DMEM was purchased from Welgene, Inc. Minimum essential medium-α (α-MEM), penicillin/streptomycin (P/S) and Dulbecco's PBS (DPBS) were obtained from Gibco; Thermo Fisher Scientific, Inc. FBS was purchased from Atlas Biologicals. RANKL was purchased from PeproTech, Inc. CellTiter 96 Aqueous non-radioactive cell proliferation (MTS) assay was purchased from Promega Corporation. Bicinchoninic acid (BCA) solution, phosphatase inhibitor cocktail, DAPI, 17β-estradiol (E₂) and alendronate (ALN) were obtained from Sigma-Aldrich; Merck KGaA. Osteo Assay Surface multiple well plates (cat. no. 3989) were obtained from Corning, Inc.. PCR primers were synthesized by Genotech Corp. Acti-stain™ 488 Fluorescent Phalloidin was purchased from Cytoskeleton, Inc.. The primary antibodies and secondary antibodies used in the present study were: β-actin (cat. no. sc-8432; Santa Cruz Biotechnology, Inc.), c-Fos (cat. no. sc-447; Santa Cruz Biotechnology, Inc.), NFATc1 (cat. no. 556602; BD Biosciences), MMP-9 (cat. no. ab38898; Abcam), CTK (cat. no. ab19027; Abcam), TRAF6 (cat. no. sc-8409; Santa Cruz Biotechnology, Inc.) and peroxidase AffiniPure Goat Anti-Mouse IgG (cat no. 115-035-062; Jackson ImmunoResearch Laboratories, Inc.) and peroxidase AffiniPure Goat Anti-Rabbit IgG (cat no. 115-035-144; Jackson ImmunoResearch Laboratories, Inc). All other reagents used were of analytical grade.

MFR was purchased from Omniherb. MFR was prepared by decocting 300 g of the dried herb in 1.5 l boiling distilled water (D.W) for 2 h. The extract was filtered using filter paper (no. 3; Whatman plc; GE Healthcare Life Sciences) and collected in a rotary evaporator at 55°C. The extract was lyophilized and 19 g dried powder was obtained (yield ratio, 12.7%).

Vitexin (cat. no. 49513; Sigma-Aldrich; Merck KGaA) is the active ingredient in MFR (29); therefore, to quantitatively evaluate the MFR extract, HPLC was performed with vitexin used as an internal standard. Vitexin was prepared in DMSO. Analysis was performed using a UV detector (2996 Waters 2695; Waters Corporation); separation was carried out on an Xbridge-C18 with a C18 guard column (250 × 4.6 mm; 5 µm; Waters Corporation) and proceeded at 30°C for 50 min at a flow rate of 1 ml/min. Vitexin was detected at 335 nm. Samples were injected in a volume of 10 µl. The mobile phase consisted of acetonitrile (A) and water (B), at a composition of 10% A from 0–10 min and 50% A from 10–30 min.

RAW 264.7 cells are mouse monocyte-macrophage like cells (30). RAW 264.7 cells were obtained from Korean Cell Line Bank; Korean Cell Line Research Foundation. RAW 264.7 cells were cultured in DMEM supplemented with 10% FBS and 1% P/S in a cell incubator at 37°C and 5% CO₂. To examine cell viability, the cells were seeded in a 96-well plate at 5×10³ cells/well and incubated for 24 h. RAW 264.7 cells were treated with MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 24 h. After treatment, 20 µl/well MTS solution was added and the cells were incubated at 37°C for 2 h. To determine cell viability of mature osteoclasts, RAW 264.7 cells were seeded in a 96-well plate at 5×10³ cells/well and were treated with or without RANKL (100 ng/ml) for 5 days. Subsequently, cells were treated with MFR (12.5, 25, 50 and 100 µg/ml) for 1 day, after which, 20 µl MTS solution was added and incubated at 37°C for 2 h. Cell viability was measured using an enzyme-linked immunosorbent assay (ELISA) reader at a wavelength of 490 nm.

