Dracorhodin perchlorate inhibits osteoclastogenesis through repressing RANKL‐stimulated NFATc1 activity

Abstract Osteolytic skeletal disorders are caused by an imbalance in the osteoclast and osteoblast function. Suppressing the differentiation and resorptive function of osteoclast is a key strategy for treating osteolytic diseases. Dracorhodin perchlorate (D.P), an active component from dragon blood resin, has been used for facilitating wound healing and anti‐cancer treatments. In this study, we determined the effect of D.P on osteoclast differentiation and function. We have found that D.P inhibited RANKL‐induced osteoclast formation and resorbed pits of hydroxyapatite‐coated plate in a dose‐dependent manner. D.P also disrupted the formation of intact actin‐rich podosome structures in mature osteoclasts and inhibited osteoclast‐specific gene and protein expressions. Further, D.P was able to suppress RANKL‐activated JNK, NF‐κB and Ca2+ signalling pathways and reduces the expression level of NFATc1 as well as the nucleus translocation of NFATc1. Overall, these results indicated a potential therapeutic effect of D.P on osteoclast‐related conditions.


| INTRODUC TI ON
Dracorhodin perchlorate (D.P), derived from dracorhodin, is a major active component of dragon blood (Sanguis draxonis or Daemonorops draco) resin, a traditional herbal medicine, which was commonly used for the therapy of injuries, such as wounds and fractures. [1][2][3] D.P has multiple applications in medicine, for example, D.P is found to enhance cutaneous wound healing by promoting fibroblasts proliferation in rats. 4,5 D.P also exerts strong angiogenic effects, which can be beneficial for wound healing. 6,7 D.P is found to have an anti-cancer function by inducing apoptosis in human cancer cells. [8][9][10][11][12] Reported in some other literatures, D.P has anti-microbial effects by reducing the production of α-toxin by Staphylococcus aureus 13 and inhibiting the formation of biofilm and virulence factors of Candida albicans. 14 We have confirmed that several compounds extracted from S draxonis resin have inhibitory effects on osteoclast differentiation. 15 However, the effect of D.P on skeletal disorders is not yet determined.
Bone diseases, such as osteoporosis, osteonecrosis and Paget's disease, are all related to the over-activation of osteoclasts. Osteoclasts, the bone resorptive cells, and osteoblasts, the bone-forming cells, interact with each other in a way that the skeletal tissue is able to undergo constant remodelling and maintain the daily life of the vertebrates. 16 However, imbalance of activities between these two types of cells can occur under certain situations, for example, postmenopausal women have higher risks of developing osteoporosis than other population due to oestrogen deficiency. 16,17 There are limited options of the treatment of osteoporosis which is of huge threat to the patients' well-being and causing enormous socio-economic burdens. [18][19][20] Osteoclasts are originated from the haematopoietic stem cells. The differentiation of osteoclasts from the precursor cells is dependent on two important signalling molecules, macrophage colony-stimulating factor (M-CSF) and receptor activated nuclear factor kappa-B ligand (RANKL). 21 RANKL belongs to the TNF family, which binds to RANK on the surface of osteoclast precursor cells and mediates the formation and function of osteoclast. Upon binding to RANK, RANKL activates several pathways to initiate the cell differentiation of osteoclast precursors to mature osteoclasts, importantly, MAPK pathway and NF-κB pathway. [22][23][24] These two pathways both activate the transcription of nuclear factor of activated T cells 1 (NFATc1), the master transcription factor of osteoclast-specific genes, such as cathepsin K, TRAcP, calcitonin receptor and c-Fos. 25 These proteins all contribute to the formation and function of osteoclast.
In this study, using in vitro experiments, we were able to show the inhibitory effect of D.P on RANKL-induced osteoclast formation and function. We found this inhibitory effect was mainly caused by impeding the activation of NF-κB and MAPK pathways, which leads to a reduction of NFATc1 expression. Furthermore, the activation of calcium pathway and NFATc1 was also suppressed by D.P. Our study demonstrated the therapeutic potential of D.P to osteoclast-related bone diseases.

