key: cord-1039321-42aye3ga
authors: Vanmechelen, Bert; Stroobants, Joren; Chiu, Winston; Schepers, Joost; Marchand, Arnaud; Chaltin, Patrick; Vermeire, Kurt; Maes, Piet
title: Identification of novel Ebola virus inhibitors using biologically contained virus
date: 2021-12-12
journal: bioRxiv
DOI: 10.1101/2021.12.09.471933
sha: 4e6b52a36243bce96083644c230e406a8b0cce11
doc_id: 1039321
cord_uid: 42aye3ga

Despite recent advancements in the development of vaccines and monoclonal antibody therapies for Ebola virus disease, treatment options remain limited. Moreover, management and containment of Ebola virus outbreaks is often hindered by the remote nature of the locations in which the outbreaks originate. Small-molecule compounds offer the advantage of being relatively cheap and easy to produce, transport and store, making them an interesting modality for the development of novel therapeutics against Ebola virus disease. Furthermore, the repurposing of small-molecule compounds, previously developed for alternative applications, can aid in reducing the time needed to bring potential therapeutics from bench to bedside. For this purpose, the Medicines for Malaria Venture provides collections of previously developed small-molecule compounds for screening against other infectious diseases. In this study, we used biologically contained Ebola virus to screen over 4,200 small-molecule drugs and drug-like compounds provided by the Medicines for Malaria Venture (i.e., the Pandemic Response Box and the COVID Box) and the Centre for Drug Design and Discovery (CD3, KU Leuven, Belgium). In addition to confirming known Ebola virus inhibitors, illustrating the validity of our screening assays, we identified eight novel selective Ebola virus inhibitors. Although the inhibitory potential of these compounds remains to be validated in vivo, they represent interesting compounds for the study of potential interventions against Ebola virus disease and might serve as a basis for the development of new therapeutics.

Ebola virus (EBOV), previously known as Zaire Ebola virus, was first discovered in 1976 in the 39 Democratic Republic of the Congo (previously called Zaire) and has since then caused several disease 40 outbreaks, predominantly in central Africa [1, 2] . Ebola virus disease (EVD) is a zoonotic hemorrhagic 41 fever that, once introduced into humans, spreads from human-to-human via direct contact [3] . The 42 incubation time varies from 2-21 days, after which symptoms develop suddenly, most frequently 43 including fever, fatigue, headache, a sore throat and muscle pain, followed by vomiting, rash and 44 diarrhea [4] . The average case fatality rate is 65%, although this varies strongly from outbreak to 45 outbreak [5] . While vaccines have been developed that have been successfully used to limit the spread 46 of EBOV outbreaks, treatment options for infected individuals are limited [6, 7] . The current 47 recommended treatment of Ebola patients is focused on early supportive care and symptomatic 48 treatment, although a recent clinical trial found early-administered single-dose injections of two 49 monoclonal antibodies, REGN-EB3 (Inmazeb) and MAb114 (Ebanga), to offer improvements in overall 50 mortality [8] . A number of small-molecule compounds have also been evaluated as potential 51 treatments for EVD, but none have proven efficacious in humans so far, potentially partly attributable 52 to difficulties in establishing scientifically sound clinical trials in the field [9, 10] . Even the nucleoside 53 analogue remdesivir, which shows strong in vitro inhibition of EBOV and other negative-stranded RNA 54 viruses, and which has been shown to efficiently protect non-human primates from EBOV challenge, 55 failed to reduce mortality in the abovementioned trial [8, 11, 12] . However, it should be noted that only 56 one dosing regimen was tested in this trial. 57

Even though the recent availability of efficacious vaccines and monoclonal antibodies have somewhat 58 improved the outlook for future outbreak management, there remains a strong need for the discovery 59 and development of more effective treatment options. However, because of the high risk these viruses 60 pose, resulting in their classification as biosafety level 4 (BSL-4) agents, research with infectious virus 61 is restricted to a limited number of BSL-4 facilities, hindering the rapid development of additional 62

