TRIzol and fetal bovine serum (FBS) were purchased from Shanghai ExCell Bio Co., Ltd., China. The Reverse Transcription Kit was obtained from Takara Biological Engineering Co., Ltd., China. Cell counting kit-8 (CCK-8) was obtained from Dojindo Laboratories, Kumamoto, Japan. Antibodies against TLR4, MyD88, and NF-κB p65, primary antibodies, and β-actin were obtained from Abcam (Cambridge, United Kingdom). ELISA kits for TNF-α, IL-1β, IL-6, and IFN-β were purchased from Boster Biological Technology Co., Ltd., China. Modified eagle’s medium (MEM) was obtained from Procell Life Science and Technology Co., Ltd., China. A 1% penicillin-streptomycin mixture and Electrochemiluminescence (ECL) reagent were purchased from Beijing Solarbio Science and Technology Co., Ltd., China.

As previously reported, the proportion of SBP was according to recommendations specified in the Pharmacopoeia of the People’s Republic of China (2020 Edition), as shown in Table 1. All the herbs were purchased from Beijing Tong Ren Tang Co., Ltd (Beijing, China). These herbs were soaked in water at a ratio of 1:10 for half an hour and heated for 1 h. The ingredients were extracted twice. These extracts were combined, filtered, and concentrated to 1 g/ml using a rotary evaporator. Ribavirin granules (Qianjin Xiangjiang Pharmaceutical Co., Ltd.) were prepared with distilled water to a concentration of 5 mg/ml.

Thirty male Wistar rats (4 week-old, 100–120 g) were purchased from the Medical Animal Test Center of Shandong University (Jinan, China). After adaptation for three days, the rats were randomly divided into three groups (ten rats per group): normal group, ribavirin group (40.5 mg/kg/d, p. o.) and SBP group (8.37 g/kg/day, p. o.). The rats in the normal group were given the corresponding amount of saline. At the end of the third day, all the rats were anesthetized. Serum samples were collected from the abdominal aorta and centrifuged at 3,000 rpm in 4°C for 10 min. A portion of the supernatant was analyzed using ultra-high-performance liquid chromatography-quadrupole/electrostatic field orbital hydrazine high resolution mass spectrometry (UPLC-Q-Orbitrap HRMS/MS), and the other portions were used for experimentation.

The analysis was performed using UPLC-Q-Orbitrap HRMS/MS (UltiMate 3000 RS) and Q-Exactive High Resolution Mass Spectrometer, which were purchased from Thermo Fisher Technology (China) Co., Ltd.). Chromatographic separation was carried out on an RP-C18 column (150 × 2.1 mm, 1.8 μm) at 35°C. The mobile phase consisted of 1): 0.1% aqueous solution of formic acid and 2): 0.1% formic acid in acetonitrile. The elution gradient was as follows: 0–1 min, 2% of B; 1–5 min, 20% of B; 5–10 min, 50% of B; 10–15 min, 80% of B; 15–25 min, 90% of B; 25–30 min 2% of B. The flow rate was 0.3 ml/min. The MS acquisition was performed using the positive and negative ion-switching modes of scanning. Mass spectrometry detection method: Full MS/dd-MS₂; ion source: electrospray ionization source. The flow rates of sheath gas and auxiliary gas was 40 arb (arbitrary units) and 15 arb respectively, at 350°C. The capillary temperature was 300°C. Spray voltage was 3.8 kV for positive ionization and 3.8 kV for negative ionization. The primary scanning resolution was 70,000 FWHM, and the secondary scanning resolution was 17,500 FWHM. The scan range was 150–2000 m/z.

Using the traditional Chinese medicine system pharmacology (TCMSP) (Ru et al., 2014) (https://tcmspw.com/tcmsp.php) and Encyclopedia of Traditional Chinese Medicine (ETCM) (Xu et al., 2018) (http://www.tcmip.cn/ETCM/index.php/Home/Index/) databases, we collected the data on the chemical components contained in the ten traditional Chinese herbs in the formulation of SBP, and selected the compounds with oral bioavailability ≥30% and drug-likeness ≥ 0.18 as the active compounds of SBP. Based on the relevant pharmacological research results in the literature, we included some pharmacologically active compounds in the study, although the inclusion criteria mentioned above was not met. The targets of these compounds were obtained through the TCMSP and SwissTarget (Daina et al., 2019) (http://www.swisstargetprediction.ch/) databases, and the target proteins were converted into gene names using the UniProt database (https://sparql.uniprot.org/). Related genes of RVE were obtained from the Genecard database (https://www.genecards.org/), and the drug-compound-target network of SBP was constructed using Cytoscape 3.7.2.

