Six-to-eight-week-old female C57BL/6 mice were purchased from SPF Biotechnology Co., Ltd. (Beijing, China). All animals were maintained under 12 h light/dark conditions at 22–24°C with unrestricted access to food and water for the duration of the experiment. All animal protocols in this study were performed according to the guidelines for care and use of laboratory animals and approved by the animal ethics committee of the Fifth Medical Center, Chinese PLA General Hospital (Beijing, China).

Nigericin, dimethyl sulfoxide, and ultrapure lipopolysaccharide (LPS) were purchased from Sigma (Munich, Germany). SolB, WeD, and colchicine were obtained from TargetMol (Boston, MA, United States). Anti-mouse-Collagen1 (1:1000), anti-mouse-Smad2/3 (1:1000), and anti-mouse-P-Smad3 (1:1000) were purchased from Cell Signaling Technology (Boston, MA, United States). Anti-mouse-P-IκBα (1:1000), anti-mouse-IκBα (1:1000), anti-mouse -P-IKK (1:1000), anti-mouse-IKK (1:1000), anti-human Smad3 (1:1000), anti-human P-Smad3 (1:1000), and anti-human α-SMA (1:1000) were purchased from Proteintech (Chicago, IL, United States). Anti-mouse-IL-1β, anti-mouse-NLRP3 (1:2000), and anti-mouse-ASC (1:1000) were purchased from Santa Cruz Biotechnology (Beijing, China). Anti-mouse -caspase-1 p20 (1:1000), anti-mouse-caspase-1 p45 (1:1000), anti-mouse-pro-IL-1β (1:1000), and anti-GAPDH (1:2000) were purchased from Proteintech (Chicago, IL, United States).

Bone marrow-derived macrophages (BMDMs) were isolated from the femoral bone marrow of 10°week-old female C57BL/6 mice and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin (P/S), and 50 ng/ml murine macrophage colony-stimulating factor. LX-2 and L0-2 cells were grown in RPMI 1640 medium. All cell lines were cultured under a humidified 5% (v/v) CO₂ atmosphere at 37°C.

Four-to-five-week-old female C57BL/6 mice were purchased from SPF Biotechnology Co., Ltd. (Beijing, China). All animals were maintained under 12 h light/dark conditions at 22–24°C with unrestricted access to food and water for the duration of the experiment, except during fasting tests. All animal protocols in this study were performed according to the guidelines for care and use of laboratory animals and approved by the animal ethics committee of the Fifth Medical Center, Chinese PLA General Hospital. For experiments with BDL, mice were randomly divided into six groups. From days 7 to 21 after surgery, the sham group and the mice with successful development of the disease were divided into groups and administered drugs by gavage with a dose of 2 ml/kg once a day. The mice were divided into the following groups: sham operation group (Control), model group (BDL/CCl₄-induced hepatic fibrosis model group), colchicine positive group (0.2 mg/kg), WeD 20 mg/kg group, SolB 40 mg/kg group, and WeD 20 mg/kg in combination with SolB 40 mg/kg group. Eight mice were included in each group. The body weight was measured daily. The animals were euthanized in random order between 9:00 am and 11:00 am after an overnight fast. Samples of serum and liver were collected for further analysis.

Alanine transaminase (ALT) and aspartate transaminase (AST) were analyzed using kits from Thermo Fisher Scientific (Cincinnati, OH, United States). Formalin-fixed tissue was embedded in paraffin, and sections were stained with Masson, hematoxylin and eosin (H&E), and Sirius red stains. Liver histology was blindly assessed for inflammation, necrosis, and bile duct proliferation. Hydroxyproline levels in the liver were measured as described.

qPCR method was used to detect gene expression in liver tissue of mice in each group. After RNA extraction, Reverse transcription kit was used to reverse transcription RNA cDNA, and the specific operation was carried out strictly in accordance with the kit instructions. Using a qPCR instrument to amplify each gene and its GAPDH, qPCR reaction system: cDNA1μL, SYBR Green Mix 5 , 0.5 μl of upstream and downstream primers, DEPC water 3 μl. Reaction conditions 95°C 3 min, 95°C 3 s, 60°C 30 s, 95°C 15 s, 60°C 1 min, 95°C 15 s, Cycle 50 times. Each sample contains two multiple holes, gene expression levels were quantitatively analyzed by 2-ΔΔCt method.