Cell lines expressing VP30 were made by lentiviral transduction. Using the NEBuilder HiFi DNA 113 assembly cloning kit (NEB), EBOV VP30 was inserted into a pLenti6.3 vector (Thermo Fisher Scientific), 114 in which the CMV promoter was replaced by an SFFV promoter derived from a pHR-SFFV-dCas9- Cat. #46911 ). An internal ribosomal entry site (IRES) cassette was 116 inserted between the VP30 gene and the blasticidin resistance marker by restriction enzyme digestion 117 of the vector with SpeI-HF and SalI-HF (NEB), and digestion of a pEF1a-IRES vector 118 (www.takarabio.com, Cat. # 631970) with NheI-HF and SalI-HF (NEB). Fragments were ligated with the 119 Quick Ligation kit (NEB). 120

For lentivirus production, 50-70% confluent HEK293FT cells in T-25 flasks were transfected with 122

containing 3 µg of lentiviral EBOV vector, 5.83 µg of psPAX2 vector, 3.17 µg of pMD2.G vector and 12 124 µL PLUS reagent were prepared in serum-free Opti-MEM (Thermo Fisher Scientific). Following a five-125 minute incubation at room temperature, solutions were mixed and incubated for an additional 20 126 minutes. Cell medium was replaced by 5 mL of fresh medium, after which transfection complexes were 127 added, followed by a 21-hour incubation at 37°C. Next, sodium butyrate (10 mM) was added and cells 128 were incubated for an additional 3 hours, after which the medium was replaced with 5 mL of fresh 129 medium. Virus-containing supernatants were harvested into 15 mL conical tubes 24 hours after sodium 130 butyrate addition and centrifuged at 2000g for 15 minutes at 4°C to pellet cell debris. Transduction of 131 cell lines with the harvested lentivirus was done according to the ViraPower HiPerform T-Rex Gateway 132

Expression System (Thermo Fisher Scientific) manufacturer's protocol. Six µg/ml Polybrene (Sigma-133

Aldrich, Saint-Louis, MO, USA) was used to increase transduction efficiency. Following transduction, 134 cell medium was supplemented with 10 µg/ml blasticidin (InvivoGen) during passaging. 135

Huh-7 cells transduced with EBOV VP30 (Huh-7-EBOV-VP30) were seeded in a 6-well plate (300.000 137 cells/well). Following overnight incubation, the cells were transfected with 1000 ng EBOV antigenome, 138 1000 ng T7 polymerase, 1000 ng pCAGGS-EBOV-NP, 2000 ng pCAGGS-EBOV-L, 500 ng pCAGGS-EBOV-139 VP35 and 500 ng pCAGGS-EBOV-VP30, using 3:1 TransIT-LT1 Transfection Reagent (Mirus Bio). Twenty-140 four hours later, the medium was replaced by fresh medium. Six days post-transfection, cells were 141 trypsinized and mixed with fresh Huh-7-EBOV-VP30 cells in a T-25 flask. After three days, supernatant 142 from flasks showing widespread eGFP expression was collected and used to infect Vero E6-EBOV-VP30 143 cells seeded one day prior in a T-25 flask. After six days, the supernatant was used to infect additional 144 T-75 flasks of Vero E6-EBOV-VP30 cells, from which, after seven days, the supernatant was collected, 145 centrifuged at 17.000g for three minutes and subsequently aliquoted and stored at -80°C. 146 RNA extraction and nanopore sequencing 147 RNA was extracted from 100 µl of virus stock using a KingFisher Flex (Thermo Fisher Scientific) in 148 combination with the MagMax Viral Pathogen kit II (Thermo Fisher Scientific), according to the 149 manufacturer's instructions. RNA was converted to cDNA and amplified by Sequence-Independent 150 Single Primer Amplification as described by Greninger et al. [18] . The resulting cDNA was prepared for 151 nanopore sequencing using the SQK-LSK110 kit (Oxford Nanopore Technologies (ONT), Oxford, UK) 152 with the EXP-NBD114 barcoding expansion (ONT). The resulting library was loaded on a R9.4.1 flow 153 cell and run on a GridION. Basecalling and barcode demultiplexing was done using the ont-guppy-for-154 gridion v4.2.3. The resulting reads were mapped against the plasmid design used for generation of the 155 antigenome construct using Minimap2 v2.17-r941, followed by Medaka v1.0.1 for consensus polishing 156 and variant calling [19] . 157

