Reversal of cholestatic liver disease by the inhibition of sphingosine 1-phosphate receptor 2 signaling

Animals and experimental design

The protocol of this study were approved by the animal research committee from the Third Military Medical University (AMWEC20224009), and they all complied with the National Research Council’s Guide for the Care and Use of Laboratory Animals. We obtained a total of 45 male Sprague-Dawley rats (weighing 180–220 g) from the Laboratory Animal Center of the Third Military University. The rats were bred in a controlled environment that was free from specific-pathogen-free (SPF). They were provided with sterilized water and a standard diet for rats. All surgical procedures were conducted utilizing an isoflurane chamber, and anesthesia was consistently administered during the entirety of the experiment using isoflurane (R510-22; RWD, Shenzhen, China) delivered through a small animal anesthesia machine’s nose cone (R500IE; RWD, Shenzhen, China). JTE-013, a selective antagonist of S1PR2, was acquired from MedChemExpress (Cat# HY-100675; Med Chem Express, Hong Kong, China). Three groups were randomly formed by dividing the animals: The first group, referred to as the sham-operation group, consisted of rats that underwent sham surgeries without bile duct ligation (n = 15). The second group, known as CBDL group, underwent CBDL model by performing bile duct ligation (BDL) near the junction of the hepatic ducts and above the entrance of the pancreatic duct using 5–0 silk. Then the ligatures were tightened, and the common bile duct was resected between the ligatures, n = 15) and the CBDL + JTE-013 group (CBDL rats treated with intraperitoneal administration of JTE-013 at a dosage of 10 mg/kg, two injections each week for a duration of 2 weeks, n = 15). At the 3 weeks after surgery, there were 15 surviving rats in the sham group, 13 in the CBDL group, and 11 in the JTE-013 group. Two rats were excluded from the study due to failed modeling, both of these rats belonged to the JTE-013 group. Four rats were excluded from the study due to inadequate fecal collection. Specifically, three rats from the sham group and one rat from the JTE-013 group were affected. Consequently,16s rRNA sequencing was conducted on a total of 33 rats, comprising 12 rats in the sham group, 13 rats in the CBDL3w group, and eight rats in the JTE-013 group).

Sample collection

Fecal samples were collected 3 weeks after surgery in each group. Fecal samples were obtained within a SPF environment, transferred into sterile 1.5 mL container, and promptly submerged in liquid nitrogen within 30 min. The frozen fecal samples were stored at −80 °C for subsequent analysis on the gut microbiota. In the 3 weeks following CBDL, rats were anesthetized by inhalation of isoflurane and sacrificed to collect blood from the abdominal aorta. Blood was centrifuged at 3,000 g for 20 min at 4 °C to separate plasma. The resulting plasma was then divided into four aliquot: one aliquot was used for liver function testing, another aliquot was used for Enzyme-linked immunosorbent assay (ELISA) assay, a third aliquot was used for serum bile acid quantification, and the remaining aliquot was stored for future use. Rat livers were extracted and a small portion of the liver was immediately preserved in 10% neutral-buffered formalin for histological examination. The remaining liver tissue was rapidly frozen in liquid nitrogen for subsequent molecular analysis.

Histology

Hematoxylin and eosin (HE) (G1120; Servicebio, Beijing, China) and Sirius Red Staining Kit (G1340; Solarbio, Beijing, China) were used to stain the rat liver sections that had been paraffin-embedded. Immunochemical staining was conducted using an anti-CD31 antibody (ab119339; Abcam, Cambridge, UK, 1:100) to provide evidence of angiogenesis. Anti-CK-19 antibody (GB11197; Servicebio, China, 1:250) was used for assessing bile duct epithelial cell proliferation. Anti-Collagen I (GB11022-3; Servicebio, China, 1:1,000) was performed to provide evidence of fibrosis. Anti-Caspase-3 Rabbit pAb (GB11009-1; Servicebio, China, 1:200 dilution) and TUNEL staining (GDP1042; Servicebio, China) were employed to assess apoptosis. To evaluate the occurrence of NETosis, a co-staining procedure was conducted using CitH3 (ab281584; Abcam, Cambridge, UK) and MPO (ab208670; Abcam, Cambridge, UK). Additionally, the anti-S1PR2 antibody (DF4921; Affinity, San Francisco, CA, USA, 1:100) was employed. The nuclei were stained using DAPI (G1012; Servicebio, Beijing, China). For each immunofluorescence slice, ten field were randomly selected using the Pannoramic MIDI (3DHISTECH, Hungary). The quantification of positive cells per high-power field was performed using Image-Pro Plus software (version 6.0; Media Cybernetics Inc, Rockville, MD, USA). For immunohistochemistry, the liver tissue was cut into 5 μm slices, which were subsequently subjected to dewaxing using xylene and rehydrated with distilled water in an ethanol gradient. Following a 30-min incubation with 3% hydrogen peroxide (H₂O₂) at room temperature, the sections were subsequently blocked with 1% bovine serum albumin. They were then treated with the following primary antibodies: anti-S1PR2 (DF4921; Affinity, San Francisco, CA, USA, 1:100), CD163 (ab182422; Abcam, Cambridge, UK, 1:500), MCP-1 (ab7202; Abcam, Cambridge, UK, 1:300), IL-6 (ab208113; Abcam, Cambridge, UK, 1:500), TNF-α (ab6671; Abcam, Cambridge, UK, 1:100), NOX2 (19013-1-AP; Proteintech, Rosemont, IL, USA, 1:1000) or NLRP3 (27458-1-AP; Proteintech, Rosemont, IL, USA, 1:200) for an entire night at 4 °C. Following this, an incubation process was carried out with, a secondary antibody. Finally, following the incubation of slices with 100 L of 3,3′-diaminobenzidine reagent, a subsequent staining procedure was performed using hematoxylin.