A total of 5×10³ RAW 264.7 cells/well were seeded in a 96-well plate. After 24 h, to induce osteoclast differentiation of RAW 264.7 cells, cells were treated with RANKL (100 ng/ml) and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 5 days. The medium was replaced every 2 days. In order to compare the inhibitory effect of MFR and vitexin on osteoclast differentiation, cells were seeded in the same manner as mentioned previously. After 24 h, the medium was replaced with RANKL (100 ng/ml) and MFR (50 and 100 µg/ml) or vitexin (0.0753 and 0.147 µg/ml) at 37°C for 5 days. Subsequently, the cells were washed and fixed with 4% formalin at room temperature for 10 min. TRAP staining was performed using a TRAP kit (Sigma-Aldrich; Merck KGaA) according to the manufacturer's protocol. The number of TRAP-positive cells was counted using an inverted light microscope (Olympus Corporation; magnification, ×100).

TRAP is secreted in large quantities by osteoclasts and TRAP levels are considered a biochemical marker of osteoclast function (31). Therefore, the present study also analyzed TRAP levels to determine the effect of MFR on osteoclast function. In order to measure the TRAP levels, 50 µl differentiation medium was obtained and transferred to a new plate. Subsequently, an equal volume of TRAP solution (4.93 mg 4-Nitrophenyl phosphate disodium salt hexahydrate in 850 µl 0.5 M acetate solution and 150 µl tartrate solution) was added to the differentiation medium and incubated at 37°C for 1 h. Subsequently, the reaction was terminated using 0.5 M NaOH and TRAP levels were measured using an ELISA reader at a wavelength of 405 nm.

To determine F-actin ring formation, 5×10³ cells/well were seeded in a 96-well plate, and treated with RANKL (100 ng/ml) and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 5 days. The medium was replaced every 2 days. The cells were fixed with 4% paraformaldehyde at room temperature for 20 min and permeabilized with 0.1% Triton X-100 in PBS at room temperature for 5 min. The cells were then stained using Acti-stain™ 488 Fluorescent Phalloidin at room temperature in the dark for 30 min. Cells were washed with PBS and nuclei were counterstained with DAPI. Images were captured using fluorescence microscopy (Cellena; Logos Biosystems; magnification, ×200).

To examine pit formation, the cells were seeded in Osteo Assay Surface multiple well plates at 5×10³ cells/well. Subsequently, the cells were treated with RANKL (100 ng/ml) and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 5 days. The differentiation medium was replaced every 2 days. Subsequently, cells were washed and removed using deionized water with 4% sodium hypochlorite. Images of resorbed areas were captured using an inverted light microscope (Olympus Corporation; magnification, ×200). Total pit areas were analyzed using ImageJ version 1.46 (National Institutes of Health).

To examine the effect of MFR on osteoclast differentiation-associated transcription factors, 5×10⁵ RAW 264.7 cells/well were seeded in a 60-mm dish and treated with RANKL (100 ng/ml) and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 1 day. In order to compare the inhibitory effect of MFR and vitexin on the expression of NFATc1 and c-Fos, 5×10⁵ RAW 264.7 cells/well were seeded in a 60-mm dish and treated with RANKL (100 ng/ml) and MFR (50 and 100 µg/ml) or vitexin (0.0753 and 0.147 µg/ml) at 37°C for 1 day. To prepare whole-cell lysates, the cells were washed with cold DPBS, and the total proteins were extracted using RIPA lysis buffer (50 mM Tris-CI, 150 mM NaCl, 1% NP-40, 0.5% Na-deoxycholate, 0.1% SDS, protease inhibitor cocktail, phosphatase inhibitor cocktail). Protein concentration was determined using a BCA assay. Proteins (30 µg) were separated by SDS-PAGE on 10% gels and transferred to a nitrocellulose membrane (Whatman plc; GE Healthcare Life Sciences) according to the manufacturer's protocol. The membranes were blocked in 5% skimmed milk for 1 h at 37°C, followed by overnight incubation at 4°C with primary antibodies against NFATc1 (1:1,000), c-Fos (1:1,000), TRAF6 (1:1,000), MMP-9 (1:1,000), CTK (1:1,000) and β-actin (1:1,000). Subsequently, membranes were incubated with secondary antibodies (1:10,000) for 1 h at room temperature. The membranes were visualized using enhanced chemiluminescence reagent (cat no. RPN2106; GE Healthcare Life Sciences) and protein expression levels were semi-quantified using ImageJ (version 1.46; National Institutes of Health).

