Development of Highly Potent Noncovalent Inhibitors of SARS-CoV-2 3CLpro

The 3C-like protease (3CLpro) is an essential enzyme for the replication of SARS-CoV-2 and other coronaviruses and thus is a target for coronavirus drug discovery. Nearly all inhibitors of coronavirus 3CLpro reported so far are covalent inhibitors. Here, we report the development of specific, noncovalent inhibitors of 3CLpro. The most potent one, WU-04, effectively blocks SARS-CoV-2 replications in human cells with EC50 values in the 10-nM range. WU-04 also inhibits the 3CLpro of SARS-CoV and MERS-CoV with high potency, indicating that it is a pan-inhibitor of coronavirus 3CLpro. WU-04 showed anti-SARS-CoV-2 activity similar to that of PF-07321332 (Nirmatrelvir) in K18-hACE2 mice when the same dose was administered orally. Thus, WU-04 is a promising drug candidate for coronavirus treatment.


Data collection and structure determination
The crystals were first transferred to a cryoprotectant solution (the well buffer plus 20 mM HEPES pH 7.4, 150 mM NaCl and 20% glycerol), then loaded onto the X-ray diffractometer (Rigaku, XtaLAB Synergy Customer) at Westlake University. The diffraction data was collected at 100 K and processed with the reduction program CrysAlisPro. The structures were solved by molecular replacement using Phaser in PHENIX. 8 The crystal structures of SARS-CoV-2 3CLpro (PDB code: 6Y2E), SARS-CoV 3CLpro (PDB code: 1UJ1), and MERS-CoV 3CLpro (PDB code: 4YLU) were used as the initial models. The structures were manually refined with Coot 9 and PHENIX. 10 Data collection and refinement statistics can be found in table S1 that was generated using the utility phenix.table_one in PHENIX. 10 Isothermal titration calorimetry (ITC) S8 ITC experiments were done with the isothermal titration calorimeter MicroCal PEAQ-ITC (Malvern Panalytical). The inhibitor with a concentration of 25 µM in the ITC buffer (20 mM HEPES, pH 7.4, 150 mM NaCl, 0.625% DMSO) was titrated by 255 µM 3CLpro in the ITC buffer at 25 °C. Control experiments in which the inhibitors were titrated by the ITC buffer were performed and the data was used to correct the 3CLpro titration data. The data was processed using the MicroCal PEAQ-ITC analysis software. Each measurement was repeated three times.
Two reporter viruses SARS-CoV-2-Nluc and SARS-CoV-2-Fluc used in the antiviral assays were engineered by inserting nanoluciferase (Nluc) or firefly luciferase gene at the ORF7 of the viral genome using an infectious clone of SARS-CoV-2 (strain 2019-nCoV/USA_WA1/2020) according to the protocols described previously. 12

