J Bone Miner Metab DOI 10.1007/s00774-015-0683-1
ORIGINAL ARTICLE
The dynamin inhibitor dynasore inhibits bone resorption by rapidly disrupting actin rings of osteoclasts Gnanasagar J. Thirukonda1 · Shunsuke Uehara2 · Takahiro Nakayama3 · Teruhito Yamashita1 · Yukio Nakamura4 · Toshihide Mizoguchi1 · Naoyuki Takahashi1 · Kimitoshi Yagami5 · Nobuyuki Udagawa2 · Yasuhiro Kobayashi1
Received: 19 January 2015 / Accepted: 9 May 2015 © The Japanese Society for Bone and Mineral Research and Springer Japan 2015
Abstract The cytoskeletal organization of osteoclasts is required for bone resorption. Binding of dynamin with guanosine triphosphate (GTP) was previously suggested to be required for the organization of the actin cytoskeleton. However, the role of the GTPase activity of dynamin in the organization of the actin cytoskeleton as well as in the bone-resorbing activity of osteoclasts remains unclear. This study investigated the effects of dynasore, an inhibitor of the GTPase activity of dynamin, on the bone-resorbing activity of and actin ring formation in mouse osteoclasts in vitro and in vivo. Dynasore inhibited the formation of resorption pits in osteoclast cultures by suppressing actin ring formation and rapidly disrupting actin rings in osteoclasts. A time-lapse image analysis showed that dynasore shrank actin rings in osteoclasts within 30 min. The intraperitoneal administration of dynasore inhibited receptor activator of nuclear factor κB ligand (RANKL)-induced trabecular bone loss in mouse femurs. These in vitro and in vivo results suggest that the GTPase activity of dynamin G. J. Thirukonda and S. Uehara contributed equally to this work. * Yasuhiro Kobayashi [email protected] 1
Institute for Oral Science, Matsumoto Dental University, 1780 Gobara, Hiro‑oka, Shiojiri‑shi, Nagano 399‑0781, Japan
2
Department of Biochemistry, Matsumoto Dental University, Nagano 399‑0781, Japan
3
Department of Periodontology, Matsumoto Dental University, Nagano 399‑0781, Japan
4
Department of Orthopaedic Surgery, School of Medicine, Shinshu University, Nagano 390‑8621, Japan
5
Department of Oral Implantology, Matsumoto Dental University, Nagano 399‑0781, Japan
is critical for the bone-resorbing activity of osteoclasts and that dynasore is a seed for the development of novel antiresorbing agents. Keywords Osteoclast · Actin ring · Bone resorption · Dynamin · Dynasore
Introduction Osteoclasts, bone-resorbing multinucleated cells, differentiate from monocyte-macrophage lineage cells [1, 2]. Once mature osteoclasts adhere to the mineralized bone matrix through αvβ3 integrins, they develop two characteristic features, ruffled borders and sealing zones [3, 4]. Protons, chloride ions, and several proteases including cathepsin K are secreted from these ruffled borders. The sealing zone, which is observed as ring-like structures containing F-actin dots (called actin rings, or podosome belts), serves in attaching osteoclasts to bone and maintaining the acidity of the resorbing lacunae. The disruption of sealing zones has been shown to suppress the bone-resorbing activity of osteoclasts [5, 6]. We previously demonstrated that bone-resorbing osteoclasts put tartrate-resistant acid phosphatase (TRAP)-marks on dentin slices [7]. These TRAP-marks were identical to those observed inside the actin rings of bone-resorbing osteoclasts. A linear relationship was found between the number of actin rings and TRAP-marks. Thus, the detection of TRAP-marks represents a simple method for identifying pre-existing bone-resorbing osteoclasts. Dynamin, a large guanosine triphosphatase (GTPase), plays an essential role in clathrin-mediated endocytosis [8– 10]. It is also involved in cell shape and cell motility by regulating the actin cytoskeleton [11]. In osteoclasts, dynamin 2
13
