Sodium cholate ameliorates nonalcoholic steatohepatitis by activation of FXR signaling

Non-alcoholic steatohepatitis (NASH) has become a major cause of liver transplantation and liver-associated death. The gut-liver axis is a potential therapy for NASH. Sodium cholate (SC) is a choleretic drug whose main component is bile acids and has anti-inflammatory, antifibrotic, and hepatoprotective effects. This study aimed to investigate whether SC exerts anti-NASH effects by the gut-liver axis. Mice were fed with an high-fat and high-cholesterol (HFHC) diet for 20 weeks to induce NASH. Mice were daily intragastric administrated with SC since the 11th week after initiation of HFHC feeding. The toxic effects of SC on normal hepatocytes were determined by CCK8 assay. The lipid accumulation in hepatocytes was virtualized by Oil Red O staining. The mRNA levels of genes were determined by real-time quantitative PCR assay. SC alleviated hepatic injury, abnormal cholesterol synthesis, and hepatic steatosis and improved serum lipid profile in NASH mice. In addition, SC decreased HFHC–induced hepatic inflammatory cell infiltration and collagen deposition. The target protein-protein interaction network was established through Cytoscape software, and NR1H4 [farnesoid x receptor (FXR)] was identified as a potential target gene for SC treatment in NASH mice. SC-activated hepatic FXR and inhibited CYP7A1 expression to reduce the levels of bile acid. In addition, high-dose SC attenuated the abnormal expression of cancer markers in NASH mouse liver. Finally, SC significantly increased the expression of FXR and FGF15 in NASH mouse intestine. Taken together, SC ameliorates steatosis, inflammation, and fibrosis in NASH mice by activating hepatic and intestinal FXR signaling so as to suppress the levels of bile acid in NASH mouse liver and intestine.


INTRODUCTION
Nonalcoholic fatty liver disease (NAFLD), a prevailing chronic liver disease, is often related to metabolic diseases including overweight, diabetes, and hyperlipidemia. [1] It mainly encompasses the non-progressive nonalcoholic fatty liver (NAFL) and the progressive NASH. NAFL connotes the presence of 5% hepatic steatosis without hepatocyte injury; however, NASH means NAFL is complicated by hepatocellular injury, including with or without fibrosis. [2] However, to date, no drug has been officially approved for the treatment of NASH. [3] Therefore, there is an urgent need to discover effective drugs for the treatment of NASH.
Sodium cholate (SC), a cholagogue, is a mixture of sodium taurocholate and sodium glycocholate extracted from bovine bile. [4] Owing to its multiple pharmacological activities, including anti-inflammatory, antifibrotic, and hepatoprotective effects, SC has been widely used to treat biliary insufficiency, cholecystitis, and cardiovascular disease. [4] Recent studies have demonstrated that sodium taurocholate reduces serum total cholesterol, low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) levels in NAFLD rats. [5] Furthermore, mixing sodium glycocholate with sodium taurocholate attenuates the absorption of bile acids from the portal vein into the liver and blocks their absorption in the enterohepatic circulation, resulting in a decrease in bile acid synthesis. [6] However, it is not clear whether SC prevents NASH and its underlying mechanism.
The aim of this study was to determine whether SC exerts anti-NASH effects through the gut-liver axis. We used a murine model established by high-fat and highcholesterol (HFHC) diet feeding, which is characterized by lipid accumulation, inflammation, and fibrosis, and to corroborate the direct effects of SC attenuates hepatocyte injury and lipid accumulation in a palmitic acid (PA)-induced cellular model. We found that SC attenuates bile acid metabolism disorder in NASH mice by activating FXR signaling in the liver and intestine, contributing to the amelioration of steatosis, inflammation, and fibrosis in NASH mice.

Preparation of SC
The SC was obtained from the Shanghai Zhonghua Pharmaceutical Co. (Shanghai, China), which belongs to the State Drug Certificate H19999122. SC was dissolved in CMC-Na solution to make a turbid solution of SC and frozen at −20°C. It was defrosted and injected i.p. at the doses indicated (90 and 180 mg/kg) before daily administration.

