Cynarin as a potent anti‑osteolytic agent: Targeting MAPK and Nrf2‑Keap1 pathways for osteoclast inhibition and bone protection

Introduction

Bone metabolism is essential for maintaining the integrity and function of the skeleton and ensuring a balanced process of bone formation and resorption. Under inflammatory conditions, bone metabolism is skewed with OC-mediated bone resorption outpacing bone formation. This imbalance accelerates bone loss and weakens the structural integrity of the skeleton (1). Therefore, the development of novel therapeutic strategies that simultaneously inhibit osteoclastogenesis and regulate inflammatory responses is crucial for alleviating inflammatory bone resorption and improving skeletal health.

OCs are essential effector cells in bone metabolism that are primarily responsible for local bone resorption and play a crucial role in inflammatory bone degradation. Multiple signalling pathways regulate OC differentiation and activity. Receptor activator of nuclear factor κB ligand (RANKL) activates the mitogen-activated protein kinase (MAPK) signalling pathway, which subsequently stimulates the expression of NFATc1, driving OC differentiation (2,3). The nuclear factor kappa-B (NF-κB) pathway (4,5) and the phosphoinositide 3-kinase (PI3K) pathway (6) also play significant roles in osteoclastogenesis and the maintenance of OC function. Inflammation is a crucial driver of bone degradation (7). Chronic inflammation induces oxidative stress, leading to increased production of reactive oxygen species (ROS), activating OCs and exacerbating bone resorption (8,9).

Current antiresorptive therapies, including bisphosphonates (10,11), effectively target OC formation but lack anti-inflammatory and antioxidant effects and have limited clinical utility (12,13). Therefore, an urgent need exists to develop novel agents with anti-inflammatory properties that inhibit abnormal OC production. Cynarin is a phenolic compound isolated from artichokes (Cynara genus) with potent antioxidant activity (14,15). In a mouse model of gouty arthritis, cynarin has been shown to inhibit the NF-κB and JNK pathways in macrophages, exerting anti-inflammatory and anti-swelling effects (16). In a mouse colitis model, it downregulated STAT3 and NF-κB (p65), thereby suppressing M1 macrophage polarisation (17). However, the mechanism of action of cynarin in inflammatory bone resorption remains unclear, and further research is required to address these questions.

The present study has evaluated the potential of cynarin as a dual-functional agent with anti-osteolytic and anti-inflammatory properties. Specifically, its effects were examined on RANKL-induced osteoclastogenesis, as well as on the MAPK signalling pathway, ROS production and Nrf2 activation in vitro. In addition, its efficacy in preventing inflammatory bone loss was evaluated in vivo. Notably, cynarin was more effective than alendronate in mitigating calvarial bone resorption. These findings highlight the ability of cynarin to inhibit OC formation and attenuate inflammation via a dual mechanism, supporting its potential as a novel therapeutic agent for treating inflammatory osteolysis.

Materials and methods

Reagents and antibodies

Cynarin (Selleck Chemicals) was used at a purity of 99.97% (high-performance liquid chromatography peak purity test and 13C NMR test data are available upon request from the corresponding author). Macrophage colony-stimulating factor (M-CSF) and RANKL were obtained from R&D Systems, Inc. Anisomycin and alendronate were purchased from MedChemExpress, and lipopolysaccharide (LPS) was sourced from MilliporeSigma. The tartrate-resistant acid phosphatase (TRAP) staining kit was acquired from Solarbio Science & Technology Co., Ltd. TRIzol reagent, PrimeScript™ RT reagent kit and TB Green® Premix Ex Taq™ II Fast qPCR kit were provided by Takara Biotechnology Co., Ltd. The bicinchoninic acid (BCA) protein assay kit, commercial nuclear and cytoplasmic protein extraction kit and phenyl-methyl-sulfonyl fluoride (PMSF) were obtained from Beyotime Institute of Biotechnology. Primary antibodies against histone H3 (cat. no. 9715), GAPDH (cat. no. 2118), phospho-(p)-JNK (cat. no. 4668), JNK (cat. no. 9252), p-ERK (cat. no. 4370), ERK (cat. no. 4695), p-p38 (cat. no. 4511), p38 (cat. no. 9212) and RANK (cat. no. 4845), as well as secondary antibodies (cat. nos. 7074 and 7076), were purchased from Cell Signalling Technology, Inc. Primary antibodies specific to heme oxygenase-1 (HO-1; cat. no. A1346) and RUNX2 (cat. no. A2851) were acquired from ABclonal Biotech Co., Ltd., while the DC-STAMP (cat. no. MABF39-I) antibody was obtained from MilliporeSigma. Antibodies targeting cathepsin K (CTSK; cat. no. ab19027), collagen I (COL1; cat. no. ab270993) and osteopontin (OPN; cat. no. ab218237) were purchased from Abcam. Antibodies against Nrf2 (cat. no. 16396-1-AP), Keap1 (cat. no. 10503-2-AP), osteocalcin (OCN; cat. no. 16157-1-AP), tumor necrosis factor-alpha (TNFα; cat. no. 17590-1-AP) and inducible nitric oxide synthase (iNOS; cat. no. 22226-1-AP) were sourced from Proteintech Group, Inc.

Cell culture and treatment

RAW264.7 cells (cat. no. TIB-71; American Type Culture Collection) were maintained in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 5% fetal bovine serum (FBS) and 100 U/ml penicillin-streptomycin (Thermo Fisher Scientific, Inc.). Bone marrow-derived macrophages (BMMs) were isolated and cultured according to the protocol established by Rucci et al (18). Primary BMMs were incubated in α-Minimum Essential Medium containing 10% FBS, 100 U/ml penicillin-streptomycin and 30 ng/ml M-CSF. Bone marrow-derived mesenchymal stem cells (BMSCs) were isolated from the femurs and tibias of 4-week-old mice, as previously described (19). BMSCs were cultured in the same medium as BMMs, excluding M-CSF supplementation. All cells were incubated under standard conditions in a humidified atmosphere at 37°C with 5% CO2 to ensure optimal proliferation and viability.