Analysis of hepatic function

The serum of the rats was utilized to collect liver biochemistry data, including aspartate aminotransferase (AST), total bile acid (TBA), alanine transaminase (ALT), and alkaline phosphatase (ALP). This data was used as evidence of liver injury and was analyzed using the Automatic Biochemistry Analyzer (AU5400; Olympus, Tokyo, Japan).

Western blot

Total proteins were extracted from liver tissue using a total protein extraction kit for animal cultured cells and tissues (SD-001/SN-002; Invent Biotechnologies, Plymouth, MN, USA). The protein concentration was determined using the BCA method (SF247582; Thermo Fisher Scientific, Waltham, MA, USA). Western blot analysis was performed using specific primary antibodies, including α-SMA (ab7817; Abcam, Cambridge, UK), and VEGFR1 (ab184784; Abcam, Cambridge, UK), and secondary antibody GAPDH (10494-1-AP; Proteintech, Rosemont, IL, USA). The immunoreactive bands were visualized using an ECL chemiluminescent kit (XF345252; Thermo Fisher Scientific, Waltham, MA, USA). After removing background noise and adjusting for loading controls, band density was quantified using Image J software (version 1.5.3; WS Rasband, National Institute of Health, Bethesda, MD, USA).

Enzyme‑linked immunosorbent assays (ELISA)

The level of growth factors, chemotactic factor, and inflammatory cytokines in the serum were assessed using a rat ELISA kit (Shanghai Jianglai Biological Technology, Shanghai, China). The specific factors measured included placental growth factor (PLGF) (JL11559), vascular endothelial growth factor (VEGF) (JL21369), Chemokine CXCL2 (CXCL2) (JL21265), Monocyte Chemoattractant Protein 1 (CCL2) (JL11450), Tumor Necrosis Factor-alpha (TNF-α) (JL13202), Nuclear Factor-kappa B (NF-κB) (JL21039), interleukin-6 (IL-6) (JL20896). All procedures were conducted in accordance with the instructions provided by the manufacturer. The color’s intensity at 450 nm of absorbance was measured using the Thermo reader (Varioskan Flash multimode reader; Thermo Scientific, Waltham, MA, USA). The concentration of these factors was expressed in picograms per milliliter (pg/ml), except for IL-6 which was expressed in nanograms per milliliter (ng/ml).

Gut microbiota 16S rRNA sequencing

A total of 200 mg of fecal samples were used for DNA extraction according to the manufacturer’s instructions. The research conducted by Zhang et al. (2023) focuses on the investigation of the hypervariable region, as well as the primers, and laboratory techniques employed for PCR amplification and quantification. The SILVA 16S rRNA database (v138) and the Naive Bayes consensus taxonomy classifier, developed in Qiime2, were employed to accurately classify ASVs into their respective taxonomic groups. To assess the abundance and diversity of gut microbiota across the three groups, the total number of sequences for each level of taxonomic (e.g., phylum and genus) were computed and aggregated. The bioinformatic study of the gut microbiota was conducted using the Majorbio Cloud platform (https://cloud.majorbio.com). Mothur version 1.30.1 was utilized to calculate rarefaction curves and alpha diversity indices using the data obtained from the Amplicon Sequence Variants (ASVs). The computed indices included observed ASVs, Chao1 richness, Shannon index, and Simpson index, ACE, observed species. By employing Vegan v2.5-3 package’s principal coordinate analysis (PCoA) utilizing Bray-Curtis dissimilarity,a comparison of microbial communities across different samples was facilitated. The Vegan v2.5-3 package was utilized to perform the PERMANOVA test,which aimed to assess the proportion of variance attributed to the therapy and its statistical significance. The raw data were uploaded to the NCBI Sequence Read Archive (SRA) database (BioProject ID: PRJNA1012444).

