Dual effects of baicalin on osteoclast differentiation and bone resorption

Abstract Osteoclasts (OC) are critical cells responsible for many bone diseases such as osteoporosis. It is of great interest to identify agents that can regulate the activity of OC to treat osteolytic bone diseases. In this study, we found that baicalin exerted a two‐way regulatory effect on OC in a concentration‐dependent manner in vitro and in vivo. In detail, baicalin at a low concentration (below 1 μmol/L) enhanced OC differentiation and bone resorption, but baicalin at a high concentration (above 2 μmol/L) exhibited inhibitory effects on OC. We demonstrated that baicalin at low concentrations enhanced the mitogen‐activated protein kinase (MAPK) (ERK) signalling pathway and activated c‐Fos and NFATc1 expression, and thus enhanced gene expression, OC differentiation and bone resorption. However, baicalin at higher levels not only suppressed ERK phosphorylation and c‐fos and NFATc1 expression, but also altered the expression of apoptosis‐related proteins, and therefore inhibiting OC function. This dual effect was further verified in an LPS‐induced mouse calvarial osteolysis model, evidenced by enhanced osteolysis at a lower concentration but reduced bone loss at a higher concentration. Overall, our findings indicate that baicalin exerts dose‐dependent effects on OC formation and function. Therefore, caution should be applied when using baicalin to treating OC‐related bone diseases.

expression of osteoclastic key transcriptional factors such as c-Fos and nuclear factor of activated T-cells 1 (NFATc1), finally inducing the formation of functional mature osteoclasts. 10 In addition, these cytokines are also required to stimulate and maintain the differentiation, survival, and apoptosis of osteoclasts by modulating the mitochondrial pathway (Bcl-2/Bax) and the death receptor pathway (caspases). [11][12][13] Many studies have demonstrated that flavonoids modify the activity of osteoclasts in vitro and in vivo. [14][15][16] Therefore, flavonoids have aroused a growing interest as bone-modifying compounds for the therapy of osteoclast-related bone diseases. 17 Among them, baicalin has been extensively used to treat different diseases owing to its anti-oxidant, 18 anti-inflammatory 19,20 and anti-cancer effects. 21,22 Baicalin has been reported to suppress inflammation through inhibiting NF-κB, which plays an essential role in differentiation and function of osteoclasts. 23,24 These findings hint that baicalin may regulate the activity of osteoclasts.
Indeed, Lu et al reported positive effects of baicalin on osteoclast formation and function at low concentrations ranging from 0 to 1 μmol/L in vitro. 25 Based on this positive effect on osteoclastogenesis, in this study, we aimed to further investigate the effects of baicalin on osteoclastogenesis at a wider concentration range from 0 to 8 μmol/L in vitro, to explore the underlying molecular mechanisms, and to verify the effect of baicalin on osteoclasts in vivo.

| Cell culture and induction of osteoclastogenesis
Mouse bone marrow-derived macrophages (BMMs) were isolated from the femurs and tibiae of mice and cultured in α-MEM supplemented with 10% FBS, 25 ng/mL M-CSF, 100 U/mL penicillin and 100 mg/mL streptomycin at 37°C in a humidified 5% CO 2 atmosphere for 4 days. This complete medium was changed every 2 days.
The M-CSF-dependent BMMs as osteoclast precursors were then resuspended and seeded in 96-well plates at a density of 8 × 10 3 cells per well for differentiation and bone resorption or in 6well plates at a density of 30 × 10 4 cells per well for protein or RNA extraction. For osteoclastogenesis, BMMs were cultured in α-MEM supplemented with 10% FBS, RANKL (100 ng/mL), and M-CSF (25 ng/mL) at 37°C in a humidified 5% CO 2 atmosphere. The complete medium was changed every 2 days. After stimulation for 8 days, images of multinucleated osteoclasts were captured by a light microscope (Eclipse TS100; Nikon, Tokyo, Japan).

