Bone remodeling occurs throughout life to maintain mineral homeostasis through a tightly regulated process of bone resorption and formation. However, an imbalance between the two processes, such as excessive bone resorption or defective bone formation, can lead to bone mineral disorders such as osteoporosis (1). Owing to the serious health consequences and social costs associated with the increasing incidence of osteoporosis, continued efforts are needed to control this disease.

Hematopoietic stem cell-derived osteoclasts are the primary multinucleated cells responsible for bone resorption (2). The receptor activator of nuclear factor κB ligand (RANKL)/RANK pathway is the major signaling pathway involved in osteoclast differentiation and activation. The binding of RANKL to RANK on the surface of osteoclast progenitor cells activates downstream signaling pathways, including nuclear factor kappa B (NF-κB), mitogen-activated protein kinase (MAPK), and phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt) pathways. Subsequently, these pathways induce the activation of transcription factors such as nuclear factor of activated T cells cytoplasmic 1 (NFATc1) and activator protein-1 (AP-1), which lead to the expression of effector molecules such as cathepsin K (CTSK) and tartrate-resistant acid phosphatase (TRAP) for osteoclastogenesis (3, 4). Therefore, the blockade of the signaling pathway induced by RANKL/RANK binding is considered a potential therapeutic target to inhibit osteoporosis (5-7).

Functionally opposite to osteoclasts in the continuous cycle of bone remodeling, osteoblasts are the primary cells responsible for bone formation by secreting organic substances essential for mineralization (8). Both osteoblasts and adipocytes are derived from MSCs; therefore, the balanced differentiation of MSCs into both cells in a reciprocal relationship plays an important role in maintaining bone homeostasis (9-11). Indeed, previous reports have shown that pathogenic conditions that cause bone loss, such as aging and menopause, coincide with increased adiposity in the bone marrow (12, 13). Therefore, enhancing the differentiation of MSCs into osteoblasts while inhibiting adipocyte differentiation represents a strategy to mitigate bone loss, in addition to inhibiting osteoclast differentiation or activity.

Natural compounds are considered promising therapeutics owing to their pharmacological effects and ability to provide long-term treatment with few side effects (5). Boeravinone B (BOB) is a natural rotenoid analog isolated from Boerhavia diffusa L., a medicinal plant used for gastrointestinal, hepatoprotective, and gynecological indications (14, 15). Recent studies have shown that BOB has inhibitory activities against inflammation, reactive oxygen species production, aging, cancer, and efflux pump (16-19). Given that BOB has been shown to have pharmacological activity, particularly in ameliorating metabolic disorders, we hypothesize that this phytochemical will have a beneficial effect on bone metabolism. Therefore, we investigated the regulatory activities of BOB in vitro and in vivo on bone remodeling. 

BOB inhibits osteoclast differentiation without cytotoxicity

First, we investigated the effect of BOB, whose chemical structure is shown in Fig. 1A, on osteoclast differentiation. Briefly, primary bone marrow macrophages (BMMs) were incubated with RANKL in the presence or absence of BOB at concentrations ranging from 5 to 40 μM. As a result, we observed a significant dose-dependent inhibition of RANKL-induced osteoclast formation, as determined by TRAP staining (Fig. 1B). The WST-8 cell viability assay showed that the concentrations of BOB used in this study did not significantly affect the survival of BMMs (Fig. 1C), thus excluding cytotoxicity as the cause of the BOB-mediated inhibition of osteoclastogenesis. These results suggest that BOB efficiently inhibits osteoclast differentiation without significant cytotoxicity.

BOB inhibits RANKL-induced osteoclast-specific gene expression

Next, we validated the efficacy of BOB in inhibiting osteoclast differentiation by altering the expression of osteoclast-specific markers. Western blot analysis showed that RANKL-mediated induction of NFATc1, a key transcription factor in osteoclastogenesis, was decreased by BOB in a dose-dependent manner (Fig. 2A). Similarly, in the presence of RANKL, the levels of both pro- and mature forms of CTSK, an osteoclast effector molecule, were reduced by BOB co-treatment (Fig. 2A). As shown in Fig. 2B and Supplementary Fig. 1, the mRNA levels of osteoclastogenesis-specific genes, such as Nfatc1, Trap, Ctsk, and Oscar, but not c-Fos, were also reduced by BOB in a dose-dependent manner. Furthermore, BOB inhibited RANKL-induced osteoclast differentiation in a time-dependent manner, which was validated by quantitative real-time PCR (qPCR) analysis of the osteoclast markers (Fig. 2C).

