Screening of Botanical Drugs against SARS-CoV-2 Entry Reveals Novel Therapeutic Agents to Treat COVID-19

An escalating pandemic caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has severely impacted global health. There is a severe lack of specific treatment options for diseases caused by SARS-CoV-2. In this study, we used a pseudotype virus (pv) containing the SARS-CoV-2 S glycoprotein to screen a botanical drug library containing 1037 botanical drugs to identify agents that prevent SARS-CoV-2 entry into the cell. Our study identified four hits, including angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid, as effective SARS-CoV-2 S pv entry inhibitors in the micromolar range. A mechanistic study revealed that these four agents inhibited SARS-CoV-2 S pv entry by blocking spike (S) protein-mediated membrane fusion. Furthermore, angeloylgomisin O and schisandrin B inhibited authentic SARS-CoV-2 with a high selective index (SI; 50% cytotoxic concentration/50% inhibition concentration). Our drug combination studies performed in cellular antiviral assays revealed that angeloylgomisin O has synergistic effects in combination with remdesivir, a drug widely used to treat SARS-CoV-2-mediated infections. We also showed that two hits could inhibit the newly emerged alpha (B.1.1.7) and beta (B.1.351) variants. Our findings collectively indicate that angeloylgomisin O and schisandrin B could inhibit SARS-CoV-2 efficiently, thereby making them potential therapeutic agents to treat the coronavirus disease of 2019.


Introduction
The coronavirus disease of 2019 , caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), poses a severe threat to global public health, affecting economic and social stability, and requires the rapid development of novel treatment and preventive measures [1]. The coronavirus is an enveloped virus with four structural proteins: spike (S), membrane (M), envelope (E), and nucleocapsid (N) proteins. SARS-CoV-2 shares a high degree of sequence identity with the previously emerged SARS-CoV and exploits the same human cell receptor, angiotensin-converting enzyme 2 (ACE2), for infection [2,3]. The mechanism by which the receptor-binding domain (RBD) in the S1 subunit of the S protein on the virion binds to the ACE2 receptor on the target cell has been elucidated. Moreover, the heptad repeat 1 (HR1) and 2 (HR2) domains in the S2 subunit of the S protein interact with each other to form a six-helix bundle fusion nucleus. This interaction brings the virus membrane and cell membrane close together to facilitate fusion and infection [4,5].

HTS Optimization and Assaying of the Botanical Drug Library
A library of 1037 botanical compounds was purchased from Weikeqi Biotech (Sichuan, China). The compounds were collected and stored in 10-mM stock solutions in DMSO at −80 • C until use. The first round of HTS was carried out, and Caco-2 cells were seeded at a density of 2.4 × 10 4 cells per well in 96-well plates. After incubation overnight, cells were treated in duplicate with the compounds (50 µM), and 1 h later, cells were infected with SARS-CoV-2 S pv (ct19) with the MOI = 5, and the supernatant was removed 1 h.p.i. Camostat 100 µM and 0.5% DMSO were used as positive and negative controls, respectively. After 24 h, the luciferase activity was measured using the Rluc assay system (Promega, Madison, WI, USA). The primary compounds were then secondarily screened using VSV pv (MOI of 0.5) to rule out VSV genome replication inhibitors and Rluc activity.

Cell Viability
Caco-2 cells were seeded at a density of 2.4 × 10 4 cells per well in 96-well plates. After incubation overnight, cells were treated in duplicate with the compounds for 24 h. The cell supernatant was removed, and 50-µL 0.5% 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT; Sigma-Aldrich) dissolved in PBS were added to the cells. In actively growing cells, MTT can utilize reduced nicotinamide adenine dinucleotide (NADH)-and reduced nicotinamide adenine dinucleotide phosphate (NADPH)-dependent cellular oxidoreductase enzymes, producing a blue formazan product that is freely soluble in DMSO. After incubation at 37 • C for 4 h, the supernatant was removed carefully, and 50-µL DMSO was added to the cells. After being gently shaken, the plates were measured at 492 nm using a spectrophotometer, and cell viability was calculated.

