A novel Chinese Medicine JY5 Formula Alleviates Hepatic Fibrosis Through Inhibiting Notch Signaling Pathway

Background: Advanced liver brosis can lead to cirrhosis, resulting in an accelerated risk of liver failure and hepatocellular carcinoma. It is necessary to develop an effective antibrotic strategy. It has been reported that Fuzheng Huayu formula (FZHY) had a remarkable anti-hepatic brosis effect. Here, We obtain a new antibrotic composition, which consists of the main active ingredients of FZHY formula and investigate its mechanism of pharmacological action. Methods: The main active ingredients of FZHY through the quantitative analysis in FZHY extracts and FZHY-treated plasma and liver in rats were investigated. The best anti-brotic composition of the main active ingredients was studied through the uniform design and validation experiments in vivo and its mechanism was evaluated in CCl 4 - and BDL-induced liver brosis models in rats and mice and TGF-β1-indued LX-2 cells activation model in vitro. Results (cid:0) A novel composition, namely JY5 formula, which consisted of Salvianolic acid B, schisantherin A and amygdalin, the main active components of FZHY, could signicantly alleviated hepatic hydroxyproline content and collagen deposition in CCl 4 - and BDL-induced brotic liver in rats and mice. Further studies showed that JY5 could inhibit the activation of hepatic stellate cells (HSCs) through inactivating Notch signaling in vivo and in vitro. Conclusions: We found a novel composition JY5 formula, which had an anti-hepatic brotic effect through inhibiting Notch signaling pathway, consequently suppressing HSCs activation. These results may provide some adequate scientic basis for the clinical research and application of JY5 formula, as a potential new therapeutic candidate for liver brosis. cells; extracellular matrix; traditional Chinese medicine; KCs, Kupffer cells; LSECs, liver sinusoidal endothelial cells; CCl 4 , carbon tetrachloride; BDL, bile duct of ligation; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; TBil, total bilirubin; DBil, direct bilirubin; TBA, total bile acids; Hyp, hydroxyproline; IHC, immunohistochemical; H&E, hematoxylin & eosin; SR, sirius red; α-SMA, α-smooth muscle actin; RBP-κB, recombination signal binding protein-κB; HCC, hepatocellular carcinoma; NICD, Notch intracellular domain.