Immunoblot analysis was used to evaluate the expression of Collagen1, p-Smad3, Smad2/3, p-IκB, IκBα, IKK, p-IKK, NLRP3, Lamin B, α-SMA, caspase-1 p20, IL-1β p17, pro-IL-1β, caspase-1 p45, NLRP3, and ASC. GAPDH and Lamin B served as loading controls in cell lysates. The samples were boiled at 105°C for 15 min. The protein samples were resolved using 12 or 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and transferred to a nitrocellulose membrane via a wet-transfer system. The membranes were subsequently incubated with 5% fat-free milk for 1 h at room temperature followed by incubation overnight with primary antibodies at 4°C. Blots were washed thrice with Tris-buffered saline Tween-20 (TBST) and incubated with corresponding horseradish peroxidase-conjugated secondary antibody (1:5000) for 1 h at room temperature. This step was followed by washing with TBST thrice, and the signals were analyzed using the enhanced chemiluminescent reagent (Promega, Beijing, China) detection system.

BMDMs were seeded at 5×10⁵ cells/well in 24-well plates overnight. The following day, the medium was replaced, and cells were stimulated with 50 ng/ml LPS for 4 h. The medium was then changed to opti-MEM containing WeD or SolB for 1 h. For inducing the activation of NLRP3 infammasome, cells were stimulated with nigericin (7.5 μmol/L) for 45 min.

BMDMs were seeded at 5×10⁵ cells/well in 24-well plates overnight. BMDMs were treated with WeD or SolB for an hour and then stimulated with LPS (50 ng/ml) for 0, 5, 15, 30, and 60 min.

We seeded BMDMs at 5×10⁵ cells/well in 24-well plates overnight. The medium was then replaced with 1640 medium without FBS, then give SolB or WeD for an hour, followed by treatment with TGF-β1 (5 ng/ml) alone or in combination with LPS (50 ng/ml).

The cell counting kit-8 (CCK-8) assay was used to detect the viability of cells. L0-2 cells were seeded in a 96-well growth-medium plate overnight at 8.5×10⁴ cells/well. The cells were then incubated at 37°C followed by individual treatments with WeD or SolB for 24 h. These cells were incubated with CCK-8 for 30 min. The optical density (O.D.) values were determined at the wavelength of 450 nm. The half-maximal inhibitory concentration (IC50) of WeD or SolB was evaluated using the software Prism 6 (GraphPad Software, San Diego, CA, United States).

To induce hepatocyte injury, L0-2 cells were seeded at 8.5×10³ cells/well in 96-well plates overnight. The following day, the medium was replaced with DMEM containing APAP (22 mmol/L) and SolB or WeD (5, 10, 20, 40 μmol/L).

Supernatants from cell culture were assayed with mouse TNF-α and mouse IL-6 according to manufacturer's instructions (Dakewe, Beijing, China).

L0-2 cells were treated with WeD and SolB separately. The release of LDH into the culture supernatants was evaluated using the LDH cytotoxicity assay kit (Beyotime, Shanghai, China) according to the manufacturer’s instructions.

Statistical analysis was performed using the software Prism 6 (GraphPad Software, San Diego, CA, United States). All experimental data were expressed as mean ± standard error of the mean (SD). One-way ANOVA was used for multiple comparisons followed by Tukey’s post hoc test or Dunnett’s test for comparison between two groups. The differences with p < 0.05 were considered statistically significant.