SARS-CoV-2 Binding Assay
The binding reaction was performed using the SARS-CoV-2 inhibitor screening ELISA kit (Sino Biological, Beijing, China). Four hits were incubated with hACE-his and then transferred into 96-well plates coated with SARS-CoV-2 RBD. One hour later, the plates were washed with the buffer and incubated with HRP-labeled anti-his antibody. After incubation for another 1 h at room temperature, the plates were washed and detected with the color regent substrate. Finally, the reaction was stopped with 1-M H2SO4, and the optical density (OD) at 450 nm was measured [20,21]. In a binding assay, the SARS-CoV-2 spike antibody (40150-D002, Sino Biological, Beijing, China) was used as the positive control.

Virucidal Assay
To study the virucidal effects, SARS-CoV-2 (10 6 PFU/mL) was incubated with hits (25 µM and 0.25 µM) or the vehicle at 37 • C for 1 h. The samples were diluted in DMEM and calculated the remaining viral titer using the plaque assay. Viral titers were determined at dilutions that the hits were not effective (more than 100× dilution).

Membrane Fusion Assay
Vero E6 cells were co-transfected with pcDNA3.1-SARS-CoV-2 S (ct19) (0.25 µg) and pEGFP-N1 (0.25 µg) by using lipo2000 in 24-well plates. After transfection for 4 h, the medium was replaced with 2% DMEM containing different concentrations of the hits. After 24 h, syncytium formation was visualized using an M-shot image system (Microshot Technology, China). A dual split protein (DSP)-based cell-cell fusion assay was used to detect SARS-CoV-2 S (ct19) inhibitory activity of the four hits [22]. Briefly, a total of 3 × 10 4 293T cells (effector cells) were seeded in a 96-well plate, and 3 × 10 5 cells/mL 293T (target cells) were seeded in a 6-well plate culture, and then, the cells were incubated at 37 • C. On the next day, the effector cells were co-transfected with SARS-CoV-2 S (ct19) and a DSP1-7 plasmid, the target cells were transfected with DSP8-11 plasmid, and then, the cells were incubated at 37°C. After 24 h, serially diluted four hits were added to the effector cells were incubated for 1 h; the target cells were resuspended at 3 × 10 5 cells/mL in a prewarmed culture medium that contained EnduRen live cell substrate (Promega) and hits. Then, target cells were transferred to the effector cells, and a mixture of cells was spun down to maximize cell-cell contract. After incubation for 6 h, the luciferase activity was measured. In the DSP-based cell-cell fusion assay, clofazimine (10 µM) was used as the positive control.

Antiviral Effect of Hits against SARS-CoV-2 and SARS-CoV-2 Variants of Concern
Caco-2 cells were seeded at a density of 2.4 × 10 4 cells per well in 96-well plates. After overnight incubation, cell monolayers were treated with hits; 1 h later, cells were infected with SARS-CoV-2 or SARS-CoV-2 variants at an MOI of 0.5. After an additional 24 h of incubation, the infection was stopped by rinsing each well, and the cells were fixed with 4% paraformaldehyde. Ten micrometers remdesivir was used as the positive control.

Drug-Drug Interactions of Remdesivir with Hits
Caco-2 cells were seeded at a density of 2.4 × 10 4 cells per well in 96-well plates. After overnight incubation, different final concentrations of remdesivir were added to each row of the 96-well plate. Simultaneously, final concentrations of the four hits (50, 25, 12.5, 6.25, 3.125, 1.5625, 0.78125, and 0 µM) and remdesivir (20, 4, 0.8, 0.16, and 0 µM) were added to each column of the plate. The cells were infected with SARS-CoV-2 at an MOI of 0.5. After an additional 24 h of incubation, the infection was stopped by rinsing each well, and the cells were fixed with 4% paraformaldehyde. The antiviral activities were determined using the IFA assay. To determine the drug-drug interactions, the results were analyzed by MacSynergy II using the manual. Differential surface plots at the 95% confidence level (CI) were generated according to the Bliss independence model. The volumes of the synergistic regions were equal to the relative quantity of synergy produced per change in the two drugs concentrations.