Vero E6-EBOV-VP30 cells were seeded in 6-well plates. Once confluent, cell medium was removed and 159 200 µl virus dilution was added to each well. A ten-fold dilution series, covering ten dilutions (1x10^-1 160 -1x10^-10) was used, with duplicate repeats for each concentration. Plates were kept in an incubator 161 (37°C, 5% CO2), gently swirling the plates every 15 minutes. After one hour, 3 ml freshly prepared 162 agarose-medium was added to each well. Agarose-medium was made by autoclaving a 17.6 µg/ml 163 SeaKem ME agarose (Lonza, Basel, Switzerland) dilution and heating it to 65°C. Once heated, the 164 agarose was added to preheated (37°C) 2X Basal Medium Eagle without Earle's salts (Thermo Fisher 165 Scientific), supplemented with 10% FBS (Biowest), 200 mM L-glutamine, 1% NEAA, 1% Penicillin-166 Streptomycin, 1% Gentamicin and 0.2% Fungizone (all Thermo Fisher Scientific), in a 1:2 ratio. After 167 cooling down to room temperature, plates were moved to an incubator for five days. Read-out was 168 performed by counting the amount of eGFP+-cell clusters. 169

Compounds, spotted in 96-well plates at 2 or 10 mM, were gifted to us by MMV and CD3. Intermediary 171 compound dilutions were made in complete cell medium directly before adding the compound to 96-172 well plates (CELLSTAR, Greiner-Bio, Vilvoorde, Belgium) in which Vero E6-EBOV-VP30 cells had been 173 seeded one day prior at 20,000 cells/well. For the MMV compound set, a dilution series of four 174 concentrations was tested for each compound, starting at 50 µM and diluting four-fold each time, 175 allowing twenty-two compounds to be tested per plate. For the CD3 set, two concentrations (1 and 10 176 µM) were tested for each compound on separate plates. Following compound addition, 200 plaque 177 forming units (PFU) virus dilution was added to each well. Medium without virus was added to the 178 negative controls. Six days post-infection, cell medium was replaced by fresh medium supplemented 179 with 5 µM Hoechst 33342 nucleic acid stain (Thermo Fisher Scientific) as a background stain for high-180 content imaging analysis. Imaging and image analysis was done using an Arrayscan XTI (Thermo Fisher 181 Scientific) and a custom Cellomics SpotDetector BioApplication protocol, as described previously [20] . 182 Further data analysis was done using Genedata Screener V17.05-Standard. GraphPad Prism v8.2.0 was 183 used for graph plotting. 184

To confirm compound activity observed in the initial screening assays, additional compound was 186 acquired. For the MMV compounds, fresh DMSO stocks were prepared from powder provided by 187 Evotec (Hamburg, Germany), while the CD3 compounds were provided as DMSO solutions. Hit 188 confirmation using these fresh stocks was done by testing each compound in triplicate in Vero E6-189 EBOV-VP30 cells over a two-fold dilution series of nine dilutions, starting at 100 µM. Cell and virus 190 quantities were identical to the ones used in the screening assay and plate handling procedures and 191 data read-out were performed as described above. In addition to Vero E6-EBOV-VP30 cells, a subset 192 of compounds was also tested in Huh-7-EBOV-VP30 cells. In this cell line, a two-fold dilution series of 193 eight dilutions, starting at 50 µM, was used. To allow adequate high-content imaging, Huh-7-EBOV-194 VP30 cells were seeded at 10,000 cells/well and infected with 0.2 PFU EBOV-VP30-eGFP per cell. 195

Assay read-out was performed four days post-infection. Other plate handling and data processing 196 procedures were performed as described above. 197

To set up a screening platform for EBOV inhibitors using infectious virus without requiring access to a 200 BSL-4 facility, we created a biologically contained EBOV system similar to the one described by 

Once validated for safety, our EBOV-VP30-eGFP assay was subsequently used to screen >4,000 233 compounds for their potential as EBOV inhibitors. 560 compounds were obtained from MMV as part 234 of the Pandemic Response Box and the Covid Box, while an additional 3,681 compounds were provided 235 by CD3. High-content imaging was used to simultaneously assess antiviral activity and toxicity. For 236 practical reasons, the initial screens of the MMV and CD3 libraries were performed separately, using 237 two different plate layouts (Figure 2 ). Both the Pandemic Response Box (400 compounds) and the 238 COVID Box (160 compounds All compounds were pre-spotted in 96-well plates at a stock concentration of 10 mM. Initial screening 250 of these compounds was performed at working concentrations of 1 and 10 µM. The global Z'-factor 251 for this screen was 0.87. Sixty-one compounds (1.7%) that showed more than 80% inhibition of eGFP-252 expression compared to the control while simultaneously showing less than 20% reduction in cell 253 number were selected as preliminary hits (Figure 3 ). Comparably to the preliminary hits of the MMV 254 screen, these 61 compounds were retested in duplicate over a wider concentration range to rule out 255 false positives. Twenty-one compounds that showed an SI >5 and an IC50 <15 µM, in this confirmation 256 assay, were retained for further analysis. 257

