Vero and porcine small intestinal epithelial cell line (IPEC-J2) cells were maintained in Dulbecco’s modified eagle medium (DMEM; Gibco, Grand Island, NE, USA) containing 10% (v/v) fetal cattle serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin mixtures at 37°C with 5% CO₂. PEDV isolate AH2012/12 (GenBank accession KU646831) was donated by researcher Li Bin of the Jiangsu Academy of Agricultural Sciences.

Cell viability was measured using the cell counting kit-8 (CCK8) (Meibio, Beijing, China) assay, following the manufacturer’s instructions. The Vero and IPEC-J2 cells were seeded into 96-well plates containing wogonin (6.25–400 μM). Subsequently, the cells were incubated with wogonin for 48 h and then washed thrice with PBS. The culture supernatant in each well was replaced with 110 µL of DMEM containing 9.09% CCK-8 reagent and incubated in the dark for 2 h. After incubation, the cells were gently shaken, and the optical density (OD450) of the culture was measured using a microplate reader. Cell viability (CC50) was calculated as follows: cell viability = (As − Ab) ÷ (Ac − Ab) × 100%, As indicates the OD450 value of the cells incubated with wogonin and CCK-8 solution, Ab denotes the OD450 values of the wells without Vero cells, incubated with wogonin and CCK-8 solution, and Ac represents the OD450 values of the cells incubated without wogonin but with CCK-8 solution.

The Vero and IPEC-J2 cells were grown to approximately 80%–90% confluence in 6-well culture plates in the absence or presence of wogonin (12.5–100 μM) and infected with PEDV at 0.01 MOI (Vero) and 0.1MOI (IPEC-J2) with 5 μg/mL trypsin (Invitrogen) for 2 h. After washing the unbound virus with PBS, the cells were cultured in fresh medium in the absence or presence of wogonin (12.5–100 μM) with 5 μg/mL of trypsin at 37°C for 48 h until the cytopathic effect (CPE) became visible. Then, the cells were collected for viral RNA detection by real-time reverse transcription PCR (RT-qPCR), immunofluorescence assay (IFA), and a median tissue culture infectious dose (TCID50) assay. The wogonin concentration that achieved 50% protection was defined as the 50% inhibitory concentration (IC50). To calculate the IC50 values, the results were transformed to a percentage of the controls, the IC50 values were obtained from the RT-qPCR results, and the percent protection achieved by wogonin in the PEDV-infected cells was calculated. The therapeutic index was defined as CC50/IC50.

To determine the effect of wogonin on viral attachment, the Vero cells were infected with PEDV in the absence or presence of wogonin (50 μM) at 0.05 MOI at 4°C for 1 h. After washing 3 times with PBS, the cell lysates were harvested for RT-qPCR assay. For the viral internalization assay, the Vero cells were incubated with PEDV (MOI = 0.05) for 1 h at 4°C. After the unbound viruses were removed by washing thrice in PBS, the cells were treated with wogonin (50 μM) at 37°C for 1 h. The cell lysates were harvested for RT-qPCR assay. To investigate the effect of wogonin on PEDV replication, the Vero cells were infected with PEDV (MOI = 0.05) for 1 h at 37°C. Then, the cells were washed with PBS to remove non-internalized viruses and incubated with wogonin (50 μM) for 2 h at 37°C. The cell lysates were harvested for RT-qPCR. To determine the effect of wogonin on PEDV release, the Vero cells were infected with PEDV (MOI = 0.05) for 10 h at 37°C and then washed thrice with PBS and incubated with wogonin (50 μM) at 37°C for 12 h. The supernatants were harvested for RT-qPCR. To examine the effect of wogonin on PEDV inactivation, PEDV (MOI = 0.05) and wogonin (50 μM) were incubated together at 37°C for 2 h. Then, the mixture was washed with PBS and ultracentrifuged at 90,000 × g for 1.5 h at 4°C in 20% sucrose buffer (w/w) to purify the virions. Subsequently, the virions were resuspended in the culture medium, and the cell lysates were harvested for RT-qPCR (Liu et al., 2022).

The viral RNA from the cell suspensions was extracted using the Viral RNA Extraction Kit (#9766, TaKaRa, Shanghai, China), following the manufacturer’s instructions. The cDNA was obtained by RT-PCR with a PrimeScript™ RT Reagent Kit with a gDNA Eraser (TaKaRa, Shanghai, China). The absolute RT-qPCR assay used to quantify the PEDV genome was performed with RealUniversal SYBR Green Premix (Tiangen, Beijing, China). As an internal reference for the quantification of the PEDV copy numbers, the PEDV 186-fragment gene was cloned into the pMD19-T vector, and primers PEDV-186-F (5′-TACTAAGCGTAACATCCTGCC-3′) and PEDV-186-R (5′-GTAGTACCAATAACAACCGAAGC-3′) were employed.