Using Bioinformatics and Evolutionary Genomics (http://bioinformatics.psb.ugent.be/webtools/Venn/), we constructed a Venn diagram to screen the targets in the intersection of the active compounds of SBP and RVE The above intersection targets were imported into Search Tool for the Retrieval of Interacting Genes/Proteins database (Szklarczyk et al., 2019) (https://string-db.org/) to construct a PPI network. Using Metascape (http://metascape.org/gp/index.html), we conducted Gene Ontology enrichment analysis and Kyoto Encyclopedia of Genes and Genomes enrichment analysis of the targets in the intersection, and prepared a bar graph for the enrichment analysis. Based on the KEGG enrichment analysis and annotation results, the target-pathway network diagram was drawn.

Caco-2 cells (Cell Bank of Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China) were cultured in an MEM supplemented with 20% FBS and 1% penicillin-streptomycin mixture at 37°C and 5% CO₂. The RV-SA11 strain (Institute of Viral disease Control and Prevention of the Chinese Center for disease Control and Prevention) was cultivated in Caco-2 cells in a medium containing 2% FBS and 1% penicillin-streptomycin mixture. The 50% tissue culture infective dose (TCID50) was calculated using the Reed–Muench method (TCID50 = 10–^4.31/100 µL, Supplementary Table S1) (Reed and Muench, 1938).

The cytotoxicity of SBP-containing serum against Caco-2 cells was determined using CCK-8 assays (Zhang et al., 2018). Caco-2 cells (3 × 10⁴ cells/well) were seeded into a 96-well plate and incubated for 24 h. The cells were then treated with the drug-containing serum at serial concentrations of 50%, 40%, 30%, 20%, 10%, 5%, and 2.5%. The samples were incubated in 5% CO₂ at 37°C for 24 h, then 10 μL CCK-8 was added into each well and incubated for an additional 2 h. The optical density was measured at 450 nm using a microplate reader, and the 50% cytotoxic concentration (CC50) of the medicated serum causing 50% of the cytopathic effect (CPE) of Caco-2 cells was calculated (Mosmann, 1983). All the experiments were performed in triplicate.

Confluent monolayers of Caco-2 cells were inoculated with 100 TCID50/mL of RV-SA11 strain virus solution in a 96-well plate (100 μL per well) and adsorbed at 37°C for 2 h. The supernatant with the virus was discarded, and different concentrations (40%, 30%, 20%, 10%, 5%, and 2.5%) of SBP-containing serum (100 μL per well) were added. All of these were below the toxic concentration of the drug according to the cytotoxicity test results. Three wells were set for each concentration, normal cell control as well as virus control simultaneously. The CPE was observed daily, and the OD was detected using the CCK-8 method when CPE of the virus control was more than 90%. According to the results, the concentration of serum containing 20% of the drug was used in the follow-up experiment.

TRIzol was used to extract the total RNA from Caco-2 cells, and the cDNA was obtained using reverse transcription of mRNA using the Reverse Transcription Kit. The qRT-PCR method was performed as follows: pre-denaturation at 95°C for 10 min, denaturation at 95°C for 20 s, annealing at 58°C for 30 s, elongation at 72°C for 20 s, 40 cycles. The sequences of primers were as follows: β-actin: Forward 5′-CCC​ATC​TAT​GAG​GGT​TAC​GC-3′, Reverse 5′-TTT​AAT​GTC​ACG​CAC​GAT​TTC-3′; TLR4: Forward 5′-AGG​TTT​CCA​TAA​AAG​CCG​AAA​G-3′, Reverse 5′-CAA​TGA​AGA​TGA​TAC​CAG​CAC​G-3′; MyD88: Forward 5′-CGC​CGC​CTG​TCT​CTG​TTC​TTG-3′, Reverse 5′-GGT​CCG​CTT​GTG​TCT​CCA​GTT​G-3′; NF-κB p65: Forward 5′-ACA​GAA​GCA​GGC​TGG​AGG​TAA​GG-3′, Reverse 5′-GGA​CAA​TGC​CAG​TGC​CAT​ACA​GG-3′. The relative expression levels of mRNA were calculated by the 2^−△△CT method.

Cells were lyzed to obtain protein samples, and the protein concentrations of the samples were detected using the BCA Protein Assay Kit (Beijing Solarbio Science and Technology Co., Ltd., China). Proteins from cell lysates were separated on 10% sodium dodecyl sulfate-polyacrylamide gel and then transferred onto a polyvinylidene fluoride membrane (Millipore, United States). The membranes were washed three times with Tris-buffered saline with 0.1% Tween^® 20 Detergent (TBST), blocked with 5% bovine serum albumin for 3 h at 25°C, and then incubated overnight with the corresponding primary antibodies at 4°C. The membranes were washed five times with TBST and incubated with secondary antibodies for 1 h at 25°C. Subsequently, after washing five times with TBST, we used an ECL reagent to capture the protein signals, and ImageJ software (NIH, United States) was used to visualize the protein bands. All experiments were performed in triplicate.

The levels of inflammatory cytokines, including TNF-α, IL-1β, IL-6, and IFN-β in the supernatant were determined using ELISA kits. The experimental steps were performed according to the manufacturer’s instructions. The OD was measured using a microplate reader at 450 nm.

