Endothelial cell LGR4 maintains liver sinusoidal endothelial cell Identity and protects against hepatic steatosis

Abstract

Aims:

Metabolic dysfunction-associated steatotic liver disease (MASLD), the most prevalent chronic liver pathological condition, develops due to complex interactions among various hepatic cell types. Emerging evidence demonstrates the pivotal role of hepatic endothelium, specifically the liver sinusoidal endothelial cell (LSEC) phenotype, in MASLD pathogenesis. While global ablation of leucine-rich repeat-containing G protein-coupled receptor 4 (LGR4) improved metabolic fitness in mice, its cell-specific deletion revealed differential phenotypes. Here, we investigated the endothelial cell (EC)-specific role of Lgr4 in regulating LSEC phenotype and MASLD development.

Materials and methods:

Control and EC-specific Lgr4 knockout mice were used. Mice were analyzed for metabolic parameters, LSEC phenotype, hepatic steatosis, and potential underlying mechanisms, following Western diet feeding.

Key findings:

Male EC Lgr4-deficient mice exhibited metabolic dysfunction as indicated by increased weight gain, fat mass, fasting blood glucose, and plasma triglycerides. Consistent with increased liver weight, male mice with EC Lgr4 deletion exhibited exacerbated hepatic steatosis and upregulated mRNA levels of scavenger receptors Cd36 and Olr1. However, no differences in fibrosis marker anti-smooth muscle actin were observed. Notably, immunostaining experiments demonstrated enhanced LSEC capillarization, evidenced by reduced LYVE-1 and increased EMCN expression in the livers of EC-restricted Lgr4 knockout mice compared with controls, which was accompanied by increased hepatic inflammation. Similarly, female EC-specific Lgr4-deficient mice had increased hepatic steatosis. Mechanistically, loss of EC Lgr4 in mice promoted expression of apoptotic markers and suppressed AMPK activation.

Significance:

These findings identify EC Lgr4 as an important regulator of hepatic lipid metabolism, LSEC capillarization, and MASLD progression.

Graphical Abstract

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD), which ranges from simple hepatic steatosis to inflammatory and fibrotic events, can progress to cirrhosis and hepatocellular carcinoma, contributing substantially to liver-related morbidity and mortality (1). MASLD is one of the most prevalent liver diseases worldwide, affecting more than 25% of adults, and its incidence is predicted to increase in the future (2). This growing disease burden necessitates newer, effective pharmacotherapies to prevent or even reverse this devastating disease. The major barrier to the development of novel therapeutic strategies is the limited understanding of the cellular and molecular mechanisms governing hepatocyte lipid accumulation and disease progression.

The pathogenesis of MASLD involves coordinated interactions among multiple hepatic cell types, including hepatocytes, liver sinusoidal endothelial cells (LSECs), resident Kupffer cells, and hepatic stellate cells (3). Most studies to date have focused on investigating hepatocyte-centered mechanisms, particularly those related to their metabolic and lipotoxic stress, inflammation, and injury (4–9). In recent years, the critical roles of LSECs have been recognized in the regulation of MASLD development (10–13). LSECs are a specialized type of endothelial cells (ECs) that facilitate the exchange of macromolecules, including lipids, nutrients, and signaling molecules, between the circulation and hepatocytes via fenestrae organized in sieve plates (14, 15). These cells also play a central role in governing hepatic immune surveillance. In chronic liver diseases, including MASLD, LSECs lose their identity by loss of fenestrae and formation of a basal membrane, which compromises their antifibrotic and anti-inflammatory functions (10, 13, 16). Although several signaling pathways and transcription factors regulating LSEC differentiation and identity have been identified (17–20), our knowledge of the endogenous molecular factors and receptors that maintain LSEC identity under hypercholesterolemic conditions is far from complete.

Leucine-rich repeat-containing G protein-coupled receptor 4 (LGR4) is expressed on multiple cell types and organs in both human and murine models. It serves a variety of roles in various organs, including the development and maturation of the eyes, kidneys, reproductive system, and hematopoietic system (21). More recently, LGR4 has been recognized as an important regulator of different pathological conditions, including cancer, bone remodeling, and metabolic disorders such as obesity (22–24). Mechanistically, LGR4 primarily functions via binding to endogenous ligands such as R-spondins (RSPOs) and nidogen-2, thereby activating both β-catenin-dependent (canonical Wnt) and -independent signaling pathways (23, 25). LGR4 gain-of-function mutation associates with an increased risk of obesity in humans, whereas its loss-of-function mutation correlates with weight loss (24, 26), hinting a potential role of LGR4 in the development of MASLD. However, previous studies utilizing hepatocyte-specific Lgr4- and Lgr4/Lgr5-deficient mice have demonstrated that loss of Lgr4 in hepatocytes exacerbates MASLD progression, indicating a protective role for hepatocyte Lgr4 in this liver pathology (27, 28). Since Lgr4 is also expressed in vascular ECs, more specifically in LSECs, no previous studies have investigated the EC-specific role of Lgr4 in hepatic lipid metabolism and MASLD pathogenesis.

