Minimum essential Eagle's medium, α-modification (α-MEM), fetal bovine serum (FBS) and penicillin/streptomycin (P/S) were supplied by Gibco; Thermo Fisher Scientific, Inc. Dulbecco's modified Eagle's medium (DMEM) was procured from Welgene, Inc. Alizarin Red S was obtained from Duksan Co., Ltd. RANKL was purchased from Peprotech, Inc. Rutin, dimethylsulfoxide (DMSO) and the TRAP assay kit were purchased from Sigma-Aldrich; Merck KGaA. Osteo strip well plates were purchased from Corning Inc. Anti-RUNX2, anti-BMP-2, anti-Ctsk and anti-MMP-9 antibodies were purchased from Abcam. Anti-phosphorylated (p)-extracellular signal-regulated kinase 1/2 (ERK1/2), anti-ERK1/2, anti-p-c-Jun N-terminal kinase (JNK), anti-JNK, anti-p-p38, anti-p38, anti-NF-κB and anti-p-NF-κB antibodies were supplied by Cell Signaling Technology, Inc. Anti-NFATc1 was procured from BD Biosciences, and anti-c-Fos, anti-actin and anti-lamin B antibodies were purchased from Santa Cruz Biotechnology, Inc. Secondary antibodies were procured from Jackson ImmunoResearch Laboratories, Inc. The reverse transcriptase kit and SYBR-Green solution was supplied by Invitrogen; Thermo Fisher Scientific, Inc. Taq polymerase was purchased from Kapa Biosystems; Roche Diagnostics. PCR primers were purchased from Genotech Corp. All of the chemicals used in the experiments were analytical grade for cell culture.

LS was purchased from Omniherb and was certified by Professor Yungmin Bu at the Herbology Laboratory, College of Korean Medicine. Voucher specimens of the plants used in the current study were stored at the Department of Anatomy Herbarium (ref. no. KHU-ANA-Et010). LS was soaked in 80% ethanol for 1 week before the extract was filtered through filter paper, and then concentrated under reduced pressure and lyophilized to obtain a powder (yield, 7.83%). The extracts were stored at -20°C until required. Prior to use in the in vitro experiments, LS was filtered through a sterile filter (pore size, 0.22 µm), diluted in DMSO and not treated with >0.1% of the total volume of cell culture medium.

To quantitatively evaluate the LS extract, HPLC was performed using rutin; a reference compound of LS. Rutin (standard stock solution, 1,000 µg/ml) were prepared in methanol. LS and rutin were analyzed using an A Waters 2695 system equipped with a Waters 2487 Dual λ absorbance detector. Separation was achieved using a C18 guard column with Xbridge-C18 (250×6 mm, 5 µm). All solvents were filtered through a 45-µm filter and the elution time was 0-30 min. The binary mobile phase was acetonitrile (A) and water (B) at a composition of 10% A from 0 to 10 min and 50% A from 10 to 30 min. The flow rate was 1.0 ml/min and rutin was detected at 254 nm.

C57BL/6 mouse calvaria MC3T3-E1 cells were purchased from the American Type Culture Collection (ATCC), and the RAW 264.7 murine macrophage cell line was purchased from the Korean Cell Line Bank. The MC3T3-E1 cells were cultured in α-MEM (without ascorbic acid) supplemented with 10% FBS and 1% P/S, and RAW 264.7 cells were cultured in DMEM supplemented with 10% FBS and 1% P/S. Cell lines were maintained in a cell incubator at 37°C and 5% CO₂. To analyze cytotoxicity levels, MC3T3-E1 cells (1×10⁴ cells/well) treated with or without LS for 3, 7 and 14 days, or RAW 264.7 cells (5×10³ cells/well) treated with or without LS and RANKL for 1 and 5 days, were seeded in 96-well-plates and then treated with MTS solution for 2 h. The absorbance was subsequently measured using an enzyme-linked immunosorbent assay (ELISA) reader (Versamax; Molecular Devices, LLC) at a wavelength of 490 nm.

The MC3T3-E1 cells were seeded in 6-well-plates at 5×10⁴ cells/well and incubated in differentiation medium consisting of α-MEM, 25 µg/ml ascorbic acid and 10 mM β-glycerophosphate with or without LS for 21 days. The culture medium was refreshed every 2 days. For Von Kossa staining, the nodules were first fixed in 80% Et-OH and washed 3 times with deionized water The fixed nodules were incubated at room temperature with 1% sliver nitrate for 40 min under ultraviolet light prior to incubation with 5% sodium thiosulfate for 10 min. The dried plate was imaged using a camera. For Alizarin Red S staining, formed mineralized nodules were fixed with 80% Et-OH and then stained with Alizarin Red S solution at room temperature for 3 min. The stained nodules were washed 3 times with Dulbecco's phosphate-buffered saline (Gibco; Thermo Fisher Scientific, Inc.). The formed nodules were subsequently photographed with a camera before the dye was extracted by the addition of 10 mM sodium phosphate (pH 7.0) diluted in 10% cetylpyridinium chloride for 15 min. The extracted dyes were measured measured using an ELISA reader at an absor-bance of 570 nm.

