Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Abstract

Objective

This study investigates the molecular mechanism by which leucine (Leu) ameliorates collagenase-induced osteoarthritis (CIOA) through macrophage polarization regulation.

Methods

A CIOA mouse model was established and evaluated by micro-computed tomography (micro-CT) and histopathological analysis. Leu intervention was administered, and its therapeutic effects on cartilage degeneration and osteophyte formation were assessed. Integrated multi-omics analyses and mechanistic assays were performed to explore the role of the mTORC1/LXRα pathway in synovial macrophage reprogramming and its regulation of the Rspo2/β-catenin axis in chondroprogenitors. Functional validation was conducted using the LXRα inhibitor GSK2033.

Results

Leu intervention demonstrated significant therapeutic effects, reducing cartilage degeneration by 42% (Osteoarthritis Research Society International [OARSI] score) and osteophyte formation by 58% (volume reduction). Integrated multi-omics and mechanistic assays indicated that Leu activated mTORC1/LXRα to reprogram synovial macrophages toward an M2-like state, suppressed Rspo2, and attenuated β-catenin signaling in chondroprogenitors, thereby improving cartilage function. Functional validation using LXRα inhibitor GSK2033 confirmed pathway specificity, reversing Leu-mediated cartilage protection and reactivating osteogenic differentiation.

Conclusion

These findings establish a novel “metabolism-immunity-cartilage” axis in which Leu coordinates mTORC1/LXRα-driven macrophage reprogramming with Rspo2/β-catenin axis suppression, offering dual-target therapeutic potential for osteoarthritis. The study redefines nutritional amino acids as immunometabolic modulators in degenerative joint diseases, proposing Leu supplementation as a viable strategy for interrupting the inflammation-bone remodeling cycle in traumatic arthritis. No clinical trials were involved in this preclinical investigation.

Graphic Abstract

Schematic illustration of leucine-mediated modulation of macrophage polarization via the mTORC1/LXRα pathway and its downstream regulation of Rspo2/β-catenin signaling to alleviate OA-induced cartilage degeneration.

Keywords

leucine macrophage polarization mammalian target of rapamycin complex 1/liver X receptor α signaling pathway R-spondin 2 β-catenin signaling pathway

Introduction

Collagenase-induced osteoarthritis (CIOA) is a prevalent degenerative joint disease following articular trauma. Globally, over 30 million new osteoarthritis (OA) cases are diagnosed annually. Notably, 30% of these patients progress to end-stage arthritis within a decade post-injury.^1
-3 The characteristic pathological features include progressive degradation of the cartilage matrix, subchondral bone sclerosis, and osteophyte formation at the joint margins, collectively leading to joint dysfunction and chronic pain.^4
-6 Current clinical management primarily relies on non-steroidal anti-inflammatory drugs (NSAIDs) and intra-articular (IA) corticosteroid injections for symptom control. These approaches provide only transient relief and fail to arrest or reverse cartilage degeneration.^7
-9 Recent studies have highlighted the pivotal role of abnormal synovial macrophage polarization in CIOA progression. Pro-inflammatory M1 macrophages exacerbate joint inflammation through the secretion of cytokines such as interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α), which activate matrix metalloproteinases (MMPs), whereas impaired M2 macrophage function compromises tissue repair.^10
-12 Although modulating macrophage polarization has been shown to improve cartilage metabolism,^10,13,14 the specific metabolic interventions required for precise regulation remain a significant scientific challenge.

Leucine (Leu), an essential branched-chain amino acid, has recently been recognized for its biological functions beyond nutritional support.¹⁵ By activating the Mammalian Target of Rapamycin Complex 1 (mTORC1), Leu regulates cellular metabolic reprogramming and exhibits dual immunomodulatory effects, promoting regulatory T cell differentiation while enhancing the antimicrobial capacity of macrophages.^16
-18 Liver X receptor α (LXRα), a master regulator of cholesterol metabolism, interacts with mTORC1 to influence immune cell metabolic phenotypes.^19
-21 For instance, in atherosclerosis models, LXRα agonists synergize with mTORC1 inhibitors to modulate macrophage lipid metabolism.^22
-25 However, the collaborative role of these pathways in regulating macrophage polarization within the osteoarthritic microenvironment remains unexplored. Concurrently, aberrant activation of the R-spondin 2 (Rspo2)/β-catenin signaling pathway has been shown to promote the differentiation of chondroprogenitor cells into osteoblasts, accelerating osteophyte formation.^12,26,27 Single-cell sequencing data reveal that Rspo2 expression in CIOA synovial tissue is 3.2-fold higher than in normal tissue. Nevertheless, the upstream regulatory mechanisms, particularly the influence of macrophage-derived factors, remain largely unknown.

Traditional research has predominantly relied on single-omics technologies to analyze specific molecular pathways, limiting the ability to fully characterize the dynamic regulatory networks underlying macrophage polarization and cartilage degeneration. For instance, conventional RNA sequencing (RNA-seq) can capture changes in gene expression profiles but fails to identify key driver modules.^28,29 Similarly, proteomics can detect protein expression levels but lacks temporal dynamic information.³⁰ In recent years, multi-omics integration technologies have emerged as a promising approach to overcome these limitations. Weighted gene co-expression network analysis (WGCNA) enables the identification of gene modules that are highly correlated with specific phenotypes, while machine learning algorithms (e.g., LASSO regression, Support Vector Machine) can screen key regulatory factors from vast datasets.^31
-33 Our previous studies have demonstrated that protein-protein interaction (PPI) analysis can effectively reveal topological associations between the mTORC1 and LXRα pathways, suggesting that these intersecting nodes may serve as critical hubs in metabolic-immune regulation. Furthermore, the development of three-dimensional organoid co-culture models has enabled the simulation of macrophage-chondrocyte interactions in vitro, providing a more physiologically relevant experimental platform for mechanistic studies.

This study aims to determine whether Leu can mitigate CIOA by modulating the local immune microenvironment in joints. Given the critical role of macrophages in synovial inflammation and cartilage degradation, this study focuses on elucidating the mechanism through which Leu regulates macrophage polarization via the mTORC1/LXRα signaling axis. This study also examines its downstream effects on the Rspo2/β-catenin signaling axis and the functionality of chondroprogenitor cells. A recent study reported that Leu modulates macrophage polarization via the mTORC1/LXRα axis to alleviate cytokine storm syndrome (CSS),²¹ but its role in chronic degenerative diseases remains unknown. Compared with the systemic immunosuppressive strategy proposed in the CSS study, the present study introduces the novel concept of “local joint metabolic–immune microenvironment remodeling,” redefining Leu’s function in joint disorders from an immunometabolic reprogramming perspective. Although clinical OA is dominated by chronic low-grade inflammation (LGI), the CIOA model displays an acute inflammatory peak and subsequent regression, in contrast to the progressive, chronic course of destabilization of the medial meniscus (DMM) models.^34
-36 Thus, our findings provide mechanistic evidence in an acute context and motivate a testable hypothesis for chronic LGI, which we explicitly treat as hypothesis-generating to be validated in long-duration models.

Materials and Methods

Establishment of the Collagenase-Induced Traumatic Osteoarthritis Model

The CIOA mouse model was established using collagenase induction. Eight-week-old male C57BL/6 mice (Cat. No. 219, Beijing Vital River Laboratory Animal Technology Co., Ltd., China) were used for the experiments. Each mouse received an IA injection of 1 U collagenase (Cat. No. C0130, Sigma-Aldrich, USA) dissolved in 50 μL sterile phosphate-buffered saline (PBS; Cat. No. P4417, Sigma-Aldrich, USA) into the right knee joint cavity, administered on two consecutive days, with a 24-hour interval between injections. After injections, mice were housed under standard conditions (22 ± 2°C, 12-hour light/dark cycle) for 7 days to allow model establishment. The sham group underwent only skin incision of the right knee without collagenase injection.³⁷ All animal experiments were approved by the Institutional Animal Ethics Committee.

Leu treatment efficacy experiment: Mice were randomly assigned to four groups: sham group (normal mice, no treatment), model group (CIOA-induced mice), low-dose Leu group (CIOA + 50 mg/kg Leu, Cat. No. L8912, Sigma-Aldrich, USA, intraperitoneally once daily), and high-dose Leu group (CIOA + 100 mg/kg Leu, intraperitoneally once daily). Intraperitoneal administration was chosen to ensure accurate dosing and consistent systemic exposure. All injections were performed under sterile conditions, and body weight and general health status were monitored throughout the treatment period.³⁸

GSK2033 intervention to reverse Leu treatment effects: Mice were randomly assigned to four groups (n = 6 per group): sham, model (CIOA), Leu (CIOA + high-dose Leu), and Leu + GSK2033 (CIOA + high-dose Leu + GSK2033). The sham group received an equivalent volume of saline. Leu was administered intraperitoneally at 100 mg/kg daily for 4 weeks. GSK2033 (MedChemExpress, Cat. No. HY-108688) was administered intraperitoneally at 30 mg/kg three times per week for 4 weeks.³⁹

Isolation of Mouse Macrophages

Following euthanasia of 6–8-week-old C57BL/6J mice by CO₂ asphyxiation, the femurs and tibias were harvested, and bone marrow was flushed with PBS to obtain a bone marrow cell suspension. The suspension was centrifuged at 300 × g for 5 min to pellet the bone marrow cells. For macrophage differentiation, bone marrow–derived cells were plated in culture dishes and maintained in αMEM medium (Gibco, 12571063) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin, with the addition of 10 ng/mL macrophage colony-stimulating factor (M-CSF) for 7 days. Adherent cells were collected and transferred to new culture dishes for subsequent experiments.

