Nepetin inhibits osteoclastogenesis by inhibiting RANKL‐induced activation of NF‐κB and MAPK signalling pathway, and autophagy

Abstract Aseptic prosthetic loosening due to wear particle–induced inflammatory osteolysis is the main cause of failure for artificial joint replacement. The inflammatory response and the production of pro‐osteoclastic factors lead to elevation of osteoclast formation and excessive activity results in extensive bone destruction around the bone‐implant interface. Here we showed that Nepetin, a natural bioactive flavonoid with proven anti‐inflammatory and anti‐proliferative properties, potently inhibited RANKL‐induced osteoclast differentiation, formation and bone resorption in vitro, and protected mice against the deleterious effects of titanium particle–induced calvarial osteolysis in vivo. Mechanistically, Nepetin attenuated RANKL‐induced activation of NF‐κB and MAPK signalling pathways and TRAF6‐dependent ubiquitination of Beclin 1 which is necessary for the induction of autophagy. In brief, our study demonstrates the potential therapeutic application of Nepetin against osteoclast‐mediated osteolytic diseases.

Osteoclasts are multinucleated giant cells that originated from the monocyte/macrophage system. 6 Osteoclasts are the unique cell in the body capable of bone resorption and hence most if not all osteolytic bone conditions are due to excessive osteoclasts activity and/or formation. 7 Two key cytokines that dictate osteoclast differentiation and are released by inflammatory cells are receptor activator of nuclear-κB ligand (RANKL) and macrophage-colony-stimulating factor (M-CSF). 8,9 M-CSF is required for osteoclast precursor survival and proliferation and up-regulated expression of surface RANK, the cognate receptor for RANKL. 10,11 The combination of RANKL and RANK activates a series of signalling cascades, via TNF receptor-associated factor (TRAF), the intracellular adaptor proteins. Activation of mitogen-activated protein kinase (MAPK) and nuclear factor-κB (NF-κB) signalling pathways that are necessary for osteoclast differentiation and function occurs via TRAF2, TRAF5 and TRAF6, with TRAF6 playing an essential role in RANKL-RANK signalling. 12 The genetic loss of TRAF6 leads to osteopetrosis phenotype in mice highlighting the important role it has in osteoclast formation and function. 13 More recently, it has also been shown that TRAF6 regulates the process of autophagy during RANKL-induced osteoclast differentiation via the ubiquitination of Beclin 1. 14 Autophagy is a highly conserved and important self-catabolic cellular process. It can degrade damaged organelles and long-lived proteins via the lysosomal system so that the raw materials could be recycled for reuse by the cell. 15 Autophagy is induced during osteoclast formation in a RANKLdependent manner and is also important for osteoclast bone resorptive process. 16 Therefore, pharmacological agents that can possess anti-inflammatory properties and the ability to suppress the RANKL/RANK/TRAF6 signalling axis have great potential in the treatment and prevention of wear particle-induced osteolysis and other osteoclast-mediated bone lytic diseases.
Nepetin, also known as Eupafolin or 6-Methoxyluteolin, is a natural active flavonoid that existed in various plants containing the common sage herb. 17 Nepetin has been proved to exhibit multiple biological effects possessing anti-apoptotic, anti-inflammatory, anti-oxidant and anti-cancer properties. 18,19 However, the effect of Nepetin on RANKL-induced osteoclast formation and function has yet to be reported. Given the profound anti-inflammatory and anti-proliferative effect of Nepetin, we hypothesized that Nepetin might be a new pharmacological agent for the attenuation of inflammation-induced osteoclast-mediated osteolysis. In this research, we showed that Nepetin potently inhibited osteoclast differentiation and bone resorption in vitro and protected mice against Ti particle-induced calvarial osteolysis in vivo.
Biochemical analyses found that Nepetin attenuated RANKLinduced activation of the NF-κB and MAPK signalling pathways, as well as TRAF6-mediated ubiquitination of Beclin 1 which is necessary for the induction of autophagy. Thus, our study provided evidence for the potential therapeutic application of Nepetin against osteoclast-mediated osteolytic diseases.