Cell culture and treatment
Human hepatocyte line L02 cell (L02) was obtained from the Chinese Academy of Sciences (Shanghai Institute of Biochemistry and Cell Biology, Shanghai, China). It was cultured in 10% fetal bovine serum (FBS) (Bioind, Israel) and sodium pyruvate solution (Gibco, USA) in RPMI-1640 medium (Thermo, USA) at 37°C with 5% CO 2 . All cell lines were confirmed to be free of mycoplasma contamination. To emulate the NAFLD model, L02 cells were treated with 0.5 mM PA (Sigma, USA) containing cell culture medium for 24 hours. Bovine serum albumin (0.5%) (BioFroxx, Germany) was used as a control.

Animal models
Mice were housed in specific pathogen free (SFP) conditions with an ambient temperature of 22-26°C and humidity of 50-65%, providing 12/12 hours of alternating day and night light and darkness. The NASH was modeled by feeding mice an HFHC (Nantong Trophic; fat, 40%; cholesterol, 2%) for 20 weeks and rendering SC (90 and 180 mg/kg) by gavage after 10 weeks. Mice feeding normal chow (NC) were taken as control. All animal protocols were approved by the Animal Ethics Committee of Guangdong Pharmaceutical University.

Cell-counting kit-8 assay
The L02 cells viability was measured by cell-counting kit-8 (CCK-8) colorimetric assay (BioSharp, China), according to the manufacturer's instructions. L02 cells were subcultured in 96-well plates with complete medium and treated with different concentrations of SC for 24 hours. After treatments, CCK8 reagent was added into each well, followed by incubating for another 1.5 hours in a 37°C incubator with 5% CO 2 . Then, the absorbance was measured by a microplate reader at 450 nm (Bio-Rad, Hercules, CA, USA ), and cell viability was expressed as percentage values, as compared with the control group.

Lactate dehydrogenase
An lactate dehydrogenase (LDH) cytotoxicity assay kit (Jiancheng Institute of Biotechnology, Nanjing, China) was used to examine the release level of LDH according to the instructions, and the optical density of the samples was measured by a microplate spectrophotometer (Bio-Rad, Hercules, CA, USA) at 450 nm. The level of cytokines was measured using ELISAbased kits, according to the manufacturer's instructions.

Cellular Oil Red O staining
L02 cells were treated with 0.5 mM PA, rinsed twice with PBS, fixed with 4% paraformaldehyde for 15 minutes, and then stained with 60% Oil Red O working solution for 5 minutes. After washing three times with deionized water, images were observed under a microscope (Olympus, Tokyo, Japan).

Glucose and insulin tolerance tests
The glucose tolerance test (GTT) was performed on mice fed with HFHC diet for 16 weeks. One week later, the same mice were performed by the insulin tolerance test (ITT). For GTT, mice were fasted for 16 hours. After measuring the baseline blood glucose level by means of a tail nick using a glucometer, 2 g/kg glucose was administered by means of intragastric injection, and glucose levels were measured 15, 30, 60, and 120 minutes after glucose injection. For ITT, mice fasted for 6 hours were injected i.p. with insulin at 0.5 U/kg and their blood glucose concentrations were determined 15, 30, 45, and 60 minutes after insulin injection.

Serum assays
Serum TG, TC, high-density lipoprotein (HDL), lowdensity lipoprotein (LDL), ALT and AST levels were measured by a commercial kit, according to the manufacturers' instruction (Jiancheng Bioengineering Institute, Nanjing, China).

Quantitative analysis of hepatic TG and total cholesterol
Hepatic TG and TC were extracted from liver tissues with a mixture of chloroform and methanol. The contents of hepatic TG and TC were measured by a commercial kit (Jiancheng Bioengineering Institute, Nanjing, China) and normalized by liver wet weight.

Quantitative analysis of hepatic free fatty acids and free cholesterol
Hepatic free fatty acid (FFA) and free cholesterol (FC) were extracted from liver tissues with saline. The contents of hepatic FFA and FC were measured by a commercial kit (Jiancheng Bioengineering Institute, Nanjing, China) and normalized by liver wet weight.

Histopathology
Liver tissue was fixed overnight in 4% paraformaldehyde solution (4°C), embedded in paraffin, and then sectioned (4 μm) for hematoxylin and eosin (H&E) staining to visualize liver ballooning, steatosis, and inflammatory cell infiltration. Picrosirius red (PSR, 26357-02; Hede Biotechnology Co., Ltd., Beijing, China) staining was performed to visualize the degree of liver fibrosis. The positive areas were quantified using the Image J software. Histologic images of section tissues were captured with a light microscope (Olympus, Tokyo, Japan).