SARS-CoV-2 antiviral assays (Figure 3a, 3b and 3c, Figure. S6)
For antiviral assay in A549-hACE2 cells, 12,000 cells per well in phenol-red free medium containing 2% FBS were plated into a white opaque 96-well plate (Corning). On the next day, 3fold serial dilutions of compounds were prepared in dimethyl sulfoxide (DMSO). The compounds were further diluted 100-fold in the phenol-red free culture medium containing 2% FBS. Cell culture fluids were removed and incubated with 50 μL of diluted compound solutions and 50 μL of SARS-CoV-2-Nluc viruses (MOI 0.025). At 48 hours postinfection, 50 μL Nano luciferase substrates (Promega) were added to each well. Luciferase signals were measured using a Synergy™ Neo2 microplate reader. The luciferase signals (Y-axis) versus the compound S10 concentrations were plotted in software Prism 9 and fitted using the built-in nonlinear regression method " [Inhibitor] vs. response -Variable slope (four parameters)" to get the "Bottom" and "Top" values in the following equation in which "Y" is the luciferase signal and "X" is the compound concentration: Y = Bottom + (Top -Bottom)/(1+(IC50/X)^HillSlope) The DMSO control was assigned a concentration of 0.1 nM. The relative luciferase signals were calculated by normalizing the luciferase signals using the following equation in which "Y" is the relative luciferase signal, "X" is the luciferase signal, "Bottom" and "Top" values are from the above equation: The relative luciferase signals (Y-axis) versus the compound concentrations (X-axis) were plotted in software Prism 9 and fitted using the built-in nonlinear regression method "[Agonist] vs. For antiviral test in NHBE cells, 1×10 5 cells per well were plated into a 24-well plate (Nunc).
On the next day, 3-fold serial dilutions of compounds were prepared in dimethyl sulfoxide (DMSO). The compounds were further diluted 1000-fold in the culture medium containing SARS-CoV-2-Fluc (MOI 10) and added to the cells. At 24 hours postinfection, cells were washed in PBS and lyzed in 300 μL 1× cell culture lysis buffer (Promega) for 20 minutes. 40 μL lysates were mixed with 100 μL firefly luciferase substrates (Promega) in a white opaque 96-well plate (Corning) and luciferase signals were measured using a Synergy™ Neo2 microplate reader. The EC50 was calculated as above. Experiments were performed with six replicates. (Figure 3d) 25,000 A549-hACE2 cells per well in medium containing 10% FBS were plated into a clear 96well plate. On the next day, 3-fold serial dilutions of compounds starting from 100 μM were prepared in the culture medium containing 2% FBS. Cell culture fluids were removed and incubated with 50 μL of diluted compound solutions and 50 μL of SARS-CoV-2 virus (nCoV-SH01, MOI 0.5). At 24 hours postinfection, cells were fixed with 4% paraformaldehyde in PBS for 30 minutes, permeablized with 0.2% Triton X-100 for 1 hour. Cells were then incubated with house-made mouse anti-SARS-CoV-2 nucleocapsid (N) protein serum (1:1000) for 2 hours at room temperature. After three washes, cells were incubated with the secondary goat anti-mouse IgG (H + L) conjugated with Alexa Fluor 555 (Thermo #A-21424, 2 μg/ml) for 1 hour at room temperature, followed by staining with 4',6-diamidino-2-phenylindole (DAPI). Images were collected using an Operetta High Content Imaging System (PerkinElmer), and processed using the PerkinElmer Harmony high-content analysis software v4.9 and ImageJ v2.0.0 (http://rsb.info.nih.gov/ij/). For each sample, the ratio of infected cells were calculated by dividing S12 the number of N protein-positive cells by the number of nucleuses stained by DAPI; next, the inhibition of virus infection by the compounds was calculated using the following equation, in which RCompound and RDMSO were the ratios of infected cells in the presence of the compounds and DMSO, respectively:

SARS-CoV-2 antiviral assays
The Inhibition (%) versus the compound concentration (X-axis) was plotted in software Prism 9 and fitted using the built-in nonlinear regression method "[Inhibitor] vs. response -Variable slope (four parameters)" to calculate the EC50. The DMSO control was assigned a concentration of 0.01 nM during data analysis, but in the figures, it was manually changed back to 0 nM. Experiments were performed with technical duplicates. (Figure 3e, 3f, 3g and 3h) A viral load reduction assay was performed on Caco-2 and A549-TMPRSS2-ACE2 cells, as described previously but with modifications. 15 30,000 cells/well of Caco-2 cells or A549-TMPRSS2-ACE2 cells in phenol-red free medium containing 2% FBS were plated into a white opaque 96-well plate (Corning). On the next day, 3-fold serial dilutions of compounds were prepared in dimethyl sulfoxide (DMSO). The 3CLpro inhibitors were further diluted 100-fold in the phenol-red free culture medium containing 2% FBS. One hour after SARS-CoV-2 Omicron or Delta variant infection (MOI 0.1), the infectious inoculum was removed and replaced with 100 μL of diluted compound solutions. At 48 h.p.i., cell lysate samples from the infected cells (MOI = 0.1) were collected for qRT-PCR analysis. Briefly, 100 μl of the cell lysate were extracted for total RNA with the RNeasy Mini Kit (Qiagen). Real-time one-step qRT-PCR was used for quantitation of SARS-CoV-2 viral load, and for quantitation of the β-actin gene of the host cells that was used as an internal control, using the QuantiNova Probe RT-PCR kit (Qiagen) with a LightCycler 480

