N‐[2‐(4‐benzoyl‐1‐piperazinyl)phenyl]‐2‐(4‐chlorophenoxy) acetamide is a novel inhibitor of resorptive bone loss in mice

Abstract The dynamic balance between bone formation and bone resorption is vital for the retention of bone mass. The abnormal activation of osteoclasts, unique cells that degrade the bone matrix, may result in many bone diseases such as osteoporosis. Osteoporosis, a bone metabolism disease, occurs when extreme osteoclast‐mediated bone resorption outstrips osteoblast‐related bone synthesis. Therefore, it is of great interest to identify agents that can regulate the activity of osteoclasts and prevent bone loss‐induced bone diseases. In this study, we found that N‐[2‐(4‐benzoyl‐1‐piperazinyl)phenyl]‐2‐(4‐chlorophenoxy) acetamide (PPOAC‐Bz) exerted a strong inhibitory effect on osteoclastogenesis. PPOAC‐Bz altered the mRNA expressions of several osteoclast‐specific marker genes and blocked the formation of mature osteoclasts, suppressing F‐actin belt formation and bone resorption activity in vitro. In addition, PPOAC‐Bz prevented OVX‐induced bone loss in vivo. These findings highlighted the potential of PPOAC‐Bz as a prospective drug for the treatment of osteolytic disorders.

integrated with bone absorption to maintain a dynamic balance to avoid excessive loss in bone mass. 2 In many pathological conditions, bone formation is exceeded by osteoclast-mediated bone resorption, which results in excessive bone loss. 5 Osteoporosis, a common bone disease caused by the excessive formation of osteoclasts and increased resorption activity, results in a decrease in oestrogen levels. 6 Changes in osteoclast formation and overactive resorption activity also contribute to the bone destruction that occurs in osteoporosis and to osteolysis-mediated osteolytic complications of metastatic tumours such as breast cancer. 7 A recent review summarized the findings related to osteoclast regulation by several cytokines, including receptor activator of nuclear facto-kappa B (RANK), receptor activator of nuclear factor-kappa B ligand (RANKL), macrophage colony-stimulating factor (M-CSF), osteoprotegerin (OPG), interleukin-1 (IL-1), tumour necrosis factor (TNF) and interleukin-1 (IL-6). 8 Of these, the interaction between RANKL and RANK is one of the most popular review topics. 8 RANKL, a member of the TNF superfamily, stimulates the differentiation of osteoclast precursor cells into osteoclasts. 9 Marrow stromal cells and osteoblasts can produce RANKL, which has been suggested to correlate with the activation of osteoclast differentiation. 10 RANK, the receptor for RANKL, is also a TNF receptor superfamily member located on the osteoclast precursor and mature osteoclast cell surface. 9 In addition to RANKL, M-CSF also has a critical function in osteoclast formation. M-CSF can induce osteoclast differentiation from osteoclast precursors and prolong the survival of mature osteoclasts. 11 In addition, M-CSF serves as a potent stimulator for the induction of RANK expression in osteoclast precursor cells. 12 After RANKL binds with RANK, the tumour necrosis factor receptor -associated factor 6 (TRAF6) is recruited, which is capable of activating RANKL-mediated signalling pathways through autoactivation and trimerization and subsequently inducing the phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt), nuclear factor-kappa B (NFκB) and mitogen-activated protein kinases (MAPKs) signalling pathways. 13 Following their activation, nuclear factor of activated T-cell cytoplasmic 1 (NFATc1), the main transcription factor of osteoclasts, can be induced, which is able to trigger the expression of osteoclast-related genes such as cathepsin K (CtsK), and matrix metalloproteinase-9 (MMP-9). [14][15][16] The currently available and most popular clinical drugs for the treatment of bone loss-related diseases are bisphosphonate and its derivatives. 17,18 However, their treatment is always associated with side effects such as jaw and hypocalcaemia osteonecrosis, 17,18 which emphasizes the need for the discovery of new preventive drugs for the treatment of bone loss-related diseases.