Cell viability assay

To evaluate the cytotoxic effects of cynarin, RAW264.7 cells and BMMs were seeded in triplicate into 96-well plates at a density of 8,000 cells per well. Cells were treated with varying concentrations of cynarin (1, 10, 50, 100, 200 and 400 μM) and incubated for 24, 48 and 96 h to assess short-, intermediate-, and long-term exposure effects. At each time point, 10 μl of Cell Counting Kit-8 (CCK-8; Proteintech Group, Inc.) solution was added to each well, and the plates were incubated for an additional 2 h. Cell viability was subsequently quantified by measuring the absorbance at 450 nm using a microplate reader.

OC formation and TRAP staining analysis

BMMs were seeded in triplicate in 96-well plates to ensure statistical robustness. After 24 h, osteoclastogenesis was induced by adding RANKL (50 ng/ml) in the presence of cynarin at 0, 10, 50, 100, or 200 μM. Culture medium was replenished every other day to maintain nutrient levels and compound stability. To preserve cellular morphology during the differentiation period, cells were fixed in 4% paraformaldehyde for 20 min at room temperature. Following fixation, TRAP staining was performed according to the manufacturer's protocol. TRAP-positive OCs were identified, counted, and images were captured using an Olympus BX53 microscope equipped for brightfield microscopy at appropriate magnification. Quantification of OC number and size was carried out with ImageJ software, ensuring objective assessment of differentiation across treatment groups.

Bone resorption assay

BMMs were cultured under the same conditions as aforementioned. To initiate OC-mediated bone resorption, cells were treated with RANKL (50 ng/ml) and cynarin (10, 50, 100, or 200 μM) until mature OC-like cells formed on bone slices. After removing the cells, the bone slices were fixed in a 4% paraformaldehyde + 2.5% glutaraldehyde mixed fixative for 24 h at 4°C. The samples after decalcification were dehydrated through a graded ethanol series, subjected to critical point drying, and sputter-coated with a 15 nm layer of gold/palladium. Images of resorption pits were captured by scanning electron microscopy (SEM; FEI Quanta 250).

Reverse transcription-quantitative PCR (RT-qPCR) analysis

For RT-qPCR, total RNA was extracted using TRIzol reagent, and its purity and concentration were assessed by measuring the A260/A280 ratios on a NanoDrop 2000/2000C spectrophotometer (Thermo Fisher Scientific, Inc.); only samples with a ratio between 1.9-2.1 were used. A total of 1 μg of RNA was reverse transcribed to cDNA using the PrimeScript™ RT reagent kit according to the manufacturer's instructions. qPCR was performed on a Light Cycler® 480 Instrument II (Roche) with TB Green® Premix Ex Taq™ II FAST Kit using the following protocol: Initial denaturation at 95°C for 30 sec; 40 cycles of 95°C for 5 sec and 60°C for 30 sec. Gene expression was calculated using the 2−ΔΔCq method (20). Primer sequences are listed in Table SI.

Western blotting (WB)

The treated cells were lysed on ice in RIPA buffer (cat. no. P0013C; Beyotime Institute of Biotechnology), and the lysates were clarified by centrifugation at 12,000 × g for 15 min at 4°C. A BCA kit was used to quantify protein concentrations. Equal amounts of protein (30 μg) were separated on 4-10% SDS-polyacrylamide gels (cat. no. M00657; GenScript) and transferred to PVDF membranes (Thermo Fisher Scientific, Inc.). Membranes were blocked in 5% skim milk at 37°C for 1 h, then incubated overnight at 4°C with primary antibodies (all at a dilution of 1:1,000). After washing, secondary antibodies (HRP-linked; 1:1,000) were administered and incubated at 37°C for 1 h to enable the detection of primary antibody-protein complexes. Following the removal of unbound secondary antibodies, bands were visualised using enhanced chemiluminescence (ECL; Thermo Fisher Scientific, Inc.) and images were captured on an Odyssey 9120 system (LI-COR Biosciences). The grey value of each band was analysed by ImageJ 24.0 software (National Institutes of Health).

Immunofluorescence (IF) analysis of podosome actin belt

OC differentiation was induced as aforementioned. Upon the appearance of multinucleated OCs in RANKL-stimulated control wells, cells were fixed in 4% paraformaldehyde for 20 min at room temperature to preserve cellular morphology. For intracellular antibody staining, cells were permeabilised with 0.5% Triton X-100 (MilliporeSigma) for 5 min, followed by washing to remove any residual detergent. F-actin structures were stained with FITC-conjugated phalloidin (MedChemExpress) for 30 min at room temperature, while nuclei were counterstained with DAPI (cat. no. C1006; Beyotime Institute of Biotechnology). All staining procedures were conducted in the dark to prevent photobleaching. After washing with phosphate-buffered saline (PBS), images of actin ring borders were captured using a Nikon Eclipse Ti inverted fluorescence microscope (Nikon Corporation). The distribution of podosome actin belts was quantified using ImageJ software.