| Cell culture and in vitro osteoclast formation assay
Cells were removed from the femur and tibia of C57BL/6 mice

| Cell viability assay
The effect of Nepetin on cell viability/proliferation was tested using the CCK-8 assay kit (Beyotime Institute of Biotechnology, Shanghai, China). In brief, BMMs cultured with complete α-MEM including M-CSF (30 ng/mL) were stimulated with different concentrations (from 0.78 μmol/L to 200 μmol/L) of Nepetin for 48, 72, 96 or 120 hours. Then, CCK-8 reagent (10 μL) was used to the culture dish and incubated for 3 hours. Then, the optical density was tested at 450 nm using the Multiskan FC Microplate Photometer (Thermo Fisher Scientific, Waltham, MA, USA).

| Bone resorption assay
Sterilized and dried bovine bone discs were soaked in serum-free α-MEM at 4°C for 2 days and transferred to a 96-well plate. BMMs were seeded onto collagen-coated plates and cultured with complete α-MEM with M-CSF (30 ng/mL) and RANKL (50 ng/mL) for 3 days to form small pre-osteoclastic cells. Cells were then removed from the collagen-coated surface and centrifuged briefly for 5 minutes at 1000 × g, and then the same number of pre-osteoclastic cells was seeded onto pre-prepared bone discs and cultured with complete α-MEM with RANKL and M-CSF. Cells were allowed to attach for 6 hours and then treated without or with 1.56, 3.125 or 6.25 μmol/L Nepetin for 3 days. After the above treatment, the bone discs were air-dried and gold-plated for analysis under a Hitachi S-4800 Field Emission Scanning Electron Microscope (Tokyo, Japan).

| Quantitative real-time polymerase chain reaction (qRT-PCR)
Bone marrow monocytes/macrophages were cultured with com-

| Immunofluorescence staining
BMM-derived osteoclasts seeded on glass coverslips were cultured and stimulated with Nepetin as described in Western blot above.
After the pre-treatment process above, cells were kept in 4% PFA for 15 minutes, permeabilized in 0.5% Triton X-100 for 30 minutes and then blocked with 5% appropriate normal animal serum for 30 minutes. After extensive washes, the primary antibodies (p65, LC3, TRAF6 or Beclin 1; diluted 1:100) were used to incubate with the fixed cells at 4°C overnight. For actin staining, cells were stained with phalloidin at 4°C for 1 hour and then proceeded with DAPI staining. The day after, cells were incubated with an appropriate fluorescent-labelled secondary antibody in a dark place for 1 hour.
Cells were stained with DAPI in the dark for 5 minutes. Coverslips were mounted, and fluorescence images were captured under by fluorescence or a confocal.

| Co-immunoprecipitation (Co-IP)
BMMs were pre-treated without or with Nepetin (6.25 μmol/L) for 2 hours followed by treatment with RANKL (50 ng/mL) for 30 minutes. After RANKL stimulation, cells were lysed by Western and IP lysis buffer supplemented with PMSF protease inhibitor for 30 minutes. Cell lysates were cleared by centrifugation, and 500 μg of total protein was treated with anti-TRAF6 (5 μg) or anti-Beclin 1 (5 μg) antibodies overnight at 4°C under constant and gentle rotation.
Total protein incubated with non-specific IgG antibody was applied as a negative control. Subsequently, 20 μL of protein A/G-sepharose beads (Cell Signaling Technology) was added to each sample and then incubated at 4°C overnight under constant and gentle rotation.
The beads were denatured by boiling in 2× SDS loading buffer for 5 minutes. After centrifugation, proteins were resolved on 10% SDS-PAGE gels.

| Titanium particle-induced calvarial osteolysis
Twenty-four C57BL/6 male mice (6-week-old) were obtained from the Laboratory Animal Center of Zhejiang University (Hangzhou, Zhejiang, China). The Animal Experimental Ethics Committee of the Taizhou Hospital of Zhejiang Province (Zhejiang, China) gave its approval to all experimental schemes. Mice were randomly divided into 4 groups (6 in each group): negative control sham group (PBS treatment), positive control Ti-particle implantation group (vehicle; PBS treatment; vehicle), Ti-particle + low-dose Nepetin group (0.5 mg/ kg) and Ti-particle + high-dose Nepetin group (1.0 mg/kg). Following anaesthesia with intraperitoneal injections of 1% pentobarbital sodium (50 mg/kg), a 1-cm sagittal incision was made in the middle of the skull with a sharp blade under sterile conditions. Thirty micrograms of Ti particles were embedded under the periosteum and onto the surface around the midline suture of the calvarial bone, and then the skin was sutured closed. The sham control group received surgical operation but no implantation of Ti particles. Two days after Ti particle implantation, Nepetin at 0.5 or 1 mg/kg was intraperitoneally injected every other day for 14 days. Meanwhile, mice in the sham control and vehicle groups were injected with an equal amount of PBS every other day. There were no adverse effects or death throughout the experimental period. After 14 days, all calvaria of experimental mice were surgically removed and kept in a 4% PFA before processed for micro-CT (computed tomography) and histological assessments. where Ti particles were implanted was chosen for further analysis. For immunohistochemical staining, dewaxed and rehydrated tissue sections were submitted to heat-induced antigen retrieval.