is localized in the inner edge of podosome belts. Dynamins, together with c-Src, c-Cbl, and Pyk2, have been shown to regulate the formation of podosomes in osteoclasts [12, 13]. Activated c-Src destabilizes the association between c-Cbl and dynamins in osteoclasts. Furthermore, dynamin was previously shown to reduce the phosphorylation of Pyk2 Y402 in a GTPase activity-dependent manner, which in turn inhibited the binding of Pyk2 to c-Src [14, 15]. These findings suggest that dynamin regulates the cytoskeletal organization in osteoclasts as well as clathrin-dependent endocytosis. Bruzzaniti et al. [12] reported that the overexpression of dynamin increased osteoclast migration and resorption, whereas the overexpression of a dominant negative mutant form of dynamin (K44A), which failed to bind GTP, shrank the cell body and attenuated the bone-resorbing activity of osteoclasts. In contrast, a previous study demonstrated that dynamin bundled F-actin even in the presence of GTPγS, a non-hydrolyzable GTP analogue, which suggested that the GTPase activity of dynamin was not necessary for regulating the actin cytoskeleton in podocytes [16]. Dynamins 1 and 2 were recently shown to be involved in the cell-fusion process in osteoclasts and myotubes, suggesting that they regulate osteoclast differentiation as well as bone-resorbing activity [17]. However, the precise role of the GTPase activity of dynamin in cytoskeletal regulation has not yet been elucidated in osteoclasts. Dynasore, a small cell-permeable compound, inhibits the GTPase activity of dynamins 1, 2 and dynamin-related protein 1 (Drp1) in a non-competitive manner [18], which in turn inhibits endocytosis in neuronal cells [18–20]. It was also shown to inhibit the polymerization of actin in Sertoli cells, thereby inhibiting phagocytosis [21]. Therefore, dynasore represents a seed for developing new antibone-resorbing agents. In the present study, we demonstrated that the treatment of osteoclasts with dynasore inhibited the formation of actin rings, the appearance of TRAP-marks, and boneresorbing activity in vitro without affecting their survival. The administration of dynasore to mice disrupted the actin rings of osteoclasts on calvariae within 60 min in vivo and inhibited receptor activator of nuclear factor κB ligand (RANKL)-induced bone loss in 7-week-old mice. Thus, the GTPase activity of dynamin plays an important role in the bone-resorbing activity of osteoclasts.
J Bone Miner Metab
All experiments were conducted in accordance with the guidelines for studies using laboratory animals by the Matsumoto Dental University Experimental Animal Committee. 1α,25-Dihydroxyvitamin D3 [1,25(OH)2D3], prostaglandin E2 (PGE2), and collagenase were obtained from Wako (Osaka, Japan). Dynasore was purchased from Sigma (St. Louis, MO, USA), dissolved in dimethyl sulfoxide (DMSO) and stored at −30 °C. Rhodamine-conjugated phalloidin was from Molecular Probes (Eugene, OR, USA). Glutathione-S-transferase-conjugated human sRANKL (GST-RANKL) was from Oriental Yeast (Shiga, Japan). All other drugs were purchased from Sigma. Osteoclast formation in a mouse co‑culture system Primary osteoblasts were prepared from newborn mouse calvariae [7]. Bone marrow cells were prepared from the tibiae of 6- to 9-week-old male mice. Primary osteoblasts (1.0 × 106 cells/dish) were co-cultured for 7 days with bone marrow cells (1.0 × 107 cells/dish) in the presence of 1,25(OH)2D3 (10−8 M) and PGE2 (10−6 M) in α-minimal essential medium (α-MEM) containing 10 % fetal bovine serum (FBS) (JRH Bioscience, Lenexa, KS, USA) in 100mm dishes pre-coated with a 0.2 % collagen gel matrix (Nitta Gelatin, Osaka, Japan) [7, 22]. After a 7-day co-culture, the dish was treated for 20 min with 4 ml of 0.2 % collagenase solution on a shaking plate. Cells recovered from the dish were suspended in 8 ml of α-MEM containing 10 % FBS and used as osteoclast preparations. In subsequent experiments, osteoclast preparations were cultured in α-MEM containing 10 % FBS without RANKL and macrophage colony-stimulating factor (M-CSF). TRAP staining of osteoclasts on dentin slices and pit formation by osteoclasts Aliquots of the osteoclast preparation (0.1 ml) were cultured on dentin slices. After culturing for 48 h, cells were fixed with 3.7 % formaldehyde in phosphate-buffered saline (PBS) for 10 min. The cells were permeabilized for 1 min with 0.1 % Triton-X 100 in PBS and incubated for 5 min with TRAP-staining solution. Cells were then removed from the dentin slices with cotton swabs, and the slices were stained with Mayer’s hematoxylin (Sigma) to identify resorption pits [7, 22]. The area of resorption pits was measured using Image J software.