In this study, we investigated the role of EC Lgr4 in regulating systemic metabolic fitness and hepatic lipid accumulation under hypercholesterolemic conditions. Using EC-specific Lgr4-deficient mice, we demonstrated that deletion of EC Lgr4 contributes to hepatic steatosis and promotes LSEC capillarization. Mechanistically, EC Lgr4 deficiency increased hepatic inflammation, impaired AMPK signaling, and elevated hepatocyte apoptosis. Collectively, these findings identify EC Lgr4 as a critical regulator of hepatic vascular function and metabolic homeostasis.

Materials and methods

Reagents and antibodies

Pierce BCA protein assay kit (23227), radio-immunoprecipitation assay (RIPA) lysis buffer (8900), protease and phosphatase inhibitor mini tablets (A32927), RevertAid RT Reverse Transcription Kit (K1691), PowerUp SYBR^™ Green Master Mix (A25742), TRIzol Reagent (15596018), Oil Red O powder (ORO, A12989.22), LGR4 (PA5-109908) and Endomucin antibodies (14-5851-85) were procured from Thermo Fisher Scientific (Rockford, IL, USA). Total ACC (3662S), pACC^Ser79 (11818S), pAMPK^Thr172 (2535S), total AMPK (2793S), TNFα (6945T), PARP (9542T), cleaved caspase 9 (20750T), CD45 (70257S), CD31 (77699S), and HSP90 (4874) antibodies were purchased from Cell Signaling Technology, Danvers, MA. Antibodies, including MCT11 (sc-515145), FASN (sc-48357), β-actin (sc-47778), and GAPDH (sc-365062) were obtained from Santa Cruz Biotechnology, Dallas, TX. ASMA (ab7817), gasdermin D (ab209845), pMLKL (ab196436), and LYVE-1 (ab14917) antibodies were bought from Abcam, Cambridge, MA. CD31 (AF3628) antibody was purchased from R&D Systems, Minneapolis, MN. Nitrocellulose membranes (926–31092) and a ready-to-use intercept blocking buffer (927–60001) to block the nitrocellulose membranes were bought from Li-Cor Biosciences (Lincoln, NE, USA). DAPI-containing Fluoromount G mounting medium (0100-20) was purchased from SouthernBiotech, Birmingham, AL. Hematoxylin (22-220-100) and eosin (22-220-104) were procured from Fisher Healthcare, Houston, TX, USA.

Animal experiments

All animal procedures were conducted after getting approval from the UTHSC Institutional Animal Care and Use Committee (protocol #22-0319/25-0594), and in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Lgr4 floxed (Lgr4^f/f) mice were kindly gifted by Dr. Jiqiu Wang (Shanghai Jiao Tong University School of Medicine, Shanghai, China). Cdh5-Cre transgenic mice (The Jackson Laboratory, stock # 006137) were kindly provided by Dr. Daniel M Collier [The University of Tennessee Health Science Center (UTHSC), Memphis, TN, USA] (29). Lgr4^f/f and Cdh5-Cre mice were crossed to generate novel EC-specific Lgr4 knockout mice (Lgr4^f/f Cdh5-Cre, referred to as Lgr4^ΔEC). The genotype of mice was determined (Supplementary Fig. 1) by performing polymerase chain reaction (PCR) utilizing tail genomic DNA samples and genotype-specific primers (Supplementary Table 1). Eight-week-old male and female Lgr4^ΔEC and littermate control Lgr4^f/f (Lgr4^WT) mice were intraperitoneally (i.p.) injected with a recombinant adeno-associated viral vector encoding a mutant form of human proprotein convertase subtilisin/kexin type 9 (pAAV/D374Y-hPCSK9, Addgene, Watertown, MA, #58379, 5 × 10¹¹ genome copies/mouse), and fed a Western diet (WD, Inotiv, Indianapolis, IN, #TD.88137) for 16 weeks to induce hypercholesterolemia and MASLD. All mice were housed in a climate-controlled vivarium with a 12-h light and 12-h dark cycle.