Background
Liver brosis is a common pathological feature of chronic liver diseases, including chronic viral hepatitis, metabolic associated fatty liver disease, cholestatic liver disease, etc. It is a consequence of an abnormal wound healing response, characterized by excessive deposition of extracellular matrix (ECM), especially of collagen. If the injury persists, liver brosis can progress to cirrhosis, hepatocellular carcinoma, ultimately leading to liver failure [1]. The effective treatment for liver brosis is critical to block the progression of chronic liver diseases. Some clinical studies [2] have shown that liver brosis, even early cirrhosis, is reversible, which provided reliable evidence for undertaking clinical researches on anti-hepatic brosis drugs. With several kinds of animal models of liver brosis becoming more and more mature, the pathogenesis of liver brosis has been deeper understanding in the last a few decades [3]. A number of anti-hepatic brosis drug studies have been conducted in recent years, some of which have been researched in clinical trials [4]. However, there are not any clinically approved or effective medical therapies aimed speci cally at hepatic brosis so far.
Traditional Chinese medicine (TCM) has remarkable clinical effects on the treatment of liver brosis. Among these, Fuzheng Huayu formula (FZHY) is the rst Chinese medicine in the eld of hepatology, nishing the phase II clinical trial of anti-hepatic brosis post hepatitis C in USA [5]. However, there are some challenges in TCM, such as multi-component, multi-target, ill-de ned active ingredients and mechanisms, to some extent, which limit their clinical application and in-depth research. In recent years, based on these questions, our group has conducted a large amount of researches to reveal the mechanisms of FZHY in the prevention and treatment of chronic liver diseases, including inhibiting in ammatory response, protecting hepatocytes (to relieve hepatocyte damage and to inhibit hepatocyte apoptosis), inhibiting hepatic stellate cells (HSCs) activation, reducing collagen deposition, inhibiting Kupffer cells (KCs) activation, inhibiting liver sinusoidal endothelial cells (LSECs) capillarization and angiogenesis, promoting liver regeneration, etc [6,7]. We further studied the effect of anti-brosis of different compounds of FZHY. For instance, phenolic acids in Danshen play a prominent role in inhibiting the in ammatory response, protecting hepatocytes and inhibiting HSCs activation [8][9][10][11][12]. Amygdalin, as one of the major active compounds of peach kernel, has a main role in inhibiting the in ammatory response, reducing collagen deposition and inhibiting HSCs activation [13,14]. The lignan compounds from Schisandrae take a prominent position in protecting hepatocytes and inhibiting HSCs activation [15,16]. These ndings suggest that the related components in FZHY may be potential anti-brotic bioactive ingredients.
Notch signaling is a highly conservative pathway during evolution. Notch receptors interact with the ligands on the surface of adjacent cells, then cleave inside of the cell membrane and translocate into the nucleus, followed by regulating transcription of multiple target genes. Previous studies [17] have shown that as an important inter-or intra-cellular signaling pathway, Notch plays an important role in liver development and pathophysiology. Notch has a great impact on the occurrence and development of hepatic brosis, which can interact with TGF-β, Hedgehog, Hippo signaling pathways to mediate cell-cell interactions. Activation of HSCs is a critical cellular event in liver brosis. Studies [18] have shown that HSCs transdifferentiate into myo broblasts accompanied by activation of Notch signaling pathway. After Notch activity levels are suppressed, this process could be reversed. In addition, with the progression of rat liver brosis induced by carbon tetrachloride (CCl 4 ) and bile duct of ligation (BDL), Notch signaling pathway is activated signi cantly. E cient inhibition of Notch pathway can signi cantly mitigate rat liver brosis, and reduce hepatocyte apoptosis to some extent [19]. Thus, targeting Notch signaling pathway can regulate activation of HSCs, and thereby suppress the occurrence and development of hepatic brosis. Therefore, Notch promises to be one of potential therapeutic targets for the treatment of hepatic brosis.
In this study, we found that salvianolic acid B, schisantherin A and amygdalin were the main active ingredients of FZHY formula through quantitative analysis in FZHY extracts and FZHY-treated plasma and liver in rats.
Then, a novel composition, namely JY5, was obtained through the uniform design and validation experiments. The anti-hepatic brosis e cacy of JY5 is comparable to that of FZHY in CCl 4 -induced hepatic brosis in rats.
Further studies demonstrated that JY5 alleviated liver brosis by inhibiting the activation of HSCs via the inhibition of Notch signaling pathway.

Animals
Adult Wistar or Sprague-Dawley (SD) rats (male, weighed 160-180g, SPF) were purchased from Shanghai Xipuer-Bikai Laboratory Animal Co., Ltd, and fed in the Laboratory Animal Center at School of Pharmacy, Fudan University. Adult C57/BL6 mice (male, aged 6-8weeks, weighed 18-20g, SPF) were purchased from Shanghai Southern Model Biotechnology Co., Ltd, and maintained in Shanghai Research Center of the Southern Model Organisms. Rats and mice were housed under constant conditions (ambient temperature 25 ± 2 °C, relative humidity 40-60% and 12/12h light-dark cycle) with free access to standard diet and water. Fuzheng Huayu (FZHY) decoction: A mixture of salvia miltiorrhiza at 533g, peach kernel at 133 g and gynostemma pentaphylla at 400g was heated to boiling with water for 2h for the rst time and 1.5h for the second time, respectively. The combined decoction was ltered, and concentrated to the relative density at 1.20 g/ml (50~55 °C). After cooling down, the decoction was precipitate by adding 70% alcohol, and then the ltrate 1 were generated after ltration and concentration. A combination of cordyceps mycelium powder at 267g, and schisandra chinensis at 133 g was heated with 70% alcohol for 2h for the rst time and 1.5h for the second time, respectively, and the combined decoction was ltered and concentrated as ltrate 2. Pollen pini at 133g were in ltrated with 50% alcohol for 4h for the rst time and 2h for the second time, respectively, and the combined decoction was ltered, and concentrated as ltrate 3. The ltrates 1-3 were combined and concentrated to 800ml at the 2g raw drug/mL.