Bile acids quantification in serum

An LC-ESI-MS/MS (UHPLC-Qtrap) was utilized in the present study to detect the target compounds in sample both qualitatively and quantitatively. A Waters BEH C18 (150*2.1 mm, 1.7 m) liquid chromatography column (AB SCIEX) was used to separate the analyte chemicals. The flow rate of the mobile phases, which were given at 0.35 mL/min, was 0.1% formic acid-acetonitrile solution (solvent B) and 0.1% formic acid-water solution (solvent A). Each ion fragment was automatically identified and integrated using default parameters in the AB Sciex quantitative software OS, and manual inspection was aided.

Statistical analysis

Continuous data was represented by the mean with standard deviation (SD) or the median (interquartile range) based on the distribution of the data. The Mann-Whitney U-test was employed to compare the species variation between the two groups. One-way ANOVA was employed to assess the variations among the three groups, and the least significant difference (LSD) method was utilized for the post hoc test. A significance level of 0.05 was adopted as the threshold for statistical significance in all two-sided statistical tests. All statistical analyses were performed using SPSS software (version 25.0), and all bar charts were generated using GraphPad PRISM (version 7.00; GraphPad Software, San Diego, CA, USA).

Alterations in the serum bile acid profile and intestinal flora in rats with cholestatic liver injury induced by CBDL

H&E staining of liver tissue demonstrated that at 3 weeks after CBDL, significant liver injuries were prominently observed in the rats, including bile duct proliferation and dilation, lobular structure destruction, cirrhotic nodule formation, and inflammatory cell infiltration (Fig. 1A). Meanwhile, the serum bile acid profile exhibited significant alterations in the CBDL rats. As anticipated, the ligation of the bile duct prevented the primary binding bile acids (TCA, TCDCA) synthesized by the liver from entering the intestine. Consequently, the enterohepatic circulation was disrupted, leading to a significant decrease in both primary and secondary bile acids in the stool due to the obstruction of their source. This process leads to a notable elevation in the levels of primary conjugated bile acids (such as TCA, TCDCA, and T-β-MCA) in the blood. The observed alterations in the spectra of serum bile acids were found to be in accordance with our initial hypothesis, as depicted in Fig. 1B.

Interestingly, the hepatic expression of S1PR2, a receptor that specifically binds to the conjugated bile acid, exhibited a significant increase the CBDL 3w group’s liver (Figs. 1C–1F). Compared to the sham group, the CBDL 3w group exhibited a significant decrease in the biodiversity of intestinal flora. This was supported by the observed decrease in α-diversity levels (Fig. 1G). Additionally, there was a significant difference in the gut microbiota between the CBDL 3w group and the sham group (Fig. 1H). Despite the fact that the phylum Firmicutes is the most abundant in both of these groups, there was a relative increase in Proteobacteria, Actinobacteriota, and Bacteroidota after CBDL (Fig. 1I). At the genus level, the CBDL 3w group exhibited a higher proportion of Clostridium_sensu_stricto_1, Bifidobacterium, and Dubosiella, while showing a lower abundance of Bacillus, UCG-005, Romboutsia, Lachnospiraceae_NK4A136_group, Ruminococcus and Christensenellaceae_R-7_group, in comparision to the sham group (Fig. 1J).