Construction of SARS-CoV-2 S pv (ct19)
Firstly, we constructed a SARS-CoV-2 S pv based on a VSV backbone to perform highthroughput screening under BSL-2 conditions. The cytoplasmic tail of the S glycoproteins of SARS-CoV-2 is highly similar to that of SARS-CoV and carries signals for their retention in the endoplasmic reticulum. Previous studies have found that 19 amino acid deletions in the cytoplasmic tail of SARS-CoV or SARS-CoV-2 S glycoprotein increased the infectivity of the single-cycle pseudotype virus [23][24][25][26].
As shown in Figure 1A, the Renilla luciferase (Rluc) activity of the cells infected with SARS-CoV-2 S pv (ct19) was higher than that of the cells infected with SARS-CoV-2 S pv (wt). Next, the VSV copy number was assayed in the different pv harvests, and a higher packaging efficiency was detected in the SARS-CoV-2 S pv (ct19) system than in SARS-CoV-2 S pv (wt) ( Figure 1A), reaching 1.15 × 10 9 copies/mL and 2.3 × 10 8 copies/mL, respectively. Immunoblotting confirmed the expression and accumulation of the SARS-CoV-2 S protein in SARS-CoV-2 S pv (ct19) ( Figure 1B), enabling us to use SARS-CoV-2 S pv (ct19) for high-throughput screening (HTS) of the potential botanical drug candidates.

SARS-CoV-2 S pv (ct19) Entry Inhibitor Screening
The HTS assay conditions were optimized to a seeding density of 2.4 × 10 4 for Caco-2 cells and an MOI of 5 for the SARS-CoV-2 S pv (ct19) infective dose per well in 96-well plates. Under these optimized conditions, the signal/background (S/B) ratio, coefficient of variation, and Z' factor were 1286.09, 7.35%, and 0.705, respectively, making this assay promising for large-scale inhibitor screening. As shown in Figure 2A, the HTS assay screened a library containing 1037 botanical drugs. Compounds with >80% inhibition and no apparent cytotoxicity at a concentration of 50 µM were defined as prime candidates, and 58 (5.59%) prime candidates (Figure 2A,B) were selected for further investigations. These prime candidates were then counter-screened to rule out the inhibition of VSV genome replication and Rluc activity. Four candidates (0.38%): angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid passed the secondary screen with a mild inhibition of VSV (16.85%, 33.65%, 21.02%, and 32.41%, respectively) at 50 µM ( Figure 2C). This result indicated that these four hits specifically inhibited SARS-CoV-2 S infection at the entry step. (wt). Next, the VSV copy number was assayed in the different pv harvests, and a higher packaging efficiency was detected in the SARS-CoV-2 S pv (ct19) system than in SARS-CoV-2 S pv (wt) ( Figure 1A), reaching 1.15 × 10 9 copies/mL and 2.3 × 10 8 copies/mL, respectively. Immunoblotting confirmed the expression and accumulation of the SARS-CoV-2 S protein in SARS-CoV-2 S pv (ct19) ( Figure 1B), enabling us to use SARS-CoV-2 S pv (ct19) for high-throughput screening (HTS) of the potential botanical drug candidates.  The left axis shows the RLU (relative light unit) value detected 24 h after pseudovirus SARS-CoV-2 S pv (ct19) and SARS-CoV-2 S pv (wt) infection. The right y-axis shows virus copy number assays of both viral particles. (B) Immunoblots verifying the incorporation of the SARS-CoV-2 spike protein in the pseudovirus. The cell control was the culture supernatant of 293T cell transfected with pcDNA3.1 48 h and infected with VSV pv. Data are presented as the means ± standard deviations (SDs) for more than 2 independent ex-pediments (* p < 0.05, ** p < 0.01).

SARS-CoV-2 S pv (ct19) Entry Inhibitor Screening
The HTS assay conditions were optimized to a seeding density of 2.4 × 10 4 for Caco-2 cells and an MOI of 5 for the SARS-CoV-2 S pv (ct19) infective dose per well in 96-well plates. Under these optimized conditions, the signal/background (S/B) ratio, coefficient of variation, and Z' factor were 1286.09, 7.35%, and 0.705, respectively, making this assay promising for large-scale inhibitor screening. As shown in Figure 2A, the HTS assay screened a library containing 1037 botanical drugs. Compounds with >80% inhibition and no apparent cytotoxicity at a concentration of 50 μM were defined as prime candidates, and 58 (5.59%) prime candidates (Figure 2 A, B) were selected for further investigations. These prime candidates were then counter-screened to rule out the inhibition of VSV genome replication and Rluc activity. Four candidates (0.38%): angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid passed the secondary screen with a mild inhibition of VSV (16.85%, 33.65%, 21.02%, and 32.41%, respectively) at 50 μM ( Figure 2C). This result indicated that these four hits specifically inhibited SARS-CoV-2 S infection at the entry step.