Real-time quantitative PCR
Total RNA of liver tissues was extracted using TRIzol reagent (Thermo, USA), followed by reverse transcription and quantitative real-time PCR (q-PCR). From the extracted mRNA, cDNA was synthesized using the PrimeScript™ RT reagent kit with gDNA Eraser (Takara, Beijing, China). All the primer sequence information was shown in Table 1. Q-PCR assay was performed using the SYBR Green Supermix (Bio-Rad, Hercules, CA, USA). The relative amount of each mRNA was calculated by using the comparative threshold cycle method. Comparative threshold values were normalized to GAPDH.

Western blot analysis
The L02 cells were lysed with ice-cold RIPA lysis buffer (65 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS) supplemented with a protease inhibitor (Roche, Basel, Switzerland) and a phosphatase inhibitor (Roche, Basel, Switzerland) and centrifuged at 12,000 rpm for 30 minutes at 4°C. Total proteins (20-40 µg) were electrophoresed on SDS-PAGE gels and transferred to polyvinylidene fluoride membranes (Millipore). Afterwards, the membranes were blocked with 10% nonfat dry milk, followed by incubation with primary and secondary antibodies. Membranes were detected by Clarity Western electrochemiluminescence (ECL) Substrate (Bio-Rad, USA) in conjunction with a chemiluminescence system (New Life Science Products, USA).

Statistical analysis
All data were statistically analyzed by using GraphPad Prism 8.3.0 (San Diego, California, USA), and the results were expressed as the mean ± SD. Differences between the means of the 2 groups were analyzed by a 2-tailed unpaired Student's t-test and were considered statistically significant when p < 0.05, while for comparisons between more than 2 groups a one-way ANOVA was performed.

SC ameliorates PA-induced hepatocyte injury and lipid accumulation in L02 cells
We examined the effect of SC on hepatocyte viability using the CCK8 assay, and the results showed that when the SC concentration was higher than 80 μM, the viability of hepatocytes was decreased ( Figure 1A). Next, L02 cells were treated with 0.5 mM PA for 24 hours to establish a cell model of NAFLD and simultaneously administered with various concentrations of SC. The results showed that the cell survival rate gradually increased as the SC concentration increased from 5 to 20 μM, whereas the cell survival rate gradually decreased when the SC concentration increased from 40 to 400 μM ( Figure 1B). Thus, 5, 10, and 20 μM of SC were selected for the subsequent cell experiments. LDH level in the supernatant of L02 cells was elevated by PA induction, suggesting hepatocyte was damaged by lipid toxicity. However, LDH level in hepatocyte was reduced by SC treatment ( Figure 1C). The TG content in L02 cells was induced by PA, whereas reduced by SC treatment ( Figure 1D). This finding was further confirmed by Oil Red O staining ( Figure 1E), suggesting that SC treatment attenuated TG accumulation in L02 cells induced by PA. Moreover, q-PCR assays showed that SC-diminished PA induced the mRNA levels of lipid synthesis such as carbohydrate response element binding protein (CHREBP), sterol regulatory element-binding protein 1C (SREBP-1C), and DGAT-1 and lipid uptake genes such as fatty acid translocase CD36 (CD36), whereas did not affect other lipid metabolism-related genes such as peroxisome proliferator-activated receptor-α (PPAR-α), adipose triglyceride lipase (ATGL), monoacylglycerol lipase (MGL), and hormonesensitive lipase (HSL) ( Figure 1F), suggesting that SC reduced lipid synthesis and uptake in L02 cells induced by PA. Collectively, these data demonstrated that SC attenuated hepatocyte injury and lipid accumulation induced by PA in hepatocytes.

