Herbal medicine Yinchenhaotang protects against α-naphthylisothiocyanate-induced cholestasis in rats

Cholestasis is a clinical disorder defined as an impairment of bile flow, and that leads to toxic bile acid (BA) accumulation in hepatocytes. Here, we investigated the hepatoprotective effect of Yinchenhaotang (YCHT), a well-known formulae for the treatment of jaundice and liver disorders, against the cholestasis using the α-naphthylisothiocyanate (ANIT)-induced cholestasis in male Wistar rats. ANIT feeding induced significant cholestasis with substantially increased intrahepatic retention of hydrophobic BAs. The dynamic changes of serum and liver BAs indicated that YCHT was able to attenuate ANIT-induced BA perturbation, which is consistent with the histopathological findings that YCHT significantly decreased the liver damage. YCHT treatment substantially reduced serum alanine aminotransferase (ALT), alkaline phosphatase (AST), total bilirubin (TBIL) and direct bilirubin (DBIL) with minimal bile duct damage in the ANIT treated rats. Elevated mRNA expression of liver IL-6, IL-17A, IL-17F, TGF-β1, α-SMA, TGR5, NTCP, OATP1a1, and ileum ASBT and decreased liver IL-10, FXR, CAR, VDR, BSEP, MRP2, MRP3, MRP4 was also observed in ANIT-induced cholestasis but were attenuated or normalized by YCHT. Our results demonstrated that the BA profiles were significantly altered with ANIT intervention and YCHT possesses the hepatoprotective potential against cholestatic liver injury induced by hepatotoxin such as ANIT.


Results
Active composition in the YCHT extract. There are four main active compounds, gallic acid, chlorogenic acid, geniposide and crocin in the YCHT extraction and their levels were measured using HPLC ( Supplementary  Fig. S1). The contents of gallic acid are 4.54 mg/g, chlorogenic acid are 42.35 mg/g, geniposide are 77.59 mg/g and crocin are 1.85 mg/g. YCHT has been safely used with few adverse events in clinic 21 . It has been reported that geniposide in Gardenia jasminoides Ellis showed liver toxicity when rats were given geniposide at a dose of 228 mg/kg body weight 22 . However, in our study, the concentration of geniposide administrated was 120 mg/kg, which is much lower than the dosage that will induce liver toxicity. Table 1, the body weight of rats in ANIT group was significantly decreased compared to the normal, while in YCHT treated group, the body weight was slightly increased compared to the rats in ANIT group at week 8.

General information of the ANIT-induced cholestasis. As shown in
The serum levels of ALT, AST, alkaline phosphatase (ALP), TBIL, DBIL, total cholesterol (TC) and liver hydroxyproline (HYP) were markedly increased at week 8 and 12 due to ANIT intervention (p < 0.05, Table 1) but subsequently decreased after YCHT treatment. The levels of triglyceride (TG) were significantly decreased in rats at week 8 and 12 due to ANIT intervention (p < 0.05, Table 1) and subsequently increased after YCHT treatment. Serum levels of ALT, AST, albumin (ALB), TBIL, DBIL, and blood urea nitrogen (BUN), liver γ-glutamyl transpeptidase (γ-GT) activity and HYP content in liver tissue were higher at week 12 than those at week 8, suggesting the gradually aggravated cholestatic liver injury.
