Drug-Repurposing Screening Identified Tropifexor as a SARS-CoV-2 Papain-like Protease Inhibitor.

The global COVID-19 pandemic underscores the dire need for effective antivirals. Encouraging progress has been made in developing small-molecule inhibitors targeting the SARS-CoV-2 RNA-dependent RNA polymerase (RdRp) and main protease (Mpro). However, the development of papain-like protease (PLpro) inhibitors faces several obstacles. Nevertheless, PLpro represents a high-profile drug target given its multifaceted roles in viral replication. PLpro is involved in not only the cleavage of viral polyprotein but also the modulation of host immune response. In this study, we conducted a drug-repurposing screening of PLpro against the MedChemExpress bioactive compound library and identified three hits, EACC, KY-226, and tropifexor, as potent PLpro inhibitors with IC50 values ranging from 3.39 to 8.28 μM. The three hits showed dose-dependent binding to PLpro in the thermal shift assay. In addition, tropifexor inhibited the cellular PLpro activity in the FlipGFP assay with an IC50 of 10.6 μM. Gratifyingly, tropifexor showed antiviral activity against SARS-CoV-2 in Calu-3 cells at noncytotoxic concentrations. Overall, tropifexor represents a novel PLpro inhibitor that can be further developed as SARS-CoV-2 antivirals.

The etiological agent of COVID-19 is SARS-CoV-2, a single-stranded, positive-sense RNA virus that belongs to the β-coronavirus genera. Given the catastrophic impact of COVID-19 on public health and global economy, researchers around the globe are working relentlessly to develop vaccines and antiviral drugs. This effort led to the approval of vaccines and antiviral drugs at a record-breaking speed. Two mRNA vaccines from Moderna and Pfizer and one adenovirus-based vaccine from Johnson & Johnson were approved by FDA.[1]

Although vaccines are the mainstay in combating the pandemic, antiviral drugs are nevertheless needed as complementary strategies. Vaccines are preventative, while antiviral drugs can be used for the treatment of COVID patients. In addition, the mRNA vaccines target the viral spike protein, which is prone to mutation as shown by the variants of concerns including the Delta variant and the most recent Omicron variant.[2] As a result, vaccines might need to be frequently updated to match the circulating strains. In comparison, small-molecule antiviral drugs targeting the conserved viral proteins are expected to have broad-spectrum antiviral activity and a high genetic barrier to drug resistance. The viral RNA-dependent RNA polymerase (RdRp) inhibitor remdesivir is the first FDA-approved COVID drug.[3] In addition, the second RdRp inhibitor molnupiravir[4−6] and the main protease (Mpro) inhibitor PF-07321332 (Nirmatrelvir in Paxlovid)[7] are FDA-approved specific oral COVID drugs.

Despite the encouraging progress, additional antiviral drugs with a novel mechanism of action are still in dire need to override the emergence of new mutations. They can be used either alone or in combination with existing RdRp inhibitors or Mpro inhibitors to combat not only the current COVID-19 pandemic but also future coronavirus outbreaks.

SARS-CoV-2 expresses two viral proteases, the Mpro and papain-like protease (PLpro), during viral replication. Both Mpro and PLpro are cysteine proteases that mediate the cleavage of viral polyprotein during viral replication.[8] In addition, PLpro dysregulates the host immune responses by cleaving ubiquitin and interferon-stimulated gene 15 protein (ISG15) from host proteins.[9] Therefore, inhibiting PLpro is a two-pronged approach to protecting host cells from viral infection.

PLpro is a 35 kDa domain withinNsp3, which is a 215 kDa multidomain protein that is a key component of the viral replication complex.[10] Compared to PLpro from SARS-CoV, SARS-CoV-2 PLpro displays decreased deubiquitination activity and enhanced deISGlyation activity.[9,11]