A pharmacokinetic study of FZHY decoction in rats
This study was consistent to the guidelines of the Institutional Animal Care and Use Committee (IACUC) of the Shanghai Institute of Materia Medica, Chinese Academy of Sciences (Shanghai, China). The experimental protocol was shown in Supplementary material: material and methods 1.1.

Instrumentation
There were 16 compounds determined in various biosamples by an UltiMate 3000 ultrahigh-performance liquid chromatograph linked to Qactive qdrupole electrostatic eld orbital trap high resolution mass spectrometer, connecting to electrospray ionization source (Thermo Scienti c, US). The operating parameters were set as shown in Supplementary material: material and methods 1.2.
Experimental liver brosis models CCl 4 -induced liver brosis rat model, BDL-induced liver brosis rat model and CCl 4 -induced liver brosis mice model were used in this study. And the experimental protocols were shown in Supplementary material: material and methods 1.3.

Histopathological and immunohistochemical (IHC) analysis
Liver injury and brosis were assessed with hematoxylin and eosin (H&E) and Sirius Red staining using 4 μm thick para n-embedded liver sections. IHC stainings of Col-I, Col-IV, α-SMA and Desmin were performed. The detailed protocols were shown in Supplementary material: material and methods 1.4.

Western blot analysis
Total proteins in Liver tissues or cells were extracted using RIPA lysis buffer containing proteinase and phosphatase inhibitor (Cat No. P0013B, Biyuntian Biotechnology Co., Ltd., Shanghai, China). Total proteins concentration was determined using BCA protein assays kit (lot.TD265229, Thermo Fisher Scienti c).
Denatured at 100 °C for 5 min, 30-50μg proteins were mixed into sample loading buffer and loaded into each lane. Proteins samples were separated by 10% SDS-PAGE, then transferred into polyvinylidene uoride (PVDF) membranes by wet transfer. Blocking was performed with 5% BSA in 1% PBST at room temperature for 60 min, before incubation with primary antibody (see Table S1) overnight at 4°C on a rocker. The following day, after incubation with uorescence-labeled second antibody (see Table S1) avoiding light for 1 h at room temperature. The PVDF membranes were scanned using the Odyssey infrared scanner (LI-COR Biosciences).
The greyscale values relative to GAPDH of the target proteins were analyzed using Image-J.1.51j8 software.
qRT-PCR analysis Total RNA in liver tissues or cells was extracted and reverse-transcribed respectively using Nucleic Acid Puri cation Kit (Code: NPK-201F, lot. 742100, TOYOBO CO., LTD. Osaka, Japan) and ReverTra Ace qPCR RT Kit (Code: FSQ-301, Lot.616800, Osaka, Japan). The Speci c operation was performed referring to the instructions. The qRT-PCR primers sequences information was listed in Table S2. The following a two-step approach PCR cycling program was: at 95°C for 60s, 40 cycles of 95°C for 15 s and 60°C for 60s, followed by melt curve analysis. The GAPDH gene was used as the internal reference for normalization of the target genes.
The relative mRNA expression of each group was calculated using the 2 -ΔΔCt method.

Statistical analysis
All data were analyzed by using the SPSS 21.0 software package. All the measurement data were expressed as means ± standard deviation (SD). Comparisons between multiple groups were analyzed by the one-way analysis of variance (ANOVA), followed by the least signi cant difference (LSD) test. p<0.05 was considered statistically signi cant. In addition, the pharmacokinetic parameters were calculated by noncompartmental analysis in WinNonlin software with a sparse sampling algorithm (Pharsight 6.2, NC, USA).

