Polydatin and resveratrol were purchased from solarbio company (Solarbio Science & Technology Company, Beijing, China). The Trans1-T1 strain, F-$\phi$80(lacZ)$△$M15$△$lacX74hsdR(rk,m)$△$recA 1398endAltonA, (TransGen Biotech, Beijing, China) was used to clone and propagate plasmid DNA. Miniprep and Maxiprep kits (Axygen, San Jose, CA, USA) were used to harvest and purify plasmid DNA. SARS-CoV-2, SARS-CoV, and MERS-CoV 3CLpro, and PLpro were purchased from novoprotein company (Novoprotein company, Beijing, China) or expressed and purified by genescript company (Genescript company, Nanjing, China). Biacore Sensor Chip CM5 and phosphate-buffered saline (PBS)-P were purchased from GE healthcare company (GE healthcare company, Uppsala, Sweden). The 12-mer fluorogenic peptide Dabcyl-KNSTLQSGLRKE-Edans substrate for the 3CLpro inhibition assay and fluorogenic peptide Dabcyl-KRLKGGAPIKGE-Edans substrate for the PLpro inhibition assay were sythesised by genescript company (Genescript company, Nanjing, China).

HCoV-OC43(OC43) strain was purchased from American Type Culture Collection (Human Coronavirus OC43(ATCC VR-1558) and propagated using RD cells (ATCC CCL-136). OC43 stock virus was obtained in RD cells in DMEM supplemented with 2% FBS at 72 h post-infection (hpi). Viral titer (PFU/mL) was determined by plaque assay.

RD cells were infected OC43 at the multiplicity of infection (MOI = 0.01 IU) in the presence of indicated compound diluted in DMEM supplemented with 2% FBS [23]. The compounds and the virus were maintained with the cells during the 48 h incubation at 37 ${}^{\circ }$C. OC43 infection efficiency in RD cell was evaluated by immunofluorescence staining against the Nucleoprotein. In general, RD cells were fixed by 100% methanol at 48 hpi, blocked by 1% BSA/PBS at 37 ${}^{\circ }$C for 1 h, stained with anti-OC43 Np mAb (Sigma, Saint Louis, MO, USA) at 1:1000 at 37 ${}^{\circ }$C for 1 h, and followed by secondary antibody incubation for another 1 h. Nuclei were stained with Hoechst 33342 (Thermo Scientific, Waltham, MA, USA) at 1:5000 for 15 min. Images were captured by an Mshot fluorescence microscrope (Olympus, Tokyo, Japan).

The median cytotoxic concentration (CC${}_{50}$) of the compounds was determined by the CellTiter-Glo 2 (Promega Corporation, Madison, WI, USA) cell viability assay kit [24]. In general, 293T-hACE2 cells were seeded in 96-well plates with three-fold serial-diluted compounds (from 100 $\mu$M to 0.137 $\mu$M). After 24 h, the cell viability was detected by CellTiter-Glo 2 reagent by following the manufacture’s instructions, and the luminescence was determined by a Spectra MaxiD3 multi-well Luminometer 458 (Molecular Devices, San Jose, CA, USA). The CC${}_{50}$ value was determined by nonlinear regression analysis [25].

The SARS-CoV-2 3CLP (pdb:6lu7) and PLP (pdb:6w9c) program database (PDB) files were downloaded from the PDB website (http://www.rcsb.org/). All heterogeneous atoms and the 3CLP ligand were removed and 6lu7 chain A was selected for subsequent molecular docking. The 3CLP docking grid was maximized for polydatin docking. All heterogeneous atoms and the PLP ligand were removed and 6w9c chain A was selected for subsequent molecular docking. The PLP docking grid was maximized for polydatin docking.

PDB file (6lu7 chain A) and (6w9c chain A) were converted to the PDBQT format as macromolecules before virtual screening. The grid (ligand docking search space) was located as described above. Then, Autodock Vina 1.1.2 [26] was used for the subsequent molecular docking. Protein–ligand interactions were visualized using Pymol version 1.7.4.5 (Schrödinger, New York, NY, USA). The amino acid residues of 3CLP protein close to the hit ligands ($\le$1 Å) were highlighted as potential interactive residues involved in the protein–ligand interaction.

To conform the binding affinities of polydatin and resveratrol to SARS-CoV-2, SARS-CoV, MERS-CoV 3CLpro and PLpro, SPR technology based Biacore T200 biosensor was used (Biacore AB, Uppsala, Sweden) [27]. At 25 ${}^{\circ }$C, the 3CLpro and PLpro proteins were diluted to a final concentration of 20 $\mu$g/mL in 10 mM sodium acetate buffer (pH 4.5) and then immobilized to CM5 by using the standard primary amine coupling method. The final immobilization levels of 3CLpro of SARS-CoV-2, SARS-CoV, and MERS-CoV were 10,125, 10,452.5, and 9856 resonance units, respectively. The final immobilization levels of PLpro of SARS-CoV-2, SARS-CoV, and MERS-CoV were 15,847, 15,358, and 10,459 resonance units, respectively.