Four Hits Specifically Inhibited SARS-CoV-2 S-Mediated Entry by Dose-Dependent Inhibition
The dose-dependent inhibition of the four hits was further investigated using Caco-2 cells. As shown in Figure 3, all four hits suppressed SARS-CoV-2 S pv (ct19) infection in a dose-dependent manner from 100 µM to 1.5625 µM. Angeloylgomisin O inhibited SARS-CoV-2 S pv (ct19) entry by 90% at 100 µM and had little effect on VSV pv, demonstrating its specificity against SARS-CoV-2 ( Figure 3A). Meanwhile, the cell viability was >80% when the inhibitors were used at an extremely high concentration (400 µM). Schisandrin B inhibited the entry of SARS-CoV-2 S pv (ct19) by 90% at 100 µM, whereas the cell viability was approximately 80% at 400 µM ( Figure 3B). At 50 µM and 100 µM, procyanidin inhibited SARS-CoV-2 S pv (ct19) by approximately 80% and 95%, respectively, while the cell viability was approximately 80% at 400 µM ( Figure 3C). Oleanolic acid (50 µM) inhibited SARS-CoV-2 S pv (ct19) entry by approximately 70% ( Figure 3D). It was found to be relatively toxic to Caco-2 cells, with a 50% cytotoxic concentration (CC50) of 54 µM. The four hits were also purchased from other commercial sources and tested to validate their antiviral effects. All the compounds showed similar antiviral effects, conforming the inhibitory effect of the four hits.

Four Hits Specifically Inhibited SARS-CoV-2 S-Mediated Entry by Dose-Dependent Inhibition
The dose-dependent inhibition of the four hits was further investigated using Caco-2 cells. As shown in Figure 3, all four hits suppressed SARS-CoV-2 S pv (ct19) infection in a dose-dependent manner from 100 μM to 1.5625 μM. Angeloylgomisin O inhibited SARS-CoV-2 S pv (ct19) entry by 90% at 100 μM and had little effect on VSV pv, demonstrating its specificity against SARS-CoV-2 ( Figure 3A). Meanwhile, the cell viability was >80% when the inhibitors were used at an extremely high concentration (400 μM). Schisandrin B inhibited the entry of SARS-CoV-2 S pv (ct19) by 90% at 100 μM, whereas the cell viability was approximately 80% at 400 μM ( Figure 3B). At 50 μM and 100 μM, procyanidin inhibited SARS-CoV-2 S pv (ct19) by approximately 80% and 95%, respectively, while the cell viability was approximately 80% at 400 μM ( Figure 3C). Oleanolic acid (50 μM) inhibited SARS-CoV-2 S pv (ct19) entry by approximately 70% ( Figure 3D). It was found to be relatively toxic to Caco-2 cells, with a 50% cytotoxic concentration (CC50) of 54 μM. The four hits were also purchased from other commercial sources and tested to validate their antiviral effects. All the compounds showed similar antiviral effects, conforming the inhibitory effect of the four hits.

Effects of Four Hits on Different Stages of SARS-CoV-2 Entry
A series of entry events were examined to dissect which step was blocked by the hits. Firstly, the effect of the inhibition on the receptor binding was tested, and as shown in Figure 4A, none of the hits had an effect on RBD binding. Next, we tested whether four hits could interact with SARS-CoV-2 and exert a virucidal effect. To this end, SARS-CoV-2 was mixed with hits at 25 μM or 0.25 μM for 1 h, and we determined the remaining viral titer using the plaque assay. SARS-CoV-2 infectivity was not reduced by incubation with

