Hepatic LGR4 aggravates cholestasis-induced liver injury in mice

Abstract

Current therapy for hepatic injury induced by the accumulation of bile acids is limited. Leucine-rich repeat G protein-coupled receptor 4 (LGR4), also known as GPR48, is critical for cytoprotection and cell proliferation. Here, we reported a novel function for the LGR4 in cholestatic liver injury. In the bile duct ligation (BDL)-induced liver injury model, hepatic LGR4 expression was significantly downregulated. Deficiency of LGR4 in hepatocytes (Lgr4^LKO) notably decreased BDL-induced liver injury measured by hepatic necrosis, fibrosis, and circulating liver enzymes and total bilirubin. Levels of total bile acids in plasma and liver were markedly reduced in these mice. However, deficiency of LGR4 in macrophages (Lyz2-Lgr4^MKO) demonstrated no significant effect on liver injury induced by BDL. Deficiency of LGR4 in hepatocytes significantly attenuated S1PR2 and the phosphorylation of protein kinase B (AKT) induced by BDL. Recombinant Rspo1 and Rspo3 potentiated the taurocholic acid (TCA)-induced upregulation in S1PR2 and phosphorylation of AKT in hepatocytes. Inhibition of S1PR2-AKT signaling by specific AKT or S1PR2 inhibitors blocked the increase of bile acid secretion induced by Rspo1/3 in hepatocytes. Our studies indicate that the R-spondins (Rspos)-LGR4 signaling in hepatocytes aggravates the cholestatic liver injury by potentiating the production of bile acids in a S1PR2-AKT-dependent manner.

NEW & NOTEWORTHY Deficiency of LGR4 in hepatocytes alleviates BDL-induced liver injury. LGR4 in macrophages demonstrates no effect on BDL-induced liver injury. Rspos-LGR4 increases bile acid synthesis and transport via potentiating S1PR2-AKT signaling in hepatocytes.

INTRODUCTION

Cholestatic liver disease is a series of chronic liver diseases including drug-induced or congenital bile duct obstruction, primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC) (1, 2). Impairment of bile formation or flow results in disordered bile acid (BA) metabolism, hepatocytes and cholangiocytes death, fibrosis, and eventually cirrhosis. Current therapy for cholestatic liver diseases is still limited because of the incomplete understanding of molecular mechanisms related to these diseases. Bile acids, the most important constituents of bile, are the end products derived from insoluble cholesterol catabolism (3). Previous studies have suggested that bile acids are involved in multiple metabolic pathways under physiological and pathological conditions, especially hepatic lipid metabolism, glucose metabolism, energy metabolism, and cholestatic liver diseases (4, 5). Intrahepatic accumulation of bile acids has been demonstrated to be the major driving force for cholestatic liver diseases (6).

Leucine-rich repeat G protein-coupled receptor 4 (LGR4), also known as GPR48, was first found with G protein-coupled receptor 5 (LGR5) in 1998 (7). LGR4 is highly conversed in mammals and widely expressed in various tissues, such as cartilage, heart, liver, and hair follicles (8, 9). LGR4 is a seven transmembrane protein consisting of 951 amino acids (10). Its endogenous ligands are the R-spondin (Rspo) protein family, a group of four secreted proteins (Rspo1-4) identified as strong potentiators of Wnt/β-catenin signaling (11–14). Rspos-LGR4 signaling has been demonstrated to be critically involved in the regulation of multiple cellular responses, including embryogenesis (15), lipid homeostasis (16), energy metabolism (17), oncogenesis (18), and other diseases. Notably, emerging studies have revealed that LGR4 coordinates multiple hepatic functions. LGR4 serves as a link between the peripheral circadian clock and hepatic lipid metabolism (19). We have previously shown that Rspos-LGR4 protects hepatocytes against ischemia/reperfusion and lipopolysaccharide/d-galactosamine (LPS/D-Gal)-induced injury (20, 21). All these results demonstrate that LGR4 is a vital protein in liver. Whether LGR4 contributes to the pathogenesis of cholestatic liver diseases remains unknown.

Activation of sphingosine-1-phosphate receptors (S1PR) by S1P is critical for the essential cellular processes such as proliferation, migration, cytoskeletal organization, and morphogenesis (22). Among five subtypes of S1PR, S1PR2 has been recently shown to mediate the effect of conjugated bile acids on the activation of ERK1/2 and protein kinase B (AKT) signaling pathways in primary hepatocytes (23). Whether S1PR2-AKT signaling is involved in cholestasis liver injury remains to be demonstrated.

The present study reports for the first time that the lack of hepatic LGR4 protects against BDL-induced liver injury. This occurs specifically in hepatocytes via potentiating the production of bile acids in a manner dependent of S1PR2-AKT signaling. Deficiency of Lgr4 in macrophages demonstrates no effect.