| Reagents
Dracorhodin perchlorate (D.P) was obtained from Chengdu Must Bio-Technology Co., Ltd, and was diluted in DMSO at a concentration of 100 mmol/L for storage, which was further diluted to the working dilutions with PBS. Alpha-Minimum essential medium Antimouse IgG-FITC antibody (Cat#F9137), anti-vinculin antibody (Cat#V9264), anti-tubulin antibody (Cat#T3526) and recombinant M-CSF (Cat#M6518) were commercially available from Sigma-Aldrich. Recombinant GST-rRANKL was produced and purified as previously described. 26

| Cell culture and osteoclast formation
Bone marrow macrophages (BMMs) were isolated by flushing the bone marrow out of the hindlimbs from C57BL/6J mice (female, 6-week-old). The resulting bone marrow cell suspension was then filtered with a 100 μm strainer, resuspended and cultured in α-MEM with 10% FBS, 100 U/mL penicillin, 100 mg/mL streptomycin (complete α-MEM). To maintain the growth of BMM culture, 50 ng/mL of M-CSF was also supplemented into the cell culture media. When the adherent cell became confluent, 6 × 10 3 BMMs were seeded into 96-well plates per well and cultured overnight at 37°C in a moist environment with 5% CO 2 . Starting the following day, the BMMs were stimulated with 50 ng/mL of RANKL, with different concentrations of D.P (0, 1, 5, 10, 20,
After 48 hours, MTS reagent was added to the cells and incubated. MTS measurement was performed as per manufacturer's instruction.

| Hydroxyapatite resorption assay
BMMs were initially plated onto collagen-coated 6-well plates After the incubation, the wells were washed with ddH 2 O, followed by that three wells were stained for TRAcP activities and the other three wells were subsequently washed with 10% bleach to get rid of the cells to expose bone resorption areas. The wells were then imaged using an inverted light microscope and analysed using ImageJ software (NIH).

| RNA extraction and real-time PCR
BMMs (1 × 10 5 cells/well) were cultured and stimulated with RANKL (50 ng/mL) in 6-well plates with the addition of different F I G U R E 3 D.P inhibited RANKLinduced osteoclast-specific gene expression. Real-time PCR was utilized to determine the expression levels of osteoclast-specific genes, Acp5, c-fos, Ctsk and Mmp9. Expression levels were normalized to the mean of Hprt1 and Hmbs. Data were presented as fold changes relative to RANKL control group. n = 3; *P < .05, **P < .01, ***P < .001 F I G U R E 4 D.P supressed osteoclast marker protein expression. (A) Representative images of Western blot on the effects of D.P on osteoclast-specific proteins during osteoclastogenesis, c-Fos, Integrin β3, MMP9 and CTSK. β-actin was used as a loading control. (B) Quantitative analysis of the band intensities relative to β-actin intensity. n = 3; *P < .05, ***P < .001

| Western blot
For the study of osteoclast-specific protein expression, the BMMs were seeded into 6-well plates at a density of 1 × 10 5 cells/well and stimulated with RANKL (50 ng/mL) and in the presence or absence and an Image-quant LAS 4000 system (GE Healthcare). The intensities of bands were measured and analysed using ImageJ software.

| Nuclear and cytoplasmic extraction
BMMs were seeded at a density of 1 × 10 5 cells/well in 6-well plates.
The cells were then stimulated with 50 ng/mL RANKL with or with-   For luciferase reporter assay, 1.5 × 10 4 cells/well were seeded into 48-well plates. After overnight incubation, the cells were serumstarved for 2 hours and pre-treated with different concentrations of D.P for 1 hour. Then, the cells were stimulated with 50 ng/mL RANKL for 24 hours. When the time-point was achieved, the cells were lysed, and luciferase reporter assay was performed using Luciferase Assay System (Promega) as per manufacturer's instructions.