Quantitative analysis in FZHY decoction and FZHY-biological samples
The chemical structure and concentration of 16 compounds in FZHY decoction (2 g/mL) were displayed in  Table S3, respectively. The concentration-time curves of compounds in the plasma and liver after oral administration of FZHY decoction (20 g/kg) in rats was shown in Fig. 1a and 1b, and the corresponding PK parameters were shown in Table 1. The summary of the contents and AUCs of compounds in the FZHY decoction or FZHY-biological samples were shown in Fig. 1c. Table 1 The PK parameters of 11 compounds in the portal vein plasma, systemic plasma and liver, following oral administration of FZHY decoction in rats. Following oral administration of FZHY decoction, there were only 5 compounds detected in the liver. The T max s of schisandrol A, schisandrin B and schisantherin A were at 0.5 h, similar to those in the plasma. In contrast, those of schisandrol B and amygdalin were at 14 h and 7 h, respectively, consistent to their multi-peak phenomenon in the concentration-time curves. Contrary to those in the plasma, amygdalin exhibited the highest hepatic exposure, followed by schisantherin A, whose exposure were more than 1000 h*ng/g. Schisantherin A belong to the middle exposure group (100 h*ng/g AUC 1000 h*ng/g), and the hepatic exposure of schisandrol A and schisandrin B was very low (AUC ≤ 100 h*ng/g). Finally, salvianolic acid B, schisantherin A and amygdalin, who were of the maximum quantity in the FZHY decoction, plasma and liver, respectively, were selected to combined to evaluate the anti-hepatic brosis effect in vivo.

JY5 signi cantly alleviate hepatic injury and collagen deposition in CCl 4 -induced rat and mouse liver brosis
Compared with the control group, the levels of serum ALT and AST were signi cantly increased in the CCl 4 group. After treatment with JY5 or FZHY, the levels of ALT and AST were signi cantly decreased (Fig. 2a, 2b and Fig. S2d, S2e). The serum AST level was decreased in the SORA group compared with the CCl 4 group (Fig. 2b).
H&E staining showed that the hepatic lobular structure was severely collapsed with more complete pseudolobules formed in the CCl 4 group. And brous tissue became denser, and hepatocytes were disordered and ballooning degeneration. There were a large number of in ammatory cells in ltration surrounding the hepatic sinusoid, central vein and portal tract. The above lesions were obviously attenuated with less pseudo-lobules and in ammatory cells in ltration, after treatment with JY5 or FZHY or SORA ( Fig. S2a and Fig. 2c, upper panel).
SR staining showed that compared to the control group, the collagen deposition was obviously increased in the CCl 4 group. The brotic septum became signi cantly widened and distributed from the portal tract to the periphery in a reticular manner, forming pseudo-lobules with varying sizes. In contrast, the collagen deposition was obviously decreased, the brotic septum became narrower, and pseudo-lobules structures were observed less in the JY5 or FZHY or SORA treated groups ( Fig. S2a and Fig. 2c, middle panel). Both the hepatic Hyp content and collagen deposition were signi cantly increased in the CCl 4 group, compared to the control group.
The above indicators were signi cantly reduced after the intervention with JY5 or FZHY or SORA (Fig. S2b, S2c and Fig. 2d, 2e). These results demonstrated that JY5 formula had a signi cantly anti-liver brosis comparably to that of FZHY.
IHC staining showed that compared with the control group, numerous Col-expression were visible in the brotic septum in the CCl 4 group. By contrast, JY5 and SORA could signi cantly reduce Col-expression in the liver tissue (Fig. 2c, lower panel). In addition, qRT-PCR results showed that Col-mRNA expression was signi cantly more elevated in the CCl 4 group than that in the control group. Whereas compared to the CCl 4 group, Col-mRNA expression was signi cantly reduced in JY5 treated group (Fig. 2f).
Consistent with the CCl 4 -induced rat liver brosis model, JY5 could signi cantly alleviate hepatic injury and collagen deposition in CCl 4 -induced liver brosis in mice (Fig. 3).

