ANIT (purity >98%) has been collected from Sigma–Aldrich Co. (St Louis, MO, USA). SRT1720 was purchased from Medscheme Express Co., Ltd. (Princeton, NJ, USA). Serum alanine-transaminase (ALT), alkaline phosphatase (ALP), total bile acid (TBA), aspartate transaminase (AST), complete bilirubin (TBIL), and gamma-glutamyl transferees (γ-GGT) had been analyzed utilizing commercial kits in accordance with the protocols provided by manufacturers (Whitman Biotech, Nanjing, China). Antibodies to BSEP (sc-74500), MRP2 (sc-5770), and HNF1α (sc-135939) were provided by Santa Cruz Biotechnology Inc. (Dallas, TX, USA). Antibody to FXR (bs-12867R) was obtained from Bioss Biotechnology (Bioss, Beijing, China).

Female C57BL/6 mice (8–9 weeks old) had been procured from SIPPR-BK Laboratory Animal Enterprise (Shanghai, China), and they were put in the 12-hour dark/light cycle. All of the mice had been sheltered in the conditions where there was not germ with constant humidity and controlled temperature (24°C ± 2°C). The mice were provided with free liquid and normal food. Previous to the tests, the mice had been acclimated to the laboratory environment for 1 week. Animal Ethics Council of China Pharmaceutical University has given permission to all animal treatment and proceedings. The tests were implemented in accordance with the Declaration of Helsinki. SRT1720 was prepared in in 0.5% CMC-Na suspension. ANIT was dissolved in olive oil. Mice had been divided into four groups casually (n = 6): (1) control group, in which mice were treated with 0.2% carboxymethylcellulose solution (ip) for 5 days; (2) ANIT group, in which the mice took 50 mg/kg dose of ANIT orally, at 48h before sacrifice; (3) ANIT+ SRT1720 (10 mg/kg), in which mice were given SRT1720 (10 mg/kg, ip) for five days; on the 3rd day, 4 h after SRT1720 or excipient medication, mice took ANIT (50 mg/kg) orally for 48h; (4) ANIT+SRT1720 (20 mg/kg), in which mice were treated with SRT1720 (20 mg/kg, ip) for 5 days; on the 3rd day, 4 h after SRT1720 or excipient medication, mice took ANIT (50 mg/kg) orally for 48 h. On the fifth day, 4 h after SRT1720 (10 mg/kg, 20 mg/kg) and vehicle treatment, mice were sacrificed to collect livers and blood. The protocol for the other animal experiment was as below. Mice had been divided into four groups casually (n = 6): (1) control group, in which mice were treated with 0.2% carboxymethylcellulose solution (ip) for 5 days; (2) ANIT group, in which the mice took 50 mg/kg dose of ANIT orally, at 24 h before sacrifice; (3) ANIT+ SRT1720 (10 mg/kg), in which mice were given SRT1720 (10 mg/kg, ip) for 5 days; on the 4th day, 4 h after SRT1720, or excipient medication, mice took ANIT (50 mg/kg) orally for 24 h; (4) ANIT+SRT1720 (20 mg/kg), in which mice were treated with SRT1720 (20 mg/kg, ip) for 5 days; on the 4th day, 4 h after SRT1720 or excipient medication, mice took ANIT (50 mg/kg) orally for 24 h. On the 5th day, 4 h after SRT1720 (10 mg/kg, 20 mg/kg) and vehicle treatment, mice were sacrificed to collect livers and blood.

The blood had been gathered in the tubes with no anticoagulation and centrifuged at room temperature to obtain serum. Livers had been put in the 10% formaldehyde solution.

After sacrificed, blood of mice was centrifuged for serum and 50 mg liver tissue was collected to homogenize in RIPA buffer. ALT, ALP, AST, TBA, TBIL, and γ-GGT in mice serum were measured by corresponding commercial kits in accordance with the producer’s protocols (Whitman Biotech, Nanjing, China). Liver samples were homogenization with RIPA buffer to measure the content of bile acid. The bile acid in supernant after centrifugation had been analyzed utilizing a commercial kit (Whitman Biotech) in accordance with the protocol. The results were normalized with total protein concentrations which were evaluated by BCA protein assay Kit (Beyotime, Shanghai, China). The results obtained were the mean of six different animal livers.

After scarified, small pieces of mice livers were collected and directly put in the 10% formaldehyde solution, then inserted in paraffin and sliced to 6 μm slices. All slices were stained with H&E for necrotic and hepatocyte degeneration variations. Images were captured with the microscope (Olympus IX81). Then pathology assessments were performed by a professional pathologist (Wenxia Bai, Jiangsu Medicine Institute, Nanjing, China).