MATERIALS AND METHODS

Chemicals

Taurocholic acid (TCA) was purchased from Sigma Aldrich (St. Louis, MO). R-spondin1 (Rspo1) and R-spondin3 (Rspo3) were obtained from R&D System (Minneapolis, MN). MK-2206 and JTE-013 were purchased from Selleck (Shanghai, China). Total triglyceride kit and total cholesterol kit were purchased from Bio-Technology and Science, Inc. (Beijing, China). Total bile acids kit and total bilirubin kit were obtained from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Kits used for the detection of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (ALP) were from Solarbio Science & Technology Company (Beijing, China). RNA trip and BCA protein quantitative assay kit were obtained from Applied Gene (Beijing, China). GoScript Reverse Transcription System was purchased from Promega (Madison, WI). Hieff qPCR SYBR Green Master Mix was obtained from Yeasen Biotechnology Company (Shanghai, China). Goat serum was purchased from ZOMANBIO Biotechnology Company (Beijing, China). Rabbit anti-LGR4 and Rabbit anti-S1PR2 were purchased from Abcam (Cambridge, MA). Rabbit anti-AKT and Rabbit anti-phosphorylation AKT were purchased from Proteintech (Wuhan, China). Rabbit anti-AMPK and Rabbit anti-phosphorylation AMPK were purchased from Cell Signaling Technology (Boston, MA). Mouse anti-β-actin was purchased from Proteintech (Wuhan, China). IRDye800 goat anti-mouse IgG and goat anti-rabbit IgG were purchased from Bioss Antibodies (Beijing, China). Rabbit anti-F4/80 was obtained from ABclonal (Wuhan, China). Fluorescein (FITC)-conjugated secondary antibodies were obtained from Huaxingbio Biotechnology Company (Beijing, China). Hematoxylin and eosin were from Zhongshan Jinqiao Biotechnology Company (Beijing, China). Adeno-associated virus (AAV)-TBG-Cre and AAV-GFP were from Vigene Biosciences (Shandong, China).

Cell Culture

AML12 cells (ATCC, Manassas, VA) are the normal mouse liver cell line with rich gene expression profiles, widely used in liver-related researches. AML12 cells were cultured in high-glucose DMEM contained with 10% fetal bovine serum (FBS), 1% penicillin and streptomycin (Macgene, Beijing), 1% insulin-transferrin-selenium (Thermo Fisher Scientific, Waltham, MA), and 1 μM dexamethasone (Sigma, St. Louis, MO) at 37°C in a humid incubator with 5% CO₂. Cells were seeded in a 12-well plate or 6-well plate until 60%–80% confluent. TCA was used to treat AML12 cells for 24 h.

Isolation and Culture of Primary Hepatocytes

Eight-week-old C57BL/6 mice were purchased from Beijing Vital River Laboratory Animal Technology Co. (Beijing, China) and anesthetized by intraperitoneal injection of pentobarbital sodium at 70 mg/kg body weight. Heparin (1,000 IU) was administrated to prevent blood coagulation. Through a midline laparectomy, liver, portal vein (PV), and inferior vena cava (IVC) were exposed. A cannula was inserted into the PV. Through this cannulation, the liver was instantly perfused with 20 mL of 37°C prewarmed D-Hanks’ buffer, followed by a perfusion with 20 mL of 0.03% prewarmed collagenase IV at a flowing rate of 5 mL/min. Liver was then carefully removed into a 10-cm dish containing high-glucose DMEM. Hepatic cells were dispersed into suspension and filtered through 100-μm nylon mesh, centrifuged for 3 min at 50 g in a swinging-arm centrifuge, and washed twice with serum-free high-glucose DMEM. Isolated hepatocytes were seeded in a 12-well plate or 6-well plate at a concentration of 1 × 10⁵ cells/mL and cultured in high-glucose DMEM supplemented with 10% FBS in a 37°C humid incubator with 5% CO₂. Rspo1 (400 ng/mL) or Rspo3 (200 ng/mL) was added to the cultured cells for 6 h. MK-2206 (2 μmol/L) or JTE-013 (10 μmol/L) was added to the cultured cells for 1 h. TCA (100 μmol/L) was used to treat AML12 cells for 24 h.

Animals

Lgr4^flox/flox mice were generated as described previously (20). Lyz2-cre mice in C57BL/6 background were purchased from Jackson Laboratory (Bar Harbor, ME). All animals were housed at a temperature of 25°C and under a 12 h light/12 h dark cycle (light 06:00–18:00, darkness 18:00–06:00), with access to food and water ad libitum. The Animal Care and Use Committee of Peking University was abided to accomplish experimental protocols (IRB00001052-16029).

