Dehydrocostus lactone attenuates osteoclastogenesis and osteoclast‐induced bone loss by modulating NF‐κB signalling pathway

Abstract Osteolysis is characterized by overactivated osteoclast formation and potent bone resorption. It is enhanced in many osteoclast‐related diseases including osteoporosis and periprosthetic osteolysis. The shortage of effective treatments for these pathological processes emphasizes the importance of screening and identifying potential regimens that could attenuate the formation and function of osteoclasts. Dehydrocostus lactone (DHE) is a natural sesquiterpene lactone containing anti‐inflammatory properties. Here, we showed that DHE suppressed receptor activator of nuclear factor‐κB ligand (RANKL)‐induced osteoclast formation and osteoclast marker gene expression. It also inhibited F‐actin ring formation and bone resorption in a dose‐dependent manner in vitro. Moreover, DHE inhibited the RANKL‐induced phosphorylation of NF‐κB, mitigated bone erosion in vivo in lipopolysaccharide‐induced inflammatory bone loss model and particle‐induced calvarial osteolysis model. Together, these results suggest that DHE reduces osteoclast‐related bone loss via the modulation of NF‐κB activation during osteoclastogenesis indicating that it might be a useful treatment for osteoclast‐related skeletal disorders.

macrophage colony-stimulating factor (M-CSF). 3 Receptor activator of nuclear factor-κB ligand belongs to the tumour necrosis factor (TNF) superfamily; binding with its receptor, RANK, on the surface of osteoclast precursors recruits adaptor TNF receptor-associated factors (TRAFs). This process is important for osteoclast formation because it activates several downstream signalling pathways and the transcription factor activator protein 1 (AP-1), leading to full activation of T-cell cytoplasmic 1 (NFATc1). 4 The expression of NFATc1 modulates osteoclast differentiation and functions via the induction of osteoclast-specific genes, including those encoding dendritic cell-specific transmembrane protein (DC-STAMP), TRAP, cathepsin K, calcitonin receptor (CTR), and matrix metalloproteinase 9 (MMP-9). 5,6 Natural products have been an increasingly important source for drug development. 7 We recently identified dehydrocostus lactone (DHE) as an effective inhibitor of osteoclastogenesis and osteoclast-related bone loss. Dehydrocostus lactone is a natural sesquiterpene lactone derived from the roots of Saussurea lappa, and it has been reported to have anti-inflammatory, anti-ulcer, anti-tumour, and immunomodulatory properties. 8,9 However, the effect of DHE on osteoclast formation has not been reported. It has been shown that proinflammatory cytokines promote osteoclastogenesis, 10 and overactivation of osteoclast activity plays an essential part in the origin of bone homeostatic imbalances. 11 Thus, osteoclasts could be a key target in the treatment of inflammatory bone loss. Lipopolysaccharide (LPS) and titanium particles are both potent inducers of the immune system, and they play critical roles in the development of osteolytic bone loss. 12 The purpose of this study was therefore to investigate the effects of DHE on osteoclastogenesis and on murine models of LPS-induced bone loss and particle-induced calvarial osteolysis, and to characterize the underlying mechanism of these processes.

| Reagents and antibodies
Foetal bovine serum and Minimum Essential Medium Eagle Alpha Modification (α-MEM) were obtained from Gibco (Sydney, Australia).

| Cell culture and osteoclast differentiation
Primary murine bone monocyte/macrophage precursors were isolated from the unfractionated long bones of 6-week-old C57BL/6 mice and differentiated into bone marrow-derived macrophages (BMMs) in α-MEM containing 30 ng/mL of M-CSF for 3-4 days.
The cells were cultured in at 37°C in 5% CO 2 incubator, and the medium was changed every 2 days. The effect of DHE on BMM viability was determined using a CCK-8 assay. Cells were plated into 96-well culture plates at 8 × 10 3 /well for 1 day, and then incubated with different concentrations of DHE for 4 days. The cells were then incubated for 4 hours after the addition of 10 μL of CCK-8 buffer to each well, and the optical density (OD) at 450 nm was measured on an ELX808 absorbance microplate reader

| Immunofluorescence staining
The effect of DHE on F-actin ring formation was visualized using rhodamine-phalloidin staining. Bone marrow-derived macrophages were plated into 96-well plates and treated with the indicated concentrations of DHE in the presence of 30 ng/mL of M-CSF and 100 ng/mL of RANKL for 5 days. The cells were then fixed in 4% paraformaldehyde at room temperature; and permeabilized with 0.1% Triton X-100 in phosphate-buffered saline. F-actin was stained with rhodamine-phalloidin in 2% bovine serum albumin for 1 hour. The nuclei were stained with 4′,6-diamidino-2-phenylindole for 5 minutes. Fluorescence images were acquired using a Leica DMI4000B fluorescence microscope (Wetzlar, Germany). Three fields were randomly selected, and the numbers and sizes of the F-actin rings were calculated using Imagej (NIH).

