virus.
Trop. biomedicine , 681–90 (2013). Antiretroviral Drugs Alter the Content of Extracellular Vesicles from HIV-1-Infected Cells.
Sci. Reports , 1–20, DOI:10.1038/s41598-018-25943-2 (2018). Boodhoo, N., Kamble, N., Sharif, S. & Behboudi, S. Glutaminolysis and Glycolysis Are Essential for Optimal Replicationof Marek’s Disease Virus.
J. Virol. , 1–13, DOI: 10.1128/jvi.01680-19 (2019). Pietschmann, T. Clinically Approved Ion Channel Inhibitors Close Gates for Hepatitis C Virus and Open Doors for DrugRepurposing in Infectious Viral Diseases.
J. Virol. , 1–9, DOI: 10.1128/jvi.01914-16 (2017). Zuo, J. et al.
Fluoxetine is a potent inhibitor of coxsackievirus replication.
Antimicrob. Agents Chemother. , 4838–4844,DOI: 10.1128/AAC.00983-12 (2012). Dodou, K. Investigations on gossypol: Past and present developments, DOI: 10.1517/13543784.14.11.1419 (2005).
Shapira, L., Ralph, M., Tomer, E., Cohen, S. & Kobiler, O. Histone deacetylase inhibitors reduce the number of herpessimplex virus-1 genomes initiating expression in individual cells.
Front. Microbiol. , 1970–1970, DOI: 10.3389/fmicb.2016.01970 (2016). Discovery of drug mode of action and drug repositioning from transcriptional responses.
Proc. Natl. Acad. Sci. , 14621–14626, DOI: 10.1073/pnas.1000138107 (2010). Mantra 2.0: An online collaborative resource for drug mode of action and repurposing by network analysis.
Bioinformatics
DOI: 10.1093/bioinformatics/btu058 (2014).
Blanco-Melo, D. et al.
SARS-CoV-2 launches a unique transcriptional signature from in vitro, ex vivo, and in vivo systems. bioRxiv
DOI: 10.1101/2020.03.24.004655 (2020).
Antiviral potential of ERK/MAPK and PI3K/AKT/mTOR signaling modulation for Middle East respiratory syndromecoronavirus infection as identified by temporal kinome analysis.
Antimicrob. Agents Chemother. , 1088–1099, DOI:10.1128/AAC.03659-14 (2015). Yang, B. et al.
Ambroxol suppresses influenza-virus proliferation in the mouse airway by increasing antiviral factor levels.
Eur. Respir. J. , 952–958, DOI: 10.1183/09031936.02.00253302 (2002). Paleari, D., Rossi, G. A., Nicolini, G. & Olivieri, D. Ambroxol: A multifaceted molecule with additional therapeuticpotentials in respiratory disorders of childhood, DOI: 10.1517/17460441.2011.629646 (2011).
Shen, L. W., Mao, H. J., Wu, Y. L., Tanaka, Y. & Zhang, W. TMPRSS2: A potential target for treatment of influenza virusand coronavirus infections, DOI: 10.1016/j.biochi.2017.07.016 (2017).
Patten, D. K., Schultz, B. G. & Berlau, D. J. The Safety and Efficacy of Low-Dose Naltrexone in the Management ofChronic Pain and Inflammation in Multiple Sclerosis, Fibromyalgia, Crohn’s Disease, and Other Chronic Pain Disorders.
Pharmacother. The J. Hum. Pharmacol. Drug Ther. , 382–389, DOI: 10.1002/phar.2086 (2018). Merino-Ramos, T., Jiménez De Oya, N., Saiz, J. C. & Martín-Acebes, M. A. Antiviral activity of nordihydroguaiareticacid and its derivative tetra-O-methyl nordihydroguaiaretic acid against West Nile virus and Zika virus.
Antimicrob. AgentsChemother. , DOI: 10.1128/AAC.00376-17 (2017). Hinte, F., van Anken, E., Tirosh, B. & Brune, W. Repression of viral gene expression and replication by the unfoldedprotein response effector XBP1u. eLife , DOI: 10.7554/eLife.51804 (2020). Coronavirus infection induces DNA replication stress partly through interaction of its nonstructural protein 13 with thep125 subunit of DNA polymerase δ . J. Biol. Chem. , 39546–39559, DOI: 10.1074/jbc.M111.242206 (2011).