The cells were lysed using radioimmunoprecipitation assay buffer (50 mM Tris-Cl, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate and 0.1% SDS) with protease inhibitor and phosphatase inhibitor 2 and 3 cocktails (Sigma Aldrich; Merck KGaA). Protein concentrations were determined using a bicinchoninic acid assay kit using bovine serum albumin as the standard. An equivalent quantity of protein for each sample (10-30 µg) was separated by 10% SDS-PAGE and transferred to nitrocellulose membranes via electrophoresis. The membranes were blocked with 5% skim milk and incubated with specific primary antibodies against RUNX2 (cat. no. ab76956; dilution, 1:1,000), BMP-2 (cat. no. ab14933; dilution, 1:1,000), β-actin (cat. no. sc-8432; dilution, 1:500), p-ERK (cat. no. 4370S; dilution, 1:1,000), ERK (cat. no. 4695S; dilution, 1:1,000), p-JNK (cat. no. 4668S; dilution, 1:1,000), JNK (cat. no. 9258S; dilution, 1:1,000), p-p38 (cat. no. 4511S; dilution, 1:500), p38 (cat. no. 9212L; dilution, 1:1,000), NF-κB (cat. no. 8242S; dilution, 1:1,000), p-NF-κB (cat. no. 3033S; dilution, 1:500), lamin B (cat. no. sc6216; dilution, 1:1,000), NFATc1 (cat. no. 556602; dilution, 1:1,000), c-Fos (cat. no. sc-447; dilution, 1:200), MMP-9 (cat. no. ab38898; dilution, 1:1,000) and Ctsk (cat. no. ab19027; dilution, 1:1,000) at 4°C overnight. The membranes were then incubated with secondary antibodies (cat. no. 111-035-045, 115-035-062; dilution, 1:10,000) at room temperature for 1 h and chemiluminescence was measured using enhanced chemiluminescence substrate (Santa Cruz Biotechnology, Inc.). Band densities were quantified using ImageJ software version 1.51j8 (National Institutes of Health).

Total RNA was extracted using TRIzol reagent (Takara Bio, Inc.) according to the manufacturer's protocol. A total of 2 µg total RNA was reverse transcribed into cDNA using a reverse transcriptase enzyme (Invitrogen; Thermo Fisher Scientific, Inc.). RT-PCR was performed using the C1000 Touch™ Thermal Cycler (Bio-Rad, Laboratories, Inc.) and Taq polymerase (Kapa Biosystems; Roche Diagnostics). The thermal cycling parameters were as follows: 30 sec at 94°C (denaturation), 30 sec at 53-58°C (annealing) and 30 sec at 72°C (extension). The temperature and number of cycles for each gene are listed in Table I. The final PCR mixtures were electrophoresed on agarose gels stained with SYBR-Green (Invitrogen; Thermo Fisher Scientific, Inc.) and bands were quantified using ImageJ software version 1.51j8.

To generate osteoclasts from RAW 264.7 cell cultures, the RAW 264.7 cells were first seeded at a density of 5×10³ cells/well in 96-well-plates containing differentiation medium (α-MEM containing 100 ng/ml RANKL with or without LS) and incubated in a cell incubator at 37°C for 5 days. The culture medium was refreshed on days 2 and 4. The differentiated cells were fixed with 10% formalin for 10 min and stained using the TRAP kit according to the manufacturer's protocol. The cells were then transferred to a new plate and an equal volume of TRAP solution [4.93 mg p-nitrophenyl phosphate (PNPP) in 0.5 M 750 ml acetate solution, mixed with 150 ml tartrate acid solution] was added followed by incubation at 37°C for 30 min. The reaction was then inhibited by the addition of 0.5 M NaOH, and the absorbance was measured using an ELISA reader at a wavelength of 405 nm. To evaluate bone resorption activity, RAW 264.7 cells (5×10³ cells/well) were cultured in osteo strip well plates containing differentiation medium (α-MEM containing 100 ng/ml RANKL with or without LS) and incubated for 5 days. The cell culture medium was refreshed on days 2 and 4. Following differentiation, the medium was removed and the cells were dissolved with NaClO before the plate was allowed to dry. The number of TRAP-positive cells and the area of the pit formation were measured using ImageJ software.