In contrast to Mpro, PLpro is a more challenging drug target mainly for two reasons. First, the protein substrate of PLpro consists of LXGG.[12] Accordingly, there is a lack of drug binding pockets in the S1 and S2 subsites. As such, a majority of reported PLpro inhibitors are noncovalent inhibitors that bind to the S3 and S4 subsites that are located more than 10 Å away from the catalytic cysteine C111.[13−15] Second, PLpro cleaves the same substrate sequence LXGG as the human deubiquitinase,[16] which presents a challenge in developing selective PLpro inhibitors. Despite extensive high-throughput screening and lead optimization,[11,13−15,17,18] GRL0617 and its analogues remain the most potent PLpro inhibitors reported so far. To identify structurally novel PLpro inhibitors, we conducted a drug-repurposing screening and identified EACC, KY-226, and tropifexor as potent PLpro inhibitors with IC50 values ranging from 3.39 to 8.28 μM. EACC is a reversible autophagy inhibitor.[19] KY-226 is a potent, selectivity, and orally bioavailable allosteric protein tyrosine phosphatase 1B (PTP1B) with an IC50 of 0.25 μM.[20] Tropifexor is a highly potent agonist of the farnesoid X receptor and is currently undergoing phase II clinical trial for nonalcoholic steatohepatitis (NASH) and liver fibrosis.[21] Their antiviral mechanism of action was further characterized in the thermal shift assay and the FlipGFP protease assay. Gratifyingly, tropifexor also had potent antiviral activity against SARS-CoV-2 in Calu-3 cells with an EC50 of 4.03 μM. Overall, tropifexor represents a potent PLpro inhibitor with a novel scaffold that can be further developed as SARS-CoV-2 antivirals.

Results and Discussion

High-Throughput Screening of SARS-CoV-2 PLpro Inhibitors

Using the previously optimized FRET assay condition,[15] we performed a high-throughput screening of SARS-CoV-2 PLpro against the MedChemExpress bioactive compound library, which consists of 9,791 compounds including FDA-approved drugs, clinical candidates, and natural products. The assay was performed in a 384-well plate with a Z′ of 0.688, and GRL0617 was included as the positive control. All compounds were originally screened at 40 μM, and hits showing more than 50% inhibition were further titrated to determine the IC50 values. GRL0617 was included as a positive control. In total, three compounds, EACC, KY-226, and tropifexor (Figure A), were identified as positive hits with IC50 values of 8.28, 3.39, and 5.11 μM, respectively (Figure B). In comparison, the IC50 value for the positive control GRL0617 was 1.66 μM (Figure B). Next, the broad-spectrum activity of the three hits was tested against SARS-CoV PLpro (Figure C) and MERS-CoV PLpro (Figure D). It was found that EACC, KY-226, and tropifexor retained potent inhibition against SARS-CoV PLpro with IC50 values of 6.28, 3.53, and 5.54 μM, respectively (Figure C). In contrast, EACC and KY-226 were weak inhibitors of MERS-CoV PLpro with IC50 values of 27.8 and 30.6 μM, while GRL0617 was inactive (IC50 > 60 μM) (Figure D). Nevertheless, tropifexor showed higher potency against MERS-CoV PLpro with an IC50 of 2.32 μM (Figure D). The hits were further counterscreened against the SARS-CoV-2 Mpro to rule out promiscuous cysteine protease inhibitors.[22−25] It was found that EACC and KY-226 were not active (IC50 ≥ 60 μM), while tropifexor had weak inhibition with an IC50 of 43.65 μM, which corresponds to a selectivity index (SI) of 8.5 (Figure E). These results suggest that the inhibition of SARS-CoV-2 PLpro by tropifexor is specific. The inhibition of PLpro’s deubiquitination and deISGlyation activities was characterized using the Ub-AMC and ISG15-AMC substrates, respectively.[14,15] While EACC and KY-226 were inactive in inhibiting the deubiquitinase activity of PLpro (IC50 > 100 μM), tropifexor showed moderate activity with an IC50 of 18.85 μM (Figure F). Similarly, EACC and KY-226 were not active in inhibiting the deISGlyation activity of PLpro (IC50 > 80 μM), tropifexor showed does-dependent inhibition with an IC50 of 27.22 μM (Figure G). Tropifexor is a hydrophobic compound with a C log P of 5.69. To rule out the possibility that the observed PLpro inhibition was due to nonspecific binding, we repeated the FRET assay against SARS-CoV-2 PLpro in the presence of 0.01% BSA, and it was found that tropifexor retained potent inhibition with an IC50 of 10.36 μM (Figure H), suggesting that the inhibition of PLpro by tropifexor is unlikely due to nonspecific hydrophobic interactions. Tropifexor had similar IC50 values against SARS-CoV-2 PLpro with and without a 30 min preincubation (Figure I), suggesting a reversible binding. The mechanism of inhibition of tropifexor was further studied in an enzymatic kinetic experiment, and GRL0617 was included as a control. The Lineweaver–Burk plots showed that both GRL0617 and tropifexor are competitive inhibitors of SARS-CoV-2 PLpro (Figure J,K).