Materials and methods Animals and drugs
Actin ring formation by osteoclasts and detection of TRAP‑marks
Six-week-old male, one-week-old and newborn ddY mice, and 7-week-old female C57/BL6J mice were obtained from the Shizuoka Laboratories Animal Center (Shizuoka, Japan).
In the actin ring formation assay, aliquots of the osteoclast preparation (0.1 ml) were cultured on dentin slices. After culturing for 48 h, cells were fixed for 10 min with
13
J Bone Miner Metab
3.7 % formaldehyde in PBS. The cells were permeabilized for 5 min with 0.1 % Triton-X 100 in PBS and incubated with rhodamine-conjugated phalloidin to visualize F-actin. The distribution of F-actin was detected using a confocal laser scanning fluorescence microscope, Axiovert 200 (Carl Zeiss Japan, Tokyo, Japan) [7, 22]. The number of actin rings was counted. Cells were removed from the dentin slices with cotton swabs, and the slices were then stained with TRAP-staining solution to identify TRAP-marks [7]. The number of TRAP-marks was counted.
with PBS three times. Cells were then treated for 30 min with 0.2 % bovine serum albumin in PBS for blocking and for 16 h with mouse anti-dynamin 2 antibodies (BD Biosciences, San Jose, CA, USA) at 4 °C. After the treatment with primary antibodies, cells were treated with FITC-conjugated secondary antibodies (donkey anti-mouse IgG, Millipore, Billerica, MA, USA). F-actin was stained with rhodamine-conjugated phalloidin. The distribution of F-actin and dynamin 2 was detected using the fluorescence microscope, Axiovert 200. Observation of actin rings in mouse calvariae
Construction of an adenovirus vector for the expression of DsRed‑actin pDsRed-monomer-actin and Adeno-X adenoviral system 3 were purchased from Clontech Laboratories (Mountain View, CA, USA). DsRed-fusion human β-actin (DsRedactin) was amplified by PCR using pDsRed-monomer-actin as a template. The fragment was linked with pAdenoX vectors using an In-Fusion enzyme according to the manufacturer’s instructions. The ligated plasmid was linearized with PacI and then transfected into HEK293T cells using the X-treme GENE 9 (Roche, Mannheim, Germany). After amplification of the adenovirus, the adenoviruses were purified using the Adeno-X maxi purification kit (Clontech) according to the method described previously [23]. Time‑lapse imaging Primary osteoblasts (4.0 × 105 cells/dish) were co-cultured for 5 days with bone marrow cells (4.0 × 106 cells/dish) in the presence of 1,25(OH)2D3 (10−8 M) and PGE2 (10−6 M) in α-MEM containing 10 % FBS on 60-mm dishes pre-coated with a 0.2 % collagen gel matrix. After co-culture for 5 days, the cells were infected with the adenovirus of DsRed-actin (multiplicity of infection 100). After being cultivated for 2 days, the dish was treated with 4 ml of 0.2 % collagenase solution on a shaking plate for 20 min. Cells recovered from the dish were suspended in 2 ml of α-MEM containing 10 % FBS and used as the infected osteoclast preparation. Osteoclast preparations were seeded on a calcium-phosphatecoated glass bottom dish. After being cultivated for 4 h, the actin rings of osteoclasts were observed using an Axiovert 200 (Carl Zeiss Japan). Time-lapse images were obtained with imaging software ZEN (Carl Zeiss Japan). Fluorescent immunostaining of dynamin 2 Aliquots of the osteoclast preparation (0.1 ml) were cultured for 30 min in the presence or absence of dynasore on dentin slices. After culturing for 30 min, cells were fixed for 15 min with 4 % paraformaldehyde in PBS. The cells were permeabilized for 30 min with 0.2 % Triton-X 100 in PBS and washed