SC attenuates HFHC-induced obesity and insulin resistance in mice
To investigate the effects of SC on NASH, we established a NASH mouse model induced by an HFHC diet. Mice were fed with HFHC diet for 20 weeks and treated with SC (90 and 180 mg/kg/d) since the 11th week ( Figure 2A). In the 10th week, there were no significant differences in the body weights between the NASH mice and the control mice. At the end of the 20th week, the body weights of the NASH mice were more than those of normal mice, whereas the body weights of the NASH mice treated with atorvastatin (ART) and SC were lower than those of the NASH mice without any treatment ( Figure 2B). To assess the effect of SC on insulin resistance in NASH mice, we performed GTT and ITT. After 20 weeks of administration of a highcalorie diet, GTT assay showed that after i.p. injection of glucose, the levels of blood glucose of the mice rapidly reached the peak after 15 minutes, whereas decreased to the baseline after 120 minutes ( Figure 2C). The AUC of GTT in NASH mice was larger than that of the normal chow mice; the AUC of GTT in NASH mice with SC treatment was smaller than that of NASH mice without any treatment ( Figure 2C). Furthermore, ITT assay showed that after the i.p. injection of insulin, the levels of blood glucose in mice continued to drop in the first 30 minutes, the blood T A B L E 1 Primer sequences for real-time PCR

Primer name
Forward primer sequence Reverse primer sequence
SODIUM CHOLATE AMELIORATES NONALCOHOLIC STEATOHEPATITIS glucose of control mice firstly rebounded after 30 minutes, and the blood glucose of NASH mice started to rebound after 45 minutes ( Figure 2D). Moreover, the AUC of ITT in NASH mice was larger than that of the normal chow mice; the AUC of ITT in NASH mice with SC treatment was smaller than that of NASH mice without any treatment ( Figure 2D). Taken together, these data suggested that SC treatment improved the rate of glucose metabolism and insulin resistance in NASH mice.

Effects of SC on hepatic injury and serum lipid profile in HFHC-fed mice
The above in vivo experiments in mice demonstrated the comprehensive protective effect of SC in improving the systemic metabolic stress status in NASH mice. Therefore, we next assayed the levels of serum glutamate-pyruvate ALT and AST, the classical indicators of clinical liver function, and the results showed that SC and ART treatment significantly reduced the abnormal rise of serum transaminases induced by HFHC ( Figure 3A). In addition, consistent with the previous results, SC and ART treatment decreased the abnormal elevation of serum TG and TC induced by HFHC, suggesting that SC restored lipid metabolism homeostasis in NASH mice ( Figure 3B). Abnormal changes in HDL and LDL are also key serum biochemical indicators in the context of NASH, reflecting the systemic metabolic stress state and the extent of vascular endothelial damage. SC and ART treatment improved the abnormal changes of lipoproteins mentioned above ( Figure 3C). These data suggested that SC attenuated hepatic injury and serum lipid profile in NASH mice.

SC ameliorates hepatic steatosis and injury in HFHC-fed mice
H&E staining exhibited significant steatosis and hepatocyte ballooning in the liver of NASH mice. Nevertheless, treatment with SC dramatically reduced these 2 lesions and its effect is similar to that of ART ( Figure 4A). On the basis the above pathological staining results, we assessed the liver injury in mice by 3 histological features, namely the degree of steatosis, the number of inflammatory lesions in the liver lobules, and the degree of ballooning of hepatocytes, by NAS score. The results showed that SC treatment significantly improved hepatic steatosis and injury in HFHC-induced NASH mice ( Figure 4B). At week 20, the liver weight and the liver/body weight ratio were increased in NASH mice. However, both were decreased in NASH mice treated with ART and SC compared with NASH mice without any treatment ( Figure 4C). Hepatic TG and TC accumulation in HFHC-fed mice was significantly decreased by SC treatment ( Figure 4D). However, when the liver is chronically exposed to high levels of FC and FFA, it increases the body's insulin resistance, [7] causing lipotoxicity and promoting the secretion of inflammatory cytokines, leading to hepatocyte damage and progressive fibrosis during NASH. Therefore, we further assayed these 2 key indicators, FFA and FC, and our data suggest that in a mouse model of NASH, SC and ART treatment significantly reduces the levels of FFA and FC ( Figure 4E). To confirm the molecular mechanism of SC for restoring hepatic lipid metabolism homeostasis, we next assayed a representative series of genes for cholesterol and lipid metabolism, including genes for cholesterol and lipid biosynthesis, lipolysis, and uptake, using q-PCR assays. The results showed that the mRNA levels of cholesterol synthesis and lipogenesis genes such as SREBP cleavage-activating protein (Scap), Srebp2, HMG-CoA Reductase (Hmgcr), Srebp-1c, and Dgat1 were increased in HFHC-fed mice, whereas decreased by SC and ART treatment ( Figure 4F). In addition, SC and ART treatment decreased the mRNA levels of hepatic lipid uptake genes such as Cd36 in HFHC-fed mice ( Figure 4F). Conversely, the hepatic mRNA levels of Ppar-α and Atgl were decreased in HFHC-fed mice, whereas increased by SC and ART treatment ( Figure 4F). These data suggested that SC attenuated HFHC-induced hepatic steatosis and disorder of hepatic lipid metabolism in mice.

