Identification and characterization of a non-peptidic cyclophilin ligand with antiviral activity against feline and porcine α-coronaviruses

24 25 Coronaviruses (CoVs) are emerging pathogens that have been extensively studied 26 over the last 20 years and can cause acute respiratory disease s in humans, as exemplified 27 by the SARS -CoV-2 pandemic. Coronaviruses are also known for their importance in 28 veterinary medicine, being responsible for severe pathologies in pets and livestock. These 29 include Feline Infectious Peritonitis Virus (FIPV), which causes a fatal disease in cats. In 30 livestock, porcine coronaviruses such as Transmissible Gastroenteritis Virus (TGEV) and 31 Porcine Epidemic Diarrhoea Virus (PEDV) are the causative agents of an acute enteric 32 disease in piglets with a high mortality rate and a significant impact on the pork industry. 33 In addition, animal coronaviruses may represent a zoonotic reservoir. Therefore, efficient 34 antiviral strategies are required to inhibit the multiplication of coronaviruses infecting 35 various animal species. Here , we synthesized 20 small -molecule ligands that target 36 cyclophilins, a family of cellular chaperons hijacked by several viruses including CoVs. 37 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 2 We screened their antiviral activity against feline and porcine alpha-CoVs, and identified 38 a compound, F83233, as a potent inhibitor of FIPV, TGEV and PEDV replication at 39 micromolar concentrations that was effective in feline, porcine and simian cells. As 40 cyclophilins are highly conserved among mammals, F83233 could be a promising antiviral 41 to treat different animal and zoonotic coronaviruses. 42 43

animal coronaviruses; FIPV; TGEV; PEDV; cyclophilins; antivirals 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 3

71 72 Coronaviruses (CoVs) are enveloped viruses with a single-stranded RNA genome 73 of positive polarity. Inside the Coronaviridae family, the sub -family Orthocoronavirinae is 74 composed of four genera: Alpha (α)-, Beta (β)-, Gamma (γ)- and Delta (δ)-coronavirus. α-CoVs 75 and β-CoVs mainly infect mammals while γ-CoVs and δ-CoVs mostly infect birds with a 76 few exceptions of mammals1. CoVs have a huge impact on public health: since 2002, they 77 have been recognized as emerging and important pathogens in humans. Indeed, three 78 zoonotic CoVs belonging to the β -CoV genus inducing severe pulmonary diseases have 79 emerged in humans: SARS -CoV in 2002, MERS -CoV in 2012 and SARS -CoV-2 in 2019 2–4. 80 The emergence of these viruses, especially SARS -CoV-2, the etiological agent of the 81 COVID-19, has led to a fantastic increase in knowledge about β -CoVs biology and 82 pathophysiology. 83 In veterinary medicine also, CoVs are well known to trigger the development of 84 various diseases, exhibiting sometimes a complex pathophysiology with a tropism which 85 is not restricted to the respiratory tract5,6. The α-CoV genus contains an important number 86 of pathogens that infect domestic and livestock animals, and effective treatments or 87 vaccines are often missing for these viruses. Among them , FIPV (Feline Infectious 88 Peritonitis Virus) causes a systematically fatal disease in cats, feline infectious perit onitis 89 (FIP)7. FIP is considered one of the leading causes of death in communal cat groups with 90 a 100% lethality rate, although treatment with GS -441524 now constitutes an excellent 91 therapeutic strategy8,9. Feline CoVs (FCoV) are classified into two biotypes: Feline enteric 92 CoV (FeCV) and Feline Infectious Peritonitis Virus (FIPV). FeCV is endemic in cats and 93 avirulent, i nducing mild or subclinical digestive symptoms. FIP disease is caused by 94 virulent FIPV strains 7. While FeCV has a strict intestinal tropism, FIPV is able to infect 95 monocytes and macrophages10, allowing the virus to spread in various organs leading to 96 a systemic infection. Beyond biotypes, FCoV is also s ubdivided, based on serological 97 responses, into two serotypes: FCoV-I, the source of most natural infection amongst cats11, 98 and FCoV-II resulting from recombination between FCoV -I and the canine coronavirus 99 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 4 CCoV-II12. In contrast to FCoV-I, FCoV-II replicates readily in feline cell lines such as CRFK 100 cells (Crandell-Reed Feline Kidney). 101 Several α-CoVs are also threatening farms and livestock animals. For example, six 102 CoVs can infect pigs, among them four induce a clinically undistinguishable severe 103 digestive disease, with case -fatality ratios of 80 -100% in piglets that are less than 10 -104 days13,14. Transmissible Gastro enteritis Virus (TGEV) is the prototype porcine CoV that 105 shares an important sequence homology with FIPV. Interestingly, TGEV also exhibits a 106 dual tropism in vivo, for intestinal and respiratory epithelia5. In the 1980’s, a deletion in the 107 N-terminal domain of the Spike protein has led to the global emergence of a TGEV variant 108 called PRCV (Porcine Respiratory Corona virus) that lost the enteric tropism and 109 virulence5. As antibodies elicited against PRCV protect against TGEV, the latter is now 110 well controlled. Porcine Epidemic Diarrhoea Virus (PEDV) has emerged in the 1970’s: 111 ancestral PEDV strains have been contained by vaccines, but PEDV is re -emerging since 112 the 2010’s with the presence of new strai ns that render the vaccines less effective15–17. Re-113 emergence of PEDV in the United States in 2013 led to the loss of 10% of the pig herd with 114 huge economi c impact 18–20. These PEDV strains now globally circulate 21. Other porcine 115 CoVs such as SADS -CoV (Swine Acute Diarrhoea Syndrome coronavirus) and PDCoV 116 (porcine Deltacoronavirus) are emerging viruses with a documented zoonotic potential22,23. 