| Histology and immunohistochemistry
Tissue sections were soaked in 0.5% Triton X-100 for 20 minutes, transferred to a wet box and soaked in 3% hydrogen peroxide/methanol solution for 30 minutes. The sections were blocked using 10% normal goat serum for 30 minutes before treated with antibody (p65 or LC3; diluted 1:400) overnight at 4°C. Tissue sections were treated with an appropriate secondary antibody for 1 h. Staining was developed by incubation with 3,5-diaminobenzidine (DAB) substrate for 10 mins, counterstained with haematoxylin for 3 minutes, differentiated with 1% hydrochloric acid, dehydrated by gradient alcohol, vitrification by dimethyl benzene, and finally sealed. The optical microscope was used for observation of sections.

| Statistical analysis
All experimental data are shown as mean ± SEM. The independent sample Student's t test was applied for comparison between two groups. One-way analysis of variance (ANOVA) was applied for comparison between multiple groups with LSD and SNK post hoc tests.
The GraphPad Prism 8.0 (San Diego, CA, USA) was applied for statistical analyses. P-values <0.05 or unless otherwise specified were considered statistically significant.

| Nepetin inhibits osteoclast differentiation in vitro
We first found the cellular cytotoxicity of Nepetin (molecular structure shown in Figure 1A) against BMM cells using the CCK-8 assay. As shown in Figure 1E,F, treatment with Nepetin during the early stages of osteoclast differentiation, that is between days 0 and 3, show that the sublethal concentration of piperidine not only effectively inhibits the formation of osteoclasts induced by RANKL, but also inhibits the expression of osteoclast-related marker genes.  Figure 2C,A marked reduction in the ability of Nepetin-treated osteoclast to resorb bone was observed. Quantitative measurement of the percentage of resorption area relative to total bone disc area further revealed a dose-dependent decrease in osteoclast bone resorption following treatment with Nepetin ( Figure 2D). Collectively, these results indicate that Nepetin also exerts anti-resorptive effects on mature osteoclast bone resorption in vitro.

| Nepetin impairs RANKL-induced early activation of the NF-κB signalling pathway and attenuates the induction of c-Fos and NFATc1
Osteoclast formation in response to RANKL requires the activation  Figure 3A,B). In a similar fashion, the activation of NF-κB after 3 days of RANKL stimulation was similarly and dose-dependently impaired following Nepetin treatment ( Figure 3C,D). Using immunofluorescence analysis, we further showed that the nuclear translocation of p65 was markedly reduced in the presence of Nepetin ( Figure 3E,F).
This was further confirmed using luciferase gene reporter assay which demonstrated a dose-dependent decrease in NF-κB activity following Nepetin treatment ( Figure 3G). Thus collectively, these results suggest that Nepetin inhibits RANKL-induced activation of NF-κB via the upstream inhibition of IKK activation and as a result prevents IκBα degradation and p65 activation and nuclear localization, which consequently impairs NF-κB transcriptional activity.

| Nepetin inhibits autophagy activation via the TRAF6-Beclin 1 pathway
Recent studies have found that the autophagic process is important for RANKL-induced osteoclast differentiation and function, 21 with TRAF6-mediated ubiquitination of Beclin 1 crucial for the induction of autophagy and osteoclast differentiation. 14 TRAF6 is a unique E3 ubiquitin ligase essential for the transduction of RANKL-RANK signalling events. 22 Using Co-IP, we confirmed that TRAF6 interacts with Beclin 1 in a RANKL-dependent manner and that binding of TRAF6 induces ubiquitination of Beclin 1 ( Figure 5A,B).