Mice body weights were recorded weekly. Whole-body fat and lean mass were measured in the final week of WD feeding using an Echo-MRI Body Composition Analyzer. Before euthanasia, mice were fasted overnight (16 h) and anesthetized with 3% isoflurane, followed by fasting blood glucose level measurement with Relion Primer Classic Blood Glucose Monitoring System (P/N 784001) using whole blood from the tail. Whole blood (cardiac puncture), liver, epididymal adipose depots (EpiWAT), and other tissues were harvested for subsequent analyses. Plasma total cholesterol levels were quantified using the Amplex Red Cholesterol Assay (Molecular Probes, Eugene, OR, A12216). Plasma/hepatic triglyceride levels were determined according to the manufacturer’s instructions (Wako Pure Chemical Industries, Richmond, VA; 994-02891, 990-02893, 990-22991, and 996-02993). Serum ALT, AST, and total protein levels were quantified using kits and standards manufactured by DiaSys Diagnostic Systems GmbH, Holzheim, Germany, utilizing DiaSys Respons 910 Vet Clinical Chemistry Analyzer.

Immunohistochemistry

For histological and molecular analyses, similar regions of the liver from each mouse were utilized. For histology, hepatic tissue samples were fixed in 4% paraformaldehyde (48 h, Santa Cruz Biotechnology, sc281692), dehydrated in 30% sucrose solution (48 h), embedded in OCT compound (Fisher Healthcare, 4585), and sectioned (7 μm). To determine lipid accumulation, frozen sections were stained with 2% ORO for 10 min at room temperature and counterstained with hematoxylin. Additionally, to visualize hepatic lipid globules and evaluate adipocyte size, paraffin-embedded liver and EpiWAT sections were stained with hematoxylin and eosin (H & E) using standard protocols (5, 30). Images were acquired using an Olympus IX73 microscope. For both liver and adipose tissue analyses, five to six randomly selected microscopic fields per section were imaged and analyzed.

For immunofluorescence studies, liver paraffin sections underwent antigen retrieval in citrate buffer (95–98 °C for 30 min), followed by permeabilization with 0.1% Triton X-100 (10 min), and blocking with 10% goat serum (SouthernBiotech, Birmingham, AL, #0060-01) for 1 h at room temperature. Sections were then incubated with primary antibodies against endomucin (EMCN; 1:100), LGR4 (1:100), CD31 (10 μg/mL, R&D Systems), LYVE-1 (1:100), and CD45 (1:80) overnight at 4 °C. After washing, sections were incubated with appropriate fluorescent secondary antibodies (Life Technologies Corporation, Eugene, OR; A11006, 1:300; A11055, 1:250; and A11037, 1:300) for 1 h at room temperature, washed, and mounted with DAPI-containing Fluoromount-G mounting medium (31). Images were captured using a Zeiss 980 inverted confocal microscope. DAB Substrate Kit (SK-4100) and ImmPRESS HRP Goat Anti-Rabbit IgG Polymer Detection Kit (MP-7451-15) from Vector Laboratories, Newark, CA, were used for staining hepatic sections with CD31 antibody (1:70) according to the manufacturer’s instructions. Images were acquired using an Olympus IX73 microscope. Quantitative image analysis was performed using Image-Pro Plus software, except for adipocyte size quantification. For adipocyte size quantification, the identification of adipocytes and nuclei from H&E-stained WAT sections was performed using an artificial intelligence (AI)-based workflow that consisted of training on the AI model (Apeer, www.apeer.com, Zeiss) and using the resulting model to quantify image data offline (Arivis Vision4D, Zeiss). At least five randomly selected microscopic fields per mouse slide were analyzed.

RNA isolation and quantitative reverse-transcriptase PCR

Total RNA was extracted from mouse liver tissues using TRIzol reagent according to the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized from 1000 ng of total RNA employing RevertAid RT Reverse Transcription Kit. Quantitative real-time PCR (qRT-PCR) was performed using PowerUp SYBR^™ Green Master Mix on a QuantStudio 3 Real-Time PCR System (Applied Biosystems). Relative gene expression was calculated using the 2^−ΔΔCt method, with Gapdh serving as the housekeeping gene. Results are shown as relative mRNA expression normalized to the control group. A list of all qRT-PCR primer sequences is provided in Supplementary Table 2.

Western blotting

Liver tissue lysates were prepared using RIPA lysis buffer supplemented with protease and phosphatase inhibitor cocktail. Equal amounts of protein (15–20 μg per lane) were separated by SDS-PAGE and transferred onto nitrocellulose membranes. Membranes were then blocked using Intercept Blocking Buffer (1 h at room temperature) and incubated overnight at 4 °C with below-mentioned primary antibodies at the indicated dilutions: pACC^Ser79 (1:1000), total ACC (1:1000), pAMPK^Thr172 (1:1000), total AMPK (1:1000), TNFα (1:1000), PARP (1:1000), cleaved caspase 9 (1:1000), FASN (1:500), MCT11 (1:500), pMLKL (1:1000), gasdermin D (1:1000), pAKT (1:1000), AKT (1:1000), HSP90 (1:1000), β-actin (1:2000), and GAPDH (1:2000). Following two washes with TBST buffer, membranes were probed with IR Dye-conjugated secondary antibodies (Li-Cor Biosciences), washed and scanned using the Odyssey DLx Infrared Imaging System (Li-Cor Biosciences). Band intensities were quantified using the NIH ImageJ software.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 10 (GraphPad Software, La Jolla, CA, USA). Data are presented as mean ± SEM. The sample size (n) for each experiment or group is indicated in the respective figure legends. Data normality was assessed using the Shapiro-Wilk test. The statistical tests used for various experiments are specified in the corresponding figure legends. A p-value < 0.05 was considered statistically significant.