| Bone marrow-derived macrophage isolation and culture
For the in vitro osteoclastogenesis assay, mice bone marrow-derived macrophages (BMMs) were collected as described previously. 13,19 Briefly, 10-week-old C57BL/6J mice were killed in a CO 2 filled box, and BMMs were isolated from the mice tibiae and femur bone via flushing the bone marrow using α-MEM. The flushed cells were collected and cultured in α-MEM supplied with 10% heat-inactivated FBS and 1% penicillin/streptomycin. On the morning of the second day, the nonadherent cells were collected, cultured in a Petri dish, and treated with M-CSF (30 ng/mL) to select the BMMs. After incubation in a cell culture incubator for 3 days, the adherent cells (BMMs) were detached using a cell-free enzyme and collected. The collected cells were further cultured in induction medium to induce osteoclast differentiation. The IACUC at Chonnam National University approved all the animal experiments (approval number: CNU IACUC-YB-2019-46).

| In vitro osteoclastogenesis and cell viability assay
BMMs (2 × 10 4 cells per well) were cultured in 48-well plates and treated with 30 ng/mL M-CSF and 50 ng/mL RANKL until the formation of mature osteoclasts was observed in the DMSO treatment group. Next, the osteoclasts were fixed in 4.0% formaldehyde (BP031) for at least 15 minutes and stained using a TRAP staining kit. The spread of the osteoclast area and the formed osteoclast

| Screening of small compound libraries
Libraries containing 52 synthetic small molecule compounds were obtained from ChemBridge (San Diego, CA, USA). Our primary screening method was image-based, using BMMs, and was performed in a 96well plate format. BMMs were treated with each molecule at 10 µmol/L or with DMSO for 3 days to allow the formation of mature osteoclasts with RANKL and M-CSF treatment by completely replacing the medium every other day. After mature osteoclasts formed in the control group, all cells were assessed using the TRAP staining assay. After airdrying for 2 days, the area of the matured osteoclasts in each group was counted using ImageJ. If the average of the total differentiated cell areas was less than that in the control group, the compounds were regarded as inhibitors of osteoclastogenesis.

| Bone resorption and F-actin belt immunofluorescence assay
A bone resorption assay kit was used to evaluate the osteoclast bone resorption activity in accordance with the manufacturer's instructions. BMMs (2 × 10 4 cells/well) were cultured in the kitsupplied coated-plate with M-CSF supplementation. On the following day, the medium was replaced, and the cells were incubated with M-CSF and RANKL and treated with or without the indicated concentrations of PPOAC-Bz until the formation of mature osteoclasts was observed. On the following day, the supernatant in each well was harvested into a black polypropylene 96-well microplate (30496; Thermo Scientific Nunc) and mixed with NaOH (S5881). Subsequently, the fluorescence intensity of each well was measured using the SpectraMax i3x fluorescence plate reader (excitation wavelength: 485 nm; emission wavelength: 535 nm).
The resorptive area was calculated based on 10 randomly selected pictures per well using the ImageJ software, as previously described. 13,19 The F-actin belts of the osteoclasts were detected using a rhodamine-conjugated phalloidin staining assay (A12379; Thermo Fisher Scientific). 20 BMMs were seeded on 12-mm cover slips

| RNA isolation and quantitative real-time PCR
BMMs were cultured in a 6-well plate in the presence or absence of PPOAC-Bz for 4 days in the induction medium. The total RNA from the BMMs was isolated using a QIAzol RNA lysis reagent (15596018; Qiagen Sciences, Valencia, CA, USA) as described in our previous study. 13 A PrimeScript™ RT reagent kit for qRT-PCR (RR420A; Takara Biotechnology, Tokyo, Japan) was used to synthesize cDNA in accordance with the manufacturer's protocol, and real-time PCR was performed with a QuantStudio 3 qRT-PCR system (Applied Biosystems, Foster City, CA, USA) together with the Power SYBR Green PCR Master Mix (4367659; Applied Biosystems, Foster City, CA, USA) and a temperature protocol provided by the company. 13,19,21 The cycle threshold values obtained were expressed as relative ratios and calculated using the 2 −ΔΔ CT method; the expression levels of the mRNA were normalized to the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression, as reported in our previous study. 13,19,21 The primers used for real-time PCR assay are listed in Table S1.