Assessment of the generation of ROS

Intracellular ROS levels were assessed using the ROS detection kit (cat. no. S0033; Beyotime Institute of Biotechnology). Following thorough washing with PBS, cells were incubated with 2 μM/l 2',7'-dichlorodihydrofluorescein diacetate (DCFH-DA) for 20 min. The non-fluorescent DCFH-DA was internalised by cells and hydrolysed by intracellular esterases to form DCFH, which was subsequently oxidised by ROS to produce the highly fluorescent compound DCF. Fluorescence intensity was measured using a microplate reader, with excitation at 488 nm and emission detection at 525 nm.

Preparation of nuclear extracts and cytoplasmic extracts

RAW264.7 cells were washed with ice-cold PBS and then centrifuged at 500 × g for 5 min at 4°C. The cell pellet was resuspended in cytoplasmic protein extraction reagent A (containing PMSF) and vortexed vigorously for 5 sec. After incubating on ice for 15 min, Reagent B was added, and the mixture was vortexed for 1 min before ultracentrifugation at 4°C. The cytoplasmic fraction was extracted from the resulting supernatant. For nuclear protein extraction, nuclear protein extraction reagent (containing PMSF) was added to the pellet. The sample was vortexed at high speed for 30 min, followed by ultracentrifugation. The nuclear fraction was collected from the supernatant. Protein concentrations were quantified as aforementioned.

RNA sequencing (RNA-seq) followed by bioinformatics evaluation

RAW264.7 cells were cultured under OC differentiation conditions for 5 days, with or without cynarin. Total RNA was extracted using TRIzol reagent and stored at -80°C. RNA quality and integrity were verified using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Inc.) to ensure A260/A280 ratios between 1.9-2.1, and an Agilent 2100 Bioanalyzer with the RNA Nano 6000 Assay Kit (Agilent Technologie, Inc.s) to confirm RNA Integrity Number values >9.0. RNA sequencing was performed by Beijing Biomarker Technologies Co., Ltd. (http://www.biomarker.com.cn/) following their standard Illumina sequencing protocols. Differential expression analysis was performed using DESeq2 (version 1.20.0). Raw counts were normalised using the median-of-ratios approach implemented in DESeq2, and differential expression was defined as |Log2FC|≥0.58 (FC ≥1.5) with unadjusted P<0.05. Volcano plots were generated to visualise distribution of differentially expressed genes (DEGs), highlighting the magnitude and statistical significance of expression changes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were conducted using the BioMarker BioCloud Platform (www.biocloud.org) (21).

LPS-induced skull osteolysis in vivo mouse model

Animal experiments were conducted in accordance with the laboratory animal management guidelines. They were approved by the Animal Ethics Committee of the Ninth People's Hospital, affiliated with Shanghai Jiao Tong University School of Medicine (approval no. SH9H-2020-A1-1; Shanghai, China). Mice were housed with a standard temperature (20-26°C) and humidity (50-60%), under a 12-h light/dark cycle and fed standard rodent diet and water ad libitum. Mice were monitored daily for weight loss and general health and euthanized when humane endpoints (signs of severe distress, unresponsiveness to treatment, and/or significant weight loss) were reached. A calvarial osteolysis model was established based on previously reported methodologies (22,23). A total of 48 male C57/BL6 mice (6-8 weeks, 18-22 g) were randomly assigned to six groups: i) Sham group (subcutaneous injection of 50 μl PBS); ii) LPS group (subcutaneous injection of 10 mg/kg LPS in 50 μl); iii) alendronate group (subcutaneous injection of 10 mg/kg LPS in 50 μl; intraperitoneal injection of 10 μg/kg alendronate in 100 μl) (24,25); iv) cynarin low-dose group (subcutaneous injection of 10 mg/kg LPS and 100 μM cynarin in 50 μl); v) cynarin medium-dose group (subcutaneous injection of 10 mg/kg LPS and 250 μM cynarin in 50 μl); and vi) cynarin high-dose group (subcutaneous injection of 10 mg/kg LPS and 500 μM cynarin in 50 μl). During the modelling process, anaesthesia was induced via intraperitoneal injection of a 3% sodium pentobarbital solution at a dose of 50 mg/kg. Collagen sponges (4×4×2 mm), soaked in either PBS or LPS, were implanted at the sagittal midline suture of the calvaria to induce bone loss. All mice received subcutaneous or intraperitoneal injections every other day for 10 days. No mice reached humane endpoints. The mice were euthanised by filling with carbon dioxide at the rate of 30% CO2 tank volume/min. Tissues were collected after death confirmation: Absent pulse and breathing, lost corneal reflexes, no response to deep toe stimulation and mucosal greying. Blood was collected from anesthetised mice's eyeballs and transferred into heparinised anticoagulant tubes. After centrifugation, the plasma was separated and used to measure biochemical indicators, including alanine aminotransferase (ALT), aspartate aminotransferase (AST) and blood urea nitrogen (BUN), for assessing liver and kidney function. The major organs (heart, liver, spleen, lungs and kidneys) were dissected and fixed for haematoxylin and eosin (H&E) staining. The calvarial bones were isolated and preserved for further analysis.

Micro-computed tomography (CT) scanning

Micro-CT scans were performed using a SkyScan 1076 system (Bruker Corporation) with a resolution of 9 μm. The scan parameters were set to 29 kV, 175 μA, and a 300 msec exposure time. Structural features of the bone were quantified using SCANCO software (version 6.5-3; Scanco Medical AG).

