key: cord-0707203-nkpmfbud
authors: Taha, Zaid; Arulanandam, Rozanne; Maznyi, Glib; Godbout, Elena; Carter-Timofte, Madalina E.; Kurmasheva, Naziia; Reinert, Line S.; Chen, Andrew; Crupi, Mathieu J.F.; Boulton, Stephen; Laroche, Geneviève; Phan, Alexandra; Rezaei, Reza; Alluqmani, Nouf; Jirovec, Anna; Acal, Alexandra; Brown, Emily E.F.; Singaravelu, Ragunath; Petryk, Julia; Idorn, Manja; Potts, Kyle G.; Todesco, Hayley; John, Cini; Mahoney, Douglas J.; Ilkow, Carolina S.; Giguère, Patrick; Alain, Tommy; Côté, Marceline; Paludan, Søren R.; Olagnier, David; Bell, John C.; Azad, Taha; Diallo, Jean-Simon
title: Identification of FDA-approved Bifonazole as SARS-CoV-2 blocking agent following a bioreporter drug screen
date: 2022-05-06
journal: Mol Ther
DOI: 10.1016/j.ymthe.2022.04.025
sha: 32c0648ecad4c6830c2bc831f15984095aac10a7
doc_id: 707203
cord_uid: nkpmfbud

We have established a split nano-luciferase complementation assay to rapidly screen for inhibitors that interfere with binding of the receptor binding domain (RBD) of the SARS-CoV-2 Spike glycoprotein with its target receptor, Angiotensin Converting Enzyme 2 (ACE2). Following a screen of 1,200 FDA-approved compounds, we identified Bifonazole, an imidazole-based antifungal, as a competitive inhibitor of RBD-ACE2 binding. Mechanistically, Bifonazole binds ACE2 around residue K353, which consequently prevents association with RBD, thereby impacting entry and replication of Spike-pseudotyped viruses as well as native SARS-CoV-2 and its variants of concern (VOC). Intranasal administration of Bifonazole reduces lethality in K18-ACE2 mice challenged with VSV-Spike by 40%, with a similar benefit after live SARS-CoV-2 challenge. Overall, our screen has identified an antiviral that is effective against SARS-CoV-2 and VOCs such as Omicron that employ the same receptor to infect cells, and therefore has high potential to be repurposed to control, treat, or prevent COVID-19.