For animal experiment, 50 mg liver tissue was prepared from each mouse. TRIzol agent (Invitrogen Life Technologies, Carlsbad, USA) had been utilized to isolate entire RNA from liver. Then 2 μg of entire RNA was reversed into cDNA by Prime Script RT reagent (Takara, Osaka, Japan). QPCR had been implemented with SYBR PCR Master Mix (Takara, Osaka, Japan) with specific PCR primers (Table 1), and reports were generated on IQTM5 Optical System Software (Version 2.1, Bio-Rad). β-actin had been normalized as an internal control. Primers set for the HNF1α, FXR, Ntcp, Oatp1b2, Bsep, Mrp2, Mrp3, Mrp4, Cyp7A1, Shp, Nrf2, SOD, GCLc, GCLm, HO-1, and Nqo1 genes are shown in Table 1.

A total protein extraction kit (KeyGEN Biotech) had been used to extract liver protein. The level of the protein was evaluated by the BCA protein assay Kit (Beyotime Biotech, Shanghai, China). Fifty microgram protein was added to the each hole of SDS-PAGE gel (10%) and then removed to PVDF membranes. After being blocked with 5% skimmed milk for 1 h, the membranes were further incubated with special antibodies (FXR, Bsep, Mrp2, HNF1α, and GAPDH) overnight at 4°C. After rinsed three times with TBST, membranes were incubated with second antibodies at room temperature for 1 h. Then Bio-Rad Imaging System (Bio-Rad, Hercules, CA, USA) was used to detect certain bands with ECL reagents (Thermo, Waltham, MA, USA).

Cut the liver tissue into appropriate size, and fixed them on the frozen embedding mold utilizing O.T.C embedding agent. Adjusted the parameters of frozen slicer and started to cut liver tissue. The thickness of the slice is about 6 μM. Then placed slices on APES treated slides. 30 min later, the slices had been put inside formaldehyde solution (4%) for 30 min and were penetrated with Triton X-100 (0.1%) for 15 mins. Then, slices were blocking with 5% BSA for 1 h to block non-specific binding and incubated with different antibodies, placed in humidified atmosphere during the night at the temperature of 4°C. The antibodies adopted in the research were Bsep and Mrp2. The slices were then wished with phosphate-buffered saline (PBS) for three times and incubated with secondary antibodies for 1 h. Finally, DAPI had been utilized for nuclear staining. After that, images were obtained from Olympus IX81 motorized inverted fluorescence microscope and analyzed by Image-Pro plus 10.0 software.

After anesthetized, C57BL/6 mouse primary hepatocytes had been isolated by a two-step collagenase digestion method (Meng et al., 2015a). The isolated hepatocytes were collected and resuspended with William’s E medium (Invitrogen, Carlsbad, CA, USA) with penethamate (100 U/mL), streptomycin (100 U/mL), dexamethasone (0.1 mM), and thyroxine (1 mM). Then, hepatocytes were seeded into collagen type I (Sigma, St. Louis, MO, USA) pre-coated cell culture dishes. After 24 h incubation, hepatocytes were coated with collagen I solution and then cultivated in complete medium to achieve sandwich culture model (Li et al., 2016).

Mouse primary hepatocytes had been transiently transfected with siRNA which targeting mouse HNF1α (sc-35568) or a negative control siRNA utilizing Lipofectamine^TM 3000 (Invitrogen, Carlsbad, CA, USA). Seven hours later, medium had been changed with fresh medium. After 48 h, DMSO or SRT1720 (10 μM) had been added to the cell culture medium for 12 h. Then, primary hepatocytes cells were collected for further QPCR experiment.

Mouse livers had been homogeneous with ice-cold 0.9% normal saline and the slurry was centrifuged at 4°C (3500 rpm/min), then supernatant had been utilized for hepatic enzymes assays. The levels of IL-6, TNF-α, and myeloperoxidase (MPO) activity in supernatant had been measured to evaluate inflammatory reactions and neutrophil activation in livers. The IL-6 and TNF-α levels were measured by ELISA kits (UNIV-bio, Shanghai, China). MPO activity was measured by a commercial assay kit (Jian cheng Bio-engineering Institution, Nanjing, China). The activities of entire antioxidant capacity (T-AOC), glutathione (GSH), malondialdehyde (MDA), and superoxide dismutase (SOD) had been determined to access the oxidant state in the hepatic constitution. Liver T-AOC, GSH, SOD, and MDA activities had been measured by their respective assay kit (Jian cheng Bioengineering Institute, Nanjing, China).