To examine the effect of MFR on osteoclast-related markers, 2×10⁵ cells/well seeded in a 6-well plate, and treated with the RANKL (100 ng/ml) and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 4 days. Total RNA of the treated cells was extracted using RNAiso Plus (cat. no. 9109; Takara Bio, Inc.) according to the manufacturer's protocol. A total of 2 µg RNA was measured using a NanoDrop 2.0 spectrophotometer (NanoDrop; Thermo Fisher Scientific, Inc.). cDNA was synthesized using a RT kit (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. The synthesized cDNA was amplified by PCR using Taq polymerase (Kapa Biosystems; Roche Diagnostics). The PCR analysis conditions were as follows: 26–40 cycles of 30 sec at 94°C (denaturation), 30 sec at 53–58°C (annealing) and 30 sec at 72°C (extension). The primer sequences are listed in Table I. β-actin was used as a loading control. The products of qPCR were assessed on a 2% agarose gel stained with SYBR-Green (Invitrogen; Thermo Fisher Scientific, Inc.). The expression levels of mRNA were semi-quantified using ImageJ version 1.46.

To further investigate the anti-osteoporotic effects of MFR in vivo, an OVX rat model of menopausal osteoporosis was used. A number of 48 female Sprague Dawley (SD)-rats (age, 12 weeks; weight, 230–250 g) were obtained from KOATECH. The rats were housed in a standard environment with a controlled temperature of 22±2°C and humidity of 55±5%, under a 12-h light/dark cycle. Animals were provided with ad libitum access to water and food. The protocol for in vivo experiments was approved by the Kyung Hee Medical Center Institutional Animal Care and Use Committee (KHMC-IACUC; approval no. KHMC-IACUC 19–017). All animals were allowed to acclimate for 1 week. For the postmenopausal osteoporosis model, rats were deeply anesthetized using 5% isoflurane (inhaled in 100% oxygen) and the bilateral ovaries of the rats were removed under 2–2.5% isoflurane anesthesia. The sham group underwent the same surgery to ensure they experienced the same stress, but the ovaries were not removed. No animals died during surgery. To prevent infection of the surgical site, the wound was sutured and gentamicin (4 mg/kg) was injected intraperitoneally for 3 days. After 1 week, the rats were randomly divided into six groups (n=8/group): i) Sham group, in which rats underwent the sham operation and were treated with vehicle (water); ii) OVX group, in which OVX rats were treated with vehicle; iii) E₂ group, in which OVX rats were treated with E₂ (100 µg/kg); iv) ALN group, in which OVX rats were treated with ALN (5 mg/kg); v) MFR-L group, in which OVX rats were treated with a low dose of MFR (16.9 mg/kg); and vi) MFR-H group, in which OVX rats were treated with a high dose of MFR (169 mg/kg) for 8 weeks. E₂ and ALN were used as the positive controls. The humane endpoints used in the present study were: Dirty hair and eye discharge; self-injury and anxiety; vomiting and hemoptysis; inactivity; or anxiety and headache. None of the animals exhibited abnormal behavior.