Lgr4^flox/flox male mice (6–8 wk old) were injected with the AAV8-TBG-Cre or AAV8-GFP (2 × 10¹¹ vg/mL) via the tail-vein to construct liver-specific Lgr4 knockout mice (Lgr4^LKO). Two weeks later, Lgr4^LKO mice and Lgr4^flox/flox littermates were subjected to sham operation or ligation of the common bile duct (BDL). Lgr4^flox/flox mice and Lyz2-cre mice were cross-bred to construct myeloid-specific Lgr4 knockout mice (Lgr4^MKO). Lgr4^MKO mice and Lgr4^flox/flox littermates also received sham or BDL surgery. Blood and hepatic tissues were harvested two weeks later.

Bile Duct Ligation

BDL was performed as described previously (24). Briefly, Lgr4^LKO, Lgr4^MKO male mice, and Lgr4^flox/flox littermates were anesthetized with pentobarbital administrated via intraperitoneal injection at a dose of 70 mg/kg body weight. Through a laparectomy, cystic duct and common bile duct were exposed and ligated using 7-0 nylon. The abdomen was closed by two-layer sutures. Animals were placed on a heating pad during the procedure. Two weeks later, the animals were euthanized. Plasma and liver tissues were harvested for quantitative real-time RT-PCR, Western blotting analysis, immunoreactive analysis, and detection of the activity of hepatic enzymes.

Biochemical Analysis

Mouse plasma was collected from whole blood samples by centrifugation at 1,600 g for 15 min at 4°C. Twenty milligrams of liver tissue were homogenized in 1 mL chloroform-methanol mix (2:1) on ice and kept at 4°C overnight to extract lipids into organic solution. Distilled water (200 μL) was added to the homogenates. The mixture was centrifuged at 3,000 rpm for 10 min at 4°C. The lyophilized powder of lipids was resolved in 5% Triton X-100 in PBS, and the supernatant was used for lipid detection. Total triglycerides and cholesterol levels in plasma and hepatic tissue were determined by the GPO/PAP method according to the manufacturer’s instructions. Values were normalized to protein concentration or tissue weight. The plasma activity of AST, ALT, and ALP was evaluated by catalytic reaction which generates pyruvic acid. Pyruvic acid reacted with 2,4-dinitrophenylhydrazine (DNPH) to form 2,4-dinitrophenylhydrazone, which presents as a brownish-red color in an alkali condition. The enzymatic activity was calculated by the optical density value obtained from the measurement of absorbance at 505 nm. Plasma total bile acids and total bilirubin levels were measured according to the manufacturer’s instructions.

Histological Examination

Hepatic tissue was harvested and washed with phosphate-buffered saline (PBS), then fixed in 4% paraformaldehyde for 24 h. The hepatic tissue was dehydrated using solutions of alcohol, followed by transparentizing, and paraffin waxing. Tissue sections with 5-μm thickness were used for hematoxylin and eosin (H&E), Masson’s trichrome, and immunohistochemical staining. Specimens were observed and images were captured by an Olympus CKX53 inverted microscope (Olympus, Tokyo, Japan).

Immunofluorescent Staining

F4/80 immunofluorescent staining was performed according to the manufacturer’s instructions. Hepatic tissue was dehydrated using series of alcohol solutions at the concentrations ranging from 70% to 100%, then blocked with 1% normal goat serum in PBS for 1 h at room temperature. After being washed with PBS, the tissues were incubated with the F4/80 antibody (ABclonal, Wuhan, China) (1:100 in 1% normal goat serum) at 4°C overnight. Next day, the sections were incubated with FITC-conjugated goat anti-rabbit antibody (Huaxingbio Biotechnology Co., Beijing, China) (1:50 in 1% normal goat serum) for 1 h at room temperature in a dark place, then stained with Hoechst 33342 (10 μg/mL) for 5 min. Fluorescent signals were observed by a fluorescent microscope (Nikon).

Western Blot Analysis

Hepatic tissue and cells were homogenized in RIPA lysis buffer and proteins were extracted. The hepatic protein was subjected to 10% SDS-PAGE running gel and subsequently transferred to a nitrocellulose (NC) membrane. After being incubated in 4% fat-free milk at room temperature for 1 h, membranes were incubated with primary antibodies at 4°C overnight. Then members were incubated with fluorescent secondary antibodies for 1 h to detect the primary antibody bond with the targeted protein. The fluorescent signals were visualized and captured by Odyssey infrared imaging systems (LI-COR Biosciences, Lincoln, NE).