JY5 signi cantly alleviate hepatic injury and collagen deposition in BDL-induced rat liver brosis
Compared with the sham group, the levels of ALT, AST, TBil, DBil, TBA and ALP were signi cantly increased in the BDL group. After JY5 or DAPT treatment, the levels of ALT, AST, TBil, DBil, TBA and ALP were signi cantly decreased ( Fig. 4a-4f).
Consistent with the CCl 4 -induced liver brosis, JY5 could reduce the hepatic Hyp content and collagen deposition, meanwhile down-regulate the expressions of Col-and Col-IV in BDL-induced rat liver brosis ( Fig. 4g-4i). These results suggested that JY5 could signi cantly alleviate hepatic injury and collagen deposition in BDL-induced liver brosis in rats.

JY5 signi cantly represses the activation of HSCs in vivo
Both in the CCl 4 -induced rat and mouse liver brosis experiments, IHC staining showed that numerous α-SMA and Desmin expression were visible in the brotic septum in the CCl 4 group. By contrast, both α-SMA (+) cells and Desmin (+) cells were decreased in the JY5 and SORA treated group (Fig. 5a, 5d). Western blot and qRT-PCR showed that the α-SMA expression was signi cantly elevated compared to the control group in the CCl 4 group.
Whereas compared to the CCl 4 group, both α-SMA protein and mRNA expressions were signi cantly reduced in JY5 and SORA treated groups (Fig. 5b, 5c, 5e and 5f). Similarly, in the BDL-induced liver brosis experiment, the treatment effect of JY5 was consistent with these results in the CCl 4 -induced liver brosis experiments (Fig. 5g-5i). These results demonstrated that JY5 signi cantly represses the activation of HSCs in CCl 4 -and BDLinduced liver brosis.
JY5 signi cantly inhibit the activation of Notch signaling pathway in vivo In the CCl 4 -induced liver brosis rat experiment, qRT-PCR showed that the mRNA expressions of Notch2, Notch3, Notch4, Jagged1, Jagged2 and recombination signal binding protein-κB (RPB-κB) were signi cantly more up-regulated in the CCl 4 group than these genes in the control group. Whereas compared to the CCl 4 group, Notch2, Notch3, Notch4, Jagged1, Jagged2 and RPB-κB mRNA expressions were signi cantly reduced in JY5 and SORA treated groups (Fig. 6a). Western blot showed that the protein expression of RPB-κB was signi cantly increased compared to the control group in the CCl 4 group. RPB-κB protein expression was signi cantly more reduced in JY5 and SORA treated groups than that in the CCl 4 group (Fig. 6b). While in the CCl 4 -induced liver brosis mice experiment, JY5 could not only decrease the expressions of Notch2, Notch3, Notch4, Jagged1 and RPB-κB, but also down-regulate the expression of Dll1 (Fig. 6c and 6d). Consistent with the CCl 4 -induced rat liver brosis model, JY5 could decrease the expressions of Notch2, Notch3, Notch4, Jagged1 and RPB-κB in the BDL-induced liver brosis model. In addition, the expressions of Dll1, Dll4 and Jagged 2 were signi cantly decreased after treatment with JY5 ( Fig. 6e and 6f). These results suggested that JY5 could signi cantly inhibit the activation of Notch signaling pathway in CCl 4 -and BDL-induced liver brosis.
JY5 might inhibit activation of LX-2 cells induced by TGF-β1 via regulating Notch signaling pathway LX-2 cells were activated by TGF-β1 to observe the effect of JY5 at various concentrations in vitro. qRT-PCR showed that the mRNA levels of α-SMA, Col-, Notch3 and Jagged1 were signi cantly elevated in the TGF-β1 treated cells compared to the control cells. Whereas a-SMA Col-Notch3 and Jagged1 mRNA expressions were signi cantly reduced after treatment with various concentrations of JY5. And Col-Notch3 and Jagged1 mRNA expressions were signi cantly decreased in the high-dose of JY5 treated group, compared to the lowdose of JY5 treated group (Fig. 7b-7e). Western blot results showed that the protein expressions of a-SMA and RBP-κB were signi cantly increased after treated with TGF-β1. Compared to the TGF-β1 group, both a-SMA and RBP-κB protein expressions were signi cantly reduced in various concentrations of JY5 treated groups. Among of these, the protein expression of RBP-κB was signi cantly reduced in the high-dose of JY5 treated group compared to the middle-dose and low-dose of JY5 treated groups ( Fig. 7a and 7f). The above results suggested that JY5 might inhibit the activation of LX-2 cells induced by TGF-β1 via regulating Notch signaling pathway.