| Western blot assays
Osteoclasts were lysed using a RIPA buffer (89900; Thermo Fisher Scientific). After lysis, the cells were centrifuged at 16 400 g for 30 minutes at 4°C; the pellets were discarded, and the supernatant was retained. The concentration of protein in the supernatant was measured using a BCA protein assay, and samples with equal protein concentrations were boiled and electrophoresed on a 12% SDS-PAGE gel. The separated proteins were transferred to PVDF membranes.
After non-specific binding to the membrane was blocked via the incubation of the membrane in 5% skim milk for 1 hour, the membranes were incubated overnight with the appropriate primary antibodies (at a 1:1000 dilution) at 4°C. After three washes with TBST, the membrane was incubated at room temperature for 1 hour with HRP-conjugated secondary antibodies (at a 1:2000 dilution), and an ECL reagent was applied to detect the chemiluminescence signals, in accordance with the manufacturer's protocol.

| OVX-induced osteoporosis mouse model
The mice were housed in a specific pathogen-free facility. To eval-

| Three-point bending test
The right femur of each mouse was removed and wrapped in 0.9% NaCl-soaked gauze and then stored at −20°C. The femurs were rehydrated overnight in 0.9% NaCl at 4°C before analysis. Mouse femurs were set on the applicable mould, and the pressure sensor was set at the maximal allowable distance for each bone without compromising the test (20.0 mm for the femur). The three-point bending test was performed with a miniature materials testing machine (Instron, MA, USA). The crosshead speed descent during testing was 1 mm/min, and the force-displacement data were collected as the maximum load and slope of the bones.

| Serum biochemical analysis
The serum calcium and phosphorus levels were analysed using The mixtures were incubated for 30 minutes at RT, and the absorbance at 620 nm was measured using a microplate reader. In addition, the serum levels of osteocalcin (MK127) and CTX-1 (AC-06F1) also were analysed in accordance with the manufacturer's protocol.

| Statistical analysis
All data are expressed as the mean ± standard deviation (SD). The results are representative examples of at least three independent experiments. Statistical analysis was performed with the unpaired t test for two groups and one-way analysis of variance (ANOVA) for multiple groups; the data showed a normal distribution, and no data points were excluded. P values <.05 were considered to indicate statistical significance.

| Identification of PPOAC-Bz as a candidate disease-modifying osteoporosis drug effective in the early stages of osteoclastogenesis
An initial screen was performed among the 52 structurally diverse molecules (Table S2) to select candidate disease-modifying osteoporosis drugs ( Figure 1A). For primary screening, 52 candidate DMOPDs, at a concentration of 10 μmol/L, were screened by performing the TRAP staining assay. After the primary screening, three candidates were chosen as initial hit compounds ( Figure 1A,B).
To further determine the anti-osteoclastogenic potential of the three compounds, a TRAP staining assay was performed, and the

| Suppression of RANKL-induced bone resorption and F-actin belt formation by PPOAC-Bz treatment in vitro
For osteoclast-mediated bone resorption, the formation of actin belts is regarded as an important visual phenotype of mature osteoclasts. Hence, we performed an immunofluorescence assay to explore the effect of PPOAC-Bz on the formation of actin belts in RANKL-induced osteoclastogenesis in vitro. As shown in Figure 3A

| PPOAC-Bz inhibits the expression of osteoclast-specific markers induced by RANKL
To explore the role of PPOAC-Bz in the process of osteoclastogenesis, a real-time PCR assay was performed. As shown in Figure 4