Actin rings in the calvariae of mice were observed as described by Kuroda et al. [24] with slight modifications. One-week-old ddY mice were injected intraperitoneally with dynasore or DMSO as a control. Sixty minutes after the injection, calvariae were dissected and fixed for 30 min with 4 % paraformaldehyde at room temperature. Calvariae were treated with 0.05 % collagenase in PBS at room temperature for 10 min and scraped with tweezers to remove the periosteal membrane from the endocranial surfaces of calvariae. Calvariae were then incubated for 30 min with rhodamine-conjugated phalloidin at 37 °C. The distribution of F-actin was detected using a fluorescence microscope. GST‑RANKL‑injected bone loss model GST-RANKL-injected bone loss in mice was performed as described previously [25] with slight modifications. Sevenweek-old female mice were divided into 3 groups: Control (n = 8, injected with saline and DMSO), RANKL (n = 7, injected with 1 mg/kg GST-RANKL and DMSO), and RANKL plus dynasore (n = 9, injected with 1 mg/kg GSTRANKL and 50 mg/kg dynasore). GST-RANKL or saline was injected intraperitoneally at 24-h intervals for 3 days. Dynasore or DMSO was injected 24 h before the first GST-RANKL injection. Mice were killed 90 min after the final injection of GST-RANKL. Blood samples and femurs were collected. A bone histomorphometric analysis was performed according to the method described previously [26]. Briefly, the femurs were stained with Villanueva bone stain and embedded. Sections of femurs were subjected to histomorphometric examination. µCT analysis and measurement of serum CTX and ALP activity A micro-computed tomography (μCT) analysis of the femurs of mice was performed according to the method described previously [27] with the ScanXmate-A080 (Comscan Tecno, Tokyo). A 3-dimensional structure was constructed using the software FanCT (Ratoc System Engineering, Tokyo, Japan) and bone volume/tissue volume (BV/TV) was analyzed with the image analysis software,
13
J Bone Miner Metab
Results
preparations obtained from co-cultures of mouse osteoblasts and bone marrow cells were placed on dentin slices and then treated with dynasore or vehicle. After culture for 48 h, cells were removed thoroughly from the dentin slices using cotton swabs. The slices were then stained with hematoxylin in order to visualize resorption pits. The treatment of osteoclasts with dynasore inhibited the formation of resorption pits on dentin slices in a dose-dependent manner (Fig. 1a). We also investigated the effects of dynasore on osteoclast survival (Fig. 1b). The number of osteoclasts remained unchanged in cultures treated with 12.5 μM dynasore, but decreased slightly on dentin slices treated with 50 μM dynasore, and no significant difference was observed between the treatment with 50 μM dynasore and DMSO (control). These results suggested that dynasore inhibited bone-resorbing activity of osteoclasts without affecting their survival.
Effects of dynasore on bone‑resorbing activity of osteoclasts in vitro
Effects of dynasore on polarization of osteoclasts in vitro
We examined the effects of dynasore on bone-resorbing activity and the survival of osteoclasts (Fig. 1). Osteoclast
Previous studies demonstrated that dynamin regulated the actin cytoskeleton including actin rings [6, 28]. Therefore,
Statistical analysis Results were expressed as the mean ± SD of three to five samples in in vitro experiments and seven to nine samples in in vivo experiments. Statistical analyses were performed using Student’s t test. Each experiment was repeated at least three times, and similar results were obtained.
A
0 µM
12.5 µM
50 µM
Area of pits (x 102 µm2)/unit
TRI/3D Bon (Ratoc System Engineering). Collagen type I cross-linked C-terminal telopeptide (CTX) and bonespecific alkaline phosphatase (ALP) activity in serum were measured using ELISA (RatLaps, Immunodiagnostic Systems, Scottsdale, AZ, USA) and the TRACP & ALP assay kit (Takara Bio, Shiga, Japan), respectively.