Culture and Treatment of ATDC5 Chondroprogenitor Cells

ATDC5 cells (Procell, Cat. No. CL-0856) were cultured in DMEM/F-12 medium (Gibco, Cat. No. 11330-032) supplemented with 10% FBS (Gibco, Cat. No. 10099-141), 1% penicillin–streptomycin (Gibco, Cat. No. 15140-122), and 10 μg/mL insulin (Sigma-Aldrich, Cat. No. I9278). Cells were maintained at 37°C in a humidified 5% CO₂ incubator and passaged at a 1:3 ratio every 2–3 days. When 70–80% confluent, cells were allocated into three groups: control (10% conditioned medium [CM] from normal macrophages), model (10% CM from lipopolysaccharide [LPS]-stimulated macrophages), and Leu-treated (10% CM from M1 macrophages treated with Leu). Treatments lasted 48 h before subsequent analyses.

RSPO2 protein stimulation experiment: Cells were assigned to a control group (equal volume of PBS) or an RSPO2-treated group (recombinant human RSPO2 protein, R&D Systems, Cat. No. 3266-RS-025, final concentration 200 ng/mL). Treatments were applied for 48 h before further analyses.

Macrophage Polarization and Drug Treatment

To determine the appropriate concentration of GSK2033, 2.5 × 10⁵ bone marrow–derived macrophages (BMDMs) were plated in 12-well plates and treated with GSK2033 at concentrations ranging from 0.5 to 5 μmol/L, with DMSO as the vehicle control.

For M1 polarization, 2.5 × 10⁵ BMDMs were stimulated with 100 ng/mL LPS (Sigma-Aldrich, Cat. No. L4391) and 20 ng/mL interferon-gamma (IFN-γ; PeproTech, Cat. No. 315-05) for 6 h, and then divided into four groups: control, M1 polarization (LPS + IFN-γ), M1 + Leu (LPS + IFN-γ + Leu), and M1 + Leu + GSK2033.

For M2 polarization, cells were stimulated with 20 ng/mL IL-4 (PeproTech, Cat. No. 214-14) for 24 h and divided into four groups: control, M2 polarization (IL-4), M2 + Leu (IL-4 + Leu), and M2 + Leu + GSK2033. GSK2033 is a small-molecule inhibitor of LXRα. Leu was used at 200 μg/mL, and GSK2033 at 2 μmol/L. Both were added to the medium 1 h prior to LPS or IL-4 stimulation.

Conditioned medium (CM) from each group was collected, centrifuged at 300 × g for 5 min to remove debris, filtered through a 0.22 μm membrane (Millipore, Cat. No. SLGP033RS), aliquoted, and stored at –80°C.

To verify the role of mTORC1 in Leu-mediated macrophage polarization, cells were treated with Leu in the presence or absence of the mTORC1-specific inhibitor rapamycin (MedChemExpress, Cat. No. HY-10219). After 6 h of treatment, cells were collected for real-time quantitative polymerase chain reaction (RT-qPCR) analysis of M1 marker genes (iNOS, TNF-α).

RNA Extraction and Sequencing

BMDMs were differentiated and divided into two groups: M1-polarized, LPS + IFN-γ, and M1-polarized + Leu (LPS + IFN-γ + Leu). After treatment, total RNA was extracted (n = 6 per group) using TRIzol reagent (Cat. No. 15596018, Invitrogen, USA). RNA quality was assessed using the Eukaryote Total RNA Nano assay (Agilent Technologies, USA), and samples with an RNA integrity number (RIN) ≥ 7.0 were considered suitable for further analysis. Sequencing libraries were prepared with the Illumina TruSeq RNA Sample Preparation Kit (Cat. No. RS-122-2001, Illumina, USA) and sequenced on the Illumina NextSeq 500 platform (Illumina, USA) to generate 150 bp paired-end reads.

Screening of Differentially Expressed Genes (DEGs)

Raw sequencing data quality was assessed using FastQC (v0.11.9). Reads were aligned to the GRCm39 reference genome with HISAT2 (v2.2.1), and gene expression levels were quantified using featureCounts (v2.0.1). Differential expression analysis was performed using DESeq2 (v1.38.3), with genes considered differentially expressed if the adjusted p-value (p.adjust) was < 0.05 and the absolute log₂ fold change (|log₂FoldChange|) exceeded 1.

Functional Enrichment Analysis Using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG)

DEGs were analyzed for functional enrichment using the R package clusterProfiler (v4.6.0). GO analysis included three annotation categories: Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). KEGG analysis was conducted to identify significantly enriched pathways. For both analyses, significance was defined as p.adjust < 0.05. Visualization was performed using ggplot2 (v3.4.0) and enrichplot (v1.18.0). GO enrichment results were presented as bar plots of significantly enriched functional terms, whereas KEGG enrichment results were displayed as bubble plots, indicating the number of enriched genes and the corresponding significance levels (p-values) for each pathway.

RT-qPCR

Total RNA was extracted from cells and tissues using TRIzol reagent (Invitrogen, Cat. No. 15596026). RNA concentration and purity were measured with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). cDNA was synthesized using the iScript cDNA Synthesis Kit (Bio-Rad, Cat. No. 1708891). Quantitative real-time PCR was performed using SYBR Green PCR Master Mix (Applied Biosystems, Cat. No. 4367659) on an ABI 7500 Real-Time PCR System (Applied Biosystems). Relative gene expression was calculated using the 2^-ΔΔCt method. Primer sequences are listed in Suppl. Table S1 .

Co-Immunoprecipitation (Co-IP)

Total protein was extracted using radio-immunoprecipitation assay (RIPA) lysis buffer (Thermo Fisher Scientific, Cat. No. 89900). The lysate was incubated overnight at 4°C with anti-mTOR antibody (Cell Signaling Technology, Cat. No. 2972, 1:1000 dilution) or normal rabbit IgG (1:50, Cell Signaling Technology, Cat. No. 2729) as a control. Protein A/G magnetic beads were then added and incubated for 4 h, followed by five washes with lysis buffer and centrifugation (3,000 × g, 5 min) at 4°C. Bead-bound proteins were eluted by boiling at 95°C and analyzed by Western blot.

Nuclear and Cytoplasmic Fractionation Assay

Nuclear and cytoplasmic proteins were isolated using the NE-PER™ Nuclear and Cytoplasmic Extraction Reagents kit (Thermo, Cat# 78833, link). Briefly, cells were collected, washed twice with cold PBS, and centrifuged at 300 × g for 5 min at 4 °C. The pellet was resuspended in 200 μL of CER I, pipetted thoroughly, and incubated on ice for 10 min, followed by the addition of 11 μL CER II. After vortexing for 5 s and incubating on ice for 1 min, samples were centrifuged at 12,000 × g for 5 min, and the supernatant was collected as the cytoplasmic fraction. The pellet was resuspended in 100 μL NER, incubated on ice for 40 min with vortexing for 15 s every 10 min, and centrifuged at 12,000 × g for 10 min to obtain the nuclear protein fraction.⁴⁰ Western blot analysis was subsequently performed.

Western Blot Analysis

Total proteins were extracted from cultured cells and synovial tissues using RIPA lysis buffer (Thermo Fisher Scientific, Cat# 89900). Protein concentrations were determined using the bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific, Cat# 23225). Equal amounts of protein (30 μg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Cat# IPVH00010). Membranes were blocked with 5% non-fat dry milk in TBST for 1 h at room temperature (RT) and incubated overnight at 4 °C with the following primary antibodies: p-STAT6 (Cell Signaling Technology, Cat# 56554, 1:1000), STAT6 (Cell Signaling Technology, Cat# 5397, 1:1000), p-mTOR (Cell Signaling Technology, Cat# 5536, 1:1000), mTOR (Cell Signaling Technology, Cat# 2972, 1:1000), LXRα (Abcam, Cat# ab176323, 1:1000), β-catenin (Cell Signaling Technology, Cat# 8480, 1:1000), Cyclin D1 (Cell Signaling Technology, Cat# 2978, 1:1000), MMP13 (Abcam, Cat# ab39012, 1:1000), and β-actin (Proteintech, Cat# 20536-1-AP, 1:5000). After washing, membranes were incubated with HRP-conjugated anti-rabbit secondary antibody (Cell Signaling Technology, Cat# 7074, 1:5000) for 1 h at RT. Bands were detected using an enhanced chemiluminescence (ECL) kit (Thermo Fisher Scientific, Cat. No. 32106) and imaged with the ChemiDoc MP Imaging System (Bio-Rad). Densitometric analysis was performed using ImageJ software (v1.53, NIH, USA).