Histopathological evaluation. H&E staining of liver tissue showed that all rats in normal group appeared normal. At week 8, obvious biliary epithelial cell proliferation were occurred in the portal area, formatting a large number of irregularly arranged small bile duct and inflammatory cell infiltration. With the prolonged action time of ANIT, the liver presented large focal necrosis, island structure of various sizes, obvious biliary epithelial cell proliferation, a large number of irregularly arranged small bile duct and inflammatory cell infiltration with significantly reduced hepatocytes ( Fig. 2A). Sirius red staining demonstrated that the original lobule was altered with a large number of collagen deposition around proliferated small bile ducts ( Fig. 2A). Compared with the model group, YCHT treatment ameliorates the ANIT induced biliary epithelial cell proliferation and the degree of liver fibrosis. Liver samples were also harvested for immunofluorescence analysis of cytokeratin-19 (CK19) and Collagen I ( Fig. 2B and C). Cytokeratins are the major structural proteins of intermediate filaments in epithelial cells; CK19 is an antigenic marker for bile duct epithelial cells and bile ductules 23 . CK19-positive cells (red fluorescence) were expressed only in the portal area in normal liver tissue. With the increased time of ANIT intervention, CK19-positive cells in the portal area increased gradually, suggesting that ANIT intervention significantly increased bile duct epithelial cell proliferation. Collagen I was found to be widely distributed in the liver tissue of ANIT model rats compared with the normal rats. The expression of Collagen I increased gradually with the prolonged time of ANIT intervention. YCHT intervention significantly reduced the expression of CK19 and Collagen I in hepatic tissue of model rats ( Fig. 2D and E).
In the livers of rats, taurine-conjugated BAs were predominated; however, ANIT intervention resulted in a significant increase in both glycine-conjugated and taurine-conjugated BAs (Supplementary Table S2 and Fig. 4C). Additionally, the levels of TCDCA, GCDCA, Tα-MCA, Tβ-MCA, and Tλ-MCA were increased in ANIT-treated rats compared to the controls but decreased after YCHT treatment (Fig. 4D). Glycine-and taurine-conjugated BAs were significantly elevated in the liver following ANIT intervention and decreased after YCHT treatment; similar to what was observed in serum samples (Fig. 4).

YCHT regulated gene expression involved in bile acid metabolism.
To further clarify the effective mechanisms of YCHT, the relative expressions of key genes involved in bile acid metabolism were measured. As shown in Fig. 5, the levels of liver farnesoid X receptor (FXR), constitutive androstane receptor (CAR), and vitamin D receptor (VDR) mRNAs were decreased (P < 0.01), and G protein-coupled bile acid receptor (TGR5) were increased (P < 0.01) in ANIT group, and all of them restored to normal levels by YCHT treatment. Furthermore, the levels of mRNAs related BA absorption such as organic anion-transporting polypeptides (OATP1) and Na + -taurocholate cotransport peptide (NTCP) were significantly up-regulated in ANIT group (P < 0.01), and related excretion such as multidrug-resistance associated protein (MRP3), MRP2, MRP4 and bile salt export pump (BSEP) were significantly down-regulated (P < 0.01). ANIT also significantly increased the expression of ileum apical sodium-dependent bile acid transporters (ASBT). The mRNA expression of FXR, BSEP and ASBT were also validated by Western blot. The protein levels of FXR and BSEP were significantly decreased while ileum ASBT levels were significantly increased in ANIT model rats, but they were all normalized after YCHT intervention ( Fig. 6 and Supplementary Fig. S2).

Correlation between BAs and bilirubin. Pearson correlation analysis showed no correlation between
serum BAs and bilirubin in model rats. A low correlation exists between serum BAs and bilirubin in model rats at week 8, and at week 12, a significant correlation was observed between serum BAs and bilirubin. The serum taurine-conjugated BAs was significantly positively correlated with TBIL (r = 0.6408, P = 0.0459), among which, TCA was the most significant (r = 0.6559, P = 0.0284). Interestingly, the ratio of TCA to CA was significantly positively correlated with TBIL (r = 0.8386, P = 0.0024) and DBIL (r = 0.8248, P = 0.0033) (Fig. 7).