Persistent hepatocyte injury or death, as an main driver of fibrosis and liver inflammation, could initiate and perpetuate sterile inflammatory responses and HSCs activation, so hepatic fibrosis can be reversed by inhibiting hepatocyte injury or eliminating the triggering factors (Cordero-Espinoza and Huch, 2018; Bansal and Chamroonkul, 2019). S. sphenanthera is a traditional hepato-protective Chinese medicine, so we test the effect of the constituents of S. sphenanthera on APAP-induced hepatocyte injury. The data showed that several constituents could improve cell viability of LO-2 cells exposed to acetaminophen, and Schisandrol B (SolB) is the most effective constituent (Figure 1A). We then tested the effect of the SolB on hepatocyte survival by the CCK8 assay and the result showed that SolB were not toxic to LO-2 cells, even at concentrations up to 200 μmol/L (Figure 1B). We next examined whether SolB inhibited hepatocyte injury at a wide range of concentrations (0, 5,10, 20, and 40 μM). The result showed that SolB treatment dose-dependently improved cell viability and inhibited LDH release in LO-2 cells exposed to acetaminophen (Figures 1C,D). The protective effects of SolB on hepatocytes was further examined using Annexin V-FITC Apoptosis detection kit, the result showed that SolB treatment significantly inhibited acetaminophen-induced apoptosis of LO-2 cells (Figure 1E). These results indicated the protective effect of SolB against hepatocyte injury.

Considering the obviously hepato-protective effect, we tested the role of SolB in mouse model of hepatic fibrosis induced by bile duct ligation (BDL) ligation. The livers of mice treated with colchicine (colchicine has been used for the treatment of liver fibrosis (Rambaldi and Gluud, 2005) and was used as a positive control in this study, 0.2 mg/kg per day) or SolB (10, 20, and 40 mg/kg per day, respectively) exhibited a remarkably lower cirrhotic appearance with varying degrees (Figures 2A–D). Furthermore, the histological assessment of liver sections by H&E staining, Sirius red, and Masson staining demonstrated that treatment with SolB inhibited hepatic injury and fibrosis in mice induced by BDL (Figures 2A,B). Consistent with the histopathology analysis, SolB treatment reversed the increase in ALT, AST, DBIL, TBIL, and TBA serum levels induced by BDL in a dose-dependent manner in mice (Figures 2C,D; Supplementary Figures S1A–C).

HSCs play a crucial role in hepatic fibrosis, which can be activated upon persistent liver injury and leads to the deposition of ECM components including collagen 1 (Shang et al., 2018). The α-smooth muscle actin (α-SMA) is a unique marker of activated HSCs (Nouchi et al., 1991). Hence, we analyzed the expression of α-SMA by immunohistochemical staining (Figure 2A) and qPCR (Figure 2E) in all groups. The results demonstrated that SolB inhibited the expression of α-SMA in mice with hepatic fibrosis induced by BDL in a dose-dependent manner. Our results also showed that treatment with SolB reduced the number of macrophages, staining by specific marker of F4/80, in mice induced by BDL. Consequently, this observation indicated that SolB inhibited the activation of HSCs in vivo. Activation and survival of HSCs are regulated by transforming growth factor-beta (TGF-β) and platelet-derived growth factor (PDGF) signaling pathways (Pinzani 2002; Kikuchi et al., 2017; Dewidar et al., 2019). The expression of TGF-β and PDGF was also inhibited by SolB treatment in mice with hepatic fibrosis induced by BDL, as evidenced by qPCR (Figures 2F,G). BDL significantly promoted collagen I gene expression compared to that in the normal mice, while SolB treatment significantly abrogated BDL-induced upregulation of collagen I, as revealed by immunoblotting (Figure 2H; Supplementary Figure S1D) and qPCR analysis (Supplementary Figure S1E). More importantly, we observed that SolB inhibited both the phosphorylation of Smad3 and the expression of Smad2/3 in a dose-dependent manner (Figure 2H; Supplementary Figure S1E). We also have evaluated macrophages (by F4/80) in the liver of the BDL-induced hepatic fibrosis mice. Our results showed that treatment with SolB reduced the number of macrophages, staining by specific marker of F4/80, in mice induced by BDL (Supplementary Figure S1F). These results demonstrated that SolB attenuates hepatic injury and fibrosis induced by BDL in mice.

