Coordination between osteoblasts, responsible for bone formation, and osteoclasts, tasked with bone resorption, is important in maintaining bone homeostasis and the structural integrity of the skeleton. The essential components, macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor κB ligand (RANKL), play a defining role in steering hematopoietic stem cells towards the osteoclast lineage for bone tissue production (1, 2). The progression of commonly observed bone disorders like osteoporosis, rheumatoid arthritis, and lytic bone metastasis is driven by the gradual and excessive bone resorption carried out by osteoclasts (3, 4). Osteoporosis is a systemic skeletal condition marked by the reduction of bone mass and diminished bone mineral density, and compromised bone strength. Osteoporosis is usually considered an age-related manifestation, notably among postmenopausal women. Several studies have indicated that osteoclast populations and activity increase further after menopause during the pathogenesis of osteoporosis (5, 6). Its management is one of the greatest challenges faced by modern medicine. Bone resorption inhibitors, such as estrogen, calcitonin, and bisphosphonates, are traditional therapeutic agents for osteoporosis. Although these drugs seem to be the most effective methods to lower the bone loss rate after menopause, they can be accompanied by serious side effects such as venous thromboembolism, rhinitis, breast cancer, or esophageal ulcers (7, 8).
Belonging to the TNF-family, RANKL holds significance as a key cytokine promoting osteoclast formation (9). Interaction between RANKL and the receptor for RANK (known as RANK) on osteoclast precursors promotes the enlistment of TNF receptor-associated factor 6 (TRAF6), subsequently initiating cascading downstream signaling pathways, including NF-κB, Akt, c-jun N-terminal protein kinase (JNK), p38, and extracellular signal-related kinase (ERK), to promote the upregulation of genes associated with osteoclastogenesis, including c-fos, osteoclast-associated receptor (Oscar), tartrate-resistant acid phosphatase (TRAP, Acp5), and nuclear factor of activated T cells c1 (Nfatc1) (10-12). Among the genes that are stimulated, NFATc1 plays a particularly significant role as a crucial regulator of osteoclastogenesis. (13).
Natural products are valuable resources for developing new herbal drugs, such as the widely used ginseng. The roots of Panax ginseng Meyer (Araliaceae) are rich in highly bioactive phytomedicine widely used in traditional Korean, Chinese, and Japanese medicine. Studies so far have identified approximately 30 different forms of ginsenosides as the dominant active component exhibiting a variety of pharmacological effects on cancer and on the cardiovascular, endocrine, nervous, and immune systems, among others (14-17). Although the anti-osteoporotic effects of ginseng have been examined (18, 19), the precise underlying molecular mechanism remains unknown. Ginsenoside Rg2 is a protopanaxatrial-type compound and one of the major active components found in the roots, stem, and leaves of ginseng. Several recent studies have suggested that ginsenoside Rg2 exerts various biological effects, including anti-apoptotic and anti-diabetic effects (20, 21). Furthermore, Rg2 plays a role in cardiovascular protection, neuroprotection, and inflammatory factor regulation (22, 23). Osteoclastogenesis is closely related to inflammatory regulation (24), but the role of Rg2 on the process of osteoclastogenesis remains unexplored.
Therefore, this study assessed the impacts of ginsenoside Rg2 on osteoclast differentiation induced by RANKL and elucidated the underlying mechanisms by which Rg2 inhibits osteoclastogenesis. The results demonstrated that Rg2 is a ginsenoside component with anti-osteoporotic activity.
First, the impact of ginsenoside Rg2 on the osteoclast differentiation of BMMs was examined under RANKL influence using tartrate-resistant acid phosphatase (TRAP) staining. RANKL-induced osteoclast differentiation was determined based on the emergence of multinucleated cells positive for TRAP staining was observed. Ginsenoside Rg2 demonstrated a dose-dependent decrease in osteoclast differentiation, as reported (Fig. 1A, B). The TRAP assay further supported the suppressive impacts of Rg2 on osteoclast differentiation induced by RANKL (Fig. 1C). These findings suggest that Rg2 strongly inhibited osteoclast differentiation of BMMs. Following this, the cytotoxicity of Rg2-treated BMMs was evaluated using the MTT assay. The results indicated that Rg2 did not display any cytotoxicity, further substantiating the inhibitory property of ginsenoside Rg2 (Fig. 1D).
Given that the pivotal osteoclast-specific transcription factors include c-Fos and NFATc1 (25, 26), their expression was analyzed to comprehend the inhibitory mechanism by which ginsenoside Rg2 counteracts osteoclastogenesis. qPCR analyses showed that administration of ginsenoside Rg2 before RANKL stimulation notably inhibited the expression of both c-fos and Nfatc1 at the transcriptional level (Fig. 2A, B), as well as osteoclast-specific gene expression of Acp5 and Oscar (Fig. 2C, D). Furthermore, Rg2 strongly attenuated c-Fos and NFATc1 protein expression (Fig. 3). These results indicate that ginsenoside Rg2 suppresses the mRNA expression of osteoclast-specific genes induced by RANKL, achieved by downregulating the expression of c-Fos and NFATc1.