BOB inhibits osteoclast-mediated bone resorption in vitro and ovariectomy (OVX)-induced bone loss in vivo

Next, to further evaluate the effect of BOB on the resorbing activity of osteoclasts, we cultured BMMs with RANKL and BOB on a calcium phosphate coated plate. The results showed that treatment with BOB reduced pit formation of the calcium phosphate matrix by resorptive osteoclasts (Fig. 3A). As CTSK is a crucial protease secreted by activated osteoclasts, we next analyzed the changes in CTSK secretion in the culture supernatants obtained during the pit formation assay. The mature form of CTSK was detected in the RANKL-treated group but not in the group co-treated with BOB, as determined by western blot analysis (Fig. 3B). To further examine the anti-osteoporotic effect of BOB, we treated ovariectomized mice with BOB. The μ-CT results showed that OVX-induced bone loss in femur tended to be reduced by BOB administration (Fig. 3C); however, there was no significant weight loss in BOB-treated mice compared to ovariectomized mice receiving vehicle (Fig. 3D). Our data suggest that BOB can prevent OVX-induced bone loss without noteworthy toxicity.

BOB inhibits RANKL-induced activation of NF-κB and MAPK signaling pathways

We further investigated how BOB inhibits osteoclast differentiation by altering intracellular signaling pathways such as NF-κB, MAPK, and PI3K/Akt, which are activated by RANKL-RANK binding and play a critical role in the initiation of osteoclastogenesis. Western blot analysis showed that pretreatment with BOB suppressed RANKL-induced phosphorylation of Ikk, IκB, and p65, implying inhibition of the NF-κB signaling pathway (Fig. 4A). Consistent with these results, BOB strongly inhibited RANKL-mediated NF-κB translocation into the nucleus (Fig. 4B). We next examined the regulatory effect of BOB on RANKL-induced MAPK activation. Western blot analysis showed that BOB also reduced the phosphorylation of ERK and p38, but not JNK, among the three MAPKs (Fig. 4C). We further determined whether the BOB-induced decrease in NF-κB signaling was causally related to the decreased activity of two MAPKs, ERK and p38. As shown in Fig. 4D, pretreatment with either ERK (U0126) or p38 (SB202190) inhibitor did not significantly affect RANKL-induced p65 phosphorylation, indicating that BOB suppresses osteoclast differentiation by separately inhibiting the MAPK and NF-κB signaling pathways. Finally, we examined the effect of BOB on PI3K/Akt activation in osteoclastogenesis. Our results showed that the strong phosphorylation of PI3K and Akt exhibited in early osteoclastogenesis was greatly attenuated by pretreatment with BOB (Fig. 4E). These results demonstrate that BOB negatively regulates all three major early signaling pathways triggered by the interaction of RANKL and RANK, resulting in the inhibition of osteoclast differentiation.

BOB enhances osteoblast differentiation but inhibits adipocyte differentiation

Considering that bone homeostasis is regulated by bone resorption and bone formation, and that osteoblasts responsible for bone formation are derived from MSCs, we next investigated the effect of BOB on the multilineage differentiation of MSCs. In osteoblast differentiation of BMSCs, the addition of BOB promoted mineral deposition, as determined by ARS staining (Supplementary Fig. 2A). BOB also enhanced the expression of OPN and DMP1, markers for the late stage of osteoblast differentiation (Supplementary Fig. 2B). To verify the effect of BOB in promoting osteoblast differentiation, MC3T3-E1 cells were differentiated into osteoblasts in the presence of BOB. The results of ALP staining and Western blot analysis further confirmed that BOB promoted ALP activity and Runx2 expression in a concentration-dependent manner (Supplementary Fig. 2C, D).

During adipocyte differentiation of BMSCs, BOB inhibited the accumulation of intracellular lipid droplets with decreased expression of GLUT4 and AdipoQ (Supplementary Fig. 3A, B). To elucidate the mechanism by which BOB inhibits adipocyte differentiation, we examined its regulatory effect on the insulin signaling pathway. As shown in Supplementary Fig. 3C, pretreatment with BOB efficiently inhibited IRS-1 and Akt phosphorylation, both of which were increased by insulin stimulation. These results suggest that BOB can direct MSC commitment toward osteoblasts rather than adipocytes, thereby contributing to bone formation. 