Results

Endothelial cell-specific Lgr4 deletion exacerbates metabolic dysfunction in hypercholesterolemic mice

Lgr4 is expressed across multiple organs and cell types, and cell-specific deletion of Lgr4 has been shown to exert differential roles in regulating metabolism (27, 28, 32–34). Among the non-parenchymal hepatic cells, ECs are the major cell population (35). Additionally, analysis of single-cell RNA sequencing data showed that Lgr4 mRNA is expressed in hepatic ECs along with hepatocytes, stromal cells, and Kupffer cells (36) (Fig. 1A). However, the role of Lgr4 present on ECs, which form a barrier between circulating blood and hepatocytes, in regulating metabolic fitness under hypercholesterolemic conditions, has never been investigated. Therefore, we generated novel EC-specific Lgr4-deficient mice (Lgr4^ΔEC) by crossing Lgr4^f/f and Cdh5-Cre mice (Fig. 1B). The genotypes of mice were confirmed by PCR using tail genomic DNA and genotype-specific primers (Supplementary Fig. 1A). In addition, double immunofluorescent staining experiments demonstrated deletion of Lgr4 from CD31-positive blood vessels in the liver of Lgr4^ΔEC mice (Supplementary Fig. 1B). To induce hypercholesterolemia and hepatic steatosis, Lgr4^ΔEC and littermate control Lgr4^WT mice were administered with hPCSK9-AAV8 and fed a WD for 16 weeks (Fig. 1B). As shown in Fig. 1C, male Lgr4^ΔEC mice gained significantly higher weight compared with control mice over the course of WD diet feeding. Consistently, Echo-MRI analysis revealed increased fat mass and decreased lean mass in male Lgr4^ΔEC mice (Fig. 1D and 1E). Moreover, EC-specific Lgr4 deficiency in male mice augmented WD diet-induced liver and EpiWAT weights (Fig. 1F and 1G), which were accompanied by elevated fasting blood glucose and plasma triglyceride levels (Fig. 1H and 1I). However, no differences in plasma total cholesterol levels were observed between the two groups (Fig. 1J). Interestingly, there were no significant differences in body weight gain, fat and lean mass, liver weight, EpiWAT weight, and plasma triglycerides among female mice (Supplementary Fig. 2A to 2F). However, female Lgr4^ΔEC had higher fasting blood glucose levels (Supplementary Fig. 2G). Collectively, these findings suggest that EC Lgr4 plays a beneficial role in maintaining metabolic homeostasis in male mice during hypercholesterolemia.

Figure 1. Endothelial cell-specific Lgr4 deletion exacerbates metabolic dysfunction in hypercholesterolemic mice.

(A) Reanalysis of scRNA-seq data showing annotation of the multiple cell populations and Lgr4-expressing cell types visualized by Uniform Manifold Approximation and Projection (UMAP). (B) The schematic diagram illustrates the experimental design. Eight-week-old male Lgr4^ΔEC and littermate control Lgr4^WT mice were injected with hPCSK9-AAV8, fed a Western diet (WD) for 16 weeks, and analyzed. (C - E) Bar diagrams represent body weight (C), whole-body fat mass (D), and lean mass (E) (n = 7 – 8). (F and G) Representative images of whole livers (F) and EpiWAT (G), scale bar 0.5 cm, and bar graphs for tissue weights. (H - J) Bar graphs representing fasting blood glucose (H), plasma triglycerides (I), and plasma total cholesterol (J) (n = 7 – 8). A two-way ANOVA followed by Sidak post hoc test for multiple comparisons (C), a two-tailed unpaired Mann-Whitney test (D - G), and a two-tailed unpaired t-test (H - J) were used for statistical analyses. Data represent mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001.