Immunofluorescence (IF) Staining for Macrophage Polarization

To evaluate changes in M1 and M2 macrophage populations in synovial tissues, IF staining was performed. Synovial tissue sections were fixed with 4% paraformaldehyde (Sigma-Aldrich, Cat# F8775, USA) and stained with rabbit anti-F4/80 antibody (Abcam, Cat# ab6640, UK, 1:200) and rabbit anti-iNOS antibody (Abcam, Cat# ab178945, UK, 1:500) to identify M1 macrophages, followed by Alexa Fluor 488-conjugated secondary antibody (Abcam, Cat# ab150077, UK, 1:400). To detect M2 macrophages in LPS-stimulated BMDMs, rabbit anti-CD206 antibody (Abcam, Cat# ab64693, UK, 1:500) was used, followed by Alexa Fluor 594-conjugated secondary antibody (Abcam, Cat# ab150080, UK, 1:400). After secondary antibody incubation, nuclei were counterstained with DAPI (Solarbio, Cat# C0060, China). Images were acquired using a confocal microscope (TCS SP8, Leica Microsystems, Germany), and the proportion of positive cells was quantified.

Immunohistochemistry (IHC)

Paraffin-embedded synovial tissue sections were deparaffinized, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked with 3% hydrogen peroxide (Sigma-Aldrich, Cat# H1009, USA), and nonspecific binding was blocked with 5% bovine serum albumin (Sigma-Aldrich, Cat# A9647, USA). Sections were incubated overnight at 4 °C with primary antibodies against CD206 (Abcam, Cat# ab64693, UK) or RSPO2 (R&D Systems, Cat# AF3474, USA), followed by HRP-conjugated secondary antibody (Sigma-Aldrich, Cat# A6154, USA) for 1 h at RT. Staining was developed using a DAB kit (Abcam, Cat# ab64238, UK). Positive staining was observed under a light microscope (Leica Microsystems, Germany), and semiquantitative analysis was performed using ImageJ software.

Toluidine Blue Staining

ATDC5 cells (3 × 10³ per well) were seeded into 96-well plates. After culture and treatment, cells were fixed with 4% paraformaldehyde for 20 min, stained with 0.1% toluidine blue at RT for 20 min, and rinsed with distilled water. Staining was observed and imaged under a light microscope (Leica DMI 3000 M, Germany) to assess cartilage matrix secretion.

Enzyme-Linked Immunosorbent Assay (ELISA)

Cell culture supernatants or mouse serum were collected, and levels of TNF-α (R&D Systems, Cat# MTA00B), IL-6 (R&D Systems, Cat# M6000B), IL-1β (R&D Systems, Cat# MLB00C), IL-10 (R&D Systems, Cat# M1000B), TGF-β (R&D Systems, Cat# MB100B), and mouse RSPO2 (Abbexa, Cat# abx258663) were measured using ELISA kits. Absorbance was read at 450 nm using a microplate reader (iMark, Bio-Rad).

Histological Analysis

Mouse right knee joints were fixed in 10% neutral-buffered formalin (Solarbio, Cat# G2161) for 48 h, decalcified in 10% EDTA (pH 7.4, Solarbio, Cat# E1170) at 4 °C for 4 weeks (solution changed every 3 days), and embedded in paraffin. Sections (5 μm) were cut using a microtome (Leica RM2235, Germany) and stained with Safranin O-Fast Green and hematoxylin–eosin (H&E). For Safranin O-Fast Green staining, sections were stained with 0.1% Safranin O (Sigma-Aldrich, Cat# S2255) for 5 min, rinsed, stained with 0.05% Fast Green (Sigma-Aldrich, Cat# F7258) for 10 min, differentiated in 1% acetic acid for 5 s, dehydrated, cleared, and mounted. For H&E staining, sections were stained with hematoxylin (Solarbio, Cat# H8070) for 5 min, differentiated in 0.3% acid ethanol for 3 s, blued in tap water, stained with 0.5% eosin (Solarbio, Cat# E6070) for 3 min, dehydrated, cleared, and mounted. Images were captured under a light microscope (BX53, Olympus, Japan). Cartilage degeneration was scored using the OARSI scoring system, and synovitis severity was evaluated using the Synovitis Score.

Micro-CT Scanning and Osteophyte Volume Measurement

Mouse hind limbs were harvested, soft tissues removed without damaging cartilage surfaces, and samples kept hydrated in 70% ethanol. Samples were stained with 1% phosphotungstic acid (PTA) in 70% ethanol for 22 h, wrapped with sealing film, and mounted on a dental wax holder. High-resolution micro-CT scanning (Bruker 1272) was performed at 50 kV, 200 μA, with a 0.5 mm Al filter, acquiring 1241 projections over 360° rotation, 2200 ms exposure per projection, averaging 3 frames, with a final resolution of 2.8 μm/pixel (scan time ~2.5 h). 3D reconstruction was performed using NRecon software (v1.7.5.9), with isotropic voxel size set to 2.8 μm. Osteophyte volume and morphology were analyzed using CTAn software (Bruker, Belgium). Histological scoring of osteophytes was performed on 3D-reconstructed images by two blinded investigators using a 0–6 scale (0 = no osteophyte; 6 = severe osteophyte), and the mean score was reported.

WGCNA

WGCNA was conducted using the WGCNA R package (v1.71) to analyze RNA-seq gene expression data. Raw counts were normalized, and low-expression genes were removed, retaining the top 50% of genes by expression level. A soft-threshold power (β) of 12, determined according to scale-free topology criteria, was applied to construct a weighted adjacency matrix, which was transformed into a topological overlap matrix (TOM). Genes were clustered into modules using the Dynamic Tree Cut algorithm, and each module was assigned a unique color. Module eigengenes (MEs) were calculated, and correlations between modules and experimental phenotypes were assessed. Modules associated with Leu regulation were identified by integrating module–trait relationships and module membership (MM) values. Hub genes were defined as those with high MM and strong correlation with MEs.

Intersection of DEGs and Hub Genes

Intersection analysis was performed between DEGs and hub genes from the grey60 module. DEGs were identified using DESeq2 with the thresholds adjusted p-value < 0.05 and |log₂FoldChange| > 1. Hub genes were selected based on MM values and significant correlation with MEs (p < 0.05). The overlapping genes were subjected to functional enrichment analysis and experimental validation.

Identification of Key Genes Using Machine Learning

Least Absolute Shrinkage and Selection Operator (LASSO) regression, known for its robustness in high-dimensional, small-sample datasets, was employed for simultaneous feature selection and regularization. LASSO analysis was performed on the intersected gene set using the glmnet R package (v4.1-4). The optimal regularization parameter (λ) was determined via 10-fold cross-validation, with model performance evaluated by binomial deviance. The resulting model achieved an average classification accuracy of approximately 90% and an area under the curve (AUC) of approximately 0.93.

Statistical Analysis

All data are presented as mean ± standard deviation (Mean ± SD). Statistical analyses were performed using GraphPad Prism 9.0. Comparisons between two groups were conducted using the t-test, while comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) followed by multiple comparison correction. For data that did not follow a normal distribution, non-parametric tests were applied. A significance level of p < 0.05 was set, and p-values below this threshold were considered statistically significant.

Results

Leu Ameliorates Cartilage Degeneration and Osteophyte Formation in CIOA Mice

A CIOA mouse model was successfully established (Fig. 1A). Mice were randomly assigned to four groups: sham (normal mice), model (CIOA), low-dose Leu (CIOA + 50 mg/kg Leu), and high-dose Leu (CIOA + 100 mg/kg Leu).

Figure 1.

Leu attenuates cartilage degeneration and osteophyte formation in CIOA mice. (A) Experimental flowchart of CIOA mouse model establishment. (B) Safranin O/Fast Green staining of cartilage tissue (scale bar: 500 μm), assessing cartilage matrix loss. (C) OARSI scoring of cartilage degeneration severity. (D, E) Micro-CT analysis of osteophyte volume and morphology in knee joints. (F, G) H&E staining of synovial tissue inflammation (scale bar: 100 μm). Data represent mean ± SD; n = 6 mice per group. AC = articular cavity; CC = calcified cartilage; CIOA = collagenase-induced osteoarthritis; F = femur; HC = hyaline cartilage; L = synovial lining macrophages; M = meniscus; SL = synovial sublining macrophages. *P < 0.05, **P < 0.01, ***P < 0.001 compared between groups.

Safranin O/Fast Green staining showed a marked reduction in cartilage matrix density in the model group, indicating substantial matrix loss. OARSI scoring confirmed significantly greater cartilage degeneration in the model group compared with the sham group, whereas high-dose Leu markedly lowered the OARSI score (Fig. 1B, C). Micro-CT scans demonstrated a pronounced increase in osteophyte volume in the model group, while Leu treatment reduced osteophyte formation, restoring joint morphology to near-normal levels (Fig. 1D, E). H&E staining revealed pronounced synovial thickening and inflammatory infiltration in the model group, both of which were alleviated by Leu (Fig. 1F, G).