Histological staining and analysis

After decalcifying in EDTA solution for 2 weeks, the skull was paraffin-embedded, and 4-μm sections were prepared. Serial sections were deparaffinized in xylene and rehydrated through a graded ethanol series. H&E and TRAP staining were performed to evaluate tissue morphology and OC activity. For immunohistochemical (IHC) staining, endogenous peroxidase activity was quenched by incubation in 3% H2O2 in methanol for 15 min at room temperature. Non-specific binding was blocked with 5% normal goat serum (Beijing Solarbio Science & Technology Co., Ltd.) in PBS for 1 h at room temperature. Sections were then incubated with primary antibodies (RANK, CTSK, OCN, RUNX2, TNFα, iNOS, p-P38, p-ERK1/2 and p-JNK1/2, all at a dilution of 1:100) overnight at 4°C. After washing, sections were incubated with HRP-conjugated secondary antibodies (1:500; Thermo Fisher Scientific, Inc.) for 1 h at room temperature. Signal was developed using a DAB substrate kit (cat. no. DA1016; Beijing Solarbio Science & Technology Co., Ltd.) and sections were counterstained with haematoxylin. For IF staining, a similar protocol was followed. After blocking, sections were incubated with primary antibodies against Nrf2 (1:100) and Keap1 (1:100) overnight at 4°C, followed by incubation with Alexa Fluor® 647-conjugated secondary antibodies (1:500; Cell Signalling Technology, Inc.) for 1 h at room temperature in the dark. Nuclei were counterstained with DAPI. Stained sections (H&E, TRAP, IHC) were examined using a Leica DM4000B light microscope. IF-stained sections were examined using the same microscope equipped for epifluorescence. Positive staining was quantified using ImageJ software.

Statistical analysis

Group data are presented as the mean ± standard deviation (SD) to indicate average values and variability. All statistical analyses were conducted using GraphPad Prism software (version 10.3.1; Dotmatics). One-way analysis of variance (ANOVA) followed by Tukey's multiple comparisons post hoc test was used for comparisons of multiple groups. P<0.05 was considered to indicate a statistically significant difference.

Results

Cynarin attenuates LPS-induced calvarial osteolysis in vivo

An LPS-induced osteolytic murine model was used to assess the therapeutic effects of cynarin on inflammatory bone loss. Throughout the experimental period, all the subjects maintained stable conditions until a specified terminal time point. Examination of histological sections from the vital organs of the mice demonstrated no discernible damage, suggesting favourable biocompatibility of the drug (Fig. S1A). Serum biochemistry showed mild elevations in hepatic indices (ALT, AST and total bilirubin) in the LPS group, which were found to be significantly improved by alendronate and a medium-dose of cynarin (P<0.05 vs. LPS for all). The renal indices (BUN, creatinine and uric acid) remained unchanged among the groups (Fig. S1B). Micro-CT demonstrated that the medium-dose cynarin (Cy-M) group had preserved bone microarchitecture, as evidenced by increased bone volume, trabecular number, trabecular thickness and decreased trabecular separation relative to LPS alone. These changes were comparable to those observed with alendronate (Fig. 1A and B). H&E staining confirmed reduced bone porosity in the Cy-M group (Fig. 1C and D), and TRAP staining revealed a significant reduction in the number of TRAP-positive OCs (Fig. 1C and E).

Figure 1 Cynarin shows therapeutic efficacy comparable to alendronate in murine model of LPS-induced osteolysis. (A) The micro-computed tomography reconstructions of each group's murine calvaria. 6 weeks old C57/B6 mice were used to establish model and were administrated by cynarin in indicated dose for 10 days. (B) BV/TV, Tb.N, Tb.Th and Tb.Sp were evaluated by quantitative analysis. (C) Quantitative analysis of porosity percentage in H&E staining and TRAP positive cells number in TRAP staining. (D) Representative images of calvarial histology stained with H&E (magnification, ×100). (E) Representative images of calvarial histology stained with TRAP (magnification, ×100). TRAP-positive cells are shown by black arrows. Animals were randomly assigned to each group using a computer-generated sequence. Histological scoring was conducted by an investigator who was blinded to the treatment groups. All experiments were performed with three independent biological replicates (n=3). The data are presented as the mean ± SD. *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001 vs. Sham and LPS group separately. LPS, lipopolysaccharide; BV/TV, bone volume to total volume ratio; Tb.N, trabecular number; Tb.Th, trabecular thickness; Tb.Sp, trabecular separation; TRAP, tartrate-resistant acid phosphatase; Alen, alendronate; Cy-M, medium-dose cynarin.

IHC further showed that cynarin reduced RANK and CTSK expression while also increasing OCN and RUNX2 expression in calvarial sections (Fig. 2A and B), suggesting the inhibition of OC differentiation and the support of osteoblast activity. Although cynarin modestly enhanced osteoblast differentiation in vitro, the mechanism was not investigated in the present study (Fig. S2). Moreover, cynarin significantly decreased TNF-α and iNOS expression in calvarial tissues (Fig. 2C and D), indicating suppression of inflammatory signalling. Inflammatory signals play a crucial role in the pathogenesis of osteolysis, and inhibition of inflammation helps alleviate osteolysis. These findings suggest that cynarin effectively preserves the bone structure and mitigates bone resorption by inhibiting OC differentiation and inflammation.

Figure 2 Cynarin exerts a more potent effect in inhibiting osteoclast differentiation and reducing inflammation in vivo. (A) Representative images of calvarial histology stained with RANK, CTSK, OCN and RUNX2 (magnification, ×100). (B) Quantitative analysis of the expression of RANK, CTSK, OCN and RUNX2 in the IHC staining. (C) Quantitative analysis of the expression of TNFα and iNOS in the IHC staining. (D) Representative images of calvarial histology stained with TNFα and iNOS (magnification, ×100). All experiments were performed with three independent biological replicates (n=3). The data are presented as the mean ± SD. *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001 vs. Sham, LPS and Alen group separately. RANK, receptor activator of nuclear factor κB; CTSK, cathepsin K; OCN, osteocalcin; IHC, immunohistochemical; iNOS; inducible nitric oxide synthase; LPS, lipopolysaccharide; Alen, alendronate; Cy-M, medium-dose cynarin.