Quantitative Real-Time Polymerase Chain Reaction

Total mRNA was isolated from hepatic tissues using RNA trip. The hepatic RNA samples (4 μg) were reverse transcribed to cDNA using a GoScript Reverse Transcription System. Quantitative PCR based on SYBR Green Master Mix was performed using an AriaMx Real-Time PCR system (Agilent Technologies, CA). The amplification procedure was as follows: 95°C for 5 min, 40 cycles of 95°C for 10 s, 60°C for 20 s, and 72°C for 24 s. The sequences of primers used for quantitative real-time polymerase chain reaction (RT-qPCR) were listed in Supplemental Table S1.

Statistical Analysis

All data were expressed as means ± SE. Graphs were constructed, and statistical analyses were performed using the GraphPad Prism7.0 software (GraphPad Prism, RRID: SCR_002798). Statistic difference was determined by two-way ANOVA and Student’s t test. P < 0.05 was considered significant.

RESULTS

Downregulation of LGR4 Induced by Bile Acids in Liver and Hepatocytes

To determine whether Rspos-LGR4 signaling is involved in the cholestatic liver diseases, we first examined whether bile acids alter the expression levels of hepatic LGR4. As shown in Fig. 1A, Lgr4 was most abundantly expressed in liver relative to its two homologs: Lgr5 and Lgr6. Furthermore, our previous study (20, 25) and current results (Fig. 1, B and C) have shown that Lgr4 has a higher expression level in liver tissue and mainly expresses in hepatocytes. Interestingly, our present study showed that BDL markedly reduced the expression levels of Lgr4 mRNA in the liver. Also reduced were the expression levels of Lgr5 and Lgr6 (Fig. 1, D and E), indicating that intrahepatic bile acids may suppress the expression of hepatic Lgrs. To further demonstrate this concept, we treated AML12 cells with different concentrations of TCA, the major bile acid in mouse serum and liver post-BDL (26), which is critical for the BDL-induced cholestatic liver injury. Real-time RT-PCR analysis showed that TCA decreased expression levels of Lgr4 mRNA in AML12 cells (Fig. 1F). These results indicate that the expression of hepatic LGR4 is altered by bile acids.

Figure 1.

Suppression of hepatic Lgr4 mRNA by bile acids. A: mRNA expression of Lgrs in liver. B: the expression of Lgr4 mRNA in different tissues. C: the expression of Lgr4 mRNA in hepatocytes, macrophages and hepatic stellate cells. D: the expression of Lgr4 mRNA in the liver of BDL mouse. E: the expression of Lgr5 and Lgr6 mRNA in the liver of BDL mouse. F: the expression of Lgr4 mRNA in AML12 cells treated with TCA at the doses of 25, 50, and 100 μmol/L for 24 h. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 4–6, *P < 0.05. BDL, bile duct ligation; TCA, taurocholic acid.

Amelioration of Cholestasis-Induced Liver Injury by Deficiency of LGR4 in Hepatocytes

To examine whether LGR4 alters the progression of cholestatic liver injury, we used the BDL mouse model. Lgr4^LKO mice and Lgr4^flox/flox littermates were subjected to sham operation or ligation of the common bile duct (BDL).

Lgr4 deficiency was evidenced by quantitative RT-PCR and Western blotting (Fig. 2A). BDL significantly induced hepatic necrosis in Lgr4^flox/flox mice. The hepatic necrosis induced by BDL was alleviated in Lgr4^LKO mice (Fig. 2D). This observation was associated with a significant reduction in hepatic weight (Fig. 2B), serum levels of aspartate transaminase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP) (Fig. 2F) in Lgr4^LKO mice relative to Lgr4^flox/flox control. Masson’s trichrome staining revealed that LGR4 deficiency rendered mice resistant to BDL-induced fibrosis (Fig. 2H). Consistently, LGR4 deficiency in hepatocytes significantly attenuated BDL-induced increase in mRNA levels of fibrosis-relevant genes such as α smooth muscle actin (α-SMA), collagen type 1 (Col1a1), and tissue inhibitor of metalloproteinase 1 (Timp1) in liver (Fig. 2I). However, matrix metallopeptidase 9 (Mmp9) demonstrated a significant increase in Lgr4^LKO mice (Fig. 2I). All these results indicate that Lgr4^LKO mice are resistant to cholestatic liver injury.

Figure 2.

Amelioration of BDL induced liver injury by Lgr4 deficiency in hepatocytes. A: expression of Lgr4 mRNA and protein in the liver of Lgr4^LKO mice. B: liver weight/body weight. C: liver weight/body weight of LGR4 deficiency after BDL surgery. D: liver H&E staining and quantitative results of necrosis area. E: liver H&E staining of LGR4 deficiency after BDL surgery. F: serum level of AST, ALT, and ALP. G: serum level of AST, ALT, and ALP of LGR4 deficiency after BDL surgery. H: Masson’s trichrome staining. I: the expression of α-SMA, Col1a1, Timp1, and Mmp9 in liver. J: Masson’s trichrome staining of LGR4 deficiency after BDL surgery. K: expression of Col1a3, Timp1, and Mmp9 in liver of LGR4 deficiency after BDL surgery. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 6–9, *P < 0.05, comparison with Sham-Lgr4^fl/fl; #P < 0.05, comparison with BDL-Lgr4^fl/fl. AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; BDL, bile duct ligation; H&E, hematoxylin and eosin; LGR4, leucine-rich repeat G protein-coupled receptor 4; SMA, smooth muscle actin.