Furthermore, IF and IHC of synovial tissue showed a significant increase in M1 macrophages (iNOS⁺) in the model group, with no notable change in M2 macrophages (CD206⁺) compared with the sham group. In contrast, the high-dose Leu group showed a marked reduction in M1 macrophages and a significant increase in M2 macrophages (Fig. 2A-F). Leu reduced synovial inflammation in parallel with a shift from M1 to M2 macrophages.

Figure 2.

Leu modulates macrophage polarization in synovial tissue of CIOA mice. (A, B, D, E) IF staining of F4/80⁺ and iNOS⁺ M1 macrophages in synovial tissue. (C and F) IHC of CD206⁺ M2 macrophage proportions. Scale bar: 50 μm. Data represent six mice per group. *P < 0.05, ***P < 0.001 compared between groups; ns, not significant.

Multi-Omics Analysis Reveals Leu-Mediated Regulation of Macrophage M1 Polarization via the mTOR Signaling Pathway

To investigate the effects of Leu on macrophage polarization, BMDMs from C57BL/6 mice were induced to undergo M1 polarization. Following a 24-hour treatment with 50 μM Leu or control conditions, total RNA was extracted from the cells and subjected to high-throughput RNA-seq (Suppl. Fig. S1A). Differential gene expression analysis identified 1,068 significantly regulated genes in the Leu group compared with controls, including 586 upregulated and 494 downregulated genes (Suppl. Fig. S1B, C). GO enrichment analysis indicated significant associations with BP and MF related to inflammatory response regulation, immune modulation, biological signaling, and cellular structural reorganization (Suppl. Fig. S1D). KEGG analysis highlighted enrichment in pathways such as TNF signaling, mTOR signaling, and lipid metabolism (Suppl. Fig. S1E).

To identify gene co-expression networks associated with M1 polarization and Leu treatment, WGCNA was performed. Scale-free topology analysis identified a soft threshold power of 12, yielding a scale-free fit index of 0.85 with acceptable connectivity (Fig. 3A). Hierarchical clustering of DEGs generated a module dendrogram, illustrating the relationships among distinct modules (Fig. 3B). Correlation heatmaps revealed that the grey60 module exhibited the highest correlation with the Leu-treated group (r = 0.98, P < 0.01) (Fig. 3C).

Figure 3.

Multi-omics analysis reveals the mechanism by which leu regulates macrophage M1 polarization. (A) Scale-free fit index analysis and mean connectivity analysis for different soft-thresholding powers; (B) Cluster dendrogram of differentially expressed gene modules; (C) Heatmap showing correlations between modules and the control/Leu-treated groups. Each cell displays the correlation coefficient and corresponding p-value; (D) Overlapping genes identified by DESeq2 and WGCNA; (E) Five key regulatory genes selected via LASSO regression machine learning model.

Hub genes from the grey60 module were intersected with significant DEGs (Fig. 3D). Combining these results with the LASSO regression machine learning model, we screened five important regulatory genes from the intersection: the transmembrane protein 5430427O19Rik (Tmem184b), implicated in neuronal axon regeneration⁴¹ and vascular endothelial homeostasis;⁴² the nuclear receptor LXRα; the DNA replication licensing factor Mcm7, with immune regulatory functions;^43,44 the regulatory protein Zfyve28, involved in intracellular transport and degradation;⁴⁵ and the stress-response kinase Sgk1.⁴⁶ Their roles in macrophage polarization were subsequently validated (Fig. 3E). Previous studies have demonstrated that Leu macrophage polarization via the mTORC1/LXRα pathway mitigates cytokine storms.²¹ However, the role of this pathway in traumatic arthritis remains unclear. Therefore, we further investigated the potential mechanism of the mTORC1/LXRα pathway in the context of traumatic arthritis.

Leu Inhibits M1 Macrophage Polarization via the mTORC1/LXRα Signaling Pathway

Previous studies have demonstrated that LXRα serves as a transcriptional regulator of Arg1 (a marker gene for M2 polarization) in macrophages, and its activation enhances Arg1 expression.⁴⁷ Nuclear translocation immunofluorescence showed that leucine directly induces LXRα nuclear translocation (Suppl. Fig. S2A, B). GSK2033, a small-molecule LXRα inhibitor, is widely used to assess LXRα function. While highly selective at low concentrations, at higher doses (10 μM), it can also affect other nuclear receptors such as ERRα and PR.³⁹ To ensure target specificity, we tested GSK2033 across a dose range (0.5–5 μM). A concentration of 2 μM produced the strongest selective suppression of LXRα (Suppl. Fig. S3A) and showed a clear dose-dependent reversal of Leu’s effects on M1 marker TNF-α and M2 marker TNF-β (Suppl. Fig. S3B, C).

In this study, BMDMs from C57BL/6 mice were polarized into M1 macrophages using LPS and IFN-γ (referred to as the LPS group) or into M2 macrophages using IL-4. The experimental groups were divided as follows: M1 polarization groups: Control, LPS, LPS + Leu, LPS + Leu + GSK2033; M2 polarization groups: Control, IL-4, IL-4 + Leu, IL-4 + Leu + GSK2033 (Fig. 4A).

Figure 4.

Leu regulates macrophage M1 polarization via the mTORC1/LXRα signaling pathway. (A) Schematic diagram illustrating Leu-mediated inhibition of macrophage M1 polarization. (B-D) RT-qPCR analysis of M1 macrophage marker genes (iNOS, TNF-α, CD86) in LPS-stimulated BMDMs. (E-G) ELISA quantification of M1-associated cytokines (TNF-α, IL-6, IL-1β) in the supernatant of LPS-stimulated BMDMs. (H-I) IF staining of iNOS⁺ M1 macrophages in LPS-stimulated BMDMs (scale bar: 25 μm). Data represent three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001 compared between groups.

RT-qPCR showed that M1 marker genes (iNOS, TNF-α, CD86) were significantly upregulated in the LPS group compared with the control. Leu markedly suppressed their expression, whereas co-treatment with the LXRα antagonist GSK2033 reversed this effect, restoring M1 marker levels (Fig. 4B-D). Consistently, ELISA revealed elevated M1-associated cytokines (TNF-α, IL-6, IL-1β) in the LPS group, which were reduced by Leu, whereas co-treatment with GSK2033 restored their secretion (Fig. 4E-G). IF further confirmed these results: the proportion of iNOS⁺ M1 macrophages increased in the LPS group, decreased with Leu, and was restored upon GSK2033 treatment (Fig. 4H-I).

Subsequently, the polarization of M2 macrophages was assessed. IF staining revealed that, compared to the control group, the proportion of CD206⁺ M2 macrophages was significantly increased in the LPS group. Furthermore, Leu treatment (LPS + Leu group) further elevated the CD206⁺ M2 macrophage proportion compared to the LPS group, whereas GSK2033 treatment (LPS + Leu + GSK2033 group) significantly reduced it (Fig. 5A). RT-qPCR analysis revealed that M2 marker genes (Arg1, CD206, IL-10) were upregulated in the IL-4 group compared with the control. Compared with the IL-4 group, the IL-4 + leucine group exhibited further significant increases in M2 gene expression, indicating that leucine can enhance IL-4-driven M2 polarization. Leu supplementation (IL-4 + Leu group) further enhanced M2 gene expression, whereas GSK2033 treatment (IL-4 + Leu + GSK2033 group) markedly suppressed it (Fig. 5B-D). Consistently, ELISA results showed that IL-10 and TGF-β levels were elevated in the IL-4 group, further increased by Leu, and significantly reduced by GSK2033 (Fig. 5E, F).

Figure 5.

Leu regulates macrophage M2 polarization via the mTORC1/LXRα signaling pathway. (A) IF staining of CD206⁺ M2 macrophages in LPS-stimulated BMDMs (scale bar: 25 μm). (B-D) RT-qPCR analysis of M2 marker genes (Arg1, CD206, IL-10) in IL-4-stimulated BMDMs. (E-F) ELISA quantification of M2-associated cytokines (IL-10, TGF-β) in IL-4-stimulated BMDMs. (G-I) Western Blot analysis of p-STAT6, STAT6, p-mTOR, mTOR, and LXRα protein expression in IL-4-stimulated BMDMs. All experiments were performed in triplicate. *P < 0.05, **P < 0.01, ***P < 0.001 compared between groups.

Western blot analysis demonstrated that LPS treatment decreased the expression of mTORC1 downstream molecules (p-STAT6, p-mTOR, LXRα) relative to the control. However, Leu treatment restored their expression levels, whereas GSK2033 administration again suppressed them (Fig. 5G-I). Notably, the total protein levels of STAT6 and mTOR remained unchanged across all groups.

Rapamycin, a specific inhibitor of mTORC1, markedly reversed the inhibitory effects of leucine on iNOS and TNF-α expression (Suppl. Fig. S4A, B), suggesting that mTORC1 is a key downstream mediator of the anti-inflammatory effect of leucine. Notably, the effects of mTORC1 inhibition by rapamycin and LXRα inhibition by GSK2033 were similar when used individually. Furthermore, dual inhibition resulted in a more pronounced blockade of the anti-inflammatory effect. Immunoprecipitation experiments further directly demonstrated that mTORC1 and LXRα can physically interact in macrophages following leucine treatment, whereas this interaction disappeared after rapamycin treatment, confirming its dependence on mTORC1 activity (Suppl. Fig. S4C). These findings support a model in which leucine modulates macrophage polarization via the mTORC1/LXRα signaling axis, with this interaction representing a potential molecular mechanism for regulating macrophage function. Future research will further explore the molecular details of this interaction and its specific role in macrophage function.