SC attenuated liver inflammation and progressive fibrosis in HFHC-fed mice
To further investigate the protective effects of SC on NASH in vivo, we evaluated the effects of SC on liver inflammation and fibrosis in NASH mice. Immunohistochemical staining of CD68 showed that NASH mice treated with SC-exhibited decreased inflammatory cell infiltration compared with NASH mice without any treatment ( Figure 5A). Furthermore, SC treatment significantly decreased the hepatic mRNA levels of inflammatory cytokines such as Il-1β, Ccl2, and Ccl5 in NASH mice ( Figure 5B). In addition, PSR staining showed that SC treatment significantly reduced the hepatic collagen deposition in NASH mice, thus slowing down the progression of liver fibrosis ( Figure 5C). Together, these data suggested that SC attenuated hepatic inflammation and fibrosis in NASH mice under a metabolic stress condition.

SC activates FXR signaling in NASH mice
To disclose the key target proteins of SC against NASH, protein-protein interaction network consisted of 97 nodes and 295 edges ( Figure 6A) was acquired from the gene expression STRING database. [8] Among the target proteins with degree > 5, TP53, TNF-α, PPARA, and NR1H4 (FXR) were reported to involve in multiple biological processes contributed to the NASH progression. [9] Among the protein-protein interaction network, the cluster containing NR1H4 (FXR) had 25 targets with an average score of 2.48; the cluster  containing TNF-α had 20 targets with an average score of 2.05; and the cluster containing PPARA had 23 targets with an average score of 1.62 (the cluster score represents the core density of nodes and topologically adjacent nodes, with higher scores representing more concentrated clusters), suggesting that NR1H4 (FXR) might be a potential direct target gene for SC treatment of NASH ( Figure 6A). GO enrichment: GO Ontology enrichment [10] showed that there were 101 shared targets were enriched with a p-value threshold of 0.01. Among the top 20 enrichment results, Response to hormone, Steroid metabolic process, Positive regulation of cytosolic calcium ion concentration and nuclear receptor activity were enriched to a relatively high degree and decreased p-values. Thus, it is suggested that nuclear receptor activity might be associated with FXR ( Figure 6B). Among 101 outstandingly enriched Kyoto encyclopedia of genes and genomes (KEGG) pathway were obtained from the metascape database, [11] top 20 ranking KEGG pathway by p-value were descripted in Figure 6C, of which the most target proteins of SC were enriched in PPAR signaling pathway, cell cycle, and bile secretion. Considering the important role of bile acids in the process of NASH, we further investigated the genes related to bile acid synthesis in the liver tissue of NASH mice. FXR, as a key gene, mediates bile acid metabolism. [12] Therefore, to verify the role of SC in bile acid metabolism, we examined genes involved in hepatic bile acid synthesis-related genes. The hepatic mRNA levels of Shp, the binding target of Fxr, was reduced in HFHC-fed mice, whereas significantly increased by SC and ART treatment ( Figure 6D). Conversely, the hepatic mRNA levels of key enzymes of the major bile acid synthesis pathway in liver such as Cyp7a1 and Cyp8b1 were increased in HFHC-fed mice, whereas decreased by SC and ART treatment ( Figure 6D). To further determine at the protein level whether SC ameliorates NASH by the activation of FXR signaling, we verified the levels of FXR in hepatocytes. SC-activated FXR signaling in hepatocytes subjected to PA stimulation ( Figure 6E). The KEGG analysis showed that the cancer pathway was the second most enriched pathway, suggesting high-dose SC treatment significantly decreased the mRNA levels of cancer markers such as Tnf-α, α-fetoprotein (Afp), and Ki67 in NASH mice ( Figure 6F). Collectively, these data suggested that SCactivated FXR and bile acid synthesis to blunt the pathogenesis of NASH in mice.  The FXR-FGF15 pathway is required for SC to ameliorate HFHC-induced intestinal inflammation in mice Next, q-PCR assays showed that the mRNA levels of Fxr and Fgf15 and bile acid efflux transporter protein [organic solute transporter β, Ostβ] in the ileum were significantly decreased in NASH mice, whereas reversed by SC treatment (Figure 7A). In contrast, the mRNA levels of apical sodium-dependent bile acid transporter (Asbt) were increased in the ileal tissues in HFHC-fed mice, whereas decreased after SC and ART treatment ( Figure 7A). We next assayed the mRNA levels of intestinal inflammatory cytokines in NASH mice. As expected, SC and ART treatment significantly decreased the intestinal mRNA levels of inflammatory cytokines such as Tlr4, Tlr2, and Il-6 in NASH mice ( Figure 7B). Collectively, these data suggested that SC treatment protected mice against intestinal inflammation by HFHC induction, suggesting that SC exerts its anti-NASH effects as least in part through intestinal FXRmediated anti-inflammatory mechanisms.