Surprisingly, Lgr4^LKO mice showed a significant increase in hepatic F4/80-positive macrophages (Fig. 3A) and the mRNA levels of inflammatory molecules, including nuclear factor κB (NFκB), tumor necrosis factor-α (Tnfa), interleukin 1β (Il1b), C-X-C motif chemokine ligand 1 (CXCL1), C-X-C motif chemokine ligand 2 (CXCL2), interleukin 10 (Il10), and arginase 1 (Arg1) (Fig. 3B).

Figure 3.

Increase of macrophages and cytokine genes in BDL mice with Lgr4 deficiency in hepatocytes. A: F4/80 immunofluorescent staining and quantitative results of F4/80-positive cells in liver. B: mRNA levels of cytokine genes. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 6, *P < 0.05 comparison with Sham-Lgr4^fl/fl; #P < 0.05, represents the comparison with BDL-Lgr4^fl/fl. BDL, bile duct ligation.

To examine the potential therapeutic relevance of targeting hepatic LGR4, we induced the BDL injury in 6- to 8-wk-old Lgr4^flox/flox male mice and then injected AAV8-TBG-Cre (2 × 10¹¹ vg/mL) via the tail-vein one week later to knock down the LGR4 in hepatocytes. AAV8-GFP virus was used as control. As shown in Fig. 2E, BDL-induced hepatic necrosis was significantly ameliorated in Lgr4^LKO mice (Fig. 2E). Compared with the Lgr4^fl/fl control mice, LGR4 deficiency after BDL surgery also significantly attenuated the BDL injury measured by reduction of liver weight (Fig. 2C), serum levels of AST, ALT, and ALP (Fig. 2G), liver fibrosis detected by Masson’s trichrome staining (Fig. 2J), and mRNA levels of collagen type 3 (Col3) and Timp1 (Fig. 2K). On the other hand, mRNA levels of Mmp9 were significantly upregulated (Fig. 2K). These results indicate that knocking down LGR4 after cholestasis is able to ameliorate cholestatic liver injury. Therefore, hepatic LGR4 is a potential target for the treatment of cholestatic liver injury.

Attenuation of BDL-Induced Increase of Plasma Bile Acids by Deficiency of LGR4 in Hepatocytes

Because the accumulation of bile acids is associated with cholestatic liver injury,(27) we next examined the plasma levels of bile acids. As shown in Fig. 4A, BDL dramatically increased plasma and hepatic total bile acid levels relative to mice with sham surgery. Post-BDL plasma and hepatic levels of total bile acids were significantly reduced in Lgr4^LKO mice. This observation was associated with a significant reduction in plasma total bilirubin level in BDL-Lgr4^LKO mice (Fig. 4A). The reduction of bile acids in BDL-Lgr4^LKO mice may be due to the suppression of bile acids synthesis because LGR4 deficiency significantly reduced mRNA levels of cytochrome P450 family 7 subfamily A member 1 (Cyp7a1), the rate-limiting enzyme in the production of bile acids (Fig. 4B). Other genes relevant to synthesis of bile acids such as cytochrome P450 family 3 subfamily A polypeptide 11 (Cyp3a11), and cytochrome P450 family 8 subfamily B polypeptide 1 (Cyp8b1) also showed a significant reduction in BDL-Lgr4^LKO mice (Fig. 4B).

Figure 4.

Lgr4 deficiency altered bile acids synthesis and transport in hepatocytes. A: total bile acids of liver and serum, total bilirubin of serum. B: mRNA levels of bile acids synthesis. C: mRNA levels of bile acids transports. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 6, * and #P < 0.05, *P < 0.05, comparison with Sham-Lgr4^fl/fl; #P < 0.05, comparison with BDL-Lgr4^fl/fl. BDL, bile duct ligation.

The reduction in bile acids content may also be due to the reduction of cholesterol, the most important substrate in the synthesis of bile acids. As shown in Supplemental Fig. S1, the BDL-induced increase of plasma total cholesterol level was significantly attenuated in Lgr4^LKO mice (Supplemental Fig. S1, A and B). Consistent with our previous report, levels of hepatic cholesterol increased slightly in Lgr4^LKO mice with or without BDL.