Discussion
Liver brosis is an abnormal repair response to tissue damage, characterized by excessive deposition of ECM, leading to the persistence and development of pathological scar. Hepatic brosis is common in most chronic liver diseases process, which is a clinically important problem to be urgently solved. In order to develop effective anti-hepatic brotic drugs, researchers have conducted a large number of basic and clinical studies. Despite achieving certain results in recent years, it has been shown that the most were stay at preclinical or clinical trials stages, even some nally ended up in failure with severe toxic side effects [20]. TCM has remarkable clinical effects in the treatment of liver brosis, which is closely correlated with its characteristics of multi-ingredients compatibility and multi-targets. But the ingredients of TCM are complex, and their mechanisms are not very clear, which somewhat increases the complexity of studies of TCM. As the intensive development of multidisciplinary crossover study, active ingredients screening, extraction and puri cation from Chinese herbs provide a new approach for TCM formula research.
Salvianolic acid B is the most main water-soluble phenolic acid compound. Numerous studies [8,9,12,21] have indicated that salvianolic acid B exerted signi cant anti-hepatic brosis effect through the following mechanisms: inhibiting the activation of HSCs via downregulating TGF-β1/Smads signaling pathway, protecting hepatocytes from apoptosis via inhibiting death receptor pathway, stabilizing the mitochondrial membrane and regulating NF-κB/IκBα signaling pathway. Amygdalin is considered to be the most major ingredient of peach kernel. It has been shown that [13,14] amygdalin can inhibit the activation of HSCs via downregulating TGF-β/CTGF signaling pathway, and induce activated HSCs apoptosis via upregulating Bax gene expression, subsequently exerting anti-hepatic brotic effect. Lignans are reported to be the main bioactive components of Schisandrae. Studies [15] have shown that these lignans could suppress in ammation, protect hepatocytes and inhibit the activation of HSCs via downregulating TGFβ/Smads and MAPK signaling pathway.
In this study, we respectively measured the content of various compounds in FZHY extract, plasma and liver in rats after intragastric administration with FZHY. We obtained three main bioactive ingredients of FZHY, salvianolic acid B, which was the highest content in FZHY extract, schisantherin A, which was the maximum exposure in plasma and amygdalin, which was the highest hepatic exposure in the liver. Then we conducted uniform design and validation experiments to explore their composition with best ratio treating in rat hepatic brosis model. Therefore, we got a new composition, namely JY5, which had an anti-brotic effect as effective as FZHY formula. Further studies have showed that JY5 could signi cantly decrease serum ALT and AST levels and inhibit in ammation reaction, at the same time, reduce collagen deposition in CCl 4 -induced or BDLinduced liver brosis models.
The activation of HSCs is a pivotal event in liver brosis. Under persistent stimulations from CCl 4 and BDL, HSCs are largely activated and transformed into myo broblasts, which causes excessive ECM accumulated in the liver, eventually leading to hepatic brosis formation. Activated HSCs, as one of the main sources of hepatic ECM, can secrete Col-and Col-proteins. It is well known that α-SMA is a speci c marker of activated HSCs. This study suggested that JY5 signi cantly reduce mRNA and protein expressions of α-SMA, and decreased Col-mRNA expression in CCl 4 -and BDL-induced liver brosis in rats and mice. The results were further con rmed in TGF-β1-indued LX-2 cells activation. JY5 could signi cantly down-regulate the mRNA and protein expressions of α-SMA in activated LX-2 cells induced by TGF-β1. Other than that, JY5 can downregulate Col-mRNA expression in a dose-dependent manner in vitro. These results suggested that JY5 signi cantly inhibit the activation of HSCs.