GSK2033 reversed Leu’s effects on macrophage markers and cytokines, supporting mTORC1/LXRα-dependent polarization.

Leucine Regulates the Macrophage–Chondrocyte Signaling Axis

To assess the impact of Leu on macrophage polarization and chondroprogenitor cell function, LPS-induced M1 macrophages were assigned to four groups: control, LPS, LPS + Leu, and LPS + Leu + GSK2033 (LXRα antagonist). CM from each group was collected, filtered, and applied to ATDC5 chondroprogenitor cells (Fig. 6A). RT-qPCR showed that LPS-CM markedly reduced chondrogenic markers (Col2a1, Acan) and increased hypertrophic markers (Runx2, Col10a1) compared with control. In contrast, Leu treatment reversed these effects, increasing Col2a1 and Acan expression while suppressing Runx2 and Col10a1. However, the addition of GSK2033 abolished the protective effects of Leu, leading to reduced Col2a1 and Acan expression and elevated Runx2 and Col10a1 levels (Fig. 6B-E).

Figure 6.

Macrophage CM regulates chondroprogenitor cell function via the β-catenin signaling pathway. (A) Experimental workflow illustrating the treatment of chondroprogenitor cells with macrophage CM and subsequent modulation of the β-catenin pathway. (B-E) RT-qPCR analysis of chondroprogenitor cell markers (Col2a1, Acan) and terminal differentiation markers (Runx2, Col10a1). (F-G) Western Blot detection of β-catenin signaling-related proteins (β-catenin, Cyclin D1, MMP13) in chondroprogenitor cells. (H) Toluidine blue staining assessing cartilage matrix secretion capacity. Experiments were performed in triplicate; *P < 0.05, **P < 0.01, ***P < 0.001 compared between groups.

Western Blot analysis demonstrated that LPS-treated CM significantly increased β-catenin pathway proteins (β-catenin, Cyclin D1, and MMP13) in chondroprogenitor cells compared to the control. Leu treatment attenuated these increases, whereas GSK2033 co-treatment restored the elevated expression of these proteins (Fig. 6F, G). Furthermore, toluidine blue staining indicated that LPS impaired extracellular matrix secretion in chondroprogenitor cells, an effect that was rescued by Leu but reversed upon GSK2033 administration (Fig. 6H).

Overall, CM from Leu-treated macrophages suppressed β-catenin activity and preserved matrix production in chondroprogenitor cells.

Rspo2 Secreted by Macrophages Mediates β-Catenin Pathway Activation

M1 macrophage polarization was induced by LPS stimulation, and cells were divided into four groups: Control, LPS, LPS + leucine, and LPS + leucine + GSK2033. CM from each group was collected, filtered, and centrifuged to remove cellular debris before use in subsequent experiments (Fig. 7A). ELISA revealed that LPS markedly increased Rspo2 secretion compared with control, while Leu (200 μg/mL) reduced Rspo2 levels, and this effect was reversed by the LXRα inhibitor GSK2033 (Fig. 7B).

Figure 7.

Rspo2 regulates chondroprogenitor cell function via activation of the β-catenin signaling pathway. (A) Experimental workflow illustrating how Leu, via a macrophage-CM, modulates Rspo2 secretion and its subsequent effects on the β-catenin pathway in chondroprogenitor cells. (B) ELISA analysis of Rspo2 secretion levels in macrophage CM. (C) RT-qPCR analysis of chondroprogenitor cell markers (Col2a1, Acan) and terminal differentiation markers (Runx2, Col10a1). (D) Western Blot analysis of β-catenin signaling-related proteins (β-catenin, Cyclin D1, MMP13) in chondroprogenitor cells. (E) Toluidine blue staining assessing cartilage matrix secretion capacity. All experiments were performed in triplicate. *P < 0.05, **P < 0.01, ***P < 0.001 compared between groups.

Treatment of ATDC5 chondroprogenitor cells with recombinant Rspo2 protein (rhRspo2, 200 ng/mL) revealed that, compared with the Control group, leucine alone increased Col2a1 and Acan expression and decreased terminal differentiation markers Runx2 and Col10a1. In contrast, Rspo2 treatment significantly reduced Col2a1 and Acan expression and increased Runx2 and Col10a1 levels (Fig. 7C). Western blotting analysis revealed that Leu markedly reduced β-catenin pathway proteins (β-catenin, Cyclin D1, MMP13), whereas the Leu + Rspo2 group displayed increased expression of these proteins, though still lower than in the Rspo2-only group (Fig. 7D). Toluidine blue staining further demonstrated enhanced extracellular matrix secretion in the Leu group, but a marked reduction in the Leu + Rspo2 group (Fig. 7E).

These results indicate that Rspo2 activated β-catenin and impaired chondroprogenitor function; exogenous Rspo2 abrogated Leu-induced protection.

Leu Attenuates Cartilage Degeneration and Osteophyte Formation in CIOA Mice Via the mTORC1/LXRα Signaling Pathway

CIOA mice were assigned to two groups: Leu (CIOA + high-dose Leu) and Leu + GSK2033 (CIOA + Leu + LXRα antagonist) (Fig. 8A). Safranin O/Fast Green staining revealed lower cartilage matrix density in the Leu + GSK2033 group than in the Leu group, indicating that GSK2033 abolished the protective effect of Leu on cartilage matrix loss (Fig. 8B). OARSI scoring further demonstrated that the Leu + GSK2033 group exhibited higher scores than the Leu group, indicating that GSK2033 markedly reverses Leu’s inhibition of cartilage degeneration (Fig. 8C). H&E staining showed increased synovial thickening and inflammatory cell infiltration in the Leu + GSK2033 group compared to the Leu group (Fig. 8D, E). Micro-CT analysis revealed a significant increase in osteophyte volume and worsened osteophyte morphology in the Leu + GSK2033 group (Fig. 8F, G).

Figure 8.

GSK2033 reverses the protective effects of leu on cartilage degeneration and osteophyte formation in CIOA mice. (A) Schematic diagram illustrating the mechanism by which Leu regulates cartilage degeneration and osteophyte formation in CIOA mice via the mTORC1/LXRα signaling pathway. (B) Safranin O/Fast Green staining assessing cartilage degradation (scale bar: 500 μm). (C) OARSI scoring system quantifying cartilage degeneration. (D-E) H&E staining evaluating synovial thickening and inflammation (scale bar: 100 μm). (F-G) Micro-CT scans demonstrating osteophyte volume and morphological changes. (H-I) Immunofluorescence and immunohistochemistry of synovial tissue detecting iNOS (M1) and CD206 (M2) expression and the M1/M2 ratio (iNOS⁺/CD206⁺ (scale bar: 25 μm). Data represent mean ± SD; n = 6 mice per group. *P < 0.05, **P < 0.01, ***P < 0.001 compared between groups.

IF and IHC revealed higher M1 marker (iNOS) and lower M2 marker (CD206) expression, with an elevated iNOS⁺/CD206⁺ ratio in the Leu + GSK2033 group. Quantitative analysis confirmed that leucine significantly reduced this ratio, indicating macrophage reprogramming toward an M2-like phenotype (Fig. 8H. I). Western blotting of synovial tissue lysates showed reduced p-STAT6, STAT6, p-mTOR, mTOR, and LXRα levels, but increased β-catenin pathway proteins (β-catenin, Cyclin D1, MMP13) in the Leu + GSK2033 group compared with the leucine group (Suppl. Fig. S5A, B). ELISA and immunohistochemistry further showed significantly elevated Rspo2 levels in serum and synovial tissue in the leucine + GSK2033 group compared with the leucine group (Suppl. Fig. S5C, D).

In vivo immunofluorescence staining of synovial tissue was performed to visualize M1 and M2 macrophage markers. CIOA mice were injected with CD45.2 antibody to label circulating monocytes, and multicolor flow cytometry distinguished CD45.2⁺ (newly recruited) from CD45.2⁻ (resident) macrophages, combined with iNOS (M1) and CD206 (M2) staining. Leucine treatment increased the proportion of CD206⁺ cells within the CD45.2⁻ population, indicating in situ polarization toward an M2 phenotype (Fig. S6A, B).