700 600 500 400 300
**
200 100 0
** 0
100 µm
0 µM
12.5 µM
50 µM
200
100 µm
Fig. 1 Effects of dynasore on bone-resorbing activity of osteoclasts. a, b Osteoclast preparations obtained from co-cultures of bone marrow cells with calvarial osteoblasts were cultured on dentin slices for 48 h in the presence of increasing concentrations of dynasore. a Bone resorbing assay—the area of resorption pits on dentin slices. After removing cells from dentin slices, the slices were stained with hematoxylin. The area of resorption pits (purple spots in the photographs)
13
50
Dynasore (µM)
Number TRAP+ cells/unit
B
12.5
150 100 50 0
0
12.5
50
Dynasore (µM)
was measured. b The number of osteoclasts on dentin slices. The dentin slices were stained for TRAP activity. Large TRAP-positive cells (>20 μm of longitudinal axis of the cell body) were counted as osteoclasts. a, b Data are expressed as the mean ± SD for three cultures. **p
ORIGINAL ARTICLE
The dynamin inhibitor dynasore inhibits bone resorption by rapidly disrupting actin rings of osteoclasts Gnanasagar J. Thirukonda1 · Shunsuke Uehara2 · Takahiro Nakayama3 · Teruhito Yamashita1 · Yukio Nakamura4 · Toshihide Mizoguchi1 · Naoyuki Takahashi1 · Kimitoshi Yagami5 · Nobuyuki Udagawa2 · Yasuhiro Kobayashi1
Received: 19 January 2015 / Accepted: 9 May 2015 © The Japanese Society for Bone and Mineral Research and Springer Japan 2015
Abstract The cytoskeletal organization of osteoclasts is required for bone resorption. Binding of dynamin with guanosine triphosphate (GTP) was previously suggested to be required for the organization of the actin cytoskeleton. However, the role of the GTPase activity of dynamin in the organization of the actin cytoskeleton as well as in the bone-resorbing activity of osteoclasts remains unclear. This study investigated the effects of dynasore, an inhibitor of the GTPase activity of dynamin, on the bone-resorbing activity of and actin ring formation in mouse osteoclasts in vitro and in vivo. Dynasore inhibited the formation of resorption pits in osteoclast cultures by suppressing actin ring formation and rapidly disrupting actin rings in osteoclasts. A time-lapse image analysis showed that dynasore shrank actin rings in osteoclasts within 30 min. The intraperitoneal administration of dynasore inhibited receptor activator of nuclear factor κB ligand (RANKL)-induced trabecular bone loss in mouse femurs. These in vitro and in vivo results suggest that the GTPase activity of dynamin G. J. Thirukonda and S. Uehara contributed equally to this work. * Yasuhiro Kobayashi [email protected] 1
Institute for Oral Science, Matsumoto Dental University, 1780 Gobara, Hiro‑oka, Shiojiri‑shi, Nagano 399‑0781, Japan
2
Department of Biochemistry, Matsumoto Dental University, Nagano 399‑0781, Japan
3
Department of Periodontology, Matsumoto Dental University, Nagano 399‑0781, Japan
4
Department of Orthopaedic Surgery, School of Medicine, Shinshu University, Nagano 390‑8621, Japan
5
Department of Oral Implantology, Matsumoto Dental University, Nagano 399‑0781, Japan
is critical for the bone-resorbing activity of osteoclasts and that dynasore is a seed for the development of novel antiresorbing agents. Keywords Osteoclast · Actin ring · Bone resorption · Dynamin · Dynasore
Introduction Osteoclasts, bone-resorbing multinucleated cells, differentiate from monocyte-macrophage lineage cells [1, 2]. Once mature osteoclasts adhere to the mineralized bone matrix through αvβ3 integrins, they develop two characteristic features, ruffled borders and sealing zones [3, 4]. Protons, chloride ions, and several proteases including cathepsin K are secreted from these ruffled borders. The sealing zone, which is observed as ring-like structures containing F-actin dots (called actin rings, or podosome belts), serves in attaching osteoclasts to bone and maintaining the acidity of the resorbing lacunae. The disruption of sealing zones has been shown to suppress the bone-resorbing activity of osteoclasts [5, 6]. We previously demonstrated that bone-resorbing osteoclasts put tartrate-resistant acid phosphatase (TRAP)-marks on dentin slices [7]. These TRAP-marks were identical to those observed inside the actin rings of bone-resorbing osteoclasts. A linear relationship was found between the number of actin rings and TRAP-marks. Thus, the detection of TRAP-marks represents a simple method for identifying pre-existing bone-resorbing osteoclasts. Dynamin, a large guanosine triphosphatase (GTPase), plays an essential role in clathrin-mediated endocytosis [8– 10]. It is also involved in cell shape and cell motility by regulating the actin cytoskeleton [11]. In osteoclasts, dynamin 2
13