Figure 1

Characterization of SARS-CoV-2 PLpro inhibitors identified from the high-throughput screening. (A) Chemical structures of the positive control GRL0617 and the three hits EACC, KY-226, and tropifexor. (B) IC50 curves of the hits in inhibiting SARS-CoV-2 PLpro with the FRET peptide substrate 1. (C) IC50 curves of the hits in inhibiting SARS-CoV PLpro with the FRET peptide substrate 1. (D) IC50 curves of the hits in inhibiting MERS-CoV PLpro with the FRET peptide substrate 1. (E) IC50 curves of the hits in inhibiting SARS-CoV-2 Mpro with the FRET peptide substrate 2. (F) IC50 curves of the hits in inhibiting SARS-CoV-2 PLpro with the Ub-AMC substrate. (G) IC50 curves of the hits in inhibiting SARS-CoV-2 PLpro with the ISG15-AMC substrate. Please refer to the Materials and Methods section for assay conditions. Values represent the average ± standard deviation of three replicates. (H) IC50 curves of tropifexor in inhibiting SARS-CoV-2 PLpro with and without the addition of 0.01% BSA. (I) IC50 curves of tropifexor in inhibiting SARS-CoV-2 PLpro with or without a 30 min preincubation. (J) Lineweaver–Burk curves of GRL0617 in inhibiting SARS-CoV-2 PLpro. (K) Lineweaver–Burk curves of tropifexor in inhibiting SARS-CoV-2 PLpro.

Overall, tropifexor appears to be the most promising hit with consistent inhibition against SARS-CoV-2, SARS-CoV, and MERS-CoV PLpros. In addition, tropifexor also inhibited the deubiquitination and deISGlyation activities of SARS-CoV-2 PLpro, albeit at lower potency.

Pharmacological Characterization of the Hits in the Thermal Shift Assay and the Cell-Based FlipGFP PLpro Assay

The mechanism of action of EACC, KY-226, and tropifexor in inhibiting SARS-CoV-2 PLpro was further characterized by the thermal shift assay and the cell-based FlipGFP PLpro assay.[15,22,23,26] Thermal shift assay measures the direct binding between the compound and the protein; therefore, it can rule out hits that might bind to the FRET substrate in the enzymatic assay. Similar to the positive control GRL0617, all three hits displayed dose-dependent binding to PLpro, as revealed by the enhanced melting temperatures with increasing drug concentrations (Figure ).

Figure 2

Thermal shift assay of SARS-CoV-2 PLpro protease against identified inhibitors. All inhibitors display a dose-dependent melting temperature (Tm) shift. Values represent the average ± standard deviation of three replicates.

Next, we tested the three hits in the FlipGFP PLpro assay.[15,22,23] The FlipGFP PLpro was recently developed by us as a surrogate assay to quantify the cellular activity of PLpro inhibitors in the biological safety level 2 facility, and we have shown that there is a positive correlation between the FlipGFP IC50 values with the SARS-CoV-2 antiviral EC50 values.[15] The FlipGFP assay is a virus-free cell-based protease assay in which the 293T cells were transfected with PLpro and the GFP reporter. The GFP reporter consists of two fragments,[27,28] the β1–9 template and the β10-11 strands that are constrained in the parallel inactive conformation through a PLpro substrate linker. Upon cleavage of the substrate linker, the β10 and β11 strands become parallel and can associate with the β1–9 template, leading to increased GFP signal. mCherry is included as an internal control to normalize transfection efficacy and compound cytotoxicity. In principle, the normalized GFP/mCherry ratio is proportional to the enzymatic activity of PLpro. The advantage of the FlipGFP assay over the FRET assay is that it can rule out compounds that are cytotoxic, membrane-impermeable, and having off-target effects that prevent cellular on-target engagement.[22,23]