In the liver, bile acids are reconjugated and then resecreted together with newly synthesized bile salts to complete one cycle of the enterohepatic circulation. Bile salt export pump (BSEP), an essential transporter for the secretion of bile salts from hepatocytes into the canaliculi, was significantly reduced in BDL-Lgr4^LKO mice (Fig. 4C). The absorbed primary and secondary bile acids are actively transported into hepatocytes by organic anion transporters (OATPs) and sodium (Na⁺)-taurocholate cotransporting polypeptide (NTCP/SLC10A1). The mRNA levels of hepatic OATPs and NTCP were significantly decreased in BDL-Lgr4^LKO mice (Fig. 4C).

All these results suggest that LGR4 in hepatocytes may contribute to pathogenesis of the cholestatic liver injury by altering bile acid synthesis and transport.

No Effect of LGR4 Deficiency in Macrophages on Cholestasis-Induced Liver Injury

To investigate whether the LGR4 of macrophages contributes to BDL-induced cholestatic liver injury, a transgene with myeloid-specific Lgr4 knockdown (Lgr4^MKO) was used. Lgr4^flox/flox littermates were used as control.

As shown in Fig. 5A, a notable reduction in the Lgr4 mRNA levels was observed in the bone marrow-derived macrophages (BMDMs), indicating the deficiency of Lgr4 in these cells. Liver weight of liver demonstrated no significant difference between Lgr4^MKO mice and Lgr4^flox/flox littermates after BDL (Fig. 5B). Neither change was the necrosis (Fig. 5C), fibrosis detected by Masson’s trichrome staining and levels of fibrosis relevant genes (Fig. 5D), and inflammation measured by F4/80-positive cells and cytokine genes (Fig. 5, E and F). These results indicate that LGR4 in macrophages has a negligible effect on BDL-induced cholestatic liver injury.

Figure 5.

No effects of myeloid-Lgr4 knockdown on BDL-induced injury. A: expression of Lgr4 in the liver of Lgr4^MKO mice. B: liver weight-to-body weight. C: liver H&E staining. D: Masson’s trichrome staining and the mRNA level of fibrosis-related genes. E: liver F4/80 immunohistochemical staining and quantitative results of F4/80-positive cells. F: mRNA level of inflammation-related genes. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 3–6, *P < 0.05, ns, not significant. BDL, bile duct ligation; H&E, hematoxylin and eosin.

Rspos-LGR4 Signaling in Hepatocytes Aggravates BDL-Induced Cholestatic Liver Injury via S1PR2-AKT Pathway

To explore the intracellular mechanism underlying the effect of Rspos-LGR4 signaling in hepatocytes on BDL-induced cholestatic liver injury, we explored several signaling pathways and identified S1PR2-AKT as the potential mechanism. Among all five S1PR subtypes, only S1PR2 mRNA levels were significantly decreased in Lgr4^LKO mice after BDL (Fig. 6A, Supplemental Fig. S2, A and B). This result indicates that S1PR2 may be involved in the regulation of bile acids metabolism by Rspos-LGR4 signaling in hepatocytes. This was further supported by the observation that the mRNA levels of S1PR2 were significantly increased by Rspo1 or Rspo3 in cultured hepatocytes treated with TCA (Fig. 6B). Both Rspo1 and Rspo3 significantly potentiated the upregulation of bile acids contents induced by TCA (Fig. 6C). Interestingly, both the peptides suppressed bile acids accumulation under basal condition (Fig. 6D). Moreover, Rspo1/3 significantly potentiated the TCA-induced upregulation in the mRNA expression of genes relevant to bile acids synthesis or transports (Fig. 6, E and F).

Figure 6.

S1PR2 mediates the effects of Rspos-LGR4 signaling in hepatocytes on BDL-induced cholestatic liver injury. A: expression of S1PRs in the liver of sham and BDL mice. B: expression of S1PR2 in primary hepatocytes treated with TCA and Rspo1/3. C: total bile acids of primary hepatocytes treated with TCA and Rspo1/3. D: total bile acids of primary hepatocytes only treated with Rspo1/3. E: mRNA levels of bile acids synthesis in primary hepatocytes treated with TCA and Rspo1/3. F: mRNA levels of bile acids transports in primary hepatocytes treated with TCA and Rspo1/3. G: expression of S1PR2 in primary hepatocytes treated with JTE-013. H: total bile acids of primary hepatocytes treated with TCA, Rspo1/3, and JTE-013. I: mRNA levels of bile acids synthesis in primary hepatocytes treated with TCA, Rspo1/3, and JTE-013. J: mRNA levels of bile acid transports in primary hepatocytes treated with TCA, Rspo1/3, and JTE-013. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 6, *P < 0.05. BDL, bile duct ligation; LGR4, leucine-rich repeat G protein-coupled receptor 4; TCA, taurocholic acid.