To evaluate F-actin ring formation, the RAW 264.7 cells (5×10³ cells/well) were cultured in 96-well plates containing differentiation medium (α-MEM containing 100 ng/ml RANKL with or without LS) and incubated in a cell incubator at 37°C for 5 days. The differentiated cells were fixed with 4% paraformaldehyde for 20 min and permeabilized with 0.1% Triton X-100 in PBS for 5 min. The cells were then stained using the Actistain™ 488 Fluorescent Phalloidin (Cytoskeleton Inc.) at room temperature in the dark for 30 min. The cells were washed with PBS, and the nuclei were then counterstained with 4′,6-diamidino-2-phenyl-indole (DAPI, Sigma Aldrich; Merck KGaA). The formation of the actin ring was captured using an immunofluorescence microscope (Cellena; Logosbio).

Institute of Cancer Research (ICR) CD-1 mice (male; age, 4 weeks; weighing, 27-29 g) were purchased from Nara Biotech, Co., Ltd. Animal experiments proceeded with permission from the Kyunghee University Animal Committee [ref. no. KHUASP(SE)-15-095]. The mice were maintained under 12 h light/dark cycle at 22-24°C at a relative humidity of 55-55% with free access to food and water. The mice were allowed to acclimatize to the laboratory conditions at the animal breeding facility for 1 week. The animal model of osteoporosis was established by an intraperitoneal injection of LPS, as previously described (18,19). The mice were divided into the following 3 groups (n=6/group): Group 1, mice treated with 100 µl PBS via intraperitoneal injection and orally administered with distilled water daily; group 2, mice treated with LPS (5 mg/kg body weight) via intraperitoneal injection and orally administered with distilled water daily; and group 3, mice treated with LPS via intraperitoneal injection and orally administered with LS at a concentration of 100 mg/kg daily. PBS or LPS administration were carried out on days 1 and 4. The humane endpoint of this experiment was as follows: Dirty hair and eye discharge, self-injury and anxiety, vomiting and hemoptysis, inactivity, anxiety and headache). No abnormal signs that signified the humane endpoints of the experiment were observed from any of the mice during the experiment. All mice were sacrificed on day 9 and the right femurs were extracted and analyzed by micro-computed tomography (micro-CT; SkyScan1176; SkyScan; Bruker Corp.). Bone volume/total volume (BV/TV), trabecular pattern factor (Tb.pf) and structure model index (SMI) calculations were performed using NRecon software (SkyScan version 1.6.10.1; SkyScan; Bruker Corp.).

The experiments were repeated at least 3 times. The results are presented as the means ± standard error of the mean. Statistical comparisons were performed one-way ANOVA tests, followed by Dunnett's post hoc analysis. A value of P<0.05 was considered to indicate a statistically significant difference.

Rutin is a bioactive marker compound used for the validation of LS. The chromatogram of the ethanol extract from LS demonstrated that a number of peaks were detected at a retention time of 0 and 30 min, and rutin was identified at the same retention time as the standards (Fig. 1).

LS at a concentration of 400 and 800 µg/ml was toxic to the MC3T3-E1 cells for 3, 7 and 14 days (Fig. 2A). Subsequently, to investigate the effects of LS on osteoblast differentiation, the MC3T3-E1 cells were treated with LS, ascorbic acid and β-glycerophosphate. Following 14 days of culture, Von Kossa and Alizarin Red S staining confirmed that calcified nodules were formed at an earlier stage in the LS treatment group when compared with the differentiated medium-treated group (Fig. 2B and C). At 17 and 21 days, LS enhanced osteoblast differentiation, as indicated by the increase in mineralized nodule formation. In addition, the absorbance of the extracted Alizarin Red S dye was increased by LS treatment (Fig. 2D). No significant difference was observed in osteoblast differentiation with LS at 50 to 200 µg/ml. Thus, the highest concentration was determined to be 50 µg/ml (data not shown). To determine the mechanisms of action of LS as regards osteoblast differentiation, the expression levels of transcription factors and osteogenic genes were analyzed in LS-treated osteoblasts. BMP-2 and RUNX2 are major transcription factors involved in osteoblast differentiation, and LS was observed to increase the expression levels of these genes (Fig. 2E and F). In addition, LS promoted the expression of osteogenic genes, including RUNX2, ALP (Alpl), OPN (Sparc), OSN (Spp1) and COL1 (Col1a1). BSP (Ibsp) expression increased with LS treatment, although the difference was not statistically significant (Fig. 2G and H).