117 Considering the impressive capacity of CoVs to jump across the species barrier 24 and the 118 regular and severe resulting epidemics, it is of utmost importance to find, not only 119 treatment for a specific pathogenic animal CoV, but also antiviral strategies that could be 120 applied to a broad spectrum of CoVs, therefore preventing future emerging viruses. 121 All along the CoV life cycle 25, viral proteins an d viral RNAs are involved in 122 interactions with host cellular factors that can facilitate or restrict the replication of CoVs26–123 30. Among them are cyclophilins, conserved cellular proteins present in both prokaryotes 124 and eukaryotes 31. Cyclophilins share a common peptidyl -prolyl cis/trans isomerase 125 domain (PPIase) 31 which catalyzes the interconversion of proline configuration 32. 126 Cyclophilins are known to play critical roles in the replication of different viruses 33, 127 including DNA viruses 34 as well as negative - and positive-sense RNA viruses 35,36. They 128 have also been shown to play a role in CoV replication, although their precise involvement 129 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 5 along the life cycle is unclear 37. Cyclosporin A (CsA), a macrocyclic inhibitor of 130 cyclophilins, as well as its non-immunosuppressive derivatives such as alisporivir (ALV) 131 inhibit CoV replication of different genera 38–40. Gene knock-out or siRNA of cyclophilins 132 render cells resistant to several CoV infections41,42. Cyclophilins have also been identified 133 as interactors of several viral proteins of CoVs including SARS-CoV43 and HCoV-229E44. 134 We present here the characterization of the antiviral effect of small -molecule 135 cyclophilin inhibitors 45 (“SMCypI”) against a feline coronavirus of major veterinary 136 interest, FIPV, and two enteric porcine CoVs, TGEV and PEDV. Following the synthesis 137 and the screening of 20 SMCypI, we identified a compound referred as F83233 as a potent 138 anti-FIPV compound in feline cells. This molecule also potently inhibited the 139 multiplication of TGEV and PEDV in cells from pigs and monkeys, demonstrating its 140 potential for the development of optimized antivirals effective to treat CoV diseases in 141 animals, and to prevent the zoonotic risk caused by these pathogens. 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 6

160 Cells and virus strains. 161 CRFK, ST, PK15 and Vero cells were cultivated at 37°C with 5% CO2 in Dulbecco’s 162 modified Eagle’s medium (DMEM, Gibco) supplemented with 10% foetal bovine serum, 163 1% penicillin-streptomycin, 1% sodium pyruvate and 1% non-essential amino acids. 164 The serotype II FIPV 79 -1146 was amplified on fe line CRFK cells. Transmissible 165 gastroenteritis virus (TGEV, Purdue strain) was amplified on porcine ST cells. Porcine 166 epidemic diarrhoea virus (PEDV, CV777 strain) was amplified on simian Vero cells. 167 Supernatants were harvested and ultracentrifugated. Infe ctious titers were determined 168 using end-point dilution assays. 169 170 PPIase enzyme assay. 171 Human cyclophilin A was purified and its PPIase activity in absence or presence 172 of SMCypI was measured as previously described45. Briefly, Cyclophilin A PPIase activity 173 was measured at 20°C using the standard chymotrypsin -coupled assay. The assay buffer 174 (25 mM Hepes and 100 mM NaCl, pH 7.8) and purified cyclophilin A were pre -cooled to 175 4°C. Then, 5 ml of 50 mg/ml chymotrypsin in 1 mM HCl was added. The reaction was 176 initiated by adding peptide substrate Suc-Ala-Ala-Cis-Pro-Phe-pNA in LiCl/TFE solution 177 with rapid inversion. The absorbance of p-nitroaniline was monitored at 390 nm until the 178 reaction was complete (around 1 min). The final concentration of LiCl in the assay was 20 179 mM, and TFE was present at a concentration of 4% (v/v). Absorbance readings were 180 collected every second using a spectrophotometer. For inhibition assessment, 5 ml of the 181 tested compound in dimethyl sulfoxide (DMSO) was added to the cyclophilin solution in 182 the assay buffer. Cyclosporine A (CsA, Sigma -Aldrich) was used as a positive control of 183 PPIase inhibition in all measurements. The percentage inhibition of cyclophilin PPIase 184 activity was calculated from the slopes, and th e IC50s values obtained represent the mean 185 ± standard deviation of at least two independent measurements. 186 187 SMCypI screening on FIPV infection. 188 CRFK cells were plated in a 96-well plate at the density of 2.104 cells/well and 189 incubated for 24 hrs. Cells were infected for 1 hr with FIPV at MOI 1 in the presence of 190 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 7 50µM or 10µM of SMCypI or CsA respectively. Inoculum was removed after 1 hr and 191 inhibitors at either 50μM for SMCypI or at 10μM for CsA were added for 24 supplemental 192 hrs. The supernatants were collected and kept at -80°C prior to end-point dilution assays. 193 194 FIPV titration by end-point dilution assay. 195 CRFK cells were plated in a 96 -well plate at the density of 2.104 cells/well and 196 incubated for 24 hrs. Dilutions of viral supernatants ranging from 10 -1 to 10 -6 in eight 197 replicates per dilution were distributed in the CRFK plate. Viral inoculum was removed 2 198 hrs after infection and cells were incubated for 2 4 supplemental hrs at 37°C. Afterwards, 199 cells were incubated for 10 minutes at room temperature in absolute ethanol, then in 70% 200 ethanol, and FIPV antigens were stained by immunofluorescence using ascites liquids 201 from an infected cat (dilution 1/500) followed by incubation with anti-cat A-488 secondary 202 antibody (Jackson Immunoresearch). Wells containing fluorescent cells were counted. The 203 50% infectious dose (ID 50) was calculated using the Spearman -Karber method. Results 204 were the means of ≥ 2 experiments performed in triplicates. 