Our results also showed that JTE-013, a specific chemical inhibitor of S1PR2, reduced the protein levels of S1PR2 in hepatocytes (Fig. 6G). JTE-013 (10 μmol/L) obviously restrained the effects of Rspo1 and Rspo3 on the TCA-induced upregulation of total bile acids and mRNA levels of genes relevant to bile acids synthesis and transports (Fig. 6, H–J).

As shown in Fig. 7A, BDL-induced cholestatic liver injury significantly increased phosphorylation of AKT (p-AKT) in Lgr4^flox/flox mice. This effect was markedly attenuated by knockdown of LGR4 in hepatocytes (Fig. 7A). On the other hand, treatment of cultured hepatocytes with 400 ng/mL Rspo1 or 200 ng/mL Rspo3 for 6 h significantly increased the phosphorylation of AKT in the presence of TCA (100 μmol/L) (Fig. 7B).

Figure 7.

S1PR2-AKT mediates the effects of Rspos-LGR4 signaling in hepatocytes on BDL-induced cholestatic liver injury. A: p-AKT/AKT in the liver of sham and BDL mice. B: p-AKT/AKT in primary hepatocytes treated with TCA and Rspo1/3. C: total bile acids of primary hepatocytes treated with TCA, Rspo1/3, and MK2206. D: p-AKT/AKT in primary hepatocytes treated with TCA, Rspo1/3, and MK2206. E: mRNA levels of bile acids synthesis in primary hepatocytes treated with TCA, Rspo1/3, and MK2206. F: mRNA levels of bile acids transports in primary hepatocytes treated with TCA, Rspo1/3, and MK2206. Results are presented as means ± SE and analyzed by two-way ANOVA and unpaired Student’s t test, n = 6, *P < 0.05. BDL, bile duct ligation; LGR4, leucine-rich repeat G protein-coupled receptor 4; p-AKT; phosphorylated protein kinase B; TCA, taurocholic acid.

To determine whether AKT signaling contributes to the effect of Rspos-LGR4 axis on bile acids metabolism, MK2206, a specific chemical inhibitor of AKT, was used. As shown in Fig. 7, C and D, MK2206 (2 μmol/L) obviously blocked the potentiating effects of Rspo1 and Rspo3 on the upregulation of total bile acids and p-AKT induced by TCA. Consistently, the potentiation of TCA-induced upregulation in mRNA levels of genes related to bile acids synthesis and transports by Rspo1 or Rspo3 was significantly attenuated by MK2206 in cultured hepatocytes (Fig. 7, E and F). Thus, our results indicate that Rspos-LGR4 signaling in hepatocytes regulates the metabolism of bile acids and contributes to BDL-induced cholestatic liver injury via S1PR2-AKT pathway.

DISCUSSION

Emerging evidence has demonstrated that R-spondins-LGR4 signaling links with liver injury. In the present study, we demonstrated that Rspos-LGR4 in hepatic parenchymal cells aggravates cholestasis-induced liver injury via S1PR2-AKT signaling pathway. Our findings thus reveal a novel function of hepatic LGR4 in BDL-induced liver injury.

Rspos-LGR4 signaling is well recognized as a pathway critical for a variety of physiological and pathophysiological functions such as embryogenesis (15), tumorigenesis(28), and control of energy metabolism (29). Our previous study extends the physiological function for this signaling pathway to the protection of hepatocytes against ischemia-reperfusion injury (20). Interestingly, our present study revealed a harmful action for the R-spondins-LGR4 axis in the cholestatic liver injury. This finding indicates that this signaling pathway may be either beneficial or harmful depending on the pathological conditions. For hepatic ischemia/reperfusion injury, R-spondins-LGR4 functions to protect hepatocytes from hypoxia-induced injury by reducing inflammation (20, 21). It has been well established that macrophages and neutrophils are recruited to the liver in acute liver injury to exacerbate inflammatory and IR-hepatocellular death (30, 31). Interestingly, our studies revealed a significant increase in F4/80-positive macrophages and inflammatory molecules in liver with hepatocytes-specific deficiency of LGR4 with the improvement in liver injury induced by BDL. This observation is in line with previous report showing that macrophages promote the neutrophil-dependent resolution of fibrosis in cholestatic rat liver (32). In addition, bone marrow-derived macrophages and neutrophils have been reported to alleviate fibrosis in the CCL4-induced mouse chronic liver injury (33, 34). Whether the distinct biological function of macrophages is dependent on liver injury mode requires further investigation.