is localized in the inner edge of podosome belts. Dynamins, together with c-Src, c-Cbl, and Pyk2, have been shown to regulate the formation of podosomes in osteoclasts [12, 13]. Activated c-Src destabilizes the association between c-Cbl and dynamins in osteoclasts. Furthermore, dynamin was previously shown to reduce the phosphorylation of Pyk2 Y402 in a GTPase activity-dependent manner, which in turn inhibited the binding of Pyk2 to c-Src [14, 15]. These findings suggest that dynamin regulates the cytoskeletal organization in osteoclasts as well as clathrin-dependent endocytosis. Bruzzaniti et al. [12] reported that the overexpression of dynamin increased osteoclast migration and resorption, whereas the overexpression of a dominant negative mutant form of dynamin (K44A), which failed to bind GTP, shrank the cell body and attenuated the bone-resorbing activity of osteoclasts. In contrast, a previous study demonstrated that dynamin bundled F-actin even in the presence of GTPγS, a non-hydrolyzable GTP analogue, which suggested that the GTPase activity of dynamin was not necessary for regulating the actin cytoskeleton in podocytes [16]. Dynamins 1 and 2 were recently shown to be involved in the cell-fusion process in osteoclasts and myotubes, suggesting that they regulate osteoclast differentiation as well as bone-resorbing activity [17]. However, the precise role of the GTPase activity of dynamin in cytoskeletal regulation has not yet been elucidated in osteoclasts. Dynasore, a small cell-permeable compound, inhibits the GTPase activity of dynamins 1, 2 and dynamin-related protein 1 (Drp1) in a non-competitive manner [18], which in turn inhibits endocytosis in neuronal cells [18–20]. It was also shown to inhibit the polymerization of actin in Sertoli cells, thereby inhibiting phagocytosis [21]. Therefore, dynasore represents a seed for developing new antibone-resorbing agents. In the present study, we demonstrated that the treatment of osteoclasts with dynasore inhibited the formation of actin rings, the appearance of TRAP-marks, and boneresorbing activity in vitro without affecting their survival. The administration of dynasore to mice disrupted the actin rings of osteoclasts on calvariae within 60 min in vivo and inhibited receptor activator of nuclear factor κB ligand (RANKL)-induced bone loss in 7-week-old mice. Thus, the GTPase activity of dynamin plays an important role in the bone-resorbing activity of osteoclasts.
J Bone Miner Metab
All experiments were conducted in accordance with the guidelines for studies using laboratory animals by the Matsumoto Dental University Experimental Animal Committee. 1α,25-Dihydroxyvitamin D3 [1,25(OH)2D3], prostaglandin E2 (PGE2), and collagenase were obtained from Wako (Osaka, Japan). Dynasore was purchased from Sigma (St. Louis, MO, USA), dissolved in dimethyl sulfoxide (DMSO) and stored at −30 °C. Rhodamine-conjugated phalloidin was from Molecular Probes (Eugene, OR, USA). Glutathione-S-transferase-conjugated human sRANKL (GST-RANKL) was from Oriental Yeast (Shiga, Japan). All other drugs were purchased from Sigma. Osteoclast formation in a mouse co‑culture system Primary osteoblasts were prepared from newborn mouse calvariae [7]. Bone marrow cells were prepared from the tibiae of 6- to 9-week-old male mice. Primary osteoblasts (1.0 × 106 cells/dish) were co-cultured for 7 days with bone marrow cells (1.0 × 107 cells/dish) in the presence of 1,25(OH)2D3 (10−8 M) and PGE2 (10−6 M) in α-minimal essential medium (α-MEM) containing 10 % fetal bovine serum (FBS) (JRH Bioscience, Lenexa, KS, USA) in 100mm dishes pre-coated with a 0.2 % collagen gel matrix (Nitta Gelatin, Osaka, Japan) [7, 22]. After a 7-day co-culture, the dish was treated for 20 min with 4 ml of 0.2 % collagenase solution on a shaking plate. Cells recovered from the dish were suspended in 8 ml of α-MEM containing 10 % FBS and used as osteoclast preparations. In subsequent experiments, osteoclast preparations were cultured in α-MEM containing 10 % FBS without RANKL and macrophage colony-stimulating factor (M-CSF). TRAP staining of osteoclasts on dentin slices and pit formation by osteoclasts Aliquots of the osteoclast preparation (0.1 ml) were cultured on dentin slices. After culturing for 48 h, cells were fixed with 3.7 % formaldehyde in phosphate-buffered saline (PBS) for 10 min. The cells were permeabilized for 1 min with 0.1 % Triton-X 100 in PBS and incubated for 5 min with TRAP-staining solution. Cells were then removed from the dentin slices with cotton swabs, and the slices were stained with Mayer’s hematoxylin (Sigma) to identify resorption pits [7, 22]. The area of resorption pits was measured using Image J software.