MFR dose was calculated as follows: In Korean medicine, based on an average adult weight of 60 kg, a single dose of 8 g is recommended. MFR (1.016 g; yield, 12.7%) was considered equivalent to 8 g; thus, 16.9 mg MFR was required per 1 kg. Therefore, the MFR-L group was treated with 16.9 mg/kg. The MFR-H group was treated as follows: Rats are well-known to exhibit 6.4-fold faster metabolism than humans (32). In vitro experiments revealed that MFR exhibited a higher inhibition of osteoclast differentiation at high concentrations than at low concentrations. Based on these findings, rats in the MFR-H groups were treated with MFR at concentrations 10-fold higher than those in the MFR-L group to induce higher pharmacological effects. Therefore, the MFR-H group was treated with 169 mg/kg MFR (33–35). Oral administration was performed every morning for 8 weeks. Body weight was measured once a week and the dose was adjusted to weight. After 8 weeks, all animals were anesthetized by 5% isoflurane of inhaled anesthetics in 100% oxygen. Blood was collected using a cardiac puncture following sacrifice by cervical vertebrae dislocation. Subsequently, the uterus and femur were collected and weighed. Femurs samples were collected and fixed in 10% neutral buffered formalin for 1 day at room temperature. Uterus samples were collected and then stored at −80°C.

Blood samples were incubated at room temperature for 30 min, and centrifuged at 29,739 × g for 10 min at 4°C. Serum samples were stored at −80°C until required. Serum levels of alkaline phosphatase (ALP), aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured by DKKorea. C-telopeptide of collagen type 1 (CTX-1) levels were measured using an ELISA kit (Elabscience; cat. no. E-EL-R1456) according to the manufacturer's protocol. TRAP levels were measured using a TRAP kit as aforementioned.

The femoral head was scanned using micro-CT (SkyScan1176; Bruker Corporation). Bone microarchitecture parameters, including bone mineral density (BMD), trabecular thickness (Tb.Th) and trabecular separation (Tb.sp) were analyzed using NRecon software (SkyScan version 1.6.10.1; Bruker Corporation).

The fixed femur samples were washed using D.W at room temperature for 1 day and decalcified using 10% EDTA at room temperature for 3 weeks. Subsequently, the femur samples were dehydrated at room temperature for 1 day and embedded in paraffin. Paraffin-embedded tissues (5 µm) were sectioned on a rotary microtome (Carl Zeiss AG) and tissue sections were mounted on slides at room temperature for 1 day. The sections on the slides were stained with hematoxylin-eosin (H&E); sections were stained with hematoxylin for 10 min and with eosin for 10 sec at room temperature. Subsequently, all slides were sealed using mounting solution and the sections were viewed under a light microscope (Olympus Corporation; magnifications, ×40 and ×100) for histological evaluation.

Paraffin-embedded tissues were deparaffinized in xylene. Endogenous peroxidases were blocked in 0.3% hydrogen peroxide at room temperature for 15 min and proteinase K (0.4 mg/ml) was used for antigen-retrieval at 37°C for 30 min. The sectioned tissues were incubated with 10% normal serum (Gibco; Thermo Fisher Scientific, Inc.) at room temperature for 1 h to block nonspecific binding, then slides were washed with PBS and incubated at 4°C for overnight with primary antibodies against NFATc1 (1:100) and CTK (1:100). The following day, the slides were washed with PBS and incubated with a secondary antibody (1:100; rabbit; cat. no: BA-1000; Vector Laboratories, Inc.) at 4°C for 1 h. The signal was visualized using an ABC kit (Vector Laboratories, Inc.) and 3, 3′-diaminobenzidine solution (Vector Laboratories, Inc.). The stained tissues were imaged using a light microscope (magnifications, ×100 and ×200).