Discussion
ANIT is a widely used chemical to mimic human intrahepatic cholestasis. ANIT is conjugated with glutathione in hepatocytes and transported through the canalicular efflux transporter (MRP2) into bile, where ANIT dissociates from glutathione. The released ANIT selectively damages bile duct epithelial cells and causes cholangitis and subsequent intrahepatic cholestasis 25 . In response to ANIT, bile duct epithelial cells release pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and monocyte chemoattractant protein-1 (MCP-1). Increases in inflammatory cytokines result in migration and infiltration of neutrophils, leading to lipid peroxidation and bile duct epithelial cell damage [26][27][28] . Early studies into the mechanisms of cholestatic liver injury strongly implicated BA-induced apoptosis as the major cause of hepatocellular injury. Recent work has focused both on the role of BAs in cell signaling as well as the role of sterile inflammation in the pathophysiology. Advances in modern analytical methodology have allowed for more accurate measuring of BA concentrations in serum and liver to very low levels of detection. This study showed both how different BAs and different BA concentrations can affect hepatocytes during cholestasis, and additionally provide insight into how these data support recent hypotheses that cholestatic liver injury may not occur through direct BA-induced apoptosis, but may involve largely inflammatory cell-mediated liver cell necrosis.
In this study, we examined the BA profiles of the liver and serum in normal rats and determined how those profiles were impacted by ANIT intervention. Serum ALT and AST, the widely used markers in evaluating the degree of liver injury 29 , and bilirubin including DBIL and TBIL, were markedly increased due to ANIT intervention and then significantly decreased with YCHT treatment. These results are in line with earlier reports showing that YCHT reduces increases in ALT, AST, DBIL and TBIL levels in ANIT-induced cholestasis model 17,30 . It has also been reported that YHCT can alleviate obstructive biliary cirrhosis induced by bile duct ligation in rats via inhibiting biliary epithelial cell proliferation and activation 31 . Our previous study also demonstrated that YHCT administration alleviated liver fibrosis via inhibiting hepatic stellate cell (HSC) activation in dimethylnitrosamine-treated rats 32,33 . Studies by Takahashi et al. showed that YHCT significantly mitigated the necroinflammation in the high fat diet induced NASH mice 34 .
Chronic ANIT intervention resulted in a global alteration of BA profiles in the liver and blood. Elevated hepatic tissue BA concentrations induce hepatocellular injury by causing hepatocyte apoptosis [35][36][37] . Retention of bile constituents within the hepatocyte during cholestasis is associated with hepatocyte apoptosis 38 . Although the mechanisms of cholestasis associated with hepatocyte apoptosis are likely complex and multifactorial, hydrophobic BAs are especially hepatotoxic and accumulated in the liver in cholestatic disorders 39 . The failure to secrete BAs into bile results in liver injury, cirrhosis, and death from liver failure 40 . The BA profiles for liver tissues Figure 3. mRNA expression of IL-6, IL-10, TGF-β1, IL-17A, IL-17F, and α-SMA in the liver. Normal, healthy rats; 8w Model, ANIT treatment for 8 weeks; 12w Model, ANIT treatment for 12 weeks; Yinchenhaotangtreated, ANIT-induced cholestasis rats treated with Yinchenhaotang from 9 th week to 12 th week. All data are presented as mean ± SEM. (n = 8-12; *P < 0.05; **P < 0.01).
in normal and the ANIT-treated rats were dominated by taurine-and glycine-conjugated BAs, with unconjugated BAs present in small amounts. Generally, ANIT intervention led to a substantial increase in glycine-and taurine-conjugated BAs in liver tissue examined, even during short-term ANIT intervention (8 weeks). In contrast, unconjugated BAs were predominant in the serum and all BAs were increased under the same conditions, and the alterations in the BA profile in the blood were normalized after YCHT treatment. Taurine-conjugated BAs are more hydrophilic and less toxic than either unconjugated or glycine-conjugated BAs. Liver BAs in normal rats were predominantly taurine conjugated, consistent with the findings of Dupont et al. 41 , who saw that one-half of the liver BAs in miniature swine were conjugated with taurine. Since YCHT intervention can have a profound impact on the gut microbiota 42,43 , it is possible that the reduced abundance of taurine-conjugated BAs in YCHT-treated rats is at least partially due to an YCHT-induced disturbance of the gut microbiota.