SARS-CoV-2 antiviral assays
Real-Time PCR System (Roche). Each 20-μl reaction mixture contained 10 μl of 2×QuantiNova S13 Probe RT-PCR Master Mix, 1.2 μl of RNase-free water, 0.2 μl of QuantiNova Probe RT-Mix, 1.6 μl each of 10 μM forward and reverse primer, 0.4 μl of 10 μM probe and 5 μl of extracted RNA as the template. Reactions were incubated at 45 °C for 10 min for reverse transcription, 95 °C for 5 min for denaturation, followed by 45 cycles of 95 °C for 5 s and 55 °C for 30 s. Signal detection was carried out and measurements were made in each cycle after the annealing step. The cycling profile ended with a cooling step at 40 °C for 30 s. The primers and probe for the viral gene were against the RNA-dependent RNA polymerase (RdRp) gene region of SARS-Cov-2; their sequences were as below: forward primer: 5′ CGCATACAGTCTTRCAGGCT-3′ reverser primer: 5′-GTGTGATGTTGAWATGACATGGTC-3′ probe: 5′-FAM TTAAGATGTGGTGCTTGCATACGTAGAC-IABkFQ-3′ The pre-designed probe and primer set for the β-actin gene were purchased from ThermoFisher (Cat#4333762).
After qRT-PCR, the relative viral gene expression (RdRp gene copy/β-actin) in each sample was calculated using the following equation in which CqRdRp and Cqβ-actin were the quantitation cycles of RdRp and β-actin, respectively: RdRp gene copies/β-actin = 2^(CqRdRp -Cqβ-actin) The "RdRp gene copies/β-actin" (Y-axis) versus the inhibitor concentrations were plotted in software Prism 9 and fitted using the built-in nonlinear regression method "[Inhibitor] vs. response -Variable slope (four parameters)" to get the "Top" value (the "Bottom" was set to 0) in the following equation in which "Y" is the "RdRp gene copies/β-actin" and "X" is the compound concentration: The DMSO control was assigned a concentration of 0.01 nM. Then the "RdRp gene copies/βactin" were normalized using the following equation: "RdRp gene copies/β-actin (%)" = 100*"RdRp gene copies/β-actin"/Top The "RdRp gene copies/β-actin (%)" versus the compound concentrations (X-axis) were plotted in software Prism 9 and fitted using the built-in nonlinear regression method "[Inhibitor] vs. response -Variable slope (four parameters)" to calculate the EC50. The DMSO control was assigned a concentration of 0.01 nM during data analysis, but in the figures, it was manually changed back to 0 nM.