| Bone resorption pit assay
The effects of DHE on osteolytic bone resorption in vitro were assessed on bone discs. An equal number of BMM-derived pre-osteoclasts were seeded onto bone discs and incubated overnight. The cells were then treated with or without 4 μmol/L DHE in osteoclastogenic medium for 2-3 days. The discs were fixed with 4% paraformaldehyde, and adhered cells were removed by gentle brushing. Three random views in each disc were visualized using a Hitachi S-3700N Scanning Electron Microscopy (Tokyo, Japan), and the resorbed areas were quantified by Imagej (NIH).

| Western blot analysis
To investigate the effects of DHE on RANKL-induced signalling pathways, RAW264.7 cells were seeded into six-well plates at 8 × 10 5 / well in complete α-MEM and incubated overnight. Cells were then pretreated with or without 4 μmol/L DHE for 4 hours before being stimulated by 50 ng/mL of RANKL for 0, 5, 15, 30, or 60 minutes.
To determine the effects of DHE on osteoclast-related protein expression, BMMs were seeded into six-well plates at 24 × 10 4  The membranes were blocked with 5% (w/v) nonfat milk in 0.1% Tris-buffered saline and Tween 20 for 1 hour at room temperature, and then incubated with primary antibodies at 4°C overnight with gentle shaking. After washing and incubation with specific secondary antibodies for 1 hour at room temperature, the bands were detected using Enhanced Chemiluminescent Detection Reagent (Fude Biological Technology) with the Bio-Rad XRS Chemiluminescence Detection System (Hercules, CA, USA).

| Murine model of LPS-induced bone loss and particle-induced calvarial osteolysis
This study was carried out in strict accordance with the protocols

| Statistical analysis
Data are expressed as the mean ± SD. All experiments were independently carried out at least three times. Unpaired Student's t tests were used to compare two groups, and one-way ANOVA tests were used for ≥three groups, with SPSS for Windows, version 19.0 (IBM Crop., Armonk, NY). A value of P < 0.05 indicated a significant difference between groups.

| DHE inhibited the differentiation of BMMs into osteoclasts without cytotoxicity
The cytotoxicity of DHE ( Figure 1A) toward osteoclast precursor cells was determined by the CCK-8 assay with a range of doses of DHE, and the results indicate that the IC 50 of DHE for BMMs was 17.5 μmol/L ( Figure 1B).
To investigate the effect of DHE on osteoclast formation, BMMs were cul-  Figure 1F,G). Together, these results showed that DHE inhibited osteoclast differentiation without cytotoxicity.

| DHE interfered with F-actin ring formation and bone resorption in vitro
The formation of F-actin rings is a prerequisite to the adhesion of osteoclasts to bone. To determine whether DHE impaired the formation of F-actin rings, BMMs were seeded into 24-well plates and

| DHE attenuated RANKL-induced NF-κB signalling activation
The downstream pathways of RANK are important for osteoclast differentiation, so we determined whether DHE modulated the phosphorylation of MAPK and NF-κB after RANKL stimulation. The phosphorylation peaked within 60 minutes after RANKL stimulation (Figure 4). The phosphorylation of MAPK pathway was unaffected. RANKL-induced NF-κB activation is also crucial for osteoclast formation. As shown in Figure, activation of the upstream IKK complex, p65, and IκBα was suppressed by DHE.
Taken together, these results suggest that DHE inhibited osteoclast formation by modulating the RANKL-induced activation of NF-κB signalling.

| DHE impaired NFATc1 induction
NFATc1 is an expression inducer of osteoclast marker genes, and a master transcription factor during osteoclast formation. 13 To determine whether DHE attenuated the expression of NFATc1, we investigated protein expression in the presence of DHE during osteoclastogenesis of BMMs. Figure 5 shows that the expression of NFATc1 reached a peak at day 3, while its expression was significantly down-regulated in the presence of DHE. In addition, it attenuated the expression of cathepsin K and c-Src in BMMs during osteoclastogenesis ( Figure 5), which is critical for osteoclast formation and function.