GSK2033 abolished these effects, consistent with LXRα involvement in Leu-mediated protection against cartilage damage and osteophyte formation

Discussion

As an essential amino acid, Leu has been extensively studied for its role in protein synthesis and metabolic regulation via the mTORC1 pathway, particularly in muscle repair and energy homeostasis.^48
-51 However, its immunomodulatory functions remain underexplored. In this study, Leu mitigated CIOA by reprogramming synovial macrophages toward an M2-like phenotype through mTORC1-dependent activation of LXRα, thereby reducing Rspo2 secretion and downstream β-catenin signaling in chondroprogenitors. Unlike previous work focusing on Leu’s effects on cell proliferation or energy metabolism,^52,53 our findings reveal a novel mechanism in which Leu attenuates pathological progression by reshaping macrophage polarization within an inflammatory microenvironment. This finding not only expands the understanding of Leu’s biological functions but also provides new insights into the crosstalk between metabolic and immune regulation. Notably, Leu’s modulation of macrophage polarization may exhibit tissue specificity, warranting further investigation into its role in other inflammatory diseases, such as rheumatoid arthritis. Our findings provide conclusive mechanistic evidence in an acute CIOA context, while implications for chronic low-grade inflammation remain hypothesis-generating.

Within an acute/early CIOA setting, our data support a conclusive mechanistic chain in which Leu activates mTORC1/LXRα to promote macrophage M2-like reprogramming, reduce Rspo2, down-tune β-catenin activity in chondrogenic progenitors, and improve joint pathology. By contrast, how these effects extend to chronic LGI remains hypothesis-generating because CIOA features a synovial inflammatory peak around day 4 that subsequently abates, whereas DMM exhibits progressive, chronic degeneration over weeks.^11,34,54 We therefore distinguish well-supported conclusions (acute) from inferences requiring chronic validation, and outline concrete next steps below.

mTORC1 and LXRα are central regulators of metabolism and inflammation, respectively. While their individual functions have been extensively studied, their cooperative mechanisms in inflammatory diseases remain unclear.^16,55
-59 This study demonstrates that Leu enhances LXRα transcriptional activity by activating mTORC1, creating a dual-pathway cascade that more effectively suppresses macrophage M1 polarization. GSK2033 partially reversed the anti-inflammatory effect of leucine, suggesting LXRα as a key downstream effector. However, the independent contribution of mTORC1 remains to be determined. Future studies using conditional mTORC1 knockout or low-toxicity mTORC1 inhibitors are warranted to clarify synergistic mechanisms. Compared to interventions targeting a single pathway, dual-pathway modulation may offer greater efficacy and specificity in OA treatment. However, the precise molecular interactions between mTORC1 and LXRα—such as phosphorylation or protein binding—require further investigation. Future studies should explore whether other metabolites engage in similar dual-pathway mechanisms to regulate immune responses, thereby advancing our understanding of the metabolic-immune crosstalk.

The role of macrophage polarization in OA progression has garnered significant attention, yet the precise mechanisms by which paracrine signaling regulates chondrocyte differentiation remain unclear.^60,61 Although Rspo2 has been implicated in OA cartilage terminal differentiation,^27,62 its cellular source and upstream regulation within the joint microenvironment remain unknown. Our multidimensional experiments reveal for the first time that Rspo2 secreted by M1 macrophages directly promotes cartilage degeneration and osteophyte formation by activating the β-catenin pathway in chondroprogenitor cells, while leucine specifically inhibits Rspo2 secretion via the mTORC1/LXRα axis. In vitro experiments confirmed that conditioned media from Leu-treated macrophages significantly suppressed aberrant β-catenin signaling, while exogenous Rspo2 supplementation completely reversed this protective effect. These findings not only identify Rspo2 as a critical mediator of macrophage-chondrocyte crosstalk but also propose a novel regulatory axis: immune cell–cytokine–metabolic reprogramming. However, Rspo2 is likely one of several mediators in leucine CM and may act synergistically with other cytokines. Unlike previous studies focusing solely on the direct damaging effects of pro-inflammatory cytokines (e.g., TNF-α, IL-1β), our work highlights the independent contribution of the Rspo2/β-catenin axis in OA pathogenesis, offering a more precise molecular target for therapeutic intervention. However, the differential expression of Rspo2 across cell types and its potential synergy with other Wnt ligands warrant further systematic investigation.

Conventional OA research has largely relied on single-omics or phenotypic association analyses, limiting the ability to resolve complex regulatory networks.^63
-67 Here, we integrated RNA-seq, WGCNA, and machine learning algorithms to delineate the core gene module underlying Leu-mediated macrophage polarization, identifying mTORC1/LXRα-Rspo2 as the key signaling axis. The multi-omics approach strengthens the mechanistic link between mTORC1/LXRα and Rspo2 rather than discovering a previously unknown pathway. Furthermore, deep learning predictions suggested miRNA involvement in Leu’s indirect regulation of Rspo2, a hypothesis preliminarily validated by functional assays. This methodological innovation not only provides a scalable framework for studying metabolic-immune interactions but also supports multi-target drug development. However, enhancing AI model interpretability remains essential, and future work should validate predicted regulatory networks through single-cell sequencing and other high-resolution approaches.

The CIOA model is widely used due to its rapid induction of cartilage degradation and osteophyte formation;⁶⁸ however, its pathological resemblance to human OA requires careful evaluation. CIOA displays acute inflammation peaking within one week, whereas clinical OA is characterized by chronic LGI—a discrepancy that may affect drug efficacy predictions. For example, while Leu rapidly alleviated inflammation by suppressing M1 polarization in the CIOA model, its long-term effects on chronic inflammation remain to be verified. Osteophyte formation in CIOA largely results from mechanical instability and localized inflammation, whereas in human OA it often develops from age-related subchondral bone remodeling. Despite these differences, the correlation between mTORC1/LXRα pathway activity and Rspo2 expression in clinical samples supports the translational relevance of the findings. Future research should incorporate aging or mechanical stress models to comprehensively evaluate Leu across different OA stages.

We use “local joint metabolic–immune microenvironment remodeling” to mean IA, joint-restricted modulation of the synovial macrophage ↔ chondroprogenitor axis (e.g., lowering Rspo2/β-catenin signaling) without broad systemic immunosuppression. Contemporary IA strategies (long-acting hydrogels/microspheres, nanomaterial platforms) can enrich drugs in synovium/macrophages, dampen inflammation, and support cartilage repair in preclinical OA, offering a translational path that aligns with our local mechanism while minimizing systemic exposure (PMID: 40843442; PMID: PMID: 40933024; 10.1016/j.cclet.2025.111723). Although our experimental data derive from an acute model, the concept of “local metabolic-immune remodeling” is introduced as a framework to guide future studies in chronic OA models.

Direct Wnt/β-catenin inhibitors have limited clinical application due to systemic toxicity. An alternative strategy is proposed, reducing joint-specific β-catenin activity through leucine-mediated suppression of Rspo2 secretion, while preserving Wnt signaling in other tissues, thereby achieving tissue specificity. Compared with conventional small-molecule inhibitors, this “metabolic intervention–immune regulation–cartilage protection” triple mechanism may reduce off-target effects but could be influenced by individual metabolic status. For example, leucine metabolism may be altered in patients with high-fat diets or liver dysfunction, affecting therapeutic response. In this study, intraperitoneal leucine administration was combined with in vitro experiments using BMDMs. Given leucine’s low molecular weight, it can readily pass through the hyperpermeable vasculature of inflamed synovium.⁴ However, BMDMs may not fully represent the metabolic state or functional phenotype of synovial macrophages, which exist in chronic hypoxia and high-lactate conditions with elevated glycolytic activity.¹⁰ Although LPS prestimulation partially mimics inflammatory metabolism, it cannot fully recapitulate the complex in vivo environment. Therefore, the optimal dosing and delivery route require further optimization. IA injection may be safer than systemic delivery but requires solutions for sustained release and technical feasibility. Future directions include developing synovium–cartilage organ-on-chip models to validate leucine’s specific effects on resident synovial macrophages, and exploring nanocarrier or gene-editing approaches to enhance intervention precision.

This study has several limitations. The CIOA mouse model replicates early cartilage degeneration and osteophyte formation characteristic of traumatic OA, but etiological and histopathological differences from human disease may limit direct clinical applicability, necessitating validation in complementary models and human tissues. CIOA features an acute synovial inflammatory peak around day 4 with systemic parameters trending toward baseline shortly thereafter; in contrast, DMM shows progressive degeneration over ~4–8 weeks.³⁴ Thus, our statements regarding chronic LGI are hypothesis-generating and require ≥8–12-week validation in chronic models, with time-course assessment of synovial macrophage subsets, Rspo2, β-catenin activity, osteophyte volume, and pain behaviors. In vitro experiments primarily used ATDC5 chondroprogenitor cells, leaving the effects of leucine on other joint-resident cell types (e.g., osteoblasts and synovial fibroblasts) insufficiently characterized. Although the data delineate crosstalk between the mTORC1/LXRα pathway and the Rspo2/β-catenin axis, potential integration with broader inflammatory or metabolic networks remains to be clarified. Limited sample size precluded rescue experiments to functionally confirm Rspo2 as a key effector. Moreover, LXRα nuclear translocation was not directly assessed; future studies will incorporate nuclear–cytoplasmic fractionation and imaging to refine the mechanistic framework. Additionally, as a nutritional supplement, Leu’s metabolic stability, tissue distribution, and optimal dosage require further investigation. Future studies should integrate clinical samples, organoid models, and multicellular co-culture systems to systematically evaluate Leu’s mechanistic precision and broader functional effects within the complex joint microenvironment. Exploring its synergistic interactions with other metabolic factors or drugs could also provide stronger experimental and theoretical support for clinical translation.