Ethics statement
All experiments with infectious SARS-CoV-2 were performed in the biosafety level 4 and animal biosafety level 4 facilities in the Harbin Veterinary Research Institute (HVRI) of the Chinese S15 Academy of Agricultural Sciences (CAAS), and approved by the Committee on the Ethics of Animal Experiments of the HVRI of CAAS (approval number 211214-01).
Compound WU-04 and PF-07321332 were first dissolved in one volume of Tween80/ethanol (v/v = 1/1), then mixed with five volumes of 10% VE-TPGS (w/v). For BALB/c mouse study, mice (n = 6 per group) were orally administered with WU-04 (250 mg/kg of body weight per dose, twice daily) or the vehicle control for 3 days, and infected intranasally with 100 PFU SARS-CoV-2 HRB26M strain in a volume of 50 μL 1 hour after the first dose. After 3 days treatment, the mice were sacrificed for viral load quantification by qPCR and PFU assay in the turbinates and lungs.
For H11-K18-hACE2 mice study, mice (n = 6 per group) were infected intranasally with 100 PFU SARS-CoV-2 HRB25 strain in a volume of 50 μL, and orally administered with WU-04 (100, 200, or 300 mg/kg of body weight per dose, twice daily), or PF-07321332 (300 mg/kg of body weight per dose, twice daily), or the vehicle control 1 hour after infection. The mice were observed daily for signs of disease, bodyweight changes for four days, and sacrificed for viral load quantification S16 in the turbinates, lungs and brain, histopathologic analysis in the lungs, and immunohistochemical analysis in brains.
qPCR Viral genomic RNA of SARS-CoV-2 was extracted by using a QIAamp vRNA Mini kit (Hilden, Germany). Reverse transcription was performed by using the HiScript II Q RT SuperMix (Nanjing, China) for qPCR. qPCR was performed to quantitate the number of viral N gene RNA copies by

Histopathologic and immunohistochemical studies
Pathology of lung was evaluated by hematoxylin and eosin (H&E) staining, and viral antigen retrieval of brain was analysis by specific anti-SARS-CoV-2 nucleoprotein monoclonal antibody staining, as previously described. 19 One-µm-thick formalin-fixed paraffin-embedded sections were prepared. The slides of lung were stained and evaluated by a pathologist in a double-blind manner. The pathological scores were calculated by a pathological scoring system established by the Harbin Veterinary Research Institute (HVRI) of the Chinese Academy of Agricultural Sciences (CAAS). The slides of brain were visualized with DAB (3,3′-Diaminobenzidine) staining and counterstained with hematoxylin after immunostaining.

Preclinical pharmacokinetics studies
The preclinical pharmacokinetics studies were performed at WuXi AppTec following the study protocol and local Standard Operating Procedures (SOPs).

Mouse pharmacokinetics
The mice were acclimated to the test facility for at least 3 days before being placed on study. Thirty

Dog pharmacokinetics
The Beagle dogs (male, ≥ 6 months, body weight 6-12 kg) were acclimated to the test facility for at least 7 days before being placed on study. The dogs were fed twice daily (approximately 220 grams of Certified Dog Diet daily). For IV dosing, compound WU-04 was dissolved in a mixed solvent containing 5% DMSO, 5% solutol and 90% water, and then administered to four Beagle dogs via IV route. Four PO dosing, the dogs were fed the afternoon (3:30 pm to 4:00 pm) prior to the day of dosing (the remaining food was removed at night) and then fed once on the day of dosing (220 g of Certified Dog Diet, 4 hours post dose); compound WU-04 was dissolved in 10% VE-TPGS to prepare a stock of 5 mg/mL, or in a mixture of Kolliphor ELP (13.47%), Transcutol HP (66.74%) and Labrafac M1944 CS (19.79%) to prepare a stock of 200 mg/mL. For the dogs S18 that were also administered RTV, a dose of RTV (3.5 mg/mL in 10% VE-TPGS) was administered 12 hours before the administration of WU-04, and a second dose of RTV (3.5 mg/mL in the WU-04 stock in 10% VE-TPGS) was administered together with WU-04; 12 hours later, the last dose of RTV (3.5 mg/mL in 10% VE-TPGS) was administered. Each blood collection (about 0.5 mL per time point) was performed from peripheral vein into pre-chilled commercial tube containing potassium (K2) EDTA (0.85-1.15 mg) and placed on wet ice until centrifugation. After centrifugation, the plasma concentrations of WU-04 were determined through LC-MS/MS analysis.

Inhibition of metabolism of WU-04 in human liver microsomes by CYP isoform-selective chemical inhibitors
The metabolism of compound WU-04 in human liver microsomes (Corning, Cat. No. 452117) was measured in the presence or absence of the CYP isoform selective inhibitors in duplicate (n = 2). CLint(liver) = CLint(mic) * mg microsomal protein/g liver weight * g liver weight/kg body weight.