205 206 Cytotoxicity assay. 207 Cells were plated in a 96-well plate at the density of 2.104 cells/well and incubated 208 for 24 hrs. Culture medium was replaced by serial dilutions of culture media containing 209 three replicates of different concentrations (ranging from 25μM to 0.9μM) of F83233, F832 210 and F833 molecules, DMSO (Dimethyl Sulfoxide) or media alone as controls. Plates were 211 incubated for 24 hrs at 37°C. Cytotoxicity was evaluated with the CellTiter -Glo Cell 212 viability Assay kit (Promega). 213 214 Dose-responses of SMCypI in CRFK, PK15 and Vero cells. 215 CRFK, PK15 and Vero cells were plated in a 96 -well plate at the density of 2.10 4 216 cells/well and incubated 18 hrs at 37°C. Cells were infected with FIPV (MOI 1), TGEV (MOI 217 0.5) and PEDV (MOI 0.5) respectively. Viral inoculation was performed in the presence of 218 F83233, F832 and F833 at concentrations ranging from 0.1μM to 25μM for FIPV assays and 219 from 0.78µM to 12.5µM for TGEV and PEDV . Two hrs after viral inoculation, fresh 220 medium was added with the sam e concentrations of inhibitors. Antiviral effect was 221 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 8 measured after further 24 hrs for FIPV and PEDV infection and 48 hrs for TGEV. The 222 experiments were repeated twice and each condition was tested in triplicates. 223 224 Time-of-drug addition assay. 225 CRFK cells were plated in a 96-well plate at the density of 2.104 cells per well and 226 incubated for 24 hrs. Cells were infected with MOI 1 of FIPV in the presence of 5µM F83233 227 for the 0 h post-infection (p.i.) condition or in presence of DMSO. After viral inoculation, 228 culture medium was removed and fresh medium containing 5μM of F83233 was added at 229 different times of the infection (1 -, 3 -, 6 -, 9 - or 12 hrs). Cells were incubated for 230 supplemental 24 hrs at 37°C. Cell supernatants were kept at -80°C prior to end -point 231 dilution assays to measure the effect of F83233 on FIPV infectivity in the different 232 conditions. The experiment was repeated three t imes and each condition was tested in 233 triplicates. 234 235 Detection of TGEV and PEDV viral antigens by immunofluorescence. 236 Fixed PK15 and Vero cells were processed similarly. TGEV infection was detected 237 with a home-made anti-Spike antibody (called 51.13, dilut ion 1/10.000) from mouse and 238 with an anti -mouse Alexa-Fluor 555 secondary antibody. PEDV infection was detected 239 with polyclonal antibodies from pig (kind gift of Drs. Y. Blanchard and M. Contrant, 240 Anses, Ploufragan, dilution 1/300) and with an anti-pig Alexa -Fluor 488 secondary 241 antibody (SouthernBiotech). Active PEDV replication was revealed with a mouse antibody 242 directed against dsRNA (J2 from Cell Signaling Technology, dilution 1/1.000) followed by 243 incubation with an anti -mouse Alexa -Fluor 555 secondary antibody. Cells were 244 counterstained with DAPI. Secondary antibodies (except the anti-pig) were used at a 1/800 245 dilution and are from Molecular Probes (Thermo Fischer Scientific). 246 247 Quantification of TGEV and PEDV infected areas and nuclei/syncytium. 248 For each concentration of F832, F833 and F83233, 3 to 5 pictures were taken with an 249 epifluorescence microscope (Zeiss) with a 10X objective . The area of staining was 250 measured with ImageJ. The number of nuclei/syncytia in PEDV -infected cells was 251 manually counted using ImageJ. 252 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 9 Alignment of amino acids sequences of cyclophilin A from different mammals. 253 The cyclophilin sequences from human (P62937), pig (P62936), monkey (P62938) 254 and cat (Q8HXS3) were recovered from UniProt (UniProt, 255 http://www.uniprot.org/uniprot/), aligned with ClustalW and processed with ESPript 3. 256 257 In silico modelling and docking. 258 The search for ligand -cyclophilin 3D crystal complexes was performed using the 259 @TOME‐3 server 46 (https://atome.cbs.cnrs.fr/ATOME_V3/index.html). Ligand files were 260 generated with MarvinSketch 6.2.2 for SMILES and Grade server for mol2 261 (https://grade.globalphasing.org/cgi-bin/grade2_server.cgi). Docking simulation of 262 F83233 in complex with pig cyclophilin A was performed using @TOME‐3 server with an 263 anchor of PDB 4J5C. The images were generated using PyMOL and MarvinSketch. 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 10

284 285 Screening of non -peptidic small -molecule cyclophilin inhibitors (SMCypI) on Feline 286 Infectious Peritonitis Virus (FIPV) infection. 287 As cellular cyclophilins facilitate the replication of numerous viruses, we 288 previously developed SMCypI to obtain broad-spectrum antiviral agents45. We have also 289 previously shown that our first SMCypI series exhibited a potent antiviral activity on 290 hepatitis C virus replication in vitro 47, but that they were only modest to reduce the 291 replication of the HCoV-229E CoV, with EC50s ranging from 7 to 71µM45. SMCypI are non-292 peptidic compounds with a common backbone on which various chemical moieties can be 293 added at three distinct sites referred as R1, R2 and R3 (Fig. 1A), thus allowing to routinely 294 generate new molecules ( Table 1 ). We thus aimed to find more potent anti -CoV 295 compounds in our enriched library. 296 Figure 1: Screening of non -peptidic small -molecule cyclophilin inhibitors (SMCypI) on Feline Infectious Peritonitis Virus (FIPV) infection. A. SMCypI are composed of a common backbone with various substitutions that could be added at the R1, R2 and R3 regions, allo wing to generate a library with an important number of compounds. B. SMCypI are more effective than CsA. FIPV viral titer was measured 24 hrs after treatment with 50µM of 20 different SMCypI . C yclosporin A (CsA) (in black) was used at the final concentrati on of 10µM .