Conclusion

This study systematically elucidates the role of Leu in mitigating cartilage degeneration and osteophyte formation in a CIOA mouse model through multi-omics approaches and in vitro/vivo experiments. Specifically, Leu suppresses macrophage M1 polarization by activating the mTORC1/LXRα signaling pathway, thereby reducing Rspo2 secretion and subsequently downregulating β-catenin signaling activation in chondroprogenitor cells. Animal experiments further confirmed Leu’s therapeutic effects, including alleviating synovial inflammation, reducing inflammatory cytokine expression, and improving cartilage function. Notably, the protective effects were reversed by the mTORC1 inhibitor GSK2033, underscoring the critical role of this pathway. This study proposes a multi-tiered mechanism—“Leu → mTORC1/LXRα → macrophages → Rspo2/β-catenin → chondrocytes”—that links metabolic regulation to joint protection. These findings broaden the scope of nutritional immunology in degenerative joint diseases and provide a mechanistic basis for interventions targeting immunometabolic pathways. Given its metabolic relevance, wide availability, and favorable safety profile, Leu warrants further exploration as a potential therapeutic agent for osteoarthritis.

Supplemental Material

sj-docx-1-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-docx-1-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Supplemental Material

sj-jpg-2-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-jpg-2-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Supplemental Material

sj-jpg-3-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-jpg-3-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Supplemental Material

sj-jpg-4-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-jpg-4-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Supplemental Material

sj-jpg-5-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-jpg-5-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Supplemental Material

sj-jpg-6-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-jpg-6-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Supplemental Material

sj-jpg-7-car-10.1177_19476035251389178 – Supplemental material for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression

Supplemental material, sj-jpg-7-car-10.1177_19476035251389178 for Leucine Attenuates Osteoarthritis via mTORC1/LXRα-Mediated Macrophage Reprogramming and Rspo2/β-Catenin Axis Suppression by Ke Zheng, Jiebin Zhang, Wenxiang Liu and Bin Xu in CARTILAGE

Footnotes

ORCID iD

Bin Xu

Ethical Considerations

All animal experiments were approved by the Animal Ethics Committee of Anhui Medical University and were conducted in accordance with institutional and national guidelines for the care and use of laboratory animals. Clinical trial number: Not applicable.

Author Contributions

K.Z.: Conceptualization, Methodology, Investigation, Writing—original draft, Writing—review & editing. B.X.: Conceptualization, Methodology, Formal Analysis, Writing—original draft, Writing—review & editing. J.Z.: Investigation, Data Curation, Formal Analysis. W.L.: Investigation, Validation.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

All data generated or analyzed during this study are included in this article and/or its supplementary material files. Further enquiries can be directed to the corresponding author.

Supplemental Material

Supplementary material for this article is available on the Cartilage website at .

References

Guo

Chen

Zhang

Osteoarthritis burden and inequality from 1990 to 2021: a systematic analysis for the global burden of disease study 2021. Sci Rep. 2025;15:8305. doi:10.1038/s41598-025-93124-z.

Ren

Yang

The global burden of osteoarthritis knee: a secondary data analysis of a population-based study. Clin Rheumatol. 2025;44(4):1769-810.

GBD 2021 Osteoarthritis Collaborators. Global, regional, and national burden of osteoarthritis, 1990–2020 and projections to 2050: a systematic analysis for the global burden of disease study 2021. Lancet Rheumatol. 2023;5(9):e508-e522.

Alexander

Ocasio-Nieves

Yocum

Lin

, et al. Pathogenesis of osteoarthritis: risk factors, regulatory pathways in chondrocytes, and experimental models. Biology. 2020;9:194.

Tong

Huang

Shen

Xiao

Chen

, et al. Current understanding of osteoarthritis pathogenesis and relevant new approaches. Bone Res. 2022;10:60. doi:10.1038/s41413-022-00226-9.

Dilley

Bello

Roman

McKinley

Sankar

Post-traumatic osteoarthritis: a review of pathogenic mechanisms and novel targets for mitigation. Bone Rep. 2023;18:101658.

Peng

Chen

Zhang

Tian

Wang

Chen

. Advances in the pathology and treatment of osteoarthritis. J Adv Res. Epub 2025 Jan 30. doi:10.1016/j.jare.2025.01.053.

Testa

Giardina

SMC

Culmone

Vescio

Turchetta

Cannavò

, et al. Intra-articular injections in knee osteoarthritis: a review of literature. J Funct Morphol Kinesiol. 2021;6:15.

Si-Hyeong Park

Kim

. Efficacy of intra-articular injections for the treatment of osteoarthritis: a narrative review. Osteoarthr Cartil Open. 2025;7:100596.

10.

Yuan

Jiang

Yang

Tao

Yang

, et al. Emerging roles of macrophage polarization in osteoarthritis: mechanisms and therapeutic strategies. Orthop Surg. 2024;16(3):532-50.

11.

Zhao

Ruan

Nie

Effects of synovial macrophages in osteoarthritis. Front Immunol. 2023;14:1164137. doi:10.3389/fimmu.2023.1164137.

12.

Zhang

Lin

Zeng

Wang

, et al. Synovial macrophage M1 polarisation exacerbates experimental osteoarthritis partially through R-spondin-2. Ann Rheum Dis. 2018;77(10):1524-34.

13.

Fernandes

Gomoll

Lattermann

Hernandez

Bueno

Amano

MT.

Macrophage: a potential target on cartilage regeneration. Front Immunol. 2020;11:111. doi:10.3389/fimmu.2020.00111.

14.

Deng

Yang

Zhao

Yang

Advances in research on M1/M2 macrophage polarization in the pathogenesis and treatment of osteoarthritis. Heliyon. 2025;11:e42881.

15.

Xia

Cheng

Zhang

Xiao

Liu

Chen

, et al. S6K1 in the central nervous system regulates energy expenditure via MC4R/CRH pathways in response to deprivation of an essential amino acid. Diabetes. 2012;61(10):2461-71.

16.

Saxton

Sabatini

DM.

mTOR signaling in growth, metabolism, and disease. Cell. 2017;168:960-76.

17.

Panwar

Singh

Bhatt

Tonk

Azizov

Raza

, et al. Multifaceted role of mTOR (mammalian target of rapamycin) signaling pathway in human health and disease. Sig Transduct Target Ther. 2023;8:375. doi:10.1038/s41392-023-01608-z.

18.

Szwed

Kim

Jacinto

Regulation and metabolic functions of mTORC1 and mTORC2. Physiol Rev. 2021;101:1371-426.

19.

Zhou

Cai

Zhong

Shen

LXRɑ participates in the mTOR/S6K1/SREBP-1c signaling pathway during sodium palmitate-induced lipogenesis in HepG2 cells. Nutr Metab. 2018;15:31. doi:10.1186/s12986-018-0268-9.

20.

Qin

Zhang

Qin

Wang

, et al. LXRα agonists ameliorates acute rejection after liver transplantation via ABCA1/MAPK and PI3K/AKT/mTOR signaling axis in macrophages. Mol Med. 2025;31:99. doi:10.1186/s10020-025-01153-1.

21.

Yan

Liu

Mao

, et al. Leucine alleviates cytokine storm syndrome by regulating macrophage polarization via the mTORC1/LXRα signaling pathway. eLife. 2024;12:RP89750. doi:10.7554/elife.89750.

22.

Yang

T-M

Miao

W-Q

Wang

Xia

F-J

Y-J

, et al. Targeting macrophages in atherosclerosis using nanocarriers loaded with liver X receptor agonists: a narrow review. Front Mol Biosci. 2023;10:1147699. doi:10.3389/fmolb.2023.1147699.

23.

Guo

Yuan

Kuai

Morin

, et al. Synthetic high-density lipoprotein-mediated targeted delivery of liver X receptors agonist promotes atherosclerosis regression. EBioMedicine. 2018;28:225-33.

24.

Verschuren

de Vries-van der Weij

Zadelaar

Kleemann

Kooistra

LXR agonist suppresses atherosclerotic lesion growth and promotes lesion regression in apoE*3Leiden mice: time course and mechanisms. J Lipid Res. 2009;50:301-11.

25.

Dragoljevic

Lee

MKS

Pernes

Morgan

Louis

Shihata

, et al. Administration of an LXR agonist promotes atherosclerotic lesion remodelling in murine inflammatory arthritis. Clin Transl Immunology. 2023;12:e1446. doi:10.1002/cti2.1446.

26.

Arima

Hasegawa

Yoshida

Mitarai

Tomokiyo

Hamano

, et al. R-spondin 2 promotes osteoblastic differentiation of immature human periodontal ligament cells through the Wnt/β-catenin signaling pathway. J Periodontal Res. 2018;54:143-53.

27.

Gong

Yuan

Sun

Wang

Liu

Guo

, et al. R-spondin 2 induces odontogenic differentiation of dental pulp stem/progenitor cells via regulation of Wnt/β-catenin signaling. Front Physiol. 2020;11:918. doi:10.3389/fphys.2020.00918.