Cynarin inhibits OC differentiation and bone resorption in vitro

The chemical formula of cynarin is illustrated in Fig. 3A. The performed cell viability test revealed no cytotoxic effects of cynarin at concentrations <200 μM (Fig. 3B). To explore the direct effects of cynarin on OC differentiation, BMMs and RAW264.7 cells treated with RANKL were used to induce osteoclastogenesis. TRAP staining revealed a significant dose-dependent reduction in the number of TRAP-positive OCs after cynarin treatment (Fig. 3C-E). The performed bone resorption assay revealed that cynarin markedly reduced resorption pits area at concentrations exceeding 50 μM after 9 days of differentiation (Fig. 3F).

Figure 3 Cynarin restrains RANKL-induced OC differentiation in vitro. (A) Chemical formula for cynarin. (B) BMMs and RAW264.7 cells were treated with gradient concentrations of cynarin for 24, 48 and 96 h, after which cells proliferation and viability were detected by Cell Counting Kit-8 assay (n=3). *P<0.05 and **P<0.01 vs. CTL group. (C) BMMs were incubated with gradient concentrations of cynarin for 5-7 days. Cells were stained for TRAP staining. Scan of a 96-well plate after TRAP staining (n=3). (D) The count of TRAP+ OCs per well was determined (n=3). *P<0.05 and **P<0.01 vs. RANKL (+) and Cynarin (-) group. (E) Representative images of TRAP staining (magnification, ×200; (n=3). (F) BMMs were incubated with gradient concentrations of cynarin for 9 days. The bone slides were visualised under the scanning electron microscope (magnification, ×50, first row; ×300, second row). Black arrows indicate osteoclastic absorption pits (n=3). The data are presented as the mean ± SD. RANKL, receptor activator of nuclear factor κB ligand; OC, osteoclast; BMMs, bone marrow-derived macrophages; TRAP, tartrate-resistant acid phosphatase.

Further molecular analysis using RT-qPCR showed that cynarin lowered the expression of Nfatc1, Dcstamp, Trap, Atp6v0d2, Ctsk and Ctr (Fig. 4A and D). Notably, the suppressive effect of cynarin on OC-related gene expression was most prominent on day four (Fig. 4D). WB revealed that cynarin reduced DC-STAMP and CTSK protein levels during RANKL-induced differentiation (Fig. 4B). Furthermore, Actin staining revealed that cynarin blocked the fusion of BMM precursor cells and development of podosomal actin structures (Fig. 4C). Collectively, these results suggested that cynarin inhibited OC differentiation and bone resorption.

Figure 4 Cynarin downregulates osteoclast-specific genes and proteins during osteoclastogenesis. (A) RAW264.7 cells were treated with RANKL and gradient concentrations of cynarin for 5 days. Dcstamp, Trap, Ctsk and Ctr expression levels were normalised to Gapdh expression by RT-qPCR (n=3). (B) RAW264.7 cells were treated with RANKL and gradient concentrations of cynarin for 5 days. Expression levels of DC-STAMP and CTSK were detected by western blotting. Quantitative densitometric analysis was performed to normalize DC-STAMP and CTSK (n=3). (C) After 5 days of macrophage colony-stimulating factor and RANKL induction, bone marrow-derived macrophages were stained for immunofluorescence analysis of podosome actin belt (magnification, ×200; (n=3). (D) RAW264.7 cells were incubated with or without 100 μM cynarin for 1, 2 and 4 days. Atp6v0d2, Trap, Ctsk, and Nfatc1 expression levels were normalised to GAPDH expression by RT-qPCR (n=3). The data are presented as the mean ± SD. *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001. RANKL, receptor activator of nuclear factor κB ligand; RT-qPCR, reverse transcription-quantitative PCR; CTSK, cathepsin K; TRAP, tartrate-resistant acid phosphatase; OC, osteoclast.

Cynarin activates Nrf2 to suppress inflammation and oxidative stress in vitro

Since inflammation plays a critical role in promoting osteoclastogenesis, the anti-inflammatory effects of cynarin on LPS-stimulated RAW264.7 macrophages were investigated. Cynarin treatment significantly reduced the expression of proinflammatory cytokines, including Tnf, IL1β, IL6 and Nos2 at the mRNA level (Fig. 5A). Additionally, ROS levels were significantly reduced in the cynarin-treated macrophages (Fig. 5B), indicating that cynarin attenuated oxidative stress, which is a key contributor to inflammation.

Figure 5 Cynarin suppresses macrophage-mediated inflammation through activation of the Nrf2-Keap1 signalling pathway. (A) RAW264.7 cells were exposed to cynarin (10, 50 or 100 μM) for 24 h and then treated with LPS (1 μg/ml) for 12 h. Expression levels of IL1β, IL6, Nos2 and Tnf were detected by RT-qPCR (n=3). (B) RAW264.7 cells and bone marrow-derived macrophages were exposed to cynarin (10, 50, 100 or 200 μM) for 24 h and then treated with LPS (1 μg/ml) for 12 h. Intracellular ROS levels were measured with a ROS assay kit (n=3). (C) RAW264.7 cells were exposed to cynarin (10, 50, 100 or 200 μM) for 6 h and then treated with LPS (1 μg/ml) for 12 h. Expression levels of Nrf2, Keap1, Hmox1 and Cat were detected by RT-qPCR (n=3). (D) RAW264.7 cells were exposed to cynarin (10, 50, 100 or 200 μM) for 24 h and then treated with LPS (1 μg/ml) for 24 h. Expression levels of Nrf2, Keap1 and HO-1 were detected by WB. (E) RAW264.7 cells were exposed to 100 μM cynarin for 24 h and then treated with LPS (1 μg/ml) for 24 h. Nuclear extracts and cytoplasmic extracts were used to test expression of Nrf2 by WB. (F) Quantitative densitometric analysis was performed to normalize Nrf2, Keap1 and HO-1 expression levels from D (n=3). (G) Quantitative densitometric analysis was performed to normalize Nrf2 expression levels from E (n=3). The data are presented as the mean ± SD. *P<0.05, **P<0.01 and ****P<0.0001 vs. CTL and LPS group separately. LPS, lipopolysaccharide; RT-qPCR, reverse transcription-quantitative PCR; ROS, reactive oxygen species; HO-1, heme oxygenase-1; WB, western blotting.