New aliquots of the antivirals in the form of compound powder (MMV) or DMSO solution (CD3) were 259 acquired to confirm the activity and selectivity of the eleven MMV and twenty-one CD3 compounds 260 that had demonstrated selective inhibition of EBOV replication. Triplicate testing of the new compound 261 stocks over a range of nine concentrations was used to accurately determine IC50, CC50 and SI values 262 for each compound (Table 1) . Twenty-seven of the thirty-two preliminary hits were confirmed to 263 inhibit EBOV-VP30-eGFP replication, although several compounds seemed to be only moderately 264 selective. The confirmed hits include two duplicates, itraconazole and retapamulin, which were 265 present in both the MMV and CD3 compound libraries. Four of the five MMV compounds that failed 266 to be confirmed (pimozide, apremilast, dabrafenib and fluconazole) were also present in the CD3 267 compound library but had not been picked up as hits in the CD3 screen, confirming their lack of 268 potency. Comparably, one of the weaker hits of the CD3 compounds (benztropine) was also present in 269 the MMV library but had failed to meet the criteria for initial hit selection. The fourteen compounds 270 that showed the highest SI (all >7) in Vero E6 cells, including the duplicates of itraconazole and 271 retapamulin, were retested using Huh-7-EBOV-VP30 cells (Table 1 Despite recent advancements in the search for EBOV therapeutics, no small-molecule compounds are 280 licensed to treat EBOV infections [9] . However, small-molecule compounds are generally easy and 281 cheap to produce, transport and store, making them interesting candidates for the treatment of 282 patients, especially in remote locations [22, 23] . Additionally, because many small-compound libraries 283 have already been developed for a variety of applications, the repurposing of existing compounds 284 forms an interesting research avenue for the rapid identification and implementation of potential 285 antivirals. In this study, we optimized a biologically contained EBOV assay and used it to screen the 286 MMV Pandemic Response box and COVID Box, two such libraries of small-molecule compounds with 287 drug-like characteristics that have been independently developed for the antimicrobial treatment of 288 various infections [16, 17] . Additionally, we screened a large in-house repurposing collection, provided 289 by CD3. In total, 4,241 compounds from these three libraries were tested for their anti-EBOV potential. 290

Thirty unique compounds were retained after the initial screens, twelve of which were ultimately 291 found to profoundly inhibit EBOV-VP30-eGFP replication with an SI >7 in at least one of the cell lines 292 tested. 293

