Hedgehog Inhibitors Suppress Osteoclastogenesis in In Vitro Cultures, and Deletion of Smo in Macrophage/Osteoclast Lineage Prevents Age-Related Bone Loss.

The functional role of the Hedgehog (Hh)-signaling pathway has been widely investigated in bone physiology/development. Previous studies have, however, focused primarily on Hh functions in bone formation, while its roles in bone resorption have not been fully elucidated. Here, we found that cyclopamine (smoothened (Smo) inhibitor), GANT-58 (GLI1 inhibitor), or GANT-61 (GLI1/2 inhibitor) significantly inhibited RANKL-induced osteoclast differentiation of bone marrow-derived macrophages. Although the inhibitory effects were exerted by cyclopamine or GANT-61 treatment during 0–48 h (early stage of osteoclast differentiation) or 48–96 h (late stage of osteoclast differentiation) after RANKL stimulation, GANT-58 suppressed osteoclast formation only during the early stage. These results suggest that the Smo-GLI1/2 axis mediates the whole process of osteoclastogenesis and that GLI1 activation is requisite only during early cellular events of osteoclastogenesis. Additionally, macrophage/osteoclast-specific deletion of Smo in mice was found to attenuate the aging phenotype characterized by trabecular low bone mass, suggesting that blockage of the Hh-signaling pathway in the osteoclast lineage plays a protective role against age-related bone loss. Our findings reveal a specific role of the Hh-signaling pathway in bone resorption and highlight that its inhibitors show potential as therapeutic agents that block osteoclast formation in the treatment of senile osteoporosis.


Introduction
Hedgehog (Hh) signaling, well-known as a mitogen and morphogen during animal development, also regulates adult tissue homeostasis and tumorigenesis [1][2][3][4]. Hh ligands in mammals are three proteins, Indian hedgehog (Ihh), Sonic hedgehog, and Desert hedgehog; although they are expressed in different cells and tissues, their functions are not considered different [5,6]. Hh ligands bind to and inhibit main receptor Patched-1 (PTCH1, encoded by the Ptch1 gene), permitting the activation of

Changes in Expression of Hedgehog (Hh) Signaling-Related Genes during Osteoclast Differentiation
First examined were the changes in the expression of Hh signaling-related genes during osteoclast differentiation. Primary cultured bone marrow-derived macrophages (BMMs) differentiated into mature osteoclasts (mOC) 96 h after RANKL stimulation ( Figure 1A). Quantitative real-time RT-PCR (qRT-PCR) analysis revealed that Ctsk mRNA expression significantly increased after RANKL stimulation ( Figure 1B), suggesting a progressing of osteoclastic differentiation. No obvious change in Gli2 and Gli3 mRNA expression, whereas Smo and Gli1 mRNA expression decreased gradually with osteoclastic differentiation, as assessed by qRT-PCR ( Figure 1C). On the other hand, Ptch1 mRNA expression significantly increased gradually with osteoclastic differentiation ( Figure 1C). These results suggest that Hh signaling in the macrophage/osteoclast lineage is associated with osteoclastic differentiation. differentiation. Moreover, macrophage/osteoclast lineage-specific Smo gene deficiency protected from age-related bone loss. Thus, we provide evidence that Hh signaling in the macrophage/osteoclast lineage mediates osteoclastogenesis in vitro and in vivo.

Changes in Expression of Hedgehog (Hh) Signaling-Related Genes during Osteoclast Differentiation
First examined were the changes in the expression of Hh signaling-related genes during osteoclast differentiation. Primary cultured bone marrow-derived macrophages (BMMs) differentiated into mature osteoclasts (mOC) 96 h after RANKL stimulation ( Figure 1A). Quantitative real-time RT-PCR (qRT-PCR) analysis revealed that Ctsk mRNA expression significantly increased after RANKL stimulation ( Figure 1B), suggesting a progressing of osteoclastic differentiation. No obvious change in Gli2 and Gli3 mRNA expression, whereas Smo and Gli1 mRNA expression decreased gradually with osteoclastic differentiation, as assessed by qRT-PCR ( Figure 1C). On the other hand, Ptch1 mRNA expression significantly increased gradually with osteoclastic differentiation ( Figure 1C). These results suggest that Hh signaling in the macrophage/osteoclast lineage is associated with osteoclastic differentiation. . The abundance of target mRNA was normalized by that of Actb mRNA. Data are represented as means ± s.d. One-way ANOVA, followed by Tukey-Kramer's test; NS, not significant; ** p < 0.01; *** p < 0.001.