Effects of Four Hits on Different Stages of SARS-CoV-2 Entry
A series of entry events were examined to dissect which step was blocked by the hits. Firstly, the effect of the inhibition on the receptor binding was tested, and as shown in Figure 4A, none of the hits had an effect on RBD binding. Next, we tested whether four hits could interact with SARS-CoV-2 and exert a virucidal effect. To this end, SARS-CoV-2 was mixed with hits at 25 µM or 0.25 µM for 1 h, and we determined the remaining viral titer using the plaque assay. SARS-CoV-2 infectivity was not reduced by incubation with angeloylgomisin O, schisandrin B, and oleanonic acid. While a reduction of >90% (>1 log10) was observed in the 25-µM procyanidin group, indicating that procyanidin exhibited a virucidal effect on the SARS-CoV-2 authentic virus. Previous evidence has shown that procyanidins can interact with synthetic membranes and protect them from oxidation and disruption [27]. Procyanidins might interact with the viral membrane and result in low infectivity. angeloylgomisin O, schisandrin B, and oleanonic acid. While a reduction of >90% (>1 log10) was observed in the 25-μM procyanidin group, indicating that procyanidin exhibited a virucidal effect on the SARS-CoV-2 authentic virus. Previous evidence has shown that procyanidins can interact with synthetic membranes and protect them from oxidation and disruption [27]. Procyanidins might interact with the viral membrane and result in low infectivity.  To further confirm the hit mechanisms, the effects of the four hits on SARS-CoV-2 Smediated membrane fusion were examined. Recent research has identified that SARS-CoV-2 induces syncytia formation in the lungs of patients with COVID-19 [28]. It has been reported that cells infected with SARS-CoV-2 exhibit a typical syncytium phenomenon [5]. In this assay, fusion activity was conducted by co-transfecting SARS-CoV-2 S (ct19) and green fluorescent protein (GFP) into Vero E6 cells. After 24 h, the S protein of SARS-CoV-2 induced cell-cell fusion, resulting in the formation of syncytia ( Figure 4C). 25-Hydrocholesterol (25HC) was used as the positive control for this assay [29,30]. As shown in Figure 4C, all four hits induced a dose-dependent reduction in syncytium size. Angeloylgomisin O blocked syncytia formation starting at a concentration of 6.25 µM, while procyanidin, oleanonic acid, and schisandrin B required higher concentrations (12.5 or 25 µM) to achieve the same ( Figure 4C). To further quantitatively evaluate the inhibition, fusion efficacy was determined using the dual split protein assay. As shown in Figure 4D, four hits exhibited a dose-dependent inhibition of SARS-CoV-2 S-mediated membrane fusion. These results indicate that angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid could inhibit SARS-CoV-2 entry by inhibiting membrane fusion.

Four Botanical Hits Inhibited Authentic SARS-CoV-2 Infection
Next, an immunofluorescence assay (IFA) was performed using an anti-SARS-CoV-2 NP antibody to evaluate the inhibition of four hits on authentic SARS-CoV-2. As shown in Figure 5, all four hits inhibited authentic SARS-CoV-2 in a dose-dependent manner. The 50% inhibition concentration values (IC 50 ) of angeloylgomisin O (3.7 µM), schisandrin B (7.3 µM), and oleanonic acid (1.4 µM) were less than 10 µM, while the IC 50 value of procyanidin was 33 µM. It seemed that angeloylgomisin O, schisandrin B, and oleanonic acid exerted a more potent inhibition of authentic SARS-CoV-2.
The selective index (SI; CC 50 /IC 50 ) was calculated using an authentic SARS-CoV-2 inhibition assay. The SI values of angeloylgomisin O and schisandrin were >61 and >27, respectively. This observation suggests that both drugs are promising candidates for COVID-19 treatment.