To determine the TCID50 values of the PEDV, the viral stock solutions were serially diluted before inoculation with the confluent Vero cell monolayers grown in 96-well plates. The cells were washed thrice with PBS, and eight wells were inoculated with 100 µL of each dilution of wogonin. The plates were incubated at 37°C in 5% CO₂ for 2 days. Wells with syncytium formation, the specific CPE caused by AH2012/12, were classified as PEDV-positive. The TCID50 of the PEDV was calculated using the Reed–Muench method.

The Vero cells infected with PEDV at different MOI were seeded onto a 6-well plate, fixed with 4% paraformaldehyde for 20 min at room temperature (RT), and permeabilized with 0.3% ice-cold TritonX-100 in PBS for 10 min at RT. After washing thrice with PBS, the cells were blocked with PBS containing 1% bovine serum albumin for 30 min at RT, incubated with anti-PEDV N protein antibodies (developed by our laboratory)(Shen et al., 2020) in PBS for 1 h at 37°C, and again washed thrice with PBS. Subsequently, the cells were incubated with Alexa 488-labeled anti-mouse antibody (Antgene, Wuhan, China) for 45 min at 37°C, washed thrice, treated with DAPI dihydrochloride at RT for 10 min in the dark (Beyotime Biotechnol, Shanghai, China) to stain the nuclei, and analyzed using an indirect immunofluorescence assay (IFA).

The RCSB protein data bank (www.rcsb.org,PDBcode:4XFQ) was employed to obtain the X-ray crystallographic structure of the pure protein, Mpro, by removing the water, ions, and ligand. The 3D structure of the ligand, wogonin, was first constructed using Gaussview 5.0, and the geometry optimization of wogonin was performed using Gaussian 09 with the ab initio calculation method at the HF/6-31G (d) level.

AutoDock Tools was used to establish the docking, and molecular docking was simulated by AutoDock software. The binding site was defined by a rectangular box with explicit coordinates, and the box was adjusted to the center of the primary binding site. Three-dimensional grids were set as 90×90×90 points with a grid spacing of 1 Å for the protein, Mpro. The center of the grid was regarded as the mass center, and the exhaustiveness was 20. The ligand, wogonin, was docked to the protein, Mpro, by turn, and the corresponding energy evaluation was also generated. The most common and stable docking conformation with the lowest docking energy was selected as the initial binding mode for the binding mode analysis.

N terminal GST-tagged Mpro expression and purification were performed as described previously (Li et al., 2020) and then digested with PreScission Protease (Beyotime, Shanghai, China) and loaded onto a Glutathione Sepharose 4B column to remove the GST tag and GST-tagged PreScission Protease. A total of 1 μg/μL purified Mpro with 40 μM wogonin was subjected to nano differential scanning fluorimetry (DSF) (Prometheus NT.48) to measure the stability of Mpro, and the results were analyzed by PR. ThermControl software. Mpro and 1% DMSO PBS were used as the control, and the temperature was increased from 25°C to 95°C.

The binding of Mpro to wogonin was examined using a microscale thermophoresis (MST) assay. The concentration of red-labeled Mpro was maintained at 200 nM, while that of the unlabeled binding partner (wogonin) varied in the range of 40 μM-1.22 nM. Wogonin was diluted in an MST-optimized buffer containing 1% DMSO and PBS, and Mpro was added to 10 µL of different concentrations of wogonin and thoroughly mixed. Subsequently, the mixture was filled into capillaries (NanoTemper technologies, Germany) and assayed after a 10 min equilibration at RT. The measurements were performed under auto LED power and medium MST power, and the results were plotted, with the x-axis representing the nanomolar concentrations. The dissociation constant (kd) was determined using MO Affinity analysis software v2.3 (Shen et al., 2022).

Mpro was immobilized on activated COOH-sensor chips serving as the ligand on an OpenSPR system (Nicoya Lifesciences Inc., Kitchener, ON, Canada). Then, 5-40 μM wogonin was added to the sensor chip with PBS as the running buffer. The kinetic constants, including the association constant (ka), kd, and affinity (KD, KD = kd/ka), were calculated according to a 1:1 binding model.

A fluorescent peptide with the sequence of Dabcyl-YNSTLQ↓AGLRKM-E-Edans (Ye et al., 2016) was employed as the substrate for the PEDV Mpro enzymatic and inhibition assays based on the fluorescence resonance energy transfer (FRET) effect. The 10 µM substrate peptides were incubated in 100 µL solution containing 20 mM Tris-HCl, pH 7.5, 100 mM NaCl, and10 mM CaCl₂ at 37°C for 1 h with 0-8µg/mL purified Mpro to detect the enzymatic activity. The fluorescence emitted by the cleaved substrate was measured using the PerkinElmer VICTOR Nivo Microplate Reader (340 nm excitation, 485 nm emission). In addition, 8ug/ml Mpro was pre-incubated with 1.28nM-100µM wogonin at 37°C for 1 h, and the 10 µM substrate peptides were added to the mixture at 37°C for 1 h to test the inhibiting effect of wogonin on the PEDV Mpro. A reaction system without wogonin was used as the control. The fluorescence of the reactions was monitored according to the method mentioned above. The inhibition ratio was calculated using the following equation: percentage of inhibition (%) = 100 × [1 − fluorescence of the experimental group/the control group].

