MiRNA negative control (miR-NC), miR-26b-5p mimics/inhibitor, adenoviral vectors expressing Jag1 (Ad-Jag1), adenoviral vectors expressing a control scrambled sequence (Ad-Ctrl), adenoviral vectors expressing shRNA against Jag1 (Ad-shJag1) and adenoviral vectors expressing the scrambled shRNA (Ad-shCtrl) were obtained from GenePharma biotechnology (Shanghai, China). The molecular weight and molecular formula of kaempferol (≥97% purity, Sigma, St Louis, MO, United States) are 286.24 and C₁₅H₁₀O₆, respectively (Figure 1A).

Liver fibrosis was produced by intraperitoneal injection with carbon tetrachloride (CCl₄) (Sigma, St Louis, MO, United States) in olive oil (10%, 7 μL/g mice) two times weekly for 4 weeks in 8-week-old C57BL/6 J mice (n = 6). Meanwhile, mice in control group were given the same volume of olive oil (n = 6). In kaempferol group, mice (n = 6) were treated with kaempferol (10 μmol/L, 1 ml) daily via intraperitoneal injection during CCl₄ period (Xu et al., 2019). This project was approved by the University Animal Care and Use Committee of Wenzhou Medical University. Mice, obtained from the Experimental Animal Center of Wenzhou Medical University, were sacrificed under anesthesia after the end of animal experiment. Finally, blood samples and liver tissues were collected for further analysis such as Sirius red staining.

Briefly, the tissues were immersed in 4% formalin for fixation, and then the formalin-fixed tissue is degreased and rehydrated. Next, the sections, blocked in 10% BSA, were in the incubation with anti-α-SMA primary antibody at 4°C for at least 12 h. Then, the sections were incubated with a horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody for 30 min. Finally, quantitative analysis for α-SMA-positive areas was performed under the microscope (Carl Zeiss, Germany).

According to the manufacturer’s instructions, the hydroxyproline content in liver tissue was measured using the Hydroxyproline Colorimetric Assay kit (BioVision, San Francisco, CA) and liver weight was used to normalize the relative abundance of hydroxyproline. Briefly, liver tissues were mixed with HCl for homogenization, and then hydrolyzed for 12 h. The supernatant was obtained from the lysate, which was centrifuged at 12,000 g for 10 min at 4°C.

According to the previously described enzyme digestion and density gradient centrifugation, isolated primary HSCs were obtained (Chang et al., 2014). The purity of primary HSCs was >98%, which was defined by α-SMA and Desmin immunocytochemical staining. Primary 1-day-old HSCs were treated with 10 μM kaempferol or 20 μM curcumin (Cur) for 24 h, respectively. Cells were transfected with 100 nM miR-26b-5p mimic using lipofectamine RNAiMAX.

The Cignal Finder 10-Pathway Reporter Array (QIAGEN, MA, United States) was performed to determine the kaempferol-related signaling pathways. Attractene (1 μL/well) was distributed into Cignal Finder 10-Pathway Reporter Array plate. Then, primary HSCs cell suspension was diluted in Opti-MEM medium and seeded in each well. Cells were treated with 10 μM kaempferol on the following day. After 24 h of treatment, reverse transfection reagent and Opti-MEM medium were removed. Dual-Glo Luciferase Reagent (75 μL) was then added to each well and plates were incubated for 10 min at room temperature. Finally, firefly and Renilla luciferase activities were measured following the manufacturer’s recommendations.

Total RNA was isolated from primary HSCs as well as liver tissues using the Tiangen RNA extraction reagent kit. Each sample (1 µg RNA) was reversely transcribed into complementary DNA (cDNA) using a reverse-transcription (RT) reagent kit (Takara Biotechnology Co., Ltd., Dalian, China). Then, Real-time PCR was performed using SYBR Premix ExTaq (Takara). As shown in Supplementary Table S1, the primers of α-SMA, alpha-1(I) collagen (Col1A1), hes family bHLH transcription factor 1 (Hes1), Hes5, Notch1-4, Delta-like 1 (Dll1), Dll3, Dll4, Jagged 1 (Jag1), Jag2 and GAPDH were designed. TaqMan MicroRNA Assays (Applied Biosystems, Foster City, CA) were performed to detect 10 miRNAs expressions. GAPDH and U6 were used as endogenous controls for mRNAs and for miRNAs, respectively.