Materials and methods Animals and drugs
Actin ring formation by osteoclasts and detection of TRAP‑marks
Six-week-old male, one-week-old and newborn ddY mice, and 7-week-old female C57/BL6J mice were obtained from the Shizuoka Laboratories Animal Center (Shizuoka, Japan).
In the actin ring formation assay, aliquots of the osteoclast preparation (0.1 ml) were cultured on dentin slices. After culturing for 48 h, cells were fixed for 10 min with
13
J Bone Miner Metab
3.7 % formaldehyde in PBS. The cells were permeabilized for 5 min with 0.1 % Triton-X 100 in PBS and incubated with rhodamine-conjugated phalloidin to visualize F-actin. The distribution of F-actin was detected using a confocal laser scanning fluorescence microscope, Axiovert 200 (Carl Zeiss Japan, Tokyo, Japan) [7, 22]. The number of actin rings was counted. Cells were removed from the dentin slices with cotton swabs, and the slices were then stained with TRAP-staining solution to identify TRAP-marks [7]. The number of TRAP-marks was counted.
with PBS three times. Cells were then treated for 30 min with 0.2 % bovine serum albumin in PBS for blocking and for 16 h with mouse anti-dynamin 2 antibodies (BD Biosciences, San Jose, CA, USA) at 4 °C. After the treatment with primary antibodies, cells were treated with FITC-conjugated secondary antibodies (donkey anti-mouse IgG, Millipore, Billerica, MA, USA). F-actin was stained with rhodamine-conjugated phalloidin. The distribution of F-actin and dynamin 2 was detected using the fluorescence microscope, Axiovert 200. Observation of actin rings in mouse calvariae
Construction of an adenovirus vector for the expression of DsRed‑actin pDsRed-monomer-actin and Adeno-X adenoviral system 3 were purchased from Clontech Laboratories (Mountain View, CA, USA). DsRed-fusion human β-actin (DsRedactin) was amplified by PCR using pDsRed-monomer-actin as a template. The fragment was linked with pAdenoX vectors using an In-Fusion enzyme according to the manufacturer’s instructions. The ligated plasmid was linearized with PacI and then transfected into HEK293T cells using the X-treme GENE 9 (Roche, Mannheim, Germany). After amplification of the adenovirus, the adenoviruses were purified using the Adeno-X maxi purification kit (Clontech) according to the method described previously [23]. Time‑lapse imaging Primary osteoblasts (4.0 × 105 cells/dish) were co-cultured for 5 days with bone marrow cells (4.0 × 106 cells/dish) in the presence of 1,25(OH)2D3 (10−8 M) and PGE2 (10−6 M) in α-MEM containing 10 % FBS on 60-mm dishes pre-coated with a 0.2 % collagen gel matrix. After co-culture for 5 days, the cells were infected with the adenovirus of DsRed-actin (multiplicity of infection 100). After being cultivated for 2 days, the dish was treated with 4 ml of 0.2 % collagenase solution on a shaking plate for 20 min. Cells recovered from the dish were suspended in 2 ml of α-MEM containing 10 % FBS and used as the infected osteoclast preparation. Osteoclast preparations were seeded on a calcium-phosphatecoated glass bottom dish. After being cultivated for 4 h, the actin rings of osteoclasts were observed using an Axiovert 200 (Carl Zeiss Japan). Time-lapse images were obtained with imaging software ZEN (Carl Zeiss Japan). Fluorescent immunostaining of dynamin 2 Aliquots of the osteoclast preparation (0.1 ml) were cultured for 30 min in the presence or absence of dynasore on dentin slices. After culturing for 30 min, cells were fixed for 15 min with 4 % paraformaldehyde in PBS. The cells were permeabilized for 30 min with 0.2 % Triton-X 100 in PBS and washed
Actin rings in the calvariae of mice were observed as described by Kuroda et al. [24] with slight modifications. One-week-old ddY mice were injected intraperitoneally with dynasore or DMSO as a control. Sixty minutes after the injection, calvariae were dissected and fixed for 30 min with 4 % paraformaldehyde at room temperature. Calvariae were treated with 0.05 % collagenase in PBS at room temperature for 10 min and scraped with tweezers to remove the periosteal membrane from the endocranial surfaces of calvariae. Calvariae were then incubated for 30 min with rhodamine-conjugated phalloidin at 37 °C. The distribution of F-actin was detected using a fluorescence