Before confirming the osteoclast experiments, the toxicity of LS in the RAW 264.7 cells was confirmed. A high concentration (800 µg/ml) of LS reduced cell viability to below 50% and the concentrations of LS used in the osteoclast experiments did not affect the level of cytotoxicity (Fig. 3A and B). Thus, the possibility that the inhibition of osteoclast formation was due to cytotoxicity was excluded. To confirm the inhibitory effects of LS on osteoclastogenesis, TRAP staining and pit formation assays were performed. As demonstrated in Fig. 3C, the number of TRAP-positive osteoclasts was increased following stimulation of the cells with RANKL, and LS decreased the number and size of osteoclasts in a dose-dependent manner (Fig. 3D). In addition, LS inhibited TRAP activity in the differentiation medium (Fig. 3E). No significant difference was observed in the inhibitory effects on osteoclastogenesis with LS at 100 to 400 µg/ml. Thus, the highest concentration was determined to be 100 µg/ml.

To evaluate the inhibitory effects of LS on resorptive activity, differentiated osteoclasts formed resorption pits through calcium absorption (Fig. 4A), and the area of the resorption pit was reduced by LS treatment (Fig. 4B). RANKL stimulation increased the formation of the well-polarized F-actin ring (Fig. 4C). However, LS was observed to decrease both the size and number of F-actin ring structures (Fig. 4D).

When osteoclast differentiation is induced by RANKL, the NF-κB and MAPKs signaling pathways are essential. RANKL stimulation increases the protein levels of NF-κB and p-NF-κB. In this study, LS was observed to decrease the translocation of NF-κB to the nucleus and inhibit the phosphorylation of NF-κB (Fig. 5A and B). In addition, LS treatment significantly inhibited the RANKL-induced phosphorylation of ERK1/2, JNK and p38 in a concentration-dependent manner (Fig. 5C and D).

To determine the mechanisms underlying LS-mediated osteoclast inhibition, the expression of NFATc1 and c-Fos was measured by western blot analysis and reverse transcription-semi-quantitative PCR. As demonstrated in Fig. 6, RANKL stimulation increased the protein and mRNA levels of Nfatc1 and FOS, whereas LS suppressed the expression of these factors. In addition, LS did not affect the expression of housekeeping genes, such as actin and GAPDH.

To examine whether LS inhibits bone resorption-related enzymes, the effects of LS on the expression of MMP-9 and Ctsk were examined by western blot analysis and reverse transcription-semi-quantitative PCR. As demonstrated in Fig. 7A-D, RANKL stimulation induced the protein and mRNA expression levels of Mmp-9 and Ctsk, whereas LS markedly suppressed the expression of these bone resorption-related enzymes. In addition, the effects of LS on markers involved in osteoclast differentiation were analyzed. As demonstrated in Fig. 7E and F, the expression of RANK (Tnfrsf11a), TRAP (Acp5), OSCAR, c-src, c-myc, PU-1 (Spi1), OC-STAMP and ATP6v0d2 was increased by RANKL treatment, whereas LS significantly inhibited the expression of these genes.

The results presented thus far indicated that LS activated osteoblast differentiation and inhibited osteoclast differentiation. To confirm these in vitro results, the potential inhibitory effects of LS on bone loss in vivo were investigated. The concentration of LS administered in this study was determined as follows: In oriental medicine, based on a typical adult weight of 60 kg, a single dose of LS is 8 g and corresponding to 0.6184 g (yield, 7.73%) of LS extract. Therefore, 10.3 mg of LS should be administered per 1 kg. In previous studies, mice are known to have a 7-fold higher metabolism than humans (20), and an LS dose of approximately 72 mg/kg is appropriate. In this study, LS was administered at a slightly higher concentration than that administered to humans. To achieve this, a mouse model of endotoxin-induced bone destruction was first generated (21,22). The following day after LPS administration, the weight of the mice was reduced, but this was recovered. In addition, no marked changes in body weight due to LS were observed (Fig. S1). As demonstrated in Fig. 8A, micro-CT images of the femur indicated a decrease in trabecular bone; however, the administration of LS significantly inhibited LPS-induced bone loss. Based on the radiographic results, the effect of LS on bone microstructure was also analyzed. This revealed that LPS administration reduced bone mineral density (BMD) and BV/TV, whereas the administration of LS significantly inhibited the LPS-induced reduction in BMD and BV/TV. In addition, the LPS-induced increase in Tb.pf was suppressed by LPS treatment. Similarly, SMI increased with LPS administration, which was suppressed by treatment with LS; however, the difference was not statistically significant (Fig. 8B-E).

Based on the results of this study, LS promoted osteogenic gene expression by upregulating RUNX2 in MC3T3-E1 cells. In addition, LS inhibited the bone resorption gene by inhibiting the expression of NFATc1/c-Fos in RANKL-induced RAW 264.7 cells. As a result, it exerted an anti-osteoporotic effect in mice with LPS-induced bone less (Fig. 9).