In order to validate the hits from our initial screen, we treated our bioreporter with the top 143 45 candidates identified at 4 µM final in quadruplicate to confirm their inhibitory activity on the 144 SARS-CoV-2 RBD-ACE2 interaction by both Methods A and B. In addition, we examined the 145 impact of our top hits on a bioreporter designed to detect the interaction of ACE2 with the larger 146 S1 domain of SARS-CoV-2 Spike. As negative controls, we also included in our validation an 147 unrelated control bioreporter (LATS biosensor), 9 as well as wildtype nanoluciferase to eliminate 148 compounds that could non-specifically inhibit the enzyme (Figure S2A-B, E) . Our results 149 confirmed that candidate Prestw-1241 (Bifonazole) led to a significant reduction in binding 150 between ACE2 and RBD with either Method A or B (Figure 2A -B, left and Z-scores in Figure   151 2C). Similar results were obtained with ACE2 and the S1 domain of Spike where a ~30 % 152 reduction was observed using Method A (cotransfected, Figure 2A-B, right) . The reduction in 153 SARS-CoV-2 bioreporter luminescence generated by some of the other candidates was much 154 lower than Prestw-1241 (< 10 % inhibition). Notably, Prestw-1405 (Ethynylestradiol) , and 155 Prestw-383 (Nicardipine) yielded a 4-6 % reduction in luminescence, specific to the 156 RBD/Spike:ACE2 interaction. While Prestw-430 (Cisapride) led to a slightly more pronounced 8 that this compound also has the potential to impede SARS-CoV viral entry (Figure S2C-D) . 166 Ethynylestradiol and Nicardipine led to a less pronounced but significant reduction in bioreporter 167 signal using both methods (3-6 % inhibition, Figure S2E ). 168 To compare head-to-head the potency of these lead compounds in impairing our SARS- 169 CoV-2 bioreporter, we treated lysates with a broader range of concentrations (2-25 µM) of newly 170 purchased drugs using Method B (Figure 2D and Figure S2F ). While we observed a dose-171 dependant reduction in bioluminescence with all four compounds, Bifonazole led to an 80 % 172 signal inhibition at 3.29 µM with IC50 of 1.3 µM. This inhibition was on par with the positive 173 control anti-RBD antibody. Nicardipine-HCl yielded a similar level of bioreporter impairment as 174 Bifonazole (IC50 = 1.9 µM), followed by Ethynylestradiol (IC50 = 2.6 µM), and Cisapride (IC50 175 = 16.5 µM), which was noticeably the least potent with minimal impact on the bioreporter. 176 To assess the effectiveness of these compounds in a biological system, we treated VeroE6 177 cells, which express detectable levels of ACE2, or VeroE6 cells stably overexpressing human 178 ACE2 and TMPRSS2 (VeroAT, Figure S2G ) cells with non-toxic concentrations (~2-50 µM) of 179 Bifonazole, Nicardipine-HCl, or Cisapride for 15 minutes. Treated cells were then infected with Bifonazole also led to a reduction in 9 efficiently than VSV-Spike-GFP (IC50 ~25-70 µM). Cisapride did not impact VSV infection at 189 all concentrations tested and was toxic at doses >20 µM. To better illustrate the specificity 190 towards Spike, the ratio of % GFP foci obtained for wtVSV over VSV-Spike was graphed across 191 all concentrations for all four compounds in both cell lines. While the curve for Bifonazole 192 trends upwards from ~20-67 µM, indicating greater inhibitory potential towards VSV-Spike, the 193 curves for Ethynylestradiol and Nicardipine trend downwards demonstrating increased inhibition 194 of wtVSV viral growth and less specificity towards Spike. 195 In sum, validation attempts using our bioreporter as well as biological assays with pseudotyped VSV present Bifonazole as a promising antiviral candidate for SARS-CoV-2 due to 197 its ability to specifically block the RBD-ACE2 interaction in model cell systems. 198 199 anti-inflammatory agent also classified as an imidazole with inhibitory potential (-21% Method 206 A and -9% Method B). The identification of multiple compounds containing an imidazole 207 pharmacophore suggests a potential mechanistic or structural relevance for RBD-ACE2 208 interaction blockade observed with the biosensor. For these reasons, we further investigated and As in Figure 2D above, we treated our bioreporter lysates with a broader range in 212 concentrations of newly purchased compounds using Method B. Reflecting initial screen results, 213 we observed superior performance by Bifonazole (IC50 of 1.6 µM), followed by Econazole 214 (IC50 of 2.3 µM), and lastly Ketoconazole (IC50 of 14.5 µM, Figure 3B and Figure S3A ). 215 To biologically assess antiviral potency, we subsequently treated both VeroE6 and VeroAT 216 cells with increasing concentrations of the three imidazoles, up to 100 µM, as in Figure 2 above. Briefly, we either pretreated VeroAT cells with Bifonazole for 15 minutes ("pre-treatment"), 233 added the drug at the same time as infection ("co-treatment") or 60 min after infection ("post-234 J o u r n a l P r e -p r o o f treatment"). VSV-Spike spread was quantified by GFP imaging (strain 1, Figure 4A ) and viral 235 output measured from supernatants 48 hours post-infection by standard plaque assay ( Figure   236 S4A). We found that pre-treatment of ACE2-expressing VeroAT cells for as little as 15 minutes 237 prior to infection with both strains of VSV-Spike led to the greatest increase in the magnitude of 238 viral blockade by Bifonazole. 239 We next compared the impact of Bifonazole on entry and spread of VSV-Spike versus 240 wtVSV. To assess entry, we treated VeroAT with increasing concentrations of Bifonazole (0, 5, 241 10, 30 uM) for 15 min prior to infection with VSV-Spike-GFP or Vero76 cells with Bifonazole 242 followed by wtVSV-GFP, with infection occuring for 1 hour at MOI 1. Virus and drug were then 243 removed, and cells were imaged for fluorescence 24 hours post-infection. GFP quantification 244 revealed that Bifonazole treatment led to a significant impairment in viral entry following a brief 245 high MOI infection with pseudotyped VSV-Spike, while wtVSV infection was largely 246 unaffected ( Figure 4B) . Upon infection at a low MOI (0.01), Bifonazole also significantly 247 impacted the spread of both VSV-Spike and wtVSV, with the impact on wtVSV occuring earlier 248 at 24 hpi due to more rapid growth kinetics ( Figure S4B) . We also observed a reduction in VSV-249 Spike plaque size, with no impact on wtVSV, following pretreatment of cells with 100 µM 250 Bifonazole and infection for 1 h prior ( Figure S4C ). 251 As a second approach to assess viral entry, VeroAT cells were pretreated with 100 µM 252 Bifonazole followed by infection with wtVSV or both strains of pseudotyped VSV-Spike. After To more conclusively probe Bifonazole's impact on VSV-Spike viral entry, we 257 interrogated its effect on fusion between Spike and ACE2 using a bimolecular fluorescence 258 complementation (BiFC) assay. 