In the FlipGFP assay, the positive control GRL0617 showed dose-dependent inhibition with an IC50 of 14.67 μM, while the negative control GC376 was not active (IC50 > 60 μM) (Figure A,B). The results from EACC and KY-226 were not conclusive due to the cell cytotoxicity of the compounds. Tropifexor had an IC50 of 10.60 μM but a low selectivity index (CC50 = 29.77 μM, SI = 2.8) (Figure A,B). Given the low selectivity, the results from the FlipGFP are not stringently conclusive. Nevertheless, tropifexor reduced the GFP/mCherry ratio by 50% at 10 μM, which was not cytotoxic. In summary, the FlipGFP assay results suggest that tropifexor might have antiviral activity against SARS-CoV-2.

Figure 3

FlipGFP SARS-CoV-2 PLpro assay to determine the cellular protease inhibitory activity of identified inhibitors. (A) Representative images of FlipGFP PLpro assay with increasing concentrations of GRL0617 (positive control), GC376 (negative control), EACC, KY-226, and tropifexor. GRL0617 showed a dose-dependent decrease of GFP signal with the increasing drug concentration, while almost no GFP signal change was observed with the increasing concentration of negative control compound GC376. (B) Dose–response curves of the GFP/mCherry ratio with increasing drug concentrations. mCherry signal alone was used to calculate the transfection efficiency and compound cytotoxicity. All three hits displayed significant cytotoxicity at high drug concentrations. Values represent the average ± standard deviation of three replicates.

Antiviral Activity of Hits against SARS-CoV-2 in Calu-3 Cells

The antiviral activity of EACC, KY-226, and tropifexor in inhibiting SARS-CoV-2 replication in Calu-3 cells was tested using the immunofluorescence assay (Figure ). Calu-3 is TMPRSS2-positive and is a close mimetic of the human respiratory epithelial cells,[29] enabling it a widely accepted cell line for SARS-CoV-2 studies.[22,30] The positive control GRL0617 had an EC50 of 31.4 μM (Figure A). EACC did not show antiviral activity at nontoxic drug concentration (EC50 > 35 μM, CC50 = 35.29 μM) (Figure B). Gratifyingly, both KY-226 and tropifexor had improved antiviral activity against SARS-CoV-2 with EC50 values of 25.0 (Figure C) and 4.03 μM (Figure D), respectively. While KY-226 had a low selectivity index (SI = 1.65), tropifexor had a moderate selectivity window (SI = 6.97) and the observed antiviral activity was likely not caused by the cytotoxicity of the compound.

Figure 4

Antiviral activity of SARS-CoV-2 PLpro inhibitors GRL0617 (A), EACC (B), KY-226 (C), and tropifexor (D) against SARS-CoV-2 in Calu-3 cells. The results were quantified by immunofluorescence assay. Values represent the average ± standard deviation of three replicates.

Molecular Docking of EACC, KY-226, and Tropifexor in SARS-CoV-2 PLpro

To gain insights into the binding mode of the three hits, we performed molecular docking with Schrödinger Glide XP (extra precision) using the wild-type SARS-CoV-2 PLpro structure we recently solved (PDB: 7JRN).[15] The binding sites were calculated by site map, and the GRL0617 binding site was identified as the top-ranked binding site; therefore, it was selected for docking. GRL0617 was included as a positive control. The docking pose of GRL0617 was superimposable with the binding mode in the X-ray crystal structure (Figure A). Tropifexor, EACC, and KY-226 all fit snuggly into the U-shaped binding pocket that is covered by the BL2 loop where GRL0617 binds (Figure B–D). Among the three hits, tropifexor showed the most favorable binding pose with a Glide score of −4.085 (Figure B). The docking poses might provide a guidance for the following lead optimization.