Effects of the Four Hits against MERS-CoV-S pv and SARS-CoV-2 S Variants pv
To verify the antiviral effects of the four hits on MERS-CoV S pv and SARS-CoV-2 S pv variants containing S protein mutations, we verified the inhibitory effects of the four hits against other coronaviruses and the SARS-CoV-2 variants. As shown in Figure 6A The selective index (SI; CC50/IC50) was calculated using an authentic SARS-CoV-2 inhibition assay. The SI values of angeloylgomisin O and schisandrin were >61 and >27, respectively. This observation suggests that both drugs are promising candidates for COVID-19 treatment. To verify the antiviral effects of the four hits on MERS-CoV S pv and SARS-CoV-2 S pv variants containing S protein mutations, we verified the inhibitory effects of the four hits against other coronaviruses and the SARS-CoV-2 variants. As shown in Figure 6A, all four hits inhibited the entry of MERS-CoV S pv into Caco-2 cells in a dose-dependent manner. The IC50 values of angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid against MERS-CoV S pv were 21 μM, 19 μM, 14 μM, and 18 μM, respectively. A SARS-CoV-2 S pv variant containing the S protein D614G substitution was constructed and used to evaluate the inhibitory effect of the selected drugs. The G614 variant appears to increase the viral transmissibility [10][11][12]. The G614 variants have become dominant in the currently circulating virus strains. The variants of concern include the D614G substitution. As shown in Figure 6B, all four hits suppressed the entry of SARS-CoV-2 S pv D614G into Caco-2 cells in a dose-dependent manner. The IC50 values of angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid, at which they inhibited the entry of SARS-CoV-2 S pv D614G, were 16 μM, 27 μM, 33 μM, and 25 μM. We also introduced three additional amino acid substitutions, K417N, E484K, and N501Y, in the S D614G protein for our next set of investigations. These mutations have been observed in the beta variant and play an essential role in immune escape [13][14][15]. The results showed that all four hits inhibited SARS-CoV-2 S pv K417N/E484K/N501Y/D614G in a dose-dependent manner ( Figure 6C). The IC50 values of angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid against SARS-CoV-2 S pv K417N/E484K/N501Y/D614G were 24 μM, 41 μM, 27 μM, and 35 μM, respectively, indicating the potential of all the four hits to inhibit new variants.

Combinatory Treatments with the Drug Pair Remdesivir-Angeloylgomisin O Showed Enhanced Antiviral Activity
As all four hits could inhibit S-mediated membrane fusion, we assessed the efficacy of the combined treatment using remdesivir, a SARS-CoV-2 viral RNA-dependent RNA polymerase inhibitor [31,32], and the four hits. According to the bliss independence model, the degree of interaction was determined using MacSynergy II software to analyze A SARS-CoV-2 S pv variant containing the S protein D614G substitution was constructed and used to evaluate the inhibitory effect of the selected drugs. The G614 variant appears to increase the viral transmissibility [10][11][12]. The G614 variants have become dominant in the currently circulating virus strains. The variants of concern include the D614G substitution. As shown in Figure 6B, all four hits suppressed the entry of SARS-CoV-2 S pv D614G into Caco-2 cells in a dose-dependent manner. The IC 50 values of angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid, at which they inhibited the entry of SARS-CoV-2 S pv D614G, were 16 µM, 27 µM, 33 µM, and 25 µM. We also introduced three additional amino acid substitutions, K417N, E484K, and N501Y, in the S D614G protein for our next set of investigations. These mutations have been observed in the beta variant and play an essential role in immune escape [13][14][15]. The results showed that all four hits inhibited SARS-CoV-2 S pv K417N/E484K/N501Y/D614G in a dose-dependent manner ( Figure 6C). The IC 50 values of angeloylgomisin O, schisandrin B, procyanidin, and oleanonic acid against SARS-CoV-2 S pv K417N/E484K/N501Y/D614G were 24 µM, 41 µM, 27 µM, and 35 µM, respectively, indicating the potential of all the four hits to inhibit new variants.

Combinatory Treatments with the Drug Pair Remdesivir-Angeloylgomisin O Showed Enhanced Antiviral Activity
As all four hits could inhibit S-mediated membrane fusion, we assessed the efficacy of the combined treatment using remdesivir, a SARS-CoV-2 viral RNA-dependent RNA polymerase inhibitor [31,32], and the four hits. According to the bliss independence model, the degree of interaction was determined using MacSynergy II software to analyze the results [33,34]. The volumes of statistically significant synergy were evaluated, and a combination volume (CV) of >25 µM 2 % was interpreted as evidence of synergy and >100 µM 2 % as strong synergy [35].
Our results showed strong synergy interactions between remdesivir and angeloylgomisin O. Combination studies with remdesivir and angeloylgomisin O led to a CV of the results [33,34]. The volumes of statistically significant synergy were evaluated, and a combination volume (CV) of >25 μM 2 % was interpreted as evidence of synergy and >100 μM 2 % as strong synergy [35].