18 Briefly, HEK293T cells expressing either hACE2, 259 hACE2/TMPRSS2, or pCaggs and Zip-Venus-1 (target cells) were seded in dishes overnight and 260 then pretreated with 0, 25, 50, 100 µM Bifonazole. Next, HEK293T cells expressing Spike and 261 Zip-Venus-2 (effector cells) were added to the wells. Cells were imaged for the following 3-6 262 hours for fusion morphology and Venus complementation through Zip dimerization as a result of 263 fusion between target and effector cells ( Figure 4D and Figure S4D ). Consistent with a 264 mechanism involving the blocking of Spike:ACE2 interaction, we observed a dose-dependent 265 reduction in fusion of ACE2 and ACE2/TMPRSS2 expressing cells with Spike expressing 266 HEK293T following treatment with Bifonazole. A similar impact was obtained when using 267 Spike from SARS-CoV ( Figure S4E) . 268 To better understand if Bifonazole's effect on viral entry was dependent on binding to 269 either RBD or ACE2, we titrated one component of our bioreporter against its pretreated counterpart at 4 µM final ( Figure 4E ). Pre-treatment of ACE2, followed by addition 271 of RBD, led to complete abrogation of bioreporter complementation and luminescence, whereas 272 the opposite order of events (as performed in Method B of our screen) led to a peak inhibition of 273 only 50%. These findings suggest a preferential impact of Bifonazole on ACE2 rather than RBD. 274 Given this observation, we next found that the impairment in ACE2:Spike interaction by 275 Bifonazole is not a result of a reduction in ACE2 protein levels in our model cell lines (VeroE6, 276 VeroAT; Figure S4F Figure S5D) . Altogether, these data support the capacity for 285 Bifonazole to specifically disrupt the ACE2:RBD interaction, likely by binding to ACE2. 286 To assess whether Bifonazole may impact ACE2 through direct binding, we first 287 performed in silico molecular docking using human ACE2 and Bifonazole ( Figure 4F ). An 288 automated in silico molecular-docking approach that uses the SwissDock web server, 19 was 289 employed to predict the preferred binding pocket of Bifonazole on human ACE2. The clusters 290 were visualized with UCSF Chimera, and the best predicted models were selected based on G 291 values. The most favourable binding models are clustered at a specified pocket of ACE2. The 292 best binding model of Bifonazole (G: -7.03 kcal/mol) to ACE2 was selected for further analysis 293 by Ligplot to identify interacting residues. Molecular docking predictions suggest that binding of 294 Bifonazole does not occur within the peptidase catalytic active site, and therefore does not likely 295 impact enzymatic activity of ACE2, as demonstrated experimentally ( Figure S4G ). The 296 prediction suggests that Bifonazole binds within the ACE2 N-terminal small lobe and that the 297 interaction involves amino acid residues crucial for maintaining stable ACE2-SARS-CoV-2 298 RBD attachment. 20 Amino acid residues forming the predicted Bifonazole binding pocket of 299 ACE2 include H34, E37, K353, G354, A386, A387, Q388, and R393 ( Figure 4G ). 300 We have previously determined critical residues within ACE2 that enable RBD binding 301 using our bioreporter through mutational analysis. 11 Based on our predicted binding model of 302 J o u r n a l P r e -p r o o f mutants: H34A, E35A, G37A, K353A or G354D. Anti-RBD was added as a positive control to 305 block the interaction of the bioreporter and RBD alone was used as a negative control. While 306 Bifonazole was able to potently inhibit wild type ACE2 binding to RBD to similar levels as 307 αRBD with concentrations as low as 3.29 µM, we found that only the K353A mutant was less 308 sensitive to the inhibitory activity of Bifonazole and required ~25 uM to achieve similar signal 309 quenching as αRBD. Bifonazole was able to reduce the bioreporter signal with the other ACE2 310 mutants as efficiently as wtACE2 ( Figure 4H and S4H) . Taken together, these data suggest that 311 Bifonazole selectively binds ACE2, in an area involving residue K353, thereby blocking its 312 interaction with Spike/RBD. 313 Finally, to confirm if Bifonazole physically engages with ACE2, we treated recombinant 314 soluble ACE2 (sACE2) with increasing concentrations of urea to induce denaturation and 315 unfolding, in the presence of 25 µM Bifonazole. SYPRO orange was used to determine changes 316 in hydrophobic surfaces as a surrogate for unfolding. Our data reveal that Bifonazole stabilizes 317 sACE2 leading to a reduction of unfolding with up to 4 M of urea ( Figure 4I ). These findings 318 together suggest that ACE2 is the primary binding site for Bifonazole required to exert its SARS-319 CoV-2 blocking effects, more specifically the area involving K353. In preparation). 9-week old mice were administered vehicle, or 125 µM Bifonazole intranasally 368 based on a previous MTD study ( Figure S6F ). The next day, mice were given the same 369 treatments followed by VSV-Spike at a lethal dose of either 1E6 or 1E7 pfu intranasally in 20 µl University of Göttingen), 31 were grown in DMEM supplemented with 5% hiFCS, 2mM L-493 glutamine, 100 U/mL penicillin, and 100 µg/mL streptomycin, with 10µg/mL blasticidin 494 (Invivogen) to maintain TMPRSS2 expression. 495 All cells were incubated at 37 °C in a 5% CO2 humidified incubator, routinely tested for mycoplasma contamination by Hoechst staining and PCR (Diamed, Mississauga, Ontario, 497 Catalog # ABMG238) and used within 3-10 passages since thaw. The impact of our library on SARS receptor binding was assessed in two ways, (1) 534 with the compounds added to SmBiT-ACE2 for 50 minutes followed by the addition of equal 535 quantity LgBiT-RBD or LgBiT-S1 for another ten minutes ("mixed lysates"), or (2) with the 536 compounds added to the preformed SmBiT-ACE2+LgBiT-RBD/S1 complex for 1 hour 537 ("cotransfected"). Following incubation, nano-luciferase substrate native coelenterazine (CTZ; For validation purposes, we tested the impact of cherry-picked compounds 542 within the drug library (n = 3-4) on SARS-CoV binding to ACE2 using a variety of plasmid 543 constructs, notably the RBD or S1 domain from SARS-CoV-2 or the RBD domain from SARS-544 CoV, or human ACE2, as described previously. 7 For validation, we also generated a doxycycline 545 inducible, stable HEK293T cell line to separately produce components of our bioreporter, which 546 we later harvested from clarified cell supernatants 72 hours post-induction. were designed using Primer 3 v 4.0, and sequences are as follows: 

Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease 843

Current Strategies of Antiviral Drug Discovery for COVID-845 19

HIV entry inhibitors and their 847 potential in HIV therapy

A high-throughput 884 chemical screen identifies novel inhibitors and enhancers of anti-inflammatory functions 885 of the glucocorticoid receptor

Identification of NLRP3(PYD) Homo-Oligomerization Inhibitors with Anti-889

Oscillation of apoptosome formation through assembly of truncated Apaf-1

Sybodies targeting the 895 SARS-CoV-2 receptor-binding domain. bioRxiv

Replication-Competent Vesicular 899

Stomatitis Virus Vaccine Vector Protects against SARS-CoV-2-Mediated Pathogenesis in 900

Identification of a High-frequency Intra-903 host SARS-CoV-2 spike Variant with Enhanced Cytopathic and Fusogenic Effect

Propagation, 951 purification, and in vivo testing of oncolytic vesicular stomatitis virus strains

A high-throughput pharmacoviral 955 approach identifies novel oncolytic virus sensitizers

BRAF(K601N) melanoma: model development, drug discovery, and clinical correlation

The strategic combination of 963 trastuzumab emtansine with oncolytic rhabdoviruses leads to therapeutic synergy

Colour deconvolution: stain 966 unmixing in histological imaging

A new mathematical model for relative quantification in real-time hACE2 mice infected with severe acute respiratory syndrome coronavirus

A) SmBiT-ACE2 and LgBiT-RBD bioreporter constructs were 1001 transfected separately (bars 1-4) or co-transfected (bars 5-7) into HEK293 cells and 1002 luminescence measured from lysates

ANOVA compared to Mock with Tukey's multiple correction test; stars directly above 1005 columns are compared to control bars 1 or 2 (RBD or ACE2 alone). (B) The bioreporter screen 1006 was performed in two ways: Method A (left) involved addition of the library to the preformed 1007 RBD:ACE2 complex to identify compounds that could disrupt this interaction

Method B also included 1010 an RBD-neutralizing Sybody positive control. (C) Flowchart of Biosensor screen depicting 1011 plasmid transfection, harvest of clarified lysates, treatment of 384-well plates with Prestwick 1012 library (n = 3 replicates per drug on separate plates, 4 uM final) followed by substrate 1013 (coelenterazine) addition and luminescence readout. (D-E) Volcano plots depicting difference in 1014 average luminescent signal for each of the 1200 compounds over q-value at 5% FDR cutoff for 1015 both method A (D) and method B (E). (F) Pie chart of top 45 hits from both Method A and B 1016 grouped by therapeutic class

selected for subsequent validation, including Z-scores obtained following Method A or B