| Tartrate-resistant acid phosphatase staining and activity assay
For tartrate-resistant acid phosphatase (TRAP) staining, multinucleated osteoclasts from BMMs were fixed with 4% paraformaldehyde.
After a 20-minutes fixation, BMMs were rinsed with 1× phosphatebuffered saline (PBS) three times and then incubated with TRAP staining solution (Sigma-Aldrich) for 30 minutes at 37°C. Images of multinucleated cells were captured by light microscopy, and then TRAP-positive multinucleated cells containing three or more nuclei F I G U R E 1 Cytotoxicity of Baicalin in Osteoclasts (OC) Precursors. A, Chemical structure of baicalin. B, 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide assay was performed to measure the cytotoxicity of baicalin in OC precursors following culture for 48 h in a concentration range of 0-32 μmol/L. Cell viability was measured at a wavelength of 570 nm. Values are the mean ± standard deviation (n = 3; *P < 0.05) were then counted. The sizes of matured osteoclasts were determined using ImageJ software (National Institutes of Health, Bethesda, MD).

| Cytotoxicity assays
BMMs were suspended in α-MEM supplemented with 10% FBS and M-CSF (25 ng/mL) and then seeded in 96-well plates at a density of The experiment was performed in triplicate.

| Pit formation assay
BMMs were plated into 96-well plates containing 100-μm bovine bone slices (Rongzhi Haida Biotech Co., Ltd., Beijing, China) on the bottom at a density of 9 × 10 3 cells per well and then treated with baicalin at 0, 1, or 8 μmol/L in medium containing RANKL (100 ng/ mL) and M-CSF (25 ng/mL). After differentiation for 8 days, the cells were removed, the resorptive pits were measured using a scanning electron microscope (Field Environmental Instruments Inc., Hillsboro, OR), and the area of the pits was measured using ImageJ software (National Institutes of Health).
Briefly, total RNA was isolated using TRIzol reagent (Life Technologies, Carlsbad, CA). First-strand cDNA was then synthesized using PrimerScript reverse transcription reagent kit (Takara, Shiga, Japan). The pre-set cycling parameters were as follows: 40 cycles of 94°C for 20 seconds, 60°C for 20 seconds and 72°C for 30 seconds.
The β-actin gene was used as an internal control to normalize the results.

| Western blot analysis
Cellular proteins were extracted following cell disintegration with radioimmunoprecipitation assay lysis buffer and centrifuged at 16 000 × g for 10 minutes at 4°C. The supernatant containing protein was collected, and the protein concentration was measured using a bicinchoninic acid (BCA) protein assay kit (Beyotime Biotechnology, Shanghai, China). The protein was then mixed with sodium dodecyl sulfate-sampling buffer, followed by incubation at 95°C for 5 minutes.
The protein samples were separated and transferred by electroblotting onto membranes, which were incubated with blocking buffer for 1 hour. The blocked membranes were then incubated with the targeted antibody overnight at 4°C, washed three times with 1× Tris-buffered saline plus Tween (TBST) for 5 minutes each time, and then incubated with the secondary antibody for 1 hour. Finally, the bands were detected via analysis of immunoreactivity using a Western Lighting Ultra Kit with a FujiFilm Las-4000 gel documentation system and quantified using a chemiluminescence imaging system (ChemiDoc XRS, Bio-Rad, Hercules, CA).

| In vivo murine calvarial model of lipopolysaccharide-induced osteolysis
Forty male 8-week-old C57/BL6 mice were purchased from Shanghai SLAC Laboratory Animal Co. (Shanghai, China). All mice were fed in a well-ventilated controlled room at 25°C on a 12-hours light/dark cycle and allowed free access to water and food. The protocol was

| Micro-computed tomography scanning
Three-dimensional reconstructions of the whole calvaria were obtained from images acquired using a high-resolution micro-com- The bone volume per total volume (BV/TV) was analysed for each sample.

| Histological analyses
After fixing in 10% formaldehyde for 3 days, calvarial bones were

| Statistics
The results are expressed as the mean ± standard deviations. The differences between two groups and multiple comparisons were evaluated using an unpaired, two-tailed Student's t-test and one-

| Cytotoxicity effect of baicalin on bone marrow-derived monocytes/macrophages
The chemical formula of baicalin is shown in Figure 1A.