| PPOAC-Bz attenuates the activation of the MAPK and PI3K/Akt signalling pathways in osteoclastogenesis
The RANKL-induced MAPK and Akt pathways are necessary for the activation of osteoclasts. 21 To further examine the mechanisms through which PPOAC-Bz exerted its suppressive effect on osteoclastogenesis, we examined the influence of PPOAC-Bz on the MAPK and Akt pathways after co-incubation with RANKL. The phosphorylation of MAPKs (p-p38, p-JNK and p-ERK1/2), as well as Akt (p-Akt), as shown in Figure 5A, was determined via Western blotting. Among the MAPKs, the phosphorylation of JNK and p38 did not change significantly; however, compared with the control group, in which phosphorylated ERK1/2 was significantly enhanced upon RANKL stimulation, the phosphorylated form of ERK1/2 was significantly reduced after PPOAC-Bz treatment. In the case of Akt ( Figure 5A), in the control group, p-AKT was strongly induced when stimulated by

| PPOAC-Bz suppresses osteoclastogenesis by blocking the NFκB and NFATc1 signalling pathways
In addition to the MAPK and Akt pathways, the RANKL-induced activation of the NFκB and NFATc1 signalling pathways is an essential step for the differentiation and function of osteoclasts. Mice lacking NFκB, c-fos and/or NFATc1 can develop osteopetrosis as they are unable to generate mature osteoclasts. 2,14,22 The activation of NFκB is modulated via four steps: the phosphorylation of IκBα, degradation of IκBα, phosphorylation of NFκB and nuclear translation of the p65 subunit of NFκB. 8,9,23,24 From our Western blotting analysis, shown in Figure 6A, the phosphorylation of IκBα and NFκB (p65) and the degradation of IκBα were significantly reduced, indicating that the activation of the NFκB signalling pathway was suppressed.
Next, we measured the relative expression of c-fos, NFATc1, the pivotal downstream transcription factors, and CtsK, in the presence and absence of PPOAC-Bz treatment, as shown in Figure 6B. The

| Attenuation of OVX-induced bone loss via PPOAC-Bz treatment in vivo
To further investigate the potential preventive effects of PPOAC-Bz  Table S1 lists the primers used in this experiment. *P < .05, **P < .01 and ***P < .001 indicate a statistically significant difference between the control and 2 µmol/L PPOAC-Bz treatment on each day

Fold change (mRNA) Fold change (mRNA)
regions of the left femurs and/or tibias, we found that the bone mass in the PPOAC-Bz-treated group was not lower than that in the control (OVX group), in the same place, as shown in Figure 7A. The BS/ TV, BV/TV, Tb.V, BMD, Po.Dn, and Po.N in the PPOAC-Bz-treated groups were higher ( Figure 7B) than those in the OVX group. In addition, the femurs were sectioned and subjected to histological analysis using H&E and TRAP staining. The H&E staining results suggested that the remaining trabecular bone was decreased by OVX but was rescued upon PPOAC-Bz treatment ( Figure S1A). The TRAP staining suggested that the amount of TRAP-positive cells was increased after OVX but decreased after treatment with PPOAC-Bz ( Figure S1B).
To define the biomechanical properties of the bones, the maximum load and slope of the bones were analysed using the threepoint bending test. As shown in Figure S2A,B, in the three-point bending test, the maximum load was lower in the OVX group than in the sham group; however, the difference was not significant.
Interestingly, the PPOAC-Bz treatment group had a greater maximum load than the OVX group. In addition, there was no significant difference in the slope among the sham, OVX and PPOAC-Bz treatment groups.
In the serum analysis, serum calcium levels were higher in the OVX and PPOAC-Bz treatment groups than in the sham group owing to osteoclast-induced bone resorption ( Figure S2C). However, there was no difference between the OVX and PPOAC-Bz treatment groups in the serum calcium levels ( Figure S2C). With regard to the serum phosphorous content, the OVX and PPOAC-Bz treatment groups showed a higher serum phosphorous content than the sham group. Furthermore, the increased serum phosphorous content was reduced to normal levels upon PPOAC-Bz treatment ( Figure S2D). In addition, the serum analysis of osteocalcin and CTX-1( Figure S2E,F) showed that OVX significantly increased the serum levels of OCN and CTX-1; however, PPOAC-Bz treatment greatly revised the serum CTX-1, without interference of the OCN level.