Mechanistically, cynarin treatment led to the upregulation of Nrf2 expression and the downregulation of Keap1, thus resulting in the induction of antioxidant enzymes, such as HO-1 and catalase (Fig. 5C, D and F). Moreover, nuclear fractionation assays confirmed that cynarin promoted Nrf2 nuclear translocation, validating its role in reducing oxidative stress and inflammation via Nrf2 activation (Fig. 5E and G). These results support the hypothesis that cynarin exerts its anti-inflammatory effects by activating the Nrf2-Keap1 pathway activation.

Cynarin suppresses MAPK signalling and engages the Nrf2-Keap1 axis

To further elucidate the mechanism of cynarin suppressing OC differentiation, cynarin was used for RNA-seq in RANKL-induced RAW264.7 cells. A volcano plot illustrates that cynarin treatment upregulated 499 genes and downregulated 726 genes (Fig. 6A). To assess the possible impact of cynarin on biological processes, GO enrichment analysis was conducted. As depicted in Fig. 6B, the regulation of OC differentiation and inflammatory response were prominently associated with the therapeutic mechanisms of cynarin. KEGG pathway enrichment analysis demonstrated profound modulation of the MAPK signalling pathway following cynarin treatment (Fig. 6C).

Figure 6 Cynarin suppresses the MAPK signalling pathway in RANKL-induced osteoclastogenesis. (A) RANKL induced RAW264.7 cells for 5 days with or without 100 μM cynarin. Total RNA was extracted and sent for RNA sequencing. Genes exhibiting |Log2FC|≥0.58 (FC ≥1.5) and unadjusted P≤0.05 were considered to be differentially expressed. Volcano plot of the distinct upregulated and downregulated genes of Cy vs. OC (n=3). (B) Gene Ontology enrichment analysis of differential genes of Cy vs. OC (n=3). (C) Kyoto Encyclopedia of Genes and Genomes pathway analysis of differential genes of Cy vs. OC (n=3). (D) RAW264,7 cells were pretreated with 100 μM cynarin for 6 h and stimulated with 50 ng/ml RANKL for 10 to 60 min. Expression levels of total and phosphorylated forms of JNK, ERK, and P38 were detected by WB. (E) RAW264.7 cells were treated with 100 μM cynarin, 10 μM anisomycin for 30 min and stimulated with 50 ng/ml RANKL for 30 min. The ratio of phosphorylated JNK to JNK, p-ERK to ERK and p-P38 to P38 were quantitatively determined by WB. (F) Quantitative densitometric analysis was performed to normalize p-JNK/JNK, p-ERK/ERK and p-P38/P38 expression levels from D (n=3). (G) Quantitative densitometric analysis was performed to normalize p-JNK/JNK, p-ERK/ERK and p-P38/P38 expression from E (n=3). (H) RAW264,7 cells were treated with 100 μM cynarin, 10 μM anisomycin for 4 days in RANKL-induced osteoclastogenesis. Expression levels of OC-specific genes were detected by reverse transcription-quantitative PCR (n=3). The data are presented as the mean ± SD. *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001. RANKL, receptor activator of nuclear factor κB ligand; OC, osteoclast; p-, phosphorylated; WB, western blotting; Ani, anisomycin.

WB demonstrated that cynarin attenuated RANKL-induced phosphorylation of JNK (10-30 min), ERK (10-20 min), and p38 (20 min) (Fig. 6D and F), indicating that cynarin suppressed osteoclastogenesis by inhibiting MAPK signalling. To confirm this mechanism, anisomycin (MAPK activator) was used. WB revealed that anisomycin counteracted the inhibitory effects of cynarin on p-JNK, p-ERK and p-P38 (Fig. 6E and G). RT-qPCR confirmed that anisomycin reversed the inhibitory effects of cynarin on OC differentiation (Fig. 6H). These findings provide strong evidence that cynarin inhibited osteoclastogenesis by suppressing MAPK signalling.

In vivo validation of the pathway regulation showed that cynarin significantly increased Nrf2 expression, decreased Keap1 expression, and promoted Nrf2 nuclear translocation in calvarial tissues (Fig. 7A and C). Furthermore, IHC staining showed that cynarin administration significantly decreased p-P38, p-ERK1/2 and p-JNK1/2 levels compared with those in the LPS group (Fig. 7B and D). Collectively, cynarin exerts dual regulatory effects in vivo by inhibiting OC differentiation by suppressing the MAPK pathway and mitigating inflammation and oxidative stress through activation of the Nrf2-Keap1 axis. These combined actions therefore underscore the potential of cynarin as a therapeutic candidate for the treatment of inflammatory bone diseases.