| Baicalin regulated RANKL-induced osteoclast fusion and bone resorption in a concentration-dependent manner
The cell fusion and bone resorption of osteoclasts treated by baicalin also present dual regulation in a concentration-dependent manner as F I G U R E 2 Effect of Baicalin on Osteoclast (OC) Formation. Bone marrow-derived macrophages as OC precursors were differentiated into mature OC by receptor activator of nuclear factor (NF)-κB ligand (RANKL, 100 ng/mL) and M-CSF (25 ng/mL) stimulation. A, TRAP-positive cells were captured using an inverted microscope. B, The number and area of baicalin-treated OC (0-8 μmol/L) were measured using Image J. C, mRNA expression of OC-specific gene markers was measured using real-time polymerase chain reaction. Values are the mean ± standard deviation (n = 3; *P < 0.05, **P < 0.01, ***P < 0.001) determined using the actin ring assay and pit formation assay. The F-actin ring formation of osteoclasts was increased in the low-concentration group (1 μmol/L), but decreased in the high-concentration group (8 μmol/L) ( Figure 3A). The number of cells with F-actin rings formed was 65.01 ± 5.01 cells per well (P < 0.01) in the 1 μmol/L group and 7.33 ± 2.51 cells per well (P < 0.001) in the 8 μmol/L group ( Figure 2B). The area of F-actin ring formation presented a similar tendency ( Figure 2B).

| Dual effect of baicalin on LPS-induced osteolysis in vivo
The dual effect of baicalin was verified for LPS-induced inflammatory osteolysis in a mouse calvaria model. Consistent with the above F I G U R E 3 Effect of Baicalin on the Fusion and Bone Resorption of Osteoclasts. Bone marrow-derived macrophages were seeded on plates or bone slices upon exposure to receptor activator of nuclear factor (NF)-κB ligand (RANKL, 100 ng/mL) and M-CSF (25 ng/mL) stimulation for 8 d. A, Image of F-actin ring fusion of osteoclasts was acquired using fluorescence microscopy. B, The number and area of F-actin rings on treatment with baicalin were measured using ImageJ. C, Images of bovine bone films were captured using scanning electron microscope. D, Area of bone resorption was measured by ImageJ software. Values are the mean ± standard deviation (n = 3; *P < 0.05, *P < 0.05, **P < 0.01, ***P < 0.001) F I G U R E 4 Baicalin Regulates Erk/c-Fos/ NFATc1. A, RANKL-induced activity of PI3K/Akt, MAPKs and NF-κB following treatment with baicalin (0-8 μmol/L) and M-CSF (25 ng/mL) for 30 min was evaluated using western blotting. B, Expression of c-Fos and NFATc1 as downstream markers following treatment with baicalin (0-8 μmol/L) for 72 h was measured using western blotting. C, D, The change in Erk activation was measured by determining phosphorylated vs unphosphorylated forms, the change in c-Fos and NFATc1 was measured by total forms vs β-actin expression which were quantified using a chemiluminescence imaging system. Values are the mean ± standard deviation (n = 3; *P < 0.05, *P < 0.05, **P < 0.01, ***P < 0.001) F I G U R E 5 Erk Inhibitor U0126 Inhibits RANKL-Induced Osteoclasts Treated by Baicalin. Bone marrow-derived macrophages were differentiated into mature osteoclasts by receptor activator of nuclear factor (NF)-κB ligand (RANKL, 100 ng/mL) and M-CSF (25 ng/mL) stimulation, supplemented with baicalin and U0126. A, Tartrate-resistant acid phosphatase (TRAP)-positive cells were captured by an inverted microscope. B, C, Alteration of Erk activation on treatment with baicalin and the Erk inhibitor U0126 was measured by determining phosphorylated vs unphosphorylated forms, which were quantified using a chemiluminescence imaging system. Values are mean ± standard deviation (n = 3; *P < 0.05, *P < 0.05, **P < 0.01, ***P < 0.001) results in vitro, the results from micro-CT showed that the LPS + baicalin (6 mg/kg body weight, low-dose group) group presented more calvarial osteolysis than the LPS group, whereas the LPS + baicalin (12 mg/kg body weight, high-dose group) group presented less calvarial osteolysis than the LPS group ( Figure 6A). The morphometric statistical analysis also suggested a dual effect for BV/TV, which was 18.67 ± 4.16% (P < 0.01) in the low-dose group and 62.33 ± 2.52% (P < 0.001) in the high-dose group compared with 33.33 ± 3.06% in the LPS group. The bone resorption area presented a similar tendency ( Figure 6B).
Histological analyses of TRAP-positive osteoclasts also showed that the low-dose group presented more osteoclasts whereas the high-dose group presented less osteoclasts than the LPS group ( Figure 6D).