JTE-013 ameliorates the CBDL-induced cholestatic liver injury

An investigation was conducted to determine if the specific intervention of S1PR2 could effectively suppress liver injury caused by CBDL and subsequently restore normal liver function. JTE-013, a selective antagonist of the S1PR2, was administered to rat models with CBDL, as depicted in Fig. 2A. As anticipated, the intervention of JTE-013 remarkably ameliorated cholestatic liver injury, as demonstrated by the decrease in levels of ALT (sham vs. CBDL vs. CBDL treated with JTE-013 = 30.5 ± 2.70 vs. 181.5 ± 45.73 vs. 98.5 ± 27.62 (U/L), P < 0.01), and ALP (148.8 ± 26.71 vs. 629.6 ± 256.96 vs. 356.0 ± 55.27 (U/L), P < 0.05) (Fig. 2B). Additionally, the histological injury was ameliorated (Fig. 2E); fibrosis was reduced (Figs. 2C–2F and 2H, P < 0.001), and the proliferation of bile duct was diminished (Figs. 2E and 2G, P < 0.001). Following the administration of JTE-013, a significant reduction in liver inflammation was observed, this was demonstrated by changes in ELISA measurements of NF-κB (sham vs. CBDL vs. CBDL treated with JTE-013 = 168.5 ± 26.9 vs. 1123.2 ± 323.8 vs. 676.5 ± 99.7 (pg/ml), P < 0.01), TNF-α (0.08 ± 0.01 vs. 0.14 ± 0.03 vs. 0.07 ± 0.01 (pg/ml), P < 0.01), CXCL2 (10.4 ± 2.6 vs. 48.9 ± 7.5 vs. 35.7 ± 7.4 (pg/ml), P < 0.01), CCL2 (41.7 ± 32.3 vs. 2068.1 ± 843.1 vs. 825.9 ± 437.7 (pg/ml), P < 0.01) (Fig. 3A). Additionally, immunohistochemical analysis revealed changes in CD163, MCP1, IL-6, and TNF-α (Figs. 3B and 3C, all P < 0.05). Simultaneously, there was a notable decrease in hepatocyte apoptosis, as indicated by the decrease in Tunel-positive cells (Figs. 3D and 3E, P < 0.001) and Caspase 3-labeled cells (Figs. 3D and 3F, P < 0.001). Furthermore, liver angiogenesis was also ameliorated after the injection of JTE-013, as evidenced by ELISA changes of PLGF (30.3 ± 6.9 vs. 1085.4 ± 257.0 vs. 632.1 ± 307, P < 0.01) and VEGF (34.4 ± 2.7 vs. 376.8 ± 189.4 vs. 104.3 ± 23.3, P < 0.01) (Fig. 4A), number changes of CD31-labelled blood vessel (Fig. 4B, P < 0.001), as well as protein level changes of VEGFR1 (Fig. 4C, P = 0.01). These results strongly suggest that JTE-013 attenuates cholestatic liver injury possibly by reducing the hepatic inflammatory response and apoptosis.

JTE-013 reverses the dysbiosis of gut microbiota induced by CBDL

According to changes in α diversity, the CBDL 3w group treated with JTE-013 showed improvements in the abundance of intestinal flora compared with the CBDL 3w group (Fig. 5A). The PCoA analysis revealed significant differences between the CBDL and JTE-013 treated CBDL groups (P = 0.01, R = 0.23) (Fig. 5B). According to the community barplot analysis, the group treated with JTE-013 for CBDL showed a reduction in both Proteobacteria and Actinobacteriota (Fig. 5C). At the genus level, the JTE-013 treated CBDL 3w group exhibited a higher abundance of the genera norank_f_Muribaculaceae, NK4A214_group, Ruminococcus, and Christensenellaceae_R-7_group, while showing a lower abundance of Enterococcus compared to the CBDL 3w group (Fig. 5D). In summary, the JTE-013 treatment resulted in an increased abundance of fecal flora, a higher amount of beneficial bacteria, and while a decreased amount of harmful bacteria.

Inhibition of S1PR2 by JTE-013 blocks NETosis and reduces the liver damage through the TCA/S1PR2/NOX2/NLRP3 pathway

Excessive NETosis has been verified to be associated with chronic inflammation. MPO and CitH3 have been used as markers for NETosis. Immunofluorescence results revealed a significant increase in the number of cells co-stained by MPO (Fig. 6F, P = 0.009) and CitH3 (Fig. 6G, P < 0.01) in the CBDL group, indicating an elevated occurrence of NETosis. Conversely, in the CBDL group treated with JTE-013, a significant decrease in the number of co-stained cells was observed. This indicates that JTE-013 has the ability to reduce the occurrence of NETosis and consequently inhibit inflammation (Fig. 6E). The expression levels of NOX2 (Figs. 6A and 6B, P < 0.05) and NLRP3 (Figs. 6C and 6D, P < 0.001) were increased in the CBDL group. However, both markers showed a significant reduction after JTE-013 treatment. Additionally, the treatment with JTE-013 also resulted in a decrease in the index of oxidative stress MPO protein, compared to the CBDL group (Figs. 6H and 6I, P < 0.001). These findings suggest that TCA might mitigate NETosis and liver injury through the S1PR2/NOX2/NLRP3 pathway.