The proteins from primary HSCs and liver tissues were extracted using RIPA extraction buffer. The protein samples of interested were separated by 10% SDS-PAGE electrophoresis, and then transferred to PVDF membranes. The primary anti-Jag1, anti-α-SMA, type I collagen and anti-GAPDH (an internal control) were added in PVDF membranes and incubated overnight at 4°C. Then, the second antibody was added and incubated at room temperature for 1 h.

In brief, cells were fixed in 4% paraformaldehyde for 20 min at room temperature, and then permeabilized with PBS-0.5% Triton X-100 for 15 min. Subsequently, blocking with 5% BSA in PBS for 1 h at 37°C and incubated with the following primary antibodies overnight at 4°C in a humidified chamber: anti-α-smooth muscle actin (α-SMA) (1:100) and anti-type I collagen (1:100). On the next day, the cells were incubated with secondary antibodies conjugated with Alexa Fluor 488 or 546 (Invitrogen), respectively. Nuclei were stained using 4,6-diamidino-2-phenylindole (DAPI), and cells were observed using a Nikon Eclipse Ti-E inverted fluorescence microscope (Nikon, Tokyo, Japan).

EdU assay was used to detect cell proliferation. After kaempferol or Cur treatment, cells were labeled with EdU for 12 h, and EdU⁺ population represented the proliferating cell population.

As described previously, miR-26b-5p or miR-NC was co-transfected with pmirGLO-Jag1 into HEK293T cells by lipofectamine-mediated gene transfer (Yu et al., 2017). After 24 h, the luciferase signals were detected using the Dual-Luciferase^® Reporter Assay System (Promega). Firefly luciferase intensity was normalized over Renilla luciferase signal.

Data were presented as the means ± SD. One-way analysis of variance (one-way ANOVA) was employed for multiple groups. Differences between two groups were compared using student’s t-test. p < 0.05 was considered significant. The SPSS software was used for all statistical analyses (version 16.0; SPSS, Chicago, IL).

The roles of kaempferol were firstly examined in CCl₄-induced mouse liver fibrosis model. As shown by Sirius red staining and immunohistochemical analysis, both collagen deposition and α-SMA expression were induced by CCl₄, respectively (Figures 1B,C). Hydroxyproline assays further confirmed enhanced collagen level by CCl₄ (Figure 1D). Liver function of CCl₄ mice was shown to be injured, associated with increased ALT and AST (Figures 1E,F). Notably, kaempferol significantly inhibited CCl₄-caused collagen (Figures 1B,D). Kaempferol resulted in the reduction in α-SMA expression (Figure 1C). Furthermore, kaempferol ameliorated liver function (Figures 1E,F). Our results demonstrate that kaempferol ameliorates CCl₄-induced liver fibrosis in vivo.

Whether kaempferol contributed to the suppression of HSC activation was subsequently investigated. Cur, which has been demonstrated to suppress HSC activation, was used as a positive control (Zhang et al., 2014). Obviously, Cur inhibited HSC proliferation, HSC transdifferentiation as well as collagen expression. Our results showed that Cur induced a significant reduction in HSC proliferation, HSC transdifferentiation and collagen level (Figures 2A–C). As shown by Edu assays, kaempferol caused no significant change in cell proliferation (Figure 2A). Kaempferol obviously inhibited HSC transdifferentiation, with reduced α-SMA mRNA level (Figure 2B). Moreover, kaempferol resulted in the suppression of Col1A1 mRNA level (Figure 2C). Accordingly, immunoblot results indicated that both α-SMA protein and type I collagen were inhibited by kaempferol (Figure 2D). Overall, the above data suggest that kaempferol plays an inhibitory role in the activation of HSCs.