All the screening assays were performed over the unmodified dextran surface and the protein surface. During the 3CLpro and PLpro binder assay, the sample concentrations were set at 100 $\mu$M, 50 $\mu$M, 25 $\mu$M, 12.5 $\mu$M, 6.25 $\mu$M, 3.125 $\mu$M, 1.5625 $\mu$M, 0.78125 $\mu$M, 0 $\mu$M. For sample dilution, the pH and the concentration of both DMSO and buffer substances in samples and running buffer were carefully matched. Each sample assay consisted of a 180-s buffer injection and a 300-s dissociation phase and the blank injection was used to check the carryover effects. The signal was adjusted for nonspecific binding of the samples to dextran matrix by subtracting the signal in the reference channel from the signal in the active channel. The experimental data were fitted and analyzed using the BIA evaluation software by steady state analysis.

To further investigate possible inhibitory activities of polydatin and resveratrol against SARS-CoV-2, SARS-CoV, and MERS-CoV 3CLpro, and PLpro, the inhibitory effects of the compounds were tested by the FRET method [28]. For this assay, the internally quenched fluorogenic substrate Dabcyl-KNSTLQSGLRKEE-Edans and Dabcyl-KRLKGGAPIKGE-Edans substrate (synthesized by genescript company) was applied and the compounds were diluted to 100 mM, 20 mM, 10 mM, 2 mM, 1 mM, 200 $\mu$M, 100 $\mu$M, and 20 $\mu$M. A 1.2 $\mu$L concentration of the solution of the compounds was taken and thoroughly mixed with 120 $\mu$L of 2 $\mu$M 3CL protease (final concentration of DMSO is 0.5%) and incubated at 4 ${}^{\circ }$C for 2 h. Thereafter, 100 $\mu$L of each concentration sample was taken into a 96-well plate, and then added 100 $\mu$L of 20 $\mu$M fluorescent substrate solution to start the reaction. At this time, the final concentration of the 3CL or PL protease is 1 $\mu$M, the final concentration of the fluorescent substrate is 10 $\mu$M, and the final concentrations of the compounds are 500 $\mu$M, 100 $\mu$M, 50 $\mu$M, 10 $\mu$M, 5 $\mu$M, 1 $\mu$M, 0.5 $\mu$M, and 0.1 $\mu$M, respectively. The enhanced fluorescence emission upon substrate cleavage was monitored at the excitation and emission wave lengths of 340 and 488 nm. The reaction was measured continuously for 1 h. According to the logistic formula, the origin software was used to perform a non-linear fitting to calculate the IC${}_{50}$ value of compounds in the inhibition of 3CL protease activity. The formula is as follows:

(1)$$A(I)/A_0=1-\frac{1}{1+{\left(I/I{C}_{50}\right)}^p}$$

In the above formula, A${}_0$ refers to the enzyme activity when 0.5% DMSO is added, A(I) refers to the enzyme activity under different concentrations of compounds, I refers to the concentration of the compound, and p refers to the u factor.

To verify the inhibitory effect of polydatin and resveratrol on coronavirus at the cellular level in vitro, we measured the antiviral activity of polydatin and resveratrol by indirect immunofluorescence assay (IFA) on RD cells infected with OC43-CoV which is an alternative model for SARS-CoV-2. As shown in Fig. 1A and 1B, polydatin and resveratrol significantly suppressed HCoV-OC43 replication, compared with that of the control (DMSO), and with over 90% inhibitory effect at a concentration of 100 $\mu$M, respectively. Cytotoxicity of these drugs to HEK293T cells was measured by CellTiter-Blue assay. As shown in Fig. 2A and 2B, cytotoxicity concentration 50% [CC${}_{50}$] of polydatin and resveratrol was 420.6 $\mu$mol/L and 182.4 $\mu$mol/L, respectively.

Fig. 1.

Confirmation of anti-coronavirus activity by immunofluorescence assay (IFA) analysis. The IFA of the human coronavirus strain OC43 (HCoV-OC43) nucleocapsid (N) protein in inhibitor-treated RD cells. RD cells in 96-well plates were infected with HCoV-OC43-wild type (multiplicity of infection = 0.01 IU) in serial dilutions of tested compounds, with dimethyl sulfoxide used as the negative controls. At 48 h post infection, the cells were analyzed by IFA for N protein expression. Nuclei (blue) were stained with Hoechst 33342.

Fig. 2.

Cytotoxicity of polydatin (A) and resveratrol (B) to HEK293T-ACE2 cells was measured by CellTiter-Blue assay. The median cytotoxic concentration value was determined by nonlinear regression analysis.