Statistics are processed with a software called Graphpad Prism (5th version) (Graph-Pad, La Jolla, CA, USA) and are shown as the mean ± SD. Through utilizing one-way variance analysis, the significance of the statistic from multiple groups is determined. The data is considered statistically significant when P < 0.05.

To determine whether SRT1720 could reverse mice cholestatic liver injury induced by ANIT, C57BL/6 mice were treated with ANIT, 0.2% CMC-Na, or SRT1720 (10 mg/kg or 20 mg/kg) for 5 days. The SRT1720 chemical structure was showed in Figure 1. Biochemical indicators in mice intraperitoneal administered vehicle, with 10 or 20 mg/kg of SRT1720, were evaluated 24 or 48 h later after taking ANIT. Taking ANIT led to the damage in the liver which was demonstrated by the remarkable increase in the serum AST and ALT expressions compared with control group. Upon SRT1720 treatment, plasma AST and ALT levels remarkably lessened by 70% compared with ANIT-treated group (Figures 2A,B). ALP is produced by osteoblasts, absorbed by the liver, and then secreted through the bile acid. The level of plasma ALP will increase significantly during cholestasis. The γ-GGT is mainly distributed in the liver and intrahepatic bile duct epithelium, when cholestasis, this enzyme may countercurrent into the blood. ANIT administration remarkably increased the plasma levels of ALP and γ-GGT and SRT1720 remarkably reduced them compared to ANIT-treated group (Figures 2C–D). Furthermore, SRT1720 dosage dependently reversed ANIT-induced increases in, TBA total bilirubin and hepatic TBA (Figures 2E–G). ANIT reduced bile acid output at 24 h after administration, and SRT1720 remarkably attenuated this variation (Figure 2H). Because all of the biochemical indicators of hepatobiliary injury were observed to be higher at 48 h than 24 h after ANIT administration, the 48 h time point had been selected for the subsequent studies. The H&E staining results demonstrated that SRT1720 treatment remarkably attenuated ANIT-induced degenerative changes and necrotic foci in the livers (Figure 3), which further demonstrated protective effects of SRT1720 on liver damage which induced by ANIT. In conclusion, all these results above demonstrated that SRT1720 could provide remarkable protective effects on cholestasis and hepatotoxicity induced by ANIT.

To explain the potential mechanisms of SRT1720 hepatoprotective effects, the gene expressions of bile acid transporters had been evaluated by real-time PCR. First, we determined the expressions of HNF1α and FXR, both of which are important nuclear receptors to regulate systemic bile acid metabolism. The gene expressions of HNF1α and FXR were reduced by 65 and 56%, respectively, following ANIT administration. SRT1720 treatment increased the mRNA levels of HNF1α and FXR (Figure 4A). Just like illustrated in Figure 4B, ANIT caused a decrease in the gene expressions of Ntcp and Oatp1b2, and SRT1720 treatment enhanced both of them. Then, we examined the gene expressions of Bsep and Mrp2. Figure 4C showed that ANIT reduced the expressions of Bsep and Mrp2 remarkably and SRT1720 treatment increased their expressions. Figure 4D illustrated that the mRNA level of Mrp3 showed a 1.5-fold increase, while SRT1720 demonstrated no effect at all. The mRNA expression of Mrp4 was reduced by ANIT, and SRT1720 increased its gene expression. Together, the above results suggested that ANIT remarkably decreased the expressions of HNF1α, FXR, Ntcp, Oatp1b2, Bsep, Mrp2, and Mrp4, which aligns with the findings of a previous research (Wang T. et al., 2014). SRT1720 treatment restored the expressions of HNF1α, FXR, Ntcp, Oatp1b2, Bsep, Mrp2, and Mrp4, which caused an increase of bile acid efflux and influx in the liver. However, SRT1720 did not have remarkable effect on Mrp3 expression.

In addition to the transporters mentioned before, bile acid synthetic enzymes are also included in bile acid homeostasis. In order to further analyze the hepatoprotective effects of SRT1720, we examined the mRNA expression of Cyp7a1, the enzyme in bile acid synthesis that limits the reaction rate. As illustrated in Figure 4E, the mRNA level of Cyp7a1 was decreased in ANIT group compared with control group. And SRT1720 therapy further decreased mRNA level of Cyp7a1. We then examined the mRNA level of Shp which is its upstream gene. Shp activation has been demonstrated to restrain Cyp7a1 transcriptionally (Kong et al., 2012b). SRT1720 treatment increased the mRNA level of Shp (Figure 4E). These findings suggested that the inhibition of Cyp7a1 by SRT1720 could be mediated by FXR-Shp signaling pathway.