Osteoporosis is a health issue that affects older people worldwide and is caused by free radicals and inflammatory mediators, which activate osteoclasts and decrease bone density (20). Plant-derived antioxidants are being studied as a means to control this condition (5, 21, 22). On this basis, we hypothesized that boeravinones may represent attractive candidates for the control of osteoporosis.

In the present study, we demonstrated that BOB, one of the boeravinones, inhibits the RANKL-RANK signaling pathway and attenuates three major signaling pathways induced by RANKL in BMMs, namely NF-κB, MAPK, and PI3K/Akt. This BOB activity inhibited osteoclast differentiation and reduced bone resorption activity, leading to an anti-osteoporotic effect. The findings suggest that boeravinones, particularly BOB, represents a promising candidate for the development of new anti-osteoporosis treatments.

Rotenoids, a class of isoflavonoid compounds found in medicinal plants belonging to the Lamiaceae family, have various pharmacological effects; however, only three specific compounds (rotenone, amorphigenin, and deguelin) have been identified as potential therapeutics for osteoporosis (23-25). Moreover, to best of our knowledge, the bone remodeling modulatory activity of boerhavinones (boerhavinones A to X), a class of rotenoids derived from Boerhaavia diffusa, has not yet been elucidated. Therefore, this study is the first to investigate the modulatory activity of boerhavinones in the pathophysiological research field. These compounds possess typical pharmacological activities of antioxidants including anti-inflammatory, antioxidant, anticancer, and anti-aging effects (26). As many antioxidants have demonstrated anti-osteoporotic effects, it is reasonable to speculate that boeravinones may also function as negative regulators of osteoclasts. Consequently, further investigation is needed to evaluate the efficacy of other boeravinones in inhibiting osteoclast differentiation and activity.

The three above-mentioned rotenoids have been shown to inhibit osteoclast differentiation by blocking RANKL activation of NF-κB and MAPK pathways (23-25). Our results showed that another rotenoid BOB can effectively inhibit both signaling pathways. Moreover, the inhibition of NF-κB activation by BOB was not the result of the inhibition of MAPK, as confirmed by studies using chemical inhibitors of ERK and p38. These results suggest that the TRAF6 complex, an upstream regulator of both pathways, is the direct target of BOB (4). However, further research is needed to confirm this hypothesis and to determine the precise target of BOB.

Akt, along with NF-κB and MAPK, is an important factor in the early stages of osteoclast differentiation; indeed, BMMs lacking Akt have been shown to exhibit defective osteoclast differentiation (27). Akt activation promotes the expression and nuclear translocation of NFATc1 through the inactivation of GSK-3β (28). Although previous studies have examined the effects of rotenoids on osteoclast differentiation, these have been limited to regulation of the NF-κB and MAPK pathways (23-25). This study extends research into the efficacy of BOB in context of Akt regulation. Interestingly, BOB at a concentration of 20 μM completely inhibited the PI3K/Akt signaling cascade induced by RANKL. BOB also showed similar inhibitory effects in Akt-mediated insulin signaling, which is critical for adipocyte differentiation. These results suggest that BOB improves bone mass by inhibiting the activation of the PI3K/Akt pathway, which contributes to osteoclast and adipocyte differentiation.

In addition to inhibiting osteoclast differentiation and activity, enhancement of osteoblast differentiation is another strategy for the treatment of osteoporosis. MSCs are multipotent cells that can differentiate into osteoblasts and adipocytes. An imbalance between the two cell lineages in the bone marrow, i.e., a decrease in osteoblasts and an excessive increase in adipocytes, has been reported to be associated with increased bone fragility (12, 13). In this study, the effects of BOB on osteoblast and adipocyte differentiation were examined using BMSCs, where the addition of BOB to BMSCs was shown to promote matrix mineralization. Although we have not performed sufficient in-depth studies to explain this enhancement, the result suggests that BOB is involved in the functional maturation of osteoblasts to induce mineral deposition. However, BOB also had a significant inhibitory effect on the in vitro adipogenic differentiation of BMSCs, as evidenced by the analysis of lipid droplet accumulation, marker expression, and insulin signaling. These results suggest that BOB contributes to the maintenance of bone marrow homeostasis by preventing MSCs from transforming and accumulating into adipocytes. As the commitment of MSCs to both lineages is regulated by crosstalk between key transcription factors, such as Runx2 and PPARγ (29), the effect of BOB on the activity of transcription factors remains to be elucidated in further studies.