Deficiency of endothelial cell Lgr4 promotes hepatic steatosis

As EC-restricted Lgr4 deficiency augmented diet-induced obesity, increased liver weight, and induced plasma triglyceride levels, we next performed phenotypic characterization of livers from these mice. H&E staining of liver sections conducted to assess the size and frequency of lipid droplets demonstrated an increased frequency and size of lipid droplets in Lgr4^ΔEC mice compared with controls (Fig. 2A–2C). Consistently, ORO staining of frozen sections to determine neutral lipid deposition revealed increased ORO-positive areas in the livers of both male and female Lgr4^ΔEC mice (Fig. 2D, 2E, and Supplementary Fig. 3A and 3B). These results are further supported by increased hepatic triglyceride levels in Lgr4^ΔEC mice compared with controls (Fig. 2F). Biochemical analysis using serum samples demonstrated increased ALT and AST levels in Lgr4^ΔEC mice, while total protein quantities were comparable to those of Lgr4^WT mice (Fig. 2G–2I). Additionally, qRT-PCR analysis for mRNA expression of various scavenger receptors and lipid metabolism-related genes showed significantly upregulated levels of fatty acid transporters Cd36 and Olr1 (encodes Lox1), along with reduced expression of Ppara in Lgr4^ΔEC mice (Fig. 2J and 2K). We also performed immunoblotting to determine expression of proteins associated with fatty acid synthesis (fatty acid synthase, FASN), pyruvate transport (monocarboxylate transporter 11, MCT11) (5, 37), and anti-smooth muscle actin (αSMA), a marker of hepatic fibrosis (38). However, no significant differences in the levels of these proteins were observed (Supplementary Fig. 4A–4D). Altogether, these results indicate that the enhanced hepatic steatosis observed in Lgr4^ΔEC mice may be attributed in part to increased hepatocyte lipid uptake via scavenger receptors.

Figure 2. Deficiency of endothelial cell Lgr4 promotes hepatic steatosis.

Male hPCSK9-AAV8-injected, WD-fed Lgr4^WT and Lgr4^ΔEC mice livers were analyzed for hepatic steatosis and mRNA expression of various genes. (A - C) Representative images of liver sections stained with H&E (lipid droplets), scale bar 100 μm (A), and bar diagrams show mean lipid droplet size (B) and frequency of lipid droplets with varying sizes (C) (n = 7 – 8). (D and E) Representative images of ORO-stained liver sections, scale bar 100 μm (D), and the bar diagram represents ORO-positive area (neutral lipid accumulation) (n = 7 – 8). (F) The bar diagram shows hepatic triglyceride levels (n = 7 – 8). (G - I) Bar diagrams showing serum ALT (G), AST (H), and total protein levels (I) (n = 7 – 8). (J and K) Bar diagrams show the relative mRNA expression of various genes involved in fatty acid transport and lipid metabolism in the livers of Lgr4^WT and Lgr4^ΔEC mice (n = 4). A two-tailed unpaired Mann-Whitney test (B, G), a 2-way ANOVA with Sidak multiple comparison test (C), two-tailed unpaired t-test (E, F, H and I), and a multiple unpaired t-test (J and K) were used for statistical analyses. Data represent mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001.

Endothelial cell-restricted Lgr4 deletion promotes LSEC capillarization and hepatic inflammation

LSECs are a specialized type of ECs present in the liver and characterized by discontinuous fenestrae that facilitate the exchange of circulating blood cells, metabolites, and nutrients (20). Further, LSEC-specific genetic deletion has been shown to regulate the pathogenesis of MASLD (39–41), and Cre recombinase driven by the Cdh5 promoter induces gene recombination in LSECs (39, 42), therefore, we next investigated LSEC phenotype in our mouse model. Double immunofluorescent staining performed utilizing LYVE-1 (LSEC marker) and EMCN (endothelial capillarization marker) antibodies, demonstrated reduced LYVE-1 expression along with elevated EMCN⁺ positive areas in the livers of Lgr4^ΔEC mice compared with control Lgr4^WT mice (Fig. 3A to 3C), suggesting increased sinusoidal capillarization in response to EC-specific Lgr4 deletion. These results were further supported by upregulated expression of CD31 in the livers of Lgr4^ΔEC mice (Fig. 3D).

Figure 3. Endothelial cell-restricted Lgr4 deletion promotes LSEC capillarization and hepatic inflammation.

(A - D) Representative images of immunofluorescence/histochemical staining and bar graphs showing LYVE1- (A and B), EMCN- (A and C), CD31- (D), and CD45- (E) positive areas in the livers of Lgr4^WT and Lgr4^ΔEC mice after WD feeding (n = 4 – 6), scale bar 20 μm. (F) The bar diagram represents relative mRNA expression of pro- (Tnfa) and anti-inflammatory (Agr1) genes in the livers of Lgr4^WT and Lgr4^ΔEC mice (n = 4). (G) Representative Western blot images for TNFα and β-actin, performed utilizing liver lysates of WD-fed Lgr4^WT and Lgr4^ΔEC mice (n = 5). A two-tailed unpaired t-test (B, C, D, and E) and a multiple unpaired t-test (F) were used for statistical analyses. Data represent mean ± SEM. * p < 0.05, ** p < 0.01.