Four of the most active compounds, remdesivir, apilimod, diphyllin and dalbavancin, were previously 294 identified as EBOV inhibitors [24, 25] . Remdesivir is an adenosine analogue monophosphoramidate 295 prodrug that is known to inhibit the polymerase activity of many mononegaviruses, including members 296 of the families Pneumoviridae, Paramyxoviridae and Filoviridae [11] . It is also a known inhibitor of 297 coronaviruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [26, 27] . In 298 addition to showing excellent in vitro activity against filoviruses, remdesivir has been reported to be 299 an effective post-exposure treatment for EBOV infection in vivo, as it was found to ameliorate disease 300 symptoms and improve survival rates in a non-human primate model [12] . However, as mentioned 301 above, despite showing excellent in vitro and in vivo anti-EBOV potential, remdesivir failed to improve 302 the survival rates of EVD patients during a clinical trial carried out during the 2018-2020 EBOV outbreak 303 in the Democratic Republic of the Congo [8]. Furthermore, also for the treatment of COVID-19, caused 304 by SARS-CoV-2, the benefit of remdesivir is heavily contested [28] [29] [30] [31] . Unlike remdesivir, the potential 305 of apilimod as an EBOV inhibitor has not yet been evaluated in vivo. Conversely, in vitro, apilimod has 306 been shown to potently inhibit EBOV replication in Vero E6, Huh-7 and primary human macrophage 307 cells [24] . Apilimod was first identified as an inhibitor of Toll-like receptor-mediate interleuking-12/-23 308 signaling and has been evaluated as a potential anti-inflammatory drug for the treatment of Crohn's 309 disease, rheumatoid arthritis and psoriasis, albeit without significant clinical success [32] [33] [34] . Later 310 research showed that apilimod works by inhibiting phosphatidylinositol-3-phosphate 5-kinase 311 (PIKfyve), a lipid kinase involved in maintaining endosome morphology and ensuring endosome 312 maturation [35] . By inhibiting PIKfyve and preventing endosome maturations, apilimod is believed to 313 block EBOV entry, as endosome maturation is a crucial process needed to allow the EBOV GP to be 314 cleaved by cathepsins L and B, exposing the GP receptor-binding domain and enabling binding of the 315 EBOV entry receptor NPC1 [36, 37] . Because it is well tolerated in humans and targets a rather unique 316 part of the virus life cycle, apilimod is an interesting candidate to be part of combination therapies for 317 the treatment of EBOV, but future research will first need to confirm its in vivo efficacy and clinical 318 benefit. Comparable to apilimod, diphyllin and its derivatives interfere with EBOV entry by preventing 319 endosome maturation [38] . Diphyllin is an inhibitor of vacuolar-type ATPase (V-ATPase), which 320 hydrolyses adenosine triphosphate and simultaneously transports protons across cellular membranes, 321 resulting in endosome acidification [39] . Lastly, dalbavancin, a glycopeptide antibiotic primarily used 322 for the treatment of skin and soft tissue infections, is known to inhibit cellular entry of many different 323 viruses, including echovirus 1, severe acute respiratory syndrome coronavirus, Middle East respiratory 324 syndrome-related coronavirus and EBOV [40] [41] [42] . Unlike apilimod and diphyllin, which target 325 endosome maturation, dalbavancin prevents virus entry by direct inhibition of cathepsin L [42] . In the 326 case of EBOV, this results in the GP being kept in its pre-cleaved state, rendering it unable to bind NPC1 327 [43] . 328

In the MMV compound library, three additional compounds, itraconazole, retapamulin and the yet-329 unnamed MMV1782214 (CHEMBL93139), were found to selectively inhibit EBOV-VP30-eGFP 330 replication. The former two compounds were also present and identified as hits in the CD3 library. 331

Itraconazole, the top hit in both compound libraries, is a triazole derivate that works as a broad-332 spectrum antifungal agent [44] . Although it can cause mild gastrointestinal disturbances, cardiotoxicity 333 and hepatotoxicity, itraconazole is generally well tolerated and it is used for the treatment of systemic 334 (histoplasmosis, aspergillosis, blastomycosis) and superficial (onychomycosis) fungal infections [45] . In 335

Vero E6 cells, itraconazole shows anti-EBOV activity in the sub-micromolar range without apparent 336 toxicity. Conversely, in a hepatocyte cell line (Huh-7) , itraconazole shows significant toxicity and no 337 selective inhibition of EBOV replication. In addition to its antifungal properties, recent research has 338 identified itraconazole as a potential cancer treatment and as an inhibitor of several viruses, including 339 influenza virus, enteroviruses and coronaviruses [46] [47] [48] [49] [50] . The mechanisms for these newly discovered 340 functions appear unrelated to its antifungal properties. Although not fully understood yet, itraconazole 341 affects angiogenesis through indirect inhibition of the mechanistic Target of Rapamycin (mTOR), an 342 important oncogene that regulates cell growth and proliferation [51] . One of the mechanisms behind 343 the inhibition of mTOR activity by itraconazole appears to be dysregulation of cholesterol trafficking, 344 which results in endosomal cholesterol accumulation [51, 52] . This disturbance of cellular cholesterol 345 homeostasis in itself is believed to contribute to the antiviral properties of itraconazole, by preventing 346 virus escape from the endosome and simultaneously interfering with virus egress [49] . Interestingly, 347 the mechanism behind impaired cholesterol trafficking has been shown to be a direct interaction 348 between the cholesterol transporter NPC1 and itraconazole [51, 53] . NPC1 is known to be an 349 indispensable entry receptor for EBOV and the interaction between NPC1 and the EBOV GP in mature 350 endolysosomes is necessary for the release of the virion contents into the cellular cytoplasm [54, 55] . 351