This is consistent with the finding by Arai and colleagues (2019).
Interestingly, the treatment of BMM cells with Nepetin diminished the TRAF6-Beclin 1 interaction and completely abolished TRAF6mediated ubiquitination of Beclin 1 ( Figure 5A,B). Inhibition of TRAF6-Beclin 1 interaction by Nepetin was further verified using F I G U R E 3 Nepetin impairs RANKL-induced activation of the NF-κB signalling pathway. (A) BMMs were pre-treated without or with 6.25 μmol/L of Nepetin for 2 h followed by treatment with RANKL (50 ng/mL) for 0, 5, 10, 20, 30 or 60 min. Representative immunoblot obtained with IKKα, IκBα, IKKβ, p65, p-IKKα/β, p-IκBα, p-p65 and β-actin antibodies. immunofluorescence co-localization analysis. Compared with RANKL only treated BMMs, which shows intense co-localization of TRAF6-Beclin 1, Nepetin treatment completely prevented the co-localization signal ( Figure 5C,D). These data suggest that the induction of autophagy by RANKL is likely to be inhibited by Nepetin treatment. In line with this, the expression of autophagy-related genes including Beclin 1, LC3, Atg5 and Atg12 was dose-dependently down-regulated ( Figure 5E). A similar dose-dependent ( Figure 5F,G) and time-dependent ( Figure 5H,I) inhibition in the protein expression of TRAF6, Beclin 1 and LC3-1/II was observed. On the other hand, the expression of TRAF3 which normally undergoes degradation via the autophagy system following RANKL stimulation was dose-dependently elevated following Nepetin treatment, which further confirms the inhibition of autophagic process ( Figure 5F-I).
Using transmission electron microscopy and immunofluorescence analyses, we further observed a significant reduction in autophagic vacuoles or autophagosomes (LC3-conjugated) following Nepetin treatment ( Figure 6A F I G U R E 5 Nepetin obstructs RANKL-induced autophagy by blocking TRAF6-mediated ubiquitination of Beclin 1. (A) BMMs were pre-incubated with Nepetin (6.25 μmol/L) for 2 h followed by treatment with RANKL (50 ng/mL) for 30 min. The effect of Nepetin on TRAF6-Beclin 1 interaction was determined by Co-IP. (B) The experiment was performed as described in A. The effect of Nepetin on ubiquitination of Beclin 1 was determined by Co-IP. (C) The co-localization ratio of TRAF6 and Beclin 1 was calculated by Pearson's correlation coefficient. (D) BMMs were treated as shown in A. BMMs were incubated with TRAF6 with anti-TRAF6 antibody (rabbit origin) and anti-Beclin 1 antibody (mouse origin) followed by incubation with Alexa Fluor 488 (green)-conjugated goat (anti-rabbit) and Alexa Fluor 594 (red)-conjugated goat (anti-mouse). The cells stained for laser scanning confocal microscopy observation. (E) BMMs cultured with RANKL (50 ng/mL) without or with Nepetin (1.56, 3.125 or 6.25 μmol/L) for 5 d. The qRT-PCR was used to evaluate the effect of Nepetin on Beclin 1, LC3, Atg5 and ATg12 mRNA expression. (F) BMMs were cultured with RANKL (50 ng/mL) with or without Nepetin