Phosphorylation of the MAPK pathway, including p38, ERK, and JNK, was observed within 5 min of RANKL treatment to BMMs (27). Earlier studies have shown that inhibition of MAPK kinases suppresses osteoclastogenesis-related factors, such as TRAP and OSCAR, indicating that MAPK signaling is positively involved in RANKL-induced osteoclastogenesis (28, 29). Therefore, we demonstrated the role of ginsenoside Rg2 on the phosphorylation of the MAPK pathway induced by RANKL. As expected, the phosphorylation of JNK, ERK, and p38 stimulated by RANKL was found to be significantly suppressed by Rg2 in BMM (Fig. 4A, B). These results suggest that Rg2 negatively contributed to osteoclastogenesis induced by RANKL through the inhibition of p38, JNK, and ERK phosphorylation.
Enhanced RANKL signaling leads to heightened osteoclast formation and bone resorption. Consequently, the reduction of RANKL expression or modulation of its downstream signaling presents a promising avenue for addressing pathological bone loss therapeutically. Multiple transcription factors, such as c-Fos, NF-κB, Spi-1 (PU.1), and NFATc1, are involved in osteoclast development, with each exerting its influence at distinct phases of osteoclast differentiation (30). In particular, the absence of c-Fos and NFATc1 significantly disrupts the process of osteoclastogenesis (31). Ginsenoside Rg2 constitutes a component found in red ginseng and belongs to the propanaxatriol saponin group, having various biological effects (32). The present study found that ginsenoside Rg2 inhibit osteoclast differentiation induced by RANKL in BMM. The findings demonstrated that Rg2 suppressed the formation of osteoclasts from their precursors induced by RANKL, without causing cytotoxic effects.
Transcription factors c-Fos and NFATc1 have demonstrated pivotal roles in osteoclast formation; their absence halts the process of osteoclastogenesis (6, 27, 33). These factors become active in response to RANKL/RANK signaling, triggering the transformation of osteoclast precursors into fully developed osteoclasts. Notably, c-Fos is a key constituent of the AP-1 transcription factor, contributes to osteoclast differentiation by inducing NFATc1 expression (26, 33, 34). NFATc1 functions as a central transcription factor in osteoclast differentiation, driving the activation of multiple osteoclast-specific genes like Acp5 and Oscar. Moreover, it has the capacity to trigger osteoclast differentiation independently of RANKL presence (35). In this study, at both mRNA and protein levels, ginsenoside Rg2 prominently suppressed the expression of c-Fos and NFATc1. These findings suggest that Rg2’s inhibitory impact on osteoclast differentiation is associated with reduced expression of c-Fos and NFATc1. While the RANKL receptor RANK lacks inherent enzymatic activity within its intracellular domain, it communicates signals by enlisting adapters like TRAF6, subsequently triggering multiple signaling cascades (36, 37). RANKL is recognized for its ability to initiate MAPK pathways, encompassing ERK, JNK, and p38 (38). Demonstrably, inhibiting ERK has been established to reduce the formation of osteoclasts (39, 40) but induce c-Fos for osteoclastogenesis (41). In contrast, dominant-negative JNK inhibits osteoclastogenesis stimulated by RANKL (42), with significant inhibition of p38 occurs during the initial phases of osteoclast formation (43). The inhibition of ERK, JNK, and p38, resulting in the suppression of osteoclastogenesis and c-Fos expression, highlights the significant involvement of MAPK signaling pathways in osteoclast formation. The downregulation of c-Fos expression by ginsenoside Rg2 (Fig. 2A) implies that ginsenoside Rg2 efficiently inhibits the MAPK pathway to regulate osteoclast differentiation.
In conclusion, ginsenoside Rg2 effectively averted osteoclastogenesis induced by RANKL in primary bone marrow cells. Furthermore, this study clarified the molecular mechanisms that underlie the inhibitory impacts of ginsenoside Rg2. The findings imply that the therapeutic influence of ginsenoside Rg2 is linked to the decrease in c-Fos and NFATc1 expression, along with ERK, JNK, and p38 phosphorylation, resulting in reduced expression of Acp5 and Oscar (Fig. 4C). Therefore, we suggest that ginsenoside Rg2 has the potential to serve as an herbal remedy for preventing and treating osteoporosis.
Alpha-modified Eagle’s minimum essential medium (α-MEM) and fetal bovine serum (FBS) were purchased from HyClone Laboratories (Waltham, MA, USA). Recombinant human M-CSF and soluble recombinant mouse RANKL were purchased from PeproTech (London, United Kingdom). Purified ginsenoside Rg2 (purity ≥ 98.0%, HPLC) was purchased from Hongjiu Biotechnology (Jilin, China). Penicillin-streptomycin solution and trypsin-EDTA were purchased from WelGENE Inc. (Daegu, Korea). The antibodies used were as follows: anti-phospho-p38 (#4631), anti-p38 (#9212), anti-phospho-JNK (#9255), anti-JNK (#9252), anti-phospho-ERK (#9101), and anti-ERK (#4695) were purchased from Cell Signaling Technology (Beverly, MA, USA); anti-NFATc1 (#sc-7294) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA); anti-c-Fos (Ab-2) (4-17) was purchased from Calbiochem (San Diego, CA, USA); and anti-actin (#A2228) was obtained from MilliporeSigma (Burlington, MA, USA). All the other chemicals were of analytical grade.