In the condition of cholestatic liver injury, R-spondins-LGR4 may deteriorate the hepatic damage induced by the accumulation of bile acids. This occurs through the potentiating effect of R-spondins-LGR4 on the upregulation of bile acids synthesis induced by BDL. Our finding further supports the notion that bile acids play a key role in cholestasis-induced liver injury. Since bile acids are derived from cholesterol, it is interesting to observe that circulating cholesterol levels were significantly reduced while hepatic cholesterol only demonstrated a slight increase in LGR4-deficient mice. The reduction in plasma cholesterol may limit the substrate supply for bile acids synthesis, thus leading to the reduction of bile acid levels in these mice. Of note, our previous studies suggest that R-spondins-LGR4 signaling inhibits hepatic cholesterol synthesis under physiological conditions (25). Whether this mechanism contributes to the maintenance of steady cholesterol levels in the liver when cholesterol supply is reduced under the cholestasis condition remains to be explored.

Deficiency of LGR4 in macrophages demonstrated no impact on BDL-induced cholestatic liver injury. There was no significant difference in the extent of necrosis, fibrosis, and inflammation induced by BDL between the Lgr4^MKO and Lgr4^flox/flox control mice. In contrast, previous report has shown a direct action of Rspo1-LGR4 signaling on the macrophages in lung tissue (35). Activation of LGR4 by Rspo3 determines interstitial macrophage transition via metabolic-epigenetic reprogramming, leading to subsequent resolution of inflammatory injury (36). Reason underlying this difference remains unclear but may be attributed to the heterogeneities of macrophages between the liver and lung.

Using both pharmacological and genetic approaches, our present study has demonstrated that S1PR2-AKT signaling may mediate the effect of R-spondin-LGR4 on bile acid metabolism in hepatocytes. Genetic deletion of LGR4 in hepatocytes completely blocked the phosphorylation of AKT induced by BDL. Recombinant R-spondin1 and R-spondin3 further augmented TCA-induced increase in phosphorylation of AKT and S1PR2 in cultured hepatocytes. Pharmacological intervention of S1PR2 and AKT reversed the potentiating effect of R-spondin1 or R-spondin3 on the upregulation of total bile acids induced by TCA. Consistently, previous studies have demonstrated that S1PR2-AKT signaling pathway regulates bile acid metabolism (37, 38). Our data thus identifies S1PR2-AKT as a novel signaling pathway mediating the modulation of bile acids metabolism by R-spondin-LGR4 in hepatocytes. Several studies have shown that HNF4α is involved in regulating the transcription of bile acids synthesis and transport genes (39–42). In addition, studies have shown that HNF4α is a downstream transcription factor of PI3K/AKT signaling pathway (43). Therefore, the change of bile acids synthesis and transport genes in this study may be mediated by AKT-HNF4α signaling. R-spondin-LGR4 signaling has been shown to potentiate the Wnt/β-catenin signaling (44). Our previous study has also identified 5′-AMP-activated protein kinase (AMPK) as the downstream target for the R-spondins-LGR4 signaling in hepatocytes. However, our studies showed that phosphorylation of AMPK is unaltered by LGR4 deficiency in our BLD model (Supplemental Fig. S2). All these findings suggest that R-spondin-LGR4 axis may induce multiple intracellular signaling pathways to exert distinct physiological functions.

Limitations still exist for our study. We only constructed the Lgr4^LKO and Lgr4^MKO mice and identified that the LGR4 in hepatic parenchymal cells but not in macrophages alters the BDL-induced liver injury. Although the expression level of LGR4 in liver sinusoidal endothelial cells (LSECs), hepatic stellate cells (HSCs), and other liver cells is relatively low, they may also play a role in cholestatic liver injury, which requires further research. In addition, how Rspos-LGR4 signaling regulates S1PR2 remains to be explored in future studies.

In summary, our study demonstrates that LGR4 in hepatocytes contributes to cholestatic liver injury by regulating the synthesis and transport of bile acids. This occurs through potentiating the bile acids-induced S1PR2 and phosphorylation of AKT. Our finding thus indicates that targeting LGR4 may provide a novel therapy for cholestatic liver diseases.

DATA AVAILABILITY

Data will be made available upon reasonable request.

SUPPLEMENTAL DATA

Supplemental Figs. S1 and S2 and Table S1: https://doi.org/10.6084/m9.figshare.23585274.

GRANTS

This research was supported by grants from the National Natural Science Foundation of China (81930015, 82070592) and National Institutes of Health Grants 1R01DK129360, R01DK112755, and 1R01DK110273.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.G., W.Z., M.W.M., Y.Y., and W.Z. conceived and designed research; Y.G., W.Z., L.S., X.D., and X.W. performed experiments; Y.G. and W.Z. analyzed data; Y.G. and W.Z. interpreted results of experiments; Y.G. and W.Z. prepared figures; Y.G. and W.Z. drafted manuscript; Y.G., W.Z., Y.Y. and W.Z. edited and revised manuscript; Y.G., W.Z., L.S., X.D., X.W., M.W.M., Y.Y., and W.Z. approved final version of manuscript.