| D ISCUSS I ON
The enhanced formation of mature osteoclasts, which accompanies excessive bone resorption, can induce several bone diseases such as rheumatoid arthritis and osteoporosis. 24,25 Therefore, the prevention or delay of the formation of mature osteoclasts has emerged as one of the main targets for anti-resorptive drugs. 12,20,24,26,27 Based on our previous studies 19,21,28  subsequently, the phosphorylation of IκBα is degraded through an ubiquitin-proteasome pathway, and NF-κB transcription occurs. 23 In the IκB-independent pathway, IKK can directly phosphorylate NF-κB, which is able to modulate NF-κB transcription. 15 PPOAC-Bz exhibited an inhibitory effect on the phosphorylation of IκB and the NF-kB p65 subunit in osteoclastogenesis, indicating that the suppression of osteoclastogenesis by PPOAC-Bz may be as a result of the inactivation of IκB and NF-κB. In addition, MAPKs have key roles downstream of TRAF6. 31 The activation of MAPKs can lead to the translocation of AP-1, a vital transcription factor for mature osteoclast formation and activation, and subsequently regulate the expression of osteoclast-related genes such as CtsK and MMP9, thereby demonstrating a unique role in the process of osteoclastogenesis. 24,25,32 The MAPK signalling analysis confirmed that the expression of p-ERK1/2 was dramatically increased in the group with well-formed mature osteoclasts but was significantly decreased upon incubation with PPOAC-Bz within 1 hour; in contrast, p-JNK and p-p38 were unaffected. It was suggested that RANKL-mediated MAPKs/activator protein-1 (AP-1) and NF-κB signalling activation occur at a very early stage of osteoclast differentiation. 25 Our results showed that PPOAC-Bz could repress mature osteoclast formation only in the early stage, as shown in Clinically, denosumab, the first drug targeting osteoclast differentiation, has been approved for the treatment of malignant osteoporosis in both the United States and Europe. 37 Although it is highly efficacious and there has been a low rate of adverse events in clinical trials, the high cost of the drug has led to further interest in potential alternatives. 37,38 Besides, other currently available clinical drugs for the treatment of osteoporosis are bisphosphonate and its derivatives; however, the side effects of treatment are jaw and hypocalcaemia osteonecrosis. 17,18 Thus, novel candidates for the prevention and treatment of osteoporosis are needed. Therefore, we propose PPOAC-Bz as a potential substitute owing to its effective inhibitory effects on osteoclast differentiation and cost-effectiveness. Naturally, further exploration and discussion of the novel findings observed in this study are warranted. Bone homeostasis is a complex phenomenon that is related to both osteoclastic bone resorption and osteoblastic bone formation, which are vital events for the treatment of bone diseases. 30,39,40 However, in the current study, we have focused mainly on an investigation of the inhibitory effects of PPOAC-Bz on the formation and activation of mature osteoclasts; in further studies, we will evaluate the effect of PPOAC-Bz on bone formation and the possible mechanism of action.
Collectively, these finding suggest that PPOAC-Bz attenuated RANKL-induced osteoclastogenesis by blocking c-Src expression and NF-κB and PI3K/Akt signalling. Reductions in the activation of NF-κB and Akt and the reduction in c-Src expression mediated the down-regulation of NFATc1, subsequently leading to a decrease in the expression of osteoclast marker genes. Hence, this study provides proof-of-concept that PPOAC-Bz is a novel inhibitor of resorptive bone loss in mice.

ACK N OWLED G EM ENTS
We thank Sang Hyun Min (DGMIF, Korea) and Nam Doo Kim (VORONOI BIO Inc, Korea) for suggesting the basic structure of lowmolecular-weight substances. We would like to thank Editage (www. edita ge.co.kr) for English language editing.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript or in the decision to publish the results.

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