| D ISCUSS I ON
Aseptic prosthetic loosening leading to implant failure is a common complication of artificial joint arthroplasty. 23 However, the cause of aseptic loosening is not fully understood but metal wear-debris particles arising from the articulating surfaces at the implant-bone interfaces have been shown to provoke the inflammatory response in the bone tissue near and around the prosthesis stimulates bone osteolysis and loosening of the implant. 24 The molecular interplay governing inflammatory response and osteoclast-mediated bone destruction is complex often involving the production of chemokines and cytokines, and cell-cell interactions that enhance osteoclast recruitment, formation and activity adjacent to and surrounding the bone-implant interface causing localized bone destruction. 25 Current pharmacological interventions including bisphosphonates, oestrogen replacements and anti-RANKL antibody (Denosumab) exhibit somewhat beneficial effects against osteoclast-mediated osteolysis. 23,26 However, serious undesirable side effects including cardiovascular events, nephrotoxicity, 27 osteonecrosis of the jaw, 28 atypical fractures 29 and malignant tumour formation are becoming increasingly prominent limiting their long-term use. Therefore, the identification of safer more effective therapeutic agents is urgently demanded.
Nepetin, also known as Eupafolin or 6-Methoxyluteolin, is a natural active flavonoid. 30 It has been confirmed to possess potent anti-inflammatory effects as well as anti-oxidant and anti-tumorigenic effects. 18,19,31 Given Nepetin's diverse biological effects of Nepetin, we thus examined whether Nepetin can affect osteoclast formation and bone resorption. In our study, Nepetin was found to potently inhibited osteoclast differentiation and bone resorption in vitro and protected mice from Ti particle-induced calvarial osteolysis in vivo by reducing the number and activity of osteoclasts. Mechanistically, we found Nepetin treatment attenuated TRAF6-induced activation of the NF-κB and MAPK signalling pathways. Interestingly, Nepetin also hindered the induction of TRAF6-mediated autophagy, a highly conserved cellular process involving the bulk degradation via the lysosomes of cytoplasmic components such as dysfunctional organelles and long-lived proteins.
Osteoclast differentiation and formation induced following the interaction of RANKL to RANK receptor (see Ref. [32] for review).
This leads to the recruitment of TRAF6, which leads to the rapid activation of signalling cascades such as MAPK and NF-κB pathways. Autophagy has gained widespread attention for its pivotal role in cell physiology and pathology processes. Autophagy is an evolutionarily conserved lysosomal dependent self-catabolic pathway critical for cell differentiation and development, and generalized maintenance of cellular homeostasis. 35 It is characterized by sequestration of cytoplasmic contents including long-lived proteins, and organelle into lysosomal vesicles termed autophagosomes for bulk degradation and subsequently degraded raw products returned to the cytosol for reuse. Recent studies have shown that autophagy is important for bone homeostasis and is involved in the regulation of bone physiology and pathophysiology. 36,37 Autophagy has been shown to be induced in response to RANKL, with the expression of autophagic proteins such as autophagy-related (Atg)4B, 5, Atg7 and Atg 12, and the ratio of LC3-II/LC3-I elevated during osteoclast differentiation and bone resorption. 21,38,39 Recently, mice with osteoclast-specific deletion of Beclin 1, an indispensable protein involved in the induction of autophagy, exhibited much better bone quality when compared to wild-type. 14 Arai et al. further showed using in vitro cellular assays that the RANKL induces autophagy during osteoclast differentiation and the loss of Beclin 1 which is required to initiate the autophagic process inhibited osteoclast differentiation. The overexpression of Beclin 1 enhanced RANKL-induced osteoclast formation. 14 Interestingly, the authors further demonstrated that the TRAF6-dependent ubiquitination of Beclin 1 is necessary for RANKL-induced autophagy and osteoclast differentiation. 14 Consistent with these findings, we also observed increased TRAF6-dependent Beclin 1 ubiquitination following RANKL stimulation and that this effect was completely abolished following Nepetin treatment. We further showed that Nepetin treatment dose-and time-dependently down-regulated the expression of LC3, Atg5 and Atg12, and inhibition of autophagy and autophagosome formation.
We confirmed the inhibition of autophagy by examining the expression of TRAF3 which was previously shown to be degraded via the autophagy system in response to RANKL stimulation. 40 TRAF3 expression was markedly elevated in BMM cells treated with Nepetin.
Interestingly, co-treatment of cells with Nepetin and rapamycin, a well-established inducer of autophagy and inhibitor of mTOR signalling, rescued the autophagy defect. This suggests that autophagy activation in osteoclasts is complex and alternative pathways can be activated to restore autophagy. However, further in-depth investigations into this alternative pathway are necessary.
In short, using computational molecular docking analysis we predicted a potential binding pocket for Nepetin in TRAF6 ( Figure   S1A). Nepetin could potentially bind leading to inhibition of TRAF6dependent ubiquitination of Beclin 1 and attenuation of autophagy. Together with the suppression of TRAF6-induced activation of NF-κB and MAPK signalling pathways, Nepetin potently inhibited osteoclast differentiation and bone resorption ( Figure S1B). We also found that nepotine had little effect on osteoblast formation ( Figure   S2). Our findings provide evidence for the use of Nepetin in the treatment of osteolysis conditions.

ACK N OWLED G EM ENTS
This study was sponsored by the National Natural Science

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no competing interests.

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.