are normalized with untreated infected cells. We performed a screen of 20 different SMCypI (Fig. 1B and Table 1) for their ability 297 to inhibit Feline Infectious Peritonitis Virus infection (FIPV, 79-1146 strain) in feline CRFK 298 cells at the final concentration of 50µM. We also used 10µM of cyclosporine A (CsA, black 299 bar), a peptidic cyclophilin inhibitor with a reported anti -CoV activity, including FIPV 38. 300 We observed that all the tested molecules modestly impacted the infectious titers of FIPV, 301 except the F83233 compound (green bar) that inhibited FIPV titer by more than 4 Log 10 302 (Fig. 1B). 303 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 11 The increased anti -FIPV effect of F83233 compared to F832 and F833 is unrelated to 304 cyclophilin A enzymatic activity inhibition. 305 SMCypI F83233 exhibits chemical moieties from F832 and F833 (blue and orange 306 bars on Fig. 1B, respectively) at R2 and R3 positions, respectively (Fig. 2A and Table 1). 307 We first confirmed the results from our screen by measuring the virus titer in cell 308 supernatants after FIPV infection in the presence of 25µM SMCypI . While F832 and F833 309 only modestly inhibited FIPV infectivity by ≈0.5 Log 10 (Fig. 2B), F83233 de creased by ≈3 310 Log10 the virus titer at this concentration. We measured the ability of these three 311 compounds to block the enzymatic activity of human cyclophilin A in an in vitro assay to 312 assess if this increased antiviral potency of F83233 was linked to an improved blockade of 313 cyclophilin function (Table 1). However, we observed no significant differences with IC50s 314 of 0.7 ± 0.5µM for F83233, 0.4 ± 0.09µM for F832 and 0.6 ± 0.1µM for F833 ( Table 1). As 315 apparent antiviral effect could sometimes be due to cellular toxicity, we treated CRFK cells 316 for 24 hrs with increasing concentrations of the three compounds, and observed a low 317 cytotoxicity of the drugs up to 25µM, that was not different between F83233 and F832/F833 318 (Fig. 2C ), an observation that was similar in the simian Vero cell line (Softic et al . 319 submitted). Overall, the increased activity of F83233 compared to its “parental” molecules 320 is not due to an ameliorated anti-PPIase activity on cyclophilin A or to cellular toxicity in 321 our experimental conditions. 322 323 Characterization of the anti-FIPV activity of F83233. 324 To characterize the anti-FIPV activity of F83233, we first quantified FIPV infectious 325 titers in the presence of escalating doses of this SMCypI ( Fig. 2D). We demonstrated a 326 strong inhibition of FIPV infectivity with a reduction in viral titer by approximately 2 Log10 327 at 5 µM. Furthermore, the maximal inhibitory effect of F83233 was observed as a decrease 328 of approximately 3-4 Log10 at 12.5 µM. Then, to better understand its mechanism of action 329 on the FIPV replication cycle, we added 5µM of F83233 at diffe rent times post -infection 330 (Fig. 2E). The maximal antiviral effect of F83233 was observed when the molecule is added 331 simultaneously with the virus or 1 hr later. Its antiviral efficiency decreased by more than 332 1 Log10 between 1 hr and 3 hrs post -infection. From 3 hrs to 12 hrs post -infection, F83233 333 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 12 continuously lost antiviral efficiency but was still effective 12 hrs post -infection with ≈1 334 Log10 of inhibition as compared to the DMSO control. Altogether these data indi cate that 335 F83233 mainly blocked an early step of the FIPV replication cycle. 336 Figure 2: Characterization of the anti -FIPV activity of SMCypI F83233. A. Chemical structures of F832, F833 and F83233 molecules. F83233 harbours, at R2 and R3 positions, chemical moieties from F832 and F833 respectively (red circles). B. Antiviral effect of 25µM of F83233 compared with those of F832 and F833 molecules was assessed by measuring FIPV viral titer in cell supernatants by TCID50 after treatment during 24 hrs with the drugs or DMSO. n = 2 independent experiments. C. Cytotoxicity assays. Measure of ATP release after treatment with increasing concentrations of F83233, F832 and F833. D. Dose-responses. FIPV viral titers were measured in cell supernatants after treatment with escalating concentrations of F83233. n = 2 independent experiments. E. Time of addition assays. F83233 (5µM) was added to the cells at various times post-infection (between 0 and 12 hrs) and FIPV titers from cell supernatants were measured in the different conditions. n = 3 independent experiments. N.T: Non-Treated. 337 F83233 inhibits infection by Transmissible Gastroenteritis Virus (TGEV) in porcine cells. 338 Figure 3: Antiviral activity of SMCypI against porcine Transmissible Gastroenteritis Virus (TGEV) in porcine cells. A. PK15 (porcine kidney) cells were infected by TGEV (Purdue strain) at a MOI of 0.5 for 48 hrs in presence of increasing concentrations of SMCypI or only DMSO . Viral infection was detected by immunofluorescence using an anti-Spike monoclonal antibody and an anti -mouse A-555 secondary antibody (red). Cell nuclei were counterstained with DAPI. Objective: 10X. B. Quantification of TGEV inhibition by SMCypI. A total of 3-5 images was captured for each condition, with two images per well. The area of infection was then quantified using ImageJ. Figure from one experiment representative of n = 2 independent experiments with 2 independent wells. N.T.: Non-Treated. *: p<0.05, **: p<0.01 (One-Way Anova followed by Kruskal-Wallis test). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 13 We next evaluate d if F83233 could also inhibit infections by other α -CoVs of 339 veterinary interest. We first studied infection by porcine Transmissible Gastro enteritis 340 Virus (TGEV, Purdue strain) in porcine kidney PK15 cells ( Fig. 3A and 3B). As for FIPV, 341 we compared the antiviral potency between F83233 and F832/F833. While F832 and F833 342 had no significant effect on TGEV infection even at 12.5µM, F83233 was active at 1.5µM 343 and was associated with an almost complete disappearance of infected cells at 6.25µM and 344 12.5µM (Fig. 3A and 3B). 345 346 F83233 inhibits infection by Porcine Epidemic Diarrhoea Virus (PEDV) in simian cells. 347 To study the broad -spectrum potential of F83233, we tested its antiviral effect on 348 the replication of Porcine Epidemic Diarrhoea Virus (PEDV, CV777 strain), which is more 349 genetically distant from FIPV. In simian Vero cells infected by PEDV, we observed a 350 modest antiviral potency of F832 and F833, while F83233 allowed to observe an important 351 reduction of infection at 3µM, and a maximal antiviral effect between 3 and 6µM (Fig. 4A 352 and 4B). 353 In cultured cells, a number of CoVs are characterized by an atypical cytopathic 354 effect, i.e. the formation of multinucleated cells referred as syncytia, as a result of a massive 355 cell-cell fusion triggered by the viral Spike protein 48–50. In Vero cells, PEDV appears to be 356 one of the most f usogenic CoV and we observed in our experimental conditions the 357 presence of several syncytia containing up to 100 nuclei (Fig. 4C, DMSO condition). Using 358 an antibody that recognizes dsRNA structures (that are formed during the replication of 359 viruses with RNA genomes of positive polarity), we observed that syncytia are an active 360 site of RNA replication, with dsRNA spots located around the packed nuclei (Fig. 4C). 361 Using 12.5µM F83233, we noticed that the SMCypI drastically decreased the size of 362 the PEDV-induced syncytia and the number of nuclei/syncytium ( Fig. 4C and 4D), and 363 markedly impacted the dsRNA staining corresponding to viral genome replication in the 364 syncytia (Fig. 4C). 365 366 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 14 Figure 4: Antiviral activity of SMCypI against Porcine Epidemic Diarrhoea Virus (PEDV) in simian cells. A. Vero (monkey kidney) cells were infected with PEDV (CV777 strain) at an MOI of 0.5 for 24 hrs in presence of increasing concentrations of SMCypI or only DMSO. Viral infection was detected by immunofluorescence using anti-PEDV porcine polyclonal antibodies and an anti -pig secondary antibody (green). Cell nuclei were counterstained with DAPI. Objective: 10X. B. Quantification of PEDV inhibition by SMCypI. A total of 3-5 images were captured for each condition, with two images per well. The area of infection was then quantified using ImageJ. Figure from one experiment representative of n = 2 independent experiments with 2 independent wells. N.T.: Non-Treated. **: p<0.01 (One-Way Anova followed by Kruskal-Wallis test). NT: Non-Treated. C. SMCypI treatment reduced drastically syncytia formation. Infected Vero cells were treated with 12.5µM of F83233 and stained after fixation with anti-PEDV polyclonal antibodies and an anti-pig secondary antibody (green), or with a mouse antibody that recognizes dsRNA structures and an anti -mouse secondary antibody (red) . Objective: 10X. D. Quantification of the number of nuclei per syncytia after treatment with increasing doses of F83233. The number of nuclei per syncytium was counted in four different fields of cells (2 pictures per well). Figure from one experiment representative of n = 2 independent experiments with 2 independent wells . ns: not significant. ***: p<0.001 (One-Way Anova followed by Kruskal-Wallis test). 367 368 369 370 371 372 373 374 375 376 377 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 15 4. Discussion 378 379 The regular emergence of new coronaviruses (CoVs) in humans and animals 380 necessitates the urgent development of antiviral tools with a broad pan -CoV spectrum. 381 Two main antiviral methodologies are commonly employed: either to target specifically 382 viral proteins or to target cellular proteins that are necessary for viral replicati on. Whilst 383 the initial approach is more frequently used, it is not without its drawbacks. Firstly, it is 384 frequently virus-specific, and secondly, it is difficult to extrapolate its antiviral efficiency 385 to other CoVs. Finally, it gives rise to the concern o f the emergence of antiviral -resistant 386 mutants. Recently, a nucleoside analogue, GS141524, a metabolite of the antiviral prodrug 387 remdesivir, has been employed to treat FIP disease, with considerable success8,9. However, 388 this strategy is not efficacious against SARS-CoV-2 for example51. In this study, we chose 389 to target the cellul ar cyclophilin A ( encoded by PPIA gene), the most abundant 390 cyclophilin. As demonstrated in previous studies, this protein has been shown to be 391 necessary for the replication of multiple CoVs from different genera 33. Furthermore, 392 evidence has demonstrated that cyclosporine A (CsA), a macrocyclic peptide composed of 393 11 amino acids, forms a complex with cyclophilin A, thereby exhibiting an antiviral effect 394 against FIPV, TGEV and PEDV at least38,42,52. Nevertheless, the immunosuppressive effect 395 of this compound prevents its utilisation as an antiviral therapeutic agent. 