Binding of WU-04 to Human, SD Rat, CD-1 Mouse, Beagle Dog, Cynomolgus Monkey Plasma
The binding of WU-04 to plasma was measured using equilibrium dialysis. A loading matrix was in which F is the analyte concentration or peak area ratio of analyte/internal standard on the buffer (receiver) side of the membrane, T is the the analyte concentration or peak area ratio of analyte/internal standard on the matrix (donor) side of the membrane, and T0 is the analyte S21 concentration or the peak area ratio of analyte/internal standard in the loading matrix sample at time zero.

Synthesis of compound WU-01
Scheme S1. The synthesis route of compound WU-01.

Preparation of compound 5-bromo-2-[3-(tert-butoxycarbonylamino) pentylamino]-3-nitrobenzoic acid (1b)
To a solution of compound 1a (384 mg, 1.11 mmol) in CH2Cl2 (6 mL compound 1a was consumed completely. The reaction mixture was extracted with a mixture of 10 mL H2O and 10 mL CH2Cl2. The aqueous phase was separated, and its pH was adjusted to 6 with saturated citric acid. Then CH2Cl2 (10 mL x 3) was added to the aqueous phase to extract the product, and the organic layers were combined, dried over Na2SO4, and concentrated under reduced pressure to give a crude product. The crude product was triturated with petroleum ether at 15 °C for 15 minutes. Compound 1b was obtained as a yellow oil (451 mg, 91.1% yield).
The crude product was purified by silica gel chromatography. Compound 1c was obtained as a yellow solid (97.3 mg, 21.4% yield).  S24 To a solution of compound 1c (97.0 mg, 211 µmol) in CH2Cl2 (3 mL), 4 M HCl in ethyl acetate (4 mL) was added. The mixture was stirred at 15 °C for 4 hours, and then concentrated under reduced pressure to give compound 1d as a yellow solid (75.0 mg, 98.9% yield).
This crude product was used in next step reaction without further purification. LC-MS: m/z = 460.9 [M+H] + .
The reaction mixture was poured into ice-water (500 mL) and extracted with ethyl acetate (200 mL x 3). The organic layers were combined, washed with brine (200 mL x 2), dried over Na2SO4, filtered. After removing the solvent by vacuum evaporation, the residue was purified by flash silica gel chromatography to give compound 3a as a yellow solid (5.0 g, 64.9% yield).

S31
To a solution of compound 4a (90.0 mg, 196 umol) in CH2Cl2 (3 mL), HATU (112 mg, 295 µmol) and DIEA (140 mg, 1.08 mmol, 188 µL) were added, followed by the addition of methanamine hydrochloride (33.2 mg, 491 µmol). Then the reaction was stirred at 25 °C for 12 hours. LC-MS showed that 4a was consumed completely. The reaction mixture was filtered through a gelite pad, and the filtrate was concentrated to remove the solvent. The crude product was purified by Perp-TLC. Compound 4b was obtained as a yellow solid (63.6 mg, 66.0% yield).

Preparation of compound tert-butyl 3-[(Z)-tert-butylsulfinyliminomethyl]azetidine-1carboxylate (5a)
The filtrate was extracted with ethyl acetate (10 mL x 3). The organic layers were combined, dried over Na2SO4, and filtered. The solvent was removed under reduced pressure and the residue was

Preparation of compound N-[(1S,2R)-2-aminocyclohexyl]isoquinoline-4-carboxamide (6b)
To a solution of compound 6a (400 mg, 1.08 mmol) in CH2Cl2 (6 mL), HCl/dioxane (4 M, 4 mL) was added. The mixture was stirred at 25 °C for 0.6 hour, and the pH was adjusted to above 7 by NaOH (3 M in H2O). Then ethyl acetate (10 mL x 3) was added into the mixture to extract the product. The ethyl acetate layers were combined, washed with saturated aqueous NaCl (50 mL x 2) and dried over Na2SO4. The solvent was removed under reduced pressure to give compound 6b as a white solid (170 mg, 55.9% yield).
The reaction mixture was diluted with ethyl acetate (100 mL) and washed with H2O (150 mL x 3).
The organic layers were combined, washed with saturated aqueous NaCl (150 mL x 3), dried over Na2SO4, and filtered. The solvent was removed by evaporation under reduced pressure, and the residue was purified by flash silica gel chromatography to give compound 7b as a brown solid (65 mg, 3.17% yield). LC-MS: m/z = 471.2 [M+H] + .