| Statistical analysis
All experiments were performed with at least 3 replicates. All data in the study were presented as mean ± standard deviation. Statistical analyses were all performed using GraphPad Prism 7 software. For the comparison of two sets of data, Student's t tests with assumption of equal variance were used, and for the comparison among 3 or more groups of data, one-way ANOVA tests were used with Dunnett's test for multiple comparisons between the groups. A P-value ≤ .05 F I G U R E 6 D.P inhibited the NFATc1 expression. (A) Representative traces and quantitative analysis of the calcium oscillation intensity. The results were presented as calcium oscillation frequency. (B) Gene expression levels of Ctr and Nfatc1 were determined using qPCR. The expression levels were normalized to the average expression of Hprt1 and Hmbs and presented as fold changes relative to the RANKL control group. (C) Representative Western blot images of NFATc1 protein level after stimulation with RANKL for 5 days, in the presence or absence of D.P. β-actin was used as loading control. (D) Quantitative analysis of the NFATc1 protein expression levels relative to the RANKL control group. n = 3; **P < .01, ***P < .001 between two groups was considered to be statistically significant (95% confidence interval).

| D.P inhibited RANKL-induced osteoclastogenesis in vitro
The chemical structure of D.P was shown in Figure 1D. To determine the effect of D.P on RANKL-induced osteoclast formation, BMMs were seeded onto 96-well plates and RANKL was then added with or without the presence of DP It was observed in the RANKL control group that TRAcP-positive multinucleated cells appeared after the stimulation with RANKL. However, with the increasing concentrations of D.P, both the size and the number of multinucleated cell were reduced by the treatment of D.P in a dose-dependent manner ( Figure 1A,B). To eliminate any potential toxicity effects of D.P to BMMs, MTS assay was performed. It was found that D.P had no significant cytotoxic effect on BMMs within the concentration range, used in our study ( Figure 1C).

| D.P attenuated RANKL-induced F-actin belt formation and osteoclastic hydroxyapatite resorption in vitro
Hydroxyapatite-coated plates were used to examine the effect of D.P on osteoclast function. As the result, the RANKL control group

| D.P suppressed the expression of osteoclastspecific genes and proteins
Real-time quantitative PCR was used to determine the effect of D.P on RANKL-induced expressions of osteoclast-specific genes.
Concluding the results, the expressions of Acp5, Ctsk and Mmp9 induced by RANKL, after 5 days of culture, were inhibited by the addition of D.P in a dose-dependent manner (Figure 3).
Western blot was utilized to examine the expression levels of osteoclast-specific proteins. Although the genetic expression level of c-fos was not interfered by D.P, the protein expression of c-Fos was suppressed. Other osteoclast-specific proteins had an increase in expression level upon the induction by RANKL during the 5-day culture, but these enhanced expressions of Integrin β3, MMP9 and CTSK were decreased by the addition of D.P (Figure 4). These results were in line with osteoclast formation assay and hydroxyapatite resorption assay.

| D.P suppressed the activation of JNK and p65
To elucidate the suppression mechanism of D.P on RANKL-induced osteoclast formation and function, Western blot was used to examine the activation of some key signalling proteins. According to the results, D.P inhibited the phosphorylation of JNK by RANKL stimulation in BMMs. It was also observed that D.P inhibited the degradation of IκB-α, and the phosphorylation of p65 subunit of NF-κB complex mediated by RANKL, indicating an inhibitory effect of D.P on the activation of NF-κB signalling pathway ( Figure 5).