| The LPS-induced bone volume loss and particle-induced calvarial osteolysis were partially rescued by DHE
To characterize the antiresorptive property of DHE in vivo, models of LPS-induced bone loss and particle-induced calvarial osteolysis were adopted. Mice that received intraperitoneal LPS showed a significant bone volume loss at the secondary spongiosa of the tibia, while mice that received LPS + DHE showed a dose-dependent increase in bone volume at this region ( Figure 6A-C). Three-dimensional micro-CT confirmed massive trabecular bone loss in LPS-treated mice compared with control mice ( Figure 6B). Consistent with the histological data, a morphometric study by micro-CT indicated significant reductions in BV/TV, Tb.Th, and Tb.N, and a remarkable increase in Tb.Sp after LPS administration. Treatment with DHE partially prevented inflammatory bone volume loss in a dose-dependent manner. What's more, protective effects of DHE were also observed in a model of particle-induced calvarial osteolysis. The resorption area of calvaria was significantly reduced in DHE treated groups ( Figure 6D-F). Together, these results confirmed the protective effects of DHE on osteoclast-related bone loss in vivo.

| D ISCUSS I ON
DHE is a naturally occurring sesquiterpene lactone reported to have many biological effects, including anticancer, bactericidal and antioxidative stress properties. 9,14,15 However, there has been no report examining the effects of DHE on osteoclasts. In this study, we investigated its effects on RANKL-induced osteoclast formation and function, and determined the underlying mechanism in vitro. We also examined its effects on LPS-induced bone loss and particle-induced calvarial osteolysis in vivo.
NFATc1 is a central regulator of the expression of marker genes, such as DC-STAMP, cathepsin K, CTR, TRAP, MMP-9 and V-ATPase a3 in the differentiation of precursors into functional osteoclasts.
In this study, we showed that treatment with DHE suppressed the mRNA expression of osteoclast marker genes, including those encoding NFATc1, DC-STAMP, cathepsin K, CTR, TRAP, MMP-9, V-ATPase a3, and OSCAR. An in vitro study also confirmed the inhibitory effects of DHE on F-actin ring formation, which is essential reported in many studies. 18,19 TRAF6 is recruited after the binding of RANKL and RANK, and it eventually forms a complex with TAK1 and TAK1-binding protein 2. 20 The complex activates TAK1, leads to the stimulation of NF-κB and AP-1 via of IKK and p38 phosphorylation, respectively. 21 The current results indicate that DHE inhibited the phosphorylation of NF-κB signalling, and attenuated the expression of c-Src, cathepsin K and NFATc1. However, DHE had little effect on the RANKL-induced phosphorylation of MAPK pathway.
The key role of osteoclasts in inflammatory bone loss makes it attractive as a potential therapeutic agent. Lipopolysaccharide is a membrane component of Gram-negative bacteria and has been used as a potent inducer of osteolytic bone loss. 22 Studies have suggested that microbial-associated molecular patterns can adhere to generate wear debris in arthroplasty cases and lead to substantial inflammation 23 ; among them, LPS is the best known. It is capable of promoting the secretion of proinflammatory cytokines 24

| CON CLUS ION
Our findings demonstrate that DHE attenuated RANKL-induced osteoclast formation and inhibited F-actin ring formation and bone resorption in vitro. Moreover, DHE partially restored LPS-induced bone loss in cancellous bone and particle-induced calvarial osteolysis in vivo. Its mechanism involved suppression of the expression of NFATc1 and other marker genes during osteoclast formation by targeting the activation of NF-κB signalling. These data suggest that DHE could be used for the treatment of osteoclast-related skeletal disorders.

ACK N OWLED G EM ENTS
This work was supported by grants from the National Science Foundation of China (81772360 to Shigui Yan, 81401785 to Xiang Zhao) and from the Natural Science Foundation of Zhejiang Province (LY17H060004, Y17H060027 to Haobo Wu).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R S' CO NTR I B UTI O N
BHu, FFW and ZLS performed the research, BHu and XZ analysed the data, BHe performed part of the in vivo study, BHu, HBW and SGY designed the research study and wrote the paper.