Given the central role of inflammation in promoting hepatocyte death and MASLD progression, we investigated immune cell infiltration and hepatic inflammation. Immunostaining experiments demonstrated increased hepatic accumulation of CD45⁺ hematopoietic immune cells in Lgr4^ΔEC mice (Fig. 3E). Consistent with this observation, qRT-PCR data showed increased hepatic expression of pro-inflammatory gene Tnfa and reduced levels of Arg1 in Lgr4^ΔEC mice (Fig. 3F). These qRT-PCR results were further supported by increased TNFα protein levels (Fig 3G). These findings suggest that endothelial cell Lgr4 deficiency associates with LSEC capillarization and enhanced hepatic inflammation.

Lgr4 deficiency in EC suppresses AMPK activation and induces apoptosis in the liver

Since EC-restricted Lgr4 deficiency promoted hepatosteatosis and inflammation, we next investigated the underlying molecular mechanisms by examining the activation status of a key metabolic pathway dysregulated in MASLD. AMP-activated protein kinase (AMPK) serves as a central cellular energy sensor, and regulates fatty acid oxidation and de novo lipogenesis in hepatocytes via regulating phosphorylation of acetyl-CoA carboxylase (ACC) (5, 43, 44). Western blot analysis performed with liver tissue lysates of WD-fed Lgr4^WT and Lgr4^ΔEC mice demonstrated reduced AMPK phosphorylation at the Thr172 residue in Lgr4^ΔEC livers compared to controls, whereas ACC phosphorylation remained unchanged between the two groups (Fig 4A–4C). These results suggest that EC Lgr4 may regulate hepatic lipid metabolism by modulating AMPK activation, independent of affecting ACC phosphorylation. Hepatocyte lipid overload is closely associated with cell apoptosis and contributes to hepatic inflammation (5, 45, 46). Consistently, additional immunoblot experiments showed increased hepatic levels of apoptotic markers, including cleaved caspase-9 and PARP, in Lgr4^ΔEC mice (Fig 4D–4F). Further experiments were performed to examine phosphorylated mixed lineage kinase domain-like protein (p-MLKL), a key marker of necroptosis, and gasdermin D (GSDMD), a marker of pyroptosis. Our immunoblotting data showed no significant differences in p-MLKL expression between Lgr4^f/f control and Lgr4^ΔEC mice (Supplementary Fig. 5A). Interestingly, total GSDMD expression was decreased in the livers of Lgr4^ΔEC mice compared with control mice (Supplementary Fig. 5B), suggesting that necroptosis and pyroptosis are unlikely to be responsible for increased hepatic cell death observed in Lgr4^ΔEC mice and instead support apoptosis as a potential predominant mechanism.

Figure 4. Lgr4 deficiency in EC suppresses AMPK activation and induces apoptosis in the liver.

Immunoblotting experiments were performed utilizing the liver tissue lysates of WD-fed Lgr4^WT and Lgr4^ΔEC mice. (A) Representative WB images. (B and C) Bar diagrams represent protein levels expressed as a ratio of phospho-to-total proteins, AMPK (B) and ACC (C) (n = 5). (D) Representative WB images. (E and F) Bar diagrams show expression of cleaved caspase-9 (E) and PARP (F) (n = 3 – 4). (G) Representative images of H&E staining of EpiWAT isolated from WD-fed Lgr4^WT and Lgr4^ΔEC mice, scale bar 100 μm. (H) The bar diagram represents the mean adipocyte size (n = 6 – 7). A two-tailed unpaired t-test (B, C, E, F, and H) was used for statistical analyses. Data denote mean ± SEM. ns: non-significant. * p < 0.05, **** p < 0.0001.

Given the role of white adipose tissue (WAT) in regulating hepatic steatosis via balancing triglyceride storage and mobilization (5), we determined adipocyte size as a measure of triglyceride storage. Consistent with the observed increase in EpiWAT weight following EC-specific Lgr4 deletion, EpiWAT from these mice exhibited increased adipocyte size compared with control mice (Fig 4G and 4H). Adipogenesis, fatty acid oxidation, and inflammation in WAT are regulated by multiple pathways, with AMPK playing a central role. Further, CREB and AKT activation promote adipocyte differentiation and enhance lipid synthesis in adipose tissue, respectively (5). To determine the role of EC Lgr4 in these pathways, we analyzed the activation status of these proteins in EpiWAT lysates. Interestingly, WB analyses revealed no significant differences in the phosphorylation levels of ACC, CREB, and AKT between the two groups, but AMPK phosphorylation was reduced in Lgr4^ΔEC mice (Supplementary Fig. 6A–6E). These systemic changes in lipid handling may further contribute to the exacerbated hepatic lipid overload observed in the absence of EC Lgr4 protein. Taken together, these findings suggest that EC Lgr4 promotes AMPK activation, limits hepatocyte apoptosis, and suppresses adipocyte hypertrophy, thereby contributing to improved metabolic homeostasis.