| D.P suppressed the activation of NFATc1 by RANKL
To examine the effect of D.P on Ca 2+ signalling pathway in the RANKL-induced osteoclastogenesis, calcium oscillation assay was performed. Shown by the results, the intensity of calcium oscillation was increased by RANKL, but D.P strongly suppressed the RANKL-induced calcium oscillation ( Figure 6A). Further to the calcium oscillation, it was found by real-time PCR that the gene expression levels of Ctr and Nfatc1 during RANKL-induced osteoclastogenesis were also suppressed by D.P ( Figure 6B). The protein level of NFATc1 was determined using Western blot. The increased expression of NFATc1 protein induced by RANKL was inhibited by D.P ( Figure 6C,D). Furthermore, we determined the attenuation of the NFATc1 signalling pathway by examining the trans-nuclear movement of NFATc1 using Western blot and immunofluorescent staining of nuclear and cytoplasmic proteins. It was seen that the NFATc1 protein was translocated into the nucleus upon RANKL stimulation; however, the translocation of NFATc1 was inhibited by D.P ( Figure 7B,C). We also examined the transcriptional activity of NFATc1 using a luciferase reporter assay, thereafter. It was shown that the activation of NFATc1 by RANKL was inhibited by D.P ( Figure 7A). Concluding the results, NFATc1 expression and activity were both inhibited by D.P.

| D ISCUSS I ON
Over-activation of osteoclast function is one of the mechanisms causing osteolytic diseases, such as osteoporosis, osteonecrosis and Paget's disease. 20,[27][28][29] Current treatments of these diseases are limited, expensive and would cause various side effects. 30 Natural compounds, found in traditional medicine, have good therapeutic potentials with fewer side effects. Therefore, determining the therapeutic effects of various natural compounds on osteoclast-related diseases could be beneficial. D.P, as mentioned in the previous content, has been used in the treatment of different diseases. In this study, we have demonstrated its inhibitory effects on osteoclast formation and function.
As a result, we found that D.P was able to inhibit RANKL-induced osteoclastogenesis in vitro, in a dose-dependent manner without causing any toxicity effects on the BMMs at any effective concentrations.
By immunofluorescent microscopy, we found that the D.P-treated cells failed to form intact actin-rich podosomes after the stimulation by RANKL for 5 days, which is essential for the resorptive function of osteoclast. 31 This result is consistent with our results that the resorptive function of mature osteoclast was affected by D.P. These results suggest that D.P might serve as a potential therapeutic agent for osteoclast-related diseases. However, the effect of D.P on osteolytic diseases should be further determined using appropriate animal models. As the downstream of the JNK and NF-κB pathways, NFATc1 is the key factor in the initiation of the transcription of osteoclastic genes. 32,35 We demonstrated that both the increase in the gene and protein levels of NFATc1 and the trans-nucleus activity of NFATc1 were inhibited by D.P. In addition, NFAT signalling pathway activates the PLCγ and leads to an increase in the intracellular Ca 2+ level, F I G U R E 7 D.P inhibited NFATc1 activity and translocation to nuclei. (A) NFATc1 activity was detected using a luciferase reporter gene assay. n = 3; ***P < .001. (B) Representative images of Western blot to determine NFATc1 translocation after stimulation with RANKL for 5 days in the presence or absence of D.P. PARP-1 and tubulin were used as cytoplasmic and nuclear loading controls, respectively. (C) Representative images of the immunofluorescent microscopy co-stained for NFATc1 (green), actin belts (red) and nuclei (blue) (Scale bar = 100 μm) resulting in calcineurin protein activation and downstream auto-amplification of NFATc1. 36 It was shown by the calcium oscillation assay that the activation of Ca 2+ pathway by RANKL was inhibited by D.P.
The reduction of the osteoclast-specific genes determined by qPCR, was a result of the inhibition of NFATc1 activation by D.P. Therefore, we conclude that D.P inhibited the differentiation and function of osteoclasts by inhibiting the activation of NFATc1, which is a result of inhibiting the JNK pathway, NF-κB signalling and Ca 2+ signalling pathways ( Figure 8). These in vitro results suggested the potential therapeutic effect against osteolytic diseases.

ACK N OWLED G EM ENTS
This study was supported in part by the Australian National Health

CO N FLI C T O F I NTE R E S T S
The authors declare no conflict of interests.

AUTH O R CO NTR I B UTI O N S
In this study, Jiake Xu and Wei He designed and supervised the experi-

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.