Figure 5

Molecular docking of SARS-CoV-2 PLpro inhibitors GRL0617 (A), tropifexor (B), EACC (C), and KY-226 (D) in PLpro (PDB: 7JRN). The Glide scores are −7.161 (GRL0617), −4.085 (tropifexor), −3.794 (EACC), and −3.332 (KY-226).

Conclusions

Although PLpro is a validated antiviral drug target, the development of PLpro inhibitors falls behind Mpro and RdRp inhibitors. As of date, no PLpro inhibitors have been advanced to the in vivo animal model studies yet. The naphthalene compounds such as GRL0617 and its analogues are the only class of validated PLpro inhibitors with antiviral activity against SARS-CoV-2. However, the low metabolic stability of this series of compounds might prevent its further development.[14,31] In this study, we aimed to identify structurally novel PLpro inhibitors that can serve as starting points for further optimization. Through screening the MedChemExpress bioactive compound library, three hits EACC, KY-226, and tropifexor were identified as SARS-CoV-2 PLpro inhibitors with IC50 values in the single-digit micromolar range. Among the three hits, tropifexor appears to be the most promising hit as it also showed potent inhibition against SARS-CoV PLpro (IC50 = 5.54 μM) and MERS-CoV PLpro (IC50 = 2.32 μM). In addition to the inhibition of the PLpro-mediated cleavage of the viral polyprotein substrate, tropifexor also inhibited the deubiquitination and deISGlyation activities of SARS-CoV-2 PLpro. Consistent with the enzymatic inhibition, tropifexor showed a dose-dependent stabilization of SARS-CoV-2 PLpro in the thermal shift assay. Importantly, tropifexor displayed cellular PLpro inhibitory activity in the FlipGFP assay and the antiviral activity against SARS-CoV-2 in Calu-3 cells. Although the low selectivity index (SI = 6.2) of tropifexor in the antiviral assay prevents its direct repurposing as a SARS-CoV-2 antiviral, the discovery of tropifexor as a novel PLpro inhibitor provides an additional scaffold for further medicinal chemistry optimization. Follow-up studies will focus on improving the target and cellular selectivity. Furthermore, tropifexor is a fairly large molecule (MW: 603.59); efforts will be made to reduce the size as well as the hydrophobicity of the compound to optimize ligand efficiency and druglikeness properties.

Materials and Methods

Protein Expression and Purification

Detailed expression and purification procedures untagged SARS-CoV-2 PLpro and SARS-CoV-2 Mpro were described in our previous publications.[15,32] SARS-CoV papain-like protease gene (ORF 1ab 1541-1855) (accession # AEA10621.1) from strain SARS coronavirus MA15 with Escherichia coli codon optimization in the pET28b-(+) vector was ordered from GenScript. Then, the SARS-CoV PLpro gene (ORF 1ab 1541-1855) was subcloned from the pET28b-(+) to pE-SUMO vector according to the manufacturer’s protocol (LifeSensors Inc., Malvern, PA). The forward primer with the Bsa I site is GCGGTCTCAAGGTGAGGTGAAGACCATCAAAGTGTTCACCACC; the reverse primer with a Bsa I site is GCGGTCTCTCTAGATTATTTAATGGTGGTGGTATAGCTGGTTTCCTTGTAG. The expression and purification protocol of SARS-CoV PLpro are identical to those of SARS-CoV-2 PLpro.[15]

MERS-CoV PLpro gene (ORF 1ab 1482-1803) (accession # KY581684) from strain MERS coronavirus Hu/UAE_002_2013 with E. coli codon optimization in the pET28b-(+) vector was ordered from GenScript. Then, the MERS-CoV PLpro gene (ORF 1ab 1482-1803) was subcloned into the pE-SUMO vector with the pair primers: GCGGTCTCAAGGTCAGCTGACCATCGAGGTGCTGGTTACCGTGG and GCGGTCTCTCTAGATTAGTTGCAATCGCTGCTATATTTTTGACCCGGGAAC. The expression and purification protocol of MERS-CoV papain-like protease are identical to those of SARS-CoV-2 PLpro.[15]

FRET Substrate Synthesis

The SARS-CoV-2 PLpro FRET substrate 1 is Dabcyl-FTLRGG/APTKV(Edans); this substrate was also used as SARS-CoV PLpro and MERS-CoV PLpro substrates. SARS-CoV-2 Mpro FRET substrate 2 is Dabcyl-KTSAVLQ/SGFRKME- (Edans). These FRET substrates were synthesized by solid-phase synthesis through iterative cycles of coupling and deprotection using the previously optimized procedure.[33] Ub-AMC and ISG15-AMC were purchased from BostonBiochem (catalog nos. U-550-050 and UL-553-050, respectively).