MC3T3-E1 cells were purchased from American Type Culture Collection. MC3T3-E1 cells were cultured in α-MEM without ascorbic acid containing 10% FBS and 1% P/S in a cell incubator at 37°C and 5% CO₂. To examine cell viability, the cells were seeded in a 24-well plate at 1×10⁴ cells/well and incubated at 37°C for 24 h. Subsequently, the cells were treated with α-MEM without ascorbic acid and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 24 h. Subsequently, 20 µl/well MTS solution was added to the cells and incubated at 37°C for 2 h. Cell viability was evaluated using an ELISA reader at a wavelength of 490 nm and was expressed as a percentage of the control.

MC3T3-E1 cells were seeded in a 24-well plate at 1×10⁴ cells/well and incubated for 24 h. Subsequently, osteoblast differentiation was induced using osteogenic medium (α-MEM without ascorbic acid supplemented with 10 mM β-glycerophosphate, 25 µg/ml ascorbic acid) and MFR (12.5, 25, 50 and 100 µg/ml) at 37°C for 14 days. The medium was replaced every 2 days. The cells were then washed with DPBS, fixed with 80% ethanol at room temperature for 1 h and stained with Alizarin red S at room temperature for 5 min. To quantify mineralization, the stained dye was extracted using 10% (v/w) cetylpyridinium chloride in sodium phosphate at room temperature for 15 min. The extracted dye was measured using an ELISA reader at a wavelength of 570 nm.

Data are presented as the mean ± standard error of the mean of three experiments. Statistical analyses were performed using GraphPad Prism version 5.01 (GraphPad Software, Inc.). In vitro and in vivo experiments were compared using a one-way ANOVA followed by a Tukey's post hoc analysis. Animal body weight was compared using a two-way ANOVA followed by Bonferroni post hoc test. P<0.05 was considered to indicate a statistically significant difference.

Vitexin was used as a standard marker of MFR, as described previously (29). The chromatogram of the MFR water extract possessed several peaks at a retention time of 0–30 min, and vitexin was observed at the same retention time as the standard (Fig. 1).

To analyze the anti-osteoporotic effect of the MFR extract in vitro, TRAP staining was performed using a TRAP kit. Following treatment with RANKL, TRAP staining revealed that the number of TRAP-positive cells was reduced by MFR treatment. Consistent with the results of TRAP staining, MFR decreased TRAP levels in the differentiation medium (Fig. 2A-C). As the actin ring is essential in osteoclast differentiation (36), the effect of MFR on actin ring formation was determined using immunocytochemistry. In addition, the effect of MFR on pit formation was determined using osteo-coated plates. The actin ring structures and the area of bone resorption pits were increased in the RANKL-treated cells, but were reduced following MFR treatment (Fig. 2D and E). Consistent with these results, MFR treatment reduced the number of actin rings in a dose-dependent manner (Fig. 2F). In addition, the area of bone resorption pits was significantly reduced following MFR treatment (Fig. 2G). To confirm whether the concentration of MFR used in the in vitro experiments affected the viability of RAW 264.7 cells, the cells were treated with 12.5–100 µg/ml MFR. MFR did not affect the viability of RAW 264.7 cells or mature osteoclasts (Fig. 2H and I). In addition, the present study compared and analyzed the inhibitory effect of MFR and vitexin on osteoclast differentiation to determine if the ability of MFR to inhibit osteoclast differentiation was associated with vitexin. The content of vitexin in MFR was 1.47 ppm (0.147%); 0.147 µg/ml vitexin in 100 µg/ml MFR and 0.0753 µg/ml vitexin in 50 µg/ml MFR. The effects of the two substances (MFR and vitexin) on TRAP staining and TRAP levels in the medium were subsequently assessed. To confirm whether the concentration of vitexin contained in MFR affects the osteoclast inhibitory effect, TRAP staining was performed. As shown in Fig. S1A and B, TRAP-positive cells and TRAP levels were increased in the RANKL-treated cells. Conversely, the number of TRAP-positive cells was considerably decreased by MFR (50 and 100 µg/ml) treatment compared with vitexin (0.0753 and 0.147 µg/ml). Similarly, MFR (50 and 100 µg/ml) further decreased TRAP levels compared with vitexin (0.0753 and 0.147 µg/ml) (Fig. S1B).