396 We previously developed low molecular weight, non -immunosuppressive 397 cyclophilin-inhibiting molecules (SMCypI). Subsequently, the anti-HCoV-229E activity of 398 SMCypI was characterised, exhibiting a modest antiviral effect 45. However, the relatively 399 simple chemistry of these molecules allows to envisage numerous substitutions on their 400 common backbone to generate a more diverse library ( Fig. 1A and Table 1). With this 401 approach, we described a new antiviral molecule, F83233, with a more efficient antiviral 402 activity (around 3 Log10 viral reduction at 10M) than CsA against FIPV (Fig 1B and 2D). 403 The potent antiviral activity of F83233 results directly from the eas y possibility to "mix" 404 groups of two “parental” compounds such as F832 and F833. These data pave the way for 405 the screening of more SMCypI, and could lead to structure -activity relationship studies 406 for the design of molecules even more effective than F83233 against CoVs (as well as other 407 viruses) multiplication. 408 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 16 For some viruses whose replication is cyclophilin -dependent, it has been 409 demonstrated that the PPIase enzymatic activity is necessary 33. For instance, the genome 410 replication of the hepatitis C virus (HCV) absolu tely requires the PPIase activity of 411 cyclophilin A 33,53. Similarly, we observed a very good correlation between the ability of 412 SMCypI to inhibit the PPIase activity of human cyclophilin A and their anti -HCV 413 potency47. In our present study, SMCypI with little or no ability to block cyclophilin A 414 PPIase function (F846, F768, F826) also showed a modest anti -FIPV effect ( Fig. 1B and 415 Table 1). By contrast, compounds with a more pronounced antiviral potency (F832, F840, 416 F799, F833, F759) blocked cyclophilin A PPIase activity with sub -micromolar IC 50s. 417 However, the massive increase in anti-CoV activity of F83233 compared to these molecules 418 is not linked to an improved potency on the PPIase ac tivity of cyclophilin A, as F83233 419 exhibits a similar IC 50 (Table 1 ). In the same line, CsA at 10µM was less effective than 420 F83233, whereas its IC 50 on the PPIase activity is 2 Log 10 lower than F83233. Thus, the 421 antiviral effect of SMCypI on CoVs multiplication may be more complex than that 422 observed on HCV, and these molecules could be very useful to decipher the interactions 423 between cyclophilins and CoVs. 424 The development of broad -spectrum molecules can help to prevent future 425 pandemic54. While specific antivirals or vaccines are being developed, this can for example 426 provide a first line of defence in the event of viral emergence. Antivirals targeting host 427 factors may thus act on a wide range of viruses and animal species thanks to their 428 conservation. In this study, the F83233 demonstrated robust antiviral efficacy against both 429 FIPV and TGEV, which are genetically related. In addition, it was found to inhibit PEDV, 430 a distinct α-CoV that exhibited only 31% and 29 % amino-acid homologies of the Spike 431 and N proteins respectively, when compared with TGEV. The broad activity of F83233 on 432 PEDV, FIPV and TGEV in simian, feline and porcine cells, respectively, demonstrates that 433 SMCypI are serious leads for considering treatments that can be generalized to severa l 434 animal species, and ultimately prevent the zoonotic risk posed by a mammalian CoV. This 435 broad activity is consistent with the very strong conservation of cyclophilin A and its 436 active PPIase site across mammals (Fig. 5A). The 3D reconstruction of the complex made 437 by F83233 and cyclophilin A shows identical interactions with cyclophilin A accross the 438 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 17 different species (human, porcine [shown on Fig. 5B]), simian and feline). This paves the 439 way for the study of SMCypI antiviral activity on animal CoVs infecting other species. 440 Figure 5: High-degree of conservation of cyclophilin A across mammals. A. Sequences of cyclophilin A (PPIA gene) were obtained from Uniprot, aligned with ClustalW and processed with ESPript 3. The alignment shows high-degree conservation of cyclophilin A from human, monkey, pig and cat. Amino -acids of the PPIase active site are in dashed rectangles while amino -acids that define the S2 “gatekeeper” pocket (that regulated the substrate specificity31) in dotted rectangles. B. In silico modelling and docking. The cyclophilin A sequences from human (P62937), pig (P62936), monkey (P62938) and cat (Q8HXS3) were recovered from UniProt. The ligand (F83233, in green )-cyclophilin 3D crystal complex was performed using the @TOME‐3 server 46 (https://atome.cbs.cnrs.fr/ATOME_V3/index.html). Ligand files were generated with MarvinSketch 6.2 .2 for SMILES and Grade server for mol2. Docking simulation was performed using @TOME‐3 server with an anchor of PDB 4J5C. The images were generated using PyMOL and MarvinSketch. The pig cyclophilin A is shown on the figure. 441 Acknowledgments: 442 This work was supported by A nses grant «CycloCoV». MD was supported by DIM1 443 Health (Region Ile de France). MK is supported by a grant from A nses. QN obtained 444 grants from INRAE Animal Health Department, Université Paris -Saclay and LSH 445 graduate school “OI Microbes”, a nd is supported by the Paris -Saclay ANR PIA 446 funding: ANR-20-IDEES-0002 grant. We are grateful to Dr Nicolas Meunier for critical 447 reading of the manuscript. 