Enzymatic Assays

The high-throughput screening was carried out in 384-well format, as described previously.[15] The bioactive compound library consisting of 9,791 compounds was purchased from MedChemExpress (catalog no. HY-L001). The enzymatic reactions for SARS-CoV-2, SARS-CoV, and MERS-CoV PLpros were carried out in a reaction buffer consisting of 50 mM HEPES pH 7.5, 5 mM DTT, and 0.01% Triton X-100. For the IC50 measurement with the FRET peptide–Edans substrate, the reaction was carried out in 96-well format with a 100 μL reaction volume. SARS-CoV-2 PLpro (200 nM), SARS-CoV PLpro (200 nM), or MERS-CoV PLpro (2 μM) was preincubated with various concentrations of testing compounds at 30 °C for 30 min before the addition of the FRET peptide substrate to initiate the reaction. The reaction was monitored in a Cytation 5 image reader with filters for excitation at 360/40 nm and emission at 460/40 nm at 30 °C for 1 h. The initial enzymatic reaction velocity was calculated from the initial 10 min enzymatic reaction via a linear regression function and was plotted against the substrate concentrations in Prism 8 with a four-parameter dose–response function. For the IC50 measurements with Ub-AMC or ISG15-AMC substrate, the reaction was carried out in 384-well format in a 50 μL reaction volume. In the Ub-AMC cleavage assay, the final SARS-CoV-2 PLpro concentration is 50 nM, and the substrate Ub-AMC concentration is 2.5 μM. In the ISG15-AMC assay, the final SARS-CoV-2 PLpro concentration is 2 nM, and the substrate ISG15-AMC concentration is 0.5 μM. The SARS-CoV-2 Mpro enzymatic assays were carried out in the reaction buffer containing 20 mM HEPES pH 6.5, 120 mM NaCl, 0.4 mM EDTA, 20% glycerol, and 4 mM DTT, as described previously.[32,34]

To rule out that the inhibition of tropifexor on PLPro is due to aggregation, 200 nM PLPro was incubated with serial concentrations of tropifexor (0, 0.1, 0.3, 1, 3, 10, 30, 100 μM) in the reaction buffer in the presence or absence of 0.01% BSA (0.1 mg/mL) at 30 °C for 30 min. The reaction was initiated by adding a 10 μM FRET substrate and monitored every 90 s for 1 h at 30 °C. The initial velocity was determined in the first 15 min by linear regression. The IC50 values were determined by fitting the curves with nonlinear regression using log (concentration of inhibitor) vs response with variable slopes in Prism 8.

To determine whether preincubation affects the IC50 value of tropifexor, 200 nM PLPro was mixed with serial concentrations of tropifexor (0, 0.1, 0.3, 1, 3, 10, 30, 100 μM) in the reaction buffer with or without preincubation at 30 °C for 30 min, and the reaction was initiated by adding a 10 μM FRET substrate. IC50 values were determined as previously described.

To determine the binding mode of tropifexor, KM and Vmax were determined at different concentrations of GRL0617 (0, 0.3, 1, 3, 10 μM) or tropifexor (0, 1, 3, 10, 30 μM). SARS-CoV-2 PLPro (200 nM) was mixed with the indicated concentrations of GRL0617 or tropifexor in the reaction buffer and incubated at 30 °C for 30 min. The reaction was initiated by adding different concentrations of FRET peptides (5, 10, 25, 50, 100, 200 μM). Michaelis–Menten and Lineweaver–Burk curves were plotted in Prism 8.