DISCUSSION
NAFLD has been the most common chronic liver disease worldwide. There are no approved drugs because of the complexity of the disease and the safety of the drugs. [5] Thus, it is urgent to explore new drugs to treat NAFLD. SC is a mixture of sodium taurocholate and sodium glycocholate extracted from bovine bile. [13] In the current study, we demonstrated that SC attenuated lipid accumulation and hepatocyte injury in a PA-induced hepatocytes and inhibited steatosis, inflammation, and fibrosis in NASH mice.
Sterol regulatory element-binding protein (SREBPs) is an essential group of transcription factors regulating lipid synthesis. [14,15] SREBP-1c is involved in hepatic lipid synthesis, and overexpression of SREBP-1c causes hepatic lipid accumulation and insulin resistance. [16] SREBP2 regulates the expression of genes related to cholesterol synthesis and uptake. [17] It was found that abnormal expression of SREBP2 would cause disorders of lipid metabolism, especially cholesterol metabolism, which would lead to excessive deposition in adipose tissue and induce NAFLD. [17] The downstream target gene of SREBP2 is HMG-CoA Reductase (HMGCR). [18] In NAFLD, an abnormal increase in SREBP2 will promote the synthesis of (HMGCR). [18] In addition, when the SREBP protein precursor binds to SREBP cleavage-activating protein (SCAP), leading to endoplasmic reticulum stress and exacerbating the imbalance of TG metabolism, which in turn leads to further intracellular lipid accumulation and induces the development of NAFLD. [19] TG deposition in hepatocytes caused by the disorders of lipid metabolism is the cornerstone of the development of NAFLD. [20] When the liver is unable to regulate the accumulation of lipids through β-oxidation, FC and FFA form lipotoxic substances, leading to endoplasmic reticulum stress, oxidative stress, and inflammatory vesicle activation. [21] PPAR-α regulates β-oxidation and the transport of FFA, which plays a crucial role in lipid metabolism. [22] Liver CHREBP-specific knockdown exacerbates obesity, hepatic steatosis, and insulin resistance in mice and vice versa. [23] Our results showed that SC treatment significantly decreased the hepatic levels of FFA and FC in NASH mice; meanwhile, the hepatic cholesterol synthesis genes such as Scap, Srebp2, and Hmgcr in NASH mice treated with the mRNA levels of high-dose SC were also reduced; in addition, the mRNA levels of lipogenic genes such as Srebp-1c was reduced, whereas the hepatic mRNA levels of Ppar-α and Atgl were significantly increased in NASH mice treated with SC and ART. These results suggested that SC treatment attenuates hepatic steatosis and lipid metabolism disorders in NASH mice.
The inflammatory process, a hallmark of NASH pathogenesis, is associated with hepatocyte injury and the release of multiple proinflammatory cytokines. The levels of proinflammatory cytokines in the liver, including IL-6, IL-1β, CCL2, and CCL5, are correlated with the severity of NASH. [24] Macrophages serve a significant role in the pathogenesis of NAFLD. In terms of liver inflammation, the number of CD68-positive cells reflected macrophage recruitment. Macrophage-mediated immune responses are an essential cause of hepatocyte injury during the development of NAFLD. [25] The accumulation of large amounts of lipids exposes macrophages to prolonged "antigenic" stimuli, inducing the production of IL-6 and causing a sustained inflammatory response. [26] In addition, inflammation of adipose tissue causes liver injury. In obese mice with severe lipid accumulation, the secretion of IL-1β by adipose tissue macrophages (ATMs) is elevated, which increases the rate of lipolysis of adipocytes and promotes hepatic steatosis. [27] CCL2 and CCL5 overexpression recruited macrophages that secrete large amounts of inflammatory cytokines and facilitate hepatic steatosis and vice versa, suggesting that the recruitment of macrophages in adipose tissue causes lipid accumulation in mice. [27] Our results showed that SC suppressed liver and intestinal inflammation in NASH mice by decreasing the mRNA levels of inflammatory cytokines such as Il-1β, Ccl2, Ccl5, Tlr4, Tlr2, and Il6. Meanwhile, immunohistochemical staining of CD68 demonstrated that treatment with SC reduced inflammatory cell infiltration in livers of NASH mice. These results suggested that SC exerts an anti-inflammatory effect on liver and intestine in NASH mice.
Severe stages of NASH might be accompanied by fibrosis that manifests in the form of excessive deposition of insoluble collagen and extracellular matrix. [28] HSC is the main source of myofibroblasts, and bone marrow-derived fibroblasts are a potential source of myofibroblasts, which are associated with the pathogenesis of liver fibrosis. [29] Our results showed that mice fed by HFHC diet developed into liver fibrosis. However, SC reduced the severity of liver fibrosis in the NASH mouse model, thereby hindered the progression of liver fibrosis, suggesting SC exerts an antifibrotic effect on NASH-associated fibrosis.