All data in the experiments were tested for normality, and were conformed to normal distribution. The data were expressed as mean ± standard deviation. SPSS 19.0 software (IBM, Chicago, IL, United States) was used for one-way ANOVA followed by the Fisher’s least significant difference (LSD) test for multiple comparisons. A p value <0.05 was considered to be a statistically significant difference between the groups.

The chemical constituents of SBP-containing serum were analyzed using UPLC-Q-Orbitrap HRMS. The results are shown in Figure 1; Supplementary Table S2. A total of nine ingredients were identified using excimer peaks, information on fragmentation, and literature. These included, N1-(1,3,5-Trimethyl-1H-pyrazol-4-yl)-2-cyano-3-(dimethylamino) acrylamide, isoliquiritigenin, two Methoxyestrone, 6-O-Pentopyranosyl-1-O-[(2,6,6-trimethyl-1-cyclohexen-1-yl)carbonyl]-β-d-glucopyranose, oleonolic acid, monoolein, 18-β-Glycyrrhetinic acid, 9(Z),11(E)-Conjugated linoleic acid, and oleic acid.

According to the oral bioavailability ≥30% and drug-likeness ≥ 0.18, a total of 219 active compounds were selected from the 10 ingredients in the SBP formulation, and they acted on 471 potential targets. The drug-compound-target network constructed using Cytoscape 3.7.2 composed of 708 nodes and 3,921 edges (Figure 2A). From this network, we identified that several compounds act on multiple targets, and vice versa. A total of 226 target genes of RVE were obtained from the GeneCards database.

The Venn diagram showed that SBP and RVE have 44 common target proteins (Figure 2B), which might be the biological basis of SBP’s effect on RVE. The PPI network showed that there is a close interaction between these 44 common targets. This showed that RGFR, EGF, IL-6, TGFB1, TNF, IL-2, and INF were the key targets of this PPI network (Figure 2C).

GO and KEGG analyses were performed using the Metascape database (Figure 2D). GO enrichment analysis showed that SBP primarily acted on membrane rafts, plasma membrane-protein complexes and lumens of vesicle, and affected cytokine-mediated signaling pathways, cell-cell adhesion of leukocytes, and other biological processes by affecting the functions of cytokine-receptor binding. To show the relationship between 44 targets and their corresponding KEGG pathways, we constructed a target-pathway network (Figure 2E), and the results showed that SBP acted on RVE by influencing several inflammatory pathways and intestinal immune networks.

To assess the antiviral efficacy of SBP, the inhibitory effects of SBP-containing serum on the replication and proliferation of RV-SA11 were tested. The CC50 of SBP-containing serum in Caco-2 cells was 68.75% (Figure 3A; Supplementary Table S3). The EC50 value of SBP-containing serum was 5.911% (Figure 3B; Supplementary Table S4)), and the SI was 11.63. The CC50 of ribavirin-containing serum in Caco-2 cells was 65.08% (Figure 3A). The EC50 value of ribavirin-containing serum was 3.467% (Figure 3B) and the SI was 18.77. The results showed that both SBP and ribavirin significantly inhibited virus replication and proliferation with less toxic effects on the viability of Caco-2 cells. Meanwhile, we examined the effects of SBP on the pathological changes in virus-infected Caco-2 cells (Figure 3C). The Caco-2 cells in the normal control group (NC) had uniform morphology with irregular and oval tight junctions and clear cell margins. The cells of the RV group were uneven in shape with vacuoles, large intercellular space, blurred margins, and dead cells and cell debris. The drug-containing serum acts on rotavirus-infected cells. As we increased the concentration of the drug-containing serum, the number of cell vacuoles and dead cells decreased, the cell margins became clearer, and the connections were tighter.

To investigate the mechanism of SBP on RV, we detected the mRNA and protein expressions of TLR4 and downstream MyD88 and NF-κB p65. The mRNA and protein expressions of TLR4, MyD88, and NF-κB p65 were significantly increased in RV-SA11-infected Caco-2 cells (p < 0.01). SBP significantly decreased the mRNA and protein expressions of TLR4, MyD88, and NF-κB p65 induced by RV-SA11 (p < 0.01). This indicates that SBP inhibits RV-SA11 virus by inhibiting the TLR4/MyD88/NF-κB signaling pathway (Figure 4; Supplementary Tables S5, 6).

To further study the effect of SBP on the inflammatory response induced by RV-SA11, ELISA was used to detect the production of cytokines in the supernatant of the infected cells. The expression levels of IL-1β, IL-6, TNF-α, and IFN-β in the RV group were significantly higher than those in the NC group (p < 0.01). The expression levels of IL-1β, IL-6, and TNF-α in the SBP group were significantly decreased (p < 0.01) when compared to the RV group (Figures 5A–C; Supplementary Table S7). Meanwhile, the expression level of IFN-β further increased in the SBP group to play its antiviral role (Figure 5D; Supplementary Table S7). These results showed that SBP was able to regulate the overexpression of various inflammatory cytokines in the RV-SA11-infected Caco-2 cells.