Figure 7 Cynarin suppresses osteoclastogenesis and inflammation via the MAPK and Nrf2-Keap1 pathways in inflammatory bone resorption. (A) Representative images of calvarial histology stained with Nrf2 and Keap1 (magnification, ×40, first row; ×200, second row; ×40, third row). (B) Representative images of calvarial histology stained with p-ERK1/2, p-JNK1/2 and p-P38 (magnification, ×100). (C) Quantitative analysis of the expression of Nrf2 and Keap1 in the immunofluorescence staining. (D) Quantitative analysis of the expression of p-ERK1/2, p-JNK1/2 and p-P38 in the immunohistochemical staining. All experiments were performed with three independent biological replicates (n=3). The data are presented as the mean ± SD. *P<0.05, **P<0.01 and ****P<0.0001. p-, phosphorylated; Cy-M, medium-dose cynarin; LPS, lipopolysaccharide; Alen, alendronate.

Discussion

The development of novel therapeutic strategies that simultaneously suppress osteoclastogenesis and modulate inflammation is critical for alleviating inflammatory bone resorption and improving bone health. Naturally derived plant compounds have garnered considerable attention because of their favourable biological activities and safety profiles. Cynarin, a bioactive compound extracted from artichoke (Cynara scolymus), shows promise in this regard. The present study aimed to evaluate the therapeutic potential of cynarin in the prevention of bone resorption and inflammation-induced bone loss. These findings demonstrate that cynarin significantly inhibits OC differentiation and activity, suppresses inflammatory responses, and modulates the key signalling pathways involved in osteoclastogenesis and inflammation.

Our in vivo findings have shown that cynarin effectively reduced LPS-induced osteolysis and preserved the bone structure, with effects comparable to those of the widely used drug alendronate sodium. Improvements in the liver and kidney function indices following cynarin treatment provide a critical basis for assessing its safety and therapeutic potential. However, further clinical translation will require comprehensive pharmacokinetic studies and the optimisation of the dosage forms to enhance bioavailability and determine the appropriate dosage. Histological evaluation showed decreased bone porosity and fewer TRAP-positive OCs in the cynarin-treated mice. Additionally, cynarin significantly reduced the expression of key OC differentiation markers, suggesting that it inhibited OC differentiation and activity. These findings align with those of previous studies that emphasised the bone protective effects of various natural compounds in inflammatory bone diseases. Well-known natural anti-inflammatory agents such as resveratrol and curcumin show similar protective effects by reducing OC activity (26). One limitation of the study is that only male mice were used. Although this reduces confounding variables such as the periodic fluctuations of female hormones during the initial efficacy screening, it is necessary to be cautious when applying our conclusions directly to female models or patients. Future work will include both sexes to rigorously assess the potential gender-specific effects of cynarin on cranial bone repair. However, the present study offers a new perspective by demonstrating that cynarin treatment significantly inhibits inflammation and downregulates the levels of inflammatory factors such as TNF-α and iNOS. Osteolysis is closely associated with proinflammatory cytokine release and OC activation (27,28). In conclusion, cynarin may prevent osteolysis and preserve the bone microstructure by regulating OC production and inflammation, providing a dual mechanism of action.

The inhibitory effect of cynarin on osteoclastogenesis observed in our in vitro study was consistent with the findings of our in vivo study. It was next discovered that cynarin concentrations >50 μM significantly suppressed OC formation as evidenced by TRAP staining and SEM. To the best of our knowledge, this finding has not been previously reported. The performed cell viability assays confirmed that cynarin concentrations <200 μM exhibited no significant cytotoxicity. It was demonstrated that varying concentrations of cynarin suppressed Nfatc1, Dcstamp, Trap, Ctsk and Atp6v0d2, and 100 μM was identified as the optimal concentration. FITC-labelled phalloidin staining indicated that 100 μM cynarin significantly inhibited OC development. This effect was most evident after four days of RANKL exposure.

Accumulating evidence has suggested that Nrf2 is a key regulator of bone oxidative homeostasis. Transcriptome analysis of Nrf2 or Keap1 knockout models revealed that Nrf2 deficiency upregulates genes associated with mitochondrial oxidative phosphorylation, including Car2, Calcr and Mmp12, and increases OC production under oxidative stress (29). Dong et al (30) found that the Nrf2 activator bitopertin blocks Keap1-Nrf2 binding, reduces intracellular iron levels, and inhibits OC formation. Nrf2 regulates the activation of genes that enhance antioxidant defence, thereby reducing ROS accumulation (31). Under normal conditions, Nrf2 is maintained at low levels because of its binding to Keap1. When Keap1 is inhibited, Nrf2 translocates to the nucleus where it activates genes involved in detoxification and antioxidant responses. It was demonstrated that cynarin inhibits Keap1 during inflammation, promotes Nrf2 nuclear translocation, and enhances the expression of HO1 and Cat. In vivo experiments confirmed that, compared with LPS and alendronate, medium-dose cynarin treatment significantly increased Nrf2 expression, decreased Keap1 expression, and activated nuclear translocation. Inflammatory conditions are commonly associated with elevated ROS, which are key regulators of bone metabolism and function (32,33). In addition, it was demonstrated that cynarin significantly reduces macrophages' inflammatory response, inhibiting the production of IL-1β, IL-6, iNOS and TNF-α while also reducing ROS production. The present results are consistent with those of previous studies showing that Nrf2 activation is critical for controlling inflammation and preventing cellular oxidative damage (34,35). The present study, to the best of our knowledge, is the first to demonstrate that cynarin regulates both oxidative stress and inflammation through the Nrf2-Keap1 signalling pathway. These results suggest that cynarin enhances the antioxidant response by activating the Nrf2 pathway, thereby inhibiting oxidative stress-driven OC differentiation. Compounds such as curcumin, resveratrol and epigallocatechin-3-gallate (EGCG) have been demonstrated to inhibit OC differentiation and activity (26,36). However, curcumin, resveratrol and EGCG exerted their effects mainly through inhibition of NF-κB and MAPK pathways, while the additional activation of Nrf2-Keap1 by cynarin provided anti-inflammatory effects. These findings, therefore, highlight the therapeutic potential of cynarin and provide valuable insights into its use as a dual-target therapeutic agent for inflammatory bone diseases.