Bile acids are synthesized from cholesterol in the liver and then transported to the intestine. [30] They promote the absorption of lipids and fat-soluble vitamins and are important signaling molecules that regulate glucolipid and energy metabolism in the body. Recent studies have shown that disorders of bile acid metabolism are an important cause of the development of NAFLD/NASH. [30] Dysregulation of bile acid metabolism affects hepatic lipid metabolism, immune microenvironment, and intestinal bacteria. The composition of BAs in the liver and intestine is regulated by BA-metabolizing enzymes and BA transporters, the majority of which are encoded by the FXR target gene that is predominantly expressed in hepatocytes and lower small intestinal epithelial cells, and by the intestinal microbiota. [31] FXR is a bile acid receptor, which regulates bile acid metabolism by regulating key target genes in all aspects of bile acid metabolism, including bile acid synthesis, metabolism, reabsorption, and transport. [32] Recent studies have shown that bile acids in the enterohepatic circulation maintain a dynamic balance of bile acids in the body by regulating FXR activity. [33] When bile acid pooling occurs to the liver, excess bile acids activate FXR and induce increased expression of small hetero dimer partner (SHP). [34] In the intestine, FXR induced the binding of FGF15/19 to the FGFR4, inhibiting CYP7A1 expression and bile acid synthesis. [35] In the intestine, conjugated bile acids are first actively reabsorbed into the small intestinal mucosal cells through ASBT in the ileum, and subsequently enter the portal vein by the bile acid efflux transporter OSTα/ OSTβ in the basolateral layer. [36] In addition, many studies have shown that FXR agonists are potential therapeutic drugs for metabolic and inflammatory diseases. [37,38] Obeticholic acid (OCA) is a selective FXR agonist with anticholestatic and hepatoprotective properties, [39] participating in BA anabolism and enterohepatic circulation, and modulate immune inflammatory and fibrotic. [40] Since the FXR agonist OCA currently developed for the treatment of NASH was denied marketing approval by the FDA because of its safety and side effects, it is sufficient to demonstrate the critical role that bile acids play in the NASH disease process. Our results showed that SC activates hepatic FXR signaling and induces the expression of downstream SHP, inhibiting the rate-limiting enzyme for bile acid synthesis (CYP7A1), leading to reduced bile acid synthesis in the liver. In addition, SC significantly increased the expression of FXR and FGF15 in NASH mouse intestine. Finally, SC activation of intestinal FXR reduced intestinal bile acid levels by inhibiting the expression of ASBT transporter to bile acid reabsorption, and inducing OSTβ transporter expression to promote bile acid efflux. KEGG analysis showed that the cancer pathway was the second most enriched F I G U R E 7 Activation of FXR-FGF15 pathway is responsible for the effect of SC on intestinal inflammation in NASH mice. (A) Quantitative real-time PCR analysis of the transcript levels of genes related to FXR-FGF15 pathway (Fxr, Fgf15, Asbt, and Ostβ ). n = 8 per group. (B) Quantitative real-time PCR analysis of the transcript levels of genes related to inflammation of the intestinal tract (Tlr4, Tlr2, and Il6). n = 8 per group. Data are represented as means ± SD. #indicates a significant difference between the NC group and the HFHC group (t test); *indicates a significant difference between the L-SC (Low dose-Sodium Cholate: 90 mg/kg)/H-SC (High dose-Sodium Cholate: 180 mg/kg)/ART group and the HFHC group (one-way ANOVA). ## p < 0.01, ### p < 0.001 versus NC mice; *p < 0.05, **p < 0.01, ***p < 0.001 versus mice fed by HFHC. Abbreviations: ART, atorvastatin; HFHC, high-fat and high-cholesterol; NC, normal chow; ns, no significance. SODIUM CHOLATE AMELIORATES NONALCOHOLIC STEATOHEPATITIS | 13 pathway, so we assayed several cancer markers. Our results suggested that high-dose SC attenuates the abnormal expression of cancer markers such as Tnf-α, Afp, and Ki67 in NASH mouse liver. Finally, SC activated FXR signaling in hepatocytes subjected to PA stimulation. Taken together, the above results indicated that SC might activate both hepatic and intestinal FXR expression and regulate bile acid synthesis to impede the pathogenesis of NASH.
Collectively, our current results demonstrated that SC attenuates bile acid metabolism disorder in NASH mice by activating FXR signaling in the liver and intestine, contributing to the amelioration of steatosis, inflammation, and fibrosis in NASH mice. Therefore, our findings will provide insight into the development of clinical treatment for NASH ( Figure 8).