Angeloylgomisin O and Schisandrin B Inhibited SARS-CoV-2 S Variants
Next, the usefulness of angeloylgomisin O and schisandrin B needs to be investigated in the context of the currently circulating SARS-CoV-2 variants. We conducted antiviral experiments on the alpha and beta variants to verify the effectiveness of angeloylgomisin O and schisandrin B against the SARS-CoV-2 variants.

Angeloylgomisin O and Schisandrin B Inhibited SARS-CoV-2 S Variants
Next, the usefulness of angeloylgomisin O and schisandrin B needs to be investigated in the context of the currently circulating SARS-CoV-2 variants. We conducted antiviral experiments on the alpha and beta variants to verify the effectiveness of angeloylgomisin O and schisandrin B against the SARS-CoV-2 variants.
The results showed that angeloylgomisin O and schisandrin B inhibited the SARS-CoV-2 variants in a dose-dependent manner ( Figure 8A,B). Angeloylgomisin O and schisandrin B showed similar IC 50 values against the variants of concern and wild-type strains ( Figure 8C). This result suggests that angeloylgomisin O and schisandrin B act as broad-spectrum antiviral agents against the emerging SARS-CoV-2 variants of concern.
Viruses 2022, 14, x FOR PEER REVIEW 13 of 16 The results showed that angeloylgomisin O and schisandrin B inhibited the SARS-CoV-2 variants in a dose-dependent manner (Figure 8 A, B). Angeloylgomisin O and schisandrin B showed similar IC50 values against the variants of concern and wild-type strains ( Figure 8C). This result suggests that angeloylgomisin O and schisandrin B act as broadspectrum antiviral agents against the emerging SARS-CoV-2 variants of concern. The IC50 values. Caco-2 cells were seeded in 96-well plates. After overnight incubation, cell monolayers were treated with hits; 1 h later, cells were infected with SARS-CoV-2 variants at an MOI of 0.5, followed by incubation for 24 h. The inhibition of SARS CoV-2 variants was measured by using the IFA assay. Data were presented as the means ± SD from two or three independent experiments.

Discussion
In this study, we screened a botanical drug library containing 1037 compounds and identified four hits, including angeloylgomisin O, procyanidin, schisandrin B, and oleanonic acid, which blocked the entry of SARS-CoV-2 S (ct19) pv by inhibiting vira membrane fusion. All four hits inhibited MERS-CoV S pv, SARS-CoV-2 S D614G pv, and SARS-CoV-2 S pv K417N/E484K/N501Y/D614G infection. Furthermore, the four hits in hibited the SARS-CoV-2 authentic virus at the micromolar level. The top two compounds angeloylgomisin O and schisandrin B, inhibited authentic SARS-CoV-2 and SARS-CoV-2 alpha and beta variants at a highly selective index (SI). The combination of remdesivirangeloylgomisin O showed more potent antiviral activity against SARS-CoV-2 than remdesivir monotherapy.
The top two hits, angeloylgomisin O and schisandrin B, were derived from Schisandra chinensis. Fruits from this plant and its extracts have been used in traditional medicines in East Asia to treat liver disorders such as hepatitis. They also possess a broad spectrum of biological and pharmacological uses, including antiviral, anti-inflammatory, and antioxidative properties, without toxicity [36]. Schisandrin B has been shown to inhibit cytochrome P450 3A (CYP3A) activity in the rat liver, thus affecting the metabolism of many drugs [37,38]. Furthermore, upon oral administration of Schisandra chinensis extract in rats, schisandrin B accumulates to a maximum plasma concentration (Cmax) of 4.54 μM [39]. This value was similar to the IC50 value for schisandrin B in the present study suggesting that this compound offers a biologically feasible treatment for SARS-CoV-2.
CD4+T cells are rapidly activated at the cellular level to produce inflammatory cytokines after SARS-CoV-2 infection, which further induces CD14+ CD16+ monocyte activation with high levels of interleukin 6 (IL-6) expression. Thus, reduced IL-6 could potentially reduce the immunopathology of SARS-CoV-2 [40,41]. Cai et al. demonstrated that a pretreatment of 8-10-week-old BALB/c mice with schisandrin B at doses of 25, 50, and 75 After overnight incubation, cell monolayers were treated with hits; 1 h later, cells were infected with SARS-CoV-2 variants at an MOI of 0.5, followed by incubation for 24 h. The inhibition of SARS-CoV-2 variants was measured by using the IFA assay. Data were presented as the means ± SD from two or three independent experiments.