FXR plays a central role in the regulation of BA synthesis, excretion, and transport 44,45 . Activation of FXR inhibits BA synthesis from cholesterol and activates secretion into bile canaliculi through BSEP, therefore protecting against the toxic accumulation of BAs in hepatocytes 46 . We observed that the FXR and BSEP expression in rat liver was significantly decreased in the ANIT model, suggesting a mechanism of down-regulating hepatic efflux transporters, thus leading to increased BA accumulation in hepatocytes and BA-induced liver injury. The higher levels of BAs in ANIT-induced model rats were also probably due to bile duct obstruction in liver impairments and impaired BAs enterohepatic circulation, with reduced transformation of BA to intestine and the feedback from intestine 47 . Although the exact mechanisms underlying the increase of FXR expression in YCHT-treated rats are unclear, we propose that YCHT-induced alterations in the gut microbiota may play a critical role in this process. In the terminal ileum and colon, BAs are reabsorbed by the enterocytes via ASBT and released into the portal vein for return to the liver. The significantly increased ASBT also led to increased BA accumulation in hepatocytes 48 . BAs can be reabsorbed into hepatocytes through both NTCP and OATP 49 . The levels of both NTCP and OATP1a1 mRNAs, were found increased in ANIT-treated rats, which may lead to increased BA levels in hepatocytes.
Gene expression analysis indicated that ANIT intervention led to alteration of genes involved in inflammation. Inflammation is a critical component of cholestasis. Abnormal expression of IL-6, IL-10, IL-17A, IL-17F, TGF-β1 and α-SMA are closely associated with the cholestasis. IL-6 is an important mediator of inflammation in vivo. The hepatocytes, Kupffer cells, and HSC secrete large quantities of IL-6 in the liver disease state. High concentrations of IL-6 can stimulate hepatocyte apoptosis and HSC proliferation, activate Kupffer cells to secrete more pro-inflammatory cytokines, and promote extracellular matrix synthesis, leading to aggravated liver fibrosis. IL-6 can directly activate signal transducers and activators of transcription (STAT3), increase the expression of RORγt and induce naive T cells to CD4 + Th17 cells differentiation 50 . Also, mice CD4 + CD25-Foxp3-T cells and CD4 + CD25 + Foxp3 + Treg cells can be transformed into Th17 cells in the presence of IL-6 without the addition of exogenous TGF-β1, resulting in reduced number of Treg and increased number of Th17 51 . In the progression of liver fibrosis, activated Kupffer and HSC cells secrete large amounts of IL-6 and TGF-β1, promote the transformation from Treg to Th17, resulting in reduced secretion of IL-10 and increased secretion of IL-17. IL-10 is an important immunosuppresive cytokine and can effectively prevent inflammation and excessive immune responses. Various types of cells can produce IL-10 52 . IL-10 can inhibit Kupffer cells to produce inflammatory cytokine, tumor necrosis factor-α (TNF-α), and reduce the activation of HSC 53 . Clinical studies have demonstrated that long-term IL-10 therapy can reduce inflammation and fibrosis score of patients with hepatitis C virus 54 . In contrast, IL-10 knockout mice showed a more significant liver fibrosis than wild-type mice 55 . Previous reports indicated that serum IL-6 was highly expressed in cholestasis, which will be attenuated when the expression of IL-6 was reduced 56 . Serum IL 10-level was negatively correlated with TBIL concentrations in chronic liver injury 57 . It was reported that serum IL-17 was highly expressed in cholestatic patients and its level decreased after immunosuppressive therapy with improved cholestasis 58 . As a result, the up-regulated IL-10 expression and It was reported that serum BA and bilirubin concentrations changed in parallel throughout cholestatic viral hepatitis, chronic active hepatitis and alcoholic hepatitis 59 , which is accordance with our findings that the serum BAs and bilirubin were significantly positively correlated.