Notch signaling is a highly conserved pathway evolutionarily, which in uences intercellular signal transduction and cell fate decisions, and regulate growth and development homeostasis of multiple tissues and organs in body, in particular, the progress and development of diseases [22]. Notch signaling pathway mainly consists of 4 Notch receptors (Notch1, Notch2, Notch3, Notch4), 5 Notch ligands (Jagged1, Jagged2, Dll1, Dll3, Dll4) and the transcriptional regulatory elements of downstream signals [23]. Previous studies [24]have shown that Notch play an important role in the progress and development of hepatic brosis, which can interact with other signaling pathways, such as TGF-β, Hedgehog, Hippo, etc. TGF-β1 can promote proliferation and activation of HSCs-T25 cells through regulating Notch signaling pathway [25]. Jagged1 gene was successfully interfered by using rAAV1-Jagged1-shRNA in CCl 4 -indued liver brosis, resulting in alleviating liver brosis [26]. In addition, the knockout of RBP-κB gene, which is considered to be the key transcription factor of Notch pathway, can inhibit the proliferation and activation of HSCs to alleviate CCl 4 -indued liver brosis in mice [27]. It can be seen that the blockade of Notch pathway can effectively inhibit activation of HSCs, which in turn attenuate liver brosis. In our study, the expressions of Jagged1, Jagged2, Notch2, Notch3, Notch4, RBP-κB were signi cantly increased in CCl 4 -and BDL-induced liver brosis in rats and mice. While after intervention with JY5, the expressions of these Notch-related genes and proteins were signi cantly decreased. This is further con rmed by LX-2 cells activation induced by TGF-β1 experiments in vitro. These results suggested that JY5 might exert anti-brotic effect through regulating Notch signaling pathway to inhibit the activation of HSCs.
In this study, we respectively used SORA and DAPT as the positive controls in the experiments. SORA, as a multi-receptor tyrosine kinase inhibitor, can inhibit the proliferation of multiple tumor cells and promote cells apoptosis, which is commonly used for treatment of hepatocellular carcinoma (HCC) clinically [28]. DAPT, as a γ-secretase inhibitor, can block the release of Notch intracellular domain (NICD) into intracellular to inhibit the activation of Notch signaling pathway [24]. In accord with reported studies and previous researches [29,30], our results show that both SORA and DAPT have signi cant anti-hepatic brosis effects. Interestingly, SORA also can signi cantly down-regulate mRNA expressions of Jagged1, Jagged2, Notch2, Notch3, Notch4, RBP-κB and the protein expression of RBP-κB. We speculate that SORA exerts anti-hepatic brosis effect, which might be related to regulation of Notch pathway. But the precise mechanism remains to be further investigated. In addition, the reported studies [19] have shown that DAPT inhibit liver brosis through blocking Notch pathway.
This conclusion is further con rmed in our study. However, it is important to note that γ-secretase inhibitors, as a nonspeci c Notch blocker, have severe side effects in the clinical trial [24]. So the clinical applications of DAPT could be limited to a certain. Compared with SORA and DAPT, JY5 may exist synergistic anti-brosis effect targeted on multiple pathways, not speci cally blocking a particular target or pathway, which confers its relatively higher safety and more effectiveness .
However, the main targets and speci c modalities, by which JY5 regulate Notch pathway, should be to further investigated. In addition, given that JY5, as a component of TCM compounds, has shown the good e cacy in two different liver brosis models, we speculate that JY5 might alleviate hepatic brosis through other mechanisms. In response to these issues, a series of related studies to elaborate the compatibility mechanisms of anti-liver brosis effect of JY5 will be further carried out. Thus, these studies will provide more adequate scienti c basis for the clinical research and application of JY5 formula.

Conclusions
In summary, we obtained a novel anti-hepatic brosis component of TCM compounds, namely JY5, through the uniform design and validation experiments in vivo, and explored its part of mechanisms for the rst time. Our study has shown that JY5 maybe exert anti-hepatic brotic effect through regulating Jagged/Notch/RBP-κB signaling pathway to inhibit the activation of HSCs. JY5 formula may be a potential new therapeutic candidate for liver brosis.

Declarations
Ethics approval and consent to participate

Consent for publication
Not applicable Availability of data and materials All data and materials in the current study are included in this published article and its additional information les.

Declaration of competing interests
The authors declare that they have no competing interests.