Differential Scanning Fluorimetry (DSF)

The thermal shift assay (TSA) was carried out using a Thermo Fisher QuantStudio 5 real-time PCR system, as described previously.[15,32] Briefly, 4 μM SARS-CoV-2 PLpro protein in the PLpro reaction buffer (50 mM HEPES pH 7.5, 5 mM DTT, and 0.01% Triton X-100) was incubated with various concentrations of testing compounds at 30 °C for 30 min. A 1× SYPRO orange dye was added, and the fluorescence of each well was monitored under a temperature gradient range from 20 to 90 °C with a 0.05 °C/s incremental step. The melting temperature (Tm) was calculated as the mid-log of the transition phase from the native to the denatured protein using a Boltzmann model in Protein Thermal Shift Software v1.3.

Cell-Based FlipGFP PLpro Assay

Plasmid pcDNA3-PLpro-flipGFP-T2A-mCherry was constructed from pcDNA3-TEV-flipGFP-T2A-mCherry.[15] SARS-CoV-2 PLpro expression plasmid pcDNA3.1-SARS2 PLpro was ordered from Genscript (Piscataway NJ) with codon optimization. For transfection, 293T cells were seeded into a 96-well Greiner plate (catalog no. 655090) overnight with 70–90% confluency; 50 ng of pcDNA3-PLPro-flipGFP-T2A-mCherry plasmid and 50 ng of protease expression plasmid pcDNA3.1-PLpro were added to each well in the presence of a transfection reagent TransIT-293 (Mirus) according to the manufacturer’s protocol. Three hours after transfection, 1 μL of the testing compound was added to each well at 100-fold dilution. Images were acquired 2 days after transfection with a Cytation 5 imaging reader (Biotek) GFP and mCherry channels and were analyzed with Gen5 3.10 software (Biotek). SARS-CoV-2 PLpro protease activity was calculated by the ratio of the GFP signal over the mCherry signal. The FlipGFP PLpro assay IC50 value was determined by plotting the GFP/mCherry signal over the compound concentration with a four-parameter dose–response function in Prism 8. The mCherry signal alone was utilized to evaluate the transfection efficiency and compound cytotoxicity.

Antiviral Assay in Calu-3 Cells

Calu-3 cells (ATCC, HTB-55) grown in minimal Eagle’s medium supplemented with 1% nonessential amino acids, 1% penicillin/streptomycin, and 10% FBS are plated in 384-well plates. The next day, 50 nL of drug suspended in DMSO is added as an 8-pt dose–response with 3-fold dilutions between test concentrations in triplicate, starting at 40 μM final concentration. The negative control (DMSO, n = 32) and positive control (10 μM remdesivir, n = 32) are included on each assay plate. Calu-3 cells are pretreated with controls and test drugs (in triplicate) for 2 h prior to infection. In BSL3 containment, SARS-CoV-2 (isolate USA-WA1/2020) diluted in a serum-free growth medium is added to plates to achieve an MOI = 0.5. Cells are incubated continuously with drugs and SARS-CoV-2 for 48 h. Cells are fixed and then immunostained with anti-dsRNA (J2), and nuclei are counterstained with Hoechst 33342 for automated microscopy. Automated image analysis quantifies the number of cells per well (toxicity) and the percentage of infected cells (dsRNA+ cells/cell number) per well. SARS-CoV-2 infection at each drug concentration was normalized to aggregated DMSO plate control wells and expressed as percentage of control (POC = % infection sample/Avg % infection DMSO cont). A nonlinear regression curve fit analysis (GraphPad Prism 8) of POC infection and cell viability versus the log 10 transformed concentration values to calculate EC50 values for infection and CC50 values for cell viability. Selectivity index (SI) was calculated as a ratio of drug’s CC50 and EC50 values (SI = CC50/IC50).

Molecular modeling of the binding of EACC, KY-226, and tropifexor to SARS-CoV-2 PLpro. Docking was performed using Schrödinger Glide extra precision (XP). The SARS-CoV-2 PLpro structure was downloaded from the PDB code 7JRN. The binding sites were calculated by the site map, and the GRL0617 binding site is the highest-scored binding site, and therefore, it was chosen for docking. The docking grid was centered around GRL0617 with the coordinates of X = 9.88, Y = −11.74, and Z = 32.55. GRL0617 was added as a positive control for the docking. The final docking poses were generated in PyMOL.