In conclusion, BOB can to comprehensively control the differentiation of cells responsible for bone remodeling, at least in an in vitro system. Our findings provide evidence to support BOB as a potential therapeutic agent for bone diseases. 

Reagents

BOB was purchased from PhytoLab and dissolved in DMSO. Recombinant mouse M-CSF and RANKL were obtained from Biolegend and Peprotech, respectively. Antibodies used in this study are described in Supplementary Material.

Cell culture and osteoclast differentiation

BMMs were isolated from C57BL/6 mouse bone marrow cells and cultured in α-MEM medium (Gibco) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin and streptomycin (Gibco), as previously described (30). To induce osteoclast differentiation, BMMs were seeded at a density of 5 × 10⁵/ml and incubated for 3 to 4 days in the medium containing 30 ng/ml M-CSF and 100 ng/ml RANKL.

TRAP staining

Osteoclast differentiation was assessed using a TRAP staining kit (Cosmo Bio), according to the manufacturer’s instructions. Images were acquired using a Lionheart FX Cell Imager (BioTek). TRAP-positive multinucleated cells with three or more nuclei were counted as osteoclasts.

qPCR analysis

Total RNA was extracted from cell samples using TRIzol reagent (Ambion). Two micrograms of total RNA were reverse-transcribed into cDNA using M-MLV reverse transcriptase (Promega). Transcriptional levels of marker genes in the cDNA samples were assessed using a QuantStudio cycler (Applied Biosystem) with Power SYBR Green PCR Master Mix (Applied Biosystem) and the primer sets described in Supplementary Material. Data are presented as the relative expression of the ΔΔCt value obtained using QuantStudio Design & Analysis software (Applied Biosystem), with 18S rRNA used as an internal control.

Western blot analysis

Whole cell lysates were prepared using Cell Lysis Buffer (Cell Signaling Technology). Specifically, cytoplasmic and nuclear cell fractions were extracted using the NE-PER Nuclear and Cytoplasmic Extraction Reagent Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. For western blot analysis, equal amounts of proteins were electrophoresed on an SDS-PAGE gel and transferred to a PVDF membrane. The membranes were blocked in blocking buffer (tris-buffered saline with 0.1% Tween-20 and 5% nonfat milk) and then incubated with primary antibodies and HRP-conjugated secondary antibodies. Immunobands were detected with ECL solution (Millipore).

OVX-induced bone loss mouse model

The animal study was conducted according to the guidelines of the Institutional Animal Care and Use Committees of Chonnam National University. Eight-week-old C57BL/6 mice were anesthetized, and a midline abdominal skin incision was made. Bilateral ovaries were excised and both ends were closed by suturing. Sham-operated mice underwent the same procedure without removal of the ovaries. All mice were allowed 1 week of recovery with the intraperitoneal administration of gentamicin (50 mg/kg) to prevent infection. After an additional 3 weeks of rest to induce bone loss, mice were randomly divided into four groups (n = 4/group) and then injected intraperitoneally every other day with either PBS containing vehicle (1% DMSO) or BOB (5 mg/kg and 20 mg/kg) for 4 weeks. The mice were sacrificed, and their femurs were fixed in formalin. Bone analysis by μ-CT scanning was performed as described in Supplementary Material.

Statistical analysis

Results were obtained from at least two independent experiments. Student’s t test was used to determine the statistical significance of the data. A P-value < 0.05 was considered statistically significant. 

We thank Sin-Hye Oh and Seung Hee Kwon for their technical assistance. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2019R1A5A2027521); the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. NRF-2020R1I1A1A01061824); the Korean Fund for Regenerative Medicine (KFRM) grant (Ministry of Science and ICT, Ministry of Health & Welfare, 22A0104L1); and by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (NRF-2021R1C1C2009626).

The authors have no conflicting interests.