microscope. GST‑RANKL‑injected bone loss model GST-RANKL-injected bone loss in mice was performed as described previously [25] with slight modifications. Sevenweek-old female mice were divided into 3 groups: Control (n = 8, injected with saline and DMSO), RANKL (n = 7, injected with 1 mg/kg GST-RANKL and DMSO), and RANKL plus dynasore (n = 9, injected with 1 mg/kg GSTRANKL and 50 mg/kg dynasore). GST-RANKL or saline was injected intraperitoneally at 24-h intervals for 3 days. Dynasore or DMSO was injected 24 h before the first GST-RANKL injection. Mice were killed 90 min after the final injection of GST-RANKL. Blood samples and femurs were collected. A bone histomorphometric analysis was performed according to the method described previously [26]. Briefly, the femurs were stained with Villanueva bone stain and embedded. Sections of femurs were subjected to histomorphometric examination. µCT analysis and measurement of serum CTX and ALP activity A micro-computed tomography (μCT) analysis of the femurs of mice was performed according to the method described previously [27] with the ScanXmate-A080 (Comscan Tecno, Tokyo). A 3-dimensional structure was constructed using the software FanCT (Ratoc System Engineering, Tokyo, Japan) and bone volume/tissue volume (BV/TV) was analyzed with the image analysis software,
13
J Bone Miner Metab
Results
preparations obtained from co-cultures of mouse osteoblasts and bone marrow cells were placed on dentin slices and then treated with dynasore or vehicle. After culture for 48 h, cells were removed thoroughly from the dentin slices using cotton swabs. The slices were then stained with hematoxylin in order to visualize resorption pits. The treatment of osteoclasts with dynasore inhibited the formation of resorption pits on dentin slices in a dose-dependent manner (Fig. 1a). We also investigated the effects of dynasore on osteoclast survival (Fig. 1b). The number of osteoclasts remained unchanged in cultures treated with 12.5 μM dynasore, but decreased slightly on dentin slices treated with 50 μM dynasore, and no significant difference was observed between the treatment with 50 μM dynasore and DMSO (control). These results suggested that dynasore inhibited bone-resorbing activity of osteoclasts without affecting their survival.
Effects of dynasore on bone‑resorbing activity of osteoclasts in vitro
Effects of dynasore on polarization of osteoclasts in vitro
We examined the effects of dynasore on bone-resorbing activity and the survival of osteoclasts (Fig. 1). Osteoclast
Previous studies demonstrated that dynamin regulated the actin cytoskeleton including actin rings [6, 28]. Therefore,
Statistical analysis Results were expressed as the mean ± SD of three to five samples in in vitro experiments and seven to nine samples in in vivo experiments. Statistical analyses were performed using Student’s t test. Each experiment was repeated at least three times, and similar results were obtained.
A
0 µM
12.5 µM
50 µM
Area of pits (x 102 µm2)/unit
TRI/3D Bon (Ratoc System Engineering). Collagen type I cross-linked C-terminal telopeptide (CTX) and bonespecific alkaline phosphatase (ALP) activity in serum were measured using ELISA (RatLaps, Immunodiagnostic Systems, Scottsdale, AZ, USA) and the TRACP & ALP assay kit (Takara Bio, Shiga, Japan), respectively.
700 600 500 400 300
**
200 100 0
** 0
100 µm
0 µM
12.5 µM
50 µM
200
100 µm
Fig. 1 Effects of dynasore on bone-resorbing activity of osteoclasts. a, b Osteoclast preparations obtained from co-cultures of bone marrow cells with calvarial osteoblasts were cultured on dentin slices for 48 h in the presence of increasing concentrations of dynasore. a Bone resorbing assay—the area of resorption pits on dentin slices. After removing cells from dentin slices, the slices were stained with hematoxylin. The area of resorption pits (purple spots in the photographs)
13
50
Dynasore (µM)
Number TRAP+ cells/unit
B
12.5
150 100 50 0
0
12.5
50
Dynasore (µM)
was measured. b The number of osteoclasts on dentin slices. The dentin slices were stained for TRAP activity. Large TRAP-positive cells (>20 μm of longitudinal axis of the cell body) were counted as osteoclasts. a, b Data are expressed as the mean ± SD for three cultures. **p