Discussion

Despite the information that LSEC capillarization, characterized by the loss of fenestrae and formation of basement membrane, contributes to the development of the most prevalent liver disease, MASLD (10–13, 47), the mechanisms underlying EC dedifferentiation and LSEC capillarization are largely unknown. In the present study, we investigated the role of EC LGR4 in regulating hepatic lipid metabolism, LSEC capillarization, and MASLD progression under hypercholesterolemic conditions using novel EC-specific Lgr4 knockout mice. Our results demonstrate, for the first time, that EC Lgr4 plays a protective role against metabolic dysfunction, LSEC capillarization, and hepatic steatosis. These findings support that EC LGR4 suppresses LSEC capillarization and MASLD development.

Previous studies have linked LGR4 gain-of-function mutations with an increased risk of obesity in humans, whereas its loss-of-function mutation correlates with weight loss (24, 26), hinting a potential role of LGR4 in the development of MASLD, a condition closely associated with weight gain. Consistently, global ablation of Lgr4 in mice has been reported to promote energy expenditure and induce brown-like adipocytes in WAT depots (48). In contrast, hepatocyte-specific Lgr4- and Lgr4/Lgr5 deletion in mice has been shown to exacerbate MASLD progression, indicating a protective role for hepatocyte Lgr4 against this disease (27, 28). Given that Lgr4 is also expressed in hepatic ECs, which are the major non-parenchymal hepatic cells, we investigated the specific role of EC Lgr4 in MASLD pathogenesis. In our study, WD-fed Lgr4^ΔEC mice exhibited metabolic dysfunction, as evidenced by increased body weight, elevated fat mass, reduced lean mass, higher fasting blood glucose, and increased plasma triglyceride levels. These data suggest that EC Lgr4 helps in regulating energy homeostasis and hepatic lipid metabolism, and EC-specific Lg4 deficiency promotes lipid storage. Further, elevated plasma triglyceride levels observed in Lgr4^ΔEC mice may result from increased de novo lipogenesis or impaired clearance of triglyceride-rich lipoproteins (49–51). However, since no differences were noted in Cpt1a and FASN protein expression in the livers of control and EC-restricted Lgr4 knockout mice in the present study, it is unlikely that translational regulation of key proteins involved in de novo lipogenesis or mitochondrial β-oxidation is responsible for the observed phenotype. Instead, impaired triglyceride clearance, potentially due to reduced lipoprotein lipase activity/expression, may contribute to elevated plasma triglyceride levels (52, 53). Moreover, increased EpiWAT mass without any changes in metabolic genes may reflect defects in energy expenditure. Therefore, in future studies, it will be interesting to investigate energy expenditure and substrate utilization in WD-fed mice with EC Lgr4 deficiency, utilizing the Comprehensive Laboratory Animal Monitoring System (5). Additionally, since global and intestinal ablation of Lgr4 in mice reduced adiposity, while EC- (present study) and hepatocyte-specific deletion promoted adiposity and MASLD, these data suggest the tissue-specific roles of Lgr4 in metabolic disease (27, 28, 32, 48). Further investigation into the function of Lgr4 in other cell types involved in systemic metabolism will be important to fully understand its context-dependent effects.

Lgr4^ΔEC mice displayed markedly enhanced steatosis, which was accompanied by upregulated expression of key scavenger receptors involved in fatty acid and oxidized LDL uptake, Cd36 and Olr1 (encodes Lox1) (54, 55), indicating increased lipid uptake by hepatocytes may be responsible for the observed steatotic effects. These observations are consistent with previous reports demonstrating that hepatic Cd36 overexpression promotes free fatty acid uptake by hepatocytes, leading to triglyceride accumulation and steatosis in mice (54, 56). However, it remains to be investigated how Lgr4 present on hepatic ECs, regulates the expression of various lipid metabolism-related genes/proteins in hepatocytes. To address this, co-culture studies using control/Lgr4-deficient LSECs treated with/without free fatty acids (FFA), in combination with hepatocytes, will be important to examine the role of LSEC Lgr4 in modulating the transcriptome of hepatocytes.