Cyclopamine Suppresses Osteoclastogenesis
Since treatment with SMO inhibitor cyclopamine increases trabecular bone volume because of reduced bone resorption in mice [27], the effect of cyclopamine treatment on primary osteoclastic cultures was examined. Treatment with cyclopamine significantly decreased formation of TRAP + multinuclear cells (MNCs) in a dose-dependent manner (Figure 2A). Although no recombinant Hh ligand proteins were added to the osteoclastic cultures, cyclopamine treatment suppressed . The abundance of target mRNA was normalized by that of Actb mRNA. Data are represented as means ± s.d. One-way ANOVA, followed by Tukey-Kramer's test; NS, not significant; ** p < 0.01; *** p < 0.001.

Cyclopamine Suppresses Osteoclastogenesis
Since treatment with SMO inhibitor cyclopamine increases trabecular bone volume because of reduced bone resorption in mice [27], the effect of cyclopamine treatment on primary osteoclastic cultures was examined. Treatment with cyclopamine significantly decreased formation of TRAP + multinuclear cells (MNCs) in a dose-dependent manner (Figure 2A). Although no recombinant Hh ligand proteins were added to the osteoclastic cultures, cyclopamine treatment suppressed osteoclastogenesis. Possibly, therefore, Hh ligands contained in FBS may be important for osteoclast formation in vitro. That the addition of recombinant Sonic hedgehog enhances osteoclast formation in RAW264.7 cells [28] suggests that Hh ligands mediate osteoclastogenesis. Next, to investigate when it suppresses osteoclast differentiation, cyclopamine treatment was implemented over terms of 0-48 h, 48-96 h, and 0-96 h after RANKL stimulation. The outcome disclosed that the suppression was induced during all terms, as assessed by TRAP staining and a TRAP activity assay ( Figure 2B). Moreover, quantitative real-time RT-PCR analysis revealed that treatment with cyclopamine significantly decreased osteoclastic marker genes Tnfrsf11a (encoding RANK), Ctsk, Acp5 (encoding TRAP), Nfatc1, and Dcstamp without affecting Calcr (encoding Calcitonin receptor) at 96 h after RANKL stimulation ( Figure 2C). These results suggest that cyclopamine induces inhibitory effects at all stages of osteoclastic differentiation. osteoclastogenesis. Possibly, therefore, Hh ligands contained in FBS may be important for osteoclast formation in vitro. That the addition of recombinant Sonic hedgehog enhances osteoclast formation in RAW264.7 cells [28] suggests that Hh ligands mediate osteoclastogenesis. Next, to investigate when it suppresses osteoclast differentiation, cyclopamine treatment was implemented over terms of 0-48 h, 48-96 h, and 0-96 h after RANKL stimulation. The outcome disclosed that the suppression was induced during all terms, as assessed by TRAP staining and a TRAP activity assay ( Figure 2B). Moreover, quantitative real-time RT-PCR analysis revealed that treatment with cyclopamine significantly decreased osteoclastic marker genes Tnfrsf11a (encoding RANK), Ctsk, Acp5 (encoding TRAP), Nfatc1, and Dcstamp without affecting Calcr (encoding Calcitonin receptor) at 96 h after RANKL stimulation ( Figure 2C). These results suggest that cyclopamine induces inhibitory effects at all stages of osteoclastic differentiation.