To elucidate the mechanism of polydatin and resveratrol on the inhibition of virus replication, we combined virtual screening with experimental studies. The chain A of 6lu7 (SARS-CoV-2 3CLP) was extracted to perform molecular docking with polydatin. For SARS-CoV-2 3CLP, polydatin binds to the PHE140, HIS163, MET165, GLU166, PRO168, and GLN189 amino acid sites (Fig. 3A). The chain A of 6w9c (SARS-CoV-2 PLP) was extracted to perform molecular docking with polydatin. For SARS-CoV-2 PLP, polydatin binds to the GLU67, GLU70, TYR71, PHE127, PRO130, and GLN133 amino acid sites (Fig. 3B).

Fig. 3.

Prediction of binding sites by virtual screening. (A) Polydatin binds to the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 3CLP (6lu7). Polydatin (gray stick) docked superposition on SARS-CoV-2 3CLP. Interactive amino acids (orange sticks) with polydatin (gray stick) on the SARS-CoV-2 3CLP (red dash line indicates the hydrogen bonds). (B) Polydatin binds to the SARS-CoV-2 PLP (6w9c). Polydatin (gray stick) docked superposition on the SARS-CoV-2 PLP, Interactive amino acids (orange sticks) with polydatin (gray stick) on the SARS-CoV-2 PLP (red dash line indicates the hydrogen bonds).

With the established FRET assay condition, we tested polydatin and resveratrol to identify its inhibition in the proteolytic activity of 3CLpro and PLpro (Table 1). Polydatin inhibited SARS-CoV-2 Mpro and PLpro with 50% inhibitory concentration (IC${}_{50}$) values of 18.66 and 125 $\mu$M, respectively (Table 1, Fig. 4A,B), MERS Mpro and PLpro with 50% inhibitory concentration (IC${}_{50}$) values of 14.6 and 25.42 $\mu$M (Table 1, Fig. 4C,D), respectively. resveratrol inhibited SARS-CoV-2 Mpro and PLpro with 50% inhibitory concentration (IC${}_{50}$) values of 29.81 and 60.86 $\mu$M, respectively (Table 1, Fig. 4A,B), MERS Mpro and PLpro with 50% inhibitory concentration (IC${}_{50}$) values of 16.35 and 19.04 $\mu$M (Table 1, Fig. 5A,B), respectively.

Fig. 4.

Determination of the 50% inhibitory concentration (IC${}_{\mathrm{50}}$) of polydatin and resveratrol for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 3CL(A), the Middle East respiratory syndrome coronavirus (MERS-CoV) 3CL(B), SARS-CoV-2 PLP(C), and MERS-CoV PLP(D) by fluorescence resonance energy transfer (FRET)-based assay.

Fig. 5.

Determination of the 50% inhibitory concentration (IC${}_{\mathrm{50}}$) of polydatin and resveratrol for the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 3CL (A), the Middle East respiratory syndrome coronavirus (MERS-CoV) 3CL (B), SARS-CoV-2 PLP (C), and MERS-CoV PLP (D) by fluorescence resonance energy transfer (FRET)-based assay.

To confirm the interaction between polydatin and resveratrol and SARS-CoV-2, SARS-CoV, MERS-CoV 3CLpro, and PLpro protein, we tested whether polydatin and resveratrol had high affinity to target protein using the SPR assay. The SARS-CoV-2, SARS-CoV, and MERS-CoV 3CLpro and PLpro protein were immobilized separately on a CM5 chip with the tested compounds flowing across their surface. We found that polydatin bound to SARS-CoV-2, SARS-CoV, and MERS-CoV 3CLpro and PLpro protein exhibiting a strong dose-dependent response, with respective KD (equilibrium dissociation constant) values of 16.01 $\mu$M, 5.428 $\mu$M, 16.44 $\mu$M, 39.53 $\mu$M, 1.976 $\mu$M, and 7.745 $\mu$M (Table 2, Fig. 6A–E); resveratrol bound to SARS-CoV-2, SARS-CoV, and MERS-CoV 3CLpro and PLpro protein exhibiting a strong dose-dependent response, with respective KD (equilibrium dissociation constant) values of 0.9812 $\mu$M, 92.51 $\mu$M, 7.312 $\mu$M, 12.35 $\mu$M, 1.555 $\mu$M, and 6.1 $\mu$M (Table 3, Fig. 7A–E). These results suggested that polydatin and resveratrol had a specific affinity to SARS-CoV-2, SARS-CoV 3CLpro and the PLpro protein, and thus can inhibit virus replication by binding to the active center of 3CLpro and the PLpro protein.

Fig. 6.

Polydatin targets 3CLpro and PLpro of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), severe acute respiratory syndrome coronavirus (SARS-CoV), and Middle East respiratory syndrome coronavirus (MERS-CoV) (A–F). The sensor image of 3CLpro and PLpro of SARS-CoV-2, SARS-CoV, and MERS-CoV with polydatin.

Fig. 7.

Resveratrol targets 3CLpro and PLpro of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), severe acute respiratory syndrome coronavirus (SARS-CoV), and Middle East respiratory syndrome coronavirus (MERS-CoV) (A–F). The sensor image of 3CLpro and PLpro of SARS-CoV-2, SARS-CoV, and MERS-CoV with resveratrol.