Phase I enzymes like Cyp2b10 and Cyp3a11, and phase II enzymes like UDP-glucuronosyltransferase 1a1 (Ugt1a1) and sulfotransferase 2a1 (Sult2a1) are bile acid detoxifying enzymes in the liver. As shown in Figure 5A, ANIT increased the mRNA levels of Cyp3a11 and Cyp2b10, and SRT1720 increased Cyp2b10 mRNA level further. ANIT decreased mRNA levels of Sult2a1 and Ugt1a1, and SRT1720 caused an increase in the mRNA expression of Sult2a1 (Figure 5B). These results indicated that mRNA levels of bile acid detoxifying enzymes could be enhanced by SRT1720 compared with ANIT group.

To check Real-time PCR results, western blotting method was used to evaluate the protein expressions of Bsep, Mrp2, and FXR. Figures 6A,B indicated that ANIT caused decreases of protein expressions of Bsep, Mrp2, and FXR, and a 20 mg/kg SRT1720 treatment caused an increase in these levels. Figures 6C,D showed that fluorescence intensities of Bsep and Mrp2 in cytomembranes were remarkably decreased by ANIT, which suggested that bile acid transporters were impaired by ANIT. SRT1720 increased fluorescence intensities of Bsep and Mrp2 in cytomembranes remarkable.

To determine whether SRT1720 could regulate the expressions of FXR and its downstream genes in vitro, FXR, Bsep and Mrp2 mRNA levels were quantified in mouse primary hepatocytes which treated with different concentrations of SRT1720. SRT1720 was incubated with ANIT for 24 h, as shown in Figures 7A,B, in control cells, Bsep staining was localized in thin continuous lines outlining the cells, and after treatment with ANIT for 24 h, the fluorescence intensity of BSEP was decreased, and SRT1720 increased Bsep fluorescence intensity compared with the cells treated with ANIT, which suggested that the inhibitory effects of ANIT could be improved by SRT1720. As shown in Figure 7C, SRT1720 was incubated with ANIT for 12 h, ANIT inhibited FXR, Bsep and Mrp2 expressions at the mRNA level, and SRT1720 increased all of them. To determine the potential mechanism of SRT1720 mediated activation of FXR, HNF1α gene silencing experiment was performed in vitro. The protein expression of HNF1α had been evaluated by western blot analysis after transfected with HNF1α siRNA and a control siRNA (Figure 7D). Figure 7E revealed that SRT1720 increased the mRNA level of FXR in normal mouse primary hepatocytes and it had no effect on it in HNF1α deficiency mouse primary hepatocytes (Figure 7E). These results demonstrated that SRT1720 mediated activation of FXR was largely through HNF1α.

Figure 8A revealed that the activities of hepatic T-AOC, SOD, and GSH in the livers were significantly inhibited by ANIT, which suggested that ANIT could cause antioxidant defense systems disruption in the liver. Comparatively, activities of T-AOC, SOD, and GSH could be restored by SRT1720 at 10 mg/kg and 20 mg/kg, respectively (Figure 8A). Forty-eight hours after ANIT-treatment, hepatic MDA level increased notably in the group with ANIT treatment compared with the control group. And both doses of SRT1720 significantly decreased MDA concentrations (Figure 8A). Taken together, SRT1720 exerted an antioxidant effect on ANIT-induced hepatotoxicity and cholestasis.

Forty-eight hours later, the hepatic MPO activity in the control group was lower than the group with ANIT treatment. And SRT1720 dose-dependently inhibited MPO activity that was induced by ANIT (Figure 8B). This was further confirmed by the Immunohistochemical staining of MPO (Supplementary Figure S1A). Multiple reports demonstrated that inflammatory reaction was related to cholestasis and SRT1720 could alleviate inflammatory reaction. ANIT increased the levels of TNF-α and IL-6 at 48 h and SRT1720 caused a decrease in the concentrations of TNF-α and IL-6. The results revealed that SRT1720 has an anti-inflammatory effect in ANIT-treated mice (Figure 8B).

Figure 9 revealed that mRNA levels of Nrf2 and its target genes were increased by SRT1720 in ANIT-treated group dose-dependently. The mRNA levels Nrf2, GCLc, GCLm, and SOD were significantly decreased by ANIT at 48 h and SRT1720 up-regulated the expressions of Nrf2 and its target genes dose-dependently (Figures 9A–D). ANIT did not inhibit expressions of HO-1 and Nqo1, and SRT1720 increased their mRNA levels dose-dependently (Figures 9E,F). Taken together, these results clearly demonstrated that SRT1720 exerted an anti-oxidation effect through activating the Nfr2/ARE pathway in ANIT-treated mice (Figure 9).