Other Hh Signaling Inhibitors that Suppress Osteoclastogenesis
Next examined was the effect of other Hh signaling inhibitors, GANT-58 (GLI1 inhibitor) and GANT-61 (GLI1/2 inhibitor), on osteoclast differentiation. Treatment with GANT-58 and GANT-61 significantly decreased TRAP + MNC formation in a dose-dependent manner, as did the treatment with cyclopamine ( Figure 3A), strongly suggesting that Hh signaling in the macrophage/osteoclast lineage is requisite for osteoclast formation. Interestingly, GANT-58 treatment during the first 48-h term (0-48 h) was sufficient to inhibit osteoclastogenesis but not during the second term (48-96 h) ( Figure 3B). On the other hand, GANT-61 treatment exerted an inhibitory effect on osteoclast differentiation during the three terms, as did cyclopamine treatment ( Figure 3C). These results suggest that GLI1 and GLI2 activation is essential for the early and late stages, respectively, of osteoclast differentiation.   3-4). Data are represented as means ± s.d. One-way ANOVA, followed by Dunnett's test; NS, not significant; * p < 0.05, ** p < 0.01, and *** p < 0.001.

Hh Signaling Inhibitors Suppress Osteoclastic Viability
MTT cell viability assay carried out to investigate the effect of Hh signaling inhibitors on osteoclast growth revealed that RANKL stimulation increased the number of BMMs and that Hh inhibitors cyclopamine, GANT-58, and GANT-61 inhibited osteoclastic viability (Figure 4), suggesting that Hh signaling mediates osteoclastic viability.  . Hydroxyurea used as DNA synthesis inhibitor. Data are represented as means ± s.d. One-way ANOVA, followed by Dunnett's test; NS, not significant; ** p < 0.01 and *** p < 0.001.

Macrophage/osteoclast Lineage-Specific Smo Knockout Mice Exhibit Resistance to Age-Related Bone Loss
Finally, to examine the function of Hh signaling in osteoclasts in vivo, macrophage/osteoclast lineage-specific Smo knockout mice were generated by crossing LysM-Cre mice [29] with Smo-floxed mice [30]. A preliminary microcomputed tomography (μCT) analysis revealed that trabecular bone mass decreased significantly with aging in wild-type C57BL/6J mice ( Figure 5A). On the other hand, deletion of Smo in the macrophage/osteoclast lineage impaired age-related bone loss ( Figure 5B), suggesting that Hh signaling in this lineage is associated with bone reduction with aging. Data are represented as means ± s.d. One-way ANOVA, followed by Dunnett's test; NS, not significant; ** p < 0.01 and *** p < 0.001.

Macrophage/osteoclast Lineage-Specific Smo Knockout Mice Exhibit Resistance to Age-Related Bone Loss
Finally, to examine the function of Hh signaling in osteoclasts in vivo, macrophage/osteoclast lineage-specific Smo knockout mice were generated by crossing LysM-Cre mice [29] with Smo-floxed mice [30]. A preliminary microcomputed tomography (µCT) analysis revealed that trabecular bone mass decreased significantly with aging in wild-type C57BL/6J mice ( Figure 5A). On the other hand, deletion of Smo in the macrophage/osteoclast lineage impaired age-related bone loss ( Figure 5B), suggesting that Hh signaling in this lineage is associated with bone reduction with aging. Int. J. Mol. Sci. 2020, 20, x 8 of 13

Discussion
This study demonstrates that Hh inhibitors directly function as suppressors of osteoclast formation in in vitro cultures, indicating that Hh signaling in the macrophage/osteoclast lineage is

Discussion
This study demonstrates that Hh inhibitors directly function as suppressors of osteoclast formation in in vitro cultures, indicating that Hh signaling in the macrophage/osteoclast lineage is indispensable for osteoclastogenesis ( Figure 6A,B). Interestingly, Gli1 mRNA expression, an important indicator of Hh signaling [8], dynamically decreased during the differentiation of pre-osteoclasts (pOCs) into mature osteoclasts (mOCs) ( Figure 1C). Additionally, treatment with the GLI1 inhibitor GANT-58 during differentiation of pOCs into mOCs did not affect osteoclast formation in vitro ( Figure 3B). These findings indicate that GLI1 is not required for the differentiation of pOCs into mOCs. Taken together with a previous study that oral treatment with cyclopamine suppresses osteoclastic function and results in increased bone mass in mice [27], our findings suggest that Hh signaling inhibitors have a potential as anti-resorptive agents.