Discussion
In this study, we screened a botanical drug library containing 1037 compounds and identified four hits, including angeloylgomisin O, procyanidin, schisandrin B, and oleanonic acid, which blocked the entry of SARS-CoV-2 S (ct19) pv by inhibiting viral membrane fusion. All four hits inhibited MERS-CoV S pv, SARS-CoV-2 S D614G pv, and SARS-CoV-2 S pv K417N/E484K/N501Y/D614G infection. Furthermore, the four hits inhibited the SARS-CoV-2 authentic virus at the micromolar level. The top two compounds, angeloylgomisin O and schisandrin B, inhibited authentic SARS-CoV-2 and SARS-CoV-2 alpha and beta variants at a highly selective index (SI). The combination of remdesivirangeloylgomisin O showed more potent antiviral activity against SARS-CoV-2 than remdesivir monotherapy.
The top two hits, angeloylgomisin O and schisandrin B, were derived from Schisandra chinensis. Fruits from this plant and its extracts have been used in traditional medicines in East Asia to treat liver disorders such as hepatitis. They also possess a broad spectrum of biological and pharmacological uses, including antiviral, anti-inflammatory, and antioxidative properties, without toxicity [36]. Schisandrin B has been shown to inhibit cytochrome P450 3A (CYP3A) activity in the rat liver, thus affecting the metabolism of many drugs [37,38]. Furthermore, upon oral administration of Schisandra chinensis extract in rats, schisandrin B accumulates to a maximum plasma concentration (C max ) of 4.54 µM [39]. This value was similar to the IC 50 value for schisandrin B in the present study, suggesting that this compound offers a biologically feasible treatment for SARS-CoV-2.
CD4+T cells are rapidly activated at the cellular level to produce inflammatory cytokines after SARS-CoV-2 infection, which further induces CD14+ CD16+ monocyte activation with high levels of interleukin 6 (IL-6) expression. Thus, reduced IL-6 could potentially reduce the immunopathology of SARS-CoV-2 [40,41]. Cai et al. demonstrated that a pretreatment of 8-10-week-old BALB/c mice with schisandrin B at doses of 25, 50, and 75 mg/kg reduced lipopolysaccharide-induced acute lung injury by lowering the number of inflammatory cells and proinflammatory cytokines, including tumor necrosis IL-6 in bronchoalveolar lavage fluid [42]. This result suggested that schisandrin B could coordinate the cytokine response to reduce the host immune response, and schisandrin B might achieve a better therapeutic effect in vivo.
Procyanidin is a pro-anthocyanidin member that belongs to the group of flavonoids, the secondary metabolites of polyphenolic plants and fungi. Polyphenolic compounds, such as flavonoids, are characterized by different biological activities, including antimicrobial, anticancer, anti-inflammatory, and antiviral properties [43]. Pro-anthocyanidins are also the most abundant polyphenolic compounds in lignin [16]. Maroli et al. used molecular docking analyses to demonstrate that procyanidin might inhibit SARS-CoV-2 entry and replication [44]. Our study demonstrated that procyanidin has virucidal and antiviral activity with a micromolar range.
Viral variants of concern may emerge with dangerous resistance to the immunity generated by the current vaccines to prevent COVID-19. For example, the alpha and beta variants are of concern because of their rapid rise to dominance and their extensive spike mutations, which may be detrimental to antiviral effectiveness and vaccine protection. Our results confirmed the broad-spectrum antiviral activity of angeloylgomisin O and schisandrin B against both variants. This result enhances the potential of angeloylgomisin O and schisandrin B as therapeutic agents for COVID-19.