Possibly, the interaction between itraconazole and the EBOV entry receptor NPC1 further potentiates 352 the antiviral activity of this compound in filovirus infections, although further research is necessary to 353 fully elucidate the interaction of itraconazole with both host and viral factors. The second hit 354 compound that was present in both libraries is retapamulin, a derivative of pleuromutilin approved for 355 use in humans [56] . This compound showed comparable selectivity in both Vero E6 and Huh-7 cells, 356

although it was roughly one order of magnitude more potent in the latter. Unlike the aforementioned 357 compounds, retapamulin has not yet been reported to possess antiviral activity and its sole known 358 function is the inhibition of the bacterial ribosome complex [57] . While retapamulin is only used for 359 topical application, other pleuromutilins can be used systemically and might be of use for EVD 360 treatment, although their antiviral mechanism of action would first need to be elucidated [58] . Lastly, 361 MMV1782214 (CHEMBL93139) showed excellent selectivity in Vero E6 cells but less so in Huh-7 cells. 362

This compound is a 1,3,4-trisubstituted pyrrolidine derivative that was developed as a C-C chemokine 363 receptor type 5 (CCR5) antagonist to be used for the treatment of human immunodeficiency virus 364 infections [59] . Other pyrrolidine derivatives have recently been shown to inhibit EBOV replication, 365 although the mechanism through which this inhibition is achieved remains to be determined [60] . 366

In the CD3 library, more than twenty compounds displayed selective anti-EBOV activity, although for 367 most compounds this selectivity was modest (SI 3-7). Aside from the aforementioned known EBOV 368 inhibitors and compounds also present in the MMV library, five compounds were found to show strong 369 selectivity towards virus inhibition: z-FA-FMK, Evans blue, UNC1999, benproperine and doramapimod. 370 Z-FA-FMK is a potent inhibitor of cysteine proteases, including cathepsin B and L [61]. As mentioned 371 above, these cathepsins are needed to cleave the EBOV GP before it can interact with NPC1. In both 372

Vero E6 and Huh-7 cells, z-FA-FMK showed only limited toxicity while inhibiting EBOV-VP30-eGFP in 373 the low-micromolar or even sub-micromolar range, presumably by preventing virus entry, making it 374 an interesting putative EBOV therapeutic. However, because of z-FA-FMK's broad and potent 375 inhibitory effect on cysteine proteases and its known function as an immunosuppressant that can 376

interfere with T-cell proliferation, detailed in vivo validation of its safety and clinical benefit would be 377 needed before it could be considered for use in humans [62] . For Evans blue, UNC1999 and 378 benproperine, the mechanism through which they might inhibit EBOV replication is less clear. Evans 379 blue or T-1824 is an azo dye that is known for its dark blue color and high affinity for albumin, and it is 380 primarily used to stain cells or tissues in a laboratory setting [63] . However, it is also known to bind 381 several glutamate receptors and transporters, and has been shown to inhibit hepatitis B virus 382 replication [64, 65] . This latter effect is in part achieved by stimulation of Ca 2+ channels by Evans blue, 383 resulting in reduced cytosolic levels of Ca 2+ . A similar mechanism might contribute to the anti-EBOV 384 effect of Evans blue, as several processes in the EBOV life cycle, including fusion and budding, are 385 affected by cytosolic Ca 2+ concentrations [66] [67] [68] . UNC1999 is an inhibitor of the lysine 386 methyltransferases enhancer of zeste homolog 1/2 (EZH1/2) [69] . It has primarily been studied as a 387 potential anti-cancer drug because of its potential to alter the differential expression of host genes 388 through epigenetic regulation [70] . Likewise, the mechanism through which UNC1999 inhibits EBOV 389 replication might be that it counteracts the pro-viral manipulation of host factor pathways during EBOV 390 infection [71] . Like UNC1999, benproperine, a clinically used antitussive drug, has also been evaluated 391 as a potential anti-cancer drug. It has been shown to inhibit Actin-related protein 2/3 complex subunit 392 2, which plays a role in actin polymerization [72] . The transport of EBOV nucleocapsids to the cellular 393 membrane prior to virion formation is dependent on actin polymerization, providing a potential 394 explanation for the anti-EBOV mechanism of benproperine [73] . CD3 compounds. Each compound was tested at 1 and 10 µM. Only compounds that showed less than 632 20% reduction in cell survival are shown. Shown on the Y-axis is the relative inhibition of eGFP-633 expression compared to the non-compound control. Compounds that reduced eGFP-expression by 634 >80% in either concentration were selected as preliminary hits. 635 