Reagents
Recombinant M-CSF protein (Cat# 416-ML-500) was purchased from R&D Systems, Minneapolis, MN, USA; GST-RANKL (Cat# 47197900) from Oriental Yeast, Tokyo, Japan; Osteoporosis is a major problem in public health because of its high morbidity and detriment to the quality of life [31]. As aging can be a risk factor in the development of osteoporosis regardless of gender [32], the mechanisms of age-related bone loss are pertinent to the pathogenesis of osteoporosis. Here, our preliminary µCT analysis demonstrated that macrophage/osteoclast lineage-specific deletion of the Smo gene prevents age-related bone loss. As all Hh signaling is transmitted through transmembrane protein SMO, Smo deficiency impairs the transmission of all Hh signaling [33]. Thus, we provide evidence that Hh signaling in the macrophage/osteoclast lineage is a regulator of age-related bone loss. Ablation of senescent cells pharmacologically or genetically in mice ameliorates age-related bone loss, suggesting that senescence-associated secretory phenotype (SASP) proteins, secreted by senescent cells, account for senile bone loss [34]. Possibly, therefore, Hh ligands may be a kind of SASP protein; nevertheless, no such evidence has been available to date. Since no other hypotheses have been tested, further comprehensive studies are needed to elucidate relevant underlying mechanisms.
In conclusion, this study demonstrates that Hh signaling in the macrophage/osteoclast lineage is requisite for osteoclastogenesis in in vitro cultures and that it mediates age-related bone loss in vivo, indicating the therapeutic potential of Hh inhibitors for senile bone loss.

Mice
C57BL/6J mice were purchased from Clea, Tokyo, Japan. Smo-floxed mouse [30] and LsyM-Cre mouse strains [29] were obtained from Jackson Laboratory, Bar Harbor, ME, USA. All strains were on a C57BL/6J background. All mice were maintained in a specific pathogen-free facility under climate-controlled conditions and a 12-h light/dark cycle and provided with water and standard diet (Oriental Yeast) ad libitum. All animals were handled according to the protocol approved by the Animal Experiment Committee of Ehime University, Japan (Permit No. 05-KU-36-16).

Generating M-CSF Overexpression Cells
To obtain M-CSF overexpression cells, mouse M-CSF cDNA in the pCAG-Neo mIgG2a-Fc vector (Wako) was transfected into mouse fibroblastic LMTK − cells (ATCC, Cat# CCL-1.3) with X-tremeGENE HP DNA transfection reagent (Roche Diagnostics, Basle, Switzerland). After selection with 5-mg mL −1 G418 (Wako), clones producing high M-CSF-Fc levels were selected by limiting dilution followed by ELISA for Fc-tag protein expression, and the cell culture supernatant of the highest M-CSF-Fc-producing cell line, LMF44-11, was harvested.

Osteoclast Culture
For in vitro osteoclast formation, whole bone marrow cells were harvested from tibias and femurs of eight ten-week-old mice, cultured for three days in α-MEM containing antibiotic-antimycotic solution (Gibco, Grand Island, NY, USA) and 10% FBS supplemented with LMF44-11-conditioned medium used as a source of M-CSF to obtain bone marrow-derived macrophages (BMMs). BMMs were further cultured for four days in medium supplemented with 50-ng mL −1 M-CSF and 50-ng mL −1 RANKL. The culture medium was changed every second day. Osteoclasts were identified by TRAP staining or TRAP activity assay, as described in [35,36].

RNA Isolation and Real-Time RT-PCR
RNA isolation and real-time RT-PCR were carried out as described in [35,36]. Briefly, total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA); cDNA was transcribed and used for quantitative RT-PCR conducted with Thunderbird SYBR qPCR Mix (Toyobo Co., Ltd., Osaka, Japan) in the 7500 fast real-time PCR system (Applied Biosystems, Carlsbad, CA, USA). The primers used for RT-PCR are shown in Table S1.

Statistical Analysis
All data are expressed as the means ± s.d. Statistical analyses were carried out with one-way ANOVA followed by Tukey-Kramer's test or Dunnett's test or unpaired two-sided Student's t-test (* p < 0.05, ** p < 0.01, and *** p < 0.001; NS, not significant).