AUTHOR CONTRIBUTIONS
Tian Lan conceived and designed the experiments. Linyu Pan, Ze Yu, Xiaolin Liang, Jiyou Yao, and Yanfang Fu carried out the experiments and wrote the manuscript. Xu He, Xiaoling Ren, Jiajia Chen, Xuejuan Li, and Minqiang Lu took part in the discussion and proofreading the manuscript.

CONFLICT OF INTEREST
The authors declare no conflict of interests for this article.
F I G U R E 8 Potential mechanism by which SC attenuated hepatic steatosis, inflammation, fibrosis, and intestinal inflammation in NASH mice. SC reduced the expression of Srebp-1c and Dgat1 in NASH mouse liver, thereby alleviating hepatic steatosis and lipid metabolism disorders. SC suppressed liver and intestinal inflammation in NASH mice by decreasing the mRNA levels of inflammatory cytokines such as Il-1β, Ccl2, Ccl5, Tlr4, Tlr2 and Il6. The protective mechanism of SC not only attributes to the downregulation of Scap, Srebp2 and Hmgcr by SC, inhibiting the synthesis of cholesterol in the NASH mouse liver. More importantly, SC activates hepatic FXR signaling and induces the expression of downstream SHP, inhibiting CYP7A1 expression, leading to reduced bile acid synthesis in the liver. Meanwhile, SC activates intestinal FXR and induces the expression of FGF15, followed by the secretion of FGF15 into the liver and inhibition of CYP7A1 expression in liver and decreased hepatic bile acid synthesis. Furthermore, SC activation of intestinal FXR reduced intestinal bile acid levels by inhibiting the expression of ASBT transporter to bile acid reabsorption, and inducing OSTβ transporter expression to promote bile acid efflux. In conclusion, SC attenuates bile acid metabolism disorder in NASH mice by activating FXR signaling in the liver and intestine, contributing to the amelioration of steatosis, inflammation, and fibrosis in NASH mice. Abbreviations: ASBT, apical sodium-dependent bile acid transporter; CHREBP, carbohydrate response element binding protein; FXR, farnesoid x receptor; HMGCR, HMG-CoA reductase; SREBP cleavage-activating protein; SREBP2, sterol regulatory element-binding protein 2; TLR, toll-like receptor.

ETHICS STATEMENT
All animal experiments were performed following the Guide for the Care and Use of Laboratory Animals, and the procedures were approved by the Research Ethical Committee of Guangdong Pharmaceutical University.