Depfenhart, M. et al.
A SARS-CoV-2 Prophylactic and Treatment ; A Counter Argument Against The Sole Use ofChloroquine.
Am. J. Biomed. Sci. & Res. , 248–251, DOI: 10.34297/AJBSR.2020.08.001283 (2020). Raynal, N. J. et al.
Targeting calcium signaling induces epigenetic reactivation of tumor suppressor genes in cancer.
Cancer Res. , 1494–1505, DOI: 10.1158/0008-5472.CAN-14-2391 (2016). Sirci, F. et al.
Comparing structural and transcriptional drug networks reveals signatures of drug activity and toxicity intranscriptional responses. npj Syst. Biol. Appl. , 1–12, DOI: 10.1038/s41540-017-0022-3 (2017). Drug ES PV 2D structure Notes1 moracizine -1.000 0.0092 Has a role as an anti-arrhythmia drug.2 Gly-His-Lys -0.995 0.0184 Has a role as a metabolite, a chelator, a vul-nerary and a hepatoprotective agent.3 carbenoxolone -0.995 0.0184 Used for the treatment of digestive tract ul-cers.4 indometacin -0.995 0.0184 Non-steroidal anti-inflammatory drug mostcommonly used in rheumatoid arthritis, anky-losing spondylitis, osteoarthritis, acute shoul-der pains, and acute gouty arthritis.5 vitexin -0.995 0.0184 platelet aggregation inhibitor, an alpha-glucosidase inhibitor, an antineoplastic agentand a plant metabolite6 nimesulide -0.991 0.0276 Nonsteroidal anti-inflammatory drug, mod-estly selective COX-2 inhibitor.7 PNU-0293363 -0.981 0.0461 N/A N/A8 chloropyramine -0.981 0.0461 Antihistamine drug with applications to thetreatment of allergic conjunctivitis, allergicrhinitis, bronchial asthma, and other atopic(allergic) conditions.9 nicardipine -0.981 0.0461 Calcium channel blocker used in the treat-ment of hypertension and stable angina pec-toris.10 remoxipride -0.981 0.0461 Antipsychotic agent specific for dopamineD2 receptors, effective in the treatment ofschizophrenia.
Table 1.
Drugs predicted to downregulate ACE2 and TMPRSS2 genes by the Gene2drug tool ( p < . Drug mean ES mean PV 2D structure Notes1 proxymetacaine -0.536 0.0002 Benzoic acid derivative anesthetic agent, withlocal anesthetic activity.2 fenoterol -0.492 0.0008 Beta2-adrenergic agonist used in the manage-ment of reversible airway obstruction.3 0179445_0000 -0.498 0.0013 N/A N/A4 cefepime 0.473 0.0015 Antibacterial drug.5 picrotoxinin -0.473 0.0023 Has a role as a plant metabolite, a GABAantagonist and a serotonergic antagonist.6 alexidine -0.485 0.0036 Amphipathic bisbiguanide with a role as anantibacterial agent.7 oxytetracycline -0.460 0.0042 Tetracycline used for treatment of infectionscaused by a variety of Gram positive andGram negative microorganisms.8 clofibrate -0.446 0.0052 Anticholesteremic drug antilipemic drug.9 ms 275 -0.439 0.0052 Inhibitor of histone deacetylase isoform 1(HDAC1) and isoform 3 (HDAC3).10 flupentixol -0.356 0.0073 Antipsychotic neuroleptic drug, used inschizophrenia.11 n6 methyladenosine -0.421 0.0084 methylated adenine residue12 gossypol -0.398 0.0086 Orally-active polyphenolic aldehyde with po-tential antineoplastic activity by inhibitingDNA replication and inducing apoptosis.
Table 2.
Drugs predicted to downregulate SARS-CoV-2 human interactors by the Gene2drug tool ( p < .