| DISCUSSION
In this study, we found that baicalin enhanced osteoclastogenesis at There are many agents that exhibit dual effects according to concentration. Among clinical medicines, aspirin is a classic drug that possesses antipyretic and analgesic effects at high doses and has antiplatelet aggregation effects at low doses. 26 In the osteoclasts, cyanidin also has been shown to have dual effects on the activity of Further studies of the downstream factors related to osteoclastogenesis showed that the expression of c-Fos and NFATc1 showed trends similar to those observed for ERK activity. 29 We deduced that baicalin affected osteoclastogenesis through dual effects on the regulation of ERK/c-Fos/NFATc1 signalling cascades. Osteoclastic-specific gene markers, such as TRAP, V-ATPase d2, cathepsin K and MMP-9, also showed these dual regulatory effects of baicalin according to concentration 29 (Figure 7).
Additionally, baicalin has been found to induce apoptosis in various cancer cells, such as ovarian cancer cells and human osteosarcoma cells. 21,30 Therefore, we measured the levels of apoptosisrelated proteins in osteoclast precursors treated with baicalin. In the apoptosis process, anti-apoptotic proteins such as Bcl-2 and proapoptotic proteins such as Bax regulate downstream proteins such as caspase-3 to induce apoptosis in cells. 31,32 In this study, the Bcl-2/Bax ratio was decreased and levels of the inactive form of caspase-3 were decreased at a high concentration of baicalin (from 2 to 8 μmol/L).
F I G U R E 6 Baicalin Regulates LPS-Induced Calvarial Osteolysis. Eight-week-old C57/BL6 mice were treated with PBS, LPS and LPS + baicalin (low-dose or high-dose) through subcutaneous injections for 7 d. A, Image of calvarial bone was captured by micro-CT. B, Per cent bone volume relative to tissue volume (BV/TV %) and per cent resorption area of calvarial bone were measured by ImageJ software. C, Images of H&E-and TRAP-stained calvarial bone were captured by Scano Microct u100. D, TRAP-positive OC number and TRAP-positive OC number/bone surface were measured by ImageJ. Values are the mean ± standard deviation (n = 3; *P < 0.05, *P < 0.05, **P < 0.01, ***P < 0.001) Although we demonstrated this dual effect of baicalin on osteoclastogenesis in vitro and in vivo, our study does not fully explain the specific molecular mechanisms underlying this dual effect of baicalin on bone remodelling. First, we can study why baicalin could dual-regulate the activity of MAPK (ERK). There may be a link between the regulation of MAPK (ERK) and apoptosis-related proteins induced by baicalin, which would require further studies on the common upstream signalling pathway. Moreover, in vivo, the capability of subcutaneous injection is limited to some degree. Intragastric administration may be a more appropriate method to study the effects of baicalin on osteoclastogenesis in vivo.
In summary, we demonstrated a dual effect of baicalin on the differentiation and function of osteoclasts in a dose-dependent manner. These dual effects of baicalin suggest its potential as a therapeutic for osteoclast-related bone diseases, although caution is necessary regarding the potential widespread side effects of a drug that can inhibit NF-κB and ERK pathways. Moreover, our findings regarding these dual effects of baicalin may guide the use of raw herbs containing baicalin in the treatment of different bone diseases.