Next, the underlying molecular mechanism of kaempferol in the regulation of HSC activation was examined. A pathway reporter array was used to rapidly identify the relevant pathways in cells with kaempferol. Clearly, kaempferol resulted in an obvious suppression in six signaling pathways including TGF-β, Notch, NF-κB, MAPK/JNK, MAPK/ERK and Wnt, with the lowest expression in Notch pathway (Figure 3A). Our results suggest that the Notch pathway may take a part in the effects of kaempferol-inhibited HSC activation (Figure 3A). Subsequently, Hes1 and Hes5, which are the downstream of the Notch pathway, were examined in cells after kaempferol treatment. In comparison with non-treated cells, kaempferol induced a significant reduction in the mRNA expressions of Hes1 and Hes5 (Figure 3B). Accordingly, the protein levels of Hes1 and Hes5 were down-regulated by kaempferol (Figure 3C). Generally, Notch signaling is initiated by the activated Notch receptors and ligands (Kovall et al., 2017). In kaempferol-treated cells, the expressions of Notch receptors as well as ligands were measured. Our results showed that the level of Jag1 was reduced after kaempferol treatment, whereas others not (Figure 3D). Consistent with the mRNA result, Jag1 protein was also decreased in kaempferol-treated cells (Figure 3E), indicating that Jag1 may participate in the effects of kaempferol on HSC activation inhibition. Thus, Jag1 expression was detected in vitro and in vivo during HSC activation. It was found that Jag1 expression was obviously enhanced in primary HSCs during cell culture, which was inhibited by kaempferol (Figures 3F,G). Similar with the results in vitro, Jag1 expression was up-regulated in CCl₄ mice, which was also suppressed by kaempferol (Figures 3H,I). Whether Jag1 plays a key role in the inhibitory effects of kaempferol on the activation of HSCs was further explored. As shown in Figure 4A and Supplementary Figure S1, overexpression of Jag1 was found in Ad-Jag1 group, while loss of Jag1 was shown in Ad-shJag1 group. Compared with kaempferol group, the change of proliferative activity between Ad-Jag1 and Ad-shJag1 group was no significant (Figure 4B). Interestingly, overexpression of Jag1 blocked down kaempferol-inhibited α-SMA and Col1A1 (Figures 4C,D). On the contrary, loss of Jag1 enhanced the inhibition of α-SMA and Col1A1 in kaempferol-treated cells (Figures 4E,F). Additionally, analysis of immunofluorescence confirmed that kaempferol-reduced type I collagen (green fluorescence) and α-SMA (red fluorescence) was blocked down by Ad-Jag1 (Figure 4G and Supplementary Figure S2A). These data suggest that the inhibitory role of kaempferol in the regulation of HSC activation is via suppressing Jag1, leading to the inactivation of Notch pathway.

Recently, miRNA-mediated Jag1 expression has been found in various human diseases (Zhang et al., 2020a; Zhang et al., 2020b). Herein, bioinformatics analysis (miRDB) was performed to identify the potential miRNAs targeting Jag1, and 10 miRNAs with the highest target score were shown in Figure 5A. Interestingly, in kaempferol-treated cells relative to the untreated cells, only miR-26b-5p was significantly up-regulated, whereas other predicted miRNAs not (Figure 5B). Therefore, miR-26b-5p was selected for the next experiments. Both in vitro and in vivo, the levels of miR-26b-5p were examined during liver fibrosis. Our results showed that reduced miR-26b-5p was found in primary HSCs at Day 4 as well as CCl₄ mice (Figure 5C), indicating that the miR-26b-5p was negatively correlated with the Jag1. Thus, the Jag1 target region was cloned in the pmirGLO plasmid to generate a Jag1 pmirGLO luciferase reporter containing the miR-26b-5p binding sites (Figure 5D). Luciferase activity assays were performed to determine whether Jag1 is a target of miR-26b-5p. Clearly, miR-26b-5p induced a significant reduction in luciferase activity driven by the wild-type 3′UTR of Jag1, and there was no significant difference in mutant-type 3′UTR of Jag1 (Figure 5E). In primary HSCs, miR-26b-5p overexpression down-regulated Jag1 mRNA as well as protein expression (Figures 5F,G). Our data suggest that miR-26b-5p induced by kaempferol targets Jag1.

Whether miR-26b-5p plays a crucial role in the inhibitory effects of kaempferol on HSC activation was further investigated. After transfection with miR-26-5p inhibitor, the proliferation of kaempferol-treated cells was detected by EdU analysis. In line with our previous results, miR-26-5p had no effect on cell proliferation in comparison with kaempferol group (Figure 6A). Notably, loss of miR-26-5p blocked down kaempferol-reduced α-SMA mRNA expression (Figure 6B). Likewise, miR-26-5p inhibitor significantly suppressed kaempferol-inhibited Col1A1 expression (Figure 6C). Accordingly, the results of immunoblot indicated that miR-26-5p inhibitor restored α-SMA and type I collagen in cells with kaempferol (Figure 6D). Results of immunofluorescence further revealed that miR-26-5p inhibitor contributed to the restoration of α-SMA and type I collagen in kaempferol-treated cells (Figure 6E and Supplementary Figure S2B). Moreover, kaempferol-inhibited Jag1 was restored by miR-26b-5p inhibitor (Figures 6F,G). Therefore, kaempferol suppresses HSC activation, at least in part, through miR-26b-5p-mediated Jag1 axis.