448 449 Authors contributions: 450 Conceptualization: SLP, QN, AAB. Investigation: MD, MK, HH, LS, LB, JFG, AAB, 451 QN and SLP. Writing: MD, AAB, QN and SLP, with input from all authors. 452 453 454 455 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 18 Compound R1 R3 R2 IC50 PPIase (µM) 456 457 Compound R1 R2 R3 IC50 PPIase CypA (µM) F716 0.8 ± 0.1 F712 3.3 ± 1.4 F846 >10 F822 6.9 ± 2.7 F767 3.2 ± 0.9 F768 >10 F826 >10 F680 2.2 ± 0.4 F816 1 ± 0.2 F834 1.5 ± 0.2 S N NH2 S S N NH2 N S NH2 S N N NH2 S N NH2 F F F S S N NH2 N S N Br N NH2 O NH2 S N N S NH2 O O (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 19 458 Table 1: Structures and anti-PPIase activities of SMCypI. 459 F83236 0,4± 0.05 F836 0.9 ± 0.1 F849 0.73 ± 0.07 F798 0.3 ± 0.1 F759 0.8 ± 0.1 F833 0.6 ± 0.1 F799 0.7 ± 0.1 F840 0.7 ± 0.05 F832 0.4 ± 0.09 F83233 0.7 ± 0.5 N S NH2 Br O O Br NH2 F S N O S N NH2 S N NH2 N S NH2 S N NH2 S N NH2 Br O O N S N S N S Br O (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 20

460 1. Millet, J. K., Jaimes, J. A. & Whittaker, G. R. Molecular diversity of coronavirus host cell entry 461 receptors. FEMS Microbiol Rev 45, fuaa057 (2021). 462 2. Drosten, C. et al. Identification of a novel coronavirus in patients with severe acute respiratory 463 syndrome. N Engl J Med 348, 1967–1976 (2003). 464 3. Zaki, A. M., van Boheemen, S., Bestebroer, T. M., Osterhaus, A. D. M. E. & Fouchier, R. A. M. 465 Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia. N Engl J Med 367, 466 1814–1820 (2012). 467 4. Zhou, P. et al. A pneumonia outbreak associated with a new coronavirus of probable bat origin. 468 Nature 579, 270–273 (2020). 469 5. Sanchez, C. M., Pascual-Iglesias, A., Sola, I., Zuñiga, S. & Enjuanes, L. Minimum Determinants of 470 Transmissible Gastroenteritis Virus Enteric Tropism Are Located in the N-Terminus of Spike 471 Protein. Pathogens 9, 2 (2019). 472 6. Li, Y. et al. An alternative pathway of enteric PEDV dissemination from nasal cavity to intestinal 473 mucosa in swine. Nat Commun 9, 3811 (2018). 474 7. Le Poder, S. Feline and canine coronaviruses: common genetic and pathobiological features. Adv 475 Virol 2011, 609465 (2011). 476 8. Green, J., Syme, H. & Tayler, S. Thirty-two cats with effusive or non-effusive feline infectious 477 peritonitis treated with a combination of remdesivir and GS-441524. J Vet Intern Med 37, 1784–478 1793 (2023). 479 9. Coggins, S. J. et al. Outcomes of treatment of cats with feline infectious peritonitis using 480 parenterally administered remdesivir, with or without transition to orally administered GS-481 441524. J Vet Intern Med 37, 1772–1783 (2023). 482 10. Rottier, P. J. M., Nakamura, K., Schellen, P., Volders, H. & Haijema, B. J. Acquisition of 483 macrophage tropism during the pathogenesis of feline infectious peritonitis is determined by 484 mutations in the feline coronavirus spike protein. J Virol 79, 14122–14130 (2005). 485 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 21 11. Kummrow, M. et al. Feline coronavirus serotypes 1 and 2: seroprevalence and association with 486 disease in Switzerland. Clin Diagn Lab Immunol 12, 1209–1215 (2005). 487 12. Herrewegh, A. A., Smeenk, I., Horzinek, M. C., Rottier, P. J. & de Groot, R. J. Feline coronavirus 488 type II strains 79-1683 and 79-1146 originate from a double recombination between feline 489 coronavirus type I and canine coronavirus. J Virol 72, 4508–4514 (1998). 490 13. Wang, Q., Vlasova, A. N., Kenney, S. P. & Saif, L. J. Emerging and re-emerging coronaviruses in 491 pigs. Curr Opin Virol 34, 39–49 (2019). 492 14. Turlewicz-Podbielska, H. & Pomorska-Mól, M. Porcine Coronaviruses: Overview of the State of 493 the Art. Virol Sin 36, 833–851 (2021). 494 15. Li, W. et al. New Variants of Porcine Epidemic Diarrhea Virus, China, 2011. Emerg. Infect. Dis. 18, 495 1350–1353 (2012). 496 16. Lin, C.-M., Saif, L. J., Marthaler, D. & Wang, Q. Evolution, antigenicity and pathogenicity of global 497 porcine epidemic diarrhea virus strains. Virus Res 226, 20–39 (2016). 498 17. Wood, E. An apparently new syndrome of porcine epidemic diarrhoea. Veterinary Record 100, 499 243–244 (1977). 500 18. Marthaler, D. et al. Rapid detection, complete genome sequencing, and phylogenetic analysis of 501 porcine deltacoronavirus. Emerg Infect Dis 20, 1347–1350 (2014). 502 19. Stevenson, G. W. et al. Emergence of Porcine epidemic diarrhea virus in the United States: 503 clinical signs, lesions, and viral genomic sequences. J VET Diagn Invest 25, 649–654 (2013). 504 20. Mole, B. Deadly pig virus slips through US borders. Nature 499, 388 (2013). 505 21. Vlasova, A. N. et al. Distinct characteristics and complex evolution of PEDV strains, North 506 America, May 2013-February 2014. Emerg Infect Dis 20, 1620–1628 (2014). 507 22. Edwards, C. E. et al. Swine acute diarrhea syndrome coronavirus replication in primary human 508 cells reveals potential susceptibility to infection. Proc Natl Acad Sci U S A 117, 26915–26925 509 (2020). 510 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 22 23. Lednicky, J. A. et al. Independent infections of porcine deltacoronavirus among Haitian children. 511 Nature 600, 133–137 (2021). 512 24. Woo, P. C. Y., Lau, S. K. P., Huang, Y. & Yuen, K.-Y. Coronavirus diversity, phylogeny and 513 interspecies jumping. Exp Biol Med (Maywood) 234, 1117–1127 (2009). 514 25. Steiner, S. et al. SARS-CoV-2 biology and host interactions. Nat Rev Microbiol 22, 206–225 (2024). 