To determine whether the concentration of vitexin contained in MFR affects the inhibitory effect on transcription factors, NFATc1 and c-Fos were measured using western blotting. The protein expression levels of NFATc1 and c-Fos were increased by RANKL treatment. MFR (50 and 100 µg/ml) inhibited the expression of NFATc1 and c-Fos compared with vitexin (0.0753 and 0.147 µg/ml) (Fig. S1C and D). To confirm whether the concentration of vitexin used in experiments affected the viability of RAW 264.7 cells, the cells were treated with vitexin (0.0753 and 0.147 µg/ml); vitexin not affect cell cytotoxicity (Fig. S1E). Taken together, it was revealed that the osteoclast inhibitory effect of MFR was not mediated by vitexin.

NFATc1 and c-Fos are important transcription factors required for mature osteoclast differentiation in RAW 264.7 cells (37,38). RANKL-induced expression of NFATc1 and c-Fos was measured using western blotting and RT-qPCR. Treatment with RANKL significantly increased the expression levels of NFATc1, whereas MFR decreased the protein and mRNA expression levels of NFATc1 (Fig. 3A and B). In addition, c-Fos expression was significantly increased by RANKL treatment, whereas this increase was reversed by MFR treatment (Fig. 3C and D). Furthermore, RANKL stimulation increased TRAF6, but this finding was not significant, and treatment of MFR attenuated the increased expression of TRAF6. In particular, in cells treated with 25 µg/ml MFR, the expression of TRAF6 was significantly decreased compared with that in RANKL-only treated cells (Fig. 3E and F).

The inhibitory effects of the MFR extract on bone resorption markers were examined using western blotting and RT-qPCR. Treatment with RANKL significantly increased the expression levels of MMP-9 and CTK, whereas MFR decreased the protein expression levels of MMP-9 and CTK (Fig. 4A and B). Consistent with the results of western blotting, MFR reduced the mRNA expression levels of Mmp9, Ctsk and carbonic anhydrase 2 (CA2/Ca2) (Fig. 4C and D). In addition, the effects of MFR on RANKL-stimulated changes in osteoclast-specific genes, including Acp5, Atp6v0d2, dendritic cell-specific transmembrane protein (DC-STAMP/Dcstamp), Oscar, c-Src (Src) and B lymphocyte-induced maturation protein-1 (Blimp-1/Prdm1), were determined, as they are crucial in promoting osteoclastogenesis (39). The mRNA expression levels of osteoclast-specific genes were increased by RANKL treatment, whereas MFR attenuated the RANKL-induced increase in the mRNA expression levels of osteoclast-specific genes, including Acp5, Atp6v0d2, Dcstamp, Oscar, Src and Prdm1, in a dose-dependent manner (Fig. 4E-H).