Combination therapy for liver fibrosis is considered to be very attractive. Targeting several vital but very different pathways to reduce chronic inflammation and ECM deposition would more effectively address liver fibrosis (Schuppan and Kim, 2013; Schuppan, 2015). So we reasoned that the combination of SolB with another compound that target HSC activation would exhibit better anti-fibrosis effect. Activated HSCs play a central role in the pathogenesis of hepatic fibrogenesis, in which a continuous synthesis of collagen is mediated by the TGF-β1/Smad signaling pathway (Lee and Friedman, 2011). We screened compounds that could inhibit TGF-β1/Smad signaling in LX-2 cells (human hepatic stellate cell line) and found that Wedelolactone (WeD) could suppress TGF-β1/Smad pathway (Supplementary Figures S2A,B). Next we tested whether WeD could inhibit TGF-β1/Smad signaling to suppress the activation of HSCs in vitro. We detected the expression and phosphorylation of Smad3 in the TGF-β1 signaling cascade. The result demonstrated a reduction in the expression of Smad3 and its phosphorylation after treatment with WeD in LX-2 cells pretreated with TGF-β1 (Figure 3A; Supplementary Figure S2C). WeD also inhibited TGF-β1-mediated induction of α-SMA (Figures 3A,B; Supplementary Figure S2C). These results demonstrated that WeD could suppress the activation of HSCs by targeting the TGF-β1/Smad signaling pathway.

Previous studies have demonstrated that TNF-α and IL-1β produced by hepatic macrophages can promote the survival of activated HSCs (Gieling et al., 2009; Osawa et al., 2013; Pradere et al., 2013). Wedelolactone has been reported to be an inhibitor of IKK that is critical for activation of NF-κB by mediating phosphorylation and degradation of IκBα (Kobori et al., 2004), activation of NF-κB signaling leads to expression of downstream target genes including TNF-α and IL-6. Hence, we next tested whether WeD affected the production of TNF-α and IL-6 in LPS-treated macrophages, the result showed that WeD dose-dependently inhibited the production of TNF-α and IL-6 in LPS-treated BMDMs (Figures 3C,D). The maturation and release of IL-1β are mediated by the inflammasome. WeD has been reported to inhibit the activation of the NLRP3 inflammasome (Wei et al., 2017; Cheng et al., 2019). We further confirmed the effect of WeD on NLRP3 inflammasme activation and release of IL-1β. Our data showed that WeD inhibited nigericin-induced caspase-1 cleavage and IL-1β production in a dose-dependent manner in LPS-primed BMDMs (Figure 3E; Supplementary Figure S2D), suggesting that WeD inhibited the production of IL-1β by directly blocking the activation of the NLRP3 inflammasome. Taken together, these results demonstrated that WeD can directly inhibit the production of TNF-α and IL-1β in macrophages.

Considering the inhibitory effect of WeD on HSC activation and inflammation, we further tested the role of WeD in liver fibrosis in a mouse model of hepatic fibrosis induced by BDL. Our result of histological assessment of liver sections by H&E staining, Sirius red, and Masson staining showed that treated with WeD significantly alleviated hepatic injury and fibrosis in BDL-induced mice (Figures 4A,B). Consistent with the histopathology analysis, WeD treatment reduced the ALT, AST, DBIL, TBIL, and TBA serum levels induced by BDL in a dose-dependent manner in mice (Figures 4C,D; Supplementary Figure S3A–C).

It has been reported that HSCs play a crucial role in hepatic fibrosis and α-SMA is a unique marker of activated HSCs (Nouchi et al., 1991; Shang et al., 2018). So we detected the expression of α-SMA by immunohistochemical staining and qPCR. Our result showed that the expression of α-SMA in mice with hepatic fibrosis induced by BDL could be reduced by WeD treatment (Figures 4A,E). Our results showed that treatment with WeD reduced the number of macrophages, staining by specific marker of F4/80, in mice induced by BDL. Consequently, this observation demonstrated that WeD also inhibited the activation of HSCs in vivo. Then we examined the expression of TGF-β and PDGF in mice. The data showed that the expression of TGF-β and PDGF was also inhibited by WeD treatment with BDL-induced hepatic fibrosis by qPCR (Figures 4F,G). WeD treatment also significantly reduced BDL-induced increase of collagen I, as revealed by immunoblotting (Figure 4H; Supplementary Figure S3H) and qPCR analysis (Supplementary Figure S3D). Consistent with the effect in vitro, WeD inhibited both the phosphorylation of Smad3 and the expression of Smad3 in a dose-dependent manner (Figure 4H; Supplementary Figure S3H). we have evaluated macrophages (by F4/80) in the liver of the BDL-induced hepatic fibrosis mice. Our results showed that treatment with WeD reduced the number of macrophages, staining by specific marker of F4/80, in mice induced by BDL (Supplementary Figure S3G). We have demonstrated that WeD dose-dependently inhibited the production of TNF-α and IL-6 in LPS-treated BMDMs and the production of IL-1β by directly blocking the activation of the NLRP3 inflammasome in BMDMs. Consistent with the effect in BMDMs, WeD treatment could inhibit the expression of IL-β and IL-6 (Supplementary Figure S3E,F). The results demonstrated that WeD could attenuate inflammation in liver in vivo. Thus, these results demonstrated that WeD attenuates hepatic injury and fibrosis induced by BDL in mice and inhibited the TGF-β/Smad-mediated activation of HSCs in vivo.