515 26. Zhang, S., Cao, Y. & Yang, Q. Transferrin receptor 1 levels at the cell surface influence the 516 susceptibility of newborn piglets to PEDV infection. PLoS Pathog 16, e1008682 (2020). 517 27. Zhu, M. et al. CMPK2 is a host restriction factor that inhibits infection of multiple coronaviruses 518 in a cell-intrinsic manner. PLoS Biol 21, e3002039 (2023). 519 28. Tse, L. V. et al. Genomewide CRISPR knockout screen identified PLAC8 as an essential factor for 520 SADS-CoVs infection. Proc Natl Acad Sci U S A 119, e2118126119 (2022). 521 29. Sun, L. et al. Genome-scale CRISPR screen identifies TMEM41B as a multi-function host factor 522 required for coronavirus replication. PLoS Pathog 17, e1010113 (2021). 523 30. Li, Y. et al. Calpain-1: a Novel Antiviral Host Factor Identified in Porcine Small Intestinal Mucus. 524 mBio 13, e0035822 (2022). 525 31. Davis, T. L. et al. Structural and biochemical characterization of the human cyclophilin family of 526 peptidyl-prolyl isomerases. PLoS Biol 8, e1000439 (2010). 527 32. Göthel, S. F. & Marahiel, M. A. Peptidyl-prolyl cis-trans isomerases, a superfamily of ubiquitous 528 folding catalysts. Cell Mol Life Sci 55, 423–436 (1999). 529 33. Baugh, J. & Gallay, P. Cyclophilin involvement in the replication of hepatitis C virus and other 530 viruses. Biol Chem 393, 579–587 (2012). 531 34. Castro, A. P. V., Carvalho, T. M. U., Moussatché, N. & Damaso, C. R. A. Redistribution of 532 cyclophilin A to viral factories during vaccinia virus infection and its incorporation into mature 533 particles. J Virol 77, 9052–9068 (2003). 534 35. Hamamoto, I. et al. Cyclosporin A inhibits the propagation of influenza virus by interfering with a 535 late event in the virus life cycle. Jpn J Infect Dis 66, 276–283 (2013). 536 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 23 36. Qing, J. et al. Cyclophilin A associates with enterovirus-71 virus capsid and plays an essential role 537 in viral infection as an uncoating regulator. PLoS Pathog 10, e1004422 (2014). 538 37. de Wilde, A. H., Pham, U., Posthuma, C. C. & Snijder, E. J. Cyclophilins and cyclophilin inhibitors in 539 nidovirus replication. Virology 522, 46–55 (2018). 540 38. Pfefferle, S. et al. The SARS-coronavirus-host interactome: identification of cyclophilins as target 541 for pan-coronavirus inhibitors. PLoS Pathog 7, e1002331 (2011). 542 39. Lee, Y. J. & Lee, C. Porcine deltacoronavirus induces caspase-dependent apoptosis through 543 activation of the cytochrome c-mediated intrinsic mitochondrial pathway. Virus Res 253, 112–544 123 (2018). 545 40. Softic, L. et al. Inhibition of SARS-CoV-2 Infection by the Cyclophilin Inhibitor Alisporivir (Debio 546 025). Antimicrob Agents Chemother 64, e00876-20 (2020). 547 41. de Wilde, A. H. et al. Cyclosporin A inhibits the replication of diverse coronaviruses. J Gen Virol 548 92, 2542–2548 (2011). 549 42. Tanaka, Y., Sato, Y. & Sasaki, T. Feline coronavirus replication is affected by both cyclophilin A 550 and cyclophilin B. J Gen Virol 98, 190–200 (2017). 551 43. Luo, C. et al. Nucleocapsid protein of SARS coronavirus tightly binds to human cyclophilin A. 552 Biochem Biophys Res Commun 321, 557–565 (2004). 553 44. Ma-Lauer, Y. et al. Influences of cyclosporin A and non-immunosuppressive derivatives on 554 cellular cyclophilins and viral nucleocapsid protein during human coronavirus 229E replication. 555 Antiviral Res 173, 104620 (2020). 556 45. Ahmed-Belkacem, A. et al. Fragment-based discovery of a new family of non-peptidic small-557 molecule cyclophilin inhibitors with potent antiviral activities. Nat Commun 7, 12777 (2016). 558 46. Pons, J.-L. et al. @TOME 3.0: Interfacing Protein Structure Modeling and Ligand Docking. J Mol 559 Biol 436, 168704 (2024). 560 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint 24 47. Nevers, Q. et al. Characterization of the Anti-Hepatitis C Virus Activity of New Nonpeptidic Small-561 Molecule Cyclophilin Inhibitors with the Potential for Broad Anti-Flaviviridae Activity. Antimicrob 562 Agents Chemother 62, e00126-18 (2018). 563 48. Rajah, M. M. et al. SARS-CoV-2 Alpha, Beta, and Delta variants display enhanced Spike-mediated 564 syncytia formation. EMBO J 40, e108944 (2021). 565 49. Xia, S. et al. Porcine deltacoronavirus resists antibody neutralization through cell-to-cell 566 transmission. Emerg Microbes Infect 12, 2207688 (2023). 567 50. Li, W. et al. A Single Point Mutation Creating a Furin Cleavage Site in the Spike Protein Renders 568 Porcine Epidemic Diarrhea Coronavirus Trypsin Independent for Cell Entry and Fusion. J Virol 89, 569 8077–8081 (2015). 570 51. WHO Solidarity Trial Consortium et al. Repurposed Antiviral Drugs for Covid-19 - Interim WHO 571 Solidarity Trial Results. N Engl J Med 384, 497–511 (2021). 572 52. Kim, Y. & Lee, C. Porcine epidemic diarrhea virus induces caspase-independent apoptosis 573 through activation of mitochondrial apoptosis-inducing factor. Virology 460–461, 180–193 574 (2014). 575 53. Coelmont, L. et al. DEB025 (Alisporivir) inhibits hepatitis C virus replication by preventing a 576 cyclophilin A induced cis-trans isomerisation in domain II of NS5A. PLoS One 5, e13687 (2010). 577 54. Jochmans, D., Laporte, M. & Neyts, J. Antiviral strategies for epidemic and pandemic 578 preparedness. Cell Host Microbe 31, 856–860 (2023). 579 580 581 582 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted May 24, 2025. ; https://doi.org/10.1101/2025.05.24.655938doi: bioRxiv preprint