LSECs function as a permeable barrier due to the presence of fenestrae, which allow transfer of FFA from the bloodstream into hepatocytes, where very low-density lipoprotein (VLDL) is synthesized and secreted into the circulation through LSECs (57). LSEC dedifferentiation, characterized by the loss of LSEC-specific proteins, including LYVE1 and Stab2, along with increased expression of vascular EC markers (EMCN, CD34, and CD31), is a hallmark of LSEC capillarization and contributes to MASLD development (10–13, 47). A key finding of the present study is that EC-specific Lgr4 deletion induces loss of LSEC identity, evidenced by reduced LYVE1 expression and increased EMCN levels. These findings suggest that EC Lgr4 is an important regulator of LSEC differentiated phenotype and a suppressor of capillarization. These capillarized LSECs may promote hyperlipidemia by blocking FFA entry into hepatocytes, and prevent the release of VLDL from the liver into the bloodstream, thereby promoting hepatic steatosis (12). Additionally, LSEC fenestrae are dynamic structures regulated by continuous actin remodeling, it is possible that EC Lgr4 deficiency disturbs the F-actin/G-actin ratio, which is crucial for maintaining LSEC fenestrae number (12, 58). Besides, the engagement of various angiocrine signals, including nitric oxide, Wnt ligands, or growth factors, cannot be excluded (59). However, we acknowledge that our data do not establish whether the observed changes in the liver are mediated directly by endothelial cell-derived paracrine factors or indirectly through changes in the hepatic sinusoidal microenvironment. Future co-culture studies with endothelial cells with/without Lgr4 deficiency and hepatocytes, as well as experiments examining the effects of endothelial cell-derived conditioned media on hepatocyte metabolism, are required to provide further insights on endothelial cell-hepatocyte crosstalk. Notably, Lgr4 is a well-known regulator of Wnt-β-catenin signaling, which plays a pivotal role in liver zonation and regeneration (60). Lgr4^ΔEC mice also displayed increased hepatic accumulation of immune cells and elevated expression of a pro-inflammatory cytokine TNFα, likely as a consequence of impaired LSEC barrier function (61). LSECs play an important role in regulating hepatic immune homeostasis and leukocyte trafficking (62). In addition to functioning as specialized sinusoidal barrier cells, LSECs regulate interactions between circulating immune cells and the hepatic microenvironment through the expression of adhesion molecules, chemokines, and inflammatory mediators (63). Loss of the LSEC phenotype may therefore disrupt these homeostatic functions, thereby facilitating leukocyte adhesion and recruitment and contributing to hepatic inflammation (63). Taken together, these results support a model in which EC-specific Lgr4 loss may disrupt LSEC-hepatocyte crosstalk, thereby exacerbating hepatic lipid accumulation and inflammation, ultimately contributing to increased MASLD severity.

AMPK serves as a central cellular energy sensor that promotes fatty acid oxidation while suppressing de novo lipogenesis, partly through the phosphorylation (Ser79) and inactivation of ACC, which facilitates CPT1-mediated mitochondrial fatty acid uptake (5, 30, 64, 65). Reduced AMPK activity is a well-recognized feature of MASLD pathogenesis, and pharmacological activation of AMPK has been shown to inhibit inflammation and attenuate hepatic steatosis (5, 65, 66). Notably, our data revealed reduced AMPK phosphorylation at Thr172 residue in the livers of Lgr4^ΔEC mice, whereas ACC phosphorylation remained unchanged. These results suggest that EC Lgr4 regulates hepatic metabolism by AMPK-dependent yet ACC-independent mechanisms. AMPK has also been shown to regulate metabolism by inhibiting 3-hydroxy-3-methylglutaryl-CoA reductase and protein kinase A in the liver. However, the role of EC Lgr4 in governing these pathways remains to be explored. Further, increased expression of apoptotic markers with EC-restricted Lgr4 deletion suggests that EC Lgr4 protects hepatocytes from lipotoxic cell death, potentially by maintaining intrahepatic metabolic balance and reducing inflammatory stress. Given the critical role of adipose tissue-liver crosstalk in metabolic homeostasis, it is important to consider the contribution of adipose tissue. Under physiological conditions, WAT serves as the primary site for lipid storage; however, dysfunctional lipid handling in WAT promotes excessive delivery of FFA to the liver and stimulates hepatic lipid deposition (67, 68). Consistent with exacerbated hepatic steatosis, adipocyte hypertrophy was observed in EpiWAT of Lgr4^ΔEC mice. These enlarged adipocytes may therefore represent both a direct metabolic consequence of EC Lgr4 deficiency and a secondary driver of hepatic steatosis by increasing FFA delivery to the liver.

In conclusion, our study using a novel EC Lgr4-deficient mouse model demonstrates that EC Lgr4 functions as a critical regulator of hepatic lipid metabolism, LSEC identity, and MASLD development. These findings warrant future research investigations to determine whether EC Lgr4 overexpression enhances LSEC porosity and protects against diet-induced or age-related metabolic syndrome and associated cardiovascular complications.

Highlights.

• Endothelial cell-specific Lgr4 deletion exacerbates metabolic dysfunction in hypercholesterolemic mice.

• Deficiency of endothelial cell Lgr4 promotes hepatic steatosis.

• Endothelial cell Lgr4 deletion promotes LSEC capillarization and hepatic inflammation.

• Endothelial cell Lgr4 deficiency suppresses AMPK activation and induces apoptosis in the liver

Data availability statement:

The data associated with this study are available upon reasonable request from the corresponding author.