To determine the cytotoxicity of wogonin ( Figure 1A ) for its potential antiviral activities against PEDV, the Vero and IPEC-J2 cells were treated with gradient concentrations of wogonin in the range of 6.25-400 μM for 48 h. As shown in Figures 1B, C , no apparent cytotoxicity of wogonin was observed on the Vero and IPEC-J2 cells at concentrations of ≤100 μM. The CC50 of wogonin was noted to be 475.3 μM and 499.41μM ( Figures 1D, E ).

To validate the antiviral activities of wogonin against PEDV, wogonin was added to PEDV-infected Vero and IPEC-J2 cells at concentrations of 12.5, 25, 50, and 100 μM, respectively, and examined by IFA. The results revealed that wogonin could inhibit PEDV proliferation until 12.5 μM in Vero cells ( Figure 2A ). Furthermore, the TCID50 assay was performed, and the dose-dependent inhibition of PEDV in Vero cells was determined ( Figure 2B ). Notably, the IC50 value of wogonin (58.87 μM and 56.56 μM) was far lower than its CC50 value (475.3 μM and 499.4 μM) obtained in the Vero ( Figures 2C, E ) and IPEC-J2 cells ( Figures 2D, F ), which indicated the ability of wogonin to suppress PEDV replication with only slight cytotoxicity. Moreover, the therapeutic index of wogonin was 8.0 and 8.8μM in IPEC-J2 cells. These findings revealed that wogonin could inhibit PEDV propagation in a dose-dependent manner, exhibiting enhanced antiviral activity with increasing wogonin concentration.

To understand the precise mechanism of the inhibitory effect of wogonin on PEDV propagation, the influence of wogonin on the PEDV at different stages of infection was assessed. The Vero cells were treated with 50 μM of wogonin during different stages of the PEDV life cycle (0.05 MOI) and then harvested to assay the viral RNA ( Figure 3A ). In the viral attachment assay, no significant difference in the level of PEDV RNA was observed between the wogonin-treated infected cells and the untreated infected cells ( Figure 3B ). However, wogonin significantly inhibited PEDV internalization, replication, and release ( Figures 3C-E ). In addition, wogonin interacted with PEDV particles, as the pre-exposure of the virus to wogonin altered the infectivity of PEDV particles ( Figure 3F ).

Binding affinity analysis (−10.0 kcal/mol) and binding mode examination suggested that Mpro can favorably combine with wogonin. This binding between Mpro (protein) and wogonin (ligand) was stable because the ligand was embedded in the groove of the active pocket owing to its small size. Figure 4 shows the optimal docked model of Mpro with wogonin. The ligand formed hydrogen bonds with E165, H41, and G142 residues, and the hydrophobic interactions of M25, Y53, I51, I140, C144, Q163, L164, and P188 also had a non-negligible role in the stabilization of wogonin. Thus, wogonin inhibited PEDV replication by blocking the binding pocket of Mpro.

The stability of Mpro was analyzed using a differential scanning fluorimetry (DSF) assay (Wright et al., 2017), and the results showed that the melting temperature (Tm) of Mpro (53.8°C) was similar to that of Mpro co-cultured with wogonin (54°C), thus suggesting that wogonin did not affect Mpro stability ( Figure 5A ). In addition, the microscale thermophoresis (MST) assay used to validate the interaction between wogonin and Mpro and quantify the binding affinity revealed that wogonin showed positive Mpro binding with a dissociation constant (kd) value of 9.589E-06 (± 6.917E-06) ( Figures 5B, C ). Furthermore, the investigation of the binding affinity between Mpro and wogonin through the use of a surface plasmon resonance (SPR) assay ( Figure 5D ) indicated a significant and dose-dependent increase in the SPR signal due to wogonin, and the affinity constant for the interaction between wogonin and Mpro was 16 μM. Overall, these results confirmed that wogonin could interact with Mpro to disrupt PEDV infection.

To confirm the inhibition of the PEDV Mpro in vitro, we performed an enzymatic inhibition assay, which specifically measures protease activity. Fluorescence increased with increasing Mpro into the substrate ( Figure 6A ), suggesting that our purified Mpro protein could effectively cleave the fluorogenic peptide substrate. In addition, the inhibition activity assay showed that wogonin results in a concentration-dependent inhibition of PEDV Mpro activity, with an approximately 54% inhibition of protease activity at 100 μM, and the IC50 was 19.4 μM ( Figure 6B ).