In conclusion, ANIT intervention had a profound effect on the BA profiles in the liver and serum. The ANIT-modulated BA profiles were characterized by a marked increase in unconjugated and taurine-and glycine-conjugated BAs. Gene expression analysis revealed several inflammation related cytokines were significantly altered including increased IL-6, IL-17A, IL-17F, TGF-β1 and α-SMA and decreased IL-10 expression. YCHT can attenuate the metabolic perturbation in ANIT-induced model rats, which was supported by liver tissue histology, and liver inflammation cytokine data that YCHT significantly reduced liver inflammation and liver injury.   60 . Artemisia annua L. (2250 g) was first boiled in 600 L of water for 1 h, and then Gardenia jasminoides Ellis (500 g) and Rheum Palmatum L. (750 g) were added in together. Finally, the filtered solutions were concentrated into the aqueous extracts containing 0.48 g/mL raw herbs. Then the YCHT extract was dissolved in methanol and filtered through a membrane syringe filter (0.2 µm) (Millipore Co., Bedford, MA, USA) and subject to HPLC analysis. An Agilent 1200 HPLC system (Santa Clara, CA, USA) coupled with a binary pump, an automatic injector, and a diode array detector were used in this study. The separation was carried out on a Phenomenex Luna C18 (250 × 4.6 mm I.D., 5 µm) (Phenomenex, Torrance, CA, USA). The column was eluted with mobile phase (A) 0.1% acetic acid in water and (B) methanol using a linear gradient of 5-62% B over 0-75 min, 62-73% B over 75-80 min, 73-100% B over 80-85 min, held at 100% B for 10 min, then equilibration at 5% B. The flow rate was 1 mL/min and the injection volume was 20 µL. All the samples were kept at room temperature during the analysis. The detection wavelength was set at 260 nm. Agilent ChemStation software was used for peak identification and integration and the content of the constituents was calculated using standard curves of gallic acid, chlorogenic acid, geniposide, and crocin.

Methods
Animal Models and YCHT Treatment. Male Wistar rats (160-180 g) were obtained from Vital River Laboratories (Beijing, China). The animal study was conducted in accordance with Chinese national legislation and local guidelines, and performed at Shanghai University of Traditional Chinese Medicine (Shanghai, China). The experimental protocol was approved by the Experiment Animal Center of Shanghai University of Traditional Chinese Medicine, Shanghai, China. The animals were housed, five per cage with controlled temperature (23-24 °C) and humidity (60 ± 10%) with 12 h/12 h light-dark cycles. All experimental rats were fed with standard rat chow and received water ad libitum throughout the experiment except those rats will be fasted for 2 h after ANIT intervention. After one week of acclimatization, the rats were randomly divided into 3 groups: the normal control group (n = 8), in which a daily dose of 10 mL/kg of body weight distilled water was given via oral gavage once a day for 4 weeks from week 9 to week 12; an ANIT treated group (model group, n = 22), in which a daily dose of 60 mg/kg of body weight ANIT (dissolved in olive oil) was given in a 0.5 mL aliquot vial oral gavage every other day for 12 weeks; YCHT treated group (n = 10), oral garage ANIT as processed in ANIT group and received a daily dose of 695 mg/kg of body weight YCHT was given in a 10 mL/kg of body weight via oral gavage once a day for 4 consecutive weeks from week 9 to week 12. The dosage of YCHT used in rat is determined according to an established formula for human−rat drug conversion 61 . At the end of 8 th week, 10 rats in ANIT treated group were sacrificed to judge liver injury. At the end of week 12, all animals (n = 30) were sacrificed. Serum samples were collected and stored at -80 °C pending biochemical and BAs analysis and tissue samples were collected for HYP and BAs measurement and histopathological examination.