2. World Health Organization. Ebola haemorrhagic fever in Zaire

The New England journal of medicine

Ebola virus disease

How severe and prevalent are Ebola and Marburg 422 viruses? A systematic review and meta-analysis of the case fatality rates and seroprevalence. BMC 423 infectious diseases

Preliminary results on the efficacy of rVSV-ZEBOV-GP Ebola vaccine 425 using the ring vaccination strategy in the control of an Ebola outbreak in the Democratic Republic of 426 the Congo: an example of integration of research into epidemic response

Efficacy and effectiveness of an rVSV-428 vectored vaccine in preventing Ebola virus disease: final results from the Guinea ring vaccination, open-429 label, cluster-randomised trial (Ebola Ca Suffit!)

Controlled Trial of Ebola Virus Disease

The New England journal of medicine

Current status of small molecule drug development for Ebola virus and 433 other filoviruses. Current opinion in virology

Experimental Treatment with Favipiravir for Ebola 435

Single-Arm Proof-of-Concept Trial in Guinea. 436 PLoS medicine

GS-5734 and its parent nucleoside analog inhibit Filo-, Pneumo-438 , and Paramyxoviruses. Scientific reports

Therapeutic efficacy of the small molecule GS-5734 against 440 Ebola virus in rhesus monkeys

et al. Biosafety 442 in microbiological and biomedical laboratories

National Institutes 444 of Health

Minigenomes, transcription and replication 446 competent virus-like particles and beyond: reverse genetics systems for filoviruses and other negative 447 stranded hemorrhagic fever viruses

Medicines for Malaria Venture. The Pandemic Response Box

Rapid metagenomic identification of viral 456 pathogens in clinical samples by real-time nanopore sequencing analysis. Genome medicine

Minimap2: pairwise alignment for nucleotide sequences

Advancing Marburg virus antiviral screening: 461 Optimization of a novel T7 polymerase-independent minigenome system

Gurevich EV, Gurevich VV. Therapeutic potential of small molecules and engineered proteins. 466 Handbook of experimental pharmacology

Macromolecular versus small-molecule therapeutics: drug discovery, 468 development and clinical considerations

The phosphatidylinositol-3-phosphate 5-kinase inhibitor 470 apilimod blocks filoviral entry and infection. PLoS neglected tropical diseases

Discovery and Synthesis of a Phosphoramidate Prodrug of a 473

Adenine C-Nucleoside (GS-5734) for the Treatment of Ebola and 474 Emerging Viruses

Remdesivir and chloroquine effectively inhibit the recently 476 emerged novel coronavirus (2019-nCoV) in vitro. Cell research

Broad-spectrum antiviral GS-5734 inhibits both 478 epidemic and zoonotic coronaviruses. Science translational medicine

Small-molecule antiviral agents in ongoing clinical 480 trials for COVID-19. Current drug targets

Clinical infectious diseases : an official publication of the Infectious Diseases 484 Society of America. 2021 Oct 1. 485 30. Consortium WHOST, Pan H, Peto R, et al. Repurposed Antiviral Drugs for Covid-19 -Interim 486 WHO Solidarity Trial Results. The New England journal of medicine

Remdesivir for the Treatment of Covid-19 -Final 488

The New England journal of medicine

Apilimod inhibits the production of IL-12 and IL-23 490 and reduces dendritic cell infiltration in psoriasis

Selective abrogation of Th1 response by STA-5326

Billich A. Drug evaluation: apilimod, an oral IL-12/IL-23 inhibitor for the treatment of 494 autoimmune diseases and common variable immunodeficiency. IDrugs: the investigational drugs 495 journal

Ebola virus and severe acute respiratory 500 syndrome coronavirus display late cell entry kinetics: evidence that transport to NPC1+ endolysosomes 501 is a rate-defining step