Next we examined if SolB and WeD would show protective effects in another hepatic fibrosis mouse model. We evaluated the anti-fibrosis effect of SolB and WeD in CCl4-induced hepatic fibrosis model mice, and the data showed that both SolB and WeD alleviated hepatic fibrosis induced by CCl₄ in mice, suggesting their potent anti-fibrosis effect. Next we tested whether the anti-fibrosis effect mediated by the combination of SolB and WeD was more potent in comparison with that mediated by individual treatments of SolB and WeD. Mice with CCl₄-induced hepatic fibrosis were randomly divided into five groups and received gavages with PBS (model group), colchicine (a positive control drug), 20 mg/kg WeD, 40 mg/kg SolB, or a combination of WeD (20 mg/kg) and SolB (40 mg/kg). Compared with control mice, mice with CCl₄-induced hepatic fibrosis showed serve hepatocyte damage and hepatic fibrosis, evidenced by HE, Sirius red staining, and Masson staining (Figures 5A–C). WeD, SolB and their combination treatment were all shown to prevent the progression of liver fibrosis in the mouse model with CCl₄-induced hepatic fibrosis. The synergistic inhibitory effect of SolB and WeD on hepatic fibrosis was significantly better than that of WeD or SolB (Figures 5A–C). Consistent with the results of liver histology assessment, combined treatment of WeD and SolB also more efficiently reduced the hydroxyproline content in liver tissues than WeD or SolB alone (Figure 5E).

To further examine the antifibrotic effect of SolB combined with WeD, the expression of α-SMA was evaluated by immunohistochemistry, results showed that CCl₄ significantly promoted α-SMA expression in the portal area of fibrotic liver tissue, while treatment with WeD, SolB, or their combination suppressed the expression of α-SMA in fibrotic liver tissues (Figures 5A,D). Furthermore, the expression of α-SMA in mice co-treated with SolB and WeD was significantly lower than that in mice treated with WeD or SolB alone (Figures 5A,D). Moreover, compared with the individual WeD or SolB treatment, the combined treatment of WeD and SolB more effectively reduced the mRNA level of α-SMA in liver tissues of mice (Figure 5F). Previous studies have shown that cytoglobin, which is the fourth globin in mammals and function as a local gas sensor, is a promising new marker that discriminates between myofibroblasts derived from stellate cells and those from portal fibroblasts (Kawada, 2015; Thuy et al., 2015). Thus, we evaluated the cytoglobin by qPCR in CCl4-induced hepatic fibrosis mice, and the results showed that combination treatment of WeD and SolB significantly reduced the expression of cytoglobin induced by CCl4 in mice compared to that by treatment with WeD and SolB alone, suggesting that the combination of WeD and SolB is more effective in the treatment of liver fibrosis (Supplementary Figure S4A). Western blot analysis also demonstrated that combination treatment of WeD and SolB significantly decreased the expression of collagen and Smad as well as phosphorylation of IκB and Smad3 induced by CCl₄ in mice compared to that by treatment with WeD and SolB alone, suggesting that WeD combined with SolB could inhibit the activity of the TGF-β1 and NF-κB pathway more effectively than WeD or SolB alone in vivo (Figure 5G; Supplementary Figure S4B). These results indicate that the combination of WeD and SolB is more effective in the treatment of liver fibrosis owing to the regulation of several signaling pathways associated with hepatic fibrosis.