Preparation of compound methyl-5-bromo-2-fluoro-3-nitrobenzoate (11a)
A solution of 5-bromo-2-fluoro-3-nitrobenzoic acid (1 g, 3.80 mmol) in MeOH (20 mL) was stirred at room temperature. After the addition of conc. H2SO4 (0.5 mL), the reaction was heated to reflux for 3 hours. Then organic solvent was removed under reduced pressure and brine was added and the mixture was extracted with ethyl acetate. The organic layer was dried over

Preparation of compound methyl 2-(((1R,2S)-2-aminocyclohexyl)amino)-5-bromo-3nitrobenzoate (11c)
To a solution of compound 11b (201 mg, 0.43 mmol) in anhydrous CH2Cl2 (6 mL) with stirring at room temperature, HCl (6 mL, 3M in ethyl acetate) was added. The reaction was stirred at room temperature for 1 hour. The mixture was concentrated under vacuum to remove the solvent. The residue was used in next step reaction without further purification.

Preparation of compound 5-bromo-2-(((1R,2S)-2-(isoquinoline-4carboxamido)cyclohexyl)amino)-3-nitrobenzoic acid (WU-12)
LiOH·H2O (36 mg, 0.87 mmol) was added to a solution of compound WU-11 (150 mg, 0.29 mmol) in a mixture of 10 mL of MeOH (10 mL) and 1 mL of water. The reaction was stirred at room temperature for 2 hours. Then the organic solvent was removed under vacuum and the residue was dissolved in 2 mL of water. After the pH was adjusted to 5 using 2N HCl, the precipitate was collected by filtration and dried to give WU-12 as a yellow solid (100 mg  Figure S1. Enzyme activity of wild type 3CLpro and 6x His-tagged 3CLpro of SARS-CoV-2. WT is the 3CLpro of SARS-CoV-2 with no additional residues at its N or C terminus. mHis is the 3CLpro of SARS-CoV-2 with a 6x His tag inserted between residues Arg222 and Phe223. N-His and C-His are the 3CLpro of SARS-CoV-2 with a 6x His tag at its N-terminus and C-terminus, respectively. The enzyme activity was evaluated using a fluorescence resonance energy transfer (FRET)-based assay (see "Materials and Methods"). The data represent the mean ± SD of four independent measurements.
S48 Figure S2. Inhibition of the 3CLpro of the Omicron variant of SARS-CoV-2.
The ability of compound WU-04 to inhibit the enzyme activity of the wild-type (WT) 3CLpro and that of the 3CLpro of the SARS-CoV-2 Omicron variant which carries a P132H mutation was evaluated using the fluorescent substrate Dabcyl-KTSAVLQSGFRKME-Edans. The data represent the mean ± SD of three independent measurements.  The binding affinity (Kd) between WU-04 and SARS-CoV 3CLpro (a) and that between WU-04 and MERS-CoV 3CLpro (b) were measured using isothermal titration calorimetry (ITC). (c, d) The antiviral activity of WU-04 against SARS-CoV (c) or against MERS-CoV (d) was measured in Vero E6 cells. The data represents the mean ± SD of three independent measurements.  CLint(liver) = CLint(mic) * mg microsomal protein/g liver weight * g liver weight/kg body weight; NCF (abbreviation of no co-factor): no NADPH is added to NCF samples (replaced by buffer) during the 60-minute incubation.