Filovirus entry: a novelty in the viral fusion world

Phenotypic Prioritization of Diphyllin Derivatives 505

That Block Filoviral Cell Entry by Vacuolar (H(+) )-ATPase Inhibition

Recent Insights into the Structure, Regulation, and 508 Function of the V-ATPases. Trends in biochemical sciences

Novel activities of safe-in-human broad-spectrum 510 antiviral agents

The role of dalbavancin in skin and soft tissue 512 infections. Current opinion in infectious diseases

Glycopeptide Antibiotics Potently Inhibit Cathepsin L in the Late 514

Endosome/Lysosome and Block the Entry of Ebola Virus, Middle East Respiratory Syndrome 515

The 516 Journal of biological chemistry

Biochemical and structural characterization of 518 cathepsin L-processed Ebola virus glycoprotein: implications for viral entry and immunogenicity. 519 Journal of virology

Treasure Island (FL)

Hedgehog pathway": a potential target of itraconazole in the 524 treatment of cancer

Antiviral activity of itraconazole against type I feline 526 coronavirus infection. Veterinary research

The clinically licensed antifungal drug itraconazole inhibits 528 influenza virus in vitro and in vivo. Emerging microbes & infections

Late Endosomal/Lysosomal Cholesterol Accumulation Is a 530 Host Cell-Protective Mechanism Inhibiting Endosomal Escape of Influenza A Virus. mBio

Antiviral Activity of Itraconazole against Echovirus 30 Infection 533 In Vitro. Osong public health and research perspectives

Simultaneous Targeting of NPC1 and VDAC1 by Itraconazole 535 Leads to Synergistic Inhibition of mTOR Signaling and Angiogenesis

Cholesterol trafficking is required for mTOR activation in endothelial 538 cells

Structural basis for itraconazole-mediated NPC1 inhibition

Cholesterol Transfer and Ebola Infection

Retapamulin: a new topical antibiotic for the treatment of uncomplicated 547 skin infections. Drugs of today

British journal of clinical pharmacology

The pleuromutilin antibiotics: a new class for human use. Current opinion 551 in investigational drugs

1,3,4-Trisubstituted pyrrolidine CCR5 receptor antagonists

Part 2: lead optimization affording selective, orally bioavailable compounds with potent anti-HIV 554 activity. Bioorganic & medicinal chemistry letters

Peptidyl fluoromethyl ketones as inhibitors of cathepsin 558 B. Implication for treatment of rheumatoid arthritis. Biochemical pharmacology

The cathepsin B inhibitor, z-FA-FMK, inhibits human T 561 cell proliferation in vitro and modulates host response to pneumococcal infection in vivo

Evans Blue Dye: A Revisit of Its Applications in Biomedicine. Contrast 564 media & molecular imaging

Evans blue antagonizes both alpha-amino-3-hydroxy-5-methyl-4-568 isoxazolepropionate and kainate receptors and modulates receptor desensitization. Molecular 569 pharmacology

Conformational changes in the Ebola virus membrane fusion 571 machine induced by pH, Ca2+, and receptor binding

Selective inhibition of Ebola entry with selective estrogen receptor 573 modulators by disrupting the endolysosomal calcium. Scientific reports

Influence of calcium/calmodulin on budding of Ebola VLPs: implications for 575 the involvement of the Ras/Raf/MEK/ERK pathway. Virus genes

An orally bioavailable chemical probe of the Lysine 577

EZH2 inhibition in multiple myeloma downregulates 579 myeloma associated oncogenes and upregulates microRNAs with potential tumor suppressor 580 functions

Single-Cell Profiling of Ebola Virus Disease In Vivo Reveals Viral 582 and Host Dynamics

Benproperine, an ARPC2 inhibitor, suppresses cancer cell 584 migration and tumor metastasis

Ebola virus proteins NP, VP35, and VP24 are essential 586 and sufficient to mediate nucleocapsid transport

Pyrazole urea-based inhibitors of p38 MAP kinase: from 589 lead compound to clinical candidate

The Ebolavirus VP24 protein blocks phosphorylation of 591 p38 mitogen-activated protein kinase. The Journal of infectious diseases

Pyridinyl imidazole inhibitors of p38 MAP kinase 594 impair viral entry and reduce cytokine induction by Zaire ebolavirus in human dendritic cells. Antiviral 595 research