05, top 20 reported).The mean enrichment score (ES) and the corresponding mean p-value (PV) across different gene-set backgrouns (see Methods)are reported for each drug.
Drug 2D structure Notes1 ambroxol Aromatic amine used in the treatment of respiratory diseasesassociated with viscid or excessive mucus.2 amprolium A veterinary coccidiostat that interferes with thiaminemetabolism.3 benzamil Potent blocker of the ENaC channel and also a sodium-calcium exchange blocker.4 chlorzoxazone Centrally acting muscle relaxant with sedative propertiesused for the symptomatic treatment of painful musclespasm.5 corticosterone Steroid hormone of the corticosteroid type.6 doxylamine A first generation ethanolamine with antiinflammatory,sedative and antihistamine properties.7 idoxuridine Antiviral drug and DNA synthesis inhibitor, with antiviralactivity against herpes simplex virus (HSV) and potentialradiosensitizing activities.8 meptazinol Opioid analgesic.9 naltrexone A mu-opioid receptor antagonist used to treat alcohol de-pendence. Has also a role as a central nervous systemdepressant, an environmental contaminant, a xenobiotic andan antidote to opioid poisoning.10 nordihydroguaiaretic acid Has a role as an antioxidant and a plant metabolite.11 sirolimus Antibiotic, has a role as immunosupressive, antineoplastic,antibacterial agent, mTOR inhibitor and a bacterial metabo-lite.
Table 3.
Drugs predicted to revert the transcriptional signature induced by SARS-CoV-2 infection through the MANTRA tool(distance to the inverse of the SARS-CoV-2 profile < . ranscription Factor / GO Term ES PV Transcription Factor TargetsNF- κ B (V$NFKAPPAB65_01) 0.704 1.08E-04ATF6 (V$ATF6_01) 0.675 2.35E-04OCT (V$OCT_Q6) 0.669 2.77E-04SREBP1 (TCANNTGAY_V$SREBP1_01) 0.668 2.82E-04NF- κ B (V$NFKAPPAB_01) 0.667 2.94E-04OCT (V$OCT_C) 0.665 3.07E-04STAT5B (V$STAT5B_01) 0.658 3.68E-04NF- κ B (V$NFKB_Q6) 0.656 3.89E-04SREBP1 (V$SREBP1_01) 0.648 4.75E-04PITX2 (V$PITX2_Q2) 0.646 5.05E-04Gene Ontology - Biological ProcessRNA modification -0.731 5.01E-05response to ischemia -0.707 9.91E-05DNA replication -0.682 1.95E-04amino acid transport 0.681 2.03E-04DNA strand elongation involved in DNA replication -0.679 2.12E-04intra-Golgi vesicle-mediated transport 0.678 2.16E-04porphyrin-containing compound biosynthetic process -0.678 2.16E-04chlorophyll biosynthetic process -0.674 2.46E-04photosynthesis -0.674 2.46E-04negative regulation of neuron projection development -0.660 3.54E-04Gene Ontology - Molecular FunctionMutLalpha complex binding -0.795 7.08E-06polypeptide N-acetylgalactosaminyltransferase activity -0.726 5.72E-05mismatched DNA binding -0.701 1.17E-04nucleotide binding -0.686 1.76E-04magnesium chelatase activity -0.674 2.46E-04single-stranded DNA binding -0.668 2.82E-04receptor activity 0.659 3.61E-04antigen binding 0.630 7.64E-04enzyme regulator activity -0.629 7.76E-04ribonuclease P activity -0.611 1.22E-03Gene Ontology - Cellular Componentnucleolus -0.715 7.94E-05DNA replication factor C complex -0.675 2.35E-04Gemini of coiled bodies -0.654 4.08E-04Cul3-RING ubiquitin ligase complex -0.638 6.27E-04nucleosome 0.608 1.29E-03apical plasma membrane 0.601 1.52E-03mitochondrial small ribosomal subunit -0.596 1.73E-03extracellular matrix 0.596 1.74E-03nucleolar ribonuclease P complex -0.584 2.31E-03COPI vesicle coat 0.577 2.72E-03
Table 4.