Mouse bone marrow cells were obtained from the femurs and tibiae of 6- to 8-week-old ICR mice and incubated in α-MEM complete medium supplemented with 10% FBS and 100 U/ml penicillin in a 100 mm dish in the presence of M-CSF (30 ng/ml) for 3 d. Adherent cells were used as BMM osteoclast precursors after the removal of non-adherent cells. To generate osteoclasts, BMMs (1 × 105 cells/well) were cultured in complete medium containing M-CSF (30 ng/ml) and RANKL (100 ng/ml) in a 96-well plate, with or without ginsenoside Rg2. Following this, the cells were fixed in 10% formalin for 15 min, permeabilized with 0.1% Triton X-100, and stained with tartrate-resistant acid phosphatase (TRAP) solution. TRAP+ multinuclear cells (MNCs) with more than three nuclei were counted as osteoclasts.
Cytotoxicity was measured using a MTT assay. BMMs (1 × 104 cells/well) were cultured with or without ginsenoside Rg2 at various concentrations (0-40 μM) for 48 h in the presence or absence of M-CSF (30 ng/ml) and RANKL (100 ng/ml) in 96-well plates (200 μl/well). MTT (100 μl/ml, 0.5 mg/ml) (MilliporeSigma) solution was added; after 4 h, 100 μl/ml of extraction buffer was added in 96-well plates (200 μl/well) and preserved for 2 h. The optical density (OD) was recorded at 570 nm with a 96-well plate reader.
Total RNA was extracted from the cultured cells using TRIZOL (Qiagen, Hilden, Germany). cDNA was synthesized from 1 μg of total RNA using Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA, USA) and random primers (Promega Corporation, Madison, WI, USA). qPCR was performed using SYBR Green PCR Master Mix (Qiagen) according to the manufacturer’s protocols. Relative differences in PCR results were evaluated using the comparative cycle threshold method. The primers used were as follows: c-fos, 5’-ATG GGC TCT CCT GTC AAC ACA-3’ and 5’-TGG CAA TCT CAG TCT GCA ACG CAG-3’; Nfatc1, 5’-CTC GAA AGA CAG CAC TGG AGC AT-3’ and 5’-CGG CTG CCT TCC GTC TCA TAG-3’; Oscar, 5’-CTG CTG GTA ACG GAT CAG CTC CCC AGA-3’ and 5’-CCA AGG AGC CAG AAC CTT CGA AAC T-3’; Acp5, 5’-CAG TTG GCA GCA GCC AAG GAG GAC-3’ and 5’-GTC CCT CAG GAG TCT AGG TAT CAC-3’; and Gapdh, 5’-TGA CCA CAG TCC ATG CCA TCA CTG-3’ and 5’-CAG GAG ACA ACC TGG TCC TCA GTG-3’. All reactions were performed in triplicate and normalized to the housekeeping gene Gapdh.
Cells were harvested after 1X PBS washing and lysed in extraction buffer (50 mM Tris-HCI [pH 8.0], 0.5% Nonidet P-40, 150 mM NaCl, 1 mM EDTA, and protease inhibitors). Protein samples were subjected to 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes (MilliporeSigma). The membranes were then incubated with primary antibodies against anti-NFATc1 (1:1,000), anti-c-Fos (1:1,000), anti-actin (1:3,000), anti-phospho-ERK (1:2,000), anti-ERK (1:1,000), anti-phospho-JNK (1:500), anti-JNK (1:1,000), anti-phospho-p38 (1:500) and anti-p38 (1:1,000). Horseradish peroxidase-conjugated secondary antibodies (Amersham Biosciences, Piscataway, NJ, USA) were probed and developed with ECL solution (MilliporeSigma). Signals were detected and analyzed using an LAS3000 luminescent image analyzer (Fuji, Tokyo, Japan).
Statistical analyses were performed using SPSS software (version 17.0). Comparisons between experimental groups were performed using one-way ANOVA with Tukey’s multiple post-hoc test or Student’s t-test. All measurements were performed at least in triplicate; all quantitative data have been expressed as mean ± standard error of the mean (SEM). P < 0.05 were considered statistically significant.
This work was supported by the Ministry of Science and ICT (MSIT), Korea, under the Innovative Human Resource Development for Local Intellectualization Support Program (IITP-2023-RS-2022-00156334) supervised by the Institute for Information & Communications Technology Planning & Evaluation (IITP), and by an NRF grant funded by the Korean Government (MSIT) (2021R1A2C1008317).
The authors have no conflicting interests.