The RNA-seq results indicated that cynarin inhibited RANKL-induced OC differentiation via the MAPK pathway. MAPK are well-known regulators of cell proliferation and differentiation (37,38). The current findings indicated that cynarin significantly inhibited P38, JNK and ERK phosphorylation. This inhibition was reversed by treatment with anisomycin (a MAPK agonist). Our in vivo experiments have also confirmed that moderate-dose cynarin treatment significantly reduced the expression of p-P38, p-ERK1/2 and p-JNK1/2 compared with LPS treatment and alendronate sodium. In conclusion, cynarin suppressed RANKL-induced OC differentiation by targeting the MAPK pathway.

Increasing evidence suggests a crosstalk between MAPK and Nrf2 signalling during osteoclastogenesis. Nrf2 activation elevates the expression of detoxifying enzymes (for example, HO-1 and NQO1) that reduce intracellular ROS, thereby attenuating ROS-dependent MAPK activation and downstream osteoclastogenic transcription factors such as NFATc1. In osseous echinococcosis, Echinococcus granulosus-driven inhibition of Nrf2 shifts the balance toward unchecked MAPK activation and OC differentiation (39). By contrast, carnosic acid-induced Nrf2 restores antioxidant defences and suppresses MAPK phosphorylation, forming a negative feedback loop that limits RANKL-driven osteoclastogenesis (40). However, the interaction between the Nrf2 pathway and MAPK signalling in the present study remains to be elucidated. In future studies, the authors plan to use Nrf2 inhibitors such as ML385 and Nrf2-specific small interfering RNA to transfect OC precursor cells. It was further elucidated whether Nrf2 knockdown acts upstream of MAPK inhibition. The effects of cynarin on osteoblast function were also investigated. The results showed that cynarin promoted the expression of key osteogenic markers, such as OCN and RUNX2, in both in vivo and in vitro models. A potential role of cynarin in supporting bone formation has also been demonstrated. However, further studies are required to fully understand how cynarin affects osteoblast activity at the molecular level.

Beyond transcriptional regulation, epigenetic alterations such as DNA methylation, histone acetylation/deacetylation and non-coding RNA expression are increasingly being recognised as diagnostic biomarkers and therapeutic entry points for OC-driven bone loss. Clinically applicable assays, such as methylation-specific PCR or targeted bisulphite sequencing panels, can quantify the promoter methylation of osteoclastogenic genes or inflammatory mediators. Circulating microRNAs (for example miR-21, miR-155 and miR-223) have been proposed as minimally invasive biomarkers that mirror OC activity and inflammatory status. Notably, the concept of leveraging nucleotide-level variation to stratify patients has already been demonstrated in other inflammatory conditions. For example, Antonino et al (41) systematically reviewed single-nucleotide polymorphisms associated with chronic rhinosinusitis, underscoring the translational value of molecular variation in diagnosis and management. Given that both the MAPK and Nrf2-Keap1 axes are subject to epigenetic regulation, integrating epigenetic testing could help identify patients who are most likely to benefit from cynarin or combination regimens.

In conclusion, the present study has identified cynarin as a potent anti-inflammatory and bone-preserving agent and demonstrated its ability to inhibit OC differentiation and modulate inflammatory processes via the MAPK and Nrf2-Keap1 pathways for the first time (Fig. 8). This dual-target therapeutic strategy effectively addresses OC-driven bone resorption while mitigating the inflammatory processes associated with diseases, such as rheumatoid arthritis and osteoporosis. Given the ability of cynarin to modulate bone resorption and inflammation, future clinical applications should explore its potential to enhance bone healing in implant therapies or in combination treatments for chronic inflammatory bone diseases.

Figure 8 Schematic representation of the mechanism by which cynarin inhibits bone loss. RANK, receptor activator of nuclear factor κB; RANKL, RANK ligand; LPS, lipopolysaccharide; ROS, reactive oxygen species.

Supplementary Data

Availability of data and materials

The data generated in the present study may be requested from the corresponding author. The data generated in the present study may be found in the NCBI under accession number PRJNA1301579 or at the following URL: https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1301579.

Authors' contributions

RC conceptualized the study, developed methodology and wrote the original draft. YXW curated data and developed methodology. ZL developed methodology and validated data. THW and YM curated data. XRX and LS validated data. WFX and XZC conceptualized the study. SYZ conceptualized and supervised the study, and acquired funding. All authors contributed to writing, reviewing and editing the manuscript. RC and SYZ confirm the authenticity of all the raw data. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

Animal experiments were conducted in accordance with the laboratory animal management guidelines. They were approved (approval no. SH9H-2020-A1-1) by the Animal Ethics Committee of the Ninth People's Hospital of Shanghai Jiao Tong University School of Medicine (Shanghai, China).

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Acknowledgements

Not applicable.

Funding

The present study was supported by the National Natural Science Foundation of China (grant nos. 82370979 and 82301108), Shanghai's Top Priority Research Center (grant no. 2022ZZ01017), CAMS Innovation Fund for Medical Sciences (grant no. 2019-I2M-5-037), Cross-disciplinary Research Fund of Shanghai Ninth People's Hospital, Shanghai JiaoTong University School of Medicine (grant no. JYJC202218) and the Program of Shanghai Academic/Technology Research Leader (grant no. 21XD1431500).

References