SARS-CoV-2 RBD lysates were incubated as in Method B with a dose range (0-25 µM

Anti-RBD was added as a positive control (dashed line) and RBD 1025 alone (LgBiT, dotted line) was used as a negative control

Dunnett's multiple correction test compared to Mock for each drug, orange asterisks signify that 1027 all subsequent values have the same level of significance). (E) VeroE6 or Vero

followed by infection of cells with SARS-CoV-2-Spike-pseudotyped VSV 1030 (VSV-GFP-Spike) or wild type VSV-GFP (MOI 1.0) followed by high 1031 content fluorescence imaging. The ratio of GFP foci obtained for wtVSV-GFP over VSV-Spike-1032 GFP infected cells was graphed across the concentrations tested for all four compounds in both 1033 cell lines. Tables on the right indicate concentrations which are able to reduce GFP foci counts 1034 from either wtVSV or VSV-Spike

toxic concentrations of the indicated drugs for 15 minutes, followed by infection of cells with 1045

VSV-Spike or wild type VSV-GFP (MOI 1.0) followed by high content fluorescence imaging

The ratio of GFP foci obtained for wtVSV over VSV-Spike-GFP infected cells was graphed 1047 across the concentrations tested for all compounds in both cell lines. Tables on the right indicate 1048 concentrations which are able to reduce GFP foci counts from either wtVSV or VSV-Spike 1049

Bifonazole impairs viral entry and interacts with ACE2 to block the RBD-binding 1052 site of SARS-CoV-2. (A) Bifonazole was added to VeroAT cells prior to infection with VSV-1053

at the time of infection (Co-treatment) or after infection (Post-1054 treatment) and GFP counts quantified 24 h later (n = 2, 2-way ANOVA with Dunnett's multiple 1055 comparison test over mock-treated). (B) Vero were pretreated with Bifonazole, infected with the 1056 indicated viruses at MOI 1 and GFP counts determined 24 hpi

Dunnett's multiple comparison test over mock-treated, and Sidak's test for counts within 1058 concentrations). (C) VeroAT cells were pretreated with Bifonazole followed by infection with the 1059 indicated viruses and immunofluorescence staining performed (magnification = 63X, bar = 20 1060 µm). (D) HEK293T cells expressing hACE2 or hACE2/TMPRSS2 and Zip-Venus-1 were treated 1061 with Bifonazole for 30 min after which HEK293T cells expressing SARS

2-way ANOVA with Dunnett's 1064 multiple comparison test over mock-treated for each condition). (E) Top: Bifonazole was added 1065 to RBD, followed by ACE2. Bottom: Bifonazole was added to ACE2, followed by RBD (n = 3 1066 for each). (F) The preferred binding pocket of Bifonazole on hACE2 was determined using an automated in silico molecular-docking approach (top). The most favourable models are clustered 1068 at a pocket of ACE2 which overlaps with RBD binding (bottom). (G) Interacting residues were 1069 identified using Ligplot. (H) RBD was incubated with Bifonazole followed by the addition 1070 of hACE2 or selected mutants (αRBD = positive control, RBD alone = negative control, n = 6, 1071 2-way ANOVA with Dunnett's multiple comparison test over mock treated). (I) sACE2 was 1072 treated with DMSO or Bifonazole, and urea was added at the indicated concentrations

ANOVA with Dunnett's multiple comparison test over mock treated for both groups

Sidak's test 1075 was used to compare means within the same concentration)

A-D) Vero-hTMPRSS2 were pre-treated with Bifonazole (100 M), infected with SARS-CoV-1079 2, and Spike levels detected 48 hpi by immunoblotting (A), viral RNA levels evaluated by qPCR 1080 (B), virus replication assessed by TCID50 assay (C) or SARS-CoV-2 spike protein visualized by 1081 immunostaining

Vero hTMPRSS2 cells 1083 were treated with Bifonazole before infection (Pre-entry), at the time of infection (Entry) or 1084 throughout the experiment (Full-Time) before challenge with SARS-CoV-2 and lysates 1085 immunoblotted for spike protein

Data are the means +/-SEM from n = 2-5 1088 with students t-test. (H) Mutant or native RBD bioreporters were incubated with Bifonazole, 1089 followed by ACE2 and luminescence measured