Biochemical analysis. Blood samples were drawn from the inferior vena cava. Serum was obtained by centrifuging the blood at 8,000 x g for 15 minutes at 4 °C. Serum ALT, AST and ALP activity, and the levels of ALB, TBIL, DBIL, TG, TC, Cr, and BUN were measured using TBA-40FR Fully Automatic Biochemical Analyzer (Toshiba, Japan). Liver γ-GT activity was detected using the colorimetry. Liver HYP content was assayed according to the previous reported method 62 .
Histopathological evaluation, immunofluorescence analysis and qRT-PCR analysis. Liver tissues were fixed in 10% formalin, and processed for paraffin embedding. Sections (5 µm) of each sample were cut and Haematoxylin Eosin (H&E) and sirius red stained for histopathological analysis.
For immunofluorescence analysis, liver tissue slides were stained with antibodies for CK19 (10712-1 Table 2. Target gene expression was normalized to β-acitin levels and presented as fold changes relative to control values (which were set at 1). For quantification of gene expression, the threshold cycle (Ct) values of all target genes were normalized to the expression of the endogenous reference (β-actin). ΔCt = Ct (target gene) -Ct (β-actin). The primers used for amplifications were listed in Table 2.
Western blot. Liver and ileum tissue lysates were obtained, and protein concentrations were determined using a BCA protein assay kit (23225#, Thermo, California, CA). Western blot analysis was processed as previously described 63 . Equal amounts of protein samples were separated by SDS-PAGE and transferred to PVDF membranes which were blocked at room temperature for 1 h in blocking solution. The membranes were then incubated with rabbit polyclonal antibodies against BSEP, FXR, and ASBT overnight at 4 °C, followed by a 1 h incubation at room temperature with HRP-conjugated goat anti-rabbit secondary antibody. The obtained bands were then scanned and analyzed with the Odyssey quantitative western blot near-infrared system and band density was analyzed using Image J software. GAPDH and β-actin served as an internal control and experiments were performed in triplicate. All antibodies were purchased from Abcam. BA quantitation in serum and liver. Serum and liver BAs were measured according to the previously reported method with some modifications 64,65 .
For serum, 50 µL serum samples were mixed with 150 μL of methanol containing internal standards (IS, 0.10 µM of CA-D4, DCA-D4, GCA-D4, and LCA-D4), and then incubated for 10 min at room temperature. After centrifugation at 20,000 g for 10 min, the supernatant was transferred to a clean tube, vacuum dried and reconstituted with 40 μL acetonitrile (with 0.1% formic acid) and 40 μL water (with 0.1% formic acid). After centrifugation, the supernatant was used for UPLC-TQMS measurement.
For liver samples, liver tissue was precisely weighed (~50 mg) and homogenized in 75 μl ice-cold 50% methanol, and then centrifuged at 20,000 g for 10 min. The supernatant was transferred to a clean tube. Then 80 μL ice-cold methanol: chloroform (3:1, v/v) was added to the remaining pellet and homogenized again. After centrifugation, the two supernatants were combined, spiked with 10 μL IS and vacuum dried. The extracts were reconstituted with 40 μL acetonitrile (with 0.1% formic acid) and 40 μL water (with 0.1% formic acid). After centrifugation, 5 μL supernatants were injected for UPLC-TQMS measurement.
The detection was performed with a Waters ACQUITY Ultra performance liquid chromatography (BEH C18 1.7 μm 2.1 × 100 mm column) coupled with Waters Xevo TQ-S triple quadrupole mass spectrometry. Data acquisition and BA quantification were performed using the TargetLynx applications manager version 4.1 (Waters Corp., Milford, MA). Data analysis. One-way analysis of variance and the least significant difference (LSD) test were used to investigate differences between the groups, and P < 0.05 was considered statistically significant. Pearson correlation analysis between BAs and bilirubin were performed with SPSS (IBM SPSS Statistics 22, USA).