28.

AlJanahi

Danielsen

Dunbar

CE.

An introduction to the analysis of single-cell RNA-sequencing data. Mol Ther Methods Clin Dev. 2018;10:189-96.

29.

Wang

Gerstein

Snyder

RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet. 2009;10(1):57-63.

30.

Yan

Sankar

Mirza

DCM

Pelletier

Huang

, et al. Missing values in longitudinal proteome dynamics studies: making a case for data multiple imputation. J Proteome Res. 2024;23:4151-62.

31.

Wang

Tian

Wang

Liu

Shang

, et al. Identification of the biomarkers and pathological process of heterotopic ossification: weighted gene co-expression network analysis. Front Endocrinol. 2020;11:581768. doi:10.3389/fendo.2020.581768.

32.

Xuan

Chen

Fan

, et al. The application of machine learning in early diagnosis of osteoarthritis: a narrative review. Ther Adv Musculoskelet Dis. 2023;15:1759720X231158198. doi:10.1177/1759720x231158198.

33.

Nielsen

Monfeuga

Kitchen

Egerod

Leal

Schreyer

ATH

, et al. Data-driven identification of predictive risk biomarkers for subgroups of osteoarthritis using interpretable machine learning. Nat Commun. 2024;15:2817. doi:10.1038/s41467-024-46663-4.

34.

Weber

Bevc

Fercher

Kauppinen

Zhang

Asadikorayem

, et al. The collagenase-induced osteoarthritis (CIOA) model: where mechanical damage meets inflammation. Osteoarthr Cartil Open. 2024;6(4):100539.

35.

Lin

Wei

Yang

Zhu

, et al. Modelling osteoarthritis in mice via surgical destabilization of the medial meniscus with or without a stereomicroscope. Bone Joint Res. 2022;11:518-27.

36.

Glasson

Blanchet

Morris

EA.

The surgical destabilization of the medial meniscus (DMM) model of osteoarthritis in the 129/SvEv mouse. Osteoarthritis Cartilage. 2007;15(9):1061-9.

37.

Zhang

Pan

Liu

, et al. Fargesin ameliorates osteoarthritis via macrophage reprogramming by downregulating MAPK and NF-κB pathways. Arthritis Res Ther. 2021;23:142. doi:10.1186/s13075-021-02512-z.

38.

Hou

Y-C

J-M

Chen

K-Y

M-H

Yang

P-J

Lee

P-C

, et al. Glutamine and leucine administration attenuates muscle atrophy in sepsis. Life Sci. 2023;314:121327.

39.

Griffett

Burris

TP.

Promiscuous activity of the LXR antagonist GSK2033 in a mouse model of fatty liver disease. Biochem Biophys Res Commun. 2016;479:424-8.

40.

Zhang

Wang

Qiao

Zhang

, et al. Glucocorticoids increase adiposity by stimulating Krüppel-like factor 9 expression in macrophages. Nat Commun. 2024;15:1190. doi:10.1038/s41467-024-45477-8.

41.

Antonelli

De Filippo

Middei

Stancheva

Pastore

Ammassari-Teule

, et al. Pin1 modulates the synaptic content of NMDA receptors via prolyl-isomerization of PSD-95. J Neurosci. 2016;36:5437-47.

42.

Broughton

Sussman

MA.

Enhancement strategies for cardiac regenerative cell therapy. Circ Res. 2018;123:177-87.

43.

Krienen

Goldman

Zhang

Del Rosario

RCH

Florio

Machold

, et al. Author correction: innovations present in the primate interneuron repertoire. Nature. 2020;588(7837):E17.

44.

Tomita

Daiho

Kusakizako

Yamashita

Ogasawara

Murata

, et al. Cryo-EM reveals mechanistic insights into lipid-facilitated polyamine export by human ATP13A2. Mol Cell. 2021;81:4799-4809.

45.

Yan

Huang

Yuan

Cui

, et al. ZFYVE28 mediates insulin resistance by promoting phosphorylated insulin receptor degradation via increasing late endosomes production. Nat Commun. 2023;14:6833. doi:10.1038/s41467-023-42657-w.

46.

J-S

Ann

E-J

Yoon

J-H

Jung

Choi

Y-H

Kim

H-Y

, et al. Serum- and glucocorticoid-inducible kinase 1 (SGK1) controls notch1 signaling by downregulation of protein stability through Fbw7 ubiquitin ligase. J Cell Sci. 2011;124:100-12.

47.

Pourcet

Feig

Vengrenyuk

Hobbs

Kepka-Lenhart

Garabedian

, et al. LXRα regulates macrophage arginase 1 through PU.1 and interferon regulatory factor 8. Circ Res. 2011;109:492-501.

48.

Rehman

Ali

Zhang

Zafar

Wang

Research progress in the role and mechanism of leucine in regulating animal growth and development. Front Physiol. 2023;14:1252089. doi:10.3389/fphys.2023.1252089.

49.

Ely

Phillips

Smith

Wilkinson

Piasecki

Breen

, et al. A focus on leucine in the nutritional regulation of human skeletal muscle metabolism in ageing, exercise and unloading states. Clin Nutr. 2023;42(10):1849-65.

50.

Yin

Nutritional and regulatory roles of leucine in muscle growth and fat reduction. Front Biosci. 2015;20:796-813.

51.

Vianna

Teodoro

GFR

Torres-Leal

Tirapegui

Protein synthesis regulation by leucine. Braz J Pharm Sci. 2010;46:29-36.

52.

da Luz Dias

Basso

Donadio

MVF

Pujol

Bartrons

Haute

, et al. Leucine reduces the proliferation of MC3T3-E1 cells through DNA damage and cell senescence. Toxicol In Vitro. 2018;48:1-10.

53.

Oliveira

Cruz

de Oliveira

SCP

Viana

Da Silva Miyaguti

Ramos

LAF

, et al. Leucine and its importance for cell signalling pathways in cancer cachexia-induced muscle wasting. In: Valarmathi

, editors. Muscle cells—recent advances and future perspectives. London: IntechOpen; 2013;64:343-350.

54.

Yin

Wang

Tang

Wang

Zhang

Research progress on macrophage polarization during osteoarthritis disease progression: a review. J Orthop Surg Res. 2024;19:584. doi:10.1186/s13018-024-05052-9.

55.

Collins

M-H

Sun

I-H

Chan-Li

Zhao

Powell

, et al. mTORC1 signaling regulates proinflammatory macrophage function and metabolism. J Immunol. 2021;207:913-22.

56.

Simcox

Lamming

DW.

The central moTOR of metabolism. Dev Cell. 2022;57:691-706.

57.

Mafi

Mansoori

Taeb

Sadeghi

Abbasi

Cho

, et al. mTOR-mediated regulation of immune responses in cancer and tumor microenvironment. Front Immunol. 2022;12:774103. doi:10.3389/fimmu.2021.774103.

58.

Glaría

Letelier

Valledor

AF.

Integrating the roles of liver X receptors in inflammation and infection: mechanisms and outcomes. Curr Opin Pharmacol. 2020;53:55-65.

59.

Zhao

Lei

Deng

Sun

Jin

, et al. The roles of liver X receptor α in inflammation and inflammation-associated diseases. J Cell Physiol. 2020;236:4807-28.

60.

Warmink

Vinod

Korthagen

Weinans

Rios

JL.

Macrophage-driven inflammation in metabolic osteoarthritis: implications for biomarker and therapy development. Int J Mol Sci. 2023;24:6112.

61.

Zhang

Cai

Bai

Macrophages regulate the progression of osteoarthritis. Osteoarthritis Cartilage. 2020;28:555-61.

62.

Zou

Qian

Macrophage polarization in the osteoarthritis pathogenesis and treatment. Orthop Surg. 2024;17:22-35.

63.

Deng

Tang

Yin

Jiang

Huang

Feng

, et al. Clinical and omics biomarkers in osteoarthritis diagnosis and treatment. J Orthop Translat. 2025;50:295-305.

64.

Wei

Qian

Zhang

Wang

Yan

Xiao

, et al. Progress in multi-omics studies of osteoarthritis. Biomark Res. 2025;13:26. doi:10.1186/s40364-025-00732-y.

65.

Kraus

Blanco

Englund

Karsdal

Lohmander

LS.

Call for standardized definitions of osteoarthritis and risk stratification for clinical trials and clinical use. Osteoarthritis Cartilage. 2015;23(8):1233-41.

66.

Tang

Zhang

Risberg

Bierma-Zeinstra

, et al. Osteoarthritis. Nat Rev Dis Primers. 2025;11:10. doi:10.1038/s41572-025-00594-6.

67.

Young

Barter

Soul

Osteoarthritis year in review: genetics, genomics, epigenetics. Osteoarthritis Cartilage. 2022;30:216-25.

68.

Alves-Simões

Rodent models of knee osteoarthritis for pain research. Osteoarthritis Cartilage. 2022;30(6):802-14.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.13 MB

0.42 MB

1.34 MB

0.00 MB

0.06 MB

0.02 MB

0.15 MB

0.09 MB