As shown in Fig. 5A, the OVX group exhibited a significant increase in body weight compared with that in the sham group from 4 weeks. The E₂ group had a significantly lower increase in body weight after 3 weeks compared with the OVX group; however, MFR-L, MFR-H and ALN treatment did not affect body weight. In addition, the OVX group exhibited significantly decreased uterus weight compared with that in the sham group, whereas E₂ treatment prevented uterus weight loss. MFR-L, MFR-H and ALN groups did not exhibit any protective effects on uterus weight loss (Fig. 5B). While E₂ and ALN groups exhibited a preventive effect on the reduction of femur weight compared with the OVX group, MFR-L and MFR-H groups did not (Fig. 5C). To investigate the extent of hepatotoxicity following treatment with MFR, E₂ and ALN, the serum levels of AST and ALT were measured. Previous studies have shown that levels above 150 U/l for AST and 40 U/l for ALT indicate hepatotoxicity in rats (40,41). The levels of AST and ALT in the MFR-L, MFR-H, E₂ and ALN groups in the present study did not exceed these values; therefore, it was indicated that MFR-L, MFR-H, E₂ and ALN did not induce hepatotoxicity (Fig. 5D and E). To determine the effect of MFR on bone formation, the serum levels of ALP were measured. ALP was significantly increased in the OVX group compared with that in the sham group; moreover, the MFR-L group exhibited decreased serum levels of ALP, whereas the MFR-H, E₂ and ALN groups did not (Fig. 5F). To examine the effect of MFR on bone resorption, serum levels of CTX-1 and TRAP were measured. CTX-1 levels were increased in the OVX group compared with those in the sham group, whereas they were significantly decreased in the MFR-L, E₂ and ALN groups compared with those in the OVX group; there were no significant changes in the MFR-H group (Fig. 5G). As shown in Fig. 5H, TRAP levels were increased due to OVX, but the difference was not significant, whereas oral administration of MFR-L, E₂ and ALN reduced TRAP levels. Furthermore, the TRAP levels were decreased in the MFR-H group; however, this finding was not significant.

In order to investigate the protective effects of MFR on ovariectomy-induced bone loss, micro-CT was used. As shown in Fig. 6A, micro-CT images of the femoral head indicated trabecular bone loss in OVX rats. BMD loss was significantly inhibited in the MFR-L, E₂ and ALN groups compared with that in the OVX group (Fig. 6B). Furthermore, Tb.Th in the OVX group was reduced compared with that in the sham group, whereas Tb.Th was significantly increased in the ALN group. However, Tb.Th in the MFR-L, MFR-H and E₂ groups did not differ significantly from that in the OVX group (Fig. 6C). As shown in Fig. 6D, Tb.sp was significantly increased in the OVX group compared with that in the sham group, whereas Tb.sp levels were decreased in the E₂ and ALN groups. However, Tb.sp levels were not affected in the MFR-L and MFR-H groups.

To investigate the histological changes in the femoral head, H&E staining was performed on the bone tissue (Fig. 7A). The trabecular area in the femoral head of OVX rats was significantly decreased compared with that in the sham group, whereas this reduction was prevented by treatment with MFR-L, E₂ and ALN (Fig. 7B).

To determine the effect of MFR on the expression levels of NFATc1 and CTK in the rat model of osteoporosis, IHC was performed (Fig. 8A and B). The OVX group exhibited increased expression levels of NFATc1 compared with those in the sham group. By contrast, the MFR-L and E₂ groups exhibited significantly reduced expression levels of NFATc1 expression; no changes were observed in the MFR-H and ALN groups compared with the OVX group (Fig. 8C). In OVX rats, the expression levels of CTK were increased compared with those in the sham group. Conversely, MFR-L, MFR-H, E₂ and ALN effectively reduced the expression levels of CTK (Fig. 8D).

MC3T3-E1 cells derived from mouse calvaria are characterized by proliferation, differentiation and mineralization of osteoblasts, and are used in studies related to bone formation (42). Alizarin red staining can be used to detect the formation of mineralization (43). Therefore, to confirm the effect of MFR on osteoblast differentiation and mineralization, Alizarin red S staining was performed. The control cells exhibited significantly increased mineralization compared with that in the normal cells. Treatment with low concentrations of MFR (12.5 and 25 µg/ml) promoted osteoblast differentiation, but high concentrations of MFR (50 and 100 µg/ml) inhibited osteoblast differentiation (Fig. 9A). To examine the effects of the MFR extract on MC3T3-E1 cell viability, cells were treated with 12.5–100 µg/ml MFR. Notably, MFR did not affect the viability of MC3T3-E1 cells (Fig. 9B). Quantification of Alizarin red staining revealed a significantly reduced effect of MFR in a dose-dependent manner (Fig. 9C).